Fossil SCM

merge trunk

jan.nijtmans 2017-03-14 12:17 openssl-1.1 merge

Commit bec0c4f04488b7993b641a68ef6ee5903071369ff2e9750332f7fc7e0f4f8f6b

Parent 28173845f9907dc…

97 files changed +1 -1 -209 -630 -972 -687 -103 +1 -1 +9 -5 +4 -4 +11 -16 +2 +12 -10 +3 -6 +1 +6 +1 +22 -12 +29 -34 +1 -1 +22 -8 +12 -4 +90 +1 -1 +1 -1 +2 -2 +1 -1 +1 -1 +11 +5 -6 +3 -3 +7 -5 +97 +13 -2 +36 +6 -1 +3 +3 +80 -80 +15 -1 +6 -5 +24 -28 +2 -2 +7 -13 +106 -94 +2 -2 +8 +79 -11 +2 +1415 -233 +14 -13 +1 -2 +5 -2 +1515 -987 +36 -16 +35 -32 +71 -4 +1 -1 +2 -6 +2 -2 +58 -1 +6 -8 +19 -30 +2 -2 +1 -1 +96 -66 +2 -5 +4 -4 +12 -6 +1 -1 +22 -4 +36 +36 +36 +36 +30 +30 +3 -3 +7 -12 +7 -12 +20 -2 +4 -4 +1 -1 +79 -84 +4 -3 +20 +2 -2 +1 +6 +7 -6 +10 -9 +1 -1 +14 -14 +5 -4 +2 -1 +1 -1 +4 -3

~ VERSION - compat/zlib/doc/algorithm.txt - compat/zlib/doc/rfc1950.txt - compat/zlib/doc/rfc1951.txt - compat/zlib/doc/rfc1952.txt - compat/zlib/doc/txtvsbin.txt ~ src/attach.c ~ src/blob.c ~ src/browse.c ~ src/bundle.c ~ src/cgi.c ~ src/checkin.c ~ src/checkout.c ~ src/clone.c ~ src/config.h ~ src/configure.c ~ src/content.c ~ src/db.c ~ src/diff.c ~ src/diffcmd.c ~ src/doc.c ~ src/encode.c ~ src/event.c ~ src/export.c ~ src/foci.c ~ src/fusefs.c ~ src/graph.c ~ src/hname.c ~ src/http_ssl.c ~ src/import.c ~ src/info.c ~ src/login.c ~ src/main.c ~ src/main.mk ~ src/makeheaders.c ~ src/makemake.tcl ~ src/makemake.tcl ~ src/manifest.c ~ src/md5.c ~ src/mkversion.c ~ src/name.c ~ src/printf.c ~ src/purge.c ~ src/rebuild.c ~ src/schema.c ~ src/setup.c ~ src/sha1.c ~ src/sha1hard.c ~ src/sha3.c ~ src/shell.c ~ src/shun.c ~ src/sitemap.c ~ src/sqlcmd.c ~ src/sqlite3.c ~ src/sqlite3.h ~ src/stash.c ~ src/stat.c ~ src/tag.c ~ src/tar.c ~ src/timeline.c ~ src/unversioned.c ~ src/verify.c ~ src/vfile.c ~ src/wiki.c ~ src/wikiformat.c ~ src/xfer.c ~ src/zip.c ~ test/json.test ~ test/wiki.test ~ tools/man_page_command_list.tcl ~ win/Makefile.dmc ~ win/Makefile.mingw ~ win/Makefile.mingw ~ win/Makefile.mingw.mistachkin ~ win/Makefile.mingw.mistachkin ~ win/Makefile.msc ~ win/Makefile.msc ~ www/branching.wiki ~ www/build.wiki ~ www/build.wiki ~ www/changes.wiki ~ www/checkin_names.wiki ~ www/customskin.md ~ www/fileformat.wiki ~ www/fossil-v-git.wiki ~ www/hashpolicy.wiki ~ www/mkdownload.tcl ~ www/mkindex.tcl ~ www/permutedindex.html ~ www/pop.wiki ~ www/selfcheck.wiki ~ www/shunning.wiki ~ www/sync.wiki ~ www/tech_overview.wiki ~ www/theory1.wiki ~ www/webpage-ex.md ~ www/whyusefossil.wiki

M VERSION

+1 -1

		--- VERSION
		+++ VERSION
		@@ -1,1 +1,1 @@
1		-1.37
	1	+2.1
2	2
3	3	DELETED compat/zlib/doc/algorithm.txt
4	4	DELETED compat/zlib/doc/rfc1950.txt
5	5	DELETED compat/zlib/doc/rfc1951.txt
6	6	DELETED compat/zlib/doc/rfc1952.txt
7	7	DELETED compat/zlib/doc/txtvsbin.txt

	--- VERSION
	+++ VERSION
	@@ -1,1 +1,1 @@
1	1.37
2
3	ELETED compat/zlib/doc/algorithm.txt
4	ELETED compat/zlib/doc/rfc1950.txt
5	ELETED compat/zlib/doc/rfc1951.txt
6	ELETED compat/zlib/doc/rfc1952.txt
7	ELETED compat/zlib/doc/txtvsbin.txt

	--- VERSION
	+++ VERSION
	@@ -1,1 +1,1 @@
1	2.1
2
3	ELETED compat/zlib/doc/algorithm.txt
4	ELETED compat/zlib/doc/rfc1950.txt
5	ELETED compat/zlib/doc/rfc1951.txt
6	ELETED compat/zlib/doc/rfc1952.txt
7	ELETED compat/zlib/doc/txtvsbin.txt

D compat/zlib/doc/algorithm.txt

-209

		--- a/compat/zlib/doc/algorithm.txt
		+++ b/compat/zlib/doc/algorithm.txt
		@@ -1,209 +0,0 @@
1		-1. Compression algorithm (deflate)
2		-
3		-The deflation algorithm used by gzip (also zip and zlib) is a variation of
4		-LZ77 (Lempel-Ziv 1977, see reference below). It finds duplicated strings in
5		-the input data. The second occurrence of a string is replaced by a
6		-pointer to the previous string, in the form of a pair (distance,
7		-length). Distances are limited to 32K bytes, and lengths are limited
8		-to 258 bytes. When a string does not occur anywhere in the previous
9		-32K bytes, it is emitted as a sequence of literal bytes. (In this
10		-description, `string' must be taken as an arbitrary sequence of bytes,
11		-and is not restricted to printable characters.)
12		-
13		-Literals or match lengths are compressed with one Huffman tree, and
14		-match distances are compressed with another tree. The trees are stored
15		-in a compact form at the start of each block. The blocks can have any
16		-size (except that the compressed data for one block must fit in
17		-available memory). A block is terminated when deflate() determines that
18		-it would be useful to start another block with fresh trees. (This is
19		-somewhat similar to the behavior of LZW-based _compress_.)
20		-
21		-Duplicated strings are found using a hash table. All input strings of
22		-length 3 are inserted in the hash table. A hash index is computed for
23		-the next 3 bytes. If the hash chain for this index is not empty, all
24		-strings in the chain are compared with the current input string, and
25		-the longest match is selected.
26		-
27		-The hash chains are searched starting with the most recent strings, to
28		-favor small distances and thus take advantage of the Huffman encoding.
29		-The hash chains are singly linked. There are no deletions from the
30		-hash chains, the algorithm simply discards matches that are too old.
31		-
32		-To avoid a worst-case situation, very long hash chains are arbitrarily
33		-truncated at a certain length, determined by a runtime option (level
34		-parameter of deflateInit). So deflate() does not always find the longest
35		-possible match but generally finds a match which is long enough.
36		-
37		-deflate() also defers the selection of matches with a lazy evaluation
38		-mechanism. After a match of length N has been found, deflate() searches for
39		-a longer match at the next input byte. If a longer match is found, the
40		-previous match is truncated to a length of one (thus producing a single
41		-literal byte) and the process of lazy evaluation begins again. Otherwise,
42		-the original match is kept, and the next match search is attempted only N
43		-steps later.
44		-
45		-The lazy match evaluation is also subject to a runtime parameter. If
46		-the current match is long enough, deflate() reduces the search for a longer
47		-match, thus speeding up the whole process. If compression ratio is more
48		-important than speed, deflate() attempts a complete second search even if
49		-the first match is already long enough.
50		-
51		-The lazy match evaluation is not performed for the fastest compression
52		-modes (level parameter 1 to 3). For these fast modes, new strings
53		-are inserted in the hash table only when no match was found, or
54		-when the match is not too long. This degrades the compression ratio
55		-but saves time since there are both fewer insertions and fewer searches.
56		-
57		-
58		-2. Decompression algorithm (inflate)
59		-
60		-2.1 Introduction
61		-
62		-The key question is how to represent a Huffman code (or any prefix code) so
63		-that you can decode fast. The most important characteristic is that shorter
64		-codes are much more common than longer codes, so pay attention to decoding the
65		-short codes fast, and let the long codes take longer to decode.
66		-
67		-inflate() sets up a first level table that covers some number of bits of
68		-input less than the length of longest code. It gets that many bits from the
69		-stream, and looks it up in the table. The table will tell if the next
70		-code is that many bits or less and how many, and if it is, it will tell
71		-the value, else it will point to the next level table for which inflate()
72		-grabs more bits and tries to decode a longer code.
73		-
74		-How many bits to make the first lookup is a tradeoff between the time it
75		-takes to decode and the time it takes to build the table. If building the
76		-table took no time (and if you had infinite memory), then there would only
77		-be a first level table to cover all the way to the longest code. However,
78		-building the table ends up taking a lot longer for more bits since short
79		-codes are replicated many times in such a table. What inflate() does is
80		-simply to make the number of bits in the first table a variable, and then
81		-to set that variable for the maximum speed.
82		-
83		-For inflate, which has 286 possible codes for the literal/length tree, the size
84		-of the first table is nine bits. Also the distance trees have 30 possible
85		-values, and the size of the first table is six bits. Note that for each of
86		-those cases, the table ended up one bit longer than the ``average'' code
87		-length, i.e. the code length of an approximately flat code which would be a
88		-little more than eight bits for 286 symbols and a little less than five bits
89		-for 30 symbols.
90		-
91		-
92		-2.2 More details on the inflate table lookup
93		-
94		-Ok, you want to know what this cleverly obfuscated inflate tree actually
95		-looks like. You are correct that it's not a Huffman tree. It is simply a
96		-lookup table for the first, let's say, nine bits of a Huffman symbol. The
97		-symbol could be as short as one bit or as long as 15 bits. If a particular
98		-symbol is shorter than nine bits, then that symbol's translation is duplicated
99		-in all those entries that start with that symbol's bits. For example, if the
100		-symbol is four bits, then it's duplicated 32 times in a nine-bit table. If a
101		-symbol is nine bits long, it appears in the table once.
102		-
103		-If the symbol is longer than nine bits, then that entry in the table points
104		-to another similar table for the remaining bits. Again, there are duplicated
105		-entries as needed. The idea is that most of the time the symbol will be short
106		-and there will only be one table look up. (That's whole idea behind data
107		-compression in the first place.) For the less frequent long symbols, there
108		-will be two lookups. If you had a compression method with really long
109		-symbols, you could have as many levels of lookups as is efficient. For
110		-inflate, two is enough.
111		-
112		-So a table entry either points to another table (in which case nine bits in
113		-the above example are gobbled), or it contains the translation for the symbol
114		-and the number of bits to gobble. Then you start again with the next
115		-ungobbled bit.
116		-
117		-You may wonder: why not just have one lookup table for how ever many bits the
118		-longest symbol is? The reason is that if you do that, you end up spending
119		-more time filling in duplicate symbol entries than you do actually decoding.
120		-At least for deflate's output that generates new trees every several 10's of
121		-kbytes. You can imagine that filling in a 2^15 entry table for a 15-bit code
122		-would take too long if you're only decoding several thousand symbols. At the
123		-other extreme, you could make a new table for every bit in the code. In fact,
124		-that's essentially a Huffman tree. But then you spend too much time
125		-traversing the tree while decoding, even for short symbols.
126		-
127		-So the number of bits for the first lookup table is a trade of the time to
128		-fill out the table vs. the time spent looking at the second level and above of
129		-the table.
130		-
131		-Here is an example, scaled down:
132		-
133		-The code being decoded, with 10 symbols, from 1 to 6 bits long:
134		-
135		-A: 0
136		-B: 10
137		-C: 1100
138		-D: 11010
139		-E: 11011
140		-F: 11100
141		-G: 11101
142		-H: 11110
143		-I: 111110
144		-J: 111111
145		-
146		-Let's make the first table three bits long (eight entries):
147		-
148		-000: A,1
149		-001: A,1
150		-010: A,1
151		-011: A,1
152		-100: B,2
153		-101: B,2
154		-110: -> table X (gobble 3 bits)
155		-111: -> table Y (gobble 3 bits)
156		-
157		-Each entry is what the bits decode as and how many bits that is, i.e. how
158		-many bits to gobble. Or the entry points to another table, with the number of
159		-bits to gobble implicit in the size of the table.
160		-
161		-Table X is two bits long since the longest code starting with 110 is five bits
162		-long:
163		-
164		-00: C,1
165		-01: C,1
166		-10: D,2
167		-11: E,2
168		-
169		-Table Y is three bits long since the longest code starting with 111 is six
170		-bits long:
171		-
172		-000: F,2
173		-001: F,2
174		-010: G,2
175		-011: G,2
176		-100: H,2
177		-101: H,2
178		-110: I,3
179		-111: J,3
180		-
181		-So what we have here are three tables with a total of 20 entries that had to
182		-be constructed. That's compared to 64 entries for a single table. Or
183		-compared to 16 entries for a Huffman tree (six two entry tables and one four
184		-entry table). Assuming that the code ideally represents the probability of
185		-the symbols, it takes on the average 1.25 lookups per symbol. That's compared
186		-to one lookup for the single table, or 1.66 lookups per symbol for the
187		-Huffman tree.
188		-
189		-There, I think that gives you a picture of what's going on. For inflate, the
190		-meaning of a particular symbol is often more than just a letter. It can be a
191		-byte (a "literal"), or it can be either a length or a distance which
192		-indicates a base value and a number of bits to fetch after the code that is
193		-added to the base value. Or it might be the special end-of-block code. The
194		-data structures created in inftrees.c try to encode all that information
195		-compactly in the tables.
196		-
197		-
198		-Jean-loup Gailly Mark Adler
199		-[email protected] [email protected]
200		-
201		-
202		-References:
203		-
204		-[LZ77] Ziv J., Lempel A., ``A Universal Algorithm for Sequential Data
205		-Compression,'' IEEE Transactions on Information Theory, Vol. 23, No. 3,
206		-pp. 337-343.
207		-
208		-``DEFLATE Compressed Data Format Specification'' available in
209		-http://tools.ietf.org/html/rfc1951

	--- a/compat/zlib/doc/algorithm.txt
	+++ b/compat/zlib/doc/algorithm.txt
	@@ -1,209 +0,0 @@
1	1. Compression algorithm (deflate)
2
3	The deflation algorithm used by gzip (also zip and zlib) is a variation of
4	LZ77 (Lempel-Ziv 1977, see reference below). It finds duplicated strings in
5	the input data. The second occurrence of a string is replaced by a
6	pointer to the previous string, in the form of a pair (distance,
7	length). Distances are limited to 32K bytes, and lengths are limited
8	to 258 bytes. When a string does not occur anywhere in the previous
9	32K bytes, it is emitted as a sequence of literal bytes. (In this
10	description, `string' must be taken as an arbitrary sequence of bytes,
11	and is not restricted to printable characters.)
12
13	Literals or match lengths are compressed with one Huffman tree, and
14	match distances are compressed with another tree. The trees are stored
15	in a compact form at the start of each block. The blocks can have any
16	size (except that the compressed data for one block must fit in
17	available memory). A block is terminated when deflate() determines that
18	it would be useful to start another block with fresh trees. (This is
19	somewhat similar to the behavior of LZW-based _compress_.)
20
21	Duplicated strings are found using a hash table. All input strings of
22	length 3 are inserted in the hash table. A hash index is computed for
23	the next 3 bytes. If the hash chain for this index is not empty, all
24	strings in the chain are compared with the current input string, and
25	the longest match is selected.
26
27	The hash chains are searched starting with the most recent strings, to
28	favor small distances and thus take advantage of the Huffman encoding.
29	The hash chains are singly linked. There are no deletions from the
30	hash chains, the algorithm simply discards matches that are too old.
31
32	To avoid a worst-case situation, very long hash chains are arbitrarily
33	truncated at a certain length, determined by a runtime option (level
34	parameter of deflateInit). So deflate() does not always find the longest
35	possible match but generally finds a match which is long enough.
36
37	deflate() also defers the selection of matches with a lazy evaluation
38	mechanism. After a match of length N has been found, deflate() searches for
39	a longer match at the next input byte. If a longer match is found, the
40	previous match is truncated to a length of one (thus producing a single
41	literal byte) and the process of lazy evaluation begins again. Otherwise,
42	the original match is kept, and the next match search is attempted only N
43	steps later.
44
45	The lazy match evaluation is also subject to a runtime parameter. If
46	the current match is long enough, deflate() reduces the search for a longer
47	match, thus speeding up the whole process. If compression ratio is more
48	important than speed, deflate() attempts a complete second search even if
49	the first match is already long enough.
50
51	The lazy match evaluation is not performed for the fastest compression
52	modes (level parameter 1 to 3). For these fast modes, new strings
53	are inserted in the hash table only when no match was found, or
54	when the match is not too long. This degrades the compression ratio
55	but saves time since there are both fewer insertions and fewer searches.
56
57
58	2. Decompression algorithm (inflate)
59
60	2.1 Introduction
61
62	The key question is how to represent a Huffman code (or any prefix code) so
63	that you can decode fast. The most important characteristic is that shorter
64	codes are much more common than longer codes, so pay attention to decoding the
65	short codes fast, and let the long codes take longer to decode.
66
67	inflate() sets up a first level table that covers some number of bits of
68	input less than the length of longest code. It gets that many bits from the
69	stream, and looks it up in the table. The table will tell if the next
70	code is that many bits or less and how many, and if it is, it will tell
71	the value, else it will point to the next level table for which inflate()
72	grabs more bits and tries to decode a longer code.
73
74	How many bits to make the first lookup is a tradeoff between the time it
75	takes to decode and the time it takes to build the table. If building the
76	table took no time (and if you had infinite memory), then there would only
77	be a first level table to cover all the way to the longest code. However,
78	building the table ends up taking a lot longer for more bits since short
79	codes are replicated many times in such a table. What inflate() does is
80	simply to make the number of bits in the first table a variable, and then
81	to set that variable for the maximum speed.
82
83	For inflate, which has 286 possible codes for the literal/length tree, the size
84	of the first table is nine bits. Also the distance trees have 30 possible
85	values, and the size of the first table is six bits. Note that for each of
86	those cases, the table ended up one bit longer than the ``average'' code
87	length, i.e. the code length of an approximately flat code which would be a
88	little more than eight bits for 286 symbols and a little less than five bits
89	for 30 symbols.
90
91
92	2.2 More details on the inflate table lookup
93
94	Ok, you want to know what this cleverly obfuscated inflate tree actually
95	looks like. You are correct that it's not a Huffman tree. It is simply a
96	lookup table for the first, let's say, nine bits of a Huffman symbol. The
97	symbol could be as short as one bit or as long as 15 bits. If a particular
98	symbol is shorter than nine bits, then that symbol's translation is duplicated
99	in all those entries that start with that symbol's bits. For example, if the
100	symbol is four bits, then it's duplicated 32 times in a nine-bit table. If a
101	symbol is nine bits long, it appears in the table once.
102
103	If the symbol is longer than nine bits, then that entry in the table points
104	to another similar table for the remaining bits. Again, there are duplicated
105	entries as needed. The idea is that most of the time the symbol will be short
106	and there will only be one table look up. (That's whole idea behind data
107	compression in the first place.) For the less frequent long symbols, there
108	will be two lookups. If you had a compression method with really long
109	symbols, you could have as many levels of lookups as is efficient. For
110	inflate, two is enough.
111
112	So a table entry either points to another table (in which case nine bits in
113	the above example are gobbled), or it contains the translation for the symbol
114	and the number of bits to gobble. Then you start again with the next
115	ungobbled bit.
116
117	You may wonder: why not just have one lookup table for how ever many bits the
118	longest symbol is? The reason is that if you do that, you end up spending
119	more time filling in duplicate symbol entries than you do actually decoding.
120	At least for deflate's output that generates new trees every several 10's of
121	kbytes. You can imagine that filling in a 2^15 entry table for a 15-bit code
122	would take too long if you're only decoding several thousand symbols. At the
123	other extreme, you could make a new table for every bit in the code. In fact,
124	that's essentially a Huffman tree. But then you spend too much time
125	traversing the tree while decoding, even for short symbols.
126
127	So the number of bits for the first lookup table is a trade of the time to
128	fill out the table vs. the time spent looking at the second level and above of
129	the table.
130
131	Here is an example, scaled down:
132
133	The code being decoded, with 10 symbols, from 1 to 6 bits long:
134
135	A: 0
136	B: 10
137	C: 1100
138	D: 11010
139	E: 11011
140	F: 11100
141	G: 11101
142	H: 11110
143	I: 111110
144	J: 111111
145
146	Let's make the first table three bits long (eight entries):
147
148	000: A,1
149	001: A,1
150	010: A,1
151	011: A,1
152	100: B,2
153	101: B,2
154	110: -> table X (gobble 3 bits)
155	111: -> table Y (gobble 3 bits)
156
157	Each entry is what the bits decode as and how many bits that is, i.e. how
158	many bits to gobble. Or the entry points to another table, with the number of
159	bits to gobble implicit in the size of the table.
160
161	Table X is two bits long since the longest code starting with 110 is five bits
162	long:
163
164	00: C,1
165	01: C,1
166	10: D,2
167	11: E,2
168
169	Table Y is three bits long since the longest code starting with 111 is six
170	bits long:
171
172	000: F,2
173	001: F,2
174	010: G,2
175	011: G,2
176	100: H,2
177	101: H,2
178	110: I,3
179	111: J,3
180
181	So what we have here are three tables with a total of 20 entries that had to
182	be constructed. That's compared to 64 entries for a single table. Or
183	compared to 16 entries for a Huffman tree (six two entry tables and one four
184	entry table). Assuming that the code ideally represents the probability of
185	the symbols, it takes on the average 1.25 lookups per symbol. That's compared
186	to one lookup for the single table, or 1.66 lookups per symbol for the
187	Huffman tree.
188
189	There, I think that gives you a picture of what's going on. For inflate, the
190	meaning of a particular symbol is often more than just a letter. It can be a
191	byte (a "literal"), or it can be either a length or a distance which
192	indicates a base value and a number of bits to fetch after the code that is
193	added to the base value. Or it might be the special end-of-block code. The
194	data structures created in inftrees.c try to encode all that information
195	compactly in the tables.
196
197
198	Jean-loup Gailly Mark Adler
199	[email protected] [email protected]
200
201
202	References:
203
204	[LZ77] Ziv J., Lempel A., ``A Universal Algorithm for Sequential Data
205	Compression,'' IEEE Transactions on Information Theory, Vol. 23, No. 3,
206	pp. 337-343.
207
208	``DEFLATE Compressed Data Format Specification'' available in
209	http://tools.ietf.org/html/rfc1951

	--- a/compat/zlib/doc/algorithm.txt
	+++ b/compat/zlib/doc/algorithm.txt
	@@ -1,209 +0,0 @@

D compat/zlib/doc/rfc1950.txt

-630

		--- a/compat/zlib/doc/rfc1950.txt
		+++ b/compat/zlib/doc/rfc1950.txt
		@@ -1,630 +0,0 @@
1		-
2		-
3		-
4		-
5		-
6		-
7		-Network Working Group P. Deutsch
8		-Request for Comments: 1950 Aladdin Enterprises
9		-Category: Informational J-L. Gailly
10		- Info-ZIP
11		- May 1996
12		-
13		-
14		- ZLIB Compressed Data Format Specification version 3.3
15		-
16		-Status of This Memo
17		-
18		- This memo provides information for the Internet community. This memo
19		- does not specify an Internet standard of any kind. Distribution of
20		- this memo is unlimited.
21		-
22		-IESG Note:
23		-
24		- The IESG takes no position on the validity of any Intellectual
25		- Property Rights statements contained in this document.
26		-
27		-Notices
28		-
29		- Copyright (c) 1996 L. Peter Deutsch and Jean-Loup Gailly
30		-
31		- Permission is granted to copy and distribute this document for any
32		- purpose and without charge, including translations into other
33		- languages and incorporation into compilations, provided that the
34		- copyright notice and this notice are preserved, and that any
35		- substantive changes or deletions from the original are clearly
36		- marked.
37		-
38		- A pointer to the latest version of this and related documentation in
39		- HTML format can be found at the URL
40		- <ftp://ftp.uu.net/graphics/png/documents/zlib/zdoc-index.html>.
41		-
42		-Abstract
43		-
44		- This specification defines a lossless compressed data format. The
45		- data can be produced or consumed, even for an arbitrarily long
46		- sequentially presented input data stream, using only an a priori
47		- bounded amount of intermediate storage. The format presently uses
48		- the DEFLATE compression method but can be easily extended to use
49		- other compression methods. It can be implemented readily in a manner
50		- not covered by patents. This specification also defines the ADLER-32
51		- checksum (an extension and improvement of the Fletcher checksum),
52		- used for detection of data corruption, and provides an algorithm for
53		- computing it.
54		-
55		-
56		-
57		-
58		-Deutsch & Gailly Informational [Page 1]
59		-
60		-
61		-RFC 1950 ZLIB Compressed Data Format Specification May 1996
62		-
63		-
64		-Table of Contents
65		-
66		- 1. Introduction ................................................... 2
67		- 1.1. Purpose ................................................... 2
68		- 1.2. Intended audience ......................................... 3
69		- 1.3. Scope ..................................................... 3
70		- 1.4. Compliance ................................................ 3
71		- 1.5. Definitions of terms and conventions used ................ 3
72		- 1.6. Changes from previous versions ............................ 3
73		- 2. Detailed specification ......................................... 3
74		- 2.1. Overall conventions ....................................... 3
75		- 2.2. Data format ............................................... 4
76		- 2.3. Compliance ................................................ 7
77		- 3. References ..................................................... 7
78		- 4. Source code .................................................... 8
79		- 5. Security Considerations ........................................ 8
80		- 6. Acknowledgements ............................................... 8
81		- 7. Authors' Addresses ............................................. 8
82		- 8. Appendix: Rationale ............................................ 9
83		- 9. Appendix: Sample code ..........................................10
84		-
85		-1. Introduction
86		-
87		- 1.1. Purpose
88		-
89		- The purpose of this specification is to define a lossless
90		- compressed data format that:
91		-
92		- * Is independent of CPU type, operating system, file system,
93		- and character set, and hence can be used for interchange;
94		-
95		- * Can be produced or consumed, even for an arbitrarily long
96		- sequentially presented input data stream, using only an a
97		- priori bounded amount of intermediate storage, and hence can
98		- be used in data communications or similar structures such as
99		- Unix filters;
100		-
101		- * Can use a number of different compression methods;
102		-
103		- * Can be implemented readily in a manner not covered by
104		- patents, and hence can be practiced freely.
105		-
106		- The data format defined by this specification does not attempt to
107		- allow random access to compressed data.
108		-
109		-
110		-
111		-
112		-
113		-
114		-
115		-Deutsch & Gailly Informational [Page 2]
116		-
117		-
118		-RFC 1950 ZLIB Compressed Data Format Specification May 1996
119		-
120		-
121		- 1.2. Intended audience
122		-
123		- This specification is intended for use by implementors of software
124		- to compress data into zlib format and/or decompress data from zlib
125		- format.
126		-
127		- The text of the specification assumes a basic background in
128		- programming at the level of bits and other primitive data
129		- representations.
130		-
131		- 1.3. Scope
132		-
133		- The specification specifies a compressed data format that can be
134		- used for in-memory compression of a sequence of arbitrary bytes.
135		-
136		- 1.4. Compliance
137		-
138		- Unless otherwise indicated below, a compliant decompressor must be
139		- able to accept and decompress any data set that conforms to all
140		- the specifications presented here; a compliant compressor must
141		- produce data sets that conform to all the specifications presented
142		- here.
143		-
144		- 1.5. Definitions of terms and conventions used
145		-
146		- byte: 8 bits stored or transmitted as a unit (same as an octet).
147		- (For this specification, a byte is exactly 8 bits, even on
148		- machines which store a character on a number of bits different
149		- from 8.) See below, for the numbering of bits within a byte.
150		-
151		- 1.6. Changes from previous versions
152		-
153		- Version 3.1 was the first public release of this specification.
154		- In version 3.2, some terminology was changed and the Adler-32
155		- sample code was rewritten for clarity. In version 3.3, the
156		- support for a preset dictionary was introduced, and the
157		- specification was converted to RFC style.
158		-
159		-2. Detailed specification
160		-
161		- 2.1. Overall conventions
162		-
163		- In the diagrams below, a box like this:
164		-
165		- +---+
166		- \| \| <-- the vertical bars might be missing
167		- +---+
168		-
169		-
170		-
171		-
172		-Deutsch & Gailly Informational [Page 3]
173		-
174		-
175		-RFC 1950 ZLIB Compressed Data Format Specification May 1996
176		-
177		-
178		- represents one byte; a box like this:
179		-
180		- +==============+
181		- \| \|
182		- +==============+
183		-
184		- represents a variable number of bytes.
185		-
186		- Bytes stored within a computer do not have a "bit order", since
187		- they are always treated as a unit. However, a byte considered as
188		- an integer between 0 and 255 does have a most- and least-
189		- significant bit, and since we write numbers with the most-
190		- significant digit on the left, we also write bytes with the most-
191		- significant bit on the left. In the diagrams below, we number the
192		- bits of a byte so that bit 0 is the least-significant bit, i.e.,
193		- the bits are numbered:
194		-
195		- +--------+
196		- \|76543210\|
197		- +--------+
198		-
199		- Within a computer, a number may occupy multiple bytes. All
200		- multi-byte numbers in the format described here are stored with
201		- the MOST-significant byte first (at the lower memory address).
202		- For example, the decimal number 520 is stored as:
203		-
204		- 0 1
205		- +--------+--------+
206		- \|00000010\|00001000\|
207		- +--------+--------+
208		- ^ ^
209		- \| \|
210		- \| + less significant byte = 8
211		- + more significant byte = 2 x 256
212		-
213		- 2.2. Data format
214		-
215		- A zlib stream has the following structure:
216		-
217		- 0 1
218		- +---+---+
219		- \|CMF\|FLG\| (more-->)
220		- +---+---+
221		-
222		-
223		-
224		-
225		-
226		-
227		-
228		-
229		-Deutsch & Gailly Informational [Page 4]
230		-
231		-
232		-RFC 1950 ZLIB Compressed Data Format Specification May 1996
233		-
234		-
235		- (if FLG.FDICT set)
236		-
237		- 0 1 2 3
238		- +---+---+---+---+
239		- \| DICTID \| (more-->)
240		- +---+---+---+---+
241		-
242		- +=====================+---+---+---+---+
243		- \|...compressed data...\| ADLER32 \|
244		- +=====================+---+---+---+---+
245		-
246		- Any data which may appear after ADLER32 are not part of the zlib
247		- stream.
248		-
249		- CMF (Compression Method and flags)
250		- This byte is divided into a 4-bit compression method and a 4-
251		- bit information field depending on the compression method.
252		-
253		- bits 0 to 3 CM Compression method
254		- bits 4 to 7 CINFO Compression info
255		-
256		- CM (Compression method)
257		- This identifies the compression method used in the file. CM = 8
258		- denotes the "deflate" compression method with a window size up
259		- to 32K. This is the method used by gzip and PNG (see
260		- references [1] and [2] in Chapter 3, below, for the reference
261		- documents). CM = 15 is reserved. It might be used in a future
262		- version of this specification to indicate the presence of an
263		- extra field before the compressed data.
264		-
265		- CINFO (Compression info)
266		- For CM = 8, CINFO is the base-2 logarithm of the LZ77 window
267		- size, minus eight (CINFO=7 indicates a 32K window size). Values
268		- of CINFO above 7 are not allowed in this version of the
269		- specification. CINFO is not defined in this specification for
270		- CM not equal to 8.
271		-
272		- FLG (FLaGs)
273		- This flag byte is divided as follows:
274		-
275		- bits 0 to 4 FCHECK (check bits for CMF and FLG)
276		- bit 5 FDICT (preset dictionary)
277		- bits 6 to 7 FLEVEL (compression level)
278		-
279		- The FCHECK value must be such that CMF and FLG, when viewed as
280		- a 16-bit unsigned integer stored in MSB order (CMF*256 + FLG),
281		- is a multiple of 31.
282		-
283		-
284		-
285		-
286		-Deutsch & Gailly Informational [Page 5]
287		-
288		-
289		-RFC 1950 ZLIB Compressed Data Format Specification May 1996
290		-
291		-
292		- FDICT (Preset dictionary)
293		- If FDICT is set, a DICT dictionary identifier is present
294		- immediately after the FLG byte. The dictionary is a sequence of
295		- bytes which are initially fed to the compressor without
296		- producing any compressed output. DICT is the Adler-32 checksum
297		- of this sequence of bytes (see the definition of ADLER32
298		- below). The decompressor can use this identifier to determine
299		- which dictionary has been used by the compressor.
300		-
301		- FLEVEL (Compression level)
302		- These flags are available for use by specific compression
303		- methods. The "deflate" method (CM = 8) sets these flags as
304		- follows:
305		-
306		- 0 - compressor used fastest algorithm
307		- 1 - compressor used fast algorithm
308		- 2 - compressor used default algorithm
309		- 3 - compressor used maximum compression, slowest algorithm
310		-
311		- The information in FLEVEL is not needed for decompression; it
312		- is there to indicate if recompression might be worthwhile.
313		-
314		- compressed data
315		- For compression method 8, the compressed data is stored in the
316		- deflate compressed data format as described in the document
317		- "DEFLATE Compressed Data Format Specification" by L. Peter
318		- Deutsch. (See reference [3] in Chapter 3, below)
319		-
320		- Other compressed data formats are not specified in this version
321		- of the zlib specification.
322		-
323		- ADLER32 (Adler-32 checksum)
324		- This contains a checksum value of the uncompressed data
325		- (excluding any dictionary data) computed according to Adler-32
326		- algorithm. This algorithm is a 32-bit extension and improvement
327		- of the Fletcher algorithm, used in the ITU-T X.224 / ISO 8073
328		- standard. See references [4] and [5] in Chapter 3, below)
329		-
330		- Adler-32 is composed of two sums accumulated per byte: s1 is
331		- the sum of all bytes, s2 is the sum of all s1 values. Both sums
332		- are done modulo 65521. s1 is initialized to 1, s2 to zero. The
333		- Adler-32 checksum is stored as s2*65536 + s1 in most-
334		- significant-byte first (network) order.
335		-
336		-
337		-
338		-
339		-
340		-
341		-
342		-
343		-Deutsch & Gailly Informational [Page 6]
344		-
345		-
346		-RFC 1950 ZLIB Compressed Data Format Specification May 1996
347		-
348		-
349		- 2.3. Compliance
350		-
351		- A compliant compressor must produce streams with correct CMF, FLG
352		- and ADLER32, but need not support preset dictionaries. When the
353		- zlib data format is used as part of another standard data format,
354		- the compressor may use only preset dictionaries that are specified
355		- by this other data format. If this other format does not use the
356		- preset dictionary feature, the compressor must not set the FDICT
357		- flag.
358		-
359		- A compliant decompressor must check CMF, FLG, and ADLER32, and
360		- provide an error indication if any of these have incorrect values.
361		- A compliant decompressor must give an error indication if CM is
362		- not one of the values defined in this specification (only the
363		- value 8 is permitted in this version), since another value could
364		- indicate the presence of new features that would cause subsequent
365		- data to be interpreted incorrectly. A compliant decompressor must
366		- give an error indication if FDICT is set and DICTID is not the
367		- identifier of a known preset dictionary. A decompressor may
368		- ignore FLEVEL and still be compliant. When the zlib data format
369		- is being used as a part of another standard format, a compliant
370		- decompressor must support all the preset dictionaries specified by
371		- the other format. When the other format does not use the preset
372		- dictionary feature, a compliant decompressor must reject any
373		- stream in which the FDICT flag is set.
374		-
375		-3. References
376		-
377		- [1] Deutsch, L.P.,"GZIP Compressed Data Format Specification",
378		- available in ftp://ftp.uu.net/pub/archiving/zip/doc/
379		-
380		- [2] Thomas Boutell, "PNG (Portable Network Graphics) specification",
381		- available in ftp://ftp.uu.net/graphics/png/documents/
382		-
383		- [3] Deutsch, L.P.,"DEFLATE Compressed Data Format Specification",
384		- available in ftp://ftp.uu.net/pub/archiving/zip/doc/
385		-
386		- [4] Fletcher, J. G., "An Arithmetic Checksum for Serial
387		- Transmissions," IEEE Transactions on Communications, Vol. COM-30,
388		- No. 1, January 1982, pp. 247-252.
389		-
390		- [5] ITU-T Recommendation X.224, Annex D, "Checksum Algorithms,"
391		- November, 1993, pp. 144, 145. (Available from
392		- gopher://info.itu.ch). ITU-T X.244 is also the same as ISO 8073.
393		-
394		-
395		-
396		-
397		-
398		-
399		-
400		-Deutsch & Gailly Informational [Page 7]
401		-
402		-
403		-RFC 1950 ZLIB Compressed Data Format Specification May 1996
404		-
405		-
406		-4. Source code
407		-
408		- Source code for a C language implementation of a "zlib" compliant
409		- library is available at ftp://ftp.uu.net/pub/archiving/zip/zlib/.
410		-
411		-5. Security Considerations
412		-
413		- A decoder that fails to check the ADLER32 checksum value may be
414		- subject to undetected data corruption.
415		-
416		-6. Acknowledgements
417		-
418		- Trademarks cited in this document are the property of their
419		- respective owners.
420		-
421		- Jean-Loup Gailly and Mark Adler designed the zlib format and wrote
422		- the related software described in this specification. Glenn
423		- Randers-Pehrson converted this document to RFC and HTML format.
424		-
425		-7. Authors' Addresses
426		-
427		- L. Peter Deutsch
428		- Aladdin Enterprises
429		- 203 Santa Margarita Ave.
430		- Menlo Park, CA 94025
431		-
432		- Phone: (415) 322-0103 (AM only)
433		- FAX: (415) 322-1734
434		- EMail: <[email protected]>
435		-
436		-
437		- Jean-Loup Gailly
438		-
439		- EMail: <[email protected]>
440		-
441		- Questions about the technical content of this specification can be
442		- sent by email to
443		-
444		- Jean-Loup Gailly <[email protected]> and
445		- Mark Adler <[email protected]>
446		-
447		- Editorial comments on this specification can be sent by email to
448		-
449		- L. Peter Deutsch <[email protected]> and
450		- Glenn Randers-Pehrson <[email protected]>
451		-
452		-
453		-
454		-
455		-
456		-
457		-Deutsch & Gailly Informational [Page 8]
458		-
459		-
460		-RFC 1950 ZLIB Compressed Data Format Specification May 1996
461		-
462		-
463		-8. Appendix: Rationale
464		-
465		- 8.1. Preset dictionaries
466		-
467		- A preset dictionary is specially useful to compress short input
468		- sequences. The compressor can take advantage of the dictionary
469		- context to encode the input in a more compact manner. The
470		- decompressor can be initialized with the appropriate context by
471		- virtually decompressing a compressed version of the dictionary
472		- without producing any output. However for certain compression
473		- algorithms such as the deflate algorithm this operation can be
474		- achieved without actually performing any decompression.
475		-
476		- The compressor and the decompressor must use exactly the same
477		- dictionary. The dictionary may be fixed or may be chosen among a
478		- certain number of predefined dictionaries, according to the kind
479		- of input data. The decompressor can determine which dictionary has
480		- been chosen by the compressor by checking the dictionary
481		- identifier. This document does not specify the contents of
482		- predefined dictionaries, since the optimal dictionaries are
483		- application specific. Standard data formats using this feature of
484		- the zlib specification must precisely define the allowed
485		- dictionaries.
486		-
487		- 8.2. The Adler-32 algorithm
488		-
489		- The Adler-32 algorithm is much faster than the CRC32 algorithm yet
490		- still provides an extremely low probability of undetected errors.
491		-
492		- The modulo on unsigned long accumulators can be delayed for 5552
493		- bytes, so the modulo operation time is negligible. If the bytes
494		- are a, b, c, the second sum is 3a + 2b + c + 3, and so is position
495		- and order sensitive, unlike the first sum, which is just a
496		- checksum. That 65521 is prime is important to avoid a possible
497		- large class of two-byte errors that leave the check unchanged.
498		- (The Fletcher checksum uses 255, which is not prime and which also
499		- makes the Fletcher check insensitive to single byte changes 0 <->
500		- 255.)
501		-
502		- The sum s1 is initialized to 1 instead of zero to make the length
503		- of the sequence part of s2, so that the length does not have to be
504		- checked separately. (Any sequence of zeroes has a Fletcher
505		- checksum of zero.)
506		-
507		-
508		-
509		-
510		-
511		-
512		-
513		-
514		-Deutsch & Gailly Informational [Page 9]
515		-
516		-
517		-RFC 1950 ZLIB Compressed Data Format Specification May 1996
518		-
519		-
520		-9. Appendix: Sample code
521		-
522		- The following C code computes the Adler-32 checksum of a data buffer.
523		- It is written for clarity, not for speed. The sample code is in the
524		- ANSI C programming language. Non C users may find it easier to read
525		- with these hints:
526		-
527		- & Bitwise AND operator.
528		- >> Bitwise right shift operator. When applied to an
529		- unsigned quantity, as here, right shift inserts zero bit(s)
530		- at the left.
531		- << Bitwise left shift operator. Left shift inserts zero
532		- bit(s) at the right.
533		- ++ "n++" increments the variable n.
534		- % modulo operator: a % b is the remainder of a divided by b.
535		-
536		- #define BASE 65521 /* largest prime smaller than 65536 */
537		-
538		- /*
539		- Update a running Adler-32 checksum with the bytes buf[0..len-1]
540		- and return the updated checksum. The Adler-32 checksum should be
541		- initialized to 1.
542		-
543		- Usage example:
544		-
545		- unsigned long adler = 1L;
546		-
547		- while (read_buffer(buffer, length) != EOF) {
548		- adler = update_adler32(adler, buffer, length);
549		- }
550		- if (adler != original_adler) error();
551		- */
552		- unsigned long update_adler32(unsigned long adler,
553		- unsigned char *buf, int len)
554		- {
555		- unsigned long s1 = adler & 0xffff;
556		- unsigned long s2 = (adler >> 16) & 0xffff;
557		- int n;
558		-
559		- for (n = 0; n < len; n++) {
560		- s1 = (s1 + buf[n]) % BASE;
561		- s2 = (s2 + s1) % BASE;
562		- }
563		- return (s2 << 16) + s1;
564		- }
565		-
566		- /* Return the adler32 of the bytes buf[0..len-1] */
567		-
568		-
569		-
570		-
571		-Deutsch & Gailly Informational [Page 10]
572		-
573		-
574		-RFC 1950 ZLIB Compressed Data Format Specification May 1996
575		-
576		-
577		- unsigned long adler32(unsigned char *buf, int len)
578		- {
579		- return update_adler32(1L, buf, len);
580		- }
581		-
582		-
583		-
584		-
585		-
586		-
587		-
588		-
589		-
590		-
591		-
592		-
593		-
594		-
595		-
596		-
597		-
598		-
599		-
600		-
601		-
602		-
603		-
604		-
605		-
606		-
607		-
608		-
609		-
610		-
611		-
612		-
613		-
614		-
615		-
616		-
617		-
618		-
619		-
620		-
621		-
622		-
623		-
624		-
625		-
626		-
627		-
628		-Deutsch & Gailly Informational [Page 11]
629		-
630		-

	--- a/compat/zlib/doc/rfc1950.txt
	+++ b/compat/zlib/doc/rfc1950.txt
	@@ -1,630 +0,0 @@
1
2
3
4
5
6
7	Network Working Group P. Deutsch
8	Request for Comments: 1950 Aladdin Enterprises
9	Category: Informational J-L. Gailly
10	Info-ZIP
11	May 1996
12
13
14	ZLIB Compressed Data Format Specification version 3.3
15
16	Status of This Memo
17
18	This memo provides information for the Internet community. This memo
19	does not specify an Internet standard of any kind. Distribution of
20	this memo is unlimited.
21
22	IESG Note:
23
24	The IESG takes no position on the validity of any Intellectual
25	Property Rights statements contained in this document.
26
27	Notices
28
29	Copyright (c) 1996 L. Peter Deutsch and Jean-Loup Gailly
30
31	Permission is granted to copy and distribute this document for any
32	purpose and without charge, including translations into other
33	languages and incorporation into compilations, provided that the
34	copyright notice and this notice are preserved, and that any
35	substantive changes or deletions from the original are clearly
36	marked.
37
38	A pointer to the latest version of this and related documentation in
39	HTML format can be found at the URL
40	<ftp://ftp.uu.net/graphics/png/documents/zlib/zdoc-index.html>.
41
42	Abstract
43
44	This specification defines a lossless compressed data format. The
45	data can be produced or consumed, even for an arbitrarily long
46	sequentially presented input data stream, using only an a priori
47	bounded amount of intermediate storage. The format presently uses
48	the DEFLATE compression method but can be easily extended to use
49	other compression methods. It can be implemented readily in a manner
50	not covered by patents. This specification also defines the ADLER-32
51	checksum (an extension and improvement of the Fletcher checksum),
52	used for detection of data corruption, and provides an algorithm for
53	computing it.
54
55
56
57
58	Deutsch & Gailly Informational [Page 1]
59
60
61	RFC 1950 ZLIB Compressed Data Format Specification May 1996
62
63
64	Table of Contents
65
66	1. Introduction ................................................... 2
67	1.1. Purpose ................................................... 2
68	1.2. Intended audience ......................................... 3
69	1.3. Scope ..................................................... 3
70	1.4. Compliance ................................................ 3
71	1.5. Definitions of terms and conventions used ................ 3
72	1.6. Changes from previous versions ............................ 3
73	2. Detailed specification ......................................... 3
74	2.1. Overall conventions ....................................... 3
75	2.2. Data format ............................................... 4
76	2.3. Compliance ................................................ 7
77	3. References ..................................................... 7
78	4. Source code .................................................... 8
79	5. Security Considerations ........................................ 8
80	6. Acknowledgements ............................................... 8
81	7. Authors' Addresses ............................................. 8
82	8. Appendix: Rationale ............................................ 9
83	9. Appendix: Sample code ..........................................10
84
85	1. Introduction
86
87	1.1. Purpose
88
89	The purpose of this specification is to define a lossless
90	compressed data format that:
91
92	* Is independent of CPU type, operating system, file system,
93	and character set, and hence can be used for interchange;
94
95	* Can be produced or consumed, even for an arbitrarily long
96	sequentially presented input data stream, using only an a
97	priori bounded amount of intermediate storage, and hence can
98	be used in data communications or similar structures such as
99	Unix filters;
100
101	* Can use a number of different compression methods;
102
103	* Can be implemented readily in a manner not covered by
104	patents, and hence can be practiced freely.
105
106	The data format defined by this specification does not attempt to
107	allow random access to compressed data.
108
109
110
111
112
113
114
115	Deutsch & Gailly Informational [Page 2]
116
117
118	RFC 1950 ZLIB Compressed Data Format Specification May 1996
119
120
121	1.2. Intended audience
122
123	This specification is intended for use by implementors of software
124	to compress data into zlib format and/or decompress data from zlib
125	format.
126
127	The text of the specification assumes a basic background in
128	programming at the level of bits and other primitive data
129	representations.
130
131	1.3. Scope
132
133	The specification specifies a compressed data format that can be
134	used for in-memory compression of a sequence of arbitrary bytes.
135
136	1.4. Compliance
137
138	Unless otherwise indicated below, a compliant decompressor must be
139	able to accept and decompress any data set that conforms to all
140	the specifications presented here; a compliant compressor must
141	produce data sets that conform to all the specifications presented
142	here.
143
144	1.5. Definitions of terms and conventions used
145
146	byte: 8 bits stored or transmitted as a unit (same as an octet).
147	(For this specification, a byte is exactly 8 bits, even on
148	machines which store a character on a number of bits different
149	from 8.) See below, for the numbering of bits within a byte.
150
151	1.6. Changes from previous versions
152
153	Version 3.1 was the first public release of this specification.
154	In version 3.2, some terminology was changed and the Adler-32
155	sample code was rewritten for clarity. In version 3.3, the
156	support for a preset dictionary was introduced, and the
157	specification was converted to RFC style.
158
159	2. Detailed specification
160
161	2.1. Overall conventions
162
163	In the diagrams below, a box like this:
164
165	+---+
166	\| \| <-- the vertical bars might be missing
167	+---+
168
169
170
171
172	Deutsch & Gailly Informational [Page 3]
173
174
175	RFC 1950 ZLIB Compressed Data Format Specification May 1996
176
177
178	represents one byte; a box like this:
179
180	+==============+
181	\| \|
182	+==============+
183
184	represents a variable number of bytes.
185
186	Bytes stored within a computer do not have a "bit order", since
187	they are always treated as a unit. However, a byte considered as
188	an integer between 0 and 255 does have a most- and least-
189	significant bit, and since we write numbers with the most-
190	significant digit on the left, we also write bytes with the most-
191	significant bit on the left. In the diagrams below, we number the
192	bits of a byte so that bit 0 is the least-significant bit, i.e.,
193	the bits are numbered:
194
195	+--------+
196	\|76543210\|
197	+--------+
198
199	Within a computer, a number may occupy multiple bytes. All
200	multi-byte numbers in the format described here are stored with
201	the MOST-significant byte first (at the lower memory address).
202	For example, the decimal number 520 is stored as:
203
204	0 1
205	+--------+--------+
206	\|00000010\|00001000\|
207	+--------+--------+
208	^ ^
209	\| \|
210	\| + less significant byte = 8
211	+ more significant byte = 2 x 256
212
213	2.2. Data format
214
215	A zlib stream has the following structure:
216
217	0 1
218	+---+---+
219	\|CMF\|FLG\| (more-->)
220	+---+---+
221
222
223
224
225
226
227
228
229	Deutsch & Gailly Informational [Page 4]
230
231
232	RFC 1950 ZLIB Compressed Data Format Specification May 1996
233
234
235	(if FLG.FDICT set)
236
237	0 1 2 3
238	+---+---+---+---+
239	\| DICTID \| (more-->)
240	+---+---+---+---+
241
242	+=====================+---+---+---+---+
243	\|...compressed data...\| ADLER32 \|
244	+=====================+---+---+---+---+
245
246	Any data which may appear after ADLER32 are not part of the zlib
247	stream.
248
249	CMF (Compression Method and flags)
250	This byte is divided into a 4-bit compression method and a 4-
251	bit information field depending on the compression method.
252
253	bits 0 to 3 CM Compression method
254	bits 4 to 7 CINFO Compression info
255
256	CM (Compression method)
257	This identifies the compression method used in the file. CM = 8
258	denotes the "deflate" compression method with a window size up
259	to 32K. This is the method used by gzip and PNG (see
260	references [1] and [2] in Chapter 3, below, for the reference
261	documents). CM = 15 is reserved. It might be used in a future
262	version of this specification to indicate the presence of an
263	extra field before the compressed data.
264
265	CINFO (Compression info)
266	For CM = 8, CINFO is the base-2 logarithm of the LZ77 window
267	size, minus eight (CINFO=7 indicates a 32K window size). Values
268	of CINFO above 7 are not allowed in this version of the
269	specification. CINFO is not defined in this specification for
270	CM not equal to 8.
271
272	FLG (FLaGs)
273	This flag byte is divided as follows:
274
275	bits 0 to 4 FCHECK (check bits for CMF and FLG)
276	bit 5 FDICT (preset dictionary)
277	bits 6 to 7 FLEVEL (compression level)
278
279	The FCHECK value must be such that CMF and FLG, when viewed as
280	a 16-bit unsigned integer stored in MSB order (CMF*256 + FLG),
281	is a multiple of 31.
282
283
284
285
286	Deutsch & Gailly Informational [Page 5]
287
288
289	RFC 1950 ZLIB Compressed Data Format Specification May 1996
290
291
292	FDICT (Preset dictionary)
293	If FDICT is set, a DICT dictionary identifier is present
294	immediately after the FLG byte. The dictionary is a sequence of
295	bytes which are initially fed to the compressor without
296	producing any compressed output. DICT is the Adler-32 checksum
297	of this sequence of bytes (see the definition of ADLER32
298	below). The decompressor can use this identifier to determine
299	which dictionary has been used by the compressor.
300
301	FLEVEL (Compression level)
302	These flags are available for use by specific compression
303	methods. The "deflate" method (CM = 8) sets these flags as
304	follows:
305
306	0 - compressor used fastest algorithm
307	1 - compressor used fast algorithm
308	2 - compressor used default algorithm
309	3 - compressor used maximum compression, slowest algorithm
310
311	The information in FLEVEL is not needed for decompression; it
312	is there to indicate if recompression might be worthwhile.
313
314	compressed data
315	For compression method 8, the compressed data is stored in the
316	deflate compressed data format as described in the document
317	"DEFLATE Compressed Data Format Specification" by L. Peter
318	Deutsch. (See reference [3] in Chapter 3, below)
319
320	Other compressed data formats are not specified in this version
321	of the zlib specification.
322
323	ADLER32 (Adler-32 checksum)
324	This contains a checksum value of the uncompressed data
325	(excluding any dictionary data) computed according to Adler-32
326	algorithm. This algorithm is a 32-bit extension and improvement
327	of the Fletcher algorithm, used in the ITU-T X.224 / ISO 8073
328	standard. See references [4] and [5] in Chapter 3, below)
329
330	Adler-32 is composed of two sums accumulated per byte: s1 is
331	the sum of all bytes, s2 is the sum of all s1 values. Both sums
332	are done modulo 65521. s1 is initialized to 1, s2 to zero. The
333	Adler-32 checksum is stored as s2*65536 + s1 in most-
334	significant-byte first (network) order.
335
336
337
338
339
340
341
342
343	Deutsch & Gailly Informational [Page 6]
344
345
346	RFC 1950 ZLIB Compressed Data Format Specification May 1996
347
348
349	2.3. Compliance
350
351	A compliant compressor must produce streams with correct CMF, FLG
352	and ADLER32, but need not support preset dictionaries. When the
353	zlib data format is used as part of another standard data format,
354	the compressor may use only preset dictionaries that are specified
355	by this other data format. If this other format does not use the
356	preset dictionary feature, the compressor must not set the FDICT
357	flag.
358
359	A compliant decompressor must check CMF, FLG, and ADLER32, and
360	provide an error indication if any of these have incorrect values.
361	A compliant decompressor must give an error indication if CM is
362	not one of the values defined in this specification (only the
363	value 8 is permitted in this version), since another value could
364	indicate the presence of new features that would cause subsequent
365	data to be interpreted incorrectly. A compliant decompressor must
366	give an error indication if FDICT is set and DICTID is not the
367	identifier of a known preset dictionary. A decompressor may
368	ignore FLEVEL and still be compliant. When the zlib data format
369	is being used as a part of another standard format, a compliant
370	decompressor must support all the preset dictionaries specified by
371	the other format. When the other format does not use the preset
372	dictionary feature, a compliant decompressor must reject any
373	stream in which the FDICT flag is set.
374
375	3. References
376
377	[1] Deutsch, L.P.,"GZIP Compressed Data Format Specification",
378	available in ftp://ftp.uu.net/pub/archiving/zip/doc/
379
380	[2] Thomas Boutell, "PNG (Portable Network Graphics) specification",
381	available in ftp://ftp.uu.net/graphics/png/documents/
382
383	[3] Deutsch, L.P.,"DEFLATE Compressed Data Format Specification",
384	available in ftp://ftp.uu.net/pub/archiving/zip/doc/
385
386	[4] Fletcher, J. G., "An Arithmetic Checksum for Serial
387	Transmissions," IEEE Transactions on Communications, Vol. COM-30,
388	No. 1, January 1982, pp. 247-252.
389
390	[5] ITU-T Recommendation X.224, Annex D, "Checksum Algorithms,"
391	November, 1993, pp. 144, 145. (Available from
392	gopher://info.itu.ch). ITU-T X.244 is also the same as ISO 8073.
393
394
395
396
397
398
399
400	Deutsch & Gailly Informational [Page 7]
401
402
403	RFC 1950 ZLIB Compressed Data Format Specification May 1996
404
405
406	4. Source code
407
408	Source code for a C language implementation of a "zlib" compliant
409	library is available at ftp://ftp.uu.net/pub/archiving/zip/zlib/.
410
411	5. Security Considerations
412
413	A decoder that fails to check the ADLER32 checksum value may be
414	subject to undetected data corruption.
415
416	6. Acknowledgements
417
418	Trademarks cited in this document are the property of their
419	respective owners.
420
421	Jean-Loup Gailly and Mark Adler designed the zlib format and wrote
422	the related software described in this specification. Glenn
423	Randers-Pehrson converted this document to RFC and HTML format.
424
425	7. Authors' Addresses
426
427	L. Peter Deutsch
428	Aladdin Enterprises
429	203 Santa Margarita Ave.
430	Menlo Park, CA 94025
431
432	Phone: (415) 322-0103 (AM only)
433	FAX: (415) 322-1734
434	EMail: <[email protected]>
435
436
437	Jean-Loup Gailly
438
439	EMail: <[email protected]>
440
441	Questions about the technical content of this specification can be
442	sent by email to
443
444	Jean-Loup Gailly <[email protected]> and
445	Mark Adler <[email protected]>
446
447	Editorial comments on this specification can be sent by email to
448
449	L. Peter Deutsch <[email protected]> and
450	Glenn Randers-Pehrson <[email protected]>
451
452
453
454
455
456
457	Deutsch & Gailly Informational [Page 8]
458
459
460	RFC 1950 ZLIB Compressed Data Format Specification May 1996
461
462
463	8. Appendix: Rationale
464
465	8.1. Preset dictionaries
466
467	A preset dictionary is specially useful to compress short input
468	sequences. The compressor can take advantage of the dictionary
469	context to encode the input in a more compact manner. The
470	decompressor can be initialized with the appropriate context by
471	virtually decompressing a compressed version of the dictionary
472	without producing any output. However for certain compression
473	algorithms such as the deflate algorithm this operation can be
474	achieved without actually performing any decompression.
475
476	The compressor and the decompressor must use exactly the same
477	dictionary. The dictionary may be fixed or may be chosen among a
478	certain number of predefined dictionaries, according to the kind
479	of input data. The decompressor can determine which dictionary has
480	been chosen by the compressor by checking the dictionary
481	identifier. This document does not specify the contents of
482	predefined dictionaries, since the optimal dictionaries are
483	application specific. Standard data formats using this feature of
484	the zlib specification must precisely define the allowed
485	dictionaries.
486
487	8.2. The Adler-32 algorithm
488
489	The Adler-32 algorithm is much faster than the CRC32 algorithm yet
490	still provides an extremely low probability of undetected errors.
491
492	The modulo on unsigned long accumulators can be delayed for 5552
493	bytes, so the modulo operation time is negligible. If the bytes
494	are a, b, c, the second sum is 3a + 2b + c + 3, and so is position
495	and order sensitive, unlike the first sum, which is just a
496	checksum. That 65521 is prime is important to avoid a possible
497	large class of two-byte errors that leave the check unchanged.
498	(The Fletcher checksum uses 255, which is not prime and which also
499	makes the Fletcher check insensitive to single byte changes 0 <->
500	255.)
501
502	The sum s1 is initialized to 1 instead of zero to make the length
503	of the sequence part of s2, so that the length does not have to be
504	checked separately. (Any sequence of zeroes has a Fletcher
505	checksum of zero.)
506
507
508
509
510
511
512
513
514	Deutsch & Gailly Informational [Page 9]
515
516
517	RFC 1950 ZLIB Compressed Data Format Specification May 1996
518
519
520	9. Appendix: Sample code
521
522	The following C code computes the Adler-32 checksum of a data buffer.
523	It is written for clarity, not for speed. The sample code is in the
524	ANSI C programming language. Non C users may find it easier to read
525	with these hints:
526
527	& Bitwise AND operator.
528	>> Bitwise right shift operator. When applied to an
529	unsigned quantity, as here, right shift inserts zero bit(s)
530	at the left.
531	<< Bitwise left shift operator. Left shift inserts zero
532	bit(s) at the right.
533	++ "n++" increments the variable n.
534	% modulo operator: a % b is the remainder of a divided by b.
535
536	#define BASE 65521 /* largest prime smaller than 65536 */
537
538	/*
539	Update a running Adler-32 checksum with the bytes buf[0..len-1]
540	and return the updated checksum. The Adler-32 checksum should be
541	initialized to 1.
542
543	Usage example:
544
545	unsigned long adler = 1L;
546
547	while (read_buffer(buffer, length) != EOF) {
548	adler = update_adler32(adler, buffer, length);
549	}
550	if (adler != original_adler) error();
551	*/
552	unsigned long update_adler32(unsigned long adler,
553	unsigned char *buf, int len)
554	{
555	unsigned long s1 = adler & 0xffff;
556	unsigned long s2 = (adler >> 16) & 0xffff;
557	int n;
558
559	for (n = 0; n < len; n++) {
560	s1 = (s1 + buf[n]) % BASE;
561	s2 = (s2 + s1) % BASE;
562	}
563	return (s2 << 16) + s1;
564	}
565
566	/* Return the adler32 of the bytes buf[0..len-1] */
567
568
569
570
571	Deutsch & Gailly Informational [Page 10]
572
573
574	RFC 1950 ZLIB Compressed Data Format Specification May 1996
575
576
577	unsigned long adler32(unsigned char *buf, int len)
578	{
579	return update_adler32(1L, buf, len);
580	}
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628	Deutsch & Gailly Informational [Page 11]
629
630

	--- a/compat/zlib/doc/rfc1950.txt
	+++ b/compat/zlib/doc/rfc1950.txt
	@@ -1,630 +0,0 @@

D compat/zlib/doc/rfc1951.txt

-972

		--- a/compat/zlib/doc/rfc1951.txt
		+++ b/compat/zlib/doc/rfc1951.txt
		@@ -1,972 +0,0 @@
1		-
2		-
3		-
4		-
5		-
6		-
7		-Network Working Group P. Deutsch
8		-Request for Comments: 1951 Aladdin Enterprises
9		-Category: Informational May 1996
10		-
11		-
12		- DEFLATE Compressed Data Format Specification version 1.3
13		-
14		-Status of This Memo
15		-
16		- This memo provides information for the Internet community. This memo
17		- does not specify an Internet standard of any kind. Distribution of
18		- this memo is unlimited.
19		-
20		-IESG Note:
21		-
22		- The IESG takes no position on the validity of any Intellectual
23		- Property Rights statements contained in this document.
24		-
25		-Notices
26		-
27		- Copyright (c) 1996 L. Peter Deutsch
28		-
29		- Permission is granted to copy and distribute this document for any
30		- purpose and without charge, including translations into other
31		- languages and incorporation into compilations, provided that the
32		- copyright notice and this notice are preserved, and that any
33		- substantive changes or deletions from the original are clearly
34		- marked.
35		-
36		- A pointer to the latest version of this and related documentation in
37		- HTML format can be found at the URL
38		- <ftp://ftp.uu.net/graphics/png/documents/zlib/zdoc-index.html>.
39		-
40		-Abstract
41		-
42		- This specification defines a lossless compressed data format that
43		- compresses data using a combination of the LZ77 algorithm and Huffman
44		- coding, with efficiency comparable to the best currently available
45		- general-purpose compression methods. The data can be produced or
46		- consumed, even for an arbitrarily long sequentially presented input
47		- data stream, using only an a priori bounded amount of intermediate
48		- storage. The format can be implemented readily in a manner not
49		- covered by patents.
50		-
51		-
52		-
53		-
54		-
55		-
56		-
57		-
58		-Deutsch Informational [Page 1]
59		-
60		-
61		-RFC 1951 DEFLATE Compressed Data Format Specification May 1996
62		-
63		-
64		-Table of Contents
65		-
66		- 1. Introduction ................................................... 2
67		- 1.1. Purpose ................................................... 2
68		- 1.2. Intended audience ......................................... 3
69		- 1.3. Scope ..................................................... 3
70		- 1.4. Compliance ................................................ 3
71		- 1.5. Definitions of terms and conventions used ................ 3
72		- 1.6. Changes from previous versions ............................ 4
73		- 2. Compressed representation overview ............................. 4
74		- 3. Detailed specification ......................................... 5
75		- 3.1. Overall conventions ....................................... 5
76		- 3.1.1. Packing into bytes .................................. 5
77		- 3.2. Compressed block format ................................... 6
78		- 3.2.1. Synopsis of prefix and Huffman coding ............... 6
79		- 3.2.2. Use of Huffman coding in the "deflate" format ....... 7
80		- 3.2.3. Details of block format ............................. 9
81		- 3.2.4. Non-compressed blocks (BTYPE=00) ................... 11
82		- 3.2.5. Compressed blocks (length and distance codes) ...... 11
83		- 3.2.6. Compression with fixed Huffman codes (BTYPE=01) .... 12
84		- 3.2.7. Compression with dynamic Huffman codes (BTYPE=10) .. 13
85		- 3.3. Compliance ............................................... 14
86		- 4. Compression algorithm details ................................. 14
87		- 5. References .................................................... 16
88		- 6. Security Considerations ....................................... 16
89		- 7. Source code ................................................... 16
90		- 8. Acknowledgements .............................................. 16
91		- 9. Author's Address .............................................. 17
92		-
93		-1. Introduction
94		-
95		- 1.1. Purpose
96		-
97		- The purpose of this specification is to define a lossless
98		- compressed data format that:
99		- * Is independent of CPU type, operating system, file system,
100		- and character set, and hence can be used for interchange;
101		- * Can be produced or consumed, even for an arbitrarily long
102		- sequentially presented input data stream, using only an a
103		- priori bounded amount of intermediate storage, and hence
104		- can be used in data communications or similar structures
105		- such as Unix filters;
106		- * Compresses data with efficiency comparable to the best
107		- currently available general-purpose compression methods,
108		- and in particular considerably better than the "compress"
109		- program;
110		- * Can be implemented readily in a manner not covered by
111		- patents, and hence can be practiced freely;
112		-
113		-
114		-
115		-Deutsch Informational [Page 2]
116		-
117		-
118		-RFC 1951 DEFLATE Compressed Data Format Specification May 1996
119		-
120		-
121		- * Is compatible with the file format produced by the current
122		- widely used gzip utility, in that conforming decompressors
123		- will be able to read data produced by the existing gzip
124		- compressor.
125		-
126		- The data format defined by this specification does not attempt to:
127		-
128		- * Allow random access to compressed data;
129		- * Compress specialized data (e.g., raster graphics) as well
130		- as the best currently available specialized algorithms.
131		-
132		- A simple counting argument shows that no lossless compression
133		- algorithm can compress every possible input data set. For the
134		- format defined here, the worst case expansion is 5 bytes per 32K-
135		- byte block, i.e., a size increase of 0.015% for large data sets.
136		- English text usually compresses by a factor of 2.5 to 3;
137		- executable files usually compress somewhat less; graphical data
138		- such as raster images may compress much more.
139		-
140		- 1.2. Intended audience
141		-
142		- This specification is intended for use by implementors of software
143		- to compress data into "deflate" format and/or decompress data from
144		- "deflate" format.
145		-
146		- The text of the specification assumes a basic background in
147		- programming at the level of bits and other primitive data
148		- representations. Familiarity with the technique of Huffman coding
149		- is helpful but not required.
150		-
151		- 1.3. Scope
152		-
153		- The specification specifies a method for representing a sequence
154		- of bytes as a (usually shorter) sequence of bits, and a method for
155		- packing the latter bit sequence into bytes.
156		-
157		- 1.4. Compliance
158		-
159		- Unless otherwise indicated below, a compliant decompressor must be
160		- able to accept and decompress any data set that conforms to all
161		- the specifications presented here; a compliant compressor must
162		- produce data sets that conform to all the specifications presented
163		- here.
164		-
165		- 1.5. Definitions of terms and conventions used
166		-
167		- Byte: 8 bits stored or transmitted as a unit (same as an octet).
168		- For this specification, a byte is exactly 8 bits, even on machines
169		-
170		-
171		-
172		-Deutsch Informational [Page 3]
173		-
174		-
175		-RFC 1951 DEFLATE Compressed Data Format Specification May 1996
176		-
177		-
178		- which store a character on a number of bits different from eight.
179		- See below, for the numbering of bits within a byte.
180		-
181		- String: a sequence of arbitrary bytes.
182		-
183		- 1.6. Changes from previous versions
184		-
185		- There have been no technical changes to the deflate format since
186		- version 1.1 of this specification. In version 1.2, some
187		- terminology was changed. Version 1.3 is a conversion of the
188		- specification to RFC style.
189		-
190		-2. Compressed representation overview
191		-
192		- A compressed data set consists of a series of blocks, corresponding
193		- to successive blocks of input data. The block sizes are arbitrary,
194		- except that non-compressible blocks are limited to 65,535 bytes.
195		-
196		- Each block is compressed using a combination of the LZ77 algorithm
197		- and Huffman coding. The Huffman trees for each block are independent
198		- of those for previous or subsequent blocks; the LZ77 algorithm may
199		- use a reference to a duplicated string occurring in a previous block,
200		- up to 32K input bytes before.
201		-
202		- Each block consists of two parts: a pair of Huffman code trees that
203		- describe the representation of the compressed data part, and a
204		- compressed data part. (The Huffman trees themselves are compressed
205		- using Huffman encoding.) The compressed data consists of a series of
206		- elements of two types: literal bytes (of strings that have not been
207		- detected as duplicated within the previous 32K input bytes), and
208		- pointers to duplicated strings, where a pointer is represented as a
209		- pair <length, backward distance>. The representation used in the
210		- "deflate" format limits distances to 32K bytes and lengths to 258
211		- bytes, but does not limit the size of a block, except for
212		- uncompressible blocks, which are limited as noted above.
213		-
214		- Each type of value (literals, distances, and lengths) in the
215		- compressed data is represented using a Huffman code, using one code
216		- tree for literals and lengths and a separate code tree for distances.
217		- The code trees for each block appear in a compact form just before
218		- the compressed data for that block.
219		-
220		-
221		-
222		-
223		-
224		-
225		-
226		-
227		-
228		-
229		-Deutsch Informational [Page 4]
230		-
231		-
232		-RFC 1951 DEFLATE Compressed Data Format Specification May 1996
233		-
234		-
235		-3. Detailed specification
236		-
237		- 3.1. Overall conventions In the diagrams below, a box like this:
238		-
239		- +---+
240		- \| \| <-- the vertical bars might be missing
241		- +---+
242		-
243		- represents one byte; a box like this:
244		-
245		- +==============+
246		- \| \|
247		- +==============+
248		-
249		- represents a variable number of bytes.
250		-
251		- Bytes stored within a computer do not have a "bit order", since
252		- they are always treated as a unit. However, a byte considered as
253		- an integer between 0 and 255 does have a most- and least-
254		- significant bit, and since we write numbers with the most-
255		- significant digit on the left, we also write bytes with the most-
256		- significant bit on the left. In the diagrams below, we number the
257		- bits of a byte so that bit 0 is the least-significant bit, i.e.,
258		- the bits are numbered:
259		-
260		- +--------+
261		- \|76543210\|
262		- +--------+
263		-
264		- Within a computer, a number may occupy multiple bytes. All
265		- multi-byte numbers in the format described here are stored with
266		- the least-significant byte first (at the lower memory address).
267		- For example, the decimal number 520 is stored as:
268		-
269		- 0 1
270		- +--------+--------+
271		- \|00001000\|00000010\|
272		- +--------+--------+
273		- ^ ^
274		- \| \|
275		- \| + more significant byte = 2 x 256
276		- + less significant byte = 8
277		-
278		- 3.1.1. Packing into bytes
279		-
280		- This document does not address the issue of the order in which
281		- bits of a byte are transmitted on a bit-sequential medium,
282		- since the final data format described here is byte- rather than
283		-
284		-
285		-
286		-Deutsch Informational [Page 5]
287		-
288		-
289		-RFC 1951 DEFLATE Compressed Data Format Specification May 1996
290		-
291		-
292		- bit-oriented. However, we describe the compressed block format
293		- in below, as a sequence of data elements of various bit
294		- lengths, not a sequence of bytes. We must therefore specify
295		- how to pack these data elements into bytes to form the final
296		- compressed byte sequence:
297		-
298		- * Data elements are packed into bytes in order of
299		- increasing bit number within the byte, i.e., starting
300		- with the least-significant bit of the byte.
301		- * Data elements other than Huffman codes are packed
302		- starting with the least-significant bit of the data
303		- element.
304		- * Huffman codes are packed starting with the most-
305		- significant bit of the code.
306		-
307		- In other words, if one were to print out the compressed data as
308		- a sequence of bytes, starting with the first byte at the
309		- right margin and proceeding to the left, with the most-
310		- significant bit of each byte on the left as usual, one would be
311		- able to parse the result from right to left, with fixed-width
312		- elements in the correct MSB-to-LSB order and Huffman codes in
313		- bit-reversed order (i.e., with the first bit of the code in the
314		- relative LSB position).
315		-
316		- 3.2. Compressed block format
317		-
318		- 3.2.1. Synopsis of prefix and Huffman coding
319		-
320		- Prefix coding represents symbols from an a priori known
321		- alphabet by bit sequences (codes), one code for each symbol, in
322		- a manner such that different symbols may be represented by bit
323		- sequences of different lengths, but a parser can always parse
324		- an encoded string unambiguously symbol-by-symbol.
325		-
326		- We define a prefix code in terms of a binary tree in which the
327		- two edges descending from each non-leaf node are labeled 0 and
328		- 1 and in which the leaf nodes correspond one-for-one with (are
329		- labeled with) the symbols of the alphabet; then the code for a
330		- symbol is the sequence of 0's and 1's on the edges leading from
331		- the root to the leaf labeled with that symbol. For example:
332		-
333		-
334		-
335		-
336		-
337		-
338		-
339		-
340		-
341		-
342		-
343		-Deutsch Informational [Page 6]
344		-
345		-
346		-RFC 1951 DEFLATE Compressed Data Format Specification May 1996
347		-
348		-
349		- /\ Symbol Code
350		- 0 1 ------ ----
351		- / \ A 00
352		- /\ B B 1
353		- 0 1 C 011
354		- / \ D 010
355		- A /\
356		- 0 1
357		- / \
358		- D C
359		-
360		- A parser can decode the next symbol from an encoded input
361		- stream by walking down the tree from the root, at each step
362		- choosing the edge corresponding to the next input bit.
363		-
364		- Given an alphabet with known symbol frequencies, the Huffman
365		- algorithm allows the construction of an optimal prefix code
366		- (one which represents strings with those symbol frequencies
367		- using the fewest bits of any possible prefix codes for that
368		- alphabet). Such a code is called a Huffman code. (See
369		- reference [1] in Chapter 5, references for additional
370		- information on Huffman codes.)
371		-
372		- Note that in the "deflate" format, the Huffman codes for the
373		- various alphabets must not exceed certain maximum code lengths.
374		- This constraint complicates the algorithm for computing code
375		- lengths from symbol frequencies. Again, see Chapter 5,
376		- references for details.
377		-
378		- 3.2.2. Use of Huffman coding in the "deflate" format
379		-
380		- The Huffman codes used for each alphabet in the "deflate"
381		- format have two additional rules:
382		-
383		- * All codes of a given bit length have lexicographically
384		- consecutive values, in the same order as the symbols
385		- they represent;
386		-
387		- * Shorter codes lexicographically precede longer codes.
388		-
389		-
390		-
391		-
392		-
393		-
394		-
395		-
396		-
397		-
398		-
399		-
400		-Deutsch Informational [Page 7]
401		-
402		-
403		-RFC 1951 DEFLATE Compressed Data Format Specification May 1996
404		-
405		-
406		- We could recode the example above to follow this rule as
407		- follows, assuming that the order of the alphabet is ABCD:
408		-
409		- Symbol Code
410		- ------ ----
411		- A 10
412		- B 0
413		- C 110
414		- D 111
415		-
416		- I.e., 0 precedes 10 which precedes 11x, and 110 and 111 are
417		- lexicographically consecutive.
418		-
419		- Given this rule, we can define the Huffman code for an alphabet
420		- just by giving the bit lengths of the codes for each symbol of
421		- the alphabet in order; this is sufficient to determine the
422		- actual codes. In our example, the code is completely defined
423		- by the sequence of bit lengths (2, 1, 3, 3). The following
424		- algorithm generates the codes as integers, intended to be read
425		- from most- to least-significant bit. The code lengths are
426		- initially in tree[I].Len; the codes are produced in
427		- tree[I].Code.
428		-
429		- 1) Count the number of codes for each code length. Let
430		- bl_count[N] be the number of codes of length N, N >= 1.
431		-
432		- 2) Find the numerical value of the smallest code for each
433		- code length:
434		-
435		- code = 0;
436		- bl_count[0] = 0;
437		- for (bits = 1; bits <= MAX_BITS; bits++) {
438		- code = (code + bl_count[bits-1]) << 1;
439		- next_code[bits] = code;
440		- }
441		-
442		- 3) Assign numerical values to all codes, using consecutive
443		- values for all codes of the same length with the base
444		- values determined at step 2. Codes that are never used
445		- (which have a bit length of zero) must not be assigned a
446		- value.
447		-
448		- for (n = 0; n <= max_code; n++) {
449		- len = tree[n].Len;
450		- if (len != 0) {
451		- tree[n].Code = next_code[len];
452		- next_code[len]++;
453		- }
454		-
455		-
456		-
457		-Deutsch Informational [Page 8]
458		-
459		-
460		-RFC 1951 DEFLATE Compressed Data Format Specification May 1996
461		-
462		-
463		- }
464		-
465		- Example:
466		-
467		- Consider the alphabet ABCDEFGH, with bit lengths (3, 3, 3, 3,
468		- 3, 2, 4, 4). After step 1, we have:
469		-
470		- N bl_count[N]
471		- - -----------
472		- 2 1
473		- 3 5
474		- 4 2
475		-
476		- Step 2 computes the following next_code values:
477		-
478		- N next_code[N]
479		- - ------------
480		- 1 0
481		- 2 0
482		- 3 2
483		- 4 14
484		-
485		- Step 3 produces the following code values:
486		-
487		- Symbol Length Code
488		- ------ ------ ----
489		- A 3 010
490		- B 3 011
491		- C 3 100
492		- D 3 101
493		- E 3 110
494		- F 2 00
495		- G 4 1110
496		- H 4 1111
497		-
498		- 3.2.3. Details of block format
499		-
500		- Each block of compressed data begins with 3 header bits
501		- containing the following data:
502		-
503		- first bit BFINAL
504		- next 2 bits BTYPE
505		-
506		- Note that the header bits do not necessarily begin on a byte
507		- boundary, since a block does not necessarily occupy an integral
508		- number of bytes.
509		-
510		-
511		-
512		-
513		-
514		-Deutsch Informational [Page 9]
515		-
516		-
517		-RFC 1951 DEFLATE Compressed Data Format Specification May 1996
518		-
519		-
520		- BFINAL is set if and only if this is the last block of the data
521		- set.
522		-
523		- BTYPE specifies how the data are compressed, as follows:
524		-
525		- 00 - no compression
526		- 01 - compressed with fixed Huffman codes
527		- 10 - compressed with dynamic Huffman codes
528		- 11 - reserved (error)
529		-
530		- The only difference between the two compressed cases is how the
531		- Huffman codes for the literal/length and distance alphabets are
532		- defined.
533		-
534		- In all cases, the decoding algorithm for the actual data is as
535		- follows:
536		-
537		- do
538		- read block header from input stream.
539		- if stored with no compression
540		- skip any remaining bits in current partially
541		- processed byte
542		- read LEN and NLEN (see next section)
543		- copy LEN bytes of data to output
544		- otherwise
545		- if compressed with dynamic Huffman codes
546		- read representation of code trees (see
547		- subsection below)
548		- loop (until end of block code recognized)
549		- decode literal/length value from input stream
550		- if value < 256
551		- copy value (literal byte) to output stream
552		- otherwise
553		- if value = end of block (256)
554		- break from loop
555		- otherwise (value = 257..285)
556		- decode distance from input stream
557		-
558		- move backwards distance bytes in the output
559		- stream, and copy length bytes from this
560		- position to the output stream.
561		- end loop
562		- while not last block
563		-
564		- Note that a duplicated string reference may refer to a string
565		- in a previous block; i.e., the backward distance may cross one
566		- or more block boundaries. However a distance cannot refer past
567		- the beginning of the output stream. (An application using a
568		-
569		-
570		-
571		-Deutsch Informational [Page 10]
572		-
573		-
574		-RFC 1951 DEFLATE Compressed Data Format Specification May 1996
575		-
576		-
577		- preset dictionary might discard part of the output stream; a
578		- distance can refer to that part of the output stream anyway)
579		- Note also that the referenced string may overlap the current
580		- position; for example, if the last 2 bytes decoded have values
581		- X and Y, a string reference with <length = 5, distance = 2>
582		- adds X,Y,X,Y,X to the output stream.
583		-
584		- We now specify each compression method in turn.
585		-
586		- 3.2.4. Non-compressed blocks (BTYPE=00)
587		-
588		- Any bits of input up to the next byte boundary are ignored.
589		- The rest of the block consists of the following information:
590		-
591		- 0 1 2 3 4...
592		- +---+---+---+---+================================+
593		- \| LEN \| NLEN \|... LEN bytes of literal data...\|
594		- +---+---+---+---+================================+
595		-
596		- LEN is the number of data bytes in the block. NLEN is the
597		- one's complement of LEN.
598		-
599		- 3.2.5. Compressed blocks (length and distance codes)
600		-
601		- As noted above, encoded data blocks in the "deflate" format
602		- consist of sequences of symbols drawn from three conceptually
603		- distinct alphabets: either literal bytes, from the alphabet of
604		- byte values (0..255), or <length, backward distance> pairs,
605		- where the length is drawn from (3..258) and the distance is
606		- drawn from (1..32,768). In fact, the literal and length
607		- alphabets are merged into a single alphabet (0..285), where
608		- values 0..255 represent literal bytes, the value 256 indicates
609		- end-of-block, and values 257..285 represent length codes
610		- (possibly in conjunction with extra bits following the symbol
611		- code) as follows:
612		-
613		-
614		-
615		-
616		-
617		-
618		-
619		-
620		-
621		-
622		-
623		-
624		-
625		-
626		-
627		-
628		-Deutsch Informational [Page 11]
629		-
630		-
631		-RFC 1951 DEFLATE Compressed Data Format Specification May 1996
632		-
633		-
634		- Extra Extra Extra
635		- Code Bits Length(s) Code Bits Lengths Code Bits Length(s)
636		- ---- ---- ------ ---- ---- ------- ---- ---- -------
637		- 257 0 3 267 1 15,16 277 4 67-82
638		- 258 0 4 268 1 17,18 278 4 83-98
639		- 259 0 5 269 2 19-22 279 4 99-114
640		- 260 0 6 270 2 23-26 280 4 115-130
641		- 261 0 7 271 2 27-30 281 5 131-162
642		- 262 0 8 272 2 31-34 282 5 163-194
643		- 263 0 9 273 3 35-42 283 5 195-226
644		- 264 0 10 274 3 43-50 284 5 227-257
645		- 265 1 11,12 275 3 51-58 285 0 258
646		- 266 1 13,14 276 3 59-66
647		-
648		- The extra bits should be interpreted as a machine integer
649		- stored with the most-significant bit first, e.g., bits 1110
650		- represent the value 14.
651		-
652		- Extra Extra Extra
653		- Code Bits Dist Code Bits Dist Code Bits Distance
654		- ---- ---- ---- ---- ---- ------ ---- ---- --------
655		- 0 0 1 10 4 33-48 20 9 1025-1536
656		- 1 0 2 11 4 49-64 21 9 1537-2048
657		- 2 0 3 12 5 65-96 22 10 2049-3072
658		- 3 0 4 13 5 97-128 23 10 3073-4096
659		- 4 1 5,6 14 6 129-192 24 11 4097-6144
660		- 5 1 7,8 15 6 193-256 25 11 6145-8192
661		- 6 2 9-12 16 7 257-384 26 12 8193-12288
662		- 7 2 13-16 17 7 385-512 27 12 12289-16384
663		- 8 3 17-24 18 8 513-768 28 13 16385-24576
664		- 9 3 25-32 19 8 769-1024 29 13 24577-32768
665		-
666		- 3.2.6. Compression with fixed Huffman codes (BTYPE=01)
667		-
668		- The Huffman codes for the two alphabets are fixed, and are not
669		- represented explicitly in the data. The Huffman code lengths
670		- for the literal/length alphabet are:
671		-
672		- Lit Value Bits Codes
673		- --------- ---- -----
674		- 0 - 143 8 00110000 through
675		- 10111111
676		- 144 - 255 9 110010000 through
677		- 111111111
678		- 256 - 279 7 0000000 through
679		- 0010111
680		- 280 - 287 8 11000000 through
681		- 11000111
682		-
683		-
684		-
685		-Deutsch Informational [Page 12]
686		-
687		-
688		-RFC 1951 DEFLATE Compressed Data Format Specification May 1996
689		-
690		-
691		- The code lengths are sufficient to generate the actual codes,
692		- as described above; we show the codes in the table for added
693		- clarity. Literal/length values 286-287 will never actually
694		- occur in the compressed data, but participate in the code
695		- construction.
696		-
697		- Distance codes 0-31 are represented by (fixed-length) 5-bit
698		- codes, with possible additional bits as shown in the table
699		- shown in Paragraph 3.2.5, above. Note that distance codes 30-
700		- 31 will never actually occur in the compressed data.
701		-
702		- 3.2.7. Compression with dynamic Huffman codes (BTYPE=10)
703		-
704		- The Huffman codes for the two alphabets appear in the block
705		- immediately after the header bits and before the actual
706		- compressed data, first the literal/length code and then the
707		- distance code. Each code is defined by a sequence of code
708		- lengths, as discussed in Paragraph 3.2.2, above. For even
709		- greater compactness, the code length sequences themselves are
710		- compressed using a Huffman code. The alphabet for code lengths
711		- is as follows:
712		-
713		- 0 - 15: Represent code lengths of 0 - 15
714		- 16: Copy the previous code length 3 - 6 times.
715		- The next 2 bits indicate repeat length
716		- (0 = 3, ... , 3 = 6)
717		- Example: Codes 8, 16 (+2 bits 11),
718		- 16 (+2 bits 10) will expand to
719		- 12 code lengths of 8 (1 + 6 + 5)
720		- 17: Repeat a code length of 0 for 3 - 10 times.
721		- (3 bits of length)
722		- 18: Repeat a code length of 0 for 11 - 138 times
723		- (7 bits of length)
724		-
725		- A code length of 0 indicates that the corresponding symbol in
726		- the literal/length or distance alphabet will not occur in the
727		- block, and should not participate in the Huffman code
728		- construction algorithm given earlier. If only one distance
729		- code is used, it is encoded using one bit, not zero bits; in
730		- this case there is a single code length of one, with one unused
731		- code. One distance code of zero bits means that there are no
732		- distance codes used at all (the data is all literals).
733		-
734		- We can now define the format of the block:
735		-
736		- 5 Bits: HLIT, # of Literal/Length codes - 257 (257 - 286)
737		- 5 Bits: HDIST, # of Distance codes - 1 (1 - 32)
738		- 4 Bits: HCLEN, # of Code Length codes - 4 (4 - 19)
739		-
740		-
741		-
742		-Deutsch Informational [Page 13]
743		-
744		-
745		-RFC 1951 DEFLATE Compressed Data Format Specification May 1996
746		-
747		-
748		- (HCLEN + 4) x 3 bits: code lengths for the code length
749		- alphabet given just above, in the order: 16, 17, 18,
750		- 0, 8, 7, 9, 6, 10, 5, 11, 4, 12, 3, 13, 2, 14, 1, 15
751		-
752		- These code lengths are interpreted as 3-bit integers
753		- (0-7); as above, a code length of 0 means the
754		- corresponding symbol (literal/length or distance code
755		- length) is not used.
756		-
757		- HLIT + 257 code lengths for the literal/length alphabet,
758		- encoded using the code length Huffman code
759		-
760		- HDIST + 1 code lengths for the distance alphabet,
761		- encoded using the code length Huffman code
762		-
763		- The actual compressed data of the block,
764		- encoded using the literal/length and distance Huffman
765		- codes
766		-
767		- The literal/length symbol 256 (end of data),
768		- encoded using the literal/length Huffman code
769		-
770		- The code length repeat codes can cross from HLIT + 257 to the
771		- HDIST + 1 code lengths. In other words, all code lengths form
772		- a single sequence of HLIT + HDIST + 258 values.
773		-
774		- 3.3. Compliance
775		-
776		- A compressor may limit further the ranges of values specified in
777		- the previous section and still be compliant; for example, it may
778		- limit the range of backward pointers to some value smaller than
779		- 32K. Similarly, a compressor may limit the size of blocks so that
780		- a compressible block fits in memory.
781		-
782		- A compliant decompressor must accept the full range of possible
783		- values defined in the previous section, and must accept blocks of
784		- arbitrary size.
785		-
786		-4. Compression algorithm details
787		-
788		- While it is the intent of this document to define the "deflate"
789		- compressed data format without reference to any particular
790		- compression algorithm, the format is related to the compressed
791		- formats produced by LZ77 (Lempel-Ziv 1977, see reference [2] below);
792		- since many variations of LZ77 are patented, it is strongly
793		- recommended that the implementor of a compressor follow the general
794		- algorithm presented here, which is known not to be patented per se.
795		- The material in this section is not part of the definition of the
796		-
797		-
798		-
799		-Deutsch Informational [Page 14]
800		-
801		-
802		-RFC 1951 DEFLATE Compressed Data Format Specification May 1996
803		-
804		-
805		- specification per se, and a compressor need not follow it in order to
806		- be compliant.
807		-
808		- The compressor terminates a block when it determines that starting a
809		- new block with fresh trees would be useful, or when the block size
810		- fills up the compressor's block buffer.
811		-
812		- The compressor uses a chained hash table to find duplicated strings,
813		- using a hash function that operates on 3-byte sequences. At any
814		- given point during compression, let XYZ be the next 3 input bytes to
815		- be examined (not necessarily all different, of course). First, the
816		- compressor examines the hash chain for XYZ. If the chain is empty,
817		- the compressor simply writes out X as a literal byte and advances one
818		- byte in the input. If the hash chain is not empty, indicating that
819		- the sequence XYZ (or, if we are unlucky, some other 3 bytes with the
820		- same hash function value) has occurred recently, the compressor
821		- compares all strings on the XYZ hash chain with the actual input data
822		- sequence starting at the current point, and selects the longest
823		- match.
824		-
825		- The compressor searches the hash chains starting with the most recent
826		- strings, to favor small distances and thus take advantage of the
827		- Huffman encoding. The hash chains are singly linked. There are no
828		- deletions from the hash chains; the algorithm simply discards matches
829		- that are too old. To avoid a worst-case situation, very long hash
830		- chains are arbitrarily truncated at a certain length, determined by a
831		- run-time parameter.
832		-
833		- To improve overall compression, the compressor optionally defers the
834		- selection of matches ("lazy matching"): after a match of length N has
835		- been found, the compressor searches for a longer match starting at
836		- the next input byte. If it finds a longer match, it truncates the
837		- previous match to a length of one (thus producing a single literal
838		- byte) and then emits the longer match. Otherwise, it emits the
839		- original match, and, as described above, advances N bytes before
840		- continuing.
841		-
842		- Run-time parameters also control this "lazy match" procedure. If
843		- compression ratio is most important, the compressor attempts a
844		- complete second search regardless of the length of the first match.
845		- In the normal case, if the current match is "long enough", the
846		- compressor reduces the search for a longer match, thus speeding up
847		- the process. If speed is most important, the compressor inserts new
848		- strings in the hash table only when no match was found, or when the
849		- match is not "too long". This degrades the compression ratio but
850		- saves time since there are both fewer insertions and fewer searches.
851		-
852		-
853		-
854		-
855		-
856		-Deutsch Informational [Page 15]
857		-
858		-
859		-RFC 1951 DEFLATE Compressed Data Format Specification May 1996
860		-
861		-
862		-5. References
863		-
864		- [1] Huffman, D. A., "A Method for the Construction of Minimum
865		- Redundancy Codes", Proceedings of the Institute of Radio
866		- Engineers, September 1952, Volume 40, Number 9, pp. 1098-1101.
867		-
868		- [2] Ziv J., Lempel A., "A Universal Algorithm for Sequential Data
869		- Compression", IEEE Transactions on Information Theory, Vol. 23,
870		- No. 3, pp. 337-343.
871		-
872		- [3] Gailly, J.-L., and Adler, M., ZLIB documentation and sources,
873		- available in ftp://ftp.uu.net/pub/archiving/zip/doc/
874		-
875		- [4] Gailly, J.-L., and Adler, M., GZIP documentation and sources,
876		- available as gzip-*.tar in ftp://prep.ai.mit.edu/pub/gnu/
877		-
878		- [5] Schwartz, E. S., and Kallick, B. "Generating a canonical prefix
879		- encoding." Comm. ACM, 7,3 (Mar. 1964), pp. 166-169.
880		-
881		- [6] Hirschberg and Lelewer, "Efficient decoding of prefix codes,"
882		- Comm. ACM, 33,4, April 1990, pp. 449-459.
883		-
884		-6. Security Considerations
885		-
886		- Any data compression method involves the reduction of redundancy in
887		- the data. Consequently, any corruption of the data is likely to have
888		- severe effects and be difficult to correct. Uncompressed text, on
889		- the other hand, will probably still be readable despite the presence
890		- of some corrupted bytes.
891		-
892		- It is recommended that systems using this data format provide some
893		- means of validating the integrity of the compressed data. See
894		- reference [3], for example.
895		-
896		-7. Source code
897		-
898		- Source code for a C language implementation of a "deflate" compliant
899		- compressor and decompressor is available within the zlib package at
900		- ftp://ftp.uu.net/pub/archiving/zip/zlib/.
901		-
902		-8. Acknowledgements
903		-
904		- Trademarks cited in this document are the property of their
905		- respective owners.
906		-
907		- Phil Katz designed the deflate format. Jean-Loup Gailly and Mark
908		- Adler wrote the related software described in this specification.
909		- Glenn Randers-Pehrson converted this document to RFC and HTML format.
910		-
911		-
912		-
913		-Deutsch Informational [Page 16]
914		-
915		-
916		-RFC 1951 DEFLATE Compressed Data Format Specification May 1996
917		-
918		-
919		-9. Author's Address
920		-
921		- L. Peter Deutsch
922		- Aladdin Enterprises
923		- 203 Santa Margarita Ave.
924		- Menlo Park, CA 94025
925		-
926		- Phone: (415) 322-0103 (AM only)
927		- FAX: (415) 322-1734
928		- EMail: <[email protected]>
929		-
930		- Questions about the technical content of this specification can be
931		- sent by email to:
932		-
933		- Jean-Loup Gailly <[email protected]> and
934		- Mark Adler <[email protected]>
935		-
936		- Editorial comments on this specification can be sent by email to:
937		-
938		- L. Peter Deutsch <[email protected]> and
939		- Glenn Randers-Pehrson <[email protected]>
940		-
941		-
942		-
943		-
944		-
945		-
946		-
947		-
948		-
949		-
950		-
951		-
952		-
953		-
954		-
955		-
956		-
957		-
958		-
959		-
960		-
961		-
962		-
963		-
964		-
965		-
966		-
967		-
968		-
969		-
970		-Deutsch Informational [Page 17]
971		-
972		-

	--- a/compat/zlib/doc/rfc1951.txt
	+++ b/compat/zlib/doc/rfc1951.txt
	@@ -1,972 +0,0 @@
1
2
3
4
5
6
7	Network Working Group P. Deutsch
8	Request for Comments: 1951 Aladdin Enterprises
9	Category: Informational May 1996
10
11
12	DEFLATE Compressed Data Format Specification version 1.3
13
14	Status of This Memo
15
16	This memo provides information for the Internet community. This memo
17	does not specify an Internet standard of any kind. Distribution of
18	this memo is unlimited.
19
20	IESG Note:
21
22	The IESG takes no position on the validity of any Intellectual
23	Property Rights statements contained in this document.
24
25	Notices
26
27	Copyright (c) 1996 L. Peter Deutsch
28
29	Permission is granted to copy and distribute this document for any
30	purpose and without charge, including translations into other
31	languages and incorporation into compilations, provided that the
32	copyright notice and this notice are preserved, and that any
33	substantive changes or deletions from the original are clearly
34	marked.
35
36	A pointer to the latest version of this and related documentation in
37	HTML format can be found at the URL
38	<ftp://ftp.uu.net/graphics/png/documents/zlib/zdoc-index.html>.
39
40	Abstract
41
42	This specification defines a lossless compressed data format that
43	compresses data using a combination of the LZ77 algorithm and Huffman
44	coding, with efficiency comparable to the best currently available
45	general-purpose compression methods. The data can be produced or
46	consumed, even for an arbitrarily long sequentially presented input
47	data stream, using only an a priori bounded amount of intermediate
48	storage. The format can be implemented readily in a manner not
49	covered by patents.
50
51
52
53
54
55
56
57
58	Deutsch Informational [Page 1]
59
60
61	RFC 1951 DEFLATE Compressed Data Format Specification May 1996
62
63
64	Table of Contents
65
66	1. Introduction ................................................... 2
67	1.1. Purpose ................................................... 2
68	1.2. Intended audience ......................................... 3
69	1.3. Scope ..................................................... 3
70	1.4. Compliance ................................................ 3
71	1.5. Definitions of terms and conventions used ................ 3
72	1.6. Changes from previous versions ............................ 4
73	2. Compressed representation overview ............................. 4
74	3. Detailed specification ......................................... 5
75	3.1. Overall conventions ....................................... 5
76	3.1.1. Packing into bytes .................................. 5
77	3.2. Compressed block format ................................... 6
78	3.2.1. Synopsis of prefix and Huffman coding ............... 6
79	3.2.2. Use of Huffman coding in the "deflate" format ....... 7
80	3.2.3. Details of block format ............................. 9
81	3.2.4. Non-compressed blocks (BTYPE=00) ................... 11
82	3.2.5. Compressed blocks (length and distance codes) ...... 11
83	3.2.6. Compression with fixed Huffman codes (BTYPE=01) .... 12
84	3.2.7. Compression with dynamic Huffman codes (BTYPE=10) .. 13
85	3.3. Compliance ............................................... 14
86	4. Compression algorithm details ................................. 14
87	5. References .................................................... 16
88	6. Security Considerations ....................................... 16
89	7. Source code ................................................... 16
90	8. Acknowledgements .............................................. 16
91	9. Author's Address .............................................. 17
92
93	1. Introduction
94
95	1.1. Purpose
96
97	The purpose of this specification is to define a lossless
98	compressed data format that:
99	* Is independent of CPU type, operating system, file system,
100	and character set, and hence can be used for interchange;
101	* Can be produced or consumed, even for an arbitrarily long
102	sequentially presented input data stream, using only an a
103	priori bounded amount of intermediate storage, and hence
104	can be used in data communications or similar structures
105	such as Unix filters;
106	* Compresses data with efficiency comparable to the best
107	currently available general-purpose compression methods,
108	and in particular considerably better than the "compress"
109	program;
110	* Can be implemented readily in a manner not covered by
111	patents, and hence can be practiced freely;
112
113
114
115	Deutsch Informational [Page 2]
116
117
118	RFC 1951 DEFLATE Compressed Data Format Specification May 1996
119
120
121	* Is compatible with the file format produced by the current
122	widely used gzip utility, in that conforming decompressors
123	will be able to read data produced by the existing gzip
124	compressor.
125
126	The data format defined by this specification does not attempt to:
127
128	* Allow random access to compressed data;
129	* Compress specialized data (e.g., raster graphics) as well
130	as the best currently available specialized algorithms.
131
132	A simple counting argument shows that no lossless compression
133	algorithm can compress every possible input data set. For the
134	format defined here, the worst case expansion is 5 bytes per 32K-
135	byte block, i.e., a size increase of 0.015% for large data sets.
136	English text usually compresses by a factor of 2.5 to 3;
137	executable files usually compress somewhat less; graphical data
138	such as raster images may compress much more.
139
140	1.2. Intended audience
141
142	This specification is intended for use by implementors of software
143	to compress data into "deflate" format and/or decompress data from
144	"deflate" format.
145
146	The text of the specification assumes a basic background in
147	programming at the level of bits and other primitive data
148	representations. Familiarity with the technique of Huffman coding
149	is helpful but not required.
150
151	1.3. Scope
152
153	The specification specifies a method for representing a sequence
154	of bytes as a (usually shorter) sequence of bits, and a method for
155	packing the latter bit sequence into bytes.
156
157	1.4. Compliance
158
159	Unless otherwise indicated below, a compliant decompressor must be
160	able to accept and decompress any data set that conforms to all
161	the specifications presented here; a compliant compressor must
162	produce data sets that conform to all the specifications presented
163	here.
164
165	1.5. Definitions of terms and conventions used
166
167	Byte: 8 bits stored or transmitted as a unit (same as an octet).
168	For this specification, a byte is exactly 8 bits, even on machines
169
170
171
172	Deutsch Informational [Page 3]
173
174
175	RFC 1951 DEFLATE Compressed Data Format Specification May 1996
176
177
178	which store a character on a number of bits different from eight.
179	See below, for the numbering of bits within a byte.
180
181	String: a sequence of arbitrary bytes.
182
183	1.6. Changes from previous versions
184
185	There have been no technical changes to the deflate format since
186	version 1.1 of this specification. In version 1.2, some
187	terminology was changed. Version 1.3 is a conversion of the
188	specification to RFC style.
189
190	2. Compressed representation overview
191
192	A compressed data set consists of a series of blocks, corresponding
193	to successive blocks of input data. The block sizes are arbitrary,
194	except that non-compressible blocks are limited to 65,535 bytes.
195
196	Each block is compressed using a combination of the LZ77 algorithm
197	and Huffman coding. The Huffman trees for each block are independent
198	of those for previous or subsequent blocks; the LZ77 algorithm may
199	use a reference to a duplicated string occurring in a previous block,
200	up to 32K input bytes before.
201
202	Each block consists of two parts: a pair of Huffman code trees that
203	describe the representation of the compressed data part, and a
204	compressed data part. (The Huffman trees themselves are compressed
205	using Huffman encoding.) The compressed data consists of a series of
206	elements of two types: literal bytes (of strings that have not been
207	detected as duplicated within the previous 32K input bytes), and
208	pointers to duplicated strings, where a pointer is represented as a
209	pair <length, backward distance>. The representation used in the
210	"deflate" format limits distances to 32K bytes and lengths to 258
211	bytes, but does not limit the size of a block, except for
212	uncompressible blocks, which are limited as noted above.
213
214	Each type of value (literals, distances, and lengths) in the
215	compressed data is represented using a Huffman code, using one code
216	tree for literals and lengths and a separate code tree for distances.
217	The code trees for each block appear in a compact form just before
218	the compressed data for that block.
219
220
221
222
223
224
225
226
227
228
229	Deutsch Informational [Page 4]
230
231
232	RFC 1951 DEFLATE Compressed Data Format Specification May 1996
233
234
235	3. Detailed specification
236
237	3.1. Overall conventions In the diagrams below, a box like this:
238
239	+---+
240	\| \| <-- the vertical bars might be missing
241	+---+
242
243	represents one byte; a box like this:
244
245	+==============+
246	\| \|
247	+==============+
248
249	represents a variable number of bytes.
250
251	Bytes stored within a computer do not have a "bit order", since
252	they are always treated as a unit. However, a byte considered as
253	an integer between 0 and 255 does have a most- and least-
254	significant bit, and since we write numbers with the most-
255	significant digit on the left, we also write bytes with the most-
256	significant bit on the left. In the diagrams below, we number the
257	bits of a byte so that bit 0 is the least-significant bit, i.e.,
258	the bits are numbered:
259
260	+--------+
261	\|76543210\|
262	+--------+
263
264	Within a computer, a number may occupy multiple bytes. All
265	multi-byte numbers in the format described here are stored with
266	the least-significant byte first (at the lower memory address).
267	For example, the decimal number 520 is stored as:
268
269	0 1
270	+--------+--------+
271	\|00001000\|00000010\|
272	+--------+--------+
273	^ ^
274	\| \|
275	\| + more significant byte = 2 x 256
276	+ less significant byte = 8
277
278	3.1.1. Packing into bytes
279
280	This document does not address the issue of the order in which
281	bits of a byte are transmitted on a bit-sequential medium,
282	since the final data format described here is byte- rather than
283
284
285
286	Deutsch Informational [Page 5]
287
288
289	RFC 1951 DEFLATE Compressed Data Format Specification May 1996
290
291
292	bit-oriented. However, we describe the compressed block format
293	in below, as a sequence of data elements of various bit
294	lengths, not a sequence of bytes. We must therefore specify
295	how to pack these data elements into bytes to form the final
296	compressed byte sequence:
297
298	* Data elements are packed into bytes in order of
299	increasing bit number within the byte, i.e., starting
300	with the least-significant bit of the byte.
301	* Data elements other than Huffman codes are packed
302	starting with the least-significant bit of the data
303	element.
304	* Huffman codes are packed starting with the most-
305	significant bit of the code.
306
307	In other words, if one were to print out the compressed data as
308	a sequence of bytes, starting with the first byte at the
309	right margin and proceeding to the left, with the most-
310	significant bit of each byte on the left as usual, one would be
311	able to parse the result from right to left, with fixed-width
312	elements in the correct MSB-to-LSB order and Huffman codes in
313	bit-reversed order (i.e., with the first bit of the code in the
314	relative LSB position).
315
316	3.2. Compressed block format
317
318	3.2.1. Synopsis of prefix and Huffman coding
319
320	Prefix coding represents symbols from an a priori known
321	alphabet by bit sequences (codes), one code for each symbol, in
322	a manner such that different symbols may be represented by bit
323	sequences of different lengths, but a parser can always parse
324	an encoded string unambiguously symbol-by-symbol.
325
326	We define a prefix code in terms of a binary tree in which the
327	two edges descending from each non-leaf node are labeled 0 and
328	1 and in which the leaf nodes correspond one-for-one with (are
329	labeled with) the symbols of the alphabet; then the code for a
330	symbol is the sequence of 0's and 1's on the edges leading from
331	the root to the leaf labeled with that symbol. For example:
332
333
334
335
336
337
338
339
340
341
342
343	Deutsch Informational [Page 6]
344
345
346	RFC 1951 DEFLATE Compressed Data Format Specification May 1996
347
348
349	/\ Symbol Code
350	0 1 ------ ----
351	/ \ A 00
352	/\ B B 1
353	0 1 C 011
354	/ \ D 010
355	A /\
356	0 1
357	/ \
358	D C
359
360	A parser can decode the next symbol from an encoded input
361	stream by walking down the tree from the root, at each step
362	choosing the edge corresponding to the next input bit.
363
364	Given an alphabet with known symbol frequencies, the Huffman
365	algorithm allows the construction of an optimal prefix code
366	(one which represents strings with those symbol frequencies
367	using the fewest bits of any possible prefix codes for that
368	alphabet). Such a code is called a Huffman code. (See
369	reference [1] in Chapter 5, references for additional
370	information on Huffman codes.)
371
372	Note that in the "deflate" format, the Huffman codes for the
373	various alphabets must not exceed certain maximum code lengths.
374	This constraint complicates the algorithm for computing code
375	lengths from symbol frequencies. Again, see Chapter 5,
376	references for details.
377
378	3.2.2. Use of Huffman coding in the "deflate" format
379
380	The Huffman codes used for each alphabet in the "deflate"
381	format have two additional rules:
382
383	* All codes of a given bit length have lexicographically
384	consecutive values, in the same order as the symbols
385	they represent;
386
387	* Shorter codes lexicographically precede longer codes.
388
389
390
391
392
393
394
395
396
397
398
399
400	Deutsch Informational [Page 7]
401
402
403	RFC 1951 DEFLATE Compressed Data Format Specification May 1996
404
405
406	We could recode the example above to follow this rule as
407	follows, assuming that the order of the alphabet is ABCD:
408
409	Symbol Code
410	------ ----
411	A 10
412	B 0
413	C 110
414	D 111
415
416	I.e., 0 precedes 10 which precedes 11x, and 110 and 111 are
417	lexicographically consecutive.
418
419	Given this rule, we can define the Huffman code for an alphabet
420	just by giving the bit lengths of the codes for each symbol of
421	the alphabet in order; this is sufficient to determine the
422	actual codes. In our example, the code is completely defined
423	by the sequence of bit lengths (2, 1, 3, 3). The following
424	algorithm generates the codes as integers, intended to be read
425	from most- to least-significant bit. The code lengths are
426	initially in tree[I].Len; the codes are produced in
427	tree[I].Code.
428
429	1) Count the number of codes for each code length. Let
430	bl_count[N] be the number of codes of length N, N >= 1.
431
432	2) Find the numerical value of the smallest code for each
433	code length:
434
435	code = 0;
436	bl_count[0] = 0;
437	for (bits = 1; bits <= MAX_BITS; bits++) {
438	code = (code + bl_count[bits-1]) << 1;
439	next_code[bits] = code;
440	}
441
442	3) Assign numerical values to all codes, using consecutive
443	values for all codes of the same length with the base
444	values determined at step 2. Codes that are never used
445	(which have a bit length of zero) must not be assigned a
446	value.
447
448	for (n = 0; n <= max_code; n++) {
449	len = tree[n].Len;
450	if (len != 0) {
451	tree[n].Code = next_code[len];
452	next_code[len]++;
453	}
454
455
456
457	Deutsch Informational [Page 8]
458
459
460	RFC 1951 DEFLATE Compressed Data Format Specification May 1996
461
462
463	}
464
465	Example:
466
467	Consider the alphabet ABCDEFGH, with bit lengths (3, 3, 3, 3,
468	3, 2, 4, 4). After step 1, we have:
469
470	N bl_count[N]
471	- -----------
472	2 1
473	3 5
474	4 2
475
476	Step 2 computes the following next_code values:
477
478	N next_code[N]
479	- ------------
480	1 0
481	2 0
482	3 2
483	4 14
484
485	Step 3 produces the following code values:
486
487	Symbol Length Code
488	------ ------ ----
489	A 3 010
490	B 3 011
491	C 3 100
492	D 3 101
493	E 3 110
494	F 2 00
495	G 4 1110
496	H 4 1111
497
498	3.2.3. Details of block format
499
500	Each block of compressed data begins with 3 header bits
501	containing the following data:
502
503	first bit BFINAL
504	next 2 bits BTYPE
505
506	Note that the header bits do not necessarily begin on a byte
507	boundary, since a block does not necessarily occupy an integral
508	number of bytes.
509
510
511
512
513
514	Deutsch Informational [Page 9]
515
516
517	RFC 1951 DEFLATE Compressed Data Format Specification May 1996
518
519
520	BFINAL is set if and only if this is the last block of the data
521	set.
522
523	BTYPE specifies how the data are compressed, as follows:
524
525	00 - no compression
526	01 - compressed with fixed Huffman codes
527	10 - compressed with dynamic Huffman codes
528	11 - reserved (error)
529
530	The only difference between the two compressed cases is how the
531	Huffman codes for the literal/length and distance alphabets are
532	defined.
533
534	In all cases, the decoding algorithm for the actual data is as
535	follows:
536
537	do
538	read block header from input stream.
539	if stored with no compression
540	skip any remaining bits in current partially
541	processed byte
542	read LEN and NLEN (see next section)
543	copy LEN bytes of data to output
544	otherwise
545	if compressed with dynamic Huffman codes
546	read representation of code trees (see
547	subsection below)
548	loop (until end of block code recognized)
549	decode literal/length value from input stream
550	if value < 256
551	copy value (literal byte) to output stream
552	otherwise
553	if value = end of block (256)
554	break from loop
555	otherwise (value = 257..285)
556	decode distance from input stream
557
558	move backwards distance bytes in the output
559	stream, and copy length bytes from this
560	position to the output stream.
561	end loop
562	while not last block
563
564	Note that a duplicated string reference may refer to a string
565	in a previous block; i.e., the backward distance may cross one
566	or more block boundaries. However a distance cannot refer past
567	the beginning of the output stream. (An application using a
568
569
570
571	Deutsch Informational [Page 10]
572
573
574	RFC 1951 DEFLATE Compressed Data Format Specification May 1996
575
576
577	preset dictionary might discard part of the output stream; a
578	distance can refer to that part of the output stream anyway)
579	Note also that the referenced string may overlap the current
580	position; for example, if the last 2 bytes decoded have values
581	X and Y, a string reference with <length = 5, distance = 2>
582	adds X,Y,X,Y,X to the output stream.
583
584	We now specify each compression method in turn.
585
586	3.2.4. Non-compressed blocks (BTYPE=00)
587
588	Any bits of input up to the next byte boundary are ignored.
589	The rest of the block consists of the following information:
590
591	0 1 2 3 4...
592	+---+---+---+---+================================+
593	\| LEN \| NLEN \|... LEN bytes of literal data...\|
594	+---+---+---+---+================================+
595
596	LEN is the number of data bytes in the block. NLEN is the
597	one's complement of LEN.
598
599	3.2.5. Compressed blocks (length and distance codes)
600
601	As noted above, encoded data blocks in the "deflate" format
602	consist of sequences of symbols drawn from three conceptually
603	distinct alphabets: either literal bytes, from the alphabet of
604	byte values (0..255), or <length, backward distance> pairs,
605	where the length is drawn from (3..258) and the distance is
606	drawn from (1..32,768). In fact, the literal and length
607	alphabets are merged into a single alphabet (0..285), where
608	values 0..255 represent literal bytes, the value 256 indicates
609	end-of-block, and values 257..285 represent length codes
610	(possibly in conjunction with extra bits following the symbol
611	code) as follows:
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628	Deutsch Informational [Page 11]
629
630
631	RFC 1951 DEFLATE Compressed Data Format Specification May 1996
632
633
634	Extra Extra Extra
635	Code Bits Length(s) Code Bits Lengths Code Bits Length(s)
636	---- ---- ------ ---- ---- ------- ---- ---- -------
637	257 0 3 267 1 15,16 277 4 67-82
638	258 0 4 268 1 17,18 278 4 83-98
639	259 0 5 269 2 19-22 279 4 99-114
640	260 0 6 270 2 23-26 280 4 115-130
641	261 0 7 271 2 27-30 281 5 131-162
642	262 0 8 272 2 31-34 282 5 163-194
643	263 0 9 273 3 35-42 283 5 195-226
644	264 0 10 274 3 43-50 284 5 227-257
645	265 1 11,12 275 3 51-58 285 0 258
646	266 1 13,14 276 3 59-66
647
648	The extra bits should be interpreted as a machine integer
649	stored with the most-significant bit first, e.g., bits 1110
650	represent the value 14.
651
652	Extra Extra Extra
653	Code Bits Dist Code Bits Dist Code Bits Distance
654	---- ---- ---- ---- ---- ------ ---- ---- --------
655	0 0 1 10 4 33-48 20 9 1025-1536
656	1 0 2 11 4 49-64 21 9 1537-2048
657	2 0 3 12 5 65-96 22 10 2049-3072
658	3 0 4 13 5 97-128 23 10 3073-4096
659	4 1 5,6 14 6 129-192 24 11 4097-6144
660	5 1 7,8 15 6 193-256 25 11 6145-8192
661	6 2 9-12 16 7 257-384 26 12 8193-12288
662	7 2 13-16 17 7 385-512 27 12 12289-16384
663	8 3 17-24 18 8 513-768 28 13 16385-24576
664	9 3 25-32 19 8 769-1024 29 13 24577-32768
665
666	3.2.6. Compression with fixed Huffman codes (BTYPE=01)
667
668	The Huffman codes for the two alphabets are fixed, and are not
669	represented explicitly in the data. The Huffman code lengths
670	for the literal/length alphabet are:
671
672	Lit Value Bits Codes
673	--------- ---- -----
674	0 - 143 8 00110000 through
675	10111111
676	144 - 255 9 110010000 through
677	111111111
678	256 - 279 7 0000000 through
679	0010111
680	280 - 287 8 11000000 through
681	11000111
682
683
684
685	Deutsch Informational [Page 12]
686
687
688	RFC 1951 DEFLATE Compressed Data Format Specification May 1996
689
690
691	The code lengths are sufficient to generate the actual codes,
692	as described above; we show the codes in the table for added
693	clarity. Literal/length values 286-287 will never actually
694	occur in the compressed data, but participate in the code
695	construction.
696
697	Distance codes 0-31 are represented by (fixed-length) 5-bit
698	codes, with possible additional bits as shown in the table
699	shown in Paragraph 3.2.5, above. Note that distance codes 30-
700	31 will never actually occur in the compressed data.
701
702	3.2.7. Compression with dynamic Huffman codes (BTYPE=10)
703
704	The Huffman codes for the two alphabets appear in the block
705	immediately after the header bits and before the actual
706	compressed data, first the literal/length code and then the
707	distance code. Each code is defined by a sequence of code
708	lengths, as discussed in Paragraph 3.2.2, above. For even
709	greater compactness, the code length sequences themselves are
710	compressed using a Huffman code. The alphabet for code lengths
711	is as follows:
712
713	0 - 15: Represent code lengths of 0 - 15
714	16: Copy the previous code length 3 - 6 times.
715	The next 2 bits indicate repeat length
716	(0 = 3, ... , 3 = 6)
717	Example: Codes 8, 16 (+2 bits 11),
718	16 (+2 bits 10) will expand to
719	12 code lengths of 8 (1 + 6 + 5)
720	17: Repeat a code length of 0 for 3 - 10 times.
721	(3 bits of length)
722	18: Repeat a code length of 0 for 11 - 138 times
723	(7 bits of length)
724
725	A code length of 0 indicates that the corresponding symbol in
726	the literal/length or distance alphabet will not occur in the
727	block, and should not participate in the Huffman code
728	construction algorithm given earlier. If only one distance
729	code is used, it is encoded using one bit, not zero bits; in
730	this case there is a single code length of one, with one unused
731	code. One distance code of zero bits means that there are no
732	distance codes used at all (the data is all literals).
733
734	We can now define the format of the block:
735
736	5 Bits: HLIT, # of Literal/Length codes - 257 (257 - 286)
737	5 Bits: HDIST, # of Distance codes - 1 (1 - 32)
738	4 Bits: HCLEN, # of Code Length codes - 4 (4 - 19)
739
740
741
742	Deutsch Informational [Page 13]
743
744
745	RFC 1951 DEFLATE Compressed Data Format Specification May 1996
746
747
748	(HCLEN + 4) x 3 bits: code lengths for the code length
749	alphabet given just above, in the order: 16, 17, 18,
750	0, 8, 7, 9, 6, 10, 5, 11, 4, 12, 3, 13, 2, 14, 1, 15
751
752	These code lengths are interpreted as 3-bit integers
753	(0-7); as above, a code length of 0 means the
754	corresponding symbol (literal/length or distance code
755	length) is not used.
756
757	HLIT + 257 code lengths for the literal/length alphabet,
758	encoded using the code length Huffman code
759
760	HDIST + 1 code lengths for the distance alphabet,
761	encoded using the code length Huffman code
762
763	The actual compressed data of the block,
764	encoded using the literal/length and distance Huffman
765	codes
766
767	The literal/length symbol 256 (end of data),
768	encoded using the literal/length Huffman code
769
770	The code length repeat codes can cross from HLIT + 257 to the
771	HDIST + 1 code lengths. In other words, all code lengths form
772	a single sequence of HLIT + HDIST + 258 values.
773
774	3.3. Compliance
775
776	A compressor may limit further the ranges of values specified in
777	the previous section and still be compliant; for example, it may
778	limit the range of backward pointers to some value smaller than
779	32K. Similarly, a compressor may limit the size of blocks so that
780	a compressible block fits in memory.
781
782	A compliant decompressor must accept the full range of possible
783	values defined in the previous section, and must accept blocks of
784	arbitrary size.
785
786	4. Compression algorithm details
787
788	While it is the intent of this document to define the "deflate"
789	compressed data format without reference to any particular
790	compression algorithm, the format is related to the compressed
791	formats produced by LZ77 (Lempel-Ziv 1977, see reference [2] below);
792	since many variations of LZ77 are patented, it is strongly
793	recommended that the implementor of a compressor follow the general
794	algorithm presented here, which is known not to be patented per se.
795	The material in this section is not part of the definition of the
796
797
798
799	Deutsch Informational [Page 14]
800
801
802	RFC 1951 DEFLATE Compressed Data Format Specification May 1996
803
804
805	specification per se, and a compressor need not follow it in order to
806	be compliant.
807
808	The compressor terminates a block when it determines that starting a
809	new block with fresh trees would be useful, or when the block size
810	fills up the compressor's block buffer.
811
812	The compressor uses a chained hash table to find duplicated strings,
813	using a hash function that operates on 3-byte sequences. At any
814	given point during compression, let XYZ be the next 3 input bytes to
815	be examined (not necessarily all different, of course). First, the
816	compressor examines the hash chain for XYZ. If the chain is empty,
817	the compressor simply writes out X as a literal byte and advances one
818	byte in the input. If the hash chain is not empty, indicating that
819	the sequence XYZ (or, if we are unlucky, some other 3 bytes with the
820	same hash function value) has occurred recently, the compressor
821	compares all strings on the XYZ hash chain with the actual input data
822	sequence starting at the current point, and selects the longest
823	match.
824
825	The compressor searches the hash chains starting with the most recent
826	strings, to favor small distances and thus take advantage of the
827	Huffman encoding. The hash chains are singly linked. There are no
828	deletions from the hash chains; the algorithm simply discards matches
829	that are too old. To avoid a worst-case situation, very long hash
830	chains are arbitrarily truncated at a certain length, determined by a
831	run-time parameter.
832
833	To improve overall compression, the compressor optionally defers the
834	selection of matches ("lazy matching"): after a match of length N has
835	been found, the compressor searches for a longer match starting at
836	the next input byte. If it finds a longer match, it truncates the
837	previous match to a length of one (thus producing a single literal
838	byte) and then emits the longer match. Otherwise, it emits the
839	original match, and, as described above, advances N bytes before
840	continuing.
841
842	Run-time parameters also control this "lazy match" procedure. If
843	compression ratio is most important, the compressor attempts a
844	complete second search regardless of the length of the first match.
845	In the normal case, if the current match is "long enough", the
846	compressor reduces the search for a longer match, thus speeding up
847	the process. If speed is most important, the compressor inserts new
848	strings in the hash table only when no match was found, or when the
849	match is not "too long". This degrades the compression ratio but
850	saves time since there are both fewer insertions and fewer searches.
851
852
853
854
855
856	Deutsch Informational [Page 15]
857
858
859	RFC 1951 DEFLATE Compressed Data Format Specification May 1996
860
861
862	5. References
863
864	[1] Huffman, D. A., "A Method for the Construction of Minimum
865	Redundancy Codes", Proceedings of the Institute of Radio
866	Engineers, September 1952, Volume 40, Number 9, pp. 1098-1101.
867
868	[2] Ziv J., Lempel A., "A Universal Algorithm for Sequential Data
869	Compression", IEEE Transactions on Information Theory, Vol. 23,
870	No. 3, pp. 337-343.
871
872	[3] Gailly, J.-L., and Adler, M., ZLIB documentation and sources,
873	available in ftp://ftp.uu.net/pub/archiving/zip/doc/
874
875	[4] Gailly, J.-L., and Adler, M., GZIP documentation and sources,
876	available as gzip-*.tar in ftp://prep.ai.mit.edu/pub/gnu/
877
878	[5] Schwartz, E. S., and Kallick, B. "Generating a canonical prefix
879	encoding." Comm. ACM, 7,3 (Mar. 1964), pp. 166-169.
880
881	[6] Hirschberg and Lelewer, "Efficient decoding of prefix codes,"
882	Comm. ACM, 33,4, April 1990, pp. 449-459.
883
884	6. Security Considerations
885
886	Any data compression method involves the reduction of redundancy in
887	the data. Consequently, any corruption of the data is likely to have
888	severe effects and be difficult to correct. Uncompressed text, on
889	the other hand, will probably still be readable despite the presence
890	of some corrupted bytes.
891
892	It is recommended that systems using this data format provide some
893	means of validating the integrity of the compressed data. See
894	reference [3], for example.
895
896	7. Source code
897
898	Source code for a C language implementation of a "deflate" compliant
899	compressor and decompressor is available within the zlib package at
900	ftp://ftp.uu.net/pub/archiving/zip/zlib/.
901
902	8. Acknowledgements
903
904	Trademarks cited in this document are the property of their
905	respective owners.
906
907	Phil Katz designed the deflate format. Jean-Loup Gailly and Mark
908	Adler wrote the related software described in this specification.
909	Glenn Randers-Pehrson converted this document to RFC and HTML format.
910
911
912
913	Deutsch Informational [Page 16]
914
915
916	RFC 1951 DEFLATE Compressed Data Format Specification May 1996
917
918
919	9. Author's Address
920
921	L. Peter Deutsch
922	Aladdin Enterprises
923	203 Santa Margarita Ave.
924	Menlo Park, CA 94025
925
926	Phone: (415) 322-0103 (AM only)
927	FAX: (415) 322-1734
928	EMail: <[email protected]>
929
930	Questions about the technical content of this specification can be
931	sent by email to:
932
933	Jean-Loup Gailly <[email protected]> and
934	Mark Adler <[email protected]>
935
936	Editorial comments on this specification can be sent by email to:
937
938	L. Peter Deutsch <[email protected]> and
939	Glenn Randers-Pehrson <[email protected]>
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970	Deutsch Informational [Page 17]
971
972

	--- a/compat/zlib/doc/rfc1951.txt
	+++ b/compat/zlib/doc/rfc1951.txt
	@@ -1,972 +0,0 @@

D compat/zlib/doc/rfc1952.txt

-687

		--- a/compat/zlib/doc/rfc1952.txt
		+++ b/compat/zlib/doc/rfc1952.txt
		@@ -1,687 +0,0 @@
1		-
2		-
3		-
4		-
5		-
6		-
7		-Network Working Group P. Deutsch
8		-Request for Comments: 1952 Aladdin Enterprises
9		-Category: Informational May 1996
10		-
11		-
12		- GZIP file format specification version 4.3
13		-
14		-Status of This Memo
15		-
16		- This memo provides information for the Internet community. This memo
17		- does not specify an Internet standard of any kind. Distribution of
18		- this memo is unlimited.
19		-
20		-IESG Note:
21		-
22		- The IESG takes no position on the validity of any Intellectual
23		- Property Rights statements contained in this document.
24		-
25		-Notices
26		-
27		- Copyright (c) 1996 L. Peter Deutsch
28		-
29		- Permission is granted to copy and distribute this document for any
30		- purpose and without charge, including translations into other
31		- languages and incorporation into compilations, provided that the
32		- copyright notice and this notice are preserved, and that any
33		- substantive changes or deletions from the original are clearly
34		- marked.
35		-
36		- A pointer to the latest version of this and related documentation in
37		- HTML format can be found at the URL
38		- <ftp://ftp.uu.net/graphics/png/documents/zlib/zdoc-index.html>.
39		-
40		-Abstract
41		-
42		- This specification defines a lossless compressed data format that is
43		- compatible with the widely used GZIP utility. The format includes a
44		- cyclic redundancy check value for detecting data corruption. The
45		- format presently uses the DEFLATE method of compression but can be
46		- easily extended to use other compression methods. The format can be
47		- implemented readily in a manner not covered by patents.
48		-
49		-
50		-
51		-
52		-
53		-
54		-
55		-
56		-
57		-
58		-Deutsch Informational [Page 1]
59		-
60		-
61		-RFC 1952 GZIP File Format Specification May 1996
62		-
63		-
64		-Table of Contents
65		-
66		- 1. Introduction ................................................... 2
67		- 1.1. Purpose ................................................... 2
68		- 1.2. Intended audience ......................................... 3
69		- 1.3. Scope ..................................................... 3
70		- 1.4. Compliance ................................................ 3
71		- 1.5. Definitions of terms and conventions used ................. 3
72		- 1.6. Changes from previous versions ............................ 3
73		- 2. Detailed specification ......................................... 4
74		- 2.1. Overall conventions ....................................... 4
75		- 2.2. File format ............................................... 5
76		- 2.3. Member format ............................................. 5
77		- 2.3.1. Member header and trailer ........................... 6
78		- 2.3.1.1. Extra field ................................... 8
79		- 2.3.1.2. Compliance .................................... 9
80		- 3. References .................................................. 9
81		- 4. Security Considerations .................................... 10
82		- 5. Acknowledgements ........................................... 10
83		- 6. Author's Address ........................................... 10
84		- 7. Appendix: Jean-Loup Gailly's gzip utility .................. 11
85		- 8. Appendix: Sample CRC Code .................................. 11
86		-
87		-1. Introduction
88		-
89		- 1.1. Purpose
90		-
91		- The purpose of this specification is to define a lossless
92		- compressed data format that:
93		-
94		- * Is independent of CPU type, operating system, file system,
95		- and character set, and hence can be used for interchange;
96		- * Can compress or decompress a data stream (as opposed to a
97		- randomly accessible file) to produce another data stream,
98		- using only an a priori bounded amount of intermediate
99		- storage, and hence can be used in data communications or
100		- similar structures such as Unix filters;
101		- * Compresses data with efficiency comparable to the best
102		- currently available general-purpose compression methods,
103		- and in particular considerably better than the "compress"
104		- program;
105		- * Can be implemented readily in a manner not covered by
106		- patents, and hence can be practiced freely;
107		- * Is compatible with the file format produced by the current
108		- widely used gzip utility, in that conforming decompressors
109		- will be able to read data produced by the existing gzip
110		- compressor.
111		-
112		-
113		-
114		-
115		-Deutsch Informational [Page 2]
116		-
117		-
118		-RFC 1952 GZIP File Format Specification May 1996
119		-
120		-
121		- The data format defined by this specification does not attempt to:
122		-
123		- * Provide random access to compressed data;
124		- * Compress specialized data (e.g., raster graphics) as well as
125		- the best currently available specialized algorithms.
126		-
127		- 1.2. Intended audience
128		-
129		- This specification is intended for use by implementors of software
130		- to compress data into gzip format and/or decompress data from gzip
131		- format.
132		-
133		- The text of the specification assumes a basic background in
134		- programming at the level of bits and other primitive data
135		- representations.
136		-
137		- 1.3. Scope
138		-
139		- The specification specifies a compression method and a file format
140		- (the latter assuming only that a file can store a sequence of
141		- arbitrary bytes). It does not specify any particular interface to
142		- a file system or anything about character sets or encodings
143		- (except for file names and comments, which are optional).
144		-
145		- 1.4. Compliance
146		-
147		- Unless otherwise indicated below, a compliant decompressor must be
148		- able to accept and decompress any file that conforms to all the
149		- specifications presented here; a compliant compressor must produce
150		- files that conform to all the specifications presented here. The
151		- material in the appendices is not part of the specification per se
152		- and is not relevant to compliance.
153		-
154		- 1.5. Definitions of terms and conventions used
155		-
156		- byte: 8 bits stored or transmitted as a unit (same as an octet).
157		- (For this specification, a byte is exactly 8 bits, even on
158		- machines which store a character on a number of bits different
159		- from 8.) See below for the numbering of bits within a byte.
160		-
161		- 1.6. Changes from previous versions
162		-
163		- There have been no technical changes to the gzip format since
164		- version 4.1 of this specification. In version 4.2, some
165		- terminology was changed, and the sample CRC code was rewritten for
166		- clarity and to eliminate the requirement for the caller to do pre-
167		- and post-conditioning. Version 4.3 is a conversion of the
168		- specification to RFC style.
169		-
170		-
171		-
172		-Deutsch Informational [Page 3]
173		-
174		-
175		-RFC 1952 GZIP File Format Specification May 1996
176		-
177		-
178		-2. Detailed specification
179		-
180		- 2.1. Overall conventions
181		-
182		- In the diagrams below, a box like this:
183		-
184		- +---+
185		- \| \| <-- the vertical bars might be missing
186		- +---+
187		-
188		- represents one byte; a box like this:
189		-
190		- +==============+
191		- \| \|
192		- +==============+
193		-
194		- represents a variable number of bytes.
195		-
196		- Bytes stored within a computer do not have a "bit order", since
197		- they are always treated as a unit. However, a byte considered as
198		- an integer between 0 and 255 does have a most- and least-
199		- significant bit, and since we write numbers with the most-
200		- significant digit on the left, we also write bytes with the most-
201		- significant bit on the left. In the diagrams below, we number the
202		- bits of a byte so that bit 0 is the least-significant bit, i.e.,
203		- the bits are numbered:
204		-
205		- +--------+
206		- \|76543210\|
207		- +--------+
208		-
209		- This document does not address the issue of the order in which
210		- bits of a byte are transmitted on a bit-sequential medium, since
211		- the data format described here is byte- rather than bit-oriented.
212		-
213		- Within a computer, a number may occupy multiple bytes. All
214		- multi-byte numbers in the format described here are stored with
215		- the least-significant byte first (at the lower memory address).
216		- For example, the decimal number 520 is stored as:
217		-
218		- 0 1
219		- +--------+--------+
220		- \|00001000\|00000010\|
221		- +--------+--------+
222		- ^ ^
223		- \| \|
224		- \| + more significant byte = 2 x 256
225		- + less significant byte = 8
226		-
227		-
228		-
229		-Deutsch Informational [Page 4]
230		-
231		-
232		-RFC 1952 GZIP File Format Specification May 1996
233		-
234		-
235		- 2.2. File format
236		-
237		- A gzip file consists of a series of "members" (compressed data
238		- sets). The format of each member is specified in the following
239		- section. The members simply appear one after another in the file,
240		- with no additional information before, between, or after them.
241		-
242		- 2.3. Member format
243		-
244		- Each member has the following structure:
245		-
246		- +---+---+---+---+---+---+---+---+---+---+
247		- \|ID1\|ID2\|CM \|FLG\| MTIME \|XFL\|OS \| (more-->)
248		- +---+---+---+---+---+---+---+---+---+---+
249		-
250		- (if FLG.FEXTRA set)
251		-
252		- +---+---+=================================+
253		- \| XLEN \|...XLEN bytes of "extra field"...\| (more-->)
254		- +---+---+=================================+
255		-
256		- (if FLG.FNAME set)
257		-
258		- +=========================================+
259		- \|...original file name, zero-terminated...\| (more-->)
260		- +=========================================+
261		-
262		- (if FLG.FCOMMENT set)
263		-
264		- +===================================+
265		- \|...file comment, zero-terminated...\| (more-->)
266		- +===================================+
267		-
268		- (if FLG.FHCRC set)
269		-
270		- +---+---+
271		- \| CRC16 \|
272		- +---+---+
273		-
274		- +=======================+
275		- \|...compressed blocks...\| (more-->)
276		- +=======================+
277		-
278		- 0 1 2 3 4 5 6 7
279		- +---+---+---+---+---+---+---+---+
280		- \| CRC32 \| ISIZE \|
281		- +---+---+---+---+---+---+---+---+
282		-
283		-
284		-
285		-
286		-Deutsch Informational [Page 5]
287		-
288		-
289		-RFC 1952 GZIP File Format Specification May 1996
290		-
291		-
292		- 2.3.1. Member header and trailer
293		-
294		- ID1 (IDentification 1)
295		- ID2 (IDentification 2)
296		- These have the fixed values ID1 = 31 (0x1f, \037), ID2 = 139
297		- (0x8b, \213), to identify the file as being in gzip format.
298		-
299		- CM (Compression Method)
300		- This identifies the compression method used in the file. CM
301		- = 0-7 are reserved. CM = 8 denotes the "deflate"
302		- compression method, which is the one customarily used by
303		- gzip and which is documented elsewhere.
304		-
305		- FLG (FLaGs)
306		- This flag byte is divided into individual bits as follows:
307		-
308		- bit 0 FTEXT
309		- bit 1 FHCRC
310		- bit 2 FEXTRA
311		- bit 3 FNAME
312		- bit 4 FCOMMENT
313		- bit 5 reserved
314		- bit 6 reserved
315		- bit 7 reserved
316		-
317		- If FTEXT is set, the file is probably ASCII text. This is
318		- an optional indication, which the compressor may set by
319		- checking a small amount of the input data to see whether any
320		- non-ASCII characters are present. In case of doubt, FTEXT
321		- is cleared, indicating binary data. For systems which have
322		- different file formats for ascii text and binary data, the
323		- decompressor can use FTEXT to choose the appropriate format.
324		- We deliberately do not specify the algorithm used to set
325		- this bit, since a compressor always has the option of
326		- leaving it cleared and a decompressor always has the option
327		- of ignoring it and letting some other program handle issues
328		- of data conversion.
329		-
330		- If FHCRC is set, a CRC16 for the gzip header is present,
331		- immediately before the compressed data. The CRC16 consists
332		- of the two least significant bytes of the CRC32 for all
333		- bytes of the gzip header up to and not including the CRC16.
334		- [The FHCRC bit was never set by versions of gzip up to
335		- 1.2.4, even though it was documented with a different
336		- meaning in gzip 1.2.4.]
337		-
338		- If FEXTRA is set, optional extra fields are present, as
339		- described in a following section.
340		-
341		-
342		-
343		-Deutsch Informational [Page 6]
344		-
345		-
346		-RFC 1952 GZIP File Format Specification May 1996
347		-
348		-
349		- If FNAME is set, an original file name is present,
350		- terminated by a zero byte. The name must consist of ISO
351		- 8859-1 (LATIN-1) characters; on operating systems using
352		- EBCDIC or any other character set for file names, the name
353		- must be translated to the ISO LATIN-1 character set. This
354		- is the original name of the file being compressed, with any
355		- directory components removed, and, if the file being
356		- compressed is on a file system with case insensitive names,
357		- forced to lower case. There is no original file name if the
358		- data was compressed from a source other than a named file;
359		- for example, if the source was stdin on a Unix system, there
360		- is no file name.
361		-
362		- If FCOMMENT is set, a zero-terminated file comment is
363		- present. This comment is not interpreted; it is only
364		- intended for human consumption. The comment must consist of
365		- ISO 8859-1 (LATIN-1) characters. Line breaks should be
366		- denoted by a single line feed character (10 decimal).
367		-
368		- Reserved FLG bits must be zero.
369		-
370		- MTIME (Modification TIME)
371		- This gives the most recent modification time of the original
372		- file being compressed. The time is in Unix format, i.e.,
373		- seconds since 00:00:00 GMT, Jan. 1, 1970. (Note that this
374		- may cause problems for MS-DOS and other systems that use
375		- local rather than Universal time.) If the compressed data
376		- did not come from a file, MTIME is set to the time at which
377		- compression started. MTIME = 0 means no time stamp is
378		- available.
379		-
380		- XFL (eXtra FLags)
381		- These flags are available for use by specific compression
382		- methods. The "deflate" method (CM = 8) sets these flags as
383		- follows:
384		-
385		- XFL = 2 - compressor used maximum compression,
386		- slowest algorithm
387		- XFL = 4 - compressor used fastest algorithm
388		-
389		- OS (Operating System)
390		- This identifies the type of file system on which compression
391		- took place. This may be useful in determining end-of-line
392		- convention for text files. The currently defined values are
393		- as follows:
394		-
395		-
396		-
397		-
398		-
399		-
400		-Deutsch Informational [Page 7]
401		-
402		-
403		-RFC 1952 GZIP File Format Specification May 1996
404		-
405		-
406		- 0 - FAT filesystem (MS-DOS, OS/2, NT/Win32)
407		- 1 - Amiga
408		- 2 - VMS (or OpenVMS)
409		- 3 - Unix
410		- 4 - VM/CMS
411		- 5 - Atari TOS
412		- 6 - HPFS filesystem (OS/2, NT)
413		- 7 - Macintosh
414		- 8 - Z-System
415		- 9 - CP/M
416		- 10 - TOPS-20
417		- 11 - NTFS filesystem (NT)
418		- 12 - QDOS
419		- 13 - Acorn RISCOS
420		- 255 - unknown
421		-
422		- XLEN (eXtra LENgth)
423		- If FLG.FEXTRA is set, this gives the length of the optional
424		- extra field. See below for details.
425		-
426		- CRC32 (CRC-32)
427		- This contains a Cyclic Redundancy Check value of the
428		- uncompressed data computed according to CRC-32 algorithm
429		- used in the ISO 3309 standard and in section 8.1.1.6.2 of
430		- ITU-T recommendation V.42. (See http://www.iso.ch for
431		- ordering ISO documents. See gopher://info.itu.ch for an
432		- online version of ITU-T V.42.)
433		-
434		- ISIZE (Input SIZE)
435		- This contains the size of the original (uncompressed) input
436		- data modulo 2^32.
437		-
438		- 2.3.1.1. Extra field
439		-
440		- If the FLG.FEXTRA bit is set, an "extra field" is present in
441		- the header, with total length XLEN bytes. It consists of a
442		- series of subfields, each of the form:
443		-
444		- +---+---+---+---+==================================+
445		- \|SI1\|SI2\| LEN \|... LEN bytes of subfield data ...\|
446		- +---+---+---+---+==================================+
447		-
448		- SI1 and SI2 provide a subfield ID, typically two ASCII letters
449		- with some mnemonic value. Jean-Loup Gailly
450		- <[email protected]> is maintaining a registry of subfield
451		- IDs; please send him any subfield ID you wish to use. Subfield
452		- IDs with SI2 = 0 are reserved for future use. The following
453		- IDs are currently defined:
454		-
455		-
456		-
457		-Deutsch Informational [Page 8]
458		-
459		-
460		-RFC 1952 GZIP File Format Specification May 1996
461		-
462		-
463		- SI1 SI2 Data
464		- ---------- ---------- ----
465		- 0x41 ('A') 0x70 ('P') Apollo file type information
466		-
467		- LEN gives the length of the subfield data, excluding the 4
468		- initial bytes.
469		-
470		- 2.3.1.2. Compliance
471		-
472		- A compliant compressor must produce files with correct ID1,
473		- ID2, CM, CRC32, and ISIZE, but may set all the other fields in
474		- the fixed-length part of the header to default values (255 for
475		- OS, 0 for all others). The compressor must set all reserved
476		- bits to zero.
477		-
478		- A compliant decompressor must check ID1, ID2, and CM, and
479		- provide an error indication if any of these have incorrect
480		- values. It must examine FEXTRA/XLEN, FNAME, FCOMMENT and FHCRC
481		- at least so it can skip over the optional fields if they are
482		- present. It need not examine any other part of the header or
483		- trailer; in particular, a decompressor may ignore FTEXT and OS
484		- and always produce binary output, and still be compliant. A
485		- compliant decompressor must give an error indication if any
486		- reserved bit is non-zero, since such a bit could indicate the
487		- presence of a new field that would cause subsequent data to be
488		- interpreted incorrectly.
489		-
490		-3. References
491		-
492		- [1] "Information Processing - 8-bit single-byte coded graphic
493		- character sets - Part 1: Latin alphabet No.1" (ISO 8859-1:1987).
494		- The ISO 8859-1 (Latin-1) character set is a superset of 7-bit
495		- ASCII. Files defining this character set are available as
496		- iso_8859-1.* in ftp://ftp.uu.net/graphics/png/documents/
497		-
498		- [2] ISO 3309
499		-
500		- [3] ITU-T recommendation V.42
501		-
502		- [4] Deutsch, L.P.,"DEFLATE Compressed Data Format Specification",
503		- available in ftp://ftp.uu.net/pub/archiving/zip/doc/
504		-
505		- [5] Gailly, J.-L., GZIP documentation, available as gzip-*.tar in
506		- ftp://prep.ai.mit.edu/pub/gnu/
507		-
508		- [6] Sarwate, D.V., "Computation of Cyclic Redundancy Checks via Table
509		- Look-Up", Communications of the ACM, 31(8), pp.1008-1013.
510		-
511		-
512		-
513		-
514		-Deutsch Informational [Page 9]
515		-
516		-
517		-RFC 1952 GZIP File Format Specification May 1996
518		-
519		-
520		- [7] Schwaderer, W.D., "CRC Calculation", April 85 PC Tech Journal,
521		- pp.118-133.
522		-
523		- [8] ftp://ftp.adelaide.edu.au/pub/rocksoft/papers/crc_v3.txt,
524		- describing the CRC concept.
525		-
526		-4. Security Considerations
527		-
528		- Any data compression method involves the reduction of redundancy in
529		- the data. Consequently, any corruption of the data is likely to have
530		- severe effects and be difficult to correct. Uncompressed text, on
531		- the other hand, will probably still be readable despite the presence
532		- of some corrupted bytes.
533		-
534		- It is recommended that systems using this data format provide some
535		- means of validating the integrity of the compressed data, such as by
536		- setting and checking the CRC-32 check value.
537		-
538		-5. Acknowledgements
539		-
540		- Trademarks cited in this document are the property of their
541		- respective owners.
542		-
543		- Jean-Loup Gailly designed the gzip format and wrote, with Mark Adler,
544		- the related software described in this specification. Glenn
545		- Randers-Pehrson converted this document to RFC and HTML format.
546		-
547		-6. Author's Address
548		-
549		- L. Peter Deutsch
550		- Aladdin Enterprises
551		- 203 Santa Margarita Ave.
552		- Menlo Park, CA 94025
553		-
554		- Phone: (415) 322-0103 (AM only)
555		- FAX: (415) 322-1734
556		- EMail: <[email protected]>
557		-
558		- Questions about the technical content of this specification can be
559		- sent by email to:
560		-
561		- Jean-Loup Gailly <[email protected]> and
562		- Mark Adler <[email protected]>
563		-
564		- Editorial comments on this specification can be sent by email to:
565		-
566		- L. Peter Deutsch <[email protected]> and
567		- Glenn Randers-Pehrson <[email protected]>
568		-
569		-
570		-
571		-Deutsch Informational [Page 10]
572		-
573		-
574		-RFC 1952 GZIP File Format Specification May 1996
575		-
576		-
577		-7. Appendix: Jean-Loup Gailly's gzip utility
578		-
579		- The most widely used implementation of gzip compression, and the
580		- original documentation on which this specification is based, were
581		- created by Jean-Loup Gailly <[email protected]>. Since this
582		- implementation is a de facto standard, we mention some more of its
583		- features here. Again, the material in this section is not part of
584		- the specification per se, and implementations need not follow it to
585		- be compliant.
586		-
587		- When compressing or decompressing a file, gzip preserves the
588		- protection, ownership, and modification time attributes on the local
589		- file system, since there is no provision for representing protection
590		- attributes in the gzip file format itself. Since the file format
591		- includes a modification time, the gzip decompressor provides a
592		- command line switch that assigns the modification time from the file,
593		- rather than the local modification time of the compressed input, to
594		- the decompressed output.
595		-
596		-8. Appendix: Sample CRC Code
597		-
598		- The following sample code represents a practical implementation of
599		- the CRC (Cyclic Redundancy Check). (See also ISO 3309 and ITU-T V.42
600		- for a formal specification.)
601		-
602		- The sample code is in the ANSI C programming language. Non C users
603		- may find it easier to read with these hints:
604		-
605		- & Bitwise AND operator.
606		- ^ Bitwise exclusive-OR operator.
607		- >> Bitwise right shift operator. When applied to an
608		- unsigned quantity, as here, right shift inserts zero
609		- bit(s) at the left.
610		- ! Logical NOT operator.
611		- ++ "n++" increments the variable n.
612		- 0xNNN 0x introduces a hexadecimal (base 16) constant.
613		- Suffix L indicates a long value (at least 32 bits).
614		-
615		- /* Table of CRCs of all 8-bit messages. */
616		- unsigned long crc_table[256];
617		-
618		- /* Flag: has the table been computed? Initially false. */
619		- int crc_table_computed = 0;
620		-
621		- /* Make the table for a fast CRC. */
622		- void make_crc_table(void)
623		- {
624		- unsigned long c;
625		-
626		-
627		-
628		-Deutsch Informational [Page 11]
629		-
630		-
631		-RFC 1952 GZIP File Format Specification May 1996
632		-
633		-
634		- int n, k;
635		- for (n = 0; n < 256; n++) {
636		- c = (unsigned long) n;
637		- for (k = 0; k < 8; k++) {
638		- if (c & 1) {
639		- c = 0xedb88320L ^ (c >> 1);
640		- } else {
641		- c = c >> 1;
642		- }
643		- }
644		- crc_table[n] = c;
645		- }
646		- crc_table_computed = 1;
647		- }
648		-
649		- /*
650		- Update a running crc with the bytes buf[0..len-1] and return
651		- the updated crc. The crc should be initialized to zero. Pre- and
652		- post-conditioning (one's complement) is performed within this
653		- function so it shouldn't be done by the caller. Usage example:
654		-
655		- unsigned long crc = 0L;
656		-
657		- while (read_buffer(buffer, length) != EOF) {
658		- crc = update_crc(crc, buffer, length);
659		- }
660		- if (crc != original_crc) error();
661		- */
662		- unsigned long update_crc(unsigned long crc,
663		- unsigned char *buf, int len)
664		- {
665		- unsigned long c = crc ^ 0xffffffffL;
666		- int n;
667		-
668		- if (!crc_table_computed)
669		- make_crc_table();
670		- for (n = 0; n < len; n++) {
671		- c = crc_table[(c ^ buf[n]) & 0xff] ^ (c >> 8);
672		- }
673		- return c ^ 0xffffffffL;
674		- }
675		-
676		- /* Return the CRC of the bytes buf[0..len-1]. */
677		- unsigned long crc(unsigned char *buf, int len)
678		- {
679		- return update_crc(0L, buf, len);
680		- }
681		-
682		-
683		-
684		-
685		-Deutsch Informational [Page 12]
686		-
687		-

	--- a/compat/zlib/doc/rfc1952.txt
	+++ b/compat/zlib/doc/rfc1952.txt
	@@ -1,687 +0,0 @@
1
2
3
4
5
6
7	Network Working Group P. Deutsch
8	Request for Comments: 1952 Aladdin Enterprises
9	Category: Informational May 1996
10
11
12	GZIP file format specification version 4.3
13
14	Status of This Memo
15
16	This memo provides information for the Internet community. This memo
17	does not specify an Internet standard of any kind. Distribution of
18	this memo is unlimited.
19
20	IESG Note:
21
22	The IESG takes no position on the validity of any Intellectual
23	Property Rights statements contained in this document.
24
25	Notices
26
27	Copyright (c) 1996 L. Peter Deutsch
28
29	Permission is granted to copy and distribute this document for any
30	purpose and without charge, including translations into other
31	languages and incorporation into compilations, provided that the
32	copyright notice and this notice are preserved, and that any
33	substantive changes or deletions from the original are clearly
34	marked.
35
36	A pointer to the latest version of this and related documentation in
37	HTML format can be found at the URL
38	<ftp://ftp.uu.net/graphics/png/documents/zlib/zdoc-index.html>.
39
40	Abstract
41
42	This specification defines a lossless compressed data format that is
43	compatible with the widely used GZIP utility. The format includes a
44	cyclic redundancy check value for detecting data corruption. The
45	format presently uses the DEFLATE method of compression but can be
46	easily extended to use other compression methods. The format can be
47	implemented readily in a manner not covered by patents.
48
49
50
51
52
53
54
55
56
57
58	Deutsch Informational [Page 1]
59
60
61	RFC 1952 GZIP File Format Specification May 1996
62
63
64	Table of Contents
65
66	1. Introduction ................................................... 2
67	1.1. Purpose ................................................... 2
68	1.2. Intended audience ......................................... 3
69	1.3. Scope ..................................................... 3
70	1.4. Compliance ................................................ 3
71	1.5. Definitions of terms and conventions used ................. 3
72	1.6. Changes from previous versions ............................ 3
73	2. Detailed specification ......................................... 4
74	2.1. Overall conventions ....................................... 4
75	2.2. File format ............................................... 5
76	2.3. Member format ............................................. 5
77	2.3.1. Member header and trailer ........................... 6
78	2.3.1.1. Extra field ................................... 8
79	2.3.1.2. Compliance .................................... 9
80	3. References .................................................. 9
81	4. Security Considerations .................................... 10
82	5. Acknowledgements ........................................... 10
83	6. Author's Address ........................................... 10
84	7. Appendix: Jean-Loup Gailly's gzip utility .................. 11
85	8. Appendix: Sample CRC Code .................................. 11
86
87	1. Introduction
88
89	1.1. Purpose
90
91	The purpose of this specification is to define a lossless
92	compressed data format that:
93
94	* Is independent of CPU type, operating system, file system,
95	and character set, and hence can be used for interchange;
96	* Can compress or decompress a data stream (as opposed to a
97	randomly accessible file) to produce another data stream,
98	using only an a priori bounded amount of intermediate
99	storage, and hence can be used in data communications or
100	similar structures such as Unix filters;
101	* Compresses data with efficiency comparable to the best
102	currently available general-purpose compression methods,
103	and in particular considerably better than the "compress"
104	program;
105	* Can be implemented readily in a manner not covered by
106	patents, and hence can be practiced freely;
107	* Is compatible with the file format produced by the current
108	widely used gzip utility, in that conforming decompressors
109	will be able to read data produced by the existing gzip
110	compressor.
111
112
113
114
115	Deutsch Informational [Page 2]
116
117
118	RFC 1952 GZIP File Format Specification May 1996
119
120
121	The data format defined by this specification does not attempt to:
122
123	* Provide random access to compressed data;
124	* Compress specialized data (e.g., raster graphics) as well as
125	the best currently available specialized algorithms.
126
127	1.2. Intended audience
128
129	This specification is intended for use by implementors of software
130	to compress data into gzip format and/or decompress data from gzip
131	format.
132
133	The text of the specification assumes a basic background in
134	programming at the level of bits and other primitive data
135	representations.
136
137	1.3. Scope
138
139	The specification specifies a compression method and a file format
140	(the latter assuming only that a file can store a sequence of
141	arbitrary bytes). It does not specify any particular interface to
142	a file system or anything about character sets or encodings
143	(except for file names and comments, which are optional).
144
145	1.4. Compliance
146
147	Unless otherwise indicated below, a compliant decompressor must be
148	able to accept and decompress any file that conforms to all the
149	specifications presented here; a compliant compressor must produce
150	files that conform to all the specifications presented here. The
151	material in the appendices is not part of the specification per se
152	and is not relevant to compliance.
153
154	1.5. Definitions of terms and conventions used
155
156	byte: 8 bits stored or transmitted as a unit (same as an octet).
157	(For this specification, a byte is exactly 8 bits, even on
158	machines which store a character on a number of bits different
159	from 8.) See below for the numbering of bits within a byte.
160
161	1.6. Changes from previous versions
162
163	There have been no technical changes to the gzip format since
164	version 4.1 of this specification. In version 4.2, some
165	terminology was changed, and the sample CRC code was rewritten for
166	clarity and to eliminate the requirement for the caller to do pre-
167	and post-conditioning. Version 4.3 is a conversion of the
168	specification to RFC style.
169
170
171
172	Deutsch Informational [Page 3]
173
174
175	RFC 1952 GZIP File Format Specification May 1996
176
177
178	2. Detailed specification
179
180	2.1. Overall conventions
181
182	In the diagrams below, a box like this:
183
184	+---+
185	\| \| <-- the vertical bars might be missing
186	+---+
187
188	represents one byte; a box like this:
189
190	+==============+
191	\| \|
192	+==============+
193
194	represents a variable number of bytes.
195
196	Bytes stored within a computer do not have a "bit order", since
197	they are always treated as a unit. However, a byte considered as
198	an integer between 0 and 255 does have a most- and least-
199	significant bit, and since we write numbers with the most-
200	significant digit on the left, we also write bytes with the most-
201	significant bit on the left. In the diagrams below, we number the
202	bits of a byte so that bit 0 is the least-significant bit, i.e.,
203	the bits are numbered:
204
205	+--------+
206	\|76543210\|
207	+--------+
208
209	This document does not address the issue of the order in which
210	bits of a byte are transmitted on a bit-sequential medium, since
211	the data format described here is byte- rather than bit-oriented.
212
213	Within a computer, a number may occupy multiple bytes. All
214	multi-byte numbers in the format described here are stored with
215	the least-significant byte first (at the lower memory address).
216	For example, the decimal number 520 is stored as:
217
218	0 1
219	+--------+--------+
220	\|00001000\|00000010\|
221	+--------+--------+
222	^ ^
223	\| \|
224	\| + more significant byte = 2 x 256
225	+ less significant byte = 8
226
227
228
229	Deutsch Informational [Page 4]
230
231
232	RFC 1952 GZIP File Format Specification May 1996
233
234
235	2.2. File format
236
237	A gzip file consists of a series of "members" (compressed data
238	sets). The format of each member is specified in the following
239	section. The members simply appear one after another in the file,
240	with no additional information before, between, or after them.
241
242	2.3. Member format
243
244	Each member has the following structure:
245
246	+---+---+---+---+---+---+---+---+---+---+
247	\|ID1\|ID2\|CM \|FLG\| MTIME \|XFL\|OS \| (more-->)
248	+---+---+---+---+---+---+---+---+---+---+
249
250	(if FLG.FEXTRA set)
251
252	+---+---+=================================+
253	\| XLEN \|...XLEN bytes of "extra field"...\| (more-->)
254	+---+---+=================================+
255
256	(if FLG.FNAME set)
257
258	+=========================================+
259	\|...original file name, zero-terminated...\| (more-->)
260	+=========================================+
261
262	(if FLG.FCOMMENT set)
263
264	+===================================+
265	\|...file comment, zero-terminated...\| (more-->)
266	+===================================+
267
268	(if FLG.FHCRC set)
269
270	+---+---+
271	\| CRC16 \|
272	+---+---+
273
274	+=======================+
275	\|...compressed blocks...\| (more-->)
276	+=======================+
277
278	0 1 2 3 4 5 6 7
279	+---+---+---+---+---+---+---+---+
280	\| CRC32 \| ISIZE \|
281	+---+---+---+---+---+---+---+---+
282
283
284
285
286	Deutsch Informational [Page 5]
287
288
289	RFC 1952 GZIP File Format Specification May 1996
290
291
292	2.3.1. Member header and trailer
293
294	ID1 (IDentification 1)
295	ID2 (IDentification 2)
296	These have the fixed values ID1 = 31 (0x1f, \037), ID2 = 139
297	(0x8b, \213), to identify the file as being in gzip format.
298
299	CM (Compression Method)
300	This identifies the compression method used in the file. CM
301	= 0-7 are reserved. CM = 8 denotes the "deflate"
302	compression method, which is the one customarily used by
303	gzip and which is documented elsewhere.
304
305	FLG (FLaGs)
306	This flag byte is divided into individual bits as follows:
307
308	bit 0 FTEXT
309	bit 1 FHCRC
310	bit 2 FEXTRA
311	bit 3 FNAME
312	bit 4 FCOMMENT
313	bit 5 reserved
314	bit 6 reserved
315	bit 7 reserved
316
317	If FTEXT is set, the file is probably ASCII text. This is
318	an optional indication, which the compressor may set by
319	checking a small amount of the input data to see whether any
320	non-ASCII characters are present. In case of doubt, FTEXT
321	is cleared, indicating binary data. For systems which have
322	different file formats for ascii text and binary data, the
323	decompressor can use FTEXT to choose the appropriate format.
324	We deliberately do not specify the algorithm used to set
325	this bit, since a compressor always has the option of
326	leaving it cleared and a decompressor always has the option
327	of ignoring it and letting some other program handle issues
328	of data conversion.
329
330	If FHCRC is set, a CRC16 for the gzip header is present,
331	immediately before the compressed data. The CRC16 consists
332	of the two least significant bytes of the CRC32 for all
333	bytes of the gzip header up to and not including the CRC16.
334	[The FHCRC bit was never set by versions of gzip up to
335	1.2.4, even though it was documented with a different
336	meaning in gzip 1.2.4.]
337
338	If FEXTRA is set, optional extra fields are present, as
339	described in a following section.
340
341
342
343	Deutsch Informational [Page 6]
344
345
346	RFC 1952 GZIP File Format Specification May 1996
347
348
349	If FNAME is set, an original file name is present,
350	terminated by a zero byte. The name must consist of ISO
351	8859-1 (LATIN-1) characters; on operating systems using
352	EBCDIC or any other character set for file names, the name
353	must be translated to the ISO LATIN-1 character set. This
354	is the original name of the file being compressed, with any
355	directory components removed, and, if the file being
356	compressed is on a file system with case insensitive names,
357	forced to lower case. There is no original file name if the
358	data was compressed from a source other than a named file;
359	for example, if the source was stdin on a Unix system, there
360	is no file name.
361
362	If FCOMMENT is set, a zero-terminated file comment is
363	present. This comment is not interpreted; it is only
364	intended for human consumption. The comment must consist of
365	ISO 8859-1 (LATIN-1) characters. Line breaks should be
366	denoted by a single line feed character (10 decimal).
367
368	Reserved FLG bits must be zero.
369
370	MTIME (Modification TIME)
371	This gives the most recent modification time of the original
372	file being compressed. The time is in Unix format, i.e.,
373	seconds since 00:00:00 GMT, Jan. 1, 1970. (Note that this
374	may cause problems for MS-DOS and other systems that use
375	local rather than Universal time.) If the compressed data
376	did not come from a file, MTIME is set to the time at which
377	compression started. MTIME = 0 means no time stamp is
378	available.
379
380	XFL (eXtra FLags)
381	These flags are available for use by specific compression
382	methods. The "deflate" method (CM = 8) sets these flags as
383	follows:
384
385	XFL = 2 - compressor used maximum compression,
386	slowest algorithm
387	XFL = 4 - compressor used fastest algorithm
388
389	OS (Operating System)
390	This identifies the type of file system on which compression
391	took place. This may be useful in determining end-of-line
392	convention for text files. The currently defined values are
393	as follows:
394
395
396
397
398
399
400	Deutsch Informational [Page 7]
401
402
403	RFC 1952 GZIP File Format Specification May 1996
404
405
406	0 - FAT filesystem (MS-DOS, OS/2, NT/Win32)
407	1 - Amiga
408	2 - VMS (or OpenVMS)
409	3 - Unix
410	4 - VM/CMS
411	5 - Atari TOS
412	6 - HPFS filesystem (OS/2, NT)
413	7 - Macintosh
414	8 - Z-System
415	9 - CP/M
416	10 - TOPS-20
417	11 - NTFS filesystem (NT)
418	12 - QDOS
419	13 - Acorn RISCOS
420	255 - unknown
421
422	XLEN (eXtra LENgth)
423	If FLG.FEXTRA is set, this gives the length of the optional
424	extra field. See below for details.
425
426	CRC32 (CRC-32)
427	This contains a Cyclic Redundancy Check value of the
428	uncompressed data computed according to CRC-32 algorithm
429	used in the ISO 3309 standard and in section 8.1.1.6.2 of
430	ITU-T recommendation V.42. (See http://www.iso.ch for
431	ordering ISO documents. See gopher://info.itu.ch for an
432	online version of ITU-T V.42.)
433
434	ISIZE (Input SIZE)
435	This contains the size of the original (uncompressed) input
436	data modulo 2^32.
437
438	2.3.1.1. Extra field
439
440	If the FLG.FEXTRA bit is set, an "extra field" is present in
441	the header, with total length XLEN bytes. It consists of a
442	series of subfields, each of the form:
443
444	+---+---+---+---+==================================+
445	\|SI1\|SI2\| LEN \|... LEN bytes of subfield data ...\|
446	+---+---+---+---+==================================+
447
448	SI1 and SI2 provide a subfield ID, typically two ASCII letters
449	with some mnemonic value. Jean-Loup Gailly
450	<[email protected]> is maintaining a registry of subfield
451	IDs; please send him any subfield ID you wish to use. Subfield
452	IDs with SI2 = 0 are reserved for future use. The following
453	IDs are currently defined:
454
455
456
457	Deutsch Informational [Page 8]
458
459
460	RFC 1952 GZIP File Format Specification May 1996
461
462
463	SI1 SI2 Data
464	---------- ---------- ----
465	0x41 ('A') 0x70 ('P') Apollo file type information
466
467	LEN gives the length of the subfield data, excluding the 4
468	initial bytes.
469
470	2.3.1.2. Compliance
471
472	A compliant compressor must produce files with correct ID1,
473	ID2, CM, CRC32, and ISIZE, but may set all the other fields in
474	the fixed-length part of the header to default values (255 for
475	OS, 0 for all others). The compressor must set all reserved
476	bits to zero.
477
478	A compliant decompressor must check ID1, ID2, and CM, and
479	provide an error indication if any of these have incorrect
480	values. It must examine FEXTRA/XLEN, FNAME, FCOMMENT and FHCRC
481	at least so it can skip over the optional fields if they are
482	present. It need not examine any other part of the header or
483	trailer; in particular, a decompressor may ignore FTEXT and OS
484	and always produce binary output, and still be compliant. A
485	compliant decompressor must give an error indication if any
486	reserved bit is non-zero, since such a bit could indicate the
487	presence of a new field that would cause subsequent data to be
488	interpreted incorrectly.
489
490	3. References
491
492	[1] "Information Processing - 8-bit single-byte coded graphic
493	character sets - Part 1: Latin alphabet No.1" (ISO 8859-1:1987).
494	The ISO 8859-1 (Latin-1) character set is a superset of 7-bit
495	ASCII. Files defining this character set are available as
496	iso_8859-1.* in ftp://ftp.uu.net/graphics/png/documents/
497
498	[2] ISO 3309
499
500	[3] ITU-T recommendation V.42
501
502	[4] Deutsch, L.P.,"DEFLATE Compressed Data Format Specification",
503	available in ftp://ftp.uu.net/pub/archiving/zip/doc/
504
505	[5] Gailly, J.-L., GZIP documentation, available as gzip-*.tar in
506	ftp://prep.ai.mit.edu/pub/gnu/
507
508	[6] Sarwate, D.V., "Computation of Cyclic Redundancy Checks via Table
509	Look-Up", Communications of the ACM, 31(8), pp.1008-1013.
510
511
512
513
514	Deutsch Informational [Page 9]
515
516
517	RFC 1952 GZIP File Format Specification May 1996
518
519
520	[7] Schwaderer, W.D., "CRC Calculation", April 85 PC Tech Journal,
521	pp.118-133.
522
523	[8] ftp://ftp.adelaide.edu.au/pub/rocksoft/papers/crc_v3.txt,
524	describing the CRC concept.
525
526	4. Security Considerations
527
528	Any data compression method involves the reduction of redundancy in
529	the data. Consequently, any corruption of the data is likely to have
530	severe effects and be difficult to correct. Uncompressed text, on
531	the other hand, will probably still be readable despite the presence
532	of some corrupted bytes.
533
534	It is recommended that systems using this data format provide some
535	means of validating the integrity of the compressed data, such as by
536	setting and checking the CRC-32 check value.
537
538	5. Acknowledgements
539
540	Trademarks cited in this document are the property of their
541	respective owners.
542
543	Jean-Loup Gailly designed the gzip format and wrote, with Mark Adler,
544	the related software described in this specification. Glenn
545	Randers-Pehrson converted this document to RFC and HTML format.
546
547	6. Author's Address
548
549	L. Peter Deutsch
550	Aladdin Enterprises
551	203 Santa Margarita Ave.
552	Menlo Park, CA 94025
553
554	Phone: (415) 322-0103 (AM only)
555	FAX: (415) 322-1734
556	EMail: <[email protected]>
557
558	Questions about the technical content of this specification can be
559	sent by email to:
560
561	Jean-Loup Gailly <[email protected]> and
562	Mark Adler <[email protected]>
563
564	Editorial comments on this specification can be sent by email to:
565
566	L. Peter Deutsch <[email protected]> and
567	Glenn Randers-Pehrson <[email protected]>
568
569
570
571	Deutsch Informational [Page 10]
572
573
574	RFC 1952 GZIP File Format Specification May 1996
575
576
577	7. Appendix: Jean-Loup Gailly's gzip utility
578
579	The most widely used implementation of gzip compression, and the
580	original documentation on which this specification is based, were
581	created by Jean-Loup Gailly <[email protected]>. Since this
582	implementation is a de facto standard, we mention some more of its
583	features here. Again, the material in this section is not part of
584	the specification per se, and implementations need not follow it to
585	be compliant.
586
587	When compressing or decompressing a file, gzip preserves the
588	protection, ownership, and modification time attributes on the local
589	file system, since there is no provision for representing protection
590	attributes in the gzip file format itself. Since the file format
591	includes a modification time, the gzip decompressor provides a
592	command line switch that assigns the modification time from the file,
593	rather than the local modification time of the compressed input, to
594	the decompressed output.
595
596	8. Appendix: Sample CRC Code
597
598	The following sample code represents a practical implementation of
599	the CRC (Cyclic Redundancy Check). (See also ISO 3309 and ITU-T V.42
600	for a formal specification.)
601
602	The sample code is in the ANSI C programming language. Non C users
603	may find it easier to read with these hints:
604
605	& Bitwise AND operator.
606	^ Bitwise exclusive-OR operator.
607	>> Bitwise right shift operator. When applied to an
608	unsigned quantity, as here, right shift inserts zero
609	bit(s) at the left.
610	! Logical NOT operator.
611	++ "n++" increments the variable n.
612	0xNNN 0x introduces a hexadecimal (base 16) constant.
613	Suffix L indicates a long value (at least 32 bits).
614
615	/* Table of CRCs of all 8-bit messages. */
616	unsigned long crc_table[256];
617
618	/* Flag: has the table been computed? Initially false. */
619	int crc_table_computed = 0;
620
621	/* Make the table for a fast CRC. */
622	void make_crc_table(void)
623	{
624	unsigned long c;
625
626
627
628	Deutsch Informational [Page 11]
629
630
631	RFC 1952 GZIP File Format Specification May 1996
632
633
634	int n, k;
635	for (n = 0; n < 256; n++) {
636	c = (unsigned long) n;
637	for (k = 0; k < 8; k++) {
638	if (c & 1) {
639	c = 0xedb88320L ^ (c >> 1);
640	} else {
641	c = c >> 1;
642	}
643	}
644	crc_table[n] = c;
645	}
646	crc_table_computed = 1;
647	}
648
649	/*
650	Update a running crc with the bytes buf[0..len-1] and return
651	the updated crc. The crc should be initialized to zero. Pre- and
652	post-conditioning (one's complement) is performed within this
653	function so it shouldn't be done by the caller. Usage example:
654
655	unsigned long crc = 0L;
656
657	while (read_buffer(buffer, length) != EOF) {
658	crc = update_crc(crc, buffer, length);
659	}
660	if (crc != original_crc) error();
661	*/
662	unsigned long update_crc(unsigned long crc,
663	unsigned char *buf, int len)
664	{
665	unsigned long c = crc ^ 0xffffffffL;
666	int n;
667
668	if (!crc_table_computed)
669	make_crc_table();
670	for (n = 0; n < len; n++) {
671	c = crc_table[(c ^ buf[n]) & 0xff] ^ (c >> 8);
672	}
673	return c ^ 0xffffffffL;
674	}
675
676	/* Return the CRC of the bytes buf[0..len-1]. */
677	unsigned long crc(unsigned char *buf, int len)
678	{
679	return update_crc(0L, buf, len);
680	}
681
682
683
684
685	Deutsch Informational [Page 12]
686
687

	--- a/compat/zlib/doc/rfc1952.txt
	+++ b/compat/zlib/doc/rfc1952.txt
	@@ -1,687 +0,0 @@

D compat/zlib/doc/txtvsbin.txt

-103

		--- a/compat/zlib/doc/txtvsbin.txt
		+++ b/compat/zlib/doc/txtvsbin.txt
		@@ -1,107 +0,0 @@
1		-A Fast Method for Identifying Plain Text Files
2		-==============================================
3		-
4		-
5		-Introduction
		-------------
6		-
7		-Given a file coming from an unknown source, it is sometimes desirable
8		-to find out whether the format of that file is plain text. Although
9		-this may appear like a simple task, a fully accurate detection of the
10		-file type requires heavy-duty semantic analysis on the file contents.
11		-It is, however, possible to obtain satisfactory results by employing
12		-various heuristics.
13		-
14		-Previous versions of PKZip and other zip-compatible compression tools
15		-were using a crude detection scheme: if more than 80% (4/5) of the bytes
16		-found in a certain buffer are within the range [7..127], the file is
17		-labeled as plain text, otherwise it is labeled as binary. A prominent
18		-limitation of this scheme is the restriction to Latin-based alphabets.
19		-Other alphabets, like Greek, Cyrillic or Asian, make extensive use of
20		-the bytes within the range [128..255], and texts using these alphabets
21		-are most often misidentified by this scheme; in other words, the rate
22		-of false negatives is sometimes too high, which means that the recall
23		-is low. Another weakness of this scheme is a reduced precision, due to
24		-the false positives that may occur when binary files containing large
25		-amounts of textual characters are misidentified as plain text.
26		-
27		-In this article we propose a new, simple detection scheme that features
28		-a much increased precision and a near-100% recall. This scheme is
29		-designed to work on ASCII, Unicode and other ASCII-derived alphabets,
30		-and it handles single-byte encodings (ISO-8859, MacRoman, KOI8, etc.)
31		-and variable-sized encodings (ISO-2022, UTF-8, etc.). Wider encodings
32		-(UCS-2/UTF-16 and UCS-4/UTF-32) are not handled, however.
33		-
34		-
35		-The Algorithm
		--------------
36		-
37		-The algorithm works by dividing the set of bytecodes [0..255] into three
38		-categories:
39		-- The white list of textual bytecodes:
40		- 9 (TAB), 10 (LF), 13 (CR), 32 (SPACE) to 255.
41		-- The gray list of tolerated bytecodes:
42		- 7 (BEL), 8 (BS), 11 (VT), 12 (FF), 26 (SUB), 27 (ESC).
43		-- The black list of undesired, non-textual bytecodes:
44		- 0 (NUL) to 6, 14 to 31.
45		-
46		-If a file contains at least one byte that belongs to the white list and
47		-no byte that belongs to the black list, then the file is categorized as
48		-plain text; otherwise, it is categorized as binary. (The boundary case,
49		-when the file is empty, automatically falls into the latter category.)
50		-
51		-
52		-Rationale
		----------
53		-
54		-The idea behind this algorithm relies on two observations.
55		-
56		-The first observation is that, although the full range of 7-bit codes
57		-[0..127] is properly specified by the ASCII standard, most control
58		-characters in the range [0..31] are not used in practice. The only
59		-widely-used, almost universally-portable control codes are 9 (TAB),
60		-10 (LF) and 13 (CR). There are a few more control codes that are
61		-recognized on a reduced range of platforms and text viewers/editors:
62		-7 (BEL), 8 (BS), 11 (VT), 12 (FF), 26 (SUB) and 27 (ESC); but these
63		-codes are rarely (if ever) used alone, without being accompanied by
64		-some printable text. Even the newer, portable text formats such as
65		-XML avoid using control characters outside the list mentioned here.
66		-
67		-The second observation is that most of the binary files tend to contain
68		-control characters, especially 0 (NUL). Even though the older text
69		-detection schemes observe the presence of non-ASCII codes from the range
70		-[128..255], the precision rarely has to suffer if this upper range is
71		-labeled as textual, because the files that are genuinely binary tend to
72		-contain both control characters and codes from the upper range. On the
73		-other hand, the upper range needs to be labeled as textual, because it
74		-is used by virtually all ASCII extensions. In particular, this range is
75		-used for encoding non-Latin scripts.
76		-
77		-Since there is no counting involved, other than simply observing the
78		-presence or the absence of some byte values, the algorithm produces
79		-consistent results, regardless what alphabet encoding is being used.
80		-(If counting were involved, it could be possible to obtain different
81		-results on a text encoded, say, using ISO-8859-16 versus UTF-8.)
82		-
83		-There is an extra category of plain text files that are "polluted" with
84		-one or more black-listed codes, either by mistake or by peculiar design
85		-considerations. In such cases, a scheme that tolerates a small fraction
86		-of black-listed codes would provide an increased recall (i.e. more true
87		-positives). This, however, incurs a reduced precision overall, since
88		-false positives are more likely to appear in binary files that contain
89		-large chunks of textual data. Furthermore, "polluted" plain text should
90		-be regarded as binary by general-purpose text detection schemes, because
91		-general-purpose text processing algorithms might not be applicable.
92		-Under this premise, it is safe to say that our detection method provides
93		-a near-100% recall.
94		-
95		-Experiments have been run on many files coming from various platforms
96		-and applications. We tried plain text files, system logs, source code,
97		-formatted office documents, compiled object code, etc. The results
98		-confirm the optimistic assumptions about the capabilities of this
99		-algorithm.
100		-
101		-
		---
102		-Cosmin Truta
103		-Last updated: 2006-May-28

	--- a/compat/zlib/doc/txtvsbin.txt
	+++ b/compat/zlib/doc/txtvsbin.txt
	@@ -1,107 +0,0 @@
1	A Fast Method for Identifying Plain Text Files
2	==============================================
3
4
5	Introduction
	-------------
6
7	Given a file coming from an unknown source, it is sometimes desirable
8	to find out whether the format of that file is plain text. Although
9	this may appear like a simple task, a fully accurate detection of the
10	file type requires heavy-duty semantic analysis on the file contents.
11	It is, however, possible to obtain satisfactory results by employing
12	various heuristics.
13
14	Previous versions of PKZip and other zip-compatible compression tools
15	were using a crude detection scheme: if more than 80% (4/5) of the bytes
16	found in a certain buffer are within the range [7..127], the file is
17	labeled as plain text, otherwise it is labeled as binary. A prominent
18	limitation of this scheme is the restriction to Latin-based alphabets.
19	Other alphabets, like Greek, Cyrillic or Asian, make extensive use of
20	the bytes within the range [128..255], and texts using these alphabets
21	are most often misidentified by this scheme; in other words, the rate
22	of false negatives is sometimes too high, which means that the recall
23	is low. Another weakness of this scheme is a reduced precision, due to
24	the false positives that may occur when binary files containing large
25	amounts of textual characters are misidentified as plain text.
26
27	In this article we propose a new, simple detection scheme that features
28	a much increased precision and a near-100% recall. This scheme is
29	designed to work on ASCII, Unicode and other ASCII-derived alphabets,
30	and it handles single-byte encodings (ISO-8859, MacRoman, KOI8, etc.)
31	and variable-sized encodings (ISO-2022, UTF-8, etc.). Wider encodings
32	(UCS-2/UTF-16 and UCS-4/UTF-32) are not handled, however.
33
34
35	The Algorithm
	--------------
36
37	The algorithm works by dividing the set of bytecodes [0..255] into three
38	categories:
39	- The white list of textual bytecodes:
40	9 (TAB), 10 (LF), 13 (CR), 32 (SPACE) to 255.
41	- The gray list of tolerated bytecodes:
42	7 (BEL), 8 (BS), 11 (VT), 12 (FF), 26 (SUB), 27 (ESC).
43	- The black list of undesired, non-textual bytecodes:
44	0 (NUL) to 6, 14 to 31.
45
46	If a file contains at least one byte that belongs to the white list and
47	no byte that belongs to the black list, then the file is categorized as
48	plain text; otherwise, it is categorized as binary. (The boundary case,
49	when the file is empty, automatically falls into the latter category.)
50
51
52	Rationale
	----------
53
54	The idea behind this algorithm relies on two observations.
55
56	The first observation is that, although the full range of 7-bit codes
57	[0..127] is properly specified by the ASCII standard, most control
58	characters in the range [0..31] are not used in practice. The only
59	widely-used, almost universally-portable control codes are 9 (TAB),
60	10 (LF) and 13 (CR). There are a few more control codes that are
61	recognized on a reduced range of platforms and text viewers/editors:
62	7 (BEL), 8 (BS), 11 (VT), 12 (FF), 26 (SUB) and 27 (ESC); but these
63	codes are rarely (if ever) used alone, without being accompanied by
64	some printable text. Even the newer, portable text formats such as
65	XML avoid using control characters outside the list mentioned here.
66
67	The second observation is that most of the binary files tend to contain
68	control characters, especially 0 (NUL). Even though the older text
69	detection schemes observe the presence of non-ASCII codes from the range
70	[128..255], the precision rarely has to suffer if this upper range is
71	labeled as textual, because the files that are genuinely binary tend to
72	contain both control characters and codes from the upper range. On the
73	other hand, the upper range needs to be labeled as textual, because it
74	is used by virtually all ASCII extensions. In particular, this range is
75	used for encoding non-Latin scripts.
76
77	Since there is no counting involved, other than simply observing the
78	presence or the absence of some byte values, the algorithm produces
79	consistent results, regardless what alphabet encoding is being used.
80	(If counting were involved, it could be possible to obtain different
81	results on a text encoded, say, using ISO-8859-16 versus UTF-8.)
82
83	There is an extra category of plain text files that are "polluted" with
84	one or more black-listed codes, either by mistake or by peculiar design
85	considerations. In such cases, a scheme that tolerates a small fraction
86	of black-listed codes would provide an increased recall (i.e. more true
87	positives). This, however, incurs a reduced precision overall, since
88	false positives are more likely to appear in binary files that contain
89	large chunks of textual data. Furthermore, "polluted" plain text should
90	be regarded as binary by general-purpose text detection schemes, because
91	general-purpose text processing algorithms might not be applicable.
92	Under this premise, it is safe to say that our detection method provides
93	a near-100% recall.
94
95	Experiments have been run on many files coming from various platforms
96	and applications. We tried plain text files, system logs, source code,
97	formatted office documents, compiled object code, etc. The results
98	confirm the optimistic assumptions about the capabilities of this
99	algorithm.
100
101
	---
102	Cosmin Truta
103	Last updated: 2006-May-28

	--- a/compat/zlib/doc/txtvsbin.txt
	+++ b/compat/zlib/doc/txtvsbin.txt
	@@ -1,107 +0,0 @@





	-------------






























	--------------

















	----------

















































	---

M src/attach.c

+1 -1

		--- src/attach.c
		+++ src/attach.c
		@@ -2,11 +2,11 @@
2	2	** Copyright (c) 2010 D. Richard Hipp
3	3	**
4	4	** This program is free software; you can redistribute it and/or
5	5	** modify it under the terms of the Simplified BSD License (also
6	6	** known as the "2-Clause License" or "FreeBSD License".)
7		-
	7	+**
8	8	** This program is distributed in the hope that it will be useful,
9	9	** but without any warranty; without even the implied warranty of
10	10	** merchantability or fitness for a particular purpose.
11	11	**
12	12	** Author contact information:
13	13

	--- src/attach.c
	+++ src/attach.c
	@@ -2,11 +2,11 @@
2	** Copyright (c) 2010 D. Richard Hipp
3	**
4	** This program is free software; you can redistribute it and/or
5	** modify it under the terms of the Simplified BSD License (also
6	** known as the "2-Clause License" or "FreeBSD License".)
7
8	** This program is distributed in the hope that it will be useful,
9	** but without any warranty; without even the implied warranty of
10	** merchantability or fitness for a particular purpose.
11	**
12	** Author contact information:
13

	--- src/attach.c
	+++ src/attach.c
	@@ -2,11 +2,11 @@
2	** Copyright (c) 2010 D. Richard Hipp
3	**
4	** This program is free software; you can redistribute it and/or
5	** modify it under the terms of the Simplified BSD License (also
6	** known as the "2-Clause License" or "FreeBSD License".)
7	**
8	** This program is distributed in the hope that it will be useful,
9	** but without any warranty; without even the implied warranty of
10	** merchantability or fitness for a particular purpose.
11	**
12	** Author contact information:
13

M src/blob.c

+9 -5

		--- src/blob.c
		+++ src/blob.c
		@@ -2,11 +2,11 @@
2	2	** Copyright (c) 2006 D. Richard Hipp
3	3	**
4	4	** This program is free software; you can redistribute it and/or
5	5	** modify it under the terms of the Simplified BSD License (also
6	6	** known as the "2-Clause License" or "FreeBSD License".)
7		-
	7	+**
8	8	** This program is distributed in the hope that it will be useful,
9	9	** but without any warranty; without even the implied warranty of
10	10	** merchantability or fitness for a particular purpose.
11	11	**
12	12	** Author contact information:
		@@ -647,15 +647,19 @@
647	647	}
648	648	pFrom->iCursor = i;
649	649	}
650	650
651	651	/*
652		-** Return true if the blob contains a valid UUID_SIZE-digit base16 identifier.
	652	+** Return true if the blob contains a valid base16 identifier artifact hash.
	653	+**
	654	+** The value returned is actually one of HNAME_SHA1 OR HNAME_K256 if the
	655	+** hash is valid. Both of these are non-zero and therefore "true".
	656	+** If the hash is not valid, then HNAME_ERROR is returned, which is zero or
	657	+** false.
653	658	*/
654		-int blob_is_uuid(Blob *pBlob){
655		- return blob_size(pBlob)==UUID_SIZE
656		- && validate16(blob_buffer(pBlob), UUID_SIZE);
	659	+int blob_is_hname(Blob *pBlob){
	660	+ return hname_validate(blob_buffer(pBlob), blob_size(pBlob));
657	661	}
658	662
659	663	/*
660	664	** Return true if the blob contains a valid filename
661	665	*/
662	666

	--- src/blob.c
	+++ src/blob.c
	@@ -2,11 +2,11 @@
2	** Copyright (c) 2006 D. Richard Hipp
3	**
4	** This program is free software; you can redistribute it and/or
5	** modify it under the terms of the Simplified BSD License (also
6	** known as the "2-Clause License" or "FreeBSD License".)
7
8	** This program is distributed in the hope that it will be useful,
9	** but without any warranty; without even the implied warranty of
10	** merchantability or fitness for a particular purpose.
11	**
12	** Author contact information:
	@@ -647,15 +647,19 @@
647	}
648	pFrom->iCursor = i;
649	}
650
651	/*
652	** Return true if the blob contains a valid UUID_SIZE-digit base16 identifier.





653	*/
654	int blob_is_uuid(Blob *pBlob){
655	return blob_size(pBlob)==UUID_SIZE
656	&& validate16(blob_buffer(pBlob), UUID_SIZE);
657	}
658
659	/*
660	** Return true if the blob contains a valid filename
661	*/
662

	--- src/blob.c
	+++ src/blob.c
	@@ -2,11 +2,11 @@
2	** Copyright (c) 2006 D. Richard Hipp
3	**
4	** This program is free software; you can redistribute it and/or
5	** modify it under the terms of the Simplified BSD License (also
6	** known as the "2-Clause License" or "FreeBSD License".)
7	**
8	** This program is distributed in the hope that it will be useful,
9	** but without any warranty; without even the implied warranty of
10	** merchantability or fitness for a particular purpose.
11	**
12	** Author contact information:
	@@ -647,15 +647,19 @@
647	}
648	pFrom->iCursor = i;
649	}
650
651	/*
652	** Return true if the blob contains a valid base16 identifier artifact hash.
653	**
654	** The value returned is actually one of HNAME_SHA1 OR HNAME_K256 if the
655	** hash is valid. Both of these are non-zero and therefore "true".
656	** If the hash is not valid, then HNAME_ERROR is returned, which is zero or
657	** false.
658	*/
659	int blob_is_hname(Blob *pBlob){
660	return hname_validate(blob_buffer(pBlob), blob_size(pBlob));

661	}
662
663	/*
664	** Return true if the blob contains a valid filename
665	*/
666

M src/browse.c

+4 -4

		--- src/browse.c
		+++ src/browse.c
		@@ -315,11 +315,11 @@
315	315	FileTreeNode pSibling; / Next element in the same subdirectory */
316	316	FileTreeNode pChild; / List of child nodes */
317	317	FileTreeNode pLastChild; / Last child on the pChild list */
318	318	char zName; / Name of this entry. The "tail" */
319	319	char zFullName; / Full pathname of this entry */
320		- char zUuid; / SHA1 hash of this file. May be NULL. */
	320	+ char zUuid; / Artifact hash of this file. May be NULL. */
321	321	double mtime; /* Modification time for this entry */
322	322	unsigned nFullName; /* Length of zFullName */
323	323	unsigned iLevel; /* Levels of parent directories */
324	324	};
325	325
		@@ -345,11 +345,11 @@
345	345	** the tree to be constructed properly.
346	346	*/
347	347	static void tree_add_node(
348	348	FileTree pTree, / Tree into which nodes are added */
349	349	const char zPath, / The full pathname of file to add */
350		- const char zUuid, / UUID of the file. Might be NULL. */
	350	+ const char zUuid, / Hash of the file. Might be NULL. */
351	351	double mtime /* Modification time for this entry */
352	352	){
353	353	int i;
354	354	FileTreeNode pParent; / Parent (directory) of the next node to insert */
355	355
		@@ -368,20 +368,20 @@
368	368	FileTreeNode *pNew;
369	369	int iStart = i;
370	370	int nByte;
371	371	while( zPath[i] && zPath[i]!='/' ){ i++; }
372	372	nByte = sizeof(*pNew) + i + 1;
373		- if( zUuid!=0 && zPath[i]==0 ) nByte += UUID_SIZE+1;
	373	+ if( zUuid!=0 && zPath[i]==0 ) nByte += HNAME_MAX+1;
374	374	pNew = fossil_malloc( nByte );
375	375	memset(pNew, 0, sizeof(*pNew));
376	376	pNew->zFullName = (char*)&pNew[1];
377	377	memcpy(pNew->zFullName, zPath, i);
378	378	pNew->zFullName[i] = 0;
379	379	pNew->nFullName = i;
380	380	if( zUuid!=0 && zPath[i]==0 ){
381	381	pNew->zUuid = pNew->zFullName + i + 1;
382		- memcpy(pNew->zUuid, zUuid, UUID_SIZE+1);
	382	+ memcpy(pNew->zUuid, zUuid, strlen(zUuid)+1);
383	383	}
384	384	pNew->zName = pNew->zFullName + iStart;
385	385	if( pTree->pLast ){
386	386	pTree->pLast->pNext = pNew;
387	387	}else{
388	388

	--- src/browse.c
	+++ src/browse.c
	@@ -315,11 +315,11 @@
315	FileTreeNode pSibling; / Next element in the same subdirectory */
316	FileTreeNode pChild; / List of child nodes */
317	FileTreeNode pLastChild; / Last child on the pChild list */
318	char zName; / Name of this entry. The "tail" */
319	char zFullName; / Full pathname of this entry */
320	char zUuid; / SHA1 hash of this file. May be NULL. */
321	double mtime; /* Modification time for this entry */
322	unsigned nFullName; /* Length of zFullName */
323	unsigned iLevel; /* Levels of parent directories */
324	};
325
	@@ -345,11 +345,11 @@
345	** the tree to be constructed properly.
346	*/
347	static void tree_add_node(
348	FileTree pTree, / Tree into which nodes are added */
349	const char zPath, / The full pathname of file to add */
350	const char zUuid, / UUID of the file. Might be NULL. */
351	double mtime /* Modification time for this entry */
352	){
353	int i;
354	FileTreeNode pParent; / Parent (directory) of the next node to insert */
355
	@@ -368,20 +368,20 @@
368	FileTreeNode *pNew;
369	int iStart = i;
370	int nByte;
371	while( zPath[i] && zPath[i]!='/' ){ i++; }
372	nByte = sizeof(*pNew) + i + 1;
373	if( zUuid!=0 && zPath[i]==0 ) nByte += UUID_SIZE+1;
374	pNew = fossil_malloc( nByte );
375	memset(pNew, 0, sizeof(*pNew));
376	pNew->zFullName = (char*)&pNew[1];
377	memcpy(pNew->zFullName, zPath, i);
378	pNew->zFullName[i] = 0;
379	pNew->nFullName = i;
380	if( zUuid!=0 && zPath[i]==0 ){
381	pNew->zUuid = pNew->zFullName + i + 1;
382	memcpy(pNew->zUuid, zUuid, UUID_SIZE+1);
383	}
384	pNew->zName = pNew->zFullName + iStart;
385	if( pTree->pLast ){
386	pTree->pLast->pNext = pNew;
387	}else{
388

	--- src/browse.c
	+++ src/browse.c
	@@ -315,11 +315,11 @@
315	FileTreeNode pSibling; / Next element in the same subdirectory */
316	FileTreeNode pChild; / List of child nodes */
317	FileTreeNode pLastChild; / Last child on the pChild list */
318	char zName; / Name of this entry. The "tail" */
319	char zFullName; / Full pathname of this entry */
320	char zUuid; / Artifact hash of this file. May be NULL. */
321	double mtime; /* Modification time for this entry */
322	unsigned nFullName; /* Length of zFullName */
323	unsigned iLevel; /* Levels of parent directories */
324	};
325
	@@ -345,11 +345,11 @@
345	** the tree to be constructed properly.
346	*/
347	static void tree_add_node(
348	FileTree pTree, / Tree into which nodes are added */
349	const char zPath, / The full pathname of file to add */
350	const char zUuid, / Hash of the file. Might be NULL. */
351	double mtime /* Modification time for this entry */
352	){
353	int i;
354	FileTreeNode pParent; / Parent (directory) of the next node to insert */
355
	@@ -368,20 +368,20 @@
368	FileTreeNode *pNew;
369	int iStart = i;
370	int nByte;
371	while( zPath[i] && zPath[i]!='/' ){ i++; }
372	nByte = sizeof(*pNew) + i + 1;
373	if( zUuid!=0 && zPath[i]==0 ) nByte += HNAME_MAX+1;
374	pNew = fossil_malloc( nByte );
375	memset(pNew, 0, sizeof(*pNew));
376	pNew->zFullName = (char*)&pNew[1];
377	memcpy(pNew->zFullName, zPath, i);
378	pNew->zFullName[i] = 0;
379	pNew->nFullName = i;
380	if( zUuid!=0 && zPath[i]==0 ){
381	pNew->zUuid = pNew->zFullName + i + 1;
382	memcpy(pNew->zUuid, zUuid, strlen(zUuid)+1);
383	}
384	pNew->zName = pNew->zFullName + iStart;
385	if( pTree->pLast ){
386	pTree->pLast->pNext = pNew;
387	}else{
388

M src/bundle.c

+11 -16

		--- src/bundle.c
		+++ src/bundle.c
		@@ -35,11 +35,11 @@
35	35	@ bcname TEXT,
36	36	@ bcvalue ANY
37	37	@ );
38	38	@ CREATE TABLE IF NOT EXISTS "%w".bblob(
39	39	@ blobid INTEGER PRIMARY KEY, -- Blob ID
40		-@ uuid TEXT NOT NULL, -- SHA1 hash of expanded blob
	40	+@ uuid TEXT NOT NULL, -- hash of expanded blob
41	41	@ sz INT NOT NULL, -- Size of blob after expansion
42	42	@ delta ANY, -- Delta compression basis, or NULL
43	43	@ notes TEXT, -- Description of content
44	44	@ data BLOB -- compressed content
45	45	@ );
		@@ -439,19 +439,19 @@
439	439	" WHERE bix.delta=%d"
440	440	" AND bix.blobid=bblob.blobid;",
441	441	iSrc
442	442	);
443	443	while( db_step(&q)==SQLITE_ROW ){
444		- Blob h1, h2, c1, c2;
	444	+ Blob h1, c1, c2;
445	445	int rid;
446	446	blob_zero(&h1);
447	447	db_column_blob(&q, 0, &h1);
448	448	blob_zero(&c1);
449	449	db_column_blob(&q, 1, &c1);
450	450	blob_uncompress(&c1, &c1);
451	451	blob_zero(&c2);
452		- if( db_column_type(&q,2)==SQLITE_TEXT && db_column_bytes(&q,2)==40 ){
	452	+ if( db_column_type(&q,2)==SQLITE_TEXT && db_column_bytes(&q,2)>=HNAME_MIN ){
453	453	Blob basis;
454	454	rid = db_int(0,"SELECT rid FROM blob WHERE uuid=%Q",
455	455	db_column_text(&q,2));
456	456	content_get(rid, &basis);
457	457	blob_delta_apply(&basis, &c1, &c2);
		@@ -461,16 +461,13 @@
461	461	blob_delta_apply(pBasis, &c1, &c2);
462	462	blob_reset(&c1);
463	463	}else{
464	464	c2 = c1;
465	465	}
466		- sha1sum_blob(&c2, &h2);
467		- if( blob_compare(&h1, &h2)!=0 ){
468		- fossil_fatal("SHA1 hash mismatch - wanted %s, got %s",
469		- blob_str(&h1), blob_str(&h2));
	466	+ if( hname_verify_hash(&c2, blob_buffer(&h1), blob_size(&h1))==0 ){
	467	+ fossil_fatal("artifact hash error on %b", &h1);
470	468	}
471		- blob_reset(&h2);
472	469	rid = content_put_ex(&c2, blob_str(&h1), 0, 0, isPriv);
473	470	if( rid==0 ){
474	471	fossil_fatal("%s", g.zErrMsg);
475	472	}else{
476	473	if( !isPriv ) content_make_public(rid);
		@@ -491,11 +488,11 @@
491	488	static void bundle_extract_item(
492	489	int blobid, /* ID of the item to extract */
493	490	Blob pOut / Write the content into this blob */
494	491	){
495	492	Stmt q;
496		- Blob x, basis, h1, h2;
	493	+ Blob x, basis, h1;
497	494	static Bag busy;
498	495
499	496	db_prepare(&q, "SELECT uuid, delta, data FROM bblob"
500	497	" WHERE blobid=%d", blobid);
501	498	if( db_step(&q)!=SQLITE_ROW ){
		@@ -525,17 +522,14 @@
525	522	}else{
526	523	*pOut = x;
527	524	}
528	525	blob_zero(&h1);
529	526	db_column_blob(&q, 0, &h1);
530		- sha1sum_blob(pOut, &h2);
531		- if( blob_compare(&h1, &h2)!=0 ){
532		- fossil_fatal("SHA1 hash mismatch - wanted %s, got %s",
533		- blob_str(&h1), blob_str(&h2));
	527	+ if( hname_verify_hash(pOut, blob_buffer(&h1), blob_size(&h1))==0 ){
	528	+ fossil_fatal("incorrect hash for artifact %b", &h1);
534	529	}
535	530	blob_reset(&h1);
536		- blob_reset(&h2);
537	531	bag_remove(&busy, blobid);
538	532	db_finalize(&q);
539	533	}
540	534
541	535	/* fossil bundle cat BUNDLE UUID...
		@@ -597,12 +591,13 @@
597	591	** repo, then the delta encodings cannot be decoded and the bundle cannot
598	592	** be extracted. */
599	593	zMissingDeltas = db_text(0,
600	594	"SELECT group_concat(substr(delta,1,10),' ')"
601	595	" FROM bblob"
602		- " WHERE typeof(delta)='text' AND length(delta)=40"
603		- " AND NOT EXISTS(SELECT 1 FROM blob WHERE uuid=bblob.delta)");
	596	+ " WHERE typeof(delta)='text' AND length(delta)>=%d"
	597	+ " AND NOT EXISTS(SELECT 1 FROM blob WHERE uuid=bblob.delta)",
	598	+ HNAME_MIN);
604	599	if( zMissingDeltas && zMissingDeltas[0] ){
605	600	fossil_fatal("delta basis artifacts not found in repository: %s",
606	601	zMissingDeltas);
607	602	}
608	603
609	604

	--- src/bundle.c
	+++ src/bundle.c
	@@ -35,11 +35,11 @@
35	@ bcname TEXT,
36	@ bcvalue ANY
37	@ );
38	@ CREATE TABLE IF NOT EXISTS "%w".bblob(
39	@ blobid INTEGER PRIMARY KEY, -- Blob ID
40	@ uuid TEXT NOT NULL, -- SHA1 hash of expanded blob
41	@ sz INT NOT NULL, -- Size of blob after expansion
42	@ delta ANY, -- Delta compression basis, or NULL
43	@ notes TEXT, -- Description of content
44	@ data BLOB -- compressed content
45	@ );
	@@ -439,19 +439,19 @@
439	" WHERE bix.delta=%d"
440	" AND bix.blobid=bblob.blobid;",
441	iSrc
442	);
443	while( db_step(&q)==SQLITE_ROW ){
444	Blob h1, h2, c1, c2;
445	int rid;
446	blob_zero(&h1);
447	db_column_blob(&q, 0, &h1);
448	blob_zero(&c1);
449	db_column_blob(&q, 1, &c1);
450	blob_uncompress(&c1, &c1);
451	blob_zero(&c2);
452	if( db_column_type(&q,2)==SQLITE_TEXT && db_column_bytes(&q,2)==40 ){
453	Blob basis;
454	rid = db_int(0,"SELECT rid FROM blob WHERE uuid=%Q",
455	db_column_text(&q,2));
456	content_get(rid, &basis);
457	blob_delta_apply(&basis, &c1, &c2);
	@@ -461,16 +461,13 @@
461	blob_delta_apply(pBasis, &c1, &c2);
462	blob_reset(&c1);
463	}else{
464	c2 = c1;
465	}
466	sha1sum_blob(&c2, &h2);
467	if( blob_compare(&h1, &h2)!=0 ){
468	fossil_fatal("SHA1 hash mismatch - wanted %s, got %s",
469	blob_str(&h1), blob_str(&h2));
470	}
471	blob_reset(&h2);
472	rid = content_put_ex(&c2, blob_str(&h1), 0, 0, isPriv);
473	if( rid==0 ){
474	fossil_fatal("%s", g.zErrMsg);
475	}else{
476	if( !isPriv ) content_make_public(rid);
	@@ -491,11 +488,11 @@
491	static void bundle_extract_item(
492	int blobid, /* ID of the item to extract */
493	Blob pOut / Write the content into this blob */
494	){
495	Stmt q;
496	Blob x, basis, h1, h2;
497	static Bag busy;
498
499	db_prepare(&q, "SELECT uuid, delta, data FROM bblob"
500	" WHERE blobid=%d", blobid);
501	if( db_step(&q)!=SQLITE_ROW ){
	@@ -525,17 +522,14 @@
525	}else{
526	*pOut = x;
527	}
528	blob_zero(&h1);
529	db_column_blob(&q, 0, &h1);
530	sha1sum_blob(pOut, &h2);
531	if( blob_compare(&h1, &h2)!=0 ){
532	fossil_fatal("SHA1 hash mismatch - wanted %s, got %s",
533	blob_str(&h1), blob_str(&h2));
534	}
535	blob_reset(&h1);
536	blob_reset(&h2);
537	bag_remove(&busy, blobid);
538	db_finalize(&q);
539	}
540
541	/* fossil bundle cat BUNDLE UUID...
	@@ -597,12 +591,13 @@
597	** repo, then the delta encodings cannot be decoded and the bundle cannot
598	** be extracted. */
599	zMissingDeltas = db_text(0,
600	"SELECT group_concat(substr(delta,1,10),' ')"
601	" FROM bblob"
602	" WHERE typeof(delta)='text' AND length(delta)=40"
603	" AND NOT EXISTS(SELECT 1 FROM blob WHERE uuid=bblob.delta)");

604	if( zMissingDeltas && zMissingDeltas[0] ){
605	fossil_fatal("delta basis artifacts not found in repository: %s",
606	zMissingDeltas);
607	}
608
609

	--- src/bundle.c
	+++ src/bundle.c
	@@ -35,11 +35,11 @@
35	@ bcname TEXT,
36	@ bcvalue ANY
37	@ );
38	@ CREATE TABLE IF NOT EXISTS "%w".bblob(
39	@ blobid INTEGER PRIMARY KEY, -- Blob ID
40	@ uuid TEXT NOT NULL, -- hash of expanded blob
41	@ sz INT NOT NULL, -- Size of blob after expansion
42	@ delta ANY, -- Delta compression basis, or NULL
43	@ notes TEXT, -- Description of content
44	@ data BLOB -- compressed content
45	@ );
	@@ -439,19 +439,19 @@
439	" WHERE bix.delta=%d"
440	" AND bix.blobid=bblob.blobid;",
441	iSrc
442	);
443	while( db_step(&q)==SQLITE_ROW ){
444	Blob h1, c1, c2;
445	int rid;
446	blob_zero(&h1);
447	db_column_blob(&q, 0, &h1);
448	blob_zero(&c1);
449	db_column_blob(&q, 1, &c1);
450	blob_uncompress(&c1, &c1);
451	blob_zero(&c2);
452	if( db_column_type(&q,2)==SQLITE_TEXT && db_column_bytes(&q,2)>=HNAME_MIN ){
453	Blob basis;
454	rid = db_int(0,"SELECT rid FROM blob WHERE uuid=%Q",
455	db_column_text(&q,2));
456	content_get(rid, &basis);
457	blob_delta_apply(&basis, &c1, &c2);
	@@ -461,16 +461,13 @@
461	blob_delta_apply(pBasis, &c1, &c2);
462	blob_reset(&c1);
463	}else{
464	c2 = c1;
465	}
466	if( hname_verify_hash(&c2, blob_buffer(&h1), blob_size(&h1))==0 ){
467	fossil_fatal("artifact hash error on %b", &h1);


468	}

469	rid = content_put_ex(&c2, blob_str(&h1), 0, 0, isPriv);
470	if( rid==0 ){
471	fossil_fatal("%s", g.zErrMsg);
472	}else{
473	if( !isPriv ) content_make_public(rid);
	@@ -491,11 +488,11 @@
488	static void bundle_extract_item(
489	int blobid, /* ID of the item to extract */
490	Blob pOut / Write the content into this blob */
491	){
492	Stmt q;
493	Blob x, basis, h1;
494	static Bag busy;
495
496	db_prepare(&q, "SELECT uuid, delta, data FROM bblob"
497	" WHERE blobid=%d", blobid);
498	if( db_step(&q)!=SQLITE_ROW ){
	@@ -525,17 +522,14 @@
522	}else{
523	*pOut = x;
524	}
525	blob_zero(&h1);
526	db_column_blob(&q, 0, &h1);
527	if( hname_verify_hash(pOut, blob_buffer(&h1), blob_size(&h1))==0 ){
528	fossil_fatal("incorrect hash for artifact %b", &h1);


529	}
530	blob_reset(&h1);

531	bag_remove(&busy, blobid);
532	db_finalize(&q);
533	}
534
535	/* fossil bundle cat BUNDLE UUID...
	@@ -597,12 +591,13 @@
591	** repo, then the delta encodings cannot be decoded and the bundle cannot
592	** be extracted. */
593	zMissingDeltas = db_text(0,
594	"SELECT group_concat(substr(delta,1,10),' ')"
595	" FROM bblob"
596	" WHERE typeof(delta)='text' AND length(delta)>=%d"
597	" AND NOT EXISTS(SELECT 1 FROM blob WHERE uuid=bblob.delta)",
598	HNAME_MIN);
599	if( zMissingDeltas && zMissingDeltas[0] ){
600	fossil_fatal("delta basis artifacts not found in repository: %s",
601	zMissingDeltas);
602	}
603
604

M src/cgi.c

		--- src/cgi.c
		+++ src/cgi.c
		@@ -1444,10 +1444,12 @@
1444	1444	cgi_setenv("HTTP_IF_MODIFIED_SINCE", zVal);
1445	1445	}else if( fossil_strcmp(zFieldName,"referer:")==0 ){
1446	1446	cgi_setenv("HTTP_REFERER", zVal);
1447	1447	}else if( fossil_strcmp(zFieldName,"user-agent:")==0 ){
1448	1448	cgi_setenv("HTTP_USER_AGENT", zVal);
	1449	+ }else if( fossil_strcmp(zFieldName,"authorization:")==0 ){
	1450	+ cgi_setenv("HTTP_AUTHORIZATION", zVal);
1449	1451	}else if( fossil_strcmp(zFieldName,"x-forwarded-for:")==0 ){
1450	1452	const char *zIpAddr = cgi_accept_forwarded_for(zVal);
1451	1453	if( zIpAddr!=0 ){
1452	1454	g.zIpAddr = mprintf("%s", zIpAddr);
1453	1455	cgi_replace_parameter("REMOTE_ADDR", g.zIpAddr);
1454	1456

	--- src/cgi.c
	+++ src/cgi.c
	@@ -1444,10 +1444,12 @@
1444	cgi_setenv("HTTP_IF_MODIFIED_SINCE", zVal);
1445	}else if( fossil_strcmp(zFieldName,"referer:")==0 ){
1446	cgi_setenv("HTTP_REFERER", zVal);
1447	}else if( fossil_strcmp(zFieldName,"user-agent:")==0 ){
1448	cgi_setenv("HTTP_USER_AGENT", zVal);


1449	}else if( fossil_strcmp(zFieldName,"x-forwarded-for:")==0 ){
1450	const char *zIpAddr = cgi_accept_forwarded_for(zVal);
1451	if( zIpAddr!=0 ){
1452	g.zIpAddr = mprintf("%s", zIpAddr);
1453	cgi_replace_parameter("REMOTE_ADDR", g.zIpAddr);
1454

	--- src/cgi.c
	+++ src/cgi.c
	@@ -1444,10 +1444,12 @@
1444	cgi_setenv("HTTP_IF_MODIFIED_SINCE", zVal);
1445	}else if( fossil_strcmp(zFieldName,"referer:")==0 ){
1446	cgi_setenv("HTTP_REFERER", zVal);
1447	}else if( fossil_strcmp(zFieldName,"user-agent:")==0 ){
1448	cgi_setenv("HTTP_USER_AGENT", zVal);
1449	}else if( fossil_strcmp(zFieldName,"authorization:")==0 ){
1450	cgi_setenv("HTTP_AUTHORIZATION", zVal);
1451	}else if( fossil_strcmp(zFieldName,"x-forwarded-for:")==0 ){
1452	const char *zIpAddr = cgi_accept_forwarded_for(zVal);
1453	if( zIpAddr!=0 ){
1454	g.zIpAddr = mprintf("%s", zIpAddr);
1455	cgi_replace_parameter("REMOTE_ADDR", g.zIpAddr);
1456

M src/checkin.c

+12 -10

		--- src/checkin.c
		+++ src/checkin.c
		@@ -377,11 +377,11 @@
377	377	** As a special case, the --no-merge option does not inhibit this default.
378	378	** This default shows exactly the set of changes that would be checked
379	379	** in by the commit command.
380	380	**
381	381	** If no filter options are used, or if the --merge option is used, the
382		-** SHA1 hash of each merge contributor check-in version is displayed at
	382	+** artifact hash of each merge contributor check-in version is displayed at
383	383	** the end of the report. The --no-merge option is useful to display the
384	384	** default set of changed files without the merge contributors.
385	385	**
386	386	** If change type classification is enabled, each output line starts with
387	387	** a code describing the file's change type, e.g. EDITED or RENAMED. It
		@@ -412,11 +412,11 @@
412	412	**
413	413	** General options:
414	414	** --abs-paths Display absolute pathnames.
415	415	** --rel-paths Display pathnames relative to the current working
416	416	** directory.
417		-** --sha1sum Verify file status using SHA1 hashing rather than
	417	+** --hash Verify file status using hashing rather than
418	418	** relying on file mtimes.
419	419	** --case-sensitive <BOOL> Override case-sensitive setting.
420	420	** --dotfiles Include unmanaged files beginning with a dot.
421	421	** --ignore <CSG> Ignore unmanaged files matching CSG glob patterns.
422	422	** --no-dir-symlinks Disables support for directory symlinks.
		@@ -464,11 +464,12 @@
464	464	{"no-merge", C_MERGE }, {"no-classify", C_CLASSIFY },
465	465	};
466	466
467	467	Blob report = BLOB_INITIALIZER;
468	468	enum {CHANGES, STATUS} command = *g.argv[1]=='s' ? STATUS : CHANGES;
469		- int useSha1sum = find_option("sha1sum", 0, 0)!=0;
	469	+ /* --sha1sum is an undocumented alias for --hash for backwards compatiblity */
	470	+ int useHash = find_option("hash",0,0)!=0 \|\| find_option("sha1sum",0,0)!=0;
470	471	int showHdr = command==CHANGES && find_option("header", 0, 0);
471	472	int verboseFlag = command==CHANGES && find_option("verbose", "v", 0);
472	473	const char *zIgnoreFlag = find_option("ignore", 0, 1);
473	474	unsigned scanFlags = 0;
474	475	unsigned flags = 0;
		@@ -528,11 +529,11 @@
528	529
529	530	/* We should be done with options. */
530	531	verify_all_options();
531	532
532	533	/* Check for changed files. */
533		- vfile_check_signature(vid, useSha1sum ? CKSIG_SHA1 : 0);
	534	+ vfile_check_signature(vid, useHash ? CKSIG_HASH : 0);
534	535
535	536	/* Search for unmanaged files if requested. */
536	537	if( flags & C_EXTRA ){
537	538	Glob *pIgnore = glob_create(zIgnoreFlag);
538	539	locate_unmanaged_files(g.argc-2, g.argv+2, scanFlags, pIgnore);
		@@ -1985,12 +1986,12 @@
1985	1986	** The --private option creates a private check-in that is never synced.
1986	1987	** Children of private check-ins are automatically private.
1987	1988	**
1988	1989	** The --tag option applies the symbolic tag name to the check-in.
1989	1990	**
1990		-** The --sha1sum option detects edited files by computing each file's
1991		-** SHA1 hash rather than just checking for changes to its size or mtime.
	1991	+** The --hash option detects edited files by computing each file's
	1992	+** artifact hash rather than just checking for changes to its size or mtime.
1992	1993	**
1993	1994	** Options:
1994	1995	** --allow-conflict allow unresolved merge conflicts
1995	1996	** --allow-empty allow a commit with no changes
1996	1997	** --allow-fork allow the commit to fork
		@@ -2010,11 +2011,11 @@
2010	2011	** input and assumes an answer of 'No' for every
2011	2012	** question.
2012	2013	** --no-warnings omit all warnings about file contents
2013	2014	** --nosign do not attempt to sign this commit with gpg
2014	2015	** --private do not sync changes and their descendants
2015		-** --sha1sum verify file status using SHA1 hashing rather
	2016	+** --hash verify file status using hashing rather
2016	2017	** than relying on file mtimes
2017	2018	** --tag TAG-NAME assign given tag TAG-NAME to the check-in
2018	2019	** --date-override DATETIME DATE to use instead of 'now'
2019	2020	** --user-override USER USER to use instead of the current default
2020	2021	**
		@@ -2033,11 +2034,11 @@
2033	2034	int nvid; /* Blob-id of the new check-in */
2034	2035	Blob comment; /* Check-in comment */
2035	2036	const char zComment; / Check-in comment */
2036	2037	Stmt q; /* Various queries */
2037	2038	char zUuid; / UUID of the new check-in */
2038		- int useSha1sum = 0; /* True to verify file status using SHA1 hashing */
	2039	+ int useHash = 0; /* True to verify file status using hashing */
2039	2040	int noSign = 0; /* True to omit signing the manifest using GPG */
2040	2041	int isAMerge = 0; /* True if checking in a merge */
2041	2042	int noWarningFlag = 0; /* True if skipping all warnings */
2042	2043	int noPrompt = 0; /* True if skipping all prompts */
2043	2044	int forceFlag = 0; /* Undocumented: Disables all checks */
		@@ -2068,11 +2069,12 @@
2068	2069	Blob ans;
2069	2070	char cReply;
2070	2071
2071	2072	memset(&sCiInfo, 0, sizeof(sCiInfo));
2072	2073	url_proxy_options();
2073		- useSha1sum = find_option("sha1sum", 0, 0)!=0;
	2074	+ /* --sha1sum is an undocumented alias for --hash for backwards compatiblity */
	2075	+ useHash = find_option("hash",0,0)!=0 \|\| find_option("sha1sum",0,0)!=0;
2074	2076	noSign = find_option("nosign",0,0)!=0;
2075	2077	forceDelta = find_option("delta",0,0)!=0;
2076	2078	forceBaseline = find_option("baseline",0,0)!=0;
2077	2079	if( forceDelta && forceBaseline ){
2078	2080	fossil_fatal("cannot use --delta and --baseline together");
		@@ -2232,11 +2234,11 @@
2232	2234	*/
2233	2235	if( !db_exists("SELECT 1 FROM user WHERE login=%Q", g.zLogin) ){
2234	2236	fossil_fatal("no such user: %s", g.zLogin);
2235	2237	}
2236	2238
2237		- hasChanges = unsaved_changes(useSha1sum ? CKSIG_SHA1 : 0);
	2239	+ hasChanges = unsaved_changes(useHash ? CKSIG_HASH : 0);
2238	2240	db_begin_transaction();
2239	2241	db_record_repository_filename(0);
2240	2242	if( hasChanges==0 && !isAMerge && !allowEmpty && !forceFlag ){
2241	2243	fossil_fatal("nothing has changed; use --allow-empty to override");
2242	2244	}
2243	2245

	--- src/checkin.c
	+++ src/checkin.c
	@@ -377,11 +377,11 @@
377	** As a special case, the --no-merge option does not inhibit this default.
378	** This default shows exactly the set of changes that would be checked
379	** in by the commit command.
380	**
381	** If no filter options are used, or if the --merge option is used, the
382	** SHA1 hash of each merge contributor check-in version is displayed at
383	** the end of the report. The --no-merge option is useful to display the
384	** default set of changed files without the merge contributors.
385	**
386	** If change type classification is enabled, each output line starts with
387	** a code describing the file's change type, e.g. EDITED or RENAMED. It
	@@ -412,11 +412,11 @@
412	**
413	** General options:
414	** --abs-paths Display absolute pathnames.
415	** --rel-paths Display pathnames relative to the current working
416	** directory.
417	** --sha1sum Verify file status using SHA1 hashing rather than
418	** relying on file mtimes.
419	** --case-sensitive <BOOL> Override case-sensitive setting.
420	** --dotfiles Include unmanaged files beginning with a dot.
421	** --ignore <CSG> Ignore unmanaged files matching CSG glob patterns.
422	** --no-dir-symlinks Disables support for directory symlinks.
	@@ -464,11 +464,12 @@
464	{"no-merge", C_MERGE }, {"no-classify", C_CLASSIFY },
465	};
466
467	Blob report = BLOB_INITIALIZER;
468	enum {CHANGES, STATUS} command = *g.argv[1]=='s' ? STATUS : CHANGES;
469	int useSha1sum = find_option("sha1sum", 0, 0)!=0;

470	int showHdr = command==CHANGES && find_option("header", 0, 0);
471	int verboseFlag = command==CHANGES && find_option("verbose", "v", 0);
472	const char *zIgnoreFlag = find_option("ignore", 0, 1);
473	unsigned scanFlags = 0;
474	unsigned flags = 0;
	@@ -528,11 +529,11 @@
528
529	/* We should be done with options. */
530	verify_all_options();
531
532	/* Check for changed files. */
533	vfile_check_signature(vid, useSha1sum ? CKSIG_SHA1 : 0);
534
535	/* Search for unmanaged files if requested. */
536	if( flags & C_EXTRA ){
537	Glob *pIgnore = glob_create(zIgnoreFlag);
538	locate_unmanaged_files(g.argc-2, g.argv+2, scanFlags, pIgnore);
	@@ -1985,12 +1986,12 @@
1985	** The --private option creates a private check-in that is never synced.
1986	** Children of private check-ins are automatically private.
1987	**
1988	** The --tag option applies the symbolic tag name to the check-in.
1989	**
1990	** The --sha1sum option detects edited files by computing each file's
1991	** SHA1 hash rather than just checking for changes to its size or mtime.
1992	**
1993	** Options:
1994	** --allow-conflict allow unresolved merge conflicts
1995	** --allow-empty allow a commit with no changes
1996	** --allow-fork allow the commit to fork
	@@ -2010,11 +2011,11 @@
2010	** input and assumes an answer of 'No' for every
2011	** question.
2012	** --no-warnings omit all warnings about file contents
2013	** --nosign do not attempt to sign this commit with gpg
2014	** --private do not sync changes and their descendants
2015	** --sha1sum verify file status using SHA1 hashing rather
2016	** than relying on file mtimes
2017	** --tag TAG-NAME assign given tag TAG-NAME to the check-in
2018	** --date-override DATETIME DATE to use instead of 'now'
2019	** --user-override USER USER to use instead of the current default
2020	**
	@@ -2033,11 +2034,11 @@
2033	int nvid; /* Blob-id of the new check-in */
2034	Blob comment; /* Check-in comment */
2035	const char zComment; / Check-in comment */
2036	Stmt q; /* Various queries */
2037	char zUuid; / UUID of the new check-in */
2038	int useSha1sum = 0; /* True to verify file status using SHA1 hashing */
2039	int noSign = 0; /* True to omit signing the manifest using GPG */
2040	int isAMerge = 0; /* True if checking in a merge */
2041	int noWarningFlag = 0; /* True if skipping all warnings */
2042	int noPrompt = 0; /* True if skipping all prompts */
2043	int forceFlag = 0; /* Undocumented: Disables all checks */
	@@ -2068,11 +2069,12 @@
2068	Blob ans;
2069	char cReply;
2070
2071	memset(&sCiInfo, 0, sizeof(sCiInfo));
2072	url_proxy_options();
2073	useSha1sum = find_option("sha1sum", 0, 0)!=0;

2074	noSign = find_option("nosign",0,0)!=0;
2075	forceDelta = find_option("delta",0,0)!=0;
2076	forceBaseline = find_option("baseline",0,0)!=0;
2077	if( forceDelta && forceBaseline ){
2078	fossil_fatal("cannot use --delta and --baseline together");
	@@ -2232,11 +2234,11 @@
2232	*/
2233	if( !db_exists("SELECT 1 FROM user WHERE login=%Q", g.zLogin) ){
2234	fossil_fatal("no such user: %s", g.zLogin);
2235	}
2236
2237	hasChanges = unsaved_changes(useSha1sum ? CKSIG_SHA1 : 0);
2238	db_begin_transaction();
2239	db_record_repository_filename(0);
2240	if( hasChanges==0 && !isAMerge && !allowEmpty && !forceFlag ){
2241	fossil_fatal("nothing has changed; use --allow-empty to override");
2242	}
2243

	--- src/checkin.c
	+++ src/checkin.c
	@@ -377,11 +377,11 @@
377	** As a special case, the --no-merge option does not inhibit this default.
378	** This default shows exactly the set of changes that would be checked
379	** in by the commit command.
380	**
381	** If no filter options are used, or if the --merge option is used, the
382	** artifact hash of each merge contributor check-in version is displayed at
383	** the end of the report. The --no-merge option is useful to display the
384	** default set of changed files without the merge contributors.
385	**
386	** If change type classification is enabled, each output line starts with
387	** a code describing the file's change type, e.g. EDITED or RENAMED. It
	@@ -412,11 +412,11 @@
412	**
413	** General options:
414	** --abs-paths Display absolute pathnames.
415	** --rel-paths Display pathnames relative to the current working
416	** directory.
417	** --hash Verify file status using hashing rather than
418	** relying on file mtimes.
419	** --case-sensitive <BOOL> Override case-sensitive setting.
420	** --dotfiles Include unmanaged files beginning with a dot.
421	** --ignore <CSG> Ignore unmanaged files matching CSG glob patterns.
422	** --no-dir-symlinks Disables support for directory symlinks.
	@@ -464,11 +464,12 @@
464	{"no-merge", C_MERGE }, {"no-classify", C_CLASSIFY },
465	};
466
467	Blob report = BLOB_INITIALIZER;
468	enum {CHANGES, STATUS} command = *g.argv[1]=='s' ? STATUS : CHANGES;
469	/* --sha1sum is an undocumented alias for --hash for backwards compatiblity */
470	int useHash = find_option("hash",0,0)!=0 \|\| find_option("sha1sum",0,0)!=0;
471	int showHdr = command==CHANGES && find_option("header", 0, 0);
472	int verboseFlag = command==CHANGES && find_option("verbose", "v", 0);
473	const char *zIgnoreFlag = find_option("ignore", 0, 1);
474	unsigned scanFlags = 0;
475	unsigned flags = 0;
	@@ -528,11 +529,11 @@
529
530	/* We should be done with options. */
531	verify_all_options();
532
533	/* Check for changed files. */
534	vfile_check_signature(vid, useHash ? CKSIG_HASH : 0);
535
536	/* Search for unmanaged files if requested. */
537	if( flags & C_EXTRA ){
538	Glob *pIgnore = glob_create(zIgnoreFlag);
539	locate_unmanaged_files(g.argc-2, g.argv+2, scanFlags, pIgnore);
	@@ -1985,12 +1986,12 @@
1986	** The --private option creates a private check-in that is never synced.
1987	** Children of private check-ins are automatically private.
1988	**
1989	** The --tag option applies the symbolic tag name to the check-in.
1990	**
1991	** The --hash option detects edited files by computing each file's
1992	** artifact hash rather than just checking for changes to its size or mtime.
1993	**
1994	** Options:
1995	** --allow-conflict allow unresolved merge conflicts
1996	** --allow-empty allow a commit with no changes
1997	** --allow-fork allow the commit to fork
	@@ -2010,11 +2011,11 @@
2011	** input and assumes an answer of 'No' for every
2012	** question.
2013	** --no-warnings omit all warnings about file contents
2014	** --nosign do not attempt to sign this commit with gpg
2015	** --private do not sync changes and their descendants
2016	** --hash verify file status using hashing rather
2017	** than relying on file mtimes
2018	** --tag TAG-NAME assign given tag TAG-NAME to the check-in
2019	** --date-override DATETIME DATE to use instead of 'now'
2020	** --user-override USER USER to use instead of the current default
2021	**
	@@ -2033,11 +2034,11 @@
2034	int nvid; /* Blob-id of the new check-in */
2035	Blob comment; /* Check-in comment */
2036	const char zComment; / Check-in comment */
2037	Stmt q; /* Various queries */
2038	char zUuid; / UUID of the new check-in */
2039	int useHash = 0; /* True to verify file status using hashing */
2040	int noSign = 0; /* True to omit signing the manifest using GPG */
2041	int isAMerge = 0; /* True if checking in a merge */
2042	int noWarningFlag = 0; /* True if skipping all warnings */
2043	int noPrompt = 0; /* True if skipping all prompts */
2044	int forceFlag = 0; /* Undocumented: Disables all checks */
	@@ -2068,11 +2069,12 @@
2069	Blob ans;
2070	char cReply;
2071
2072	memset(&sCiInfo, 0, sizeof(sCiInfo));
2073	url_proxy_options();
2074	/* --sha1sum is an undocumented alias for --hash for backwards compatiblity */
2075	useHash = find_option("hash",0,0)!=0 \|\| find_option("sha1sum",0,0)!=0;
2076	noSign = find_option("nosign",0,0)!=0;
2077	forceDelta = find_option("delta",0,0)!=0;
2078	forceBaseline = find_option("baseline",0,0)!=0;
2079	if( forceDelta && forceBaseline ){
2080	fossil_fatal("cannot use --delta and --baseline together");
	@@ -2232,11 +2234,11 @@
2234	*/
2235	if( !db_exists("SELECT 1 FROM user WHERE login=%Q", g.zLogin) ){
2236	fossil_fatal("no such user: %s", g.zLogin);
2237	}
2238
2239	hasChanges = unsaved_changes(useHash ? CKSIG_HASH : 0);
2240	db_begin_transaction();
2241	db_record_repository_filename(0);
2242	if( hasChanges==0 && !isAMerge && !allowEmpty && !forceFlag ){
2243	fossil_fatal("nothing has changed; use --allow-empty to override");
2244	}
2245

M src/checkout.c

+3 -6

		--- src/checkout.c
		+++ src/checkout.c
		@@ -134,24 +134,19 @@
134	134	** "manifest.tags".
135	135	*/
136	136	void manifest_to_disk(int vid){
137	137	char *zManFile;
138	138	Blob manifest;
139		- Blob hash;
140	139	Blob taglist;
141	140	int flg;
142	141
143	142	flg = db_get_manifest_setting();
144	143
145		- if( flg & (MFESTFLG_RAW\|MFESTFLG_UUID) ){
	144	+ if( flg & MFESTFLG_RAW ){
146	145	blob_zero(&manifest);
147	146	content_get(vid, &manifest);
148		- blob_zero(&hash);
149		- sha1sum_blob(&manifest, &hash);
150	147	sterilize_manifest(&manifest);
151		- }
152		- if( flg & MFESTFLG_RAW ){
153	148	zManFile = mprintf("%smanifest", g.zLocalRoot);
154	149	blob_write_to_file(&manifest, zManFile);
155	150	free(zManFile);
156	151	}else{
157	152	if( !db_exists("SELECT 1 FROM vfile WHERE pathname='manifest'") ){
		@@ -159,11 +154,13 @@
159	154	file_delete(zManFile);
160	155	free(zManFile);
161	156	}
162	157	}
163	158	if( flg & MFESTFLG_UUID ){
	159	+ Blob hash;
164	160	zManFile = mprintf("%smanifest.uuid", g.zLocalRoot);
	161	+ blob_set_dynamic(&hash, rid_to_uuid(vid));
165	162	blob_append(&hash, "\n", 1);
166	163	blob_write_to_file(&hash, zManFile);
167	164	free(zManFile);
168	165	blob_reset(&hash);
169	166	}else{
170	167

	--- src/checkout.c
	+++ src/checkout.c
	@@ -134,24 +134,19 @@
134	** "manifest.tags".
135	*/
136	void manifest_to_disk(int vid){
137	char *zManFile;
138	Blob manifest;
139	Blob hash;
140	Blob taglist;
141	int flg;
142
143	flg = db_get_manifest_setting();
144
145	if( flg & (MFESTFLG_RAW\|MFESTFLG_UUID) ){
146	blob_zero(&manifest);
147	content_get(vid, &manifest);
148	blob_zero(&hash);
149	sha1sum_blob(&manifest, &hash);
150	sterilize_manifest(&manifest);
151	}
152	if( flg & MFESTFLG_RAW ){
153	zManFile = mprintf("%smanifest", g.zLocalRoot);
154	blob_write_to_file(&manifest, zManFile);
155	free(zManFile);
156	}else{
157	if( !db_exists("SELECT 1 FROM vfile WHERE pathname='manifest'") ){
	@@ -159,11 +154,13 @@
159	file_delete(zManFile);
160	free(zManFile);
161	}
162	}
163	if( flg & MFESTFLG_UUID ){

164	zManFile = mprintf("%smanifest.uuid", g.zLocalRoot);

165	blob_append(&hash, "\n", 1);
166	blob_write_to_file(&hash, zManFile);
167	free(zManFile);
168	blob_reset(&hash);
169	}else{
170

	--- src/checkout.c
	+++ src/checkout.c
	@@ -134,24 +134,19 @@
134	** "manifest.tags".
135	*/
136	void manifest_to_disk(int vid){
137	char *zManFile;
138	Blob manifest;

139	Blob taglist;
140	int flg;
141
142	flg = db_get_manifest_setting();
143
144	if( flg & MFESTFLG_RAW ){
145	blob_zero(&manifest);
146	content_get(vid, &manifest);


147	sterilize_manifest(&manifest);


148	zManFile = mprintf("%smanifest", g.zLocalRoot);
149	blob_write_to_file(&manifest, zManFile);
150	free(zManFile);
151	}else{
152	if( !db_exists("SELECT 1 FROM vfile WHERE pathname='manifest'") ){
	@@ -159,11 +154,13 @@
154	file_delete(zManFile);
155	free(zManFile);
156	}
157	}
158	if( flg & MFESTFLG_UUID ){
159	Blob hash;
160	zManFile = mprintf("%smanifest.uuid", g.zLocalRoot);
161	blob_set_dynamic(&hash, rid_to_uuid(vid));
162	blob_append(&hash, "\n", 1);
163	blob_write_to_file(&hash, zManFile);
164	free(zManFile);
165	blob_reset(&hash);
166	}else{
167

M src/clone.c

		--- src/clone.c
		+++ src/clone.c
		@@ -175,10 +175,11 @@
175	175	db_initial_setup(0, 0, zDefaultUser);
176	176	user_select();
177	177	db_set("content-schema", CONTENT_SCHEMA, 0);
178	178	db_set("aux-schema", AUX_SCHEMA_MAX, 0);
179	179	db_set("rebuilt", get_version(), 0);
	180	+ db_unset("hash-policy", 0);
180	181	remember_or_get_http_auth(zHttpAuth, urlFlags & URL_REMEMBER, g.argv[2]);
181	182	url_remember();
182	183	if( g.zSSLIdentity!=0 ){
183	184	/* If the --ssl-identity option was specified, store it as a setting */
184	185	Blob fn;
185	186

	--- src/clone.c
	+++ src/clone.c
	@@ -175,10 +175,11 @@
175	db_initial_setup(0, 0, zDefaultUser);
176	user_select();
177	db_set("content-schema", CONTENT_SCHEMA, 0);
178	db_set("aux-schema", AUX_SCHEMA_MAX, 0);
179	db_set("rebuilt", get_version(), 0);

180	remember_or_get_http_auth(zHttpAuth, urlFlags & URL_REMEMBER, g.argv[2]);
181	url_remember();
182	if( g.zSSLIdentity!=0 ){
183	/* If the --ssl-identity option was specified, store it as a setting */
184	Blob fn;
185

	--- src/clone.c
	+++ src/clone.c
	@@ -175,10 +175,11 @@
175	db_initial_setup(0, 0, zDefaultUser);
176	user_select();
177	db_set("content-schema", CONTENT_SCHEMA, 0);
178	db_set("aux-schema", AUX_SCHEMA_MAX, 0);
179	db_set("rebuilt", get_version(), 0);
180	db_unset("hash-policy", 0);
181	remember_or_get_http_auth(zHttpAuth, urlFlags & URL_REMEMBER, g.argv[2]);
182	url_remember();
183	if( g.zSSLIdentity!=0 ){
184	/* If the --ssl-identity option was specified, store it as a setting */
185	Blob fn;
186

M src/config.h

		--- src/config.h
		+++ src/config.h
		@@ -38,10 +38,15 @@
38	38	#endif
39	39
40	40	#ifdef HAVE_AUTOCONFIG_H
41	41	#include "autoconfig.h"
42	42	#endif
	43	+
	44	+/* Enable the hardened SHA1 implemenation by default */
	45	+#ifndef FOSSIL_HARDENED_SHA1
	46	+# define FOSSIL_HARDENED_SHA1 1
	47	+#endif
43	48
44	49	#ifndef _RC_COMPILE_
45	50
46	51	/*
47	52	** System header files used by all modules
		@@ -168,10 +173,11 @@
168	173	# endif
169	174	#endif
170	175
171	176	#if !defined(_RC_COMPILE_) && !defined(SQLITE_AMALGAMATION)
172	177
	178	+#include <stdint.h>
173	179	#include "sqlite3.h"
174	180
175	181	/*
176	182	** On Solaris, getpass() will only return up to 8 characters. getpassphrase() returns up to 257.
177	183	*/
178	184

	--- src/config.h
	+++ src/config.h
	@@ -38,10 +38,15 @@
38	#endif
39
40	#ifdef HAVE_AUTOCONFIG_H
41	#include "autoconfig.h"
42	#endif





43
44	#ifndef _RC_COMPILE_
45
46	/*
47	** System header files used by all modules
	@@ -168,10 +173,11 @@
168	# endif
169	#endif
170
171	#if !defined(_RC_COMPILE_) && !defined(SQLITE_AMALGAMATION)
172

173	#include "sqlite3.h"
174
175	/*
176	** On Solaris, getpass() will only return up to 8 characters. getpassphrase() returns up to 257.
177	*/
178

	--- src/config.h
	+++ src/config.h
	@@ -38,10 +38,15 @@
38	#endif
39
40	#ifdef HAVE_AUTOCONFIG_H
41	#include "autoconfig.h"
42	#endif
43
44	/* Enable the hardened SHA1 implemenation by default */
45	#ifndef FOSSIL_HARDENED_SHA1
46	# define FOSSIL_HARDENED_SHA1 1
47	#endif
48
49	#ifndef _RC_COMPILE_
50
51	/*
52	** System header files used by all modules
	@@ -168,10 +173,11 @@
173	# endif
174	#endif
175
176	#if !defined(_RC_COMPILE_) && !defined(SQLITE_AMALGAMATION)
177
178	#include <stdint.h>
179	#include "sqlite3.h"
180
181	/*
182	** On Solaris, getpass() will only return up to 8 characters. getpassphrase() returns up to 257.
183	*/
184

M src/configure.c

		--- src/configure.c
		+++ src/configure.c
		@@ -129,10 +129,11 @@
129	129	{ "empty-dirs", CONFIGSET_PROJ },
130	130	{ "allow-symlinks", CONFIGSET_PROJ },
131	131	{ "dotfiles", CONFIGSET_PROJ },
132	132	{ "parent-project-code", CONFIGSET_PROJ },
133	133	{ "parent-project-name", CONFIGSET_PROJ },
	134	+ { "hash-policy", CONFIGSET_PROJ },
134	135
135	136	#ifdef FOSSIL_ENABLE_LEGACY_MV_RM
136	137	{ "mv-rm-files", CONFIGSET_PROJ },
137	138	#endif
138	139
139	140

	--- src/configure.c
	+++ src/configure.c
	@@ -129,10 +129,11 @@
129	{ "empty-dirs", CONFIGSET_PROJ },
130	{ "allow-symlinks", CONFIGSET_PROJ },
131	{ "dotfiles", CONFIGSET_PROJ },
132	{ "parent-project-code", CONFIGSET_PROJ },
133	{ "parent-project-name", CONFIGSET_PROJ },

134
135	#ifdef FOSSIL_ENABLE_LEGACY_MV_RM
136	{ "mv-rm-files", CONFIGSET_PROJ },
137	#endif
138
139

	--- src/configure.c
	+++ src/configure.c
	@@ -129,10 +129,11 @@
129	{ "empty-dirs", CONFIGSET_PROJ },
130	{ "allow-symlinks", CONFIGSET_PROJ },
131	{ "dotfiles", CONFIGSET_PROJ },
132	{ "parent-project-code", CONFIGSET_PROJ },
133	{ "parent-project-name", CONFIGSET_PROJ },
134	{ "hash-policy", CONFIGSET_PROJ },
135
136	#ifdef FOSSIL_ENABLE_LEGACY_MV_RM
137	{ "mv-rm-files", CONFIGSET_PROJ },
138	#endif
139
140

M src/content.c

+22 -12

		--- src/content.c
		+++ src/content.c
		@@ -312,11 +312,11 @@
312	312	/*
313	313	** COMMAND: artifact*
314	314	**
315	315	** Usage: %fossil artifact ARTIFACT-ID ?OUTPUT-FILENAME? ?OPTIONS?
316	316	**
317		-** Extract an artifact by its SHA1 hash and write the results on
	317	+** Extract an artifact by its artifact hash and write the results on
318	318	** standard output, or if the optional 4th argument is given, in
319	319	** the named output file.
320	320	**
321	321	** Options:
322	322	** -R\|--repository FILE Extract artifacts from repository FILE
		@@ -497,11 +497,11 @@
497	497	** to be responsible for pBlob. This routine does not take over
498	498	** responsibility for freeing pBlob.
499	499	*/
500	500	int content_put_ex(
501	501	Blob pBlob, / Content to add to the repository */
502		- const char zUuid, / SHA1 hash of reconstructed pBlob */
	502	+ const char zUuid, / artifact hash of reconstructed pBlob */
503	503	int srcId, /* pBlob is a delta from this entry */
504	504	int nBlob, /* pBlob is compressed. Original size is this */
505	505	int isPrivate /* The content should be marked private */
506	506	){
507	507	int size;
		@@ -513,25 +513,38 @@
513	513	int isDephantomize = 0;
514	514
515	515	assert( g.repositoryOpen );
516	516	assert( pBlob!=0 );
517	517	assert( srcId==0 \|\| zUuid!=0 );
	518	+ db_begin_transaction();
518	519	if( zUuid==0 ){
519	520	assert( nBlob==0 );
520		- sha1sum_blob(pBlob, &hash);
	521	+ /* First check the auxiliary hash to see if there is already an artifact
	522	+ ** that uses the auxiliary hash name */
	523	+ hname_hash(pBlob, 1, &hash);
	524	+ rid = fast_uuid_to_rid(blob_str(&hash));
	525	+ if( rid==0 ){
	526	+ /* No existing artifact with the auxiliary hash name. Therefore, use
	527	+ ** the primary hash name. */
	528	+ blob_reset(&hash);
	529	+ hname_hash(pBlob, 0, &hash);
	530	+ }
521	531	}else{
522	532	blob_init(&hash, zUuid, -1);
	533	+ }
	534	+ if( g.eHashPolicy==HPOLICY_AUTO && blob_size(&hash)>HNAME_LEN_SHA1 ){
	535	+ g.eHashPolicy = HPOLICY_SHA3;
	536	+ db_set_int("hash-policy", HPOLICY_SHA3, 0);
523	537	}
524	538	if( nBlob ){
525	539	size = nBlob;
526	540	}else{
527	541	size = blob_size(pBlob);
528	542	if( srcId ){
529	543	size = delta_output_size(blob_buffer(pBlob), size);
530	544	}
531	545	}
532		- db_begin_transaction();
533	546
534	547	/* Check to see if the entry already exists and if it does whether
535	548	** or not the entry is a phantom
536	549	*/
537	550	db_prepare(&s1, "SELECT rid, size FROM blob WHERE uuid=%B", &hash);
		@@ -868,11 +881,10 @@
868	881	** so forth, reporting any errors found.
869	882	*/
870	883	void test_integrity(void){
871	884	Stmt q;
872	885	Blob content;
873		- Blob cksum;
874	886	int n1 = 0;
875	887	int n2 = 0;
876	888	int nErr = 0;
877	889	int total;
878	890	int nCA = 0;
		@@ -905,10 +917,11 @@
905	917	db_prepare(&q, "SELECT rid, uuid, size FROM blob ORDER BY rid");
906	918	total = db_int(0, "SELECT max(rid) FROM blob");
907	919	while( db_step(&q)==SQLITE_ROW ){
908	920	int rid = db_column_int(&q, 0);
909	921	const char *zUuid = db_column_text(&q, 1);
	922	+ int nUuid = db_column_bytes(&q, 1);
910	923	int size = db_column_int(&q, 2);
911	924	n1++;
912	925	fossil_print(" %d/%d\r", n1, total);
913	926	fflush(stdout);
914	927	if( size<0 ){
		@@ -919,14 +932,12 @@
919	932	if( blob_size(&content)!=size ){
920	933	fossil_print("size mismatch on artifact %d: wanted %d but got %d\n",
921	934	rid, size, blob_size(&content));
922	935	nErr++;
923	936	}
924		- sha1sum_blob(&content, &cksum);
925		- if( fossil_strcmp(blob_str(&cksum), zUuid)!=0 ){
926		- fossil_print("wrong hash on artifact %d: wanted %s but got %s\n",
927		- rid, zUuid, blob_str(&cksum));
	937	+ if( !hname_verify_hash(&content, zUuid, nUuid) ){
	938	+ fossil_print("wrong hash on artifact %d\n",rid);
928	939	nErr++;
929	940	}
930	941	if( bParse && looks_like_control_artifact(&content) ){
931	942	Blob err;
932	943	int i, n;
		@@ -941,11 +952,11 @@
941	952	memcpy(zFirstLine, z, i);
942	953	zFirstLine[i] = 0;
943	954	p = manifest_parse(&content, 0, &err);
944	955	if( p==0 ){
945	956	fossil_print("manifest_parse failed for %s:\n%s\n",
946		- blob_str(&cksum), blob_str(&err));
	957	+ zUuid, blob_str(&err));
947	958	if( strncmp(blob_str(&err), "line 1:", 7)==0 ){
948	959	fossil_print("\"%s\"\n", zFirstLine);
949	960	}
950	961	}else{
951	962	anCA[p->type]++;
		@@ -954,11 +965,10 @@
954	965	}
955	966	blob_reset(&err);
956	967	}else{
957	968	blob_reset(&content);
958	969	}
959		- blob_reset(&cksum);
960	970	n2++;
961	971	}
962	972	db_finalize(&q);
963	973	fossil_print("%d non-phantom blobs (out of %d total) checked: %d errors\n",
964	974	n2, n1, nErr);
		@@ -1157,11 +1167,11 @@
1157	1167	**
1158	1168	** WARNING: You must run "fossil rebuild" after this command to rebuild
1159	1169	** the metadata.
1160	1170	**
1161	1171	** Note that the arguments are the integer raw RID values from the BLOB table,
1162		-** not SHA1 hashs or labels.
	1172	+** not artifact hashs or labels.
1163	1173	*/
1164	1174	void test_content_erase(void){
1165	1175	int i;
1166	1176	Blob x;
1167	1177	char c;
1168	1178

	--- src/content.c
	+++ src/content.c
	@@ -312,11 +312,11 @@
312	/*
313	** COMMAND: artifact*
314	**
315	** Usage: %fossil artifact ARTIFACT-ID ?OUTPUT-FILENAME? ?OPTIONS?
316	**
317	** Extract an artifact by its SHA1 hash and write the results on
318	** standard output, or if the optional 4th argument is given, in
319	** the named output file.
320	**
321	** Options:
322	** -R\|--repository FILE Extract artifacts from repository FILE
	@@ -497,11 +497,11 @@
497	** to be responsible for pBlob. This routine does not take over
498	** responsibility for freeing pBlob.
499	*/
500	int content_put_ex(
501	Blob pBlob, / Content to add to the repository */
502	const char zUuid, / SHA1 hash of reconstructed pBlob */
503	int srcId, /* pBlob is a delta from this entry */
504	int nBlob, /* pBlob is compressed. Original size is this */
505	int isPrivate /* The content should be marked private */
506	){
507	int size;
	@@ -513,25 +513,38 @@
513	int isDephantomize = 0;
514
515	assert( g.repositoryOpen );
516	assert( pBlob!=0 );
517	assert( srcId==0 \|\| zUuid!=0 );

518	if( zUuid==0 ){
519	assert( nBlob==0 );
520	sha1sum_blob(pBlob, &hash);









521	}else{
522	blob_init(&hash, zUuid, -1);




523	}
524	if( nBlob ){
525	size = nBlob;
526	}else{
527	size = blob_size(pBlob);
528	if( srcId ){
529	size = delta_output_size(blob_buffer(pBlob), size);
530	}
531	}
532	db_begin_transaction();
533
534	/* Check to see if the entry already exists and if it does whether
535	** or not the entry is a phantom
536	*/
537	db_prepare(&s1, "SELECT rid, size FROM blob WHERE uuid=%B", &hash);
	@@ -868,11 +881,10 @@
868	** so forth, reporting any errors found.
869	*/
870	void test_integrity(void){
871	Stmt q;
872	Blob content;
873	Blob cksum;
874	int n1 = 0;
875	int n2 = 0;
876	int nErr = 0;
877	int total;
878	int nCA = 0;
	@@ -905,10 +917,11 @@
905	db_prepare(&q, "SELECT rid, uuid, size FROM blob ORDER BY rid");
906	total = db_int(0, "SELECT max(rid) FROM blob");
907	while( db_step(&q)==SQLITE_ROW ){
908	int rid = db_column_int(&q, 0);
909	const char *zUuid = db_column_text(&q, 1);

910	int size = db_column_int(&q, 2);
911	n1++;
912	fossil_print(" %d/%d\r", n1, total);
913	fflush(stdout);
914	if( size<0 ){
	@@ -919,14 +932,12 @@
919	if( blob_size(&content)!=size ){
920	fossil_print("size mismatch on artifact %d: wanted %d but got %d\n",
921	rid, size, blob_size(&content));
922	nErr++;
923	}
924	sha1sum_blob(&content, &cksum);
925	if( fossil_strcmp(blob_str(&cksum), zUuid)!=0 ){
926	fossil_print("wrong hash on artifact %d: wanted %s but got %s\n",
927	rid, zUuid, blob_str(&cksum));
928	nErr++;
929	}
930	if( bParse && looks_like_control_artifact(&content) ){
931	Blob err;
932	int i, n;
	@@ -941,11 +952,11 @@
941	memcpy(zFirstLine, z, i);
942	zFirstLine[i] = 0;
943	p = manifest_parse(&content, 0, &err);
944	if( p==0 ){
945	fossil_print("manifest_parse failed for %s:\n%s\n",
946	blob_str(&cksum), blob_str(&err));
947	if( strncmp(blob_str(&err), "line 1:", 7)==0 ){
948	fossil_print("\"%s\"\n", zFirstLine);
949	}
950	}else{
951	anCA[p->type]++;
	@@ -954,11 +965,10 @@
954	}
955	blob_reset(&err);
956	}else{
957	blob_reset(&content);
958	}
959	blob_reset(&cksum);
960	n2++;
961	}
962	db_finalize(&q);
963	fossil_print("%d non-phantom blobs (out of %d total) checked: %d errors\n",
964	n2, n1, nErr);
	@@ -1157,11 +1167,11 @@
1157	**
1158	** WARNING: You must run "fossil rebuild" after this command to rebuild
1159	** the metadata.
1160	**
1161	** Note that the arguments are the integer raw RID values from the BLOB table,
1162	** not SHA1 hashs or labels.
1163	*/
1164	void test_content_erase(void){
1165	int i;
1166	Blob x;
1167	char c;
1168

	--- src/content.c
	+++ src/content.c
	@@ -312,11 +312,11 @@
312	/*
313	** COMMAND: artifact*
314	**
315	** Usage: %fossil artifact ARTIFACT-ID ?OUTPUT-FILENAME? ?OPTIONS?
316	**
317	** Extract an artifact by its artifact hash and write the results on
318	** standard output, or if the optional 4th argument is given, in
319	** the named output file.
320	**
321	** Options:
322	** -R\|--repository FILE Extract artifacts from repository FILE
	@@ -497,11 +497,11 @@
497	** to be responsible for pBlob. This routine does not take over
498	** responsibility for freeing pBlob.
499	*/
500	int content_put_ex(
501	Blob pBlob, / Content to add to the repository */
502	const char zUuid, / artifact hash of reconstructed pBlob */
503	int srcId, /* pBlob is a delta from this entry */
504	int nBlob, /* pBlob is compressed. Original size is this */
505	int isPrivate /* The content should be marked private */
506	){
507	int size;
	@@ -513,25 +513,38 @@
513	int isDephantomize = 0;
514
515	assert( g.repositoryOpen );
516	assert( pBlob!=0 );
517	assert( srcId==0 \|\| zUuid!=0 );
518	db_begin_transaction();
519	if( zUuid==0 ){
520	assert( nBlob==0 );
521	/* First check the auxiliary hash to see if there is already an artifact
522	** that uses the auxiliary hash name */
523	hname_hash(pBlob, 1, &hash);
524	rid = fast_uuid_to_rid(blob_str(&hash));
525	if( rid==0 ){
526	/* No existing artifact with the auxiliary hash name. Therefore, use
527	** the primary hash name. */
528	blob_reset(&hash);
529	hname_hash(pBlob, 0, &hash);
530	}
531	}else{
532	blob_init(&hash, zUuid, -1);
533	}
534	if( g.eHashPolicy==HPOLICY_AUTO && blob_size(&hash)>HNAME_LEN_SHA1 ){
535	g.eHashPolicy = HPOLICY_SHA3;
536	db_set_int("hash-policy", HPOLICY_SHA3, 0);
537	}
538	if( nBlob ){
539	size = nBlob;
540	}else{
541	size = blob_size(pBlob);
542	if( srcId ){
543	size = delta_output_size(blob_buffer(pBlob), size);
544	}
545	}

546
547	/* Check to see if the entry already exists and if it does whether
548	** or not the entry is a phantom
549	*/
550	db_prepare(&s1, "SELECT rid, size FROM blob WHERE uuid=%B", &hash);
	@@ -868,11 +881,10 @@
881	** so forth, reporting any errors found.
882	*/
883	void test_integrity(void){
884	Stmt q;
885	Blob content;

886	int n1 = 0;
887	int n2 = 0;
888	int nErr = 0;
889	int total;
890	int nCA = 0;
	@@ -905,10 +917,11 @@
917	db_prepare(&q, "SELECT rid, uuid, size FROM blob ORDER BY rid");
918	total = db_int(0, "SELECT max(rid) FROM blob");
919	while( db_step(&q)==SQLITE_ROW ){
920	int rid = db_column_int(&q, 0);
921	const char *zUuid = db_column_text(&q, 1);
922	int nUuid = db_column_bytes(&q, 1);
923	int size = db_column_int(&q, 2);
924	n1++;
925	fossil_print(" %d/%d\r", n1, total);
926	fflush(stdout);
927	if( size<0 ){
	@@ -919,14 +932,12 @@
932	if( blob_size(&content)!=size ){
933	fossil_print("size mismatch on artifact %d: wanted %d but got %d\n",
934	rid, size, blob_size(&content));
935	nErr++;
936	}
937	if( !hname_verify_hash(&content, zUuid, nUuid) ){
938	fossil_print("wrong hash on artifact %d\n",rid);


939	nErr++;
940	}
941	if( bParse && looks_like_control_artifact(&content) ){
942	Blob err;
943	int i, n;
	@@ -941,11 +952,11 @@
952	memcpy(zFirstLine, z, i);
953	zFirstLine[i] = 0;
954	p = manifest_parse(&content, 0, &err);
955	if( p==0 ){
956	fossil_print("manifest_parse failed for %s:\n%s\n",
957	zUuid, blob_str(&err));
958	if( strncmp(blob_str(&err), "line 1:", 7)==0 ){
959	fossil_print("\"%s\"\n", zFirstLine);
960	}
961	}else{
962	anCA[p->type]++;
	@@ -954,11 +965,10 @@
965	}
966	blob_reset(&err);
967	}else{
968	blob_reset(&content);
969	}

970	n2++;
971	}
972	db_finalize(&q);
973	fossil_print("%d non-phantom blobs (out of %d total) checked: %d errors\n",
974	n2, n1, nErr);
	@@ -1157,11 +1167,11 @@
1167	**
1168	** WARNING: You must run "fossil rebuild" after this command to rebuild
1169	** the metadata.
1170	**
1171	** Note that the arguments are the integer raw RID values from the BLOB table,
1172	** not artifact hashs or labels.
1173	*/
1174	void test_content_erase(void){
1175	int i;
1176	Blob x;
1177	char c;
1178

M src/db.c

+29 -34

		--- src/db.c
		+++ src/db.c
		@@ -2,11 +2,11 @@
2	2	** Copyright (c) 2006 D. Richard Hipp
3	3	**
4	4	** This program is free software; you can redistribute it and/or
5	5	** modify it under the terms of the Simplified BSD License (also
6	6	** known as the "2-Clause License" or "FreeBSD License".)
7		-
	7	+**
8	8	** This program is distributed in the hope that it will be useful,
9	9	** but without any warranty; without even the implied warranty of
10	10	** merchantability or fitness for a particular purpose.
11	11	**
12	12	** Author contact information:
		@@ -1485,35 +1485,20 @@
1485	1485	g.repositoryOpen = 1;
1486	1486	/* Cache "allow-symlinks" option, because we'll need it on every stat call */
1487	1487	g.allowSymlinks = db_get_boolean("allow-symlinks",
1488	1488	db_allow_symlinks_by_default());
1489	1489	g.zAuxSchema = db_get("aux-schema","");
1490		-
1491		- /* Verify that the PLINK table has a new column added by the
1492		- ** 2014-11-28 schema change. Create it if necessary. This code
1493		- ** can be removed in the future, once all users have upgraded to the
1494		- ** 2014-11-28 or later schema.
1495		- */
1496		- if( !db_table_has_column("repository","plink","baseid") ){
1497		- db_multi_exec(
1498		- "ALTER TABLE repository.plink ADD COLUMN baseid;"
1499		- );
1500		- }
1501		-
1502		- /* Verify that the MLINK table has the newer columns added by the
1503		- ** 2015-01-24 schema change. Create them if necessary. This code
1504		- ** can be removed in the future, once all users have upgraded to the
1505		- ** 2015-01-24 or later schema.
1506		- */
1507		- if( !db_table_has_column("repository","mlink","isaux") ){
1508		- db_begin_transaction();
1509		- db_multi_exec(
1510		- "ALTER TABLE repository.mlink ADD COLUMN pmid INTEGER DEFAULT 0;"
1511		- "ALTER TABLE repository.mlink ADD COLUMN isaux BOOLEAN DEFAULT 0;"
1512		- );
1513		- db_end_transaction(0);
1514		- }
	1490	+ g.eHashPolicy = db_get_int("hash-policy",-1);
	1491	+ if( g.eHashPolicy<0 ){
	1492	+ g.eHashPolicy = hname_default_policy();
	1493	+ db_set_int("hash-policy", g.eHashPolicy, 0);
	1494	+ }
	1495	+
	1496	+ /* Make a change to the CHECK constraint on the BLOB table for
	1497	+ ** version 2.0 and later.
	1498	+ */
	1499	+ rebuild_schema_update_2_0(); /* Do the Fossil-2.0 schema updates */
1515	1500	}
1516	1501
1517	1502	/*
1518	1503	** Flags for the db_find_and_open_repository() function.
1519	1504	*/
		@@ -1677,10 +1662,11 @@
1677	1662	while( db.pAllStmt ){
1678	1663	db_finalize(db.pAllStmt);
1679	1664	}
1680	1665	db_end_transaction(1);
1681	1666	pStmt = 0;
	1667	+ sqlite3_exec(g.db, "PRAGMA optimize", 0, 0, 0);
1682	1668	db_close_config();
1683	1669
1684	1670	/* If the localdb has a lot of unused free space,
1685	1671	** then VACUUM it as we shut down.
1686	1672	*/
		@@ -1848,10 +1834,11 @@
1848	1834	" AND name NOT GLOB 'project-*'"
1849	1835	" AND name NOT GLOB 'short-project-*';",
1850	1836	configure_inop_rhs(CONFIGSET_ALL),
1851	1837	db_setting_inop_rhs()
1852	1838	);
	1839	+ g.eHashPolicy = db_get_int("hash-policy", g.eHashPolicy);
1853	1840	db_multi_exec(
1854	1841	"REPLACE INTO reportfmt SELECT * FROM settingSrc.reportfmt;"
1855	1842	);
1856	1843
1857	1844	/*
		@@ -1920,13 +1907,14 @@
1920	1907	** their associated permissions will not be copied; however, the system
1921	1908	** default users "anonymous", "nobody", "reader", "developer", and their
1922	1909	** associated permissions will be copied.
1923	1910	**
1924	1911	** Options:
1925		-** --template FILE copy settings from repository file
1926		-** --admin-user\|-A USERNAME select given USERNAME as admin user
1927		-** --date-override DATETIME use DATETIME as time of the initial check-in
	1912	+** --template FILE Copy settings from repository file
	1913	+** --admin-user\|-A USERNAME Select given USERNAME as admin user
	1914	+** --date-override DATETIME Use DATETIME as time of the initial check-in
	1915	+** --sha1 Use a initial hash policy of "sha1"
1928	1916	**
1929	1917	** DATETIME may be "now" or "YYYY-MM-DDTHH:MM:SS.SSS". If in
1930	1918	** year-month-day form, it may be truncated, the "T" may be replaced by
1931	1919	** a space, and it may also name a timezone offset from UTC as "-HH:MM"
1932	1920	** (westward) or "+HH:MM" (eastward). Either no timezone suffix or "Z"
		@@ -1937,14 +1925,17 @@
1937	1925	void create_repository_cmd(void){
1938	1926	char *zPassword;
1939	1927	const char zTemplate; / Repository from which to copy settings */
1940	1928	const char zDate; / Date of the initial check-in */
1941	1929	const char zDefaultUser; / Optional name of the default user */
	1930	+ int bUseSha1 = 0; /* True to set the hash-policy to sha1 */
	1931	+
1942	1932
1943	1933	zTemplate = find_option("template",0,1);
1944	1934	zDate = find_option("date-override",0,1);
1945	1935	zDefaultUser = find_option("admin-user","A",1);
	1936	+ bUseSha1 = find_option("sha1",0,0)!=0;
1946	1937	/* We should be done with options.. */
1947	1938	verify_all_options();
1948	1939
1949	1940	if( g.argc!=3 ){
1950	1941	usage("REPOSITORY-NAME");
		@@ -1957,10 +1948,14 @@
1957	1948	db_create_repository(g.argv[2]);
1958	1949	db_open_repository(g.argv[2]);
1959	1950	db_open_config(0, 0);
1960	1951	if( zTemplate ) db_attach(zTemplate, "settingSrc");
1961	1952	db_begin_transaction();
	1953	+ if( bUseSha1 ){
	1954	+ g.eHashPolicy = HPOLICY_SHA1;
	1955	+ db_set_int("hash-policy", HPOLICY_SHA1, 0);
	1956	+ }
1962	1957	if( zDate==0 ) zDate = "now";
1963	1958	db_initial_setup(zTemplate, zDate, zDefaultUser);
1964	1959	db_end_transaction(0);
1965	1960	if( zTemplate ) db_detach("settingSrc");
1966	1961	fossil_print("project-id: %s\n", db_get("project-code", 0));
		@@ -2082,29 +2077,29 @@
2082	2077	sqlite3_result_value(context, argv[2-rc]);
2083	2078	}
2084	2079	}
2085	2080
2086	2081	/*
2087		-** Convert the input string into an SHA1. Make a notation in the
	2082	+** Convert the input string into a artifact hash. Make a notation in the
2088	2083	** CONCEALED table so that the hash can be undo using the db_reveal()
2089	2084	** function at some later time.
2090	2085	**
2091	2086	** The value returned is stored in static space and will be overwritten
2092	2087	** on subsequent calls.
2093	2088	**
2094		-** If zContent is already a well-formed SHA1 hash, then return a copy
	2089	+** If zContent is already a well-formed artifact hash, then return a copy
2095	2090	** of that hash, not a hash of the hash.
2096	2091	**
2097	2092	** The CONCEALED table is meant to obscure email addresses. Every valid
2098	2093	** email address will contain a "@" character and "@" is not valid within
2099		-** an SHA1 hash so there is no chance that a valid email address will go
	2094	+** a SHA1 hash so there is no chance that a valid email address will go
2100	2095	** unconcealed.
2101	2096	*/
2102	2097	char db_conceal(const char zContent, int n){
2103		- static char zHash[42];
	2098	+ static char zHash[HNAME_MAX+1];
2104	2099	Blob out;
2105		- if( n==40 && validate16(zContent, n) ){
	2100	+ if( hname_validate(zContent, n) ){
2106	2101	memcpy(zHash, zContent, n);
2107	2102	zHash[n] = 0;
2108	2103	}else{
2109	2104	sha1sum_step_text(zContent, n);
2110	2105	sha1sum_finish(&out);
2111	2106

	--- src/db.c
	+++ src/db.c
	@@ -2,11 +2,11 @@
2	** Copyright (c) 2006 D. Richard Hipp
3	**
4	** This program is free software; you can redistribute it and/or
5	** modify it under the terms of the Simplified BSD License (also
6	** known as the "2-Clause License" or "FreeBSD License".)
7
8	** This program is distributed in the hope that it will be useful,
9	** but without any warranty; without even the implied warranty of
10	** merchantability or fitness for a particular purpose.
11	**
12	** Author contact information:
	@@ -1485,35 +1485,20 @@
1485	g.repositoryOpen = 1;
1486	/* Cache "allow-symlinks" option, because we'll need it on every stat call */
1487	g.allowSymlinks = db_get_boolean("allow-symlinks",
1488	db_allow_symlinks_by_default());
1489	g.zAuxSchema = db_get("aux-schema","");
1490
1491	/* Verify that the PLINK table has a new column added by the
1492	** 2014-11-28 schema change. Create it if necessary. This code
1493	** can be removed in the future, once all users have upgraded to the
1494	** 2014-11-28 or later schema.
1495	*/
1496	if( !db_table_has_column("repository","plink","baseid") ){
1497	db_multi_exec(
1498	"ALTER TABLE repository.plink ADD COLUMN baseid;"
1499	);
1500	}
1501
1502	/* Verify that the MLINK table has the newer columns added by the
1503	** 2015-01-24 schema change. Create them if necessary. This code
1504	** can be removed in the future, once all users have upgraded to the
1505	** 2015-01-24 or later schema.
1506	*/
1507	if( !db_table_has_column("repository","mlink","isaux") ){
1508	db_begin_transaction();
1509	db_multi_exec(
1510	"ALTER TABLE repository.mlink ADD COLUMN pmid INTEGER DEFAULT 0;"
1511	"ALTER TABLE repository.mlink ADD COLUMN isaux BOOLEAN DEFAULT 0;"
1512	);
1513	db_end_transaction(0);
1514	}
1515	}
1516
1517	/*
1518	** Flags for the db_find_and_open_repository() function.
1519	*/
	@@ -1677,10 +1662,11 @@
1677	while( db.pAllStmt ){
1678	db_finalize(db.pAllStmt);
1679	}
1680	db_end_transaction(1);
1681	pStmt = 0;

1682	db_close_config();
1683
1684	/* If the localdb has a lot of unused free space,
1685	** then VACUUM it as we shut down.
1686	*/
	@@ -1848,10 +1834,11 @@
1848	" AND name NOT GLOB 'project-*'"
1849	" AND name NOT GLOB 'short-project-*';",
1850	configure_inop_rhs(CONFIGSET_ALL),
1851	db_setting_inop_rhs()
1852	);

1853	db_multi_exec(
1854	"REPLACE INTO reportfmt SELECT * FROM settingSrc.reportfmt;"
1855	);
1856
1857	/*
	@@ -1920,13 +1907,14 @@
1920	** their associated permissions will not be copied; however, the system
1921	** default users "anonymous", "nobody", "reader", "developer", and their
1922	** associated permissions will be copied.
1923	**
1924	** Options:
1925	** --template FILE copy settings from repository file
1926	** --admin-user\|-A USERNAME select given USERNAME as admin user
1927	** --date-override DATETIME use DATETIME as time of the initial check-in

1928	**
1929	** DATETIME may be "now" or "YYYY-MM-DDTHH:MM:SS.SSS". If in
1930	** year-month-day form, it may be truncated, the "T" may be replaced by
1931	** a space, and it may also name a timezone offset from UTC as "-HH:MM"
1932	** (westward) or "+HH:MM" (eastward). Either no timezone suffix or "Z"
	@@ -1937,14 +1925,17 @@
1937	void create_repository_cmd(void){
1938	char *zPassword;
1939	const char zTemplate; / Repository from which to copy settings */
1940	const char zDate; / Date of the initial check-in */
1941	const char zDefaultUser; / Optional name of the default user */


1942
1943	zTemplate = find_option("template",0,1);
1944	zDate = find_option("date-override",0,1);
1945	zDefaultUser = find_option("admin-user","A",1);

1946	/* We should be done with options.. */
1947	verify_all_options();
1948
1949	if( g.argc!=3 ){
1950	usage("REPOSITORY-NAME");
	@@ -1957,10 +1948,14 @@
1957	db_create_repository(g.argv[2]);
1958	db_open_repository(g.argv[2]);
1959	db_open_config(0, 0);
1960	if( zTemplate ) db_attach(zTemplate, "settingSrc");
1961	db_begin_transaction();




1962	if( zDate==0 ) zDate = "now";
1963	db_initial_setup(zTemplate, zDate, zDefaultUser);
1964	db_end_transaction(0);
1965	if( zTemplate ) db_detach("settingSrc");
1966	fossil_print("project-id: %s\n", db_get("project-code", 0));
	@@ -2082,29 +2077,29 @@
2082	sqlite3_result_value(context, argv[2-rc]);
2083	}
2084	}
2085
2086	/*
2087	** Convert the input string into an SHA1. Make a notation in the
2088	** CONCEALED table so that the hash can be undo using the db_reveal()
2089	** function at some later time.
2090	**
2091	** The value returned is stored in static space and will be overwritten
2092	** on subsequent calls.
2093	**
2094	** If zContent is already a well-formed SHA1 hash, then return a copy
2095	** of that hash, not a hash of the hash.
2096	**
2097	** The CONCEALED table is meant to obscure email addresses. Every valid
2098	** email address will contain a "@" character and "@" is not valid within
2099	** an SHA1 hash so there is no chance that a valid email address will go
2100	** unconcealed.
2101	*/
2102	char db_conceal(const char zContent, int n){
2103	static char zHash[42];
2104	Blob out;
2105	if( n==40 && validate16(zContent, n) ){
2106	memcpy(zHash, zContent, n);
2107	zHash[n] = 0;
2108	}else{
2109	sha1sum_step_text(zContent, n);
2110	sha1sum_finish(&out);
2111

	--- src/db.c
	+++ src/db.c
	@@ -2,11 +2,11 @@
2	** Copyright (c) 2006 D. Richard Hipp
3	**
4	** This program is free software; you can redistribute it and/or
5	** modify it under the terms of the Simplified BSD License (also
6	** known as the "2-Clause License" or "FreeBSD License".)
7	**
8	** This program is distributed in the hope that it will be useful,
9	** but without any warranty; without even the implied warranty of
10	** merchantability or fitness for a particular purpose.
11	**
12	** Author contact information:
	@@ -1485,35 +1485,20 @@
1485	g.repositoryOpen = 1;
1486	/* Cache "allow-symlinks" option, because we'll need it on every stat call */
1487	g.allowSymlinks = db_get_boolean("allow-symlinks",
1488	db_allow_symlinks_by_default());
1489	g.zAuxSchema = db_get("aux-schema","");
1490	g.eHashPolicy = db_get_int("hash-policy",-1);
1491	if( g.eHashPolicy<0 ){
1492	g.eHashPolicy = hname_default_policy();
1493	db_set_int("hash-policy", g.eHashPolicy, 0);
1494	}
1495
1496	/* Make a change to the CHECK constraint on the BLOB table for
1497	** version 2.0 and later.
1498	*/
1499	rebuild_schema_update_2_0(); /* Do the Fossil-2.0 schema updates */















1500	}
1501
1502	/*
1503	** Flags for the db_find_and_open_repository() function.
1504	*/
	@@ -1677,10 +1662,11 @@
1662	while( db.pAllStmt ){
1663	db_finalize(db.pAllStmt);
1664	}
1665	db_end_transaction(1);
1666	pStmt = 0;
1667	sqlite3_exec(g.db, "PRAGMA optimize", 0, 0, 0);
1668	db_close_config();
1669
1670	/* If the localdb has a lot of unused free space,
1671	** then VACUUM it as we shut down.
1672	*/
	@@ -1848,10 +1834,11 @@
1834	" AND name NOT GLOB 'project-*'"
1835	" AND name NOT GLOB 'short-project-*';",
1836	configure_inop_rhs(CONFIGSET_ALL),
1837	db_setting_inop_rhs()
1838	);
1839	g.eHashPolicy = db_get_int("hash-policy", g.eHashPolicy);
1840	db_multi_exec(
1841	"REPLACE INTO reportfmt SELECT * FROM settingSrc.reportfmt;"
1842	);
1843
1844	/*
	@@ -1920,13 +1907,14 @@
1907	** their associated permissions will not be copied; however, the system
1908	** default users "anonymous", "nobody", "reader", "developer", and their
1909	** associated permissions will be copied.
1910	**
1911	** Options:
1912	** --template FILE Copy settings from repository file
1913	** --admin-user\|-A USERNAME Select given USERNAME as admin user
1914	** --date-override DATETIME Use DATETIME as time of the initial check-in
1915	** --sha1 Use a initial hash policy of "sha1"
1916	**
1917	** DATETIME may be "now" or "YYYY-MM-DDTHH:MM:SS.SSS". If in
1918	** year-month-day form, it may be truncated, the "T" may be replaced by
1919	** a space, and it may also name a timezone offset from UTC as "-HH:MM"
1920	** (westward) or "+HH:MM" (eastward). Either no timezone suffix or "Z"
	@@ -1937,14 +1925,17 @@
1925	void create_repository_cmd(void){
1926	char *zPassword;
1927	const char zTemplate; / Repository from which to copy settings */
1928	const char zDate; / Date of the initial check-in */
1929	const char zDefaultUser; / Optional name of the default user */
1930	int bUseSha1 = 0; /* True to set the hash-policy to sha1 */
1931
1932
1933	zTemplate = find_option("template",0,1);
1934	zDate = find_option("date-override",0,1);
1935	zDefaultUser = find_option("admin-user","A",1);
1936	bUseSha1 = find_option("sha1",0,0)!=0;
1937	/* We should be done with options.. */
1938	verify_all_options();
1939
1940	if( g.argc!=3 ){
1941	usage("REPOSITORY-NAME");
	@@ -1957,10 +1948,14 @@
1948	db_create_repository(g.argv[2]);
1949	db_open_repository(g.argv[2]);
1950	db_open_config(0, 0);
1951	if( zTemplate ) db_attach(zTemplate, "settingSrc");
1952	db_begin_transaction();
1953	if( bUseSha1 ){
1954	g.eHashPolicy = HPOLICY_SHA1;
1955	db_set_int("hash-policy", HPOLICY_SHA1, 0);
1956	}
1957	if( zDate==0 ) zDate = "now";
1958	db_initial_setup(zTemplate, zDate, zDefaultUser);
1959	db_end_transaction(0);
1960	if( zTemplate ) db_detach("settingSrc");
1961	fossil_print("project-id: %s\n", db_get("project-code", 0));
	@@ -2082,29 +2077,29 @@
2077	sqlite3_result_value(context, argv[2-rc]);
2078	}
2079	}
2080
2081	/*
2082	** Convert the input string into a artifact hash. Make a notation in the
2083	** CONCEALED table so that the hash can be undo using the db_reveal()
2084	** function at some later time.
2085	**
2086	** The value returned is stored in static space and will be overwritten
2087	** on subsequent calls.
2088	**
2089	** If zContent is already a well-formed artifact hash, then return a copy
2090	** of that hash, not a hash of the hash.
2091	**
2092	** The CONCEALED table is meant to obscure email addresses. Every valid
2093	** email address will contain a "@" character and "@" is not valid within
2094	** a SHA1 hash so there is no chance that a valid email address will go
2095	** unconcealed.
2096	*/
2097	char db_conceal(const char zContent, int n){
2098	static char zHash[HNAME_MAX+1];
2099	Blob out;
2100	if( hname_validate(zContent, n) ){
2101	memcpy(zHash, zContent, n);
2102	zHash[n] = 0;
2103	}else{
2104	sha1sum_step_text(zContent, n);
2105	sha1sum_finish(&out);
2106

M src/diff.c

+1 -1

		--- src/diff.c
		+++ src/diff.c
		@@ -2278,11 +2278,11 @@
2278	2278	** URL: /annotate?checkin=ID&filename=FILENAME
2279	2279	** URL: /blame?checkin=ID&filename=FILENAME
2280	2280	** URL: /praise?checkin=ID&filename=FILENAME
2281	2281	**
2282	2282	** Show the most recent change to each line of a text file. /annotate shows
2283		-** the date of the changes and the check-in SHA1 hash (with a link to the
	2283	+** the date of the changes and the check-in hash (with a link to the
2284	2284	** check-in). /blame and /praise also show the user who made the check-in.
2285	2285	**
2286	2286	** Query parameters:
2287	2287	**
2288	2288	** checkin=ID The manifest ID at which to start the annotation
2289	2289

	--- src/diff.c
	+++ src/diff.c
	@@ -2278,11 +2278,11 @@
2278	** URL: /annotate?checkin=ID&filename=FILENAME
2279	** URL: /blame?checkin=ID&filename=FILENAME
2280	** URL: /praise?checkin=ID&filename=FILENAME
2281	**
2282	** Show the most recent change to each line of a text file. /annotate shows
2283	** the date of the changes and the check-in SHA1 hash (with a link to the
2284	** check-in). /blame and /praise also show the user who made the check-in.
2285	**
2286	** Query parameters:
2287	**
2288	** checkin=ID The manifest ID at which to start the annotation
2289

	--- src/diff.c
	+++ src/diff.c
	@@ -2278,11 +2278,11 @@
2278	** URL: /annotate?checkin=ID&filename=FILENAME
2279	** URL: /blame?checkin=ID&filename=FILENAME
2280	** URL: /praise?checkin=ID&filename=FILENAME
2281	**
2282	** Show the most recent change to each line of a text file. /annotate shows
2283	** the date of the changes and the check-in hash (with a link to the
2284	** check-in). /blame and /praise also show the user who made the check-in.
2285	**
2286	** Query parameters:
2287	**
2288	** checkin=ID The manifest ID at which to start the annotation
2289

M src/diffcmd.c

+22 -8

		--- src/diffcmd.c
		+++ src/diffcmd.c
		@@ -151,10 +151,13 @@
151	151	/*
152	152	** Show the difference between two files, one in memory and one on disk.
153	153	**
154	154	** The difference is the set of edits needed to transform pFile1 into
155	155	** zFile2. The content of pFile1 is in memory. zFile2 exists on disk.
	156	+**
	157	+** If fSwapDiff is 1, show the set of edits to transform zFile2 into pFile1
	158	+** instead of the opposite.
156	159	**
157	160	** Use the internal diff logic if zDiffCmd is NULL. Otherwise call the
158	161	** command zDiffCmd to do the diffing.
159	162	**
160	163	** When using an external diff program, zBinGlob contains the GLOB patterns
		@@ -167,11 +170,12 @@
167	170	const char zFile2, / On disk content to compare to */
168	171	const char zName, / Display name of the file */
169	172	const char zDiffCmd, / Command for comparison */
170	173	const char zBinGlob, / Treat file names matching this as binary */
171	174	int fIncludeBinary, /* Include binary files for external diff */
172		- u64 diffFlags /* Flags to control the diff */
	175	+ u64 diffFlags, /* Flags to control the diff */
	176	+ int fSwapDiff /* Diff from Zfile2 to Pfile1 */
173	177	){
174	178	if( zDiffCmd==0 ){
175	179	Blob out; /* Diff output text */
176	180	Blob file2; /* Content of zFile2 */
177	181	const char zName2; / Name of zFile2 for display */
		@@ -194,11 +198,15 @@
194	198	if( blob_compare(pFile1, &file2) ){
195	199	fossil_print("CHANGED %s\n", zName);
196	200	}
197	201	}else{
198	202	blob_zero(&out);
199		- text_diff(pFile1, &file2, &out, 0, diffFlags);
	203	+ if( fSwapDiff ){
	204	+ text_diff(&file2, pFile1, &out, 0, diffFlags);
	205	+ }else{
	206	+ text_diff(pFile1, &file2, &out, 0, diffFlags);
	207	+ }
200	208	if( blob_size(&out) ){
201	209	diff_print_filenames(zName, zName2, diffFlags);
202	210	fossil_print("%s\n", blob_str(&out));
203	211	}
204	212	blob_reset(&out);
		@@ -252,13 +260,19 @@
252	260	blob_write_to_file(pFile1, blob_str(&nameFile1));
253	261
254	262	/* Construct the external diff command */
255	263	blob_zero(&cmd);
256	264	blob_appendf(&cmd, "%s ", zDiffCmd);
257		- shell_escape(&cmd, blob_str(&nameFile1));
258		- blob_append(&cmd, " ", 1);
259		- shell_escape(&cmd, zFile2);
	265	+ if( fSwapDiff ){
	266	+ shell_escape(&cmd, zFile2);
	267	+ blob_append(&cmd, " ", 1);
	268	+ shell_escape(&cmd, blob_str(&nameFile1));
	269	+ }else{
	270	+ shell_escape(&cmd, blob_str(&nameFile1));
	271	+ blob_append(&cmd, " ", 1);
	272	+ shell_escape(&cmd, zFile2);
	273	+ }
260	274
261	275	/* Run the external diff command */
262	276	fossil_system(blob_str(&cmd));
263	277
264	278	/* Delete the temporary file and clean up memory used */
		@@ -482,11 +496,11 @@
482	496	blob_zero(&content);
483	497	}
484	498	isBin = fIncludeBinary ? 0 : looks_like_binary(&content);
485	499	diff_print_index(zPathname, diffFlags);
486	500	diff_file(&content, isBin, zFullName, zPathname, zDiffCmd,
487		- zBinGlob, fIncludeBinary, diffFlags);
	501	+ zBinGlob, fIncludeBinary, diffFlags, 0);
488	502	blob_reset(&content);
489	503	}
490	504	blob_reset(&fname);
491	505	}
492	506	db_finalize(&q);
		@@ -519,11 +533,11 @@
519	533	const char zFile = (const char)db_column_text(&q, 0);
520	534	if( !file_dir_match(pFileDir, zFile) ) continue;
521	535	zFullName = mprintf("%s%s", g.zLocalRoot, zFile);
522	536	db_column_blob(&q, 1, &content);
523	537	diff_file(&content, 0, zFullName, zFile,
524		- zDiffCmd, zBinGlob, fIncludeBinary, diffFlags);
	538	+ zDiffCmd, zBinGlob, fIncludeBinary, diffFlags, 0);
525	539	fossil_free(zFullName);
526	540	blob_reset(&content);
527	541	}
528	542	db_finalize(&q);
529	543	}
		@@ -750,11 +764,11 @@
750	764	** Returns non-zero if files that may be binary should be used with external
751	765	** diff programs.
752	766	*/
753	767	int diff_include_binary_files(void){
754	768	const char* zArgIncludeBinary = find_option("diff-binary", 0, 1);
755		-
	769	+
756	770	/* Command line argument have priority on settings */
757	771	if( zArgIncludeBinary ){
758	772	return is_truth(zArgIncludeBinary);
759	773	}else{
760	774	return db_get_boolean("diff-binary", 1);
761	775

	--- src/diffcmd.c
	+++ src/diffcmd.c
	@@ -151,10 +151,13 @@
151	/*
152	** Show the difference between two files, one in memory and one on disk.
153	**
154	** The difference is the set of edits needed to transform pFile1 into
155	** zFile2. The content of pFile1 is in memory. zFile2 exists on disk.



156	**
157	** Use the internal diff logic if zDiffCmd is NULL. Otherwise call the
158	** command zDiffCmd to do the diffing.
159	**
160	** When using an external diff program, zBinGlob contains the GLOB patterns
	@@ -167,11 +170,12 @@
167	const char zFile2, / On disk content to compare to */
168	const char zName, / Display name of the file */
169	const char zDiffCmd, / Command for comparison */
170	const char zBinGlob, / Treat file names matching this as binary */
171	int fIncludeBinary, /* Include binary files for external diff */
172	u64 diffFlags /* Flags to control the diff */

173	){
174	if( zDiffCmd==0 ){
175	Blob out; /* Diff output text */
176	Blob file2; /* Content of zFile2 */
177	const char zName2; / Name of zFile2 for display */
	@@ -194,11 +198,15 @@
194	if( blob_compare(pFile1, &file2) ){
195	fossil_print("CHANGED %s\n", zName);
196	}
197	}else{
198	blob_zero(&out);
199	text_diff(pFile1, &file2, &out, 0, diffFlags);




200	if( blob_size(&out) ){
201	diff_print_filenames(zName, zName2, diffFlags);
202	fossil_print("%s\n", blob_str(&out));
203	}
204	blob_reset(&out);
	@@ -252,13 +260,19 @@
252	blob_write_to_file(pFile1, blob_str(&nameFile1));
253
254	/* Construct the external diff command */
255	blob_zero(&cmd);
256	blob_appendf(&cmd, "%s ", zDiffCmd);
257	shell_escape(&cmd, blob_str(&nameFile1));
258	blob_append(&cmd, " ", 1);
259	shell_escape(&cmd, zFile2);






260
261	/* Run the external diff command */
262	fossil_system(blob_str(&cmd));
263
264	/* Delete the temporary file and clean up memory used */
	@@ -482,11 +496,11 @@
482	blob_zero(&content);
483	}
484	isBin = fIncludeBinary ? 0 : looks_like_binary(&content);
485	diff_print_index(zPathname, diffFlags);
486	diff_file(&content, isBin, zFullName, zPathname, zDiffCmd,
487	zBinGlob, fIncludeBinary, diffFlags);
488	blob_reset(&content);
489	}
490	blob_reset(&fname);
491	}
492	db_finalize(&q);
	@@ -519,11 +533,11 @@
519	const char zFile = (const char)db_column_text(&q, 0);
520	if( !file_dir_match(pFileDir, zFile) ) continue;
521	zFullName = mprintf("%s%s", g.zLocalRoot, zFile);
522	db_column_blob(&q, 1, &content);
523	diff_file(&content, 0, zFullName, zFile,
524	zDiffCmd, zBinGlob, fIncludeBinary, diffFlags);
525	fossil_free(zFullName);
526	blob_reset(&content);
527	}
528	db_finalize(&q);
529	}
	@@ -750,11 +764,11 @@
750	** Returns non-zero if files that may be binary should be used with external
751	** diff programs.
752	*/
753	int diff_include_binary_files(void){
754	const char* zArgIncludeBinary = find_option("diff-binary", 0, 1);
755
756	/* Command line argument have priority on settings */
757	if( zArgIncludeBinary ){
758	return is_truth(zArgIncludeBinary);
759	}else{
760	return db_get_boolean("diff-binary", 1);
761

	--- src/diffcmd.c
	+++ src/diffcmd.c
	@@ -151,10 +151,13 @@
151	/*
152	** Show the difference between two files, one in memory and one on disk.
153	**
154	** The difference is the set of edits needed to transform pFile1 into
155	** zFile2. The content of pFile1 is in memory. zFile2 exists on disk.
156	**
157	** If fSwapDiff is 1, show the set of edits to transform zFile2 into pFile1
158	** instead of the opposite.
159	**
160	** Use the internal diff logic if zDiffCmd is NULL. Otherwise call the
161	** command zDiffCmd to do the diffing.
162	**
163	** When using an external diff program, zBinGlob contains the GLOB patterns
	@@ -167,11 +170,12 @@
170	const char zFile2, / On disk content to compare to */
171	const char zName, / Display name of the file */
172	const char zDiffCmd, / Command for comparison */
173	const char zBinGlob, / Treat file names matching this as binary */
174	int fIncludeBinary, /* Include binary files for external diff */
175	u64 diffFlags, /* Flags to control the diff */
176	int fSwapDiff /* Diff from Zfile2 to Pfile1 */
177	){
178	if( zDiffCmd==0 ){
179	Blob out; /* Diff output text */
180	Blob file2; /* Content of zFile2 */
181	const char zName2; / Name of zFile2 for display */
	@@ -194,11 +198,15 @@
198	if( blob_compare(pFile1, &file2) ){
199	fossil_print("CHANGED %s\n", zName);
200	}
201	}else{
202	blob_zero(&out);
203	if( fSwapDiff ){
204	text_diff(&file2, pFile1, &out, 0, diffFlags);
205	}else{
206	text_diff(pFile1, &file2, &out, 0, diffFlags);
207	}
208	if( blob_size(&out) ){
209	diff_print_filenames(zName, zName2, diffFlags);
210	fossil_print("%s\n", blob_str(&out));
211	}
212	blob_reset(&out);
	@@ -252,13 +260,19 @@
260	blob_write_to_file(pFile1, blob_str(&nameFile1));
261
262	/* Construct the external diff command */
263	blob_zero(&cmd);
264	blob_appendf(&cmd, "%s ", zDiffCmd);
265	if( fSwapDiff ){
266	shell_escape(&cmd, zFile2);
267	blob_append(&cmd, " ", 1);
268	shell_escape(&cmd, blob_str(&nameFile1));
269	}else{
270	shell_escape(&cmd, blob_str(&nameFile1));
271	blob_append(&cmd, " ", 1);
272	shell_escape(&cmd, zFile2);
273	}
274
275	/* Run the external diff command */
276	fossil_system(blob_str(&cmd));
277
278	/* Delete the temporary file and clean up memory used */
	@@ -482,11 +496,11 @@
496	blob_zero(&content);
497	}
498	isBin = fIncludeBinary ? 0 : looks_like_binary(&content);
499	diff_print_index(zPathname, diffFlags);
500	diff_file(&content, isBin, zFullName, zPathname, zDiffCmd,
501	zBinGlob, fIncludeBinary, diffFlags, 0);
502	blob_reset(&content);
503	}
504	blob_reset(&fname);
505	}
506	db_finalize(&q);
	@@ -519,11 +533,11 @@
533	const char zFile = (const char)db_column_text(&q, 0);
534	if( !file_dir_match(pFileDir, zFile) ) continue;
535	zFullName = mprintf("%s%s", g.zLocalRoot, zFile);
536	db_column_blob(&q, 1, &content);
537	diff_file(&content, 0, zFullName, zFile,
538	zDiffCmd, zBinGlob, fIncludeBinary, diffFlags, 0);
539	fossil_free(zFullName);
540	blob_reset(&content);
541	}
542	db_finalize(&q);
543	}
	@@ -750,11 +764,11 @@
764	** Returns non-zero if files that may be binary should be used with external
765	** diff programs.
766	*/
767	int diff_include_binary_files(void){
768	const char* zArgIncludeBinary = find_option("diff-binary", 0, 1);
769
770	/* Command line argument have priority on settings */
771	if( zArgIncludeBinary ){
772	return is_truth(zArgIncludeBinary);
773	}else{
774	return db_get_boolean("diff-binary", 1);
775

M src/doc.c

+12 -4

		--- src/doc.c
		+++ src/doc.c
		@@ -527,11 +527,11 @@
527	527	** WEBPAGE: uv
528	528	** WEBPAGE: doc
529	529	** URL: /uv/FILE
530	530	** URL: /doc/CHECKIN/FILE
531	531	**
532		-** CHECKIN can be either tag or SHA1 hash or timestamp identifying a
	532	+** CHECKIN can be either tag or hash prefix or timestamp identifying a
533	533	** particular check, or the name of a branch (meaning the most recent
534	534	** check-in on that branch) or one of various magic words:
535	535	**
536	536	** "tip" means the most recent check-in
537	537	**
		@@ -708,13 +708,21 @@
708	708	#ifdef FOSSIL_ENABLE_TH1_DOCS
709	709	}else if( Th_AreDocsEnabled() &&
710	710	fossil_strcmp(zMime, "application/x-th1")==0 ){
711	711	int raw = P("raw")!=0;
712	712	if( !raw ){
713		- style_header("%h", zName);
	713	+ Blob tail;
	714	+ blob_zero(&tail);
	715	+ if( wiki_find_title(&filebody, &title, &tail) ){
	716	+ style_header("%s", blob_str(&title));
	717	+ Th_Render(blob_str(&tail));
	718	+ blob_reset(&tail);
	719	+ }else{
	720	+ style_header("%h", zName);
	721	+ Th_Render(blob_str(&filebody));
	722	+ }
714	723	}
715		- Th_Render(blob_str(&filebody));
716	724	if( !raw ){
717	725	style_footer();
718	726	}
719	727	#endif
720	728	}else{
		@@ -727,11 +735,11 @@
727	735
728	736	/* Jump here when unable to locate the document */
729	737	doc_not_found:
730	738	db_end_transaction(0);
731	739	if( isUV && P("name")==0 ){
732		- uvstat_page();
	740	+ uvlist_page();
733	741	return;
734	742	}
735	743	cgi_set_status(404, "Not Found");
736	744	style_header("Not Found");
737	745	@ <p>Document %h(zOrigName) not found
738	746

	--- src/doc.c
	+++ src/doc.c
	@@ -527,11 +527,11 @@
527	** WEBPAGE: uv
528	** WEBPAGE: doc
529	** URL: /uv/FILE
530	** URL: /doc/CHECKIN/FILE
531	**
532	** CHECKIN can be either tag or SHA1 hash or timestamp identifying a
533	** particular check, or the name of a branch (meaning the most recent
534	** check-in on that branch) or one of various magic words:
535	**
536	** "tip" means the most recent check-in
537	**
	@@ -708,13 +708,21 @@
708	#ifdef FOSSIL_ENABLE_TH1_DOCS
709	}else if( Th_AreDocsEnabled() &&
710	fossil_strcmp(zMime, "application/x-th1")==0 ){
711	int raw = P("raw")!=0;
712	if( !raw ){
713	style_header("%h", zName);









714	}
715	Th_Render(blob_str(&filebody));
716	if( !raw ){
717	style_footer();
718	}
719	#endif
720	}else{
	@@ -727,11 +735,11 @@
727
728	/* Jump here when unable to locate the document */
729	doc_not_found:
730	db_end_transaction(0);
731	if( isUV && P("name")==0 ){
732	uvstat_page();
733	return;
734	}
735	cgi_set_status(404, "Not Found");
736	style_header("Not Found");
737	@ <p>Document %h(zOrigName) not found
738

	--- src/doc.c
	+++ src/doc.c
	@@ -527,11 +527,11 @@
527	** WEBPAGE: uv
528	** WEBPAGE: doc
529	** URL: /uv/FILE
530	** URL: /doc/CHECKIN/FILE
531	**
532	** CHECKIN can be either tag or hash prefix or timestamp identifying a
533	** particular check, or the name of a branch (meaning the most recent
534	** check-in on that branch) or one of various magic words:
535	**
536	** "tip" means the most recent check-in
537	**
	@@ -708,13 +708,21 @@
708	#ifdef FOSSIL_ENABLE_TH1_DOCS
709	}else if( Th_AreDocsEnabled() &&
710	fossil_strcmp(zMime, "application/x-th1")==0 ){
711	int raw = P("raw")!=0;
712	if( !raw ){
713	Blob tail;
714	blob_zero(&tail);
715	if( wiki_find_title(&filebody, &title, &tail) ){
716	style_header("%s", blob_str(&title));
717	Th_Render(blob_str(&tail));
718	blob_reset(&tail);
719	}else{
720	style_header("%h", zName);
721	Th_Render(blob_str(&filebody));
722	}
723	}

724	if( !raw ){
725	style_footer();
726	}
727	#endif
728	}else{
	@@ -727,11 +735,11 @@
735
736	/* Jump here when unable to locate the document */
737	doc_not_found:
738	db_end_transaction(0);
739	if( isUV && P("name")==0 ){
740	uvlist_page();
741	return;
742	}
743	cgi_set_status(404, "Not Found");
744	style_header("Not Found");
745	@ <p>Document %h(zOrigName) not found
746

M src/encode.c

+90

		--- src/encode.c
		+++ src/encode.c
		@@ -336,10 +336,100 @@
336	336	z[j++] = c;
337	337	}
338	338	if( z[j] ) z[j] = 0;
339	339	}
340	340
	341	+
	342	+/*
	343	+** The *pz variable points to a UTF8 string. Read the next character
	344	+** off of that string and return its codepoint value. Advance *pz to the
	345	+** next character
	346	+*/
	347	+u32 fossil_utf8_read(
	348	+ const unsigned char *pz / Pointer to string from which to read char */
	349	+){
	350	+ unsigned int c;
	351	+
	352	+ /*
	353	+ ** This lookup table is used to help decode the first byte of
	354	+ ** a multi-byte UTF8 character.
	355	+ */
	356	+ static const unsigned char utf8Trans1[] = {
	357	+ 0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
	358	+ 0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f,
	359	+ 0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17,
	360	+ 0x18, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f,
	361	+ 0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
	362	+ 0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f,
	363	+ 0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
	364	+ 0x00, 0x01, 0x02, 0x03, 0x00, 0x01, 0x00, 0x00,
	365	+ };
	366	+
	367	+ c = ((pz)++);
	368	+ if( c>=0xc0 ){
	369	+ c = utf8Trans1[c-0xc0];
	370	+ while( ((pz) & 0xc0)==0x80 ){
	371	+ c = (c<<6) + (0x3f & ((pz)++));
	372	+ }
	373	+ if( c<0x80
	374	+ \|\| (c&0xFFFFF800)==0xD800
	375	+ \|\| (c&0xFFFFFFFE)==0xFFFE ){ c = 0xFFFD; }
	376	+ }
	377	+ return c;
	378	+}
	379	+
	380	+/*
	381	+** Encode a UTF8 string for JSON. All special characters are escaped.
	382	+*/
	383	+void blob_append_json_string(Blob pBlob, const char zStr){
	384	+ const unsigned char *z;
	385	+ char *zOut;
	386	+ u32 c;
	387	+ int n, i, j;
	388	+ z = (const unsigned char*)zStr;
	389	+ n = 0;
	390	+ while( (c = fossil_utf8_read(&z))!=0 ){
	391	+ if( c=='\\' \|\| c=='"' ){
	392	+ n += 2;
	393	+ }else if( c<' ' \|\| c>=0x7f ){
	394	+ if( c=='\n' \|\| c=='\r' ){
	395	+ n += 2;
	396	+ }else{
	397	+ n += 6;
	398	+ }
	399	+ }else{
	400	+ n++;
	401	+ }
	402	+ }
	403	+ i = blob_size(pBlob);
	404	+ blob_resize(pBlob, i+n);
	405	+ zOut = blob_buffer(pBlob);
	406	+ z = (const unsigned char*)zStr;
	407	+ while( (c = fossil_utf8_read(&z))!=0 ){
	408	+ if( c=='\\' ){
	409	+ zOut[i++] = '\\';
	410	+ zOut[i++] = c;
	411	+ }else if( c<' ' \|\| c>=0x7f ){
	412	+ zOut[i++] = '\\';
	413	+ if( c=='\n' ){
	414	+ zOut[i++] = 'n';
	415	+ }else if( c=='\r' ){
	416	+ zOut[i++] = 'r';
	417	+ }else{
	418	+ zOut[i++] = 'u';
	419	+ for(j=3; j>=0; j--){
	420	+ zOut[i+j] = "0123456789abcdef"[c&0xf];
	421	+ c >>= 4;
	422	+ }
	423	+ i += 4;
	424	+ }
	425	+ }else{
	426	+ zOut[i++] = c;
	427	+ }
	428	+ }
	429	+ zOut[i] = 0;
	430	+}
341	431
342	432	/*
343	433	** The characters used for HTTP base64 encoding.
344	434	*/
345	435	static unsigned char zBase[] =
346	436

	--- src/encode.c
	+++ src/encode.c
	@@ -336,10 +336,100 @@
336	z[j++] = c;
337	}
338	if( z[j] ) z[j] = 0;
339	}
340


























































































341
342	/*
343	** The characters used for HTTP base64 encoding.
344	*/
345	static unsigned char zBase[] =
346

	--- src/encode.c
	+++ src/encode.c
	@@ -336,10 +336,100 @@
336	z[j++] = c;
337	}
338	if( z[j] ) z[j] = 0;
339	}
340
341
342	/*
343	** The *pz variable points to a UTF8 string. Read the next character
344	** off of that string and return its codepoint value. Advance *pz to the
345	** next character
346	*/
347	u32 fossil_utf8_read(
348	const unsigned char *pz / Pointer to string from which to read char */
349	){
350	unsigned int c;
351
352	/*
353	** This lookup table is used to help decode the first byte of
354	** a multi-byte UTF8 character.
355	*/
356	static const unsigned char utf8Trans1[] = {
357	0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
358	0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f,
359	0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17,
360	0x18, 0x19, 0x1a, 0x1b, 0x1c, 0x1d, 0x1e, 0x1f,
361	0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
362	0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f,
363	0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
364	0x00, 0x01, 0x02, 0x03, 0x00, 0x01, 0x00, 0x00,
365	};
366
367	c = ((pz)++);
368	if( c>=0xc0 ){
369	c = utf8Trans1[c-0xc0];
370	while( ((pz) & 0xc0)==0x80 ){
371	c = (c<<6) + (0x3f & ((pz)++));
372	}
373	if( c<0x80
374	\|\| (c&0xFFFFF800)==0xD800
375	\|\| (c&0xFFFFFFFE)==0xFFFE ){ c = 0xFFFD; }
376	}
377	return c;
378	}
379
380	/*
381	** Encode a UTF8 string for JSON. All special characters are escaped.
382	*/
383	void blob_append_json_string(Blob pBlob, const char zStr){
384	const unsigned char *z;
385	char *zOut;
386	u32 c;
387	int n, i, j;
388	z = (const unsigned char*)zStr;
389	n = 0;
390	while( (c = fossil_utf8_read(&z))!=0 ){
391	if( c=='\\' \|\| c=='"' ){
392	n += 2;
393	}else if( c<' ' \|\| c>=0x7f ){
394	if( c=='\n' \|\| c=='\r' ){
395	n += 2;
396	}else{
397	n += 6;
398	}
399	}else{
400	n++;
401	}
402	}
403	i = blob_size(pBlob);
404	blob_resize(pBlob, i+n);
405	zOut = blob_buffer(pBlob);
406	z = (const unsigned char*)zStr;
407	while( (c = fossil_utf8_read(&z))!=0 ){
408	if( c=='\\' ){
409	zOut[i++] = '\\';
410	zOut[i++] = c;
411	}else if( c<' ' \|\| c>=0x7f ){
412	zOut[i++] = '\\';
413	if( c=='\n' ){
414	zOut[i++] = 'n';
415	}else if( c=='\r' ){
416	zOut[i++] = 'r';
417	}else{
418	zOut[i++] = 'u';
419	for(j=3; j>=0; j--){
420	zOut[i+j] = "0123456789abcdef"[c&0xf];
421	c >>= 4;
422	}
423	i += 4;
424	}
425	}else{
426	zOut[i++] = c;
427	}
428	}
429	zOut[i] = 0;
430	}
431
432	/*
433	** The characters used for HTTP base64 encoding.
434	*/
435	static unsigned char zBase[] =
436

M src/event.c

+1 -1

		--- src/event.c
		+++ src/event.c
		@@ -384,11 +384,11 @@
384	384	rid = db_int(0,
385	385	"SELECT rid FROM tagxref"
386	386	" WHERE tagid=(SELECT tagid FROM tag WHERE tagname GLOB '%q*')"
387	387	" ORDER BY mtime DESC", zTag
388	388	);
389		- if( rid && strlen(zId)<40 ){
	389	+ if( rid && strlen(zId)<HNAME_MIN ){
390	390	zId = db_text(0,
391	391	"SELECT substr(tagname,7) FROM tag WHERE tagname GLOB '%q*'",
392	392	zTag
393	393	);
394	394	}
395	395

	--- src/event.c
	+++ src/event.c
	@@ -384,11 +384,11 @@
384	rid = db_int(0,
385	"SELECT rid FROM tagxref"
386	" WHERE tagid=(SELECT tagid FROM tag WHERE tagname GLOB '%q*')"
387	" ORDER BY mtime DESC", zTag
388	);
389	if( rid && strlen(zId)<40 ){
390	zId = db_text(0,
391	"SELECT substr(tagname,7) FROM tag WHERE tagname GLOB '%q*'",
392	zTag
393	);
394	}
395

	--- src/event.c
	+++ src/event.c
	@@ -384,11 +384,11 @@
384	rid = db_int(0,
385	"SELECT rid FROM tagxref"
386	" WHERE tagid=(SELECT tagid FROM tag WHERE tagname GLOB '%q*')"
387	" ORDER BY mtime DESC", zTag
388	);
389	if( rid && strlen(zId)<HNAME_MIN ){
390	zId = db_text(0,
391	"SELECT substr(tagname,7) FROM tag WHERE tagname GLOB '%q*'",
392	zTag
393	);
394	}
395

M src/export.c

+1 -1

		--- src/export.c
		+++ src/export.c
		@@ -261,11 +261,11 @@
261	261	cur_tok = strtok(NULL, " \t");
262	262	if( !cur_tok ){
263	263	/* This mark was generated by an older version of Fossil and doesn't
264	264	** include the mark name and uuid. create_mark() will name the new mark
265	265	** exactly as it was when exported to git, so that we should have a
266		- ** valid mapping from git sha1<->mark name<->fossil sha1. */
	266	+ ** valid mapping from git hash<->mark name<->fossil hash. */
267	267	unsigned int mid;
268	268	if( type_=='c' ){
269	269	mid = COMMITMARK(mark->rid);
270	270	}
271	271	else{
272	272

	--- src/export.c
	+++ src/export.c
	@@ -261,11 +261,11 @@
261	cur_tok = strtok(NULL, " \t");
262	if( !cur_tok ){
263	/* This mark was generated by an older version of Fossil and doesn't
264	** include the mark name and uuid. create_mark() will name the new mark
265	** exactly as it was when exported to git, so that we should have a
266	** valid mapping from git sha1<->mark name<->fossil sha1. */
267	unsigned int mid;
268	if( type_=='c' ){
269	mid = COMMITMARK(mark->rid);
270	}
271	else{
272

	--- src/export.c
	+++ src/export.c
	@@ -261,11 +261,11 @@
261	cur_tok = strtok(NULL, " \t");
262	if( !cur_tok ){
263	/* This mark was generated by an older version of Fossil and doesn't
264	** include the mark name and uuid. create_mark() will name the new mark
265	** exactly as it was when exported to git, so that we should have a
266	** valid mapping from git hash<->mark name<->fossil hash. */
267	unsigned int mid;
268	if( type_=='c' ){
269	mid = COMMITMARK(mark->rid);
270	}
271	else{
272

M src/foci.c

+2 -2

		--- src/foci.c
		+++ src/foci.c
		@@ -28,11 +28,11 @@
28	28	** The "schema" for the temp.foci table is:
29	29	**
30	30	** CREATE TABLE files_of_checkin(
31	31	** checkinID INTEGER, -- RID for the check-in manifest
32	32	** filename TEXT, -- Name of a file
33		-** uuid TEXT, -- SHA1 hash of the file
	33	+** uuid TEXT, -- hash of the file
34	34	** previousName TEXT, -- Name of the file in previous check-in
35	35	** perm TEXT, -- Permissions on the file
36	36	** symname TEXT HIDDEN -- Symbolic name of the check-in.
37	37	** );
38	38	**
		@@ -54,11 +54,11 @@
54	54	*/
55	55	static const char zFociSchema[] =
56	56	@ CREATE TABLE files_of_checkin(
57	57	@ checkinID INTEGER, -- RID for the check-in manifest
58	58	@ filename TEXT, -- Name of a file
59		-@ uuid TEXT, -- SHA1 hash of the file
	59	+@ uuid TEXT, -- hash of the file
60	60	@ previousName TEXT, -- Name of the file in previous check-in
61	61	@ perm TEXT, -- Permissions on the file
62	62	@ symname TEXT HIDDEN -- Symbolic name of the check-in
63	63	@ );
64	64	;
65	65

	--- src/foci.c
	+++ src/foci.c
	@@ -28,11 +28,11 @@
28	** The "schema" for the temp.foci table is:
29	**
30	** CREATE TABLE files_of_checkin(
31	** checkinID INTEGER, -- RID for the check-in manifest
32	** filename TEXT, -- Name of a file
33	** uuid TEXT, -- SHA1 hash of the file
34	** previousName TEXT, -- Name of the file in previous check-in
35	** perm TEXT, -- Permissions on the file
36	** symname TEXT HIDDEN -- Symbolic name of the check-in.
37	** );
38	**
	@@ -54,11 +54,11 @@
54	*/
55	static const char zFociSchema[] =
56	@ CREATE TABLE files_of_checkin(
57	@ checkinID INTEGER, -- RID for the check-in manifest
58	@ filename TEXT, -- Name of a file
59	@ uuid TEXT, -- SHA1 hash of the file
60	@ previousName TEXT, -- Name of the file in previous check-in
61	@ perm TEXT, -- Permissions on the file
62	@ symname TEXT HIDDEN -- Symbolic name of the check-in
63	@ );
64	;
65

	--- src/foci.c
	+++ src/foci.c
	@@ -28,11 +28,11 @@
28	** The "schema" for the temp.foci table is:
29	**
30	** CREATE TABLE files_of_checkin(
31	** checkinID INTEGER, -- RID for the check-in manifest
32	** filename TEXT, -- Name of a file
33	** uuid TEXT, -- hash of the file
34	** previousName TEXT, -- Name of the file in previous check-in
35	** perm TEXT, -- Permissions on the file
36	** symname TEXT HIDDEN -- Symbolic name of the check-in.
37	** );
38	**
	@@ -54,11 +54,11 @@
54	*/
55	static const char zFociSchema[] =
56	@ CREATE TABLE files_of_checkin(
57	@ checkinID INTEGER, -- RID for the check-in manifest
58	@ filename TEXT, -- Name of a file
59	@ uuid TEXT, -- hash of the file
60	@ previousName TEXT, -- Name of the file in previous check-in
61	@ perm TEXT, -- Permissions on the file
62	@ symname TEXT HIDDEN -- Symbolic name of the check-in
63	@ );
64	;
65

M src/fusefs.c

+1 -1

		--- src/fusefs.c
		+++ src/fusefs.c
		@@ -294,11 +294,11 @@
294	294	** This command uses the Fuse Filesystem (FuseFS) to mount a directory
295	295	** at DIRECTORY that contains the content of all check-ins in the
296	296	** repository. The names of files are DIRECTORY/checkins/VERSION/PATH
297	297	** where DIRECTORY is the root of the mount, VERSION is any valid
298	298	** check-in name (examples: "trunk" or "tip" or a tag or any unique
299		-** prefix of a SHA1 hash, etc) and PATH is the pathname of the file in
	299	+** prefix of an artifact hash, etc) and PATH is the pathname of the file in
300	300	** the check-in. If DIRECTORY does not exist, then an attempt is made
301	301	** to create it.
302	302	**
303	303	** The DIRECTORY/checkins directory is not searchable so one cannot
304	304	** do "ls DIRECTORY/checkins" to get a listing of all possible check-in
305	305

	--- src/fusefs.c
	+++ src/fusefs.c
	@@ -294,11 +294,11 @@
294	** This command uses the Fuse Filesystem (FuseFS) to mount a directory
295	** at DIRECTORY that contains the content of all check-ins in the
296	** repository. The names of files are DIRECTORY/checkins/VERSION/PATH
297	** where DIRECTORY is the root of the mount, VERSION is any valid
298	** check-in name (examples: "trunk" or "tip" or a tag or any unique
299	** prefix of a SHA1 hash, etc) and PATH is the pathname of the file in
300	** the check-in. If DIRECTORY does not exist, then an attempt is made
301	** to create it.
302	**
303	** The DIRECTORY/checkins directory is not searchable so one cannot
304	** do "ls DIRECTORY/checkins" to get a listing of all possible check-in
305

	--- src/fusefs.c
	+++ src/fusefs.c
	@@ -294,11 +294,11 @@
294	** This command uses the Fuse Filesystem (FuseFS) to mount a directory
295	** at DIRECTORY that contains the content of all check-ins in the
296	** repository. The names of files are DIRECTORY/checkins/VERSION/PATH
297	** where DIRECTORY is the root of the mount, VERSION is any valid
298	** check-in name (examples: "trunk" or "tip" or a tag or any unique
299	** prefix of an artifact hash, etc) and PATH is the pathname of the file in
300	** the check-in. If DIRECTORY does not exist, then an attempt is made
301	** to create it.
302	**
303	** The DIRECTORY/checkins directory is not searchable so one cannot
304	** do "ls DIRECTORY/checkins" to get a listing of all possible check-in
305

M src/graph.c

+1 -1

		--- src/graph.c
		+++ src/graph.c
		@@ -179,11 +179,11 @@
179	179	int rid, /* RID for the check-in */
180	180	int nParent, /* Number of parents */
181	181	int aParent, / Array of parents */
182	182	const char zBranch, / Branch for this check-in */
183	183	const char zBgClr, / Background color. NULL or "" for white. */
184		- const char zUuid, / SHA1 hash of the object being graphed */
	184	+ const char zUuid, / hash name of the object being graphed */
185	185	int isLeaf /* True if this row is a leaf */
186	186	){
187	187	GraphRow *pRow;
188	188	int nByte;
189	189
190	190
191	191	ADDED src/hname.c

	--- src/graph.c
	+++ src/graph.c
	@@ -179,11 +179,11 @@
179	int rid, /* RID for the check-in */
180	int nParent, /* Number of parents */
181	int aParent, / Array of parents */
182	const char zBranch, / Branch for this check-in */
183	const char zBgClr, / Background color. NULL or "" for white. */
184	const char zUuid, / SHA1 hash of the object being graphed */
185	int isLeaf /* True if this row is a leaf */
186	){
187	GraphRow *pRow;
188	int nByte;
189
190
191	DDED src/hname.c

	--- src/graph.c
	+++ src/graph.c
	@@ -179,11 +179,11 @@
179	int rid, /* RID for the check-in */
180	int nParent, /* Number of parents */
181	int aParent, / Array of parents */
182	const char zBranch, / Branch for this check-in */
183	const char zBgClr, / Background color. NULL or "" for white. */
184	const char zUuid, / hash name of the object being graphed */
185	int isLeaf /* True if this row is a leaf */
186	){
187	GraphRow *pRow;
188	int nByte;
189
190
191	DDED src/hname.c

M src/hname.c

+11

		--- a/src/hname.c
		+++ b/src/hname.c
		@@ -0,0 +1,11 @@
	1	+/*
	2	+** Copyright (c) 2017 D. Richard Hipp
	3	+**
	4	+** This program is free software; you can redistribute it and/or
	5	+** modify it under the termoftware; you can redistribute it and/or
	6	+** modify it under the term/*
	7	+** Copyright (c) 2017 D. Richard Hipp
	8	+**
	9	+** This program is free software; you can redistribute it and/or
	10	+** modify it under the termsha1sum_file(zFileBlob hash;
	11	+

	--- a/src/hname.c
	+++ b/src/hname.c
	@@ -0,0 +1,11 @@

	--- a/src/hname.c
	+++ b/src/hname.c
	@@ -0,0 +1,11 @@
1	/*
2	** Copyright (c) 2017 D. Richard Hipp
3	**
4	** This program is free software; you can redistribute it and/or
5	** modify it under the termoftware; you can redistribute it and/or
6	** modify it under the term/*
7	** Copyright (c) 2017 D. Richard Hipp
8	**
9	** This program is free software; you can redistribute it and/or
10	** modify it under the termsha1sum_file(zFileBlob hash;
11

M src/http_ssl.c

+5 -6

		--- src/http_ssl.c
		+++ src/http_ssl.c
		@@ -252,14 +252,12 @@
252	252	hasSavedCertificate = 1;
253	253	}
254	254
255	255	if( pUrlData->useProxy ){
256	256	int rc;
257		- BIO *sBio;
258		- char *connStr;
259		- connStr = mprintf("%s:%d", g.url.name, pUrlData->port);
260		- sBio = BIO_new_connect(connStr);
	257	+ char *connStr = mprintf("%s:%d", g.url.name, pUrlData->port);
	258	+ BIO *sBio = BIO_new_connect(connStr);
261	259	free(connStr);
262	260	if( BIO_do_connect(sBio)<=0 ){
263	261	ssl_set_errmsg("SSL: cannot connect to proxy %s:%d (%s)",
264	262	pUrlData->name, pUrlData->port, ERR_reason_error_string(ERR_get_error()));
265	263	ssl_close();
		@@ -293,12 +291,13 @@
293	291	#endif
294	292
295	293	SSL_set_mode(ssl, SSL_MODE_AUTO_RETRY);
296	294
297	295	if( !pUrlData->useProxy ){
298		- BIO_set_conn_hostname(iBio, pUrlData->name);
299		- BIO_ctrl(iBio,BIO_C_SET_CONNECT,3,(char *)&pUrlData->port);
	296	+ char *connStr = mprintf("%s:%d", pUrlData->name, pUrlData->port);
	297	+ BIO_set_conn_hostname(iBio, connStr);
	298	+ free(connStr);
300	299	if( BIO_do_connect(iBio)<=0 ){
301	300	ssl_set_errmsg("SSL: cannot connect to host %s:%d (%s)",
302	301	pUrlData->name, pUrlData->port, ERR_reason_error_string(ERR_get_error()));
303	302	ssl_close();
304	303	return 1;
305	304

	--- src/http_ssl.c
	+++ src/http_ssl.c
	@@ -252,14 +252,12 @@
252	hasSavedCertificate = 1;
253	}
254
255	if( pUrlData->useProxy ){
256	int rc;
257	BIO *sBio;
258	char *connStr;
259	connStr = mprintf("%s:%d", g.url.name, pUrlData->port);
260	sBio = BIO_new_connect(connStr);
261	free(connStr);
262	if( BIO_do_connect(sBio)<=0 ){
263	ssl_set_errmsg("SSL: cannot connect to proxy %s:%d (%s)",
264	pUrlData->name, pUrlData->port, ERR_reason_error_string(ERR_get_error()));
265	ssl_close();
	@@ -293,12 +291,13 @@
293	#endif
294
295	SSL_set_mode(ssl, SSL_MODE_AUTO_RETRY);
296
297	if( !pUrlData->useProxy ){
298	BIO_set_conn_hostname(iBio, pUrlData->name);
299	BIO_ctrl(iBio,BIO_C_SET_CONNECT,3,(char *)&pUrlData->port);

300	if( BIO_do_connect(iBio)<=0 ){
301	ssl_set_errmsg("SSL: cannot connect to host %s:%d (%s)",
302	pUrlData->name, pUrlData->port, ERR_reason_error_string(ERR_get_error()));
303	ssl_close();
304	return 1;
305

	--- src/http_ssl.c
	+++ src/http_ssl.c
	@@ -252,14 +252,12 @@
252	hasSavedCertificate = 1;
253	}
254
255	if( pUrlData->useProxy ){
256	int rc;
257	char *connStr = mprintf("%s:%d", g.url.name, pUrlData->port);
258	BIO *sBio = BIO_new_connect(connStr);


259	free(connStr);
260	if( BIO_do_connect(sBio)<=0 ){
261	ssl_set_errmsg("SSL: cannot connect to proxy %s:%d (%s)",
262	pUrlData->name, pUrlData->port, ERR_reason_error_string(ERR_get_error()));
263	ssl_close();
	@@ -293,12 +291,13 @@
291	#endif
292
293	SSL_set_mode(ssl, SSL_MODE_AUTO_RETRY);
294
295	if( !pUrlData->useProxy ){
296	char *connStr = mprintf("%s:%d", pUrlData->name, pUrlData->port);
297	BIO_set_conn_hostname(iBio, connStr);
298	free(connStr);
299	if( BIO_do_connect(iBio)<=0 ){
300	ssl_set_errmsg("SSL: cannot connect to host %s:%d (%s)",
301	pUrlData->name, pUrlData->port, ERR_reason_error_string(ERR_get_error()));
302	ssl_close();
303	return 1;
304

M src/import.c

+3 -3

		--- src/import.c
		+++ src/import.c
		@@ -149,18 +149,18 @@
149	149	** UUID in gg.zPrevCheckin.
150	150	*/
151	151	static int fast_insert_content(
152	152	Blob pContent, / Content to insert */
153	153	const char zMark, / Label using this mark, if not NULL */
154		- int saveUuid, /* Save SHA1 hash in gg.zPrevCheckin */
	154	+ int saveUuid, /* Save artifact hash in gg.zPrevCheckin */
155	155	int doParse /* Invoke manifest_crosslink() */
156	156	){
157	157	Blob hash;
158	158	Blob cmpr;
159	159	int rid;
160	160
161		- sha1sum_blob(pContent, &hash);
	161	+ hname_hash(pContent, 0, &hash);
162	162	rid = db_int(0, "SELECT rid FROM blob WHERE uuid=%B", &hash);
163	163	if( rid==0 ){
164	164	static Stmt ins;
165	165	db_static_prepare(&ins,
166	166	"INSERT INTO blob(uuid, size, content) VALUES(:uuid, :size, :content)"
		@@ -1652,11 +1652,11 @@
1652	1652	*/
1653	1653	const char *zIgnTree;
1654	1654	unsigned nIgnTree = 0;
1655	1655	while( (zIgnTree = find_option("ignore-tree", 0, 1)) ){
1656	1656	if ( *zIgnTree ){
1657		- gsvn.azIgnTree = fossil_realloc(gsvn.azIgnTree,
	1657	+ gsvn.azIgnTree = fossil_realloc((void *)gsvn.azIgnTree,
1658	1658	sizeof(gsvn.azIgnTree) (nIgnTree + 2));
1659	1659	gsvn.azIgnTree[nIgnTree++] = zIgnTree;
1660	1660	gsvn.azIgnTree[nIgnTree] = 0;
1661	1661	}
1662	1662	}
1663	1663

	--- src/import.c
	+++ src/import.c
	@@ -149,18 +149,18 @@
149	** UUID in gg.zPrevCheckin.
150	*/
151	static int fast_insert_content(
152	Blob pContent, / Content to insert */
153	const char zMark, / Label using this mark, if not NULL */
154	int saveUuid, /* Save SHA1 hash in gg.zPrevCheckin */
155	int doParse /* Invoke manifest_crosslink() */
156	){
157	Blob hash;
158	Blob cmpr;
159	int rid;
160
161	sha1sum_blob(pContent, &hash);
162	rid = db_int(0, "SELECT rid FROM blob WHERE uuid=%B", &hash);
163	if( rid==0 ){
164	static Stmt ins;
165	db_static_prepare(&ins,
166	"INSERT INTO blob(uuid, size, content) VALUES(:uuid, :size, :content)"
	@@ -1652,11 +1652,11 @@
1652	*/
1653	const char *zIgnTree;
1654	unsigned nIgnTree = 0;
1655	while( (zIgnTree = find_option("ignore-tree", 0, 1)) ){
1656	if ( *zIgnTree ){
1657	gsvn.azIgnTree = fossil_realloc(gsvn.azIgnTree,
1658	sizeof(gsvn.azIgnTree) (nIgnTree + 2));
1659	gsvn.azIgnTree[nIgnTree++] = zIgnTree;
1660	gsvn.azIgnTree[nIgnTree] = 0;
1661	}
1662	}
1663

	--- src/import.c
	+++ src/import.c
	@@ -149,18 +149,18 @@
149	** UUID in gg.zPrevCheckin.
150	*/
151	static int fast_insert_content(
152	Blob pContent, / Content to insert */
153	const char zMark, / Label using this mark, if not NULL */
154	int saveUuid, /* Save artifact hash in gg.zPrevCheckin */
155	int doParse /* Invoke manifest_crosslink() */
156	){
157	Blob hash;
158	Blob cmpr;
159	int rid;
160
161	hname_hash(pContent, 0, &hash);
162	rid = db_int(0, "SELECT rid FROM blob WHERE uuid=%B", &hash);
163	if( rid==0 ){
164	static Stmt ins;
165	db_static_prepare(&ins,
166	"INSERT INTO blob(uuid, size, content) VALUES(:uuid, :size, :content)"
	@@ -1652,11 +1652,11 @@
1652	*/
1653	const char *zIgnTree;
1654	unsigned nIgnTree = 0;
1655	while( (zIgnTree = find_option("ignore-tree", 0, 1)) ){
1656	if ( *zIgnTree ){
1657	gsvn.azIgnTree = fossil_realloc((void *)gsvn.azIgnTree,
1658	sizeof(gsvn.azIgnTree) (nIgnTree + 2));
1659	gsvn.azIgnTree[nIgnTree++] = zIgnTree;
1660	gsvn.azIgnTree[nIgnTree] = 0;
1661	}
1662	}
1663

M src/info.c

+7 -5

		--- src/info.c
		+++ src/info.c
		@@ -68,11 +68,11 @@
68	68	zDate = db_text(0,
69	69	"SELECT datetime(mtime) \|\| ' UTC' FROM event WHERE objid=%d",
70	70	rid
71	71	);
72	72	/* 01234567890123 */
73		- fossil_print("%-13s %s %s\n", zUuidName, zUuid, zDate ? zDate : "");
	73	+ fossil_print("%-13s %.40s %s\n", zUuidName, zUuid, zDate ? zDate : "");
74	74	free(zDate);
75	75	if( showComment ){
76	76	zComment = db_text(0,
77	77	"SELECT coalesce(ecomment,comment) \|\| "
78	78	" ' (user: ' \|\| coalesce(euser,user,'?') \|\| ')' "
		@@ -91,11 +91,11 @@
91	91	const char *zType = db_column_int(&q, 2) ? "parent:" : "merged-from:";
92	92	zDate = db_text("",
93	93	"SELECT datetime(mtime) \|\| ' UTC' FROM event WHERE objid=%d",
94	94	db_column_int(&q, 1)
95	95	);
96		- fossil_print("%-13s %s %s\n", zType, zUuid, zDate);
	96	+ fossil_print("%-13s %.40s %s\n", zType, zUuid, zDate);
97	97	free(zDate);
98	98	}
99	99	db_finalize(&q);
100	100	db_prepare(&q, "SELECT uuid, cid, isprim FROM plink JOIN blob ON cid=rid "
101	101	" WHERE pid=%d"
		@@ -105,11 +105,11 @@
105	105	const char *zType = db_column_int(&q, 2) ? "child:" : "merged-into:";
106	106	zDate = db_text("",
107	107	"SELECT datetime(mtime) \|\| ' UTC' FROM event WHERE objid=%d",
108	108	db_column_int(&q, 1)
109	109	);
110		- fossil_print("%-13s %s %s\n", zType, zUuid, zDate);
	110	+ fossil_print("%-13s %.40s %s\n", zType, zUuid, zDate);
111	111	free(zDate);
112	112	}
113	113	db_finalize(&q);
114	114	}
115	115	zTags = info_tags_of_checkin(rid, 0);
		@@ -572,10 +572,11 @@
572	572	rid, rid
573	573	);
574	574	sideBySide = !is_false(PD("sbs","1"));
575	575	if( db_step(&q1)==SQLITE_ROW ){
576	576	const char *zUuid = db_column_text(&q1, 0);
	577	+ int nUuid = db_column_bytes(&q1, 0);
577	578	char zEUser, zEComment;
578	579	const char *zUser;
579	580	const char *zComment;
580	581	const char *zDate;
581	582	const char *zOrigDate;
		@@ -593,11 +594,11 @@
593	594	zComment = db_column_text(&q1, 3);
594	595	zDate = db_column_text(&q1,1);
595	596	zOrigDate = db_column_text(&q1, 4);
596	597	@ <div class="section">Overview</div>
597	598	@ <table class="label-value">
598		- @ <tr><th>SHA1 Hash:</th><td>%s(zUuid)
	599	+ @ <tr><th>%s(hname_alg(nUuid)):</th><td>%s(zUuid)
599	600	if( g.perm.Setup ){
600	601	@ (Record ID: %d(rid))
601	602	}
602	603	@ </td></tr>
603	604	@ <tr><th>Date:</th><td>
		@@ -1385,10 +1386,11 @@
1385	1386	" ORDER BY mtime DESC /sort/",
1386	1387	rid
1387	1388	);
1388	1389	while( db_step(&q)==SQLITE_ROW ){
1389	1390	const char *zTarget = db_column_text(&q, 0);
	1391	+ int nTarget = db_column_bytes(&q, 0);
1390	1392	const char *zFilename = db_column_text(&q, 1);
1391	1393	const char *zDate = db_column_text(&q, 2);
1392	1394	const char *zUser = db_column_text(&q, 3);
1393	1395	/* const char zSrc = db_column_text(&q, 4); /
1394	1396	if( cnt>0 ){
		@@ -1395,11 +1397,11 @@
1395	1397	@ Also attachment "%h(zFilename)" to
1396	1398	}else{
1397	1399	@ Attachment "%h(zFilename)" to
1398	1400	}
1399	1401	objType \|= OBJTYPE_ATTACHMENT;
1400		- if( strlen(zTarget)==UUID_SIZE && validate16(zTarget,UUID_SIZE) ){
	1402	+ if( nTarget==UUID_SIZE && validate16(zTarget,UUID_SIZE) ){
1401	1403	if ( db_exists("SELECT 1 FROM tag WHERE tagname='tkt-%q'",
1402	1404	zTarget)
1403	1405	){
1404	1406	if( g.perm.Hyperlink && g.anon.RdTkt ){
1405	1407	@ ticket [%z(href("%R/tktview?name=%!S",zTarget))%S(zTarget)</a>]
1406	1408

	--- src/info.c
	+++ src/info.c
	@@ -68,11 +68,11 @@
68	zDate = db_text(0,
69	"SELECT datetime(mtime) \|\| ' UTC' FROM event WHERE objid=%d",
70	rid
71	);
72	/* 01234567890123 */
73	fossil_print("%-13s %s %s\n", zUuidName, zUuid, zDate ? zDate : "");
74	free(zDate);
75	if( showComment ){
76	zComment = db_text(0,
77	"SELECT coalesce(ecomment,comment) \|\| "
78	" ' (user: ' \|\| coalesce(euser,user,'?') \|\| ')' "
	@@ -91,11 +91,11 @@
91	const char *zType = db_column_int(&q, 2) ? "parent:" : "merged-from:";
92	zDate = db_text("",
93	"SELECT datetime(mtime) \|\| ' UTC' FROM event WHERE objid=%d",
94	db_column_int(&q, 1)
95	);
96	fossil_print("%-13s %s %s\n", zType, zUuid, zDate);
97	free(zDate);
98	}
99	db_finalize(&q);
100	db_prepare(&q, "SELECT uuid, cid, isprim FROM plink JOIN blob ON cid=rid "
101	" WHERE pid=%d"
	@@ -105,11 +105,11 @@
105	const char *zType = db_column_int(&q, 2) ? "child:" : "merged-into:";
106	zDate = db_text("",
107	"SELECT datetime(mtime) \|\| ' UTC' FROM event WHERE objid=%d",
108	db_column_int(&q, 1)
109	);
110	fossil_print("%-13s %s %s\n", zType, zUuid, zDate);
111	free(zDate);
112	}
113	db_finalize(&q);
114	}
115	zTags = info_tags_of_checkin(rid, 0);
	@@ -572,10 +572,11 @@
572	rid, rid
573	);
574	sideBySide = !is_false(PD("sbs","1"));
575	if( db_step(&q1)==SQLITE_ROW ){
576	const char *zUuid = db_column_text(&q1, 0);

577	char zEUser, zEComment;
578	const char *zUser;
579	const char *zComment;
580	const char *zDate;
581	const char *zOrigDate;
	@@ -593,11 +594,11 @@
593	zComment = db_column_text(&q1, 3);
594	zDate = db_column_text(&q1,1);
595	zOrigDate = db_column_text(&q1, 4);
596	@ <div class="section">Overview</div>
597	@ <table class="label-value">
598	@ <tr><th>SHA1 Hash:</th><td>%s(zUuid)
599	if( g.perm.Setup ){
600	@ (Record ID: %d(rid))
601	}
602	@ </td></tr>
603	@ <tr><th>Date:</th><td>
	@@ -1385,10 +1386,11 @@
1385	" ORDER BY mtime DESC /sort/",
1386	rid
1387	);
1388	while( db_step(&q)==SQLITE_ROW ){
1389	const char *zTarget = db_column_text(&q, 0);

1390	const char *zFilename = db_column_text(&q, 1);
1391	const char *zDate = db_column_text(&q, 2);
1392	const char *zUser = db_column_text(&q, 3);
1393	/* const char zSrc = db_column_text(&q, 4); /
1394	if( cnt>0 ){
	@@ -1395,11 +1397,11 @@
1395	@ Also attachment "%h(zFilename)" to
1396	}else{
1397	@ Attachment "%h(zFilename)" to
1398	}
1399	objType \|= OBJTYPE_ATTACHMENT;
1400	if( strlen(zTarget)==UUID_SIZE && validate16(zTarget,UUID_SIZE) ){
1401	if ( db_exists("SELECT 1 FROM tag WHERE tagname='tkt-%q'",
1402	zTarget)
1403	){
1404	if( g.perm.Hyperlink && g.anon.RdTkt ){
1405	@ ticket [%z(href("%R/tktview?name=%!S",zTarget))%S(zTarget)</a>]
1406

	--- src/info.c
	+++ src/info.c
	@@ -68,11 +68,11 @@
68	zDate = db_text(0,
69	"SELECT datetime(mtime) \|\| ' UTC' FROM event WHERE objid=%d",
70	rid
71	);
72	/* 01234567890123 */
73	fossil_print("%-13s %.40s %s\n", zUuidName, zUuid, zDate ? zDate : "");
74	free(zDate);
75	if( showComment ){
76	zComment = db_text(0,
77	"SELECT coalesce(ecomment,comment) \|\| "
78	" ' (user: ' \|\| coalesce(euser,user,'?') \|\| ')' "
	@@ -91,11 +91,11 @@
91	const char *zType = db_column_int(&q, 2) ? "parent:" : "merged-from:";
92	zDate = db_text("",
93	"SELECT datetime(mtime) \|\| ' UTC' FROM event WHERE objid=%d",
94	db_column_int(&q, 1)
95	);
96	fossil_print("%-13s %.40s %s\n", zType, zUuid, zDate);
97	free(zDate);
98	}
99	db_finalize(&q);
100	db_prepare(&q, "SELECT uuid, cid, isprim FROM plink JOIN blob ON cid=rid "
101	" WHERE pid=%d"
	@@ -105,11 +105,11 @@
105	const char *zType = db_column_int(&q, 2) ? "child:" : "merged-into:";
106	zDate = db_text("",
107	"SELECT datetime(mtime) \|\| ' UTC' FROM event WHERE objid=%d",
108	db_column_int(&q, 1)
109	);
110	fossil_print("%-13s %.40s %s\n", zType, zUuid, zDate);
111	free(zDate);
112	}
113	db_finalize(&q);
114	}
115	zTags = info_tags_of_checkin(rid, 0);
	@@ -572,10 +572,11 @@
572	rid, rid
573	);
574	sideBySide = !is_false(PD("sbs","1"));
575	if( db_step(&q1)==SQLITE_ROW ){
576	const char *zUuid = db_column_text(&q1, 0);
577	int nUuid = db_column_bytes(&q1, 0);
578	char zEUser, zEComment;
579	const char *zUser;
580	const char *zComment;
581	const char *zDate;
582	const char *zOrigDate;
	@@ -593,11 +594,11 @@
594	zComment = db_column_text(&q1, 3);
595	zDate = db_column_text(&q1,1);
596	zOrigDate = db_column_text(&q1, 4);
597	@ <div class="section">Overview</div>
598	@ <table class="label-value">
599	@ <tr><th>%s(hname_alg(nUuid)):</th><td>%s(zUuid)
600	if( g.perm.Setup ){
601	@ (Record ID: %d(rid))
602	}
603	@ </td></tr>
604	@ <tr><th>Date:</th><td>
	@@ -1385,10 +1386,11 @@
1386	" ORDER BY mtime DESC /sort/",
1387	rid
1388	);
1389	while( db_step(&q)==SQLITE_ROW ){
1390	const char *zTarget = db_column_text(&q, 0);
1391	int nTarget = db_column_bytes(&q, 0);
1392	const char *zFilename = db_column_text(&q, 1);
1393	const char *zDate = db_column_text(&q, 2);
1394	const char *zUser = db_column_text(&q, 3);
1395	/* const char zSrc = db_column_text(&q, 4); /
1396	if( cnt>0 ){
	@@ -1395,11 +1397,11 @@
1397	@ Also attachment "%h(zFilename)" to
1398	}else{
1399	@ Attachment "%h(zFilename)" to
1400	}
1401	objType \|= OBJTYPE_ATTACHMENT;
1402	if( nTarget==UUID_SIZE && validate16(zTarget,UUID_SIZE) ){
1403	if ( db_exists("SELECT 1 FROM tag WHERE tagname='tkt-%q'",
1404	zTarget)
1405	){
1406	if( g.perm.Hyperlink && g.anon.RdTkt ){
1407	@ ticket [%z(href("%R/tktview?name=%!S",zTarget))%S(zTarget)</a>]
1408

M src/login.c

+97

		--- src/login.c
		+++ src/login.c
		@@ -49,10 +49,48 @@
49	49	# define sleep Sleep /* windows does not have sleep, but Sleep */
50	50	# endif
51	51	#endif
52	52	#include <time.h>
53	53
	54	+#if defined(__MINGW32__) \|\| defined(_MSC_VER)
	55	+/*
	56	+** MinGW doesn't have strtok_r in its libc. Here's a public domain one
	57	+** found at StackOverflow as a work-around, with formatting adjusted to
	58	+** make it more like the usual style here. This is certainly the wrong
	59	+** place for it, which is emphasized by making the function static.
	60	+**
	61	+** See http://stackoverflow.com/a/12979321/68204
	62	+**
	63	+** public domain strtok_r() by Charlie Gordon
	64	+** from comp.lang.c 9/14/2007
	65	+** http://groups.google.com/group/comp.lang.c/msg/2ab1ecbb86646684
	66	+** (Declaration that it's public domain):
	67	+** http://groups.google.com/group/comp.lang.c/msg/7c7b39328fefab9c
	68	+*/
	69	+static char* strtok_r(
	70	+ char *str,
	71	+ const char *delim,
	72	+ char **nextp
	73	+){
	74	+ char *ret;
	75	+ if( str == NULL ){
	76	+ str = *nextp;
	77	+ }
	78	+ str += strspn(str, delim);
	79	+ if( *str == '\0' ){
	80	+ return NULL;
	81	+ }
	82	+ ret = str;
	83	+ str += strcspn(str, delim);
	84	+ if( *str ){
	85	+ *str++ = '\0';
	86	+ }
	87	+ *nextp = str;
	88	+ return ret;
	89	+}
	90	+#endif
	91	+
54	92	/*
55	93	** Return the login-group name. Or return 0 if this repository is
56	94	** not a member of a login-group.
57	95	*/
58	96	const char *login_group_name(void){
		@@ -966,10 +1004,69 @@
966	1004	if( zRemoteUser && db_get_boolean("remote_user_ok",0) ){
967	1005	uid = db_int(0, "SELECT uid FROM user WHERE login=%Q"
968	1006	" AND length(cap)>0 AND length(pw)>0", zRemoteUser);
969	1007	}
970	1008	}
	1009	+
	1010	+ /* If the request didn't provide a login cookie or the login cookie didn't
	1011	+ ** match a known valid user, check the HTTP "Authorization" header and
	1012	+ ** see if those credentials are valid for a known user.
	1013	+ */
	1014	+ if( uid==0 ){
	1015	+ const char *zHTTPAuth = PD("HTTP_AUTHORIZATION", 0);
	1016	+
	1017	+ /* Check to see if the HTTP "Authorization" header is present
	1018	+ */
	1019	+ if( zHTTPAuth!=0 && zHTTPAuth[0]!=0
	1020	+ && db_get_boolean("http_authentication_ok",0)
	1021	+ ){
	1022	+ char *zBuf = fossil_strdup(zHTTPAuth);
	1023	+
	1024	+ if( zBuf!=0 ){
	1025	+ char *zPos;
	1026	+ char *zTok = strtok_r(zBuf, " ", &zPos);
	1027	+
	1028	+ if( zTok != 0 ){
	1029	+ /* Check to see if the authorization scheme is HTTP
	1030	+ ** basic auth.
	1031	+ */
	1032	+ if (strncmp(zTok, "Basic", zTok - zBuf) == 0) {
	1033	+ zTok = strtok_r(NULL, " ", &zPos);
	1034	+ int zBytesDecoded = 0;
	1035	+ char *zDecodedAuth = decode64(zTok, &zBytesDecoded);
	1036	+
	1037	+ char *zUsername = strtok_r(zDecodedAuth, ":", &zPos);
	1038	+ char *zPasswd = strtok_r(NULL, ":", &zPos);
	1039	+
	1040	+ if( zUsername!=0 && zPasswd!=0 && zPasswd[0]!=0 ){
	1041	+ /* Attempting to log in as the user provided by HTTP
	1042	+ ** basic auth
	1043	+ */
	1044	+ uid = login_search_uid(zUsername, zPasswd);
	1045	+ if( uid>0 ){
	1046	+ record_login_attempt(zUsername, zIpAddr, 1);
	1047	+ }else{
	1048	+ record_login_attempt(zUsername, zIpAddr, 0);
	1049	+
	1050	+ /* The user attempted to login specifically with HTTP basic
	1051	+ ** auth, but provided invalid credentials. Inform them of
	1052	+ ** the failed login attempt via 401.
	1053	+ */
	1054	+ cgi_set_status(401, "Unauthorized");
	1055	+ cgi_reply();
	1056	+ fossil_exit(0);
	1057	+ }
	1058	+ }
	1059	+
	1060	+ fossil_free(zDecodedAuth);
	1061	+ }
	1062	+ }
	1063	+
	1064	+ fossil_free(zBuf);
	1065	+ }
	1066	+ }
	1067	+ }
971	1068
972	1069	/* If no user found yet, try to log in as "nobody" */
973	1070	if( uid==0 ){
974	1071	uid = db_int(0, "SELECT uid FROM user WHERE login='nobody'");
975	1072	if( uid==0 ){
976	1073

	--- src/login.c
	+++ src/login.c
	@@ -49,10 +49,48 @@
49	# define sleep Sleep /* windows does not have sleep, but Sleep */
50	# endif
51	#endif
52	#include <time.h>
53






































54	/*
55	** Return the login-group name. Or return 0 if this repository is
56	** not a member of a login-group.
57	*/
58	const char *login_group_name(void){
	@@ -966,10 +1004,69 @@
966	if( zRemoteUser && db_get_boolean("remote_user_ok",0) ){
967	uid = db_int(0, "SELECT uid FROM user WHERE login=%Q"
968	" AND length(cap)>0 AND length(pw)>0", zRemoteUser);
969	}
970	}



























































971
972	/* If no user found yet, try to log in as "nobody" */
973	if( uid==0 ){
974	uid = db_int(0, "SELECT uid FROM user WHERE login='nobody'");
975	if( uid==0 ){
976

	--- src/login.c
	+++ src/login.c
	@@ -49,10 +49,48 @@
49	# define sleep Sleep /* windows does not have sleep, but Sleep */
50	# endif
51	#endif
52	#include <time.h>
53
54	#if defined(__MINGW32__) \|\| defined(_MSC_VER)
55	/*
56	** MinGW doesn't have strtok_r in its libc. Here's a public domain one
57	** found at StackOverflow as a work-around, with formatting adjusted to
58	** make it more like the usual style here. This is certainly the wrong
59	** place for it, which is emphasized by making the function static.
60	**
61	** See http://stackoverflow.com/a/12979321/68204
62	**
63	** public domain strtok_r() by Charlie Gordon
64	** from comp.lang.c 9/14/2007
65	** http://groups.google.com/group/comp.lang.c/msg/2ab1ecbb86646684
66	** (Declaration that it's public domain):
67	** http://groups.google.com/group/comp.lang.c/msg/7c7b39328fefab9c
68	*/
69	static char* strtok_r(
70	char *str,
71	const char *delim,
72	char **nextp
73	){
74	char *ret;
75	if( str == NULL ){
76	str = *nextp;
77	}
78	str += strspn(str, delim);
79	if( *str == '\0' ){
80	return NULL;
81	}
82	ret = str;
83	str += strcspn(str, delim);
84	if( *str ){
85	*str++ = '\0';
86	}
87	*nextp = str;
88	return ret;
89	}
90	#endif
91
92	/*
93	** Return the login-group name. Or return 0 if this repository is
94	** not a member of a login-group.
95	*/
96	const char *login_group_name(void){
	@@ -966,10 +1004,69 @@
1004	if( zRemoteUser && db_get_boolean("remote_user_ok",0) ){
1005	uid = db_int(0, "SELECT uid FROM user WHERE login=%Q"
1006	" AND length(cap)>0 AND length(pw)>0", zRemoteUser);
1007	}
1008	}
1009
1010	/* If the request didn't provide a login cookie or the login cookie didn't
1011	** match a known valid user, check the HTTP "Authorization" header and
1012	** see if those credentials are valid for a known user.
1013	*/
1014	if( uid==0 ){
1015	const char *zHTTPAuth = PD("HTTP_AUTHORIZATION", 0);
1016
1017	/* Check to see if the HTTP "Authorization" header is present
1018	*/
1019	if( zHTTPAuth!=0 && zHTTPAuth[0]!=0
1020	&& db_get_boolean("http_authentication_ok",0)
1021	){
1022	char *zBuf = fossil_strdup(zHTTPAuth);
1023
1024	if( zBuf!=0 ){
1025	char *zPos;
1026	char *zTok = strtok_r(zBuf, " ", &zPos);
1027
1028	if( zTok != 0 ){
1029	/* Check to see if the authorization scheme is HTTP
1030	** basic auth.
1031	*/
1032	if (strncmp(zTok, "Basic", zTok - zBuf) == 0) {
1033	zTok = strtok_r(NULL, " ", &zPos);
1034	int zBytesDecoded = 0;
1035	char *zDecodedAuth = decode64(zTok, &zBytesDecoded);
1036
1037	char *zUsername = strtok_r(zDecodedAuth, ":", &zPos);
1038	char *zPasswd = strtok_r(NULL, ":", &zPos);
1039
1040	if( zUsername!=0 && zPasswd!=0 && zPasswd[0]!=0 ){
1041	/* Attempting to log in as the user provided by HTTP
1042	** basic auth
1043	*/
1044	uid = login_search_uid(zUsername, zPasswd);
1045	if( uid>0 ){
1046	record_login_attempt(zUsername, zIpAddr, 1);
1047	}else{
1048	record_login_attempt(zUsername, zIpAddr, 0);
1049
1050	/* The user attempted to login specifically with HTTP basic
1051	** auth, but provided invalid credentials. Inform them of
1052	** the failed login attempt via 401.
1053	*/
1054	cgi_set_status(401, "Unauthorized");
1055	cgi_reply();
1056	fossil_exit(0);
1057	}
1058	}
1059
1060	fossil_free(zDecodedAuth);
1061	}
1062	}
1063
1064	fossil_free(zBuf);
1065	}
1066	}
1067	}
1068
1069	/* If no user found yet, try to log in as "nobody" */
1070	if( uid==0 ){
1071	uid = db_int(0, "SELECT uid FROM user WHERE login='nobody'");
1072	if( uid==0 ){
1073

M src/main.c

+13 -2

		--- src/main.c
		+++ src/main.c
		@@ -50,11 +50,16 @@
50	50	# include "cson_amalgamation.h" /* JSON API. */
51	51	# include "json_detail.h"
52	52	#endif
53	53
54	54	/*
55		-** Size of a UUID in characters
	55	+** Size of a UUID in characters. A UUID is a randomly generated
	56	+** lower-case hexadecimal number used to identify tickets.
	57	+**
	58	+** In Fossil 1.x, UUID also referred to a SHA1 artifact hash. But that
	59	+** usage is now obsolete. The term UUID should now mean only a very large
	60	+** random number used as a unique identifier for tickets or other objects.
56	61	*/
57	62	#define UUID_SIZE 40
58	63
59	64	/*
60	65	** Maximum number of auxiliary parameters on reports
		@@ -135,10 +140,11 @@
135	140	char zLocalDbName; / Name of the local database file */
136	141	char zOpenRevision; / Check-in version to use during database open */
137	142	int localOpen; /* True if the local database is open */
138	143	char zLocalRoot; / The directory holding the local database */
139	144	int minPrefix; /* Number of digits needed for a distinct UUID */
	145	+ int eHashPolicy; /* Current hash policy. One of HPOLICY_* */
140	146	int fNoDirSymlinks; /* True if --no-dir-symlinks flag is present */
141	147	int fSqlTrace; /* True if --sqltrace flag is present */
142	148	int fSqlStats; /* True if --sqltrace or --sqlstats are present */
143	149	int fSqlPrint; /* True if -sqlprint flag is present */
144	150	int fQuiet; /* True if -quiet flag is present */
		@@ -930,10 +936,13 @@
930	936	#if defined(FOSSIL_ENABLE_MINIZ)
931	937	blob_appendf(pOut, "miniz %s, loaded %s\n", MZ_VERSION, mz_version());
932	938	#else
933	939	blob_appendf(pOut, "zlib %s, loaded %s\n", ZLIB_VERSION, zlibVersion());
934	940	#endif
	941	+#if FOSSIL_HARDENED_SHA1
	942	+ blob_appendf(pOut, "hardened-SHA1 by Marc Stevens and Dan Shumow\n");
	943	+#endif
935	944	#if defined(FOSSIL_ENABLE_SSL)
936	945	blob_appendf(pOut, "SSL (%s)\n", SSLeay_version(SSLEAY_VERSION));
937	946	#endif
938	947	#if defined(FOSSIL_HAVE_FUSEFS)
939	948	blob_appendf(pOut, "libfuse %s, loaded %s\n", fusefs_inc_version(),
		@@ -1997,11 +2006,11 @@
1997	2006	** the name of that directory and the specific repository will be
1998	2007	** opened later by process_one_web_page() based on the content of
1999	2008	** the PATH_INFO variable.
2000	2009	**
2001	2010	** If the fCreate flag is set, then create the repository if it
2002		-** does not already exist.
	2011	+** does not already exist. Always use "auto" hash-policy in this case.
2003	2012	*/
2004	2013	static void find_server_repository(int arg, int fCreate){
2005	2014	if( g.argc<=arg ){
2006	2015	db_must_be_within_tree();
2007	2016	}else{
		@@ -2014,10 +2023,12 @@
2014	2023	if( isDir==0 && fCreate ){
2015	2024	const char *zPassword;
2016	2025	db_create_repository(zRepo);
2017	2026	db_open_repository(zRepo);
2018	2027	db_begin_transaction();
	2028	+ g.eHashPolicy = HPOLICY_AUTO;
	2029	+ db_set_int("hash-policy", HPOLICY_AUTO, 0);
2019	2030	db_initial_setup(0, "now", g.zLogin);
2020	2031	db_end_transaction(0);
2021	2032	fossil_print("project-id: %s\n", db_get("project-code", 0));
2022	2033	fossil_print("server-id: %s\n", db_get("server-code", 0));
2023	2034	zPassword = db_text(0, "SELECT pw FROM user WHERE login=%Q", g.zLogin);
2024	2035

	--- src/main.c
	+++ src/main.c
	@@ -50,11 +50,16 @@
50	# include "cson_amalgamation.h" /* JSON API. */
51	# include "json_detail.h"
52	#endif
53
54	/*
55	** Size of a UUID in characters





56	*/
57	#define UUID_SIZE 40
58
59	/*
60	** Maximum number of auxiliary parameters on reports
	@@ -135,10 +140,11 @@
135	char zLocalDbName; / Name of the local database file */
136	char zOpenRevision; / Check-in version to use during database open */
137	int localOpen; /* True if the local database is open */
138	char zLocalRoot; / The directory holding the local database */
139	int minPrefix; /* Number of digits needed for a distinct UUID */

140	int fNoDirSymlinks; /* True if --no-dir-symlinks flag is present */
141	int fSqlTrace; /* True if --sqltrace flag is present */
142	int fSqlStats; /* True if --sqltrace or --sqlstats are present */
143	int fSqlPrint; /* True if -sqlprint flag is present */
144	int fQuiet; /* True if -quiet flag is present */
	@@ -930,10 +936,13 @@
930	#if defined(FOSSIL_ENABLE_MINIZ)
931	blob_appendf(pOut, "miniz %s, loaded %s\n", MZ_VERSION, mz_version());
932	#else
933	blob_appendf(pOut, "zlib %s, loaded %s\n", ZLIB_VERSION, zlibVersion());
934	#endif



935	#if defined(FOSSIL_ENABLE_SSL)
936	blob_appendf(pOut, "SSL (%s)\n", SSLeay_version(SSLEAY_VERSION));
937	#endif
938	#if defined(FOSSIL_HAVE_FUSEFS)
939	blob_appendf(pOut, "libfuse %s, loaded %s\n", fusefs_inc_version(),
	@@ -1997,11 +2006,11 @@
1997	** the name of that directory and the specific repository will be
1998	** opened later by process_one_web_page() based on the content of
1999	** the PATH_INFO variable.
2000	**
2001	** If the fCreate flag is set, then create the repository if it
2002	** does not already exist.
2003	*/
2004	static void find_server_repository(int arg, int fCreate){
2005	if( g.argc<=arg ){
2006	db_must_be_within_tree();
2007	}else{
	@@ -2014,10 +2023,12 @@
2014	if( isDir==0 && fCreate ){
2015	const char *zPassword;
2016	db_create_repository(zRepo);
2017	db_open_repository(zRepo);
2018	db_begin_transaction();


2019	db_initial_setup(0, "now", g.zLogin);
2020	db_end_transaction(0);
2021	fossil_print("project-id: %s\n", db_get("project-code", 0));
2022	fossil_print("server-id: %s\n", db_get("server-code", 0));
2023	zPassword = db_text(0, "SELECT pw FROM user WHERE login=%Q", g.zLogin);
2024

	--- src/main.c
	+++ src/main.c
	@@ -50,11 +50,16 @@
50	# include "cson_amalgamation.h" /* JSON API. */
51	# include "json_detail.h"
52	#endif
53
54	/*
55	** Size of a UUID in characters. A UUID is a randomly generated
56	** lower-case hexadecimal number used to identify tickets.
57	**
58	** In Fossil 1.x, UUID also referred to a SHA1 artifact hash. But that
59	** usage is now obsolete. The term UUID should now mean only a very large
60	** random number used as a unique identifier for tickets or other objects.
61	*/
62	#define UUID_SIZE 40
63
64	/*
65	** Maximum number of auxiliary parameters on reports
	@@ -135,10 +140,11 @@
140	char zLocalDbName; / Name of the local database file */
141	char zOpenRevision; / Check-in version to use during database open */
142	int localOpen; /* True if the local database is open */
143	char zLocalRoot; / The directory holding the local database */
144	int minPrefix; /* Number of digits needed for a distinct UUID */
145	int eHashPolicy; /* Current hash policy. One of HPOLICY_* */
146	int fNoDirSymlinks; /* True if --no-dir-symlinks flag is present */
147	int fSqlTrace; /* True if --sqltrace flag is present */
148	int fSqlStats; /* True if --sqltrace or --sqlstats are present */
149	int fSqlPrint; /* True if -sqlprint flag is present */
150	int fQuiet; /* True if -quiet flag is present */
	@@ -930,10 +936,13 @@
936	#if defined(FOSSIL_ENABLE_MINIZ)
937	blob_appendf(pOut, "miniz %s, loaded %s\n", MZ_VERSION, mz_version());
938	#else
939	blob_appendf(pOut, "zlib %s, loaded %s\n", ZLIB_VERSION, zlibVersion());
940	#endif
941	#if FOSSIL_HARDENED_SHA1
942	blob_appendf(pOut, "hardened-SHA1 by Marc Stevens and Dan Shumow\n");
943	#endif
944	#if defined(FOSSIL_ENABLE_SSL)
945	blob_appendf(pOut, "SSL (%s)\n", SSLeay_version(SSLEAY_VERSION));
946	#endif
947	#if defined(FOSSIL_HAVE_FUSEFS)
948	blob_appendf(pOut, "libfuse %s, loaded %s\n", fusefs_inc_version(),
	@@ -1997,11 +2006,11 @@
2006	** the name of that directory and the specific repository will be
2007	** opened later by process_one_web_page() based on the content of
2008	** the PATH_INFO variable.
2009	**
2010	** If the fCreate flag is set, then create the repository if it
2011	** does not already exist. Always use "auto" hash-policy in this case.
2012	*/
2013	static void find_server_repository(int arg, int fCreate){
2014	if( g.argc<=arg ){
2015	db_must_be_within_tree();
2016	}else{
	@@ -2014,10 +2023,12 @@
2023	if( isDir==0 && fCreate ){
2024	const char *zPassword;
2025	db_create_repository(zRepo);
2026	db_open_repository(zRepo);
2027	db_begin_transaction();
2028	g.eHashPolicy = HPOLICY_AUTO;
2029	db_set_int("hash-policy", HPOLICY_AUTO, 0);
2030	db_initial_setup(0, "now", g.zLogin);
2031	db_end_transaction(0);
2032	fossil_print("project-id: %s\n", db_get("project-code", 0));
2033	fossil_print("server-id: %s\n", db_get("server-code", 0));
2034	zPassword = db_text(0, "SELECT pw FROM user WHERE login=%Q", g.zLogin);
2035

M src/main.mk

+36

		--- src/main.mk
		+++ src/main.mk
		@@ -52,10 +52,11 @@
52	52	$(SRCDIR)/fshell.c \
53	53	$(SRCDIR)/fusefs.c \
54	54	$(SRCDIR)/glob.c \
55	55	$(SRCDIR)/graph.c \
56	56	$(SRCDIR)/gzip.c \
	57	+ $(SRCDIR)/hname.c \
57	58	$(SRCDIR)/http.c \
58	59	$(SRCDIR)/http_socket.c \
59	60	$(SRCDIR)/http_ssl.c \
60	61	$(SRCDIR)/http_transport.c \
61	62	$(SRCDIR)/import.c \
		@@ -102,10 +103,12 @@
102	103	$(SRCDIR)/rss.c \
103	104	$(SRCDIR)/schema.c \
104	105	$(SRCDIR)/search.c \
105	106	$(SRCDIR)/setup.c \
106	107	$(SRCDIR)/sha1.c \
	108	+ $(SRCDIR)/sha1hard.c \
	109	+ $(SRCDIR)/sha3.c \
107	110	$(SRCDIR)/shun.c \
108	111	$(SRCDIR)/sitemap.c \
109	112	$(SRCDIR)/skins.c \
110	113	$(SRCDIR)/sqlcmd.c \
111	114	$(SRCDIR)/stash.c \
		@@ -227,10 +230,11 @@
227	230	$(OBJDIR)/fshell_.c \
228	231	$(OBJDIR)/fusefs_.c \
229	232	$(OBJDIR)/glob_.c \
230	233	$(OBJDIR)/graph_.c \
231	234	$(OBJDIR)/gzip_.c \
	235	+ $(OBJDIR)/hname_.c \
232	236	$(OBJDIR)/http_.c \
233	237	$(OBJDIR)/http_socket_.c \
234	238	$(OBJDIR)/http_ssl_.c \
235	239	$(OBJDIR)/http_transport_.c \
236	240	$(OBJDIR)/import_.c \
		@@ -277,10 +281,12 @@
277	281	$(OBJDIR)/rss_.c \
278	282	$(OBJDIR)/schema_.c \
279	283	$(OBJDIR)/search_.c \
280	284	$(OBJDIR)/setup_.c \
281	285	$(OBJDIR)/sha1_.c \
	286	+ $(OBJDIR)/sha1hard_.c \
	287	+ $(OBJDIR)/sha3_.c \
282	288	$(OBJDIR)/shun_.c \
283	289	$(OBJDIR)/sitemap_.c \
284	290	$(OBJDIR)/skins_.c \
285	291	$(OBJDIR)/sqlcmd_.c \
286	292	$(OBJDIR)/stash_.c \
		@@ -351,10 +357,11 @@
351	357	$(OBJDIR)/fshell.o \
352	358	$(OBJDIR)/fusefs.o \
353	359	$(OBJDIR)/glob.o \
354	360	$(OBJDIR)/graph.o \
355	361	$(OBJDIR)/gzip.o \
	362	+ $(OBJDIR)/hname.o \
356	363	$(OBJDIR)/http.o \
357	364	$(OBJDIR)/http_socket.o \
358	365	$(OBJDIR)/http_ssl.o \
359	366	$(OBJDIR)/http_transport.o \
360	367	$(OBJDIR)/import.o \
		@@ -401,10 +408,12 @@
401	408	$(OBJDIR)/rss.o \
402	409	$(OBJDIR)/schema.o \
403	410	$(OBJDIR)/search.o \
404	411	$(OBJDIR)/setup.o \
405	412	$(OBJDIR)/sha1.o \
	413	+ $(OBJDIR)/sha1hard.o \
	414	+ $(OBJDIR)/sha3.o \
406	415	$(OBJDIR)/shun.o \
407	416	$(OBJDIR)/sitemap.o \
408	417	$(OBJDIR)/skins.o \
409	418	$(OBJDIR)/sqlcmd.o \
410	419	$(OBJDIR)/stash.o \
		@@ -636,10 +645,11 @@
636	645	$(OBJDIR)/fshell_.c:$(OBJDIR)/fshell.h \
637	646	$(OBJDIR)/fusefs_.c:$(OBJDIR)/fusefs.h \
638	647	$(OBJDIR)/glob_.c:$(OBJDIR)/glob.h \
639	648	$(OBJDIR)/graph_.c:$(OBJDIR)/graph.h \
640	649	$(OBJDIR)/gzip_.c:$(OBJDIR)/gzip.h \
	650	+ $(OBJDIR)/hname_.c:$(OBJDIR)/hname.h \
641	651	$(OBJDIR)/http_.c:$(OBJDIR)/http.h \
642	652	$(OBJDIR)/http_socket_.c:$(OBJDIR)/http_socket.h \
643	653	$(OBJDIR)/http_ssl_.c:$(OBJDIR)/http_ssl.h \
644	654	$(OBJDIR)/http_transport_.c:$(OBJDIR)/http_transport.h \
645	655	$(OBJDIR)/import_.c:$(OBJDIR)/import.h \
		@@ -686,10 +696,12 @@
686	696	$(OBJDIR)/rss_.c:$(OBJDIR)/rss.h \
687	697	$(OBJDIR)/schema_.c:$(OBJDIR)/schema.h \
688	698	$(OBJDIR)/search_.c:$(OBJDIR)/search.h \
689	699	$(OBJDIR)/setup_.c:$(OBJDIR)/setup.h \
690	700	$(OBJDIR)/sha1_.c:$(OBJDIR)/sha1.h \
	701	+ $(OBJDIR)/sha1hard_.c:$(OBJDIR)/sha1hard.h \
	702	+ $(OBJDIR)/sha3_.c:$(OBJDIR)/sha3.h \
691	703	$(OBJDIR)/shun_.c:$(OBJDIR)/shun.h \
692	704	$(OBJDIR)/sitemap_.c:$(OBJDIR)/sitemap.h \
693	705	$(OBJDIR)/skins_.c:$(OBJDIR)/skins.h \
694	706	$(OBJDIR)/sqlcmd_.c:$(OBJDIR)/sqlcmd.h \
695	707	$(OBJDIR)/stash_.c:$(OBJDIR)/stash.h \
		@@ -1037,10 +1049,18 @@
1037	1049
1038	1050	$(OBJDIR)/gzip.o: $(OBJDIR)/gzip_.c $(OBJDIR)/gzip.h $(SRCDIR)/config.h
1039	1051	$(XTCC) -o $(OBJDIR)/gzip.o -c $(OBJDIR)/gzip_.c
1040	1052
1041	1053	$(OBJDIR)/gzip.h: $(OBJDIR)/headers
	1054	+
	1055	+$(OBJDIR)/hname_.c: $(SRCDIR)/hname.c $(OBJDIR)/translate
	1056	+ $(OBJDIR)/translate $(SRCDIR)/hname.c >$@
	1057	+
	1058	+$(OBJDIR)/hname.o: $(OBJDIR)/hname_.c $(OBJDIR)/hname.h $(SRCDIR)/config.h
	1059	+ $(XTCC) -o $(OBJDIR)/hname.o -c $(OBJDIR)/hname_.c
	1060	+
	1061	+$(OBJDIR)/hname.h: $(OBJDIR)/headers
1042	1062
1043	1063	$(OBJDIR)/http_.c: $(SRCDIR)/http.c $(OBJDIR)/translate
1044	1064	$(OBJDIR)/translate $(SRCDIR)/http.c >$@
1045	1065
1046	1066	$(OBJDIR)/http.o: $(OBJDIR)/http_.c $(OBJDIR)/http.h $(SRCDIR)/config.h
		@@ -1437,10 +1457,26 @@
1437	1457
1438	1458	$(OBJDIR)/sha1.o: $(OBJDIR)/sha1_.c $(OBJDIR)/sha1.h $(SRCDIR)/config.h
1439	1459	$(XTCC) -o $(OBJDIR)/sha1.o -c $(OBJDIR)/sha1_.c
1440	1460
1441	1461	$(OBJDIR)/sha1.h: $(OBJDIR)/headers
	1462	+
	1463	+$(OBJDIR)/sha1hard_.c: $(SRCDIR)/sha1hard.c $(OBJDIR)/translate
	1464	+ $(OBJDIR)/translate $(SRCDIR)/sha1hard.c >$@
	1465	+
	1466	+$(OBJDIR)/sha1hard.o: $(OBJDIR)/sha1hard_.c $(OBJDIR)/sha1hard.h $(SRCDIR)/config.h
	1467	+ $(XTCC) -o $(OBJDIR)/sha1hard.o -c $(OBJDIR)/sha1hard_.c
	1468	+
	1469	+$(OBJDIR)/sha1hard.h: $(OBJDIR)/headers
	1470	+
	1471	+$(OBJDIR)/sha3_.c: $(SRCDIR)/sha3.c $(OBJDIR)/translate
	1472	+ $(OBJDIR)/translate $(SRCDIR)/sha3.c >$@
	1473	+
	1474	+$(OBJDIR)/sha3.o: $(OBJDIR)/sha3_.c $(OBJDIR)/sha3.h $(SRCDIR)/config.h
	1475	+ $(XTCC) -o $(OBJDIR)/sha3.o -c $(OBJDIR)/sha3_.c
	1476	+
	1477	+$(OBJDIR)/sha3.h: $(OBJDIR)/headers
1442	1478
1443	1479	$(OBJDIR)/shun_.c: $(SRCDIR)/shun.c $(OBJDIR)/translate
1444	1480	$(OBJDIR)/translate $(SRCDIR)/shun.c >$@
1445	1481
1446	1482	$(OBJDIR)/shun.o: $(OBJDIR)/shun_.c $(OBJDIR)/shun.h $(SRCDIR)/config.h
1447	1483

	--- src/main.mk
	+++ src/main.mk
	@@ -52,10 +52,11 @@
52	$(SRCDIR)/fshell.c \
53	$(SRCDIR)/fusefs.c \
54	$(SRCDIR)/glob.c \
55	$(SRCDIR)/graph.c \
56	$(SRCDIR)/gzip.c \

57	$(SRCDIR)/http.c \
58	$(SRCDIR)/http_socket.c \
59	$(SRCDIR)/http_ssl.c \
60	$(SRCDIR)/http_transport.c \
61	$(SRCDIR)/import.c \
	@@ -102,10 +103,12 @@
102	$(SRCDIR)/rss.c \
103	$(SRCDIR)/schema.c \
104	$(SRCDIR)/search.c \
105	$(SRCDIR)/setup.c \
106	$(SRCDIR)/sha1.c \


107	$(SRCDIR)/shun.c \
108	$(SRCDIR)/sitemap.c \
109	$(SRCDIR)/skins.c \
110	$(SRCDIR)/sqlcmd.c \
111	$(SRCDIR)/stash.c \
	@@ -227,10 +230,11 @@
227	$(OBJDIR)/fshell_.c \
228	$(OBJDIR)/fusefs_.c \
229	$(OBJDIR)/glob_.c \
230	$(OBJDIR)/graph_.c \
231	$(OBJDIR)/gzip_.c \

232	$(OBJDIR)/http_.c \
233	$(OBJDIR)/http_socket_.c \
234	$(OBJDIR)/http_ssl_.c \
235	$(OBJDIR)/http_transport_.c \
236	$(OBJDIR)/import_.c \
	@@ -277,10 +281,12 @@
277	$(OBJDIR)/rss_.c \
278	$(OBJDIR)/schema_.c \
279	$(OBJDIR)/search_.c \
280	$(OBJDIR)/setup_.c \
281	$(OBJDIR)/sha1_.c \


282	$(OBJDIR)/shun_.c \
283	$(OBJDIR)/sitemap_.c \
284	$(OBJDIR)/skins_.c \
285	$(OBJDIR)/sqlcmd_.c \
286	$(OBJDIR)/stash_.c \
	@@ -351,10 +357,11 @@
351	$(OBJDIR)/fshell.o \
352	$(OBJDIR)/fusefs.o \
353	$(OBJDIR)/glob.o \
354	$(OBJDIR)/graph.o \
355	$(OBJDIR)/gzip.o \

356	$(OBJDIR)/http.o \
357	$(OBJDIR)/http_socket.o \
358	$(OBJDIR)/http_ssl.o \
359	$(OBJDIR)/http_transport.o \
360	$(OBJDIR)/import.o \
	@@ -401,10 +408,12 @@
401	$(OBJDIR)/rss.o \
402	$(OBJDIR)/schema.o \
403	$(OBJDIR)/search.o \
404	$(OBJDIR)/setup.o \
405	$(OBJDIR)/sha1.o \


406	$(OBJDIR)/shun.o \
407	$(OBJDIR)/sitemap.o \
408	$(OBJDIR)/skins.o \
409	$(OBJDIR)/sqlcmd.o \
410	$(OBJDIR)/stash.o \
	@@ -636,10 +645,11 @@
636	$(OBJDIR)/fshell_.c:$(OBJDIR)/fshell.h \
637	$(OBJDIR)/fusefs_.c:$(OBJDIR)/fusefs.h \
638	$(OBJDIR)/glob_.c:$(OBJDIR)/glob.h \
639	$(OBJDIR)/graph_.c:$(OBJDIR)/graph.h \
640	$(OBJDIR)/gzip_.c:$(OBJDIR)/gzip.h \

641	$(OBJDIR)/http_.c:$(OBJDIR)/http.h \
642	$(OBJDIR)/http_socket_.c:$(OBJDIR)/http_socket.h \
643	$(OBJDIR)/http_ssl_.c:$(OBJDIR)/http_ssl.h \
644	$(OBJDIR)/http_transport_.c:$(OBJDIR)/http_transport.h \
645	$(OBJDIR)/import_.c:$(OBJDIR)/import.h \
	@@ -686,10 +696,12 @@
686	$(OBJDIR)/rss_.c:$(OBJDIR)/rss.h \
687	$(OBJDIR)/schema_.c:$(OBJDIR)/schema.h \
688	$(OBJDIR)/search_.c:$(OBJDIR)/search.h \
689	$(OBJDIR)/setup_.c:$(OBJDIR)/setup.h \
690	$(OBJDIR)/sha1_.c:$(OBJDIR)/sha1.h \


691	$(OBJDIR)/shun_.c:$(OBJDIR)/shun.h \
692	$(OBJDIR)/sitemap_.c:$(OBJDIR)/sitemap.h \
693	$(OBJDIR)/skins_.c:$(OBJDIR)/skins.h \
694	$(OBJDIR)/sqlcmd_.c:$(OBJDIR)/sqlcmd.h \
695	$(OBJDIR)/stash_.c:$(OBJDIR)/stash.h \
	@@ -1037,10 +1049,18 @@
1037
1038	$(OBJDIR)/gzip.o: $(OBJDIR)/gzip_.c $(OBJDIR)/gzip.h $(SRCDIR)/config.h
1039	$(XTCC) -o $(OBJDIR)/gzip.o -c $(OBJDIR)/gzip_.c
1040
1041	$(OBJDIR)/gzip.h: $(OBJDIR)/headers








1042
1043	$(OBJDIR)/http_.c: $(SRCDIR)/http.c $(OBJDIR)/translate
1044	$(OBJDIR)/translate $(SRCDIR)/http.c >$@
1045
1046	$(OBJDIR)/http.o: $(OBJDIR)/http_.c $(OBJDIR)/http.h $(SRCDIR)/config.h
	@@ -1437,10 +1457,26 @@
1437
1438	$(OBJDIR)/sha1.o: $(OBJDIR)/sha1_.c $(OBJDIR)/sha1.h $(SRCDIR)/config.h
1439	$(XTCC) -o $(OBJDIR)/sha1.o -c $(OBJDIR)/sha1_.c
1440
1441	$(OBJDIR)/sha1.h: $(OBJDIR)/headers
















1442
1443	$(OBJDIR)/shun_.c: $(SRCDIR)/shun.c $(OBJDIR)/translate
1444	$(OBJDIR)/translate $(SRCDIR)/shun.c >$@
1445
1446	$(OBJDIR)/shun.o: $(OBJDIR)/shun_.c $(OBJDIR)/shun.h $(SRCDIR)/config.h
1447

	--- src/main.mk
	+++ src/main.mk
	@@ -52,10 +52,11 @@
52	$(SRCDIR)/fshell.c \
53	$(SRCDIR)/fusefs.c \
54	$(SRCDIR)/glob.c \
55	$(SRCDIR)/graph.c \
56	$(SRCDIR)/gzip.c \
57	$(SRCDIR)/hname.c \
58	$(SRCDIR)/http.c \
59	$(SRCDIR)/http_socket.c \
60	$(SRCDIR)/http_ssl.c \
61	$(SRCDIR)/http_transport.c \
62	$(SRCDIR)/import.c \
	@@ -102,10 +103,12 @@
103	$(SRCDIR)/rss.c \
104	$(SRCDIR)/schema.c \
105	$(SRCDIR)/search.c \
106	$(SRCDIR)/setup.c \
107	$(SRCDIR)/sha1.c \
108	$(SRCDIR)/sha1hard.c \
109	$(SRCDIR)/sha3.c \
110	$(SRCDIR)/shun.c \
111	$(SRCDIR)/sitemap.c \
112	$(SRCDIR)/skins.c \
113	$(SRCDIR)/sqlcmd.c \
114	$(SRCDIR)/stash.c \
	@@ -227,10 +230,11 @@
230	$(OBJDIR)/fshell_.c \
231	$(OBJDIR)/fusefs_.c \
232	$(OBJDIR)/glob_.c \
233	$(OBJDIR)/graph_.c \
234	$(OBJDIR)/gzip_.c \
235	$(OBJDIR)/hname_.c \
236	$(OBJDIR)/http_.c \
237	$(OBJDIR)/http_socket_.c \
238	$(OBJDIR)/http_ssl_.c \
239	$(OBJDIR)/http_transport_.c \
240	$(OBJDIR)/import_.c \
	@@ -277,10 +281,12 @@
281	$(OBJDIR)/rss_.c \
282	$(OBJDIR)/schema_.c \
283	$(OBJDIR)/search_.c \
284	$(OBJDIR)/setup_.c \
285	$(OBJDIR)/sha1_.c \
286	$(OBJDIR)/sha1hard_.c \
287	$(OBJDIR)/sha3_.c \
288	$(OBJDIR)/shun_.c \
289	$(OBJDIR)/sitemap_.c \
290	$(OBJDIR)/skins_.c \
291	$(OBJDIR)/sqlcmd_.c \
292	$(OBJDIR)/stash_.c \
	@@ -351,10 +357,11 @@
357	$(OBJDIR)/fshell.o \
358	$(OBJDIR)/fusefs.o \
359	$(OBJDIR)/glob.o \
360	$(OBJDIR)/graph.o \
361	$(OBJDIR)/gzip.o \
362	$(OBJDIR)/hname.o \
363	$(OBJDIR)/http.o \
364	$(OBJDIR)/http_socket.o \
365	$(OBJDIR)/http_ssl.o \
366	$(OBJDIR)/http_transport.o \
367	$(OBJDIR)/import.o \
	@@ -401,10 +408,12 @@
408	$(OBJDIR)/rss.o \
409	$(OBJDIR)/schema.o \
410	$(OBJDIR)/search.o \
411	$(OBJDIR)/setup.o \
412	$(OBJDIR)/sha1.o \
413	$(OBJDIR)/sha1hard.o \
414	$(OBJDIR)/sha3.o \
415	$(OBJDIR)/shun.o \
416	$(OBJDIR)/sitemap.o \
417	$(OBJDIR)/skins.o \
418	$(OBJDIR)/sqlcmd.o \
419	$(OBJDIR)/stash.o \
	@@ -636,10 +645,11 @@
645	$(OBJDIR)/fshell_.c:$(OBJDIR)/fshell.h \
646	$(OBJDIR)/fusefs_.c:$(OBJDIR)/fusefs.h \
647	$(OBJDIR)/glob_.c:$(OBJDIR)/glob.h \
648	$(OBJDIR)/graph_.c:$(OBJDIR)/graph.h \
649	$(OBJDIR)/gzip_.c:$(OBJDIR)/gzip.h \
650	$(OBJDIR)/hname_.c:$(OBJDIR)/hname.h \
651	$(OBJDIR)/http_.c:$(OBJDIR)/http.h \
652	$(OBJDIR)/http_socket_.c:$(OBJDIR)/http_socket.h \
653	$(OBJDIR)/http_ssl_.c:$(OBJDIR)/http_ssl.h \
654	$(OBJDIR)/http_transport_.c:$(OBJDIR)/http_transport.h \
655	$(OBJDIR)/import_.c:$(OBJDIR)/import.h \
	@@ -686,10 +696,12 @@
696	$(OBJDIR)/rss_.c:$(OBJDIR)/rss.h \
697	$(OBJDIR)/schema_.c:$(OBJDIR)/schema.h \
698	$(OBJDIR)/search_.c:$(OBJDIR)/search.h \
699	$(OBJDIR)/setup_.c:$(OBJDIR)/setup.h \
700	$(OBJDIR)/sha1_.c:$(OBJDIR)/sha1.h \
701	$(OBJDIR)/sha1hard_.c:$(OBJDIR)/sha1hard.h \
702	$(OBJDIR)/sha3_.c:$(OBJDIR)/sha3.h \
703	$(OBJDIR)/shun_.c:$(OBJDIR)/shun.h \
704	$(OBJDIR)/sitemap_.c:$(OBJDIR)/sitemap.h \
705	$(OBJDIR)/skins_.c:$(OBJDIR)/skins.h \
706	$(OBJDIR)/sqlcmd_.c:$(OBJDIR)/sqlcmd.h \
707	$(OBJDIR)/stash_.c:$(OBJDIR)/stash.h \
	@@ -1037,10 +1049,18 @@
1049
1050	$(OBJDIR)/gzip.o: $(OBJDIR)/gzip_.c $(OBJDIR)/gzip.h $(SRCDIR)/config.h
1051	$(XTCC) -o $(OBJDIR)/gzip.o -c $(OBJDIR)/gzip_.c
1052
1053	$(OBJDIR)/gzip.h: $(OBJDIR)/headers
1054
1055	$(OBJDIR)/hname_.c: $(SRCDIR)/hname.c $(OBJDIR)/translate
1056	$(OBJDIR)/translate $(SRCDIR)/hname.c >$@
1057
1058	$(OBJDIR)/hname.o: $(OBJDIR)/hname_.c $(OBJDIR)/hname.h $(SRCDIR)/config.h
1059	$(XTCC) -o $(OBJDIR)/hname.o -c $(OBJDIR)/hname_.c
1060
1061	$(OBJDIR)/hname.h: $(OBJDIR)/headers
1062
1063	$(OBJDIR)/http_.c: $(SRCDIR)/http.c $(OBJDIR)/translate
1064	$(OBJDIR)/translate $(SRCDIR)/http.c >$@
1065
1066	$(OBJDIR)/http.o: $(OBJDIR)/http_.c $(OBJDIR)/http.h $(SRCDIR)/config.h
	@@ -1437,10 +1457,26 @@
1457
1458	$(OBJDIR)/sha1.o: $(OBJDIR)/sha1_.c $(OBJDIR)/sha1.h $(SRCDIR)/config.h
1459	$(XTCC) -o $(OBJDIR)/sha1.o -c $(OBJDIR)/sha1_.c
1460
1461	$(OBJDIR)/sha1.h: $(OBJDIR)/headers
1462
1463	$(OBJDIR)/sha1hard_.c: $(SRCDIR)/sha1hard.c $(OBJDIR)/translate
1464	$(OBJDIR)/translate $(SRCDIR)/sha1hard.c >$@
1465
1466	$(OBJDIR)/sha1hard.o: $(OBJDIR)/sha1hard_.c $(OBJDIR)/sha1hard.h $(SRCDIR)/config.h
1467	$(XTCC) -o $(OBJDIR)/sha1hard.o -c $(OBJDIR)/sha1hard_.c
1468
1469	$(OBJDIR)/sha1hard.h: $(OBJDIR)/headers
1470
1471	$(OBJDIR)/sha3_.c: $(SRCDIR)/sha3.c $(OBJDIR)/translate
1472	$(OBJDIR)/translate $(SRCDIR)/sha3.c >$@
1473
1474	$(OBJDIR)/sha3.o: $(OBJDIR)/sha3_.c $(OBJDIR)/sha3.h $(SRCDIR)/config.h
1475	$(XTCC) -o $(OBJDIR)/sha3.o -c $(OBJDIR)/sha3_.c
1476
1477	$(OBJDIR)/sha3.h: $(OBJDIR)/headers
1478
1479	$(OBJDIR)/shun_.c: $(SRCDIR)/shun.c $(OBJDIR)/translate
1480	$(OBJDIR)/translate $(SRCDIR)/shun.c >$@
1481
1482	$(OBJDIR)/shun.o: $(OBJDIR)/shun_.c $(OBJDIR)/shun.h $(SRCDIR)/config.h
1483

M src/makeheaders.c

+6 -1

		--- src/makeheaders.c
		+++ src/makeheaders.c
		@@ -1017,11 +1017,14 @@
1017	1017	/* printf("%04d: Type=%d nIf=%d [%.*s]\n",
1018	1018	pToken->nLine,pToken->eType,nIf,pToken->nText,
1019	1019	pToken->eType!=TT_Space ? pToken->zText : "<space>"); */
1020	1020	pToken->pComment = blockComment;
1021	1021	switch( pToken->eType ){
1022		- case TT_Comment:
	1022	+ case TT_Comment: /0123456789 12345678 /
	1023	+ if( strncmp(pToken->zText, "/*MAKEHEADERS-STOP", 18)==0 ) return nErr;
	1024	+ break;
	1025	+
1023	1026	case TT_Space:
1024	1027	break;
1025	1028
1026	1029	case TT_BlockComment:
1027	1030	if( doc_flag ){
		@@ -2181,10 +2184,12 @@
2181	2184	if( nArg==9 && strncmp(zArg,"INTERFACE",9)==0 ){
2182	2185	PushIfMacro(0,0,0,pToken->nLine,PS_Interface);
2183	2186	}else if( nArg==16 && strncmp(zArg,"EXPORT_INTERFACE",16)==0 ){
2184	2187	PushIfMacro(0,0,0,pToken->nLine,PS_Export);
2185	2188	}else if( nArg==15 && strncmp(zArg,"LOCAL_INTERFACE",15)==0 ){
	2189	+ PushIfMacro(0,0,0,pToken->nLine,PS_Local);
	2190	+ }else if( nArg==15 && strncmp(zArg,"MAKEHEADERS_STOPLOCAL_INTERFACE",15)==0 ){
2186	2191	PushIfMacro(0,0,0,pToken->nLine,PS_Local);
2187	2192	}else{
2188	2193	PushIfMacro(0,zArg,nArg,pToken->nLine,0);
2189	2194	}
2190	2195	}else if( nCmd==5 && strncmp(zCmd,"ifdef",5)==0 ){
2191	2196

	--- src/makeheaders.c
	+++ src/makeheaders.c
	@@ -1017,11 +1017,14 @@
1017	/* printf("%04d: Type=%d nIf=%d [%.*s]\n",
1018	pToken->nLine,pToken->eType,nIf,pToken->nText,
1019	pToken->eType!=TT_Space ? pToken->zText : "<space>"); */
1020	pToken->pComment = blockComment;
1021	switch( pToken->eType ){
1022	case TT_Comment:



1023	case TT_Space:
1024	break;
1025
1026	case TT_BlockComment:
1027	if( doc_flag ){
	@@ -2181,10 +2184,12 @@
2181	if( nArg==9 && strncmp(zArg,"INTERFACE",9)==0 ){
2182	PushIfMacro(0,0,0,pToken->nLine,PS_Interface);
2183	}else if( nArg==16 && strncmp(zArg,"EXPORT_INTERFACE",16)==0 ){
2184	PushIfMacro(0,0,0,pToken->nLine,PS_Export);
2185	}else if( nArg==15 && strncmp(zArg,"LOCAL_INTERFACE",15)==0 ){


2186	PushIfMacro(0,0,0,pToken->nLine,PS_Local);
2187	}else{
2188	PushIfMacro(0,zArg,nArg,pToken->nLine,0);
2189	}
2190	}else if( nCmd==5 && strncmp(zCmd,"ifdef",5)==0 ){
2191

	--- src/makeheaders.c
	+++ src/makeheaders.c
	@@ -1017,11 +1017,14 @@
1017	/* printf("%04d: Type=%d nIf=%d [%.*s]\n",
1018	pToken->nLine,pToken->eType,nIf,pToken->nText,
1019	pToken->eType!=TT_Space ? pToken->zText : "<space>"); */
1020	pToken->pComment = blockComment;
1021	switch( pToken->eType ){
1022	case TT_Comment: /0123456789 12345678 /
1023	if( strncmp(pToken->zText, "/*MAKEHEADERS-STOP", 18)==0 ) return nErr;
1024	break;
1025
1026	case TT_Space:
1027	break;
1028
1029	case TT_BlockComment:
1030	if( doc_flag ){
	@@ -2181,10 +2184,12 @@
2184	if( nArg==9 && strncmp(zArg,"INTERFACE",9)==0 ){
2185	PushIfMacro(0,0,0,pToken->nLine,PS_Interface);
2186	}else if( nArg==16 && strncmp(zArg,"EXPORT_INTERFACE",16)==0 ){
2187	PushIfMacro(0,0,0,pToken->nLine,PS_Export);
2188	}else if( nArg==15 && strncmp(zArg,"LOCAL_INTERFACE",15)==0 ){
2189	PushIfMacro(0,0,0,pToken->nLine,PS_Local);
2190	}else if( nArg==15 && strncmp(zArg,"MAKEHEADERS_STOPLOCAL_INTERFACE",15)==0 ){
2191	PushIfMacro(0,0,0,pToken->nLine,PS_Local);
2192	}else{
2193	PushIfMacro(0,zArg,nArg,pToken->nLine,0);
2194	}
2195	}else if( nCmd==5 && strncmp(zCmd,"ifdef",5)==0 ){
2196

M src/makemake.tcl

		--- src/makemake.tcl
		+++ src/makemake.tcl
		@@ -58,10 +58,11 @@
58	58	fshell
59	59	fusefs
60	60	glob
61	61	graph
62	62	gzip
	63	+ hname
63	64	http
64	65	http_socket
65	66	http_transport
66	67	import
67	68	info
		@@ -107,10 +108,12 @@
107	108	rss
108	109	schema
109	110	search
110	111	setup
111	112	sha1
	113	+ sha1hard
	114	+ sha3
112	115	shun
113	116	sitemap
114	117	skins
115	118	sqlcmd
116	119	stash
117	120

	--- src/makemake.tcl
	+++ src/makemake.tcl
	@@ -58,10 +58,11 @@
58	fshell
59	fusefs
60	glob
61	graph
62	gzip

63	http
64	http_socket
65	http_transport
66	import
67	info
	@@ -107,10 +108,12 @@
107	rss
108	schema
109	search
110	setup
111	sha1


112	shun
113	sitemap
114	skins
115	sqlcmd
116	stash
117

	--- src/makemake.tcl
	+++ src/makemake.tcl
	@@ -58,10 +58,11 @@
58	fshell
59	fusefs
60	glob
61	graph
62	gzip
63	hname
64	http
65	http_socket
66	http_transport
67	import
68	info
	@@ -107,10 +108,12 @@
108	rss
109	schema
110	search
111	setup
112	sha1
113	sha1hard
114	sha3
115	shun
116	sitemap
117	skins
118	sqlcmd
119	stash
120

M src/makemake.tcl

		--- src/makemake.tcl
		+++ src/makemake.tcl
		@@ -58,10 +58,11 @@
58	58	fshell
59	59	fusefs
60	60	glob
61	61	graph
62	62	gzip
	63	+ hname
63	64	http
64	65	http_socket
65	66	http_transport
66	67	import
67	68	info
		@@ -107,10 +108,12 @@
107	108	rss
108	109	schema
109	110	search
110	111	setup
111	112	sha1
	113	+ sha1hard
	114	+ sha3
112	115	shun
113	116	sitemap
114	117	skins
115	118	sqlcmd
116	119	stash
117	120

	--- src/makemake.tcl
	+++ src/makemake.tcl
	@@ -58,10 +58,11 @@
58	fshell
59	fusefs
60	glob
61	graph
62	gzip

63	http
64	http_socket
65	http_transport
66	import
67	info
	@@ -107,10 +108,12 @@
107	rss
108	schema
109	search
110	setup
111	sha1


112	shun
113	sitemap
114	skins
115	sqlcmd
116	stash
117

	--- src/makemake.tcl
	+++ src/makemake.tcl
	@@ -58,10 +58,11 @@
58	fshell
59	fusefs
60	glob
61	graph
62	gzip
63	hname
64	http
65	http_socket
66	http_transport
67	import
68	info
	@@ -107,10 +108,12 @@
108	rss
109	schema
110	search
111	setup
112	sha1
113	sha1hard
114	sha3
115	shun
116	sitemap
117	skins
118	sqlcmd
119	stash
120

M src/manifest.c

+80 -80

		--- src/manifest.c
		+++ src/manifest.c
		@@ -54,11 +54,11 @@
54	54	/*
55	55	** A single F-card within a manifest
56	56	*/
57	57	struct ManifestFile {
58	58	char zName; / Name of a file */
59		- char zUuid; / UUID of the file */
	59	+ char zUuid; / Artifact hash for the file */
60	60	char zPerm; / File permissions */
61	61	char zPrior; / Prior name if the name was changed */
62	62	};
63	63
64	64
		@@ -77,35 +77,35 @@
77	77	char zRepoCksum; / MD5 checksum of the baseline content. R card. */
78	78	char zWiki; / Text of the wiki page. W card. */
79	79	char zWikiTitle; / Name of the wiki page. L card. */
80	80	char zMimetype; / Mime type of wiki or comment text. N card. */
81	81	double rEventDate; /* Date of an event. E card. */
82		- char zEventId; / UUID for an event. E card. */
	82	+ char zEventId; / Artifact hash for an event. E card. */
83	83	char zTicketUuid; / UUID for a ticket. K card. */
84	84	char zAttachName; / Filename of an attachment. A card. */
85		- char zAttachSrc; / UUID of document being attached. A card. */
	85	+ char zAttachSrc; / Artifact hash for document being attached. A card. */
86	86	char zAttachTarget; / Ticket or wiki that attachment applies to. A card */
87	87	int nFile; /* Number of F cards */
88	88	int nFileAlloc; /* Slots allocated in aFile[] */
89	89	int iFile; /* Index of current file in iterator */
90	90	ManifestFile aFile; / One entry for each F-card */
91	91	int nParent; /* Number of parents. */
92	92	int nParentAlloc; /* Slots allocated in azParent[] */
93		- char *azParent; / UUIDs of parents. One for each P card argument */
	93	+ char *azParent; / Hashes of parents. One for each P card argument */
94	94	int nCherrypick; /* Number of entries in aCherrypick[] */
95	95	struct {
96		- char zCPTarget; / UUID of cherry-picked version w/ +\|- prefix */
97		- char zCPBase; / UUID of cherry-pick baseline. NULL for singletons */
	96	+ char zCPTarget; / Hash for cherry-picked version w/ +\|- prefix */
	97	+ char zCPBase; / Hash for cherry-pick baseline. NULL for singletons */
98	98	} *aCherrypick;
99	99	int nCChild; /* Number of cluster children */
100	100	int nCChildAlloc; /* Number of closts allocated in azCChild[] */
101		- char *azCChild; / UUIDs of referenced objects in a cluster. M cards */
	101	+ char *azCChild; / Hashes of referenced objects in a cluster. M cards */
102	102	int nTag; /* Number of T Cards */
103	103	int nTagAlloc; /* Slots allocated in aTag[] */
104	104	struct TagType {
105	105	char zName; / Name of the tag */
106		- char zUuid; / UUID that the tag is applied to */
	106	+ char zUuid; / Hash of artifact that the tag is applied to */
107	107	char zValue; / Value if the tag is really a property */
108	108	} aTag; / One for each T card */
109	109	int nField; /* Number of J cards */
110	110	int nFieldAlloc; /* Slots allocated in aField[] */
111	111	struct {
		@@ -327,26 +327,27 @@
327	327	** takes over the input blob and will free it when the
328	328	** Manifest object is freed. Zeros are inserted into the blob
329	329	** as string terminators so that blob should not be used again.
330	330	**
331	331	** Return a pointer to an allocated Manifest object if the content
332		-** really is a control file of some kind. This object needs to be
333		-** freed by a subsequent call to manifest_destroy(). Return NULL
334		-** if there are syntax errors.
	332	+** really is a structural artifact of some kind. The returned Manifest
	333	+** object needs to be freed by a subsequent call to manifest_destroy().
	334	+** Return NULL if there are syntax errors or if the input blob does
	335	+** not describe a valid structural artifact.
335	336	**
336		-** This routine is strict about the format of a control file.
	337	+** This routine is strict about the format of a structural artifacts.
337	338	** The format must match exactly or else it is rejected. This
338		-** rule minimizes the risk that a content file will be mistaken
339		-** for a control file simply because they look the same.
	339	+** rule minimizes the risk that a content artifact will be mistaken
	340	+** for a structural artifact simply because they look the same.
340	341	**
341	342	** The pContent is reset. If a pointer is returned, then pContent will
342	343	** be reset when the Manifest object is cleared. If NULL is
343	344	** returned then the Manifest object is cleared automatically
344	345	** and pContent is reset before the return.
345	346	**
346		-** The entire file can be PGP clear-signed. The signature is ignored.
347		-** The file consists of zero or more cards, one card per line.
	347	+** The entire input blob can be PGP clear-signed. The signature is ignored.
	348	+** The artifact consists of zero or more cards, one card per line.
348	349	** (Except: the content of the W card can extend of multiple lines.)
349	350	** Each card is divided into tokens by a single space character.
350	351	** The first token is a single upper-case letter which is the card type.
351	352	** The card type determines the other parameters to the card.
352	353	** Cards must occur in lexicographical order.
		@@ -361,11 +362,10 @@
361	362	char *z;
362	363	int n;
363	364	char *zUuid;
364	365	int sz = 0;
365	366	int isRepeat, hasSelfRefTag = 0;
366		- Blob bUuid = BLOB_INITIALIZER;
367	367	static Bag seen;
368	368	const char *zErr = 0;
369	369
370	370	if( rid==0 ){
371	371	isRepeat = 1;
		@@ -374,11 +374,11 @@
374	374	}else{
375	375	isRepeat = 0;
376	376	bag_insert(&seen, rid);
377	377	}
378	378
379		- /* Every control artifact ends with a '\n' character. Exit early
	379	+ /* Every structural artifact ends with a '\n' character. Exit early
380	380	** if that is not the case for this artifact.
381	381	*/
382	382	if( !isRepeat ) g.parseCnt[0]++;
383	383	z = blob_materialize(pContent);
384	384	n = blob_size(pContent);
		@@ -406,15 +406,10 @@
406	406	blob_reset(pContent);
407	407	blob_appendf(pErr, "incorrect Z-card cksum");
408	408	return 0;
409	409	}
410	410
411		- /* Store the UUID (before modifying the blob) only for error
412		- ** reporting purposes.
413		- */
414		- sha1sum_blob(pContent, &bUuid);
415		-
416	411	/* Allocate a Manifest object to hold the parsed control artifact.
417	412	*/
418	413	p = fossil_malloc( sizeof(*p) );
419	414	memset(p, 0, sizeof(*p));
420	415	memcpy(&p->content, pContent, sizeof(p->content));
		@@ -449,15 +444,15 @@
449	444	defossilize(zName);
450	445	if( !file_is_simple_pathname(zName, 0) ){
451	446	SYNTAX("invalid filename on A-card");
452	447	}
453	448	defossilize(zTarget);
454		- if( (nTarget!=UUID_SIZE \|\| !validate16(zTarget, UUID_SIZE))
	449	+ if( !hname_validate(zTarget,nTarget)
455	450	&& !wiki_name_is_wellformed((const unsigned char *)zTarget) ){
456	451	SYNTAX("invalid target on A-card");
457	452	}
458		- if( zSrc && (nSrc!=UUID_SIZE \|\| !validate16(zSrc, UUID_SIZE)) ){
	453	+ if( zSrc && !hname_validate(zSrc,nSrc) ){
459	454	SYNTAX("invalid source on A-card");
460	455	}
461	456	p->zAttachName = (char*)file_tail(zName);
462	457	p->zAttachSrc = zSrc;
463	458	p->zAttachTarget = zTarget;
		@@ -465,18 +460,18 @@
465	460	}
466	461
467	462	/*
468	463	** B <uuid>
469	464	**
470		- ** A B-line gives the UUID for the baseline of a delta-manifest.
	465	+ ** A B-line gives the artifact hash for the baseline of a delta-manifest.
471	466	*/
472	467	case 'B': {
473	468	if( p->zBaseline ) SYNTAX("more than one B-card");
474	469	p->zBaseline = next_token(&x, &sz);
475		- if( p->zBaseline==0 ) SYNTAX("missing UUID on B-card");
476		- if( sz!=UUID_SIZE \|\| !validate16(p->zBaseline, UUID_SIZE) ){
477		- SYNTAX("invalid UUID on B-card");
	470	+ if( p->zBaseline==0 ) SYNTAX("missing hash on B-card");
	471	+ if( !hname_validate(p->zBaseline,sz) ){
	472	+ SYNTAX("invalid hash on B-card");
478	473	}
479	474	break;
480	475	}
481	476
482	477
		@@ -522,12 +517,12 @@
522	517	case 'E': {
523	518	if( p->rEventDate>0.0 ) SYNTAX("more than one E-card");
524	519	p->rEventDate = db_double(0.0,"SELECT julianday(%Q)", next_token(&x,0));
525	520	if( p->rEventDate<=0.0 ) SYNTAX("malformed date on E-card");
526	521	p->zEventId = next_token(&x, &sz);
527		- if( sz!=UUID_SIZE \|\| !validate16(p->zEventId, UUID_SIZE) ){
528		- SYNTAX("malformed UUID on E-card");
	522	+ if( !hname_validate(p->zEventId, sz) ){
	523	+ SYNTAX("malformed hash on E-card");
529	524	}
530	525	break;
531	526	}
532	527
533	528	/*
		@@ -545,12 +540,13 @@
545	540	if( !file_is_simple_pathname(zName, 0) ){
546	541	SYNTAX("F-card filename is not a simple path");
547	542	}
548	543	zUuid = next_token(&x, &sz);
549	544	if( p->zBaseline==0 \|\| zUuid!=0 ){
550		- if( sz!=UUID_SIZE ) SYNTAX("F-card UUID is the wrong size");
551		- if( !validate16(zUuid, UUID_SIZE) ) SYNTAX("F-card UUID invalid");
	545	+ if( !hname_validate(zUuid,sz) ){
	546	+ SYNTAX("F-card hash invalid");
	547	+ }
552	548	}
553	549	zPerm = next_token(&x,0);
554	550	zPriorName = next_token(&x,0);
555	551	if( zPriorName ){
556	552	defossilize(zPriorName);
		@@ -636,20 +632,21 @@
636	632	}
637	633	break;
638	634	}
639	635
640	636	/*
641		- ** M <uuid>
	637	+ ** M <hash>
642	638	**
643		- ** An M-line identifies another artifact by its UUID. M-lines
	639	+ ** An M-line identifies another artifact by its hash. M-lines
644	640	** occur in clusters only.
645	641	*/
646	642	case 'M': {
647	643	zUuid = next_token(&x, &sz);
648		- if( zUuid==0 ) SYNTAX("missing UUID on M-card");
649		- if( sz!=UUID_SIZE ) SYNTAX("wrong size for UUID on M-card");
650		- if( !validate16(zUuid, UUID_SIZE) ) SYNTAX("UUID invalid on M-card");
	644	+ if( zUuid==0 ) SYNTAX("missing hash on M-card");
	645	+ if( !hname_validate(zUuid,sz) ){
	646	+ SYNTAX("Invalid hash on M-card");
	647	+ }
651	648	if( p->nCChild>=p->nCChildAlloc ){
652	649	p->nCChildAlloc = p->nCChildAlloc*2 + 10;
653	650	p->azCChild = fossil_realloc(p->azCChild
654	651	, p->nCChildAlloc*sizeof(p->azCChild[0]) );
655	652	}
		@@ -683,12 +680,13 @@
683	680	** check-in historically has an empty P-card, so empty P-cards
684	681	** must be accepted.
685	682	*/
686	683	case 'P': {
687	684	while( (zUuid = next_token(&x, &sz))!=0 ){
688		- if( sz!=UUID_SIZE ) SYNTAX("wrong size UUID on P-card");
689		- if( !validate16(zUuid, UUID_SIZE) )SYNTAX("invalid UUID on P-card");
	685	+ if( !hname_validate(zUuid, sz) ){
	686	+ SYNTAX("invalid hash on P-card");
	687	+ }
690	688	if( p->nParent>=p->nParentAlloc ){
691	689	p->nParentAlloc = p->nParentAlloc*2 + 5;
692	690	p->azParent = fossil_realloc(p->azParent,
693	691	p->nParentAllocsizeof(char));
694	692	}
		@@ -703,29 +701,25 @@
703	701	**
704	702	** Specify one or a range of check-ins that are cherrypicked into
705	703	** this check-in ("+") or backed out of this check-in ("-").
706	704	*/
707	705	case 'Q': {
708		- if( (zUuid=next_token(&x, &sz))==0 ) SYNTAX("missing UUID on Q-card");
709		- if( sz!=UUID_SIZE+1 ) SYNTAX("wrong size UUID on Q-card");
	706	+ if( (zUuid=next_token(&x, &sz))==0 ) SYNTAX("missing hash on Q-card");
710	707	if( zUuid[0]!='+' && zUuid[0]!='-' ){
711	708	SYNTAX("Q-card does not begin with '+' or '-'");
712	709	}
713		- if( !validate16(&zUuid[1], UUID_SIZE) ){
714		- SYNTAX("invalid UUID on Q-card");
	710	+ if( !hname_validate(&zUuid[1], sz-1) ){
	711	+ SYNTAX("invalid hash on Q-card");
715	712	}
716	713	n = p->nCherrypick;
717	714	p->nCherrypick++;
718	715	p->aCherrypick = fossil_realloc(p->aCherrypick,
719	716	p->nCherrypick*sizeof(p->aCherrypick[0]));
720	717	p->aCherrypick[n].zCPTarget = zUuid;
721	718	p->aCherrypick[n].zCPBase = zUuid = next_token(&x, &sz);
722		- if( zUuid ){
723		- if( sz!=UUID_SIZE ) SYNTAX("wrong size second UUID in Q-card");
724		- if( !validate16(zUuid, UUID_SIZE) ){
725		- SYNTAX("invalid second UUID on Q-card");
726		- }
	719	+ if( zUuid && !hname_validate(zUuid,sz) ){
	720	+ SYNTAX("invalid second hash on Q-card");
727	721	}
728	722	break;
729	723	}
730	724
731	725	/*
		@@ -760,32 +754,32 @@
760	754	case 'T': {
761	755	char zName, zValue;
762	756	zName = next_token(&x, 0);
763	757	if( zName==0 ) SYNTAX("missing name on T-card");
764	758	zUuid = next_token(&x, &sz);
765		- if( zUuid==0 ) SYNTAX("missing UUID on T-card");
	759	+ if( zUuid==0 ) SYNTAX("missing artifact hash on T-card");
766	760	zValue = next_token(&x, 0);
767	761	if( zValue ) defossilize(zValue);
768		- if( sz==UUID_SIZE && validate16(zUuid, UUID_SIZE) ){
769		- /* A valid uuid */
	762	+ if( hname_validate(zUuid, sz) ){
	763	+ /* A valid artifact hash */
770	764	if( p->zEventId ) SYNTAX("non-self-referential T-card in event");
771	765	}else if( sz==1 && zUuid[0]=='*' ){
772	766	zUuid = 0;
773	767	hasSelfRefTag = 1;
774	768	if( p->zEventId && zName[0]!='+' ){
775	769	SYNTAX("propagating T-card in event");
776	770	}
777	771	}else{
778		- SYNTAX("malformed UUID on T-card");
	772	+ SYNTAX("malformed artifact hash on T-card");
779	773	}
780	774	defossilize(zName);
781	775	if( zName[0]!='-' && zName[0]!='+' && zName[0]!='*' ){
782	776	SYNTAX("T-card name does not begin with '-', '+', or '*'");
783	777	}
784	778	if( validate16(&zName[1], strlen(&zName[1])) ){
785		- /* Do not allow tags whose names look like UUIDs */
786		- SYNTAX("T-card name looks like a UUID");
	779	+ /* Do not allow tags whose names look like a hash */
	780	+ SYNTAX("T-card name looks like a hexadecimal hash");
787	781	}
788	782	if( p->nTag>=p->nTagAlloc ){
789	783	p->nTagAlloc = p->nTagAlloc*2 + 10;
790	784	p->aTag = fossil_realloc(p->aTag, p->nTagAlloc*sizeof(p->aTag[0]) );
791	785	}
		@@ -951,17 +945,19 @@
951	945	if( !seenZ ) SYNTAX("missing Z-card on control");
952	946	p->type = CFTYPE_CONTROL;
953	947	}
954	948	md5sum_init();
955	949	if( !isRepeat ) g.parseCnt[p->type]++;
956		- blob_reset(&bUuid);
957	950	return p;
958	951
959	952	manifest_syntax_error:
960		- if(bUuid.nUsed){
961		- blob_appendf(pErr, "manifest [%.40s] ", blob_str(&bUuid));
962		- blob_reset(&bUuid);
	953	+ {
	954	+ char *zUuid = db_text(0, "SELECT uuid FROM blob WHERE rid=%d", rid);
	955	+ if( zUuid ){
	956	+ blob_appendf(pErr, "manifest [%s] ", zUuid);
	957	+ fossil_free(zUuid);
	958	+ }
963	959	}
964	960	if( zErr ){
965	961	blob_appendf(pErr, "line %d: %s", lineNo, zErr);
966	962	}else{
967	963	blob_appendf(pErr, "unknown error on line %d", lineNo);
		@@ -1193,13 +1189,13 @@
1193	1189	** then the mlink entry is only created if there is already an mlink
1194	1190	** from primary parent for the same file.
1195	1191	*/
1196	1192	static void add_one_mlink(
1197	1193	int pmid, /* The parent manifest */
1198		- const char zFromUuid, / UUID for content in parent */
	1194	+ const char zFromUuid, / Artifact hash for content in parent */
1199	1195	int mid, /* The record ID of the manifest */
1200		- const char zToUuid, / UUID for content in child */
	1196	+ const char zToUuid, / artifact hash for content in child */
1201	1197	const char zFilename, / Filename */
1202	1198	const char zPrior, / Previous filename. NULL if unchanged */
1203	1199	int isPublic, /* True if mid is not a private manifest */
1204	1200	int isPrimary, /* pmid is the primary parent of mid */
1205	1201	int mperm /* 1: exec, 2: symlink */
		@@ -1545,22 +1541,22 @@
1545	1541	}
1546	1542	}
1547	1543
1548	1544	/*
1549	1545	** For a check-in with RID "rid" that has nParent parent check-ins given
1550		-** by the UUIDs in azParent[], create all appropriate plink and mlink table
	1546	+** by the hashes in azParent[], create all appropriate plink and mlink table
1551	1547	** entries.
1552	1548	**
1553		-** The primary parent is the first UUID on the azParent[] list.
	1549	+** The primary parent is the first hash on the azParent[] list.
1554	1550	**
1555	1551	** Return the RID of the primary parent.
1556	1552	*/
1557	1553	static int manifest_add_checkin_linkages(
1558	1554	int rid, /* The RID of the check-in */
1559	1555	Manifest p, / Manifest for this check-in */
1560	1556	int nParent, /* Number of parents for this check-in */
1561		- char *azParent / UUIDs for each parent */
	1557	+ char *azParent / hashes for each parent */
1562	1558	){
1563	1559	int i;
1564	1560	int parentid = 0;
1565	1561	char zBaseId[30]; /* Baseline manifest RID for deltas. "NULL" otherwise */
1566	1562	Stmt q;
		@@ -1611,33 +1607,36 @@
1611	1607	return parentid;
1612	1608	}
1613	1609
1614	1610	/*
1615	1611	** There exists a "parent" tag against checkin rid that has value zValue.
1616		-** If value is well-formed (meaning that it is a list of UUIDs), then use
	1612	+** If value is well-formed (meaning that it is a list of hashes), then use
1617	1613	** zValue to reparent check-in rid.
1618	1614	*/
1619	1615	void manifest_reparent_checkin(int rid, const char *zValue){
1620		- int nParent;
	1616	+ int nParent = 0;
1621	1617	char *zCopy = 0;
1622	1618	char **azParent = 0;
1623	1619	Manifest *p = 0;
1624		- int i;
	1620	+ int i, j;
1625	1621	int n = (int)strlen(zValue);
1626		- nParent = (n+1)/(UUID_SIZE+1);
1627		- if( nParent*(UUID_SIZE+1) - 1 !=n ) return;
1628		- if( nParent<1 ) return;
	1622	+ int mxParent = (n+1)/(HNAME_MIN+1);
	1623	+
	1624	+ if( mxParent<1 ) return;
1629	1625	zCopy = fossil_strdup(zValue);
1630		- azParent = fossil_malloc( sizeof(azParent[0])*nParent );
1631		- for(i=0; i<nParent; i++){
1632		- azParent[i] = &zCopy[i*(UUID_SIZE+1)];
1633		- if( i<nParent-1 && azParent[i][UUID_SIZE]!=' ' ) break;
1634		- azParent[i][UUID_SIZE] = 0;
1635		- if( !validate16(azParent[i],UUID_SIZE) ) break;
1636		- }
1637		- if( i==nParent
1638		- && !db_exists("SELECT 1 FROM plink WHERE cid=%d AND pid=%d",
	1626	+ azParent = fossil_malloc( sizeof(azParent[0])*mxParent );
	1627	+ for(nParent=0, i=0; zCopy[i]; i++){
	1628	+ char *z = &zCopy[i];
	1629	+ azParent[nParent++] = z;
	1630	+ if( nParent>mxParent ) goto reparent_abort;
	1631	+ for(j=HNAME_MIN; z[j]>' '; j++){}
	1632	+ if( !hname_validate(z, j) ) goto reparent_abort;
	1633	+ if( z[j]==0 ) break;
	1634	+ z[j] = 0;
	1635	+ i += j;
	1636	+ }
	1637	+ if( !db_exists("SELECT 1 FROM plink WHERE cid=%d AND pid=%d",
1639	1638	rid, uuid_to_rid(azParent[0],0))
1640	1639	){
1641	1640	p = manifest_get(rid, CFTYPE_MANIFEST, 0);
1642	1641	}
1643	1642	if( p!=0 ){
		@@ -1647,10 +1646,11 @@
1647	1646	rid, rid
1648	1647	);
1649	1648	manifest_add_checkin_linkages(rid,p,nParent,azParent);
1650	1649	}
1651	1650	manifest_destroy(p);
	1651	+reparent_abort:
1652	1652	fossil_free(azParent);
1653	1653	fossil_free(zCopy);
1654	1654	}
1655	1655
1656	1656	/*
		@@ -2280,11 +2280,11 @@
2280	2280	int branchMove = 0;
2281	2281	blob_zero(&comment);
2282	2282	if( p->zComment ){
2283	2283	blob_appendf(&comment, " %s.", p->zComment);
2284	2284	}
2285		- /* Next loop expects tags to be sorted on UUID, so sort it. */
	2285	+ /* Next loop expects tags to be sorted on hash, so sort it. */
2286	2286	qsort(p->aTag, p->nTag, sizeof(p->aTag[0]), tag_compare);
2287	2287	for(i=0; i<p->nTag; i++){
2288	2288	zTagUuid = p->aTag[i].zUuid;
2289	2289	if( !zTagUuid ) continue;
2290	2290	if( i==0 \|\| fossil_strcmp(zTagUuid, p->aTag[i-1].zUuid)!=0 ){
2291	2291

	--- src/manifest.c
	+++ src/manifest.c
	@@ -54,11 +54,11 @@
54	/*
55	** A single F-card within a manifest
56	*/
57	struct ManifestFile {
58	char zName; / Name of a file */
59	char zUuid; / UUID of the file */
60	char zPerm; / File permissions */
61	char zPrior; / Prior name if the name was changed */
62	};
63
64
	@@ -77,35 +77,35 @@
77	char zRepoCksum; / MD5 checksum of the baseline content. R card. */
78	char zWiki; / Text of the wiki page. W card. */
79	char zWikiTitle; / Name of the wiki page. L card. */
80	char zMimetype; / Mime type of wiki or comment text. N card. */
81	double rEventDate; /* Date of an event. E card. */
82	char zEventId; / UUID for an event. E card. */
83	char zTicketUuid; / UUID for a ticket. K card. */
84	char zAttachName; / Filename of an attachment. A card. */
85	char zAttachSrc; / UUID of document being attached. A card. */
86	char zAttachTarget; / Ticket or wiki that attachment applies to. A card */
87	int nFile; /* Number of F cards */
88	int nFileAlloc; /* Slots allocated in aFile[] */
89	int iFile; /* Index of current file in iterator */
90	ManifestFile aFile; / One entry for each F-card */
91	int nParent; /* Number of parents. */
92	int nParentAlloc; /* Slots allocated in azParent[] */
93	char *azParent; / UUIDs of parents. One for each P card argument */
94	int nCherrypick; /* Number of entries in aCherrypick[] */
95	struct {
96	char zCPTarget; / UUID of cherry-picked version w/ +\|- prefix */
97	char zCPBase; / UUID of cherry-pick baseline. NULL for singletons */
98	} *aCherrypick;
99	int nCChild; /* Number of cluster children */
100	int nCChildAlloc; /* Number of closts allocated in azCChild[] */
101	char *azCChild; / UUIDs of referenced objects in a cluster. M cards */
102	int nTag; /* Number of T Cards */
103	int nTagAlloc; /* Slots allocated in aTag[] */
104	struct TagType {
105	char zName; / Name of the tag */
106	char zUuid; / UUID that the tag is applied to */
107	char zValue; / Value if the tag is really a property */
108	} aTag; / One for each T card */
109	int nField; /* Number of J cards */
110	int nFieldAlloc; /* Slots allocated in aField[] */
111	struct {
	@@ -327,26 +327,27 @@
327	** takes over the input blob and will free it when the
328	** Manifest object is freed. Zeros are inserted into the blob
329	** as string terminators so that blob should not be used again.
330	**
331	** Return a pointer to an allocated Manifest object if the content
332	** really is a control file of some kind. This object needs to be
333	** freed by a subsequent call to manifest_destroy(). Return NULL
334	** if there are syntax errors.

335	**
336	** This routine is strict about the format of a control file.
337	** The format must match exactly or else it is rejected. This
338	** rule minimizes the risk that a content file will be mistaken
339	** for a control file simply because they look the same.
340	**
341	** The pContent is reset. If a pointer is returned, then pContent will
342	** be reset when the Manifest object is cleared. If NULL is
343	** returned then the Manifest object is cleared automatically
344	** and pContent is reset before the return.
345	**
346	** The entire file can be PGP clear-signed. The signature is ignored.
347	** The file consists of zero or more cards, one card per line.
348	** (Except: the content of the W card can extend of multiple lines.)
349	** Each card is divided into tokens by a single space character.
350	** The first token is a single upper-case letter which is the card type.
351	** The card type determines the other parameters to the card.
352	** Cards must occur in lexicographical order.
	@@ -361,11 +362,10 @@
361	char *z;
362	int n;
363	char *zUuid;
364	int sz = 0;
365	int isRepeat, hasSelfRefTag = 0;
366	Blob bUuid = BLOB_INITIALIZER;
367	static Bag seen;
368	const char *zErr = 0;
369
370	if( rid==0 ){
371	isRepeat = 1;
	@@ -374,11 +374,11 @@
374	}else{
375	isRepeat = 0;
376	bag_insert(&seen, rid);
377	}
378
379	/* Every control artifact ends with a '\n' character. Exit early
380	** if that is not the case for this artifact.
381	*/
382	if( !isRepeat ) g.parseCnt[0]++;
383	z = blob_materialize(pContent);
384	n = blob_size(pContent);
	@@ -406,15 +406,10 @@
406	blob_reset(pContent);
407	blob_appendf(pErr, "incorrect Z-card cksum");
408	return 0;
409	}
410
411	/* Store the UUID (before modifying the blob) only for error
412	** reporting purposes.
413	*/
414	sha1sum_blob(pContent, &bUuid);
415
416	/* Allocate a Manifest object to hold the parsed control artifact.
417	*/
418	p = fossil_malloc( sizeof(*p) );
419	memset(p, 0, sizeof(*p));
420	memcpy(&p->content, pContent, sizeof(p->content));
	@@ -449,15 +444,15 @@
449	defossilize(zName);
450	if( !file_is_simple_pathname(zName, 0) ){
451	SYNTAX("invalid filename on A-card");
452	}
453	defossilize(zTarget);
454	if( (nTarget!=UUID_SIZE \|\| !validate16(zTarget, UUID_SIZE))
455	&& !wiki_name_is_wellformed((const unsigned char *)zTarget) ){
456	SYNTAX("invalid target on A-card");
457	}
458	if( zSrc && (nSrc!=UUID_SIZE \|\| !validate16(zSrc, UUID_SIZE)) ){
459	SYNTAX("invalid source on A-card");
460	}
461	p->zAttachName = (char*)file_tail(zName);
462	p->zAttachSrc = zSrc;
463	p->zAttachTarget = zTarget;
	@@ -465,18 +460,18 @@
465	}
466
467	/*
468	** B <uuid>
469	**
470	** A B-line gives the UUID for the baseline of a delta-manifest.
471	*/
472	case 'B': {
473	if( p->zBaseline ) SYNTAX("more than one B-card");
474	p->zBaseline = next_token(&x, &sz);
475	if( p->zBaseline==0 ) SYNTAX("missing UUID on B-card");
476	if( sz!=UUID_SIZE \|\| !validate16(p->zBaseline, UUID_SIZE) ){
477	SYNTAX("invalid UUID on B-card");
478	}
479	break;
480	}
481
482
	@@ -522,12 +517,12 @@
522	case 'E': {
523	if( p->rEventDate>0.0 ) SYNTAX("more than one E-card");
524	p->rEventDate = db_double(0.0,"SELECT julianday(%Q)", next_token(&x,0));
525	if( p->rEventDate<=0.0 ) SYNTAX("malformed date on E-card");
526	p->zEventId = next_token(&x, &sz);
527	if( sz!=UUID_SIZE \|\| !validate16(p->zEventId, UUID_SIZE) ){
528	SYNTAX("malformed UUID on E-card");
529	}
530	break;
531	}
532
533	/*
	@@ -545,12 +540,13 @@
545	if( !file_is_simple_pathname(zName, 0) ){
546	SYNTAX("F-card filename is not a simple path");
547	}
548	zUuid = next_token(&x, &sz);
549	if( p->zBaseline==0 \|\| zUuid!=0 ){
550	if( sz!=UUID_SIZE ) SYNTAX("F-card UUID is the wrong size");
551	if( !validate16(zUuid, UUID_SIZE) ) SYNTAX("F-card UUID invalid");

552	}
553	zPerm = next_token(&x,0);
554	zPriorName = next_token(&x,0);
555	if( zPriorName ){
556	defossilize(zPriorName);
	@@ -636,20 +632,21 @@
636	}
637	break;
638	}
639
640	/*
641	** M <uuid>
642	**
643	** An M-line identifies another artifact by its UUID. M-lines
644	** occur in clusters only.
645	*/
646	case 'M': {
647	zUuid = next_token(&x, &sz);
648	if( zUuid==0 ) SYNTAX("missing UUID on M-card");
649	if( sz!=UUID_SIZE ) SYNTAX("wrong size for UUID on M-card");
650	if( !validate16(zUuid, UUID_SIZE) ) SYNTAX("UUID invalid on M-card");

651	if( p->nCChild>=p->nCChildAlloc ){
652	p->nCChildAlloc = p->nCChildAlloc*2 + 10;
653	p->azCChild = fossil_realloc(p->azCChild
654	, p->nCChildAlloc*sizeof(p->azCChild[0]) );
655	}
	@@ -683,12 +680,13 @@
683	** check-in historically has an empty P-card, so empty P-cards
684	** must be accepted.
685	*/
686	case 'P': {
687	while( (zUuid = next_token(&x, &sz))!=0 ){
688	if( sz!=UUID_SIZE ) SYNTAX("wrong size UUID on P-card");
689	if( !validate16(zUuid, UUID_SIZE) )SYNTAX("invalid UUID on P-card");

690	if( p->nParent>=p->nParentAlloc ){
691	p->nParentAlloc = p->nParentAlloc*2 + 5;
692	p->azParent = fossil_realloc(p->azParent,
693	p->nParentAllocsizeof(char));
694	}
	@@ -703,29 +701,25 @@
703	**
704	** Specify one or a range of check-ins that are cherrypicked into
705	** this check-in ("+") or backed out of this check-in ("-").
706	*/
707	case 'Q': {
708	if( (zUuid=next_token(&x, &sz))==0 ) SYNTAX("missing UUID on Q-card");
709	if( sz!=UUID_SIZE+1 ) SYNTAX("wrong size UUID on Q-card");
710	if( zUuid[0]!='+' && zUuid[0]!='-' ){
711	SYNTAX("Q-card does not begin with '+' or '-'");
712	}
713	if( !validate16(&zUuid[1], UUID_SIZE) ){
714	SYNTAX("invalid UUID on Q-card");
715	}
716	n = p->nCherrypick;
717	p->nCherrypick++;
718	p->aCherrypick = fossil_realloc(p->aCherrypick,
719	p->nCherrypick*sizeof(p->aCherrypick[0]));
720	p->aCherrypick[n].zCPTarget = zUuid;
721	p->aCherrypick[n].zCPBase = zUuid = next_token(&x, &sz);
722	if( zUuid ){
723	if( sz!=UUID_SIZE ) SYNTAX("wrong size second UUID in Q-card");
724	if( !validate16(zUuid, UUID_SIZE) ){
725	SYNTAX("invalid second UUID on Q-card");
726	}
727	}
728	break;
729	}
730
731	/*
	@@ -760,32 +754,32 @@
760	case 'T': {
761	char zName, zValue;
762	zName = next_token(&x, 0);
763	if( zName==0 ) SYNTAX("missing name on T-card");
764	zUuid = next_token(&x, &sz);
765	if( zUuid==0 ) SYNTAX("missing UUID on T-card");
766	zValue = next_token(&x, 0);
767	if( zValue ) defossilize(zValue);
768	if( sz==UUID_SIZE && validate16(zUuid, UUID_SIZE) ){
769	/* A valid uuid */
770	if( p->zEventId ) SYNTAX("non-self-referential T-card in event");
771	}else if( sz==1 && zUuid[0]=='*' ){
772	zUuid = 0;
773	hasSelfRefTag = 1;
774	if( p->zEventId && zName[0]!='+' ){
775	SYNTAX("propagating T-card in event");
776	}
777	}else{
778	SYNTAX("malformed UUID on T-card");
779	}
780	defossilize(zName);
781	if( zName[0]!='-' && zName[0]!='+' && zName[0]!='*' ){
782	SYNTAX("T-card name does not begin with '-', '+', or '*'");
783	}
784	if( validate16(&zName[1], strlen(&zName[1])) ){
785	/* Do not allow tags whose names look like UUIDs */
786	SYNTAX("T-card name looks like a UUID");
787	}
788	if( p->nTag>=p->nTagAlloc ){
789	p->nTagAlloc = p->nTagAlloc*2 + 10;
790	p->aTag = fossil_realloc(p->aTag, p->nTagAlloc*sizeof(p->aTag[0]) );
791	}
	@@ -951,17 +945,19 @@
951	if( !seenZ ) SYNTAX("missing Z-card on control");
952	p->type = CFTYPE_CONTROL;
953	}
954	md5sum_init();
955	if( !isRepeat ) g.parseCnt[p->type]++;
956	blob_reset(&bUuid);
957	return p;
958
959	manifest_syntax_error:
960	if(bUuid.nUsed){
961	blob_appendf(pErr, "manifest [%.40s] ", blob_str(&bUuid));
962	blob_reset(&bUuid);



963	}
964	if( zErr ){
965	blob_appendf(pErr, "line %d: %s", lineNo, zErr);
966	}else{
967	blob_appendf(pErr, "unknown error on line %d", lineNo);
	@@ -1193,13 +1189,13 @@
1193	** then the mlink entry is only created if there is already an mlink
1194	** from primary parent for the same file.
1195	*/
1196	static void add_one_mlink(
1197	int pmid, /* The parent manifest */
1198	const char zFromUuid, / UUID for content in parent */
1199	int mid, /* The record ID of the manifest */
1200	const char zToUuid, / UUID for content in child */
1201	const char zFilename, / Filename */
1202	const char zPrior, / Previous filename. NULL if unchanged */
1203	int isPublic, /* True if mid is not a private manifest */
1204	int isPrimary, /* pmid is the primary parent of mid */
1205	int mperm /* 1: exec, 2: symlink */
	@@ -1545,22 +1541,22 @@
1545	}
1546	}
1547
1548	/*
1549	** For a check-in with RID "rid" that has nParent parent check-ins given
1550	** by the UUIDs in azParent[], create all appropriate plink and mlink table
1551	** entries.
1552	**
1553	** The primary parent is the first UUID on the azParent[] list.
1554	**
1555	** Return the RID of the primary parent.
1556	*/
1557	static int manifest_add_checkin_linkages(
1558	int rid, /* The RID of the check-in */
1559	Manifest p, / Manifest for this check-in */
1560	int nParent, /* Number of parents for this check-in */
1561	char *azParent / UUIDs for each parent */
1562	){
1563	int i;
1564	int parentid = 0;
1565	char zBaseId[30]; /* Baseline manifest RID for deltas. "NULL" otherwise */
1566	Stmt q;
	@@ -1611,33 +1607,36 @@
1611	return parentid;
1612	}
1613
1614	/*
1615	** There exists a "parent" tag against checkin rid that has value zValue.
1616	** If value is well-formed (meaning that it is a list of UUIDs), then use
1617	** zValue to reparent check-in rid.
1618	*/
1619	void manifest_reparent_checkin(int rid, const char *zValue){
1620	int nParent;
1621	char *zCopy = 0;
1622	char **azParent = 0;
1623	Manifest *p = 0;
1624	int i;
1625	int n = (int)strlen(zValue);
1626	nParent = (n+1)/(UUID_SIZE+1);
1627	if( nParent*(UUID_SIZE+1) - 1 !=n ) return;
1628	if( nParent<1 ) return;
1629	zCopy = fossil_strdup(zValue);
1630	azParent = fossil_malloc( sizeof(azParent[0])*nParent );
1631	for(i=0; i<nParent; i++){
1632	azParent[i] = &zCopy[i*(UUID_SIZE+1)];
1633	if( i<nParent-1 && azParent[i][UUID_SIZE]!=' ' ) break;
1634	azParent[i][UUID_SIZE] = 0;
1635	if( !validate16(azParent[i],UUID_SIZE) ) break;
1636	}
1637	if( i==nParent
1638	&& !db_exists("SELECT 1 FROM plink WHERE cid=%d AND pid=%d",



1639	rid, uuid_to_rid(azParent[0],0))
1640	){
1641	p = manifest_get(rid, CFTYPE_MANIFEST, 0);
1642	}
1643	if( p!=0 ){
	@@ -1647,10 +1646,11 @@
1647	rid, rid
1648	);
1649	manifest_add_checkin_linkages(rid,p,nParent,azParent);
1650	}
1651	manifest_destroy(p);

1652	fossil_free(azParent);
1653	fossil_free(zCopy);
1654	}
1655
1656	/*
	@@ -2280,11 +2280,11 @@
2280	int branchMove = 0;
2281	blob_zero(&comment);
2282	if( p->zComment ){
2283	blob_appendf(&comment, " %s.", p->zComment);
2284	}
2285	/* Next loop expects tags to be sorted on UUID, so sort it. */
2286	qsort(p->aTag, p->nTag, sizeof(p->aTag[0]), tag_compare);
2287	for(i=0; i<p->nTag; i++){
2288	zTagUuid = p->aTag[i].zUuid;
2289	if( !zTagUuid ) continue;
2290	if( i==0 \|\| fossil_strcmp(zTagUuid, p->aTag[i-1].zUuid)!=0 ){
2291

	--- src/manifest.c
	+++ src/manifest.c
	@@ -54,11 +54,11 @@
54	/*
55	** A single F-card within a manifest
56	*/
57	struct ManifestFile {
58	char zName; / Name of a file */
59	char zUuid; / Artifact hash for the file */
60	char zPerm; / File permissions */
61	char zPrior; / Prior name if the name was changed */
62	};
63
64
	@@ -77,35 +77,35 @@
77	char zRepoCksum; / MD5 checksum of the baseline content. R card. */
78	char zWiki; / Text of the wiki page. W card. */
79	char zWikiTitle; / Name of the wiki page. L card. */
80	char zMimetype; / Mime type of wiki or comment text. N card. */
81	double rEventDate; /* Date of an event. E card. */
82	char zEventId; / Artifact hash for an event. E card. */
83	char zTicketUuid; / UUID for a ticket. K card. */
84	char zAttachName; / Filename of an attachment. A card. */
85	char zAttachSrc; / Artifact hash for document being attached. A card. */
86	char zAttachTarget; / Ticket or wiki that attachment applies to. A card */
87	int nFile; /* Number of F cards */
88	int nFileAlloc; /* Slots allocated in aFile[] */
89	int iFile; /* Index of current file in iterator */
90	ManifestFile aFile; / One entry for each F-card */
91	int nParent; /* Number of parents. */
92	int nParentAlloc; /* Slots allocated in azParent[] */
93	char *azParent; / Hashes of parents. One for each P card argument */
94	int nCherrypick; /* Number of entries in aCherrypick[] */
95	struct {
96	char zCPTarget; / Hash for cherry-picked version w/ +\|- prefix */
97	char zCPBase; / Hash for cherry-pick baseline. NULL for singletons */
98	} *aCherrypick;
99	int nCChild; /* Number of cluster children */
100	int nCChildAlloc; /* Number of closts allocated in azCChild[] */
101	char *azCChild; / Hashes of referenced objects in a cluster. M cards */
102	int nTag; /* Number of T Cards */
103	int nTagAlloc; /* Slots allocated in aTag[] */
104	struct TagType {
105	char zName; / Name of the tag */
106	char zUuid; / Hash of artifact that the tag is applied to */
107	char zValue; / Value if the tag is really a property */
108	} aTag; / One for each T card */
109	int nField; /* Number of J cards */
110	int nFieldAlloc; /* Slots allocated in aField[] */
111	struct {
	@@ -327,26 +327,27 @@
327	** takes over the input blob and will free it when the
328	** Manifest object is freed. Zeros are inserted into the blob
329	** as string terminators so that blob should not be used again.
330	**
331	** Return a pointer to an allocated Manifest object if the content
332	** really is a structural artifact of some kind. The returned Manifest
333	** object needs to be freed by a subsequent call to manifest_destroy().
334	** Return NULL if there are syntax errors or if the input blob does
335	** not describe a valid structural artifact.
336	**
337	** This routine is strict about the format of a structural artifacts.
338	** The format must match exactly or else it is rejected. This
339	** rule minimizes the risk that a content artifact will be mistaken
340	** for a structural artifact simply because they look the same.
341	**
342	** The pContent is reset. If a pointer is returned, then pContent will
343	** be reset when the Manifest object is cleared. If NULL is
344	** returned then the Manifest object is cleared automatically
345	** and pContent is reset before the return.
346	**
347	** The entire input blob can be PGP clear-signed. The signature is ignored.
348	** The artifact consists of zero or more cards, one card per line.
349	** (Except: the content of the W card can extend of multiple lines.)
350	** Each card is divided into tokens by a single space character.
351	** The first token is a single upper-case letter which is the card type.
352	** The card type determines the other parameters to the card.
353	** Cards must occur in lexicographical order.
	@@ -361,11 +362,10 @@
362	char *z;
363	int n;
364	char *zUuid;
365	int sz = 0;
366	int isRepeat, hasSelfRefTag = 0;

367	static Bag seen;
368	const char *zErr = 0;
369
370	if( rid==0 ){
371	isRepeat = 1;
	@@ -374,11 +374,11 @@
374	}else{
375	isRepeat = 0;
376	bag_insert(&seen, rid);
377	}
378
379	/* Every structural artifact ends with a '\n' character. Exit early
380	** if that is not the case for this artifact.
381	*/
382	if( !isRepeat ) g.parseCnt[0]++;
383	z = blob_materialize(pContent);
384	n = blob_size(pContent);
	@@ -406,15 +406,10 @@
406	blob_reset(pContent);
407	blob_appendf(pErr, "incorrect Z-card cksum");
408	return 0;
409	}
410





411	/* Allocate a Manifest object to hold the parsed control artifact.
412	*/
413	p = fossil_malloc( sizeof(*p) );
414	memset(p, 0, sizeof(*p));
415	memcpy(&p->content, pContent, sizeof(p->content));
	@@ -449,15 +444,15 @@
444	defossilize(zName);
445	if( !file_is_simple_pathname(zName, 0) ){
446	SYNTAX("invalid filename on A-card");
447	}
448	defossilize(zTarget);
449	if( !hname_validate(zTarget,nTarget)
450	&& !wiki_name_is_wellformed((const unsigned char *)zTarget) ){
451	SYNTAX("invalid target on A-card");
452	}
453	if( zSrc && !hname_validate(zSrc,nSrc) ){
454	SYNTAX("invalid source on A-card");
455	}
456	p->zAttachName = (char*)file_tail(zName);
457	p->zAttachSrc = zSrc;
458	p->zAttachTarget = zTarget;
	@@ -465,18 +460,18 @@
460	}
461
462	/*
463	** B <uuid>
464	**
465	** A B-line gives the artifact hash for the baseline of a delta-manifest.
466	*/
467	case 'B': {
468	if( p->zBaseline ) SYNTAX("more than one B-card");
469	p->zBaseline = next_token(&x, &sz);
470	if( p->zBaseline==0 ) SYNTAX("missing hash on B-card");
471	if( !hname_validate(p->zBaseline,sz) ){
472	SYNTAX("invalid hash on B-card");
473	}
474	break;
475	}
476
477
	@@ -522,12 +517,12 @@
517	case 'E': {
518	if( p->rEventDate>0.0 ) SYNTAX("more than one E-card");
519	p->rEventDate = db_double(0.0,"SELECT julianday(%Q)", next_token(&x,0));
520	if( p->rEventDate<=0.0 ) SYNTAX("malformed date on E-card");
521	p->zEventId = next_token(&x, &sz);
522	if( !hname_validate(p->zEventId, sz) ){
523	SYNTAX("malformed hash on E-card");
524	}
525	break;
526	}
527
528	/*
	@@ -545,12 +540,13 @@
540	if( !file_is_simple_pathname(zName, 0) ){
541	SYNTAX("F-card filename is not a simple path");
542	}
543	zUuid = next_token(&x, &sz);
544	if( p->zBaseline==0 \|\| zUuid!=0 ){
545	if( !hname_validate(zUuid,sz) ){
546	SYNTAX("F-card hash invalid");
547	}
548	}
549	zPerm = next_token(&x,0);
550	zPriorName = next_token(&x,0);
551	if( zPriorName ){
552	defossilize(zPriorName);
	@@ -636,20 +632,21 @@
632	}
633	break;
634	}
635
636	/*
637	** M <hash>
638	**
639	** An M-line identifies another artifact by its hash. M-lines
640	** occur in clusters only.
641	*/
642	case 'M': {
643	zUuid = next_token(&x, &sz);
644	if( zUuid==0 ) SYNTAX("missing hash on M-card");
645	if( !hname_validate(zUuid,sz) ){
646	SYNTAX("Invalid hash on M-card");
647	}
648	if( p->nCChild>=p->nCChildAlloc ){
649	p->nCChildAlloc = p->nCChildAlloc*2 + 10;
650	p->azCChild = fossil_realloc(p->azCChild
651	, p->nCChildAlloc*sizeof(p->azCChild[0]) );
652	}
	@@ -683,12 +680,13 @@
680	** check-in historically has an empty P-card, so empty P-cards
681	** must be accepted.
682	*/
683	case 'P': {
684	while( (zUuid = next_token(&x, &sz))!=0 ){
685	if( !hname_validate(zUuid, sz) ){
686	SYNTAX("invalid hash on P-card");
687	}
688	if( p->nParent>=p->nParentAlloc ){
689	p->nParentAlloc = p->nParentAlloc*2 + 5;
690	p->azParent = fossil_realloc(p->azParent,
691	p->nParentAllocsizeof(char));
692	}
	@@ -703,29 +701,25 @@
701	**
702	** Specify one or a range of check-ins that are cherrypicked into
703	** this check-in ("+") or backed out of this check-in ("-").
704	*/
705	case 'Q': {
706	if( (zUuid=next_token(&x, &sz))==0 ) SYNTAX("missing hash on Q-card");

707	if( zUuid[0]!='+' && zUuid[0]!='-' ){
708	SYNTAX("Q-card does not begin with '+' or '-'");
709	}
710	if( !hname_validate(&zUuid[1], sz-1) ){
711	SYNTAX("invalid hash on Q-card");
712	}
713	n = p->nCherrypick;
714	p->nCherrypick++;
715	p->aCherrypick = fossil_realloc(p->aCherrypick,
716	p->nCherrypick*sizeof(p->aCherrypick[0]));
717	p->aCherrypick[n].zCPTarget = zUuid;
718	p->aCherrypick[n].zCPBase = zUuid = next_token(&x, &sz);
719	if( zUuid && !hname_validate(zUuid,sz) ){
720	SYNTAX("invalid second hash on Q-card");



721	}
722	break;
723	}
724
725	/*
	@@ -760,32 +754,32 @@
754	case 'T': {
755	char zName, zValue;
756	zName = next_token(&x, 0);
757	if( zName==0 ) SYNTAX("missing name on T-card");
758	zUuid = next_token(&x, &sz);
759	if( zUuid==0 ) SYNTAX("missing artifact hash on T-card");
760	zValue = next_token(&x, 0);
761	if( zValue ) defossilize(zValue);
762	if( hname_validate(zUuid, sz) ){
763	/* A valid artifact hash */
764	if( p->zEventId ) SYNTAX("non-self-referential T-card in event");
765	}else if( sz==1 && zUuid[0]=='*' ){
766	zUuid = 0;
767	hasSelfRefTag = 1;
768	if( p->zEventId && zName[0]!='+' ){
769	SYNTAX("propagating T-card in event");
770	}
771	}else{
772	SYNTAX("malformed artifact hash on T-card");
773	}
774	defossilize(zName);
775	if( zName[0]!='-' && zName[0]!='+' && zName[0]!='*' ){
776	SYNTAX("T-card name does not begin with '-', '+', or '*'");
777	}
778	if( validate16(&zName[1], strlen(&zName[1])) ){
779	/* Do not allow tags whose names look like a hash */
780	SYNTAX("T-card name looks like a hexadecimal hash");
781	}
782	if( p->nTag>=p->nTagAlloc ){
783	p->nTagAlloc = p->nTagAlloc*2 + 10;
784	p->aTag = fossil_realloc(p->aTag, p->nTagAlloc*sizeof(p->aTag[0]) );
785	}
	@@ -951,17 +945,19 @@
945	if( !seenZ ) SYNTAX("missing Z-card on control");
946	p->type = CFTYPE_CONTROL;
947	}
948	md5sum_init();
949	if( !isRepeat ) g.parseCnt[p->type]++;

950	return p;
951
952	manifest_syntax_error:
953	{
954	char *zUuid = db_text(0, "SELECT uuid FROM blob WHERE rid=%d", rid);
955	if( zUuid ){
956	blob_appendf(pErr, "manifest [%s] ", zUuid);
957	fossil_free(zUuid);
958	}
959	}
960	if( zErr ){
961	blob_appendf(pErr, "line %d: %s", lineNo, zErr);
962	}else{
963	blob_appendf(pErr, "unknown error on line %d", lineNo);
	@@ -1193,13 +1189,13 @@
1189	** then the mlink entry is only created if there is already an mlink
1190	** from primary parent for the same file.
1191	*/
1192	static void add_one_mlink(
1193	int pmid, /* The parent manifest */
1194	const char zFromUuid, / Artifact hash for content in parent */
1195	int mid, /* The record ID of the manifest */
1196	const char zToUuid, / artifact hash for content in child */
1197	const char zFilename, / Filename */
1198	const char zPrior, / Previous filename. NULL if unchanged */
1199	int isPublic, /* True if mid is not a private manifest */
1200	int isPrimary, /* pmid is the primary parent of mid */
1201	int mperm /* 1: exec, 2: symlink */
	@@ -1545,22 +1541,22 @@
1541	}
1542	}
1543
1544	/*
1545	** For a check-in with RID "rid" that has nParent parent check-ins given
1546	** by the hashes in azParent[], create all appropriate plink and mlink table
1547	** entries.
1548	**
1549	** The primary parent is the first hash on the azParent[] list.
1550	**
1551	** Return the RID of the primary parent.
1552	*/
1553	static int manifest_add_checkin_linkages(
1554	int rid, /* The RID of the check-in */
1555	Manifest p, / Manifest for this check-in */
1556	int nParent, /* Number of parents for this check-in */
1557	char *azParent / hashes for each parent */
1558	){
1559	int i;
1560	int parentid = 0;
1561	char zBaseId[30]; /* Baseline manifest RID for deltas. "NULL" otherwise */
1562	Stmt q;
	@@ -1611,33 +1607,36 @@
1607	return parentid;
1608	}
1609
1610	/*
1611	** There exists a "parent" tag against checkin rid that has value zValue.
1612	** If value is well-formed (meaning that it is a list of hashes), then use
1613	** zValue to reparent check-in rid.
1614	*/
1615	void manifest_reparent_checkin(int rid, const char *zValue){
1616	int nParent = 0;
1617	char *zCopy = 0;
1618	char **azParent = 0;
1619	Manifest *p = 0;
1620	int i, j;
1621	int n = (int)strlen(zValue);
1622	int mxParent = (n+1)/(HNAME_MIN+1);
1623
1624	if( mxParent<1 ) return;
1625	zCopy = fossil_strdup(zValue);
1626	azParent = fossil_malloc( sizeof(azParent[0])*mxParent );
1627	for(nParent=0, i=0; zCopy[i]; i++){
1628	char *z = &zCopy[i];
1629	azParent[nParent++] = z;
1630	if( nParent>mxParent ) goto reparent_abort;
1631	for(j=HNAME_MIN; z[j]>' '; j++){}
1632	if( !hname_validate(z, j) ) goto reparent_abort;
1633	if( z[j]==0 ) break;
1634	z[j] = 0;
1635	i += j;
1636	}
1637	if( !db_exists("SELECT 1 FROM plink WHERE cid=%d AND pid=%d",
1638	rid, uuid_to_rid(azParent[0],0))
1639	){
1640	p = manifest_get(rid, CFTYPE_MANIFEST, 0);
1641	}
1642	if( p!=0 ){
	@@ -1647,10 +1646,11 @@
1646	rid, rid
1647	);
1648	manifest_add_checkin_linkages(rid,p,nParent,azParent);
1649	}
1650	manifest_destroy(p);
1651	reparent_abort:
1652	fossil_free(azParent);
1653	fossil_free(zCopy);
1654	}
1655
1656	/*
	@@ -2280,11 +2280,11 @@
2280	int branchMove = 0;
2281	blob_zero(&comment);
2282	if( p->zComment ){
2283	blob_appendf(&comment, " %s.", p->zComment);
2284	}
2285	/* Next loop expects tags to be sorted on hash, so sort it. */
2286	qsort(p->aTag, p->nTag, sizeof(p->aTag[0]), tag_compare);
2287	for(i=0; i<p->nTag; i++){
2288	zTagUuid = p->aTag[i].zUuid;
2289	if( !zTagUuid ) continue;
2290	if( i==0 \|\| fossil_strcmp(zTagUuid, p->aTag[i-1].zUuid)!=0 ){
2291

M src/md5.c

+15 -1

		--- src/md5.c
		+++ src/md5.c
		@@ -22,10 +22,20 @@
22	22	#include <string.h>
23	23	#include <stdio.h>
24	24	#include <sqlite3.h>
25	25	#include "md5.h"
26	26
	27	+#ifdef FOSSIL_ENABLE_SSL
	28	+
	29	+# include <openssl/md5.h>
	30	+# define MD5Context MD5_CTX
	31	+# define MD5Init MD5_Init
	32	+# define MD5Update MD5_Update
	33	+# define MD5Final MD5_Final
	34	+
	35	+#else
	36	+
27	37	/*
28	38	* If compiled on a machine that doesn't have a 32-bit integer,
29	39	* you just set "uint32" to the appropriate datatype for an
30	40	* unsigned 32-bit integer. For example:
31	41	*
		@@ -42,11 +52,14 @@
42	52	uint32 bits[2];
43	53	unsigned char in[64];
44	54	};
45	55	typedef struct Context MD5Context;
46	56
47		-#if defined(__i386__) \|\| defined(__x86_64__) \|\| defined(_WIN32)
	57	+#if defined(i386) \|\| defined(__i386__) \|\| defined(_M_IX86) \|\| \
	58	+ defined(__x86_64) \|\| defined(__x86_64__) \|\| defined(_M_X64) \|\| \
	59	+ defined(_M_AMD64) \|\| defined(_M_ARM) \|\| defined(__x86) \|\| \
	60	+ defined(__arm__) \|\| defined(_WIN32)
48	61	# define byteReverse(A,B)
49	62	#else
50	63	/*
51	64	* Convert an array of integers to little-endian.
52	65	* Note: this code is a no-op on little-endian machines.
		@@ -266,10 +279,11 @@
266	279	MD5Transform(ctx->buf, (uint32 *)ctx->in);
267	280	byteReverse((unsigned char *)ctx->buf, 4);
268	281	memcpy(digest, ctx->buf, 16);
269	282	memset(ctx, 0, sizeof(ctx)); / In case it's sensitive */
270	283	}
	284	+#endif
271	285
272	286	/*
273	287	** Convert a digest into base-16. digest should be declared as
274	288	** "unsigned char digest[16]" in the calling function. The MD5
275	289	** digest is stored in the first 16 bytes. zBuf should
276	290

	--- src/md5.c
	+++ src/md5.c
	@@ -22,10 +22,20 @@
22	#include <string.h>
23	#include <stdio.h>
24	#include <sqlite3.h>
25	#include "md5.h"
26










27	/*
28	* If compiled on a machine that doesn't have a 32-bit integer,
29	* you just set "uint32" to the appropriate datatype for an
30	* unsigned 32-bit integer. For example:
31	*
	@@ -42,11 +52,14 @@
42	uint32 bits[2];
43	unsigned char in[64];
44	};
45	typedef struct Context MD5Context;
46
47	#if defined(__i386__) \|\| defined(__x86_64__) \|\| defined(_WIN32)



48	# define byteReverse(A,B)
49	#else
50	/*
51	* Convert an array of integers to little-endian.
52	* Note: this code is a no-op on little-endian machines.
	@@ -266,10 +279,11 @@
266	MD5Transform(ctx->buf, (uint32 *)ctx->in);
267	byteReverse((unsigned char *)ctx->buf, 4);
268	memcpy(digest, ctx->buf, 16);
269	memset(ctx, 0, sizeof(ctx)); / In case it's sensitive */
270	}

271
272	/*
273	** Convert a digest into base-16. digest should be declared as
274	** "unsigned char digest[16]" in the calling function. The MD5
275	** digest is stored in the first 16 bytes. zBuf should
276

	--- src/md5.c
	+++ src/md5.c
	@@ -22,10 +22,20 @@
22	#include <string.h>
23	#include <stdio.h>
24	#include <sqlite3.h>
25	#include "md5.h"
26
27	#ifdef FOSSIL_ENABLE_SSL
28
29	# include <openssl/md5.h>
30	# define MD5Context MD5_CTX
31	# define MD5Init MD5_Init
32	# define MD5Update MD5_Update
33	# define MD5Final MD5_Final
34
35	#else
36
37	/*
38	* If compiled on a machine that doesn't have a 32-bit integer,
39	* you just set "uint32" to the appropriate datatype for an
40	* unsigned 32-bit integer. For example:
41	*
	@@ -42,11 +52,14 @@
52	uint32 bits[2];
53	unsigned char in[64];
54	};
55	typedef struct Context MD5Context;
56
57	#if defined(i386) \|\| defined(__i386__) \|\| defined(_M_IX86) \|\| \
58	defined(__x86_64) \|\| defined(__x86_64__) \|\| defined(_M_X64) \|\| \
59	defined(_M_AMD64) \|\| defined(_M_ARM) \|\| defined(__x86) \|\| \
60	defined(__arm__) \|\| defined(_WIN32)
61	# define byteReverse(A,B)
62	#else
63	/*
64	* Convert an array of integers to little-endian.
65	* Note: this code is a no-op on little-endian machines.
	@@ -266,10 +279,11 @@
279	MD5Transform(ctx->buf, (uint32 *)ctx->in);
280	byteReverse((unsigned char *)ctx->buf, 4);
281	memcpy(digest, ctx->buf, 16);
282	memset(ctx, 0, sizeof(ctx)); / In case it's sensitive */
283	}
284	#endif
285
286	/*
287	** Convert a digest into base-16. digest should be declared as
288	** "unsigned char digest[16]" in the calling function. The MD5
289	** digest is stored in the first 16 bytes. zBuf should
290

M src/mkversion.c

+6 -5

		--- src/mkversion.c
		+++ src/mkversion.c
		@@ -12,11 +12,12 @@
12	12	#include <stdlib.h>
13	13
14	14	int main(int argc, char *argv[]){
15	15	FILE m,u,*v;
16	16	char *z;
17		- int i, x, d;
	17	+ int i, j, x, d;
	18	+ int vn[3];
18	19	char b[1000];
19	20	char vx[1000];
20	21	memset(b,0,sizeof(b));
21	22	memset(vx,0,sizeof(vx));
22	23	u = fopen(argv[1],"r");
		@@ -45,25 +46,25 @@
45	46	fclose(v);
46	47	for(z=b; z[0] && z[0]!='\r' && z[0]!='\n'; z++){}
47	48	*z = 0;
48	49	printf("#define RELEASE_VERSION \"%s\"\n", b);
49	50	x=0;
50		- i=0;
	51	+ i=j=0;
51	52	z=b;
	53	+ vn[0] = vn[1] = vn[2] = 0;
52	54	while(1){
53	55	if( z[0]>='0' && z[0]<='9' ){
54	56	x = x*10 + z[0] - '0';
55	57	}else{
56		- sprintf(&vx[i],"%02d",x);
57		- i += 2;
	58	+ if( j<3 ) vn[j++] = x;
58	59	x = 0;
59	60	if( z[0]==0 ) break;
60	61	}
61	62	z++;
62	63	}
63	64	for(z=vx; z[0]=='0'; z++){}
64		- printf("#define RELEASE_VERSION_NUMBER %s\n", z);
	65	+ printf("#define RELEASE_VERSION_NUMBER %d%02d%02d\n", vn[0], vn[1], vn[2]);
65	66	memset(vx,0,sizeof(vx));
66	67	strcpy(vx,b);
67	68	d = 0;
68	69	for(z=vx; z[0]; z++){
69	70	if( z[0]=='-' ){
70	71

	--- src/mkversion.c
	+++ src/mkversion.c
	@@ -12,11 +12,12 @@
12	#include <stdlib.h>
13
14	int main(int argc, char *argv[]){
15	FILE m,u,*v;
16	char *z;
17	int i, x, d;

18	char b[1000];
19	char vx[1000];
20	memset(b,0,sizeof(b));
21	memset(vx,0,sizeof(vx));
22	u = fopen(argv[1],"r");
	@@ -45,25 +46,25 @@
45	fclose(v);
46	for(z=b; z[0] && z[0]!='\r' && z[0]!='\n'; z++){}
47	*z = 0;
48	printf("#define RELEASE_VERSION \"%s\"\n", b);
49	x=0;
50	i=0;
51	z=b;

52	while(1){
53	if( z[0]>='0' && z[0]<='9' ){
54	x = x*10 + z[0] - '0';
55	}else{
56	sprintf(&vx[i],"%02d",x);
57	i += 2;
58	x = 0;
59	if( z[0]==0 ) break;
60	}
61	z++;
62	}
63	for(z=vx; z[0]=='0'; z++){}
64	printf("#define RELEASE_VERSION_NUMBER %s\n", z);
65	memset(vx,0,sizeof(vx));
66	strcpy(vx,b);
67	d = 0;
68	for(z=vx; z[0]; z++){
69	if( z[0]=='-' ){
70

	--- src/mkversion.c
	+++ src/mkversion.c
	@@ -12,11 +12,12 @@
12	#include <stdlib.h>
13
14	int main(int argc, char *argv[]){
15	FILE m,u,*v;
16	char *z;
17	int i, j, x, d;
18	int vn[3];
19	char b[1000];
20	char vx[1000];
21	memset(b,0,sizeof(b));
22	memset(vx,0,sizeof(vx));
23	u = fopen(argv[1],"r");
	@@ -45,25 +46,25 @@
46	fclose(v);
47	for(z=b; z[0] && z[0]!='\r' && z[0]!='\n'; z++){}
48	*z = 0;
49	printf("#define RELEASE_VERSION \"%s\"\n", b);
50	x=0;
51	i=j=0;
52	z=b;
53	vn[0] = vn[1] = vn[2] = 0;
54	while(1){
55	if( z[0]>='0' && z[0]<='9' ){
56	x = x*10 + z[0] - '0';
57	}else{
58	if( j<3 ) vn[j++] = x;

59	x = 0;
60	if( z[0]==0 ) break;
61	}
62	z++;
63	}
64	for(z=vx; z[0]=='0'; z++){}
65	printf("#define RELEASE_VERSION_NUMBER %d%02d%02d\n", vn[0], vn[1], vn[2]);
66	memset(vx,0,sizeof(vx));
67	strcpy(vx,b);
68	d = 0;
69	for(z=vx; z[0]; z++){
70	if( z[0]=='-' ){
71

M src/name.c

+24 -28

		--- src/name.c
		+++ src/name.c
		@@ -2,11 +2,11 @@
2	2	** Copyright (c) 2006 D. Richard Hipp
3	3	**
4	4	** This program is free software; you can redistribute it and/or
5	5	** modify it under the terms of the Simplified BSD License (also
6	6	** known as the "2-Clause License" or "FreeBSD License".)
7		-
	7	+**
8	8	** This program is distributed in the hope that it will be useful,
9	9	** but without any warranty; without even the implied warranty of
10	10	** merchantability or fitness for a particular purpose.
11	11	**
12	12	** Author contact information:
		@@ -13,15 +13,11 @@
13	13	** [email protected]
14	14	** http://www.hwaci.com/drh/
15	15	**
16	16	*******************************************************************************
17	17	**
18		-** This file contains code used to convert user-supplied object names into
19		-** canonical UUIDs.
20		-**
21		-** A user-supplied object name is any unique prefix of a valid UUID but
22		-** not necessarily in canonical form.
	18	+** This file contains code used to resolved user-supplied object names.
23	19	*/
24	20	#include "config.h"
25	21	#include "name.h"
26	22	#include <assert.h>
27	23
		@@ -80,12 +76,12 @@
80	76	}
81	77
82	78	/*
83	79	** Convert a symbolic name into a RID. Acceptable forms:
84	80	**
85		-** * SHA1 hash
86		-** * SHA1 hash prefix of at least 4 characters
	81	+** * artifact hash
	82	+** * 4-character or larger prefix of a artifact
87	83	** * Symbolic Name
88	84	** * "tag:" + symbolic name
89	85	** * Date or date-time
90	86	** * "date:" + Date or date-time
91	87	** * symbolic-name ":" date-time
		@@ -105,11 +101,11 @@
105	101	** If zType is "br" then find the first check-in of the named branch
106	102	** rather than the last.
107	103	** zType is "ci" in most use cases since we are usually searching for
108	104	** a check-in.
109	105	**
110		-** Note that the input zTag for types "t" and "e" is the SHA1 hash of
	106	+** Note that the input zTag for types "t" and "e" is the artifact hash of
111	107	** the ticket-change or event-change artifact, not the randomly generated
112	108	** hexadecimal identifier assigned to tickets and events. Those identifiers
113	109	** live in a separate namespace.
114	110	*/
115	111	int symbolic_name_to_rid(const char zTag, const char zType){
		@@ -232,14 +228,14 @@
232	228	zTagBase, zDate, zType
233	229	);
234	230	return rid;
235	231	}
236	232
237		- /* SHA1 hash or prefix */
238		- if( nTag>=4 && nTag<=UUID_SIZE && validate16(zTag, nTag) ){
	233	+ /* artifact hash or prefix */
	234	+ if( nTag>=4 && nTag<=HNAME_MAX && validate16(zTag, nTag) ){
239	235	Stmt q;
240		- char zUuid[UUID_SIZE+1];
	236	+ char zUuid[HNAME_MAX+1];
241	237	memcpy(zUuid, zTag, nTag+1);
242	238	canonical16(zUuid, nTag);
243	239	rid = 0;
244	240	if( zType[0]=='*' ){
245	241	db_prepare(&q, "SELECT rid FROM blob WHERE uuid GLOB '%q*'", zUuid);
		@@ -354,11 +350,11 @@
354	350	*/
355	351	int name_collisions(const char *zName){
356	352	int c = 0; /* count of collisions for zName */
357	353	int nLen; /* length of zName */
358	354	nLen = strlen(zName);
359		- if( nLen>=4 && nLen<=UUID_SIZE && validate16(zName, nLen) ){
	355	+ if( nLen>=4 && nLen<=HNAME_MAX && validate16(zName, nLen) ){
360	356	c = db_int(0,
361	357	"SELECT"
362	358	" (SELECT count(*) FROM ticket"
363	359	" WHERE tkt_uuid GLOB '%q*') +"
364	360	" (SELECT count(*) FROM tag"
		@@ -396,11 +392,11 @@
396	392
397	393	/*
398	394	** Convert a name to a rid. If the name can be any of the various forms
399	395	** accepted:
400	396	**
401		-** * SHA1 hash or prefix thereof
	397	+** * artifact hash or prefix thereof
402	398	** * symbolic name
403	399	** * date
404	400	** * label:date
405	401	** * prev, previous
406	402	** * next
		@@ -425,13 +421,13 @@
425	421	return name_to_typed_rid(zName, "*");
426	422	}
427	423
428	424	/*
429	425	** WEBPAGE: ambiguous
430		-** URL: /ambiguous?name=UUID&src=WEBPAGE
	426	+** URL: /ambiguous?name=NAME&src=WEBPAGE
431	427	**
432		-** The UUID given by the name parameter is ambiguous. Display a page
	428	+** The NAME given by the name parameter is ambiguous. Display a page
433	429	** that shows all possible choices and let the user select between them.
434	430	*/
435	431	void ambiguous_page(void){
436	432	Stmt q;
437	433	const char *zName = P("name");
		@@ -668,11 +664,11 @@
668	664	/*
669	665	** COMMAND: whatis*
670	666	**
671	667	** Usage: %fossil whatis NAME
672	668	**
673		-** Resolve the symbol NAME into its canonical 40-character SHA1-hash
	669	+** Resolve the symbol NAME into its canonical artifact hash
674	670	** artifact name and provide a description of what role that artifact
675	671	** plays.
676	672	**
677	673	** Options:
678	674	**
		@@ -744,11 +740,11 @@
744	740	/*
745	741	** COMMAND: test-ambiguous
746	742	**
747	743	** Usage: %fossil test-ambiguous [--minsize N]
748	744	**
749		-** Show a list of ambiguous SHA1-hash abbreviations of N characters or
	745	+** Show a list of ambiguous artifact hash abbreviations of N characters or
750	746	** more where N defaults to 4. Change N to a different value using
751	747	** the "--minsize N" command-line option.
752	748	*/
753	749	void test_ambiguous_cmd(void){
754	750	Stmt q, ins;
		@@ -794,11 +790,11 @@
794	790	** Schema for the description table
795	791	*/
796	792	static const char zDescTab[] =
797	793	@ CREATE TEMP TABLE IF NOT EXISTS description(
798	794	@ rid INTEGER PRIMARY KEY, -- RID of the object
799		-@ uuid TEXT, -- SHA1 hash of the object
	795	+@ uuid TEXT, -- hash of the object
800	796	@ ctime DATETIME, -- Time of creation
801	797	@ isPrivate BOOLEAN DEFAULT 0, -- True for unpublished artifacts
802	798	@ type TEXT, -- file, checkin, wiki, ticket, etc.
803	799	@ summary TEXT, -- Summary comment for the object
804	800	@ detail TEXT -- File name, check-in comment, etc
		@@ -1084,11 +1080,11 @@
1084	1080	" WHERE description.rid=blob.rid"
1085	1081	" ORDER BY length(content) DESC"
1086	1082	);
1087	1083	@ <table cellpadding="2" cellspacing="0" border="1" id="bigblobtab">
1088	1084	@ <thead><tr><th align="right">Size<th align="right">RID
1089		- @ <th align="right">Delta From<th>SHA1<th>Description<th>Date</tr></thead>
	1085	+ @ <th align="right">Delta From<th>Hash<th>Description<th>Date</tr></thead>
1090	1086	@ <tbody>
1091	1087	while( db_step(&q)==SQLITE_ROW ){
1092	1088	int rid = db_column_int(&q,0);
1093	1089	const char *zUuid = db_column_text(&q, 1);
1094	1090	const char *zDesc = db_column_text(&q, 2);
		@@ -1147,33 +1143,33 @@
1147	1143
1148	1144	/* Maximum number of collision examples to remember */
1149	1145	#define MAX_COLLIDE 25
1150	1146
1151	1147	/*
1152		-** Generate a report on the number of collisions in SHA1 hashes
	1148	+** Generate a report on the number of collisions in artifact hashes
1153	1149	** generated by the SQL given in the argument.
1154	1150	*/
1155	1151	static void collision_report(const char *zSql){
1156	1152	int i, j, kk;
1157	1153	int nHash = 0;
1158	1154	Stmt q;
1159		- char zPrev[UUID_SIZE+1];
	1155	+ char zPrev[HNAME_MAX+1];
1160	1156	struct {
1161	1157	int cnt;
1162	1158	char *azHit[MAX_COLLIDE];
1163		- char z[UUID_SIZE+1];
1164		- } aCollide[UUID_SIZE+1];
	1159	+ char z[HNAME_MAX+1];
	1160	+ } aCollide[HNAME_MAX+1];
1165	1161	memset(aCollide, 0, sizeof(aCollide));
1166	1162	memset(zPrev, 0, sizeof(zPrev));
1167	1163	db_prepare(&q,"%s",zSql/safe-for-%s/);
1168	1164	while( db_step(&q)==SQLITE_ROW ){
1169	1165	const char *zUuid = db_column_text(&q,0);
1170	1166	int n = db_column_bytes(&q,0);
1171	1167	int i;
1172	1168	nHash++;
1173	1169	for(i=0; zPrev[i] && zPrev[i]==zUuid[i]; i++){}
1174		- if( i>0 && i<=UUID_SIZE ){
	1170	+ if( i>0 && i<=HNAME_MAX ){
1175	1171	if( i>=4 && aCollide[i].cnt<MAX_COLLIDE ){
1176	1172	aCollide[i].azHit[aCollide[i].cnt] = mprintf("%.*s", i, zPrev);
1177	1173	}
1178	1174	aCollide[i].cnt++;
1179	1175	if( aCollide[i].z[0]==0 ) memcpy(aCollide[i].z, zPrev, n+1);
		@@ -1182,18 +1178,18 @@
1182	1178	}
1183	1179	db_finalize(&q);
1184	1180	@ <table border=1><thead>
1185	1181	@ <tr><th>Length<th>Instances<th>First Instance</tr>
1186	1182	@ </thead><tbody>
1187		- for(i=1; i<=UUID_SIZE; i++){
	1183	+ for(i=1; i<=HNAME_MAX; i++){
1188	1184	if( aCollide[i].cnt==0 ) continue;
1189	1185	@ <tr><td>%d(i)<td>%d(aCollide[i].cnt)<td>%h(aCollide[i].z)</tr>
1190	1186	}
1191	1187	@ </tbody></table>
1192	1188	@ <p>Total number of hashes: %d(nHash)</p>
1193	1189	kk = 0;
1194		- for(i=UUID_SIZE; i>=4; i--){
	1190	+ for(i=HNAME_MAX; i>=4; i--){
1195	1191	if( aCollide[i].cnt==0 ) continue;
1196	1192	if( aCollide[i].cnt>200 ) break;
1197	1193	kk += aCollide[i].cnt;
1198	1194	if( aCollide[i].cnt<25 ){
1199	1195	@ <p>Collisions of length %d(i):
		@@ -1219,11 +1215,11 @@
1219	1215	** Show the number of hash collisions for hash prefixes of various lengths.
1220	1216	*/
1221	1217	void hash_collisions_webpage(void){
1222	1218	login_check_credentials();
1223	1219	if( !g.perm.Read ){ login_needed(g.anon.Read); return; }
1224		- style_header("SHA1 Prefix Collisions");
	1220	+ style_header("Hash Prefix Collisions");
1225	1221	style_submenu_element("Activity Reports", "reports");
1226	1222	style_submenu_element("Stats", "stat");
1227	1223	@ <h1>Hash Prefix Collisions on Check-ins</h1>
1228	1224	collision_report("SELECT (SELECT uuid FROM blob WHERE rid=objid)"
1229	1225	" FROM event WHERE event.type='ci'"
1230	1226

	--- src/name.c
	+++ src/name.c
	@@ -2,11 +2,11 @@
2	** Copyright (c) 2006 D. Richard Hipp
3	**
4	** This program is free software; you can redistribute it and/or
5	** modify it under the terms of the Simplified BSD License (also
6	** known as the "2-Clause License" or "FreeBSD License".)
7
8	** This program is distributed in the hope that it will be useful,
9	** but without any warranty; without even the implied warranty of
10	** merchantability or fitness for a particular purpose.
11	**
12	** Author contact information:
	@@ -13,15 +13,11 @@
13	** [email protected]
14	** http://www.hwaci.com/drh/
15	**
16	*******************************************************************************
17	**
18	** This file contains code used to convert user-supplied object names into
19	** canonical UUIDs.
20	**
21	** A user-supplied object name is any unique prefix of a valid UUID but
22	** not necessarily in canonical form.
23	*/
24	#include "config.h"
25	#include "name.h"
26	#include <assert.h>
27
	@@ -80,12 +76,12 @@
80	}
81
82	/*
83	** Convert a symbolic name into a RID. Acceptable forms:
84	**
85	** * SHA1 hash
86	** * SHA1 hash prefix of at least 4 characters
87	** * Symbolic Name
88	** * "tag:" + symbolic name
89	** * Date or date-time
90	** * "date:" + Date or date-time
91	** * symbolic-name ":" date-time
	@@ -105,11 +101,11 @@
105	** If zType is "br" then find the first check-in of the named branch
106	** rather than the last.
107	** zType is "ci" in most use cases since we are usually searching for
108	** a check-in.
109	**
110	** Note that the input zTag for types "t" and "e" is the SHA1 hash of
111	** the ticket-change or event-change artifact, not the randomly generated
112	** hexadecimal identifier assigned to tickets and events. Those identifiers
113	** live in a separate namespace.
114	*/
115	int symbolic_name_to_rid(const char zTag, const char zType){
	@@ -232,14 +228,14 @@
232	zTagBase, zDate, zType
233	);
234	return rid;
235	}
236
237	/* SHA1 hash or prefix */
238	if( nTag>=4 && nTag<=UUID_SIZE && validate16(zTag, nTag) ){
239	Stmt q;
240	char zUuid[UUID_SIZE+1];
241	memcpy(zUuid, zTag, nTag+1);
242	canonical16(zUuid, nTag);
243	rid = 0;
244	if( zType[0]=='*' ){
245	db_prepare(&q, "SELECT rid FROM blob WHERE uuid GLOB '%q*'", zUuid);
	@@ -354,11 +350,11 @@
354	*/
355	int name_collisions(const char *zName){
356	int c = 0; /* count of collisions for zName */
357	int nLen; /* length of zName */
358	nLen = strlen(zName);
359	if( nLen>=4 && nLen<=UUID_SIZE && validate16(zName, nLen) ){
360	c = db_int(0,
361	"SELECT"
362	" (SELECT count(*) FROM ticket"
363	" WHERE tkt_uuid GLOB '%q*') +"
364	" (SELECT count(*) FROM tag"
	@@ -396,11 +392,11 @@
396
397	/*
398	** Convert a name to a rid. If the name can be any of the various forms
399	** accepted:
400	**
401	** * SHA1 hash or prefix thereof
402	** * symbolic name
403	** * date
404	** * label:date
405	** * prev, previous
406	** * next
	@@ -425,13 +421,13 @@
425	return name_to_typed_rid(zName, "*");
426	}
427
428	/*
429	** WEBPAGE: ambiguous
430	** URL: /ambiguous?name=UUID&src=WEBPAGE
431	**
432	** The UUID given by the name parameter is ambiguous. Display a page
433	** that shows all possible choices and let the user select between them.
434	*/
435	void ambiguous_page(void){
436	Stmt q;
437	const char *zName = P("name");
	@@ -668,11 +664,11 @@
668	/*
669	** COMMAND: whatis*
670	**
671	** Usage: %fossil whatis NAME
672	**
673	** Resolve the symbol NAME into its canonical 40-character SHA1-hash
674	** artifact name and provide a description of what role that artifact
675	** plays.
676	**
677	** Options:
678	**
	@@ -744,11 +740,11 @@
744	/*
745	** COMMAND: test-ambiguous
746	**
747	** Usage: %fossil test-ambiguous [--minsize N]
748	**
749	** Show a list of ambiguous SHA1-hash abbreviations of N characters or
750	** more where N defaults to 4. Change N to a different value using
751	** the "--minsize N" command-line option.
752	*/
753	void test_ambiguous_cmd(void){
754	Stmt q, ins;
	@@ -794,11 +790,11 @@
794	** Schema for the description table
795	*/
796	static const char zDescTab[] =
797	@ CREATE TEMP TABLE IF NOT EXISTS description(
798	@ rid INTEGER PRIMARY KEY, -- RID of the object
799	@ uuid TEXT, -- SHA1 hash of the object
800	@ ctime DATETIME, -- Time of creation
801	@ isPrivate BOOLEAN DEFAULT 0, -- True for unpublished artifacts
802	@ type TEXT, -- file, checkin, wiki, ticket, etc.
803	@ summary TEXT, -- Summary comment for the object
804	@ detail TEXT -- File name, check-in comment, etc
	@@ -1084,11 +1080,11 @@
1084	" WHERE description.rid=blob.rid"
1085	" ORDER BY length(content) DESC"
1086	);
1087	@ <table cellpadding="2" cellspacing="0" border="1" id="bigblobtab">
1088	@ <thead><tr><th align="right">Size<th align="right">RID
1089	@ <th align="right">Delta From<th>SHA1<th>Description<th>Date</tr></thead>
1090	@ <tbody>
1091	while( db_step(&q)==SQLITE_ROW ){
1092	int rid = db_column_int(&q,0);
1093	const char *zUuid = db_column_text(&q, 1);
1094	const char *zDesc = db_column_text(&q, 2);
	@@ -1147,33 +1143,33 @@
1147
1148	/* Maximum number of collision examples to remember */
1149	#define MAX_COLLIDE 25
1150
1151	/*
1152	** Generate a report on the number of collisions in SHA1 hashes
1153	** generated by the SQL given in the argument.
1154	*/
1155	static void collision_report(const char *zSql){
1156	int i, j, kk;
1157	int nHash = 0;
1158	Stmt q;
1159	char zPrev[UUID_SIZE+1];
1160	struct {
1161	int cnt;
1162	char *azHit[MAX_COLLIDE];
1163	char z[UUID_SIZE+1];
1164	} aCollide[UUID_SIZE+1];
1165	memset(aCollide, 0, sizeof(aCollide));
1166	memset(zPrev, 0, sizeof(zPrev));
1167	db_prepare(&q,"%s",zSql/safe-for-%s/);
1168	while( db_step(&q)==SQLITE_ROW ){
1169	const char *zUuid = db_column_text(&q,0);
1170	int n = db_column_bytes(&q,0);
1171	int i;
1172	nHash++;
1173	for(i=0; zPrev[i] && zPrev[i]==zUuid[i]; i++){}
1174	if( i>0 && i<=UUID_SIZE ){
1175	if( i>=4 && aCollide[i].cnt<MAX_COLLIDE ){
1176	aCollide[i].azHit[aCollide[i].cnt] = mprintf("%.*s", i, zPrev);
1177	}
1178	aCollide[i].cnt++;
1179	if( aCollide[i].z[0]==0 ) memcpy(aCollide[i].z, zPrev, n+1);
	@@ -1182,18 +1178,18 @@
1182	}
1183	db_finalize(&q);
1184	@ <table border=1><thead>
1185	@ <tr><th>Length<th>Instances<th>First Instance</tr>
1186	@ </thead><tbody>
1187	for(i=1; i<=UUID_SIZE; i++){
1188	if( aCollide[i].cnt==0 ) continue;
1189	@ <tr><td>%d(i)<td>%d(aCollide[i].cnt)<td>%h(aCollide[i].z)</tr>
1190	}
1191	@ </tbody></table>
1192	@ <p>Total number of hashes: %d(nHash)</p>
1193	kk = 0;
1194	for(i=UUID_SIZE; i>=4; i--){
1195	if( aCollide[i].cnt==0 ) continue;
1196	if( aCollide[i].cnt>200 ) break;
1197	kk += aCollide[i].cnt;
1198	if( aCollide[i].cnt<25 ){
1199	@ <p>Collisions of length %d(i):
	@@ -1219,11 +1215,11 @@
1219	** Show the number of hash collisions for hash prefixes of various lengths.
1220	*/
1221	void hash_collisions_webpage(void){
1222	login_check_credentials();
1223	if( !g.perm.Read ){ login_needed(g.anon.Read); return; }
1224	style_header("SHA1 Prefix Collisions");
1225	style_submenu_element("Activity Reports", "reports");
1226	style_submenu_element("Stats", "stat");
1227	@ <h1>Hash Prefix Collisions on Check-ins</h1>
1228	collision_report("SELECT (SELECT uuid FROM blob WHERE rid=objid)"
1229	" FROM event WHERE event.type='ci'"
1230

	--- src/name.c
	+++ src/name.c
	@@ -2,11 +2,11 @@
2	** Copyright (c) 2006 D. Richard Hipp
3	**
4	** This program is free software; you can redistribute it and/or
5	** modify it under the terms of the Simplified BSD License (also
6	** known as the "2-Clause License" or "FreeBSD License".)
7	**
8	** This program is distributed in the hope that it will be useful,
9	** but without any warranty; without even the implied warranty of
10	** merchantability or fitness for a particular purpose.
11	**
12	** Author contact information:
	@@ -13,15 +13,11 @@
13	** [email protected]
14	** http://www.hwaci.com/drh/
15	**
16	*******************************************************************************
17	**
18	** This file contains code used to resolved user-supplied object names.




19	*/
20	#include "config.h"
21	#include "name.h"
22	#include <assert.h>
23
	@@ -80,12 +76,12 @@
76	}
77
78	/*
79	** Convert a symbolic name into a RID. Acceptable forms:
80	**
81	** * artifact hash
82	** * 4-character or larger prefix of a artifact
83	** * Symbolic Name
84	** * "tag:" + symbolic name
85	** * Date or date-time
86	** * "date:" + Date or date-time
87	** * symbolic-name ":" date-time
	@@ -105,11 +101,11 @@
101	** If zType is "br" then find the first check-in of the named branch
102	** rather than the last.
103	** zType is "ci" in most use cases since we are usually searching for
104	** a check-in.
105	**
106	** Note that the input zTag for types "t" and "e" is the artifact hash of
107	** the ticket-change or event-change artifact, not the randomly generated
108	** hexadecimal identifier assigned to tickets and events. Those identifiers
109	** live in a separate namespace.
110	*/
111	int symbolic_name_to_rid(const char zTag, const char zType){
	@@ -232,14 +228,14 @@
228	zTagBase, zDate, zType
229	);
230	return rid;
231	}
232
233	/* artifact hash or prefix */
234	if( nTag>=4 && nTag<=HNAME_MAX && validate16(zTag, nTag) ){
235	Stmt q;
236	char zUuid[HNAME_MAX+1];
237	memcpy(zUuid, zTag, nTag+1);
238	canonical16(zUuid, nTag);
239	rid = 0;
240	if( zType[0]=='*' ){
241	db_prepare(&q, "SELECT rid FROM blob WHERE uuid GLOB '%q*'", zUuid);
	@@ -354,11 +350,11 @@
350	*/
351	int name_collisions(const char *zName){
352	int c = 0; /* count of collisions for zName */
353	int nLen; /* length of zName */
354	nLen = strlen(zName);
355	if( nLen>=4 && nLen<=HNAME_MAX && validate16(zName, nLen) ){
356	c = db_int(0,
357	"SELECT"
358	" (SELECT count(*) FROM ticket"
359	" WHERE tkt_uuid GLOB '%q*') +"
360	" (SELECT count(*) FROM tag"
	@@ -396,11 +392,11 @@
392
393	/*
394	** Convert a name to a rid. If the name can be any of the various forms
395	** accepted:
396	**
397	** * artifact hash or prefix thereof
398	** * symbolic name
399	** * date
400	** * label:date
401	** * prev, previous
402	** * next
	@@ -425,13 +421,13 @@
421	return name_to_typed_rid(zName, "*");
422	}
423
424	/*
425	** WEBPAGE: ambiguous
426	** URL: /ambiguous?name=NAME&src=WEBPAGE
427	**
428	** The NAME given by the name parameter is ambiguous. Display a page
429	** that shows all possible choices and let the user select between them.
430	*/
431	void ambiguous_page(void){
432	Stmt q;
433	const char *zName = P("name");
	@@ -668,11 +664,11 @@
664	/*
665	** COMMAND: whatis*
666	**
667	** Usage: %fossil whatis NAME
668	**
669	** Resolve the symbol NAME into its canonical artifact hash
670	** artifact name and provide a description of what role that artifact
671	** plays.
672	**
673	** Options:
674	**
	@@ -744,11 +740,11 @@
740	/*
741	** COMMAND: test-ambiguous
742	**
743	** Usage: %fossil test-ambiguous [--minsize N]
744	**
745	** Show a list of ambiguous artifact hash abbreviations of N characters or
746	** more where N defaults to 4. Change N to a different value using
747	** the "--minsize N" command-line option.
748	*/
749	void test_ambiguous_cmd(void){
750	Stmt q, ins;
	@@ -794,11 +790,11 @@
790	** Schema for the description table
791	*/
792	static const char zDescTab[] =
793	@ CREATE TEMP TABLE IF NOT EXISTS description(
794	@ rid INTEGER PRIMARY KEY, -- RID of the object
795	@ uuid TEXT, -- hash of the object
796	@ ctime DATETIME, -- Time of creation
797	@ isPrivate BOOLEAN DEFAULT 0, -- True for unpublished artifacts
798	@ type TEXT, -- file, checkin, wiki, ticket, etc.
799	@ summary TEXT, -- Summary comment for the object
800	@ detail TEXT -- File name, check-in comment, etc
	@@ -1084,11 +1080,11 @@
1080	" WHERE description.rid=blob.rid"
1081	" ORDER BY length(content) DESC"
1082	);
1083	@ <table cellpadding="2" cellspacing="0" border="1" id="bigblobtab">
1084	@ <thead><tr><th align="right">Size<th align="right">RID
1085	@ <th align="right">Delta From<th>Hash<th>Description<th>Date</tr></thead>
1086	@ <tbody>
1087	while( db_step(&q)==SQLITE_ROW ){
1088	int rid = db_column_int(&q,0);
1089	const char *zUuid = db_column_text(&q, 1);
1090	const char *zDesc = db_column_text(&q, 2);
	@@ -1147,33 +1143,33 @@
1143
1144	/* Maximum number of collision examples to remember */
1145	#define MAX_COLLIDE 25
1146
1147	/*
1148	** Generate a report on the number of collisions in artifact hashes
1149	** generated by the SQL given in the argument.
1150	*/
1151	static void collision_report(const char *zSql){
1152	int i, j, kk;
1153	int nHash = 0;
1154	Stmt q;
1155	char zPrev[HNAME_MAX+1];
1156	struct {
1157	int cnt;
1158	char *azHit[MAX_COLLIDE];
1159	char z[HNAME_MAX+1];
1160	} aCollide[HNAME_MAX+1];
1161	memset(aCollide, 0, sizeof(aCollide));
1162	memset(zPrev, 0, sizeof(zPrev));
1163	db_prepare(&q,"%s",zSql/safe-for-%s/);
1164	while( db_step(&q)==SQLITE_ROW ){
1165	const char *zUuid = db_column_text(&q,0);
1166	int n = db_column_bytes(&q,0);
1167	int i;
1168	nHash++;
1169	for(i=0; zPrev[i] && zPrev[i]==zUuid[i]; i++){}
1170	if( i>0 && i<=HNAME_MAX ){
1171	if( i>=4 && aCollide[i].cnt<MAX_COLLIDE ){
1172	aCollide[i].azHit[aCollide[i].cnt] = mprintf("%.*s", i, zPrev);
1173	}
1174	aCollide[i].cnt++;
1175	if( aCollide[i].z[0]==0 ) memcpy(aCollide[i].z, zPrev, n+1);
	@@ -1182,18 +1178,18 @@
1178	}
1179	db_finalize(&q);
1180	@ <table border=1><thead>
1181	@ <tr><th>Length<th>Instances<th>First Instance</tr>
1182	@ </thead><tbody>
1183	for(i=1; i<=HNAME_MAX; i++){
1184	if( aCollide[i].cnt==0 ) continue;
1185	@ <tr><td>%d(i)<td>%d(aCollide[i].cnt)<td>%h(aCollide[i].z)</tr>
1186	}
1187	@ </tbody></table>
1188	@ <p>Total number of hashes: %d(nHash)</p>
1189	kk = 0;
1190	for(i=HNAME_MAX; i>=4; i--){
1191	if( aCollide[i].cnt==0 ) continue;
1192	if( aCollide[i].cnt>200 ) break;
1193	kk += aCollide[i].cnt;
1194	if( aCollide[i].cnt<25 ){
1195	@ <p>Collisions of length %d(i):
	@@ -1219,11 +1215,11 @@
1215	** Show the number of hash collisions for hash prefixes of various lengths.
1216	*/
1217	void hash_collisions_webpage(void){
1218	login_check_credentials();
1219	if( !g.perm.Read ){ login_needed(g.anon.Read); return; }
1220	style_header("Hash Prefix Collisions");
1221	style_submenu_element("Activity Reports", "reports");
1222	style_submenu_element("Stats", "stat");
1223	@ <h1>Hash Prefix Collisions on Check-ins</h1>
1224	collision_report("SELECT (SELECT uuid FROM blob WHERE rid=objid)"
1225	" FROM event WHERE event.type='ci'"
1226

M src/printf.c

+2 -2

		--- src/printf.c
		+++ src/printf.c
		@@ -24,11 +24,11 @@
24	24	# include <io.h>
25	25	# include <fcntl.h>
26	26	#endif
27	27	#include <time.h>
28	28
29		-/* Two custom conversions are used to show a prefix of SHA1 hashes:
	29	+/* Two custom conversions are used to show a prefix of artifact hashes:
30	30	**
31	31	** %!S Prefix of a length appropriate for URLs
32	32	** %S Prefix of a length appropriate for human display
33	33	**
34	34	** The following macros help determine those lengths. FOSSIL_HASH_DIGITS
		@@ -44,11 +44,11 @@
44	44	#ifndef FOSSIL_HASH_DIGITS_URL
45	45	# define FOSSIL_HASH_DIGITS_URL 16 /* For %!S (embedded in URLs) */
46	46	#endif
47	47
48	48	/*
49		-** Return the number of SHA1 hash digits to display. The number is for
	49	+** Return the number of artifact hash digits to display. The number is for
50	50	** human output if the bForUrl is false and is destined for a URL if
51	51	** bForUrl is false.
52	52	*/
53	53	static int hashDigits(int bForUrl){
54	54	static int nDigitHuman = 0;
55	55

	--- src/printf.c
	+++ src/printf.c
	@@ -24,11 +24,11 @@
24	# include <io.h>
25	# include <fcntl.h>
26	#endif
27	#include <time.h>
28
29	/* Two custom conversions are used to show a prefix of SHA1 hashes:
30	**
31	** %!S Prefix of a length appropriate for URLs
32	** %S Prefix of a length appropriate for human display
33	**
34	** The following macros help determine those lengths. FOSSIL_HASH_DIGITS
	@@ -44,11 +44,11 @@
44	#ifndef FOSSIL_HASH_DIGITS_URL
45	# define FOSSIL_HASH_DIGITS_URL 16 /* For %!S (embedded in URLs) */
46	#endif
47
48	/*
49	** Return the number of SHA1 hash digits to display. The number is for
50	** human output if the bForUrl is false and is destined for a URL if
51	** bForUrl is false.
52	*/
53	static int hashDigits(int bForUrl){
54	static int nDigitHuman = 0;
55

	--- src/printf.c
	+++ src/printf.c
	@@ -24,11 +24,11 @@
24	# include <io.h>
25	# include <fcntl.h>
26	#endif
27	#include <time.h>
28
29	/* Two custom conversions are used to show a prefix of artifact hashes:
30	**
31	** %!S Prefix of a length appropriate for URLs
32	** %S Prefix of a length appropriate for human display
33	**
34	** The following macros help determine those lengths. FOSSIL_HASH_DIGITS
	@@ -44,11 +44,11 @@
44	#ifndef FOSSIL_HASH_DIGITS_URL
45	# define FOSSIL_HASH_DIGITS_URL 16 /* For %!S (embedded in URLs) */
46	#endif
47
48	/*
49	** Return the number of artifact hash digits to display. The number is for
50	** human output if the bForUrl is false and is destined for a URL if
51	** bForUrl is false.
52	*/
53	static int hashDigits(int bForUrl){
54	static int nDigitHuman = 0;
55

M src/purge.c

+7 -13

		--- src/purge.c
		+++ src/purge.c
		@@ -43,11 +43,11 @@
43	43	@ );
44	44	@ CREATE TABLE IF NOT EXISTS "%w".purgeitem(
45	45	@ piid INTEGER PRIMARY KEY, -- ID for the purge item
46	46	@ peid INTEGER REFERENCES purgeevent ON DELETE CASCADE, -- Purge event
47	47	@ orid INTEGER, -- Original RID before purged
48		-@ uuid TEXT NOT NULL, -- SHA1 hash of the purged artifact
	48	+@ uuid TEXT NOT NULL, -- hash of the purged artifact
49	49	@ srcid INTEGER, -- Basis purgeitem for delta compression
50	50	@ isPrivate BOOLEAN, -- True if artifact was originally private
51	51	@ sz INT NOT NULL, -- Uncompressed size of the purged artifact
52	52	@ desc TEXT, -- Brief description of this artifact
53	53	@ data BLOB -- Compressed artifact content
		@@ -347,11 +347,11 @@
347	347	int piid, /* ID of the item to extract */
348	348	Blob pOut / Write the content into this blob */
349	349	){
350	350	Stmt q;
351	351	int srcid;
352		- Blob h1, h2, x;
	352	+ Blob h1, x;
353	353	static Bag busy;
354	354
355	355	db_prepare(&q, "SELECT uuid, srcid, data FROM purgeitem"
356	356	" WHERE piid=%d", piid);
357	357	if( db_step(&q)!=SQLITE_ROW ){
		@@ -376,17 +376,14 @@
376	376	blob_reset(&baseline);
377	377	}
378	378	bag_remove(&busy, piid);
379	379	blob_zero(&h1);
380	380	db_column_blob(&q, 0, &h1);
381		- sha1sum_blob(pOut, &h2);
382		- if( blob_compare(&h1, &h2)!=0 ){
383		- fossil_fatal("SHA1 hash mismatch - wanted %s, got %s",
384		- blob_str(&h1), blob_str(&h2));
	381	+ if( hname_verify_hash(pOut, blob_buffer(&h1), blob_size(&h1))==0 ){
	382	+ fossil_fatal("incorrect artifact hash on %b", &h1);
385	383	}
386	384	blob_reset(&h1);
387		- blob_reset(&h2);
388	385	db_finalize(&q);
389	386	return 0;
390	387	}
391	388
392	389	/*
		@@ -411,11 +408,11 @@
411	408	" WHERE ix.srcid=%d"
412	409	" AND ix.piid=purgeitem.piid;",
413	410	iSrc
414	411	);
415	412	while( db_step(&q)==SQLITE_ROW ){
416		- Blob h1, h2, c1, c2;
	413	+ Blob h1, c1, c2;
417	414	int isPriv, rid;
418	415	blob_zero(&h1);
419	416	db_column_blob(&q, 0, &h1);
420	417	blob_zero(&c1);
421	418	db_column_blob(&q, 1, &c1);
		@@ -425,16 +422,13 @@
425	422	blob_delta_apply(pBasis, &c1, &c2);
426	423	blob_reset(&c1);
427	424	}else{
428	425	c2 = c1;
429	426	}
430		- sha1sum_blob(&c2, &h2);
431		- if( blob_compare(&h1, &h2)!=0 ){
432		- fossil_fatal("SHA1 hash mismatch - wanted %s, got %s",
433		- blob_str(&h1), blob_str(&h2));
	427	+ if( hname_verify_hash(&c2, blob_buffer(&h1), blob_size(&h1))==0 ){
	428	+ fossil_fatal("incorrect hash on %b", &h1);
434	429	}
435		- blob_reset(&h2);
436	430	isPriv = db_column_int(&q, 2);
437	431	rid = content_put_ex(&c2, blob_str(&h1), 0, 0, isPriv);
438	432	if( rid==0 ){
439	433	fossil_fatal("%s", g.zErrMsg);
440	434	}else{
441	435

	--- src/purge.c
	+++ src/purge.c
	@@ -43,11 +43,11 @@
43	@ );
44	@ CREATE TABLE IF NOT EXISTS "%w".purgeitem(
45	@ piid INTEGER PRIMARY KEY, -- ID for the purge item
46	@ peid INTEGER REFERENCES purgeevent ON DELETE CASCADE, -- Purge event
47	@ orid INTEGER, -- Original RID before purged
48	@ uuid TEXT NOT NULL, -- SHA1 hash of the purged artifact
49	@ srcid INTEGER, -- Basis purgeitem for delta compression
50	@ isPrivate BOOLEAN, -- True if artifact was originally private
51	@ sz INT NOT NULL, -- Uncompressed size of the purged artifact
52	@ desc TEXT, -- Brief description of this artifact
53	@ data BLOB -- Compressed artifact content
	@@ -347,11 +347,11 @@
347	int piid, /* ID of the item to extract */
348	Blob pOut / Write the content into this blob */
349	){
350	Stmt q;
351	int srcid;
352	Blob h1, h2, x;
353	static Bag busy;
354
355	db_prepare(&q, "SELECT uuid, srcid, data FROM purgeitem"
356	" WHERE piid=%d", piid);
357	if( db_step(&q)!=SQLITE_ROW ){
	@@ -376,17 +376,14 @@
376	blob_reset(&baseline);
377	}
378	bag_remove(&busy, piid);
379	blob_zero(&h1);
380	db_column_blob(&q, 0, &h1);
381	sha1sum_blob(pOut, &h2);
382	if( blob_compare(&h1, &h2)!=0 ){
383	fossil_fatal("SHA1 hash mismatch - wanted %s, got %s",
384	blob_str(&h1), blob_str(&h2));
385	}
386	blob_reset(&h1);
387	blob_reset(&h2);
388	db_finalize(&q);
389	return 0;
390	}
391
392	/*
	@@ -411,11 +408,11 @@
411	" WHERE ix.srcid=%d"
412	" AND ix.piid=purgeitem.piid;",
413	iSrc
414	);
415	while( db_step(&q)==SQLITE_ROW ){
416	Blob h1, h2, c1, c2;
417	int isPriv, rid;
418	blob_zero(&h1);
419	db_column_blob(&q, 0, &h1);
420	blob_zero(&c1);
421	db_column_blob(&q, 1, &c1);
	@@ -425,16 +422,13 @@
425	blob_delta_apply(pBasis, &c1, &c2);
426	blob_reset(&c1);
427	}else{
428	c2 = c1;
429	}
430	sha1sum_blob(&c2, &h2);
431	if( blob_compare(&h1, &h2)!=0 ){
432	fossil_fatal("SHA1 hash mismatch - wanted %s, got %s",
433	blob_str(&h1), blob_str(&h2));
434	}
435	blob_reset(&h2);
436	isPriv = db_column_int(&q, 2);
437	rid = content_put_ex(&c2, blob_str(&h1), 0, 0, isPriv);
438	if( rid==0 ){
439	fossil_fatal("%s", g.zErrMsg);
440	}else{
441

	--- src/purge.c
	+++ src/purge.c
	@@ -43,11 +43,11 @@
43	@ );
44	@ CREATE TABLE IF NOT EXISTS "%w".purgeitem(
45	@ piid INTEGER PRIMARY KEY, -- ID for the purge item
46	@ peid INTEGER REFERENCES purgeevent ON DELETE CASCADE, -- Purge event
47	@ orid INTEGER, -- Original RID before purged
48	@ uuid TEXT NOT NULL, -- hash of the purged artifact
49	@ srcid INTEGER, -- Basis purgeitem for delta compression
50	@ isPrivate BOOLEAN, -- True if artifact was originally private
51	@ sz INT NOT NULL, -- Uncompressed size of the purged artifact
52	@ desc TEXT, -- Brief description of this artifact
53	@ data BLOB -- Compressed artifact content
	@@ -347,11 +347,11 @@
347	int piid, /* ID of the item to extract */
348	Blob pOut / Write the content into this blob */
349	){
350	Stmt q;
351	int srcid;
352	Blob h1, x;
353	static Bag busy;
354
355	db_prepare(&q, "SELECT uuid, srcid, data FROM purgeitem"
356	" WHERE piid=%d", piid);
357	if( db_step(&q)!=SQLITE_ROW ){
	@@ -376,17 +376,14 @@
376	blob_reset(&baseline);
377	}
378	bag_remove(&busy, piid);
379	blob_zero(&h1);
380	db_column_blob(&q, 0, &h1);
381	if( hname_verify_hash(pOut, blob_buffer(&h1), blob_size(&h1))==0 ){
382	fossil_fatal("incorrect artifact hash on %b", &h1);


383	}
384	blob_reset(&h1);

385	db_finalize(&q);
386	return 0;
387	}
388
389	/*
	@@ -411,11 +408,11 @@
408	" WHERE ix.srcid=%d"
409	" AND ix.piid=purgeitem.piid;",
410	iSrc
411	);
412	while( db_step(&q)==SQLITE_ROW ){
413	Blob h1, c1, c2;
414	int isPriv, rid;
415	blob_zero(&h1);
416	db_column_blob(&q, 0, &h1);
417	blob_zero(&c1);
418	db_column_blob(&q, 1, &c1);
	@@ -425,16 +422,13 @@
422	blob_delta_apply(pBasis, &c1, &c2);
423	blob_reset(&c1);
424	}else{
425	c2 = c1;
426	}
427	if( hname_verify_hash(&c2, blob_buffer(&h1), blob_size(&h1))==0 ){
428	fossil_fatal("incorrect hash on %b", &h1);


429	}

430	isPriv = db_column_int(&q, 2);
431	rid = content_put_ex(&c2, blob_str(&h1), 0, 0, isPriv);
432	if( rid==0 ){
433	fossil_fatal("%s", g.zErrMsg);
434	}else{
435

M src/rebuild.c

+106 -94

		--- src/rebuild.c
		+++ src/rebuild.c
		@@ -21,76 +21,41 @@
21	21	#include "rebuild.h"
22	22	#include <assert.h>
23	23	#include <errno.h>
24	24
25	25	/*
26		-** Make changes to the stable part of the schema (the part that is not
27		-** simply deleted and reconstructed on a rebuild) to bring the schema
28		-** up to the latest.
	26	+** Update the schema as necessary
29	27	*/
30		-static const char zSchemaUpdates1[] =
31		-@ -- Index on the delta table
32		-@ --
33		-@ CREATE INDEX IF NOT EXISTS delta_i1 ON delta(srcid);
34		-@
35		-@ -- Artifacts that should not be processed are identified in the
36		-@ -- "shun" table. Artifacts that are control-file forgeries or
37		-@ -- spam or artifacts whose contents violate administrative policy
38		-@ -- can be shunned in order to prevent them from contaminating
39		-@ -- the repository.
40		-@ --
41		-@ -- Shunned artifacts do not exist in the blob table. Hence they
42		-@ -- have not artifact ID (rid) and we thus must store their full
43		-@ -- UUID.
44		-@ --
45		-@ CREATE TABLE IF NOT EXISTS shun(
46		-@ uuid UNIQUE, -- UUID of artifact to be shunned. Canonical form
47		-@ mtime INTEGER, -- When added. Seconds since 1970
48		-@ scom TEXT -- Optional text explaining why the shun occurred
49		-@ );
50		-@
51		-@ -- Artifacts that should not be pushed are stored in the "private"
52		-@ -- table.
53		-@ --
54		-@ CREATE TABLE IF NOT EXISTS private(rid INTEGER PRIMARY KEY);
55		-@
56		-@ -- Some ticket content (such as the originators email address or contact
57		-@ -- information) needs to be obscured to protect privacy. This is achieved
58		-@ -- by storing an SHA1 hash of the content. For display, the hash is
59		-@ -- mapped back into the original text using this table.
60		-@ --
61		-@ -- This table contains sensitive information and should not be shared
62		-@ -- with unauthorized users.
63		-@ --
64		-@ CREATE TABLE IF NOT EXISTS concealed(
65		-@ hash TEXT PRIMARY KEY, -- The SHA1 hash of content
66		-@ mtime INTEGER, -- Time created. Seconds since 1970
67		-@ content TEXT -- Content intended to be concealed
68		-@ );
69		-;
70		-static const char zSchemaUpdates2[] =
71		-@ -- An entry in this table describes a database query that generates a
72		-@ -- table of tickets.
73		-@ --
74		-@ CREATE TABLE IF NOT EXISTS reportfmt(
75		-@ rn INTEGER PRIMARY KEY, -- Report number
76		-@ owner TEXT, -- Owner of this report format (not used)
77		-@ title TEXT UNIQUE, -- Title of this report
78		-@ mtime INTEGER, -- Time last modified. Seconds since 1970
79		-@ cols TEXT, -- A color-key specification
80		-@ sqlcode TEXT -- An SQL SELECT statement for this report
81		-@ );
82		-;
83		-
84	28	static void rebuild_update_schema(void){
85		- int rc;
86		- db_multi_exec("%s", zSchemaUpdates1 /safe-for-%s/);
87		- db_multi_exec("%s", zSchemaUpdates2 /safe-for-%s/);
	29	+ /* Verify that the PLINK table has a new column added by the
	30	+ ** 2014-11-28 schema change. Create it if necessary. This code
	31	+ ** can be removed in the future, once all users have upgraded to the
	32	+ ** 2014-11-28 or later schema.
	33	+ */
	34	+ if( !db_table_has_column("repository","plink","baseid") ){
	35	+ db_multi_exec(
	36	+ "ALTER TABLE repository.plink ADD COLUMN baseid;"
	37	+ );
	38	+ }
	39	+
	40	+ /* Verify that the MLINK table has the newer columns added by the
	41	+ ** 2015-01-24 schema change. Create them if necessary. This code
	42	+ ** can be removed in the future, once all users have upgraded to the
	43	+ ** 2015-01-24 or later schema.
	44	+ */
	45	+ if( !db_table_has_column("repository","mlink","isaux") ){
	46	+ db_begin_transaction();
	47	+ db_multi_exec(
	48	+ "ALTER TABLE repository.mlink ADD COLUMN pmid INTEGER DEFAULT 0;"
	49	+ "ALTER TABLE repository.mlink ADD COLUMN isaux BOOLEAN DEFAULT 0;"
	50	+ );
	51	+ db_end_transaction(0);
	52	+ }
88	53
89		- rc = db_exists("SELECT 1 FROM sqlite_master"
90		- " WHERE name='user' AND sql GLOB '* mtime *'");
91		- if( rc==0 ){
	54	+ /* Add the user.mtime column if it is missing. (2011-04-27)
	55	+ */
	56	+ if( !db_table_has_column("repository", "user", "mtime") ){
92	57	db_multi_exec(
93	58	"CREATE TEMP TABLE temp_user AS SELECT * FROM user;"
94	59	"DROP TABLE user;"
95	60	"CREATE TABLE user(\n"
96	61	" uid INTEGER PRIMARY KEY,\n"
		@@ -109,54 +74,99 @@
109	74	" ipaddr, cexpire, info, now(), photo FROM temp_user;"
110	75	"DROP TABLE temp_user;"
111	76	);
112	77	}
113	78
114		- rc = db_exists("SELECT 1 FROM sqlite_master"
115		- " WHERE name='config' AND sql GLOB '* mtime *'");
116		- if( rc==0 ){
	79	+ /* Add the config.mtime column if it is missing. (2011-04-27)
	80	+ */
	81	+ if( !db_table_has_column("repository", "config", "mtime") ){
117	82	db_multi_exec(
118	83	"ALTER TABLE config ADD COLUMN mtime INTEGER;"
119	84	"UPDATE config SET mtime=now();"
120	85	);
121	86	}
122	87
123		- rc = db_exists("SELECT 1 FROM sqlite_master"
124		- " WHERE name='shun' AND sql GLOB '* mtime *'");
125		- if( rc==0 ){
	88	+ /* Add the shun.mtime and shun.scom columns if they are missing.
	89	+ ** (2011-04-27)
	90	+ */
	91	+ if( !db_table_has_column("repository", "shun", "mtime") ){
126	92	db_multi_exec(
127	93	"ALTER TABLE shun ADD COLUMN mtime INTEGER;"
128	94	"ALTER TABLE shun ADD COLUMN scom TEXT;"
129	95	"UPDATE shun SET mtime=now();"
130	96	);
131	97	}
132	98
133		- rc = db_exists("SELECT 1 FROM sqlite_master"
134		- " WHERE name='reportfmt' AND sql GLOB '* mtime *'");
135		- if( rc==0 ){
	99	+ /* Add the reportfmt.mtime column if it is missing. (2011-04-27)
	100	+ */
	101	+ if( !db_table_has_column("repository", "reportfmt", "mtime") ){
	102	+ static const char zCreateReportFmtTable[] =
	103	+ @ -- An entry in this table describes a database query that generates a
	104	+ @ -- table of tickets.
	105	+ @ --
	106	+ @ CREATE TABLE IF NOT EXISTS reportfmt(
	107	+ @ rn INTEGER PRIMARY KEY, -- Report number
	108	+ @ owner TEXT, -- Owner of this report format (not used)
	109	+ @ title TEXT UNIQUE, -- Title of this report
	110	+ @ mtime INTEGER, -- Time last modified. Seconds since 1970
	111	+ @ cols TEXT, -- A color-key specification
	112	+ @ sqlcode TEXT -- An SQL SELECT statement for this report
	113	+ @ );
	114	+ ;
136	115	db_multi_exec(
137	116	"CREATE TEMP TABLE old_fmt AS SELECT * FROM reportfmt;"
138	117	"DROP TABLE reportfmt;"
139	118	);
140		- db_multi_exec("%s", zSchemaUpdates2/safe-for-%s/);
	119	+ db_multi_exec("%s", zCreateReportFmtTable/safe-for-%s/);
141	120	db_multi_exec(
142	121	"INSERT OR IGNORE INTO reportfmt(rn,owner,title,cols,sqlcode,mtime)"
143	122	" SELECT rn, owner, title, cols, sqlcode, now() FROM old_fmt;"
144	123	"INSERT OR IGNORE INTO reportfmt(rn,owner,title,cols,sqlcode,mtime)"
145	124	" SELECT rn, owner, title \|\| ' (' \|\| rn \|\| ')', cols, sqlcode, now()"
146	125	" FROM old_fmt;"
147	126	);
148	127	}
149	128
150		- rc = db_exists("SELECT 1 FROM sqlite_master"
151		- " WHERE name='concealed' AND sql GLOB '* mtime *'");
152		- if( rc==0 ){
	129	+ /* Add the concealed.mtime column if it is missing. (2011-04-27)
	130	+ */
	131	+ if( !db_table_has_column("repository", "concealed", "mtime") ){
153	132	db_multi_exec(
154	133	"ALTER TABLE concealed ADD COLUMN mtime INTEGER;"
155	134	"UPDATE concealed SET mtime=now();"
156	135	);
157	136	}
	137	+
	138	+ /* Do the fossil-2.0 updates to the schema. (2017-02-28)
	139	+ */
	140	+ rebuild_schema_update_2_0();
	141	+}
	142	+
	143	+/*
	144	+** Update the repository schema for Fossil version 2.0. (2017-02-28)
	145	+** (1) Change the CHECK constraint on BLOB.UUID so that the length
	146	+** is greater than or equal to 40, not exactly equal to 40.
	147	+*/
	148	+void rebuild_schema_update_2_0(void){
	149	+ char *z = db_text(0, "SELECT sql FROM repository.sqlite_master WHERE name='blob'");
	150	+ if( z ){
	151	+ /* Search for: length(uuid)==40
	152	+ ** 0123456789 12345 */
	153	+ int i;
	154	+ for(i=10; z[i]; i++){
	155	+ if( z[i]=='=' && strncmp(&z[i-6],"(uuid)==40",10)==0 ){
	156	+ z[i] = '>';
	157	+ db_multi_exec(
	158	+ "PRAGMA writable_schema=ON;"
	159	+ "UPDATE repository.sqlite_master SET sql=%Q WHERE name LIKE 'blob';"
	160	+ "PRAGMA writable_schema=OFF;",
	161	+ z
	162	+ );
	163	+ break;
	164	+ }
	165	+ }
	166	+ fossil_free(z);
	167	+ }
158	168	}
159	169
160	170	/*
161	171	** Variables used to store state information about an on-going "rebuild"
162	172	** or "deconstruct".
		@@ -330,37 +340,39 @@
330	340	** extracted in a random order. This feature is used to test the
331	341	** ability of fossil to accept records in any order and still
332	342	** construct a sane repository.
333	343	*/
334	344	int rebuild_db(int randomize, int doOut, int doClustering){
335		- Stmt s;
	345	+ Stmt s, q;
336	346	int errCnt = 0;
337		- char *zTable;
338	347	int incrSize;
	348	+ Blob sql;
339	349
340	350	bag_init(&bagDone);
341	351	ttyOutput = doOut;
342	352	processCnt = 0;
343	353	if (ttyOutput && !g.fQuiet) {
344	354	percent_complete(0);
345	355	}
346	356	rebuild_update_schema();
347		- for(;;){
348		- zTable = db_text(0,
349		- "SELECT name FROM sqlite_master /scan/"
350		- " WHERE type='table'"
351		- " AND name NOT IN ('admin_log', 'blob','delta','rcvfrom','user',"
352		- "'config','shun','private','reportfmt',"
353		- "'concealed','accesslog','modreq',"
354		- "'purgeevent','purgeitem','unversioned')"
355		- " AND name NOT GLOB 'sqlite_*'"
356		- " AND name NOT GLOB 'fx_*'"
357		- );
358		- if( zTable==0 ) break;
359		- db_multi_exec("DROP TABLE %Q", zTable);
360		- free(zTable);
361		- }
	357	+ blob_init(&sql, 0, 0);
	358	+ db_prepare(&q,
	359	+ "SELECT name FROM sqlite_master /scan/"
	360	+ " WHERE type='table'"
	361	+ " AND name NOT IN ('admin_log', 'blob','delta','rcvfrom','user','alias',"
	362	+ "'config','shun','private','reportfmt',"
	363	+ "'concealed','accesslog','modreq',"
	364	+ "'purgeevent','purgeitem','unversioned')"
	365	+ " AND name NOT GLOB 'sqlite_*'"
	366	+ " AND name NOT GLOB 'fx_*'"
	367	+ );
	368	+ while( db_step(&q)==SQLITE_ROW ){
	369	+ blob_appendf(&sql, "DROP TABLE \"%w\";\n", db_column_text(&q,0));
	370	+ }
	371	+ db_finalize(&q);
	372	+ db_multi_exec("%s", blob_str(&sql)/safe-for-%s/);
	373	+ blob_reset(&sql);
362	374	db_multi_exec("%s", zRepositorySchema2/safe-for-%s/);
363	375	ticket_create_table(0);
364	376	shun_artifacts();
365	377
366	378	db_multi_exec(
		@@ -994,11 +1006,11 @@
994	1006	**
995	1007	**
996	1008	** This command exports all artifacts of a given repository and
997	1009	** writes all artifacts to the file system. The DESTINATION directory
998	1010	** will be populated with subdirectories AA and files AA/BBBBBBBBB.., where
999		-** AABBBBBBBBB.. is the 40 character artifact ID, AA the first 2 characters.
	1011	+** AABBBBBBBBB.. is the 40+ character artifact ID, AA the first 2 characters.
1000	1012	** If -L\|--prefixlength is given, the length (default 2) of the directory
1001	1013	** prefix can be set to 0,1,..,9 characters.
1002	1014	**
1003	1015	** Options:
1004	1016	** -R\|--repository REPOSITORY deconstruct given REPOSITORY
1005	1017

	--- src/rebuild.c
	+++ src/rebuild.c
	@@ -21,76 +21,41 @@
21	#include "rebuild.h"
22	#include <assert.h>
23	#include <errno.h>
24
25	/*
26	** Make changes to the stable part of the schema (the part that is not
27	** simply deleted and reconstructed on a rebuild) to bring the schema
28	** up to the latest.
29	*/
30	static const char zSchemaUpdates1[] =
31	@ -- Index on the delta table
32	@ --
33	@ CREATE INDEX IF NOT EXISTS delta_i1 ON delta(srcid);
34	@
35	@ -- Artifacts that should not be processed are identified in the
36	@ -- "shun" table. Artifacts that are control-file forgeries or
37	@ -- spam or artifacts whose contents violate administrative policy
38	@ -- can be shunned in order to prevent them from contaminating
39	@ -- the repository.
40	@ --
41	@ -- Shunned artifacts do not exist in the blob table. Hence they
42	@ -- have not artifact ID (rid) and we thus must store their full
43	@ -- UUID.
44	@ --
45	@ CREATE TABLE IF NOT EXISTS shun(
46	@ uuid UNIQUE, -- UUID of artifact to be shunned. Canonical form
47	@ mtime INTEGER, -- When added. Seconds since 1970
48	@ scom TEXT -- Optional text explaining why the shun occurred
49	@ );
50	@
51	@ -- Artifacts that should not be pushed are stored in the "private"
52	@ -- table.
53	@ --
54	@ CREATE TABLE IF NOT EXISTS private(rid INTEGER PRIMARY KEY);
55	@
56	@ -- Some ticket content (such as the originators email address or contact
57	@ -- information) needs to be obscured to protect privacy. This is achieved
58	@ -- by storing an SHA1 hash of the content. For display, the hash is
59	@ -- mapped back into the original text using this table.
60	@ --
61	@ -- This table contains sensitive information and should not be shared
62	@ -- with unauthorized users.
63	@ --
64	@ CREATE TABLE IF NOT EXISTS concealed(
65	@ hash TEXT PRIMARY KEY, -- The SHA1 hash of content
66	@ mtime INTEGER, -- Time created. Seconds since 1970
67	@ content TEXT -- Content intended to be concealed
68	@ );
69	;
70	static const char zSchemaUpdates2[] =
71	@ -- An entry in this table describes a database query that generates a
72	@ -- table of tickets.
73	@ --
74	@ CREATE TABLE IF NOT EXISTS reportfmt(
75	@ rn INTEGER PRIMARY KEY, -- Report number
76	@ owner TEXT, -- Owner of this report format (not used)
77	@ title TEXT UNIQUE, -- Title of this report
78	@ mtime INTEGER, -- Time last modified. Seconds since 1970
79	@ cols TEXT, -- A color-key specification
80	@ sqlcode TEXT -- An SQL SELECT statement for this report
81	@ );
82	;
83
84	static void rebuild_update_schema(void){
85	int rc;
86	db_multi_exec("%s", zSchemaUpdates1 /safe-for-%s/);
87	db_multi_exec("%s", zSchemaUpdates2 /safe-for-%s/);





















88
89	rc = db_exists("SELECT 1 FROM sqlite_master"
90	" WHERE name='user' AND sql GLOB '* mtime *'");
91	if( rc==0 ){
92	db_multi_exec(
93	"CREATE TEMP TABLE temp_user AS SELECT * FROM user;"
94	"DROP TABLE user;"
95	"CREATE TABLE user(\n"
96	" uid INTEGER PRIMARY KEY,\n"
	@@ -109,54 +74,99 @@
109	" ipaddr, cexpire, info, now(), photo FROM temp_user;"
110	"DROP TABLE temp_user;"
111	);
112	}
113
114	rc = db_exists("SELECT 1 FROM sqlite_master"
115	" WHERE name='config' AND sql GLOB '* mtime *'");
116	if( rc==0 ){
117	db_multi_exec(
118	"ALTER TABLE config ADD COLUMN mtime INTEGER;"
119	"UPDATE config SET mtime=now();"
120	);
121	}
122
123	rc = db_exists("SELECT 1 FROM sqlite_master"
124	" WHERE name='shun' AND sql GLOB '* mtime *'");
125	if( rc==0 ){

126	db_multi_exec(
127	"ALTER TABLE shun ADD COLUMN mtime INTEGER;"
128	"ALTER TABLE shun ADD COLUMN scom TEXT;"
129	"UPDATE shun SET mtime=now();"
130	);
131	}
132
133	rc = db_exists("SELECT 1 FROM sqlite_master"
134	" WHERE name='reportfmt' AND sql GLOB '* mtime *'");
135	if( rc==0 ){













136	db_multi_exec(
137	"CREATE TEMP TABLE old_fmt AS SELECT * FROM reportfmt;"
138	"DROP TABLE reportfmt;"
139	);
140	db_multi_exec("%s", zSchemaUpdates2/safe-for-%s/);
141	db_multi_exec(
142	"INSERT OR IGNORE INTO reportfmt(rn,owner,title,cols,sqlcode,mtime)"
143	" SELECT rn, owner, title, cols, sqlcode, now() FROM old_fmt;"
144	"INSERT OR IGNORE INTO reportfmt(rn,owner,title,cols,sqlcode,mtime)"
145	" SELECT rn, owner, title \|\| ' (' \|\| rn \|\| ')', cols, sqlcode, now()"
146	" FROM old_fmt;"
147	);
148	}
149
150	rc = db_exists("SELECT 1 FROM sqlite_master"
151	" WHERE name='concealed' AND sql GLOB '* mtime *'");
152	if( rc==0 ){
153	db_multi_exec(
154	"ALTER TABLE concealed ADD COLUMN mtime INTEGER;"
155	"UPDATE concealed SET mtime=now();"
156	);
157	}































158	}
159
160	/*
161	** Variables used to store state information about an on-going "rebuild"
162	** or "deconstruct".
	@@ -330,37 +340,39 @@
330	** extracted in a random order. This feature is used to test the
331	** ability of fossil to accept records in any order and still
332	** construct a sane repository.
333	*/
334	int rebuild_db(int randomize, int doOut, int doClustering){
335	Stmt s;
336	int errCnt = 0;
337	char *zTable;
338	int incrSize;

339
340	bag_init(&bagDone);
341	ttyOutput = doOut;
342	processCnt = 0;
343	if (ttyOutput && !g.fQuiet) {
344	percent_complete(0);
345	}
346	rebuild_update_schema();
347	for(;;){
348	zTable = db_text(0,
349	"SELECT name FROM sqlite_master /scan/"
350	" WHERE type='table'"
351	" AND name NOT IN ('admin_log', 'blob','delta','rcvfrom','user',"
352	"'config','shun','private','reportfmt',"
353	"'concealed','accesslog','modreq',"
354	"'purgeevent','purgeitem','unversioned')"
355	" AND name NOT GLOB 'sqlite_*'"
356	" AND name NOT GLOB 'fx_*'"
357	);
358	if( zTable==0 ) break;
359	db_multi_exec("DROP TABLE %Q", zTable);
360	free(zTable);
361	}


362	db_multi_exec("%s", zRepositorySchema2/safe-for-%s/);
363	ticket_create_table(0);
364	shun_artifacts();
365
366	db_multi_exec(
	@@ -994,11 +1006,11 @@
994	**
995	**
996	** This command exports all artifacts of a given repository and
997	** writes all artifacts to the file system. The DESTINATION directory
998	** will be populated with subdirectories AA and files AA/BBBBBBBBB.., where
999	** AABBBBBBBBB.. is the 40 character artifact ID, AA the first 2 characters.
1000	** If -L\|--prefixlength is given, the length (default 2) of the directory
1001	** prefix can be set to 0,1,..,9 characters.
1002	**
1003	** Options:
1004	** -R\|--repository REPOSITORY deconstruct given REPOSITORY
1005

	--- src/rebuild.c
	+++ src/rebuild.c
	@@ -21,76 +21,41 @@
21	#include "rebuild.h"
22	#include <assert.h>
23	#include <errno.h>
24
25	/*
26	** Update the schema as necessary


27	*/






















































28	static void rebuild_update_schema(void){
29	/* Verify that the PLINK table has a new column added by the
30	** 2014-11-28 schema change. Create it if necessary. This code
31	** can be removed in the future, once all users have upgraded to the
32	** 2014-11-28 or later schema.
33	*/
34	if( !db_table_has_column("repository","plink","baseid") ){
35	db_multi_exec(
36	"ALTER TABLE repository.plink ADD COLUMN baseid;"
37	);
38	}
39
40	/* Verify that the MLINK table has the newer columns added by the
41	** 2015-01-24 schema change. Create them if necessary. This code
42	** can be removed in the future, once all users have upgraded to the
43	** 2015-01-24 or later schema.
44	*/
45	if( !db_table_has_column("repository","mlink","isaux") ){
46	db_begin_transaction();
47	db_multi_exec(
48	"ALTER TABLE repository.mlink ADD COLUMN pmid INTEGER DEFAULT 0;"
49	"ALTER TABLE repository.mlink ADD COLUMN isaux BOOLEAN DEFAULT 0;"
50	);
51	db_end_transaction(0);
52	}
53
54	/* Add the user.mtime column if it is missing. (2011-04-27)
55	*/
56	if( !db_table_has_column("repository", "user", "mtime") ){
57	db_multi_exec(
58	"CREATE TEMP TABLE temp_user AS SELECT * FROM user;"
59	"DROP TABLE user;"
60	"CREATE TABLE user(\n"
61	" uid INTEGER PRIMARY KEY,\n"
	@@ -109,54 +74,99 @@
74	" ipaddr, cexpire, info, now(), photo FROM temp_user;"
75	"DROP TABLE temp_user;"
76	);
77	}
78
79	/* Add the config.mtime column if it is missing. (2011-04-27)
80	*/
81	if( !db_table_has_column("repository", "config", "mtime") ){
82	db_multi_exec(
83	"ALTER TABLE config ADD COLUMN mtime INTEGER;"
84	"UPDATE config SET mtime=now();"
85	);
86	}
87
88	/* Add the shun.mtime and shun.scom columns if they are missing.
89	** (2011-04-27)
90	*/
91	if( !db_table_has_column("repository", "shun", "mtime") ){
92	db_multi_exec(
93	"ALTER TABLE shun ADD COLUMN mtime INTEGER;"
94	"ALTER TABLE shun ADD COLUMN scom TEXT;"
95	"UPDATE shun SET mtime=now();"
96	);
97	}
98
99	/* Add the reportfmt.mtime column if it is missing. (2011-04-27)
100	*/
101	if( !db_table_has_column("repository", "reportfmt", "mtime") ){
102	static const char zCreateReportFmtTable[] =
103	@ -- An entry in this table describes a database query that generates a
104	@ -- table of tickets.
105	@ --
106	@ CREATE TABLE IF NOT EXISTS reportfmt(
107	@ rn INTEGER PRIMARY KEY, -- Report number
108	@ owner TEXT, -- Owner of this report format (not used)
109	@ title TEXT UNIQUE, -- Title of this report
110	@ mtime INTEGER, -- Time last modified. Seconds since 1970
111	@ cols TEXT, -- A color-key specification
112	@ sqlcode TEXT -- An SQL SELECT statement for this report
113	@ );
114	;
115	db_multi_exec(
116	"CREATE TEMP TABLE old_fmt AS SELECT * FROM reportfmt;"
117	"DROP TABLE reportfmt;"
118	);
119	db_multi_exec("%s", zCreateReportFmtTable/safe-for-%s/);
120	db_multi_exec(
121	"INSERT OR IGNORE INTO reportfmt(rn,owner,title,cols,sqlcode,mtime)"
122	" SELECT rn, owner, title, cols, sqlcode, now() FROM old_fmt;"
123	"INSERT OR IGNORE INTO reportfmt(rn,owner,title,cols,sqlcode,mtime)"
124	" SELECT rn, owner, title \|\| ' (' \|\| rn \|\| ')', cols, sqlcode, now()"
125	" FROM old_fmt;"
126	);
127	}
128
129	/* Add the concealed.mtime column if it is missing. (2011-04-27)
130	*/
131	if( !db_table_has_column("repository", "concealed", "mtime") ){
132	db_multi_exec(
133	"ALTER TABLE concealed ADD COLUMN mtime INTEGER;"
134	"UPDATE concealed SET mtime=now();"
135	);
136	}
137
138	/* Do the fossil-2.0 updates to the schema. (2017-02-28)
139	*/
140	rebuild_schema_update_2_0();
141	}
142
143	/*
144	** Update the repository schema for Fossil version 2.0. (2017-02-28)
145	** (1) Change the CHECK constraint on BLOB.UUID so that the length
146	** is greater than or equal to 40, not exactly equal to 40.
147	*/
148	void rebuild_schema_update_2_0(void){
149	char *z = db_text(0, "SELECT sql FROM repository.sqlite_master WHERE name='blob'");
150	if( z ){
151	/* Search for: length(uuid)==40
152	** 0123456789 12345 */
153	int i;
154	for(i=10; z[i]; i++){
155	if( z[i]=='=' && strncmp(&z[i-6],"(uuid)==40",10)==0 ){
156	z[i] = '>';
157	db_multi_exec(
158	"PRAGMA writable_schema=ON;"
159	"UPDATE repository.sqlite_master SET sql=%Q WHERE name LIKE 'blob';"
160	"PRAGMA writable_schema=OFF;",
161	z
162	);
163	break;
164	}
165	}
166	fossil_free(z);
167	}
168	}
169
170	/*
171	** Variables used to store state information about an on-going "rebuild"
172	** or "deconstruct".
	@@ -330,37 +340,39 @@
340	** extracted in a random order. This feature is used to test the
341	** ability of fossil to accept records in any order and still
342	** construct a sane repository.
343	*/
344	int rebuild_db(int randomize, int doOut, int doClustering){
345	Stmt s, q;
346	int errCnt = 0;

347	int incrSize;
348	Blob sql;
349
350	bag_init(&bagDone);
351	ttyOutput = doOut;
352	processCnt = 0;
353	if (ttyOutput && !g.fQuiet) {
354	percent_complete(0);
355	}
356	rebuild_update_schema();
357	blob_init(&sql, 0, 0);
358	db_prepare(&q,
359	"SELECT name FROM sqlite_master /scan/"
360	" WHERE type='table'"
361	" AND name NOT IN ('admin_log', 'blob','delta','rcvfrom','user','alias',"
362	"'config','shun','private','reportfmt',"
363	"'concealed','accesslog','modreq',"
364	"'purgeevent','purgeitem','unversioned')"
365	" AND name NOT GLOB 'sqlite_*'"
366	" AND name NOT GLOB 'fx_*'"
367	);
368	while( db_step(&q)==SQLITE_ROW ){
369	blob_appendf(&sql, "DROP TABLE \"%w\";\n", db_column_text(&q,0));
370	}
371	db_finalize(&q);
372	db_multi_exec("%s", blob_str(&sql)/safe-for-%s/);
373	blob_reset(&sql);
374	db_multi_exec("%s", zRepositorySchema2/safe-for-%s/);
375	ticket_create_table(0);
376	shun_artifacts();
377
378	db_multi_exec(
	@@ -994,11 +1006,11 @@
1006	**
1007	**
1008	** This command exports all artifacts of a given repository and
1009	** writes all artifacts to the file system. The DESTINATION directory
1010	** will be populated with subdirectories AA and files AA/BBBBBBBBB.., where
1011	** AABBBBBBBBB.. is the 40+ character artifact ID, AA the first 2 characters.
1012	** If -L\|--prefixlength is given, the length (default 2) of the directory
1013	** prefix can be set to 0,1,..,9 characters.
1014	**
1015	** Options:
1016	** -R\|--repository REPOSITORY deconstruct given REPOSITORY
1017

M src/schema.c

+2 -2

		--- src/schema.c
		+++ src/schema.c
		@@ -79,13 +79,13 @@
79	79	@ --
80	80	@ CREATE TABLE blob(
81	81	@ rid INTEGER PRIMARY KEY, -- Record ID
82	82	@ rcvid INTEGER, -- Origin of this record
83	83	@ size INTEGER, -- Size of content. -1 for a phantom.
84		-@ uuid TEXT UNIQUE NOT NULL, -- SHA1 hash of the content
	84	+@ uuid TEXT UNIQUE NOT NULL, -- hash of the content
85	85	@ content BLOB, -- Compressed content of this record
86		-@ CHECK( length(uuid)==40 AND rid>0 )
	86	+@ CHECK( length(uuid)>=40 AND rid>0 )
87	87	@ );
88	88	@ CREATE TABLE delta(
89	89	@ rid INTEGER PRIMARY KEY, -- BLOB that is delta-compressed
90	90	@ srcid INTEGER NOT NULL REFERENCES blob -- Baseline for delta-compression
91	91	@ );
92	92

	--- src/schema.c
	+++ src/schema.c
	@@ -79,13 +79,13 @@
79	@ --
80	@ CREATE TABLE blob(
81	@ rid INTEGER PRIMARY KEY, -- Record ID
82	@ rcvid INTEGER, -- Origin of this record
83	@ size INTEGER, -- Size of content. -1 for a phantom.
84	@ uuid TEXT UNIQUE NOT NULL, -- SHA1 hash of the content
85	@ content BLOB, -- Compressed content of this record
86	@ CHECK( length(uuid)==40 AND rid>0 )
87	@ );
88	@ CREATE TABLE delta(
89	@ rid INTEGER PRIMARY KEY, -- BLOB that is delta-compressed
90	@ srcid INTEGER NOT NULL REFERENCES blob -- Baseline for delta-compression
91	@ );
92

	--- src/schema.c
	+++ src/schema.c
	@@ -79,13 +79,13 @@
79	@ --
80	@ CREATE TABLE blob(
81	@ rid INTEGER PRIMARY KEY, -- Record ID
82	@ rcvid INTEGER, -- Origin of this record
83	@ size INTEGER, -- Size of content. -1 for a phantom.
84	@ uuid TEXT UNIQUE NOT NULL, -- hash of the content
85	@ content BLOB, -- Compressed content of this record
86	@ CHECK( length(uuid)>=40 AND rid>0 )
87	@ );
88	@ CREATE TABLE delta(
89	@ rid INTEGER PRIMARY KEY, -- BLOB that is delta-compressed
90	@ srcid INTEGER NOT NULL REFERENCES blob -- Baseline for delta-compression
91	@ );
92

M src/setup.c

		--- src/setup.c
		+++ src/setup.c
		@@ -1108,10 +1108,18 @@
1108	1108	"remote_user_ok", "remote_user_ok", 0, 0);
1109	1109	@ <p>When enabled, if the REMOTE_USER environment variable is set to the
1110	1110	@ login name of a valid user and no other login credentials are available,
1111	1111	@ then the REMOTE_USER is accepted as an authenticated user.
1112	1112	@ </p>
	1113	+ @
	1114	+ @ <hr />
	1115	+ onoff_attribute("Allow HTTP_AUTHENTICATION authentication",
	1116	+ "http_authentication_ok", "http_authentication_ok", 0, 0);
	1117	+ @ <p>When enabled, allow the use of the HTTP_AUTHENTICATION environment
	1118	+ @ variable or the "Authentication:" HTTP header to find the username and
	1119	+ @ password. This is another way of supporting Basic Authenitication.
	1120	+ @ </p>
1113	1121	@
1114	1122	@ <hr />
1115	1123	entry_attribute("IP address terms used in login cookie", 3,
1116	1124	"ip-prefix-terms", "ipt", "2", 0);
1117	1125	@ <p>The number of octets of of the IP address used in the login cookie.
1118	1126

	--- src/setup.c
	+++ src/setup.c
	@@ -1108,10 +1108,18 @@
1108	"remote_user_ok", "remote_user_ok", 0, 0);
1109	@ <p>When enabled, if the REMOTE_USER environment variable is set to the
1110	@ login name of a valid user and no other login credentials are available,
1111	@ then the REMOTE_USER is accepted as an authenticated user.
1112	@ </p>








1113	@
1114	@ <hr />
1115	entry_attribute("IP address terms used in login cookie", 3,
1116	"ip-prefix-terms", "ipt", "2", 0);
1117	@ <p>The number of octets of of the IP address used in the login cookie.
1118

	--- src/setup.c
	+++ src/setup.c
	@@ -1108,10 +1108,18 @@
1108	"remote_user_ok", "remote_user_ok", 0, 0);
1109	@ <p>When enabled, if the REMOTE_USER environment variable is set to the
1110	@ login name of a valid user and no other login credentials are available,
1111	@ then the REMOTE_USER is accepted as an authenticated user.
1112	@ </p>
1113	@
1114	@ <hr />
1115	onoff_attribute("Allow HTTP_AUTHENTICATION authentication",
1116	"http_authentication_ok", "http_authentication_ok", 0, 0);
1117	@ <p>When enabled, allow the use of the HTTP_AUTHENTICATION environment
1118	@ variable or the "Authentication:" HTTP header to find the username and
1119	@ password. This is another way of supporting Basic Authenitication.
1120	@ </p>
1121	@
1122	@ <hr />
1123	entry_attribute("IP address terms used in login cookie", 3,
1124	"ip-prefix-terms", "ipt", "2", 0);
1125	@ <p>The number of octets of of the IP address used in the login cookie.
1126

M src/sha1.c

+79 -11

		--- src/sha1.c
		+++ src/sha1.c
		@@ -1,22 +1,88 @@
1	1	/*
	2	+** Copyright (c) 2006 D. Richard Hipp
	3	+**
	4	+** This program is free software; you can redistribute it and/or
	5	+** modify it under the terms of the Simplified BSD License (also
	6	+** known as the "2-Clause License" or "FreeBSD License".)
	7	+**
	8	+** This program is distributed in the hope that it will be useful,
	9	+** but without any warranty; without even the implied warranty of
	10	+** merchantability or fitness for a particular purpose.
	11	+**
	12	+** Author contact information:
	13	+** [email protected]
	14	+** http://www.hwaci.com/drh/
	15	+**
	16	+*******************************************************************************
	17	+**
2	18	** This implementation of SHA1.
3	19	*/
4	20	#include "config.h"
5	21	#include <sys/types.h>
	22	+#include <stdint.h>
6	23	#include "sha1.h"
7	24
8		-#ifdef FOSSIL_ENABLE_SSL
	25	+
	26	+/*
	27	+** SHA1 Implementation #1 is the hardened SHA1 implementation by
	28	+** Marc Stevens. Code obtained from GitHub
	29	+**
	30	+** https://github.com/cr-marcstevens/sha1collisiondetection
	31	+**
	32	+** Downloaded on 2017-03-01 then repackaged to work with Fossil
	33	+** and makeheaders.
	34	+*/
	35	+#if FOSSIL_HARDENED_SHA1
	36	+
	37	+#if INTERFACE
	38	+typedef void(collision_block_callback)(uint64_t, const uint32_t, const uint32_t, const uint32_t, const uint32_t*);
	39	+struct SHA1_CTX {
	40	+ uint64_t total;
	41	+ uint32_t ihv[5];
	42	+ unsigned char buffer[64];
	43	+ int bigendian;
	44	+ int found_collision;
	45	+ int safe_hash;
	46	+ int detect_coll;
	47	+ int ubc_check;
	48	+ int reduced_round_coll;
	49	+ collision_block_callback callback;
	50	+
	51	+ uint32_t ihv1[5];
	52	+ uint32_t ihv2[5];
	53	+ uint32_t m1[80];
	54	+ uint32_t m2[80];
	55	+ uint32_t states[80][5];
	56	+};
	57	+#endif
	58	+void SHA1DCInit(SHA1_CTX*);
	59	+void SHA1DCUpdate(SHA1_CTX, const unsigned char, unsigned);
	60	+int SHA1DCFinal(unsigned char[20], SHA1_CTX*);
	61	+
	62	+#define SHA1Context SHA1_CTX
	63	+#define SHA1Init SHA1DCInit
	64	+#define SHA1Update SHA1DCUpdate
	65	+#define SHA1Final SHA1DCFinal
	66	+
	67	+/*
	68	+** SHA1 Implementation #2: use the SHA1 algorithm built into SSL
	69	+*/
	70	+#elif defined(FOSSIL_ENABLE_SSL)
9	71
10	72	# include <openssl/sha.h>
11	73	# define SHA1Context SHA_CTX
12	74	# define SHA1Init SHA1_Init
13	75	# define SHA1Update SHA1_Update
14		-# define SHA1Final(a,b) SHA1_Final(b,a)
	76	+# define SHA1Final SHA1_Final
15	77
	78	+/*
	79	+** SHA1 Implementation #3: If none of the previous two SHA1
	80	+** algorithms work, there is this built-in. This built-in was the
	81	+** original implementation used by Fossil.
	82	+*/
16	83	#else
17		-
18	84	/*
19	85	** The SHA1 implementation below is adapted from:
20	86	**
21	87	** $NetBSD: sha1.c,v 1.6 2009/11/06 20:31:18 joerg Exp $
22	88	** $OpenBSD: sha1.c,v 1.9 1997/07/23 21:12:32 kstailey Exp $
		@@ -53,10 +119,13 @@
53	119	/* Generic C equivalent */
54	120	#define SHA_ROT(x,l,r) ((x) << (l) \| (x) >> (r))
55	121	#define rol(x,k) SHA_ROT(x,k,32-(k))
56	122	#define ror(x,k) SHA_ROT(x,32-(k),k)
57	123	#endif
	124	+
	125	+
	126	+
58	127
59	128
60	129	#define blk0le(i) (block[i] = (ror(block[i],8)&0xFF00FF00) \
61	130	\|(rol(block[i],8)&0x00FF00FF))
62	131	#define blk0be(i) block[i]
		@@ -188,11 +257,11 @@
188	257
189	258
190	259	/*
191	260	* Add padding and return the message digest.
192	261	*/
193		-static void SHA1Final(SHA1Context *context, unsigned char digest[20]){
	262	+static void SHA1Final(unsigned char digest, SHA1Context context){
194	263	unsigned int i;
195	264	unsigned char finalcount[8];
196	265
197	266	for (i = 0; i < 8; i++) {
198	267	finalcount[i] = (unsigned char)((context->count[(i >= 4 ? 0 : 1)]
		@@ -207,12 +276,11 @@
207	276	for (i = 0; i < 20; i++)
208	277	digest[i] = (unsigned char)
209	278	((context->state[i>>2] >> ((3-(i & 3)) * 8) ) & 255);
210	279	}
211	280	}
212		-#endif
213		-
	281	+#endif /* Built-in SHA1 implemenation */
214	282
215	283	/*
216	284	** Convert a digest into base-16. digest should be declared as
217	285	** "unsigned char digest[20]" in the calling function. The SHA1
218	286	** digest is stored in the first 20 bytes. zBuf should
		@@ -268,11 +336,11 @@
268	336	*/
269	337	char sha1sum_finish(Blob pOut){
270	338	unsigned char zResult[20];
271	339	static char zOut[41];
272	340	sha1sum_step_text(0,0);
273		- SHA1Final(&incrCtx, zResult);
	341	+ SHA1Final(zResult, &incrCtx);
274	342	incrInit = 0;
275	343	DigestToBase16(zResult, zOut);
276	344	if( pOut ){
277	345	blob_zero(pOut);
278	346	blob_append(pOut, zOut, 40);
		@@ -316,11 +384,11 @@
316	384	SHA1Update(&ctx, (unsigned char*)zBuf, (unsigned)n);
317	385	}
318	386	fclose(in);
319	387	blob_zero(pCksum);
320	388	blob_resize(pCksum, 40);
321		- SHA1Final(&ctx, zResult);
	389	+ SHA1Final(zResult, &ctx);
322	390	DigestToBase16(zResult, blob_buffer(pCksum));
323	391	return 0;
324	392	}
325	393
326	394	/*
		@@ -340,11 +408,11 @@
340	408	blob_reset(pCksum);
341	409	}else{
342	410	blob_zero(pCksum);
343	411	}
344	412	blob_resize(pCksum, 40);
345		- SHA1Final(&ctx, zResult);
	413	+ SHA1Final(zResult, &ctx);
346	414	DigestToBase16(zResult, blob_buffer(pCksum));
347	415	return 0;
348	416	}
349	417
350	418	/*
		@@ -356,11 +424,11 @@
356	424	unsigned char zResult[20];
357	425	char zDigest[41];
358	426
359	427	SHA1Init(&ctx);
360	428	SHA1Update(&ctx, (unsigned const char*)zIn, strlen(zIn));
361		- SHA1Final(&ctx, zResult);
	429	+ SHA1Final(zResult, &ctx);
362	430	DigestToBase16(zResult, zDigest);
363	431	return mprintf("%s", zDigest);
364	432	}
365	433
366	434	/*
		@@ -407,11 +475,11 @@
407	475	SHA1Update(&ctx, (unsigned char*)zProjCode, strlen(zProjCode));
408	476	SHA1Update(&ctx, (unsigned char*)"/", 1);
409	477	SHA1Update(&ctx, (unsigned char*)zLogin, strlen(zLogin));
410	478	SHA1Update(&ctx, (unsigned char*)"/", 1);
411	479	SHA1Update(&ctx, (unsigned const char*)zPw, strlen(zPw));
412		- SHA1Final(&ctx, zResult);
	480	+ SHA1Final(zResult, &ctx);
413	481	DigestToBase16(zResult, zDigest);
414	482	return mprintf("%s", zDigest);
415	483	}
416	484
417	485	/*
418	486
419	487	ADDED src/sha1hard.c
420	488	ADDED src/sha3.c

	--- src/sha1.c
	+++ src/sha1.c
	@@ -1,22 +1,88 @@
1	/*
















2	** This implementation of SHA1.
3	*/
4	#include "config.h"
5	#include <sys/types.h>

6	#include "sha1.h"
7
8	#ifdef FOSSIL_ENABLE_SSL













































9
10	# include <openssl/sha.h>
11	# define SHA1Context SHA_CTX
12	# define SHA1Init SHA1_Init
13	# define SHA1Update SHA1_Update
14	# define SHA1Final(a,b) SHA1_Final(b,a)
15





16	#else
17
18	/*
19	** The SHA1 implementation below is adapted from:
20	**
21	** $NetBSD: sha1.c,v 1.6 2009/11/06 20:31:18 joerg Exp $
22	** $OpenBSD: sha1.c,v 1.9 1997/07/23 21:12:32 kstailey Exp $
	@@ -53,10 +119,13 @@
53	/* Generic C equivalent */
54	#define SHA_ROT(x,l,r) ((x) << (l) \| (x) >> (r))
55	#define rol(x,k) SHA_ROT(x,k,32-(k))
56	#define ror(x,k) SHA_ROT(x,32-(k),k)
57	#endif



58
59
60	#define blk0le(i) (block[i] = (ror(block[i],8)&0xFF00FF00) \
61	\|(rol(block[i],8)&0x00FF00FF))
62	#define blk0be(i) block[i]
	@@ -188,11 +257,11 @@
188
189
190	/*
191	* Add padding and return the message digest.
192	*/
193	static void SHA1Final(SHA1Context *context, unsigned char digest[20]){
194	unsigned int i;
195	unsigned char finalcount[8];
196
197	for (i = 0; i < 8; i++) {
198	finalcount[i] = (unsigned char)((context->count[(i >= 4 ? 0 : 1)]
	@@ -207,12 +276,11 @@
207	for (i = 0; i < 20; i++)
208	digest[i] = (unsigned char)
209	((context->state[i>>2] >> ((3-(i & 3)) * 8) ) & 255);
210	}
211	}
212	#endif
213
214
215	/*
216	** Convert a digest into base-16. digest should be declared as
217	** "unsigned char digest[20]" in the calling function. The SHA1
218	** digest is stored in the first 20 bytes. zBuf should
	@@ -268,11 +336,11 @@
268	*/
269	char sha1sum_finish(Blob pOut){
270	unsigned char zResult[20];
271	static char zOut[41];
272	sha1sum_step_text(0,0);
273	SHA1Final(&incrCtx, zResult);
274	incrInit = 0;
275	DigestToBase16(zResult, zOut);
276	if( pOut ){
277	blob_zero(pOut);
278	blob_append(pOut, zOut, 40);
	@@ -316,11 +384,11 @@
316	SHA1Update(&ctx, (unsigned char*)zBuf, (unsigned)n);
317	}
318	fclose(in);
319	blob_zero(pCksum);
320	blob_resize(pCksum, 40);
321	SHA1Final(&ctx, zResult);
322	DigestToBase16(zResult, blob_buffer(pCksum));
323	return 0;
324	}
325
326	/*
	@@ -340,11 +408,11 @@
340	blob_reset(pCksum);
341	}else{
342	blob_zero(pCksum);
343	}
344	blob_resize(pCksum, 40);
345	SHA1Final(&ctx, zResult);
346	DigestToBase16(zResult, blob_buffer(pCksum));
347	return 0;
348	}
349
350	/*
	@@ -356,11 +424,11 @@
356	unsigned char zResult[20];
357	char zDigest[41];
358
359	SHA1Init(&ctx);
360	SHA1Update(&ctx, (unsigned const char*)zIn, strlen(zIn));
361	SHA1Final(&ctx, zResult);
362	DigestToBase16(zResult, zDigest);
363	return mprintf("%s", zDigest);
364	}
365
366	/*
	@@ -407,11 +475,11 @@
407	SHA1Update(&ctx, (unsigned char*)zProjCode, strlen(zProjCode));
408	SHA1Update(&ctx, (unsigned char*)"/", 1);
409	SHA1Update(&ctx, (unsigned char*)zLogin, strlen(zLogin));
410	SHA1Update(&ctx, (unsigned char*)"/", 1);
411	SHA1Update(&ctx, (unsigned const char*)zPw, strlen(zPw));
412	SHA1Final(&ctx, zResult);
413	DigestToBase16(zResult, zDigest);
414	return mprintf("%s", zDigest);
415	}
416
417	/*
418
419	DDED src/sha1hard.c
420	DDED src/sha3.c

	--- src/sha1.c
	+++ src/sha1.c
	@@ -1,22 +1,88 @@
1	/*
2	** Copyright (c) 2006 D. Richard Hipp
3	**
4	** This program is free software; you can redistribute it and/or
5	** modify it under the terms of the Simplified BSD License (also
6	** known as the "2-Clause License" or "FreeBSD License".)
7	**
8	** This program is distributed in the hope that it will be useful,
9	** but without any warranty; without even the implied warranty of
10	** merchantability or fitness for a particular purpose.
11	**
12	** Author contact information:
13	** [email protected]
14	** http://www.hwaci.com/drh/
15	**
16	*******************************************************************************
17	**
18	** This implementation of SHA1.
19	*/
20	#include "config.h"
21	#include <sys/types.h>
22	#include <stdint.h>
23	#include "sha1.h"
24
25
26	/*
27	** SHA1 Implementation #1 is the hardened SHA1 implementation by
28	** Marc Stevens. Code obtained from GitHub
29	**
30	** https://github.com/cr-marcstevens/sha1collisiondetection
31	**
32	** Downloaded on 2017-03-01 then repackaged to work with Fossil
33	** and makeheaders.
34	*/
35	#if FOSSIL_HARDENED_SHA1
36
37	#if INTERFACE
38	typedef void(collision_block_callback)(uint64_t, const uint32_t, const uint32_t, const uint32_t, const uint32_t*);
39	struct SHA1_CTX {
40	uint64_t total;
41	uint32_t ihv[5];
42	unsigned char buffer[64];
43	int bigendian;
44	int found_collision;
45	int safe_hash;
46	int detect_coll;
47	int ubc_check;
48	int reduced_round_coll;
49	collision_block_callback callback;
50
51	uint32_t ihv1[5];
52	uint32_t ihv2[5];
53	uint32_t m1[80];
54	uint32_t m2[80];
55	uint32_t states[80][5];
56	};
57	#endif
58	void SHA1DCInit(SHA1_CTX*);
59	void SHA1DCUpdate(SHA1_CTX, const unsigned char, unsigned);
60	int SHA1DCFinal(unsigned char[20], SHA1_CTX*);
61
62	#define SHA1Context SHA1_CTX
63	#define SHA1Init SHA1DCInit
64	#define SHA1Update SHA1DCUpdate
65	#define SHA1Final SHA1DCFinal
66
67	/*
68	** SHA1 Implementation #2: use the SHA1 algorithm built into SSL
69	*/
70	#elif defined(FOSSIL_ENABLE_SSL)
71
72	# include <openssl/sha.h>
73	# define SHA1Context SHA_CTX
74	# define SHA1Init SHA1_Init
75	# define SHA1Update SHA1_Update
76	# define SHA1Final SHA1_Final
77
78	/*
79	** SHA1 Implementation #3: If none of the previous two SHA1
80	** algorithms work, there is this built-in. This built-in was the
81	** original implementation used by Fossil.
82	*/
83	#else

84	/*
85	** The SHA1 implementation below is adapted from:
86	**
87	** $NetBSD: sha1.c,v 1.6 2009/11/06 20:31:18 joerg Exp $
88	** $OpenBSD: sha1.c,v 1.9 1997/07/23 21:12:32 kstailey Exp $
	@@ -53,10 +119,13 @@
119	/* Generic C equivalent */
120	#define SHA_ROT(x,l,r) ((x) << (l) \| (x) >> (r))
121	#define rol(x,k) SHA_ROT(x,k,32-(k))
122	#define ror(x,k) SHA_ROT(x,32-(k),k)
123	#endif
124
125
126
127
128
129	#define blk0le(i) (block[i] = (ror(block[i],8)&0xFF00FF00) \
130	\|(rol(block[i],8)&0x00FF00FF))
131	#define blk0be(i) block[i]
	@@ -188,11 +257,11 @@
257
258
259	/*
260	* Add padding and return the message digest.
261	*/
262	static void SHA1Final(unsigned char digest, SHA1Context context){
263	unsigned int i;
264	unsigned char finalcount[8];
265
266	for (i = 0; i < 8; i++) {
267	finalcount[i] = (unsigned char)((context->count[(i >= 4 ? 0 : 1)]
	@@ -207,12 +276,11 @@
276	for (i = 0; i < 20; i++)
277	digest[i] = (unsigned char)
278	((context->state[i>>2] >> ((3-(i & 3)) * 8) ) & 255);
279	}
280	}
281	#endif /* Built-in SHA1 implemenation */

282
283	/*
284	** Convert a digest into base-16. digest should be declared as
285	** "unsigned char digest[20]" in the calling function. The SHA1
286	** digest is stored in the first 20 bytes. zBuf should
	@@ -268,11 +336,11 @@
336	*/
337	char sha1sum_finish(Blob pOut){
338	unsigned char zResult[20];
339	static char zOut[41];
340	sha1sum_step_text(0,0);
341	SHA1Final(zResult, &incrCtx);
342	incrInit = 0;
343	DigestToBase16(zResult, zOut);
344	if( pOut ){
345	blob_zero(pOut);
346	blob_append(pOut, zOut, 40);
	@@ -316,11 +384,11 @@
384	SHA1Update(&ctx, (unsigned char*)zBuf, (unsigned)n);
385	}
386	fclose(in);
387	blob_zero(pCksum);
388	blob_resize(pCksum, 40);
389	SHA1Final(zResult, &ctx);
390	DigestToBase16(zResult, blob_buffer(pCksum));
391	return 0;
392	}
393
394	/*
	@@ -340,11 +408,11 @@
408	blob_reset(pCksum);
409	}else{
410	blob_zero(pCksum);
411	}
412	blob_resize(pCksum, 40);
413	SHA1Final(zResult, &ctx);
414	DigestToBase16(zResult, blob_buffer(pCksum));
415	return 0;
416	}
417
418	/*
	@@ -356,11 +424,11 @@
424	unsigned char zResult[20];
425	char zDigest[41];
426
427	SHA1Init(&ctx);
428	SHA1Update(&ctx, (unsigned const char*)zIn, strlen(zIn));
429	SHA1Final(zResult, &ctx);
430	DigestToBase16(zResult, zDigest);
431	return mprintf("%s", zDigest);
432	}
433
434	/*
	@@ -407,11 +475,11 @@
475	SHA1Update(&ctx, (unsigned char*)zProjCode, strlen(zProjCode));
476	SHA1Update(&ctx, (unsigned char*)"/", 1);
477	SHA1Update(&ctx, (unsigned char*)zLogin, strlen(zLogin));
478	SHA1Update(&ctx, (unsigned char*)"/", 1);
479	SHA1Update(&ctx, (unsigned const char*)zPw, strlen(zPw));
480	SHA1Final(zResult, &ctx);
481	DigestToBase16(zResult, zDigest);
482	return mprintf("%s", zDigest);
483	}
484
485	/*
486
487	DDED src/sha1hard.c
488	DDED src/sha3.c

M src/sha1hard.c

No diff available

M src/sha3.c

		--- a/src/sha3.c
		+++ b/src/sha3.c
		@@ -0,0 +1,2 @@
	1	+/*
	2	+*testfile_wd 128and 5124

	--- a/src/sha3.c
	+++ b/src/sha3.c
	@@ -0,0 +1,2 @@

	--- a/src/sha3.c
	+++ b/src/sha3.c
	@@ -0,0 +1,2 @@
1	/*
2	*testfile_wd 128and 5124

M src/shell.c

+1415 -233

		--- src/shell.c
		+++ src/shell.c
		@@ -452,32 +452,10 @@
452	452	while( IsDigit(*z) ){ z++; }
453	453	if( realnum ) *realnum = 1;
454	454	}
455	455	return *z==0;
456	456	}
457		-
458		-/*
459		-** A global char* and an SQL function to access its current value
460		-** from within an SQL statement. This program used to use the
461		-** sqlite_exec_printf() API to substitue a string into an SQL statement.
462		-** The correct way to do this with sqlite3 is to use the bind API, but
463		-** since the shell is built around the callback paradigm it would be a lot
464		-** of work. Instead just use this hack, which is quite harmless.
465		-*/
466		-static const char *zShellStatic = 0;
467		-static void shellstaticFunc(
468		- sqlite3_context *context,
469		- int argc,
470		- sqlite3_value **argv
471		-){
472		- assert( 0==argc );
473		- assert( zShellStatic );
474		- UNUSED_PARAMETER(argc);
475		- UNUSED_PARAMETER(argv);
476		- sqlite3_result_text(context, zShellStatic, -1, SQLITE_STATIC);
477		-}
478		-
479	457
480	458	/*
481	459	** Compute a string length that is limited to what can be stored in
482	460	** lower 30 bits of a 32-bit signed integer.
483	461	*/
		@@ -575,10 +553,783 @@
575	553	if( zResult && *zResult ) shell_add_history(zResult);
576	554	#endif
577	555	}
578	556	return zResult;
579	557	}
	558	+/*
	559	+** A variable length string to which one can append text.
	560	+*/
	561	+typedef struct ShellText ShellText;
	562	+struct ShellText {
	563	+ char *z;
	564	+ int n;
	565	+ int nAlloc;
	566	+};
	567	+
	568	+/*
	569	+** Initialize and destroy a ShellText object
	570	+*/
	571	+static void initText(ShellText *p){
	572	+ memset(p, 0, sizeof(*p));
	573	+}
	574	+static void freeText(ShellText *p){
	575	+ free(p->z);
	576	+ initText(p);
	577	+}
	578	+
	579	+/* zIn is either a pointer to a NULL-terminated string in memory obtained
	580	+** from malloc(), or a NULL pointer. The string pointed to by zAppend is
	581	+** added to zIn, and the result returned in memory obtained from malloc().
	582	+** zIn, if it was not NULL, is freed.
	583	+**
	584	+** If the third argument, quote, is not '\0', then it is used as a
	585	+** quote character for zAppend.
	586	+*/
	587	+static void appendText(ShellText p, char const zAppend, char quote){
	588	+ int len;
	589	+ int i;
	590	+ int nAppend = strlen30(zAppend);
	591	+
	592	+ len = nAppend+p->n+1;
	593	+ if( quote ){
	594	+ len += 2;
	595	+ for(i=0; i<nAppend; i++){
	596	+ if( zAppend[i]==quote ) len++;
	597	+ }
	598	+ }
	599	+
	600	+ if( p->n+len>=p->nAlloc ){
	601	+ p->nAlloc = p->nAlloc*2 + len + 20;
	602	+ p->z = realloc(p->z, p->nAlloc);
	603	+ if( p->z==0 ){
	604	+ memset(p, 0, sizeof(*p));
	605	+ return;
	606	+ }
	607	+ }
	608	+
	609	+ if( quote ){
	610	+ char *zCsr = p->z+p->n;
	611	+ *zCsr++ = quote;
	612	+ for(i=0; i<nAppend; i++){
	613	+ *zCsr++ = zAppend[i];
	614	+ if( zAppend[i]==quote ) *zCsr++ = quote;
	615	+ }
	616	+ *zCsr++ = quote;
	617	+ p->n = (int)(zCsr - p->z);
	618	+ *zCsr = '\0';
	619	+ }else{
	620	+ memcpy(p->z+p->n, zAppend, nAppend);
	621	+ p->n += nAppend;
	622	+ p->z[p->n] = '\0';
	623	+ }
	624	+}
	625	+
	626	+/*
	627	+** Attempt to determine if identifier zName needs to be quoted, either
	628	+** because it contains non-alphanumeric characters, or because it is an
	629	+** SQLite keyword. Be conservative in this estimate: When in doubt assume
	630	+** that quoting is required.
	631	+**
	632	+** Return '"' if quoting is required. Return 0 if no quoting is required.
	633	+*/
	634	+static char quoteChar(const char *zName){
	635	+ /* All SQLite keywords, in alphabetical order */
	636	+ static const char *azKeywords[] = {
	637	+ "ABORT", "ACTION", "ADD", "AFTER", "ALL", "ALTER", "ANALYZE", "AND", "AS",
	638	+ "ASC", "ATTACH", "AUTOINCREMENT", "BEFORE", "BEGIN", "BETWEEN", "BY",
	639	+ "CASCADE", "CASE", "CAST", "CHECK", "COLLATE", "COLUMN", "COMMIT",
	640	+ "CONFLICT", "CONSTRAINT", "CREATE", "CROSS", "CURRENT_DATE",
	641	+ "CURRENT_TIME", "CURRENT_TIMESTAMP", "DATABASE", "DEFAULT", "DEFERRABLE",
	642	+ "DEFERRED", "DELETE", "DESC", "DETACH", "DISTINCT", "DROP", "EACH",
	643	+ "ELSE", "END", "ESCAPE", "EXCEPT", "EXCLUSIVE", "EXISTS", "EXPLAIN",
	644	+ "FAIL", "FOR", "FOREIGN", "FROM", "FULL", "GLOB", "GROUP", "HAVING", "IF",
	645	+ "IGNORE", "IMMEDIATE", "IN", "INDEX", "INDEXED", "INITIALLY", "INNER",
	646	+ "INSERT", "INSTEAD", "INTERSECT", "INTO", "IS", "ISNULL", "JOIN", "KEY",
	647	+ "LEFT", "LIKE", "LIMIT", "MATCH", "NATURAL", "NO", "NOT", "NOTNULL",
	648	+ "NULL", "OF", "OFFSET", "ON", "OR", "ORDER", "OUTER", "PLAN", "PRAGMA",
	649	+ "PRIMARY", "QUERY", "RAISE", "RECURSIVE", "REFERENCES", "REGEXP",
	650	+ "REINDEX", "RELEASE", "RENAME", "REPLACE", "RESTRICT", "RIGHT",
	651	+ "ROLLBACK", "ROW", "SAVEPOINT", "SELECT", "SET", "TABLE", "TEMP",
	652	+ "TEMPORARY", "THEN", "TO", "TRANSACTION", "TRIGGER", "UNION", "UNIQUE",
	653	+ "UPDATE", "USING", "VACUUM", "VALUES", "VIEW", "VIRTUAL", "WHEN", "WHERE",
	654	+ "WITH", "WITHOUT",
	655	+ };
	656	+ int i, lwr, upr, mid, c;
	657	+ if( !isalpha((unsigned char)zName[0]) && zName[0]!='_' ) return '"';
	658	+ for(i=0; zName[i]; i++){
	659	+ if( !isalnum((unsigned char)zName[i]) && zName[i]!='_' ) return '"';
	660	+ }
	661	+ lwr = 0;
	662	+ upr = sizeof(azKeywords)/sizeof(azKeywords[0]) - 1;
	663	+ while( lwr<=upr ){
	664	+ mid = (lwr+upr)/2;
	665	+ c = sqlite3_stricmp(azKeywords[mid], zName);
	666	+ if( c==0 ) return '"';
	667	+ if( c<0 ){
	668	+ lwr = mid+1;
	669	+ }else{
	670	+ upr = mid-1;
	671	+ }
	672	+ }
	673	+ return 0;
	674	+}
	675	+
	676	+/******************************************************************************
	677	+** SHA3 hash implementation copied from ../ext/misc/shathree.c
	678	+*/
	679	+typedef sqlite3_uint64 u64;
	680	+/*
	681	+** Macros to determine whether the machine is big or little endian,
	682	+** and whether or not that determination is run-time or compile-time.
	683	+**
	684	+** For best performance, an attempt is made to guess at the byte-order
	685	+** using C-preprocessor macros. If that is unsuccessful, or if
	686	+** -DSHA3_BYTEORDER=0 is set, then byte-order is determined
	687	+** at run-time.
	688	+*/
	689	+#ifndef SHA3_BYTEORDER
	690	+# if defined(i386) \|\| defined(__i386__) \|\| defined(_M_IX86) \|\| \
	691	+ defined(__x86_64) \|\| defined(__x86_64__) \|\| defined(_M_X64) \|\| \
	692	+ defined(_M_AMD64) \|\| defined(_M_ARM) \|\| defined(__x86) \|\| \
	693	+ defined(__arm__)
	694	+# define SHA3_BYTEORDER 1234
	695	+# elif defined(sparc) \|\| defined(__ppc__)
	696	+# define SHA3_BYTEORDER 4321
	697	+# else
	698	+# define SHA3_BYTEORDER 0
	699	+# endif
	700	+#endif
	701	+
	702	+
	703	+/*
	704	+** State structure for a SHA3 hash in progress
	705	+*/
	706	+typedef struct SHA3Context SHA3Context;
	707	+struct SHA3Context {
	708	+ union {
	709	+ u64 s[25]; /* Keccak state. 5x5 lines of 64 bits each */
	710	+ unsigned char x[1600]; /* ... or 1600 bytes */
	711	+ } u;
	712	+ unsigned nRate; /* Bytes of input accepted per Keccak iteration */
	713	+ unsigned nLoaded; /* Input bytes loaded into u.x[] so far this cycle */
	714	+ unsigned ixMask; /* Insert next input into u.x[nLoaded^ixMask]. */
	715	+};
	716	+
	717	+/*
	718	+** A single step of the Keccak mixing function for a 1600-bit state
	719	+*/
	720	+static void KeccakF1600Step(SHA3Context *p){
	721	+ int i;
	722	+ u64 B0, B1, B2, B3, B4;
	723	+ u64 C0, C1, C2, C3, C4;
	724	+ u64 D0, D1, D2, D3, D4;
	725	+ static const u64 RC[] = {
	726	+ 0x0000000000000001ULL, 0x0000000000008082ULL,
	727	+ 0x800000000000808aULL, 0x8000000080008000ULL,
	728	+ 0x000000000000808bULL, 0x0000000080000001ULL,
	729	+ 0x8000000080008081ULL, 0x8000000000008009ULL,
	730	+ 0x000000000000008aULL, 0x0000000000000088ULL,
	731	+ 0x0000000080008009ULL, 0x000000008000000aULL,
	732	+ 0x000000008000808bULL, 0x800000000000008bULL,
	733	+ 0x8000000000008089ULL, 0x8000000000008003ULL,
	734	+ 0x8000000000008002ULL, 0x8000000000000080ULL,
	735	+ 0x000000000000800aULL, 0x800000008000000aULL,
	736	+ 0x8000000080008081ULL, 0x8000000000008080ULL,
	737	+ 0x0000000080000001ULL, 0x8000000080008008ULL
	738	+ };
	739	+# define A00 (p->u.s[0])
	740	+# define A01 (p->u.s[1])
	741	+# define A02 (p->u.s[2])
	742	+# define A03 (p->u.s[3])
	743	+# define A04 (p->u.s[4])
	744	+# define A10 (p->u.s[5])
	745	+# define A11 (p->u.s[6])
	746	+# define A12 (p->u.s[7])
	747	+# define A13 (p->u.s[8])
	748	+# define A14 (p->u.s[9])
	749	+# define A20 (p->u.s[10])
	750	+# define A21 (p->u.s[11])
	751	+# define A22 (p->u.s[12])
	752	+# define A23 (p->u.s[13])
	753	+# define A24 (p->u.s[14])
	754	+# define A30 (p->u.s[15])
	755	+# define A31 (p->u.s[16])
	756	+# define A32 (p->u.s[17])
	757	+# define A33 (p->u.s[18])
	758	+# define A34 (p->u.s[19])
	759	+# define A40 (p->u.s[20])
	760	+# define A41 (p->u.s[21])
	761	+# define A42 (p->u.s[22])
	762	+# define A43 (p->u.s[23])
	763	+# define A44 (p->u.s[24])
	764	+# define ROL64(a,x) ((a<<x)\|(a>>(64-x)))
	765	+
	766	+ for(i=0; i<24; i+=4){
	767	+ C0 = A00^A10^A20^A30^A40;
	768	+ C1 = A01^A11^A21^A31^A41;
	769	+ C2 = A02^A12^A22^A32^A42;
	770	+ C3 = A03^A13^A23^A33^A43;
	771	+ C4 = A04^A14^A24^A34^A44;
	772	+ D0 = C4^ROL64(C1, 1);
	773	+ D1 = C0^ROL64(C2, 1);
	774	+ D2 = C1^ROL64(C3, 1);
	775	+ D3 = C2^ROL64(C4, 1);
	776	+ D4 = C3^ROL64(C0, 1);
	777	+
	778	+ B0 = (A00^D0);
	779	+ B1 = ROL64((A11^D1), 44);
	780	+ B2 = ROL64((A22^D2), 43);
	781	+ B3 = ROL64((A33^D3), 21);
	782	+ B4 = ROL64((A44^D4), 14);
	783	+ A00 = B0 ^((~B1)& B2 );
	784	+ A00 ^= RC[i];
	785	+ A11 = B1 ^((~B2)& B3 );
	786	+ A22 = B2 ^((~B3)& B4 );
	787	+ A33 = B3 ^((~B4)& B0 );
	788	+ A44 = B4 ^((~B0)& B1 );
	789	+
	790	+ B2 = ROL64((A20^D0), 3);
	791	+ B3 = ROL64((A31^D1), 45);
	792	+ B4 = ROL64((A42^D2), 61);
	793	+ B0 = ROL64((A03^D3), 28);
	794	+ B1 = ROL64((A14^D4), 20);
	795	+ A20 = B0 ^((~B1)& B2 );
	796	+ A31 = B1 ^((~B2)& B3 );
	797	+ A42 = B2 ^((~B3)& B4 );
	798	+ A03 = B3 ^((~B4)& B0 );
	799	+ A14 = B4 ^((~B0)& B1 );
	800	+
	801	+ B4 = ROL64((A40^D0), 18);
	802	+ B0 = ROL64((A01^D1), 1);
	803	+ B1 = ROL64((A12^D2), 6);
	804	+ B2 = ROL64((A23^D3), 25);
	805	+ B3 = ROL64((A34^D4), 8);
	806	+ A40 = B0 ^((~B1)& B2 );
	807	+ A01 = B1 ^((~B2)& B3 );
	808	+ A12 = B2 ^((~B3)& B4 );
	809	+ A23 = B3 ^((~B4)& B0 );
	810	+ A34 = B4 ^((~B0)& B1 );
	811	+
	812	+ B1 = ROL64((A10^D0), 36);
	813	+ B2 = ROL64((A21^D1), 10);
	814	+ B3 = ROL64((A32^D2), 15);
	815	+ B4 = ROL64((A43^D3), 56);
	816	+ B0 = ROL64((A04^D4), 27);
	817	+ A10 = B0 ^((~B1)& B2 );
	818	+ A21 = B1 ^((~B2)& B3 );
	819	+ A32 = B2 ^((~B3)& B4 );
	820	+ A43 = B3 ^((~B4)& B0 );
	821	+ A04 = B4 ^((~B0)& B1 );
	822	+
	823	+ B3 = ROL64((A30^D0), 41);
	824	+ B4 = ROL64((A41^D1), 2);
	825	+ B0 = ROL64((A02^D2), 62);
	826	+ B1 = ROL64((A13^D3), 55);
	827	+ B2 = ROL64((A24^D4), 39);
	828	+ A30 = B0 ^((~B1)& B2 );
	829	+ A41 = B1 ^((~B2)& B3 );
	830	+ A02 = B2 ^((~B3)& B4 );
	831	+ A13 = B3 ^((~B4)& B0 );
	832	+ A24 = B4 ^((~B0)& B1 );
	833	+
	834	+ C0 = A00^A20^A40^A10^A30;
	835	+ C1 = A11^A31^A01^A21^A41;
	836	+ C2 = A22^A42^A12^A32^A02;
	837	+ C3 = A33^A03^A23^A43^A13;
	838	+ C4 = A44^A14^A34^A04^A24;
	839	+ D0 = C4^ROL64(C1, 1);
	840	+ D1 = C0^ROL64(C2, 1);
	841	+ D2 = C1^ROL64(C3, 1);
	842	+ D3 = C2^ROL64(C4, 1);
	843	+ D4 = C3^ROL64(C0, 1);
	844	+
	845	+ B0 = (A00^D0);
	846	+ B1 = ROL64((A31^D1), 44);
	847	+ B2 = ROL64((A12^D2), 43);
	848	+ B3 = ROL64((A43^D3), 21);
	849	+ B4 = ROL64((A24^D4), 14);
	850	+ A00 = B0 ^((~B1)& B2 );
	851	+ A00 ^= RC[i+1];
	852	+ A31 = B1 ^((~B2)& B3 );
	853	+ A12 = B2 ^((~B3)& B4 );
	854	+ A43 = B3 ^((~B4)& B0 );
	855	+ A24 = B4 ^((~B0)& B1 );
	856	+
	857	+ B2 = ROL64((A40^D0), 3);
	858	+ B3 = ROL64((A21^D1), 45);
	859	+ B4 = ROL64((A02^D2), 61);
	860	+ B0 = ROL64((A33^D3), 28);
	861	+ B1 = ROL64((A14^D4), 20);
	862	+ A40 = B0 ^((~B1)& B2 );
	863	+ A21 = B1 ^((~B2)& B3 );
	864	+ A02 = B2 ^((~B3)& B4 );
	865	+ A33 = B3 ^((~B4)& B0 );
	866	+ A14 = B4 ^((~B0)& B1 );
	867	+
	868	+ B4 = ROL64((A30^D0), 18);
	869	+ B0 = ROL64((A11^D1), 1);
	870	+ B1 = ROL64((A42^D2), 6);
	871	+ B2 = ROL64((A23^D3), 25);
	872	+ B3 = ROL64((A04^D4), 8);
	873	+ A30 = B0 ^((~B1)& B2 );
	874	+ A11 = B1 ^((~B2)& B3 );
	875	+ A42 = B2 ^((~B3)& B4 );
	876	+ A23 = B3 ^((~B4)& B0 );
	877	+ A04 = B4 ^((~B0)& B1 );
	878	+
	879	+ B1 = ROL64((A20^D0), 36);
	880	+ B2 = ROL64((A01^D1), 10);
	881	+ B3 = ROL64((A32^D2), 15);
	882	+ B4 = ROL64((A13^D3), 56);
	883	+ B0 = ROL64((A44^D4), 27);
	884	+ A20 = B0 ^((~B1)& B2 );
	885	+ A01 = B1 ^((~B2)& B3 );
	886	+ A32 = B2 ^((~B3)& B4 );
	887	+ A13 = B3 ^((~B4)& B0 );
	888	+ A44 = B4 ^((~B0)& B1 );
	889	+
	890	+ B3 = ROL64((A10^D0), 41);
	891	+ B4 = ROL64((A41^D1), 2);
	892	+ B0 = ROL64((A22^D2), 62);
	893	+ B1 = ROL64((A03^D3), 55);
	894	+ B2 = ROL64((A34^D4), 39);
	895	+ A10 = B0 ^((~B1)& B2 );
	896	+ A41 = B1 ^((~B2)& B3 );
	897	+ A22 = B2 ^((~B3)& B4 );
	898	+ A03 = B3 ^((~B4)& B0 );
	899	+ A34 = B4 ^((~B0)& B1 );
	900	+
	901	+ C0 = A00^A40^A30^A20^A10;
	902	+ C1 = A31^A21^A11^A01^A41;
	903	+ C2 = A12^A02^A42^A32^A22;
	904	+ C3 = A43^A33^A23^A13^A03;
	905	+ C4 = A24^A14^A04^A44^A34;
	906	+ D0 = C4^ROL64(C1, 1);
	907	+ D1 = C0^ROL64(C2, 1);
	908	+ D2 = C1^ROL64(C3, 1);
	909	+ D3 = C2^ROL64(C4, 1);
	910	+ D4 = C3^ROL64(C0, 1);
	911	+
	912	+ B0 = (A00^D0);
	913	+ B1 = ROL64((A21^D1), 44);
	914	+ B2 = ROL64((A42^D2), 43);
	915	+ B3 = ROL64((A13^D3), 21);
	916	+ B4 = ROL64((A34^D4), 14);
	917	+ A00 = B0 ^((~B1)& B2 );
	918	+ A00 ^= RC[i+2];
	919	+ A21 = B1 ^((~B2)& B3 );
	920	+ A42 = B2 ^((~B3)& B4 );
	921	+ A13 = B3 ^((~B4)& B0 );
	922	+ A34 = B4 ^((~B0)& B1 );
	923	+
	924	+ B2 = ROL64((A30^D0), 3);
	925	+ B3 = ROL64((A01^D1), 45);
	926	+ B4 = ROL64((A22^D2), 61);
	927	+ B0 = ROL64((A43^D3), 28);
	928	+ B1 = ROL64((A14^D4), 20);
	929	+ A30 = B0 ^((~B1)& B2 );
	930	+ A01 = B1 ^((~B2)& B3 );
	931	+ A22 = B2 ^((~B3)& B4 );
	932	+ A43 = B3 ^((~B4)& B0 );
	933	+ A14 = B4 ^((~B0)& B1 );
	934	+
	935	+ B4 = ROL64((A10^D0), 18);
	936	+ B0 = ROL64((A31^D1), 1);
	937	+ B1 = ROL64((A02^D2), 6);
	938	+ B2 = ROL64((A23^D3), 25);
	939	+ B3 = ROL64((A44^D4), 8);
	940	+ A10 = B0 ^((~B1)& B2 );
	941	+ A31 = B1 ^((~B2)& B3 );
	942	+ A02 = B2 ^((~B3)& B4 );
	943	+ A23 = B3 ^((~B4)& B0 );
	944	+ A44 = B4 ^((~B0)& B1 );
	945	+
	946	+ B1 = ROL64((A40^D0), 36);
	947	+ B2 = ROL64((A11^D1), 10);
	948	+ B3 = ROL64((A32^D2), 15);
	949	+ B4 = ROL64((A03^D3), 56);
	950	+ B0 = ROL64((A24^D4), 27);
	951	+ A40 = B0 ^((~B1)& B2 );
	952	+ A11 = B1 ^((~B2)& B3 );
	953	+ A32 = B2 ^((~B3)& B4 );
	954	+ A03 = B3 ^((~B4)& B0 );
	955	+ A24 = B4 ^((~B0)& B1 );
	956	+
	957	+ B3 = ROL64((A20^D0), 41);
	958	+ B4 = ROL64((A41^D1), 2);
	959	+ B0 = ROL64((A12^D2), 62);
	960	+ B1 = ROL64((A33^D3), 55);
	961	+ B2 = ROL64((A04^D4), 39);
	962	+ A20 = B0 ^((~B1)& B2 );
	963	+ A41 = B1 ^((~B2)& B3 );
	964	+ A12 = B2 ^((~B3)& B4 );
	965	+ A33 = B3 ^((~B4)& B0 );
	966	+ A04 = B4 ^((~B0)& B1 );
	967	+
	968	+ C0 = A00^A30^A10^A40^A20;
	969	+ C1 = A21^A01^A31^A11^A41;
	970	+ C2 = A42^A22^A02^A32^A12;
	971	+ C3 = A13^A43^A23^A03^A33;
	972	+ C4 = A34^A14^A44^A24^A04;
	973	+ D0 = C4^ROL64(C1, 1);
	974	+ D1 = C0^ROL64(C2, 1);
	975	+ D2 = C1^ROL64(C3, 1);
	976	+ D3 = C2^ROL64(C4, 1);
	977	+ D4 = C3^ROL64(C0, 1);
	978	+
	979	+ B0 = (A00^D0);
	980	+ B1 = ROL64((A01^D1), 44);
	981	+ B2 = ROL64((A02^D2), 43);
	982	+ B3 = ROL64((A03^D3), 21);
	983	+ B4 = ROL64((A04^D4), 14);
	984	+ A00 = B0 ^((~B1)& B2 );
	985	+ A00 ^= RC[i+3];
	986	+ A01 = B1 ^((~B2)& B3 );
	987	+ A02 = B2 ^((~B3)& B4 );
	988	+ A03 = B3 ^((~B4)& B0 );
	989	+ A04 = B4 ^((~B0)& B1 );
	990	+
	991	+ B2 = ROL64((A10^D0), 3);
	992	+ B3 = ROL64((A11^D1), 45);
	993	+ B4 = ROL64((A12^D2), 61);
	994	+ B0 = ROL64((A13^D3), 28);
	995	+ B1 = ROL64((A14^D4), 20);
	996	+ A10 = B0 ^((~B1)& B2 );
	997	+ A11 = B1 ^((~B2)& B3 );
	998	+ A12 = B2 ^((~B3)& B4 );
	999	+ A13 = B3 ^((~B4)& B0 );
	1000	+ A14 = B4 ^((~B0)& B1 );
	1001	+
	1002	+ B4 = ROL64((A20^D0), 18);
	1003	+ B0 = ROL64((A21^D1), 1);
	1004	+ B1 = ROL64((A22^D2), 6);
	1005	+ B2 = ROL64((A23^D3), 25);
	1006	+ B3 = ROL64((A24^D4), 8);
	1007	+ A20 = B0 ^((~B1)& B2 );
	1008	+ A21 = B1 ^((~B2)& B3 );
	1009	+ A22 = B2 ^((~B3)& B4 );
	1010	+ A23 = B3 ^((~B4)& B0 );
	1011	+ A24 = B4 ^((~B0)& B1 );
	1012	+
	1013	+ B1 = ROL64((A30^D0), 36);
	1014	+ B2 = ROL64((A31^D1), 10);
	1015	+ B3 = ROL64((A32^D2), 15);
	1016	+ B4 = ROL64((A33^D3), 56);
	1017	+ B0 = ROL64((A34^D4), 27);
	1018	+ A30 = B0 ^((~B1)& B2 );
	1019	+ A31 = B1 ^((~B2)& B3 );
	1020	+ A32 = B2 ^((~B3)& B4 );
	1021	+ A33 = B3 ^((~B4)& B0 );
	1022	+ A34 = B4 ^((~B0)& B1 );
	1023	+
	1024	+ B3 = ROL64((A40^D0), 41);
	1025	+ B4 = ROL64((A41^D1), 2);
	1026	+ B0 = ROL64((A42^D2), 62);
	1027	+ B1 = ROL64((A43^D3), 55);
	1028	+ B2 = ROL64((A44^D4), 39);
	1029	+ A40 = B0 ^((~B1)& B2 );
	1030	+ A41 = B1 ^((~B2)& B3 );
	1031	+ A42 = B2 ^((~B3)& B4 );
	1032	+ A43 = B3 ^((~B4)& B0 );
	1033	+ A44 = B4 ^((~B0)& B1 );
	1034	+ }
	1035	+}
	1036	+
	1037	+/*
	1038	+** Initialize a new hash. iSize determines the size of the hash
	1039	+** in bits and should be one of 224, 256, 384, or 512. Or iSize
	1040	+** can be zero to use the default hash size of 256 bits.
	1041	+*/
	1042	+static void SHA3Init(SHA3Context *p, int iSize){
	1043	+ memset(p, 0, sizeof(*p));
	1044	+ if( iSize>=128 && iSize<=512 ){
	1045	+ p->nRate = (1600 - ((iSize + 31)&~31)*2)/8;
	1046	+ }else{
	1047	+ p->nRate = (1600 - 2*256)/8;
	1048	+ }
	1049	+#if SHA3_BYTEORDER==1234
	1050	+ /* Known to be little-endian at compile-time. No-op */
	1051	+#elif SHA3_BYTEORDER==4321
	1052	+ p->ixMask = 7; /* Big-endian */
	1053	+#else
	1054	+ {
	1055	+ static unsigned int one = 1;
	1056	+ if( 1==(unsigned char)&one ){
	1057	+ /* Little endian. No byte swapping. */
	1058	+ p->ixMask = 0;
	1059	+ }else{
	1060	+ /* Big endian. Byte swap. */
	1061	+ p->ixMask = 7;
	1062	+ }
	1063	+ }
	1064	+#endif
	1065	+}
	1066	+
	1067	+/*
	1068	+** Make consecutive calls to the SHA3Update function to add new content
	1069	+** to the hash
	1070	+*/
	1071	+static void SHA3Update(
	1072	+ SHA3Context *p,
	1073	+ const unsigned char *aData,
	1074	+ unsigned int nData
	1075	+){
	1076	+ unsigned int i = 0;
	1077	+#if SHA3_BYTEORDER==1234
	1078	+ if( (p->nLoaded % 8)==0 && ((aData - (const unsigned char*)0)&7)==0 ){
	1079	+ for(; i+7<nData; i+=8){
	1080	+ p->u.s[p->nLoaded/8] ^= (u64)&aData[i];
	1081	+ p->nLoaded += 8;
	1082	+ if( p->nLoaded>=p->nRate ){
	1083	+ KeccakF1600Step(p);
	1084	+ p->nLoaded = 0;
	1085	+ }
	1086	+ }
	1087	+ }
	1088	+#endif
	1089	+ for(; i<nData; i++){
	1090	+#if SHA3_BYTEORDER==1234
	1091	+ p->u.x[p->nLoaded] ^= aData[i];
	1092	+#elif SHA3_BYTEORDER==4321
	1093	+ p->u.x[p->nLoaded^0x07] ^= aData[i];
	1094	+#else
	1095	+ p->u.x[p->nLoaded^p->ixMask] ^= aData[i];
	1096	+#endif
	1097	+ p->nLoaded++;
	1098	+ if( p->nLoaded==p->nRate ){
	1099	+ KeccakF1600Step(p);
	1100	+ p->nLoaded = 0;
	1101	+ }
	1102	+ }
	1103	+}
	1104	+
	1105	+/*
	1106	+** After all content has been added, invoke SHA3Final() to compute
	1107	+** the final hash. The function returns a pointer to the binary
	1108	+** hash value.
	1109	+*/
	1110	+static unsigned char SHA3Final(SHA3Context p){
	1111	+ unsigned int i;
	1112	+ if( p->nLoaded==p->nRate-1 ){
	1113	+ const unsigned char c1 = 0x86;
	1114	+ SHA3Update(p, &c1, 1);
	1115	+ }else{
	1116	+ const unsigned char c2 = 0x06;
	1117	+ const unsigned char c3 = 0x80;
	1118	+ SHA3Update(p, &c2, 1);
	1119	+ p->nLoaded = p->nRate - 1;
	1120	+ SHA3Update(p, &c3, 1);
	1121	+ }
	1122	+ for(i=0; i<p->nRate; i++){
	1123	+ p->u.x[i+p->nRate] = p->u.x[i^p->ixMask];
	1124	+ }
	1125	+ return &p->u.x[p->nRate];
	1126	+}
	1127	+
	1128	+/*
	1129	+** Implementation of the sha3(X,SIZE) function.
	1130	+**
	1131	+** Return a BLOB which is the SIZE-bit SHA3 hash of X. The default
	1132	+** size is 256. If X is a BLOB, it is hashed as is.
	1133	+** For all other non-NULL types of input, X is converted into a UTF-8 string
	1134	+** and the string is hashed without the trailing 0x00 terminator. The hash
	1135	+** of a NULL value is NULL.
	1136	+*/
	1137	+static void sha3Func(
	1138	+ sqlite3_context *context,
	1139	+ int argc,
	1140	+ sqlite3_value **argv
	1141	+){
	1142	+ SHA3Context cx;
	1143	+ int eType = sqlite3_value_type(argv[0]);
	1144	+ int nByte = sqlite3_value_bytes(argv[0]);
	1145	+ int iSize;
	1146	+ if( argc==1 ){
	1147	+ iSize = 256;
	1148	+ }else{
	1149	+ iSize = sqlite3_value_int(argv[1]);
	1150	+ if( iSize!=224 && iSize!=256 && iSize!=384 && iSize!=512 ){
	1151	+ sqlite3_result_error(context, "SHA3 size should be one of: 224 256 "
	1152	+ "384 512", -1);
	1153	+ return;
	1154	+ }
	1155	+ }
	1156	+ if( eType==SQLITE_NULL ) return;
	1157	+ SHA3Init(&cx, iSize);
	1158	+ if( eType==SQLITE_BLOB ){
	1159	+ SHA3Update(&cx, sqlite3_value_blob(argv[0]), nByte);
	1160	+ }else{
	1161	+ SHA3Update(&cx, sqlite3_value_text(argv[0]), nByte);
	1162	+ }
	1163	+ sqlite3_result_blob(context, SHA3Final(&cx), iSize/8, SQLITE_TRANSIENT);
	1164	+}
	1165	+
	1166	+/* Compute a string using sqlite3_vsnprintf() with a maximum length
	1167	+** of 50 bytes and add it to the hash.
	1168	+*/
	1169	+static void hash_step_vformat(
	1170	+ SHA3Context p, / Add content to this context */
	1171	+ const char *zFormat,
	1172	+ ...
	1173	+){
	1174	+ va_list ap;
	1175	+ int n;
	1176	+ char zBuf[50];
	1177	+ va_start(ap, zFormat);
	1178	+ sqlite3_vsnprintf(sizeof(zBuf),zBuf,zFormat,ap);
	1179	+ va_end(ap);
	1180	+ n = (int)strlen(zBuf);
	1181	+ SHA3Update(p, (unsigned char*)zBuf, n);
	1182	+}
	1183	+
	1184	+/*
	1185	+** Implementation of the sha3_query(SQL,SIZE) function.
	1186	+**
	1187	+** This function compiles and runs the SQL statement(s) given in the
	1188	+** argument. The results are hashed using a SIZE-bit SHA3. The default
	1189	+** size is 256.
	1190	+**
	1191	+** The format of the byte stream that is hashed is summarized as follows:
	1192	+**
	1193	+** S<n>:<sql>
	1194	+** R
	1195	+** N
	1196	+** I<int>
	1197	+** F<ieee-float>
	1198	+** B<size>:<bytes>
	1199	+** T<size>:<text>
	1200	+**
	1201	+** <sql> is the original SQL text for each statement run and <n> is
	1202	+** the size of that text. The SQL text is UTF-8. A single R character
	1203	+** occurs before the start of each row. N means a NULL value.
	1204	+** I mean an 8-byte little-endian integer <int>. F is a floating point
	1205	+** number with an 8-byte little-endian IEEE floating point value <ieee-float>.
	1206	+** B means blobs of <size> bytes. T means text rendered as <size>
	1207	+** bytes of UTF-8. The <n> and <size> values are expressed as an ASCII
	1208	+** text integers.
	1209	+**
	1210	+** For each SQL statement in the X input, there is one S segment. Each
	1211	+** S segment is followed by zero or more R segments, one for each row in the
	1212	+** result set. After each R, there are one or more N, I, F, B, or T segments,
	1213	+** one for each column in the result set. Segments are concatentated directly
	1214	+** with no delimiters of any kind.
	1215	+*/
	1216	+static void sha3QueryFunc(
	1217	+ sqlite3_context *context,
	1218	+ int argc,
	1219	+ sqlite3_value **argv
	1220	+){
	1221	+ sqlite3 *db = sqlite3_context_db_handle(context);
	1222	+ const char zSql = (const char)sqlite3_value_text(argv[0]);
	1223	+ sqlite3_stmt *pStmt = 0;
	1224	+ int nCol; /* Number of columns in the result set */
	1225	+ int i; /* Loop counter */
	1226	+ int rc;
	1227	+ int n;
	1228	+ const char *z;
	1229	+ SHA3Context cx;
	1230	+ int iSize;
	1231	+
	1232	+ if( argc==1 ){
	1233	+ iSize = 256;
	1234	+ }else{
	1235	+ iSize = sqlite3_value_int(argv[1]);
	1236	+ if( iSize!=224 && iSize!=256 && iSize!=384 && iSize!=512 ){
	1237	+ sqlite3_result_error(context, "SHA3 size should be one of: 224 256 "
	1238	+ "384 512", -1);
	1239	+ return;
	1240	+ }
	1241	+ }
	1242	+ if( zSql==0 ) return;
	1243	+ SHA3Init(&cx, iSize);
	1244	+ while( zSql[0] ){
	1245	+ rc = sqlite3_prepare_v2(db, zSql, -1, &pStmt, &zSql);
	1246	+ if( rc ){
	1247	+ char *zMsg = sqlite3_mprintf("error SQL statement [%s]: %s",
	1248	+ zSql, sqlite3_errmsg(db));
	1249	+ sqlite3_finalize(pStmt);
	1250	+ sqlite3_result_error(context, zMsg, -1);
	1251	+ sqlite3_free(zMsg);
	1252	+ return;
	1253	+ }
	1254	+ if( !sqlite3_stmt_readonly(pStmt) ){
	1255	+ char *zMsg = sqlite3_mprintf("non-query: [%s]", sqlite3_sql(pStmt));
	1256	+ sqlite3_finalize(pStmt);
	1257	+ sqlite3_result_error(context, zMsg, -1);
	1258	+ sqlite3_free(zMsg);
	1259	+ return;
	1260	+ }
	1261	+ nCol = sqlite3_column_count(pStmt);
	1262	+ z = sqlite3_sql(pStmt);
	1263	+ if( z==0 ){
	1264	+ sqlite3_finalize(pStmt);
	1265	+ continue;
	1266	+ }
	1267	+ n = (int)strlen(z);
	1268	+ hash_step_vformat(&cx,"S%d:",n);
	1269	+ SHA3Update(&cx,(unsigned char*)z,n);
	1270	+
	1271	+ /* Compute a hash over the result of the query */
	1272	+ while( SQLITE_ROW==sqlite3_step(pStmt) ){
	1273	+ SHA3Update(&cx,(const unsigned char*)"R",1);
	1274	+ for(i=0; i<nCol; i++){
	1275	+ switch( sqlite3_column_type(pStmt,i) ){
	1276	+ case SQLITE_NULL: {
	1277	+ SHA3Update(&cx, (const unsigned char*)"N",1);
	1278	+ break;
	1279	+ }
	1280	+ case SQLITE_INTEGER: {
	1281	+ sqlite3_uint64 u;
	1282	+ int j;
	1283	+ unsigned char x[9];
	1284	+ sqlite3_int64 v = sqlite3_column_int64(pStmt,i);
	1285	+ memcpy(&u, &v, 8);
	1286	+ for(j=8; j>=1; j--){
	1287	+ x[j] = u & 0xff;
	1288	+ u >>= 8;
	1289	+ }
	1290	+ x[0] = 'I';
	1291	+ SHA3Update(&cx, x, 9);
	1292	+ break;
	1293	+ }
	1294	+ case SQLITE_FLOAT: {
	1295	+ sqlite3_uint64 u;
	1296	+ int j;
	1297	+ unsigned char x[9];
	1298	+ double r = sqlite3_column_double(pStmt,i);
	1299	+ memcpy(&u, &r, 8);
	1300	+ for(j=8; j>=1; j--){
	1301	+ x[j] = u & 0xff;
	1302	+ u >>= 8;
	1303	+ }
	1304	+ x[0] = 'F';
	1305	+ SHA3Update(&cx,x,9);
	1306	+ break;
	1307	+ }
	1308	+ case SQLITE_TEXT: {
	1309	+ int n2 = sqlite3_column_bytes(pStmt, i);
	1310	+ const unsigned char *z2 = sqlite3_column_text(pStmt, i);
	1311	+ hash_step_vformat(&cx,"T%d:",n2);
	1312	+ SHA3Update(&cx, z2, n2);
	1313	+ break;
	1314	+ }
	1315	+ case SQLITE_BLOB: {
	1316	+ int n2 = sqlite3_column_bytes(pStmt, i);
	1317	+ const unsigned char *z2 = sqlite3_column_blob(pStmt, i);
	1318	+ hash_step_vformat(&cx,"B%d:",n2);
	1319	+ SHA3Update(&cx, z2, n2);
	1320	+ break;
	1321	+ }
	1322	+ }
	1323	+ }
	1324	+ }
	1325	+ sqlite3_finalize(pStmt);
	1326	+ }
	1327	+ sqlite3_result_blob(context, SHA3Final(&cx), iSize/8, SQLITE_TRANSIENT);
	1328	+}
	1329	+/* End of SHA3 hashing logic copy/pasted from ../ext/misc/shathree.c
	1330	+********************************************************************************/
580	1331
581	1332	#if defined(SQLITE_ENABLE_SESSION)
582	1333	/*
583	1334	** State information for a single open session
584	1335	*/
		@@ -608,17 +1359,14 @@
608	1359	** instance of the following structure.
609	1360	*/
610	1361	typedef struct ShellState ShellState;
611	1362	struct ShellState {
612	1363	sqlite3 db; / The database */
613		- int echoOn; /* True to echo input commands */
614	1364	int autoExplain; /* Automatically turn on .explain mode */
615	1365	int autoEQP; /* Run EXPLAIN QUERY PLAN prior to seach SQL stmt */
616	1366	int statsOn; /* True to display memory stats before each finalize */
617	1367	int scanstatsOn; /* True to display scan stats before each finalize */
618		- int countChanges; /* True to display change counts */
619		- int backslashOn; /* Resolve C-style \x escapes in SQL input text */
620	1368	int outCount; /* Revert to stdout when reaching zero */
621	1369	int cnt; /* Number of records displayed so far */
622	1370	FILE out; / Write results here */
623	1371	FILE traceOut; / Output for sqlite3_trace() */
624	1372	int nErr; /* Number of errors seen */
		@@ -653,13 +1401,24 @@
653	1401	};
654	1402
655	1403	/*
656	1404	** These are the allowed shellFlgs values
657	1405	*/
658		-#define SHFLG_Scratch 0x00001 /* The --scratch option is used */
659		-#define SHFLG_Pagecache 0x00002 /* The --pagecache option is used */
660		-#define SHFLG_Lookaside 0x00004 /* Lookaside memory is used */
	1406	+#define SHFLG_Scratch 0x00000001 /* The --scratch option is used */
	1407	+#define SHFLG_Pagecache 0x00000002 /* The --pagecache option is used */
	1408	+#define SHFLG_Lookaside 0x00000004 /* Lookaside memory is used */
	1409	+#define SHFLG_Backslash 0x00000008 /* The --backslash option is used */
	1410	+#define SHFLG_PreserveRowid 0x00000010 /* .dump preserves rowid values */
	1411	+#define SHFLG_CountChanges 0x00000020 /* .changes setting */
	1412	+#define SHFLG_Echo 0x00000040 /* .echo or --echo setting */
	1413	+
	1414	+/*
	1415	+** Macros for testing and setting shellFlgs
	1416	+*/
	1417	+#define ShellHasFlag(P,X) (((P)->shellFlgs & (X))!=0)
	1418	+#define ShellSetFlag(P,X) ((P)->shellFlgs\|=(X))
	1419	+#define ShellClearFlag(P,X) ((P)->shellFlgs&=(~(X)))
661	1420
662	1421	/*
663	1422	** These are the allowed modes.
664	1423	*/
665	1424	#define MODE_Line 0 /* One column per line. Blank line between records */
		@@ -1292,92 +2051,106 @@
1292	2051	*/
1293	2052	static int callback(void pArg, int nArg, char azArg, char *azCol){
1294	2053	/* since we don't have type info, call the shell_callback with a NULL value */
1295	2054	return shell_callback(pArg, nArg, azArg, azCol, NULL);
1296	2055	}
	2056	+
	2057	+/*
	2058	+** This is the callback routine from sqlite3_exec() that appends all
	2059	+** output onto the end of a ShellText object.
	2060	+*/
	2061	+static int captureOutputCallback(void pArg, int nArg, char azArg, char *az){
	2062	+ ShellText p = (ShellText)pArg;
	2063	+ int i;
	2064	+ if( p->n ) appendText(p, "\|", 0);
	2065	+ for(i=0; i<nArg; i++){
	2066	+ if( i ) appendText(p, ",", 0);
	2067	+ if( azArg[i] ) appendText(p, azArg[i], 0);
	2068	+ }
	2069	+ return 0;
	2070	+}
	2071	+
	2072	+/*
	2073	+** Generate an appropriate SELFTEST table in the main database.
	2074	+*/
	2075	+static void createSelftestTable(ShellState *p){
	2076	+ char *zErrMsg = 0;
	2077	+ sqlite3_exec(p->db,
	2078	+ "SAVEPOINT selftest_init;\n"
	2079	+ "CREATE TABLE IF NOT EXISTS selftest(\n"
	2080	+ " tno INTEGER PRIMARY KEY,\n" /* Test number */
	2081	+ " op TEXT,\n" /* Operator: memo run */
	2082	+ " cmd TEXT,\n" /* Command text */
	2083	+ " ans TEXT\n" /* Desired answer */
	2084	+ ");"
	2085	+ "CREATE TEMP TABLE [_shell$self](op,cmd,ans);\n"
	2086	+ "INSERT INTO [_shell$self](rowid,op,cmd)\n"
	2087	+ " VALUES(coalesce((SELECT (max(tno)+100)/10 FROM selftest),10),\n"
	2088	+ " 'memo','Tests generated by --init');\n"
	2089	+ "INSERT INTO [_shell$self]\n"
	2090	+ " SELECT 'run',\n"
	2091	+ " 'SELECT hex(sha3_query(''SELECT type,name,tbl_name,sql "
	2092	+ "FROM sqlite_master ORDER BY 2'',224))',\n"
	2093	+ " hex(sha3_query('SELECT type,name,tbl_name,sql "
	2094	+ "FROM sqlite_master ORDER BY 2',224));\n"
	2095	+ "INSERT INTO [_shell$self]\n"
	2096	+ " SELECT 'run',"
	2097	+ " 'SELECT hex(sha3_query(''SELECT * FROM \"' \|\|"
	2098	+ " printf('%w',name) \|\| '\" NOT INDEXED'',224))',\n"
	2099	+ " hex(sha3_query(printf('SELECT * FROM \"%w\" NOT INDEXED',name),224))\n"
	2100	+ " FROM (\n"
	2101	+ " SELECT name FROM sqlite_master\n"
	2102	+ " WHERE type='table'\n"
	2103	+ " AND name<>'selftest'\n"
	2104	+ " AND coalesce(rootpage,0)>0\n"
	2105	+ " )\n"
	2106	+ " ORDER BY name;\n"
	2107	+ "INSERT INTO [_shell$self]\n"
	2108	+ " VALUES('run','PRAGMA integrity_check','ok');\n"
	2109	+ "INSERT INTO selftest(tno,op,cmd,ans)"
	2110	+ " SELECT rowid*10,op,cmd,ans FROM [_shell$self];\n"
	2111	+ "DROP TABLE [_shell$self];"
	2112	+ ,0,0,&zErrMsg);
	2113	+ if( zErrMsg ){
	2114	+ utf8_printf(stderr, "SELFTEST initialization failure: %s\n", zErrMsg);
	2115	+ sqlite3_free(zErrMsg);
	2116	+ }
	2117	+ sqlite3_exec(p->db, "RELEASE selftest_init",0,0,0);
	2118	+}
	2119	+
1297	2120
1298	2121	/*
1299	2122	** Set the destination table field of the ShellState structure to
1300	2123	** the name of the table given. Escape any quote characters in the
1301	2124	** table name.
1302	2125	*/
1303	2126	static void set_table_name(ShellState p, const char zName){
1304	2127	int i, n;
1305		- int needQuote;
	2128	+ int cQuote;
1306	2129	char *z;
1307	2130
1308	2131	if( p->zDestTable ){
1309	2132	free(p->zDestTable);
1310	2133	p->zDestTable = 0;
1311	2134	}
1312	2135	if( zName==0 ) return;
1313		- needQuote = !isalpha((unsigned char)zName) && zName!='_';
1314		- for(i=n=0; zName[i]; i++, n++){
1315		- if( !isalnum((unsigned char)zName[i]) && zName[i]!='_' ){
1316		- needQuote = 1;
1317		- if( zName[i]=='\'' ) n++;
1318		- }
1319		- }
1320		- if( needQuote ) n += 2;
	2136	+ cQuote = quoteChar(zName);
	2137	+ n = strlen30(zName);
	2138	+ if( cQuote ) n += 2;
1321	2139	z = p->zDestTable = malloc( n+1 );
1322	2140	if( z==0 ){
1323	2141	raw_printf(stderr,"Error: out of memory\n");
1324	2142	exit(1);
1325	2143	}
1326	2144	n = 0;
1327		- if( needQuote ) z[n++] = '\'';
	2145	+ if( cQuote ) z[n++] = cQuote;
1328	2146	for(i=0; zName[i]; i++){
1329	2147	z[n++] = zName[i];
1330		- if( zName[i]=='\'' ) z[n++] = '\'';
1331		- }
1332		- if( needQuote ) z[n++] = '\'';
1333		- z[n] = 0;
1334		-}
1335		-
1336		-/* zIn is either a pointer to a NULL-terminated string in memory obtained
1337		-** from malloc(), or a NULL pointer. The string pointed to by zAppend is
1338		-** added to zIn, and the result returned in memory obtained from malloc().
1339		-** zIn, if it was not NULL, is freed.
1340		-**
1341		-** If the third argument, quote, is not '\0', then it is used as a
1342		-** quote character for zAppend.
1343		-*/
1344		-static char appendText(char zIn, char const *zAppend, char quote){
1345		- int len;
1346		- int i;
1347		- int nAppend = strlen30(zAppend);
1348		- int nIn = (zIn?strlen30(zIn):0);
1349		-
1350		- len = nAppend+nIn+1;
1351		- if( quote ){
1352		- len += 2;
1353		- for(i=0; i<nAppend; i++){
1354		- if( zAppend[i]==quote ) len++;
1355		- }
1356		- }
1357		-
1358		- zIn = (char *)realloc(zIn, len);
1359		- if( !zIn ){
1360		- return 0;
1361		- }
1362		-
1363		- if( quote ){
1364		- char *zCsr = &zIn[nIn];
1365		- *zCsr++ = quote;
1366		- for(i=0; i<nAppend; i++){
1367		- *zCsr++ = zAppend[i];
1368		- if( zAppend[i]==quote ) *zCsr++ = quote;
1369		- }
1370		- *zCsr++ = quote;
1371		- *zCsr++ = '\0';
1372		- assert( (zCsr-zIn)==len );
1373		- }else{
1374		- memcpy(&zIn[nIn], zAppend, nAppend);
1375		- zIn[len-1] = '\0';
1376		- }
1377		-
1378		- return zIn;
	2148	+ if( zName[i]==cQuote ) z[n++] = cQuote;
	2149	+ }
	2150	+ if( cQuote ) z[n++] = cQuote;
	2151	+ z[n] = 0;
1379	2152	}
1380	2153
1381	2154
1382	2155	/*
1383	2156	** Execute a query statement that will generate SQL output. Print
		@@ -1484,10 +2257,35 @@
1484	2257	}
1485	2258	fclose(in);
1486	2259	}
1487	2260	#endif
1488	2261
	2262	+/*
	2263	+** Display a single line of status using 64-bit values.
	2264	+*/
	2265	+static void displayStatLine(
	2266	+ ShellState p, / The shell context */
	2267	+ char zLabel, / Label for this one line */
	2268	+ char zFormat, / Format for the result */
	2269	+ int iStatusCtrl, /* Which status to display */
	2270	+ int bReset /* True to reset the stats */
	2271	+){
	2272	+ sqlite3_int64 iCur = -1;
	2273	+ sqlite3_int64 iHiwtr = -1;
	2274	+ int i, nPercent;
	2275	+ char zLine[200];
	2276	+ sqlite3_status64(iStatusCtrl, &iCur, &iHiwtr, bReset);
	2277	+ for(i=0, nPercent=0; zFormat[i]; i++){
	2278	+ if( zFormat[i]=='%' ) nPercent++;
	2279	+ }
	2280	+ if( nPercent>1 ){
	2281	+ sqlite3_snprintf(sizeof(zLine), zLine, zFormat, iCur, iHiwtr);
	2282	+ }else{
	2283	+ sqlite3_snprintf(sizeof(zLine), zLine, zFormat, iHiwtr);
	2284	+ }
	2285	+ raw_printf(p->out, "%-36s %s\n", zLabel, zLine);
	2286	+}
1489	2287
1490	2288	/*
1491	2289	** Display memory stats.
1492	2290	*/
1493	2291	static int display_stats(
		@@ -1497,61 +2295,35 @@
1497	2295	){
1498	2296	int iCur;
1499	2297	int iHiwtr;
1500	2298
1501	2299	if( pArg && pArg->out ){
1502		-
1503		- iHiwtr = iCur = -1;
1504		- sqlite3_status(SQLITE_STATUS_MEMORY_USED, &iCur, &iHiwtr, bReset);
1505		- raw_printf(pArg->out,
1506		- "Memory Used: %d (max %d) bytes\n",
1507		- iCur, iHiwtr);
1508		- iHiwtr = iCur = -1;
1509		- sqlite3_status(SQLITE_STATUS_MALLOC_COUNT, &iCur, &iHiwtr, bReset);
1510		- raw_printf(pArg->out, "Number of Outstanding Allocations: %d (max %d)\n",
1511		- iCur, iHiwtr);
	2300	+ displayStatLine(pArg, "Memory Used:",
	2301	+ "%lld (max %lld) bytes", SQLITE_STATUS_MEMORY_USED, bReset);
	2302	+ displayStatLine(pArg, "Number of Outstanding Allocations:",
	2303	+ "%lld (max %lld)", SQLITE_STATUS_MALLOC_COUNT, bReset);
1512	2304	if( pArg->shellFlgs & SHFLG_Pagecache ){
1513		- iHiwtr = iCur = -1;
1514		- sqlite3_status(SQLITE_STATUS_PAGECACHE_USED, &iCur, &iHiwtr, bReset);
1515		- raw_printf(pArg->out,
1516		- "Number of Pcache Pages Used: %d (max %d) pages\n",
1517		- iCur, iHiwtr);
1518		- }
1519		- iHiwtr = iCur = -1;
1520		- sqlite3_status(SQLITE_STATUS_PAGECACHE_OVERFLOW, &iCur, &iHiwtr, bReset);
1521		- raw_printf(pArg->out,
1522		- "Number of Pcache Overflow Bytes: %d (max %d) bytes\n",
1523		- iCur, iHiwtr);
	2305	+ displayStatLine(pArg, "Number of Pcache Pages Used:",
	2306	+ "%lld (max %lld) pages", SQLITE_STATUS_PAGECACHE_USED, bReset);
	2307	+ }
	2308	+ displayStatLine(pArg, "Number of Pcache Overflow Bytes:",
	2309	+ "%lld (max %lld) bytes", SQLITE_STATUS_PAGECACHE_OVERFLOW, bReset);
1524	2310	if( pArg->shellFlgs & SHFLG_Scratch ){
1525		- iHiwtr = iCur = -1;
1526		- sqlite3_status(SQLITE_STATUS_SCRATCH_USED, &iCur, &iHiwtr, bReset);
1527		- raw_printf(pArg->out,
1528		- "Number of Scratch Allocations Used: %d (max %d)\n",
1529		- iCur, iHiwtr);
1530		- }
1531		- iHiwtr = iCur = -1;
1532		- sqlite3_status(SQLITE_STATUS_SCRATCH_OVERFLOW, &iCur, &iHiwtr, bReset);
1533		- raw_printf(pArg->out,
1534		- "Number of Scratch Overflow Bytes: %d (max %d) bytes\n",
1535		- iCur, iHiwtr);
1536		- iHiwtr = iCur = -1;
1537		- sqlite3_status(SQLITE_STATUS_MALLOC_SIZE, &iCur, &iHiwtr, bReset);
1538		- raw_printf(pArg->out, "Largest Allocation: %d bytes\n",
1539		- iHiwtr);
1540		- iHiwtr = iCur = -1;
1541		- sqlite3_status(SQLITE_STATUS_PAGECACHE_SIZE, &iCur, &iHiwtr, bReset);
1542		- raw_printf(pArg->out, "Largest Pcache Allocation: %d bytes\n",
1543		- iHiwtr);
1544		- iHiwtr = iCur = -1;
1545		- sqlite3_status(SQLITE_STATUS_SCRATCH_SIZE, &iCur, &iHiwtr, bReset);
1546		- raw_printf(pArg->out, "Largest Scratch Allocation: %d bytes\n",
1547		- iHiwtr);
	2311	+ displayStatLine(pArg, "Number of Scratch Allocations Used:",
	2312	+ "%lld (max %lld)", SQLITE_STATUS_SCRATCH_USED, bReset);
	2313	+ }
	2314	+ displayStatLine(pArg, "Number of Scratch Overflow Bytes:",
	2315	+ "%lld (max %lld) bytes", SQLITE_STATUS_SCRATCH_OVERFLOW, bReset);
	2316	+ displayStatLine(pArg, "Largest Allocation:",
	2317	+ "%lld bytes", SQLITE_STATUS_MALLOC_SIZE, bReset);
	2318	+ displayStatLine(pArg, "Largest Pcache Allocation:",
	2319	+ "%lld bytes", SQLITE_STATUS_PAGECACHE_SIZE, bReset);
	2320	+ displayStatLine(pArg, "Largest Scratch Allocation:",
	2321	+ "%lld bytes", SQLITE_STATUS_SCRATCH_SIZE, bReset);
1548	2322	#ifdef YYTRACKMAXSTACKDEPTH
1549		- iHiwtr = iCur = -1;
1550		- sqlite3_status(SQLITE_STATUS_PARSER_STACK, &iCur, &iHiwtr, bReset);
1551		- raw_printf(pArg->out, "Deepest Parser Stack: %d (max %d)\n",
1552		- iCur, iHiwtr);
	2323	+ displayStatLine(pArg, "Deepest Parser Stack:",
	2324	+ "%lld (max %lld)", SQLITE_STATUS_PARSER_STACK, bReset);
1553	2325	#endif
1554	2326	}
1555	2327
1556	2328	if( pArg && pArg->out && db ){
1557	2329	if( pArg->shellFlgs & SHFLG_Lookaside ){
		@@ -1934,11 +2706,11 @@
1934	2706	pArg->pStmt = pStmt;
1935	2707	pArg->cnt = 0;
1936	2708	}
1937	2709
1938	2710	/* echo the sql statement if echo on */
1939		- if( pArg && pArg->echoOn ){
	2711	+ if( pArg && ShellHasFlag(pArg, SHFLG_Echo) ){
1940	2712	utf8_printf(pArg->out, "%s\n", zStmtSql ? zStmtSql : zSql);
1941	2713	}
1942	2714
1943	2715	/* Show the EXPLAIN QUERY PLAN if .eqp is on */
1944	2716	if( pArg && pArg->autoEQP && sqlite3_strlike("EXPLAIN%",zStmtSql,0)!=0 ){
		@@ -2022,10 +2794,142 @@
2022	2794	} /* end while */
2023	2795
2024	2796	return rc;
2025	2797	}
2026	2798
	2799	+/*
	2800	+** Release memory previously allocated by tableColumnList().
	2801	+*/
	2802	+static void freeColumnList(char **azCol){
	2803	+ int i;
	2804	+ for(i=1; azCol[i]; i++){
	2805	+ sqlite3_free(azCol[i]);
	2806	+ }
	2807	+ /* azCol[0] is a static string */
	2808	+ sqlite3_free(azCol);
	2809	+}
	2810	+
	2811	+/*
	2812	+** Return a list of pointers to strings which are the names of all
	2813	+** columns in table zTab. The memory to hold the names is dynamically
	2814	+** allocated and must be released by the caller using a subsequent call
	2815	+** to freeColumnList().
	2816	+**
	2817	+** The azCol[0] entry is usually NULL. However, if zTab contains a rowid
	2818	+** value that needs to be preserved, then azCol[0] is filled in with the
	2819	+** name of the rowid column.
	2820	+**
	2821	+** The first regular column in the table is azCol[1]. The list is terminated
	2822	+** by an entry with azCol[i]==0.
	2823	+*/
	2824	+static char *tableColumnList(ShellState p, const char *zTab){
	2825	+ char **azCol = 0;
	2826	+ sqlite3_stmt *pStmt;
	2827	+ char *zSql;
	2828	+ int nCol = 0;
	2829	+ int nAlloc = 0;
	2830	+ int nPK = 0; /* Number of PRIMARY KEY columns seen */
	2831	+ int isIPK = 0; /* True if one PRIMARY KEY column of type INTEGER */
	2832	+ int preserveRowid = ShellHasFlag(p, SHFLG_PreserveRowid);
	2833	+ int rc;
	2834	+
	2835	+ zSql = sqlite3_mprintf("PRAGMA table_info=%Q", zTab);
	2836	+ rc = sqlite3_prepare_v2(p->db, zSql, -1, &pStmt, 0);
	2837	+ sqlite3_free(zSql);
	2838	+ if( rc ) return 0;
	2839	+ while( sqlite3_step(pStmt)==SQLITE_ROW ){
	2840	+ if( nCol>=nAlloc-2 ){
	2841	+ nAlloc = nAlloc*2 + nCol + 10;
	2842	+ azCol = sqlite3_realloc(azCol, nAlloc*sizeof(azCol[0]));
	2843	+ if( azCol==0 ){
	2844	+ raw_printf(stderr, "Error: out of memory\n");
	2845	+ exit(1);
	2846	+ }
	2847	+ }
	2848	+ azCol[++nCol] = sqlite3_mprintf("%s", sqlite3_column_text(pStmt, 1));
	2849	+ if( sqlite3_column_int(pStmt, 5) ){
	2850	+ nPK++;
	2851	+ if( nPK==1
	2852	+ && sqlite3_stricmp((const char*)sqlite3_column_text(pStmt,2),
	2853	+ "INTEGER")==0
	2854	+ ){
	2855	+ isIPK = 1;
	2856	+ }else{
	2857	+ isIPK = 0;
	2858	+ }
	2859	+ }
	2860	+ }
	2861	+ sqlite3_finalize(pStmt);
	2862	+ azCol[0] = 0;
	2863	+ azCol[nCol+1] = 0;
	2864	+
	2865	+ /* The decision of whether or not a rowid really needs to be preserved
	2866	+ ** is tricky. We never need to preserve a rowid for a WITHOUT ROWID table
	2867	+ ** or a table with an INTEGER PRIMARY KEY. We are unable to preserve
	2868	+ ** rowids on tables where the rowid is inaccessible because there are other
	2869	+ ** columns in the table named "rowid", "_rowid_", and "oid".
	2870	+ */
	2871	+ if( preserveRowid && isIPK ){
	2872	+ /* If a single PRIMARY KEY column with type INTEGER was seen, then it
	2873	+ ** might be an alise for the ROWID. But it might also be a WITHOUT ROWID
	2874	+ ** table or a INTEGER PRIMARY KEY DESC column, neither of which are
	2875	+ ** ROWID aliases. To distinguish these cases, check to see if
	2876	+ ** there is a "pk" entry in "PRAGMA index_list". There will be
	2877	+ ** no "pk" index if the PRIMARY KEY really is an alias for the ROWID.
	2878	+ */
	2879	+ zSql = sqlite3_mprintf("SELECT 1 FROM pragma_index_list(%Q)"
	2880	+ " WHERE origin='pk'", zTab);
	2881	+ rc = sqlite3_prepare_v2(p->db, zSql, -1, &pStmt, 0);
	2882	+ sqlite3_free(zSql);
	2883	+ if( rc ){
	2884	+ freeColumnList(azCol);
	2885	+ return 0;
	2886	+ }
	2887	+ rc = sqlite3_step(pStmt);
	2888	+ sqlite3_finalize(pStmt);
	2889	+ preserveRowid = rc==SQLITE_ROW;
	2890	+ }
	2891	+ if( preserveRowid ){
	2892	+ /* Only preserve the rowid if we can find a name to use for the
	2893	+ ** rowid */
	2894	+ static char *azRowid[] = { "rowid", "_rowid_", "oid" };
	2895	+ int i, j;
	2896	+ for(j=0; j<3; j++){
	2897	+ for(i=1; i<=nCol; i++){
	2898	+ if( sqlite3_stricmp(azRowid[j],azCol[i])==0 ) break;
	2899	+ }
	2900	+ if( i>nCol ){
	2901	+ /* At this point, we know that azRowid[j] is not the name of any
	2902	+ ** ordinary column in the table. Verify that azRowid[j] is a valid
	2903	+ ** name for the rowid before adding it to azCol[0]. WITHOUT ROWID
	2904	+ ** tables will fail this last check */
	2905	+ int rc;
	2906	+ rc = sqlite3_table_column_metadata(p->db,0,zTab,azRowid[j],0,0,0,0,0);
	2907	+ if( rc==SQLITE_OK ) azCol[0] = azRowid[j];
	2908	+ break;
	2909	+ }
	2910	+ }
	2911	+ }
	2912	+ return azCol;
	2913	+}
	2914	+
	2915	+/*
	2916	+** Toggle the reverse_unordered_selects setting.
	2917	+*/
	2918	+static void toggleSelectOrder(sqlite3 *db){
	2919	+ sqlite3_stmt *pStmt = 0;
	2920	+ int iSetting = 0;
	2921	+ char zStmt[100];
	2922	+ sqlite3_prepare_v2(db, "PRAGMA reverse_unordered_selects", -1, &pStmt, 0);
	2923	+ if( sqlite3_step(pStmt)==SQLITE_ROW ){
	2924	+ iSetting = sqlite3_column_int(pStmt, 0);
	2925	+ }
	2926	+ sqlite3_finalize(pStmt);
	2927	+ sqlite3_snprintf(sizeof(zStmt), zStmt,
	2928	+ "PRAGMA reverse_unordered_selects(%d)", !iSetting);
	2929	+ sqlite3_exec(db, zStmt, 0, 0, 0);
	2930	+}
2027	2931
2028	2932	/*
2029	2933	** This is a different callback routine used for dumping the database.
2030	2934	** Each row received by this callback consists of a table name,
2031	2935	** the table type ("index" or "table") and SQL to create the table.
		@@ -2034,21 +2938,20 @@
2034	2938	static int dump_callback(void pArg, int nArg, char azArg, char *azCol){
2035	2939	int rc;
2036	2940	const char *zTable;
2037	2941	const char *zType;
2038	2942	const char *zSql;
2039		- const char *zPrepStmt = 0;
2040	2943	ShellState p = (ShellState )pArg;
2041	2944
2042	2945	UNUSED_PARAMETER(azCol);
2043	2946	if( nArg!=3 ) return 1;
2044	2947	zTable = azArg[0];
2045	2948	zType = azArg[1];
2046	2949	zSql = azArg[2];
2047	2950
2048	2951	if( strcmp(zTable, "sqlite_sequence")==0 ){
2049		- zPrepStmt = "DELETE FROM sqlite_sequence;\n";
	2952	+ raw_printf(p->out, "DELETE FROM sqlite_sequence;\n");
2050	2953	}else if( sqlite3_strglob("sqlite_stat?", zTable)==0 ){
2051	2954	raw_printf(p->out, "ANALYZE sqlite_master;\n");
2052	2955	}else if( strncmp(zTable, "sqlite_", 7)==0 ){
2053	2956	return 0;
2054	2957	}else if( strncmp(zSql, "CREATE VIRTUAL TABLE", 20)==0 ){
		@@ -2067,62 +2970,74 @@
2067	2970	}else{
2068	2971	printSchemaLine(p->out, zSql, ";\n");
2069	2972	}
2070	2973
2071	2974	if( strcmp(zType, "table")==0 ){
2072		- sqlite3_stmt *pTableInfo = 0;
2073		- char *zSelect = 0;
2074		- char *zTableInfo = 0;
2075		- char *zTmp = 0;
2076		- int nRow = 0;
2077		-
2078		- zTableInfo = appendText(zTableInfo, "PRAGMA table_info(", 0);
2079		- zTableInfo = appendText(zTableInfo, zTable, '"');
2080		- zTableInfo = appendText(zTableInfo, ");", 0);
2081		-
2082		- rc = sqlite3_prepare_v2(p->db, zTableInfo, -1, &pTableInfo, 0);
2083		- free(zTableInfo);
2084		- if( rc!=SQLITE_OK \|\| !pTableInfo ){
2085		- return 1;
2086		- }
2087		-
2088		- zSelect = appendText(zSelect, "SELECT 'INSERT INTO ' \|\| ", 0);
	2975	+ ShellText sSelect;
	2976	+ ShellText sTable;
	2977	+ char **azCol;
	2978	+ int i;
	2979	+ char *savedDestTable;
	2980	+ int savedMode;
	2981	+
	2982	+ azCol = tableColumnList(p, zTable);
	2983	+ if( azCol==0 ){
	2984	+ p->nErr++;
	2985	+ return 0;
	2986	+ }
	2987	+
2089	2988	/* Always quote the table name, even if it appears to be pure ascii,
2090	2989	** in case it is a keyword. Ex: INSERT INTO "table" ... */
2091		- zTmp = appendText(zTmp, zTable, '"');
2092		- if( zTmp ){
2093		- zSelect = appendText(zSelect, zTmp, '\'');
2094		- free(zTmp);
2095		- }
2096		- zSelect = appendText(zSelect, " \|\| ' VALUES(' \|\| ", 0);
2097		- rc = sqlite3_step(pTableInfo);
2098		- while( rc==SQLITE_ROW ){
2099		- const char zText = (const char )sqlite3_column_text(pTableInfo, 1);
2100		- zSelect = appendText(zSelect, "quote(", 0);
2101		- zSelect = appendText(zSelect, zText, '"');
2102		- rc = sqlite3_step(pTableInfo);
2103		- if( rc==SQLITE_ROW ){
2104		- zSelect = appendText(zSelect, "), ", 0);
2105		- }else{
2106		- zSelect = appendText(zSelect, ") ", 0);
2107		- }
2108		- nRow++;
2109		- }
2110		- rc = sqlite3_finalize(pTableInfo);
2111		- if( rc!=SQLITE_OK \|\| nRow==0 ){
2112		- free(zSelect);
2113		- return 1;
2114		- }
2115		- zSelect = appendText(zSelect, "\|\| ')' FROM ", 0);
2116		- zSelect = appendText(zSelect, zTable, '"');
2117		-
2118		- rc = run_table_dump_query(p, zSelect, zPrepStmt);
2119		- if( rc==SQLITE_CORRUPT ){
2120		- zSelect = appendText(zSelect, " ORDER BY rowid DESC", 0);
2121		- run_table_dump_query(p, zSelect, 0);
2122		- }
2123		- free(zSelect);
	2990	+ initText(&sTable);
	2991	+ appendText(&sTable, zTable, quoteChar(zTable));
	2992	+ /* If preserving the rowid, add a column list after the table name.
	2993	+ ** In other words: "INSERT INTO tab(rowid,a,b,c,...) VALUES(...)"
	2994	+ ** instead of the usual "INSERT INTO tab VALUES(...)".
	2995	+ */
	2996	+ if( azCol[0] ){
	2997	+ appendText(&sTable, "(", 0);
	2998	+ appendText(&sTable, azCol[0], 0);
	2999	+ for(i=1; azCol[i]; i++){
	3000	+ appendText(&sTable, ",", 0);
	3001	+ appendText(&sTable, azCol[i], quoteChar(azCol[i]));
	3002	+ }
	3003	+ appendText(&sTable, ")", 0);
	3004	+ }
	3005	+
	3006	+ /* Build an appropriate SELECT statement */
	3007	+ initText(&sSelect);
	3008	+ appendText(&sSelect, "SELECT ", 0);
	3009	+ if( azCol[0] ){
	3010	+ appendText(&sSelect, azCol[0], 0);
	3011	+ appendText(&sSelect, ",", 0);
	3012	+ }
	3013	+ for(i=1; azCol[i]; i++){
	3014	+ appendText(&sSelect, azCol[i], quoteChar(azCol[i]));
	3015	+ if( azCol[i+1] ){
	3016	+ appendText(&sSelect, ",", 0);
	3017	+ }
	3018	+ }
	3019	+ freeColumnList(azCol);
	3020	+ appendText(&sSelect, " FROM ", 0);
	3021	+ appendText(&sSelect, zTable, quoteChar(zTable));
	3022	+
	3023	+ savedDestTable = p->zDestTable;
	3024	+ savedMode = p->mode;
	3025	+ p->zDestTable = sTable.z;
	3026	+ p->mode = p->cMode = MODE_Insert;
	3027	+ rc = shell_exec(p->db, sSelect.z, shell_callback, p, 0);
	3028	+ if( (rc&0xff)==SQLITE_CORRUPT ){
	3029	+ raw_printf(p->out, "/**** CORRUPTION ERROR *****/\n");
	3030	+ toggleSelectOrder(p->db);
	3031	+ shell_exec(p->db, sSelect.z, shell_callback, p, 0);
	3032	+ toggleSelectOrder(p->db);
	3033	+ }
	3034	+ p->zDestTable = savedDestTable;
	3035	+ p->mode = savedMode;
	3036	+ freeText(&sTable);
	3037	+ freeText(&sSelect);
	3038	+ if( rc ) p->nErr++;
2124	3039	}
2125	3040	return 0;
2126	3041	}
2127	3042
2128	3043	/*
		@@ -2210,11 +3125,11 @@
2210	3125	" csv Comma-separated values\n"
2211	3126	" column Left-aligned columns. (See .width)\n"
2212	3127	" html HTML <table> code\n"
2213	3128	" insert SQL insert statements for TABLE\n"
2214	3129	" line One value per line\n"
2215		- " list Values delimited by .separator strings\n"
	3130	+ " list Values delimited by \"\|\"\n"
2216	3131	" quote Escape answers as for SQL\n"
2217	3132	" tabs Tab-separated values\n"
2218	3133	" tcl TCL list elements\n"
2219	3134	".nullvalue STRING Use STRING in place of NULL values\n"
2220	3135	".once FILENAME Output for the next SQL command only to FILENAME\n"
		@@ -2233,10 +3148,11 @@
2233	3148	".separator COL ?ROW? Change the column separator and optionally the row\n"
2234	3149	" separator for both the output mode and .import\n"
2235	3150	#if defined(SQLITE_ENABLE_SESSION)
2236	3151	".session CMD ... Create or control sessions\n"
2237	3152	#endif
	3153	+ ".sha3sum ?OPTIONS...? Compute a SHA3 hash of database content\n"
2238	3154	".shell CMD ARGS... Run CMD ARGS... in a system shell\n"
2239	3155	".show Show the current values for various settings\n"
2240	3156	".stats ?on\|off? Show stats or turn stats on or off\n"
2241	3157	".system CMD ARGS... Run CMD ARGS... in a system shell\n"
2242	3158	".tables ?TABLE? List names of tables\n"
		@@ -2423,14 +3339,10 @@
2423	3339	static void open_db(ShellState *p, int keepAlive){
2424	3340	if( p->db==0 ){
2425	3341	sqlite3_initialize();
2426	3342	sqlite3_open(p->zDbFilename, &p->db);
2427	3343	globalDb = p->db;
2428		- if( p->db && sqlite3_errcode(p->db)==SQLITE_OK ){
2429		- sqlite3_create_function(p->db, "shellstatic", 0, SQLITE_UTF8, 0,
2430		- shellstaticFunc, 0, 0);
2431		- }
2432	3344	if( p->db==0 \|\| SQLITE_OK!=sqlite3_errcode(p->db) ){
2433	3345	utf8_printf(stderr,"Error: unable to open database \"%s\": %s\n",
2434	3346	p->zDbFilename, sqlite3_errmsg(p->db));
2435	3347	if( keepAlive ) return;
2436	3348	exit(1);
		@@ -2440,10 +3352,18 @@
2440	3352	#endif
2441	3353	sqlite3_create_function(p->db, "readfile", 1, SQLITE_UTF8, 0,
2442	3354	readfileFunc, 0, 0);
2443	3355	sqlite3_create_function(p->db, "writefile", 2, SQLITE_UTF8, 0,
2444	3356	writefileFunc, 0, 0);
	3357	+ sqlite3_create_function(p->db, "sha3", 1, SQLITE_UTF8, 0,
	3358	+ sha3Func, 0, 0);
	3359	+ sqlite3_create_function(p->db, "sha3", 2, SQLITE_UTF8, 0,
	3360	+ sha3Func, 0, 0);
	3361	+ sqlite3_create_function(p->db, "sha3_query", 1, SQLITE_UTF8, 0,
	3362	+ sha3QueryFunc, 0, 0);
	3363	+ sqlite3_create_function(p->db, "sha3_query", 2, SQLITE_UTF8, 0,
	3364	+ sha3QueryFunc, 0, 0);
2445	3365	}
2446	3366	}
2447	3367
2448	3368	/*
2449	3369	** Do C-language style dequoting.
		@@ -2564,11 +3484,11 @@
2564	3484
2565	3485	/*
2566	3486	** Interpret zArg as either an integer or a boolean value. Return 1 or 0
2567	3487	** for TRUE and FALSE. Return the integer value if appropriate.
2568	3488	*/
2569		-static int booleanValue(char *zArg){
	3489	+static int booleanValue(const char *zArg){
2570	3490	int i;
2571	3491	if( zArg[0]=='0' && zArg[1]=='x' ){
2572	3492	for(i=2; hexDigitValue(zArg[i])>=0; i++){}
2573	3493	}else{
2574	3494	for(i=0; zArg[i]>='0' && zArg[i]<='9'; i++){}
		@@ -2582,10 +3502,21 @@
2582	3502	}
2583	3503	utf8_printf(stderr, "ERROR: Not a boolean value: \"%s\". Assuming \"no\".\n",
2584	3504	zArg);
2585	3505	return 0;
2586	3506	}
	3507	+
	3508	+/*
	3509	+** Set or clear a shell flag according to a boolean value.
	3510	+*/
	3511	+static void setOrClearFlag(ShellState p, unsigned mFlag, const char zArg){
	3512	+ if( booleanValue(zArg) ){
	3513	+ ShellSetFlag(p, mFlag);
	3514	+ }else{
	3515	+ ShellClearFlag(p, mFlag);
	3516	+ }
	3517	+}
2587	3518
2588	3519	/*
2589	3520	** Close an output file, assuming it is not stderr or stdout
2590	3521	*/
2591	3522	static void output_file_close(FILE *f){
		@@ -3656,11 +4587,11 @@
3656	4587	test_breakpoint();
3657	4588	}else
3658	4589
3659	4590	if( c=='c' && n>=3 && strncmp(azArg[0], "changes", n)==0 ){
3660	4591	if( nArg==2 ){
3661		- p->countChanges = booleanValue(azArg[1]);
	4592	+ setOrClearFlag(p, SHFLG_CountChanges, azArg[1]);
3662	4593	}else{
3663	4594	raw_printf(stderr, "Usage: .changes on\|off\n");
3664	4595	rc = 1;
3665	4596	}
3666	4597	}else
		@@ -3720,25 +4651,46 @@
3720	4651	if( c=='d' && strncmp(azArg[0], "dbinfo", n)==0 ){
3721	4652	rc = shell_dbinfo_command(p, nArg, azArg);
3722	4653	}else
3723	4654
3724	4655	if( c=='d' && strncmp(azArg[0], "dump", n)==0 ){
	4656	+ const char *zLike = 0;
	4657	+ int i;
	4658	+ ShellClearFlag(p, SHFLG_PreserveRowid);
	4659	+ for(i=1; i<nArg; i++){
	4660	+ if( azArg[i][0]=='-' ){
	4661	+ const char *z = azArg[i]+1;
	4662	+ if( z[0]=='-' ) z++;
	4663	+ if( strcmp(z,"preserve-rowids")==0 ){
	4664	+ ShellSetFlag(p, SHFLG_PreserveRowid);
	4665	+ }else
	4666	+ {
	4667	+ raw_printf(stderr, "Unknown option \"%s\" on \".dump\"\n", azArg[i]);
	4668	+ rc = 1;
	4669	+ goto meta_command_exit;
	4670	+ }
	4671	+ }else if( zLike ){
	4672	+ raw_printf(stderr, "Usage: .dump ?--preserve-rowids? ?LIKE-PATTERN?\n");
	4673	+ rc = 1;
	4674	+ goto meta_command_exit;
	4675	+ }else{
	4676	+ zLike = azArg[i];
	4677	+ }
	4678	+ }
3725	4679	open_db(p, 0);
3726	4680	/* When playing back a "dump", the content might appear in an order
3727	4681	** which causes immediate foreign key constraints to be violated.
3728	4682	** So disable foreign-key constraint enforcement to prevent problems. */
3729		- if( nArg!=1 && nArg!=2 ){
3730		- raw_printf(stderr, "Usage: .dump ?LIKE-PATTERN?\n");
3731		- rc = 1;
3732		- goto meta_command_exit;
3733		- }
3734	4683	raw_printf(p->out, "PRAGMA foreign_keys=OFF;\n");
3735	4684	raw_printf(p->out, "BEGIN TRANSACTION;\n");
3736	4685	p->writableSchema = 0;
	4686	+ /* Set writable_schema=ON since doing so forces SQLite to initialize
	4687	+ ** as much of the schema as it can even if the sqlite_master table is
	4688	+ ** corrupt. */
3737	4689	sqlite3_exec(p->db, "SAVEPOINT dump; PRAGMA writable_schema=ON", 0, 0, 0);
3738	4690	p->nErr = 0;
3739		- if( nArg==1 ){
	4691	+ if( zLike==0 ){
3740	4692	run_schema_dump_query(p,
3741	4693	"SELECT name, type, sql FROM sqlite_master "
3742	4694	"WHERE sql NOT NULL AND type=='table' AND name!='sqlite_sequence'"
3743	4695	);
3744	4696	run_schema_dump_query(p,
		@@ -3748,25 +4700,24 @@
3748	4700	run_table_dump_query(p,
3749	4701	"SELECT sql FROM sqlite_master "
3750	4702	"WHERE sql NOT NULL AND type IN ('index','trigger','view')", 0
3751	4703	);
3752	4704	}else{
3753		- int i;
3754		- for(i=1; i<nArg; i++){
3755		- zShellStatic = azArg[i];
3756		- run_schema_dump_query(p,
3757		- "SELECT name, type, sql FROM sqlite_master "
3758		- "WHERE tbl_name LIKE shellstatic() AND type=='table'"
3759		- " AND sql NOT NULL");
3760		- run_table_dump_query(p,
3761		- "SELECT sql FROM sqlite_master "
3762		- "WHERE sql NOT NULL"
3763		- " AND type IN ('index','trigger','view')"
3764		- " AND tbl_name LIKE shellstatic()", 0
3765		- );
3766		- zShellStatic = 0;
3767		- }
	4705	+ char *zSql;
	4706	+ zSql = sqlite3_mprintf(
	4707	+ "SELECT name, type, sql FROM sqlite_master "
	4708	+ "WHERE tbl_name LIKE %Q AND type=='table'"
	4709	+ " AND sql NOT NULL", zLike);
	4710	+ run_schema_dump_query(p,zSql);
	4711	+ sqlite3_free(zSql);
	4712	+ zSql = sqlite3_mprintf(
	4713	+ "SELECT sql FROM sqlite_master "
	4714	+ "WHERE sql NOT NULL"
	4715	+ " AND type IN ('index','trigger','view')"
	4716	+ " AND tbl_name LIKE %Q", zLike);
	4717	+ run_table_dump_query(p, zSql, 0);
	4718	+ sqlite3_free(zSql);
3768	4719	}
3769	4720	if( p->writableSchema ){
3770	4721	raw_printf(p->out, "PRAGMA writable_schema=OFF;\n");
3771	4722	p->writableSchema = 0;
3772	4723	}
		@@ -3775,11 +4726,11 @@
3775	4726	raw_printf(p->out, p->nErr ? "ROLLBACK; -- due to errors\n" : "COMMIT;\n");
3776	4727	}else
3777	4728
3778	4729	if( c=='e' && strncmp(azArg[0], "echo", n)==0 ){
3779	4730	if( nArg==2 ){
3780		- p->echoOn = booleanValue(azArg[1]);
	4731	+ setOrClearFlag(p, SHFLG_Echo, azArg[1]);
3781	4732	}else{
3782	4733	raw_printf(stderr, "Usage: .echo on\|off\n");
3783	4734	rc = 1;
3784	4735	}
3785	4736	}else
		@@ -4342,10 +5293,11 @@
4342	5293	int newFlag = 0; /* True to delete file before opening */
4343	5294	/* Close the existing database */
4344	5295	session_close_all(p);
4345	5296	sqlite3_close(p->db);
4346	5297	p->db = 0;
	5298	+ p->zDbFilename = 0;
4347	5299	sqlite3_free(p->zFreeOnClose);
4348	5300	p->zFreeOnClose = 0;
4349	5301	/* Check for command-line arguments */
4350	5302	for(iName=1; iName<nArg && azArg[iName][0]=='-'; iName++){
4351	5303	const char *z = azArg[iName];
		@@ -4571,21 +5523,21 @@
4571	5523	new_colv[0] = "sql";
4572	5524	new_colv[1] = 0;
4573	5525	callback(&data, 1, new_argv, new_colv);
4574	5526	rc = SQLITE_OK;
4575	5527	}else{
4576		- zShellStatic = azArg[1];
4577		- rc = sqlite3_exec(p->db,
	5528	+ char *zSql;
	5529	+ zSql = sqlite3_mprintf(
4578	5530	"SELECT sql FROM "
4579	5531	" (SELECT sql sql, type type, tbl_name tbl_name, name name, rowid x"
4580	5532	" FROM sqlite_master UNION ALL"
4581	5533	" SELECT sql, type, tbl_name, name, rowid FROM sqlite_temp_master) "
4582		- "WHERE lower(tbl_name) LIKE shellstatic()"
	5534	+ "WHERE lower(tbl_name) LIKE %Q"
4583	5535	" AND type!='meta' AND sql NOTNULL "
4584		- "ORDER BY rowid",
4585		- callback, &data, &zErrMsg);
4586		- zShellStatic = 0;
	5536	+ "ORDER BY rowid", azArg[1]);
	5537	+ rc = sqlite3_exec(p->db, zSql, callback, &data, &zErrMsg);
	5538	+ sqlite3_free(zSql);
4587	5539	}
4588	5540	}else if( nArg==1 ){
4589	5541	rc = sqlite3_exec(p->db,
4590	5542	"SELECT sql FROM "
4591	5543	" (SELECT sql sql, type type, tbl_name tbl_name, name name, rowid x"
		@@ -4612,11 +5564,11 @@
4612	5564	}
4613	5565	}else
4614	5566
4615	5567	#if defined(SQLITE_DEBUG) && defined(SQLITE_ENABLE_SELECTTRACE)
4616	5568	if( c=='s' && n==11 && strncmp(azArg[0], "selecttrace", n)==0 ){
4617		- sqlite3SelectTrace = integerValue(azArg[1]);
	5569	+ sqlite3SelectTrace = (int)integerValue(azArg[1]);
4618	5570	}else
4619	5571	#endif
4620	5572
4621	5573	#if defined(SQLITE_ENABLE_SESSION)
4622	5574	if( c=='s' && strncmp(azArg[0],"session",n)==0 && n>=3 ){
		@@ -4834,10 +5786,123 @@
4834	5786	utf8_printf(p->out, "%s", zBuf);
4835	5787	}
4836	5788	}
4837	5789	}else
4838	5790	#endif
	5791	+
	5792	+ if( c=='s' && n>=4 && strncmp(azArg[0],"selftest",n)==0 ){
	5793	+ int bIsInit = 0; /* True to initialize the SELFTEST table */
	5794	+ int bVerbose = 0; /* Verbose output */
	5795	+ int bSelftestExists; /* True if SELFTEST already exists */
	5796	+ char *azTest = 0; / Content of the SELFTEST table */
	5797	+ int nRow = 0; /* Number of rows in the SELFTEST table */
	5798	+ int nCol = 4; /* Number of columns in the SELFTEST table */
	5799	+ int i; /* Loop counter */
	5800	+ int nTest = 0; /* Number of tests runs */
	5801	+ int nErr = 0; /* Number of errors seen */
	5802	+ ShellText str; /* Answer for a query */
	5803	+ static char *azDefaultTest[] = {
	5804	+ 0, 0, 0, 0,
	5805	+ "0", "memo", "Missing SELFTEST table - default checks only", "",
	5806	+ "1", "run", "PRAGMA integrity_check", "ok"
	5807	+ };
	5808	+ static const int nDefaultRow = 2;
	5809	+
	5810	+ open_db(p,0);
	5811	+ for(i=1; i<nArg; i++){
	5812	+ const char *z = azArg[i];
	5813	+ if( z[0]=='-' && z[1]=='-' ) z++;
	5814	+ if( strcmp(z,"-init")==0 ){
	5815	+ bIsInit = 1;
	5816	+ }else
	5817	+ if( strcmp(z,"-v")==0 ){
	5818	+ bVerbose++;
	5819	+ }else
	5820	+ {
	5821	+ utf8_printf(stderr, "Unknown option \"%s\" on \"%s\"\n",
	5822	+ azArg[i], azArg[0]);
	5823	+ raw_printf(stderr, "Should be one of: --init -v\n");
	5824	+ rc = 1;
	5825	+ goto meta_command_exit;
	5826	+ }
	5827	+ }
	5828	+ if( sqlite3_table_column_metadata(p->db,"main","selftest",0,0,0,0,0,0)
	5829	+ != SQLITE_OK ){
	5830	+ bSelftestExists = 0;
	5831	+ }else{
	5832	+ bSelftestExists = 1;
	5833	+ }
	5834	+ if( bIsInit ){
	5835	+ createSelftestTable(p);
	5836	+ bSelftestExists = 1;
	5837	+ }
	5838	+ if( bSelftestExists ){
	5839	+ rc = sqlite3_get_table(p->db,
	5840	+ "SELECT tno,op,cmd,ans FROM selftest ORDER BY tno",
	5841	+ &azTest, &nRow, &nCol, 0);
	5842	+ if( rc ){
	5843	+ raw_printf(stderr, "Error querying the selftest table\n");
	5844	+ rc = 1;
	5845	+ sqlite3_free_table(azTest);
	5846	+ goto meta_command_exit;
	5847	+ }else if( nRow==0 ){
	5848	+ sqlite3_free_table(azTest);
	5849	+ azTest = azDefaultTest;
	5850	+ nRow = nDefaultRow;
	5851	+ }
	5852	+ }else{
	5853	+ azTest = azDefaultTest;
	5854	+ nRow = nDefaultRow;
	5855	+ }
	5856	+ initText(&str);
	5857	+ appendText(&str, "x", 0);
	5858	+ for(i=1; i<=nRow; i++){
	5859	+ int tno = atoi(azTest[i*nCol]);
	5860	+ const char zOp = azTest[inCol+1];
	5861	+ const char zSql = azTest[inCol+2];
	5862	+ const char zAns = azTest[inCol+3];
	5863	+
	5864	+ if( bVerbose>0 ){
	5865	+ char *zQuote = sqlite3_mprintf("%q", zSql);
	5866	+ printf("%d: %s %s\n", tno, zOp, zSql);
	5867	+ sqlite3_free(zQuote);
	5868	+ }
	5869	+ if( strcmp(zOp,"memo")==0 ){
	5870	+ utf8_printf(p->out, "%s\n", zSql);
	5871	+ }else
	5872	+ if( strcmp(zOp,"run")==0 ){
	5873	+ char *zErrMsg = 0;
	5874	+ str.n = 0;
	5875	+ str.z[0] = 0;
	5876	+ rc = sqlite3_exec(p->db, zSql, captureOutputCallback, &str, &zErrMsg);
	5877	+ nTest++;
	5878	+ if( bVerbose ){
	5879	+ utf8_printf(p->out, "Result: %s\n", str.z);
	5880	+ }
	5881	+ if( rc \|\| zErrMsg ){
	5882	+ nErr++;
	5883	+ rc = 1;
	5884	+ utf8_printf(p->out, "%d: error-code-%d: %s\n", tno, rc, zErrMsg);
	5885	+ sqlite3_free(zErrMsg);
	5886	+ }else if( strcmp(zAns,str.z)!=0 ){
	5887	+ nErr++;
	5888	+ rc = 1;
	5889	+ utf8_printf(p->out, "%d: Expected: [%s]\n", tno, zAns);
	5890	+ utf8_printf(p->out, "%d: Got: [%s]\n", tno, str.z);
	5891	+ }
	5892	+ }else
	5893	+ {
	5894	+ utf8_printf(stderr,
	5895	+ "Unknown operation \"%s\" on selftest line %d\n", zOp, tno);
	5896	+ rc = 1;
	5897	+ break;
	5898	+ }
	5899	+ }
	5900	+ freeText(&str);
	5901	+ if( azTest!=azDefaultTest ) sqlite3_free_table(azTest);
	5902	+ utf8_printf(p->out, "%d errors out of %d tests\n", nErr, nTest);
	5903	+ }else
4839	5904
4840	5905	if( c=='s' && strncmp(azArg[0], "separator", n)==0 ){
4841	5906	if( nArg<2 \|\| nArg>3 ){
4842	5907	raw_printf(stderr, "Usage: .separator COL ?ROW?\n");
4843	5908	rc = 1;
		@@ -4849,10 +5914,126 @@
4849	5914	if( nArg>=3 ){
4850	5915	sqlite3_snprintf(sizeof(p->rowSeparator), p->rowSeparator,
4851	5916	"%.*s", (int)ArraySize(p->rowSeparator)-1, azArg[2]);
4852	5917	}
4853	5918	}else
	5919	+
	5920	+ if( c=='s' && n>=4 && strncmp(azArg[0],"sha3sum",n)==0 ){
	5921	+ const char zLike = 0; / Which table to checksum. 0 means everything */
	5922	+ int i; /* Loop counter */
	5923	+ int bSchema = 0; /* Also hash the schema */
	5924	+ int bSeparate = 0; /* Hash each table separately */
	5925	+ int iSize = 224; /* Hash algorithm to use */
	5926	+ int bDebug = 0; /* Only show the query that would have run */
	5927	+ sqlite3_stmt pStmt; / For querying tables names */
	5928	+ char zSql; / SQL to be run */
	5929	+ char zSep; / Separator */
	5930	+ ShellText sSql; /* Complete SQL for the query to run the hash */
	5931	+ ShellText sQuery; /* Set of queries used to read all content */
	5932	+ open_db(p, 0);
	5933	+ for(i=1; i<nArg; i++){
	5934	+ const char *z = azArg[i];
	5935	+ if( z[0]=='-' ){
	5936	+ z++;
	5937	+ if( z[0]=='-' ) z++;
	5938	+ if( strcmp(z,"schema")==0 ){
	5939	+ bSchema = 1;
	5940	+ }else
	5941	+ if( strcmp(z,"sha3-224")==0 \|\| strcmp(z,"sha3-256")==0
	5942	+ \|\| strcmp(z,"sha3-384")==0 \|\| strcmp(z,"sha3-512")==0
	5943	+ ){
	5944	+ iSize = atoi(&z[5]);
	5945	+ }else
	5946	+ if( strcmp(z,"debug")==0 ){
	5947	+ bDebug = 1;
	5948	+ }else
	5949	+ {
	5950	+ utf8_printf(stderr, "Unknown option \"%s\" on \"%s\"\n",
	5951	+ azArg[i], azArg[0]);
	5952	+ raw_printf(stderr, "Should be one of: --schema"
	5953	+ " --sha3-224 --sha3-255 --sha3-384 --sha3-512\n");
	5954	+ rc = 1;
	5955	+ goto meta_command_exit;
	5956	+ }
	5957	+ }else if( zLike ){
	5958	+ raw_printf(stderr, "Usage: .sha3sum ?OPTIONS? ?LIKE-PATTERN?\n");
	5959	+ rc = 1;
	5960	+ goto meta_command_exit;
	5961	+ }else{
	5962	+ zLike = z;
	5963	+ bSeparate = 1;
	5964	+ if( sqlite3_strlike("sqlite_%", zLike, 0)==0 ) bSchema = 1;
	5965	+ }
	5966	+ }
	5967	+ if( bSchema ){
	5968	+ zSql = "SELECT lower(name) FROM sqlite_master"
	5969	+ " WHERE type='table' AND coalesce(rootpage,0)>1"
	5970	+ " UNION ALL SELECT 'sqlite_master'"
	5971	+ " ORDER BY 1 collate nocase";
	5972	+ }else{
	5973	+ zSql = "SELECT lower(name) FROM sqlite_master"
	5974	+ " WHERE type='table' AND coalesce(rootpage,0)>1"
	5975	+ " AND name NOT LIKE 'sqlite_%'"
	5976	+ " ORDER BY 1 collate nocase";
	5977	+ }
	5978	+ sqlite3_prepare_v2(p->db, zSql, -1, &pStmt, 0);
	5979	+ initText(&sQuery);
	5980	+ initText(&sSql);
	5981	+ appendText(&sSql, "WITH [sha3sum$query](a,b) AS(",0);
	5982	+ zSep = "VALUES(";
	5983	+ while( SQLITE_ROW==sqlite3_step(pStmt) ){
	5984	+ const char zTab = (const char)sqlite3_column_text(pStmt,0);
	5985	+ if( zLike && sqlite3_strlike(zLike, zTab, 0)!=0 ) continue;
	5986	+ if( strncmp(zTab, "sqlite_",7)!=0 ){
	5987	+ appendText(&sQuery,"SELECT * FROM ", 0);
	5988	+ appendText(&sQuery,zTab,'"');
	5989	+ appendText(&sQuery," NOT INDEXED;", 0);
	5990	+ }else if( strcmp(zTab, "sqlite_master")==0 ){
	5991	+ appendText(&sQuery,"SELECT type,name,tbl_name,sql FROM sqlite_master"
	5992	+ " ORDER BY name;", 0);
	5993	+ }else if( strcmp(zTab, "sqlite_sequence")==0 ){
	5994	+ appendText(&sQuery,"SELECT name,seq FROM sqlite_sequence"
	5995	+ " ORDER BY name;", 0);
	5996	+ }else if( strcmp(zTab, "sqlite_stat1")==0 ){
	5997	+ appendText(&sQuery,"SELECT tbl,idx,stat FROM sqlite_stat1"
	5998	+ " ORDER BY tbl,idx;", 0);
	5999	+ }else if( strcmp(zTab, "sqlite_stat3")==0
	6000	+ \|\| strcmp(zTab, "sqlite_stat4")==0 ){
	6001	+ appendText(&sQuery, "SELECT * FROM ", 0);
	6002	+ appendText(&sQuery, zTab, 0);
	6003	+ appendText(&sQuery, " ORDER BY tbl, idx, rowid;\n", 0);
	6004	+ }
	6005	+ appendText(&sSql, zSep, 0);
	6006	+ appendText(&sSql, sQuery.z, '\'');
	6007	+ sQuery.n = 0;
	6008	+ appendText(&sSql, ",", 0);
	6009	+ appendText(&sSql, zTab, '\'');
	6010	+ zSep = "),(";
	6011	+ }
	6012	+ sqlite3_finalize(pStmt);
	6013	+ if( bSeparate ){
	6014	+ zSql = sqlite3_mprintf(
	6015	+ "%s))"
	6016	+ " SELECT lower(hex(sha3_query(a,%d))) AS hash, b AS label"
	6017	+ " FROM [sha3sum$query]",
	6018	+ sSql.z, iSize);
	6019	+ }else{
	6020	+ zSql = sqlite3_mprintf(
	6021	+ "%s))"
	6022	+ " SELECT lower(hex(sha3_query(group_concat(a,''),%d))) AS hash"
	6023	+ " FROM [sha3sum$query]",
	6024	+ sSql.z, iSize);
	6025	+ }
	6026	+ freeText(&sQuery);
	6027	+ freeText(&sSql);
	6028	+ if( bDebug ){
	6029	+ utf8_printf(p->out, "%s\n", zSql);
	6030	+ }else{
	6031	+ shell_exec(p->db, zSql, shell_callback, p, 0);
	6032	+ }
	6033	+ sqlite3_free(zSql);
	6034	+ }else
4854	6035
4855	6036	if( c=='s'
4856	6037	&& (strncmp(azArg[0], "shell", n)==0 \|\| strncmp(azArg[0],"system",n)==0)
4857	6038	){
4858	6039	char *zCmd;
		@@ -4878,11 +6059,12 @@
4878	6059	if( nArg!=1 ){
4879	6060	raw_printf(stderr, "Usage: .show\n");
4880	6061	rc = 1;
4881	6062	goto meta_command_exit;
4882	6063	}
4883		- utf8_printf(p->out, "%12.12s: %s\n","echo", azBool[p->echoOn!=0]);
	6064	+ utf8_printf(p->out, "%12.12s: %s\n","echo",
	6065	+ azBool[ShellHasFlag(p, SHFLG_Echo)]);
4884	6066	utf8_printf(p->out, "%12.12s: %s\n","eqp", azBool[p->autoEQP&3]);
4885	6067	utf8_printf(p->out, "%12.12s: %s\n","explain",
4886	6068	p->mode==MODE_Explain ? "on" : p->autoExplain ? "auto" : "off");
4887	6069	utf8_printf(p->out,"%12.12s: %s\n","headers", azBool[p->showHeader!=0]);
4888	6070	utf8_printf(p->out, "%12.12s: %s\n","mode", modeDescr[p->mode]);
		@@ -5190,10 +6372,11 @@
5190	6372	azArg[1]);
5191	6373	break;
5192	6374	}
5193	6375	}
5194	6376	}else
	6377	+#endif /* !defined(SQLITE_UNTESTABLE) */
5195	6378
5196	6379	if( c=='t' && n>4 && strncmp(azArg[0], "timeout", n)==0 ){
5197	6380	open_db(p, 0);
5198	6381	sqlite3_busy_timeout(p->db, nArg>=2 ? (int)integerValue(azArg[1]) : 0);
5199	6382	}else
		@@ -5226,11 +6409,10 @@
5226	6409	}else{
5227	6410	sqlite3_trace_v2(p->db, SQLITE_TRACE_STMT, sql_trace_callback,p->traceOut);
5228	6411	}
5229	6412	#endif
5230	6413	}else
5231		-#endif /* !defined(SQLITE_UNTESTABLE) */
5232	6414
5233	6415	#if SQLITE_USER_AUTHENTICATION
5234	6416	if( c=='u' && strncmp(azArg[0], "user", n)==0 ){
5235	6417	if( nArg<2 ){
5236	6418	raw_printf(stderr, "Usage: .user SUBCOMMAND ...\n");
		@@ -5443,11 +6625,11 @@
5443	6625	static int runOneSqlLine(ShellState p, char zSql, FILE *in, int startline){
5444	6626	int rc;
5445	6627	char *zErrMsg = 0;
5446	6628
5447	6629	open_db(p, 0);
5448		- if( p->backslashOn ) resolve_backslashes(zSql);
	6630	+ if( ShellHasFlag(p,SHFLG_Backslash) ) resolve_backslashes(zSql);
5449	6631	BEGIN_TIMER;
5450	6632	rc = shell_exec(p->db, zSql, shell_callback, p, &zErrMsg);
5451	6633	END_TIMER;
5452	6634	if( rc \|\| zErrMsg ){
5453	6635	char zPrefix[100];
		@@ -5463,11 +6645,11 @@
5463	6645	zErrMsg = 0;
5464	6646	}else{
5465	6647	utf8_printf(stderr, "%s %s\n", zPrefix, sqlite3_errmsg(p->db));
5466	6648	}
5467	6649	return 1;
5468		- }else if( p->countChanges ){
	6650	+ }else if( ShellHasFlag(p, SHFLG_CountChanges) ){
5469	6651	raw_printf(p->out, "changes: %3d total_changes: %d\n",
5470	6652	sqlite3_changes(p->db), sqlite3_total_changes(p->db));
5471	6653	}
5472	6654	return 0;
5473	6655	}
		@@ -5506,15 +6688,15 @@
5506	6688	if( in!=0 ) break;
5507	6689	seenInterrupt = 0;
5508	6690	}
5509	6691	lineno++;
5510	6692	if( nSql==0 && _all_whitespace(zLine) ){
5511		- if( p->echoOn ) printf("%s\n", zLine);
	6693	+ if( ShellHasFlag(p, SHFLG_Echo) ) printf("%s\n", zLine);
5512	6694	continue;
5513	6695	}
5514	6696	if( zLine && zLine[0]=='.' && nSql==0 ){
5515		- if( p->echoOn ) printf("%s\n", zLine);
	6697	+ if( ShellHasFlag(p, SHFLG_Echo) ) printf("%s\n", zLine);
5516	6698	rc = do_meta_command(zLine, p);
5517	6699	if( rc==2 ){ /* exit requested */
5518	6700	break;
5519	6701	}else if( rc ){
5520	6702	errCnt++;
		@@ -5553,11 +6735,11 @@
5553	6735	if( p->outCount ){
5554	6736	output_reset(p);
5555	6737	p->outCount = 0;
5556	6738	}
5557	6739	}else if( nSql && _all_whitespace(zSql) ){
5558		- if( p->echoOn ) printf("%s\n", zSql);
	6740	+ if( ShellHasFlag(p, SHFLG_Echo) ) printf("%s\n", zSql);
5559	6741	nSql = 0;
5560	6742	}
5561	6743	}
5562	6744	if( nSql && !_all_whitespace(zSql) ){
5563	6745	runOneSqlLine(p, zSql, in, startline);
		@@ -6022,11 +7204,11 @@
6022	7204	}else if( strcmp(z,"-header")==0 ){
6023	7205	data.showHeader = 1;
6024	7206	}else if( strcmp(z,"-noheader")==0 ){
6025	7207	data.showHeader = 0;
6026	7208	}else if( strcmp(z,"-echo")==0 ){
6027		- data.echoOn = 1;
	7209	+ ShellSetFlag(&data, SHFLG_Echo);
6028	7210	}else if( strcmp(z,"-eqp")==0 ){
6029	7211	data.autoEQP = 1;
6030	7212	}else if( strcmp(z,"-eqpfull")==0 ){
6031	7213	data.autoEQP = 2;
6032	7214	}else if( strcmp(z,"-stats")==0 ){
		@@ -6037,11 +7219,11 @@
6037	7219	/* Undocumented command-line option: -backslash
6038	7220	** Causes C-style backslash escapes to be evaluated in SQL statements
6039	7221	** prior to sending the SQL into SQLite. Useful for injecting
6040	7222	** crazy bytes in the middle of SQL statements for testing and debugging.
6041	7223	*/
6042		- data.backslashOn = 1;
	7224	+ ShellSetFlag(&data, SHFLG_Backslash);
6043	7225	}else if( strcmp(z,"-bail")==0 ){
6044	7226	bail_on_error = 1;
6045	7227	}else if( strcmp(z,"-version")==0 ){
6046	7228	printf("%s %s\n", sqlite3_libversion(), sqlite3_sourceid());
6047	7229	return 0;
6048	7230

	--- src/shell.c
	+++ src/shell.c
	@@ -452,32 +452,10 @@
452	while( IsDigit(*z) ){ z++; }
453	if( realnum ) *realnum = 1;
454	}
455	return *z==0;
456	}
457
458	/*
459	** A global char* and an SQL function to access its current value
460	** from within an SQL statement. This program used to use the
461	** sqlite_exec_printf() API to substitue a string into an SQL statement.
462	** The correct way to do this with sqlite3 is to use the bind API, but
463	** since the shell is built around the callback paradigm it would be a lot
464	** of work. Instead just use this hack, which is quite harmless.
465	*/
466	static const char *zShellStatic = 0;
467	static void shellstaticFunc(
468	sqlite3_context *context,
469	int argc,
470	sqlite3_value **argv
471	){
472	assert( 0==argc );
473	assert( zShellStatic );
474	UNUSED_PARAMETER(argc);
475	UNUSED_PARAMETER(argv);
476	sqlite3_result_text(context, zShellStatic, -1, SQLITE_STATIC);
477	}
478
479
480	/*
481	** Compute a string length that is limited to what can be stored in
482	** lower 30 bits of a 32-bit signed integer.
483	*/
	@@ -575,10 +553,783 @@
575	if( zResult && *zResult ) shell_add_history(zResult);
576	#endif
577	}
578	return zResult;
579	}





































































































































































































































































































































































































































































































































































































































































































































































































580
581	#if defined(SQLITE_ENABLE_SESSION)
582	/*
583	** State information for a single open session
584	*/
	@@ -608,17 +1359,14 @@
608	** instance of the following structure.
609	*/
610	typedef struct ShellState ShellState;
611	struct ShellState {
612	sqlite3 db; / The database */
613	int echoOn; /* True to echo input commands */
614	int autoExplain; /* Automatically turn on .explain mode */
615	int autoEQP; /* Run EXPLAIN QUERY PLAN prior to seach SQL stmt */
616	int statsOn; /* True to display memory stats before each finalize */
617	int scanstatsOn; /* True to display scan stats before each finalize */
618	int countChanges; /* True to display change counts */
619	int backslashOn; /* Resolve C-style \x escapes in SQL input text */
620	int outCount; /* Revert to stdout when reaching zero */
621	int cnt; /* Number of records displayed so far */
622	FILE out; / Write results here */
623	FILE traceOut; / Output for sqlite3_trace() */
624	int nErr; /* Number of errors seen */
	@@ -653,13 +1401,24 @@
653	};
654
655	/*
656	** These are the allowed shellFlgs values
657	*/
658	#define SHFLG_Scratch 0x00001 /* The --scratch option is used */
659	#define SHFLG_Pagecache 0x00002 /* The --pagecache option is used */
660	#define SHFLG_Lookaside 0x00004 /* Lookaside memory is used */











661
662	/*
663	** These are the allowed modes.
664	*/
665	#define MODE_Line 0 /* One column per line. Blank line between records */
	@@ -1292,92 +2051,106 @@
1292	*/
1293	static int callback(void pArg, int nArg, char azArg, char *azCol){
1294	/* since we don't have type info, call the shell_callback with a NULL value */
1295	return shell_callback(pArg, nArg, azArg, azCol, NULL);
1296	}
































































1297
1298	/*
1299	** Set the destination table field of the ShellState structure to
1300	** the name of the table given. Escape any quote characters in the
1301	** table name.
1302	*/
1303	static void set_table_name(ShellState p, const char zName){
1304	int i, n;
1305	int needQuote;
1306	char *z;
1307
1308	if( p->zDestTable ){
1309	free(p->zDestTable);
1310	p->zDestTable = 0;
1311	}
1312	if( zName==0 ) return;
1313	needQuote = !isalpha((unsigned char)zName) && zName!='_';
1314	for(i=n=0; zName[i]; i++, n++){
1315	if( !isalnum((unsigned char)zName[i]) && zName[i]!='_' ){
1316	needQuote = 1;
1317	if( zName[i]=='\'' ) n++;
1318	}
1319	}
1320	if( needQuote ) n += 2;
1321	z = p->zDestTable = malloc( n+1 );
1322	if( z==0 ){
1323	raw_printf(stderr,"Error: out of memory\n");
1324	exit(1);
1325	}
1326	n = 0;
1327	if( needQuote ) z[n++] = '\'';
1328	for(i=0; zName[i]; i++){
1329	z[n++] = zName[i];
1330	if( zName[i]=='\'' ) z[n++] = '\'';
1331	}
1332	if( needQuote ) z[n++] = '\'';
1333	z[n] = 0;
1334	}
1335
1336	/* zIn is either a pointer to a NULL-terminated string in memory obtained
1337	** from malloc(), or a NULL pointer. The string pointed to by zAppend is
1338	** added to zIn, and the result returned in memory obtained from malloc().
1339	** zIn, if it was not NULL, is freed.
1340	**
1341	** If the third argument, quote, is not '\0', then it is used as a
1342	** quote character for zAppend.
1343	*/
1344	static char appendText(char zIn, char const *zAppend, char quote){
1345	int len;
1346	int i;
1347	int nAppend = strlen30(zAppend);
1348	int nIn = (zIn?strlen30(zIn):0);
1349
1350	len = nAppend+nIn+1;
1351	if( quote ){
1352	len += 2;
1353	for(i=0; i<nAppend; i++){
1354	if( zAppend[i]==quote ) len++;
1355	}
1356	}
1357
1358	zIn = (char *)realloc(zIn, len);
1359	if( !zIn ){
1360	return 0;
1361	}
1362
1363	if( quote ){
1364	char *zCsr = &zIn[nIn];
1365	*zCsr++ = quote;
1366	for(i=0; i<nAppend; i++){
1367	*zCsr++ = zAppend[i];
1368	if( zAppend[i]==quote ) *zCsr++ = quote;
1369	}
1370	*zCsr++ = quote;
1371	*zCsr++ = '\0';
1372	assert( (zCsr-zIn)==len );
1373	}else{
1374	memcpy(&zIn[nIn], zAppend, nAppend);
1375	zIn[len-1] = '\0';
1376	}
1377
1378	return zIn;
1379	}
1380
1381
1382	/*
1383	** Execute a query statement that will generate SQL output. Print
	@@ -1484,10 +2257,35 @@
1484	}
1485	fclose(in);
1486	}
1487	#endif
1488

























1489
1490	/*
1491	** Display memory stats.
1492	*/
1493	static int display_stats(
	@@ -1497,61 +2295,35 @@
1497	){
1498	int iCur;
1499	int iHiwtr;
1500
1501	if( pArg && pArg->out ){
1502
1503	iHiwtr = iCur = -1;
1504	sqlite3_status(SQLITE_STATUS_MEMORY_USED, &iCur, &iHiwtr, bReset);
1505	raw_printf(pArg->out,
1506	"Memory Used: %d (max %d) bytes\n",
1507	iCur, iHiwtr);
1508	iHiwtr = iCur = -1;
1509	sqlite3_status(SQLITE_STATUS_MALLOC_COUNT, &iCur, &iHiwtr, bReset);
1510	raw_printf(pArg->out, "Number of Outstanding Allocations: %d (max %d)\n",
1511	iCur, iHiwtr);
1512	if( pArg->shellFlgs & SHFLG_Pagecache ){
1513	iHiwtr = iCur = -1;
1514	sqlite3_status(SQLITE_STATUS_PAGECACHE_USED, &iCur, &iHiwtr, bReset);
1515	raw_printf(pArg->out,
1516	"Number of Pcache Pages Used: %d (max %d) pages\n",
1517	iCur, iHiwtr);
1518	}
1519	iHiwtr = iCur = -1;
1520	sqlite3_status(SQLITE_STATUS_PAGECACHE_OVERFLOW, &iCur, &iHiwtr, bReset);
1521	raw_printf(pArg->out,
1522	"Number of Pcache Overflow Bytes: %d (max %d) bytes\n",
1523	iCur, iHiwtr);
1524	if( pArg->shellFlgs & SHFLG_Scratch ){
1525	iHiwtr = iCur = -1;
1526	sqlite3_status(SQLITE_STATUS_SCRATCH_USED, &iCur, &iHiwtr, bReset);
1527	raw_printf(pArg->out,
1528	"Number of Scratch Allocations Used: %d (max %d)\n",
1529	iCur, iHiwtr);
1530	}
1531	iHiwtr = iCur = -1;
1532	sqlite3_status(SQLITE_STATUS_SCRATCH_OVERFLOW, &iCur, &iHiwtr, bReset);
1533	raw_printf(pArg->out,
1534	"Number of Scratch Overflow Bytes: %d (max %d) bytes\n",
1535	iCur, iHiwtr);
1536	iHiwtr = iCur = -1;
1537	sqlite3_status(SQLITE_STATUS_MALLOC_SIZE, &iCur, &iHiwtr, bReset);
1538	raw_printf(pArg->out, "Largest Allocation: %d bytes\n",
1539	iHiwtr);
1540	iHiwtr = iCur = -1;
1541	sqlite3_status(SQLITE_STATUS_PAGECACHE_SIZE, &iCur, &iHiwtr, bReset);
1542	raw_printf(pArg->out, "Largest Pcache Allocation: %d bytes\n",
1543	iHiwtr);
1544	iHiwtr = iCur = -1;
1545	sqlite3_status(SQLITE_STATUS_SCRATCH_SIZE, &iCur, &iHiwtr, bReset);
1546	raw_printf(pArg->out, "Largest Scratch Allocation: %d bytes\n",
1547	iHiwtr);
1548	#ifdef YYTRACKMAXSTACKDEPTH
1549	iHiwtr = iCur = -1;
1550	sqlite3_status(SQLITE_STATUS_PARSER_STACK, &iCur, &iHiwtr, bReset);
1551	raw_printf(pArg->out, "Deepest Parser Stack: %d (max %d)\n",
1552	iCur, iHiwtr);
1553	#endif
1554	}
1555
1556	if( pArg && pArg->out && db ){
1557	if( pArg->shellFlgs & SHFLG_Lookaside ){
	@@ -1934,11 +2706,11 @@
1934	pArg->pStmt = pStmt;
1935	pArg->cnt = 0;
1936	}
1937
1938	/* echo the sql statement if echo on */
1939	if( pArg && pArg->echoOn ){
1940	utf8_printf(pArg->out, "%s\n", zStmtSql ? zStmtSql : zSql);
1941	}
1942
1943	/* Show the EXPLAIN QUERY PLAN if .eqp is on */
1944	if( pArg && pArg->autoEQP && sqlite3_strlike("EXPLAIN%",zStmtSql,0)!=0 ){
	@@ -2022,10 +2794,142 @@
2022	} /* end while */
2023
2024	return rc;
2025	}
2026




































































































































2027
2028	/*
2029	** This is a different callback routine used for dumping the database.
2030	** Each row received by this callback consists of a table name,
2031	** the table type ("index" or "table") and SQL to create the table.
	@@ -2034,21 +2938,20 @@
2034	static int dump_callback(void pArg, int nArg, char azArg, char *azCol){
2035	int rc;
2036	const char *zTable;
2037	const char *zType;
2038	const char *zSql;
2039	const char *zPrepStmt = 0;
2040	ShellState p = (ShellState )pArg;
2041
2042	UNUSED_PARAMETER(azCol);
2043	if( nArg!=3 ) return 1;
2044	zTable = azArg[0];
2045	zType = azArg[1];
2046	zSql = azArg[2];
2047
2048	if( strcmp(zTable, "sqlite_sequence")==0 ){
2049	zPrepStmt = "DELETE FROM sqlite_sequence;\n";
2050	}else if( sqlite3_strglob("sqlite_stat?", zTable)==0 ){
2051	raw_printf(p->out, "ANALYZE sqlite_master;\n");
2052	}else if( strncmp(zTable, "sqlite_", 7)==0 ){
2053	return 0;
2054	}else if( strncmp(zSql, "CREATE VIRTUAL TABLE", 20)==0 ){
	@@ -2067,62 +2970,74 @@
2067	}else{
2068	printSchemaLine(p->out, zSql, ";\n");
2069	}
2070
2071	if( strcmp(zType, "table")==0 ){
2072	sqlite3_stmt *pTableInfo = 0;
2073	char *zSelect = 0;
2074	char *zTableInfo = 0;
2075	char *zTmp = 0;
2076	int nRow = 0;
2077
2078	zTableInfo = appendText(zTableInfo, "PRAGMA table_info(", 0);
2079	zTableInfo = appendText(zTableInfo, zTable, '"');
2080	zTableInfo = appendText(zTableInfo, ");", 0);
2081
2082	rc = sqlite3_prepare_v2(p->db, zTableInfo, -1, &pTableInfo, 0);
2083	free(zTableInfo);
2084	if( rc!=SQLITE_OK \|\| !pTableInfo ){
2085	return 1;
2086	}
2087
2088	zSelect = appendText(zSelect, "SELECT 'INSERT INTO ' \|\| ", 0);
2089	/* Always quote the table name, even if it appears to be pure ascii,
2090	** in case it is a keyword. Ex: INSERT INTO "table" ... */
2091	zTmp = appendText(zTmp, zTable, '"');
2092	if( zTmp ){
2093	zSelect = appendText(zSelect, zTmp, '\'');
2094	free(zTmp);
2095	}
2096	zSelect = appendText(zSelect, " \|\| ' VALUES(' \|\| ", 0);
2097	rc = sqlite3_step(pTableInfo);
2098	while( rc==SQLITE_ROW ){
2099	const char zText = (const char )sqlite3_column_text(pTableInfo, 1);
2100	zSelect = appendText(zSelect, "quote(", 0);
2101	zSelect = appendText(zSelect, zText, '"');
2102	rc = sqlite3_step(pTableInfo);
2103	if( rc==SQLITE_ROW ){
2104	zSelect = appendText(zSelect, "), ", 0);
2105	}else{
2106	zSelect = appendText(zSelect, ") ", 0);
2107	}
2108	nRow++;
2109	}
2110	rc = sqlite3_finalize(pTableInfo);
2111	if( rc!=SQLITE_OK \|\| nRow==0 ){
2112	free(zSelect);
2113	return 1;
2114	}
2115	zSelect = appendText(zSelect, "\|\| ')' FROM ", 0);
2116	zSelect = appendText(zSelect, zTable, '"');
2117
2118	rc = run_table_dump_query(p, zSelect, zPrepStmt);
2119	if( rc==SQLITE_CORRUPT ){
2120	zSelect = appendText(zSelect, " ORDER BY rowid DESC", 0);
2121	run_table_dump_query(p, zSelect, 0);
2122	}
2123	free(zSelect);
















2124	}
2125	return 0;
2126	}
2127
2128	/*
	@@ -2210,11 +3125,11 @@
2210	" csv Comma-separated values\n"
2211	" column Left-aligned columns. (See .width)\n"
2212	" html HTML <table> code\n"
2213	" insert SQL insert statements for TABLE\n"
2214	" line One value per line\n"
2215	" list Values delimited by .separator strings\n"
2216	" quote Escape answers as for SQL\n"
2217	" tabs Tab-separated values\n"
2218	" tcl TCL list elements\n"
2219	".nullvalue STRING Use STRING in place of NULL values\n"
2220	".once FILENAME Output for the next SQL command only to FILENAME\n"
	@@ -2233,10 +3148,11 @@
2233	".separator COL ?ROW? Change the column separator and optionally the row\n"
2234	" separator for both the output mode and .import\n"
2235	#if defined(SQLITE_ENABLE_SESSION)
2236	".session CMD ... Create or control sessions\n"
2237	#endif

2238	".shell CMD ARGS... Run CMD ARGS... in a system shell\n"
2239	".show Show the current values for various settings\n"
2240	".stats ?on\|off? Show stats or turn stats on or off\n"
2241	".system CMD ARGS... Run CMD ARGS... in a system shell\n"
2242	".tables ?TABLE? List names of tables\n"
	@@ -2423,14 +3339,10 @@
2423	static void open_db(ShellState *p, int keepAlive){
2424	if( p->db==0 ){
2425	sqlite3_initialize();
2426	sqlite3_open(p->zDbFilename, &p->db);
2427	globalDb = p->db;
2428	if( p->db && sqlite3_errcode(p->db)==SQLITE_OK ){
2429	sqlite3_create_function(p->db, "shellstatic", 0, SQLITE_UTF8, 0,
2430	shellstaticFunc, 0, 0);
2431	}
2432	if( p->db==0 \|\| SQLITE_OK!=sqlite3_errcode(p->db) ){
2433	utf8_printf(stderr,"Error: unable to open database \"%s\": %s\n",
2434	p->zDbFilename, sqlite3_errmsg(p->db));
2435	if( keepAlive ) return;
2436	exit(1);
	@@ -2440,10 +3352,18 @@
2440	#endif
2441	sqlite3_create_function(p->db, "readfile", 1, SQLITE_UTF8, 0,
2442	readfileFunc, 0, 0);
2443	sqlite3_create_function(p->db, "writefile", 2, SQLITE_UTF8, 0,
2444	writefileFunc, 0, 0);








2445	}
2446	}
2447
2448	/*
2449	** Do C-language style dequoting.
	@@ -2564,11 +3484,11 @@
2564
2565	/*
2566	** Interpret zArg as either an integer or a boolean value. Return 1 or 0
2567	** for TRUE and FALSE. Return the integer value if appropriate.
2568	*/
2569	static int booleanValue(char *zArg){
2570	int i;
2571	if( zArg[0]=='0' && zArg[1]=='x' ){
2572	for(i=2; hexDigitValue(zArg[i])>=0; i++){}
2573	}else{
2574	for(i=0; zArg[i]>='0' && zArg[i]<='9'; i++){}
	@@ -2582,10 +3502,21 @@
2582	}
2583	utf8_printf(stderr, "ERROR: Not a boolean value: \"%s\". Assuming \"no\".\n",
2584	zArg);
2585	return 0;
2586	}











2587
2588	/*
2589	** Close an output file, assuming it is not stderr or stdout
2590	*/
2591	static void output_file_close(FILE *f){
	@@ -3656,11 +4587,11 @@
3656	test_breakpoint();
3657	}else
3658
3659	if( c=='c' && n>=3 && strncmp(azArg[0], "changes", n)==0 ){
3660	if( nArg==2 ){
3661	p->countChanges = booleanValue(azArg[1]);
3662	}else{
3663	raw_printf(stderr, "Usage: .changes on\|off\n");
3664	rc = 1;
3665	}
3666	}else
	@@ -3720,25 +4651,46 @@
3720	if( c=='d' && strncmp(azArg[0], "dbinfo", n)==0 ){
3721	rc = shell_dbinfo_command(p, nArg, azArg);
3722	}else
3723
3724	if( c=='d' && strncmp(azArg[0], "dump", n)==0 ){























3725	open_db(p, 0);
3726	/* When playing back a "dump", the content might appear in an order
3727	** which causes immediate foreign key constraints to be violated.
3728	** So disable foreign-key constraint enforcement to prevent problems. */
3729	if( nArg!=1 && nArg!=2 ){
3730	raw_printf(stderr, "Usage: .dump ?LIKE-PATTERN?\n");
3731	rc = 1;
3732	goto meta_command_exit;
3733	}
3734	raw_printf(p->out, "PRAGMA foreign_keys=OFF;\n");
3735	raw_printf(p->out, "BEGIN TRANSACTION;\n");
3736	p->writableSchema = 0;



3737	sqlite3_exec(p->db, "SAVEPOINT dump; PRAGMA writable_schema=ON", 0, 0, 0);
3738	p->nErr = 0;
3739	if( nArg==1 ){
3740	run_schema_dump_query(p,
3741	"SELECT name, type, sql FROM sqlite_master "
3742	"WHERE sql NOT NULL AND type=='table' AND name!='sqlite_sequence'"
3743	);
3744	run_schema_dump_query(p,
	@@ -3748,25 +4700,24 @@
3748	run_table_dump_query(p,
3749	"SELECT sql FROM sqlite_master "
3750	"WHERE sql NOT NULL AND type IN ('index','trigger','view')", 0
3751	);
3752	}else{
3753	int i;
3754	for(i=1; i<nArg; i++){
3755	zShellStatic = azArg[i];
3756	run_schema_dump_query(p,
3757	"SELECT name, type, sql FROM sqlite_master "
3758	"WHERE tbl_name LIKE shellstatic() AND type=='table'"
3759	" AND sql NOT NULL");
3760	run_table_dump_query(p,
3761	"SELECT sql FROM sqlite_master "
3762	"WHERE sql NOT NULL"
3763	" AND type IN ('index','trigger','view')"
3764	" AND tbl_name LIKE shellstatic()", 0
3765	);
3766	zShellStatic = 0;
3767	}
3768	}
3769	if( p->writableSchema ){
3770	raw_printf(p->out, "PRAGMA writable_schema=OFF;\n");
3771	p->writableSchema = 0;
3772	}
	@@ -3775,11 +4726,11 @@
3775	raw_printf(p->out, p->nErr ? "ROLLBACK; -- due to errors\n" : "COMMIT;\n");
3776	}else
3777
3778	if( c=='e' && strncmp(azArg[0], "echo", n)==0 ){
3779	if( nArg==2 ){
3780	p->echoOn = booleanValue(azArg[1]);
3781	}else{
3782	raw_printf(stderr, "Usage: .echo on\|off\n");
3783	rc = 1;
3784	}
3785	}else
	@@ -4342,10 +5293,11 @@
4342	int newFlag = 0; /* True to delete file before opening */
4343	/* Close the existing database */
4344	session_close_all(p);
4345	sqlite3_close(p->db);
4346	p->db = 0;

4347	sqlite3_free(p->zFreeOnClose);
4348	p->zFreeOnClose = 0;
4349	/* Check for command-line arguments */
4350	for(iName=1; iName<nArg && azArg[iName][0]=='-'; iName++){
4351	const char *z = azArg[iName];
	@@ -4571,21 +5523,21 @@
4571	new_colv[0] = "sql";
4572	new_colv[1] = 0;
4573	callback(&data, 1, new_argv, new_colv);
4574	rc = SQLITE_OK;
4575	}else{
4576	zShellStatic = azArg[1];
4577	rc = sqlite3_exec(p->db,
4578	"SELECT sql FROM "
4579	" (SELECT sql sql, type type, tbl_name tbl_name, name name, rowid x"
4580	" FROM sqlite_master UNION ALL"
4581	" SELECT sql, type, tbl_name, name, rowid FROM sqlite_temp_master) "
4582	"WHERE lower(tbl_name) LIKE shellstatic()"
4583	" AND type!='meta' AND sql NOTNULL "
4584	"ORDER BY rowid",
4585	callback, &data, &zErrMsg);
4586	zShellStatic = 0;
4587	}
4588	}else if( nArg==1 ){
4589	rc = sqlite3_exec(p->db,
4590	"SELECT sql FROM "
4591	" (SELECT sql sql, type type, tbl_name tbl_name, name name, rowid x"
	@@ -4612,11 +5564,11 @@
4612	}
4613	}else
4614
4615	#if defined(SQLITE_DEBUG) && defined(SQLITE_ENABLE_SELECTTRACE)
4616	if( c=='s' && n==11 && strncmp(azArg[0], "selecttrace", n)==0 ){
4617	sqlite3SelectTrace = integerValue(azArg[1]);
4618	}else
4619	#endif
4620
4621	#if defined(SQLITE_ENABLE_SESSION)
4622	if( c=='s' && strncmp(azArg[0],"session",n)==0 && n>=3 ){
	@@ -4834,10 +5786,123 @@
4834	utf8_printf(p->out, "%s", zBuf);
4835	}
4836	}
4837	}else
4838	#endif

















































































































4839
4840	if( c=='s' && strncmp(azArg[0], "separator", n)==0 ){
4841	if( nArg<2 \|\| nArg>3 ){
4842	raw_printf(stderr, "Usage: .separator COL ?ROW?\n");
4843	rc = 1;
	@@ -4849,10 +5914,126 @@
4849	if( nArg>=3 ){
4850	sqlite3_snprintf(sizeof(p->rowSeparator), p->rowSeparator,
4851	"%.*s", (int)ArraySize(p->rowSeparator)-1, azArg[2]);
4852	}
4853	}else




















































































































4854
4855	if( c=='s'
4856	&& (strncmp(azArg[0], "shell", n)==0 \|\| strncmp(azArg[0],"system",n)==0)
4857	){
4858	char *zCmd;
	@@ -4878,11 +6059,12 @@
4878	if( nArg!=1 ){
4879	raw_printf(stderr, "Usage: .show\n");
4880	rc = 1;
4881	goto meta_command_exit;
4882	}
4883	utf8_printf(p->out, "%12.12s: %s\n","echo", azBool[p->echoOn!=0]);

4884	utf8_printf(p->out, "%12.12s: %s\n","eqp", azBool[p->autoEQP&3]);
4885	utf8_printf(p->out, "%12.12s: %s\n","explain",
4886	p->mode==MODE_Explain ? "on" : p->autoExplain ? "auto" : "off");
4887	utf8_printf(p->out,"%12.12s: %s\n","headers", azBool[p->showHeader!=0]);
4888	utf8_printf(p->out, "%12.12s: %s\n","mode", modeDescr[p->mode]);
	@@ -5190,10 +6372,11 @@
5190	azArg[1]);
5191	break;
5192	}
5193	}
5194	}else

5195
5196	if( c=='t' && n>4 && strncmp(azArg[0], "timeout", n)==0 ){
5197	open_db(p, 0);
5198	sqlite3_busy_timeout(p->db, nArg>=2 ? (int)integerValue(azArg[1]) : 0);
5199	}else
	@@ -5226,11 +6409,10 @@
5226	}else{
5227	sqlite3_trace_v2(p->db, SQLITE_TRACE_STMT, sql_trace_callback,p->traceOut);
5228	}
5229	#endif
5230	}else
5231	#endif /* !defined(SQLITE_UNTESTABLE) */
5232
5233	#if SQLITE_USER_AUTHENTICATION
5234	if( c=='u' && strncmp(azArg[0], "user", n)==0 ){
5235	if( nArg<2 ){
5236	raw_printf(stderr, "Usage: .user SUBCOMMAND ...\n");
	@@ -5443,11 +6625,11 @@
5443	static int runOneSqlLine(ShellState p, char zSql, FILE *in, int startline){
5444	int rc;
5445	char *zErrMsg = 0;
5446
5447	open_db(p, 0);
5448	if( p->backslashOn ) resolve_backslashes(zSql);
5449	BEGIN_TIMER;
5450	rc = shell_exec(p->db, zSql, shell_callback, p, &zErrMsg);
5451	END_TIMER;
5452	if( rc \|\| zErrMsg ){
5453	char zPrefix[100];
	@@ -5463,11 +6645,11 @@
5463	zErrMsg = 0;
5464	}else{
5465	utf8_printf(stderr, "%s %s\n", zPrefix, sqlite3_errmsg(p->db));
5466	}
5467	return 1;
5468	}else if( p->countChanges ){
5469	raw_printf(p->out, "changes: %3d total_changes: %d\n",
5470	sqlite3_changes(p->db), sqlite3_total_changes(p->db));
5471	}
5472	return 0;
5473	}
	@@ -5506,15 +6688,15 @@
5506	if( in!=0 ) break;
5507	seenInterrupt = 0;
5508	}
5509	lineno++;
5510	if( nSql==0 && _all_whitespace(zLine) ){
5511	if( p->echoOn ) printf("%s\n", zLine);
5512	continue;
5513	}
5514	if( zLine && zLine[0]=='.' && nSql==0 ){
5515	if( p->echoOn ) printf("%s\n", zLine);
5516	rc = do_meta_command(zLine, p);
5517	if( rc==2 ){ /* exit requested */
5518	break;
5519	}else if( rc ){
5520	errCnt++;
	@@ -5553,11 +6735,11 @@
5553	if( p->outCount ){
5554	output_reset(p);
5555	p->outCount = 0;
5556	}
5557	}else if( nSql && _all_whitespace(zSql) ){
5558	if( p->echoOn ) printf("%s\n", zSql);
5559	nSql = 0;
5560	}
5561	}
5562	if( nSql && !_all_whitespace(zSql) ){
5563	runOneSqlLine(p, zSql, in, startline);
	@@ -6022,11 +7204,11 @@
6022	}else if( strcmp(z,"-header")==0 ){
6023	data.showHeader = 1;
6024	}else if( strcmp(z,"-noheader")==0 ){
6025	data.showHeader = 0;
6026	}else if( strcmp(z,"-echo")==0 ){
6027	data.echoOn = 1;
6028	}else if( strcmp(z,"-eqp")==0 ){
6029	data.autoEQP = 1;
6030	}else if( strcmp(z,"-eqpfull")==0 ){
6031	data.autoEQP = 2;
6032	}else if( strcmp(z,"-stats")==0 ){
	@@ -6037,11 +7219,11 @@
6037	/* Undocumented command-line option: -backslash
6038	** Causes C-style backslash escapes to be evaluated in SQL statements
6039	** prior to sending the SQL into SQLite. Useful for injecting
6040	** crazy bytes in the middle of SQL statements for testing and debugging.
6041	*/
6042	data.backslashOn = 1;
6043	}else if( strcmp(z,"-bail")==0 ){
6044	bail_on_error = 1;
6045	}else if( strcmp(z,"-version")==0 ){
6046	printf("%s %s\n", sqlite3_libversion(), sqlite3_sourceid());
6047	return 0;
6048

	--- src/shell.c
	+++ src/shell.c
	@@ -452,32 +452,10 @@
452	while( IsDigit(*z) ){ z++; }
453	if( realnum ) *realnum = 1;
454	}
455	return *z==0;
456	}






















457
458	/*
459	** Compute a string length that is limited to what can be stored in
460	** lower 30 bits of a 32-bit signed integer.
461	*/
	@@ -575,10 +553,783 @@
553	if( zResult && *zResult ) shell_add_history(zResult);
554	#endif
555	}
556	return zResult;
557	}
558	/*
559	** A variable length string to which one can append text.
560	*/
561	typedef struct ShellText ShellText;
562	struct ShellText {
563	char *z;
564	int n;
565	int nAlloc;
566	};
567
568	/*
569	** Initialize and destroy a ShellText object
570	*/
571	static void initText(ShellText *p){
572	memset(p, 0, sizeof(*p));
573	}
574	static void freeText(ShellText *p){
575	free(p->z);
576	initText(p);
577	}
578
579	/* zIn is either a pointer to a NULL-terminated string in memory obtained
580	** from malloc(), or a NULL pointer. The string pointed to by zAppend is
581	** added to zIn, and the result returned in memory obtained from malloc().
582	** zIn, if it was not NULL, is freed.
583	**
584	** If the third argument, quote, is not '\0', then it is used as a
585	** quote character for zAppend.
586	*/
587	static void appendText(ShellText p, char const zAppend, char quote){
588	int len;
589	int i;
590	int nAppend = strlen30(zAppend);
591
592	len = nAppend+p->n+1;
593	if( quote ){
594	len += 2;
595	for(i=0; i<nAppend; i++){
596	if( zAppend[i]==quote ) len++;
597	}
598	}
599
600	if( p->n+len>=p->nAlloc ){
601	p->nAlloc = p->nAlloc*2 + len + 20;
602	p->z = realloc(p->z, p->nAlloc);
603	if( p->z==0 ){
604	memset(p, 0, sizeof(*p));
605	return;
606	}
607	}
608
609	if( quote ){
610	char *zCsr = p->z+p->n;
611	*zCsr++ = quote;
612	for(i=0; i<nAppend; i++){
613	*zCsr++ = zAppend[i];
614	if( zAppend[i]==quote ) *zCsr++ = quote;
615	}
616	*zCsr++ = quote;
617	p->n = (int)(zCsr - p->z);
618	*zCsr = '\0';
619	}else{
620	memcpy(p->z+p->n, zAppend, nAppend);
621	p->n += nAppend;
622	p->z[p->n] = '\0';
623	}
624	}
625
626	/*
627	** Attempt to determine if identifier zName needs to be quoted, either
628	** because it contains non-alphanumeric characters, or because it is an
629	** SQLite keyword. Be conservative in this estimate: When in doubt assume
630	** that quoting is required.
631	**
632	** Return '"' if quoting is required. Return 0 if no quoting is required.
633	*/
634	static char quoteChar(const char *zName){
635	/* All SQLite keywords, in alphabetical order */
636	static const char *azKeywords[] = {
637	"ABORT", "ACTION", "ADD", "AFTER", "ALL", "ALTER", "ANALYZE", "AND", "AS",
638	"ASC", "ATTACH", "AUTOINCREMENT", "BEFORE", "BEGIN", "BETWEEN", "BY",
639	"CASCADE", "CASE", "CAST", "CHECK", "COLLATE", "COLUMN", "COMMIT",
640	"CONFLICT", "CONSTRAINT", "CREATE", "CROSS", "CURRENT_DATE",
641	"CURRENT_TIME", "CURRENT_TIMESTAMP", "DATABASE", "DEFAULT", "DEFERRABLE",
642	"DEFERRED", "DELETE", "DESC", "DETACH", "DISTINCT", "DROP", "EACH",
643	"ELSE", "END", "ESCAPE", "EXCEPT", "EXCLUSIVE", "EXISTS", "EXPLAIN",
644	"FAIL", "FOR", "FOREIGN", "FROM", "FULL", "GLOB", "GROUP", "HAVING", "IF",
645	"IGNORE", "IMMEDIATE", "IN", "INDEX", "INDEXED", "INITIALLY", "INNER",
646	"INSERT", "INSTEAD", "INTERSECT", "INTO", "IS", "ISNULL", "JOIN", "KEY",
647	"LEFT", "LIKE", "LIMIT", "MATCH", "NATURAL", "NO", "NOT", "NOTNULL",
648	"NULL", "OF", "OFFSET", "ON", "OR", "ORDER", "OUTER", "PLAN", "PRAGMA",
649	"PRIMARY", "QUERY", "RAISE", "RECURSIVE", "REFERENCES", "REGEXP",
650	"REINDEX", "RELEASE", "RENAME", "REPLACE", "RESTRICT", "RIGHT",
651	"ROLLBACK", "ROW", "SAVEPOINT", "SELECT", "SET", "TABLE", "TEMP",
652	"TEMPORARY", "THEN", "TO", "TRANSACTION", "TRIGGER", "UNION", "UNIQUE",
653	"UPDATE", "USING", "VACUUM", "VALUES", "VIEW", "VIRTUAL", "WHEN", "WHERE",
654	"WITH", "WITHOUT",
655	};
656	int i, lwr, upr, mid, c;
657	if( !isalpha((unsigned char)zName[0]) && zName[0]!='_' ) return '"';
658	for(i=0; zName[i]; i++){
659	if( !isalnum((unsigned char)zName[i]) && zName[i]!='_' ) return '"';
660	}
661	lwr = 0;
662	upr = sizeof(azKeywords)/sizeof(azKeywords[0]) - 1;
663	while( lwr<=upr ){
664	mid = (lwr+upr)/2;
665	c = sqlite3_stricmp(azKeywords[mid], zName);
666	if( c==0 ) return '"';
667	if( c<0 ){
668	lwr = mid+1;
669	}else{
670	upr = mid-1;
671	}
672	}
673	return 0;
674	}
675
676	/******************************************************************************
677	** SHA3 hash implementation copied from ../ext/misc/shathree.c
678	*/
679	typedef sqlite3_uint64 u64;
680	/*
681	** Macros to determine whether the machine is big or little endian,
682	** and whether or not that determination is run-time or compile-time.
683	**
684	** For best performance, an attempt is made to guess at the byte-order
685	** using C-preprocessor macros. If that is unsuccessful, or if
686	** -DSHA3_BYTEORDER=0 is set, then byte-order is determined
687	** at run-time.
688	*/
689	#ifndef SHA3_BYTEORDER
690	# if defined(i386) \|\| defined(__i386__) \|\| defined(_M_IX86) \|\| \
691	defined(__x86_64) \|\| defined(__x86_64__) \|\| defined(_M_X64) \|\| \
692	defined(_M_AMD64) \|\| defined(_M_ARM) \|\| defined(__x86) \|\| \
693	defined(__arm__)
694	# define SHA3_BYTEORDER 1234
695	# elif defined(sparc) \|\| defined(__ppc__)
696	# define SHA3_BYTEORDER 4321
697	# else
698	# define SHA3_BYTEORDER 0
699	# endif
700	#endif
701
702
703	/*
704	** State structure for a SHA3 hash in progress
705	*/
706	typedef struct SHA3Context SHA3Context;
707	struct SHA3Context {
708	union {
709	u64 s[25]; /* Keccak state. 5x5 lines of 64 bits each */
710	unsigned char x[1600]; /* ... or 1600 bytes */
711	} u;
712	unsigned nRate; /* Bytes of input accepted per Keccak iteration */
713	unsigned nLoaded; /* Input bytes loaded into u.x[] so far this cycle */
714	unsigned ixMask; /* Insert next input into u.x[nLoaded^ixMask]. */
715	};
716
717	/*
718	** A single step of the Keccak mixing function for a 1600-bit state
719	*/
720	static void KeccakF1600Step(SHA3Context *p){
721	int i;
722	u64 B0, B1, B2, B3, B4;
723	u64 C0, C1, C2, C3, C4;
724	u64 D0, D1, D2, D3, D4;
725	static const u64 RC[] = {
726	0x0000000000000001ULL, 0x0000000000008082ULL,
727	0x800000000000808aULL, 0x8000000080008000ULL,
728	0x000000000000808bULL, 0x0000000080000001ULL,
729	0x8000000080008081ULL, 0x8000000000008009ULL,
730	0x000000000000008aULL, 0x0000000000000088ULL,
731	0x0000000080008009ULL, 0x000000008000000aULL,
732	0x000000008000808bULL, 0x800000000000008bULL,
733	0x8000000000008089ULL, 0x8000000000008003ULL,
734	0x8000000000008002ULL, 0x8000000000000080ULL,
735	0x000000000000800aULL, 0x800000008000000aULL,
736	0x8000000080008081ULL, 0x8000000000008080ULL,
737	0x0000000080000001ULL, 0x8000000080008008ULL
738	};
739	# define A00 (p->u.s[0])
740	# define A01 (p->u.s[1])
741	# define A02 (p->u.s[2])
742	# define A03 (p->u.s[3])
743	# define A04 (p->u.s[4])
744	# define A10 (p->u.s[5])
745	# define A11 (p->u.s[6])
746	# define A12 (p->u.s[7])
747	# define A13 (p->u.s[8])
748	# define A14 (p->u.s[9])
749	# define A20 (p->u.s[10])
750	# define A21 (p->u.s[11])
751	# define A22 (p->u.s[12])
752	# define A23 (p->u.s[13])
753	# define A24 (p->u.s[14])
754	# define A30 (p->u.s[15])
755	# define A31 (p->u.s[16])
756	# define A32 (p->u.s[17])
757	# define A33 (p->u.s[18])
758	# define A34 (p->u.s[19])
759	# define A40 (p->u.s[20])
760	# define A41 (p->u.s[21])
761	# define A42 (p->u.s[22])
762	# define A43 (p->u.s[23])
763	# define A44 (p->u.s[24])
764	# define ROL64(a,x) ((a<<x)\|(a>>(64-x)))
765
766	for(i=0; i<24; i+=4){
767	C0 = A00^A10^A20^A30^A40;
768	C1 = A01^A11^A21^A31^A41;
769	C2 = A02^A12^A22^A32^A42;
770	C3 = A03^A13^A23^A33^A43;
771	C4 = A04^A14^A24^A34^A44;
772	D0 = C4^ROL64(C1, 1);
773	D1 = C0^ROL64(C2, 1);
774	D2 = C1^ROL64(C3, 1);
775	D3 = C2^ROL64(C4, 1);
776	D4 = C3^ROL64(C0, 1);
777
778	B0 = (A00^D0);
779	B1 = ROL64((A11^D1), 44);
780	B2 = ROL64((A22^D2), 43);
781	B3 = ROL64((A33^D3), 21);
782	B4 = ROL64((A44^D4), 14);
783	A00 = B0 ^((~B1)& B2 );
784	A00 ^= RC[i];
785	A11 = B1 ^((~B2)& B3 );
786	A22 = B2 ^((~B3)& B4 );
787	A33 = B3 ^((~B4)& B0 );
788	A44 = B4 ^((~B0)& B1 );
789
790	B2 = ROL64((A20^D0), 3);
791	B3 = ROL64((A31^D1), 45);
792	B4 = ROL64((A42^D2), 61);
793	B0 = ROL64((A03^D3), 28);
794	B1 = ROL64((A14^D4), 20);
795	A20 = B0 ^((~B1)& B2 );
796	A31 = B1 ^((~B2)& B3 );
797	A42 = B2 ^((~B3)& B4 );
798	A03 = B3 ^((~B4)& B0 );
799	A14 = B4 ^((~B0)& B1 );
800
801	B4 = ROL64((A40^D0), 18);
802	B0 = ROL64((A01^D1), 1);
803	B1 = ROL64((A12^D2), 6);
804	B2 = ROL64((A23^D3), 25);
805	B3 = ROL64((A34^D4), 8);
806	A40 = B0 ^((~B1)& B2 );
807	A01 = B1 ^((~B2)& B3 );
808	A12 = B2 ^((~B3)& B4 );
809	A23 = B3 ^((~B4)& B0 );
810	A34 = B4 ^((~B0)& B1 );
811
812	B1 = ROL64((A10^D0), 36);
813	B2 = ROL64((A21^D1), 10);
814	B3 = ROL64((A32^D2), 15);
815	B4 = ROL64((A43^D3), 56);
816	B0 = ROL64((A04^D4), 27);
817	A10 = B0 ^((~B1)& B2 );
818	A21 = B1 ^((~B2)& B3 );
819	A32 = B2 ^((~B3)& B4 );
820	A43 = B3 ^((~B4)& B0 );
821	A04 = B4 ^((~B0)& B1 );
822
823	B3 = ROL64((A30^D0), 41);
824	B4 = ROL64((A41^D1), 2);
825	B0 = ROL64((A02^D2), 62);
826	B1 = ROL64((A13^D3), 55);
827	B2 = ROL64((A24^D4), 39);
828	A30 = B0 ^((~B1)& B2 );
829	A41 = B1 ^((~B2)& B3 );
830	A02 = B2 ^((~B3)& B4 );
831	A13 = B3 ^((~B4)& B0 );
832	A24 = B4 ^((~B0)& B1 );
833
834	C0 = A00^A20^A40^A10^A30;
835	C1 = A11^A31^A01^A21^A41;
836	C2 = A22^A42^A12^A32^A02;
837	C3 = A33^A03^A23^A43^A13;
838	C4 = A44^A14^A34^A04^A24;
839	D0 = C4^ROL64(C1, 1);
840	D1 = C0^ROL64(C2, 1);
841	D2 = C1^ROL64(C3, 1);
842	D3 = C2^ROL64(C4, 1);
843	D4 = C3^ROL64(C0, 1);
844
845	B0 = (A00^D0);
846	B1 = ROL64((A31^D1), 44);
847	B2 = ROL64((A12^D2), 43);
848	B3 = ROL64((A43^D3), 21);
849	B4 = ROL64((A24^D4), 14);
850	A00 = B0 ^((~B1)& B2 );
851	A00 ^= RC[i+1];
852	A31 = B1 ^((~B2)& B3 );
853	A12 = B2 ^((~B3)& B4 );
854	A43 = B3 ^((~B4)& B0 );
855	A24 = B4 ^((~B0)& B1 );
856
857	B2 = ROL64((A40^D0), 3);
858	B3 = ROL64((A21^D1), 45);
859	B4 = ROL64((A02^D2), 61);
860	B0 = ROL64((A33^D3), 28);
861	B1 = ROL64((A14^D4), 20);
862	A40 = B0 ^((~B1)& B2 );
863	A21 = B1 ^((~B2)& B3 );
864	A02 = B2 ^((~B3)& B4 );
865	A33 = B3 ^((~B4)& B0 );
866	A14 = B4 ^((~B0)& B1 );
867
868	B4 = ROL64((A30^D0), 18);
869	B0 = ROL64((A11^D1), 1);
870	B1 = ROL64((A42^D2), 6);
871	B2 = ROL64((A23^D3), 25);
872	B3 = ROL64((A04^D4), 8);
873	A30 = B0 ^((~B1)& B2 );
874	A11 = B1 ^((~B2)& B3 );
875	A42 = B2 ^((~B3)& B4 );
876	A23 = B3 ^((~B4)& B0 );
877	A04 = B4 ^((~B0)& B1 );
878
879	B1 = ROL64((A20^D0), 36);
880	B2 = ROL64((A01^D1), 10);
881	B3 = ROL64((A32^D2), 15);
882	B4 = ROL64((A13^D3), 56);
883	B0 = ROL64((A44^D4), 27);
884	A20 = B0 ^((~B1)& B2 );
885	A01 = B1 ^((~B2)& B3 );
886	A32 = B2 ^((~B3)& B4 );
887	A13 = B3 ^((~B4)& B0 );
888	A44 = B4 ^((~B0)& B1 );
889
890	B3 = ROL64((A10^D0), 41);
891	B4 = ROL64((A41^D1), 2);
892	B0 = ROL64((A22^D2), 62);
893	B1 = ROL64((A03^D3), 55);
894	B2 = ROL64((A34^D4), 39);
895	A10 = B0 ^((~B1)& B2 );
896	A41 = B1 ^((~B2)& B3 );
897	A22 = B2 ^((~B3)& B4 );
898	A03 = B3 ^((~B4)& B0 );
899	A34 = B4 ^((~B0)& B1 );
900
901	C0 = A00^A40^A30^A20^A10;
902	C1 = A31^A21^A11^A01^A41;
903	C2 = A12^A02^A42^A32^A22;
904	C3 = A43^A33^A23^A13^A03;
905	C4 = A24^A14^A04^A44^A34;
906	D0 = C4^ROL64(C1, 1);
907	D1 = C0^ROL64(C2, 1);
908	D2 = C1^ROL64(C3, 1);
909	D3 = C2^ROL64(C4, 1);
910	D4 = C3^ROL64(C0, 1);
911
912	B0 = (A00^D0);
913	B1 = ROL64((A21^D1), 44);
914	B2 = ROL64((A42^D2), 43);
915	B3 = ROL64((A13^D3), 21);
916	B4 = ROL64((A34^D4), 14);
917	A00 = B0 ^((~B1)& B2 );
918	A00 ^= RC[i+2];
919	A21 = B1 ^((~B2)& B3 );
920	A42 = B2 ^((~B3)& B4 );
921	A13 = B3 ^((~B4)& B0 );
922	A34 = B4 ^((~B0)& B1 );
923
924	B2 = ROL64((A30^D0), 3);
925	B3 = ROL64((A01^D1), 45);
926	B4 = ROL64((A22^D2), 61);
927	B0 = ROL64((A43^D3), 28);
928	B1 = ROL64((A14^D4), 20);
929	A30 = B0 ^((~B1)& B2 );
930	A01 = B1 ^((~B2)& B3 );
931	A22 = B2 ^((~B3)& B4 );
932	A43 = B3 ^((~B4)& B0 );
933	A14 = B4 ^((~B0)& B1 );
934
935	B4 = ROL64((A10^D0), 18);
936	B0 = ROL64((A31^D1), 1);
937	B1 = ROL64((A02^D2), 6);
938	B2 = ROL64((A23^D3), 25);
939	B3 = ROL64((A44^D4), 8);
940	A10 = B0 ^((~B1)& B2 );
941	A31 = B1 ^((~B2)& B3 );
942	A02 = B2 ^((~B3)& B4 );
943	A23 = B3 ^((~B4)& B0 );
944	A44 = B4 ^((~B0)& B1 );
945
946	B1 = ROL64((A40^D0), 36);
947	B2 = ROL64((A11^D1), 10);
948	B3 = ROL64((A32^D2), 15);
949	B4 = ROL64((A03^D3), 56);
950	B0 = ROL64((A24^D4), 27);
951	A40 = B0 ^((~B1)& B2 );
952	A11 = B1 ^((~B2)& B3 );
953	A32 = B2 ^((~B3)& B4 );
954	A03 = B3 ^((~B4)& B0 );
955	A24 = B4 ^((~B0)& B1 );
956
957	B3 = ROL64((A20^D0), 41);
958	B4 = ROL64((A41^D1), 2);
959	B0 = ROL64((A12^D2), 62);
960	B1 = ROL64((A33^D3), 55);
961	B2 = ROL64((A04^D4), 39);
962	A20 = B0 ^((~B1)& B2 );
963	A41 = B1 ^((~B2)& B3 );
964	A12 = B2 ^((~B3)& B4 );
965	A33 = B3 ^((~B4)& B0 );
966	A04 = B4 ^((~B0)& B1 );
967
968	C0 = A00^A30^A10^A40^A20;
969	C1 = A21^A01^A31^A11^A41;
970	C2 = A42^A22^A02^A32^A12;
971	C3 = A13^A43^A23^A03^A33;
972	C4 = A34^A14^A44^A24^A04;
973	D0 = C4^ROL64(C1, 1);
974	D1 = C0^ROL64(C2, 1);
975	D2 = C1^ROL64(C3, 1);
976	D3 = C2^ROL64(C4, 1);
977	D4 = C3^ROL64(C0, 1);
978
979	B0 = (A00^D0);
980	B1 = ROL64((A01^D1), 44);
981	B2 = ROL64((A02^D2), 43);
982	B3 = ROL64((A03^D3), 21);
983	B4 = ROL64((A04^D4), 14);
984	A00 = B0 ^((~B1)& B2 );
985	A00 ^= RC[i+3];
986	A01 = B1 ^((~B2)& B3 );
987	A02 = B2 ^((~B3)& B4 );
988	A03 = B3 ^((~B4)& B0 );
989	A04 = B4 ^((~B0)& B1 );
990
991	B2 = ROL64((A10^D0), 3);
992	B3 = ROL64((A11^D1), 45);
993	B4 = ROL64((A12^D2), 61);
994	B0 = ROL64((A13^D3), 28);
995	B1 = ROL64((A14^D4), 20);
996	A10 = B0 ^((~B1)& B2 );
997	A11 = B1 ^((~B2)& B3 );
998	A12 = B2 ^((~B3)& B4 );
999	A13 = B3 ^((~B4)& B0 );
1000	A14 = B4 ^((~B0)& B1 );
1001
1002	B4 = ROL64((A20^D0), 18);
1003	B0 = ROL64((A21^D1), 1);
1004	B1 = ROL64((A22^D2), 6);
1005	B2 = ROL64((A23^D3), 25);
1006	B3 = ROL64((A24^D4), 8);
1007	A20 = B0 ^((~B1)& B2 );
1008	A21 = B1 ^((~B2)& B3 );
1009	A22 = B2 ^((~B3)& B4 );
1010	A23 = B3 ^((~B4)& B0 );
1011	A24 = B4 ^((~B0)& B1 );
1012
1013	B1 = ROL64((A30^D0), 36);
1014	B2 = ROL64((A31^D1), 10);
1015	B3 = ROL64((A32^D2), 15);
1016	B4 = ROL64((A33^D3), 56);
1017	B0 = ROL64((A34^D4), 27);
1018	A30 = B0 ^((~B1)& B2 );
1019	A31 = B1 ^((~B2)& B3 );
1020	A32 = B2 ^((~B3)& B4 );
1021	A33 = B3 ^((~B4)& B0 );
1022	A34 = B4 ^((~B0)& B1 );
1023
1024	B3 = ROL64((A40^D0), 41);
1025	B4 = ROL64((A41^D1), 2);
1026	B0 = ROL64((A42^D2), 62);
1027	B1 = ROL64((A43^D3), 55);
1028	B2 = ROL64((A44^D4), 39);
1029	A40 = B0 ^((~B1)& B2 );
1030	A41 = B1 ^((~B2)& B3 );
1031	A42 = B2 ^((~B3)& B4 );
1032	A43 = B3 ^((~B4)& B0 );
1033	A44 = B4 ^((~B0)& B1 );
1034	}
1035	}
1036
1037	/*
1038	** Initialize a new hash. iSize determines the size of the hash
1039	** in bits and should be one of 224, 256, 384, or 512. Or iSize
1040	** can be zero to use the default hash size of 256 bits.
1041	*/
1042	static void SHA3Init(SHA3Context *p, int iSize){
1043	memset(p, 0, sizeof(*p));
1044	if( iSize>=128 && iSize<=512 ){
1045	p->nRate = (1600 - ((iSize + 31)&~31)*2)/8;
1046	}else{
1047	p->nRate = (1600 - 2*256)/8;
1048	}
1049	#if SHA3_BYTEORDER==1234
1050	/* Known to be little-endian at compile-time. No-op */
1051	#elif SHA3_BYTEORDER==4321
1052	p->ixMask = 7; /* Big-endian */
1053	#else
1054	{
1055	static unsigned int one = 1;
1056	if( 1==(unsigned char)&one ){
1057	/* Little endian. No byte swapping. */
1058	p->ixMask = 0;
1059	}else{
1060	/* Big endian. Byte swap. */
1061	p->ixMask = 7;
1062	}
1063	}
1064	#endif
1065	}
1066
1067	/*
1068	** Make consecutive calls to the SHA3Update function to add new content
1069	** to the hash
1070	*/
1071	static void SHA3Update(
1072	SHA3Context *p,
1073	const unsigned char *aData,
1074	unsigned int nData
1075	){
1076	unsigned int i = 0;
1077	#if SHA3_BYTEORDER==1234
1078	if( (p->nLoaded % 8)==0 && ((aData - (const unsigned char*)0)&7)==0 ){
1079	for(; i+7<nData; i+=8){
1080	p->u.s[p->nLoaded/8] ^= (u64)&aData[i];
1081	p->nLoaded += 8;
1082	if( p->nLoaded>=p->nRate ){
1083	KeccakF1600Step(p);
1084	p->nLoaded = 0;
1085	}
1086	}
1087	}
1088	#endif
1089	for(; i<nData; i++){
1090	#if SHA3_BYTEORDER==1234
1091	p->u.x[p->nLoaded] ^= aData[i];
1092	#elif SHA3_BYTEORDER==4321
1093	p->u.x[p->nLoaded^0x07] ^= aData[i];
1094	#else
1095	p->u.x[p->nLoaded^p->ixMask] ^= aData[i];
1096	#endif
1097	p->nLoaded++;
1098	if( p->nLoaded==p->nRate ){
1099	KeccakF1600Step(p);
1100	p->nLoaded = 0;
1101	}
1102	}
1103	}
1104
1105	/*
1106	** After all content has been added, invoke SHA3Final() to compute
1107	** the final hash. The function returns a pointer to the binary
1108	** hash value.
1109	*/
1110	static unsigned char SHA3Final(SHA3Context p){
1111	unsigned int i;
1112	if( p->nLoaded==p->nRate-1 ){
1113	const unsigned char c1 = 0x86;
1114	SHA3Update(p, &c1, 1);
1115	}else{
1116	const unsigned char c2 = 0x06;
1117	const unsigned char c3 = 0x80;
1118	SHA3Update(p, &c2, 1);
1119	p->nLoaded = p->nRate - 1;
1120	SHA3Update(p, &c3, 1);
1121	}
1122	for(i=0; i<p->nRate; i++){
1123	p->u.x[i+p->nRate] = p->u.x[i^p->ixMask];
1124	}
1125	return &p->u.x[p->nRate];
1126	}
1127
1128	/*
1129	** Implementation of the sha3(X,SIZE) function.
1130	**
1131	** Return a BLOB which is the SIZE-bit SHA3 hash of X. The default
1132	** size is 256. If X is a BLOB, it is hashed as is.
1133	** For all other non-NULL types of input, X is converted into a UTF-8 string
1134	** and the string is hashed without the trailing 0x00 terminator. The hash
1135	** of a NULL value is NULL.
1136	*/
1137	static void sha3Func(
1138	sqlite3_context *context,
1139	int argc,
1140	sqlite3_value **argv
1141	){
1142	SHA3Context cx;
1143	int eType = sqlite3_value_type(argv[0]);
1144	int nByte = sqlite3_value_bytes(argv[0]);
1145	int iSize;
1146	if( argc==1 ){
1147	iSize = 256;
1148	}else{
1149	iSize = sqlite3_value_int(argv[1]);
1150	if( iSize!=224 && iSize!=256 && iSize!=384 && iSize!=512 ){
1151	sqlite3_result_error(context, "SHA3 size should be one of: 224 256 "
1152	"384 512", -1);
1153	return;
1154	}
1155	}
1156	if( eType==SQLITE_NULL ) return;
1157	SHA3Init(&cx, iSize);
1158	if( eType==SQLITE_BLOB ){
1159	SHA3Update(&cx, sqlite3_value_blob(argv[0]), nByte);
1160	}else{
1161	SHA3Update(&cx, sqlite3_value_text(argv[0]), nByte);
1162	}
1163	sqlite3_result_blob(context, SHA3Final(&cx), iSize/8, SQLITE_TRANSIENT);
1164	}
1165
1166	/* Compute a string using sqlite3_vsnprintf() with a maximum length
1167	** of 50 bytes and add it to the hash.
1168	*/
1169	static void hash_step_vformat(
1170	SHA3Context p, / Add content to this context */
1171	const char *zFormat,
1172	...
1173	){
1174	va_list ap;
1175	int n;
1176	char zBuf[50];
1177	va_start(ap, zFormat);
1178	sqlite3_vsnprintf(sizeof(zBuf),zBuf,zFormat,ap);
1179	va_end(ap);
1180	n = (int)strlen(zBuf);
1181	SHA3Update(p, (unsigned char*)zBuf, n);
1182	}
1183
1184	/*
1185	** Implementation of the sha3_query(SQL,SIZE) function.
1186	**
1187	** This function compiles and runs the SQL statement(s) given in the
1188	** argument. The results are hashed using a SIZE-bit SHA3. The default
1189	** size is 256.
1190	**
1191	** The format of the byte stream that is hashed is summarized as follows:
1192	**
1193	** S<n>:<sql>
1194	** R
1195	** N
1196	** I<int>
1197	** F<ieee-float>
1198	** B<size>:<bytes>
1199	** T<size>:<text>
1200	**
1201	** <sql> is the original SQL text for each statement run and <n> is
1202	** the size of that text. The SQL text is UTF-8. A single R character
1203	** occurs before the start of each row. N means a NULL value.
1204	** I mean an 8-byte little-endian integer <int>. F is a floating point
1205	** number with an 8-byte little-endian IEEE floating point value <ieee-float>.
1206	** B means blobs of <size> bytes. T means text rendered as <size>
1207	** bytes of UTF-8. The <n> and <size> values are expressed as an ASCII
1208	** text integers.
1209	**
1210	** For each SQL statement in the X input, there is one S segment. Each
1211	** S segment is followed by zero or more R segments, one for each row in the
1212	** result set. After each R, there are one or more N, I, F, B, or T segments,
1213	** one for each column in the result set. Segments are concatentated directly
1214	** with no delimiters of any kind.
1215	*/
1216	static void sha3QueryFunc(
1217	sqlite3_context *context,
1218	int argc,
1219	sqlite3_value **argv
1220	){
1221	sqlite3 *db = sqlite3_context_db_handle(context);
1222	const char zSql = (const char)sqlite3_value_text(argv[0]);
1223	sqlite3_stmt *pStmt = 0;
1224	int nCol; /* Number of columns in the result set */
1225	int i; /* Loop counter */
1226	int rc;
1227	int n;
1228	const char *z;
1229	SHA3Context cx;
1230	int iSize;
1231
1232	if( argc==1 ){
1233	iSize = 256;
1234	}else{
1235	iSize = sqlite3_value_int(argv[1]);
1236	if( iSize!=224 && iSize!=256 && iSize!=384 && iSize!=512 ){
1237	sqlite3_result_error(context, "SHA3 size should be one of: 224 256 "
1238	"384 512", -1);
1239	return;
1240	}
1241	}
1242	if( zSql==0 ) return;
1243	SHA3Init(&cx, iSize);
1244	while( zSql[0] ){
1245	rc = sqlite3_prepare_v2(db, zSql, -1, &pStmt, &zSql);
1246	if( rc ){
1247	char *zMsg = sqlite3_mprintf("error SQL statement [%s]: %s",
1248	zSql, sqlite3_errmsg(db));
1249	sqlite3_finalize(pStmt);
1250	sqlite3_result_error(context, zMsg, -1);
1251	sqlite3_free(zMsg);
1252	return;
1253	}
1254	if( !sqlite3_stmt_readonly(pStmt) ){
1255	char *zMsg = sqlite3_mprintf("non-query: [%s]", sqlite3_sql(pStmt));
1256	sqlite3_finalize(pStmt);
1257	sqlite3_result_error(context, zMsg, -1);
1258	sqlite3_free(zMsg);
1259	return;
1260	}
1261	nCol = sqlite3_column_count(pStmt);
1262	z = sqlite3_sql(pStmt);
1263	if( z==0 ){
1264	sqlite3_finalize(pStmt);
1265	continue;
1266	}
1267	n = (int)strlen(z);
1268	hash_step_vformat(&cx,"S%d:",n);
1269	SHA3Update(&cx,(unsigned char*)z,n);
1270
1271	/* Compute a hash over the result of the query */
1272	while( SQLITE_ROW==sqlite3_step(pStmt) ){
1273	SHA3Update(&cx,(const unsigned char*)"R",1);
1274	for(i=0; i<nCol; i++){
1275	switch( sqlite3_column_type(pStmt,i) ){
1276	case SQLITE_NULL: {
1277	SHA3Update(&cx, (const unsigned char*)"N",1);
1278	break;
1279	}
1280	case SQLITE_INTEGER: {
1281	sqlite3_uint64 u;
1282	int j;
1283	unsigned char x[9];
1284	sqlite3_int64 v = sqlite3_column_int64(pStmt,i);
1285	memcpy(&u, &v, 8);
1286	for(j=8; j>=1; j--){
1287	x[j] = u & 0xff;
1288	u >>= 8;
1289	}
1290	x[0] = 'I';
1291	SHA3Update(&cx, x, 9);
1292	break;
1293	}
1294	case SQLITE_FLOAT: {
1295	sqlite3_uint64 u;
1296	int j;
1297	unsigned char x[9];
1298	double r = sqlite3_column_double(pStmt,i);
1299	memcpy(&u, &r, 8);
1300	for(j=8; j>=1; j--){
1301	x[j] = u & 0xff;
1302	u >>= 8;
1303	}
1304	x[0] = 'F';
1305	SHA3Update(&cx,x,9);
1306	break;
1307	}
1308	case SQLITE_TEXT: {
1309	int n2 = sqlite3_column_bytes(pStmt, i);
1310	const unsigned char *z2 = sqlite3_column_text(pStmt, i);
1311	hash_step_vformat(&cx,"T%d:",n2);
1312	SHA3Update(&cx, z2, n2);
1313	break;
1314	}
1315	case SQLITE_BLOB: {
1316	int n2 = sqlite3_column_bytes(pStmt, i);
1317	const unsigned char *z2 = sqlite3_column_blob(pStmt, i);
1318	hash_step_vformat(&cx,"B%d:",n2);
1319	SHA3Update(&cx, z2, n2);
1320	break;
1321	}
1322	}
1323	}
1324	}
1325	sqlite3_finalize(pStmt);
1326	}
1327	sqlite3_result_blob(context, SHA3Final(&cx), iSize/8, SQLITE_TRANSIENT);
1328	}
1329	/* End of SHA3 hashing logic copy/pasted from ../ext/misc/shathree.c
1330	********************************************************************************/
1331
1332	#if defined(SQLITE_ENABLE_SESSION)
1333	/*
1334	** State information for a single open session
1335	*/
	@@ -608,17 +1359,14 @@
1359	** instance of the following structure.
1360	*/
1361	typedef struct ShellState ShellState;
1362	struct ShellState {
1363	sqlite3 db; / The database */

1364	int autoExplain; /* Automatically turn on .explain mode */
1365	int autoEQP; /* Run EXPLAIN QUERY PLAN prior to seach SQL stmt */
1366	int statsOn; /* True to display memory stats before each finalize */
1367	int scanstatsOn; /* True to display scan stats before each finalize */


1368	int outCount; /* Revert to stdout when reaching zero */
1369	int cnt; /* Number of records displayed so far */
1370	FILE out; / Write results here */
1371	FILE traceOut; / Output for sqlite3_trace() */
1372	int nErr; /* Number of errors seen */
	@@ -653,13 +1401,24 @@
1401	};
1402
1403	/*
1404	** These are the allowed shellFlgs values
1405	*/
1406	#define SHFLG_Scratch 0x00000001 /* The --scratch option is used */
1407	#define SHFLG_Pagecache 0x00000002 /* The --pagecache option is used */
1408	#define SHFLG_Lookaside 0x00000004 /* Lookaside memory is used */
1409	#define SHFLG_Backslash 0x00000008 /* The --backslash option is used */
1410	#define SHFLG_PreserveRowid 0x00000010 /* .dump preserves rowid values */
1411	#define SHFLG_CountChanges 0x00000020 /* .changes setting */
1412	#define SHFLG_Echo 0x00000040 /* .echo or --echo setting */
1413
1414	/*
1415	** Macros for testing and setting shellFlgs
1416	*/
1417	#define ShellHasFlag(P,X) (((P)->shellFlgs & (X))!=0)
1418	#define ShellSetFlag(P,X) ((P)->shellFlgs\|=(X))
1419	#define ShellClearFlag(P,X) ((P)->shellFlgs&=(~(X)))
1420
1421	/*
1422	** These are the allowed modes.
1423	*/
1424	#define MODE_Line 0 /* One column per line. Blank line between records */
	@@ -1292,92 +2051,106 @@
2051	*/
2052	static int callback(void pArg, int nArg, char azArg, char *azCol){
2053	/* since we don't have type info, call the shell_callback with a NULL value */
2054	return shell_callback(pArg, nArg, azArg, azCol, NULL);
2055	}
2056
2057	/*
2058	** This is the callback routine from sqlite3_exec() that appends all
2059	** output onto the end of a ShellText object.
2060	*/
2061	static int captureOutputCallback(void pArg, int nArg, char azArg, char *az){
2062	ShellText p = (ShellText)pArg;
2063	int i;
2064	if( p->n ) appendText(p, "\|", 0);
2065	for(i=0; i<nArg; i++){
2066	if( i ) appendText(p, ",", 0);
2067	if( azArg[i] ) appendText(p, azArg[i], 0);
2068	}
2069	return 0;
2070	}
2071
2072	/*
2073	** Generate an appropriate SELFTEST table in the main database.
2074	*/
2075	static void createSelftestTable(ShellState *p){
2076	char *zErrMsg = 0;
2077	sqlite3_exec(p->db,
2078	"SAVEPOINT selftest_init;\n"
2079	"CREATE TABLE IF NOT EXISTS selftest(\n"
2080	" tno INTEGER PRIMARY KEY,\n" /* Test number */
2081	" op TEXT,\n" /* Operator: memo run */
2082	" cmd TEXT,\n" /* Command text */
2083	" ans TEXT\n" /* Desired answer */
2084	");"
2085	"CREATE TEMP TABLE [_shell$self](op,cmd,ans);\n"
2086	"INSERT INTO [_shell$self](rowid,op,cmd)\n"
2087	" VALUES(coalesce((SELECT (max(tno)+100)/10 FROM selftest),10),\n"
2088	" 'memo','Tests generated by --init');\n"
2089	"INSERT INTO [_shell$self]\n"
2090	" SELECT 'run',\n"
2091	" 'SELECT hex(sha3_query(''SELECT type,name,tbl_name,sql "
2092	"FROM sqlite_master ORDER BY 2'',224))',\n"
2093	" hex(sha3_query('SELECT type,name,tbl_name,sql "
2094	"FROM sqlite_master ORDER BY 2',224));\n"
2095	"INSERT INTO [_shell$self]\n"
2096	" SELECT 'run',"
2097	" 'SELECT hex(sha3_query(''SELECT * FROM \"' \|\|"
2098	" printf('%w',name) \|\| '\" NOT INDEXED'',224))',\n"
2099	" hex(sha3_query(printf('SELECT * FROM \"%w\" NOT INDEXED',name),224))\n"
2100	" FROM (\n"
2101	" SELECT name FROM sqlite_master\n"
2102	" WHERE type='table'\n"
2103	" AND name<>'selftest'\n"
2104	" AND coalesce(rootpage,0)>0\n"
2105	" )\n"
2106	" ORDER BY name;\n"
2107	"INSERT INTO [_shell$self]\n"
2108	" VALUES('run','PRAGMA integrity_check','ok');\n"
2109	"INSERT INTO selftest(tno,op,cmd,ans)"
2110	" SELECT rowid*10,op,cmd,ans FROM [_shell$self];\n"
2111	"DROP TABLE [_shell$self];"
2112	,0,0,&zErrMsg);
2113	if( zErrMsg ){
2114	utf8_printf(stderr, "SELFTEST initialization failure: %s\n", zErrMsg);
2115	sqlite3_free(zErrMsg);
2116	}
2117	sqlite3_exec(p->db, "RELEASE selftest_init",0,0,0);
2118	}
2119
2120
2121	/*
2122	** Set the destination table field of the ShellState structure to
2123	** the name of the table given. Escape any quote characters in the
2124	** table name.
2125	*/
2126	static void set_table_name(ShellState p, const char zName){
2127	int i, n;
2128	int cQuote;
2129	char *z;
2130
2131	if( p->zDestTable ){
2132	free(p->zDestTable);
2133	p->zDestTable = 0;
2134	}
2135	if( zName==0 ) return;
2136	cQuote = quoteChar(zName);
2137	n = strlen30(zName);
2138	if( cQuote ) n += 2;





2139	z = p->zDestTable = malloc( n+1 );
2140	if( z==0 ){
2141	raw_printf(stderr,"Error: out of memory\n");
2142	exit(1);
2143	}
2144	n = 0;
2145	if( cQuote ) z[n++] = cQuote;
2146	for(i=0; zName[i]; i++){
2147	z[n++] = zName[i];
2148	if( zName[i]==cQuote ) z[n++] = cQuote;
2149	}
2150	if( cQuote ) z[n++] = cQuote;
2151	z[n] = 0;













































2152	}
2153
2154
2155	/*
2156	** Execute a query statement that will generate SQL output. Print
	@@ -1484,10 +2257,35 @@
2257	}
2258	fclose(in);
2259	}
2260	#endif
2261
2262	/*
2263	** Display a single line of status using 64-bit values.
2264	*/
2265	static void displayStatLine(
2266	ShellState p, / The shell context */
2267	char zLabel, / Label for this one line */
2268	char zFormat, / Format for the result */
2269	int iStatusCtrl, /* Which status to display */
2270	int bReset /* True to reset the stats */
2271	){
2272	sqlite3_int64 iCur = -1;
2273	sqlite3_int64 iHiwtr = -1;
2274	int i, nPercent;
2275	char zLine[200];
2276	sqlite3_status64(iStatusCtrl, &iCur, &iHiwtr, bReset);
2277	for(i=0, nPercent=0; zFormat[i]; i++){
2278	if( zFormat[i]=='%' ) nPercent++;
2279	}
2280	if( nPercent>1 ){
2281	sqlite3_snprintf(sizeof(zLine), zLine, zFormat, iCur, iHiwtr);
2282	}else{
2283	sqlite3_snprintf(sizeof(zLine), zLine, zFormat, iHiwtr);
2284	}
2285	raw_printf(p->out, "%-36s %s\n", zLabel, zLine);
2286	}
2287
2288	/*
2289	** Display memory stats.
2290	*/
2291	static int display_stats(
	@@ -1497,61 +2295,35 @@
2295	){
2296	int iCur;
2297	int iHiwtr;
2298
2299	if( pArg && pArg->out ){
2300	displayStatLine(pArg, "Memory Used:",
2301	"%lld (max %lld) bytes", SQLITE_STATUS_MEMORY_USED, bReset);
2302	displayStatLine(pArg, "Number of Outstanding Allocations:",
2303	"%lld (max %lld)", SQLITE_STATUS_MALLOC_COUNT, bReset);






2304	if( pArg->shellFlgs & SHFLG_Pagecache ){
2305	displayStatLine(pArg, "Number of Pcache Pages Used:",
2306	"%lld (max %lld) pages", SQLITE_STATUS_PAGECACHE_USED, bReset);
2307	}
2308	displayStatLine(pArg, "Number of Pcache Overflow Bytes:",
2309	"%lld (max %lld) bytes", SQLITE_STATUS_PAGECACHE_OVERFLOW, bReset);






2310	if( pArg->shellFlgs & SHFLG_Scratch ){
2311	displayStatLine(pArg, "Number of Scratch Allocations Used:",
2312	"%lld (max %lld)", SQLITE_STATUS_SCRATCH_USED, bReset);
2313	}
2314	displayStatLine(pArg, "Number of Scratch Overflow Bytes:",
2315	"%lld (max %lld) bytes", SQLITE_STATUS_SCRATCH_OVERFLOW, bReset);
2316	displayStatLine(pArg, "Largest Allocation:",
2317	"%lld bytes", SQLITE_STATUS_MALLOC_SIZE, bReset);
2318	displayStatLine(pArg, "Largest Pcache Allocation:",
2319	"%lld bytes", SQLITE_STATUS_PAGECACHE_SIZE, bReset);
2320	displayStatLine(pArg, "Largest Scratch Allocation:",
2321	"%lld bytes", SQLITE_STATUS_SCRATCH_SIZE, bReset);












2322	#ifdef YYTRACKMAXSTACKDEPTH
2323	displayStatLine(pArg, "Deepest Parser Stack:",
2324	"%lld (max %lld)", SQLITE_STATUS_PARSER_STACK, bReset);


2325	#endif
2326	}
2327
2328	if( pArg && pArg->out && db ){
2329	if( pArg->shellFlgs & SHFLG_Lookaside ){
	@@ -1934,11 +2706,11 @@
2706	pArg->pStmt = pStmt;
2707	pArg->cnt = 0;
2708	}
2709
2710	/* echo the sql statement if echo on */
2711	if( pArg && ShellHasFlag(pArg, SHFLG_Echo) ){
2712	utf8_printf(pArg->out, "%s\n", zStmtSql ? zStmtSql : zSql);
2713	}
2714
2715	/* Show the EXPLAIN QUERY PLAN if .eqp is on */
2716	if( pArg && pArg->autoEQP && sqlite3_strlike("EXPLAIN%",zStmtSql,0)!=0 ){
	@@ -2022,10 +2794,142 @@
2794	} /* end while */
2795
2796	return rc;
2797	}
2798
2799	/*
2800	** Release memory previously allocated by tableColumnList().
2801	*/
2802	static void freeColumnList(char **azCol){
2803	int i;
2804	for(i=1; azCol[i]; i++){
2805	sqlite3_free(azCol[i]);
2806	}
2807	/* azCol[0] is a static string */
2808	sqlite3_free(azCol);
2809	}
2810
2811	/*
2812	** Return a list of pointers to strings which are the names of all
2813	** columns in table zTab. The memory to hold the names is dynamically
2814	** allocated and must be released by the caller using a subsequent call
2815	** to freeColumnList().
2816	**
2817	** The azCol[0] entry is usually NULL. However, if zTab contains a rowid
2818	** value that needs to be preserved, then azCol[0] is filled in with the
2819	** name of the rowid column.
2820	**
2821	** The first regular column in the table is azCol[1]. The list is terminated
2822	** by an entry with azCol[i]==0.
2823	*/
2824	static char *tableColumnList(ShellState p, const char *zTab){
2825	char **azCol = 0;
2826	sqlite3_stmt *pStmt;
2827	char *zSql;
2828	int nCol = 0;
2829	int nAlloc = 0;
2830	int nPK = 0; /* Number of PRIMARY KEY columns seen */
2831	int isIPK = 0; /* True if one PRIMARY KEY column of type INTEGER */
2832	int preserveRowid = ShellHasFlag(p, SHFLG_PreserveRowid);
2833	int rc;
2834
2835	zSql = sqlite3_mprintf("PRAGMA table_info=%Q", zTab);
2836	rc = sqlite3_prepare_v2(p->db, zSql, -1, &pStmt, 0);
2837	sqlite3_free(zSql);
2838	if( rc ) return 0;
2839	while( sqlite3_step(pStmt)==SQLITE_ROW ){
2840	if( nCol>=nAlloc-2 ){
2841	nAlloc = nAlloc*2 + nCol + 10;
2842	azCol = sqlite3_realloc(azCol, nAlloc*sizeof(azCol[0]));
2843	if( azCol==0 ){
2844	raw_printf(stderr, "Error: out of memory\n");
2845	exit(1);
2846	}
2847	}
2848	azCol[++nCol] = sqlite3_mprintf("%s", sqlite3_column_text(pStmt, 1));
2849	if( sqlite3_column_int(pStmt, 5) ){
2850	nPK++;
2851	if( nPK==1
2852	&& sqlite3_stricmp((const char*)sqlite3_column_text(pStmt,2),
2853	"INTEGER")==0
2854	){
2855	isIPK = 1;
2856	}else{
2857	isIPK = 0;
2858	}
2859	}
2860	}
2861	sqlite3_finalize(pStmt);
2862	azCol[0] = 0;
2863	azCol[nCol+1] = 0;
2864
2865	/* The decision of whether or not a rowid really needs to be preserved
2866	** is tricky. We never need to preserve a rowid for a WITHOUT ROWID table
2867	** or a table with an INTEGER PRIMARY KEY. We are unable to preserve
2868	** rowids on tables where the rowid is inaccessible because there are other
2869	** columns in the table named "rowid", "_rowid_", and "oid".
2870	*/
2871	if( preserveRowid && isIPK ){
2872	/* If a single PRIMARY KEY column with type INTEGER was seen, then it
2873	** might be an alise for the ROWID. But it might also be a WITHOUT ROWID
2874	** table or a INTEGER PRIMARY KEY DESC column, neither of which are
2875	** ROWID aliases. To distinguish these cases, check to see if
2876	** there is a "pk" entry in "PRAGMA index_list". There will be
2877	** no "pk" index if the PRIMARY KEY really is an alias for the ROWID.
2878	*/
2879	zSql = sqlite3_mprintf("SELECT 1 FROM pragma_index_list(%Q)"
2880	" WHERE origin='pk'", zTab);
2881	rc = sqlite3_prepare_v2(p->db, zSql, -1, &pStmt, 0);
2882	sqlite3_free(zSql);
2883	if( rc ){
2884	freeColumnList(azCol);
2885	return 0;
2886	}
2887	rc = sqlite3_step(pStmt);
2888	sqlite3_finalize(pStmt);
2889	preserveRowid = rc==SQLITE_ROW;
2890	}
2891	if( preserveRowid ){
2892	/* Only preserve the rowid if we can find a name to use for the
2893	** rowid */
2894	static char *azRowid[] = { "rowid", "_rowid_", "oid" };
2895	int i, j;
2896	for(j=0; j<3; j++){
2897	for(i=1; i<=nCol; i++){
2898	if( sqlite3_stricmp(azRowid[j],azCol[i])==0 ) break;
2899	}
2900	if( i>nCol ){
2901	/* At this point, we know that azRowid[j] is not the name of any
2902	** ordinary column in the table. Verify that azRowid[j] is a valid
2903	** name for the rowid before adding it to azCol[0]. WITHOUT ROWID
2904	** tables will fail this last check */
2905	int rc;
2906	rc = sqlite3_table_column_metadata(p->db,0,zTab,azRowid[j],0,0,0,0,0);
2907	if( rc==SQLITE_OK ) azCol[0] = azRowid[j];
2908	break;
2909	}
2910	}
2911	}
2912	return azCol;
2913	}
2914
2915	/*
2916	** Toggle the reverse_unordered_selects setting.
2917	*/
2918	static void toggleSelectOrder(sqlite3 *db){
2919	sqlite3_stmt *pStmt = 0;
2920	int iSetting = 0;
2921	char zStmt[100];
2922	sqlite3_prepare_v2(db, "PRAGMA reverse_unordered_selects", -1, &pStmt, 0);
2923	if( sqlite3_step(pStmt)==SQLITE_ROW ){
2924	iSetting = sqlite3_column_int(pStmt, 0);
2925	}
2926	sqlite3_finalize(pStmt);
2927	sqlite3_snprintf(sizeof(zStmt), zStmt,
2928	"PRAGMA reverse_unordered_selects(%d)", !iSetting);
2929	sqlite3_exec(db, zStmt, 0, 0, 0);
2930	}
2931
2932	/*
2933	** This is a different callback routine used for dumping the database.
2934	** Each row received by this callback consists of a table name,
2935	** the table type ("index" or "table") and SQL to create the table.
	@@ -2034,21 +2938,20 @@
2938	static int dump_callback(void pArg, int nArg, char azArg, char *azCol){
2939	int rc;
2940	const char *zTable;
2941	const char *zType;
2942	const char *zSql;

2943	ShellState p = (ShellState )pArg;
2944
2945	UNUSED_PARAMETER(azCol);
2946	if( nArg!=3 ) return 1;
2947	zTable = azArg[0];
2948	zType = azArg[1];
2949	zSql = azArg[2];
2950
2951	if( strcmp(zTable, "sqlite_sequence")==0 ){
2952	raw_printf(p->out, "DELETE FROM sqlite_sequence;\n");
2953	}else if( sqlite3_strglob("sqlite_stat?", zTable)==0 ){
2954	raw_printf(p->out, "ANALYZE sqlite_master;\n");
2955	}else if( strncmp(zTable, "sqlite_", 7)==0 ){
2956	return 0;
2957	}else if( strncmp(zSql, "CREATE VIRTUAL TABLE", 20)==0 ){
	@@ -2067,62 +2970,74 @@
2970	}else{
2971	printSchemaLine(p->out, zSql, ";\n");
2972	}
2973
2974	if( strcmp(zType, "table")==0 ){
2975	ShellText sSelect;
2976	ShellText sTable;
2977	char **azCol;
2978	int i;
2979	char *savedDestTable;
2980	int savedMode;
2981
2982	azCol = tableColumnList(p, zTable);
2983	if( azCol==0 ){
2984	p->nErr++;
2985	return 0;
2986	}
2987




2988	/* Always quote the table name, even if it appears to be pure ascii,
2989	** in case it is a keyword. Ex: INSERT INTO "table" ... */
2990	initText(&sTable);
2991	appendText(&sTable, zTable, quoteChar(zTable));
2992	/* If preserving the rowid, add a column list after the table name.
2993	** In other words: "INSERT INTO tab(rowid,a,b,c,...) VALUES(...)"
2994	** instead of the usual "INSERT INTO tab VALUES(...)".
2995	*/
2996	if( azCol[0] ){
2997	appendText(&sTable, "(", 0);
2998	appendText(&sTable, azCol[0], 0);
2999	for(i=1; azCol[i]; i++){
3000	appendText(&sTable, ",", 0);
3001	appendText(&sTable, azCol[i], quoteChar(azCol[i]));
3002	}
3003	appendText(&sTable, ")", 0);
3004	}
3005
3006	/* Build an appropriate SELECT statement */
3007	initText(&sSelect);
3008	appendText(&sSelect, "SELECT ", 0);
3009	if( azCol[0] ){
3010	appendText(&sSelect, azCol[0], 0);
3011	appendText(&sSelect, ",", 0);
3012	}
3013	for(i=1; azCol[i]; i++){
3014	appendText(&sSelect, azCol[i], quoteChar(azCol[i]));
3015	if( azCol[i+1] ){
3016	appendText(&sSelect, ",", 0);
3017	}
3018	}
3019	freeColumnList(azCol);
3020	appendText(&sSelect, " FROM ", 0);
3021	appendText(&sSelect, zTable, quoteChar(zTable));
3022
3023	savedDestTable = p->zDestTable;
3024	savedMode = p->mode;
3025	p->zDestTable = sTable.z;
3026	p->mode = p->cMode = MODE_Insert;
3027	rc = shell_exec(p->db, sSelect.z, shell_callback, p, 0);
3028	if( (rc&0xff)==SQLITE_CORRUPT ){
3029	raw_printf(p->out, "/**** CORRUPTION ERROR *****/\n");
3030	toggleSelectOrder(p->db);
3031	shell_exec(p->db, sSelect.z, shell_callback, p, 0);
3032	toggleSelectOrder(p->db);
3033	}
3034	p->zDestTable = savedDestTable;
3035	p->mode = savedMode;
3036	freeText(&sTable);
3037	freeText(&sSelect);
3038	if( rc ) p->nErr++;
3039	}
3040	return 0;
3041	}
3042
3043	/*
	@@ -2210,11 +3125,11 @@
3125	" csv Comma-separated values\n"
3126	" column Left-aligned columns. (See .width)\n"
3127	" html HTML <table> code\n"
3128	" insert SQL insert statements for TABLE\n"
3129	" line One value per line\n"
3130	" list Values delimited by \"\|\"\n"
3131	" quote Escape answers as for SQL\n"
3132	" tabs Tab-separated values\n"
3133	" tcl TCL list elements\n"
3134	".nullvalue STRING Use STRING in place of NULL values\n"
3135	".once FILENAME Output for the next SQL command only to FILENAME\n"
	@@ -2233,10 +3148,11 @@
3148	".separator COL ?ROW? Change the column separator and optionally the row\n"
3149	" separator for both the output mode and .import\n"
3150	#if defined(SQLITE_ENABLE_SESSION)
3151	".session CMD ... Create or control sessions\n"
3152	#endif
3153	".sha3sum ?OPTIONS...? Compute a SHA3 hash of database content\n"
3154	".shell CMD ARGS... Run CMD ARGS... in a system shell\n"
3155	".show Show the current values for various settings\n"
3156	".stats ?on\|off? Show stats or turn stats on or off\n"
3157	".system CMD ARGS... Run CMD ARGS... in a system shell\n"
3158	".tables ?TABLE? List names of tables\n"
	@@ -2423,14 +3339,10 @@
3339	static void open_db(ShellState *p, int keepAlive){
3340	if( p->db==0 ){
3341	sqlite3_initialize();
3342	sqlite3_open(p->zDbFilename, &p->db);
3343	globalDb = p->db;




3344	if( p->db==0 \|\| SQLITE_OK!=sqlite3_errcode(p->db) ){
3345	utf8_printf(stderr,"Error: unable to open database \"%s\": %s\n",
3346	p->zDbFilename, sqlite3_errmsg(p->db));
3347	if( keepAlive ) return;
3348	exit(1);
	@@ -2440,10 +3352,18 @@
3352	#endif
3353	sqlite3_create_function(p->db, "readfile", 1, SQLITE_UTF8, 0,
3354	readfileFunc, 0, 0);
3355	sqlite3_create_function(p->db, "writefile", 2, SQLITE_UTF8, 0,
3356	writefileFunc, 0, 0);
3357	sqlite3_create_function(p->db, "sha3", 1, SQLITE_UTF8, 0,
3358	sha3Func, 0, 0);
3359	sqlite3_create_function(p->db, "sha3", 2, SQLITE_UTF8, 0,
3360	sha3Func, 0, 0);
3361	sqlite3_create_function(p->db, "sha3_query", 1, SQLITE_UTF8, 0,
3362	sha3QueryFunc, 0, 0);
3363	sqlite3_create_function(p->db, "sha3_query", 2, SQLITE_UTF8, 0,
3364	sha3QueryFunc, 0, 0);
3365	}
3366	}
3367
3368	/*
3369	** Do C-language style dequoting.
	@@ -2564,11 +3484,11 @@
3484
3485	/*
3486	** Interpret zArg as either an integer or a boolean value. Return 1 or 0
3487	** for TRUE and FALSE. Return the integer value if appropriate.
3488	*/
3489	static int booleanValue(const char *zArg){
3490	int i;
3491	if( zArg[0]=='0' && zArg[1]=='x' ){
3492	for(i=2; hexDigitValue(zArg[i])>=0; i++){}
3493	}else{
3494	for(i=0; zArg[i]>='0' && zArg[i]<='9'; i++){}
	@@ -2582,10 +3502,21 @@
3502	}
3503	utf8_printf(stderr, "ERROR: Not a boolean value: \"%s\". Assuming \"no\".\n",
3504	zArg);
3505	return 0;
3506	}
3507
3508	/*
3509	** Set or clear a shell flag according to a boolean value.
3510	*/
3511	static void setOrClearFlag(ShellState p, unsigned mFlag, const char zArg){
3512	if( booleanValue(zArg) ){
3513	ShellSetFlag(p, mFlag);
3514	}else{
3515	ShellClearFlag(p, mFlag);
3516	}
3517	}
3518
3519	/*
3520	** Close an output file, assuming it is not stderr or stdout
3521	*/
3522	static void output_file_close(FILE *f){
	@@ -3656,11 +4587,11 @@
4587	test_breakpoint();
4588	}else
4589
4590	if( c=='c' && n>=3 && strncmp(azArg[0], "changes", n)==0 ){
4591	if( nArg==2 ){
4592	setOrClearFlag(p, SHFLG_CountChanges, azArg[1]);
4593	}else{
4594	raw_printf(stderr, "Usage: .changes on\|off\n");
4595	rc = 1;
4596	}
4597	}else
	@@ -3720,25 +4651,46 @@
4651	if( c=='d' && strncmp(azArg[0], "dbinfo", n)==0 ){
4652	rc = shell_dbinfo_command(p, nArg, azArg);
4653	}else
4654
4655	if( c=='d' && strncmp(azArg[0], "dump", n)==0 ){
4656	const char *zLike = 0;
4657	int i;
4658	ShellClearFlag(p, SHFLG_PreserveRowid);
4659	for(i=1; i<nArg; i++){
4660	if( azArg[i][0]=='-' ){
4661	const char *z = azArg[i]+1;
4662	if( z[0]=='-' ) z++;
4663	if( strcmp(z,"preserve-rowids")==0 ){
4664	ShellSetFlag(p, SHFLG_PreserveRowid);
4665	}else
4666	{
4667	raw_printf(stderr, "Unknown option \"%s\" on \".dump\"\n", azArg[i]);
4668	rc = 1;
4669	goto meta_command_exit;
4670	}
4671	}else if( zLike ){
4672	raw_printf(stderr, "Usage: .dump ?--preserve-rowids? ?LIKE-PATTERN?\n");
4673	rc = 1;
4674	goto meta_command_exit;
4675	}else{
4676	zLike = azArg[i];
4677	}
4678	}
4679	open_db(p, 0);
4680	/* When playing back a "dump", the content might appear in an order
4681	** which causes immediate foreign key constraints to be violated.
4682	** So disable foreign-key constraint enforcement to prevent problems. */





4683	raw_printf(p->out, "PRAGMA foreign_keys=OFF;\n");
4684	raw_printf(p->out, "BEGIN TRANSACTION;\n");
4685	p->writableSchema = 0;
4686	/* Set writable_schema=ON since doing so forces SQLite to initialize
4687	** as much of the schema as it can even if the sqlite_master table is
4688	** corrupt. */
4689	sqlite3_exec(p->db, "SAVEPOINT dump; PRAGMA writable_schema=ON", 0, 0, 0);
4690	p->nErr = 0;
4691	if( zLike==0 ){
4692	run_schema_dump_query(p,
4693	"SELECT name, type, sql FROM sqlite_master "
4694	"WHERE sql NOT NULL AND type=='table' AND name!='sqlite_sequence'"
4695	);
4696	run_schema_dump_query(p,
	@@ -3748,25 +4700,24 @@
4700	run_table_dump_query(p,
4701	"SELECT sql FROM sqlite_master "
4702	"WHERE sql NOT NULL AND type IN ('index','trigger','view')", 0
4703	);
4704	}else{
4705	char *zSql;
4706	zSql = sqlite3_mprintf(
4707	"SELECT name, type, sql FROM sqlite_master "
4708	"WHERE tbl_name LIKE %Q AND type=='table'"
4709	" AND sql NOT NULL", zLike);
4710	run_schema_dump_query(p,zSql);
4711	sqlite3_free(zSql);
4712	zSql = sqlite3_mprintf(
4713	"SELECT sql FROM sqlite_master "
4714	"WHERE sql NOT NULL"
4715	" AND type IN ('index','trigger','view')"
4716	" AND tbl_name LIKE %Q", zLike);
4717	run_table_dump_query(p, zSql, 0);
4718	sqlite3_free(zSql);

4719	}
4720	if( p->writableSchema ){
4721	raw_printf(p->out, "PRAGMA writable_schema=OFF;\n");
4722	p->writableSchema = 0;
4723	}
	@@ -3775,11 +4726,11 @@
4726	raw_printf(p->out, p->nErr ? "ROLLBACK; -- due to errors\n" : "COMMIT;\n");
4727	}else
4728
4729	if( c=='e' && strncmp(azArg[0], "echo", n)==0 ){
4730	if( nArg==2 ){
4731	setOrClearFlag(p, SHFLG_Echo, azArg[1]);
4732	}else{
4733	raw_printf(stderr, "Usage: .echo on\|off\n");
4734	rc = 1;
4735	}
4736	}else
	@@ -4342,10 +5293,11 @@
5293	int newFlag = 0; /* True to delete file before opening */
5294	/* Close the existing database */
5295	session_close_all(p);
5296	sqlite3_close(p->db);
5297	p->db = 0;
5298	p->zDbFilename = 0;
5299	sqlite3_free(p->zFreeOnClose);
5300	p->zFreeOnClose = 0;
5301	/* Check for command-line arguments */
5302	for(iName=1; iName<nArg && azArg[iName][0]=='-'; iName++){
5303	const char *z = azArg[iName];
	@@ -4571,21 +5523,21 @@
5523	new_colv[0] = "sql";
5524	new_colv[1] = 0;
5525	callback(&data, 1, new_argv, new_colv);
5526	rc = SQLITE_OK;
5527	}else{
5528	char *zSql;
5529	zSql = sqlite3_mprintf(
5530	"SELECT sql FROM "
5531	" (SELECT sql sql, type type, tbl_name tbl_name, name name, rowid x"
5532	" FROM sqlite_master UNION ALL"
5533	" SELECT sql, type, tbl_name, name, rowid FROM sqlite_temp_master) "
5534	"WHERE lower(tbl_name) LIKE %Q"
5535	" AND type!='meta' AND sql NOTNULL "
5536	"ORDER BY rowid", azArg[1]);
5537	rc = sqlite3_exec(p->db, zSql, callback, &data, &zErrMsg);
5538	sqlite3_free(zSql);
5539	}
5540	}else if( nArg==1 ){
5541	rc = sqlite3_exec(p->db,
5542	"SELECT sql FROM "
5543	" (SELECT sql sql, type type, tbl_name tbl_name, name name, rowid x"
	@@ -4612,11 +5564,11 @@
5564	}
5565	}else
5566
5567	#if defined(SQLITE_DEBUG) && defined(SQLITE_ENABLE_SELECTTRACE)
5568	if( c=='s' && n==11 && strncmp(azArg[0], "selecttrace", n)==0 ){
5569	sqlite3SelectTrace = (int)integerValue(azArg[1]);
5570	}else
5571	#endif
5572
5573	#if defined(SQLITE_ENABLE_SESSION)
5574	if( c=='s' && strncmp(azArg[0],"session",n)==0 && n>=3 ){
	@@ -4834,10 +5786,123 @@
5786	utf8_printf(p->out, "%s", zBuf);
5787	}
5788	}
5789	}else
5790	#endif
5791
5792	if( c=='s' && n>=4 && strncmp(azArg[0],"selftest",n)==0 ){
5793	int bIsInit = 0; /* True to initialize the SELFTEST table */
5794	int bVerbose = 0; /* Verbose output */
5795	int bSelftestExists; /* True if SELFTEST already exists */
5796	char *azTest = 0; / Content of the SELFTEST table */
5797	int nRow = 0; /* Number of rows in the SELFTEST table */
5798	int nCol = 4; /* Number of columns in the SELFTEST table */
5799	int i; /* Loop counter */
5800	int nTest = 0; /* Number of tests runs */
5801	int nErr = 0; /* Number of errors seen */
5802	ShellText str; /* Answer for a query */
5803	static char *azDefaultTest[] = {
5804	0, 0, 0, 0,
5805	"0", "memo", "Missing SELFTEST table - default checks only", "",
5806	"1", "run", "PRAGMA integrity_check", "ok"
5807	};
5808	static const int nDefaultRow = 2;
5809
5810	open_db(p,0);
5811	for(i=1; i<nArg; i++){
5812	const char *z = azArg[i];
5813	if( z[0]=='-' && z[1]=='-' ) z++;
5814	if( strcmp(z,"-init")==0 ){
5815	bIsInit = 1;
5816	}else
5817	if( strcmp(z,"-v")==0 ){
5818	bVerbose++;
5819	}else
5820	{
5821	utf8_printf(stderr, "Unknown option \"%s\" on \"%s\"\n",
5822	azArg[i], azArg[0]);
5823	raw_printf(stderr, "Should be one of: --init -v\n");
5824	rc = 1;
5825	goto meta_command_exit;
5826	}
5827	}
5828	if( sqlite3_table_column_metadata(p->db,"main","selftest",0,0,0,0,0,0)
5829	!= SQLITE_OK ){
5830	bSelftestExists = 0;
5831	}else{
5832	bSelftestExists = 1;
5833	}
5834	if( bIsInit ){
5835	createSelftestTable(p);
5836	bSelftestExists = 1;
5837	}
5838	if( bSelftestExists ){
5839	rc = sqlite3_get_table(p->db,
5840	"SELECT tno,op,cmd,ans FROM selftest ORDER BY tno",
5841	&azTest, &nRow, &nCol, 0);
5842	if( rc ){
5843	raw_printf(stderr, "Error querying the selftest table\n");
5844	rc = 1;
5845	sqlite3_free_table(azTest);
5846	goto meta_command_exit;
5847	}else if( nRow==0 ){
5848	sqlite3_free_table(azTest);
5849	azTest = azDefaultTest;
5850	nRow = nDefaultRow;
5851	}
5852	}else{
5853	azTest = azDefaultTest;
5854	nRow = nDefaultRow;
5855	}
5856	initText(&str);
5857	appendText(&str, "x", 0);
5858	for(i=1; i<=nRow; i++){
5859	int tno = atoi(azTest[i*nCol]);
5860	const char zOp = azTest[inCol+1];
5861	const char zSql = azTest[inCol+2];
5862	const char zAns = azTest[inCol+3];
5863
5864	if( bVerbose>0 ){
5865	char *zQuote = sqlite3_mprintf("%q", zSql);
5866	printf("%d: %s %s\n", tno, zOp, zSql);
5867	sqlite3_free(zQuote);
5868	}
5869	if( strcmp(zOp,"memo")==0 ){
5870	utf8_printf(p->out, "%s\n", zSql);
5871	}else
5872	if( strcmp(zOp,"run")==0 ){
5873	char *zErrMsg = 0;
5874	str.n = 0;
5875	str.z[0] = 0;
5876	rc = sqlite3_exec(p->db, zSql, captureOutputCallback, &str, &zErrMsg);
5877	nTest++;
5878	if( bVerbose ){
5879	utf8_printf(p->out, "Result: %s\n", str.z);
5880	}
5881	if( rc \|\| zErrMsg ){
5882	nErr++;
5883	rc = 1;
5884	utf8_printf(p->out, "%d: error-code-%d: %s\n", tno, rc, zErrMsg);
5885	sqlite3_free(zErrMsg);
5886	}else if( strcmp(zAns,str.z)!=0 ){
5887	nErr++;
5888	rc = 1;
5889	utf8_printf(p->out, "%d: Expected: [%s]\n", tno, zAns);
5890	utf8_printf(p->out, "%d: Got: [%s]\n", tno, str.z);
5891	}
5892	}else
5893	{
5894	utf8_printf(stderr,
5895	"Unknown operation \"%s\" on selftest line %d\n", zOp, tno);
5896	rc = 1;
5897	break;
5898	}
5899	}
5900	freeText(&str);
5901	if( azTest!=azDefaultTest ) sqlite3_free_table(azTest);
5902	utf8_printf(p->out, "%d errors out of %d tests\n", nErr, nTest);
5903	}else
5904
5905	if( c=='s' && strncmp(azArg[0], "separator", n)==0 ){
5906	if( nArg<2 \|\| nArg>3 ){
5907	raw_printf(stderr, "Usage: .separator COL ?ROW?\n");
5908	rc = 1;
	@@ -4849,10 +5914,126 @@
5914	if( nArg>=3 ){
5915	sqlite3_snprintf(sizeof(p->rowSeparator), p->rowSeparator,
5916	"%.*s", (int)ArraySize(p->rowSeparator)-1, azArg[2]);
5917	}
5918	}else
5919
5920	if( c=='s' && n>=4 && strncmp(azArg[0],"sha3sum",n)==0 ){
5921	const char zLike = 0; / Which table to checksum. 0 means everything */
5922	int i; /* Loop counter */
5923	int bSchema = 0; /* Also hash the schema */
5924	int bSeparate = 0; /* Hash each table separately */
5925	int iSize = 224; /* Hash algorithm to use */
5926	int bDebug = 0; /* Only show the query that would have run */
5927	sqlite3_stmt pStmt; / For querying tables names */
5928	char zSql; / SQL to be run */
5929	char zSep; / Separator */
5930	ShellText sSql; /* Complete SQL for the query to run the hash */
5931	ShellText sQuery; /* Set of queries used to read all content */
5932	open_db(p, 0);
5933	for(i=1; i<nArg; i++){
5934	const char *z = azArg[i];
5935	if( z[0]=='-' ){
5936	z++;
5937	if( z[0]=='-' ) z++;
5938	if( strcmp(z,"schema")==0 ){
5939	bSchema = 1;
5940	}else
5941	if( strcmp(z,"sha3-224")==0 \|\| strcmp(z,"sha3-256")==0
5942	\|\| strcmp(z,"sha3-384")==0 \|\| strcmp(z,"sha3-512")==0
5943	){
5944	iSize = atoi(&z[5]);
5945	}else
5946	if( strcmp(z,"debug")==0 ){
5947	bDebug = 1;
5948	}else
5949	{
5950	utf8_printf(stderr, "Unknown option \"%s\" on \"%s\"\n",
5951	azArg[i], azArg[0]);
5952	raw_printf(stderr, "Should be one of: --schema"
5953	" --sha3-224 --sha3-255 --sha3-384 --sha3-512\n");
5954	rc = 1;
5955	goto meta_command_exit;
5956	}
5957	}else if( zLike ){
5958	raw_printf(stderr, "Usage: .sha3sum ?OPTIONS? ?LIKE-PATTERN?\n");
5959	rc = 1;
5960	goto meta_command_exit;
5961	}else{
5962	zLike = z;
5963	bSeparate = 1;
5964	if( sqlite3_strlike("sqlite_%", zLike, 0)==0 ) bSchema = 1;
5965	}
5966	}
5967	if( bSchema ){
5968	zSql = "SELECT lower(name) FROM sqlite_master"
5969	" WHERE type='table' AND coalesce(rootpage,0)>1"
5970	" UNION ALL SELECT 'sqlite_master'"
5971	" ORDER BY 1 collate nocase";
5972	}else{
5973	zSql = "SELECT lower(name) FROM sqlite_master"
5974	" WHERE type='table' AND coalesce(rootpage,0)>1"
5975	" AND name NOT LIKE 'sqlite_%'"
5976	" ORDER BY 1 collate nocase";
5977	}
5978	sqlite3_prepare_v2(p->db, zSql, -1, &pStmt, 0);
5979	initText(&sQuery);
5980	initText(&sSql);
5981	appendText(&sSql, "WITH [sha3sum$query](a,b) AS(",0);
5982	zSep = "VALUES(";
5983	while( SQLITE_ROW==sqlite3_step(pStmt) ){
5984	const char zTab = (const char)sqlite3_column_text(pStmt,0);
5985	if( zLike && sqlite3_strlike(zLike, zTab, 0)!=0 ) continue;
5986	if( strncmp(zTab, "sqlite_",7)!=0 ){
5987	appendText(&sQuery,"SELECT * FROM ", 0);
5988	appendText(&sQuery,zTab,'"');
5989	appendText(&sQuery," NOT INDEXED;", 0);
5990	}else if( strcmp(zTab, "sqlite_master")==0 ){
5991	appendText(&sQuery,"SELECT type,name,tbl_name,sql FROM sqlite_master"
5992	" ORDER BY name;", 0);
5993	}else if( strcmp(zTab, "sqlite_sequence")==0 ){
5994	appendText(&sQuery,"SELECT name,seq FROM sqlite_sequence"
5995	" ORDER BY name;", 0);
5996	}else if( strcmp(zTab, "sqlite_stat1")==0 ){
5997	appendText(&sQuery,"SELECT tbl,idx,stat FROM sqlite_stat1"
5998	" ORDER BY tbl,idx;", 0);
5999	}else if( strcmp(zTab, "sqlite_stat3")==0
6000	\|\| strcmp(zTab, "sqlite_stat4")==0 ){
6001	appendText(&sQuery, "SELECT * FROM ", 0);
6002	appendText(&sQuery, zTab, 0);
6003	appendText(&sQuery, " ORDER BY tbl, idx, rowid;\n", 0);
6004	}
6005	appendText(&sSql, zSep, 0);
6006	appendText(&sSql, sQuery.z, '\'');
6007	sQuery.n = 0;
6008	appendText(&sSql, ",", 0);
6009	appendText(&sSql, zTab, '\'');
6010	zSep = "),(";
6011	}
6012	sqlite3_finalize(pStmt);
6013	if( bSeparate ){
6014	zSql = sqlite3_mprintf(
6015	"%s))"
6016	" SELECT lower(hex(sha3_query(a,%d))) AS hash, b AS label"
6017	" FROM [sha3sum$query]",
6018	sSql.z, iSize);
6019	}else{
6020	zSql = sqlite3_mprintf(
6021	"%s))"
6022	" SELECT lower(hex(sha3_query(group_concat(a,''),%d))) AS hash"
6023	" FROM [sha3sum$query]",
6024	sSql.z, iSize);
6025	}
6026	freeText(&sQuery);
6027	freeText(&sSql);
6028	if( bDebug ){
6029	utf8_printf(p->out, "%s\n", zSql);
6030	}else{
6031	shell_exec(p->db, zSql, shell_callback, p, 0);
6032	}
6033	sqlite3_free(zSql);
6034	}else
6035
6036	if( c=='s'
6037	&& (strncmp(azArg[0], "shell", n)==0 \|\| strncmp(azArg[0],"system",n)==0)
6038	){
6039	char *zCmd;
	@@ -4878,11 +6059,12 @@
6059	if( nArg!=1 ){
6060	raw_printf(stderr, "Usage: .show\n");
6061	rc = 1;
6062	goto meta_command_exit;
6063	}
6064	utf8_printf(p->out, "%12.12s: %s\n","echo",
6065	azBool[ShellHasFlag(p, SHFLG_Echo)]);
6066	utf8_printf(p->out, "%12.12s: %s\n","eqp", azBool[p->autoEQP&3]);
6067	utf8_printf(p->out, "%12.12s: %s\n","explain",
6068	p->mode==MODE_Explain ? "on" : p->autoExplain ? "auto" : "off");
6069	utf8_printf(p->out,"%12.12s: %s\n","headers", azBool[p->showHeader!=0]);
6070	utf8_printf(p->out, "%12.12s: %s\n","mode", modeDescr[p->mode]);
	@@ -5190,10 +6372,11 @@
6372	azArg[1]);
6373	break;
6374	}
6375	}
6376	}else
6377	#endif /* !defined(SQLITE_UNTESTABLE) */
6378
6379	if( c=='t' && n>4 && strncmp(azArg[0], "timeout", n)==0 ){
6380	open_db(p, 0);
6381	sqlite3_busy_timeout(p->db, nArg>=2 ? (int)integerValue(azArg[1]) : 0);
6382	}else
	@@ -5226,11 +6409,10 @@
6409	}else{
6410	sqlite3_trace_v2(p->db, SQLITE_TRACE_STMT, sql_trace_callback,p->traceOut);
6411	}
6412	#endif
6413	}else

6414
6415	#if SQLITE_USER_AUTHENTICATION
6416	if( c=='u' && strncmp(azArg[0], "user", n)==0 ){
6417	if( nArg<2 ){
6418	raw_printf(stderr, "Usage: .user SUBCOMMAND ...\n");
	@@ -5443,11 +6625,11 @@
6625	static int runOneSqlLine(ShellState p, char zSql, FILE *in, int startline){
6626	int rc;
6627	char *zErrMsg = 0;
6628
6629	open_db(p, 0);
6630	if( ShellHasFlag(p,SHFLG_Backslash) ) resolve_backslashes(zSql);
6631	BEGIN_TIMER;
6632	rc = shell_exec(p->db, zSql, shell_callback, p, &zErrMsg);
6633	END_TIMER;
6634	if( rc \|\| zErrMsg ){
6635	char zPrefix[100];
	@@ -5463,11 +6645,11 @@
6645	zErrMsg = 0;
6646	}else{
6647	utf8_printf(stderr, "%s %s\n", zPrefix, sqlite3_errmsg(p->db));
6648	}
6649	return 1;
6650	}else if( ShellHasFlag(p, SHFLG_CountChanges) ){
6651	raw_printf(p->out, "changes: %3d total_changes: %d\n",
6652	sqlite3_changes(p->db), sqlite3_total_changes(p->db));
6653	}
6654	return 0;
6655	}
	@@ -5506,15 +6688,15 @@
6688	if( in!=0 ) break;
6689	seenInterrupt = 0;
6690	}
6691	lineno++;
6692	if( nSql==0 && _all_whitespace(zLine) ){
6693	if( ShellHasFlag(p, SHFLG_Echo) ) printf("%s\n", zLine);
6694	continue;
6695	}
6696	if( zLine && zLine[0]=='.' && nSql==0 ){
6697	if( ShellHasFlag(p, SHFLG_Echo) ) printf("%s\n", zLine);
6698	rc = do_meta_command(zLine, p);
6699	if( rc==2 ){ /* exit requested */
6700	break;
6701	}else if( rc ){
6702	errCnt++;
	@@ -5553,11 +6735,11 @@
6735	if( p->outCount ){
6736	output_reset(p);
6737	p->outCount = 0;
6738	}
6739	}else if( nSql && _all_whitespace(zSql) ){
6740	if( ShellHasFlag(p, SHFLG_Echo) ) printf("%s\n", zSql);
6741	nSql = 0;
6742	}
6743	}
6744	if( nSql && !_all_whitespace(zSql) ){
6745	runOneSqlLine(p, zSql, in, startline);
	@@ -6022,11 +7204,11 @@
7204	}else if( strcmp(z,"-header")==0 ){
7205	data.showHeader = 1;
7206	}else if( strcmp(z,"-noheader")==0 ){
7207	data.showHeader = 0;
7208	}else if( strcmp(z,"-echo")==0 ){
7209	ShellSetFlag(&data, SHFLG_Echo);
7210	}else if( strcmp(z,"-eqp")==0 ){
7211	data.autoEQP = 1;
7212	}else if( strcmp(z,"-eqpfull")==0 ){
7213	data.autoEQP = 2;
7214	}else if( strcmp(z,"-stats")==0 ){
	@@ -6037,11 +7219,11 @@
7219	/* Undocumented command-line option: -backslash
7220	** Causes C-style backslash escapes to be evaluated in SQL statements
7221	** prior to sending the SQL into SQLite. Useful for injecting
7222	** crazy bytes in the middle of SQL statements for testing and debugging.
7223	*/
7224	ShellSetFlag(&data, SHFLG_Backslash);
7225	}else if( strcmp(z,"-bail")==0 ){
7226	bail_on_error = 1;
7227	}else if( strcmp(z,"-version")==0 ){
7228	printf("%s %s\n", sqlite3_libversion(), sqlite3_sourceid());
7229	return 0;
7230

M src/shun.c

+14 -13

		--- src/shun.c
		+++ src/shun.c
		@@ -26,10 +26,11 @@
26	26	*/
27	27	int uuid_is_shunned(const char *zUuid){
28	28	static Stmt q;
29	29	int rc;
30	30	if( zUuid==0 \|\| zUuid[0]==0 ) return 0;
	31	+ if( g.eHashPolicy==HPOLICY_SHUN_SHA1 && zUuid[HNAME_LEN_SHA1]==0 ) return 1;
31	32	db_static_prepare(&q, "SELECT 1 FROM shun WHERE uuid=:uuid");
32	33	db_bind_text(&q, ":uuid", zUuid);
33	34	rc = db_step(&q);
34	35	db_reset(&q);
35	36	return rc==SQLITE_ROW;
		@@ -36,11 +37,11 @@
36	37	}
37	38
38	39	/*
39	40	** WEBPAGE: shun
40	41	**
41		-** View the SHA1 hashes of all shunned artifacts. Add new hashes
	42	+** View the hashes of all shunned artifacts. Add new hashes
42	43	** to the shun set. Requires Admin privilege.
43	44	*/
44	45	void shun_page(void){
45	46	Stmt q;
46	47	int cnt = 0;
		@@ -84,18 +85,18 @@
84	85	}
85	86	zCanonical[j+1] = zCanonical[j] = 0;
86	87	p = zCanonical;
87	88	while( *p ){
88	89	int nUuid = strlen(p);
89		- if( nUuid!=UUID_SIZE \|\| !validate16(p, nUuid) ){
	90	+ if( !hname_validate(p, nUuid) ){
90	91	@ <p class="generalError">Error: Bad artifact IDs.</p>
91	92	fossil_free(zCanonical);
92	93	zCanonical = 0;
93	94	break;
94	95	}else{
95		- canonical16(p, UUID_SIZE);
96		- p += UUID_SIZE+1;
	96	+ canonical16(p, nUuid);
	97	+ p += nUuid+1;
97	98	}
98	99	}
99	100	zUuid = zCanonical;
100	101	}
101	102	style_header("Shunned Artifacts");
		@@ -107,21 +108,21 @@
107	108	db_multi_exec("DELETE FROM shun WHERE uuid=%Q", p);
108	109	if( !db_exists("SELECT 1 FROM blob WHERE uuid=%Q", p) ){
109	110	allExist = 0;
110	111	}
111	112	admin_log("Unshunned %Q", p);
112		- p += UUID_SIZE+1;
	113	+ p += strlen(p)+1;
113	114	}
114	115	if( allExist ){
115	116	@ <p class="noMoreShun">Artifact(s)<br />
116		- for( p = zUuid ; *p ; p += UUID_SIZE+1 ){
	117	+ for( p = zUuid ; *p ; p += strlen(p)+1 ){
117	118	@ <a href="%R/artifact/%s(p)">%s(p)</a><br />
118	119	}
119	120	@ are no longer being shunned.</p>
120	121	}else{
121	122	@ <p class="noMoreShun">Artifact(s)<br />
122		- for( p = zUuid ; *p ; p += UUID_SIZE+1 ){
	123	+ for( p = zUuid ; *p ; p += strlen(p)+1 ){
123	124	@ %s(p)<br />
124	125	}
125	126	@ will no longer be shunned. But they may not exist in the repository.
126	127	@ It may be necessary to rebuild the repository using the
127	128	@ <b>fossil rebuild</b> command-line before the artifact content
		@@ -146,14 +147,14 @@
146	147	db_multi_exec("DELETE FROM ticket WHERE tkt_uuid=%Q", p);
147	148	db_multi_exec("DELETE FROM tag WHERE tagid=%d", tagid);
148	149	db_multi_exec("DELETE FROM tagxref WHERE tagid=%d", tagid);
149	150	}
150	151	admin_log("Shunned %Q", p);
151		- p += UUID_SIZE+1;
	152	+ p += strlen(p)+1;
152	153	}
153	154	@ <p class="shunned">Artifact(s)<br />
154		- for( p = zUuid ; *p ; p += UUID_SIZE+1 ){
	155	+ for( p = zUuid ; *p ; p += strlen(p)+1 ){
155	156	@ <a href="%R/artifact/%s(p)">%s(p)</a><br />
156	157	}
157	158	@ have been shunned. They will no longer be pushed.
158	159	@ They will be removed from the repository the next time the repository
159	160	@ is rebuilt using the <b>fossil rebuild</b> command-line</p>
		@@ -167,18 +168,18 @@
167	168	@ artifact content will be purged from the repository the next time the
168	169	@ repository is rebuilt. A list of shunned artifacts can be seen at the
169	170	@ bottom of this page.</p>
170	171	@
171	172	@ <a name="addshun"></a>
172		- @ <p>To shun artifacts, enter their artifact IDs (the 40-character SHA1
173		- @ hash of the artifacts) in the
	173	+ @ <p>To shun artifacts, enter their artifact hashes (the 40- or
	174	+ @ 64-character lowercase hexadecimal hash of the artifact content) in the
174	175	@ following box and press the "Shun" button. This will cause the artifacts
175	176	@ to be removed from the repository and will prevent the artifacts from being
176	177	@ readded to the repository by subsequent sync operation.</p>
177	178	@
178		- @ <p>Note that you must enter the full 40-character artifact IDs, not
179		- @ an abbreviation or a symbolic tag.</p>
	179	+ @ <p>Note that you must enter the full 40- or 64-character artifact hashes,
	180	+ @ not an abbreviation or a symbolic tag.</p>
180	181	@
181	182	@ <p>Warning: Shunning should only be used to remove inappropriate content
182	183	@ from the repository. Inappropriate content includes such things as
183	184	@ spam added to Wiki, files that violate copyright or patent agreements,
184	185	@ or artifacts that by design or accident interfere with the processing
185	186

	--- src/shun.c
	+++ src/shun.c
	@@ -26,10 +26,11 @@
26	*/
27	int uuid_is_shunned(const char *zUuid){
28	static Stmt q;
29	int rc;
30	if( zUuid==0 \|\| zUuid[0]==0 ) return 0;

31	db_static_prepare(&q, "SELECT 1 FROM shun WHERE uuid=:uuid");
32	db_bind_text(&q, ":uuid", zUuid);
33	rc = db_step(&q);
34	db_reset(&q);
35	return rc==SQLITE_ROW;
	@@ -36,11 +37,11 @@
36	}
37
38	/*
39	** WEBPAGE: shun
40	**
41	** View the SHA1 hashes of all shunned artifacts. Add new hashes
42	** to the shun set. Requires Admin privilege.
43	*/
44	void shun_page(void){
45	Stmt q;
46	int cnt = 0;
	@@ -84,18 +85,18 @@
84	}
85	zCanonical[j+1] = zCanonical[j] = 0;
86	p = zCanonical;
87	while( *p ){
88	int nUuid = strlen(p);
89	if( nUuid!=UUID_SIZE \|\| !validate16(p, nUuid) ){
90	@ <p class="generalError">Error: Bad artifact IDs.</p>
91	fossil_free(zCanonical);
92	zCanonical = 0;
93	break;
94	}else{
95	canonical16(p, UUID_SIZE);
96	p += UUID_SIZE+1;
97	}
98	}
99	zUuid = zCanonical;
100	}
101	style_header("Shunned Artifacts");
	@@ -107,21 +108,21 @@
107	db_multi_exec("DELETE FROM shun WHERE uuid=%Q", p);
108	if( !db_exists("SELECT 1 FROM blob WHERE uuid=%Q", p) ){
109	allExist = 0;
110	}
111	admin_log("Unshunned %Q", p);
112	p += UUID_SIZE+1;
113	}
114	if( allExist ){
115	@ <p class="noMoreShun">Artifact(s)<br />
116	for( p = zUuid ; *p ; p += UUID_SIZE+1 ){
117	@ <a href="%R/artifact/%s(p)">%s(p)</a><br />
118	}
119	@ are no longer being shunned.</p>
120	}else{
121	@ <p class="noMoreShun">Artifact(s)<br />
122	for( p = zUuid ; *p ; p += UUID_SIZE+1 ){
123	@ %s(p)<br />
124	}
125	@ will no longer be shunned. But they may not exist in the repository.
126	@ It may be necessary to rebuild the repository using the
127	@ <b>fossil rebuild</b> command-line before the artifact content
	@@ -146,14 +147,14 @@
146	db_multi_exec("DELETE FROM ticket WHERE tkt_uuid=%Q", p);
147	db_multi_exec("DELETE FROM tag WHERE tagid=%d", tagid);
148	db_multi_exec("DELETE FROM tagxref WHERE tagid=%d", tagid);
149	}
150	admin_log("Shunned %Q", p);
151	p += UUID_SIZE+1;
152	}
153	@ <p class="shunned">Artifact(s)<br />
154	for( p = zUuid ; *p ; p += UUID_SIZE+1 ){
155	@ <a href="%R/artifact/%s(p)">%s(p)</a><br />
156	}
157	@ have been shunned. They will no longer be pushed.
158	@ They will be removed from the repository the next time the repository
159	@ is rebuilt using the <b>fossil rebuild</b> command-line</p>
	@@ -167,18 +168,18 @@
167	@ artifact content will be purged from the repository the next time the
168	@ repository is rebuilt. A list of shunned artifacts can be seen at the
169	@ bottom of this page.</p>
170	@
171	@ <a name="addshun"></a>
172	@ <p>To shun artifacts, enter their artifact IDs (the 40-character SHA1
173	@ hash of the artifacts) in the
174	@ following box and press the "Shun" button. This will cause the artifacts
175	@ to be removed from the repository and will prevent the artifacts from being
176	@ readded to the repository by subsequent sync operation.</p>
177	@
178	@ <p>Note that you must enter the full 40-character artifact IDs, not
179	@ an abbreviation or a symbolic tag.</p>
180	@
181	@ <p>Warning: Shunning should only be used to remove inappropriate content
182	@ from the repository. Inappropriate content includes such things as
183	@ spam added to Wiki, files that violate copyright or patent agreements,
184	@ or artifacts that by design or accident interfere with the processing
185

	--- src/shun.c
	+++ src/shun.c
	@@ -26,10 +26,11 @@
26	*/
27	int uuid_is_shunned(const char *zUuid){
28	static Stmt q;
29	int rc;
30	if( zUuid==0 \|\| zUuid[0]==0 ) return 0;
31	if( g.eHashPolicy==HPOLICY_SHUN_SHA1 && zUuid[HNAME_LEN_SHA1]==0 ) return 1;
32	db_static_prepare(&q, "SELECT 1 FROM shun WHERE uuid=:uuid");
33	db_bind_text(&q, ":uuid", zUuid);
34	rc = db_step(&q);
35	db_reset(&q);
36	return rc==SQLITE_ROW;
	@@ -36,11 +37,11 @@
37	}
38
39	/*
40	** WEBPAGE: shun
41	**
42	** View the hashes of all shunned artifacts. Add new hashes
43	** to the shun set. Requires Admin privilege.
44	*/
45	void shun_page(void){
46	Stmt q;
47	int cnt = 0;
	@@ -84,18 +85,18 @@
85	}
86	zCanonical[j+1] = zCanonical[j] = 0;
87	p = zCanonical;
88	while( *p ){
89	int nUuid = strlen(p);
90	if( !hname_validate(p, nUuid) ){
91	@ <p class="generalError">Error: Bad artifact IDs.</p>
92	fossil_free(zCanonical);
93	zCanonical = 0;
94	break;
95	}else{
96	canonical16(p, nUuid);
97	p += nUuid+1;
98	}
99	}
100	zUuid = zCanonical;
101	}
102	style_header("Shunned Artifacts");
	@@ -107,21 +108,21 @@
108	db_multi_exec("DELETE FROM shun WHERE uuid=%Q", p);
109	if( !db_exists("SELECT 1 FROM blob WHERE uuid=%Q", p) ){
110	allExist = 0;
111	}
112	admin_log("Unshunned %Q", p);
113	p += strlen(p)+1;
114	}
115	if( allExist ){
116	@ <p class="noMoreShun">Artifact(s)<br />
117	for( p = zUuid ; *p ; p += strlen(p)+1 ){
118	@ <a href="%R/artifact/%s(p)">%s(p)</a><br />
119	}
120	@ are no longer being shunned.</p>
121	}else{
122	@ <p class="noMoreShun">Artifact(s)<br />
123	for( p = zUuid ; *p ; p += strlen(p)+1 ){
124	@ %s(p)<br />
125	}
126	@ will no longer be shunned. But they may not exist in the repository.
127	@ It may be necessary to rebuild the repository using the
128	@ <b>fossil rebuild</b> command-line before the artifact content
	@@ -146,14 +147,14 @@
147	db_multi_exec("DELETE FROM ticket WHERE tkt_uuid=%Q", p);
148	db_multi_exec("DELETE FROM tag WHERE tagid=%d", tagid);
149	db_multi_exec("DELETE FROM tagxref WHERE tagid=%d", tagid);
150	}
151	admin_log("Shunned %Q", p);
152	p += strlen(p)+1;
153	}
154	@ <p class="shunned">Artifact(s)<br />
155	for( p = zUuid ; *p ; p += strlen(p)+1 ){
156	@ <a href="%R/artifact/%s(p)">%s(p)</a><br />
157	}
158	@ have been shunned. They will no longer be pushed.
159	@ They will be removed from the repository the next time the repository
160	@ is rebuilt using the <b>fossil rebuild</b> command-line</p>
	@@ -167,18 +168,18 @@
168	@ artifact content will be purged from the repository the next time the
169	@ repository is rebuilt. A list of shunned artifacts can be seen at the
170	@ bottom of this page.</p>
171	@
172	@ <a name="addshun"></a>
173	@ <p>To shun artifacts, enter their artifact hashes (the 40- or
174	@ 64-character lowercase hexadecimal hash of the artifact content) in the
175	@ following box and press the "Shun" button. This will cause the artifacts
176	@ to be removed from the repository and will prevent the artifacts from being
177	@ readded to the repository by subsequent sync operation.</p>
178	@
179	@ <p>Note that you must enter the full 40- or 64-character artifact hashes,
180	@ not an abbreviation or a symbolic tag.</p>
181	@
182	@ <p>Warning: Shunning should only be used to remove inappropriate content
183	@ from the repository. Inappropriate content includes such things as
184	@ spam added to Wiki, files that violate copyright or patent agreements,
185	@ or artifacts that by design or accident interfere with the processing
186

M src/sitemap.c

+1 -2

		--- src/sitemap.c
		+++ src/sitemap.c
		@@ -111,12 +111,11 @@
111	111	}
112	112	@ <li>%z(href("%R/login"))Login/Logout/Change Password</a></li>
113	113	if( g.perm.Read ){
114	114	@ <li>%z(href("%R/stat"))Repository Status</a>
115	115	@ <ul>
116		- @ <li>%z(href("%R/hash-collisions"))Collisions on SHA1 hash
117		- @ prefixes</a></li>
	116	+ @ <li>%z(href("%R/hash-collisions"))Collisions on hash prefixes</a></li>
118	117	if( g.perm.Admin ){
119	118	@ <li>%z(href("%R/urllist"))List of URLs used to access
120	119	@ this repository</a></li>
121	120	}
122	121	@ <li>%z(href("%R/bloblist"))List of Artifacts</a></li>
123	122

	--- src/sitemap.c
	+++ src/sitemap.c
	@@ -111,12 +111,11 @@
111	}
112	@ <li>%z(href("%R/login"))Login/Logout/Change Password</a></li>
113	if( g.perm.Read ){
114	@ <li>%z(href("%R/stat"))Repository Status</a>
115	@ <ul>
116	@ <li>%z(href("%R/hash-collisions"))Collisions on SHA1 hash
117	@ prefixes</a></li>
118	if( g.perm.Admin ){
119	@ <li>%z(href("%R/urllist"))List of URLs used to access
120	@ this repository</a></li>
121	}
122	@ <li>%z(href("%R/bloblist"))List of Artifacts</a></li>
123

	--- src/sitemap.c
	+++ src/sitemap.c
	@@ -111,12 +111,11 @@
111	}
112	@ <li>%z(href("%R/login"))Login/Logout/Change Password</a></li>
113	if( g.perm.Read ){
114	@ <li>%z(href("%R/stat"))Repository Status</a>
115	@ <ul>
116	@ <li>%z(href("%R/hash-collisions"))Collisions on hash prefixes</a></li>

117	if( g.perm.Admin ){
118	@ <li>%z(href("%R/urllist"))List of URLs used to access
119	@ this repository</a></li>
120	}
121	@ <li>%z(href("%R/bloblist"))List of Artifacts</a></li>
122

M src/sqlcmd.c

+5 -2

		--- src/sqlcmd.c
		+++ src/sqlcmd.c
		@@ -185,12 +185,12 @@
185	185	** in ways that are unrecoverable. Be sure you know what you are doing before
186	186	** running any SQL commands that modify the repository database.
187	187	**
188	188	** The following extensions to the usual SQLite commands are provided:
189	189	**
190		-** content(X) Return the content of artifact X. X can be a
191		-** SHA1 hash or prefix or a tag.
	190	+** content(X) Return the content of artifact X. X can be an
	191	+** artifact hash or prefix or a tag.
192	192	**
193	193	** compress(X) Compress text X.
194	194	**
195	195	** decompress(X) Decompress text X. Undoes the work of
196	196	** compress(X).
		@@ -212,10 +212,13 @@
212	212	*/
213	213	void cmd_sqlite3(void){
214	214	int noRepository;
215	215	const char *zConfigDb;
216	216	extern int sqlite3_shell(int, char**);
	217	+#ifdef FOSSIL_ENABLE_TH1_HOOKS
	218	+ g.fNoThHook = 1;
	219	+#endif
217	220	noRepository = find_option("no-repository", 0, 0)!=0;
218	221	if( !noRepository ){
219	222	db_find_and_open_repository(OPEN_ANY_SCHEMA, 0);
220	223	}
221	224	db_open_config(1,0);
222	225

	--- src/sqlcmd.c
	+++ src/sqlcmd.c
	@@ -185,12 +185,12 @@
185	** in ways that are unrecoverable. Be sure you know what you are doing before
186	** running any SQL commands that modify the repository database.
187	**
188	** The following extensions to the usual SQLite commands are provided:
189	**
190	** content(X) Return the content of artifact X. X can be a
191	** SHA1 hash or prefix or a tag.
192	**
193	** compress(X) Compress text X.
194	**
195	** decompress(X) Decompress text X. Undoes the work of
196	** compress(X).
	@@ -212,10 +212,13 @@
212	*/
213	void cmd_sqlite3(void){
214	int noRepository;
215	const char *zConfigDb;
216	extern int sqlite3_shell(int, char**);



217	noRepository = find_option("no-repository", 0, 0)!=0;
218	if( !noRepository ){
219	db_find_and_open_repository(OPEN_ANY_SCHEMA, 0);
220	}
221	db_open_config(1,0);
222

	--- src/sqlcmd.c
	+++ src/sqlcmd.c
	@@ -185,12 +185,12 @@
185	** in ways that are unrecoverable. Be sure you know what you are doing before
186	** running any SQL commands that modify the repository database.
187	**
188	** The following extensions to the usual SQLite commands are provided:
189	**
190	** content(X) Return the content of artifact X. X can be an
191	** artifact hash or prefix or a tag.
192	**
193	** compress(X) Compress text X.
194	**
195	** decompress(X) Decompress text X. Undoes the work of
196	** compress(X).
	@@ -212,10 +212,13 @@
212	*/
213	void cmd_sqlite3(void){
214	int noRepository;
215	const char *zConfigDb;
216	extern int sqlite3_shell(int, char**);
217	#ifdef FOSSIL_ENABLE_TH1_HOOKS
218	g.fNoThHook = 1;
219	#endif
220	noRepository = find_option("no-repository", 0, 0)!=0;
221	if( !noRepository ){
222	db_find_and_open_repository(OPEN_ANY_SCHEMA, 0);
223	}
224	db_open_config(1,0);
225

M src/sqlite3.c

+1515 -987

		--- src/sqlite3.c
		+++ src/sqlite3.c
		@@ -1,8 +1,8 @@
1	1	/******************************************************************************
2	2	** This file is an amalgamation of many separate C source files from SQLite
3		-** version 3.17.0. By combining all the individual C code files into this
	3	+** version 3.18.0. By combining all the individual C code files into this
4	4	** single large file, the entire code can be compiled as a single translation
5	5	** unit. This allows many compilers to do optimizations that would not be
6	6	** possible if the files were compiled separately. Performance improvements
7	7	** of 5% or more are commonly seen when SQLite is compiled as a single
8	8	** translation unit.
		@@ -396,13 +396,13 @@
396	396	**
397	397	** See also: [sqlite3_libversion()],
398	398	** [sqlite3_libversion_number()], [sqlite3_sourceid()],
399	399	** [sqlite_version()] and [sqlite_source_id()].
400	400	*/
401		-#define SQLITE_VERSION "3.17.0"
402		-#define SQLITE_VERSION_NUMBER 3017000
403		-#define SQLITE_SOURCE_ID "2017-02-13 16:02:40 ada05cfa86ad7f5645450ac7a2a21c9aa6e57d2c"
	401	+#define SQLITE_VERSION "3.18.0"
	402	+#define SQLITE_VERSION_NUMBER 3018000
	403	+#define SQLITE_SOURCE_ID "2017-03-10 17:03:11 f8560c60d10c0365b33342ab05b5a953987b0471"
404	404
405	405	/*
406	406	** CAPI3REF: Run-Time Library Version Numbers
407	407	** KEYWORDS: sqlite3_version sqlite3_sourceid
408	408	**
		@@ -2315,24 +2315,34 @@
2315	2315	** as an undeclared column named ROWID, OID, or _ROWID_ as long as those
2316	2316	** names are not also used by explicitly declared columns. ^If
2317	2317	** the table has a column of type [INTEGER PRIMARY KEY] then that column
2318	2318	** is another alias for the rowid.
2319	2319	**
2320		-** ^The sqlite3_last_insert_rowid(D) interface returns the [rowid] of the
2321		-** most recent successful [INSERT] into a rowid table or [virtual table]
2322		-** on database connection D.
2323		-** ^Inserts into [WITHOUT ROWID] tables are not recorded.
2324		-** ^If no successful [INSERT]s into rowid tables
2325		-** have ever occurred on the database connection D,
2326		-** then sqlite3_last_insert_rowid(D) returns zero.
	2320	+** ^The sqlite3_last_insert_rowid(D) interface usually returns the [rowid] of
	2321	+** the most recent successful [INSERT] into a rowid table or [virtual table]
	2322	+** on database connection D. ^Inserts into [WITHOUT ROWID] tables are not
	2323	+** recorded. ^If no successful [INSERT]s into rowid tables have ever occurred
	2324	+** on the database connection D, then sqlite3_last_insert_rowid(D) returns
	2325	+** zero.
2327	2326	**
2328		-** ^(If an [INSERT] occurs within a trigger or within a [virtual table]
2329		-** method, then this routine will return the [rowid] of the inserted
2330		-** row as long as the trigger or virtual table method is running.
2331		-** But once the trigger or virtual table method ends, the value returned
2332		-** by this routine reverts to what it was before the trigger or virtual
2333		-** table method began.)^
	2327	+** As well as being set automatically as rows are inserted into database
	2328	+** tables, the value returned by this function may be set explicitly by
	2329	+** [sqlite3_set_last_insert_rowid()]
	2330	+**
	2331	+** Some virtual table implementations may INSERT rows into rowid tables as
	2332	+** part of committing a transaction (e.g. to flush data accumulated in memory
	2333	+** to disk). In this case subsequent calls to this function return the rowid
	2334	+** associated with these internal INSERT operations, which leads to
	2335	+** unintuitive results. Virtual table implementations that do write to rowid
	2336	+** tables in this way can avoid this problem by restoring the original
	2337	+** rowid value using [sqlite3_set_last_insert_rowid()] before returning
	2338	+** control to the user.
	2339	+**
	2340	+** ^(If an [INSERT] occurs within a trigger then this routine will
	2341	+** return the [rowid] of the inserted row as long as the trigger is
	2342	+** running. Once the trigger program ends, the value returned
	2343	+** by this routine reverts to what it was before the trigger was fired.)^
2334	2344	**
2335	2345	** ^An [INSERT] that fails due to a constraint violation is not a
2336	2346	** successful [INSERT] and does not change the value returned by this
2337	2347	** routine. ^Thus INSERT OR FAIL, INSERT OR IGNORE, INSERT OR ROLLBACK,
2338	2348	** and INSERT OR ABORT make no changes to the return value of this
		@@ -2355,10 +2365,20 @@
2355	2365	** unpredictable and might not equal either the old or the new
2356	2366	** last insert [rowid].
2357	2367	*/
2358	2368	SQLITE_API sqlite3_int64 sqlite3_last_insert_rowid(sqlite3*);
2359	2369
	2370	+/*
	2371	+** CAPI3REF: Set the Last Insert Rowid value.
	2372	+** METHOD: sqlite3
	2373	+**
	2374	+** The sqlite3_set_last_insert_rowid(D, R) method allows the application to
	2375	+** set the value returned by calling sqlite3_last_insert_rowid(D) to R
	2376	+** without inserting a row into the database.
	2377	+*/
	2378	+SQLITE_API void sqlite3_set_last_insert_rowid(sqlite3*,sqlite3_int64);
	2379	+
2360	2380	/*
2361	2381	** CAPI3REF: Count The Number Of Rows Modified
2362	2382	** METHOD: sqlite3
2363	2383	**
2364	2384	** ^This function returns the number of rows modified, inserted or
		@@ -12457,10 +12477,11 @@
12457	12477	SQLITE_PRIVATE const void sqlite3BtreePayloadFetch(BtCursor, u32 *pAmt);
12458	12478	SQLITE_PRIVATE u32 sqlite3BtreePayloadSize(BtCursor*);
12459	12479
12460	12480	SQLITE_PRIVATE char sqlite3BtreeIntegrityCheck(Btree, int aRoot, int nRoot, int, int);
12461	12481	SQLITE_PRIVATE struct Pager sqlite3BtreePager(Btree);
	12482	+SQLITE_PRIVATE i64 sqlite3BtreeRowCountEst(BtCursor*);
12462	12483
12463	12484	#ifndef SQLITE_OMIT_INCRBLOB
12464	12485	SQLITE_PRIVATE int sqlite3BtreePayloadChecked(BtCursor, u32 offset, u32 amt, void);
12465	12486	SQLITE_PRIVATE int sqlite3BtreePutData(BtCursor, u32 offset, u32 amt, void);
12466	12487	SQLITE_PRIVATE void sqlite3BtreeIncrblobCursor(BtCursor *);
		@@ -12757,117 +12778,119 @@
12757	12778	#define OP_Divide 50 /* same as TK_SLASH, synopsis: r[P3]=r[P2]/r[P1] */
12758	12779	#define OP_Remainder 51 /* same as TK_REM, synopsis: r[P3]=r[P2]%r[P1] */
12759	12780	#define OP_Concat 52 /* same as TK_CONCAT, synopsis: r[P3]=r[P2]+r[P1] */
12760	12781	#define OP_Last 53
12761	12782	#define OP_BitNot 54 /* same as TK_BITNOT, synopsis: r[P1]= ~r[P1] */
12762		-#define OP_SorterSort 55
12763		-#define OP_Sort 56
12764		-#define OP_Rewind 57
12765		-#define OP_IdxLE 58 /* synopsis: key=r[P3@P4] */
12766		-#define OP_IdxGT 59 /* synopsis: key=r[P3@P4] */
12767		-#define OP_IdxLT 60 /* synopsis: key=r[P3@P4] */
12768		-#define OP_IdxGE 61 /* synopsis: key=r[P3@P4] */
12769		-#define OP_RowSetRead 62 /* synopsis: r[P3]=rowset(P1) */
12770		-#define OP_RowSetTest 63 /* synopsis: if r[P3] in rowset(P1) goto P2 */
12771		-#define OP_Program 64
12772		-#define OP_FkIfZero 65 /* synopsis: if fkctr[P1]==0 goto P2 */
12773		-#define OP_IfPos 66 /* synopsis: if r[P1]>0 then r[P1]-=P3, goto P2 */
12774		-#define OP_IfNotZero 67 /* synopsis: if r[P1]!=0 then r[P1]--, goto P2 */
12775		-#define OP_DecrJumpZero 68 /* synopsis: if (--r[P1])==0 goto P2 */
12776		-#define OP_IncrVacuum 69
12777		-#define OP_VNext 70
12778		-#define OP_Init 71 /* synopsis: Start at P2 */
12779		-#define OP_Return 72
12780		-#define OP_EndCoroutine 73
12781		-#define OP_HaltIfNull 74 /* synopsis: if r[P3]=null halt */
12782		-#define OP_Halt 75
12783		-#define OP_Integer 76 /* synopsis: r[P2]=P1 */
12784		-#define OP_Int64 77 /* synopsis: r[P2]=P4 */
12785		-#define OP_String 78 /* synopsis: r[P2]='P4' (len=P1) */
12786		-#define OP_Null 79 /* synopsis: r[P2..P3]=NULL */
12787		-#define OP_SoftNull 80 /* synopsis: r[P1]=NULL */
12788		-#define OP_Blob 81 /* synopsis: r[P2]=P4 (len=P1) */
12789		-#define OP_Variable 82 /* synopsis: r[P2]=parameter(P1,P4) */
12790		-#define OP_Move 83 /* synopsis: r[P2@P3]=r[P1@P3] */
12791		-#define OP_Copy 84 /* synopsis: r[P2@P3+1]=r[P1@P3+1] */
12792		-#define OP_SCopy 85 /* synopsis: r[P2]=r[P1] */
12793		-#define OP_IntCopy 86 /* synopsis: r[P2]=r[P1] */
12794		-#define OP_ResultRow 87 /* synopsis: output=r[P1@P2] */
12795		-#define OP_CollSeq 88
12796		-#define OP_Function0 89 /* synopsis: r[P3]=func(r[P2@P5]) */
12797		-#define OP_Function 90 /* synopsis: r[P3]=func(r[P2@P5]) */
12798		-#define OP_AddImm 91 /* synopsis: r[P1]=r[P1]+P2 */
12799		-#define OP_RealAffinity 92
12800		-#define OP_Cast 93 /* synopsis: affinity(r[P1]) */
12801		-#define OP_Permutation 94
12802		-#define OP_Compare 95 /* synopsis: r[P1@P3] <-> r[P2@P3] */
12803		-#define OP_Column 96 /* synopsis: r[P3]=PX */
	12783	+#define OP_IfSmaller 55
	12784	+#define OP_SorterSort 56
	12785	+#define OP_Sort 57
	12786	+#define OP_Rewind 58
	12787	+#define OP_IdxLE 59 /* synopsis: key=r[P3@P4] */
	12788	+#define OP_IdxGT 60 /* synopsis: key=r[P3@P4] */
	12789	+#define OP_IdxLT 61 /* synopsis: key=r[P3@P4] */
	12790	+#define OP_IdxGE 62 /* synopsis: key=r[P3@P4] */
	12791	+#define OP_RowSetRead 63 /* synopsis: r[P3]=rowset(P1) */
	12792	+#define OP_RowSetTest 64 /* synopsis: if r[P3] in rowset(P1) goto P2 */
	12793	+#define OP_Program 65
	12794	+#define OP_FkIfZero 66 /* synopsis: if fkctr[P1]==0 goto P2 */
	12795	+#define OP_IfPos 67 /* synopsis: if r[P1]>0 then r[P1]-=P3, goto P2 */
	12796	+#define OP_IfNotZero 68 /* synopsis: if r[P1]!=0 then r[P1]--, goto P2 */
	12797	+#define OP_DecrJumpZero 69 /* synopsis: if (--r[P1])==0 goto P2 */
	12798	+#define OP_IncrVacuum 70
	12799	+#define OP_VNext 71
	12800	+#define OP_Init 72 /* synopsis: Start at P2 */
	12801	+#define OP_Return 73
	12802	+#define OP_EndCoroutine 74
	12803	+#define OP_HaltIfNull 75 /* synopsis: if r[P3]=null halt */
	12804	+#define OP_Halt 76
	12805	+#define OP_Integer 77 /* synopsis: r[P2]=P1 */
	12806	+#define OP_Int64 78 /* synopsis: r[P2]=P4 */
	12807	+#define OP_String 79 /* synopsis: r[P2]='P4' (len=P1) */
	12808	+#define OP_Null 80 /* synopsis: r[P2..P3]=NULL */
	12809	+#define OP_SoftNull 81 /* synopsis: r[P1]=NULL */
	12810	+#define OP_Blob 82 /* synopsis: r[P2]=P4 (len=P1) */
	12811	+#define OP_Variable 83 /* synopsis: r[P2]=parameter(P1,P4) */
	12812	+#define OP_Move 84 /* synopsis: r[P2@P3]=r[P1@P3] */
	12813	+#define OP_Copy 85 /* synopsis: r[P2@P3+1]=r[P1@P3+1] */
	12814	+#define OP_SCopy 86 /* synopsis: r[P2]=r[P1] */
	12815	+#define OP_IntCopy 87 /* synopsis: r[P2]=r[P1] */
	12816	+#define OP_ResultRow 88 /* synopsis: output=r[P1@P2] */
	12817	+#define OP_CollSeq 89
	12818	+#define OP_Function0 90 /* synopsis: r[P3]=func(r[P2@P5]) */
	12819	+#define OP_Function 91 /* synopsis: r[P3]=func(r[P2@P5]) */
	12820	+#define OP_AddImm 92 /* synopsis: r[P1]=r[P1]+P2 */
	12821	+#define OP_RealAffinity 93
	12822	+#define OP_Cast 94 /* synopsis: affinity(r[P1]) */
	12823	+#define OP_Permutation 95
	12824	+#define OP_Compare 96 /* synopsis: r[P1@P3] <-> r[P2@P3] */
12804	12825	#define OP_String8 97 /* same as TK_STRING, synopsis: r[P2]='P4' */
12805		-#define OP_Affinity 98 /* synopsis: affinity(r[P1@P2]) */
12806		-#define OP_MakeRecord 99 /* synopsis: r[P3]=mkrec(r[P1@P2]) */
12807		-#define OP_Count 100 /* synopsis: r[P2]=count() */
12808		-#define OP_ReadCookie 101
12809		-#define OP_SetCookie 102
12810		-#define OP_ReopenIdx 103 /* synopsis: root=P2 iDb=P3 */
12811		-#define OP_OpenRead 104 /* synopsis: root=P2 iDb=P3 */
12812		-#define OP_OpenWrite 105 /* synopsis: root=P2 iDb=P3 */
12813		-#define OP_OpenAutoindex 106 /* synopsis: nColumn=P2 */
12814		-#define OP_OpenEphemeral 107 /* synopsis: nColumn=P2 */
12815		-#define OP_SorterOpen 108
12816		-#define OP_SequenceTest 109 /* synopsis: if( cursor[P1].ctr++ ) pc = P2 */
12817		-#define OP_OpenPseudo 110 /* synopsis: P3 columns in r[P2] */
12818		-#define OP_Close 111
12819		-#define OP_ColumnsUsed 112
12820		-#define OP_Sequence 113 /* synopsis: r[P2]=cursor[P1].ctr++ */
12821		-#define OP_NewRowid 114 /* synopsis: r[P2]=rowid */
12822		-#define OP_Insert 115 /* synopsis: intkey=r[P3] data=r[P2] */
12823		-#define OP_InsertInt 116 /* synopsis: intkey=P3 data=r[P2] */
12824		-#define OP_Delete 117
12825		-#define OP_ResetCount 118
12826		-#define OP_SorterCompare 119 /* synopsis: if key(P1)!=trim(r[P3],P4) goto P2 */
12827		-#define OP_SorterData 120 /* synopsis: r[P2]=data */
12828		-#define OP_RowData 121 /* synopsis: r[P2]=data */
12829		-#define OP_Rowid 122 /* synopsis: r[P2]=rowid */
12830		-#define OP_NullRow 123
12831		-#define OP_SorterInsert 124 /* synopsis: key=r[P2] */
12832		-#define OP_IdxInsert 125 /* synopsis: key=r[P2] */
12833		-#define OP_IdxDelete 126 /* synopsis: key=r[P2@P3] */
12834		-#define OP_Seek 127 /* synopsis: Move P3 to P1.rowid */
12835		-#define OP_IdxRowid 128 /* synopsis: r[P2]=rowid */
12836		-#define OP_Destroy 129
12837		-#define OP_Clear 130
12838		-#define OP_ResetSorter 131
	12826	+#define OP_Column 98 /* synopsis: r[P3]=PX */
	12827	+#define OP_Affinity 99 /* synopsis: affinity(r[P1@P2]) */
	12828	+#define OP_MakeRecord 100 /* synopsis: r[P3]=mkrec(r[P1@P2]) */
	12829	+#define OP_Count 101 /* synopsis: r[P2]=count() */
	12830	+#define OP_ReadCookie 102
	12831	+#define OP_SetCookie 103
	12832	+#define OP_ReopenIdx 104 /* synopsis: root=P2 iDb=P3 */
	12833	+#define OP_OpenRead 105 /* synopsis: root=P2 iDb=P3 */
	12834	+#define OP_OpenWrite 106 /* synopsis: root=P2 iDb=P3 */
	12835	+#define OP_OpenAutoindex 107 /* synopsis: nColumn=P2 */
	12836	+#define OP_OpenEphemeral 108 /* synopsis: nColumn=P2 */
	12837	+#define OP_SorterOpen 109
	12838	+#define OP_SequenceTest 110 /* synopsis: if( cursor[P1].ctr++ ) pc = P2 */
	12839	+#define OP_OpenPseudo 111 /* synopsis: P3 columns in r[P2] */
	12840	+#define OP_Close 112
	12841	+#define OP_ColumnsUsed 113
	12842	+#define OP_Sequence 114 /* synopsis: r[P2]=cursor[P1].ctr++ */
	12843	+#define OP_NewRowid 115 /* synopsis: r[P2]=rowid */
	12844	+#define OP_Insert 116 /* synopsis: intkey=r[P3] data=r[P2] */
	12845	+#define OP_InsertInt 117 /* synopsis: intkey=P3 data=r[P2] */
	12846	+#define OP_Delete 118
	12847	+#define OP_ResetCount 119
	12848	+#define OP_SorterCompare 120 /* synopsis: if key(P1)!=trim(r[P3],P4) goto P2 */
	12849	+#define OP_SorterData 121 /* synopsis: r[P2]=data */
	12850	+#define OP_RowData 122 /* synopsis: r[P2]=data */
	12851	+#define OP_Rowid 123 /* synopsis: r[P2]=rowid */
	12852	+#define OP_NullRow 124
	12853	+#define OP_SorterInsert 125 /* synopsis: key=r[P2] */
	12854	+#define OP_IdxInsert 126 /* synopsis: key=r[P2] */
	12855	+#define OP_IdxDelete 127 /* synopsis: key=r[P2@P3] */
	12856	+#define OP_Seek 128 /* synopsis: Move P3 to P1.rowid */
	12857	+#define OP_IdxRowid 129 /* synopsis: r[P2]=rowid */
	12858	+#define OP_Destroy 130
	12859	+#define OP_Clear 131
12839	12860	#define OP_Real 132 /* same as TK_FLOAT, synopsis: r[P2]=P4 */
12840		-#define OP_CreateIndex 133 /* synopsis: r[P2]=root iDb=P1 */
12841		-#define OP_CreateTable 134 /* synopsis: r[P2]=root iDb=P1 */
12842		-#define OP_ParseSchema 135
12843		-#define OP_LoadAnalysis 136
12844		-#define OP_DropTable 137
12845		-#define OP_DropIndex 138
12846		-#define OP_DropTrigger 139
12847		-#define OP_IntegrityCk 140
12848		-#define OP_RowSetAdd 141 /* synopsis: rowset(P1)=r[P2] */
12849		-#define OP_Param 142
12850		-#define OP_FkCounter 143 /* synopsis: fkctr[P1]+=P2 */
12851		-#define OP_MemMax 144 /* synopsis: r[P1]=max(r[P1],r[P2]) */
12852		-#define OP_OffsetLimit 145 /* synopsis: if r[P1]>0 then r[P2]=r[P1]+max(0,r[P3]) else r[P2]=(-1) */
12853		-#define OP_AggStep0 146 /* synopsis: accum=r[P3] step(r[P2@P5]) */
12854		-#define OP_AggStep 147 /* synopsis: accum=r[P3] step(r[P2@P5]) */
12855		-#define OP_AggFinal 148 /* synopsis: accum=r[P1] N=P2 */
12856		-#define OP_Expire 149
12857		-#define OP_TableLock 150 /* synopsis: iDb=P1 root=P2 write=P3 */
12858		-#define OP_VBegin 151
12859		-#define OP_VCreate 152
12860		-#define OP_VDestroy 153
12861		-#define OP_VOpen 154
12862		-#define OP_VColumn 155 /* synopsis: r[P3]=vcolumn(P2) */
12863		-#define OP_VRename 156
12864		-#define OP_Pagecount 157
12865		-#define OP_MaxPgcnt 158
12866		-#define OP_CursorHint 159
12867		-#define OP_Noop 160
12868		-#define OP_Explain 161
	12861	+#define OP_ResetSorter 133
	12862	+#define OP_CreateIndex 134 /* synopsis: r[P2]=root iDb=P1 */
	12863	+#define OP_CreateTable 135 /* synopsis: r[P2]=root iDb=P1 */
	12864	+#define OP_SqlExec 136
	12865	+#define OP_ParseSchema 137
	12866	+#define OP_LoadAnalysis 138
	12867	+#define OP_DropTable 139
	12868	+#define OP_DropIndex 140
	12869	+#define OP_DropTrigger 141
	12870	+#define OP_IntegrityCk 142
	12871	+#define OP_RowSetAdd 143 /* synopsis: rowset(P1)=r[P2] */
	12872	+#define OP_Param 144
	12873	+#define OP_FkCounter 145 /* synopsis: fkctr[P1]+=P2 */
	12874	+#define OP_MemMax 146 /* synopsis: r[P1]=max(r[P1],r[P2]) */
	12875	+#define OP_OffsetLimit 147 /* synopsis: if r[P1]>0 then r[P2]=r[P1]+max(0,r[P3]) else r[P2]=(-1) */
	12876	+#define OP_AggStep0 148 /* synopsis: accum=r[P3] step(r[P2@P5]) */
	12877	+#define OP_AggStep 149 /* synopsis: accum=r[P3] step(r[P2@P5]) */
	12878	+#define OP_AggFinal 150 /* synopsis: accum=r[P1] N=P2 */
	12879	+#define OP_Expire 151
	12880	+#define OP_TableLock 152 /* synopsis: iDb=P1 root=P2 write=P3 */
	12881	+#define OP_VBegin 153
	12882	+#define OP_VCreate 154
	12883	+#define OP_VDestroy 155
	12884	+#define OP_VOpen 156
	12885	+#define OP_VColumn 157 /* synopsis: r[P3]=vcolumn(P2) */
	12886	+#define OP_VRename 158
	12887	+#define OP_Pagecount 159
	12888	+#define OP_MaxPgcnt 160
	12889	+#define OP_CursorHint 161
	12890	+#define OP_Noop 162
	12891	+#define OP_Explain 163
12869	12892
12870	12893	/* Properties such as "out2" or "jump" that are specified in
12871	12894	** comments following the "case" for each opcode in the vdbe.c
12872	12895	** are encoded into bitvectors as follows:
12873	12896	*/
		@@ -12883,32 +12906,32 @@
12883	12906	/* 16 */ 0x03, 0x03, 0x01, 0x12, 0x01, 0x03, 0x03, 0x09,\
12884	12907	/* 24 */ 0x09, 0x09, 0x09, 0x26, 0x26, 0x09, 0x09, 0x09,\
12885	12908	/* 32 */ 0x09, 0x09, 0x03, 0x03, 0x0b, 0x0b, 0x0b, 0x0b,\
12886	12909	/* 40 */ 0x0b, 0x0b, 0x01, 0x26, 0x26, 0x26, 0x26, 0x26,\
12887	12910	/* 48 */ 0x26, 0x26, 0x26, 0x26, 0x26, 0x01, 0x12, 0x01,\
12888		-/* 56 */ 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x23, 0x0b,\
12889		-/* 64 */ 0x01, 0x01, 0x03, 0x03, 0x03, 0x01, 0x01, 0x01,\
12890		-/* 72 */ 0x02, 0x02, 0x08, 0x00, 0x10, 0x10, 0x10, 0x10,\
12891		-/* 80 */ 0x00, 0x10, 0x10, 0x00, 0x00, 0x10, 0x10, 0x00,\
12892		-/* 88 */ 0x00, 0x00, 0x00, 0x02, 0x02, 0x02, 0x00, 0x00,\
12893		-/* 96 */ 0x00, 0x10, 0x00, 0x00, 0x10, 0x10, 0x00, 0x00,\
	12911	+/* 56 */ 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x23,\
	12912	+/* 64 */ 0x0b, 0x01, 0x01, 0x03, 0x03, 0x03, 0x01, 0x01,\
	12913	+/* 72 */ 0x01, 0x02, 0x02, 0x08, 0x00, 0x10, 0x10, 0x10,\
	12914	+/* 80 */ 0x10, 0x00, 0x10, 0x10, 0x00, 0x00, 0x10, 0x10,\
	12915	+/* 88 */ 0x00, 0x00, 0x00, 0x00, 0x02, 0x02, 0x02, 0x00,\
	12916	+/* 96 */ 0x00, 0x10, 0x00, 0x00, 0x00, 0x10, 0x10, 0x00,\
12894	12917	/* 104 */ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,\
12895		-/* 112 */ 0x00, 0x10, 0x10, 0x00, 0x00, 0x00, 0x00, 0x00,\
12896		-/* 120 */ 0x00, 0x00, 0x10, 0x00, 0x04, 0x04, 0x00, 0x00,\
12897		-/* 128 */ 0x10, 0x10, 0x00, 0x00, 0x10, 0x10, 0x10, 0x00,\
12898		-/* 136 */ 0x00, 0x00, 0x00, 0x00, 0x00, 0x06, 0x10, 0x00,\
12899		-/* 144 */ 0x04, 0x1a, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,\
12900		-/* 152 */ 0x00, 0x00, 0x00, 0x00, 0x00, 0x10, 0x10, 0x00,\
12901		-/* 160 */ 0x00, 0x00,}
	12918	+/* 112 */ 0x00, 0x00, 0x10, 0x10, 0x00, 0x00, 0x00, 0x00,\
	12919	+/* 120 */ 0x00, 0x00, 0x00, 0x10, 0x00, 0x04, 0x04, 0x00,\
	12920	+/* 128 */ 0x00, 0x10, 0x10, 0x00, 0x10, 0x00, 0x10, 0x10,\
	12921	+/* 136 */ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x06,\
	12922	+/* 144 */ 0x10, 0x00, 0x04, 0x1a, 0x00, 0x00, 0x00, 0x00,\
	12923	+/* 152 */ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x10,\
	12924	+/* 160 */ 0x10, 0x00, 0x00, 0x00,}
12902	12925
12903	12926	/* The sqlite3P2Values() routine is able to run faster if it knows
12904	12927	** the value of the largest JUMP opcode. The smaller the maximum
12905	12928	** JUMP opcode the better, so the mkopcodeh.tcl script that
12906	12929	** generated this include file strives to group all JUMP opcodes
12907	12930	** together near the beginning of the list.
12908	12931	*/
12909		-#define SQLITE_MX_JUMP_OPCODE 71 /* Maximum JUMP opcode */
	12932	+#define SQLITE_MX_JUMP_OPCODE 72 /* Maximum JUMP opcode */
12910	12933
12911	12934	/************ End of opcodes.h *******************************************/
12912	12935	/************ Continuing where we left off in vdbe.h *********************/
12913	12936
12914	12937	/*
		@@ -14074,10 +14097,11 @@
14074	14097	u8 suppressErr; /* Do not issue error messages if true */
14075	14098	u8 vtabOnConflict; /* Value to return for s3_vtab_on_conflict() */
14076	14099	u8 isTransactionSavepoint; /* True if the outermost savepoint is a TS */
14077	14100	u8 mTrace; /* zero or more SQLITE_TRACE flags */
14078	14101	u8 skipBtreeMutex; /* True if no shared-cache backends */
	14102	+ u8 nSqlExec; /* Number of pending OP_SqlExec opcodes */
14079	14103	int nextPagesize; /* Pagesize after VACUUM if >0 */
14080	14104	u32 magic; /* Magic number for detect library misuse */
14081	14105	int nChange; /* Value returned by sqlite3_changes() */
14082	14106	int nTotalChange; /* Value returned by sqlite3_total_changes() */
14083	14107	int aLimit[SQLITE_N_LIMIT]; /* Limits */
		@@ -14589,18 +14613,18 @@
14589	14613	char zColAff; / String defining the affinity of each column */
14590	14614	ExprList pCheck; / All CHECK constraints */
14591	14615	/* ... also used as column name list in a VIEW */
14592	14616	int tnum; /* Root BTree page for this table */
14593	14617	u32 nTabRef; /* Number of pointers to this Table */
	14618	+ u32 tabFlags; /* Mask of TF_* values */
14594	14619	i16 iPKey; /* If not negative, use aCol[iPKey] as the rowid */
14595	14620	i16 nCol; /* Number of columns in this table */
14596	14621	LogEst nRowLogEst; /* Estimated rows in table - from sqlite_stat1 table */
14597	14622	LogEst szTabRow; /* Estimated size of each table row in bytes */
14598	14623	#ifdef SQLITE_ENABLE_COSTMULT
14599	14624	LogEst costMult; /* Cost multiplier for using this table */
14600	14625	#endif
14601		- u8 tabFlags; /* Mask of TF_* values */
14602	14626	u8 keyConf; /* What to do in case of uniqueness conflict on iPKey */
14603	14627	#ifndef SQLITE_OMIT_ALTERTABLE
14604	14628	int addColOffset; /* Offset in CREATE TABLE stmt to add a new column */
14605	14629	#endif
14606	14630	#ifndef SQLITE_OMIT_VIRTUALTABLE
		@@ -14620,27 +14644,29 @@
14620	14644	** followed by non-hidden columns. Example: "CREATE VIRTUAL TABLE x USING
14621	14645	** vtab1(a HIDDEN, b);". Since "b" is a non-hidden column but "a" is hidden,
14622	14646	** the TF_OOOHidden attribute would apply in this case. Such tables require
14623	14647	** special handling during INSERT processing.
14624	14648	*/
14625		-#define TF_Readonly 0x01 /* Read-only system table */
14626		-#define TF_Ephemeral 0x02 /* An ephemeral table */
14627		-#define TF_HasPrimaryKey 0x04 /* Table has a primary key */
14628		-#define TF_Autoincrement 0x08 /* Integer primary key is autoincrement */
14629		-#define TF_Virtual 0x10 /* Is a virtual table */
14630		-#define TF_WithoutRowid 0x20 /* No rowid. PRIMARY KEY is the key */
14631		-#define TF_NoVisibleRowid 0x40 /* No user-visible "rowid" column */
14632		-#define TF_OOOHidden 0x80 /* Out-of-Order hidden columns */
14633		-
	14649	+#define TF_Readonly 0x0001 /* Read-only system table */
	14650	+#define TF_Ephemeral 0x0002 /* An ephemeral table */
	14651	+#define TF_HasPrimaryKey 0x0004 /* Table has a primary key */
	14652	+#define TF_Autoincrement 0x0008 /* Integer primary key is autoincrement */
	14653	+#define TF_HasStat1 0x0010 /* nRowLogEst set from sqlite_stat1 */
	14654	+#define TF_WithoutRowid 0x0020 /* No rowid. PRIMARY KEY is the key */
	14655	+#define TF_NoVisibleRowid 0x0040 /* No user-visible "rowid" column */
	14656	+#define TF_OOOHidden 0x0080 /* Out-of-Order hidden columns */
	14657	+#define TF_StatsUsed 0x0100 /* Query planner decisions affected by
	14658	+ ** Index.aiRowLogEst[] values */
	14659	+#define TF_HasNotNull 0x0200 /* Contains NOT NULL constraints */
14634	14660
14635	14661	/*
14636	14662	** Test to see whether or not a table is a virtual table. This is
14637	14663	** done as a macro so that it will be optimized out when virtual
14638	14664	** table support is omitted from the build.
14639	14665	*/
14640	14666	#ifndef SQLITE_OMIT_VIRTUALTABLE
14641		-# define IsVirtual(X) (((X)->tabFlags & TF_Virtual)!=0)
	14667	+# define IsVirtual(X) ((X)->nModuleArg)
14642	14668	#else
14643	14669	# define IsVirtual(X) 0
14644	14670	#endif
14645	14671
14646	14672	/*
		@@ -14871,10 +14897,11 @@
14871	14897	unsigned bUnordered:1; /* Use this index for == or IN queries only */
14872	14898	unsigned uniqNotNull:1; /* True if UNIQUE and NOT NULL for all columns */
14873	14899	unsigned isResized:1; /* True if resizeIndexObject() has been called */
14874	14900	unsigned isCovering:1; /* True if this is a covering index */
14875	14901	unsigned noSkipScan:1; /* Do not try to use skip-scan if true */
	14902	+ unsigned hasStat1:1; /* aiRowLogEst values come from sqlite_stat1 */
14876	14903	#ifdef SQLITE_ENABLE_STAT3_OR_STAT4
14877	14904	int nSample; /* Number of elements in aSample[] */
14878	14905	int nSampleCol; /* Size of IndexSample.anEq[] and so on */
14879	14906	tRowcnt aAvgEq; / Average nEq values for keys not in aSample */
14880	14907	IndexSample aSample; / Samples of the left-most key */
		@@ -15181,11 +15208,11 @@
15181	15208	** form is used for name resolution with nested FROM clauses.
15182	15209	*/
15183	15210	struct ExprList {
15184	15211	int nExpr; /* Number of expressions on the list */
15185	15212	struct ExprList_item { /* For each expression in the list */
15186		- Expr pExpr; / The list of expressions */
	15213	+ Expr pExpr; / The parse tree for this expression */
15187	15214	char zName; / Token associated with this expression */
15188	15215	char zSpan; / Original text of the expression */
15189	15216	u8 sortOrder; /* 1 for DESC or 0 for ASC */
15190	15217	unsigned done :1; /* A flag to indicate when processing is finished */
15191	15218	unsigned bSpanIsTab :1; /* zSpan holds DB.TABLE.COLUMN */
		@@ -16504,10 +16531,11 @@
16504	16531	SQLITE_PRIVATE void sqlite3UnlinkAndDeleteIndex(sqlite3,int,const char);
16505	16532	SQLITE_PRIVATE void sqlite3Vacuum(Parse,Token);
16506	16533	SQLITE_PRIVATE int sqlite3RunVacuum(char*, sqlite3, int);
16507	16534	SQLITE_PRIVATE char sqlite3NameFromToken(sqlite3, Token*);
16508	16535	SQLITE_PRIVATE int sqlite3ExprCompare(Expr, Expr, int);
	16536	+SQLITE_PRIVATE int sqlite3ExprCompareSkip(Expr, Expr, int);
16509	16537	SQLITE_PRIVATE int sqlite3ExprListCompare(ExprList, ExprList, int);
16510	16538	SQLITE_PRIVATE int sqlite3ExprImpliesExpr(Expr, Expr, int);
16511	16539	SQLITE_PRIVATE void sqlite3ExprAnalyzeAggregates(NameContext, Expr);
16512	16540	SQLITE_PRIVATE void sqlite3ExprAnalyzeAggList(NameContext,ExprList);
16513	16541	SQLITE_PRIVATE int sqlite3ExprCoveredByIndex(Expr, int iCur, Index pIdx);
		@@ -17450,11 +17478,11 @@
17450	17478	"COMPILER=gcc-" __VERSION__,
17451	17479	#endif
17452	17480	#if SQLITE_COVERAGE_TEST
17453	17481	"COVERAGE_TEST",
17454	17482	#endif
17455		-#if SQLITE_DEBUG
	17483	+#ifdef SQLITE_DEBUG
17456	17484	"DEBUG",
17457	17485	#endif
17458	17486	#if SQLITE_DEFAULT_LOCKING_MODE
17459	17487	"DEFAULT_LOCKING_MODE=" CTIMEOPT_VAL(SQLITE_DEFAULT_LOCKING_MODE),
17460	17488	#endif
		@@ -19505,22 +19533,23 @@
19505	19533	**
19506	19534	** Move the date backwards to the beginning of the current day,
19507	19535	** or month or year.
19508	19536	*/
19509	19537	if( sqlite3_strnicmp(z, "start of ", 9)!=0 ) break;
	19538	+ if( !p->validJD && !p->validYMD && !p->validHMS ) break;
19510	19539	z += 9;
19511	19540	computeYMD(p);
19512	19541	p->validHMS = 1;
19513	19542	p->h = p->m = 0;
19514	19543	p->s = 0.0;
	19544	+ p->rawS = 0;
19515	19545	p->validTZ = 0;
19516	19546	p->validJD = 0;
19517	19547	if( sqlite3_stricmp(z,"month")==0 ){
19518	19548	p->D = 1;
19519	19549	rc = 0;
19520	19550	}else if( sqlite3_stricmp(z,"year")==0 ){
19521		- computeYMD(p);
19522	19551	p->M = 1;
19523	19552	p->D = 1;
19524	19553	rc = 0;
19525	19554	}else if( sqlite3_stricmp(z,"day")==0 ){
19526	19555	rc = 0;
		@@ -24050,12 +24079,12 @@
24050	24079	#ifdef SQLITE_DEBUG
24051	24080	assert( p->nRef>0 \|\| p->owner==0 );
24052	24081	p->owner = tid;
24053	24082	p->nRef++;
24054	24083	if( p->trace ){
24055		- OSTRACE(("ENTER-MUTEX tid=%lu, mutex=%p (%d), nRef=%d\n",
24056		- tid, p, p->trace, p->nRef));
	24084	+ OSTRACE(("ENTER-MUTEX tid=%lu, mutex(%d)=%p (%d), nRef=%d\n",
	24085	+ tid, p->id, p, p->trace, p->nRef));
24057	24086	}
24058	24087	#endif
24059	24088	}
24060	24089
24061	24090	static int winMutexTry(sqlite3_mutex *p){
		@@ -24093,12 +24122,12 @@
24093	24122	#else
24094	24123	UNUSED_PARAMETER(p);
24095	24124	#endif
24096	24125	#ifdef SQLITE_DEBUG
24097	24126	if( p->trace ){
24098		- OSTRACE(("TRY-MUTEX tid=%lu, mutex=%p (%d), owner=%lu, nRef=%d, rc=%s\n",
24099		- tid, p, p->trace, p->owner, p->nRef, sqlite3ErrName(rc)));
	24127	+ OSTRACE(("TRY-MUTEX tid=%lu, mutex(%d)=%p (%d), owner=%lu, nRef=%d, rc=%s\n",
	24128	+ tid, p->id, p, p->trace, p->owner, p->nRef, sqlite3ErrName(rc)));
24100	24129	}
24101	24130	#endif
24102	24131	return rc;
24103	24132	}
24104	24133
		@@ -24122,12 +24151,12 @@
24122	24151	#endif
24123	24152	assert( winMutex_isInit==1 );
24124	24153	LeaveCriticalSection(&p->mutex);
24125	24154	#ifdef SQLITE_DEBUG
24126	24155	if( p->trace ){
24127		- OSTRACE(("LEAVE-MUTEX tid=%lu, mutex=%p (%d), nRef=%d\n",
24128		- tid, p, p->trace, p->nRef));
	24156	+ OSTRACE(("LEAVE-MUTEX tid=%lu, mutex(%d)=%p (%d), nRef=%d\n",
	24157	+ tid, p->id, p, p->trace, p->nRef));
24129	24158	}
24130	24159	#endif
24131	24160	}
24132	24161
24133	24162	SQLITE_PRIVATE sqlite3_mutex_methods const *sqlite3DefaultMutex(void){
		@@ -24383,10 +24412,17 @@
24383	24412	** implementation of malloc_good_size(), which must be called in debug
24384	24413	** mode and specifically when the DMD "Dark Matter Detector" is enabled
24385	24414	** or else a crash results. Hence, do not attempt to optimize out the
24386	24415	** following xRoundup() call. */
24387	24416	nFull = sqlite3GlobalConfig.m.xRoundup(n);
	24417	+
	24418	+#ifdef SQLITE_MAX_MEMORY
	24419	+ if( sqlite3StatusValue(SQLITE_STATUS_MEMORY_USED)+nFull>SQLITE_MAX_MEMORY ){
	24420	+ *pp = 0;
	24421	+ return;
	24422	+ }
	24423	+#endif
24388	24424
24389	24425	sqlite3StatusHighwater(SQLITE_STATUS_MALLOC_SIZE, n);
24390	24426	if( mem0.alarmThreshold>0 ){
24391	24427	sqlite3_int64 nUsed = sqlite3StatusValue(SQLITE_STATUS_MEMORY_USED);
24392	24428	if( nUsed >= mem0.alarmThreshold - nFull ){
		@@ -24572,11 +24608,11 @@
24572	24608	return sqlite3GlobalConfig.m.xSize(p);
24573	24609	}
24574	24610	SQLITE_PRIVATE int sqlite3DbMallocSize(sqlite3 db, void p){
24575	24611	assert( p!=0 );
24576	24612	if( db==0 \|\| !isLookaside(db,p) ){
24577		-#if SQLITE_DEBUG
	24613	+#ifdef SQLITE_DEBUG
24578	24614	if( db==0 ){
24579	24615	assert( sqlite3MemdebugNoType(p, (u8)~MEMTYPE_HEAP) );
24580	24616	assert( sqlite3MemdebugHasType(p, MEMTYPE_HEAP) );
24581	24617	}else{
24582	24618	assert( sqlite3MemdebugHasType(p, (MEMTYPE_LOOKASIDE\|MEMTYPE_HEAP)) );
		@@ -24633,11 +24669,11 @@
24633	24669	measureAllocationSize(db, p);
24634	24670	return;
24635	24671	}
24636	24672	if( isLookaside(db, p) ){
24637	24673	LookasideSlot pBuf = (LookasideSlot)p;
24638		-#if SQLITE_DEBUG
	24674	+#ifdef SQLITE_DEBUG
24639	24675	/* Trash all content in the buffer being freed */
24640	24676	memset(p, 0xaa, db->lookaside.sz);
24641	24677	#endif
24642	24678	pBuf->pNext = db->lookaside.pFree;
24643	24679	db->lookaside.pFree = pBuf;
		@@ -25002,11 +25038,11 @@
25002	25038
25003	25039	/*
25004	25040	** Conversion types fall into various categories as defined by the
25005	25041	** following enumeration.
25006	25042	*/
25007		-#define etRADIX 0 /* Integer types. %d, %x, %o, and so forth */
	25043	+#define etRADIX 0 /* non-decimal integer types. %x %o */
25008	25044	#define etFLOAT 1 /* Floating point. %f */
25009	25045	#define etEXP 2 /* Exponentional notation. %e and %E */
25010	25046	#define etGENERIC 3 /* Floating or exponential, depending on exponent. %g */
25011	25047	#define etSIZE 4 /* Return number of characters processed so far. %n */
25012	25048	#define etSTRING 5 /* Strings. %s */
		@@ -25020,12 +25056,13 @@
25020	25056	#define etTOKEN 11 /* a pointer to a Token structure */
25021	25057	#define etSRCLIST 12 /* a pointer to a SrcList */
25022	25058	#define etPOINTER 13 /* The %p conversion */
25023	25059	#define etSQLESCAPE3 14 /* %w -> Strings with '\"' doubled */
25024	25060	#define etORDINAL 15 /* %r -> 1st, 2nd, 3rd, 4th, etc. English only */
	25061	+#define etDECIMAL 16 /* %d or %u, but not %x, %o */
25025	25062
25026		-#define etINVALID 16 /* Any unrecognized conversion type */
	25063	+#define etINVALID 17 /* Any unrecognized conversion type */
25027	25064
25028	25065
25029	25066	/*
25030	25067	** An "etByte" is an 8-bit unsigned value.
25031	25068	*/
		@@ -25045,40 +25082,40 @@
25045	25082	} et_info;
25046	25083
25047	25084	/*
25048	25085	** Allowed values for et_info.flags
25049	25086	*/
25050		-#define FLAG_SIGNED 1 /* True if the value to convert is signed */
25051		-#define FLAG_STRING 4 /* Allow infinity precision */
	25087	+#define FLAG_SIGNED 1 /* True if the value to convert is signed */
	25088	+#define FLAG_STRING 4 /* Allow infinite precision */
25052	25089
25053	25090
25054	25091	/*
25055	25092	** The following table is searched linearly, so it is good to put the
25056	25093	** most frequently used conversion types first.
25057	25094	*/
25058	25095	static const char aDigits[] = "0123456789ABCDEF0123456789abcdef";
25059	25096	static const char aPrefix[] = "-x0\000X0";
25060	25097	static const et_info fmtinfo[] = {
25061		- { 'd', 10, 1, etRADIX, 0, 0 },
	25098	+ { 'd', 10, 1, etDECIMAL, 0, 0 },
25062	25099	{ 's', 0, 4, etSTRING, 0, 0 },
25063	25100	{ 'g', 0, 1, etGENERIC, 30, 0 },
25064	25101	{ 'z', 0, 4, etDYNSTRING, 0, 0 },
25065	25102	{ 'q', 0, 4, etSQLESCAPE, 0, 0 },
25066	25103	{ 'Q', 0, 4, etSQLESCAPE2, 0, 0 },
25067	25104	{ 'w', 0, 4, etSQLESCAPE3, 0, 0 },
25068	25105	{ 'c', 0, 0, etCHARX, 0, 0 },
25069	25106	{ 'o', 8, 0, etRADIX, 0, 2 },
25070		- { 'u', 10, 0, etRADIX, 0, 0 },
	25107	+ { 'u', 10, 0, etDECIMAL, 0, 0 },
25071	25108	{ 'x', 16, 0, etRADIX, 16, 1 },
25072	25109	{ 'X', 16, 0, etRADIX, 0, 4 },
25073	25110	#ifndef SQLITE_OMIT_FLOATING_POINT
25074	25111	{ 'f', 0, 1, etFLOAT, 0, 0 },
25075	25112	{ 'e', 0, 1, etEXP, 30, 0 },
25076	25113	{ 'E', 0, 1, etEXP, 14, 0 },
25077	25114	{ 'G', 0, 1, etGENERIC, 14, 0 },
25078	25115	#endif
25079		- { 'i', 10, 1, etRADIX, 0, 0 },
	25116	+ { 'i', 10, 1, etDECIMAL, 0, 0 },
25080	25117	{ 'n', 0, 0, etSIZE, 0, 0 },
25081	25118	{ '%', 0, 0, etPERCENT, 0, 0 },
25082	25119	{ 'p', 16, 0, etPOINTER, 0, 1 },
25083	25120
25084	25121	/* All the rest are undocumented and are for internal use only */
		@@ -25166,18 +25203,17 @@
25166	25203	int precision; /* Precision of the current field */
25167	25204	int length; /* Length of the field */
25168	25205	int idx; /* A general purpose loop counter */
25169	25206	int width; /* Width of the current field */
25170	25207	etByte flag_leftjustify; /* True if "-" flag is present */
25171		- etByte flag_plussign; /* True if "+" flag is present */
25172		- etByte flag_blanksign; /* True if " " flag is present */
	25208	+ etByte flag_prefix; /* '+' or ' ' or 0 for prefix */
25173	25209	etByte flag_alternateform; /* True if "#" flag is present */
25174	25210	etByte flag_altform2; /* True if "!" flag is present */
25175	25211	etByte flag_zeropad; /* True if field width constant starts with zero */
25176		- etByte flag_long; /* True if "l" flag is present */
25177		- etByte flag_longlong; /* True if the "ll" flag is present */
	25212	+ etByte flag_long; /* 1 for the "l" flag, 2 for "ll", 0 by default */
25178	25213	etByte done; /* Loop termination flag */
	25214	+ etByte cThousand; /* Thousands separator for %d and %u */
25179	25215	etByte xtype = etINVALID; /* Conversion paradigm */
25180	25216	u8 bArgList; /* True for SQLITE_PRINTF_SQLFUNC */
25181	25217	char prefix; /* Prefix character. "+" or "-" or " " or '\0'. */
25182	25218	sqlite_uint64 longvalue; /* Value for integer types */
25183	25219	LONGDOUBLE_TYPE realvalue; /* Value for real types */
		@@ -25216,21 +25252,22 @@
25216	25252	if( (c=(*++fmt))==0 ){
25217	25253	sqlite3StrAccumAppend(pAccum, "%", 1);
25218	25254	break;
25219	25255	}
25220	25256	/* Find out what flags are present */
25221		- flag_leftjustify = flag_plussign = flag_blanksign =
	25257	+ flag_leftjustify = flag_prefix = cThousand =
25222	25258	flag_alternateform = flag_altform2 = flag_zeropad = 0;
25223	25259	done = 0;
25224	25260	do{
25225	25261	switch( c ){
25226	25262	case '-': flag_leftjustify = 1; break;
25227		- case '+': flag_plussign = 1; break;
25228		- case ' ': flag_blanksign = 1; break;
	25263	+ case '+': flag_prefix = '+'; break;
	25264	+ case ' ': flag_prefix = ' '; break;
25229	25265	case '#': flag_alternateform = 1; break;
25230	25266	case '!': flag_altform2 = 1; break;
25231	25267	case '0': flag_zeropad = 1; break;
	25268	+ case ',': cThousand = ','; break;
25232	25269	default: done = 1; break;
25233	25270	}
25234	25271	}while( !done && (c=(*++fmt))!=0 );
25235	25272	/* Get the field width */
25236	25273	if( c=='*' ){
		@@ -25296,17 +25333,15 @@
25296	25333	/* Get the conversion type modifier */
25297	25334	if( c=='l' ){
25298	25335	flag_long = 1;
25299	25336	c = *++fmt;
25300	25337	if( c=='l' ){
25301		- flag_longlong = 1;
	25338	+ flag_long = 2;
25302	25339	c = *++fmt;
25303		- }else{
25304		- flag_longlong = 0;
25305	25340	}
25306	25341	}else{
25307		- flag_long = flag_longlong = 0;
	25342	+ flag_long = 0;
25308	25343	}
25309	25344	/* Fetch the info entry for the field */
25310	25345	infop = &fmtinfo[0];
25311	25346	xtype = etINVALID;
25312	25347	for(idx=0; idx<ArraySize(fmtinfo); idx++){
		@@ -25320,41 +25355,42 @@
25320	25355	/*
25321	25356	** At this point, variables are initialized as follows:
25322	25357	**
25323	25358	** flag_alternateform TRUE if a '#' is present.
25324	25359	** flag_altform2 TRUE if a '!' is present.
25325		- ** flag_plussign TRUE if a '+' is present.
	25360	+ ** flag_prefix '+' or ' ' or zero
25326	25361	** flag_leftjustify TRUE if a '-' is present or if the
25327	25362	** field width was negative.
25328	25363	** flag_zeropad TRUE if the width began with 0.
25329		- ** flag_long TRUE if the letter 'l' (ell) prefixed
25330		- ** the conversion character.
25331		- ** flag_longlong TRUE if the letter 'll' (ell ell) prefixed
25332		- ** the conversion character.
25333		- ** flag_blanksign TRUE if a ' ' is present.
	25364	+ ** flag_long 1 for "l", 2 for "ll"
25334	25365	** width The specified field width. This is
25335	25366	** always non-negative. Zero is the default.
25336	25367	** precision The specified precision. The default
25337	25368	** is -1.
25338	25369	** xtype The class of the conversion.
25339	25370	** infop Pointer to the appropriate info struct.
25340	25371	*/
25341	25372	switch( xtype ){
25342	25373	case etPOINTER:
25343		- flag_longlong = sizeof(char*)==sizeof(i64);
25344		- flag_long = sizeof(char*)==sizeof(long int);
	25374	+ flag_long = sizeof(char*)==sizeof(i64) ? 2 :
	25375	+ sizeof(char*)==sizeof(long int) ? 1 : 0;
25345	25376	/* Fall through into the next case */
25346	25377	case etORDINAL:
25347		- case etRADIX:
	25378	+ case etRADIX:
	25379	+ cThousand = 0;
	25380	+ /* Fall through into the next case */
	25381	+ case etDECIMAL:
25348	25382	if( infop->flags & FLAG_SIGNED ){
25349	25383	i64 v;
25350	25384	if( bArgList ){
25351	25385	v = getIntArg(pArgList);
25352		- }else if( flag_longlong ){
25353		- v = va_arg(ap,i64);
25354	25386	}else if( flag_long ){
25355		- v = va_arg(ap,long int);
	25387	+ if( flag_long==2 ){
	25388	+ v = va_arg(ap,i64) ;
	25389	+ }else{
	25390	+ v = va_arg(ap,long int);
	25391	+ }
25356	25392	}else{
25357	25393	v = va_arg(ap,int);
25358	25394	}
25359	25395	if( v<0 ){
25360	25396	if( v==SMALLEST_INT64 ){
		@@ -25363,35 +25399,35 @@
25363	25399	longvalue = -v;
25364	25400	}
25365	25401	prefix = '-';
25366	25402	}else{
25367	25403	longvalue = v;
25368		- if( flag_plussign ) prefix = '+';
25369		- else if( flag_blanksign ) prefix = ' ';
25370		- else prefix = 0;
	25404	+ prefix = flag_prefix;
25371	25405	}
25372	25406	}else{
25373	25407	if( bArgList ){
25374	25408	longvalue = (u64)getIntArg(pArgList);
25375		- }else if( flag_longlong ){
25376		- longvalue = va_arg(ap,u64);
25377	25409	}else if( flag_long ){
25378		- longvalue = va_arg(ap,unsigned long int);
	25410	+ if( flag_long==2 ){
	25411	+ longvalue = va_arg(ap,u64);
	25412	+ }else{
	25413	+ longvalue = va_arg(ap,unsigned long int);
	25414	+ }
25379	25415	}else{
25380	25416	longvalue = va_arg(ap,unsigned int);
25381	25417	}
25382	25418	prefix = 0;
25383	25419	}
25384	25420	if( longvalue==0 ) flag_alternateform = 0;
25385	25421	if( flag_zeropad && precision<width-(prefix!=0) ){
25386	25422	precision = width-(prefix!=0);
25387	25423	}
25388		- if( precision<etBUFSIZE-10 ){
	25424	+ if( precision<etBUFSIZE-10-etBUFSIZE/3 ){
25389	25425	nOut = etBUFSIZE;
25390	25426	zOut = buf;
25391	25427	}else{
25392		- nOut = precision + 10;
	25428	+ nOut = precision + 10 + precision/3;
25393	25429	zOut = zExtra = sqlite3Malloc( nOut );
25394	25430	if( zOut==0 ){
25395	25431	setStrAccumError(pAccum, STRACCUM_NOMEM);
25396	25432	return;
25397	25433	}
		@@ -25413,12 +25449,27 @@
25413	25449	*(--bufpt) = cset[longvalue%base];
25414	25450	longvalue = longvalue/base;
25415	25451	}while( longvalue>0 );
25416	25452	}
25417	25453	length = (int)(&zOut[nOut-1]-bufpt);
25418		- for(idx=precision-length; idx>0; idx--){
	25454	+ while( precision>length ){
25419	25455	(--bufpt) = '0'; / Zero pad */
	25456	+ length++;
	25457	+ }
	25458	+ if( cThousand ){
	25459	+ int nn = (length - 1)/3; /* Number of "," to insert */
	25460	+ int ix = (length - 1)%3 + 1;
	25461	+ bufpt -= nn;
	25462	+ for(idx=0; nn>0; idx++){
	25463	+ bufpt[idx] = bufpt[idx+nn];
	25464	+ ix--;
	25465	+ if( ix==0 ){
	25466	+ bufpt[++idx] = cThousand;
	25467	+ nn--;
	25468	+ ix = 3;
	25469	+ }
	25470	+ }
25420	25471	}
25421	25472	if( prefix ) (--bufpt) = prefix; / Add sign */
25422	25473	if( flag_alternateform && infop->prefix ){ /* Add "0" or "0x" */
25423	25474	const char *pre;
25424	25475	char x;
		@@ -25441,13 +25492,11 @@
25441	25492	if( precision<0 ) precision = 6; /* Set default precision */
25442	25493	if( realvalue<0.0 ){
25443	25494	realvalue = -realvalue;
25444	25495	prefix = '-';
25445	25496	}else{
25446		- if( flag_plussign ) prefix = '+';
25447		- else if( flag_blanksign ) prefix = ' ';
25448		- else prefix = 0;
	25497	+ prefix = flag_prefix;
25449	25498	}
25450	25499	if( xtype==etGENERIC && precision>0 ) precision--;
25451	25500	testcase( precision>0xfff );
25452	25501	for(idx=precision&0xfff, rounder=0.5; idx>0; idx--, rounder*=0.1){}
25453	25502	if( xtype==etFLOAT ) realvalue += rounder;
		@@ -29464,117 +29513,119 @@
29464	29513	/* 50 */ "Divide" OpHelp("r[P3]=r[P2]/r[P1]"),
29465	29514	/* 51 */ "Remainder" OpHelp("r[P3]=r[P2]%r[P1]"),
29466	29515	/* 52 */ "Concat" OpHelp("r[P3]=r[P2]+r[P1]"),
29467	29516	/* 53 */ "Last" OpHelp(""),
29468	29517	/* 54 */ "BitNot" OpHelp("r[P1]= ~r[P1]"),
29469		- /* 55 */ "SorterSort" OpHelp(""),
29470		- /* 56 */ "Sort" OpHelp(""),
29471		- /* 57 */ "Rewind" OpHelp(""),
29472		- /* 58 */ "IdxLE" OpHelp("key=r[P3@P4]"),
29473		- /* 59 */ "IdxGT" OpHelp("key=r[P3@P4]"),
29474		- /* 60 */ "IdxLT" OpHelp("key=r[P3@P4]"),
29475		- /* 61 */ "IdxGE" OpHelp("key=r[P3@P4]"),
29476		- /* 62 */ "RowSetRead" OpHelp("r[P3]=rowset(P1)"),
29477		- /* 63 */ "RowSetTest" OpHelp("if r[P3] in rowset(P1) goto P2"),
29478		- /* 64 */ "Program" OpHelp(""),
29479		- /* 65 */ "FkIfZero" OpHelp("if fkctr[P1]==0 goto P2"),
29480		- /* 66 */ "IfPos" OpHelp("if r[P1]>0 then r[P1]-=P3, goto P2"),
29481		- /* 67 */ "IfNotZero" OpHelp("if r[P1]!=0 then r[P1]--, goto P2"),
29482		- /* 68 */ "DecrJumpZero" OpHelp("if (--r[P1])==0 goto P2"),
29483		- /* 69 */ "IncrVacuum" OpHelp(""),
29484		- /* 70 */ "VNext" OpHelp(""),
29485		- /* 71 */ "Init" OpHelp("Start at P2"),
29486		- /* 72 */ "Return" OpHelp(""),
29487		- /* 73 */ "EndCoroutine" OpHelp(""),
29488		- /* 74 */ "HaltIfNull" OpHelp("if r[P3]=null halt"),
29489		- /* 75 */ "Halt" OpHelp(""),
29490		- /* 76 */ "Integer" OpHelp("r[P2]=P1"),
29491		- /* 77 */ "Int64" OpHelp("r[P2]=P4"),
29492		- /* 78 */ "String" OpHelp("r[P2]='P4' (len=P1)"),
29493		- /* 79 */ "Null" OpHelp("r[P2..P3]=NULL"),
29494		- /* 80 */ "SoftNull" OpHelp("r[P1]=NULL"),
29495		- /* 81 */ "Blob" OpHelp("r[P2]=P4 (len=P1)"),
29496		- /* 82 */ "Variable" OpHelp("r[P2]=parameter(P1,P4)"),
29497		- /* 83 */ "Move" OpHelp("r[P2@P3]=r[P1@P3]"),
29498		- /* 84 */ "Copy" OpHelp("r[P2@P3+1]=r[P1@P3+1]"),
29499		- /* 85 */ "SCopy" OpHelp("r[P2]=r[P1]"),
29500		- /* 86 */ "IntCopy" OpHelp("r[P2]=r[P1]"),
29501		- /* 87 */ "ResultRow" OpHelp("output=r[P1@P2]"),
29502		- /* 88 */ "CollSeq" OpHelp(""),
29503		- /* 89 */ "Function0" OpHelp("r[P3]=func(r[P2@P5])"),
29504		- /* 90 */ "Function" OpHelp("r[P3]=func(r[P2@P5])"),
29505		- /* 91 */ "AddImm" OpHelp("r[P1]=r[P1]+P2"),
29506		- /* 92 */ "RealAffinity" OpHelp(""),
29507		- /* 93 */ "Cast" OpHelp("affinity(r[P1])"),
29508		- /* 94 */ "Permutation" OpHelp(""),
29509		- /* 95 */ "Compare" OpHelp("r[P1@P3] <-> r[P2@P3]"),
29510		- /* 96 */ "Column" OpHelp("r[P3]=PX"),
	29518	+ /* 55 */ "IfSmaller" OpHelp(""),
	29519	+ /* 56 */ "SorterSort" OpHelp(""),
	29520	+ /* 57 */ "Sort" OpHelp(""),
	29521	+ /* 58 */ "Rewind" OpHelp(""),
	29522	+ /* 59 */ "IdxLE" OpHelp("key=r[P3@P4]"),
	29523	+ /* 60 */ "IdxGT" OpHelp("key=r[P3@P4]"),
	29524	+ /* 61 */ "IdxLT" OpHelp("key=r[P3@P4]"),
	29525	+ /* 62 */ "IdxGE" OpHelp("key=r[P3@P4]"),
	29526	+ /* 63 */ "RowSetRead" OpHelp("r[P3]=rowset(P1)"),
	29527	+ /* 64 */ "RowSetTest" OpHelp("if r[P3] in rowset(P1) goto P2"),
	29528	+ /* 65 */ "Program" OpHelp(""),
	29529	+ /* 66 */ "FkIfZero" OpHelp("if fkctr[P1]==0 goto P2"),
	29530	+ /* 67 */ "IfPos" OpHelp("if r[P1]>0 then r[P1]-=P3, goto P2"),
	29531	+ /* 68 */ "IfNotZero" OpHelp("if r[P1]!=0 then r[P1]--, goto P2"),
	29532	+ /* 69 */ "DecrJumpZero" OpHelp("if (--r[P1])==0 goto P2"),
	29533	+ /* 70 */ "IncrVacuum" OpHelp(""),
	29534	+ /* 71 */ "VNext" OpHelp(""),
	29535	+ /* 72 */ "Init" OpHelp("Start at P2"),
	29536	+ /* 73 */ "Return" OpHelp(""),
	29537	+ /* 74 */ "EndCoroutine" OpHelp(""),
	29538	+ /* 75 */ "HaltIfNull" OpHelp("if r[P3]=null halt"),
	29539	+ /* 76 */ "Halt" OpHelp(""),
	29540	+ /* 77 */ "Integer" OpHelp("r[P2]=P1"),
	29541	+ /* 78 */ "Int64" OpHelp("r[P2]=P4"),
	29542	+ /* 79 */ "String" OpHelp("r[P2]='P4' (len=P1)"),
	29543	+ /* 80 */ "Null" OpHelp("r[P2..P3]=NULL"),
	29544	+ /* 81 */ "SoftNull" OpHelp("r[P1]=NULL"),
	29545	+ /* 82 */ "Blob" OpHelp("r[P2]=P4 (len=P1)"),
	29546	+ /* 83 */ "Variable" OpHelp("r[P2]=parameter(P1,P4)"),
	29547	+ /* 84 */ "Move" OpHelp("r[P2@P3]=r[P1@P3]"),
	29548	+ /* 85 */ "Copy" OpHelp("r[P2@P3+1]=r[P1@P3+1]"),
	29549	+ /* 86 */ "SCopy" OpHelp("r[P2]=r[P1]"),
	29550	+ /* 87 */ "IntCopy" OpHelp("r[P2]=r[P1]"),
	29551	+ /* 88 */ "ResultRow" OpHelp("output=r[P1@P2]"),
	29552	+ /* 89 */ "CollSeq" OpHelp(""),
	29553	+ /* 90 */ "Function0" OpHelp("r[P3]=func(r[P2@P5])"),
	29554	+ /* 91 */ "Function" OpHelp("r[P3]=func(r[P2@P5])"),
	29555	+ /* 92 */ "AddImm" OpHelp("r[P1]=r[P1]+P2"),
	29556	+ /* 93 */ "RealAffinity" OpHelp(""),
	29557	+ /* 94 */ "Cast" OpHelp("affinity(r[P1])"),
	29558	+ /* 95 */ "Permutation" OpHelp(""),
	29559	+ /* 96 */ "Compare" OpHelp("r[P1@P3] <-> r[P2@P3]"),
29511	29560	/* 97 */ "String8" OpHelp("r[P2]='P4'"),
29512		- /* 98 */ "Affinity" OpHelp("affinity(r[P1@P2])"),
29513		- /* 99 */ "MakeRecord" OpHelp("r[P3]=mkrec(r[P1@P2])"),
29514		- /* 100 */ "Count" OpHelp("r[P2]=count()"),
29515		- /* 101 */ "ReadCookie" OpHelp(""),
29516		- /* 102 */ "SetCookie" OpHelp(""),
29517		- /* 103 */ "ReopenIdx" OpHelp("root=P2 iDb=P3"),
29518		- /* 104 */ "OpenRead" OpHelp("root=P2 iDb=P3"),
29519		- /* 105 */ "OpenWrite" OpHelp("root=P2 iDb=P3"),
29520		- /* 106 */ "OpenAutoindex" OpHelp("nColumn=P2"),
29521		- /* 107 */ "OpenEphemeral" OpHelp("nColumn=P2"),
29522		- /* 108 */ "SorterOpen" OpHelp(""),
29523		- /* 109 */ "SequenceTest" OpHelp("if( cursor[P1].ctr++ ) pc = P2"),
29524		- /* 110 */ "OpenPseudo" OpHelp("P3 columns in r[P2]"),
29525		- /* 111 */ "Close" OpHelp(""),
29526		- /* 112 */ "ColumnsUsed" OpHelp(""),
29527		- /* 113 */ "Sequence" OpHelp("r[P2]=cursor[P1].ctr++"),
29528		- /* 114 */ "NewRowid" OpHelp("r[P2]=rowid"),
29529		- /* 115 */ "Insert" OpHelp("intkey=r[P3] data=r[P2]"),
29530		- /* 116 */ "InsertInt" OpHelp("intkey=P3 data=r[P2]"),
29531		- /* 117 */ "Delete" OpHelp(""),
29532		- /* 118 */ "ResetCount" OpHelp(""),
29533		- /* 119 */ "SorterCompare" OpHelp("if key(P1)!=trim(r[P3],P4) goto P2"),
29534		- /* 120 */ "SorterData" OpHelp("r[P2]=data"),
29535		- /* 121 */ "RowData" OpHelp("r[P2]=data"),
29536		- /* 122 */ "Rowid" OpHelp("r[P2]=rowid"),
29537		- /* 123 */ "NullRow" OpHelp(""),
29538		- /* 124 */ "SorterInsert" OpHelp("key=r[P2]"),
29539		- /* 125 */ "IdxInsert" OpHelp("key=r[P2]"),
29540		- /* 126 */ "IdxDelete" OpHelp("key=r[P2@P3]"),
29541		- /* 127 */ "Seek" OpHelp("Move P3 to P1.rowid"),
29542		- /* 128 */ "IdxRowid" OpHelp("r[P2]=rowid"),
29543		- /* 129 */ "Destroy" OpHelp(""),
29544		- /* 130 */ "Clear" OpHelp(""),
29545		- /* 131 */ "ResetSorter" OpHelp(""),
	29561	+ /* 98 */ "Column" OpHelp("r[P3]=PX"),
	29562	+ /* 99 */ "Affinity" OpHelp("affinity(r[P1@P2])"),
	29563	+ /* 100 */ "MakeRecord" OpHelp("r[P3]=mkrec(r[P1@P2])"),
	29564	+ /* 101 */ "Count" OpHelp("r[P2]=count()"),
	29565	+ /* 102 */ "ReadCookie" OpHelp(""),
	29566	+ /* 103 */ "SetCookie" OpHelp(""),
	29567	+ /* 104 */ "ReopenIdx" OpHelp("root=P2 iDb=P3"),
	29568	+ /* 105 */ "OpenRead" OpHelp("root=P2 iDb=P3"),
	29569	+ /* 106 */ "OpenWrite" OpHelp("root=P2 iDb=P3"),
	29570	+ /* 107 */ "OpenAutoindex" OpHelp("nColumn=P2"),
	29571	+ /* 108 */ "OpenEphemeral" OpHelp("nColumn=P2"),
	29572	+ /* 109 */ "SorterOpen" OpHelp(""),
	29573	+ /* 110 */ "SequenceTest" OpHelp("if( cursor[P1].ctr++ ) pc = P2"),
	29574	+ /* 111 */ "OpenPseudo" OpHelp("P3 columns in r[P2]"),
	29575	+ /* 112 */ "Close" OpHelp(""),
	29576	+ /* 113 */ "ColumnsUsed" OpHelp(""),
	29577	+ /* 114 */ "Sequence" OpHelp("r[P2]=cursor[P1].ctr++"),
	29578	+ /* 115 */ "NewRowid" OpHelp("r[P2]=rowid"),
	29579	+ /* 116 */ "Insert" OpHelp("intkey=r[P3] data=r[P2]"),
	29580	+ /* 117 */ "InsertInt" OpHelp("intkey=P3 data=r[P2]"),
	29581	+ /* 118 */ "Delete" OpHelp(""),
	29582	+ /* 119 */ "ResetCount" OpHelp(""),
	29583	+ /* 120 */ "SorterCompare" OpHelp("if key(P1)!=trim(r[P3],P4) goto P2"),
	29584	+ /* 121 */ "SorterData" OpHelp("r[P2]=data"),
	29585	+ /* 122 */ "RowData" OpHelp("r[P2]=data"),
	29586	+ /* 123 */ "Rowid" OpHelp("r[P2]=rowid"),
	29587	+ /* 124 */ "NullRow" OpHelp(""),
	29588	+ /* 125 */ "SorterInsert" OpHelp("key=r[P2]"),
	29589	+ /* 126 */ "IdxInsert" OpHelp("key=r[P2]"),
	29590	+ /* 127 */ "IdxDelete" OpHelp("key=r[P2@P3]"),
	29591	+ /* 128 */ "Seek" OpHelp("Move P3 to P1.rowid"),
	29592	+ /* 129 */ "IdxRowid" OpHelp("r[P2]=rowid"),
	29593	+ /* 130 */ "Destroy" OpHelp(""),
	29594	+ /* 131 */ "Clear" OpHelp(""),
29546	29595	/* 132 */ "Real" OpHelp("r[P2]=P4"),
29547		- /* 133 */ "CreateIndex" OpHelp("r[P2]=root iDb=P1"),
29548		- /* 134 */ "CreateTable" OpHelp("r[P2]=root iDb=P1"),
29549		- /* 135 */ "ParseSchema" OpHelp(""),
29550		- /* 136 */ "LoadAnalysis" OpHelp(""),
29551		- /* 137 */ "DropTable" OpHelp(""),
29552		- /* 138 */ "DropIndex" OpHelp(""),
29553		- /* 139 */ "DropTrigger" OpHelp(""),
29554		- /* 140 */ "IntegrityCk" OpHelp(""),
29555		- /* 141 */ "RowSetAdd" OpHelp("rowset(P1)=r[P2]"),
29556		- /* 142 */ "Param" OpHelp(""),
29557		- /* 143 */ "FkCounter" OpHelp("fkctr[P1]+=P2"),
29558		- /* 144 */ "MemMax" OpHelp("r[P1]=max(r[P1],r[P2])"),
29559		- /* 145 */ "OffsetLimit" OpHelp("if r[P1]>0 then r[P2]=r[P1]+max(0,r[P3]) else r[P2]=(-1)"),
29560		- /* 146 */ "AggStep0" OpHelp("accum=r[P3] step(r[P2@P5])"),
29561		- /* 147 */ "AggStep" OpHelp("accum=r[P3] step(r[P2@P5])"),
29562		- /* 148 */ "AggFinal" OpHelp("accum=r[P1] N=P2"),
29563		- /* 149 */ "Expire" OpHelp(""),
29564		- /* 150 */ "TableLock" OpHelp("iDb=P1 root=P2 write=P3"),
29565		- /* 151 */ "VBegin" OpHelp(""),
29566		- /* 152 */ "VCreate" OpHelp(""),
29567		- /* 153 */ "VDestroy" OpHelp(""),
29568		- /* 154 */ "VOpen" OpHelp(""),
29569		- /* 155 */ "VColumn" OpHelp("r[P3]=vcolumn(P2)"),
29570		- /* 156 */ "VRename" OpHelp(""),
29571		- /* 157 */ "Pagecount" OpHelp(""),
29572		- /* 158 */ "MaxPgcnt" OpHelp(""),
29573		- /* 159 */ "CursorHint" OpHelp(""),
29574		- /* 160 */ "Noop" OpHelp(""),
29575		- /* 161 */ "Explain" OpHelp(""),
	29596	+ /* 133 */ "ResetSorter" OpHelp(""),
	29597	+ /* 134 */ "CreateIndex" OpHelp("r[P2]=root iDb=P1"),
	29598	+ /* 135 */ "CreateTable" OpHelp("r[P2]=root iDb=P1"),
	29599	+ /* 136 */ "SqlExec" OpHelp(""),
	29600	+ /* 137 */ "ParseSchema" OpHelp(""),
	29601	+ /* 138 */ "LoadAnalysis" OpHelp(""),
	29602	+ /* 139 */ "DropTable" OpHelp(""),
	29603	+ /* 140 */ "DropIndex" OpHelp(""),
	29604	+ /* 141 */ "DropTrigger" OpHelp(""),
	29605	+ /* 142 */ "IntegrityCk" OpHelp(""),
	29606	+ /* 143 */ "RowSetAdd" OpHelp("rowset(P1)=r[P2]"),
	29607	+ /* 144 */ "Param" OpHelp(""),
	29608	+ /* 145 */ "FkCounter" OpHelp("fkctr[P1]+=P2"),
	29609	+ /* 146 */ "MemMax" OpHelp("r[P1]=max(r[P1],r[P2])"),
	29610	+ /* 147 */ "OffsetLimit" OpHelp("if r[P1]>0 then r[P2]=r[P1]+max(0,r[P3]) else r[P2]=(-1)"),
	29611	+ /* 148 */ "AggStep0" OpHelp("accum=r[P3] step(r[P2@P5])"),
	29612	+ /* 149 */ "AggStep" OpHelp("accum=r[P3] step(r[P2@P5])"),
	29613	+ /* 150 */ "AggFinal" OpHelp("accum=r[P1] N=P2"),
	29614	+ /* 151 */ "Expire" OpHelp(""),
	29615	+ /* 152 */ "TableLock" OpHelp("iDb=P1 root=P2 write=P3"),
	29616	+ /* 153 */ "VBegin" OpHelp(""),
	29617	+ /* 154 */ "VCreate" OpHelp(""),
	29618	+ /* 155 */ "VDestroy" OpHelp(""),
	29619	+ /* 156 */ "VOpen" OpHelp(""),
	29620	+ /* 157 */ "VColumn" OpHelp("r[P3]=vcolumn(P2)"),
	29621	+ /* 158 */ "VRename" OpHelp(""),
	29622	+ /* 159 */ "Pagecount" OpHelp(""),
	29623	+ /* 160 */ "MaxPgcnt" OpHelp(""),
	29624	+ /* 161 */ "CursorHint" OpHelp(""),
	29625	+ /* 162 */ "Noop" OpHelp(""),
	29626	+ /* 163 */ "Explain" OpHelp(""),
29576	29627	};
29577	29628	return azName[i];
29578	29629	}
29579	29630	#endif
29580	29631
		@@ -37952,46 +38003,84 @@
37952	38003	* be freed immediately and any attempt to access any of that freed
37953	38004	* data will almost certainly result in an immediate access violation.
37954	38005	******************************************************************************
37955	38006	*/
37956	38007	#ifndef SQLITE_WIN32_HEAP_CREATE
37957		-# define SQLITE_WIN32_HEAP_CREATE (TRUE)
	38008	+# define SQLITE_WIN32_HEAP_CREATE (TRUE)
	38009	+#endif
	38010	+
	38011	+/*
	38012	+ * This is the maximum possible initial size of the Win32-specific heap, in
	38013	+ * bytes.
	38014	+ */
	38015	+#ifndef SQLITE_WIN32_HEAP_MAX_INIT_SIZE
	38016	+# define SQLITE_WIN32_HEAP_MAX_INIT_SIZE (4294967295U)
	38017	+#endif
	38018	+
	38019	+/*
	38020	+ * This is the extra space for the initial size of the Win32-specific heap,
	38021	+ * in bytes. This value may be zero.
	38022	+ */
	38023	+#ifndef SQLITE_WIN32_HEAP_INIT_EXTRA
	38024	+# define SQLITE_WIN32_HEAP_INIT_EXTRA (4194304)
	38025	+#endif
	38026	+
	38027	+/*
	38028	+ * Calculate the maximum legal cache size, in pages, based on the maximum
	38029	+ * possible initial heap size and the default page size, setting aside the
	38030	+ * needed extra space.
	38031	+ */
	38032	+#ifndef SQLITE_WIN32_MAX_CACHE_SIZE
	38033	+# define SQLITE_WIN32_MAX_CACHE_SIZE (((SQLITE_WIN32_HEAP_MAX_INIT_SIZE) - \
	38034	+ (SQLITE_WIN32_HEAP_INIT_EXTRA)) / \
	38035	+ (SQLITE_DEFAULT_PAGE_SIZE))
37958	38036	#endif
37959	38037
37960	38038	/*
37961	38039	* This is cache size used in the calculation of the initial size of the
37962	38040	* Win32-specific heap. It cannot be negative.
37963	38041	*/
37964	38042	#ifndef SQLITE_WIN32_CACHE_SIZE
37965	38043	# if SQLITE_DEFAULT_CACHE_SIZE>=0
37966		-# define SQLITE_WIN32_CACHE_SIZE (SQLITE_DEFAULT_CACHE_SIZE)
	38044	+# define SQLITE_WIN32_CACHE_SIZE (SQLITE_DEFAULT_CACHE_SIZE)
37967	38045	# else
37968		-# define SQLITE_WIN32_CACHE_SIZE (-(SQLITE_DEFAULT_CACHE_SIZE))
	38046	+# define SQLITE_WIN32_CACHE_SIZE (-(SQLITE_DEFAULT_CACHE_SIZE))
37969	38047	# endif
37970	38048	#endif
37971	38049
	38050	+/*
	38051	+ * Make sure that the calculated cache size, in pages, cannot cause the
	38052	+ * initial size of the Win32-specific heap to exceed the maximum amount
	38053	+ * of memory that can be specified in the call to HeapCreate.
	38054	+ */
	38055	+#if SQLITE_WIN32_CACHE_SIZE>SQLITE_WIN32_MAX_CACHE_SIZE
	38056	+# undef SQLITE_WIN32_CACHE_SIZE
	38057	+# define SQLITE_WIN32_CACHE_SIZE (2000)
	38058	+#endif
	38059	+
37972	38060	/*
37973	38061	* The initial size of the Win32-specific heap. This value may be zero.
37974	38062	*/
37975	38063	#ifndef SQLITE_WIN32_HEAP_INIT_SIZE
37976		-# define SQLITE_WIN32_HEAP_INIT_SIZE ((SQLITE_WIN32_CACHE_SIZE) * \
37977		- (SQLITE_DEFAULT_PAGE_SIZE) + 4194304)
	38064	+# define SQLITE_WIN32_HEAP_INIT_SIZE ((SQLITE_WIN32_CACHE_SIZE) * \
	38065	+ (SQLITE_DEFAULT_PAGE_SIZE) + \
	38066	+ (SQLITE_WIN32_HEAP_INIT_EXTRA))
37978	38067	#endif
37979	38068
37980	38069	/*
37981	38070	* The maximum size of the Win32-specific heap. This value may be zero.
37982	38071	*/
37983	38072	#ifndef SQLITE_WIN32_HEAP_MAX_SIZE
37984		-# define SQLITE_WIN32_HEAP_MAX_SIZE (0)
	38073	+# define SQLITE_WIN32_HEAP_MAX_SIZE (0)
37985	38074	#endif
37986	38075
37987	38076	/*
37988	38077	* The extra flags to use in calls to the Win32 heap APIs. This value may be
37989	38078	* zero for the default behavior.
37990	38079	*/
37991	38080	#ifndef SQLITE_WIN32_HEAP_FLAGS
37992		-# define SQLITE_WIN32_HEAP_FLAGS (0)
	38081	+# define SQLITE_WIN32_HEAP_FLAGS (0)
37993	38082	#endif
37994	38083
37995	38084
37996	38085	/*
37997	38086	** The winMemData structure stores information required by the Win32-specific
		@@ -44084,11 +44173,11 @@
44084	44173	** This routine is for use inside of assert() statements only. For
44085	44174	** example:
44086	44175	**
44087	44176	** assert( sqlite3PcachePageSanity(pPg) );
44088	44177	*/
44089		-#if SQLITE_DEBUG
	44178	+#ifdef SQLITE_DEBUG
44090	44179	SQLITE_PRIVATE int sqlite3PcachePageSanity(PgHdr *pPg){
44091	44180	PCache *pCache;
44092	44181	assert( pPg!=0 );
44093	44182	assert( pPg->pgno>0 \|\| pPg->pPager==0 ); /* Page number is 1 or more */
44094	44183	pCache = pPg->pCache;
		@@ -60177,21 +60266,22 @@
60177	60266	}
60178	60267	#endif
60179	60268
60180	60269
60181	60270	/*
60182		-** Defragment the page given. All Cells are moved to the
60183		-** end of the page and all free space is collected into one
60184		-** big FreeBlk that occurs in between the header and cell
60185		-** pointer array and the cell content area.
	60271	+** Defragment the page given. This routine reorganizes cells within the
	60272	+** page so that there are no free-blocks on the free-block list.
	60273	+**
	60274	+** Parameter nMaxFrag is the maximum amount of fragmented space that may be
	60275	+** present in the page after this routine returns.
60186	60276	**
60187	60277	** EVIDENCE-OF: R-44582-60138 SQLite may from time to time reorganize a
60188	60278	** b-tree page so that there are no freeblocks or fragment bytes, all
60189	60279	** unused bytes are contained in the unallocated space region, and all
60190	60280	** cells are packed tightly at the end of the page.
60191	60281	*/
60192		-static int defragmentPage(MemPage *pPage){
	60282	+static int defragmentPage(MemPage *pPage, int nMaxFrag){
60193	60283	int i; /* Loop counter */
60194	60284	int pc; /* Address of the i-th cell */
60195	60285	int hdr; /* Offset to the page header */
60196	60286	int size; /* Size of a cell */
60197	60287	int usableSize; /* Number of usable bytes on a page */
		@@ -60202,11 +60292,10 @@
60202	60292	unsigned char temp; / Temp area for cell content */
60203	60293	unsigned char src; / Source of content */
60204	60294	int iCellFirst; /* First allowable cell index */
60205	60295	int iCellLast; /* Last possible cell index */
60206	60296
60207		-
60208	60297	assert( sqlite3PagerIswriteable(pPage->pDbPage) );
60209	60298	assert( pPage->pBt!=0 );
60210	60299	assert( pPage->pBt->usableSize <= SQLITE_MAX_PAGE_SIZE );
60211	60300	assert( pPage->nOverflow==0 );
60212	60301	assert( sqlite3_mutex_held(pPage->pBt->mutex) );
		@@ -60214,13 +60303,60 @@
60214	60303	src = data = pPage->aData;
60215	60304	hdr = pPage->hdrOffset;
60216	60305	cellOffset = pPage->cellOffset;
60217	60306	nCell = pPage->nCell;
60218	60307	assert( nCell==get2byte(&data[hdr+3]) );
	60308	+ iCellFirst = cellOffset + 2*nCell;
60219	60309	usableSize = pPage->pBt->usableSize;
	60310	+
	60311	+ /* This block handles pages with two or fewer free blocks and nMaxFrag
	60312	+ ** or fewer fragmented bytes. In this case it is faster to move the
	60313	+ ** two (or one) blocks of cells using memmove() and add the required
	60314	+ ** offsets to each pointer in the cell-pointer array than it is to
	60315	+ ** reconstruct the entire page. */
	60316	+ if( (int)data[hdr+7]<=nMaxFrag ){
	60317	+ int iFree = get2byte(&data[hdr+1]);
	60318	+ if( iFree ){
	60319	+ int iFree2 = get2byte(&data[iFree]);
	60320	+
	60321	+ /* pageFindSlot() has already verified that free blocks are sorted
	60322	+ ** in order of offset within the page, and that no block extends
	60323	+ ** past the end of the page. Provided the two free slots do not
	60324	+ ** overlap, this guarantees that the memmove() calls below will not
	60325	+ ** overwrite the usableSize byte buffer, even if the database page
	60326	+ ** is corrupt. */
	60327	+ assert( iFree2==0 \|\| iFree2>iFree );
	60328	+ assert( iFree+get2byte(&data[iFree+2]) <= usableSize );
	60329	+ assert( iFree2==0 \|\| iFree2+get2byte(&data[iFree2+2]) <= usableSize );
	60330	+
	60331	+ if( 0==iFree2 \|\| (data[iFree2]==0 && data[iFree2+1]==0) ){
	60332	+ u8 pEnd = &data[cellOffset + nCell2];
	60333	+ u8 *pAddr;
	60334	+ int sz2 = 0;
	60335	+ int sz = get2byte(&data[iFree+2]);
	60336	+ int top = get2byte(&data[hdr+5]);
	60337	+ if( iFree2 ){
	60338	+ if( iFree+sz>iFree2 ) return SQLITE_CORRUPT_BKPT;
	60339	+ sz2 = get2byte(&data[iFree2+2]);
	60340	+ assert( iFree+sz+sz2+iFree2-(iFree+sz) <= usableSize );
	60341	+ memmove(&data[iFree+sz+sz2], &data[iFree+sz], iFree2-(iFree+sz));
	60342	+ sz += sz2;
	60343	+ }
	60344	+ cbrk = top+sz;
	60345	+ assert( cbrk+(iFree-top) <= usableSize );
	60346	+ memmove(&data[cbrk], &data[top], iFree-top);
	60347	+ for(pAddr=&data[cellOffset]; pAddr<pEnd; pAddr+=2){
	60348	+ pc = get2byte(pAddr);
	60349	+ if( pc<iFree ){ put2byte(pAddr, pc+sz); }
	60350	+ else if( pc<iFree2 ){ put2byte(pAddr, pc+sz2); }
	60351	+ }
	60352	+ goto defragment_out;
	60353	+ }
	60354	+ }
	60355	+ }
	60356	+
60220	60357	cbrk = usableSize;
60221		- iCellFirst = cellOffset + 2*nCell;
60222	60358	iCellLast = usableSize - 4;
60223	60359	for(i=0; i<nCell; i++){
60224	60360	u8 pAddr; / The i-th cell pointer */
60225	60361	pAddr = &data[cellOffset + i*2];
60226	60362	pc = get2byte(pAddr);
		@@ -60250,20 +60386,22 @@
60250	60386	memcpy(&temp[x], &data[x], (cbrk+size) - x);
60251	60387	src = temp;
60252	60388	}
60253	60389	memcpy(&data[cbrk], &src[pc], size);
60254	60390	}
	60391	+ data[hdr+7] = 0;
	60392	+
	60393	+ defragment_out:
	60394	+ if( data[hdr+7]+cbrk-iCellFirst!=pPage->nFree ){
	60395	+ return SQLITE_CORRUPT_BKPT;
	60396	+ }
60255	60397	assert( cbrk>=iCellFirst );
60256	60398	put2byte(&data[hdr+5], cbrk);
60257	60399	data[hdr+1] = 0;
60258	60400	data[hdr+2] = 0;
60259		- data[hdr+7] = 0;
60260	60401	memset(&data[iCellFirst], 0, cbrk-iCellFirst);
60261	60402	assert( sqlite3PagerIswriteable(pPage->pDbPage) );
60262		- if( cbrk-iCellFirst!=pPage->nFree ){
60263		- return SQLITE_CORRUPT_BKPT;
60264		- }
60265	60403	return SQLITE_OK;
60266	60404	}
60267	60405
60268	60406	/*
60269	60407	** Search the free-list on page pPg for space to store a cell nByte bytes in
		@@ -60397,14 +60535,14 @@
60397	60535	** to see if defragmentation is necessary.
60398	60536	*/
60399	60537	testcase( gap+2+nByte==top );
60400	60538	if( gap+2+nByte>top ){
60401	60539	assert( pPage->nCell>0 \|\| CORRUPT_DB );
60402		- rc = defragmentPage(pPage);
	60540	+ rc = defragmentPage(pPage, MIN(4, pPage->nFree - (2+nByte)));
60403	60541	if( rc ) return rc;
60404	60542	top = get2byteNotZero(&data[hdr+5]);
60405		- assert( gap+nByte<=top );
	60543	+ assert( gap+2+nByte<=top );
60406	60544	}
60407	60545
60408	60546
60409	60547	/* Allocate memory from the gap in between the cell pointer array
60410	60548	** and the cell content area. The btreeInitPage() call has already
		@@ -63636,11 +63774,11 @@
63636	63774	pCur->aiIdx[pCur->iPage] = 0;
63637	63775	return getAndInitPage(pBt, newPgno, &pCur->apPage[pCur->iPage],
63638	63776	pCur, pCur->curPagerFlags);
63639	63777	}
63640	63778
63641		-#if SQLITE_DEBUG
	63779	+#ifdef SQLITE_DEBUG
63642	63780	/*
63643	63781	** Page pParent is an internal (non-leaf) tree page. This function
63644	63782	** asserts that page number iChild is the left-child if the iIdx'th
63645	63783	** cell in page pParent. Or, if iIdx is equal to the total number of
63646	63784	** cells in pParent, that page number iChild is the right-child of
		@@ -64178,10 +64316,34 @@
64178	64316	** have been deleted? This API will need to change to return an error code
64179	64317	** as well as the boolean result value.
64180	64318	*/
64181	64319	return (CURSOR_VALID!=pCur->eState);
64182	64320	}
	64321	+
	64322	+/*
	64323	+** Return an estimate for the number of rows in the table that pCur is
	64324	+** pointing to. Return a negative number if no estimate is currently
	64325	+** available.
	64326	+*/
	64327	+SQLITE_PRIVATE i64 sqlite3BtreeRowCountEst(BtCursor *pCur){
	64328	+ i64 n;
	64329	+ u8 i;
	64330	+
	64331	+ assert( cursorOwnsBtShared(pCur) );
	64332	+ assert( sqlite3_mutex_held(pCur->pBtree->db->mutex) );
	64333	+
	64334	+ /* Currently this interface is only called by the OP_IfSmaller
	64335	+ ** opcode, and it that case the cursor will always be valid and
	64336	+ ** will always point to a leaf node. */
	64337	+ if( NEVER(pCur->eState!=CURSOR_VALID) ) return -1;
	64338	+ if( NEVER(pCur->apPage[pCur->iPage]->leaf==0) ) return -1;
	64339	+
	64340	+ for(n=1, i=0; i<=pCur->iPage; i++){
	64341	+ n *= pCur->apPage[i]->nCell;
	64342	+ }
	64343	+ return n;
	64344	+}
64183	64345
64184	64346	/*
64185	64347	** Advance the cursor to the next entry in the database. If
64186	64348	** successful then set *pRes=0. If the cursor
64187	64349	** was already pointing to the last entry in the database before
		@@ -65023,11 +65185,11 @@
65023	65185	** pPrior Where to write the pgno of the first overflow page
65024	65186	**
65025	65187	** Use a call to btreeParseCellPtr() to verify that the values above
65026	65188	** were computed correctly.
65027	65189	*/
65028		-#if SQLITE_DEBUG
	65190	+#ifdef SQLITE_DEBUG
65029	65191	{
65030	65192	CellInfo info;
65031	65193	pPage->xParseCell(pPage, pCell, &info);
65032	65194	assert( nHeader==(int)(info.pPayload - pCell) );
65033	65195	assert( info.nKey==pX->nKey );
		@@ -66549,11 +66711,11 @@
66549	66711	** copied into the parent, because if the parent is page 1 then it will
66550	66712	** by smaller than the child due to the database header, and so all the
66551	66713	** free space needs to be up front.
66552	66714	*/
66553	66715	assert( nNew==1 \|\| CORRUPT_DB );
66554		- rc = defragmentPage(apNew[0]);
	66716	+ rc = defragmentPage(apNew[0], -1);
66555	66717	testcase( rc!=SQLITE_OK );
66556	66718	assert( apNew[0]->nFree ==
66557	66719	(get2byte(&apNew[0]->aData[5])-apNew[0]->cellOffset-apNew[0]->nCell*2)
66558	66720	\|\| rc!=SQLITE_OK
66559	66721	);
		@@ -71201,10 +71363,11 @@
71201	71363	** Remember the SQL string for a prepared statement.
71202	71364	*/
71203	71365	SQLITE_PRIVATE void sqlite3VdbeSetSql(Vdbe p, const char z, int n, int isPrepareV2){
71204	71366	assert( isPrepareV2==1 \|\| isPrepareV2==0 );
71205	71367	if( p==0 ) return;
	71368	+ if( !isPrepareV2 ) p->expmask = 0;
71206	71369	#if defined(SQLITE_OMIT_TRACE) && !defined(SQLITE_ENABLE_SQLLOG)
71207	71370	if( !isPrepareV2 ) return;
71208	71371	#endif
71209	71372	assert( p->zSql==0 );
71210	71373	p->zSql = sqlite3DbStrNDup(p->db, z, n);
		@@ -71229,10 +71392,11 @@
71229	71392	pB->pPrev = pTmp;
71230	71393	zTmp = pA->zSql;
71231	71394	pA->zSql = pB->zSql;
71232	71395	pB->zSql = zTmp;
71233	71396	pB->isPrepareV2 = pA->isPrepareV2;
	71397	+ pB->expmask = pA->expmask;
71234	71398	}
71235	71399
71236	71400	/*
71237	71401	** Resize the Vdbe.aOp array so that it is at least nOp elements larger
71238	71402	** than its current size. nOp is guaranteed to be less than or equal
		@@ -73753,17 +73917,17 @@
73753	73917	** Then the internal cache might have been left in an inconsistent
73754	73918	** state. We need to rollback the statement transaction, if there is
73755	73919	** one, or the complete transaction if there is no statement transaction.
73756	73920	*/
73757	73921
	73922	+ if( p->magic!=VDBE_MAGIC_RUN ){
	73923	+ return SQLITE_OK;
	73924	+ }
73758	73925	if( db->mallocFailed ){
73759	73926	p->rc = SQLITE_NOMEM_BKPT;
73760	73927	}
73761	73928	closeAllCursors(p);
73762		- if( p->magic!=VDBE_MAGIC_RUN ){
73763		- return SQLITE_OK;
73764		- }
73765	73929	checkActiveVdbeCnt(db);
73766	73930
73767	73931	/* No commit or rollback needed if the program never started or if the
73768	73932	** SQL statement does not read or write a database file. */
73769	73933	if( p->pc>=0 && p->bIsReader ){
		@@ -74706,11 +74870,11 @@
74706	74870	}
74707	74871	assert( u<=pKeyInfo->nField + 1 );
74708	74872	p->nField = u;
74709	74873	}
74710	74874
74711		-#if SQLITE_DEBUG
	74875	+#ifdef SQLITE_DEBUG
74712	74876	/*
74713	74877	** This function compares two index or table record keys in the same way
74714	74878	** as the sqlite3VdbeRecordCompare() routine. Unlike VdbeRecordCompare(),
74715	74879	** this function deserializes and compares values using the
74716	74880	** sqlite3VdbeSerialGet() and sqlite3MemCompare() functions. It is used
		@@ -74811,11 +74975,11 @@
74811	74975	if( pKeyInfo->db->mallocFailed ) return 1;
74812	74976	return 0;
74813	74977	}
74814	74978	#endif
74815	74979
74816		-#if SQLITE_DEBUG
	74980	+#ifdef SQLITE_DEBUG
74817	74981	/*
74818	74982	** Count the number of fields (a.k.a. columns) in the record given by
74819	74983	** pKey,nKey. The verify that this count is less than or equal to the
74820	74984	** limit given by pKeyInfo->nField + pKeyInfo->nXField.
74821	74985	**
		@@ -75694,12 +75858,12 @@
75694	75858	** to sqlite3_reoptimize() that re-preparing the statement may result
75695	75859	** in a better query plan.
75696	75860	*/
75697	75861	SQLITE_PRIVATE void sqlite3VdbeSetVarmask(Vdbe *v, int iVar){
75698	75862	assert( iVar>0 );
75699		- if( iVar>32 ){
75700		- v->expmask = 0xffffffff;
	75863	+ if( iVar>=32 ){
	75864	+ v->expmask \|= 0x80000000;
75701	75865	}else{
75702	75866	v->expmask \|= ((u32)1 << (iVar-1));
75703	75867	}
75704	75868	}
75705	75869
		@@ -75965,11 +76129,12 @@
75965	76129	sqlite3_mutex_enter(mutex);
75966	76130	for(i=0; i<p->nVar; i++){
75967	76131	sqlite3VdbeMemRelease(&p->aVar[i]);
75968	76132	p->aVar[i].flags = MEM_Null;
75969	76133	}
75970		- if( p->isPrepareV2 && p->expmask ){
	76134	+ assert( p->isPrepareV2 \|\| p->expmask==0 );
	76135	+ if( p->expmask ){
75971	76136	p->expired = 1;
75972	76137	}
75973	76138	sqlite3_mutex_leave(mutex);
75974	76139	return rc;
75975	76140	}
		@@ -77069,13 +77234,12 @@
77069	77234	** parameter in the WHERE clause might influence the choice of query plan
77070	77235	** for a statement, then the statement will be automatically recompiled,
77071	77236	** as if there had been a schema change, on the first sqlite3_step() call
77072	77237	** following any change to the bindings of that parameter.
77073	77238	*/
77074		- if( p->isPrepareV2 &&
77075		- ((i<32 && p->expmask & ((u32)1 << i)) \|\| p->expmask==0xffffffff)
77076		- ){
	77239	+ assert( p->isPrepareV2 \|\| p->expmask==0 );
	77240	+ if( p->expmask!=0 && (p->expmask & (i>=31 ? 0x80000000 : (u32)1<<i))!=0 ){
77077	77241	p->expired = 1;
77078	77242	}
77079	77243	return SQLITE_OK;
77080	77244	}
77081	77245
		@@ -77334,14 +77498,16 @@
77334	77498	Vdbe pFrom = (Vdbe)pFromStmt;
77335	77499	Vdbe pTo = (Vdbe)pToStmt;
77336	77500	if( pFrom->nVar!=pTo->nVar ){
77337	77501	return SQLITE_ERROR;
77338	77502	}
77339		- if( pTo->isPrepareV2 && pTo->expmask ){
	77503	+ assert( pTo->isPrepareV2 \|\| pTo->expmask==0 );
	77504	+ if( pTo->expmask ){
77340	77505	pTo->expired = 1;
77341	77506	}
77342		- if( pFrom->isPrepareV2 && pFrom->expmask ){
	77507	+ assert( pFrom->isPrepareV2 \|\| pFrom->expmask==0 );
	77508	+ if( pFrom->expmask ){
77343	77509	pFrom->expired = 1;
77344	77510	}
77345	77511	return sqlite3TransferBindings(pFromStmt, pToStmt);
77346	77512	}
77347	77513	#endif
		@@ -79658,39 +79824,39 @@
79658	79824	if( pCtx->pOut != pOut ){
79659	79825	pCtx->pOut = pOut;
79660	79826	for(i=pCtx->argc-1; i>=0; i--) pCtx->argv[i] = &aMem[pOp->p2+i];
79661	79827	}
79662	79828
79663		- memAboutToChange(p, pCtx->pOut);
	79829	+ memAboutToChange(p, pOut);
79664	79830	#ifdef SQLITE_DEBUG
79665	79831	for(i=0; i<pCtx->argc; i++){
79666	79832	assert( memIsValid(pCtx->argv[i]) );
79667	79833	REGISTER_TRACE(pOp->p2+i, pCtx->argv[i]);
79668	79834	}
79669	79835	#endif
79670		- MemSetTypeFlag(pCtx->pOut, MEM_Null);
	79836	+ MemSetTypeFlag(pOut, MEM_Null);
79671	79837	pCtx->fErrorOrAux = 0;
79672	79838	(pCtx->pFunc->xSFunc)(pCtx, pCtx->argc, pCtx->argv);/ IMP: R-24505-23230 */
79673	79839
79674	79840	/* If the function returned an error, throw an exception */
79675	79841	if( pCtx->fErrorOrAux ){
79676	79842	if( pCtx->isError ){
79677		- sqlite3VdbeError(p, "%s", sqlite3_value_text(pCtx->pOut));
	79843	+ sqlite3VdbeError(p, "%s", sqlite3_value_text(pOut));
79678	79844	rc = pCtx->isError;
79679	79845	}
79680	79846	sqlite3VdbeDeleteAuxData(db, &p->pAuxData, pCtx->iOp, pOp->p1);
79681	79847	if( rc ) goto abort_due_to_error;
79682	79848	}
79683	79849
79684	79850	/* Copy the result of the function into register P3 */
79685	79851	if( pOut->flags & (MEM_Str\|MEM_Blob) ){
79686		- sqlite3VdbeChangeEncoding(pCtx->pOut, encoding);
79687		- if( sqlite3VdbeMemTooBig(pCtx->pOut) ) goto too_big;
	79852	+ sqlite3VdbeChangeEncoding(pOut, encoding);
	79853	+ if( sqlite3VdbeMemTooBig(pOut) ) goto too_big;
79688	79854	}
79689	79855
79690		- REGISTER_TRACE(pOp->p3, pCtx->pOut);
79691		- UPDATE_MAX_BLOBSIZE(pCtx->pOut);
	79856	+ REGISTER_TRACE(pOp->p3, pOut);
	79857	+ UPDATE_MAX_BLOBSIZE(pOut);
79692	79858	break;
79693	79859	}
79694	79860
79695	79861	/* Opcode: BitAnd P1 P2 P3 * *
79696	79862	** Synopsis: r[P3]=r[P1]&r[P2]
		@@ -80187,11 +80353,11 @@
80187	80353	pKeyInfo = pOp->p4.pKeyInfo;
80188	80354	assert( n>0 );
80189	80355	assert( pKeyInfo!=0 );
80190	80356	p1 = pOp->p1;
80191	80357	p2 = pOp->p2;
80192		-#if SQLITE_DEBUG
	80358	+#ifdef SQLITE_DEBUG
80193	80359	if( aPermute ){
80194	80360	int k, mx = 0;
80195	80361	for(k=0; k<n; k++) if( aPermute[k]>mx ) mx = aPermute[k];
80196	80362	assert( p1>0 && p1+mx<=(p->nMem+1 - p->nCursor)+1 );
80197	80363	assert( p2>0 && p2+mx<=(p->nMem+1 - p->nCursor)+1 );
		@@ -82836,10 +83002,37 @@
82836	83002	assert( pOp->p2==0 );
82837	83003	}
82838	83004	break;
82839	83005	}
82840	83006
	83007	+/* Opcode: IfSmaller P1 P2 P3 * *
	83008	+**
	83009	+** Estimate the number of rows in the table P1. Jump to P2 if that
	83010	+ estimate is less than approximately 2(0.1*P3).
	83011	+*/
	83012	+case OP_IfSmaller: { /* jump */
	83013	+ VdbeCursor *pC;
	83014	+ BtCursor *pCrsr;
	83015	+ int res;
	83016	+ i64 sz;
	83017	+
	83018	+ assert( pOp->p1>=0 && pOp->p1<p->nCursor );
	83019	+ pC = p->apCsr[pOp->p1];
	83020	+ assert( pC!=0 );
	83021	+ pCrsr = pC->uc.pCursor;
	83022	+ assert( pCrsr );
	83023	+ rc = sqlite3BtreeFirst(pCrsr, &res);
	83024	+ if( rc ) goto abort_due_to_error;
	83025	+ if( res==0 ){
	83026	+ sz = sqlite3BtreeRowCountEst(pCrsr);
	83027	+ if( ALWAYS(sz>=0) && sqlite3LogEst((u64)sz)<pOp->p3 ) res = 1;
	83028	+ }
	83029	+ VdbeBranchTaken(res!=0,2);
	83030	+ if( res ) goto jump_to_p2;
	83031	+ break;
	83032	+}
	83033	+
82841	83034
82842	83035	/* Opcode: SorterSort P1 P2 * * *
82843	83036	**
82844	83037	** After all records have been inserted into the Sorter object
82845	83038	** identified by P1, invoke this opcode to actually do the sorting.
		@@ -83479,10 +83672,22 @@
83479	83672	rc = sqlite3BtreeCreateTable(pDb->pBt, &pgno, flags);
83480	83673	if( rc ) goto abort_due_to_error;
83481	83674	pOut->u.i = pgno;
83482	83675	break;
83483	83676	}
	83677	+
	83678	+/* Opcode: SqlExec * * * P4 *
	83679	+**
	83680	+** Run the SQL statement or statements specified in the P4 string.
	83681	+*/
	83682	+case OP_SqlExec: {
	83683	+ db->nSqlExec++;
	83684	+ rc = sqlite3_exec(db, pOp->p4.z, 0, 0, 0);
	83685	+ db->nSqlExec--;
	83686	+ if( rc ) goto abort_due_to_error;
	83687	+ break;
	83688	+}
83484	83689
83485	83690	/* Opcode: ParseSchema P1 * * P4 *
83486	83691	**
83487	83692	** Read and parse all entries from the SQLITE_MASTER table of database P1
83488	83693	** that match the WHERE clause P4.
		@@ -83600,11 +83805,11 @@
83600	83805	**
83601	83806	** Do an analysis of the currently open database. Store in
83602	83807	** register P1 the text of an error message describing any problems.
83603	83808	** If no problems are found, store a NULL in register P1.
83604	83809	**
83605		-** The register P3 contains the maximum number of allowed errors.
	83810	+** The register P3 contains one less than the maximum number of allowed errors.
83606	83811	** At most reg(P3) errors will be reported.
83607	83812	** In other words, the analysis stops as soon as reg(P1) errors are
83608	83813	** seen. Reg(P1) is updated with the number of errors remaining.
83609	83814	**
83610	83815	** The root page numbers of all tables in the database are integers
		@@ -83633,18 +83838,18 @@
83633	83838	assert( (pnErr->flags & (MEM_Str\|MEM_Blob))==0 );
83634	83839	pIn1 = &aMem[pOp->p1];
83635	83840	assert( pOp->p5<db->nDb );
83636	83841	assert( DbMaskTest(p->btreeMask, pOp->p5) );
83637	83842	z = sqlite3BtreeIntegrityCheck(db->aDb[pOp->p5].pBt, aRoot, nRoot,
83638		- (int)pnErr->u.i, &nErr);
83639		- pnErr->u.i -= nErr;
	83843	+ (int)pnErr->u.i+1, &nErr);
83640	83844	sqlite3VdbeMemSetNull(pIn1);
83641	83845	if( nErr==0 ){
83642	83846	assert( z==0 );
83643	83847	}else if( z==0 ){
83644	83848	goto no_mem;
83645	83849	}else{
	83850	+ pnErr->u.i -= nErr-1;
83646	83851	sqlite3VdbeMemSetStr(pIn1, z, -1, SQLITE_UTF8, sqlite3_free);
83647	83852	}
83648	83853	UPDATE_MAX_BLOBSIZE(pIn1);
83649	83854	sqlite3VdbeChangeEncoding(pIn1, encoding);
83650	83855	break;
		@@ -88787,15 +88992,15 @@
88787	88992	** The return value from the callback should be one of the WRC_*
88788	88993	** constants to specify how to proceed with the walk.
88789	88994	**
88790	88995	** WRC_Continue Continue descending down the tree.
88791	88996	**
88792		-** WRC_Prune Do not descend into child nodes. But allow
	88997	+** WRC_Prune Do not descend into child nodes, but allow
88793	88998	** the walk to continue with sibling nodes.
88794	88999	**
88795	89000	** WRC_Abort Do no more callbacks. Unwind the stack and
88796		-** return the top-level walk call.
	89001	+** return from the top-level walk call.
88797	89002	**
88798	89003	** The return value from this routine is WRC_Abort to abandon the tree walk
88799	89004	** and WRC_Continue to continue.
88800	89005	*/
88801	89006	static SQLITE_NOINLINE int walkExpr(Walker pWalker, Expr pExpr){
		@@ -89153,11 +89358,12 @@
89153	89358	}
89154	89359	}
89155	89360	}
89156	89361
89157	89362	/* Start at the inner-most context and move outward until a match is found */
89158		- while( pNC && cnt==0 ){
	89363	+ assert( pNC && cnt==0 );
	89364	+ do{
89159	89365	ExprList *pEList;
89160	89366	SrcList *pSrcList = pNC->pSrcList;
89161	89367
89162	89368	if( pSrcList ){
89163	89369	for(i=0, pItem=pSrcList->a; i<pSrcList->nSrc; i++, pItem++){
		@@ -89338,15 +89544,15 @@
89338	89544	}
89339	89545
89340	89546	/* Advance to the next name context. The loop will exit when either
89341	89547	** we have a match (cnt>0) or when we run out of name contexts.
89342	89548	*/
89343		- if( cnt==0 ){
89344		- pNC = pNC->pNext;
89345		- nSubquery++;
89346		- }
89347		- }
	89549	+ if( cnt ) break;
	89550	+ pNC = pNC->pNext;
	89551	+ nSubquery++;
	89552	+ }while( pNC );
	89553	+
89348	89554
89349	89555	/*
89350	89556	** If X and Y are NULL (in other words if only the column name Z is
89351	89557	** supplied) and the value of Z is enclosed in double-quotes, then
89352	89558	** Z is a string literal if it doesn't match any column names. In that
		@@ -89532,37 +89738,42 @@
89532	89738	break;
89533	89739	}
89534	89740	#endif /* defined(SQLITE_ENABLE_UPDATE_DELETE_LIMIT)
89535	89741	&& !defined(SQLITE_OMIT_SUBQUERY) */
89536	89742
89537		- /* A lone identifier is the name of a column.
89538		- */
89539		- case TK_ID: {
89540		- return lookupName(pParse, 0, 0, pExpr->u.zToken, pNC, pExpr);
89541		- }
89542		-
89543		- /* A table name and column name: ID.ID
	89743	+ /* A column name: ID
	89744	+ ** Or table name and column name: ID.ID
89544	89745	** Or a database, table and column: ID.ID.ID
	89746	+ **
	89747	+ ** The TK_ID and TK_OUT cases are combined so that there will only
	89748	+ ** be one call to lookupName(). Then the compiler will in-line
	89749	+ ** lookupName() for a size reduction and performance increase.
89545	89750	*/
	89751	+ case TK_ID:
89546	89752	case TK_DOT: {
89547	89753	const char *zColumn;
89548	89754	const char *zTable;
89549	89755	const char *zDb;
89550	89756	Expr *pRight;
89551	89757
89552		- /* if( pSrcList==0 ) break; */
89553		- notValid(pParse, pNC, "the \".\" operator", NC_IdxExpr);
89554		- pRight = pExpr->pRight;
89555		- if( pRight->op==TK_ID ){
	89758	+ if( pExpr->op==TK_ID ){
89556	89759	zDb = 0;
89557		- zTable = pExpr->pLeft->u.zToken;
89558		- zColumn = pRight->u.zToken;
	89760	+ zTable = 0;
	89761	+ zColumn = pExpr->u.zToken;
89559	89762	}else{
89560		- assert( pRight->op==TK_DOT );
89561		- zDb = pExpr->pLeft->u.zToken;
89562		- zTable = pRight->pLeft->u.zToken;
89563		- zColumn = pRight->pRight->u.zToken;
	89763	+ notValid(pParse, pNC, "the \".\" operator", NC_IdxExpr);
	89764	+ pRight = pExpr->pRight;
	89765	+ if( pRight->op==TK_ID ){
	89766	+ zDb = 0;
	89767	+ zTable = pExpr->pLeft->u.zToken;
	89768	+ zColumn = pRight->u.zToken;
	89769	+ }else{
	89770	+ assert( pRight->op==TK_DOT );
	89771	+ zDb = pExpr->pLeft->u.zToken;
	89772	+ zTable = pRight->pLeft->u.zToken;
	89773	+ zColumn = pRight->pRight->u.zToken;
	89774	+ }
89564	89775	}
89565	89776	return lookupName(pParse, zDb, zTable, zColumn, pNC, pExpr);
89566	89777	}
89567	89778
89568	89779	/* Resolve function names
		@@ -92833,11 +93044,11 @@
92833	93044	int nVal = sqlite3ExprVectorSize(pLeft);
92834	93045	Select *pSelect = (pExpr->flags & EP_xIsSelect) ? pExpr->x.pSelect : 0;
92835	93046	char *zRet;
92836	93047
92837	93048	assert( pExpr->op==TK_IN );
92838		- zRet = sqlite3DbMallocZero(pParse->db, nVal+1);
	93049	+ zRet = sqlite3DbMallocRaw(pParse->db, nVal+1);
92839	93050	if( zRet ){
92840	93051	int i;
92841	93052	for(i=0; i<nVal; i++){
92842	93053	Expr *pA = sqlite3VectorFieldSubexpr(pLeft, i);
92843	93054	char a = sqlite3ExprAffinity(pA);
		@@ -93738,11 +93949,11 @@
93738	93949	** Clear all column cache entries.
93739	93950	*/
93740	93951	SQLITE_PRIVATE void sqlite3ExprCacheClear(Parse *pParse){
93741	93952	int i;
93742	93953
93743		-#if SQLITE_DEBUG
	93954	+#ifdef SQLITE_DEBUG
93744	93955	if( pParse->db->flags & SQLITE_VdbeAddopTrace ){
93745	93956	printf("CLEAR\n");
93746	93957	}
93747	93958	#endif
93748	93959	for(i=0; i<pParse->nColCache; i++){
		@@ -95145,10 +95356,21 @@
95145	95356	if( pA->a[i].sortOrder!=pB->a[i].sortOrder ) return 1;
95146	95357	if( sqlite3ExprCompare(pExprA, pExprB, iTab) ) return 1;
95147	95358	}
95148	95359	return 0;
95149	95360	}
	95361	+
	95362	+/*
	95363	+** Like sqlite3ExprCompare() except COLLATE operators at the top-level
	95364	+** are ignored.
	95365	+*/
	95366	+SQLITE_PRIVATE int sqlite3ExprCompareSkip(Expr pA, Expr pB, int iTab){
	95367	+ return sqlite3ExprCompare(
	95368	+ sqlite3ExprSkipCollate(pA),
	95369	+ sqlite3ExprSkipCollate(pB),
	95370	+ iTab);
	95371	+}
95150	95372
95151	95373	/*
95152	95374	** Return true if we can prove the pE2 will always be true if pE1 is
95153	95375	** true. Return false if we cannot complete the proof or if pE2 might
95154	95376	** be false. Examples:
		@@ -96698,10 +96920,11 @@
96698	96920	Stat4Sample current; /* Current row as a Stat4Sample */
96699	96921	u32 iPrn; /* Pseudo-random number used for sampling */
96700	96922	Stat4Sample aBest; / Array of nCol best samples */
96701	96923	int iMin; /* Index in a[] of entry with minimum score */
96702	96924	int nSample; /* Current number of samples */
	96925	+ int nMaxEqZero; /* Max leading 0 in anEq[] for any a[] entry */
96703	96926	int iGet; /* Index of current sample accessed by stat_get() */
96704	96927	Stat4Sample a; / Array of mxSample Stat4Sample objects */
96705	96928	sqlite3 db; / Database connection, for malloc() */
96706	96929	};
96707	96930
		@@ -96962,10 +97185,17 @@
96962	97185	int i;
96963	97186
96964	97187	assert( IsStat4 \|\| nEqZero==0 );
96965	97188
96966	97189	#ifdef SQLITE_ENABLE_STAT4
	97190	+ /* Stat4Accum.nMaxEqZero is set to the maximum number of leading 0
	97191	+ ** values in the anEq[] array of any sample in Stat4Accum.a[]. In
	97192	+ ** other words, if nMaxEqZero is n, then it is guaranteed that there
	97193	+ ** are no samples with Stat4Sample.anEq[m]==0 for (m>=n). */
	97194	+ if( nEqZero>p->nMaxEqZero ){
	97195	+ p->nMaxEqZero = nEqZero;
	97196	+ }
96967	97197	if( pNew->isPSample==0 ){
96968	97198	Stat4Sample *pUpgrade = 0;
96969	97199	assert( pNew->anEq[pNew->iCol]>0 );
96970	97200
96971	97201	/* This sample is being added because the prefix that ends in column
		@@ -97059,16 +97289,26 @@
97059	97289	if( p->nSample<p->mxSample \|\| sampleIsBetter(p, pBest, &p->a[p->iMin]) ){
97060	97290	sampleInsert(p, pBest, i);
97061	97291	}
97062	97292	}
97063	97293
97064		- /* Update the anEq[] fields of any samples already collected. */
	97294	+ /* Check that no sample contains an anEq[] entry with an index of
	97295	+ ** p->nMaxEqZero or greater set to zero. */
97065	97296	for(i=p->nSample-1; i>=0; i--){
97066	97297	int j;
97067		- for(j=iChng; j<p->nCol; j++){
97068		- if( p->a[i].anEq[j]==0 ) p->a[i].anEq[j] = p->current.anEq[j];
	97298	+ for(j=p->nMaxEqZero; j<p->nCol; j++) assert( p->a[i].anEq[j]>0 );
	97299	+ }
	97300	+
	97301	+ /* Update the anEq[] fields of any samples already collected. */
	97302	+ if( iChng<p->nMaxEqZero ){
	97303	+ for(i=p->nSample-1; i>=0; i--){
	97304	+ int j;
	97305	+ for(j=iChng; j<p->nCol; j++){
	97306	+ if( p->a[i].anEq[j]==0 ) p->a[i].anEq[j] = p->current.anEq[j];
	97307	+ }
97069	97308	}
	97309	+ p->nMaxEqZero = iChng;
97070	97310	}
97071	97311	#endif
97072	97312
97073	97313	#if defined(SQLITE_ENABLE_STAT3) && !defined(SQLITE_ENABLE_STAT4)
97074	97314	if( iChng==0 ){
		@@ -97778,44 +98018,32 @@
97778	98018	/* Form 1: Analyze everything */
97779	98019	for(i=0; i<db->nDb; i++){
97780	98020	if( i==1 ) continue; /* Do not analyze the TEMP database */
97781	98021	analyzeDatabase(pParse, i);
97782	98022	}
97783		- }else if( pName2->n==0 ){
97784		- /* Form 2: Analyze the database or table named */
97785		- iDb = sqlite3FindDb(db, pName1);
97786		- if( iDb>=0 ){
97787		- analyzeDatabase(pParse, iDb);
97788		- }else{
97789		- z = sqlite3NameFromToken(db, pName1);
97790		- if( z ){
97791		- if( (pIdx = sqlite3FindIndex(db, z, 0))!=0 ){
97792		- analyzeTable(pParse, pIdx->pTable, pIdx);
97793		- }else if( (pTab = sqlite3LocateTable(pParse, 0, z, 0))!=0 ){
97794		- analyzeTable(pParse, pTab, 0);
97795		- }
97796		- sqlite3DbFree(db, z);
97797		- }
97798		- }
97799		- }else{
97800		- /* Form 3: Analyze the fully qualified table name */
	98023	+ }else if( pName2->n==0 && (iDb = sqlite3FindDb(db, pName1))>=0 ){
	98024	+ /* Analyze the schema named as the argument */
	98025	+ analyzeDatabase(pParse, iDb);
	98026	+ }else{
	98027	+ /* Form 3: Analyze the table or index named as an argument */
97801	98028	iDb = sqlite3TwoPartName(pParse, pName1, pName2, &pTableName);
97802	98029	if( iDb>=0 ){
97803		- zDb = db->aDb[iDb].zDbSName;
	98030	+ zDb = pName2->n ? db->aDb[iDb].zDbSName : 0;
97804	98031	z = sqlite3NameFromToken(db, pTableName);
97805	98032	if( z ){
97806	98033	if( (pIdx = sqlite3FindIndex(db, z, zDb))!=0 ){
97807	98034	analyzeTable(pParse, pIdx->pTable, pIdx);
97808	98035	}else if( (pTab = sqlite3LocateTable(pParse, 0, z, zDb))!=0 ){
97809	98036	analyzeTable(pParse, pTab, 0);
97810	98037	}
97811	98038	sqlite3DbFree(db, z);
97812	98039	}
97813		- }
	98040	+ }
97814	98041	}
97815		- v = sqlite3GetVdbe(pParse);
97816		- if( v ) sqlite3VdbeAddOp0(v, OP_Expire);
	98042	+ if( db->nSqlExec==0 && (v = sqlite3GetVdbe(pParse))!=0 ){
	98043	+ sqlite3VdbeAddOp0(v, OP_Expire);
	98044	+ }
97817	98045	}
97818	98046
97819	98047	/*
97820	98048	** Used to pass information from the analyzer reader through to the
97821	98049	** callback routine.
		@@ -97940,19 +98168,24 @@
97940	98168	}
97941	98169	aiRowEst = pIndex->aiRowEst;
97942	98170	#endif
97943	98171	pIndex->bUnordered = 0;
97944	98172	decodeIntArray((char*)z, nCol, aiRowEst, pIndex->aiRowLogEst, pIndex);
97945		- if( pIndex->pPartIdxWhere==0 ) pTable->nRowLogEst = pIndex->aiRowLogEst[0];
	98173	+ pIndex->hasStat1 = 1;
	98174	+ if( pIndex->pPartIdxWhere==0 ){
	98175	+ pTable->nRowLogEst = pIndex->aiRowLogEst[0];
	98176	+ pTable->tabFlags \|= TF_HasStat1;
	98177	+ }
97946	98178	}else{
97947	98179	Index fakeIdx;
97948	98180	fakeIdx.szIdxRow = pTable->szTabRow;
97949	98181	#ifdef SQLITE_ENABLE_COSTMULT
97950	98182	fakeIdx.pTable = pTable;
97951	98183	#endif
97952	98184	decodeIntArray((char*)z, 1, 0, &pTable->nRowLogEst, &fakeIdx);
97953	98185	pTable->szTabRow = fakeIdx.szIdxRow;
	98186	+ pTable->tabFlags \|= TF_HasStat1;
97954	98187	}
97955	98188
97956	98189	return 0;
97957	98190	}
97958	98191
		@@ -98243,19 +98476,24 @@
98243	98476	SQLITE_PRIVATE int sqlite3AnalysisLoad(sqlite3 *db, int iDb){
98244	98477	analysisInfo sInfo;
98245	98478	HashElem *i;
98246	98479	char *zSql;
98247	98480	int rc = SQLITE_OK;
	98481	+ Schema *pSchema = db->aDb[iDb].pSchema;
98248	98482
98249	98483	assert( iDb>=0 && iDb<db->nDb );
98250	98484	assert( db->aDb[iDb].pBt!=0 );
98251	98485
98252	98486	/* Clear any prior statistics */
98253	98487	assert( sqlite3SchemaMutexHeld(db, iDb, 0) );
98254		- for(i=sqliteHashFirst(&db->aDb[iDb].pSchema->idxHash);i;i=sqliteHashNext(i)){
	98488	+ for(i=sqliteHashFirst(&pSchema->tblHash); i; i=sqliteHashNext(i)){
	98489	+ Table *pTab = sqliteHashData(i);
	98490	+ pTab->tabFlags &= ~TF_HasStat1;
	98491	+ }
	98492	+ for(i=sqliteHashFirst(&pSchema->idxHash); i; i=sqliteHashNext(i)){
98255	98493	Index *pIdx = sqliteHashData(i);
98256		- pIdx->aiRowLogEst[0] = 0;
	98494	+ pIdx->hasStat1 = 0;
98257	98495	#ifdef SQLITE_ENABLE_STAT3_OR_STAT4
98258	98496	sqlite3DeleteIndexSamples(db, pIdx);
98259	98497	pIdx->aSample = 0;
98260	98498	#endif
98261	98499	}
		@@ -98274,23 +98512,23 @@
98274	98512	}
98275	98513	}
98276	98514
98277	98515	/* Set appropriate defaults on all indexes not in the sqlite_stat1 table */
98278	98516	assert( sqlite3SchemaMutexHeld(db, iDb, 0) );
98279		- for(i=sqliteHashFirst(&db->aDb[iDb].pSchema->idxHash);i;i=sqliteHashNext(i)){
	98517	+ for(i=sqliteHashFirst(&pSchema->idxHash); i; i=sqliteHashNext(i)){
98280	98518	Index *pIdx = sqliteHashData(i);
98281		- if( pIdx->aiRowLogEst[0]==0 ) sqlite3DefaultRowEst(pIdx);
	98519	+ if( !pIdx->hasStat1 ) sqlite3DefaultRowEst(pIdx);
98282	98520	}
98283	98521
98284	98522	/* Load the statistics from the sqlite_stat4 table. */
98285	98523	#ifdef SQLITE_ENABLE_STAT3_OR_STAT4
98286	98524	if( rc==SQLITE_OK && OptimizationEnabled(db, SQLITE_Stat34) ){
98287	98525	db->lookaside.bDisable++;
98288	98526	rc = loadStat4(db, sInfo.zDatabase);
98289	98527	db->lookaside.bDisable--;
98290	98528	}
98291		- for(i=sqliteHashFirst(&db->aDb[iDb].pSchema->idxHash);i;i=sqliteHashNext(i)){
	98529	+ for(i=sqliteHashFirst(&pSchema->idxHash); i; i=sqliteHashNext(i)){
98292	98530	Index *pIdx = sqliteHashData(i);
98293	98531	sqlite3_free(pIdx->aiRowEst);
98294	98532	pIdx->aiRowEst = 0;
98295	98533	}
98296	98534	#endif
		@@ -100271,10 +100509,11 @@
100271	100509	SQLITE_PRIVATE void sqlite3AddNotNull(Parse *pParse, int onError){
100272	100510	Table *p;
100273	100511	p = pParse->pNewTable;
100274	100512	if( p==0 \|\| NEVER(p->nCol<1) ) return;
100275	100513	p->aCol[p->nCol-1].notNull = (u8)onError;
	100514	+ p->tabFlags \|= TF_HasNotNull;
100276	100515	}
100277	100516
100278	100517	/*
100279	100518	** Scan the column type name zType (length nType) and return the
100280	100519	** associated affinity type.
		@@ -102609,10 +102848,13 @@
102609	102848	/* 10, 9, 8, 7, 6 */
102610	102849	LogEst aVal[] = { 33, 32, 30, 28, 26 };
102611	102850	LogEst *a = pIdx->aiRowLogEst;
102612	102851	int nCopy = MIN(ArraySize(aVal), pIdx->nKeyCol);
102613	102852	int i;
	102853	+
	102854	+ /* Indexes with default row estimates should not have stat1 data */
	102855	+ assert( !pIdx->hasStat1 );
102614	102856
102615	102857	/* Set the first entry (number of rows in the index) to the estimated
102616	102858	** number of rows in the table, or half the number of rows in the table
102617	102859	** for a partial index. But do not let the estimate drop below 10. */
102618	102860	a[0] = pIdx->pTable->nRowLogEst;
		@@ -109798,10 +110040,13 @@
109798	110040	VdbeComment((v, "%s", iField<0 ? "rowid" : pTab->aCol[iField].zName));
109799	110041	}
109800	110042	}
109801	110043	sqlite3VdbeAddOp3(v, OP_MakeRecord, regIdx, pIdx->nColumn, aRegIdx[ix]);
109802	110044	VdbeComment((v, "for %s", pIdx->zName));
	110045	+#ifdef SQLITE_ENABLE_NULL_TRIM
	110046	+ if( pIdx->idxType==2 ) sqlite3SetMakeRecordP5(v, pIdx->pTable);
	110047	+#endif
109803	110048
109804	110049	/* In an UPDATE operation, if this index is the PRIMARY KEY index
109805	110050	** of a WITHOUT ROWID table and there has been no change the
109806	110051	** primary key, then no collision is possible. The collision detection
109807	110052	** logic below can all be skipped. */
		@@ -109953,12 +110198,15 @@
109953	110198
109954	110199	/* Records with omitted columns are only allowed for schema format
109955	110200	** version 2 and later (SQLite version 3.1.4, 2005-02-20). */
109956	110201	if( pTab->pSchema->file_format<2 ) return;
109957	110202
109958		- for(i=pTab->nCol; i>1 && pTab->aCol[i-1].pDflt==0; i--){}
109959		- sqlite3VdbeChangeP5(v, i);
	110203	+ for(i=pTab->nCol-1; i>0; i--){
	110204	+ if( pTab->aCol[i].pDflt!=0 ) break;
	110205	+ if( pTab->aCol[i].colFlags & COLFLAG_PRIMKEY ) break;
	110206	+ }
	110207	+ sqlite3VdbeChangeP5(v, i+1);
109960	110208	}
109961	110209	#endif
109962	110210
109963	110211	/*
109964	110212	** This routine generates code to finish the INSERT or UPDATE operation
		@@ -110244,11 +110492,11 @@
110244	110492	}
110245	110493	if( sqlite3TriggerList(pParse, pDest) ){
110246	110494	return 0; /* tab1 must not have triggers */
110247	110495	}
110248	110496	#ifndef SQLITE_OMIT_VIRTUALTABLE
110249		- if( pDest->tabFlags & TF_Virtual ){
	110497	+ if( IsVirtual(pDest) ){
110250	110498	return 0; /* tab1 must not be a virtual table */
110251	110499	}
110252	110500	#endif
110253	110501	if( onError==OE_Default ){
110254	110502	if( pDest->iPKey>=0 ) onError = pDest->keyConf;
		@@ -110306,11 +110554,11 @@
110306	110554	}
110307	110555	if( HasRowid(pDest)!=HasRowid(pSrc) ){
110308	110556	return 0; /* source and destination must both be WITHOUT ROWID or not */
110309	110557	}
110310	110558	#ifndef SQLITE_OMIT_VIRTUALTABLE
110311		- if( pSrc->tabFlags & TF_Virtual ){
	110559	+ if( IsVirtual(pSrc) ){
110312	110560	return 0; /* tab2 must not be a virtual table */
110313	110561	}
110314	110562	#endif
110315	110563	if( pSrc->pSelect ){
110316	110564	return 0; /* tab2 may not be a view */
		@@ -110603,11 +110851,11 @@
110603	110851	/* Invoke the callback function if required */
110604	110852	if( xCallback && (SQLITE_ROW==rc \|\|
110605	110853	(SQLITE_DONE==rc && !callbackIsInit
110606	110854	&& db->flags&SQLITE_NullCallback)) ){
110607	110855	if( !callbackIsInit ){
110608		- azCols = sqlite3DbMallocZero(db, 2nColsizeof(const char*) + 1);
	110856	+ azCols = sqlite3DbMallocRaw(db, (2nCol+1)sizeof(const char*));
110609	110857	if( azCols==0 ){
110610	110858	goto exec_out;
110611	110859	}
110612	110860	for(i=0; i<nCol; i++){
110613	110861	azCols[i] = (char *)sqlite3_column_name(pStmt, i);
		@@ -110624,10 +110872,11 @@
110624	110872	if( !azVals[i] && sqlite3_column_type(pStmt, i)!=SQLITE_NULL ){
110625	110873	sqlite3OomFault(db);
110626	110874	goto exec_out;
110627	110875	}
110628	110876	}
	110877	+ azVals[i] = 0;
110629	110878	}
110630	110879	if( xCallback(pArg, nCol, azVals, azCols) ){
110631	110880	/* EVIDENCE-OF: R-38229-40159 If the callback function to
110632	110881	** sqlite3_exec() returns non-zero, then sqlite3_exec() will
110633	110882	** return SQLITE_ABORT. */
		@@ -112094,15 +112343,15 @@
112094	112343	#define PragTyp_JOURNAL_SIZE_LIMIT 20
112095	112344	#define PragTyp_LOCK_PROXY_FILE 21
112096	112345	#define PragTyp_LOCKING_MODE 22
112097	112346	#define PragTyp_PAGE_COUNT 23
112098	112347	#define PragTyp_MMAP_SIZE 24
112099		-#define PragTyp_PAGE_SIZE 25
112100		-#define PragTyp_SECURE_DELETE 26
112101		-#define PragTyp_SHRINK_MEMORY 27
112102		-#define PragTyp_SOFT_HEAP_LIMIT 28
112103		-#define PragTyp_STATS 29
	112348	+#define PragTyp_OPTIMIZE 25
	112349	+#define PragTyp_PAGE_SIZE 26
	112350	+#define PragTyp_SECURE_DELETE 27
	112351	+#define PragTyp_SHRINK_MEMORY 28
	112352	+#define PragTyp_SOFT_HEAP_LIMIT 29
112104	112353	#define PragTyp_SYNCHRONOUS 30
112105	112354	#define PragTyp_TABLE_INFO 31
112106	112355	#define PragTyp_TEMP_STORE 32
112107	112356	#define PragTyp_TEMP_STORE_DIRECTORY 33
112108	112357	#define PragTyp_THREADS 34
		@@ -112112,10 +112361,11 @@
112112	112361	#define PragTyp_HEXKEY 38
112113	112362	#define PragTyp_KEY 39
112114	112363	#define PragTyp_REKEY 40
112115	112364	#define PragTyp_LOCK_STATUS 41
112116	112365	#define PragTyp_PARSER_TRACE 42
	112366	+#define PragTyp_STATS 43
112117	112367
112118	112368	/* Property flags associated with various pragma. */
112119	112369	#define PragFlg_NeedSchema 0x01 /* Force schema load before running */
112120	112370	#define PragFlg_NoColumns 0x02 /* OP_ResultRow called with zero columns */
112121	112371	#define PragFlg_NoColumns1 0x04 /* zero columns if RHS argument is present */
		@@ -112135,51 +112385,52 @@
112135	112385	/* 2 */ "name",
112136	112386	/* 3 */ "type",
112137	112387	/* 4 */ "notnull",
112138	112388	/* 5 */ "dflt_value",
112139	112389	/* 6 */ "pk",
112140		- /* 7 / "table", / Used by: stats */
112141		- /* 8 */ "index",
112142		- /* 9 */ "width",
112143		- /* 10 */ "height",
112144		- /* 11 / "seqno", / Used by: index_info */
112145		- /* 12 */ "cid",
112146		- /* 13 */ "name",
112147		- /* 14 / "seqno", / Used by: index_xinfo */
112148		- /* 15 */ "cid",
112149		- /* 16 */ "name",
112150		- /* 17 */ "desc",
112151		- /* 18 */ "coll",
112152		- /* 19 */ "key",
112153		- /* 20 / "seq", / Used by: index_list */
112154		- /* 21 */ "name",
112155		- /* 22 */ "unique",
112156		- /* 23 */ "origin",
112157		- /* 24 */ "partial",
112158		- /* 25 / "seq", / Used by: database_list */
112159		- /* 26 */ "name",
112160		- /* 27 */ "file",
112161		- /* 28 / "seq", / Used by: collation_list */
112162		- /* 29 */ "name",
112163		- /* 30 / "id", / Used by: foreign_key_list */
112164		- /* 31 */ "seq",
112165		- /* 32 */ "table",
112166		- /* 33 */ "from",
112167		- /* 34 */ "to",
112168		- /* 35 */ "on_update",
112169		- /* 36 */ "on_delete",
112170		- /* 37 */ "match",
112171		- /* 38 / "table", / Used by: foreign_key_check */
112172		- /* 39 */ "rowid",
112173		- /* 40 */ "parent",
112174		- /* 41 */ "fkid",
112175		- /* 42 / "busy", / Used by: wal_checkpoint */
112176		- /* 43 */ "log",
112177		- /* 44 */ "checkpointed",
112178		- /* 45 / "timeout", / Used by: busy_timeout */
112179		- /* 46 / "database", / Used by: lock_status */
112180		- /* 47 */ "status",
	112390	+ /* 7 / "tbl", / Used by: stats */
	112391	+ /* 8 */ "idx",
	112392	+ /* 9 */ "wdth",
	112393	+ /* 10 */ "hght",
	112394	+ /* 11 */ "flgs",
	112395	+ /* 12 / "seqno", / Used by: index_info */
	112396	+ /* 13 */ "cid",
	112397	+ /* 14 */ "name",
	112398	+ /* 15 / "seqno", / Used by: index_xinfo */
	112399	+ /* 16 */ "cid",
	112400	+ /* 17 */ "name",
	112401	+ /* 18 */ "desc",
	112402	+ /* 19 */ "coll",
	112403	+ /* 20 */ "key",
	112404	+ /* 21 / "seq", / Used by: index_list */
	112405	+ /* 22 */ "name",
	112406	+ /* 23 */ "unique",
	112407	+ /* 24 */ "origin",
	112408	+ /* 25 */ "partial",
	112409	+ /* 26 / "seq", / Used by: database_list */
	112410	+ /* 27 */ "name",
	112411	+ /* 28 */ "file",
	112412	+ /* 29 / "seq", / Used by: collation_list */
	112413	+ /* 30 */ "name",
	112414	+ /* 31 / "id", / Used by: foreign_key_list */
	112415	+ /* 32 */ "seq",
	112416	+ /* 33 */ "table",
	112417	+ /* 34 */ "from",
	112418	+ /* 35 */ "to",
	112419	+ /* 36 */ "on_update",
	112420	+ /* 37 */ "on_delete",
	112421	+ /* 38 */ "match",
	112422	+ /* 39 / "table", / Used by: foreign_key_check */
	112423	+ /* 40 */ "rowid",
	112424	+ /* 41 */ "parent",
	112425	+ /* 42 */ "fkid",
	112426	+ /* 43 / "busy", / Used by: wal_checkpoint */
	112427	+ /* 44 */ "log",
	112428	+ /* 45 */ "checkpointed",
	112429	+ /* 46 / "timeout", / Used by: busy_timeout */
	112430	+ /* 47 / "database", / Used by: lock_status */
	112431	+ /* 48 */ "status",
112181	112432	};
112182	112433
112183	112434	/* Definitions of all built-in pragmas */
112184	112435	typedef struct PragmaName {
112185	112436	const char const zName; / Name of pragma */
		@@ -112221,11 +112472,11 @@
112221	112472	#endif
112222	112473	#endif
112223	112474	{/* zName: */ "busy_timeout",
112224	112475	/* ePragTyp: */ PragTyp_BUSY_TIMEOUT,
112225	112476	/* ePragFlg: */ PragFlg_Result0,
112226		- /* ColNames: */ 45, 1,
	112477	+ /* ColNames: */ 46, 1,
112227	112478	/* iArg: */ 0 },
112228	112479	#if !defined(SQLITE_OMIT_PAGER_PRAGMAS)
112229	112480	{/* zName: */ "cache_size",
112230	112481	/* ePragTyp: */ PragTyp_CACHE_SIZE,
112231	112482	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result0\|PragFlg_SchemaReq\|PragFlg_NoColumns1,
		@@ -112258,11 +112509,11 @@
112258	112509	#endif
112259	112510	#if !defined(SQLITE_OMIT_SCHEMA_PRAGMAS)
112260	112511	{/* zName: */ "collation_list",
112261	112512	/* ePragTyp: */ PragTyp_COLLATION_LIST,
112262	112513	/* ePragFlg: */ PragFlg_Result0,
112263		- /* ColNames: */ 28, 2,
	112514	+ /* ColNames: */ 29, 2,
112264	112515	/* iArg: */ 0 },
112265	112516	#endif
112266	112517	#if !defined(SQLITE_OMIT_COMPILEOPTION_DIAGS)
112267	112518	{/* zName: */ "compile_options",
112268	112519	/* ePragTyp: */ PragTyp_COMPILE_OPTIONS,
		@@ -112293,11 +112544,11 @@
112293	112544	#endif
112294	112545	#if !defined(SQLITE_OMIT_SCHEMA_PRAGMAS)
112295	112546	{/* zName: */ "database_list",
112296	112547	/* ePragTyp: */ PragTyp_DATABASE_LIST,
112297	112548	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result0,
112298		- /* ColNames: */ 25, 3,
	112549	+ /* ColNames: */ 26, 3,
112299	112550	/* iArg: */ 0 },
112300	112551	#endif
112301	112552	#if !defined(SQLITE_OMIT_PAGER_PRAGMAS) && !defined(SQLITE_OMIT_DEPRECATED)
112302	112553	{/* zName: */ "default_cache_size",
112303	112554	/* ePragTyp: */ PragTyp_DEFAULT_CACHE_SIZE,
		@@ -112330,18 +112581,18 @@
112330	112581	#endif
112331	112582	#if !defined(SQLITE_OMIT_FOREIGN_KEY) && !defined(SQLITE_OMIT_TRIGGER)
112332	112583	{/* zName: */ "foreign_key_check",
112333	112584	/* ePragTyp: */ PragTyp_FOREIGN_KEY_CHECK,
112334	112585	/* ePragFlg: */ PragFlg_NeedSchema,
112335		- /* ColNames: */ 38, 4,
	112586	+ /* ColNames: */ 39, 4,
112336	112587	/* iArg: */ 0 },
112337	112588	#endif
112338	112589	#if !defined(SQLITE_OMIT_FOREIGN_KEY)
112339	112590	{/* zName: */ "foreign_key_list",
112340	112591	/* ePragTyp: */ PragTyp_FOREIGN_KEY_LIST,
112341	112592	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result1\|PragFlg_SchemaOpt,
112342		- /* ColNames: */ 30, 8,
	112593	+ /* ColNames: */ 31, 8,
112343	112594	/* iArg: */ 0 },
112344	112595	#endif
112345	112596	#if !defined(SQLITE_OMIT_FLAG_PRAGMAS)
112346	112597	#if !defined(SQLITE_OMIT_FOREIGN_KEY) && !defined(SQLITE_OMIT_TRIGGER)
112347	112598	{/* zName: */ "foreign_keys",
		@@ -112400,21 +112651,21 @@
112400	112651	#endif
112401	112652	#if !defined(SQLITE_OMIT_SCHEMA_PRAGMAS)
112402	112653	{/* zName: */ "index_info",
112403	112654	/* ePragTyp: */ PragTyp_INDEX_INFO,
112404	112655	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result1\|PragFlg_SchemaOpt,
112405		- /* ColNames: */ 11, 3,
	112656	+ /* ColNames: */ 12, 3,
112406	112657	/* iArg: */ 0 },
112407	112658	{/* zName: */ "index_list",
112408	112659	/* ePragTyp: */ PragTyp_INDEX_LIST,
112409	112660	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result1\|PragFlg_SchemaOpt,
112410		- /* ColNames: */ 20, 5,
	112661	+ /* ColNames: */ 21, 5,
112411	112662	/* iArg: */ 0 },
112412	112663	{/* zName: */ "index_xinfo",
112413	112664	/* ePragTyp: */ PragTyp_INDEX_INFO,
112414	112665	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result1\|PragFlg_SchemaOpt,
112415		- /* ColNames: */ 14, 6,
	112666	+ /* ColNames: */ 15, 6,
112416	112667	/* iArg: */ 1 },
112417	112668	#endif
112418	112669	#if !defined(SQLITE_OMIT_INTEGRITY_CHECK)
112419	112670	{/* zName: */ "integrity_check",
112420	112671	/* ePragTyp: */ PragTyp_INTEGRITY_CHECK,
		@@ -112457,11 +112708,11 @@
112457	112708	#endif
112458	112709	#if defined(SQLITE_DEBUG) \|\| defined(SQLITE_TEST)
112459	112710	{/* zName: */ "lock_status",
112460	112711	/* ePragTyp: */ PragTyp_LOCK_STATUS,
112461	112712	/* ePragFlg: */ PragFlg_Result0,
112462		- /* ColNames: */ 46, 2,
	112713	+ /* ColNames: */ 47, 2,
112463	112714	/* iArg: */ 0 },
112464	112715	#endif
112465	112716	#if !defined(SQLITE_OMIT_PAGER_PRAGMAS)
112466	112717	{/* zName: */ "locking_mode",
112467	112718	/* ePragTyp: */ PragTyp_LOCKING_MODE,
		@@ -112476,10 +112727,17 @@
112476	112727	{/* zName: */ "mmap_size",
112477	112728	/* ePragTyp: */ PragTyp_MMAP_SIZE,
112478	112729	/* ePragFlg: */ 0,
112479	112730	/* ColNames: */ 0, 0,
112480	112731	/* iArg: */ 0 },
	112732	+#endif
	112733	+ {/* zName: */ "optimize",
	112734	+ /* ePragTyp: */ PragTyp_OPTIMIZE,
	112735	+ /* ePragFlg: */ PragFlg_Result1,
	112736	+ /* ColNames: */ 0, 0,
	112737	+ /* iArg: */ 0 },
	112738	+#if !defined(SQLITE_OMIT_PAGER_PRAGMAS)
112481	112739	{/* zName: */ "page_count",
112482	112740	/* ePragTyp: */ PragTyp_PAGE_COUNT,
112483	112741	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result0\|PragFlg_SchemaReq,
112484	112742	/* ColNames: */ 0, 0,
112485	112743	/* iArg: */ 0 },
		@@ -112574,15 +112832,15 @@
112574	112832	/* ePragFlg: */ PragFlg_Result0\|PragFlg_NoColumns1,
112575	112833	/* ColNames: */ 0, 0,
112576	112834	/* iArg: */ SQLITE_SqlTrace },
112577	112835	#endif
112578	112836	#endif
112579		-#if !defined(SQLITE_OMIT_SCHEMA_PRAGMAS)
	112837	+#if !defined(SQLITE_OMIT_SCHEMA_PRAGMAS) && defined(SQLITE_DEBUG)
112580	112838	{/* zName: */ "stats",
112581	112839	/* ePragTyp: */ PragTyp_STATS,
112582	112840	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result0\|PragFlg_SchemaReq,
112583		- /* ColNames: */ 7, 4,
	112841	+ /* ColNames: */ 7, 5,
112584	112842	/* iArg: */ 0 },
112585	112843	#endif
112586	112844	#if !defined(SQLITE_OMIT_PAGER_PRAGMAS)
112587	112845	{/* zName: */ "synchronous",
112588	112846	/* ePragTyp: */ PragTyp_SYNCHRONOUS,
		@@ -112657,11 +112915,11 @@
112657	112915	/* ColNames: */ 0, 0,
112658	112916	/* iArg: */ 0 },
112659	112917	{/* zName: */ "wal_checkpoint",
112660	112918	/* ePragTyp: */ PragTyp_WAL_CHECKPOINT,
112661	112919	/* ePragFlg: */ PragFlg_NeedSchema,
112662		- /* ColNames: */ 42, 3,
	112920	+ /* ColNames: */ 43, 3,
112663	112921	/* iArg: */ 0 },
112664	112922	#endif
112665	112923	#if !defined(SQLITE_OMIT_FLAG_PRAGMAS)
112666	112924	{/* zName: */ "writable_schema",
112667	112925	/* ePragTyp: */ PragTyp_FLAG,
		@@ -112668,11 +112926,11 @@
112668	112926	/* ePragFlg: */ PragFlg_Result0\|PragFlg_NoColumns1,
112669	112927	/* ColNames: */ 0, 0,
112670	112928	/* iArg: */ SQLITE_WriteSchema\|SQLITE_RecoveryMode },
112671	112929	#endif
112672	112930	};
112673		-/* Number of pragmas: 60 on by default, 73 total. */
	112931	+/* Number of pragmas: 60 on by default, 74 total. */
112674	112932
112675	112933	/************ End of pragma.h ********************************************/
112676	112934	/************ Continuing where we left off in pragma.c *******************/
112677	112935
112678	112936	/*
		@@ -112937,10 +113195,26 @@
112937	113195	lwr = mid + 1;
112938	113196	}
112939	113197	}
112940	113198	return lwr>upr ? 0 : &aPragmaName[mid];
112941	113199	}
	113200	+
	113201	+/*
	113202	+** Helper subroutine for PRAGMA integrity_check:
	113203	+**
	113204	+** Generate code to output a single-column result row with the result
	113205	+** held in register regResult. Decrement the result count and halt if
	113206	+** the maximum number of result rows have been issued.
	113207	+*/
	113208	+static int integrityCheckResultRow(Vdbe *v, int regResult){
	113209	+ int addr;
	113210	+ sqlite3VdbeAddOp2(v, OP_ResultRow, regResult, 1);
	113211	+ addr = sqlite3VdbeAddOp3(v, OP_IfPos, 1, sqlite3VdbeCurrentAddr(v)+2, 1);
	113212	+ VdbeCoverage(v);
	113213	+ sqlite3VdbeAddOp2(v, OP_Halt, 0, 0);
	113214	+ return addr;
	113215	+}
112942	113216
112943	113217	/*
112944	113218	** Process a pragma statement.
112945	113219	**
112946	113220	** Pragmas are of this form:
		@@ -113741,33 +114015,37 @@
113741	114015	}
113742	114016	}
113743	114017	}
113744	114018	break;
113745	114019
	114020	+#ifdef SQLITE_DEBUG
113746	114021	case PragTyp_STATS: {
113747	114022	Index *pIdx;
113748	114023	HashElem *i;
113749		- pParse->nMem = 4;
	114024	+ pParse->nMem = 5;
113750	114025	sqlite3CodeVerifySchema(pParse, iDb);
113751	114026	for(i=sqliteHashFirst(&pDb->pSchema->tblHash); i; i=sqliteHashNext(i)){
113752	114027	Table *pTab = sqliteHashData(i);
113753		- sqlite3VdbeMultiLoad(v, 1, "ssii",
	114028	+ sqlite3VdbeMultiLoad(v, 1, "ssiii",
113754	114029	pTab->zName,
113755	114030	0,
113756	114031	pTab->szTabRow,
113757		- pTab->nRowLogEst);
113758		- sqlite3VdbeAddOp2(v, OP_ResultRow, 1, 4);
	114032	+ pTab->nRowLogEst,
	114033	+ pTab->tabFlags);
	114034	+ sqlite3VdbeAddOp2(v, OP_ResultRow, 1, 5);
113759	114035	for(pIdx=pTab->pIndex; pIdx; pIdx=pIdx->pNext){
113760		- sqlite3VdbeMultiLoad(v, 2, "sii",
	114036	+ sqlite3VdbeMultiLoad(v, 2, "siii",
113761	114037	pIdx->zName,
113762	114038	pIdx->szIdxRow,
113763		- pIdx->aiRowLogEst[0]);
113764		- sqlite3VdbeAddOp2(v, OP_ResultRow, 1, 4);
	114039	+ pIdx->aiRowLogEst[0],
	114040	+ pIdx->hasStat1);
	114041	+ sqlite3VdbeAddOp2(v, OP_ResultRow, 1, 5);
113765	114042	}
113766	114043	}
113767	114044	}
113768	114045	break;
	114046	+#endif
113769	114047
113770	114048	case PragTyp_INDEX_INFO: if( zRight ){
113771	114049	Index *pIdx;
113772	114050	Table *pTab;
113773	114051	pIdx = sqlite3FindIndex(db, zRight, zDb);
		@@ -114020,13 +114298,21 @@
114020	114298	#ifndef SQLITE_INTEGRITY_CHECK_ERROR_MAX
114021	114299	# define SQLITE_INTEGRITY_CHECK_ERROR_MAX 100
114022	114300	#endif
114023	114301
114024	114302	#ifndef SQLITE_OMIT_INTEGRITY_CHECK
114025		- /* Pragma "quick_check" is reduced version of
	114303	+ /* PRAGMA integrity_check
	114304	+ ** PRAGMA integrity_check(N)
	114305	+ ** PRAGMA quick_check
	114306	+ ** PRAGMA quick_check(N)
	114307	+ **
	114308	+ ** Verify the integrity of the database.
	114309	+ **
	114310	+ ** The "quick_check" is reduced version of
114026	114311	** integrity_check designed to detect most database corruption
114027		- ** without most of the overhead of a full integrity-check.
	114312	+ ** without the overhead of cross-checking indexes. Quick_check
	114313	+ ** is linear time wherease integrity_check is O(NlogN).
114028	114314	*/
114029	114315	case PragTyp_INTEGRITY_CHECK: {
114030	114316	int i, j, addr, mxErr;
114031	114317
114032	114318	int isQuick = (sqlite3Tolower(zLeft[0])=='q');
		@@ -114053,11 +114339,11 @@
114053	114339	sqlite3GetInt32(zRight, &mxErr);
114054	114340	if( mxErr<=0 ){
114055	114341	mxErr = SQLITE_INTEGRITY_CHECK_ERROR_MAX;
114056	114342	}
114057	114343	}
114058		- sqlite3VdbeAddOp2(v, OP_Integer, mxErr, 1); /* reg[1] holds errors left */
	114344	+ sqlite3VdbeAddOp2(v, OP_Integer, mxErr-1, 1); /* reg[1] holds errors left */
114059	114345
114060	114346	/* Do an integrity check on each database file */
114061	114347	for(i=0; i<db->nDb; i++){
114062	114348	HashElem *x;
114063	114349	Hash *pTbls;
		@@ -114068,14 +114354,10 @@
114068	114354
114069	114355	if( OMIT_TEMPDB && i==1 ) continue;
114070	114356	if( iDb>=0 && i!=iDb ) continue;
114071	114357
114072	114358	sqlite3CodeVerifySchema(pParse, i);
114073		- addr = sqlite3VdbeAddOp1(v, OP_IfPos, 1); /* Halt if out of errors */
114074		- VdbeCoverage(v);
114075		- sqlite3VdbeAddOp2(v, OP_Halt, 0, 0);
114076		- sqlite3VdbeJumpHere(v, addr);
114077	114359
114078	114360	/* Do an integrity check of the B-Tree
114079	114361	**
114080	114362	** Begin by finding the root pages numbers
114081	114363	** for all tables and indices in the database.
		@@ -114111,29 +114393,30 @@
114111	114393	sqlite3VdbeAddOp4(v, OP_String8, 0, 3, 0,
114112	114394	sqlite3MPrintf(db, "* in database %s *\n", db->aDb[i].zDbSName),
114113	114395	P4_DYNAMIC);
114114	114396	sqlite3VdbeAddOp3(v, OP_Move, 2, 4, 1);
114115	114397	sqlite3VdbeAddOp3(v, OP_Concat, 4, 3, 2);
114116		- sqlite3VdbeAddOp2(v, OP_ResultRow, 2, 1);
	114398	+ integrityCheckResultRow(v, 2);
114117	114399	sqlite3VdbeJumpHere(v, addr);
114118	114400
114119	114401	/* Make sure all the indices are constructed correctly.
114120	114402	*/
114121		- for(x=sqliteHashFirst(pTbls); x && !isQuick; x=sqliteHashNext(x)){
	114403	+ for(x=sqliteHashFirst(pTbls); x; x=sqliteHashNext(x)){
114122	114404	Table *pTab = sqliteHashData(x);
114123	114405	Index pIdx, pPk;
114124	114406	Index *pPrior = 0;
114125	114407	int loopTop;
114126	114408	int iDataCur, iIdxCur;
114127	114409	int r1 = -1;
114128	114410
114129		- if( pTab->pIndex==0 ) continue;
	114411	+ if( pTab->pCheck==0
	114412	+ && (pTab->tabFlags & TF_HasNotNull)==0
	114413	+ && (pTab->pIndex==0 \|\| isQuick)
	114414	+ ){
	114415	+ continue; /* No additional checks needed for this table */
	114416	+ }
114130	114417	pPk = HasRowid(pTab) ? 0 : sqlite3PrimaryKeyIndex(pTab);
114131		- addr = sqlite3VdbeAddOp1(v, OP_IfPos, 1); /* Stop if out of errors */
114132		- VdbeCoverage(v);
114133		- sqlite3VdbeAddOp2(v, OP_Halt, 0, 0);
114134		- sqlite3VdbeJumpHere(v, addr);
114135	114418	sqlite3ExprCacheClear(pParse);
114136	114419	sqlite3OpenTableAndIndices(pParse, pTab, OP_OpenRead, 0,
114137	114420	1, 0, &iDataCur, &iIdxCur);
114138	114421	sqlite3VdbeAddOp2(v, OP_Integer, 0, 7);
114139	114422	for(j=0, pIdx=pTab->pIndex; pIdx; pIdx=pIdx->pNext, j++){
		@@ -114144,28 +114427,46 @@
114144	114427	sqlite3VdbeAddOp2(v, OP_Rewind, iDataCur, 0); VdbeCoverage(v);
114145	114428	loopTop = sqlite3VdbeAddOp2(v, OP_AddImm, 7, 1);
114146	114429	/* Verify that all NOT NULL columns really are NOT NULL */
114147	114430	for(j=0; j<pTab->nCol; j++){
114148	114431	char *zErr;
114149		- int jmp2, jmp3;
	114432	+ int jmp2;
114150	114433	if( j==pTab->iPKey ) continue;
114151	114434	if( pTab->aCol[j].notNull==0 ) continue;
114152	114435	sqlite3ExprCodeGetColumnOfTable(v, pTab, iDataCur, j, 3);
114153	114436	sqlite3VdbeChangeP5(v, OPFLAG_TYPEOFARG);
114154	114437	jmp2 = sqlite3VdbeAddOp1(v, OP_NotNull, 3); VdbeCoverage(v);
114155		- sqlite3VdbeAddOp2(v, OP_AddImm, 1, -1); /* Decrement error limit */
114156	114438	zErr = sqlite3MPrintf(db, "NULL value in %s.%s", pTab->zName,
114157	114439	pTab->aCol[j].zName);
114158	114440	sqlite3VdbeAddOp4(v, OP_String8, 0, 3, 0, zErr, P4_DYNAMIC);
114159		- sqlite3VdbeAddOp2(v, OP_ResultRow, 3, 1);
114160		- jmp3 = sqlite3VdbeAddOp1(v, OP_IfPos, 1); VdbeCoverage(v);
114161		- sqlite3VdbeAddOp0(v, OP_Halt);
	114441	+ integrityCheckResultRow(v, 3);
114162	114442	sqlite3VdbeJumpHere(v, jmp2);
114163		- sqlite3VdbeJumpHere(v, jmp3);
	114443	+ }
	114444	+ /* Verify CHECK constraints */
	114445	+ if( pTab->pCheck && (db->flags & SQLITE_IgnoreChecks)==0 ){
	114446	+ int addrCkFault = sqlite3VdbeMakeLabel(v);
	114447	+ int addrCkOk = sqlite3VdbeMakeLabel(v);
	114448	+ ExprList *pCheck = pTab->pCheck;
	114449	+ char *zErr;
	114450	+ int k;
	114451	+ pParse->iSelfTab = iDataCur;
	114452	+ sqlite3ExprCachePush(pParse);
	114453	+ for(k=pCheck->nExpr-1; k>0; k--){
	114454	+ sqlite3ExprIfFalse(pParse, pCheck->a[k].pExpr, addrCkFault, 0);
	114455	+ }
	114456	+ sqlite3ExprIfTrue(pParse, pCheck->a[0].pExpr, addrCkOk,
	114457	+ SQLITE_JUMPIFNULL);
	114458	+ sqlite3VdbeResolveLabel(v, addrCkFault);
	114459	+ zErr = sqlite3MPrintf(db, "CHECK constraint failed in %s",
	114460	+ pTab->zName);
	114461	+ sqlite3VdbeAddOp4(v, OP_String8, 0, 3, 0, zErr, P4_DYNAMIC);
	114462	+ integrityCheckResultRow(v, 3);
	114463	+ sqlite3VdbeResolveLabel(v, addrCkOk);
	114464	+ sqlite3ExprCachePop(pParse);
114164	114465	}
114165	114466	/* Validate index entries for the current row */
114166		- for(j=0, pIdx=pTab->pIndex; pIdx; pIdx=pIdx->pNext, j++){
	114467	+ for(j=0, pIdx=pTab->pIndex; pIdx && !isQuick; pIdx=pIdx->pNext, j++){
114167	114468	int jmp2, jmp3, jmp4, jmp5;
114168	114469	int ckUniq = sqlite3VdbeMakeLabel(v);
114169	114470	if( pPk==pIdx ) continue;
114170	114471	r1 = sqlite3GenerateIndexKey(pParse, pIdx, iDataCur, 0, 0, &jmp3,
114171	114472	pPrior, r1);
		@@ -114172,20 +114473,17 @@
114172	114473	pPrior = pIdx;
114173	114474	sqlite3VdbeAddOp2(v, OP_AddImm, 8+j, 1); /* increment entry count */
114174	114475	/* Verify that an index entry exists for the current table row */
114175	114476	jmp2 = sqlite3VdbeAddOp4Int(v, OP_Found, iIdxCur+j, ckUniq, r1,
114176	114477	pIdx->nColumn); VdbeCoverage(v);
114177		- sqlite3VdbeAddOp2(v, OP_AddImm, 1, -1); /* Decrement error limit */
114178	114478	sqlite3VdbeLoadString(v, 3, "row ");
114179	114479	sqlite3VdbeAddOp3(v, OP_Concat, 7, 3, 3);
114180	114480	sqlite3VdbeLoadString(v, 4, " missing from index ");
114181	114481	sqlite3VdbeAddOp3(v, OP_Concat, 4, 3, 3);
114182	114482	jmp5 = sqlite3VdbeLoadString(v, 4, pIdx->zName);
114183	114483	sqlite3VdbeAddOp3(v, OP_Concat, 4, 3, 3);
114184		- sqlite3VdbeAddOp2(v, OP_ResultRow, 3, 1);
114185		- jmp4 = sqlite3VdbeAddOp1(v, OP_IfPos, 1); VdbeCoverage(v);
114186		- sqlite3VdbeAddOp0(v, OP_Halt);
	114484	+ jmp4 = integrityCheckResultRow(v, 3);
114187	114485	sqlite3VdbeJumpHere(v, jmp2);
114188	114486	/* For UNIQUE indexes, verify that only one entry exists with the
114189	114487	** current key. The entry is unique if (1) any column is NULL
114190	114488	** or (2) the next entry has a different key */
114191	114489	if( IsUniqueIndex(pIdx) ){
		@@ -114202,11 +114500,10 @@
114202	114500	jmp6 = sqlite3VdbeAddOp1(v, OP_Next, iIdxCur+j); VdbeCoverage(v);
114203	114501	sqlite3VdbeGoto(v, uniqOk);
114204	114502	sqlite3VdbeJumpHere(v, jmp6);
114205	114503	sqlite3VdbeAddOp4Int(v, OP_IdxGT, iIdxCur+j, uniqOk, r1,
114206	114504	pIdx->nKeyCol); VdbeCoverage(v);
114207		- sqlite3VdbeAddOp2(v, OP_AddImm, 1, -1); /* Decrement error limit */
114208	114505	sqlite3VdbeLoadString(v, 3, "non-unique entry in index ");
114209	114506	sqlite3VdbeGoto(v, jmp5);
114210	114507	sqlite3VdbeResolveLabel(v, uniqOk);
114211	114508	}
114212	114509	sqlite3VdbeJumpHere(v, jmp4);
		@@ -114213,40 +114510,39 @@
114213	114510	sqlite3ResolvePartIdxLabel(pParse, jmp3);
114214	114511	}
114215	114512	sqlite3VdbeAddOp2(v, OP_Next, iDataCur, loopTop); VdbeCoverage(v);
114216	114513	sqlite3VdbeJumpHere(v, loopTop-1);
114217	114514	#ifndef SQLITE_OMIT_BTREECOUNT
114218		- sqlite3VdbeLoadString(v, 2, "wrong # of entries in index ");
114219		- for(j=0, pIdx=pTab->pIndex; pIdx; pIdx=pIdx->pNext, j++){
114220		- if( pPk==pIdx ) continue;
114221		- addr = sqlite3VdbeCurrentAddr(v);
114222		- sqlite3VdbeAddOp2(v, OP_IfPos, 1, addr+2); VdbeCoverage(v);
114223		- sqlite3VdbeAddOp2(v, OP_Halt, 0, 0);
114224		- sqlite3VdbeAddOp2(v, OP_Count, iIdxCur+j, 3);
114225		- sqlite3VdbeAddOp3(v, OP_Eq, 8+j, addr+8, 3); VdbeCoverage(v);
114226		- sqlite3VdbeChangeP5(v, SQLITE_NOTNULL);
114227		- sqlite3VdbeAddOp2(v, OP_AddImm, 1, -1);
114228		- sqlite3VdbeLoadString(v, 3, pIdx->zName);
114229		- sqlite3VdbeAddOp3(v, OP_Concat, 3, 2, 7);
114230		- sqlite3VdbeAddOp2(v, OP_ResultRow, 7, 1);
	114515	+ if( !isQuick ){
	114516	+ sqlite3VdbeLoadString(v, 2, "wrong # of entries in index ");
	114517	+ for(j=0, pIdx=pTab->pIndex; pIdx; pIdx=pIdx->pNext, j++){
	114518	+ if( pPk==pIdx ) continue;
	114519	+ sqlite3VdbeAddOp2(v, OP_Count, iIdxCur+j, 3);
	114520	+ addr = sqlite3VdbeAddOp3(v, OP_Eq, 8+j, 0, 3); VdbeCoverage(v);
	114521	+ sqlite3VdbeChangeP5(v, SQLITE_NOTNULL);
	114522	+ sqlite3VdbeLoadString(v, 3, pIdx->zName);
	114523	+ sqlite3VdbeAddOp3(v, OP_Concat, 3, 2, 7);
	114524	+ integrityCheckResultRow(v, 7);
	114525	+ sqlite3VdbeJumpHere(v, addr);
	114526	+ }
114231	114527	}
114232	114528	#endif /* SQLITE_OMIT_BTREECOUNT */
114233	114529	}
114234	114530	}
114235	114531	{
114236	114532	static const int iLn = VDBE_OFFSET_LINENO(2);
114237	114533	static const VdbeOpList endCode[] = {
114238	114534	{ OP_AddImm, 1, 0, 0}, /* 0 */
114239		- { OP_If, 1, 4, 0}, /* 1 */
	114535	+ { OP_IfNotZero, 1, 4, 0}, /* 1 */
114240	114536	{ OP_String8, 0, 3, 0}, /* 2 */
114241	114537	{ OP_ResultRow, 3, 1, 0}, /* 3 */
114242	114538	};
114243	114539	VdbeOp *aOp;
114244	114540
114245	114541	aOp = sqlite3VdbeAddOpList(v, ArraySize(endCode), endCode, iLn);
114246	114542	if( aOp ){
114247		- aOp[0].p2 = -mxErr;
	114543	+ aOp[0].p2 = 1-mxErr;
114248	114544	aOp[2].p4type = P4_STATIC;
114249	114545	aOp[2].p4.z = "ok";
114250	114546	}
114251	114547	}
114252	114548	}
		@@ -114466,10 +114762,122 @@
114466	114762	*/
114467	114763	case PragTyp_SHRINK_MEMORY: {
114468	114764	sqlite3_db_release_memory(db);
114469	114765	break;
114470	114766	}
	114767	+
	114768	+ /*
	114769	+ ** PRAGMA optimize
	114770	+ ** PRAGMA optimize(MASK)
	114771	+ ** PRAGMA schema.optimize
	114772	+ ** PRAGMA schema.optimize(MASK)
	114773	+ **
	114774	+ ** Attempt to optimize the database. All schemas are optimized in the first
	114775	+ ** two forms, and only the specified schema is optimized in the latter two.
	114776	+ **
	114777	+ ** The details of optimizations performed by this pragma are expected
	114778	+ ** to change and improve over time. Applications should anticipate that
	114779	+ ** this pragma will perform new optimizations in future releases.
	114780	+ **
	114781	+ ** The optional argument is a bitmask of optimizations to perform:
	114782	+ **
	114783	+ ** 0x0001 Debugging mode. Do not actually perform any optimizations
	114784	+ ** but instead return one line of text for each optimization
	114785	+ ** that would have been done. Off by default.
	114786	+ **
	114787	+ ** 0x0002 Run ANALYZE on tables that might benefit. On by default.
	114788	+ ** See below for additional information.
	114789	+ **
	114790	+ ** 0x0004 (Not yet implemented) Record usage and performance
	114791	+ ** information from the current session in the
	114792	+ ** database file so that it will be available to "optimize"
	114793	+ ** pragmas run by future database connections.
	114794	+ **
	114795	+ ** 0x0008 (Not yet implemented) Create indexes that might have
	114796	+ ** been helpful to recent queries
	114797	+ **
	114798	+ The default MASK is and always shall be 0xfffe. 0xfffe means perform all of the optimizations listed above except Debug Mode, including new
	114799	+ ** optimizations that have not yet been invented. If new optimizations are
	114800	+ ** ever added that should be off by default, those off-by-default
	114801	+ ** optimizations will have bitmasks of 0x10000 or larger.
	114802	+ **
	114803	+ ** DETERMINATION OF WHEN TO RUN ANALYZE
	114804	+ **
	114805	+ ** In the current implementation, a table is analyzed if only if all of
	114806	+ ** the following are true:
	114807	+ **
	114808	+ ** (1) MASK bit 0x02 is set.
	114809	+ **
	114810	+ ** (2) The query planner used sqlite_stat1-style statistics for one or
	114811	+ ** more indexes of the table at some point during the lifetime of
	114812	+ ** the current connection.
	114813	+ **
	114814	+ ** (3) One or more indexes of the table are currently unanalyzed OR
	114815	+ ** the number of rows in the table has increased by 25 times or more
	114816	+ ** since the last time ANALYZE was run.
	114817	+ **
	114818	+ ** The rules for when tables are analyzed are likely to change in
	114819	+ ** future releases.
	114820	+ */
	114821	+ case PragTyp_OPTIMIZE: {
	114822	+ int iDbLast; /* Loop termination point for the schema loop */
	114823	+ int iTabCur; /* Cursor for a table whose size needs checking */
	114824	+ HashElem k; / Loop over tables of a schema */
	114825	+ Schema pSchema; / The current schema */
	114826	+ Table pTab; / A table in the schema */
	114827	+ Index pIdx; / An index of the table */
	114828	+ LogEst szThreshold; /* Size threshold above which reanalysis is needd */
	114829	+ char zSubSql; / SQL statement for the OP_SqlExec opcode */
	114830	+ u32 opMask; /* Mask of operations to perform */
	114831	+
	114832	+ if( zRight ){
	114833	+ opMask = (u32)sqlite3Atoi(zRight);
	114834	+ if( (opMask & 0x02)==0 ) break;
	114835	+ }else{
	114836	+ opMask = 0xfffe;
	114837	+ }
	114838	+ iTabCur = pParse->nTab++;
	114839	+ for(iDbLast = zDb?iDb:db->nDb-1; iDb<=iDbLast; iDb++){
	114840	+ if( iDb==1 ) continue;
	114841	+ sqlite3CodeVerifySchema(pParse, iDb);
	114842	+ pSchema = db->aDb[iDb].pSchema;
	114843	+ for(k=sqliteHashFirst(&pSchema->tblHash); k; k=sqliteHashNext(k)){
	114844	+ pTab = (Table*)sqliteHashData(k);
	114845	+
	114846	+ /* If table pTab has not been used in a way that would benefit from
	114847	+ ** having analysis statistics during the current session, then skip it.
	114848	+ ** This also has the effect of skipping virtual tables and views */
	114849	+ if( (pTab->tabFlags & TF_StatsUsed)==0 ) continue;
	114850	+
	114851	+ /* Reanalyze if the table is 25 times larger than the last analysis */
	114852	+ szThreshold = pTab->nRowLogEst + 46; assert( sqlite3LogEst(25)==46 );
	114853	+ for(pIdx=pTab->pIndex; pIdx; pIdx=pIdx->pNext){
	114854	+ if( !pIdx->hasStat1 ){
	114855	+ szThreshold = 0; /* Always analyze if any index lacks statistics */
	114856	+ break;
	114857	+ }
	114858	+ }
	114859	+ if( szThreshold ){
	114860	+ sqlite3OpenTable(pParse, iTabCur, iDb, pTab, OP_OpenRead);
	114861	+ sqlite3VdbeAddOp3(v, OP_IfSmaller, iTabCur,
	114862	+ sqlite3VdbeCurrentAddr(v)+2+(opMask&1), szThreshold);
	114863	+ VdbeCoverage(v);
	114864	+ }
	114865	+ zSubSql = sqlite3MPrintf(db, "ANALYZE \"%w\".\"%w\"",
	114866	+ db->aDb[iDb].zDbSName, pTab->zName);
	114867	+ if( opMask & 0x01 ){
	114868	+ int r1 = sqlite3GetTempReg(pParse);
	114869	+ sqlite3VdbeAddOp4(v, OP_String8, 0, r1, 0, zSubSql, P4_DYNAMIC);
	114870	+ sqlite3VdbeAddOp2(v, OP_ResultRow, r1, 1);
	114871	+ }else{
	114872	+ sqlite3VdbeAddOp4(v, OP_SqlExec, 0, 0, 0, zSubSql, P4_DYNAMIC);
	114873	+ }
	114874	+ }
	114875	+ }
	114876	+ sqlite3VdbeAddOp0(v, OP_Expire);
	114877	+ break;
	114878	+ }
114471	114879
114472	114880	/*
114473	114881	** PRAGMA busy_timeout
114474	114882	** PRAGMA busy_timeout = N
114475	114883	**
		@@ -123782,12 +124190,29 @@
123782	124190	** transient would cause the database file to appear to be deleted
123783	124191	** following reboot.
123784	124192	*/
123785	124193	SQLITE_PRIVATE void sqlite3Vacuum(Parse pParse, Token pNm){
123786	124194	Vdbe *v = sqlite3GetVdbe(pParse);
123787		- int iDb = pNm ? sqlite3TwoPartName(pParse, pNm, pNm, &pNm) : 0;
123788		- if( v && (iDb>=2 \|\| iDb==0) ){
	124195	+ int iDb = 0;
	124196	+ if( v==0 ) return;
	124197	+ if( pNm ){
	124198	+#ifndef SQLITE_BUG_COMPATIBLE_20160819
	124199	+ /* Default behavior: Report an error if the argument to VACUUM is
	124200	+ ** not recognized */
	124201	+ iDb = sqlite3TwoPartName(pParse, pNm, pNm, &pNm);
	124202	+ if( iDb<0 ) return;
	124203	+#else
	124204	+ /* When SQLITE_BUG_COMPATIBLE_20160819 is defined, unrecognized arguments
	124205	+ ** to VACUUM are silently ignored. This is a back-out of a bug fix that
	124206	+ ** occurred on 2016-08-19 (https://www.sqlite.org/src/info/083f9e6270).
	124207	+ ** The buggy behavior is required for binary compatibility with some
	124208	+ ** legacy applications. */
	124209	+ iDb = sqlite3FindDb(pParse->db, pNm);
	124210	+ if( iDb<0 ) iDb = 0;
	124211	+#endif
	124212	+ }
	124213	+ if( iDb!=1 ){
123789	124214	sqlite3VdbeAddOp1(v, OP_Vacuum, iDb);
123790	124215	sqlite3VdbeUsesBtree(v, iDb);
123791	124216	}
123792	124217	return;
123793	124218	}
		@@ -124377,12 +124802,11 @@
124377	124802
124378	124803	db = pParse->db;
124379	124804	iDb = sqlite3SchemaToIndex(db, pTable->pSchema);
124380	124805	assert( iDb>=0 );
124381	124806
124382		- pTable->tabFlags \|= TF_Virtual;
124383		- pTable->nModuleArg = 0;
	124807	+ assert( pTable->nModuleArg==0 );
124384	124808	addModuleArgument(db, pTable, sqlite3NameFromToken(db, pModuleName));
124385	124809	addModuleArgument(db, pTable, 0);
124386	124810	addModuleArgument(db, pTable, sqlite3DbStrDup(db, pTable->zName));
124387	124811	assert( (pParse->sNameToken.z==pName2->z && pName2->z!=0)
124388	124812	\|\| (pParse->sNameToken.z==pName1->z && pName2->z==0)
		@@ -124666,11 +125090,11 @@
124666	125090	const char *zMod;
124667	125091	Module *pMod;
124668	125092	int rc;
124669	125093
124670	125094	assert( pTab );
124671		- if( (pTab->tabFlags & TF_Virtual)==0 \|\| sqlite3GetVTable(db, pTab) ){
	125095	+ if( !IsVirtual(pTab) \|\| sqlite3GetVTable(db, pTab) ){
124672	125096	return SQLITE_OK;
124673	125097	}
124674	125098
124675	125099	/* Locate the required virtual table module */
124676	125100	zMod = pTab->azModuleArg[0];
		@@ -124736,11 +125160,11 @@
124736	125160	Table *pTab;
124737	125161	Module *pMod;
124738	125162	const char *zMod;
124739	125163
124740	125164	pTab = sqlite3FindTable(db, zTab, db->aDb[iDb].zDbSName);
124741		- assert( pTab && (pTab->tabFlags & TF_Virtual)!=0 && !pTab->pVTable );
	125165	+ assert( pTab && IsVirtual(pTab) && !pTab->pVTable );
124742	125166
124743	125167	/* Locate the required virtual table module */
124744	125168	zMod = pTab->azModuleArg[0];
124745	125169	pMod = (Module*)sqlite3HashFind(&db->aModule, zMod);
124746	125170
		@@ -124790,11 +125214,11 @@
124790	125214	sqlite3Error(db, SQLITE_MISUSE);
124791	125215	sqlite3_mutex_leave(db->mutex);
124792	125216	return SQLITE_MISUSE_BKPT;
124793	125217	}
124794	125218	pTab = pCtx->pTab;
124795		- assert( (pTab->tabFlags & TF_Virtual)!=0 );
	125219	+ assert( IsVirtual(pTab) );
124796	125220
124797	125221	pParse = sqlite3StackAllocZero(db, sizeof(*pParse));
124798	125222	if( pParse==0 ){
124799	125223	rc = SQLITE_NOMEM_BKPT;
124800	125224	}else{
		@@ -124804,11 +125228,11 @@
124804	125228
124805	125229	if( SQLITE_OK==sqlite3RunParser(pParse, zCreateTable, &zErr)
124806	125230	&& pParse->pNewTable
124807	125231	&& !db->mallocFailed
124808	125232	&& !pParse->pNewTable->pSelect
124809		- && (pParse->pNewTable->tabFlags & TF_Virtual)==0
	125233	+ && !IsVirtual(pParse->pNewTable)
124810	125234	){
124811	125235	if( !pTab->aCol ){
124812	125236	Table *pNew = pParse->pNewTable;
124813	125237	Index *pIdx;
124814	125238	pTab->aCol = pNew->aCol;
		@@ -125093,11 +125517,11 @@
125093	125517	/* Check to see the left operand is a column in a virtual table */
125094	125518	if( NEVER(pExpr==0) ) return pDef;
125095	125519	if( pExpr->op!=TK_COLUMN ) return pDef;
125096	125520	pTab = pExpr->pTab;
125097	125521	if( NEVER(pTab==0) ) return pDef;
125098		- if( (pTab->tabFlags & TF_Virtual)==0 ) return pDef;
	125522	+ if( !IsVirtual(pTab) ) return pDef;
125099	125523	pVtab = sqlite3GetVTable(db, pTab)->pVtab;
125100	125524	assert( pVtab!=0 );
125101	125525	assert( pVtab->pModule!=0 );
125102	125526	pMod = (sqlite3_module *)pVtab->pModule;
125103	125527	if( pMod->xFindFunction==0 ) return pDef;
		@@ -125188,12 +125612,11 @@
125188	125612	return 0;
125189	125613	}
125190	125614	pMod->pEpoTab = pTab;
125191	125615	pTab->nTabRef = 1;
125192	125616	pTab->pSchema = db->aDb[0].pSchema;
125193		- pTab->tabFlags \|= TF_Virtual;
125194		- pTab->nModuleArg = 0;
	125617	+ assert( pTab->nModuleArg==0 );
125195	125618	pTab->iPKey = -1;
125196	125619	addModuleArgument(db, pTab, sqlite3DbStrDup(db, pTab->zName));
125197	125620	addModuleArgument(db, pTab, 0);
125198	125621	addModuleArgument(db, pTab, sqlite3DbStrDup(db, pTab->zName));
125199	125622	rc = vtabCallConstructor(db, pTab, pMod, pModule->xConnect, &zErr);
		@@ -125260,11 +125683,11 @@
125260	125683	case SQLITE_VTAB_CONSTRAINT_SUPPORT: {
125261	125684	VtabCtx *p = db->pVtabCtx;
125262	125685	if( !p ){
125263	125686	rc = SQLITE_MISUSE_BKPT;
125264	125687	}else{
125265		- assert( p->pTab==0 \|\| (p->pTab->tabFlags & TF_Virtual)!=0 );
	125688	+ assert( p->pTab==0 \|\| IsVirtual(p->pTab) );
125266	125689	p->pVTable->bConstraint = (u8)va_arg(ap, int);
125267	125690	}
125268	125691	break;
125269	125692	}
125270	125693	default:
		@@ -125699,12 +126122,17 @@
125699	126122	WhereOrSet pOrSet; / Record best loops here, if not NULL */
125700	126123	#ifdef SQLITE_ENABLE_STAT3_OR_STAT4
125701	126124	UnpackedRecord pRec; / Probe for stat4 (if required) */
125702	126125	int nRecValid; /* Number of valid fields currently in pRec */
125703	126126	#endif
	126127	+ unsigned int bldFlags; /* SQLITE_BLDF_* flags */
125704	126128	};
125705	126129
	126130	+/* Allowed values for WhereLoopBuider.bldFlags */
	126131	+#define SQLITE_BLDF_INDEXED 0x0001 /* An index is used */
	126132	+#define SQLITE_BLDF_UNIQUE 0x0002 /* All keys of a UNIQUE index used */
	126133	+
125706	126134	/*
125707	126135	** The WHERE clause processing routine has two halves. The
125708	126136	** first part does the start of the WHERE loop and the second
125709	126137	** half does the tail of the WHERE loop. An instance of
125710	126138	** this structure is returned by the first half and passed
		@@ -125715,11 +126143,11 @@
125715	126143	*/
125716	126144	struct WhereInfo {
125717	126145	Parse pParse; / Parsing and code generating context */
125718	126146	SrcList pTabList; / List of tables in the join */
125719	126147	ExprList pOrderBy; / The ORDER BY clause or NULL */
125720		- ExprList pDistinctSet; / DISTINCT over all these values */
	126148	+ ExprList pResultSet; / Result set of the query */
125721	126149	LogEst iLimit; /* LIMIT if wctrlFlags has WHERE_USE_LIMIT */
125722	126150	int aiCurOnePass[2]; /* OP_OpenWrite cursors for the ONEPASS opt */
125723	126151	int iContinue; /* Jump here to continue with next record */
125724	126152	int iBreak; /* Jump here to break out of the loop */
125725	126153	int savedNQueryLoop; /* pParse->nQueryLoop outside the WHERE loop */
		@@ -126899,10 +127327,11 @@
126899	127327	Parse pParse; / Parsing context */
126900	127328	sqlite3 db; / Database connection */
126901	127329	Vdbe v; / The prepared stmt under constructions */
126902	127330	struct SrcList_item pTabItem; / FROM clause term being coded */
126903	127331	int addrBrk; /* Jump here to break out of the loop */
	127332	+ int addrHalt; /* addrBrk for the outermost loop */
126904	127333	int addrCont; /* Jump here to continue with next cycle */
126905	127334	int iRowidReg = 0; /* Rowid is stored in this register, if not zero */
126906	127335	int iReleaseReg = 0; /* Temp register to free before returning */
126907	127336
126908	127337	pParse = pWInfo->pParse;
		@@ -126939,10 +127368,15 @@
126939	127368	if( pLevel->iFrom>0 && (pTabItem[0].fg.jointype & JT_LEFT)!=0 ){
126940	127369	pLevel->iLeftJoin = ++pParse->nMem;
126941	127370	sqlite3VdbeAddOp2(v, OP_Integer, 0, pLevel->iLeftJoin);
126942	127371	VdbeComment((v, "init LEFT JOIN no-match flag"));
126943	127372	}
	127373	+
	127374	+ /* Compute a safe address to jump to if we discover that the table for
	127375	+ ** this loop is empty and can never contribute content. */
	127376	+ for(j=iLevel; j>0 && pWInfo->a[j].iLeftJoin==0; j--){}
	127377	+ addrHalt = pWInfo->a[j].addrBrk;
126944	127378
126945	127379	/* Special case of a FROM clause subquery implemented as a co-routine */
126946	127380	if( pTabItem->fg.viaCoroutine ){
126947	127381	int regYield = pTabItem->regReturn;
126948	127382	sqlite3VdbeAddOp3(v, OP_InitCoroutine, regYield, 0, pTabItem->addrFillSub);
		@@ -127124,11 +127558,11 @@
127124	127558	VdbeCoverageIf(v, pX->op==TK_LT);
127125	127559	VdbeCoverageIf(v, pX->op==TK_GE);
127126	127560	sqlite3ExprCacheAffinityChange(pParse, r1, 1);
127127	127561	sqlite3ReleaseTempReg(pParse, rTemp);
127128	127562	}else{
127129		- sqlite3VdbeAddOp2(v, bRev ? OP_Last : OP_Rewind, iCur, addrBrk);
	127563	+ sqlite3VdbeAddOp2(v, bRev ? OP_Last : OP_Rewind, iCur, addrHalt);
127130	127564	VdbeCoverageIf(v, bRev==0);
127131	127565	VdbeCoverageIf(v, bRev!=0);
127132	127566	}
127133	127567	if( pEnd ){
127134	127568	Expr *pX;
		@@ -127770,11 +128204,11 @@
127770	128204	pLevel->op = OP_Noop;
127771	128205	}else{
127772	128206	codeCursorHint(pTabItem, pWInfo, pLevel, 0);
127773	128207	pLevel->op = aStep[bRev];
127774	128208	pLevel->p1 = iCur;
127775		- pLevel->p2 = 1 + sqlite3VdbeAddOp2(v, aStart[bRev], iCur, addrBrk);
	128209	+ pLevel->p2 = 1 + sqlite3VdbeAddOp2(v, aStart[bRev], iCur, addrHalt);
127776	128210	VdbeCoverageIf(v, bRev==0);
127777	128211	VdbeCoverageIf(v, bRev!=0);
127778	128212	pLevel->p5 = SQLITE_STMTSTATUS_FULLSCAN_STEP;
127779	128213	}
127780	128214	}
		@@ -128095,19 +128529,10 @@
128095	128529	#ifdef SQLITE_EBCDIC
128096	128530	if( *pnoCase ) return 0;
128097	128531	#endif
128098	128532	pList = pExpr->x.pList;
128099	128533	pLeft = pList->a[1].pExpr;
128100		- if( pLeft->op!=TK_COLUMN
128101		- \|\| sqlite3ExprAffinity(pLeft)!=SQLITE_AFF_TEXT
128102		- \|\| IsVirtual(pLeft->pTab) /* Value might be numeric */
128103		- ){
128104		- /* IMP: R-02065-49465 The left-hand side of the LIKE or GLOB operator must
128105		- ** be the name of an indexed column with TEXT affinity. */
128106		- return 0;
128107		- }
128108		- assert( pLeft->iColumn!=(-1) ); /* Because IPK never has AFF_TEXT */
128109	128534
128110	128535	pRight = sqlite3ExprSkipCollate(pList->a[0].pExpr);
128111	128536	op = pRight->op;
128112	128537	if( op==TK_VARIABLE ){
128113	128538	Vdbe *pReprepare = pParse->pReprepare;
		@@ -128120,10 +128545,27 @@
128120	128545	assert( pRight->op==TK_VARIABLE \|\| pRight->op==TK_REGISTER );
128121	128546	}else if( op==TK_STRING ){
128122	128547	z = pRight->u.zToken;
128123	128548	}
128124	128549	if( z ){
	128550	+
	128551	+ /* If the RHS begins with a digit or a minus sign, then the LHS must
	128552	+ ** be an ordinary column (not a virtual table column) with TEXT affinity.
	128553	+ ** Otherwise the LHS might be numeric and "lhs >= rhs" would be false
	128554	+ ** even though "lhs LIKE rhs" is true. But if the RHS does not start
	128555	+ ** with a digit or '-', then "lhs LIKE rhs" will always be false if
	128556	+ ** the LHS is numeric and so the optimization still works.
	128557	+ */
	128558	+ if( sqlite3Isdigit(z[0]) \|\| z[0]=='-' ){
	128559	+ if( pLeft->op!=TK_COLUMN
	128560	+ \|\| sqlite3ExprAffinity(pLeft)!=SQLITE_AFF_TEXT
	128561	+ \|\| IsVirtual(pLeft->pTab) /* Value might be numeric */
	128562	+ ){
	128563	+ sqlite3ValueFree(pVal);
	128564	+ return 0;
	128565	+ }
	128566	+ }
128125	128567	cnt = 0;
128126	128568	while( (c=z[cnt])!=0 && c!=wc[0] && c!=wc[1] && c!=wc[2] ){
128127	128569	cnt++;
128128	128570	}
128129	128571	if( cnt!=0 && 255!=(u8)z[cnt-1] ){
		@@ -128748,11 +129190,11 @@
128748	129190	iCur = pFrom->a[i].iCursor;
128749	129191	for(pIdx=pFrom->a[i].pTab->pIndex; pIdx; pIdx=pIdx->pNext){
128750	129192	if( pIdx->aColExpr==0 ) continue;
128751	129193	for(i=0; i<pIdx->nKeyCol; i++){
128752	129194	if( pIdx->aiColumn[i]!=XN_EXPR ) continue;
128753		- if( sqlite3ExprCompare(pExpr, pIdx->aColExpr->a[i].pExpr, iCur)==0 ){
	129195	+ if( sqlite3ExprCompareSkip(pExpr, pIdx->aColExpr->a[i].pExpr, iCur)==0 ){
128754	129196	*piCur = iCur;
128755	129197	*piColumn = XN_EXPR;
128756	129198	return 1;
128757	129199	}
128758	129200	}
		@@ -129539,11 +129981,12 @@
129539	129981	do{
129540	129982	for(pTerm=pWC->a+k; k<pWC->nTerm; k++, pTerm++){
129541	129983	if( pTerm->leftCursor==iCur
129542	129984	&& pTerm->u.leftColumn==iColumn
129543	129985	&& (iColumn!=XN_EXPR
129544		- \|\| sqlite3ExprCompare(pTerm->pExpr->pLeft,pScan->pIdxExpr,iCur)==0)
	129986	+ \|\| sqlite3ExprCompareSkip(pTerm->pExpr->pLeft,
	129987	+ pScan->pIdxExpr,iCur)==0)
129545	129988	&& (pScan->iEquiv<=1 \|\| !ExprHasProperty(pTerm->pExpr, EP_FromJoin))
129546	129989	){
129547	129990	if( (pTerm->eOperator & WO_EQUIV)!=0
129548	129991	&& pScan->nEquiv<ArraySize(pScan->aiCur)
129549	129992	&& (pX = sqlite3ExprSkipCollate(pTerm->pExpr->pRight))->op==TK_COLUMN
		@@ -131716,10 +132159,15 @@
131716	132159	testcase( eOp & WO_IS );
131717	132160	testcase( eOp & WO_ISNULL );
131718	132161	continue;
131719	132162	}
131720	132163
	132164	+ if( IsUniqueIndex(pProbe) && saved_nEq==pProbe->nKeyCol-1 ){
	132165	+ pBuilder->bldFlags \|= SQLITE_BLDF_UNIQUE;
	132166	+ }else{
	132167	+ pBuilder->bldFlags \|= SQLITE_BLDF_INDEXED;
	132168	+ }
131721	132169	pNew->wsFlags = saved_wsFlags;
131722	132170	pNew->u.btree.nEq = saved_nEq;
131723	132171	pNew->u.btree.nBtm = saved_nBtm;
131724	132172	pNew->u.btree.nTop = saved_nTop;
131725	132173	pNew->nLTerm = saved_nLTerm;
		@@ -132263,11 +132711,19 @@
132263	132711	pNew->nOut = rSize;
132264	132712	if( rc ) break;
132265	132713	}
132266	132714	}
132267	132715
	132716	+ pBuilder->bldFlags = 0;
132268	132717	rc = whereLoopAddBtreeIndex(pBuilder, pSrc, pProbe, 0);
	132718	+ if( pBuilder->bldFlags==SQLITE_BLDF_INDEXED ){
	132719	+ /* If a non-unique index is used, or if a prefix of the key for
	132720	+ ** unique index is used (making the index functionally non-unique)
	132721	+ ** then the sqlite_stat1 data becomes important for scoring the
	132722	+ ** plan */
	132723	+ pTab->tabFlags \|= TF_StatsUsed;
	132724	+ }
132269	132725	#ifdef SQLITE_ENABLE_STAT3_OR_STAT4
132270	132726	sqlite3Stat4ProbeFree(pBuilder->pRec);
132271	132727	pBuilder->nRecValid = 0;
132272	132728	pBuilder->pRec = 0;
132273	132729	#endif
		@@ -133443,13 +133899,13 @@
133443	133899	&& (pWInfo->wctrlFlags & WHERE_DISTINCTBY)==0
133444	133900	&& pWInfo->eDistinct==WHERE_DISTINCT_NOOP
133445	133901	&& nRowEst
133446	133902	){
133447	133903	Bitmask notUsed;
133448		- int rc = wherePathSatisfiesOrderBy(pWInfo, pWInfo->pDistinctSet, pFrom,
	133904	+ int rc = wherePathSatisfiesOrderBy(pWInfo, pWInfo->pResultSet, pFrom,
133449	133905	WHERE_DISTINCTBY, nLoop-1, pFrom->aLoop[nLoop-1], &notUsed);
133450		- if( rc==pWInfo->pDistinctSet->nExpr ){
	133906	+ if( rc==pWInfo->pResultSet->nExpr ){
133451	133907	pWInfo->eDistinct = WHERE_DISTINCT_ORDERED;
133452	133908	}
133453	133909	}
133454	133910	if( pWInfo->pOrderBy ){
133455	133911	if( pWInfo->wctrlFlags & WHERE_DISTINCTBY ){
		@@ -133682,11 +134138,11 @@
133682	134138	SQLITE_PRIVATE WhereInfo *sqlite3WhereBegin(
133683	134139	Parse pParse, / The parser context */
133684	134140	SrcList pTabList, / FROM clause: A list of all tables to be scanned */
133685	134141	Expr pWhere, / The WHERE clause */
133686	134142	ExprList pOrderBy, / An ORDER BY (or GROUP BY) clause, or NULL */
133687		- ExprList pDistinctSet, / Try not to output two rows that duplicate these */
	134143	+ ExprList pResultSet, / Query result set. Req'd for DISTINCT */
133688	134144	u16 wctrlFlags, /* The WHERE_* flags defined in sqliteInt.h */
133689	134145	int iAuxArg /* If WHERE_OR_SUBCLAUSE is set, index cursor number
133690	134146	** If WHERE_USE_LIMIT, then the limit amount */
133691	134147	){
133692	134148	int nByteWInfo; /* Num. bytes allocated for WhereInfo struct */
		@@ -133758,11 +134214,11 @@
133758	134214	goto whereBeginError;
133759	134215	}
133760	134216	pWInfo->pParse = pParse;
133761	134217	pWInfo->pTabList = pTabList;
133762	134218	pWInfo->pOrderBy = pOrderBy;
133763		- pWInfo->pDistinctSet = pDistinctSet;
	134219	+ pWInfo->pResultSet = pResultSet;
133764	134220	pWInfo->aiCurOnePass[0] = pWInfo->aiCurOnePass[1] = -1;
133765	134221	pWInfo->nLevel = nTabList;
133766	134222	pWInfo->iBreak = pWInfo->iContinue = sqlite3VdbeMakeLabel(v);
133767	134223	pWInfo->wctrlFlags = wctrlFlags;
133768	134224	pWInfo->iLimit = iAuxArg;
		@@ -133836,17 +134292,17 @@
133836	134292	/* Analyze all of the subexpressions. */
133837	134293	sqlite3WhereExprAnalyze(pTabList, &pWInfo->sWC);
133838	134294	if( db->mallocFailed ) goto whereBeginError;
133839	134295
133840	134296	if( wctrlFlags & WHERE_WANT_DISTINCT ){
133841		- if( isDistinctRedundant(pParse, pTabList, &pWInfo->sWC, pDistinctSet) ){
	134297	+ if( isDistinctRedundant(pParse, pTabList, &pWInfo->sWC, pResultSet) ){
133842	134298	/* The DISTINCT marking is pointless. Ignore it. */
133843	134299	pWInfo->eDistinct = WHERE_DISTINCT_UNIQUE;
133844	134300	}else if( pOrderBy==0 ){
133845	134301	/* Try to ORDER BY the result set to make distinct processing easier */
133846	134302	pWInfo->wctrlFlags \|= WHERE_DISTINCTBY;
133847		- pWInfo->pOrderBy = pDistinctSet;
	134303	+ pWInfo->pOrderBy = pResultSet;
133848	134304	}
133849	134305	}
133850	134306
133851	134307	/* Construct the WhereLoop objects */
133852	134308	#if defined(WHERETRACE_ENABLED)
		@@ -133918,14 +134374,14 @@
133918	134374	}
133919	134375	}
133920	134376	#endif
133921	134377	/* Attempt to omit tables from the join that do not effect the result */
133922	134378	if( pWInfo->nLevel>=2
133923		- && pDistinctSet!=0
	134379	+ && pResultSet!=0
133924	134380	&& OptimizationEnabled(db, SQLITE_OmitNoopJoin)
133925	134381	){
133926		- Bitmask tabUsed = sqlite3WhereExprListUsage(pMaskSet, pDistinctSet);
	134382	+ Bitmask tabUsed = sqlite3WhereExprListUsage(pMaskSet, pResultSet);
133927	134383	if( sWLB.pOrderBy ){
133928	134384	tabUsed \|= sqlite3WhereExprListUsage(pMaskSet, sWLB.pOrderBy);
133929	134385	}
133930	134386	while( pWInfo->nLevel>=2 ){
133931	134387	WhereTerm pTerm, pEnd;
		@@ -134704,166 +135160,166 @@
134704	135160	**
134705	135161	********* Begin parsing tables ********************************************/
134706	135162	#define YY_ACTTAB_COUNT (1567)
134707	135163	static const YYACTIONTYPE yy_action[] = {
134708	135164	/* 0 */ 325, 832, 351, 825, 5, 203, 203, 819, 99, 100,
134709		- /* 10 */ 90, 842, 842, 854, 857, 846, 846, 97, 97, 98,
	135165	+ /* 10 */ 90, 978, 978, 853, 856, 845, 845, 97, 97, 98,
134710	135166	/* 20 */ 98, 98, 98, 301, 96, 96, 96, 96, 95, 95,
134711		- /* 30 */ 94, 94, 94, 93, 351, 325, 977, 977, 824, 824,
134712		- /* 40 */ 826, 947, 354, 99, 100, 90, 842, 842, 854, 857,
134713		- /* 50 */ 846, 846, 97, 97, 98, 98, 98, 98, 338, 96,
	135167	+ /* 30 */ 94, 94, 94, 93, 351, 325, 976, 976, 824, 824,
	135168	+ /* 40 */ 826, 946, 354, 99, 100, 90, 978, 978, 853, 856,
	135169	+ /* 50 */ 845, 845, 97, 97, 98, 98, 98, 98, 338, 96,
134714	135170	/* 60 */ 96, 96, 96, 95, 95, 94, 94, 94, 93, 351,
134715		- /* 70 */ 95, 95, 94, 94, 94, 93, 351, 791, 977, 977,
	135171	+ /* 70 */ 95, 95, 94, 94, 94, 93, 351, 791, 976, 976,
134716	135172	/* 80 */ 325, 94, 94, 94, 93, 351, 792, 75, 99, 100,
134717		- /* 90 */ 90, 842, 842, 854, 857, 846, 846, 97, 97, 98,
	135173	+ /* 90 */ 90, 978, 978, 853, 856, 845, 845, 97, 97, 98,
134718	135174	/* 100 */ 98, 98, 98, 450, 96, 96, 96, 96, 95, 95,
134719	135175	/* 110 */ 94, 94, 94, 93, 351, 1333, 155, 155, 2, 325,
134720	135176	/* 120 */ 275, 146, 132, 52, 52, 93, 351, 99, 100, 90,
134721		- /* 130 */ 842, 842, 854, 857, 846, 846, 97, 97, 98, 98,
	135177	+ /* 130 */ 978, 978, 853, 856, 845, 845, 97, 97, 98, 98,
134722	135178	/* 140 */ 98, 98, 101, 96, 96, 96, 96, 95, 95, 94,
134723		- /* 150 */ 94, 94, 93, 351, 958, 958, 325, 268, 428, 413,
134724		- /* 160 */ 411, 61, 752, 752, 99, 100, 90, 842, 842, 854,
134725		- /* 170 */ 857, 846, 846, 97, 97, 98, 98, 98, 98, 60,
	135179	+ /* 150 */ 94, 94, 93, 351, 957, 957, 325, 268, 428, 413,
	135180	+ /* 160 */ 411, 61, 752, 752, 99, 100, 90, 978, 978, 853,
	135181	+ /* 170 */ 856, 845, 845, 97, 97, 98, 98, 98, 98, 60,
134726	135182	/* 180 */ 96, 96, 96, 96, 95, 95, 94, 94, 94, 93,
134727		- /* 190 */ 351, 325, 270, 329, 273, 277, 959, 960, 250, 99,
134728		- /* 200 */ 100, 90, 842, 842, 854, 857, 846, 846, 97, 97,
	135183	+ /* 190 */ 351, 325, 270, 329, 273, 277, 958, 959, 250, 99,
	135184	+ /* 200 */ 100, 90, 978, 978, 853, 856, 845, 845, 97, 97,
134729	135185	/* 210 */ 98, 98, 98, 98, 301, 96, 96, 96, 96, 95,
134730		- /* 220 */ 95, 94, 94, 94, 93, 351, 325, 938, 1326, 698,
134731		- /* 230 */ 706, 1326, 242, 412, 99, 100, 90, 842, 842, 854,
134732		- /* 240 */ 857, 846, 846, 97, 97, 98, 98, 98, 98, 347,
	135186	+ /* 220 */ 95, 94, 94, 94, 93, 351, 325, 937, 1326, 698,
	135187	+ /* 230 */ 706, 1326, 242, 412, 99, 100, 90, 978, 978, 853,
	135188	+ /* 240 */ 856, 845, 845, 97, 97, 98, 98, 98, 98, 347,
134733	135189	/* 250 */ 96, 96, 96, 96, 95, 95, 94, 94, 94, 93,
134734		- /* 260 */ 351, 325, 938, 1327, 384, 699, 1327, 381, 379, 99,
134735		- /* 270 */ 100, 90, 842, 842, 854, 857, 846, 846, 97, 97,
	135190	+ /* 260 */ 351, 325, 937, 1327, 384, 699, 1327, 381, 379, 99,
	135191	+ /* 270 */ 100, 90, 978, 978, 853, 856, 845, 845, 97, 97,
134736	135192	/* 280 */ 98, 98, 98, 98, 701, 96, 96, 96, 96, 95,
134737	135193	/* 290 */ 95, 94, 94, 94, 93, 351, 325, 92, 89, 178,
134738		- /* 300 */ 833, 936, 373, 700, 99, 100, 90, 842, 842, 854,
134739		- /* 310 */ 857, 846, 846, 97, 97, 98, 98, 98, 98, 375,
	135194	+ /* 300 */ 833, 935, 373, 700, 99, 100, 90, 978, 978, 853,
	135195	+ /* 310 */ 856, 845, 845, 97, 97, 98, 98, 98, 98, 375,
134740	135196	/* 320 */ 96, 96, 96, 96, 95, 95, 94, 94, 94, 93,
134741		- /* 330 */ 351, 325, 1276, 947, 354, 818, 936, 739, 739, 99,
134742		- /* 340 */ 100, 90, 842, 842, 854, 857, 846, 846, 97, 97,
	135197	+ /* 330 */ 351, 325, 1275, 946, 354, 818, 935, 739, 739, 99,
	135198	+ /* 340 */ 100, 90, 978, 978, 853, 856, 845, 845, 97, 97,
134743	135199	/* 350 */ 98, 98, 98, 98, 230, 96, 96, 96, 96, 95,
134744		- /* 360 */ 95, 94, 94, 94, 93, 351, 325, 969, 227, 92,
134745		- /* 370 */ 89, 178, 373, 300, 99, 100, 90, 842, 842, 854,
134746		- /* 380 */ 857, 846, 846, 97, 97, 98, 98, 98, 98, 921,
	135200	+ /* 360 */ 95, 94, 94, 94, 93, 351, 325, 968, 227, 92,
	135201	+ /* 370 */ 89, 178, 373, 300, 99, 100, 90, 978, 978, 853,
	135202	+ /* 380 */ 856, 845, 845, 97, 97, 98, 98, 98, 98, 920,
134747	135203	/* 390 */ 96, 96, 96, 96, 95, 95, 94, 94, 94, 93,
134748	135204	/* 400 */ 351, 325, 449, 447, 447, 447, 147, 737, 737, 99,
134749		- /* 410 */ 100, 90, 842, 842, 854, 857, 846, 846, 97, 97,
	135205	+ /* 410 */ 100, 90, 978, 978, 853, 856, 845, 845, 97, 97,
134750	135206	/* 420 */ 98, 98, 98, 98, 296, 96, 96, 96, 96, 95,
134751		- /* 430 */ 95, 94, 94, 94, 93, 351, 325, 419, 231, 958,
134752		- /* 440 */ 958, 158, 25, 422, 99, 100, 90, 842, 842, 854,
134753		- /* 450 */ 857, 846, 846, 97, 97, 98, 98, 98, 98, 450,
	135207	+ /* 430 */ 95, 94, 94, 94, 93, 351, 325, 419, 231, 957,
	135208	+ /* 440 */ 957, 158, 25, 422, 99, 100, 90, 978, 978, 853,
	135209	+ /* 450 */ 856, 845, 845, 97, 97, 98, 98, 98, 98, 450,
134754	135210	/* 460 */ 96, 96, 96, 96, 95, 95, 94, 94, 94, 93,
134755		- /* 470 */ 351, 443, 224, 224, 420, 958, 958, 962, 325, 52,
134756		- /* 480 */ 52, 959, 960, 176, 415, 78, 99, 100, 90, 842,
134757		- /* 490 */ 842, 854, 857, 846, 846, 97, 97, 98, 98, 98,
	135211	+ /* 470 */ 351, 443, 224, 224, 420, 957, 957, 961, 325, 52,
	135212	+ /* 480 */ 52, 958, 959, 176, 415, 78, 99, 100, 90, 978,
	135213	+ /* 490 */ 978, 853, 856, 845, 845, 97, 97, 98, 98, 98,
134758	135214	/* 500 */ 98, 379, 96, 96, 96, 96, 95, 95, 94, 94,
134759		- /* 510 */ 94, 93, 351, 325, 428, 418, 298, 959, 960, 962,
134760		- /* 520 */ 81, 99, 88, 90, 842, 842, 854, 857, 846, 846,
	135215	+ /* 510 */ 94, 93, 351, 325, 428, 418, 298, 958, 959, 961,
	135216	+ /* 520 */ 81, 99, 88, 90, 978, 978, 853, 856, 845, 845,
134761	135217	/* 530 */ 97, 97, 98, 98, 98, 98, 717, 96, 96, 96,
134762		- /* 540 */ 96, 95, 95, 94, 94, 94, 93, 351, 325, 843,
134763		- /* 550 */ 843, 855, 858, 996, 318, 343, 379, 100, 90, 842,
134764		- /* 560 */ 842, 854, 857, 846, 846, 97, 97, 98, 98, 98,
	135218	+ /* 540 */ 96, 95, 95, 94, 94, 94, 93, 351, 325, 842,
	135219	+ /* 550 */ 842, 854, 857, 996, 318, 343, 379, 100, 90, 978,
	135220	+ /* 560 */ 978, 853, 856, 845, 845, 97, 97, 98, 98, 98,
134765	135221	/* 570 */ 98, 450, 96, 96, 96, 96, 95, 95, 94, 94,
134766	135222	/* 580 */ 94, 93, 351, 325, 350, 350, 350, 260, 377, 340,
134767		- /* 590 */ 929, 52, 52, 90, 842, 842, 854, 857, 846, 846,
	135223	+ /* 590 */ 928, 52, 52, 90, 978, 978, 853, 856, 845, 845,
134768	135224	/* 600 */ 97, 97, 98, 98, 98, 98, 361, 96, 96, 96,
134769	135225	/* 610 */ 96, 95, 95, 94, 94, 94, 93, 351, 86, 445,
134770		- /* 620 */ 847, 3, 1203, 361, 360, 378, 344, 813, 958, 958,
134771		- /* 630 */ 1300, 86, 445, 729, 3, 212, 169, 287, 405, 282,
	135226	+ /* 620 */ 846, 3, 1202, 361, 360, 378, 344, 813, 957, 957,
	135227	+ /* 630 */ 1299, 86, 445, 729, 3, 212, 169, 287, 405, 282,
134772	135228	/* 640 */ 404, 199, 232, 450, 300, 760, 83, 84, 280, 245,
134773	135229	/* 650 */ 262, 365, 251, 85, 352, 352, 92, 89, 178, 83,
134774	135230	/* 660 */ 84, 242, 412, 52, 52, 448, 85, 352, 352, 246,
134775		- /* 670 */ 959, 960, 194, 455, 670, 402, 399, 398, 448, 243,
	135231	+ /* 670 */ 958, 959, 194, 455, 670, 402, 399, 398, 448, 243,
134776	135232	/* 680 */ 221, 114, 434, 776, 361, 450, 397, 268, 747, 224,
134777	135233	/* 690 */ 224, 132, 132, 198, 832, 434, 452, 451, 428, 427,
134778	135234	/* 700 */ 819, 415, 734, 713, 132, 52, 52, 832, 268, 452,
134779		- /* 710 */ 451, 734, 194, 819, 363, 402, 399, 398, 450, 1271,
134780		- /* 720 */ 1271, 23, 958, 958, 86, 445, 397, 3, 228, 429,
134781		- /* 730 */ 895, 824, 824, 826, 827, 19, 203, 720, 52, 52,
	135235	+ /* 710 */ 451, 734, 194, 819, 363, 402, 399, 398, 450, 1270,
	135236	+ /* 720 */ 1270, 23, 957, 957, 86, 445, 397, 3, 228, 429,
	135237	+ /* 730 */ 894, 824, 824, 826, 827, 19, 203, 720, 52, 52,
134782	135238	/* 740 */ 428, 408, 439, 249, 824, 824, 826, 827, 19, 229,
134783	135239	/* 750 */ 403, 153, 83, 84, 761, 177, 241, 450, 721, 85,
134784		- /* 760 */ 352, 352, 120, 157, 959, 960, 58, 977, 409, 355,
	135240	+ /* 760 */ 352, 352, 120, 157, 958, 959, 58, 976, 409, 355,
134785	135241	/* 770 */ 330, 448, 268, 428, 430, 320, 790, 32, 32, 86,
134786	135242	/* 780 */ 445, 776, 3, 341, 98, 98, 98, 98, 434, 96,
134787	135243	/* 790 */ 96, 96, 96, 95, 95, 94, 94, 94, 93, 351,
134788		- /* 800 */ 832, 120, 452, 451, 813, 887, 819, 83, 84, 977,
134789		- /* 810 */ 813, 132, 410, 920, 85, 352, 352, 132, 407, 789,
134790		- /* 820 */ 958, 958, 92, 89, 178, 917, 448, 262, 370, 261,
134791		- /* 830 */ 82, 914, 80, 262, 370, 261, 776, 824, 824, 826,
134792		- /* 840 */ 827, 19, 934, 434, 96, 96, 96, 96, 95, 95,
134793		- /* 850 */ 94, 94, 94, 93, 351, 832, 74, 452, 451, 958,
134794		- /* 860 */ 958, 819, 959, 960, 120, 92, 89, 178, 945, 2,
134795		- /* 870 */ 918, 965, 268, 1, 976, 76, 445, 762, 3, 708,
134796		- /* 880 */ 901, 901, 387, 958, 958, 757, 919, 371, 740, 778,
	135244	+ /* 800 */ 832, 120, 452, 451, 813, 886, 819, 83, 84, 976,
	135245	+ /* 810 */ 813, 132, 410, 919, 85, 352, 352, 132, 407, 789,
	135246	+ /* 820 */ 957, 957, 92, 89, 178, 916, 448, 262, 370, 261,
	135247	+ /* 830 */ 82, 913, 80, 262, 370, 261, 776, 824, 824, 826,
	135248	+ /* 840 */ 827, 19, 933, 434, 96, 96, 96, 96, 95, 95,
	135249	+ /* 850 */ 94, 94, 94, 93, 351, 832, 74, 452, 451, 957,
	135250	+ /* 860 */ 957, 819, 958, 959, 120, 92, 89, 178, 944, 2,
	135251	+ /* 870 */ 917, 964, 268, 1, 975, 76, 445, 762, 3, 708,
	135252	+ /* 880 */ 900, 900, 387, 957, 957, 757, 918, 371, 740, 778,
134797	135253	/* 890 */ 756, 257, 824, 824, 826, 827, 19, 417, 741, 450,
134798		- /* 900 */ 24, 959, 960, 83, 84, 369, 958, 958, 177, 226,
134799		- /* 910 */ 85, 352, 352, 885, 315, 314, 313, 215, 311, 10,
134800		- /* 920 */ 10, 683, 448, 349, 348, 959, 960, 909, 777, 157,
134801		- /* 930 */ 120, 958, 958, 337, 776, 416, 711, 310, 450, 434,
134802		- /* 940 */ 450, 321, 450, 791, 103, 200, 175, 450, 959, 960,
134803		- /* 950 */ 908, 832, 792, 452, 451, 9, 9, 819, 10, 10,
	135254	+ /* 900 */ 24, 958, 959, 83, 84, 369, 957, 957, 177, 226,
	135255	+ /* 910 */ 85, 352, 352, 884, 315, 314, 313, 215, 311, 10,
	135256	+ /* 920 */ 10, 683, 448, 349, 348, 958, 959, 908, 777, 157,
	135257	+ /* 930 */ 120, 957, 957, 337, 776, 416, 711, 310, 450, 434,
	135258	+ /* 940 */ 450, 321, 450, 791, 103, 200, 175, 450, 958, 959,
	135259	+ /* 950 */ 907, 832, 792, 452, 451, 9, 9, 819, 10, 10,
134804	135260	/* 960 */ 52, 52, 51, 51, 180, 716, 248, 10, 10, 171,
134805		- /* 970 */ 170, 167, 339, 959, 960, 247, 984, 702, 702, 450,
134806		- /* 980 */ 715, 233, 686, 982, 889, 983, 182, 914, 824, 824,
	135261	+ /* 970 */ 170, 167, 339, 958, 959, 247, 984, 702, 702, 450,
	135262	+ /* 980 */ 715, 233, 686, 982, 888, 983, 182, 913, 824, 824,
134807	135263	/* 990 */ 826, 827, 19, 183, 256, 423, 132, 181, 394, 10,
134808		- /* 1000 */ 10, 889, 891, 749, 958, 958, 917, 268, 985, 198,
	135264	+ /* 1000 */ 10, 888, 890, 749, 957, 957, 916, 268, 985, 198,
134809	135265	/* 1010 */ 985, 349, 348, 425, 415, 299, 817, 832, 326, 825,
134810	135266	/* 1020 */ 120, 332, 133, 819, 268, 98, 98, 98, 98, 91,
134811	135267	/* 1030 */ 96, 96, 96, 96, 95, 95, 94, 94, 94, 93,
134812		- /* 1040 */ 351, 157, 810, 371, 382, 359, 959, 960, 358, 268,
134813		- /* 1050 */ 450, 918, 368, 324, 824, 824, 826, 450, 709, 450,
134814		- /* 1060 */ 264, 380, 889, 450, 877, 746, 253, 919, 255, 433,
	135268	+ /* 1040 */ 351, 157, 810, 371, 382, 359, 958, 959, 358, 268,
	135269	+ /* 1050 */ 450, 917, 368, 324, 824, 824, 826, 450, 709, 450,
	135270	+ /* 1060 */ 264, 380, 888, 450, 876, 746, 253, 918, 255, 433,
134815	135271	/* 1070 */ 36, 36, 234, 450, 234, 120, 269, 37, 37, 12,
134816	135272	/* 1080 */ 12, 334, 272, 27, 27, 450, 330, 118, 450, 162,
134817	135273	/* 1090 */ 742, 280, 450, 38, 38, 450, 985, 356, 985, 450,
134818		- /* 1100 */ 709, 1210, 450, 132, 450, 39, 39, 450, 40, 40,
	135274	+ /* 1100 */ 709, 1209, 450, 132, 450, 39, 39, 450, 40, 40,
134819	135275	/* 1110 */ 450, 362, 41, 41, 450, 42, 42, 450, 254, 28,
134820	135276	/* 1120 */ 28, 450, 29, 29, 31, 31, 450, 43, 43, 450,
134821	135277	/* 1130 */ 44, 44, 450, 714, 45, 45, 450, 11, 11, 767,
134822	135278	/* 1140 */ 450, 46, 46, 450, 268, 450, 105, 105, 450, 47,
134823	135279	/* 1150 */ 47, 450, 48, 48, 450, 237, 33, 33, 450, 172,
134824	135280	/* 1160 */ 49, 49, 450, 50, 50, 34, 34, 274, 122, 122,
134825		- /* 1170 */ 450, 123, 123, 450, 124, 124, 450, 898, 56, 56,
134826		- /* 1180 */ 450, 897, 35, 35, 450, 267, 450, 817, 450, 817,
	135281	+ /* 1170 */ 450, 123, 123, 450, 124, 124, 450, 897, 56, 56,
	135282	+ /* 1180 */ 450, 896, 35, 35, 450, 267, 450, 817, 450, 817,
134827	135283	/* 1190 */ 106, 106, 450, 53, 53, 385, 107, 107, 450, 817,
134828	135284	/* 1200 */ 108, 108, 817, 450, 104, 104, 121, 121, 119, 119,
134829	135285	/* 1210 */ 450, 117, 112, 112, 450, 276, 450, 225, 111, 111,
134830		- /* 1220 */ 450, 730, 450, 109, 109, 450, 673, 674, 675, 912,
	135286	+ /* 1220 */ 450, 730, 450, 109, 109, 450, 673, 674, 675, 911,
134831	135287	/* 1230 */ 110, 110, 317, 998, 55, 55, 57, 57, 692, 331,
134832		- /* 1240 */ 54, 54, 26, 26, 696, 30, 30, 317, 937, 197,
	135288	+ /* 1240 */ 54, 54, 26, 26, 696, 30, 30, 317, 936, 197,
134833	135289	/* 1250 */ 196, 195, 335, 281, 336, 446, 331, 745, 689, 436,
134834		- /* 1260 */ 440, 444, 120, 72, 386, 223, 175, 345, 757, 933,
	135290	+ /* 1260 */ 440, 444, 120, 72, 386, 223, 175, 345, 757, 932,
134835	135291	/* 1270 */ 20, 286, 319, 756, 815, 372, 374, 202, 202, 202,
134836		- /* 1280 */ 263, 395, 285, 74, 208, 21, 696, 719, 718, 884,
	135292	+ /* 1280 */ 263, 395, 285, 74, 208, 21, 696, 719, 718, 883,
134837	135293	/* 1290 */ 120, 120, 120, 120, 120, 754, 278, 828, 77, 74,
134838		- /* 1300 */ 726, 727, 785, 783, 880, 202, 999, 208, 894, 893,
134839		- /* 1310 */ 894, 893, 694, 816, 763, 116, 774, 1290, 431, 432,
	135294	+ /* 1300 */ 726, 727, 785, 783, 879, 202, 999, 208, 893, 892,
	135295	+ /* 1310 */ 893, 892, 694, 816, 763, 116, 774, 1289, 431, 432,
134840	135296	/* 1320 */ 302, 999, 390, 303, 823, 697, 691, 680, 159, 289,
134841		- /* 1330 */ 679, 884, 681, 952, 291, 218, 293, 7, 316, 828,
134842		- /* 1340 */ 173, 805, 259, 364, 252, 911, 376, 713, 295, 435,
134843		- /* 1350 */ 308, 168, 955, 993, 135, 400, 990, 284, 882, 881,
134844		- /* 1360 */ 205, 928, 926, 59, 333, 62, 144, 156, 130, 72,
	135297	+ /* 1330 */ 679, 883, 681, 951, 291, 218, 293, 7, 316, 828,
	135298	+ /* 1340 */ 173, 805, 259, 364, 252, 910, 376, 713, 295, 435,
	135299	+ /* 1350 */ 308, 168, 954, 993, 135, 400, 990, 284, 881, 880,
	135300	+ /* 1360 */ 205, 927, 925, 59, 333, 62, 144, 156, 130, 72,
134845	135301	/* 1370 */ 802, 366, 367, 393, 137, 185, 189, 160, 139, 383,
134846		- /* 1380 */ 67, 896, 140, 141, 142, 148, 389, 812, 775, 266,
134847		- /* 1390 */ 219, 190, 154, 391, 913, 876, 271, 406, 191, 322,
	135302	+ /* 1380 */ 67, 895, 140, 141, 142, 148, 389, 812, 775, 266,
	135303	+ /* 1390 */ 219, 190, 154, 391, 912, 875, 271, 406, 191, 322,
134848	135304	/* 1400 */ 682, 733, 192, 342, 732, 724, 731, 711, 723, 421,
134849	135305	/* 1410 */ 705, 71, 323, 6, 204, 771, 288, 79, 297, 346,
134850		- /* 1420 */ 772, 704, 290, 283, 703, 770, 292, 294, 967, 239,
134851		- /* 1430 */ 769, 102, 862, 438, 426, 240, 424, 442, 73, 213,
134852		- /* 1440 */ 688, 238, 22, 453, 953, 214, 217, 216, 454, 677,
	135306	+ /* 1420 */ 772, 704, 290, 283, 703, 770, 292, 294, 966, 239,
	135307	+ /* 1430 */ 769, 102, 861, 438, 426, 240, 424, 442, 73, 213,
	135308	+ /* 1440 */ 688, 238, 22, 453, 952, 214, 217, 216, 454, 677,
134853	135309	/* 1450 */ 676, 671, 753, 125, 115, 235, 126, 669, 353, 166,
134854		- /* 1460 */ 127, 244, 179, 357, 306, 304, 305, 307, 113, 892,
134855		- /* 1470 */ 327, 890, 811, 328, 134, 128, 136, 138, 743, 258,
134856		- /* 1480 */ 907, 184, 143, 129, 910, 186, 63, 64, 145, 187,
134857		- /* 1490 */ 906, 65, 8, 66, 13, 188, 202, 899, 265, 149,
	135310	+ /* 1460 */ 127, 244, 179, 357, 306, 304, 305, 307, 113, 891,
	135311	+ /* 1470 */ 327, 889, 811, 328, 134, 128, 136, 138, 743, 258,
	135312	+ /* 1480 */ 906, 184, 143, 129, 909, 186, 63, 64, 145, 187,
	135313	+ /* 1490 */ 905, 65, 8, 66, 13, 188, 202, 898, 265, 149,
134858	135314	/* 1500 */ 987, 388, 150, 685, 161, 392, 285, 193, 279, 396,
134859	135315	/* 1510 */ 151, 401, 68, 14, 15, 722, 69, 236, 831, 131,
134860		- /* 1520 */ 830, 860, 70, 751, 16, 414, 755, 4, 174, 220,
134861		- /* 1530 */ 222, 784, 201, 152, 779, 77, 74, 17, 18, 875,
134862		- /* 1540 */ 861, 859, 916, 864, 915, 207, 206, 942, 163, 437,
134863		- /* 1550 */ 948, 943, 164, 209, 1002, 441, 863, 165, 210, 829,
134864		- /* 1560 */ 695, 87, 312, 211, 1292, 1291, 309,
	135316	+ /* 1520 */ 830, 859, 70, 751, 16, 414, 755, 4, 174, 220,
	135317	+ /* 1530 */ 222, 784, 201, 152, 779, 77, 74, 17, 18, 874,
	135318	+ /* 1540 */ 860, 858, 915, 863, 914, 207, 206, 941, 163, 437,
	135319	+ /* 1550 */ 947, 942, 164, 209, 1002, 441, 862, 165, 210, 829,
	135320	+ /* 1560 */ 695, 87, 312, 211, 1291, 1290, 309,
134865	135321	};
134866	135322	static const YYCODETYPE yy_lookahead[] = {
134867	135323	/* 0 */ 19, 95, 53, 97, 22, 24, 24, 101, 27, 28,
134868	135324	/* 10 */ 29, 30, 31, 32, 33, 34, 35, 36, 37, 38,
134869	135325	/* 20 */ 39, 40, 41, 152, 43, 44, 45, 46, 47, 48,
		@@ -135112,55 +135568,55 @@
135112	135568	/* 300 */ 1256, 1200, 1206, 1260, 1247, 1261, 1263, 1262, 1266, 1278,
135113	135569	/* 310 */ 1282, 1292, 1294, 1297, 1298, 1299, 1300, 1221, 1224, 1228,
135114	135570	/* 320 */ 1288, 1291, 1276, 1277, 1295,
135115	135571	};
135116	135572	static const YYACTIONTYPE yy_default[] = {
135117		- /* 0 */ 1281, 1271, 1271, 1271, 1203, 1203, 1203, 1203, 1271, 1096,
135118		- /* 10 */ 1125, 1125, 1255, 1332, 1332, 1332, 1332, 1332, 1332, 1202,
135119		- /* 20 */ 1332, 1332, 1332, 1332, 1271, 1100, 1131, 1332, 1332, 1332,
135120		- /* 30 */ 1332, 1204, 1205, 1332, 1332, 1332, 1254, 1256, 1141, 1140,
135121		- /* 40 */ 1139, 1138, 1237, 1112, 1136, 1129, 1133, 1204, 1198, 1199,
135122		- /* 50 */ 1197, 1201, 1205, 1332, 1132, 1167, 1182, 1166, 1332, 1332,
	135573	+ /* 0 */ 1280, 1270, 1270, 1270, 1202, 1202, 1202, 1202, 1270, 1096,
	135574	+ /* 10 */ 1125, 1125, 1254, 1332, 1332, 1332, 1332, 1332, 1332, 1201,
	135575	+ /* 20 */ 1332, 1332, 1332, 1332, 1270, 1100, 1131, 1332, 1332, 1332,
	135576	+ /* 30 */ 1332, 1203, 1204, 1332, 1332, 1332, 1253, 1255, 1141, 1140,
	135577	+ /* 40 */ 1139, 1138, 1236, 1112, 1136, 1129, 1133, 1203, 1197, 1198,
	135578	+ /* 50 */ 1196, 1200, 1204, 1332, 1132, 1167, 1181, 1166, 1332, 1332,
135123	135579	/* 60 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135124	135580	/* 70 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135125	135581	/* 80 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135126	135582	/* 90 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135127		- /* 100 */ 1332, 1332, 1332, 1332, 1176, 1181, 1188, 1180, 1177, 1169,
	135583	+ /* 100 */ 1332, 1332, 1332, 1332, 1175, 1180, 1187, 1179, 1176, 1169,
135128	135584	/* 110 */ 1168, 1170, 1171, 1332, 1019, 1067, 1332, 1332, 1332, 1172,
135129		- /* 120 */ 1332, 1173, 1185, 1184, 1183, 1262, 1289, 1288, 1332, 1332,
	135585	+ /* 120 */ 1332, 1173, 1184, 1183, 1182, 1261, 1288, 1287, 1332, 1332,
135130	135586	/* 130 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135131	135587	/* 140 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135132		- /* 150 */ 1332, 1332, 1332, 1332, 1332, 1281, 1271, 1025, 1025, 1332,
135133		- /* 160 */ 1271, 1271, 1271, 1271, 1271, 1271, 1267, 1100, 1091, 1332,
	135588	+ /* 150 */ 1332, 1332, 1332, 1332, 1332, 1280, 1270, 1025, 1025, 1332,
	135589	+ /* 160 */ 1270, 1270, 1270, 1270, 1270, 1270, 1266, 1100, 1091, 1332,
135134	135590	/* 170 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135135		- /* 180 */ 1259, 1257, 1332, 1218, 1332, 1332, 1332, 1332, 1332, 1332,
	135591	+ /* 180 */ 1258, 1256, 1332, 1217, 1332, 1332, 1332, 1332, 1332, 1332,
135136	135592	/* 190 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135137	135593	/* 200 */ 1332, 1332, 1332, 1332, 1096, 1332, 1332, 1332, 1332, 1332,
135138		- /* 210 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1283, 1332, 1232,
	135594	+ /* 210 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1282, 1332, 1231,
135139	135595	/* 220 */ 1096, 1096, 1096, 1098, 1080, 1090, 1004, 1135, 1114, 1114,
135140		- /* 230 */ 1321, 1135, 1321, 1042, 1303, 1039, 1125, 1114, 1200, 1125,
	135596	+ /* 230 */ 1321, 1135, 1321, 1042, 1302, 1039, 1125, 1114, 1199, 1125,
135141	135597	/* 240 */ 1125, 1097, 1090, 1332, 1324, 1105, 1105, 1323, 1323, 1105,
135142	135598	/* 250 */ 1146, 1070, 1135, 1076, 1076, 1076, 1076, 1105, 1016, 1135,
135143		- /* 260 */ 1146, 1070, 1070, 1135, 1105, 1016, 1236, 1318, 1105, 1105,
135144		- /* 270 */ 1016, 1211, 1105, 1016, 1105, 1016, 1211, 1068, 1068, 1068,
135145		- /* 280 */ 1057, 1211, 1068, 1042, 1068, 1057, 1068, 1068, 1118, 1113,
135146		- /* 290 */ 1118, 1113, 1118, 1113, 1118, 1113, 1105, 1206, 1105, 1332,
135147		- /* 300 */ 1211, 1215, 1215, 1211, 1130, 1119, 1128, 1126, 1135, 1022,
135148		- /* 310 */ 1060, 1286, 1286, 1282, 1282, 1282, 1282, 1329, 1329, 1267,
135149		- /* 320 */ 1298, 1298, 1044, 1044, 1298, 1332, 1332, 1332, 1332, 1332,
135150		- /* 330 */ 1332, 1293, 1332, 1220, 1332, 1332, 1332, 1332, 1332, 1332,
	135599	+ /* 260 */ 1146, 1070, 1070, 1135, 1105, 1016, 1235, 1318, 1105, 1105,
	135600	+ /* 270 */ 1016, 1210, 1105, 1016, 1105, 1016, 1210, 1068, 1068, 1068,
	135601	+ /* 280 */ 1057, 1210, 1068, 1042, 1068, 1057, 1068, 1068, 1118, 1113,
	135602	+ /* 290 */ 1118, 1113, 1118, 1113, 1118, 1113, 1105, 1205, 1105, 1332,
	135603	+ /* 300 */ 1210, 1214, 1214, 1210, 1130, 1119, 1128, 1126, 1135, 1022,
	135604	+ /* 310 */ 1060, 1285, 1285, 1281, 1281, 1281, 1281, 1329, 1329, 1266,
	135605	+ /* 320 */ 1297, 1297, 1044, 1044, 1297, 1332, 1332, 1332, 1332, 1332,
	135606	+ /* 330 */ 1332, 1292, 1332, 1219, 1332, 1332, 1332, 1332, 1332, 1332,
135151	135607	/* 340 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135152		- /* 350 */ 1332, 1332, 1152, 1332, 1000, 1264, 1332, 1332, 1263, 1332,
	135608	+ /* 350 */ 1332, 1332, 1152, 1332, 1000, 1263, 1332, 1332, 1262, 1332,
135153	135609	/* 360 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135154	135610	/* 370 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1320,
135155		- /* 380 */ 1332, 1332, 1332, 1332, 1332, 1332, 1235, 1234, 1332, 1332,
	135611	+ /* 380 */ 1332, 1332, 1332, 1332, 1332, 1332, 1234, 1233, 1332, 1332,
135156	135612	/* 390 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135157	135613	/* 400 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135158		- /* 410 */ 1332, 1082, 1332, 1332, 1332, 1307, 1332, 1332, 1332, 1332,
	135614	+ /* 410 */ 1332, 1082, 1332, 1332, 1332, 1306, 1332, 1332, 1332, 1332,
135159	135615	/* 420 */ 1332, 1332, 1332, 1127, 1332, 1120, 1332, 1332, 1311, 1332,
135160		- /* 430 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1273,
135161		- /* 440 */ 1332, 1332, 1332, 1272, 1332, 1332, 1332, 1332, 1332, 1154,
	135616	+ /* 430 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1272,
	135617	+ /* 440 */ 1332, 1332, 1332, 1271, 1332, 1332, 1332, 1332, 1332, 1154,
135162	135618	/* 450 */ 1332, 1153, 1157, 1332, 1010, 1332,
135163	135619	};
135164	135620	/******** End of lemon-generated parsing tables ***************************/
135165	135621
135166	135622	/* The next table maps tokens (terminal symbols) into fallback tokens.
		@@ -135602,147 +136058,147 @@
135602	136058	/* 169 */ "expr ::= expr EQ\|NE expr",
135603	136059	/* 170 */ "expr ::= expr BITAND\|BITOR\|LSHIFT\|RSHIFT expr",
135604	136060	/* 171 */ "expr ::= expr PLUS\|MINUS expr",
135605	136061	/* 172 */ "expr ::= expr STAR\|SLASH\|REM expr",
135606	136062	/* 173 */ "expr ::= expr CONCAT expr",
135607		- /* 174 */ "likeop ::= LIKE_KW\|MATCH",
135608		- /* 175 */ "likeop ::= NOT LIKE_KW\|MATCH",
135609		- /* 176 */ "expr ::= expr likeop expr",
135610		- /* 177 */ "expr ::= expr likeop expr ESCAPE expr",
135611		- /* 178 */ "expr ::= expr ISNULL\|NOTNULL",
135612		- /* 179 */ "expr ::= expr NOT NULL",
135613		- /* 180 */ "expr ::= expr IS expr",
135614		- /* 181 */ "expr ::= expr IS NOT expr",
135615		- /* 182 */ "expr ::= NOT expr",
135616		- /* 183 */ "expr ::= BITNOT expr",
135617		- /* 184 */ "expr ::= MINUS expr",
135618		- /* 185 */ "expr ::= PLUS expr",
135619		- /* 186 */ "between_op ::= BETWEEN",
135620		- /* 187 */ "between_op ::= NOT BETWEEN",
135621		- /* 188 */ "expr ::= expr between_op expr AND expr",
135622		- /* 189 */ "in_op ::= IN",
135623		- /* 190 */ "in_op ::= NOT IN",
135624		- /* 191 */ "expr ::= expr in_op LP exprlist RP",
135625		- /* 192 */ "expr ::= LP select RP",
135626		- /* 193 */ "expr ::= expr in_op LP select RP",
135627		- /* 194 */ "expr ::= expr in_op nm dbnm paren_exprlist",
135628		- /* 195 */ "expr ::= EXISTS LP select RP",
135629		- /* 196 */ "expr ::= CASE case_operand case_exprlist case_else END",
135630		- /* 197 */ "case_exprlist ::= case_exprlist WHEN expr THEN expr",
135631		- /* 198 */ "case_exprlist ::= WHEN expr THEN expr",
135632		- /* 199 */ "case_else ::= ELSE expr",
135633		- /* 200 */ "case_else ::=",
135634		- /* 201 */ "case_operand ::= expr",
135635		- /* 202 */ "case_operand ::=",
135636		- /* 203 */ "exprlist ::=",
135637		- /* 204 */ "nexprlist ::= nexprlist COMMA expr",
135638		- /* 205 */ "nexprlist ::= expr",
135639		- /* 206 */ "paren_exprlist ::=",
135640		- /* 207 */ "paren_exprlist ::= LP exprlist RP",
135641		- /* 208 */ "cmd ::= createkw uniqueflag INDEX ifnotexists nm dbnm ON nm LP sortlist RP where_opt",
135642		- /* 209 */ "uniqueflag ::= UNIQUE",
135643		- /* 210 */ "uniqueflag ::=",
135644		- /* 211 */ "eidlist_opt ::=",
135645		- /* 212 */ "eidlist_opt ::= LP eidlist RP",
135646		- /* 213 */ "eidlist ::= eidlist COMMA nm collate sortorder",
135647		- /* 214 */ "eidlist ::= nm collate sortorder",
135648		- /* 215 */ "collate ::=",
135649		- /* 216 */ "collate ::= COLLATE ID\|STRING",
135650		- /* 217 */ "cmd ::= DROP INDEX ifexists fullname",
135651		- /* 218 */ "cmd ::= VACUUM",
135652		- /* 219 */ "cmd ::= VACUUM nm",
135653		- /* 220 */ "cmd ::= PRAGMA nm dbnm",
135654		- /* 221 */ "cmd ::= PRAGMA nm dbnm EQ nmnum",
135655		- /* 222 */ "cmd ::= PRAGMA nm dbnm LP nmnum RP",
135656		- /* 223 */ "cmd ::= PRAGMA nm dbnm EQ minus_num",
135657		- /* 224 */ "cmd ::= PRAGMA nm dbnm LP minus_num RP",
135658		- /* 225 */ "plus_num ::= PLUS INTEGER\|FLOAT",
135659		- /* 226 */ "minus_num ::= MINUS INTEGER\|FLOAT",
135660		- /* 227 */ "cmd ::= createkw trigger_decl BEGIN trigger_cmd_list END",
135661		- /* 228 */ "trigger_decl ::= temp TRIGGER ifnotexists nm dbnm trigger_time trigger_event ON fullname foreach_clause when_clause",
135662		- /* 229 */ "trigger_time ::= BEFORE",
135663		- /* 230 */ "trigger_time ::= AFTER",
135664		- /* 231 */ "trigger_time ::= INSTEAD OF",
135665		- /* 232 */ "trigger_time ::=",
135666		- /* 233 */ "trigger_event ::= DELETE\|INSERT",
135667		- /* 234 */ "trigger_event ::= UPDATE",
135668		- /* 235 */ "trigger_event ::= UPDATE OF idlist",
135669		- /* 236 */ "when_clause ::=",
135670		- /* 237 */ "when_clause ::= WHEN expr",
135671		- /* 238 */ "trigger_cmd_list ::= trigger_cmd_list trigger_cmd SEMI",
135672		- /* 239 */ "trigger_cmd_list ::= trigger_cmd SEMI",
135673		- /* 240 */ "trnm ::= nm DOT nm",
135674		- /* 241 */ "tridxby ::= INDEXED BY nm",
135675		- /* 242 */ "tridxby ::= NOT INDEXED",
135676		- /* 243 */ "trigger_cmd ::= UPDATE orconf trnm tridxby SET setlist where_opt",
135677		- /* 244 */ "trigger_cmd ::= insert_cmd INTO trnm idlist_opt select",
135678		- /* 245 */ "trigger_cmd ::= DELETE FROM trnm tridxby where_opt",
135679		- /* 246 */ "trigger_cmd ::= select",
135680		- /* 247 */ "expr ::= RAISE LP IGNORE RP",
135681		- /* 248 */ "expr ::= RAISE LP raisetype COMMA nm RP",
135682		- /* 249 */ "raisetype ::= ROLLBACK",
135683		- /* 250 */ "raisetype ::= ABORT",
135684		- /* 251 */ "raisetype ::= FAIL",
135685		- /* 252 */ "cmd ::= DROP TRIGGER ifexists fullname",
135686		- /* 253 */ "cmd ::= ATTACH database_kw_opt expr AS expr key_opt",
135687		- /* 254 */ "cmd ::= DETACH database_kw_opt expr",
135688		- /* 255 */ "key_opt ::=",
135689		- /* 256 */ "key_opt ::= KEY expr",
135690		- /* 257 */ "cmd ::= REINDEX",
135691		- /* 258 */ "cmd ::= REINDEX nm dbnm",
135692		- /* 259 */ "cmd ::= ANALYZE",
135693		- /* 260 */ "cmd ::= ANALYZE nm dbnm",
135694		- /* 261 */ "cmd ::= ALTER TABLE fullname RENAME TO nm",
135695		- /* 262 */ "cmd ::= ALTER TABLE add_column_fullname ADD kwcolumn_opt columnname carglist",
135696		- /* 263 */ "add_column_fullname ::= fullname",
135697		- /* 264 */ "cmd ::= create_vtab",
135698		- /* 265 */ "cmd ::= create_vtab LP vtabarglist RP",
135699		- /* 266 */ "create_vtab ::= createkw VIRTUAL TABLE ifnotexists nm dbnm USING nm",
135700		- /* 267 */ "vtabarg ::=",
135701		- /* 268 */ "vtabargtoken ::= ANY",
135702		- /* 269 */ "vtabargtoken ::= lp anylist RP",
135703		- /* 270 */ "lp ::= LP",
135704		- /* 271 */ "with ::=",
135705		- /* 272 */ "with ::= WITH wqlist",
135706		- /* 273 */ "with ::= WITH RECURSIVE wqlist",
135707		- /* 274 */ "wqlist ::= nm eidlist_opt AS LP select RP",
135708		- /* 275 */ "wqlist ::= wqlist COMMA nm eidlist_opt AS LP select RP",
135709		- /* 276 */ "input ::= cmdlist",
135710		- /* 277 */ "cmdlist ::= cmdlist ecmd",
135711		- /* 278 */ "cmdlist ::= ecmd",
135712		- /* 279 */ "ecmd ::= SEMI",
135713		- /* 280 */ "ecmd ::= explain cmdx SEMI",
135714		- /* 281 */ "explain ::=",
135715		- /* 282 */ "trans_opt ::=",
135716		- /* 283 */ "trans_opt ::= TRANSACTION",
135717		- /* 284 */ "trans_opt ::= TRANSACTION nm",
135718		- /* 285 */ "savepoint_opt ::= SAVEPOINT",
135719		- /* 286 */ "savepoint_opt ::=",
135720		- /* 287 */ "cmd ::= create_table create_table_args",
135721		- /* 288 */ "columnlist ::= columnlist COMMA columnname carglist",
135722		- /* 289 */ "columnlist ::= columnname carglist",
135723		- /* 290 */ "nm ::= ID\|INDEXED",
135724		- /* 291 */ "nm ::= STRING",
135725		- /* 292 */ "nm ::= JOIN_KW",
135726		- /* 293 */ "typetoken ::= typename",
135727		- /* 294 */ "typename ::= ID\|STRING",
135728		- /* 295 */ "signed ::= plus_num",
135729		- /* 296 */ "signed ::= minus_num",
135730		- /* 297 */ "carglist ::= carglist ccons",
135731		- /* 298 */ "carglist ::=",
135732		- /* 299 */ "ccons ::= NULL onconf",
135733		- /* 300 */ "conslist_opt ::= COMMA conslist",
135734		- /* 301 */ "conslist ::= conslist tconscomma tcons",
135735		- /* 302 */ "conslist ::= tcons",
135736		- /* 303 */ "tconscomma ::=",
135737		- /* 304 */ "defer_subclause_opt ::= defer_subclause",
135738		- /* 305 */ "resolvetype ::= raisetype",
135739		- /* 306 */ "selectnowith ::= oneselect",
135740		- /* 307 */ "oneselect ::= values",
135741		- /* 308 */ "sclp ::= selcollist COMMA",
135742		- /* 309 */ "as ::= ID\|STRING",
135743		- /* 310 */ "expr ::= term",
	136063	+ /* 174 */ "likeop ::= NOT LIKE_KW\|MATCH",
	136064	+ /* 175 */ "expr ::= expr likeop expr",
	136065	+ /* 176 */ "expr ::= expr likeop expr ESCAPE expr",
	136066	+ /* 177 */ "expr ::= expr ISNULL\|NOTNULL",
	136067	+ /* 178 */ "expr ::= expr NOT NULL",
	136068	+ /* 179 */ "expr ::= expr IS expr",
	136069	+ /* 180 */ "expr ::= expr IS NOT expr",
	136070	+ /* 181 */ "expr ::= NOT expr",
	136071	+ /* 182 */ "expr ::= BITNOT expr",
	136072	+ /* 183 */ "expr ::= MINUS expr",
	136073	+ /* 184 */ "expr ::= PLUS expr",
	136074	+ /* 185 */ "between_op ::= BETWEEN",
	136075	+ /* 186 */ "between_op ::= NOT BETWEEN",
	136076	+ /* 187 */ "expr ::= expr between_op expr AND expr",
	136077	+ /* 188 */ "in_op ::= IN",
	136078	+ /* 189 */ "in_op ::= NOT IN",
	136079	+ /* 190 */ "expr ::= expr in_op LP exprlist RP",
	136080	+ /* 191 */ "expr ::= LP select RP",
	136081	+ /* 192 */ "expr ::= expr in_op LP select RP",
	136082	+ /* 193 */ "expr ::= expr in_op nm dbnm paren_exprlist",
	136083	+ /* 194 */ "expr ::= EXISTS LP select RP",
	136084	+ /* 195 */ "expr ::= CASE case_operand case_exprlist case_else END",
	136085	+ /* 196 */ "case_exprlist ::= case_exprlist WHEN expr THEN expr",
	136086	+ /* 197 */ "case_exprlist ::= WHEN expr THEN expr",
	136087	+ /* 198 */ "case_else ::= ELSE expr",
	136088	+ /* 199 */ "case_else ::=",
	136089	+ /* 200 */ "case_operand ::= expr",
	136090	+ /* 201 */ "case_operand ::=",
	136091	+ /* 202 */ "exprlist ::=",
	136092	+ /* 203 */ "nexprlist ::= nexprlist COMMA expr",
	136093	+ /* 204 */ "nexprlist ::= expr",
	136094	+ /* 205 */ "paren_exprlist ::=",
	136095	+ /* 206 */ "paren_exprlist ::= LP exprlist RP",
	136096	+ /* 207 */ "cmd ::= createkw uniqueflag INDEX ifnotexists nm dbnm ON nm LP sortlist RP where_opt",
	136097	+ /* 208 */ "uniqueflag ::= UNIQUE",
	136098	+ /* 209 */ "uniqueflag ::=",
	136099	+ /* 210 */ "eidlist_opt ::=",
	136100	+ /* 211 */ "eidlist_opt ::= LP eidlist RP",
	136101	+ /* 212 */ "eidlist ::= eidlist COMMA nm collate sortorder",
	136102	+ /* 213 */ "eidlist ::= nm collate sortorder",
	136103	+ /* 214 */ "collate ::=",
	136104	+ /* 215 */ "collate ::= COLLATE ID\|STRING",
	136105	+ /* 216 */ "cmd ::= DROP INDEX ifexists fullname",
	136106	+ /* 217 */ "cmd ::= VACUUM",
	136107	+ /* 218 */ "cmd ::= VACUUM nm",
	136108	+ /* 219 */ "cmd ::= PRAGMA nm dbnm",
	136109	+ /* 220 */ "cmd ::= PRAGMA nm dbnm EQ nmnum",
	136110	+ /* 221 */ "cmd ::= PRAGMA nm dbnm LP nmnum RP",
	136111	+ /* 222 */ "cmd ::= PRAGMA nm dbnm EQ minus_num",
	136112	+ /* 223 */ "cmd ::= PRAGMA nm dbnm LP minus_num RP",
	136113	+ /* 224 */ "plus_num ::= PLUS INTEGER\|FLOAT",
	136114	+ /* 225 */ "minus_num ::= MINUS INTEGER\|FLOAT",
	136115	+ /* 226 */ "cmd ::= createkw trigger_decl BEGIN trigger_cmd_list END",
	136116	+ /* 227 */ "trigger_decl ::= temp TRIGGER ifnotexists nm dbnm trigger_time trigger_event ON fullname foreach_clause when_clause",
	136117	+ /* 228 */ "trigger_time ::= BEFORE",
	136118	+ /* 229 */ "trigger_time ::= AFTER",
	136119	+ /* 230 */ "trigger_time ::= INSTEAD OF",
	136120	+ /* 231 */ "trigger_time ::=",
	136121	+ /* 232 */ "trigger_event ::= DELETE\|INSERT",
	136122	+ /* 233 */ "trigger_event ::= UPDATE",
	136123	+ /* 234 */ "trigger_event ::= UPDATE OF idlist",
	136124	+ /* 235 */ "when_clause ::=",
	136125	+ /* 236 */ "when_clause ::= WHEN expr",
	136126	+ /* 237 */ "trigger_cmd_list ::= trigger_cmd_list trigger_cmd SEMI",
	136127	+ /* 238 */ "trigger_cmd_list ::= trigger_cmd SEMI",
	136128	+ /* 239 */ "trnm ::= nm DOT nm",
	136129	+ /* 240 */ "tridxby ::= INDEXED BY nm",
	136130	+ /* 241 */ "tridxby ::= NOT INDEXED",
	136131	+ /* 242 */ "trigger_cmd ::= UPDATE orconf trnm tridxby SET setlist where_opt",
	136132	+ /* 243 */ "trigger_cmd ::= insert_cmd INTO trnm idlist_opt select",
	136133	+ /* 244 */ "trigger_cmd ::= DELETE FROM trnm tridxby where_opt",
	136134	+ /* 245 */ "trigger_cmd ::= select",
	136135	+ /* 246 */ "expr ::= RAISE LP IGNORE RP",
	136136	+ /* 247 */ "expr ::= RAISE LP raisetype COMMA nm RP",
	136137	+ /* 248 */ "raisetype ::= ROLLBACK",
	136138	+ /* 249 */ "raisetype ::= ABORT",
	136139	+ /* 250 */ "raisetype ::= FAIL",
	136140	+ /* 251 */ "cmd ::= DROP TRIGGER ifexists fullname",
	136141	+ /* 252 */ "cmd ::= ATTACH database_kw_opt expr AS expr key_opt",
	136142	+ /* 253 */ "cmd ::= DETACH database_kw_opt expr",
	136143	+ /* 254 */ "key_opt ::=",
	136144	+ /* 255 */ "key_opt ::= KEY expr",
	136145	+ /* 256 */ "cmd ::= REINDEX",
	136146	+ /* 257 */ "cmd ::= REINDEX nm dbnm",
	136147	+ /* 258 */ "cmd ::= ANALYZE",
	136148	+ /* 259 */ "cmd ::= ANALYZE nm dbnm",
	136149	+ /* 260 */ "cmd ::= ALTER TABLE fullname RENAME TO nm",
	136150	+ /* 261 */ "cmd ::= ALTER TABLE add_column_fullname ADD kwcolumn_opt columnname carglist",
	136151	+ /* 262 */ "add_column_fullname ::= fullname",
	136152	+ /* 263 */ "cmd ::= create_vtab",
	136153	+ /* 264 */ "cmd ::= create_vtab LP vtabarglist RP",
	136154	+ /* 265 */ "create_vtab ::= createkw VIRTUAL TABLE ifnotexists nm dbnm USING nm",
	136155	+ /* 266 */ "vtabarg ::=",
	136156	+ /* 267 */ "vtabargtoken ::= ANY",
	136157	+ /* 268 */ "vtabargtoken ::= lp anylist RP",
	136158	+ /* 269 */ "lp ::= LP",
	136159	+ /* 270 */ "with ::=",
	136160	+ /* 271 */ "with ::= WITH wqlist",
	136161	+ /* 272 */ "with ::= WITH RECURSIVE wqlist",
	136162	+ /* 273 */ "wqlist ::= nm eidlist_opt AS LP select RP",
	136163	+ /* 274 */ "wqlist ::= wqlist COMMA nm eidlist_opt AS LP select RP",
	136164	+ /* 275 */ "input ::= cmdlist",
	136165	+ /* 276 */ "cmdlist ::= cmdlist ecmd",
	136166	+ /* 277 */ "cmdlist ::= ecmd",
	136167	+ /* 278 */ "ecmd ::= SEMI",
	136168	+ /* 279 */ "ecmd ::= explain cmdx SEMI",
	136169	+ /* 280 */ "explain ::=",
	136170	+ /* 281 */ "trans_opt ::=",
	136171	+ /* 282 */ "trans_opt ::= TRANSACTION",
	136172	+ /* 283 */ "trans_opt ::= TRANSACTION nm",
	136173	+ /* 284 */ "savepoint_opt ::= SAVEPOINT",
	136174	+ /* 285 */ "savepoint_opt ::=",
	136175	+ /* 286 */ "cmd ::= create_table create_table_args",
	136176	+ /* 287 */ "columnlist ::= columnlist COMMA columnname carglist",
	136177	+ /* 288 */ "columnlist ::= columnname carglist",
	136178	+ /* 289 */ "nm ::= ID\|INDEXED",
	136179	+ /* 290 */ "nm ::= STRING",
	136180	+ /* 291 */ "nm ::= JOIN_KW",
	136181	+ /* 292 */ "typetoken ::= typename",
	136182	+ /* 293 */ "typename ::= ID\|STRING",
	136183	+ /* 294 */ "signed ::= plus_num",
	136184	+ /* 295 */ "signed ::= minus_num",
	136185	+ /* 296 */ "carglist ::= carglist ccons",
	136186	+ /* 297 */ "carglist ::=",
	136187	+ /* 298 */ "ccons ::= NULL onconf",
	136188	+ /* 299 */ "conslist_opt ::= COMMA conslist",
	136189	+ /* 300 */ "conslist ::= conslist tconscomma tcons",
	136190	+ /* 301 */ "conslist ::= tcons",
	136191	+ /* 302 */ "tconscomma ::=",
	136192	+ /* 303 */ "defer_subclause_opt ::= defer_subclause",
	136193	+ /* 304 */ "resolvetype ::= raisetype",
	136194	+ /* 305 */ "selectnowith ::= oneselect",
	136195	+ /* 306 */ "oneselect ::= values",
	136196	+ /* 307 */ "sclp ::= selcollist COMMA",
	136197	+ /* 308 */ "as ::= ID\|STRING",
	136198	+ /* 309 */ "expr ::= term",
	136199	+ /* 310 */ "likeop ::= LIKE_KW\|MATCH",
135744	136200	/* 311 */ "exprlist ::= nexprlist",
135745	136201	/* 312 */ "nmnum ::= plus_num",
135746	136202	/* 313 */ "nmnum ::= nm",
135747	136203	/* 314 */ "nmnum ::= ON",
135748	136204	/* 315 */ "nmnum ::= DELETE",
		@@ -136375,11 +136831,10 @@
136375	136831	{ 173, 3 },
136376	136832	{ 173, 3 },
136377	136833	{ 173, 3 },
136378	136834	{ 173, 3 },
136379	136835	{ 173, 3 },
136380		- { 221, 1 },
136381	136836	{ 221, 2 },
136382	136837	{ 173, 3 },
136383	136838	{ 173, 5 },
136384	136839	{ 173, 2 },
136385	136840	{ 173, 3 },
		@@ -136512,10 +136967,11 @@
136512	136967	{ 194, 1 },
136513	136968	{ 195, 1 },
136514	136969	{ 209, 2 },
136515	136970	{ 210, 1 },
136516	136971	{ 173, 1 },
	136972	+ { 221, 1 },
136517	136973	{ 208, 1 },
136518	136974	{ 230, 1 },
136519	136975	{ 230, 1 },
136520	136976	{ 230, 1 },
136521	136977	{ 230, 1 },
		@@ -136654,11 +137110,11 @@
136654	137110	case 42: /* autoinc ::= */ yytestcase(yyruleno==42);
136655	137111	case 57: /* init_deferred_pred_opt ::= */ yytestcase(yyruleno==57);
136656	137112	case 67: /* defer_subclause_opt ::= */ yytestcase(yyruleno==67);
136657	137113	case 76: /* ifexists ::= */ yytestcase(yyruleno==76);
136658	137114	case 90: /* distinct ::= */ yytestcase(yyruleno==90);
136659		- case 215: /* collate ::= */ yytestcase(yyruleno==215);
	137115	+ case 214: /* collate ::= */ yytestcase(yyruleno==214);
136660	137116	{yymsp[1].minor.yy194 = 0;}
136661	137117	break;
136662	137118	case 17: /* ifnotexists ::= IF NOT EXISTS */
136663	137119	{yymsp[-2].minor.yy194 = 1;}
136664	137120	break;
		@@ -136798,13 +137254,13 @@
136798	137254	case 144: /* insert_cmd ::= INSERT orconf */ yytestcase(yyruleno==144);
136799	137255	{yymsp[-1].minor.yy194 = yymsp[0].minor.yy194;}
136800	137256	break;
136801	137257	case 58: /* init_deferred_pred_opt ::= INITIALLY DEFERRED */
136802	137258	case 75: /* ifexists ::= IF EXISTS */ yytestcase(yyruleno==75);
136803		- case 187: /* between_op ::= NOT BETWEEN */ yytestcase(yyruleno==187);
136804		- case 190: /* in_op ::= NOT IN */ yytestcase(yyruleno==190);
136805		- case 216: /* collate ::= COLLATE ID\|STRING */ yytestcase(yyruleno==216);
	137259	+ case 186: /* between_op ::= NOT BETWEEN */ yytestcase(yyruleno==186);
	137260	+ case 189: /* in_op ::= NOT IN */ yytestcase(yyruleno==189);
	137261	+ case 215: /* collate ::= COLLATE ID\|STRING */ yytestcase(yyruleno==215);
136806	137262	{yymsp[-1].minor.yy194 = 1;}
136807	137263	break;
136808	137264	case 59: /* init_deferred_pred_opt ::= INITIALLY IMMEDIATE */
136809	137265	{yymsp[-1].minor.yy194 = 0;}
136810	137266	break;
		@@ -136964,13 +137420,13 @@
136964	137420	{yymsp[0].minor.yy194 = SF_All;}
136965	137421	break;
136966	137422	case 91: /* sclp ::= */
136967	137423	case 119: /* orderby_opt ::= */ yytestcase(yyruleno==119);
136968	137424	case 126: /* groupby_opt ::= */ yytestcase(yyruleno==126);
136969		- case 203: /* exprlist ::= */ yytestcase(yyruleno==203);
136970		- case 206: /* paren_exprlist ::= */ yytestcase(yyruleno==206);
136971		- case 211: /* eidlist_opt ::= */ yytestcase(yyruleno==211);
	137425	+ case 202: /* exprlist ::= */ yytestcase(yyruleno==202);
	137426	+ case 205: /* paren_exprlist ::= */ yytestcase(yyruleno==205);
	137427	+ case 210: /* eidlist_opt ::= */ yytestcase(yyruleno==210);
136972	137428	{yymsp[1].minor.yy148 = 0;}
136973	137429	break;
136974	137430	case 92: /* selcollist ::= sclp expr as */
136975	137431	{
136976	137432	yymsp[-2].minor.yy148 = sqlite3ExprListAppend(pParse, yymsp[-2].minor.yy148, yymsp[-1].minor.yy190.pExpr);
		@@ -136992,12 +137448,12 @@
136992	137448	yymsp[-3].minor.yy148 = sqlite3ExprListAppend(pParse,yymsp[-3].minor.yy148, pDot);
136993	137449	}
136994	137450	break;
136995	137451	case 95: /* as ::= AS nm */
136996	137452	case 106: /* dbnm ::= DOT nm */ yytestcase(yyruleno==106);
136997		- case 225: /* plus_num ::= PLUS INTEGER\|FLOAT */ yytestcase(yyruleno==225);
136998		- case 226: /* minus_num ::= MINUS INTEGER\|FLOAT */ yytestcase(yyruleno==226);
	137453	+ case 224: /* plus_num ::= PLUS INTEGER\|FLOAT */ yytestcase(yyruleno==224);
	137454	+ case 225: /* minus_num ::= MINUS INTEGER\|FLOAT */ yytestcase(yyruleno==225);
136999	137455	{yymsp[-1].minor.yy0 = yymsp[0].minor.yy0;}
137000	137456	break;
137001	137457	case 97: /* from ::= */
137002	137458	{yymsp[1].minor.yy185 = sqlite3DbMallocZero(pParse->db, sizeof(*yymsp[1].minor.yy185));}
137003	137459	break;
		@@ -137076,18 +137532,18 @@
137076	137532	{yymsp[-3].minor.yy194 = sqlite3JoinType(pParse,&yymsp[-3].minor.yy0,&yymsp[-2].minor.yy0,&yymsp[-1].minor.yy0);/X-overwrites-A/}
137077	137533	break;
137078	137534	case 112: /* on_opt ::= ON expr */
137079	137535	case 129: /* having_opt ::= HAVING expr */ yytestcase(yyruleno==129);
137080	137536	case 136: /* where_opt ::= WHERE expr */ yytestcase(yyruleno==136);
137081		- case 199: /* case_else ::= ELSE expr */ yytestcase(yyruleno==199);
	137537	+ case 198: /* case_else ::= ELSE expr */ yytestcase(yyruleno==198);
137082	137538	{yymsp[-1].minor.yy72 = yymsp[0].minor.yy190.pExpr;}
137083	137539	break;
137084	137540	case 113: /* on_opt ::= */
137085	137541	case 128: /* having_opt ::= */ yytestcase(yyruleno==128);
137086	137542	case 135: /* where_opt ::= */ yytestcase(yyruleno==135);
137087		- case 200: /* case_else ::= */ yytestcase(yyruleno==200);
137088		- case 202: /* case_operand ::= */ yytestcase(yyruleno==202);
	137543	+ case 199: /* case_else ::= */ yytestcase(yyruleno==199);
	137544	+ case 201: /* case_operand ::= */ yytestcase(yyruleno==201);
137089	137545	{yymsp[1].minor.yy72 = 0;}
137090	137546	break;
137091	137547	case 115: /* indexed_opt ::= INDEXED BY nm */
137092	137548	{yymsp[-2].minor.yy0 = yymsp[0].minor.yy0;}
137093	137549	break;
		@@ -137230,11 +137686,11 @@
137230	137686	case 158: /* term ::= INTEGER */
137231	137687	{
137232	137688	yylhsminor.yy190.pExpr = sqlite3ExprAlloc(pParse->db, TK_INTEGER, &yymsp[0].minor.yy0, 1);
137233	137689	yylhsminor.yy190.zStart = yymsp[0].minor.yy0.z;
137234	137690	yylhsminor.yy190.zEnd = yymsp[0].minor.yy0.z + yymsp[0].minor.yy0.n;
137235		- if( yylhsminor.yy190.pExpr ) yylhsminor.yy190.pExpr->flags \|= EP_Leaf;
	137691	+ if( yylhsminor.yy190.pExpr ) yylhsminor.yy190.pExpr->flags \|= EP_Leaf\|EP_Resolved;
137236	137692	}
137237	137693	yymsp[0].minor.yy190 = yylhsminor.yy190;
137238	137694	break;
137239	137695	case 159: /* expr ::= VARIABLE */
137240	137696	{
		@@ -137320,17 +137776,14 @@
137320	137776	case 171: /* expr ::= expr PLUS\|MINUS expr */ yytestcase(yyruleno==171);
137321	137777	case 172: /* expr ::= expr STAR\|SLASH\|REM expr */ yytestcase(yyruleno==172);
137322	137778	case 173: /* expr ::= expr CONCAT expr */ yytestcase(yyruleno==173);
137323	137779	{spanBinaryExpr(pParse,yymsp[-1].major,&yymsp[-2].minor.yy190,&yymsp[0].minor.yy190);}
137324	137780	break;
137325		- case 174: /* likeop ::= LIKE_KW\|MATCH */
137326		-{yymsp[0].minor.yy0=yymsp[0].minor.yy0;/A-overwrites-X/}
137327		- break;
137328		- case 175: /* likeop ::= NOT LIKE_KW\|MATCH */
	137781	+ case 174: /* likeop ::= NOT LIKE_KW\|MATCH */
137329	137782	{yymsp[-1].minor.yy0=yymsp[0].minor.yy0; yymsp[-1].minor.yy0.n\|=0x80000000; /yymsp[-1].minor.yy0-overwrite-yymsp[0].minor.yy0/}
137330	137783	break;
137331		- case 176: /* expr ::= expr likeop expr */
	137784	+ case 175: /* expr ::= expr likeop expr */
137332	137785	{
137333	137786	ExprList *pList;
137334	137787	int bNot = yymsp[-1].minor.yy0.n & 0x80000000;
137335	137788	yymsp[-1].minor.yy0.n &= 0x7fffffff;
137336	137789	pList = sqlite3ExprListAppend(pParse,0, yymsp[0].minor.yy190.pExpr);
		@@ -137339,11 +137792,11 @@
137339	137792	exprNot(pParse, bNot, &yymsp[-2].minor.yy190);
137340	137793	yymsp[-2].minor.yy190.zEnd = yymsp[0].minor.yy190.zEnd;
137341	137794	if( yymsp[-2].minor.yy190.pExpr ) yymsp[-2].minor.yy190.pExpr->flags \|= EP_InfixFunc;
137342	137795	}
137343	137796	break;
137344		- case 177: /* expr ::= expr likeop expr ESCAPE expr */
	137797	+ case 176: /* expr ::= expr likeop expr ESCAPE expr */
137345	137798	{
137346	137799	ExprList *pList;
137347	137800	int bNot = yymsp[-3].minor.yy0.n & 0x80000000;
137348	137801	yymsp[-3].minor.yy0.n &= 0x7fffffff;
137349	137802	pList = sqlite3ExprListAppend(pParse,0, yymsp[-2].minor.yy190.pExpr);
		@@ -137353,43 +137806,43 @@
137353	137806	exprNot(pParse, bNot, &yymsp[-4].minor.yy190);
137354	137807	yymsp[-4].minor.yy190.zEnd = yymsp[0].minor.yy190.zEnd;
137355	137808	if( yymsp[-4].minor.yy190.pExpr ) yymsp[-4].minor.yy190.pExpr->flags \|= EP_InfixFunc;
137356	137809	}
137357	137810	break;
137358		- case 178: /* expr ::= expr ISNULL\|NOTNULL */
	137811	+ case 177: /* expr ::= expr ISNULL\|NOTNULL */
137359	137812	{spanUnaryPostfix(pParse,yymsp[0].major,&yymsp[-1].minor.yy190,&yymsp[0].minor.yy0);}
137360	137813	break;
137361		- case 179: /* expr ::= expr NOT NULL */
	137814	+ case 178: /* expr ::= expr NOT NULL */
137362	137815	{spanUnaryPostfix(pParse,TK_NOTNULL,&yymsp[-2].minor.yy190,&yymsp[0].minor.yy0);}
137363	137816	break;
137364		- case 180: /* expr ::= expr IS expr */
	137817	+ case 179: /* expr ::= expr IS expr */
137365	137818	{
137366	137819	spanBinaryExpr(pParse,TK_IS,&yymsp[-2].minor.yy190,&yymsp[0].minor.yy190);
137367	137820	binaryToUnaryIfNull(pParse, yymsp[0].minor.yy190.pExpr, yymsp[-2].minor.yy190.pExpr, TK_ISNULL);
137368	137821	}
137369	137822	break;
137370		- case 181: /* expr ::= expr IS NOT expr */
	137823	+ case 180: /* expr ::= expr IS NOT expr */
137371	137824	{
137372	137825	spanBinaryExpr(pParse,TK_ISNOT,&yymsp[-3].minor.yy190,&yymsp[0].minor.yy190);
137373	137826	binaryToUnaryIfNull(pParse, yymsp[0].minor.yy190.pExpr, yymsp[-3].minor.yy190.pExpr, TK_NOTNULL);
137374	137827	}
137375	137828	break;
137376		- case 182: /* expr ::= NOT expr */
137377		- case 183: /* expr ::= BITNOT expr */ yytestcase(yyruleno==183);
	137829	+ case 181: /* expr ::= NOT expr */
	137830	+ case 182: /* expr ::= BITNOT expr */ yytestcase(yyruleno==182);
137378	137831	{spanUnaryPrefix(&yymsp[-1].minor.yy190,pParse,yymsp[-1].major,&yymsp[0].minor.yy190,&yymsp[-1].minor.yy0);/A-overwrites-B/}
137379	137832	break;
137380		- case 184: /* expr ::= MINUS expr */
	137833	+ case 183: /* expr ::= MINUS expr */
137381	137834	{spanUnaryPrefix(&yymsp[-1].minor.yy190,pParse,TK_UMINUS,&yymsp[0].minor.yy190,&yymsp[-1].minor.yy0);/A-overwrites-B/}
137382	137835	break;
137383		- case 185: /* expr ::= PLUS expr */
	137836	+ case 184: /* expr ::= PLUS expr */
137384	137837	{spanUnaryPrefix(&yymsp[-1].minor.yy190,pParse,TK_UPLUS,&yymsp[0].minor.yy190,&yymsp[-1].minor.yy0);/A-overwrites-B/}
137385	137838	break;
137386		- case 186: /* between_op ::= BETWEEN */
137387		- case 189: /* in_op ::= IN */ yytestcase(yyruleno==189);
	137839	+ case 185: /* between_op ::= BETWEEN */
	137840	+ case 188: /* in_op ::= IN */ yytestcase(yyruleno==188);
137388	137841	{yymsp[0].minor.yy194 = 0;}
137389	137842	break;
137390		- case 188: /* expr ::= expr between_op expr AND expr */
	137843	+ case 187: /* expr ::= expr between_op expr AND expr */
137391	137844	{
137392	137845	ExprList *pList = sqlite3ExprListAppend(pParse,0, yymsp[-2].minor.yy190.pExpr);
137393	137846	pList = sqlite3ExprListAppend(pParse,pList, yymsp[0].minor.yy190.pExpr);
137394	137847	yymsp[-4].minor.yy190.pExpr = sqlite3PExpr(pParse, TK_BETWEEN, yymsp[-4].minor.yy190.pExpr, 0);
137395	137848	if( yymsp[-4].minor.yy190.pExpr ){
		@@ -137399,11 +137852,11 @@
137399	137852	}
137400	137853	exprNot(pParse, yymsp[-3].minor.yy194, &yymsp[-4].minor.yy190);
137401	137854	yymsp[-4].minor.yy190.zEnd = yymsp[0].minor.yy190.zEnd;
137402	137855	}
137403	137856	break;
137404		- case 191: /* expr ::= expr in_op LP exprlist RP */
	137857	+ case 190: /* expr ::= expr in_op LP exprlist RP */
137405	137858	{
137406	137859	if( yymsp[-1].minor.yy148==0 ){
137407	137860	/* Expressions of the form
137408	137861	**
137409	137862	** expr1 IN ()
		@@ -137452,26 +137905,26 @@
137452	137905	exprNot(pParse, yymsp[-3].minor.yy194, &yymsp[-4].minor.yy190);
137453	137906	}
137454	137907	yymsp[-4].minor.yy190.zEnd = &yymsp[0].minor.yy0.z[yymsp[0].minor.yy0.n];
137455	137908	}
137456	137909	break;
137457		- case 192: /* expr ::= LP select RP */
	137910	+ case 191: /* expr ::= LP select RP */
137458	137911	{
137459	137912	spanSet(&yymsp[-2].minor.yy190,&yymsp[-2].minor.yy0,&yymsp[0].minor.yy0); /A-overwrites-B/
137460	137913	yymsp[-2].minor.yy190.pExpr = sqlite3PExpr(pParse, TK_SELECT, 0, 0);
137461	137914	sqlite3PExprAddSelect(pParse, yymsp[-2].minor.yy190.pExpr, yymsp[-1].minor.yy243);
137462	137915	}
137463	137916	break;
137464		- case 193: /* expr ::= expr in_op LP select RP */
	137917	+ case 192: /* expr ::= expr in_op LP select RP */
137465	137918	{
137466	137919	yymsp[-4].minor.yy190.pExpr = sqlite3PExpr(pParse, TK_IN, yymsp[-4].minor.yy190.pExpr, 0);
137467	137920	sqlite3PExprAddSelect(pParse, yymsp[-4].minor.yy190.pExpr, yymsp[-1].minor.yy243);
137468	137921	exprNot(pParse, yymsp[-3].minor.yy194, &yymsp[-4].minor.yy190);
137469	137922	yymsp[-4].minor.yy190.zEnd = &yymsp[0].minor.yy0.z[yymsp[0].minor.yy0.n];
137470	137923	}
137471	137924	break;
137472		- case 194: /* expr ::= expr in_op nm dbnm paren_exprlist */
	137925	+ case 193: /* expr ::= expr in_op nm dbnm paren_exprlist */
137473	137926	{
137474	137927	SrcList *pSrc = sqlite3SrcListAppend(pParse->db, 0,&yymsp[-2].minor.yy0,&yymsp[-1].minor.yy0);
137475	137928	Select *pSelect = sqlite3SelectNew(pParse, 0,pSrc,0,0,0,0,0,0,0);
137476	137929	if( yymsp[0].minor.yy148 ) sqlite3SrcListFuncArgs(pParse, pSelect ? pSrc : 0, yymsp[0].minor.yy148);
137477	137930	yymsp[-4].minor.yy190.pExpr = sqlite3PExpr(pParse, TK_IN, yymsp[-4].minor.yy190.pExpr, 0);
		@@ -137478,19 +137931,19 @@
137478	137931	sqlite3PExprAddSelect(pParse, yymsp[-4].minor.yy190.pExpr, pSelect);
137479	137932	exprNot(pParse, yymsp[-3].minor.yy194, &yymsp[-4].minor.yy190);
137480	137933	yymsp[-4].minor.yy190.zEnd = yymsp[-1].minor.yy0.z ? &yymsp[-1].minor.yy0.z[yymsp[-1].minor.yy0.n] : &yymsp[-2].minor.yy0.z[yymsp[-2].minor.yy0.n];
137481	137934	}
137482	137935	break;
137483		- case 195: /* expr ::= EXISTS LP select RP */
	137936	+ case 194: /* expr ::= EXISTS LP select RP */
137484	137937	{
137485	137938	Expr *p;
137486	137939	spanSet(&yymsp[-3].minor.yy190,&yymsp[-3].minor.yy0,&yymsp[0].minor.yy0); /A-overwrites-B/
137487	137940	p = yymsp[-3].minor.yy190.pExpr = sqlite3PExpr(pParse, TK_EXISTS, 0, 0);
137488	137941	sqlite3PExprAddSelect(pParse, p, yymsp[-1].minor.yy243);
137489	137942	}
137490	137943	break;
137491		- case 196: /* expr ::= CASE case_operand case_exprlist case_else END */
	137944	+ case 195: /* expr ::= CASE case_operand case_exprlist case_else END */
137492	137945	{
137493	137946	spanSet(&yymsp[-4].minor.yy190,&yymsp[-4].minor.yy0,&yymsp[0].minor.yy0); /A-overwrites-C/
137494	137947	yymsp[-4].minor.yy190.pExpr = sqlite3PExpr(pParse, TK_CASE, yymsp[-3].minor.yy72, 0);
137495	137948	if( yymsp[-4].minor.yy190.pExpr ){
137496	137949	yymsp[-4].minor.yy190.pExpr->x.pList = yymsp[-1].minor.yy72 ? sqlite3ExprListAppend(pParse,yymsp[-2].minor.yy148,yymsp[-1].minor.yy72) : yymsp[-2].minor.yy148;
		@@ -137499,313 +137952,314 @@
137499	137952	sqlite3ExprListDelete(pParse->db, yymsp[-2].minor.yy148);
137500	137953	sqlite3ExprDelete(pParse->db, yymsp[-1].minor.yy72);
137501	137954	}
137502	137955	}
137503	137956	break;
137504		- case 197: /* case_exprlist ::= case_exprlist WHEN expr THEN expr */
	137957	+ case 196: /* case_exprlist ::= case_exprlist WHEN expr THEN expr */
137505	137958	{
137506	137959	yymsp[-4].minor.yy148 = sqlite3ExprListAppend(pParse,yymsp[-4].minor.yy148, yymsp[-2].minor.yy190.pExpr);
137507	137960	yymsp[-4].minor.yy148 = sqlite3ExprListAppend(pParse,yymsp[-4].minor.yy148, yymsp[0].minor.yy190.pExpr);
137508	137961	}
137509	137962	break;
137510		- case 198: /* case_exprlist ::= WHEN expr THEN expr */
	137963	+ case 197: /* case_exprlist ::= WHEN expr THEN expr */
137511	137964	{
137512	137965	yymsp[-3].minor.yy148 = sqlite3ExprListAppend(pParse,0, yymsp[-2].minor.yy190.pExpr);
137513	137966	yymsp[-3].minor.yy148 = sqlite3ExprListAppend(pParse,yymsp[-3].minor.yy148, yymsp[0].minor.yy190.pExpr);
137514	137967	}
137515	137968	break;
137516		- case 201: /* case_operand ::= expr */
	137969	+ case 200: /* case_operand ::= expr */
137517	137970	{yymsp[0].minor.yy72 = yymsp[0].minor.yy190.pExpr; /A-overwrites-X/}
137518	137971	break;
137519		- case 204: /* nexprlist ::= nexprlist COMMA expr */
	137972	+ case 203: /* nexprlist ::= nexprlist COMMA expr */
137520	137973	{yymsp[-2].minor.yy148 = sqlite3ExprListAppend(pParse,yymsp[-2].minor.yy148,yymsp[0].minor.yy190.pExpr);}
137521	137974	break;
137522		- case 205: /* nexprlist ::= expr */
	137975	+ case 204: /* nexprlist ::= expr */
137523	137976	{yymsp[0].minor.yy148 = sqlite3ExprListAppend(pParse,0,yymsp[0].minor.yy190.pExpr); /A-overwrites-Y/}
137524	137977	break;
137525		- case 207: /* paren_exprlist ::= LP exprlist RP */
137526		- case 212: /* eidlist_opt ::= LP eidlist RP */ yytestcase(yyruleno==212);
	137978	+ case 206: /* paren_exprlist ::= LP exprlist RP */
	137979	+ case 211: /* eidlist_opt ::= LP eidlist RP */ yytestcase(yyruleno==211);
137527	137980	{yymsp[-2].minor.yy148 = yymsp[-1].minor.yy148;}
137528	137981	break;
137529		- case 208: /* cmd ::= createkw uniqueflag INDEX ifnotexists nm dbnm ON nm LP sortlist RP where_opt */
	137982	+ case 207: /* cmd ::= createkw uniqueflag INDEX ifnotexists nm dbnm ON nm LP sortlist RP where_opt */
137530	137983	{
137531	137984	sqlite3CreateIndex(pParse, &yymsp[-7].minor.yy0, &yymsp[-6].minor.yy0,
137532	137985	sqlite3SrcListAppend(pParse->db,0,&yymsp[-4].minor.yy0,0), yymsp[-2].minor.yy148, yymsp[-10].minor.yy194,
137533	137986	&yymsp[-11].minor.yy0, yymsp[0].minor.yy72, SQLITE_SO_ASC, yymsp[-8].minor.yy194, SQLITE_IDXTYPE_APPDEF);
137534	137987	}
137535	137988	break;
137536		- case 209: /* uniqueflag ::= UNIQUE */
137537		- case 250: /* raisetype ::= ABORT */ yytestcase(yyruleno==250);
	137989	+ case 208: /* uniqueflag ::= UNIQUE */
	137990	+ case 249: /* raisetype ::= ABORT */ yytestcase(yyruleno==249);
137538	137991	{yymsp[0].minor.yy194 = OE_Abort;}
137539	137992	break;
137540		- case 210: /* uniqueflag ::= */
	137993	+ case 209: /* uniqueflag ::= */
137541	137994	{yymsp[1].minor.yy194 = OE_None;}
137542	137995	break;
137543		- case 213: /* eidlist ::= eidlist COMMA nm collate sortorder */
	137996	+ case 212: /* eidlist ::= eidlist COMMA nm collate sortorder */
137544	137997	{
137545	137998	yymsp[-4].minor.yy148 = parserAddExprIdListTerm(pParse, yymsp[-4].minor.yy148, &yymsp[-2].minor.yy0, yymsp[-1].minor.yy194, yymsp[0].minor.yy194);
137546	137999	}
137547	138000	break;
137548		- case 214: /* eidlist ::= nm collate sortorder */
	138001	+ case 213: /* eidlist ::= nm collate sortorder */
137549	138002	{
137550	138003	yymsp[-2].minor.yy148 = parserAddExprIdListTerm(pParse, 0, &yymsp[-2].minor.yy0, yymsp[-1].minor.yy194, yymsp[0].minor.yy194); /A-overwrites-Y/
137551	138004	}
137552	138005	break;
137553		- case 217: /* cmd ::= DROP INDEX ifexists fullname */
	138006	+ case 216: /* cmd ::= DROP INDEX ifexists fullname */
137554	138007	{sqlite3DropIndex(pParse, yymsp[0].minor.yy185, yymsp[-1].minor.yy194);}
137555	138008	break;
137556		- case 218: /* cmd ::= VACUUM */
	138009	+ case 217: /* cmd ::= VACUUM */
137557	138010	{sqlite3Vacuum(pParse,0);}
137558	138011	break;
137559		- case 219: /* cmd ::= VACUUM nm */
	138012	+ case 218: /* cmd ::= VACUUM nm */
137560	138013	{sqlite3Vacuum(pParse,&yymsp[0].minor.yy0);}
137561	138014	break;
137562		- case 220: /* cmd ::= PRAGMA nm dbnm */
	138015	+ case 219: /* cmd ::= PRAGMA nm dbnm */
137563	138016	{sqlite3Pragma(pParse,&yymsp[-1].minor.yy0,&yymsp[0].minor.yy0,0,0);}
137564	138017	break;
137565		- case 221: /* cmd ::= PRAGMA nm dbnm EQ nmnum */
	138018	+ case 220: /* cmd ::= PRAGMA nm dbnm EQ nmnum */
137566	138019	{sqlite3Pragma(pParse,&yymsp[-3].minor.yy0,&yymsp[-2].minor.yy0,&yymsp[0].minor.yy0,0);}
137567	138020	break;
137568		- case 222: /* cmd ::= PRAGMA nm dbnm LP nmnum RP */
	138021	+ case 221: /* cmd ::= PRAGMA nm dbnm LP nmnum RP */
137569	138022	{sqlite3Pragma(pParse,&yymsp[-4].minor.yy0,&yymsp[-3].minor.yy0,&yymsp[-1].minor.yy0,0);}
137570	138023	break;
137571		- case 223: /* cmd ::= PRAGMA nm dbnm EQ minus_num */
	138024	+ case 222: /* cmd ::= PRAGMA nm dbnm EQ minus_num */
137572	138025	{sqlite3Pragma(pParse,&yymsp[-3].minor.yy0,&yymsp[-2].minor.yy0,&yymsp[0].minor.yy0,1);}
137573	138026	break;
137574		- case 224: /* cmd ::= PRAGMA nm dbnm LP minus_num RP */
	138027	+ case 223: /* cmd ::= PRAGMA nm dbnm LP minus_num RP */
137575	138028	{sqlite3Pragma(pParse,&yymsp[-4].minor.yy0,&yymsp[-3].minor.yy0,&yymsp[-1].minor.yy0,1);}
137576	138029	break;
137577		- case 227: /* cmd ::= createkw trigger_decl BEGIN trigger_cmd_list END */
	138030	+ case 226: /* cmd ::= createkw trigger_decl BEGIN trigger_cmd_list END */
137578	138031	{
137579	138032	Token all;
137580	138033	all.z = yymsp[-3].minor.yy0.z;
137581	138034	all.n = (int)(yymsp[0].minor.yy0.z - yymsp[-3].minor.yy0.z) + yymsp[0].minor.yy0.n;
137582	138035	sqlite3FinishTrigger(pParse, yymsp[-1].minor.yy145, &all);
137583	138036	}
137584	138037	break;
137585		- case 228: /* trigger_decl ::= temp TRIGGER ifnotexists nm dbnm trigger_time trigger_event ON fullname foreach_clause when_clause */
	138038	+ case 227: /* trigger_decl ::= temp TRIGGER ifnotexists nm dbnm trigger_time trigger_event ON fullname foreach_clause when_clause */
137586	138039	{
137587	138040	sqlite3BeginTrigger(pParse, &yymsp[-7].minor.yy0, &yymsp[-6].minor.yy0, yymsp[-5].minor.yy194, yymsp[-4].minor.yy332.a, yymsp[-4].minor.yy332.b, yymsp[-2].minor.yy185, yymsp[0].minor.yy72, yymsp[-10].minor.yy194, yymsp[-8].minor.yy194);
137588	138041	yymsp[-10].minor.yy0 = (yymsp[-6].minor.yy0.n==0?yymsp[-7].minor.yy0:yymsp[-6].minor.yy0); /A-overwrites-T/
137589	138042	}
137590	138043	break;
137591		- case 229: /* trigger_time ::= BEFORE */
	138044	+ case 228: /* trigger_time ::= BEFORE */
137592	138045	{ yymsp[0].minor.yy194 = TK_BEFORE; }
137593	138046	break;
137594		- case 230: /* trigger_time ::= AFTER */
	138047	+ case 229: /* trigger_time ::= AFTER */
137595	138048	{ yymsp[0].minor.yy194 = TK_AFTER; }
137596	138049	break;
137597		- case 231: /* trigger_time ::= INSTEAD OF */
	138050	+ case 230: /* trigger_time ::= INSTEAD OF */
137598	138051	{ yymsp[-1].minor.yy194 = TK_INSTEAD;}
137599	138052	break;
137600		- case 232: /* trigger_time ::= */
	138053	+ case 231: /* trigger_time ::= */
137601	138054	{ yymsp[1].minor.yy194 = TK_BEFORE; }
137602	138055	break;
137603		- case 233: /* trigger_event ::= DELETE\|INSERT */
137604		- case 234: /* trigger_event ::= UPDATE */ yytestcase(yyruleno==234);
	138056	+ case 232: /* trigger_event ::= DELETE\|INSERT */
	138057	+ case 233: /* trigger_event ::= UPDATE */ yytestcase(yyruleno==233);
137605	138058	{yymsp[0].minor.yy332.a = yymsp[0].major; /A-overwrites-X/ yymsp[0].minor.yy332.b = 0;}
137606	138059	break;
137607		- case 235: /* trigger_event ::= UPDATE OF idlist */
	138060	+ case 234: /* trigger_event ::= UPDATE OF idlist */
137608	138061	{yymsp[-2].minor.yy332.a = TK_UPDATE; yymsp[-2].minor.yy332.b = yymsp[0].minor.yy254;}
137609	138062	break;
137610		- case 236: /* when_clause ::= */
137611		- case 255: /* key_opt ::= */ yytestcase(yyruleno==255);
	138063	+ case 235: /* when_clause ::= */
	138064	+ case 254: /* key_opt ::= */ yytestcase(yyruleno==254);
137612	138065	{ yymsp[1].minor.yy72 = 0; }
137613	138066	break;
137614		- case 237: /* when_clause ::= WHEN expr */
137615		- case 256: /* key_opt ::= KEY expr */ yytestcase(yyruleno==256);
	138067	+ case 236: /* when_clause ::= WHEN expr */
	138068	+ case 255: /* key_opt ::= KEY expr */ yytestcase(yyruleno==255);
137616	138069	{ yymsp[-1].minor.yy72 = yymsp[0].minor.yy190.pExpr; }
137617	138070	break;
137618		- case 238: /* trigger_cmd_list ::= trigger_cmd_list trigger_cmd SEMI */
	138071	+ case 237: /* trigger_cmd_list ::= trigger_cmd_list trigger_cmd SEMI */
137619	138072	{
137620	138073	assert( yymsp[-2].minor.yy145!=0 );
137621	138074	yymsp[-2].minor.yy145->pLast->pNext = yymsp[-1].minor.yy145;
137622	138075	yymsp[-2].minor.yy145->pLast = yymsp[-1].minor.yy145;
137623	138076	}
137624	138077	break;
137625		- case 239: /* trigger_cmd_list ::= trigger_cmd SEMI */
	138078	+ case 238: /* trigger_cmd_list ::= trigger_cmd SEMI */
137626	138079	{
137627	138080	assert( yymsp[-1].minor.yy145!=0 );
137628	138081	yymsp[-1].minor.yy145->pLast = yymsp[-1].minor.yy145;
137629	138082	}
137630	138083	break;
137631		- case 240: /* trnm ::= nm DOT nm */
	138084	+ case 239: /* trnm ::= nm DOT nm */
137632	138085	{
137633	138086	yymsp[-2].minor.yy0 = yymsp[0].minor.yy0;
137634	138087	sqlite3ErrorMsg(pParse,
137635	138088	"qualified table names are not allowed on INSERT, UPDATE, and DELETE "
137636	138089	"statements within triggers");
137637	138090	}
137638	138091	break;
137639		- case 241: /* tridxby ::= INDEXED BY nm */
	138092	+ case 240: /* tridxby ::= INDEXED BY nm */
137640	138093	{
137641	138094	sqlite3ErrorMsg(pParse,
137642	138095	"the INDEXED BY clause is not allowed on UPDATE or DELETE statements "
137643	138096	"within triggers");
137644	138097	}
137645	138098	break;
137646		- case 242: /* tridxby ::= NOT INDEXED */
	138099	+ case 241: /* tridxby ::= NOT INDEXED */
137647	138100	{
137648	138101	sqlite3ErrorMsg(pParse,
137649	138102	"the NOT INDEXED clause is not allowed on UPDATE or DELETE statements "
137650	138103	"within triggers");
137651	138104	}
137652	138105	break;
137653		- case 243: /* trigger_cmd ::= UPDATE orconf trnm tridxby SET setlist where_opt */
	138106	+ case 242: /* trigger_cmd ::= UPDATE orconf trnm tridxby SET setlist where_opt */
137654	138107	{yymsp[-6].minor.yy145 = sqlite3TriggerUpdateStep(pParse->db, &yymsp[-4].minor.yy0, yymsp[-1].minor.yy148, yymsp[0].minor.yy72, yymsp[-5].minor.yy194);}
137655	138108	break;
137656		- case 244: /* trigger_cmd ::= insert_cmd INTO trnm idlist_opt select */
	138109	+ case 243: /* trigger_cmd ::= insert_cmd INTO trnm idlist_opt select */
137657	138110	{yymsp[-4].minor.yy145 = sqlite3TriggerInsertStep(pParse->db, &yymsp[-2].minor.yy0, yymsp[-1].minor.yy254, yymsp[0].minor.yy243, yymsp[-4].minor.yy194);/A-overwrites-R/}
137658	138111	break;
137659		- case 245: /* trigger_cmd ::= DELETE FROM trnm tridxby where_opt */
	138112	+ case 244: /* trigger_cmd ::= DELETE FROM trnm tridxby where_opt */
137660	138113	{yymsp[-4].minor.yy145 = sqlite3TriggerDeleteStep(pParse->db, &yymsp[-2].minor.yy0, yymsp[0].minor.yy72);}
137661	138114	break;
137662		- case 246: /* trigger_cmd ::= select */
	138115	+ case 245: /* trigger_cmd ::= select */
137663	138116	{yymsp[0].minor.yy145 = sqlite3TriggerSelectStep(pParse->db, yymsp[0].minor.yy243); /A-overwrites-X/}
137664	138117	break;
137665		- case 247: /* expr ::= RAISE LP IGNORE RP */
	138118	+ case 246: /* expr ::= RAISE LP IGNORE RP */
137666	138119	{
137667	138120	spanSet(&yymsp[-3].minor.yy190,&yymsp[-3].minor.yy0,&yymsp[0].minor.yy0); /A-overwrites-X/
137668	138121	yymsp[-3].minor.yy190.pExpr = sqlite3PExpr(pParse, TK_RAISE, 0, 0);
137669	138122	if( yymsp[-3].minor.yy190.pExpr ){
137670	138123	yymsp[-3].minor.yy190.pExpr->affinity = OE_Ignore;
137671	138124	}
137672	138125	}
137673	138126	break;
137674		- case 248: /* expr ::= RAISE LP raisetype COMMA nm RP */
	138127	+ case 247: /* expr ::= RAISE LP raisetype COMMA nm RP */
137675	138128	{
137676	138129	spanSet(&yymsp[-5].minor.yy190,&yymsp[-5].minor.yy0,&yymsp[0].minor.yy0); /A-overwrites-X/
137677	138130	yymsp[-5].minor.yy190.pExpr = sqlite3ExprAlloc(pParse->db, TK_RAISE, &yymsp[-1].minor.yy0, 1);
137678	138131	if( yymsp[-5].minor.yy190.pExpr ) {
137679	138132	yymsp[-5].minor.yy190.pExpr->affinity = (char)yymsp[-3].minor.yy194;
137680	138133	}
137681	138134	}
137682	138135	break;
137683		- case 249: /* raisetype ::= ROLLBACK */
	138136	+ case 248: /* raisetype ::= ROLLBACK */
137684	138137	{yymsp[0].minor.yy194 = OE_Rollback;}
137685	138138	break;
137686		- case 251: /* raisetype ::= FAIL */
	138139	+ case 250: /* raisetype ::= FAIL */
137687	138140	{yymsp[0].minor.yy194 = OE_Fail;}
137688	138141	break;
137689		- case 252: /* cmd ::= DROP TRIGGER ifexists fullname */
	138142	+ case 251: /* cmd ::= DROP TRIGGER ifexists fullname */
137690	138143	{
137691	138144	sqlite3DropTrigger(pParse,yymsp[0].minor.yy185,yymsp[-1].minor.yy194);
137692	138145	}
137693	138146	break;
137694		- case 253: /* cmd ::= ATTACH database_kw_opt expr AS expr key_opt */
	138147	+ case 252: /* cmd ::= ATTACH database_kw_opt expr AS expr key_opt */
137695	138148	{
137696	138149	sqlite3Attach(pParse, yymsp[-3].minor.yy190.pExpr, yymsp[-1].minor.yy190.pExpr, yymsp[0].minor.yy72);
137697	138150	}
137698	138151	break;
137699		- case 254: /* cmd ::= DETACH database_kw_opt expr */
	138152	+ case 253: /* cmd ::= DETACH database_kw_opt expr */
137700	138153	{
137701	138154	sqlite3Detach(pParse, yymsp[0].minor.yy190.pExpr);
137702	138155	}
137703	138156	break;
137704		- case 257: /* cmd ::= REINDEX */
	138157	+ case 256: /* cmd ::= REINDEX */
137705	138158	{sqlite3Reindex(pParse, 0, 0);}
137706	138159	break;
137707		- case 258: /* cmd ::= REINDEX nm dbnm */
	138160	+ case 257: /* cmd ::= REINDEX nm dbnm */
137708	138161	{sqlite3Reindex(pParse, &yymsp[-1].minor.yy0, &yymsp[0].minor.yy0);}
137709	138162	break;
137710		- case 259: /* cmd ::= ANALYZE */
	138163	+ case 258: /* cmd ::= ANALYZE */
137711	138164	{sqlite3Analyze(pParse, 0, 0);}
137712	138165	break;
137713		- case 260: /* cmd ::= ANALYZE nm dbnm */
	138166	+ case 259: /* cmd ::= ANALYZE nm dbnm */
137714	138167	{sqlite3Analyze(pParse, &yymsp[-1].minor.yy0, &yymsp[0].minor.yy0);}
137715	138168	break;
137716		- case 261: /* cmd ::= ALTER TABLE fullname RENAME TO nm */
	138169	+ case 260: /* cmd ::= ALTER TABLE fullname RENAME TO nm */
137717	138170	{
137718	138171	sqlite3AlterRenameTable(pParse,yymsp[-3].minor.yy185,&yymsp[0].minor.yy0);
137719	138172	}
137720	138173	break;
137721		- case 262: /* cmd ::= ALTER TABLE add_column_fullname ADD kwcolumn_opt columnname carglist */
	138174	+ case 261: /* cmd ::= ALTER TABLE add_column_fullname ADD kwcolumn_opt columnname carglist */
137722	138175	{
137723	138176	yymsp[-1].minor.yy0.n = (int)(pParse->sLastToken.z-yymsp[-1].minor.yy0.z) + pParse->sLastToken.n;
137724	138177	sqlite3AlterFinishAddColumn(pParse, &yymsp[-1].minor.yy0);
137725	138178	}
137726	138179	break;
137727		- case 263: /* add_column_fullname ::= fullname */
	138180	+ case 262: /* add_column_fullname ::= fullname */
137728	138181	{
137729	138182	disableLookaside(pParse);
137730	138183	sqlite3AlterBeginAddColumn(pParse, yymsp[0].minor.yy185);
137731	138184	}
137732	138185	break;
137733		- case 264: /* cmd ::= create_vtab */
	138186	+ case 263: /* cmd ::= create_vtab */
137734	138187	{sqlite3VtabFinishParse(pParse,0);}
137735	138188	break;
137736		- case 265: /* cmd ::= create_vtab LP vtabarglist RP */
	138189	+ case 264: /* cmd ::= create_vtab LP vtabarglist RP */
137737	138190	{sqlite3VtabFinishParse(pParse,&yymsp[0].minor.yy0);}
137738	138191	break;
137739		- case 266: /* create_vtab ::= createkw VIRTUAL TABLE ifnotexists nm dbnm USING nm */
	138192	+ case 265: /* create_vtab ::= createkw VIRTUAL TABLE ifnotexists nm dbnm USING nm */
137740	138193	{
137741	138194	sqlite3VtabBeginParse(pParse, &yymsp[-3].minor.yy0, &yymsp[-2].minor.yy0, &yymsp[0].minor.yy0, yymsp[-4].minor.yy194);
137742	138195	}
137743	138196	break;
137744		- case 267: /* vtabarg ::= */
	138197	+ case 266: /* vtabarg ::= */
137745	138198	{sqlite3VtabArgInit(pParse);}
137746	138199	break;
137747		- case 268: /* vtabargtoken ::= ANY */
137748		- case 269: /* vtabargtoken ::= lp anylist RP */ yytestcase(yyruleno==269);
137749		- case 270: /* lp ::= LP */ yytestcase(yyruleno==270);
	138200	+ case 267: /* vtabargtoken ::= ANY */
	138201	+ case 268: /* vtabargtoken ::= lp anylist RP */ yytestcase(yyruleno==268);
	138202	+ case 269: /* lp ::= LP */ yytestcase(yyruleno==269);
137750	138203	{sqlite3VtabArgExtend(pParse,&yymsp[0].minor.yy0);}
137751	138204	break;
137752		- case 271: /* with ::= */
	138205	+ case 270: /* with ::= */
137753	138206	{yymsp[1].minor.yy285 = 0;}
137754	138207	break;
137755		- case 272: /* with ::= WITH wqlist */
	138208	+ case 271: /* with ::= WITH wqlist */
137756	138209	{ yymsp[-1].minor.yy285 = yymsp[0].minor.yy285; }
137757	138210	break;
137758		- case 273: /* with ::= WITH RECURSIVE wqlist */
	138211	+ case 272: /* with ::= WITH RECURSIVE wqlist */
137759	138212	{ yymsp[-2].minor.yy285 = yymsp[0].minor.yy285; }
137760	138213	break;
137761		- case 274: /* wqlist ::= nm eidlist_opt AS LP select RP */
	138214	+ case 273: /* wqlist ::= nm eidlist_opt AS LP select RP */
137762	138215	{
137763	138216	yymsp[-5].minor.yy285 = sqlite3WithAdd(pParse, 0, &yymsp[-5].minor.yy0, yymsp[-4].minor.yy148, yymsp[-1].minor.yy243); /A-overwrites-X/
137764	138217	}
137765	138218	break;
137766		- case 275: /* wqlist ::= wqlist COMMA nm eidlist_opt AS LP select RP */
	138219	+ case 274: /* wqlist ::= wqlist COMMA nm eidlist_opt AS LP select RP */
137767	138220	{
137768	138221	yymsp[-7].minor.yy285 = sqlite3WithAdd(pParse, yymsp[-7].minor.yy285, &yymsp[-5].minor.yy0, yymsp[-4].minor.yy148, yymsp[-1].minor.yy243);
137769	138222	}
137770	138223	break;
137771	138224	default:
137772		- /* (276) input ::= cmdlist */ yytestcase(yyruleno==276);
137773		- /* (277) cmdlist ::= cmdlist ecmd */ yytestcase(yyruleno==277);
137774		- /* (278) cmdlist ::= ecmd (OPTIMIZED OUT) */ assert(yyruleno!=278);
137775		- /* (279) ecmd ::= SEMI */ yytestcase(yyruleno==279);
137776		- /* (280) ecmd ::= explain cmdx SEMI */ yytestcase(yyruleno==280);
137777		- /* (281) explain ::= */ yytestcase(yyruleno==281);
137778		- /* (282) trans_opt ::= */ yytestcase(yyruleno==282);
137779		- /* (283) trans_opt ::= TRANSACTION */ yytestcase(yyruleno==283);
137780		- /* (284) trans_opt ::= TRANSACTION nm */ yytestcase(yyruleno==284);
137781		- /* (285) savepoint_opt ::= SAVEPOINT */ yytestcase(yyruleno==285);
137782		- /* (286) savepoint_opt ::= */ yytestcase(yyruleno==286);
137783		- /* (287) cmd ::= create_table create_table_args */ yytestcase(yyruleno==287);
137784		- /* (288) columnlist ::= columnlist COMMA columnname carglist */ yytestcase(yyruleno==288);
137785		- /* (289) columnlist ::= columnname carglist */ yytestcase(yyruleno==289);
137786		- /* (290) nm ::= ID\|INDEXED */ yytestcase(yyruleno==290);
137787		- /* (291) nm ::= STRING */ yytestcase(yyruleno==291);
137788		- /* (292) nm ::= JOIN_KW */ yytestcase(yyruleno==292);
137789		- /* (293) typetoken ::= typename */ yytestcase(yyruleno==293);
137790		- /* (294) typename ::= ID\|STRING */ yytestcase(yyruleno==294);
137791		- /* (295) signed ::= plus_num (OPTIMIZED OUT) */ assert(yyruleno!=295);
137792		- /* (296) signed ::= minus_num (OPTIMIZED OUT) */ assert(yyruleno!=296);
137793		- /* (297) carglist ::= carglist ccons */ yytestcase(yyruleno==297);
137794		- /* (298) carglist ::= */ yytestcase(yyruleno==298);
137795		- /* (299) ccons ::= NULL onconf */ yytestcase(yyruleno==299);
137796		- /* (300) conslist_opt ::= COMMA conslist */ yytestcase(yyruleno==300);
137797		- /* (301) conslist ::= conslist tconscomma tcons */ yytestcase(yyruleno==301);
137798		- /* (302) conslist ::= tcons (OPTIMIZED OUT) */ assert(yyruleno!=302);
137799		- /* (303) tconscomma ::= */ yytestcase(yyruleno==303);
137800		- /* (304) defer_subclause_opt ::= defer_subclause (OPTIMIZED OUT) */ assert(yyruleno!=304);
137801		- /* (305) resolvetype ::= raisetype (OPTIMIZED OUT) */ assert(yyruleno!=305);
137802		- /* (306) selectnowith ::= oneselect (OPTIMIZED OUT) */ assert(yyruleno!=306);
137803		- /* (307) oneselect ::= values */ yytestcase(yyruleno==307);
137804		- /* (308) sclp ::= selcollist COMMA */ yytestcase(yyruleno==308);
137805		- /* (309) as ::= ID\|STRING */ yytestcase(yyruleno==309);
137806		- /* (310) expr ::= term (OPTIMIZED OUT) */ assert(yyruleno!=310);
	138225	+ /* (275) input ::= cmdlist */ yytestcase(yyruleno==275);
	138226	+ /* (276) cmdlist ::= cmdlist ecmd */ yytestcase(yyruleno==276);
	138227	+ /* (277) cmdlist ::= ecmd (OPTIMIZED OUT) */ assert(yyruleno!=277);
	138228	+ /* (278) ecmd ::= SEMI */ yytestcase(yyruleno==278);
	138229	+ /* (279) ecmd ::= explain cmdx SEMI */ yytestcase(yyruleno==279);
	138230	+ /* (280) explain ::= */ yytestcase(yyruleno==280);
	138231	+ /* (281) trans_opt ::= */ yytestcase(yyruleno==281);
	138232	+ /* (282) trans_opt ::= TRANSACTION */ yytestcase(yyruleno==282);
	138233	+ /* (283) trans_opt ::= TRANSACTION nm */ yytestcase(yyruleno==283);
	138234	+ /* (284) savepoint_opt ::= SAVEPOINT */ yytestcase(yyruleno==284);
	138235	+ /* (285) savepoint_opt ::= */ yytestcase(yyruleno==285);
	138236	+ /* (286) cmd ::= create_table create_table_args */ yytestcase(yyruleno==286);
	138237	+ /* (287) columnlist ::= columnlist COMMA columnname carglist */ yytestcase(yyruleno==287);
	138238	+ /* (288) columnlist ::= columnname carglist */ yytestcase(yyruleno==288);
	138239	+ /* (289) nm ::= ID\|INDEXED */ yytestcase(yyruleno==289);
	138240	+ /* (290) nm ::= STRING */ yytestcase(yyruleno==290);
	138241	+ /* (291) nm ::= JOIN_KW */ yytestcase(yyruleno==291);
	138242	+ /* (292) typetoken ::= typename */ yytestcase(yyruleno==292);
	138243	+ /* (293) typename ::= ID\|STRING */ yytestcase(yyruleno==293);
	138244	+ /* (294) signed ::= plus_num (OPTIMIZED OUT) */ assert(yyruleno!=294);
	138245	+ /* (295) signed ::= minus_num (OPTIMIZED OUT) */ assert(yyruleno!=295);
	138246	+ /* (296) carglist ::= carglist ccons */ yytestcase(yyruleno==296);
	138247	+ /* (297) carglist ::= */ yytestcase(yyruleno==297);
	138248	+ /* (298) ccons ::= NULL onconf */ yytestcase(yyruleno==298);
	138249	+ /* (299) conslist_opt ::= COMMA conslist */ yytestcase(yyruleno==299);
	138250	+ /* (300) conslist ::= conslist tconscomma tcons */ yytestcase(yyruleno==300);
	138251	+ /* (301) conslist ::= tcons (OPTIMIZED OUT) */ assert(yyruleno!=301);
	138252	+ /* (302) tconscomma ::= */ yytestcase(yyruleno==302);
	138253	+ /* (303) defer_subclause_opt ::= defer_subclause (OPTIMIZED OUT) */ assert(yyruleno!=303);
	138254	+ /* (304) resolvetype ::= raisetype (OPTIMIZED OUT) */ assert(yyruleno!=304);
	138255	+ /* (305) selectnowith ::= oneselect (OPTIMIZED OUT) */ assert(yyruleno!=305);
	138256	+ /* (306) oneselect ::= values */ yytestcase(yyruleno==306);
	138257	+ /* (307) sclp ::= selcollist COMMA */ yytestcase(yyruleno==307);
	138258	+ /* (308) as ::= ID\|STRING */ yytestcase(yyruleno==308);
	138259	+ /* (309) expr ::= term (OPTIMIZED OUT) */ assert(yyruleno!=309);
	138260	+ /* (310) likeop ::= LIKE_KW\|MATCH */ yytestcase(yyruleno==310);
137807	138261	/* (311) exprlist ::= nexprlist */ yytestcase(yyruleno==311);
137808	138262	/* (312) nmnum ::= plus_num (OPTIMIZED OUT) */ assert(yyruleno!=312);
137809	138263	/* (313) nmnum ::= nm (OPTIMIZED OUT) */ assert(yyruleno!=313);
137810	138264	/* (314) nmnum ::= ON */ yytestcase(yyruleno==314);
137811	138265	/* (315) nmnum ::= DELETE */ yytestcase(yyruleno==315);
		@@ -138854,12 +139308,12 @@
138854	139308	** memory obtained from sqlite3_malloc() and to make *pzErrMsg point to that
138855	139309	** error message.
138856	139310	*/
138857	139311	SQLITE_PRIVATE int sqlite3RunParser(Parse pParse, const char zSql, char **pzErrMsg){
138858	139312	int nErr = 0; /* Number of errors encountered */
138859		- int i; /* Loop counter */
138860	139313	void pEngine; / The LEMON-generated LALR(1) parser */
	139314	+ int n = 0; /* Length of the next token token */
138861	139315	int tokenType; /* type of the next token */
138862	139316	int lastTokenParsed = -1; /* type of the previous token */
138863	139317	sqlite3 db = pParse->db; / The database connection */
138864	139318	int mxSqlLen; /* Max length of an SQL string */
138865	139319	#ifdef sqlite3Parser_ENGINEALWAYSONSTACK
		@@ -138871,11 +139325,10 @@
138871	139325	if( db->nVdbeActive==0 ){
138872	139326	db->u1.isInterrupted = 0;
138873	139327	}
138874	139328	pParse->rc = SQLITE_OK;
138875	139329	pParse->zTail = zSql;
138876		- i = 0;
138877	139330	assert( pzErrMsg!=0 );
138878	139331	/* sqlite3ParserTrace(stdout, "parser: "); */
138879	139332	#ifdef sqlite3Parser_ENGINEALWAYSONSTACK
138880	139333	pEngine = zSpace;
138881	139334	sqlite3ParserInit(pEngine);
		@@ -138889,16 +139342,14 @@
138889	139342	assert( pParse->pNewTable==0 );
138890	139343	assert( pParse->pNewTrigger==0 );
138891	139344	assert( pParse->nVar==0 );
138892	139345	assert( pParse->pVList==0 );
138893	139346	while( 1 ){
138894		- assert( i>=0 );
138895		- if( zSql[i]!=0 ){
138896		- pParse->sLastToken.z = &zSql[i];
138897		- pParse->sLastToken.n = sqlite3GetToken((u8*)&zSql[i],&tokenType);
138898		- i += pParse->sLastToken.n;
138899		- if( i>mxSqlLen ){
	139347	+ if( zSql[0]!=0 ){
	139348	+ n = sqlite3GetToken((u8*)zSql, &tokenType);
	139349	+ mxSqlLen -= n;
	139350	+ if( mxSqlLen<0 ){
138900	139351	pParse->rc = SQLITE_TOOBIG;
138901	139352	break;
138902	139353	}
138903	139354	}else{
138904	139355	/* Upon reaching the end of input, call the parser two more times
		@@ -138908,30 +139359,34 @@
138908	139359	}else if( lastTokenParsed==0 ){
138909	139360	break;
138910	139361	}else{
138911	139362	tokenType = TK_SEMI;
138912	139363	}
	139364	+ zSql -= n;
138913	139365	}
138914	139366	if( tokenType>=TK_SPACE ){
138915	139367	assert( tokenType==TK_SPACE \|\| tokenType==TK_ILLEGAL );
138916	139368	if( db->u1.isInterrupted ){
138917	139369	pParse->rc = SQLITE_INTERRUPT;
138918	139370	break;
138919	139371	}
138920	139372	if( tokenType==TK_ILLEGAL ){
138921		- sqlite3ErrorMsg(pParse, "unrecognized token: \"%T\"",
138922		- &pParse->sLastToken);
	139373	+ sqlite3ErrorMsg(pParse, "unrecognized token: \"%.*s\"", n, zSql);
138923	139374	break;
138924	139375	}
	139376	+ zSql += n;
138925	139377	}else{
	139378	+ pParse->sLastToken.z = zSql;
	139379	+ pParse->sLastToken.n = n;
138926	139380	sqlite3Parser(pEngine, tokenType, pParse->sLastToken, pParse);
138927	139381	lastTokenParsed = tokenType;
	139382	+ zSql += n;
138928	139383	if( pParse->rc!=SQLITE_OK \|\| db->mallocFailed ) break;
138929	139384	}
138930	139385	}
138931	139386	assert( nErr==0 );
138932		- pParse->zTail = &zSql[i];
	139387	+ pParse->zTail = zSql;
138933	139388	#ifdef YYTRACKMAXSTACKDEPTH
138934	139389	sqlite3_mutex_enter(sqlite3MallocMutex());
138935	139390	sqlite3StatusHighwater(SQLITE_STATUS_PARSER_STACK,
138936	139391	sqlite3ParserStackPeak(pEngine)
138937	139392	);
		@@ -140301,10 +140756,25 @@
140301	140756	return 0;
140302	140757	}
140303	140758	#endif
140304	140759	return db->lastRowid;
140305	140760	}
	140761	+
	140762	+/*
	140763	+** Set the value returned by the sqlite3_last_insert_rowid() API function.
	140764	+*/
	140765	+SQLITE_API void sqlite3_set_last_insert_rowid(sqlite3 *db, sqlite3_int64 iRowid){
	140766	+#ifdef SQLITE_ENABLE_API_ARMOR
	140767	+ if( !sqlite3SafetyCheckOk(db) ){
	140768	+ (void)SQLITE_MISUSE_BKPT;
	140769	+ return;
	140770	+ }
	140771	+#endif
	140772	+ sqlite3_mutex_enter(db->mutex);
	140773	+ db->lastRowid = iRowid;
	140774	+ sqlite3_mutex_leave(db->mutex);
	140775	+}
140306	140776
140307	140777	/*
140308	140778	** Return the number of changes in the most recent call to sqlite3_exec().
140309	140779	*/
140310	140780	SQLITE_API int sqlite3_changes(sqlite3 *db){
		@@ -148111,12 +148581,14 @@
148111	148581	** segments.
148112	148582	*/
148113	148583	const u32 nMinMerge = 64; /* Minimum amount of incr-merge work to do */
148114	148584
148115	148585	Fts3Table p = (Fts3Table)pVtab;
148116		- int rc = sqlite3Fts3PendingTermsFlush(p);
	148586	+ int rc;
	148587	+ i64 iLastRowid = sqlite3_last_insert_rowid(p->db);
148117	148588
	148589	+ rc = sqlite3Fts3PendingTermsFlush(p);
148118	148590	if( rc==SQLITE_OK
148119	148591	&& p->nLeafAdd>(nMinMerge/16)
148120	148592	&& p->nAutoincrmerge && p->nAutoincrmerge!=0xff
148121	148593	){
148122	148594	int mxLevel = 0; /* Maximum relative level value in db */
		@@ -148127,10 +148599,11 @@
148127	148599	A = p->nLeafAdd * mxLevel;
148128	148600	A += (A/2);
148129	148601	if( A>(int)nMinMerge ) rc = sqlite3Fts3Incrmerge(p, A, p->nAutoincrmerge);
148130	148602	}
148131	148603	sqlite3Fts3SegmentsClose(p);
	148604	+ sqlite3_set_last_insert_rowid(p->db, iLastRowid);
148132	148605	return rc;
148133	148606	}
148134	148607
148135	148608	/*
148136	148609	** If it is currently unknown whether or not the FTS table has an %_stat
		@@ -168462,10 +168935,11 @@
168462	168935	int nStep; /* Rows processed for current object */
168463	168936	int nProgress; /* Rows processed for all objects */
168464	168937	RbuObjIter objiter; /* Iterator for skipping through tbl/idx */
168465	168938	const char zVfsName; / Name of automatically created rbu vfs */
168466	168939	rbu_file pTargetFd; / File handle open on target db */
	168940	+ int nPagePerSector; /* Pages per sector for pTargetFd */
168467	168941	i64 iOalSz;
168468	168942	i64 nPhaseOneStep;
168469	168943
168470	168944	/* The following state variables are used as part of the incremental
168471	168945	** checkpoint stage (eStage==RBU_STAGE_CKPT). See comments surrounding
		@@ -170726,10 +171200,27 @@
170726	171200
170727	171201	if( p->rc==SQLITE_OK ){
170728	171202	if( p->nFrame==0 \|\| (pState && pState->iWalCksum!=p->iWalCksum) ){
170729	171203	p->rc = SQLITE_DONE;
170730	171204	p->eStage = RBU_STAGE_DONE;
	171205	+ }else{
	171206	+ int nSectorSize;
	171207	+ sqlite3_file *pDb = p->pTargetFd->pReal;
	171208	+ sqlite3_file *pWal = p->pTargetFd->pWalFd->pReal;
	171209	+ assert( p->nPagePerSector==0 );
	171210	+ nSectorSize = pDb->pMethods->xSectorSize(pDb);
	171211	+ if( nSectorSize>p->pgsz ){
	171212	+ p->nPagePerSector = nSectorSize / p->pgsz;
	171213	+ }else{
	171214	+ p->nPagePerSector = 1;
	171215	+ }
	171216	+
	171217	+ /* Call xSync() on the wal file. This causes SQLite to sync the
	171218	+ ** directory in which the target database and the wal file reside, in
	171219	+ ** case it has not been synced since the rename() call in
	171220	+ ** rbuMoveOalFile(). */
	171221	+ p->rc = pWal->pMethods->xSync(pWal, SQLITE_SYNC_NORMAL);
170731	171222	}
170732	171223	}
170733	171224	}
170734	171225
170735	171226	/*
		@@ -171381,13 +171872,30 @@
171381	171872	if( p->rc==SQLITE_OK ){
171382	171873	p->eStage = RBU_STAGE_DONE;
171383	171874	p->rc = SQLITE_DONE;
171384	171875	}
171385	171876	}else{
171386		- RbuFrame *pFrame = &p->aFrame[p->nStep];
171387		- rbuCheckpointFrame(p, pFrame);
171388		- p->nStep++;
	171877	+ /* At one point the following block copied a single frame from the
	171878	+ ** wal file to the database file. So that one call to sqlite3rbu_step()
	171879	+ ** checkpointed a single frame.
	171880	+ **
	171881	+ ** However, if the sector-size is larger than the page-size, and the
	171882	+ ** application calls sqlite3rbu_savestate() or close() immediately
	171883	+ ** after this step, then rbu_step() again, then a power failure occurs,
	171884	+ ** then the database page written here may be damaged. Work around
	171885	+ ** this by checkpointing frames until the next page in the aFrame[]
	171886	+ ** lies on a different disk sector to the current one. */
	171887	+ u32 iSector;
	171888	+ do{
	171889	+ RbuFrame *pFrame = &p->aFrame[p->nStep];
	171890	+ iSector = (pFrame->iDbPage-1) / p->nPagePerSector;
	171891	+ rbuCheckpointFrame(p, pFrame);
	171892	+ p->nStep++;
	171893	+ }while( p->nStep<p->nFrame
	171894	+ && iSector==((p->aFrame[p->nStep].iDbPage-1) / p->nPagePerSector)
	171895	+ && p->rc==SQLITE_OK
	171896	+ );
171389	171897	}
171390	171898	p->nProgress++;
171391	171899	}
171392	171900	break;
171393	171901	}
		@@ -171823,10 +172331,16 @@
171823	172331
171824	172332	/* Commit the transaction to the -oal file. /
171825	172333	if( p->rc==SQLITE_OK && p->eStage==RBU_STAGE_OAL ){
171826	172334	p->rc = sqlite3_exec(p->dbMain, "COMMIT", 0, 0, &p->zErrmsg);
171827	172335	}
	172336	+
	172337	+ /* Sync the db file if currently doing an incremental checkpoint */
	172338	+ if( p->rc==SQLITE_OK && p->eStage==RBU_STAGE_CKPT ){
	172339	+ sqlite3_file *pDb = p->pTargetFd->pReal;
	172340	+ p->rc = pDb->pMethods->xSync(pDb, SQLITE_SYNC_NORMAL);
	172341	+ }
171828	172342
171829	172343	rbuSaveState(p, p->eStage);
171830	172344
171831	172345	if( p->rc==SQLITE_OK && p->eStage==RBU_STAGE_OAL ){
171832	172346	p->rc = sqlite3_exec(p->dbRbu, "COMMIT", 0, 0, &p->zErrmsg);
		@@ -171947,10 +172461,16 @@
171947	172461	assert( p->eStage>=RBU_STAGE_OAL && p->eStage<=RBU_STAGE_DONE );
171948	172462	if( p->eStage==RBU_STAGE_OAL ){
171949	172463	assert( rc!=SQLITE_DONE );
171950	172464	if( rc==SQLITE_OK ) rc = sqlite3_exec(p->dbMain, "COMMIT", 0, 0, 0);
171951	172465	}
	172466	+
	172467	+ /* Sync the db file */
	172468	+ if( rc==SQLITE_OK && p->eStage==RBU_STAGE_CKPT ){
	172469	+ sqlite3_file *pDb = p->pTargetFd->pReal;
	172470	+ rc = pDb->pMethods->xSync(pDb, SQLITE_SYNC_NORMAL);
	172471	+ }
171952	172472
171953	172473	p->rc = rc;
171954	172474	rbuSaveState(p, p->eStage);
171955	172475	rc = p->rc;
171956	172476
		@@ -181079,11 +181599,13 @@
181079	181599	typedef unsigned short u16;
181080	181600	typedef short i16;
181081	181601	typedef sqlite3_int64 i64;
181082	181602	typedef sqlite3_uint64 u64;
181083	181603
181084		-#define ArraySize(x) ((int)(sizeof(x) / sizeof(x[0])))
	181604	+#ifndef ArraySize
	181605	+# define ArraySize(x) ((int)(sizeof(x) / sizeof(x[0])))
	181606	+#endif
181085	181607
181086	181608	#define testcase(x)
181087	181609	#define ALWAYS(x) 1
181088	181610	#define NEVER(x) 0
181089	181611
		@@ -186302,11 +186824,14 @@
186302	186824	if( p1->bEof==0 ){
186303	186825	if( (p1->iRowid==iLast)
186304	186826	\|\| (bFromValid && fts5RowidCmp(pExpr, p1->iRowid, iFrom)<0)
186305	186827	){
186306	186828	int rc = fts5ExprNodeNext(pExpr, p1, bFromValid, iFrom);
186307		- if( rc!=SQLITE_OK ) return rc;
	186829	+ if( rc!=SQLITE_OK ){
	186830	+ pNode->bNomatch = 0;
	186831	+ return rc;
	186832	+ }
186308	186833	}
186309	186834	}
186310	186835	}
186311	186836
186312	186837	fts5ExprNodeTest_OR(pExpr, pNode);
		@@ -186333,11 +186858,14 @@
186333	186858	Fts5ExprNode *pChild = pAnd->apChild[iChild];
186334	186859	int cmp = fts5RowidCmp(pExpr, iLast, pChild->iRowid);
186335	186860	if( cmp>0 ){
186336	186861	/* Advance pChild until it points to iLast or laster */
186337	186862	rc = fts5ExprNodeNext(pExpr, pChild, 1, iLast);
186338		- if( rc!=SQLITE_OK ) return rc;
	186863	+ if( rc!=SQLITE_OK ){
	186864	+ pAnd->bNomatch = 0;
	186865	+ return rc;
	186866	+ }
186339	186867	}
186340	186868
186341	186869	/* If the child node is now at EOF, so is the parent AND node. Otherwise,
186342	186870	** the child node is guaranteed to have advanced at least as far as
186343	186871	** rowid iLast. So if it is not at exactly iLast, pChild->iRowid is the
		@@ -186372,10 +186900,12 @@
186372	186900	i64 iFrom
186373	186901	){
186374	186902	int rc = fts5ExprNodeNext(pExpr, pNode->apChild[0], bFromValid, iFrom);
186375	186903	if( rc==SQLITE_OK ){
186376	186904	rc = fts5ExprNodeTest_AND(pExpr, pNode);
	186905	+ }else{
	186906	+ pNode->bNomatch = 0;
186377	186907	}
186378	186908	return rc;
186379	186909	}
186380	186910
186381	186911	static int fts5ExprNodeTest_NOT(
		@@ -186414,10 +186944,13 @@
186414	186944	){
186415	186945	int rc = fts5ExprNodeNext(pExpr, pNode->apChild[0], bFromValid, iFrom);
186416	186946	if( rc==SQLITE_OK ){
186417	186947	rc = fts5ExprNodeTest_NOT(pExpr, pNode);
186418	186948	}
	186949	+ if( rc!=SQLITE_OK ){
	186950	+ pNode->bNomatch = 0;
	186951	+ }
186419	186952	return rc;
186420	186953	}
186421	186954
186422	186955	/*
186423	186956	** If pNode currently points to a match, this function returns SQLITE_OK
		@@ -197534,11 +198067,11 @@
197534	198067	int nArg, /* Number of args */
197535	198068	sqlite3_value *apUnused / Function arguments */
197536	198069	){
197537	198070	assert( nArg==0 );
197538	198071	UNUSED_PARAM2(nArg, apUnused);
197539		- sqlite3_result_text(pCtx, "fts5: 2017-02-13 16:02:40 ada05cfa86ad7f5645450ac7a2a21c9aa6e57d2c", -1, SQLITE_TRANSIENT);
	198072	+ sqlite3_result_text(pCtx, "fts5: 2017-03-10 17:03:11 f8560c60d10c0365b33342ab05b5a953987b0471", -1, SQLITE_TRANSIENT);
197540	198073	}
197541	198074
197542	198075	static int fts5Init(sqlite3 *db){
197543	198076	static const sqlite3_module fts5Mod = {
197544	198077	/* iVersion */ 2,
		@@ -198197,15 +198730,10 @@
198197	198730	sqlite3_step(pDel);
198198	198731	rc = sqlite3_reset(pDel);
198199	198732	}
198200	198733	}
198201	198734
198202		- /* Write the averages record */
198203		- if( rc==SQLITE_OK ){
198204		- rc = fts5StorageSaveTotals(p);
198205		- }
198206		-
198207	198735	return rc;
198208	198736	}
198209	198737
198210	198738	/*
198211	198739	** Delete all entries in the FTS5 index.
		@@ -198405,15 +198933,10 @@
198405	198933	if( rc==SQLITE_OK ){
198406	198934	rc = fts5StorageInsertDocsize(p, iRowid, &buf);
198407	198935	}
198408	198936	sqlite3_free(buf.p);
198409	198937
198410		- /* Write the averages record */
198411		- if( rc==SQLITE_OK ){
198412		- rc = fts5StorageSaveTotals(p);
198413		- }
198414		-
198415	198938	return rc;
198416	198939	}
198417	198940
198418	198941	static int fts5StorageCount(Fts5Storage p, const char zSuffix, i64 *pnRow){
198419	198942	Fts5Config *pConfig = p->pConfig;
		@@ -198744,16 +199267,21 @@
198744	199267
198745	199268	/*
198746	199269	** Flush any data currently held in-memory to disk.
198747	199270	*/
198748	199271	static int sqlite3Fts5StorageSync(Fts5Storage *p, int bCommit){
198749		- if( bCommit && p->bTotalsValid ){
198750		- int rc = fts5StorageSaveTotals(p);
198751		- p->bTotalsValid = 0;
198752		- if( rc!=SQLITE_OK ) return rc;
	199272	+ int rc = SQLITE_OK;
	199273	+ i64 iLastRowid = sqlite3_last_insert_rowid(p->pConfig->db);
	199274	+ if( p->bTotalsValid ){
	199275	+ rc = fts5StorageSaveTotals(p);
	199276	+ if( bCommit ) p->bTotalsValid = 0;
198753	199277	}
198754		- return sqlite3Fts5IndexSync(p->pIndex, bCommit);
	199278	+ if( rc==SQLITE_OK ){
	199279	+ rc = sqlite3Fts5IndexSync(p->pIndex, bCommit);
	199280	+ }
	199281	+ sqlite3_set_last_insert_rowid(p->pConfig->db, iLastRowid);
	199282	+ return rc;
198755	199283	}
198756	199284
198757	199285	static int sqlite3Fts5StorageRollback(Fts5Storage *p){
198758	199286	p->bTotalsValid = 0;
198759	199287	return sqlite3Fts5IndexRollback(p->pIndex);
198760	199288

	--- src/sqlite3.c
	+++ src/sqlite3.c
	@@ -1,8 +1,8 @@
1	/******************************************************************************
2	** This file is an amalgamation of many separate C source files from SQLite
3	** version 3.17.0. By combining all the individual C code files into this
4	** single large file, the entire code can be compiled as a single translation
5	** unit. This allows many compilers to do optimizations that would not be
6	** possible if the files were compiled separately. Performance improvements
7	** of 5% or more are commonly seen when SQLite is compiled as a single
8	** translation unit.
	@@ -396,13 +396,13 @@
396	**
397	** See also: [sqlite3_libversion()],
398	** [sqlite3_libversion_number()], [sqlite3_sourceid()],
399	** [sqlite_version()] and [sqlite_source_id()].
400	*/
401	#define SQLITE_VERSION "3.17.0"
402	#define SQLITE_VERSION_NUMBER 3017000
403	#define SQLITE_SOURCE_ID "2017-02-13 16:02:40 ada05cfa86ad7f5645450ac7a2a21c9aa6e57d2c"
404
405	/*
406	** CAPI3REF: Run-Time Library Version Numbers
407	** KEYWORDS: sqlite3_version sqlite3_sourceid
408	**
	@@ -2315,24 +2315,34 @@
2315	** as an undeclared column named ROWID, OID, or _ROWID_ as long as those
2316	** names are not also used by explicitly declared columns. ^If
2317	** the table has a column of type [INTEGER PRIMARY KEY] then that column
2318	** is another alias for the rowid.
2319	**
2320	** ^The sqlite3_last_insert_rowid(D) interface returns the [rowid] of the
2321	** most recent successful [INSERT] into a rowid table or [virtual table]
2322	** on database connection D.
2323	** ^Inserts into [WITHOUT ROWID] tables are not recorded.
2324	** ^If no successful [INSERT]s into rowid tables
2325	** have ever occurred on the database connection D,
2326	** then sqlite3_last_insert_rowid(D) returns zero.
2327	**
2328	** ^(If an [INSERT] occurs within a trigger or within a [virtual table]
2329	** method, then this routine will return the [rowid] of the inserted
2330	** row as long as the trigger or virtual table method is running.
2331	** But once the trigger or virtual table method ends, the value returned
2332	** by this routine reverts to what it was before the trigger or virtual
2333	** table method began.)^











2334	**
2335	** ^An [INSERT] that fails due to a constraint violation is not a
2336	** successful [INSERT] and does not change the value returned by this
2337	** routine. ^Thus INSERT OR FAIL, INSERT OR IGNORE, INSERT OR ROLLBACK,
2338	** and INSERT OR ABORT make no changes to the return value of this
	@@ -2355,10 +2365,20 @@
2355	** unpredictable and might not equal either the old or the new
2356	** last insert [rowid].
2357	*/
2358	SQLITE_API sqlite3_int64 sqlite3_last_insert_rowid(sqlite3*);
2359










2360	/*
2361	** CAPI3REF: Count The Number Of Rows Modified
2362	** METHOD: sqlite3
2363	**
2364	** ^This function returns the number of rows modified, inserted or
	@@ -12457,10 +12477,11 @@
12457	SQLITE_PRIVATE const void sqlite3BtreePayloadFetch(BtCursor, u32 *pAmt);
12458	SQLITE_PRIVATE u32 sqlite3BtreePayloadSize(BtCursor*);
12459
12460	SQLITE_PRIVATE char sqlite3BtreeIntegrityCheck(Btree, int aRoot, int nRoot, int, int);
12461	SQLITE_PRIVATE struct Pager sqlite3BtreePager(Btree);

12462
12463	#ifndef SQLITE_OMIT_INCRBLOB
12464	SQLITE_PRIVATE int sqlite3BtreePayloadChecked(BtCursor, u32 offset, u32 amt, void);
12465	SQLITE_PRIVATE int sqlite3BtreePutData(BtCursor, u32 offset, u32 amt, void);
12466	SQLITE_PRIVATE void sqlite3BtreeIncrblobCursor(BtCursor *);
	@@ -12757,117 +12778,119 @@
12757	#define OP_Divide 50 /* same as TK_SLASH, synopsis: r[P3]=r[P2]/r[P1] */
12758	#define OP_Remainder 51 /* same as TK_REM, synopsis: r[P3]=r[P2]%r[P1] */
12759	#define OP_Concat 52 /* same as TK_CONCAT, synopsis: r[P3]=r[P2]+r[P1] */
12760	#define OP_Last 53
12761	#define OP_BitNot 54 /* same as TK_BITNOT, synopsis: r[P1]= ~r[P1] */
12762	#define OP_SorterSort 55
12763	#define OP_Sort 56
12764	#define OP_Rewind 57
12765	#define OP_IdxLE 58 /* synopsis: key=r[P3@P4] */
12766	#define OP_IdxGT 59 /* synopsis: key=r[P3@P4] */
12767	#define OP_IdxLT 60 /* synopsis: key=r[P3@P4] */
12768	#define OP_IdxGE 61 /* synopsis: key=r[P3@P4] */
12769	#define OP_RowSetRead 62 /* synopsis: r[P3]=rowset(P1) */
12770	#define OP_RowSetTest 63 /* synopsis: if r[P3] in rowset(P1) goto P2 */
12771	#define OP_Program 64
12772	#define OP_FkIfZero 65 /* synopsis: if fkctr[P1]==0 goto P2 */
12773	#define OP_IfPos 66 /* synopsis: if r[P1]>0 then r[P1]-=P3, goto P2 */
12774	#define OP_IfNotZero 67 /* synopsis: if r[P1]!=0 then r[P1]--, goto P2 */
12775	#define OP_DecrJumpZero 68 /* synopsis: if (--r[P1])==0 goto P2 */
12776	#define OP_IncrVacuum 69
12777	#define OP_VNext 70
12778	#define OP_Init 71 /* synopsis: Start at P2 */
12779	#define OP_Return 72
12780	#define OP_EndCoroutine 73
12781	#define OP_HaltIfNull 74 /* synopsis: if r[P3]=null halt */
12782	#define OP_Halt 75
12783	#define OP_Integer 76 /* synopsis: r[P2]=P1 */
12784	#define OP_Int64 77 /* synopsis: r[P2]=P4 */
12785	#define OP_String 78 /* synopsis: r[P2]='P4' (len=P1) */
12786	#define OP_Null 79 /* synopsis: r[P2..P3]=NULL */
12787	#define OP_SoftNull 80 /* synopsis: r[P1]=NULL */
12788	#define OP_Blob 81 /* synopsis: r[P2]=P4 (len=P1) */
12789	#define OP_Variable 82 /* synopsis: r[P2]=parameter(P1,P4) */
12790	#define OP_Move 83 /* synopsis: r[P2@P3]=r[P1@P3] */
12791	#define OP_Copy 84 /* synopsis: r[P2@P3+1]=r[P1@P3+1] */
12792	#define OP_SCopy 85 /* synopsis: r[P2]=r[P1] */
12793	#define OP_IntCopy 86 /* synopsis: r[P2]=r[P1] */
12794	#define OP_ResultRow 87 /* synopsis: output=r[P1@P2] */
12795	#define OP_CollSeq 88
12796	#define OP_Function0 89 /* synopsis: r[P3]=func(r[P2@P5]) */
12797	#define OP_Function 90 /* synopsis: r[P3]=func(r[P2@P5]) */
12798	#define OP_AddImm 91 /* synopsis: r[P1]=r[P1]+P2 */
12799	#define OP_RealAffinity 92
12800	#define OP_Cast 93 /* synopsis: affinity(r[P1]) */
12801	#define OP_Permutation 94
12802	#define OP_Compare 95 /* synopsis: r[P1@P3] <-> r[P2@P3] */
12803	#define OP_Column 96 /* synopsis: r[P3]=PX */
12804	#define OP_String8 97 /* same as TK_STRING, synopsis: r[P2]='P4' */
12805	#define OP_Affinity 98 /* synopsis: affinity(r[P1@P2]) */
12806	#define OP_MakeRecord 99 /* synopsis: r[P3]=mkrec(r[P1@P2]) */
12807	#define OP_Count 100 /* synopsis: r[P2]=count() */
12808	#define OP_ReadCookie 101
12809	#define OP_SetCookie 102
12810	#define OP_ReopenIdx 103 /* synopsis: root=P2 iDb=P3 */
12811	#define OP_OpenRead 104 /* synopsis: root=P2 iDb=P3 */
12812	#define OP_OpenWrite 105 /* synopsis: root=P2 iDb=P3 */
12813	#define OP_OpenAutoindex 106 /* synopsis: nColumn=P2 */
12814	#define OP_OpenEphemeral 107 /* synopsis: nColumn=P2 */
12815	#define OP_SorterOpen 108
12816	#define OP_SequenceTest 109 /* synopsis: if( cursor[P1].ctr++ ) pc = P2 */
12817	#define OP_OpenPseudo 110 /* synopsis: P3 columns in r[P2] */
12818	#define OP_Close 111
12819	#define OP_ColumnsUsed 112
12820	#define OP_Sequence 113 /* synopsis: r[P2]=cursor[P1].ctr++ */
12821	#define OP_NewRowid 114 /* synopsis: r[P2]=rowid */
12822	#define OP_Insert 115 /* synopsis: intkey=r[P3] data=r[P2] */
12823	#define OP_InsertInt 116 /* synopsis: intkey=P3 data=r[P2] */
12824	#define OP_Delete 117
12825	#define OP_ResetCount 118
12826	#define OP_SorterCompare 119 /* synopsis: if key(P1)!=trim(r[P3],P4) goto P2 */
12827	#define OP_SorterData 120 /* synopsis: r[P2]=data */
12828	#define OP_RowData 121 /* synopsis: r[P2]=data */
12829	#define OP_Rowid 122 /* synopsis: r[P2]=rowid */
12830	#define OP_NullRow 123
12831	#define OP_SorterInsert 124 /* synopsis: key=r[P2] */
12832	#define OP_IdxInsert 125 /* synopsis: key=r[P2] */
12833	#define OP_IdxDelete 126 /* synopsis: key=r[P2@P3] */
12834	#define OP_Seek 127 /* synopsis: Move P3 to P1.rowid */
12835	#define OP_IdxRowid 128 /* synopsis: r[P2]=rowid */
12836	#define OP_Destroy 129
12837	#define OP_Clear 130
12838	#define OP_ResetSorter 131
12839	#define OP_Real 132 /* same as TK_FLOAT, synopsis: r[P2]=P4 */
12840	#define OP_CreateIndex 133 /* synopsis: r[P2]=root iDb=P1 */
12841	#define OP_CreateTable 134 /* synopsis: r[P2]=root iDb=P1 */
12842	#define OP_ParseSchema 135
12843	#define OP_LoadAnalysis 136
12844	#define OP_DropTable 137
12845	#define OP_DropIndex 138
12846	#define OP_DropTrigger 139
12847	#define OP_IntegrityCk 140
12848	#define OP_RowSetAdd 141 /* synopsis: rowset(P1)=r[P2] */
12849	#define OP_Param 142
12850	#define OP_FkCounter 143 /* synopsis: fkctr[P1]+=P2 */
12851	#define OP_MemMax 144 /* synopsis: r[P1]=max(r[P1],r[P2]) */
12852	#define OP_OffsetLimit 145 /* synopsis: if r[P1]>0 then r[P2]=r[P1]+max(0,r[P3]) else r[P2]=(-1) */
12853	#define OP_AggStep0 146 /* synopsis: accum=r[P3] step(r[P2@P5]) */
12854	#define OP_AggStep 147 /* synopsis: accum=r[P3] step(r[P2@P5]) */
12855	#define OP_AggFinal 148 /* synopsis: accum=r[P1] N=P2 */
12856	#define OP_Expire 149
12857	#define OP_TableLock 150 /* synopsis: iDb=P1 root=P2 write=P3 */
12858	#define OP_VBegin 151
12859	#define OP_VCreate 152
12860	#define OP_VDestroy 153
12861	#define OP_VOpen 154
12862	#define OP_VColumn 155 /* synopsis: r[P3]=vcolumn(P2) */
12863	#define OP_VRename 156
12864	#define OP_Pagecount 157
12865	#define OP_MaxPgcnt 158
12866	#define OP_CursorHint 159
12867	#define OP_Noop 160
12868	#define OP_Explain 161


12869
12870	/* Properties such as "out2" or "jump" that are specified in
12871	** comments following the "case" for each opcode in the vdbe.c
12872	** are encoded into bitvectors as follows:
12873	*/
	@@ -12883,32 +12906,32 @@
12883	/* 16 */ 0x03, 0x03, 0x01, 0x12, 0x01, 0x03, 0x03, 0x09,\
12884	/* 24 */ 0x09, 0x09, 0x09, 0x26, 0x26, 0x09, 0x09, 0x09,\
12885	/* 32 */ 0x09, 0x09, 0x03, 0x03, 0x0b, 0x0b, 0x0b, 0x0b,\
12886	/* 40 */ 0x0b, 0x0b, 0x01, 0x26, 0x26, 0x26, 0x26, 0x26,\
12887	/* 48 */ 0x26, 0x26, 0x26, 0x26, 0x26, 0x01, 0x12, 0x01,\
12888	/* 56 */ 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x23, 0x0b,\
12889	/* 64 */ 0x01, 0x01, 0x03, 0x03, 0x03, 0x01, 0x01, 0x01,\
12890	/* 72 */ 0x02, 0x02, 0x08, 0x00, 0x10, 0x10, 0x10, 0x10,\
12891	/* 80 */ 0x00, 0x10, 0x10, 0x00, 0x00, 0x10, 0x10, 0x00,\
12892	/* 88 */ 0x00, 0x00, 0x00, 0x02, 0x02, 0x02, 0x00, 0x00,\
12893	/* 96 */ 0x00, 0x10, 0x00, 0x00, 0x10, 0x10, 0x00, 0x00,\
12894	/* 104 */ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,\
12895	/* 112 */ 0x00, 0x10, 0x10, 0x00, 0x00, 0x00, 0x00, 0x00,\
12896	/* 120 */ 0x00, 0x00, 0x10, 0x00, 0x04, 0x04, 0x00, 0x00,\
12897	/* 128 */ 0x10, 0x10, 0x00, 0x00, 0x10, 0x10, 0x10, 0x00,\
12898	/* 136 */ 0x00, 0x00, 0x00, 0x00, 0x00, 0x06, 0x10, 0x00,\
12899	/* 144 */ 0x04, 0x1a, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,\
12900	/* 152 */ 0x00, 0x00, 0x00, 0x00, 0x00, 0x10, 0x10, 0x00,\
12901	/* 160 */ 0x00, 0x00,}
12902
12903	/* The sqlite3P2Values() routine is able to run faster if it knows
12904	** the value of the largest JUMP opcode. The smaller the maximum
12905	** JUMP opcode the better, so the mkopcodeh.tcl script that
12906	** generated this include file strives to group all JUMP opcodes
12907	** together near the beginning of the list.
12908	*/
12909	#define SQLITE_MX_JUMP_OPCODE 71 /* Maximum JUMP opcode */
12910
12911	/************ End of opcodes.h *******************************************/
12912	/************ Continuing where we left off in vdbe.h *********************/
12913
12914	/*
	@@ -14074,10 +14097,11 @@
14074	u8 suppressErr; /* Do not issue error messages if true */
14075	u8 vtabOnConflict; /* Value to return for s3_vtab_on_conflict() */
14076	u8 isTransactionSavepoint; /* True if the outermost savepoint is a TS */
14077	u8 mTrace; /* zero or more SQLITE_TRACE flags */
14078	u8 skipBtreeMutex; /* True if no shared-cache backends */

14079	int nextPagesize; /* Pagesize after VACUUM if >0 */
14080	u32 magic; /* Magic number for detect library misuse */
14081	int nChange; /* Value returned by sqlite3_changes() */
14082	int nTotalChange; /* Value returned by sqlite3_total_changes() */
14083	int aLimit[SQLITE_N_LIMIT]; /* Limits */
	@@ -14589,18 +14613,18 @@
14589	char zColAff; / String defining the affinity of each column */
14590	ExprList pCheck; / All CHECK constraints */
14591	/* ... also used as column name list in a VIEW */
14592	int tnum; /* Root BTree page for this table */
14593	u32 nTabRef; /* Number of pointers to this Table */

14594	i16 iPKey; /* If not negative, use aCol[iPKey] as the rowid */
14595	i16 nCol; /* Number of columns in this table */
14596	LogEst nRowLogEst; /* Estimated rows in table - from sqlite_stat1 table */
14597	LogEst szTabRow; /* Estimated size of each table row in bytes */
14598	#ifdef SQLITE_ENABLE_COSTMULT
14599	LogEst costMult; /* Cost multiplier for using this table */
14600	#endif
14601	u8 tabFlags; /* Mask of TF_* values */
14602	u8 keyConf; /* What to do in case of uniqueness conflict on iPKey */
14603	#ifndef SQLITE_OMIT_ALTERTABLE
14604	int addColOffset; /* Offset in CREATE TABLE stmt to add a new column */
14605	#endif
14606	#ifndef SQLITE_OMIT_VIRTUALTABLE
	@@ -14620,27 +14644,29 @@
14620	** followed by non-hidden columns. Example: "CREATE VIRTUAL TABLE x USING
14621	** vtab1(a HIDDEN, b);". Since "b" is a non-hidden column but "a" is hidden,
14622	** the TF_OOOHidden attribute would apply in this case. Such tables require
14623	** special handling during INSERT processing.
14624	*/
14625	#define TF_Readonly 0x01 /* Read-only system table */
14626	#define TF_Ephemeral 0x02 /* An ephemeral table */
14627	#define TF_HasPrimaryKey 0x04 /* Table has a primary key */
14628	#define TF_Autoincrement 0x08 /* Integer primary key is autoincrement */
14629	#define TF_Virtual 0x10 /* Is a virtual table */
14630	#define TF_WithoutRowid 0x20 /* No rowid. PRIMARY KEY is the key */
14631	#define TF_NoVisibleRowid 0x40 /* No user-visible "rowid" column */
14632	#define TF_OOOHidden 0x80 /* Out-of-Order hidden columns */
14633


14634
14635	/*
14636	** Test to see whether or not a table is a virtual table. This is
14637	** done as a macro so that it will be optimized out when virtual
14638	** table support is omitted from the build.
14639	*/
14640	#ifndef SQLITE_OMIT_VIRTUALTABLE
14641	# define IsVirtual(X) (((X)->tabFlags & TF_Virtual)!=0)
14642	#else
14643	# define IsVirtual(X) 0
14644	#endif
14645
14646	/*
	@@ -14871,10 +14897,11 @@
14871	unsigned bUnordered:1; /* Use this index for == or IN queries only */
14872	unsigned uniqNotNull:1; /* True if UNIQUE and NOT NULL for all columns */
14873	unsigned isResized:1; /* True if resizeIndexObject() has been called */
14874	unsigned isCovering:1; /* True if this is a covering index */
14875	unsigned noSkipScan:1; /* Do not try to use skip-scan if true */

14876	#ifdef SQLITE_ENABLE_STAT3_OR_STAT4
14877	int nSample; /* Number of elements in aSample[] */
14878	int nSampleCol; /* Size of IndexSample.anEq[] and so on */
14879	tRowcnt aAvgEq; / Average nEq values for keys not in aSample */
14880	IndexSample aSample; / Samples of the left-most key */
	@@ -15181,11 +15208,11 @@
15181	** form is used for name resolution with nested FROM clauses.
15182	*/
15183	struct ExprList {
15184	int nExpr; /* Number of expressions on the list */
15185	struct ExprList_item { /* For each expression in the list */
15186	Expr pExpr; / The list of expressions */
15187	char zName; / Token associated with this expression */
15188	char zSpan; / Original text of the expression */
15189	u8 sortOrder; /* 1 for DESC or 0 for ASC */
15190	unsigned done :1; /* A flag to indicate when processing is finished */
15191	unsigned bSpanIsTab :1; /* zSpan holds DB.TABLE.COLUMN */
	@@ -16504,10 +16531,11 @@
16504	SQLITE_PRIVATE void sqlite3UnlinkAndDeleteIndex(sqlite3,int,const char);
16505	SQLITE_PRIVATE void sqlite3Vacuum(Parse,Token);
16506	SQLITE_PRIVATE int sqlite3RunVacuum(char*, sqlite3, int);
16507	SQLITE_PRIVATE char sqlite3NameFromToken(sqlite3, Token*);
16508	SQLITE_PRIVATE int sqlite3ExprCompare(Expr, Expr, int);

16509	SQLITE_PRIVATE int sqlite3ExprListCompare(ExprList, ExprList, int);
16510	SQLITE_PRIVATE int sqlite3ExprImpliesExpr(Expr, Expr, int);
16511	SQLITE_PRIVATE void sqlite3ExprAnalyzeAggregates(NameContext, Expr);
16512	SQLITE_PRIVATE void sqlite3ExprAnalyzeAggList(NameContext,ExprList);
16513	SQLITE_PRIVATE int sqlite3ExprCoveredByIndex(Expr, int iCur, Index pIdx);
	@@ -17450,11 +17478,11 @@
17450	"COMPILER=gcc-" __VERSION__,
17451	#endif
17452	#if SQLITE_COVERAGE_TEST
17453	"COVERAGE_TEST",
17454	#endif
17455	#if SQLITE_DEBUG
17456	"DEBUG",
17457	#endif
17458	#if SQLITE_DEFAULT_LOCKING_MODE
17459	"DEFAULT_LOCKING_MODE=" CTIMEOPT_VAL(SQLITE_DEFAULT_LOCKING_MODE),
17460	#endif
	@@ -19505,22 +19533,23 @@
19505	**
19506	** Move the date backwards to the beginning of the current day,
19507	** or month or year.
19508	*/
19509	if( sqlite3_strnicmp(z, "start of ", 9)!=0 ) break;

19510	z += 9;
19511	computeYMD(p);
19512	p->validHMS = 1;
19513	p->h = p->m = 0;
19514	p->s = 0.0;

19515	p->validTZ = 0;
19516	p->validJD = 0;
19517	if( sqlite3_stricmp(z,"month")==0 ){
19518	p->D = 1;
19519	rc = 0;
19520	}else if( sqlite3_stricmp(z,"year")==0 ){
19521	computeYMD(p);
19522	p->M = 1;
19523	p->D = 1;
19524	rc = 0;
19525	}else if( sqlite3_stricmp(z,"day")==0 ){
19526	rc = 0;
	@@ -24050,12 +24079,12 @@
24050	#ifdef SQLITE_DEBUG
24051	assert( p->nRef>0 \|\| p->owner==0 );
24052	p->owner = tid;
24053	p->nRef++;
24054	if( p->trace ){
24055	OSTRACE(("ENTER-MUTEX tid=%lu, mutex=%p (%d), nRef=%d\n",
24056	tid, p, p->trace, p->nRef));
24057	}
24058	#endif
24059	}
24060
24061	static int winMutexTry(sqlite3_mutex *p){
	@@ -24093,12 +24122,12 @@
24093	#else
24094	UNUSED_PARAMETER(p);
24095	#endif
24096	#ifdef SQLITE_DEBUG
24097	if( p->trace ){
24098	OSTRACE(("TRY-MUTEX tid=%lu, mutex=%p (%d), owner=%lu, nRef=%d, rc=%s\n",
24099	tid, p, p->trace, p->owner, p->nRef, sqlite3ErrName(rc)));
24100	}
24101	#endif
24102	return rc;
24103	}
24104
	@@ -24122,12 +24151,12 @@
24122	#endif
24123	assert( winMutex_isInit==1 );
24124	LeaveCriticalSection(&p->mutex);
24125	#ifdef SQLITE_DEBUG
24126	if( p->trace ){
24127	OSTRACE(("LEAVE-MUTEX tid=%lu, mutex=%p (%d), nRef=%d\n",
24128	tid, p, p->trace, p->nRef));
24129	}
24130	#endif
24131	}
24132
24133	SQLITE_PRIVATE sqlite3_mutex_methods const *sqlite3DefaultMutex(void){
	@@ -24383,10 +24412,17 @@
24383	** implementation of malloc_good_size(), which must be called in debug
24384	** mode and specifically when the DMD "Dark Matter Detector" is enabled
24385	** or else a crash results. Hence, do not attempt to optimize out the
24386	** following xRoundup() call. */
24387	nFull = sqlite3GlobalConfig.m.xRoundup(n);







24388
24389	sqlite3StatusHighwater(SQLITE_STATUS_MALLOC_SIZE, n);
24390	if( mem0.alarmThreshold>0 ){
24391	sqlite3_int64 nUsed = sqlite3StatusValue(SQLITE_STATUS_MEMORY_USED);
24392	if( nUsed >= mem0.alarmThreshold - nFull ){
	@@ -24572,11 +24608,11 @@
24572	return sqlite3GlobalConfig.m.xSize(p);
24573	}
24574	SQLITE_PRIVATE int sqlite3DbMallocSize(sqlite3 db, void p){
24575	assert( p!=0 );
24576	if( db==0 \|\| !isLookaside(db,p) ){
24577	#if SQLITE_DEBUG
24578	if( db==0 ){
24579	assert( sqlite3MemdebugNoType(p, (u8)~MEMTYPE_HEAP) );
24580	assert( sqlite3MemdebugHasType(p, MEMTYPE_HEAP) );
24581	}else{
24582	assert( sqlite3MemdebugHasType(p, (MEMTYPE_LOOKASIDE\|MEMTYPE_HEAP)) );
	@@ -24633,11 +24669,11 @@
24633	measureAllocationSize(db, p);
24634	return;
24635	}
24636	if( isLookaside(db, p) ){
24637	LookasideSlot pBuf = (LookasideSlot)p;
24638	#if SQLITE_DEBUG
24639	/* Trash all content in the buffer being freed */
24640	memset(p, 0xaa, db->lookaside.sz);
24641	#endif
24642	pBuf->pNext = db->lookaside.pFree;
24643	db->lookaside.pFree = pBuf;
	@@ -25002,11 +25038,11 @@
25002
25003	/*
25004	** Conversion types fall into various categories as defined by the
25005	** following enumeration.
25006	*/
25007	#define etRADIX 0 /* Integer types. %d, %x, %o, and so forth */
25008	#define etFLOAT 1 /* Floating point. %f */
25009	#define etEXP 2 /* Exponentional notation. %e and %E */
25010	#define etGENERIC 3 /* Floating or exponential, depending on exponent. %g */
25011	#define etSIZE 4 /* Return number of characters processed so far. %n */
25012	#define etSTRING 5 /* Strings. %s */
	@@ -25020,12 +25056,13 @@
25020	#define etTOKEN 11 /* a pointer to a Token structure */
25021	#define etSRCLIST 12 /* a pointer to a SrcList */
25022	#define etPOINTER 13 /* The %p conversion */
25023	#define etSQLESCAPE3 14 /* %w -> Strings with '\"' doubled */
25024	#define etORDINAL 15 /* %r -> 1st, 2nd, 3rd, 4th, etc. English only */

25025
25026	#define etINVALID 16 /* Any unrecognized conversion type */
25027
25028
25029	/*
25030	** An "etByte" is an 8-bit unsigned value.
25031	*/
	@@ -25045,40 +25082,40 @@
25045	} et_info;
25046
25047	/*
25048	** Allowed values for et_info.flags
25049	*/
25050	#define FLAG_SIGNED 1 /* True if the value to convert is signed */
25051	#define FLAG_STRING 4 /* Allow infinity precision */
25052
25053
25054	/*
25055	** The following table is searched linearly, so it is good to put the
25056	** most frequently used conversion types first.
25057	*/
25058	static const char aDigits[] = "0123456789ABCDEF0123456789abcdef";
25059	static const char aPrefix[] = "-x0\000X0";
25060	static const et_info fmtinfo[] = {
25061	{ 'd', 10, 1, etRADIX, 0, 0 },
25062	{ 's', 0, 4, etSTRING, 0, 0 },
25063	{ 'g', 0, 1, etGENERIC, 30, 0 },
25064	{ 'z', 0, 4, etDYNSTRING, 0, 0 },
25065	{ 'q', 0, 4, etSQLESCAPE, 0, 0 },
25066	{ 'Q', 0, 4, etSQLESCAPE2, 0, 0 },
25067	{ 'w', 0, 4, etSQLESCAPE3, 0, 0 },
25068	{ 'c', 0, 0, etCHARX, 0, 0 },
25069	{ 'o', 8, 0, etRADIX, 0, 2 },
25070	{ 'u', 10, 0, etRADIX, 0, 0 },
25071	{ 'x', 16, 0, etRADIX, 16, 1 },
25072	{ 'X', 16, 0, etRADIX, 0, 4 },
25073	#ifndef SQLITE_OMIT_FLOATING_POINT
25074	{ 'f', 0, 1, etFLOAT, 0, 0 },
25075	{ 'e', 0, 1, etEXP, 30, 0 },
25076	{ 'E', 0, 1, etEXP, 14, 0 },
25077	{ 'G', 0, 1, etGENERIC, 14, 0 },
25078	#endif
25079	{ 'i', 10, 1, etRADIX, 0, 0 },
25080	{ 'n', 0, 0, etSIZE, 0, 0 },
25081	{ '%', 0, 0, etPERCENT, 0, 0 },
25082	{ 'p', 16, 0, etPOINTER, 0, 1 },
25083
25084	/* All the rest are undocumented and are for internal use only */
	@@ -25166,18 +25203,17 @@
25166	int precision; /* Precision of the current field */
25167	int length; /* Length of the field */
25168	int idx; /* A general purpose loop counter */
25169	int width; /* Width of the current field */
25170	etByte flag_leftjustify; /* True if "-" flag is present */
25171	etByte flag_plussign; /* True if "+" flag is present */
25172	etByte flag_blanksign; /* True if " " flag is present */
25173	etByte flag_alternateform; /* True if "#" flag is present */
25174	etByte flag_altform2; /* True if "!" flag is present */
25175	etByte flag_zeropad; /* True if field width constant starts with zero */
25176	etByte flag_long; /* True if "l" flag is present */
25177	etByte flag_longlong; /* True if the "ll" flag is present */
25178	etByte done; /* Loop termination flag */

25179	etByte xtype = etINVALID; /* Conversion paradigm */
25180	u8 bArgList; /* True for SQLITE_PRINTF_SQLFUNC */
25181	char prefix; /* Prefix character. "+" or "-" or " " or '\0'. */
25182	sqlite_uint64 longvalue; /* Value for integer types */
25183	LONGDOUBLE_TYPE realvalue; /* Value for real types */
	@@ -25216,21 +25252,22 @@
25216	if( (c=(*++fmt))==0 ){
25217	sqlite3StrAccumAppend(pAccum, "%", 1);
25218	break;
25219	}
25220	/* Find out what flags are present */
25221	flag_leftjustify = flag_plussign = flag_blanksign =
25222	flag_alternateform = flag_altform2 = flag_zeropad = 0;
25223	done = 0;
25224	do{
25225	switch( c ){
25226	case '-': flag_leftjustify = 1; break;
25227	case '+': flag_plussign = 1; break;
25228	case ' ': flag_blanksign = 1; break;
25229	case '#': flag_alternateform = 1; break;
25230	case '!': flag_altform2 = 1; break;
25231	case '0': flag_zeropad = 1; break;

25232	default: done = 1; break;
25233	}
25234	}while( !done && (c=(*++fmt))!=0 );
25235	/* Get the field width */
25236	if( c=='*' ){
	@@ -25296,17 +25333,15 @@
25296	/* Get the conversion type modifier */
25297	if( c=='l' ){
25298	flag_long = 1;
25299	c = *++fmt;
25300	if( c=='l' ){
25301	flag_longlong = 1;
25302	c = *++fmt;
25303	}else{
25304	flag_longlong = 0;
25305	}
25306	}else{
25307	flag_long = flag_longlong = 0;
25308	}
25309	/* Fetch the info entry for the field */
25310	infop = &fmtinfo[0];
25311	xtype = etINVALID;
25312	for(idx=0; idx<ArraySize(fmtinfo); idx++){
	@@ -25320,41 +25355,42 @@
25320	/*
25321	** At this point, variables are initialized as follows:
25322	**
25323	** flag_alternateform TRUE if a '#' is present.
25324	** flag_altform2 TRUE if a '!' is present.
25325	** flag_plussign TRUE if a '+' is present.
25326	** flag_leftjustify TRUE if a '-' is present or if the
25327	** field width was negative.
25328	** flag_zeropad TRUE if the width began with 0.
25329	** flag_long TRUE if the letter 'l' (ell) prefixed
25330	** the conversion character.
25331	** flag_longlong TRUE if the letter 'll' (ell ell) prefixed
25332	** the conversion character.
25333	** flag_blanksign TRUE if a ' ' is present.
25334	** width The specified field width. This is
25335	** always non-negative. Zero is the default.
25336	** precision The specified precision. The default
25337	** is -1.
25338	** xtype The class of the conversion.
25339	** infop Pointer to the appropriate info struct.
25340	*/
25341	switch( xtype ){
25342	case etPOINTER:
25343	flag_longlong = sizeof(char*)==sizeof(i64);
25344	flag_long = sizeof(char*)==sizeof(long int);
25345	/* Fall through into the next case */
25346	case etORDINAL:
25347	case etRADIX:



25348	if( infop->flags & FLAG_SIGNED ){
25349	i64 v;
25350	if( bArgList ){
25351	v = getIntArg(pArgList);
25352	}else if( flag_longlong ){
25353	v = va_arg(ap,i64);
25354	}else if( flag_long ){
25355	v = va_arg(ap,long int);




25356	}else{
25357	v = va_arg(ap,int);
25358	}
25359	if( v<0 ){
25360	if( v==SMALLEST_INT64 ){
	@@ -25363,35 +25399,35 @@
25363	longvalue = -v;
25364	}
25365	prefix = '-';
25366	}else{
25367	longvalue = v;
25368	if( flag_plussign ) prefix = '+';
25369	else if( flag_blanksign ) prefix = ' ';
25370	else prefix = 0;
25371	}
25372	}else{
25373	if( bArgList ){
25374	longvalue = (u64)getIntArg(pArgList);
25375	}else if( flag_longlong ){
25376	longvalue = va_arg(ap,u64);
25377	}else if( flag_long ){
25378	longvalue = va_arg(ap,unsigned long int);




25379	}else{
25380	longvalue = va_arg(ap,unsigned int);
25381	}
25382	prefix = 0;
25383	}
25384	if( longvalue==0 ) flag_alternateform = 0;
25385	if( flag_zeropad && precision<width-(prefix!=0) ){
25386	precision = width-(prefix!=0);
25387	}
25388	if( precision<etBUFSIZE-10 ){
25389	nOut = etBUFSIZE;
25390	zOut = buf;
25391	}else{
25392	nOut = precision + 10;
25393	zOut = zExtra = sqlite3Malloc( nOut );
25394	if( zOut==0 ){
25395	setStrAccumError(pAccum, STRACCUM_NOMEM);
25396	return;
25397	}
	@@ -25413,12 +25449,27 @@
25413	*(--bufpt) = cset[longvalue%base];
25414	longvalue = longvalue/base;
25415	}while( longvalue>0 );
25416	}
25417	length = (int)(&zOut[nOut-1]-bufpt);
25418	for(idx=precision-length; idx>0; idx--){
25419	(--bufpt) = '0'; / Zero pad */















25420	}
25421	if( prefix ) (--bufpt) = prefix; / Add sign */
25422	if( flag_alternateform && infop->prefix ){ /* Add "0" or "0x" */
25423	const char *pre;
25424	char x;
	@@ -25441,13 +25492,11 @@
25441	if( precision<0 ) precision = 6; /* Set default precision */
25442	if( realvalue<0.0 ){
25443	realvalue = -realvalue;
25444	prefix = '-';
25445	}else{
25446	if( flag_plussign ) prefix = '+';
25447	else if( flag_blanksign ) prefix = ' ';
25448	else prefix = 0;
25449	}
25450	if( xtype==etGENERIC && precision>0 ) precision--;
25451	testcase( precision>0xfff );
25452	for(idx=precision&0xfff, rounder=0.5; idx>0; idx--, rounder*=0.1){}
25453	if( xtype==etFLOAT ) realvalue += rounder;
	@@ -29464,117 +29513,119 @@
29464	/* 50 */ "Divide" OpHelp("r[P3]=r[P2]/r[P1]"),
29465	/* 51 */ "Remainder" OpHelp("r[P3]=r[P2]%r[P1]"),
29466	/* 52 */ "Concat" OpHelp("r[P3]=r[P2]+r[P1]"),
29467	/* 53 */ "Last" OpHelp(""),
29468	/* 54 */ "BitNot" OpHelp("r[P1]= ~r[P1]"),
29469	/* 55 */ "SorterSort" OpHelp(""),
29470	/* 56 */ "Sort" OpHelp(""),
29471	/* 57 */ "Rewind" OpHelp(""),
29472	/* 58 */ "IdxLE" OpHelp("key=r[P3@P4]"),
29473	/* 59 */ "IdxGT" OpHelp("key=r[P3@P4]"),
29474	/* 60 */ "IdxLT" OpHelp("key=r[P3@P4]"),
29475	/* 61 */ "IdxGE" OpHelp("key=r[P3@P4]"),
29476	/* 62 */ "RowSetRead" OpHelp("r[P3]=rowset(P1)"),
29477	/* 63 */ "RowSetTest" OpHelp("if r[P3] in rowset(P1) goto P2"),
29478	/* 64 */ "Program" OpHelp(""),
29479	/* 65 */ "FkIfZero" OpHelp("if fkctr[P1]==0 goto P2"),
29480	/* 66 */ "IfPos" OpHelp("if r[P1]>0 then r[P1]-=P3, goto P2"),
29481	/* 67 */ "IfNotZero" OpHelp("if r[P1]!=0 then r[P1]--, goto P2"),
29482	/* 68 */ "DecrJumpZero" OpHelp("if (--r[P1])==0 goto P2"),
29483	/* 69 */ "IncrVacuum" OpHelp(""),
29484	/* 70 */ "VNext" OpHelp(""),
29485	/* 71 */ "Init" OpHelp("Start at P2"),
29486	/* 72 */ "Return" OpHelp(""),
29487	/* 73 */ "EndCoroutine" OpHelp(""),
29488	/* 74 */ "HaltIfNull" OpHelp("if r[P3]=null halt"),
29489	/* 75 */ "Halt" OpHelp(""),
29490	/* 76 */ "Integer" OpHelp("r[P2]=P1"),
29491	/* 77 */ "Int64" OpHelp("r[P2]=P4"),
29492	/* 78 */ "String" OpHelp("r[P2]='P4' (len=P1)"),
29493	/* 79 */ "Null" OpHelp("r[P2..P3]=NULL"),
29494	/* 80 */ "SoftNull" OpHelp("r[P1]=NULL"),
29495	/* 81 */ "Blob" OpHelp("r[P2]=P4 (len=P1)"),
29496	/* 82 */ "Variable" OpHelp("r[P2]=parameter(P1,P4)"),
29497	/* 83 */ "Move" OpHelp("r[P2@P3]=r[P1@P3]"),
29498	/* 84 */ "Copy" OpHelp("r[P2@P3+1]=r[P1@P3+1]"),
29499	/* 85 */ "SCopy" OpHelp("r[P2]=r[P1]"),
29500	/* 86 */ "IntCopy" OpHelp("r[P2]=r[P1]"),
29501	/* 87 */ "ResultRow" OpHelp("output=r[P1@P2]"),
29502	/* 88 */ "CollSeq" OpHelp(""),
29503	/* 89 */ "Function0" OpHelp("r[P3]=func(r[P2@P5])"),
29504	/* 90 */ "Function" OpHelp("r[P3]=func(r[P2@P5])"),
29505	/* 91 */ "AddImm" OpHelp("r[P1]=r[P1]+P2"),
29506	/* 92 */ "RealAffinity" OpHelp(""),
29507	/* 93 */ "Cast" OpHelp("affinity(r[P1])"),
29508	/* 94 */ "Permutation" OpHelp(""),
29509	/* 95 */ "Compare" OpHelp("r[P1@P3] <-> r[P2@P3]"),
29510	/* 96 */ "Column" OpHelp("r[P3]=PX"),
29511	/* 97 */ "String8" OpHelp("r[P2]='P4'"),
29512	/* 98 */ "Affinity" OpHelp("affinity(r[P1@P2])"),
29513	/* 99 */ "MakeRecord" OpHelp("r[P3]=mkrec(r[P1@P2])"),
29514	/* 100 */ "Count" OpHelp("r[P2]=count()"),
29515	/* 101 */ "ReadCookie" OpHelp(""),
29516	/* 102 */ "SetCookie" OpHelp(""),
29517	/* 103 */ "ReopenIdx" OpHelp("root=P2 iDb=P3"),
29518	/* 104 */ "OpenRead" OpHelp("root=P2 iDb=P3"),
29519	/* 105 */ "OpenWrite" OpHelp("root=P2 iDb=P3"),
29520	/* 106 */ "OpenAutoindex" OpHelp("nColumn=P2"),
29521	/* 107 */ "OpenEphemeral" OpHelp("nColumn=P2"),
29522	/* 108 */ "SorterOpen" OpHelp(""),
29523	/* 109 */ "SequenceTest" OpHelp("if( cursor[P1].ctr++ ) pc = P2"),
29524	/* 110 */ "OpenPseudo" OpHelp("P3 columns in r[P2]"),
29525	/* 111 */ "Close" OpHelp(""),
29526	/* 112 */ "ColumnsUsed" OpHelp(""),
29527	/* 113 */ "Sequence" OpHelp("r[P2]=cursor[P1].ctr++"),
29528	/* 114 */ "NewRowid" OpHelp("r[P2]=rowid"),
29529	/* 115 */ "Insert" OpHelp("intkey=r[P3] data=r[P2]"),
29530	/* 116 */ "InsertInt" OpHelp("intkey=P3 data=r[P2]"),
29531	/* 117 */ "Delete" OpHelp(""),
29532	/* 118 */ "ResetCount" OpHelp(""),
29533	/* 119 */ "SorterCompare" OpHelp("if key(P1)!=trim(r[P3],P4) goto P2"),
29534	/* 120 */ "SorterData" OpHelp("r[P2]=data"),
29535	/* 121 */ "RowData" OpHelp("r[P2]=data"),
29536	/* 122 */ "Rowid" OpHelp("r[P2]=rowid"),
29537	/* 123 */ "NullRow" OpHelp(""),
29538	/* 124 */ "SorterInsert" OpHelp("key=r[P2]"),
29539	/* 125 */ "IdxInsert" OpHelp("key=r[P2]"),
29540	/* 126 */ "IdxDelete" OpHelp("key=r[P2@P3]"),
29541	/* 127 */ "Seek" OpHelp("Move P3 to P1.rowid"),
29542	/* 128 */ "IdxRowid" OpHelp("r[P2]=rowid"),
29543	/* 129 */ "Destroy" OpHelp(""),
29544	/* 130 */ "Clear" OpHelp(""),
29545	/* 131 */ "ResetSorter" OpHelp(""),
29546	/* 132 */ "Real" OpHelp("r[P2]=P4"),
29547	/* 133 */ "CreateIndex" OpHelp("r[P2]=root iDb=P1"),
29548	/* 134 */ "CreateTable" OpHelp("r[P2]=root iDb=P1"),
29549	/* 135 */ "ParseSchema" OpHelp(""),
29550	/* 136 */ "LoadAnalysis" OpHelp(""),
29551	/* 137 */ "DropTable" OpHelp(""),
29552	/* 138 */ "DropIndex" OpHelp(""),
29553	/* 139 */ "DropTrigger" OpHelp(""),
29554	/* 140 */ "IntegrityCk" OpHelp(""),
29555	/* 141 */ "RowSetAdd" OpHelp("rowset(P1)=r[P2]"),
29556	/* 142 */ "Param" OpHelp(""),
29557	/* 143 */ "FkCounter" OpHelp("fkctr[P1]+=P2"),
29558	/* 144 */ "MemMax" OpHelp("r[P1]=max(r[P1],r[P2])"),
29559	/* 145 */ "OffsetLimit" OpHelp("if r[P1]>0 then r[P2]=r[P1]+max(0,r[P3]) else r[P2]=(-1)"),
29560	/* 146 */ "AggStep0" OpHelp("accum=r[P3] step(r[P2@P5])"),
29561	/* 147 */ "AggStep" OpHelp("accum=r[P3] step(r[P2@P5])"),
29562	/* 148 */ "AggFinal" OpHelp("accum=r[P1] N=P2"),
29563	/* 149 */ "Expire" OpHelp(""),
29564	/* 150 */ "TableLock" OpHelp("iDb=P1 root=P2 write=P3"),
29565	/* 151 */ "VBegin" OpHelp(""),
29566	/* 152 */ "VCreate" OpHelp(""),
29567	/* 153 */ "VDestroy" OpHelp(""),
29568	/* 154 */ "VOpen" OpHelp(""),
29569	/* 155 */ "VColumn" OpHelp("r[P3]=vcolumn(P2)"),
29570	/* 156 */ "VRename" OpHelp(""),
29571	/* 157 */ "Pagecount" OpHelp(""),
29572	/* 158 */ "MaxPgcnt" OpHelp(""),
29573	/* 159 */ "CursorHint" OpHelp(""),
29574	/* 160 */ "Noop" OpHelp(""),
29575	/* 161 */ "Explain" OpHelp(""),


29576	};
29577	return azName[i];
29578	}
29579	#endif
29580
	@@ -37952,46 +38003,84 @@
37952	* be freed immediately and any attempt to access any of that freed
37953	* data will almost certainly result in an immediate access violation.
37954	******************************************************************************
37955	*/
37956	#ifndef SQLITE_WIN32_HEAP_CREATE
37957	# define SQLITE_WIN32_HEAP_CREATE (TRUE)



























37958	#endif
37959
37960	/*
37961	* This is cache size used in the calculation of the initial size of the
37962	* Win32-specific heap. It cannot be negative.
37963	*/
37964	#ifndef SQLITE_WIN32_CACHE_SIZE
37965	# if SQLITE_DEFAULT_CACHE_SIZE>=0
37966	# define SQLITE_WIN32_CACHE_SIZE (SQLITE_DEFAULT_CACHE_SIZE)
37967	# else
37968	# define SQLITE_WIN32_CACHE_SIZE (-(SQLITE_DEFAULT_CACHE_SIZE))
37969	# endif
37970	#endif
37971










37972	/*
37973	* The initial size of the Win32-specific heap. This value may be zero.
37974	*/
37975	#ifndef SQLITE_WIN32_HEAP_INIT_SIZE
37976	# define SQLITE_WIN32_HEAP_INIT_SIZE ((SQLITE_WIN32_CACHE_SIZE) * \
37977	(SQLITE_DEFAULT_PAGE_SIZE) + 4194304)

37978	#endif
37979
37980	/*
37981	* The maximum size of the Win32-specific heap. This value may be zero.
37982	*/
37983	#ifndef SQLITE_WIN32_HEAP_MAX_SIZE
37984	# define SQLITE_WIN32_HEAP_MAX_SIZE (0)
37985	#endif
37986
37987	/*
37988	* The extra flags to use in calls to the Win32 heap APIs. This value may be
37989	* zero for the default behavior.
37990	*/
37991	#ifndef SQLITE_WIN32_HEAP_FLAGS
37992	# define SQLITE_WIN32_HEAP_FLAGS (0)
37993	#endif
37994
37995
37996	/*
37997	** The winMemData structure stores information required by the Win32-specific
	@@ -44084,11 +44173,11 @@
44084	** This routine is for use inside of assert() statements only. For
44085	** example:
44086	**
44087	** assert( sqlite3PcachePageSanity(pPg) );
44088	*/
44089	#if SQLITE_DEBUG
44090	SQLITE_PRIVATE int sqlite3PcachePageSanity(PgHdr *pPg){
44091	PCache *pCache;
44092	assert( pPg!=0 );
44093	assert( pPg->pgno>0 \|\| pPg->pPager==0 ); /* Page number is 1 or more */
44094	pCache = pPg->pCache;
	@@ -60177,21 +60266,22 @@
60177	}
60178	#endif
60179
60180
60181	/*
60182	** Defragment the page given. All Cells are moved to the
60183	** end of the page and all free space is collected into one
60184	** big FreeBlk that occurs in between the header and cell
60185	** pointer array and the cell content area.

60186	**
60187	** EVIDENCE-OF: R-44582-60138 SQLite may from time to time reorganize a
60188	** b-tree page so that there are no freeblocks or fragment bytes, all
60189	** unused bytes are contained in the unallocated space region, and all
60190	** cells are packed tightly at the end of the page.
60191	*/
60192	static int defragmentPage(MemPage *pPage){
60193	int i; /* Loop counter */
60194	int pc; /* Address of the i-th cell */
60195	int hdr; /* Offset to the page header */
60196	int size; /* Size of a cell */
60197	int usableSize; /* Number of usable bytes on a page */
	@@ -60202,11 +60292,10 @@
60202	unsigned char temp; / Temp area for cell content */
60203	unsigned char src; / Source of content */
60204	int iCellFirst; /* First allowable cell index */
60205	int iCellLast; /* Last possible cell index */
60206
60207
60208	assert( sqlite3PagerIswriteable(pPage->pDbPage) );
60209	assert( pPage->pBt!=0 );
60210	assert( pPage->pBt->usableSize <= SQLITE_MAX_PAGE_SIZE );
60211	assert( pPage->nOverflow==0 );
60212	assert( sqlite3_mutex_held(pPage->pBt->mutex) );
	@@ -60214,13 +60303,60 @@
60214	src = data = pPage->aData;
60215	hdr = pPage->hdrOffset;
60216	cellOffset = pPage->cellOffset;
60217	nCell = pPage->nCell;
60218	assert( nCell==get2byte(&data[hdr+3]) );

60219	usableSize = pPage->pBt->usableSize;















































60220	cbrk = usableSize;
60221	iCellFirst = cellOffset + 2*nCell;
60222	iCellLast = usableSize - 4;
60223	for(i=0; i<nCell; i++){
60224	u8 pAddr; / The i-th cell pointer */
60225	pAddr = &data[cellOffset + i*2];
60226	pc = get2byte(pAddr);
	@@ -60250,20 +60386,22 @@
60250	memcpy(&temp[x], &data[x], (cbrk+size) - x);
60251	src = temp;
60252	}
60253	memcpy(&data[cbrk], &src[pc], size);
60254	}






60255	assert( cbrk>=iCellFirst );
60256	put2byte(&data[hdr+5], cbrk);
60257	data[hdr+1] = 0;
60258	data[hdr+2] = 0;
60259	data[hdr+7] = 0;
60260	memset(&data[iCellFirst], 0, cbrk-iCellFirst);
60261	assert( sqlite3PagerIswriteable(pPage->pDbPage) );
60262	if( cbrk-iCellFirst!=pPage->nFree ){
60263	return SQLITE_CORRUPT_BKPT;
60264	}
60265	return SQLITE_OK;
60266	}
60267
60268	/*
60269	** Search the free-list on page pPg for space to store a cell nByte bytes in
	@@ -60397,14 +60535,14 @@
60397	** to see if defragmentation is necessary.
60398	*/
60399	testcase( gap+2+nByte==top );
60400	if( gap+2+nByte>top ){
60401	assert( pPage->nCell>0 \|\| CORRUPT_DB );
60402	rc = defragmentPage(pPage);
60403	if( rc ) return rc;
60404	top = get2byteNotZero(&data[hdr+5]);
60405	assert( gap+nByte<=top );
60406	}
60407
60408
60409	/* Allocate memory from the gap in between the cell pointer array
60410	** and the cell content area. The btreeInitPage() call has already
	@@ -63636,11 +63774,11 @@
63636	pCur->aiIdx[pCur->iPage] = 0;
63637	return getAndInitPage(pBt, newPgno, &pCur->apPage[pCur->iPage],
63638	pCur, pCur->curPagerFlags);
63639	}
63640
63641	#if SQLITE_DEBUG
63642	/*
63643	** Page pParent is an internal (non-leaf) tree page. This function
63644	** asserts that page number iChild is the left-child if the iIdx'th
63645	** cell in page pParent. Or, if iIdx is equal to the total number of
63646	** cells in pParent, that page number iChild is the right-child of
	@@ -64178,10 +64316,34 @@
64178	** have been deleted? This API will need to change to return an error code
64179	** as well as the boolean result value.
64180	*/
64181	return (CURSOR_VALID!=pCur->eState);
64182	}
























64183
64184	/*
64185	** Advance the cursor to the next entry in the database. If
64186	** successful then set *pRes=0. If the cursor
64187	** was already pointing to the last entry in the database before
	@@ -65023,11 +65185,11 @@
65023	** pPrior Where to write the pgno of the first overflow page
65024	**
65025	** Use a call to btreeParseCellPtr() to verify that the values above
65026	** were computed correctly.
65027	*/
65028	#if SQLITE_DEBUG
65029	{
65030	CellInfo info;
65031	pPage->xParseCell(pPage, pCell, &info);
65032	assert( nHeader==(int)(info.pPayload - pCell) );
65033	assert( info.nKey==pX->nKey );
	@@ -66549,11 +66711,11 @@
66549	** copied into the parent, because if the parent is page 1 then it will
66550	** by smaller than the child due to the database header, and so all the
66551	** free space needs to be up front.
66552	*/
66553	assert( nNew==1 \|\| CORRUPT_DB );
66554	rc = defragmentPage(apNew[0]);
66555	testcase( rc!=SQLITE_OK );
66556	assert( apNew[0]->nFree ==
66557	(get2byte(&apNew[0]->aData[5])-apNew[0]->cellOffset-apNew[0]->nCell*2)
66558	\|\| rc!=SQLITE_OK
66559	);
	@@ -71201,10 +71363,11 @@
71201	** Remember the SQL string for a prepared statement.
71202	*/
71203	SQLITE_PRIVATE void sqlite3VdbeSetSql(Vdbe p, const char z, int n, int isPrepareV2){
71204	assert( isPrepareV2==1 \|\| isPrepareV2==0 );
71205	if( p==0 ) return;

71206	#if defined(SQLITE_OMIT_TRACE) && !defined(SQLITE_ENABLE_SQLLOG)
71207	if( !isPrepareV2 ) return;
71208	#endif
71209	assert( p->zSql==0 );
71210	p->zSql = sqlite3DbStrNDup(p->db, z, n);
	@@ -71229,10 +71392,11 @@
71229	pB->pPrev = pTmp;
71230	zTmp = pA->zSql;
71231	pA->zSql = pB->zSql;
71232	pB->zSql = zTmp;
71233	pB->isPrepareV2 = pA->isPrepareV2;

71234	}
71235
71236	/*
71237	** Resize the Vdbe.aOp array so that it is at least nOp elements larger
71238	** than its current size. nOp is guaranteed to be less than or equal
	@@ -73753,17 +73917,17 @@
73753	** Then the internal cache might have been left in an inconsistent
73754	** state. We need to rollback the statement transaction, if there is
73755	** one, or the complete transaction if there is no statement transaction.
73756	*/
73757



73758	if( db->mallocFailed ){
73759	p->rc = SQLITE_NOMEM_BKPT;
73760	}
73761	closeAllCursors(p);
73762	if( p->magic!=VDBE_MAGIC_RUN ){
73763	return SQLITE_OK;
73764	}
73765	checkActiveVdbeCnt(db);
73766
73767	/* No commit or rollback needed if the program never started or if the
73768	** SQL statement does not read or write a database file. */
73769	if( p->pc>=0 && p->bIsReader ){
	@@ -74706,11 +74870,11 @@
74706	}
74707	assert( u<=pKeyInfo->nField + 1 );
74708	p->nField = u;
74709	}
74710
74711	#if SQLITE_DEBUG
74712	/*
74713	** This function compares two index or table record keys in the same way
74714	** as the sqlite3VdbeRecordCompare() routine. Unlike VdbeRecordCompare(),
74715	** this function deserializes and compares values using the
74716	** sqlite3VdbeSerialGet() and sqlite3MemCompare() functions. It is used
	@@ -74811,11 +74975,11 @@
74811	if( pKeyInfo->db->mallocFailed ) return 1;
74812	return 0;
74813	}
74814	#endif
74815
74816	#if SQLITE_DEBUG
74817	/*
74818	** Count the number of fields (a.k.a. columns) in the record given by
74819	** pKey,nKey. The verify that this count is less than or equal to the
74820	** limit given by pKeyInfo->nField + pKeyInfo->nXField.
74821	**
	@@ -75694,12 +75858,12 @@
75694	** to sqlite3_reoptimize() that re-preparing the statement may result
75695	** in a better query plan.
75696	*/
75697	SQLITE_PRIVATE void sqlite3VdbeSetVarmask(Vdbe *v, int iVar){
75698	assert( iVar>0 );
75699	if( iVar>32 ){
75700	v->expmask = 0xffffffff;
75701	}else{
75702	v->expmask \|= ((u32)1 << (iVar-1));
75703	}
75704	}
75705
	@@ -75965,11 +76129,12 @@
75965	sqlite3_mutex_enter(mutex);
75966	for(i=0; i<p->nVar; i++){
75967	sqlite3VdbeMemRelease(&p->aVar[i]);
75968	p->aVar[i].flags = MEM_Null;
75969	}
75970	if( p->isPrepareV2 && p->expmask ){

75971	p->expired = 1;
75972	}
75973	sqlite3_mutex_leave(mutex);
75974	return rc;
75975	}
	@@ -77069,13 +77234,12 @@
77069	** parameter in the WHERE clause might influence the choice of query plan
77070	** for a statement, then the statement will be automatically recompiled,
77071	** as if there had been a schema change, on the first sqlite3_step() call
77072	** following any change to the bindings of that parameter.
77073	*/
77074	if( p->isPrepareV2 &&
77075	((i<32 && p->expmask & ((u32)1 << i)) \|\| p->expmask==0xffffffff)
77076	){
77077	p->expired = 1;
77078	}
77079	return SQLITE_OK;
77080	}
77081
	@@ -77334,14 +77498,16 @@
77334	Vdbe pFrom = (Vdbe)pFromStmt;
77335	Vdbe pTo = (Vdbe)pToStmt;
77336	if( pFrom->nVar!=pTo->nVar ){
77337	return SQLITE_ERROR;
77338	}
77339	if( pTo->isPrepareV2 && pTo->expmask ){

77340	pTo->expired = 1;
77341	}
77342	if( pFrom->isPrepareV2 && pFrom->expmask ){

77343	pFrom->expired = 1;
77344	}
77345	return sqlite3TransferBindings(pFromStmt, pToStmt);
77346	}
77347	#endif
	@@ -79658,39 +79824,39 @@
79658	if( pCtx->pOut != pOut ){
79659	pCtx->pOut = pOut;
79660	for(i=pCtx->argc-1; i>=0; i--) pCtx->argv[i] = &aMem[pOp->p2+i];
79661	}
79662
79663	memAboutToChange(p, pCtx->pOut);
79664	#ifdef SQLITE_DEBUG
79665	for(i=0; i<pCtx->argc; i++){
79666	assert( memIsValid(pCtx->argv[i]) );
79667	REGISTER_TRACE(pOp->p2+i, pCtx->argv[i]);
79668	}
79669	#endif
79670	MemSetTypeFlag(pCtx->pOut, MEM_Null);
79671	pCtx->fErrorOrAux = 0;
79672	(pCtx->pFunc->xSFunc)(pCtx, pCtx->argc, pCtx->argv);/ IMP: R-24505-23230 */
79673
79674	/* If the function returned an error, throw an exception */
79675	if( pCtx->fErrorOrAux ){
79676	if( pCtx->isError ){
79677	sqlite3VdbeError(p, "%s", sqlite3_value_text(pCtx->pOut));
79678	rc = pCtx->isError;
79679	}
79680	sqlite3VdbeDeleteAuxData(db, &p->pAuxData, pCtx->iOp, pOp->p1);
79681	if( rc ) goto abort_due_to_error;
79682	}
79683
79684	/* Copy the result of the function into register P3 */
79685	if( pOut->flags & (MEM_Str\|MEM_Blob) ){
79686	sqlite3VdbeChangeEncoding(pCtx->pOut, encoding);
79687	if( sqlite3VdbeMemTooBig(pCtx->pOut) ) goto too_big;
79688	}
79689
79690	REGISTER_TRACE(pOp->p3, pCtx->pOut);
79691	UPDATE_MAX_BLOBSIZE(pCtx->pOut);
79692	break;
79693	}
79694
79695	/* Opcode: BitAnd P1 P2 P3 * *
79696	** Synopsis: r[P3]=r[P1]&r[P2]
	@@ -80187,11 +80353,11 @@
80187	pKeyInfo = pOp->p4.pKeyInfo;
80188	assert( n>0 );
80189	assert( pKeyInfo!=0 );
80190	p1 = pOp->p1;
80191	p2 = pOp->p2;
80192	#if SQLITE_DEBUG
80193	if( aPermute ){
80194	int k, mx = 0;
80195	for(k=0; k<n; k++) if( aPermute[k]>mx ) mx = aPermute[k];
80196	assert( p1>0 && p1+mx<=(p->nMem+1 - p->nCursor)+1 );
80197	assert( p2>0 && p2+mx<=(p->nMem+1 - p->nCursor)+1 );
	@@ -82836,10 +83002,37 @@
82836	assert( pOp->p2==0 );
82837	}
82838	break;
82839	}
82840



























82841
82842	/* Opcode: SorterSort P1 P2 * * *
82843	**
82844	** After all records have been inserted into the Sorter object
82845	** identified by P1, invoke this opcode to actually do the sorting.
	@@ -83479,10 +83672,22 @@
83479	rc = sqlite3BtreeCreateTable(pDb->pBt, &pgno, flags);
83480	if( rc ) goto abort_due_to_error;
83481	pOut->u.i = pgno;
83482	break;
83483	}












83484
83485	/* Opcode: ParseSchema P1 * * P4 *
83486	**
83487	** Read and parse all entries from the SQLITE_MASTER table of database P1
83488	** that match the WHERE clause P4.
	@@ -83600,11 +83805,11 @@
83600	**
83601	** Do an analysis of the currently open database. Store in
83602	** register P1 the text of an error message describing any problems.
83603	** If no problems are found, store a NULL in register P1.
83604	**
83605	** The register P3 contains the maximum number of allowed errors.
83606	** At most reg(P3) errors will be reported.
83607	** In other words, the analysis stops as soon as reg(P1) errors are
83608	** seen. Reg(P1) is updated with the number of errors remaining.
83609	**
83610	** The root page numbers of all tables in the database are integers
	@@ -83633,18 +83838,18 @@
83633	assert( (pnErr->flags & (MEM_Str\|MEM_Blob))==0 );
83634	pIn1 = &aMem[pOp->p1];
83635	assert( pOp->p5<db->nDb );
83636	assert( DbMaskTest(p->btreeMask, pOp->p5) );
83637	z = sqlite3BtreeIntegrityCheck(db->aDb[pOp->p5].pBt, aRoot, nRoot,
83638	(int)pnErr->u.i, &nErr);
83639	pnErr->u.i -= nErr;
83640	sqlite3VdbeMemSetNull(pIn1);
83641	if( nErr==0 ){
83642	assert( z==0 );
83643	}else if( z==0 ){
83644	goto no_mem;
83645	}else{

83646	sqlite3VdbeMemSetStr(pIn1, z, -1, SQLITE_UTF8, sqlite3_free);
83647	}
83648	UPDATE_MAX_BLOBSIZE(pIn1);
83649	sqlite3VdbeChangeEncoding(pIn1, encoding);
83650	break;
	@@ -88787,15 +88992,15 @@
88787	** The return value from the callback should be one of the WRC_*
88788	** constants to specify how to proceed with the walk.
88789	**
88790	** WRC_Continue Continue descending down the tree.
88791	**
88792	** WRC_Prune Do not descend into child nodes. But allow
88793	** the walk to continue with sibling nodes.
88794	**
88795	** WRC_Abort Do no more callbacks. Unwind the stack and
88796	** return the top-level walk call.
88797	**
88798	** The return value from this routine is WRC_Abort to abandon the tree walk
88799	** and WRC_Continue to continue.
88800	*/
88801	static SQLITE_NOINLINE int walkExpr(Walker pWalker, Expr pExpr){
	@@ -89153,11 +89358,12 @@
89153	}
89154	}
89155	}
89156
89157	/* Start at the inner-most context and move outward until a match is found */
89158	while( pNC && cnt==0 ){

89159	ExprList *pEList;
89160	SrcList *pSrcList = pNC->pSrcList;
89161
89162	if( pSrcList ){
89163	for(i=0, pItem=pSrcList->a; i<pSrcList->nSrc; i++, pItem++){
	@@ -89338,15 +89544,15 @@
89338	}
89339
89340	/* Advance to the next name context. The loop will exit when either
89341	** we have a match (cnt>0) or when we run out of name contexts.
89342	*/
89343	if( cnt==0 ){
89344	pNC = pNC->pNext;
89345	nSubquery++;
89346	}
89347	}
89348
89349	/*
89350	** If X and Y are NULL (in other words if only the column name Z is
89351	** supplied) and the value of Z is enclosed in double-quotes, then
89352	** Z is a string literal if it doesn't match any column names. In that
	@@ -89532,37 +89738,42 @@
89532	break;
89533	}
89534	#endif /* defined(SQLITE_ENABLE_UPDATE_DELETE_LIMIT)
89535	&& !defined(SQLITE_OMIT_SUBQUERY) */
89536
89537	/* A lone identifier is the name of a column.
89538	*/
89539	case TK_ID: {
89540	return lookupName(pParse, 0, 0, pExpr->u.zToken, pNC, pExpr);
89541	}
89542
89543	/* A table name and column name: ID.ID
89544	** Or a database, table and column: ID.ID.ID




89545	*/

89546	case TK_DOT: {
89547	const char *zColumn;
89548	const char *zTable;
89549	const char *zDb;
89550	Expr *pRight;
89551
89552	/* if( pSrcList==0 ) break; */
89553	notValid(pParse, pNC, "the \".\" operator", NC_IdxExpr);
89554	pRight = pExpr->pRight;
89555	if( pRight->op==TK_ID ){
89556	zDb = 0;
89557	zTable = pExpr->pLeft->u.zToken;
89558	zColumn = pRight->u.zToken;
89559	}else{
89560	assert( pRight->op==TK_DOT );
89561	zDb = pExpr->pLeft->u.zToken;
89562	zTable = pRight->pLeft->u.zToken;
89563	zColumn = pRight->pRight->u.zToken;








89564	}
89565	return lookupName(pParse, zDb, zTable, zColumn, pNC, pExpr);
89566	}
89567
89568	/* Resolve function names
	@@ -92833,11 +93044,11 @@
92833	int nVal = sqlite3ExprVectorSize(pLeft);
92834	Select *pSelect = (pExpr->flags & EP_xIsSelect) ? pExpr->x.pSelect : 0;
92835	char *zRet;
92836
92837	assert( pExpr->op==TK_IN );
92838	zRet = sqlite3DbMallocZero(pParse->db, nVal+1);
92839	if( zRet ){
92840	int i;
92841	for(i=0; i<nVal; i++){
92842	Expr *pA = sqlite3VectorFieldSubexpr(pLeft, i);
92843	char a = sqlite3ExprAffinity(pA);
	@@ -93738,11 +93949,11 @@
93738	** Clear all column cache entries.
93739	*/
93740	SQLITE_PRIVATE void sqlite3ExprCacheClear(Parse *pParse){
93741	int i;
93742
93743	#if SQLITE_DEBUG
93744	if( pParse->db->flags & SQLITE_VdbeAddopTrace ){
93745	printf("CLEAR\n");
93746	}
93747	#endif
93748	for(i=0; i<pParse->nColCache; i++){
	@@ -95145,10 +95356,21 @@
95145	if( pA->a[i].sortOrder!=pB->a[i].sortOrder ) return 1;
95146	if( sqlite3ExprCompare(pExprA, pExprB, iTab) ) return 1;
95147	}
95148	return 0;
95149	}











95150
95151	/*
95152	** Return true if we can prove the pE2 will always be true if pE1 is
95153	** true. Return false if we cannot complete the proof or if pE2 might
95154	** be false. Examples:
	@@ -96698,10 +96920,11 @@
96698	Stat4Sample current; /* Current row as a Stat4Sample */
96699	u32 iPrn; /* Pseudo-random number used for sampling */
96700	Stat4Sample aBest; / Array of nCol best samples */
96701	int iMin; /* Index in a[] of entry with minimum score */
96702	int nSample; /* Current number of samples */

96703	int iGet; /* Index of current sample accessed by stat_get() */
96704	Stat4Sample a; / Array of mxSample Stat4Sample objects */
96705	sqlite3 db; / Database connection, for malloc() */
96706	};
96707
	@@ -96962,10 +97185,17 @@
96962	int i;
96963
96964	assert( IsStat4 \|\| nEqZero==0 );
96965
96966	#ifdef SQLITE_ENABLE_STAT4







96967	if( pNew->isPSample==0 ){
96968	Stat4Sample *pUpgrade = 0;
96969	assert( pNew->anEq[pNew->iCol]>0 );
96970
96971	/* This sample is being added because the prefix that ends in column
	@@ -97059,16 +97289,26 @@
97059	if( p->nSample<p->mxSample \|\| sampleIsBetter(p, pBest, &p->a[p->iMin]) ){
97060	sampleInsert(p, pBest, i);
97061	}
97062	}
97063
97064	/* Update the anEq[] fields of any samples already collected. */

97065	for(i=p->nSample-1; i>=0; i--){
97066	int j;
97067	for(j=iChng; j<p->nCol; j++){
97068	if( p->a[i].anEq[j]==0 ) p->a[i].anEq[j] = p->current.anEq[j];








97069	}

97070	}
97071	#endif
97072
97073	#if defined(SQLITE_ENABLE_STAT3) && !defined(SQLITE_ENABLE_STAT4)
97074	if( iChng==0 ){
	@@ -97778,44 +98018,32 @@
97778	/* Form 1: Analyze everything */
97779	for(i=0; i<db->nDb; i++){
97780	if( i==1 ) continue; /* Do not analyze the TEMP database */
97781	analyzeDatabase(pParse, i);
97782	}
97783	}else if( pName2->n==0 ){
97784	/* Form 2: Analyze the database or table named */
97785	iDb = sqlite3FindDb(db, pName1);
97786	if( iDb>=0 ){
97787	analyzeDatabase(pParse, iDb);
97788	}else{
97789	z = sqlite3NameFromToken(db, pName1);
97790	if( z ){
97791	if( (pIdx = sqlite3FindIndex(db, z, 0))!=0 ){
97792	analyzeTable(pParse, pIdx->pTable, pIdx);
97793	}else if( (pTab = sqlite3LocateTable(pParse, 0, z, 0))!=0 ){
97794	analyzeTable(pParse, pTab, 0);
97795	}
97796	sqlite3DbFree(db, z);
97797	}
97798	}
97799	}else{
97800	/* Form 3: Analyze the fully qualified table name */
97801	iDb = sqlite3TwoPartName(pParse, pName1, pName2, &pTableName);
97802	if( iDb>=0 ){
97803	zDb = db->aDb[iDb].zDbSName;
97804	z = sqlite3NameFromToken(db, pTableName);
97805	if( z ){
97806	if( (pIdx = sqlite3FindIndex(db, z, zDb))!=0 ){
97807	analyzeTable(pParse, pIdx->pTable, pIdx);
97808	}else if( (pTab = sqlite3LocateTable(pParse, 0, z, zDb))!=0 ){
97809	analyzeTable(pParse, pTab, 0);
97810	}
97811	sqlite3DbFree(db, z);
97812	}
97813	}
97814	}
97815	v = sqlite3GetVdbe(pParse);
97816	if( v ) sqlite3VdbeAddOp0(v, OP_Expire);

97817	}
97818
97819	/*
97820	** Used to pass information from the analyzer reader through to the
97821	** callback routine.
	@@ -97940,19 +98168,24 @@
97940	}
97941	aiRowEst = pIndex->aiRowEst;
97942	#endif
97943	pIndex->bUnordered = 0;
97944	decodeIntArray((char*)z, nCol, aiRowEst, pIndex->aiRowLogEst, pIndex);
97945	if( pIndex->pPartIdxWhere==0 ) pTable->nRowLogEst = pIndex->aiRowLogEst[0];




97946	}else{
97947	Index fakeIdx;
97948	fakeIdx.szIdxRow = pTable->szTabRow;
97949	#ifdef SQLITE_ENABLE_COSTMULT
97950	fakeIdx.pTable = pTable;
97951	#endif
97952	decodeIntArray((char*)z, 1, 0, &pTable->nRowLogEst, &fakeIdx);
97953	pTable->szTabRow = fakeIdx.szIdxRow;

97954	}
97955
97956	return 0;
97957	}
97958
	@@ -98243,19 +98476,24 @@
98243	SQLITE_PRIVATE int sqlite3AnalysisLoad(sqlite3 *db, int iDb){
98244	analysisInfo sInfo;
98245	HashElem *i;
98246	char *zSql;
98247	int rc = SQLITE_OK;

98248
98249	assert( iDb>=0 && iDb<db->nDb );
98250	assert( db->aDb[iDb].pBt!=0 );
98251
98252	/* Clear any prior statistics */
98253	assert( sqlite3SchemaMutexHeld(db, iDb, 0) );
98254	for(i=sqliteHashFirst(&db->aDb[iDb].pSchema->idxHash);i;i=sqliteHashNext(i)){




98255	Index *pIdx = sqliteHashData(i);
98256	pIdx->aiRowLogEst[0] = 0;
98257	#ifdef SQLITE_ENABLE_STAT3_OR_STAT4
98258	sqlite3DeleteIndexSamples(db, pIdx);
98259	pIdx->aSample = 0;
98260	#endif
98261	}
	@@ -98274,23 +98512,23 @@
98274	}
98275	}
98276
98277	/* Set appropriate defaults on all indexes not in the sqlite_stat1 table */
98278	assert( sqlite3SchemaMutexHeld(db, iDb, 0) );
98279	for(i=sqliteHashFirst(&db->aDb[iDb].pSchema->idxHash);i;i=sqliteHashNext(i)){
98280	Index *pIdx = sqliteHashData(i);
98281	if( pIdx->aiRowLogEst[0]==0 ) sqlite3DefaultRowEst(pIdx);
98282	}
98283
98284	/* Load the statistics from the sqlite_stat4 table. */
98285	#ifdef SQLITE_ENABLE_STAT3_OR_STAT4
98286	if( rc==SQLITE_OK && OptimizationEnabled(db, SQLITE_Stat34) ){
98287	db->lookaside.bDisable++;
98288	rc = loadStat4(db, sInfo.zDatabase);
98289	db->lookaside.bDisable--;
98290	}
98291	for(i=sqliteHashFirst(&db->aDb[iDb].pSchema->idxHash);i;i=sqliteHashNext(i)){
98292	Index *pIdx = sqliteHashData(i);
98293	sqlite3_free(pIdx->aiRowEst);
98294	pIdx->aiRowEst = 0;
98295	}
98296	#endif
	@@ -100271,10 +100509,11 @@
100271	SQLITE_PRIVATE void sqlite3AddNotNull(Parse *pParse, int onError){
100272	Table *p;
100273	p = pParse->pNewTable;
100274	if( p==0 \|\| NEVER(p->nCol<1) ) return;
100275	p->aCol[p->nCol-1].notNull = (u8)onError;

100276	}
100277
100278	/*
100279	** Scan the column type name zType (length nType) and return the
100280	** associated affinity type.
	@@ -102609,10 +102848,13 @@
102609	/* 10, 9, 8, 7, 6 */
102610	LogEst aVal[] = { 33, 32, 30, 28, 26 };
102611	LogEst *a = pIdx->aiRowLogEst;
102612	int nCopy = MIN(ArraySize(aVal), pIdx->nKeyCol);
102613	int i;



102614
102615	/* Set the first entry (number of rows in the index) to the estimated
102616	** number of rows in the table, or half the number of rows in the table
102617	** for a partial index. But do not let the estimate drop below 10. */
102618	a[0] = pIdx->pTable->nRowLogEst;
	@@ -109798,10 +110040,13 @@
109798	VdbeComment((v, "%s", iField<0 ? "rowid" : pTab->aCol[iField].zName));
109799	}
109800	}
109801	sqlite3VdbeAddOp3(v, OP_MakeRecord, regIdx, pIdx->nColumn, aRegIdx[ix]);
109802	VdbeComment((v, "for %s", pIdx->zName));



109803
109804	/* In an UPDATE operation, if this index is the PRIMARY KEY index
109805	** of a WITHOUT ROWID table and there has been no change the
109806	** primary key, then no collision is possible. The collision detection
109807	** logic below can all be skipped. */
	@@ -109953,12 +110198,15 @@
109953
109954	/* Records with omitted columns are only allowed for schema format
109955	** version 2 and later (SQLite version 3.1.4, 2005-02-20). */
109956	if( pTab->pSchema->file_format<2 ) return;
109957
109958	for(i=pTab->nCol; i>1 && pTab->aCol[i-1].pDflt==0; i--){}
109959	sqlite3VdbeChangeP5(v, i);



109960	}
109961	#endif
109962
109963	/*
109964	** This routine generates code to finish the INSERT or UPDATE operation
	@@ -110244,11 +110492,11 @@
110244	}
110245	if( sqlite3TriggerList(pParse, pDest) ){
110246	return 0; /* tab1 must not have triggers */
110247	}
110248	#ifndef SQLITE_OMIT_VIRTUALTABLE
110249	if( pDest->tabFlags & TF_Virtual ){
110250	return 0; /* tab1 must not be a virtual table */
110251	}
110252	#endif
110253	if( onError==OE_Default ){
110254	if( pDest->iPKey>=0 ) onError = pDest->keyConf;
	@@ -110306,11 +110554,11 @@
110306	}
110307	if( HasRowid(pDest)!=HasRowid(pSrc) ){
110308	return 0; /* source and destination must both be WITHOUT ROWID or not */
110309	}
110310	#ifndef SQLITE_OMIT_VIRTUALTABLE
110311	if( pSrc->tabFlags & TF_Virtual ){
110312	return 0; /* tab2 must not be a virtual table */
110313	}
110314	#endif
110315	if( pSrc->pSelect ){
110316	return 0; /* tab2 may not be a view */
	@@ -110603,11 +110851,11 @@
110603	/* Invoke the callback function if required */
110604	if( xCallback && (SQLITE_ROW==rc \|\|
110605	(SQLITE_DONE==rc && !callbackIsInit
110606	&& db->flags&SQLITE_NullCallback)) ){
110607	if( !callbackIsInit ){
110608	azCols = sqlite3DbMallocZero(db, 2nColsizeof(const char*) + 1);
110609	if( azCols==0 ){
110610	goto exec_out;
110611	}
110612	for(i=0; i<nCol; i++){
110613	azCols[i] = (char *)sqlite3_column_name(pStmt, i);
	@@ -110624,10 +110872,11 @@
110624	if( !azVals[i] && sqlite3_column_type(pStmt, i)!=SQLITE_NULL ){
110625	sqlite3OomFault(db);
110626	goto exec_out;
110627	}
110628	}

110629	}
110630	if( xCallback(pArg, nCol, azVals, azCols) ){
110631	/* EVIDENCE-OF: R-38229-40159 If the callback function to
110632	** sqlite3_exec() returns non-zero, then sqlite3_exec() will
110633	** return SQLITE_ABORT. */
	@@ -112094,15 +112343,15 @@
112094	#define PragTyp_JOURNAL_SIZE_LIMIT 20
112095	#define PragTyp_LOCK_PROXY_FILE 21
112096	#define PragTyp_LOCKING_MODE 22
112097	#define PragTyp_PAGE_COUNT 23
112098	#define PragTyp_MMAP_SIZE 24
112099	#define PragTyp_PAGE_SIZE 25
112100	#define PragTyp_SECURE_DELETE 26
112101	#define PragTyp_SHRINK_MEMORY 27
112102	#define PragTyp_SOFT_HEAP_LIMIT 28
112103	#define PragTyp_STATS 29
112104	#define PragTyp_SYNCHRONOUS 30
112105	#define PragTyp_TABLE_INFO 31
112106	#define PragTyp_TEMP_STORE 32
112107	#define PragTyp_TEMP_STORE_DIRECTORY 33
112108	#define PragTyp_THREADS 34
	@@ -112112,10 +112361,11 @@
112112	#define PragTyp_HEXKEY 38
112113	#define PragTyp_KEY 39
112114	#define PragTyp_REKEY 40
112115	#define PragTyp_LOCK_STATUS 41
112116	#define PragTyp_PARSER_TRACE 42

112117
112118	/* Property flags associated with various pragma. */
112119	#define PragFlg_NeedSchema 0x01 /* Force schema load before running */
112120	#define PragFlg_NoColumns 0x02 /* OP_ResultRow called with zero columns */
112121	#define PragFlg_NoColumns1 0x04 /* zero columns if RHS argument is present */
	@@ -112135,51 +112385,52 @@
112135	/* 2 */ "name",
112136	/* 3 */ "type",
112137	/* 4 */ "notnull",
112138	/* 5 */ "dflt_value",
112139	/* 6 */ "pk",
112140	/* 7 / "table", / Used by: stats */
112141	/* 8 */ "index",
112142	/* 9 */ "width",
112143	/* 10 */ "height",
112144	/* 11 / "seqno", / Used by: index_info */
112145	/* 12 */ "cid",
112146	/* 13 */ "name",
112147	/* 14 / "seqno", / Used by: index_xinfo */
112148	/* 15 */ "cid",
112149	/* 16 */ "name",
112150	/* 17 */ "desc",
112151	/* 18 */ "coll",
112152	/* 19 */ "key",
112153	/* 20 / "seq", / Used by: index_list */
112154	/* 21 */ "name",
112155	/* 22 */ "unique",
112156	/* 23 */ "origin",
112157	/* 24 */ "partial",
112158	/* 25 / "seq", / Used by: database_list */
112159	/* 26 */ "name",
112160	/* 27 */ "file",
112161	/* 28 / "seq", / Used by: collation_list */
112162	/* 29 */ "name",
112163	/* 30 / "id", / Used by: foreign_key_list */
112164	/* 31 */ "seq",
112165	/* 32 */ "table",
112166	/* 33 */ "from",
112167	/* 34 */ "to",
112168	/* 35 */ "on_update",
112169	/* 36 */ "on_delete",
112170	/* 37 */ "match",
112171	/* 38 / "table", / Used by: foreign_key_check */
112172	/* 39 */ "rowid",
112173	/* 40 */ "parent",
112174	/* 41 */ "fkid",
112175	/* 42 / "busy", / Used by: wal_checkpoint */
112176	/* 43 */ "log",
112177	/* 44 */ "checkpointed",
112178	/* 45 / "timeout", / Used by: busy_timeout */
112179	/* 46 / "database", / Used by: lock_status */
112180	/* 47 */ "status",

112181	};
112182
112183	/* Definitions of all built-in pragmas */
112184	typedef struct PragmaName {
112185	const char const zName; / Name of pragma */
	@@ -112221,11 +112472,11 @@
112221	#endif
112222	#endif
112223	{/* zName: */ "busy_timeout",
112224	/* ePragTyp: */ PragTyp_BUSY_TIMEOUT,
112225	/* ePragFlg: */ PragFlg_Result0,
112226	/* ColNames: */ 45, 1,
112227	/* iArg: */ 0 },
112228	#if !defined(SQLITE_OMIT_PAGER_PRAGMAS)
112229	{/* zName: */ "cache_size",
112230	/* ePragTyp: */ PragTyp_CACHE_SIZE,
112231	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result0\|PragFlg_SchemaReq\|PragFlg_NoColumns1,
	@@ -112258,11 +112509,11 @@
112258	#endif
112259	#if !defined(SQLITE_OMIT_SCHEMA_PRAGMAS)
112260	{/* zName: */ "collation_list",
112261	/* ePragTyp: */ PragTyp_COLLATION_LIST,
112262	/* ePragFlg: */ PragFlg_Result0,
112263	/* ColNames: */ 28, 2,
112264	/* iArg: */ 0 },
112265	#endif
112266	#if !defined(SQLITE_OMIT_COMPILEOPTION_DIAGS)
112267	{/* zName: */ "compile_options",
112268	/* ePragTyp: */ PragTyp_COMPILE_OPTIONS,
	@@ -112293,11 +112544,11 @@
112293	#endif
112294	#if !defined(SQLITE_OMIT_SCHEMA_PRAGMAS)
112295	{/* zName: */ "database_list",
112296	/* ePragTyp: */ PragTyp_DATABASE_LIST,
112297	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result0,
112298	/* ColNames: */ 25, 3,
112299	/* iArg: */ 0 },
112300	#endif
112301	#if !defined(SQLITE_OMIT_PAGER_PRAGMAS) && !defined(SQLITE_OMIT_DEPRECATED)
112302	{/* zName: */ "default_cache_size",
112303	/* ePragTyp: */ PragTyp_DEFAULT_CACHE_SIZE,
	@@ -112330,18 +112581,18 @@
112330	#endif
112331	#if !defined(SQLITE_OMIT_FOREIGN_KEY) && !defined(SQLITE_OMIT_TRIGGER)
112332	{/* zName: */ "foreign_key_check",
112333	/* ePragTyp: */ PragTyp_FOREIGN_KEY_CHECK,
112334	/* ePragFlg: */ PragFlg_NeedSchema,
112335	/* ColNames: */ 38, 4,
112336	/* iArg: */ 0 },
112337	#endif
112338	#if !defined(SQLITE_OMIT_FOREIGN_KEY)
112339	{/* zName: */ "foreign_key_list",
112340	/* ePragTyp: */ PragTyp_FOREIGN_KEY_LIST,
112341	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result1\|PragFlg_SchemaOpt,
112342	/* ColNames: */ 30, 8,
112343	/* iArg: */ 0 },
112344	#endif
112345	#if !defined(SQLITE_OMIT_FLAG_PRAGMAS)
112346	#if !defined(SQLITE_OMIT_FOREIGN_KEY) && !defined(SQLITE_OMIT_TRIGGER)
112347	{/* zName: */ "foreign_keys",
	@@ -112400,21 +112651,21 @@
112400	#endif
112401	#if !defined(SQLITE_OMIT_SCHEMA_PRAGMAS)
112402	{/* zName: */ "index_info",
112403	/* ePragTyp: */ PragTyp_INDEX_INFO,
112404	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result1\|PragFlg_SchemaOpt,
112405	/* ColNames: */ 11, 3,
112406	/* iArg: */ 0 },
112407	{/* zName: */ "index_list",
112408	/* ePragTyp: */ PragTyp_INDEX_LIST,
112409	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result1\|PragFlg_SchemaOpt,
112410	/* ColNames: */ 20, 5,
112411	/* iArg: */ 0 },
112412	{/* zName: */ "index_xinfo",
112413	/* ePragTyp: */ PragTyp_INDEX_INFO,
112414	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result1\|PragFlg_SchemaOpt,
112415	/* ColNames: */ 14, 6,
112416	/* iArg: */ 1 },
112417	#endif
112418	#if !defined(SQLITE_OMIT_INTEGRITY_CHECK)
112419	{/* zName: */ "integrity_check",
112420	/* ePragTyp: */ PragTyp_INTEGRITY_CHECK,
	@@ -112457,11 +112708,11 @@
112457	#endif
112458	#if defined(SQLITE_DEBUG) \|\| defined(SQLITE_TEST)
112459	{/* zName: */ "lock_status",
112460	/* ePragTyp: */ PragTyp_LOCK_STATUS,
112461	/* ePragFlg: */ PragFlg_Result0,
112462	/* ColNames: */ 46, 2,
112463	/* iArg: */ 0 },
112464	#endif
112465	#if !defined(SQLITE_OMIT_PAGER_PRAGMAS)
112466	{/* zName: */ "locking_mode",
112467	/* ePragTyp: */ PragTyp_LOCKING_MODE,
	@@ -112476,10 +112727,17 @@
112476	{/* zName: */ "mmap_size",
112477	/* ePragTyp: */ PragTyp_MMAP_SIZE,
112478	/* ePragFlg: */ 0,
112479	/* ColNames: */ 0, 0,
112480	/* iArg: */ 0 },







112481	{/* zName: */ "page_count",
112482	/* ePragTyp: */ PragTyp_PAGE_COUNT,
112483	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result0\|PragFlg_SchemaReq,
112484	/* ColNames: */ 0, 0,
112485	/* iArg: */ 0 },
	@@ -112574,15 +112832,15 @@
112574	/* ePragFlg: */ PragFlg_Result0\|PragFlg_NoColumns1,
112575	/* ColNames: */ 0, 0,
112576	/* iArg: */ SQLITE_SqlTrace },
112577	#endif
112578	#endif
112579	#if !defined(SQLITE_OMIT_SCHEMA_PRAGMAS)
112580	{/* zName: */ "stats",
112581	/* ePragTyp: */ PragTyp_STATS,
112582	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result0\|PragFlg_SchemaReq,
112583	/* ColNames: */ 7, 4,
112584	/* iArg: */ 0 },
112585	#endif
112586	#if !defined(SQLITE_OMIT_PAGER_PRAGMAS)
112587	{/* zName: */ "synchronous",
112588	/* ePragTyp: */ PragTyp_SYNCHRONOUS,
	@@ -112657,11 +112915,11 @@
112657	/* ColNames: */ 0, 0,
112658	/* iArg: */ 0 },
112659	{/* zName: */ "wal_checkpoint",
112660	/* ePragTyp: */ PragTyp_WAL_CHECKPOINT,
112661	/* ePragFlg: */ PragFlg_NeedSchema,
112662	/* ColNames: */ 42, 3,
112663	/* iArg: */ 0 },
112664	#endif
112665	#if !defined(SQLITE_OMIT_FLAG_PRAGMAS)
112666	{/* zName: */ "writable_schema",
112667	/* ePragTyp: */ PragTyp_FLAG,
	@@ -112668,11 +112926,11 @@
112668	/* ePragFlg: */ PragFlg_Result0\|PragFlg_NoColumns1,
112669	/* ColNames: */ 0, 0,
112670	/* iArg: */ SQLITE_WriteSchema\|SQLITE_RecoveryMode },
112671	#endif
112672	};
112673	/* Number of pragmas: 60 on by default, 73 total. */
112674
112675	/************ End of pragma.h ********************************************/
112676	/************ Continuing where we left off in pragma.c *******************/
112677
112678	/*
	@@ -112937,10 +113195,26 @@
112937	lwr = mid + 1;
112938	}
112939	}
112940	return lwr>upr ? 0 : &aPragmaName[mid];
112941	}
















112942
112943	/*
112944	** Process a pragma statement.
112945	**
112946	** Pragmas are of this form:
	@@ -113741,33 +114015,37 @@
113741	}
113742	}
113743	}
113744	break;
113745

113746	case PragTyp_STATS: {
113747	Index *pIdx;
113748	HashElem *i;
113749	pParse->nMem = 4;
113750	sqlite3CodeVerifySchema(pParse, iDb);
113751	for(i=sqliteHashFirst(&pDb->pSchema->tblHash); i; i=sqliteHashNext(i)){
113752	Table *pTab = sqliteHashData(i);
113753	sqlite3VdbeMultiLoad(v, 1, "ssii",
113754	pTab->zName,
113755	0,
113756	pTab->szTabRow,
113757	pTab->nRowLogEst);
113758	sqlite3VdbeAddOp2(v, OP_ResultRow, 1, 4);

113759	for(pIdx=pTab->pIndex; pIdx; pIdx=pIdx->pNext){
113760	sqlite3VdbeMultiLoad(v, 2, "sii",
113761	pIdx->zName,
113762	pIdx->szIdxRow,
113763	pIdx->aiRowLogEst[0]);
113764	sqlite3VdbeAddOp2(v, OP_ResultRow, 1, 4);

113765	}
113766	}
113767	}
113768	break;

113769
113770	case PragTyp_INDEX_INFO: if( zRight ){
113771	Index *pIdx;
113772	Table *pTab;
113773	pIdx = sqlite3FindIndex(db, zRight, zDb);
	@@ -114020,13 +114298,21 @@
114020	#ifndef SQLITE_INTEGRITY_CHECK_ERROR_MAX
114021	# define SQLITE_INTEGRITY_CHECK_ERROR_MAX 100
114022	#endif
114023
114024	#ifndef SQLITE_OMIT_INTEGRITY_CHECK
114025	/* Pragma "quick_check" is reduced version of







114026	** integrity_check designed to detect most database corruption
114027	** without most of the overhead of a full integrity-check.

114028	*/
114029	case PragTyp_INTEGRITY_CHECK: {
114030	int i, j, addr, mxErr;
114031
114032	int isQuick = (sqlite3Tolower(zLeft[0])=='q');
	@@ -114053,11 +114339,11 @@
114053	sqlite3GetInt32(zRight, &mxErr);
114054	if( mxErr<=0 ){
114055	mxErr = SQLITE_INTEGRITY_CHECK_ERROR_MAX;
114056	}
114057	}
114058	sqlite3VdbeAddOp2(v, OP_Integer, mxErr, 1); /* reg[1] holds errors left */
114059
114060	/* Do an integrity check on each database file */
114061	for(i=0; i<db->nDb; i++){
114062	HashElem *x;
114063	Hash *pTbls;
	@@ -114068,14 +114354,10 @@
114068
114069	if( OMIT_TEMPDB && i==1 ) continue;
114070	if( iDb>=0 && i!=iDb ) continue;
114071
114072	sqlite3CodeVerifySchema(pParse, i);
114073	addr = sqlite3VdbeAddOp1(v, OP_IfPos, 1); /* Halt if out of errors */
114074	VdbeCoverage(v);
114075	sqlite3VdbeAddOp2(v, OP_Halt, 0, 0);
114076	sqlite3VdbeJumpHere(v, addr);
114077
114078	/* Do an integrity check of the B-Tree
114079	**
114080	** Begin by finding the root pages numbers
114081	** for all tables and indices in the database.
	@@ -114111,29 +114393,30 @@
114111	sqlite3VdbeAddOp4(v, OP_String8, 0, 3, 0,
114112	sqlite3MPrintf(db, "* in database %s *\n", db->aDb[i].zDbSName),
114113	P4_DYNAMIC);
114114	sqlite3VdbeAddOp3(v, OP_Move, 2, 4, 1);
114115	sqlite3VdbeAddOp3(v, OP_Concat, 4, 3, 2);
114116	sqlite3VdbeAddOp2(v, OP_ResultRow, 2, 1);
114117	sqlite3VdbeJumpHere(v, addr);
114118
114119	/* Make sure all the indices are constructed correctly.
114120	*/
114121	for(x=sqliteHashFirst(pTbls); x && !isQuick; x=sqliteHashNext(x)){
114122	Table *pTab = sqliteHashData(x);
114123	Index pIdx, pPk;
114124	Index *pPrior = 0;
114125	int loopTop;
114126	int iDataCur, iIdxCur;
114127	int r1 = -1;
114128
114129	if( pTab->pIndex==0 ) continue;





114130	pPk = HasRowid(pTab) ? 0 : sqlite3PrimaryKeyIndex(pTab);
114131	addr = sqlite3VdbeAddOp1(v, OP_IfPos, 1); /* Stop if out of errors */
114132	VdbeCoverage(v);
114133	sqlite3VdbeAddOp2(v, OP_Halt, 0, 0);
114134	sqlite3VdbeJumpHere(v, addr);
114135	sqlite3ExprCacheClear(pParse);
114136	sqlite3OpenTableAndIndices(pParse, pTab, OP_OpenRead, 0,
114137	1, 0, &iDataCur, &iIdxCur);
114138	sqlite3VdbeAddOp2(v, OP_Integer, 0, 7);
114139	for(j=0, pIdx=pTab->pIndex; pIdx; pIdx=pIdx->pNext, j++){
	@@ -114144,28 +114427,46 @@
114144	sqlite3VdbeAddOp2(v, OP_Rewind, iDataCur, 0); VdbeCoverage(v);
114145	loopTop = sqlite3VdbeAddOp2(v, OP_AddImm, 7, 1);
114146	/* Verify that all NOT NULL columns really are NOT NULL */
114147	for(j=0; j<pTab->nCol; j++){
114148	char *zErr;
114149	int jmp2, jmp3;
114150	if( j==pTab->iPKey ) continue;
114151	if( pTab->aCol[j].notNull==0 ) continue;
114152	sqlite3ExprCodeGetColumnOfTable(v, pTab, iDataCur, j, 3);
114153	sqlite3VdbeChangeP5(v, OPFLAG_TYPEOFARG);
114154	jmp2 = sqlite3VdbeAddOp1(v, OP_NotNull, 3); VdbeCoverage(v);
114155	sqlite3VdbeAddOp2(v, OP_AddImm, 1, -1); /* Decrement error limit */
114156	zErr = sqlite3MPrintf(db, "NULL value in %s.%s", pTab->zName,
114157	pTab->aCol[j].zName);
114158	sqlite3VdbeAddOp4(v, OP_String8, 0, 3, 0, zErr, P4_DYNAMIC);
114159	sqlite3VdbeAddOp2(v, OP_ResultRow, 3, 1);
114160	jmp3 = sqlite3VdbeAddOp1(v, OP_IfPos, 1); VdbeCoverage(v);
114161	sqlite3VdbeAddOp0(v, OP_Halt);
114162	sqlite3VdbeJumpHere(v, jmp2);
114163	sqlite3VdbeJumpHere(v, jmp3);





















114164	}
114165	/* Validate index entries for the current row */
114166	for(j=0, pIdx=pTab->pIndex; pIdx; pIdx=pIdx->pNext, j++){
114167	int jmp2, jmp3, jmp4, jmp5;
114168	int ckUniq = sqlite3VdbeMakeLabel(v);
114169	if( pPk==pIdx ) continue;
114170	r1 = sqlite3GenerateIndexKey(pParse, pIdx, iDataCur, 0, 0, &jmp3,
114171	pPrior, r1);
	@@ -114172,20 +114473,17 @@
114172	pPrior = pIdx;
114173	sqlite3VdbeAddOp2(v, OP_AddImm, 8+j, 1); /* increment entry count */
114174	/* Verify that an index entry exists for the current table row */
114175	jmp2 = sqlite3VdbeAddOp4Int(v, OP_Found, iIdxCur+j, ckUniq, r1,
114176	pIdx->nColumn); VdbeCoverage(v);
114177	sqlite3VdbeAddOp2(v, OP_AddImm, 1, -1); /* Decrement error limit */
114178	sqlite3VdbeLoadString(v, 3, "row ");
114179	sqlite3VdbeAddOp3(v, OP_Concat, 7, 3, 3);
114180	sqlite3VdbeLoadString(v, 4, " missing from index ");
114181	sqlite3VdbeAddOp3(v, OP_Concat, 4, 3, 3);
114182	jmp5 = sqlite3VdbeLoadString(v, 4, pIdx->zName);
114183	sqlite3VdbeAddOp3(v, OP_Concat, 4, 3, 3);
114184	sqlite3VdbeAddOp2(v, OP_ResultRow, 3, 1);
114185	jmp4 = sqlite3VdbeAddOp1(v, OP_IfPos, 1); VdbeCoverage(v);
114186	sqlite3VdbeAddOp0(v, OP_Halt);
114187	sqlite3VdbeJumpHere(v, jmp2);
114188	/* For UNIQUE indexes, verify that only one entry exists with the
114189	** current key. The entry is unique if (1) any column is NULL
114190	** or (2) the next entry has a different key */
114191	if( IsUniqueIndex(pIdx) ){
	@@ -114202,11 +114500,10 @@
114202	jmp6 = sqlite3VdbeAddOp1(v, OP_Next, iIdxCur+j); VdbeCoverage(v);
114203	sqlite3VdbeGoto(v, uniqOk);
114204	sqlite3VdbeJumpHere(v, jmp6);
114205	sqlite3VdbeAddOp4Int(v, OP_IdxGT, iIdxCur+j, uniqOk, r1,
114206	pIdx->nKeyCol); VdbeCoverage(v);
114207	sqlite3VdbeAddOp2(v, OP_AddImm, 1, -1); /* Decrement error limit */
114208	sqlite3VdbeLoadString(v, 3, "non-unique entry in index ");
114209	sqlite3VdbeGoto(v, jmp5);
114210	sqlite3VdbeResolveLabel(v, uniqOk);
114211	}
114212	sqlite3VdbeJumpHere(v, jmp4);
	@@ -114213,40 +114510,39 @@
114213	sqlite3ResolvePartIdxLabel(pParse, jmp3);
114214	}
114215	sqlite3VdbeAddOp2(v, OP_Next, iDataCur, loopTop); VdbeCoverage(v);
114216	sqlite3VdbeJumpHere(v, loopTop-1);
114217	#ifndef SQLITE_OMIT_BTREECOUNT
114218	sqlite3VdbeLoadString(v, 2, "wrong # of entries in index ");
114219	for(j=0, pIdx=pTab->pIndex; pIdx; pIdx=pIdx->pNext, j++){
114220	if( pPk==pIdx ) continue;
114221	addr = sqlite3VdbeCurrentAddr(v);
114222	sqlite3VdbeAddOp2(v, OP_IfPos, 1, addr+2); VdbeCoverage(v);
114223	sqlite3VdbeAddOp2(v, OP_Halt, 0, 0);
114224	sqlite3VdbeAddOp2(v, OP_Count, iIdxCur+j, 3);
114225	sqlite3VdbeAddOp3(v, OP_Eq, 8+j, addr+8, 3); VdbeCoverage(v);
114226	sqlite3VdbeChangeP5(v, SQLITE_NOTNULL);
114227	sqlite3VdbeAddOp2(v, OP_AddImm, 1, -1);
114228	sqlite3VdbeLoadString(v, 3, pIdx->zName);
114229	sqlite3VdbeAddOp3(v, OP_Concat, 3, 2, 7);
114230	sqlite3VdbeAddOp2(v, OP_ResultRow, 7, 1);
114231	}
114232	#endif /* SQLITE_OMIT_BTREECOUNT */
114233	}
114234	}
114235	{
114236	static const int iLn = VDBE_OFFSET_LINENO(2);
114237	static const VdbeOpList endCode[] = {
114238	{ OP_AddImm, 1, 0, 0}, /* 0 */
114239	{ OP_If, 1, 4, 0}, /* 1 */
114240	{ OP_String8, 0, 3, 0}, /* 2 */
114241	{ OP_ResultRow, 3, 1, 0}, /* 3 */
114242	};
114243	VdbeOp *aOp;
114244
114245	aOp = sqlite3VdbeAddOpList(v, ArraySize(endCode), endCode, iLn);
114246	if( aOp ){
114247	aOp[0].p2 = -mxErr;
114248	aOp[2].p4type = P4_STATIC;
114249	aOp[2].p4.z = "ok";
114250	}
114251	}
114252	}
	@@ -114466,10 +114762,122 @@
114466	*/
114467	case PragTyp_SHRINK_MEMORY: {
114468	sqlite3_db_release_memory(db);
114469	break;
114470	}
















































































































114471
114472	/*
114473	** PRAGMA busy_timeout
114474	** PRAGMA busy_timeout = N
114475	**
	@@ -123782,12 +124190,29 @@
123782	** transient would cause the database file to appear to be deleted
123783	** following reboot.
123784	*/
123785	SQLITE_PRIVATE void sqlite3Vacuum(Parse pParse, Token pNm){
123786	Vdbe *v = sqlite3GetVdbe(pParse);
123787	int iDb = pNm ? sqlite3TwoPartName(pParse, pNm, pNm, &pNm) : 0;
123788	if( v && (iDb>=2 \|\| iDb==0) ){

















123789	sqlite3VdbeAddOp1(v, OP_Vacuum, iDb);
123790	sqlite3VdbeUsesBtree(v, iDb);
123791	}
123792	return;
123793	}
	@@ -124377,12 +124802,11 @@
124377
124378	db = pParse->db;
124379	iDb = sqlite3SchemaToIndex(db, pTable->pSchema);
124380	assert( iDb>=0 );
124381
124382	pTable->tabFlags \|= TF_Virtual;
124383	pTable->nModuleArg = 0;
124384	addModuleArgument(db, pTable, sqlite3NameFromToken(db, pModuleName));
124385	addModuleArgument(db, pTable, 0);
124386	addModuleArgument(db, pTable, sqlite3DbStrDup(db, pTable->zName));
124387	assert( (pParse->sNameToken.z==pName2->z && pName2->z!=0)
124388	\|\| (pParse->sNameToken.z==pName1->z && pName2->z==0)
	@@ -124666,11 +125090,11 @@
124666	const char *zMod;
124667	Module *pMod;
124668	int rc;
124669
124670	assert( pTab );
124671	if( (pTab->tabFlags & TF_Virtual)==0 \|\| sqlite3GetVTable(db, pTab) ){
124672	return SQLITE_OK;
124673	}
124674
124675	/* Locate the required virtual table module */
124676	zMod = pTab->azModuleArg[0];
	@@ -124736,11 +125160,11 @@
124736	Table *pTab;
124737	Module *pMod;
124738	const char *zMod;
124739
124740	pTab = sqlite3FindTable(db, zTab, db->aDb[iDb].zDbSName);
124741	assert( pTab && (pTab->tabFlags & TF_Virtual)!=0 && !pTab->pVTable );
124742
124743	/* Locate the required virtual table module */
124744	zMod = pTab->azModuleArg[0];
124745	pMod = (Module*)sqlite3HashFind(&db->aModule, zMod);
124746
	@@ -124790,11 +125214,11 @@
124790	sqlite3Error(db, SQLITE_MISUSE);
124791	sqlite3_mutex_leave(db->mutex);
124792	return SQLITE_MISUSE_BKPT;
124793	}
124794	pTab = pCtx->pTab;
124795	assert( (pTab->tabFlags & TF_Virtual)!=0 );
124796
124797	pParse = sqlite3StackAllocZero(db, sizeof(*pParse));
124798	if( pParse==0 ){
124799	rc = SQLITE_NOMEM_BKPT;
124800	}else{
	@@ -124804,11 +125228,11 @@
124804
124805	if( SQLITE_OK==sqlite3RunParser(pParse, zCreateTable, &zErr)
124806	&& pParse->pNewTable
124807	&& !db->mallocFailed
124808	&& !pParse->pNewTable->pSelect
124809	&& (pParse->pNewTable->tabFlags & TF_Virtual)==0
124810	){
124811	if( !pTab->aCol ){
124812	Table *pNew = pParse->pNewTable;
124813	Index *pIdx;
124814	pTab->aCol = pNew->aCol;
	@@ -125093,11 +125517,11 @@
125093	/* Check to see the left operand is a column in a virtual table */
125094	if( NEVER(pExpr==0) ) return pDef;
125095	if( pExpr->op!=TK_COLUMN ) return pDef;
125096	pTab = pExpr->pTab;
125097	if( NEVER(pTab==0) ) return pDef;
125098	if( (pTab->tabFlags & TF_Virtual)==0 ) return pDef;
125099	pVtab = sqlite3GetVTable(db, pTab)->pVtab;
125100	assert( pVtab!=0 );
125101	assert( pVtab->pModule!=0 );
125102	pMod = (sqlite3_module *)pVtab->pModule;
125103	if( pMod->xFindFunction==0 ) return pDef;
	@@ -125188,12 +125612,11 @@
125188	return 0;
125189	}
125190	pMod->pEpoTab = pTab;
125191	pTab->nTabRef = 1;
125192	pTab->pSchema = db->aDb[0].pSchema;
125193	pTab->tabFlags \|= TF_Virtual;
125194	pTab->nModuleArg = 0;
125195	pTab->iPKey = -1;
125196	addModuleArgument(db, pTab, sqlite3DbStrDup(db, pTab->zName));
125197	addModuleArgument(db, pTab, 0);
125198	addModuleArgument(db, pTab, sqlite3DbStrDup(db, pTab->zName));
125199	rc = vtabCallConstructor(db, pTab, pMod, pModule->xConnect, &zErr);
	@@ -125260,11 +125683,11 @@
125260	case SQLITE_VTAB_CONSTRAINT_SUPPORT: {
125261	VtabCtx *p = db->pVtabCtx;
125262	if( !p ){
125263	rc = SQLITE_MISUSE_BKPT;
125264	}else{
125265	assert( p->pTab==0 \|\| (p->pTab->tabFlags & TF_Virtual)!=0 );
125266	p->pVTable->bConstraint = (u8)va_arg(ap, int);
125267	}
125268	break;
125269	}
125270	default:
	@@ -125699,12 +126122,17 @@
125699	WhereOrSet pOrSet; / Record best loops here, if not NULL */
125700	#ifdef SQLITE_ENABLE_STAT3_OR_STAT4
125701	UnpackedRecord pRec; / Probe for stat4 (if required) */
125702	int nRecValid; /* Number of valid fields currently in pRec */
125703	#endif

125704	};
125705




125706	/*
125707	** The WHERE clause processing routine has two halves. The
125708	** first part does the start of the WHERE loop and the second
125709	** half does the tail of the WHERE loop. An instance of
125710	** this structure is returned by the first half and passed
	@@ -125715,11 +126143,11 @@
125715	*/
125716	struct WhereInfo {
125717	Parse pParse; / Parsing and code generating context */
125718	SrcList pTabList; / List of tables in the join */
125719	ExprList pOrderBy; / The ORDER BY clause or NULL */
125720	ExprList pDistinctSet; / DISTINCT over all these values */
125721	LogEst iLimit; /* LIMIT if wctrlFlags has WHERE_USE_LIMIT */
125722	int aiCurOnePass[2]; /* OP_OpenWrite cursors for the ONEPASS opt */
125723	int iContinue; /* Jump here to continue with next record */
125724	int iBreak; /* Jump here to break out of the loop */
125725	int savedNQueryLoop; /* pParse->nQueryLoop outside the WHERE loop */
	@@ -126899,10 +127327,11 @@
126899	Parse pParse; / Parsing context */
126900	sqlite3 db; / Database connection */
126901	Vdbe v; / The prepared stmt under constructions */
126902	struct SrcList_item pTabItem; / FROM clause term being coded */
126903	int addrBrk; /* Jump here to break out of the loop */

126904	int addrCont; /* Jump here to continue with next cycle */
126905	int iRowidReg = 0; /* Rowid is stored in this register, if not zero */
126906	int iReleaseReg = 0; /* Temp register to free before returning */
126907
126908	pParse = pWInfo->pParse;
	@@ -126939,10 +127368,15 @@
126939	if( pLevel->iFrom>0 && (pTabItem[0].fg.jointype & JT_LEFT)!=0 ){
126940	pLevel->iLeftJoin = ++pParse->nMem;
126941	sqlite3VdbeAddOp2(v, OP_Integer, 0, pLevel->iLeftJoin);
126942	VdbeComment((v, "init LEFT JOIN no-match flag"));
126943	}





126944
126945	/* Special case of a FROM clause subquery implemented as a co-routine */
126946	if( pTabItem->fg.viaCoroutine ){
126947	int regYield = pTabItem->regReturn;
126948	sqlite3VdbeAddOp3(v, OP_InitCoroutine, regYield, 0, pTabItem->addrFillSub);
	@@ -127124,11 +127558,11 @@
127124	VdbeCoverageIf(v, pX->op==TK_LT);
127125	VdbeCoverageIf(v, pX->op==TK_GE);
127126	sqlite3ExprCacheAffinityChange(pParse, r1, 1);
127127	sqlite3ReleaseTempReg(pParse, rTemp);
127128	}else{
127129	sqlite3VdbeAddOp2(v, bRev ? OP_Last : OP_Rewind, iCur, addrBrk);
127130	VdbeCoverageIf(v, bRev==0);
127131	VdbeCoverageIf(v, bRev!=0);
127132	}
127133	if( pEnd ){
127134	Expr *pX;
	@@ -127770,11 +128204,11 @@
127770	pLevel->op = OP_Noop;
127771	}else{
127772	codeCursorHint(pTabItem, pWInfo, pLevel, 0);
127773	pLevel->op = aStep[bRev];
127774	pLevel->p1 = iCur;
127775	pLevel->p2 = 1 + sqlite3VdbeAddOp2(v, aStart[bRev], iCur, addrBrk);
127776	VdbeCoverageIf(v, bRev==0);
127777	VdbeCoverageIf(v, bRev!=0);
127778	pLevel->p5 = SQLITE_STMTSTATUS_FULLSCAN_STEP;
127779	}
127780	}
	@@ -128095,19 +128529,10 @@
128095	#ifdef SQLITE_EBCDIC
128096	if( *pnoCase ) return 0;
128097	#endif
128098	pList = pExpr->x.pList;
128099	pLeft = pList->a[1].pExpr;
128100	if( pLeft->op!=TK_COLUMN
128101	\|\| sqlite3ExprAffinity(pLeft)!=SQLITE_AFF_TEXT
128102	\|\| IsVirtual(pLeft->pTab) /* Value might be numeric */
128103	){
128104	/* IMP: R-02065-49465 The left-hand side of the LIKE or GLOB operator must
128105	** be the name of an indexed column with TEXT affinity. */
128106	return 0;
128107	}
128108	assert( pLeft->iColumn!=(-1) ); /* Because IPK never has AFF_TEXT */
128109
128110	pRight = sqlite3ExprSkipCollate(pList->a[0].pExpr);
128111	op = pRight->op;
128112	if( op==TK_VARIABLE ){
128113	Vdbe *pReprepare = pParse->pReprepare;
	@@ -128120,10 +128545,27 @@
128120	assert( pRight->op==TK_VARIABLE \|\| pRight->op==TK_REGISTER );
128121	}else if( op==TK_STRING ){
128122	z = pRight->u.zToken;
128123	}
128124	if( z ){

















128125	cnt = 0;
128126	while( (c=z[cnt])!=0 && c!=wc[0] && c!=wc[1] && c!=wc[2] ){
128127	cnt++;
128128	}
128129	if( cnt!=0 && 255!=(u8)z[cnt-1] ){
	@@ -128748,11 +129190,11 @@
128748	iCur = pFrom->a[i].iCursor;
128749	for(pIdx=pFrom->a[i].pTab->pIndex; pIdx; pIdx=pIdx->pNext){
128750	if( pIdx->aColExpr==0 ) continue;
128751	for(i=0; i<pIdx->nKeyCol; i++){
128752	if( pIdx->aiColumn[i]!=XN_EXPR ) continue;
128753	if( sqlite3ExprCompare(pExpr, pIdx->aColExpr->a[i].pExpr, iCur)==0 ){
128754	*piCur = iCur;
128755	*piColumn = XN_EXPR;
128756	return 1;
128757	}
128758	}
	@@ -129539,11 +129981,12 @@
129539	do{
129540	for(pTerm=pWC->a+k; k<pWC->nTerm; k++, pTerm++){
129541	if( pTerm->leftCursor==iCur
129542	&& pTerm->u.leftColumn==iColumn
129543	&& (iColumn!=XN_EXPR
129544	\|\| sqlite3ExprCompare(pTerm->pExpr->pLeft,pScan->pIdxExpr,iCur)==0)

129545	&& (pScan->iEquiv<=1 \|\| !ExprHasProperty(pTerm->pExpr, EP_FromJoin))
129546	){
129547	if( (pTerm->eOperator & WO_EQUIV)!=0
129548	&& pScan->nEquiv<ArraySize(pScan->aiCur)
129549	&& (pX = sqlite3ExprSkipCollate(pTerm->pExpr->pRight))->op==TK_COLUMN
	@@ -131716,10 +132159,15 @@
131716	testcase( eOp & WO_IS );
131717	testcase( eOp & WO_ISNULL );
131718	continue;
131719	}
131720





131721	pNew->wsFlags = saved_wsFlags;
131722	pNew->u.btree.nEq = saved_nEq;
131723	pNew->u.btree.nBtm = saved_nBtm;
131724	pNew->u.btree.nTop = saved_nTop;
131725	pNew->nLTerm = saved_nLTerm;
	@@ -132263,11 +132711,19 @@
132263	pNew->nOut = rSize;
132264	if( rc ) break;
132265	}
132266	}
132267

132268	rc = whereLoopAddBtreeIndex(pBuilder, pSrc, pProbe, 0);







132269	#ifdef SQLITE_ENABLE_STAT3_OR_STAT4
132270	sqlite3Stat4ProbeFree(pBuilder->pRec);
132271	pBuilder->nRecValid = 0;
132272	pBuilder->pRec = 0;
132273	#endif
	@@ -133443,13 +133899,13 @@
133443	&& (pWInfo->wctrlFlags & WHERE_DISTINCTBY)==0
133444	&& pWInfo->eDistinct==WHERE_DISTINCT_NOOP
133445	&& nRowEst
133446	){
133447	Bitmask notUsed;
133448	int rc = wherePathSatisfiesOrderBy(pWInfo, pWInfo->pDistinctSet, pFrom,
133449	WHERE_DISTINCTBY, nLoop-1, pFrom->aLoop[nLoop-1], &notUsed);
133450	if( rc==pWInfo->pDistinctSet->nExpr ){
133451	pWInfo->eDistinct = WHERE_DISTINCT_ORDERED;
133452	}
133453	}
133454	if( pWInfo->pOrderBy ){
133455	if( pWInfo->wctrlFlags & WHERE_DISTINCTBY ){
	@@ -133682,11 +134138,11 @@
133682	SQLITE_PRIVATE WhereInfo *sqlite3WhereBegin(
133683	Parse pParse, / The parser context */
133684	SrcList pTabList, / FROM clause: A list of all tables to be scanned */
133685	Expr pWhere, / The WHERE clause */
133686	ExprList pOrderBy, / An ORDER BY (or GROUP BY) clause, or NULL */
133687	ExprList pDistinctSet, / Try not to output two rows that duplicate these */
133688	u16 wctrlFlags, /* The WHERE_* flags defined in sqliteInt.h */
133689	int iAuxArg /* If WHERE_OR_SUBCLAUSE is set, index cursor number
133690	** If WHERE_USE_LIMIT, then the limit amount */
133691	){
133692	int nByteWInfo; /* Num. bytes allocated for WhereInfo struct */
	@@ -133758,11 +134214,11 @@
133758	goto whereBeginError;
133759	}
133760	pWInfo->pParse = pParse;
133761	pWInfo->pTabList = pTabList;
133762	pWInfo->pOrderBy = pOrderBy;
133763	pWInfo->pDistinctSet = pDistinctSet;
133764	pWInfo->aiCurOnePass[0] = pWInfo->aiCurOnePass[1] = -1;
133765	pWInfo->nLevel = nTabList;
133766	pWInfo->iBreak = pWInfo->iContinue = sqlite3VdbeMakeLabel(v);
133767	pWInfo->wctrlFlags = wctrlFlags;
133768	pWInfo->iLimit = iAuxArg;
	@@ -133836,17 +134292,17 @@
133836	/* Analyze all of the subexpressions. */
133837	sqlite3WhereExprAnalyze(pTabList, &pWInfo->sWC);
133838	if( db->mallocFailed ) goto whereBeginError;
133839
133840	if( wctrlFlags & WHERE_WANT_DISTINCT ){
133841	if( isDistinctRedundant(pParse, pTabList, &pWInfo->sWC, pDistinctSet) ){
133842	/* The DISTINCT marking is pointless. Ignore it. */
133843	pWInfo->eDistinct = WHERE_DISTINCT_UNIQUE;
133844	}else if( pOrderBy==0 ){
133845	/* Try to ORDER BY the result set to make distinct processing easier */
133846	pWInfo->wctrlFlags \|= WHERE_DISTINCTBY;
133847	pWInfo->pOrderBy = pDistinctSet;
133848	}
133849	}
133850
133851	/* Construct the WhereLoop objects */
133852	#if defined(WHERETRACE_ENABLED)
	@@ -133918,14 +134374,14 @@
133918	}
133919	}
133920	#endif
133921	/* Attempt to omit tables from the join that do not effect the result */
133922	if( pWInfo->nLevel>=2
133923	&& pDistinctSet!=0
133924	&& OptimizationEnabled(db, SQLITE_OmitNoopJoin)
133925	){
133926	Bitmask tabUsed = sqlite3WhereExprListUsage(pMaskSet, pDistinctSet);
133927	if( sWLB.pOrderBy ){
133928	tabUsed \|= sqlite3WhereExprListUsage(pMaskSet, sWLB.pOrderBy);
133929	}
133930	while( pWInfo->nLevel>=2 ){
133931	WhereTerm pTerm, pEnd;
	@@ -134704,166 +135160,166 @@
134704	**
134705	********* Begin parsing tables ********************************************/
134706	#define YY_ACTTAB_COUNT (1567)
134707	static const YYACTIONTYPE yy_action[] = {
134708	/* 0 */ 325, 832, 351, 825, 5, 203, 203, 819, 99, 100,
134709	/* 10 */ 90, 842, 842, 854, 857, 846, 846, 97, 97, 98,
134710	/* 20 */ 98, 98, 98, 301, 96, 96, 96, 96, 95, 95,
134711	/* 30 */ 94, 94, 94, 93, 351, 325, 977, 977, 824, 824,
134712	/* 40 */ 826, 947, 354, 99, 100, 90, 842, 842, 854, 857,
134713	/* 50 */ 846, 846, 97, 97, 98, 98, 98, 98, 338, 96,
134714	/* 60 */ 96, 96, 96, 95, 95, 94, 94, 94, 93, 351,
134715	/* 70 */ 95, 95, 94, 94, 94, 93, 351, 791, 977, 977,
134716	/* 80 */ 325, 94, 94, 94, 93, 351, 792, 75, 99, 100,
134717	/* 90 */ 90, 842, 842, 854, 857, 846, 846, 97, 97, 98,
134718	/* 100 */ 98, 98, 98, 450, 96, 96, 96, 96, 95, 95,
134719	/* 110 */ 94, 94, 94, 93, 351, 1333, 155, 155, 2, 325,
134720	/* 120 */ 275, 146, 132, 52, 52, 93, 351, 99, 100, 90,
134721	/* 130 */ 842, 842, 854, 857, 846, 846, 97, 97, 98, 98,
134722	/* 140 */ 98, 98, 101, 96, 96, 96, 96, 95, 95, 94,
134723	/* 150 */ 94, 94, 93, 351, 958, 958, 325, 268, 428, 413,
134724	/* 160 */ 411, 61, 752, 752, 99, 100, 90, 842, 842, 854,
134725	/* 170 */ 857, 846, 846, 97, 97, 98, 98, 98, 98, 60,
134726	/* 180 */ 96, 96, 96, 96, 95, 95, 94, 94, 94, 93,
134727	/* 190 */ 351, 325, 270, 329, 273, 277, 959, 960, 250, 99,
134728	/* 200 */ 100, 90, 842, 842, 854, 857, 846, 846, 97, 97,
134729	/* 210 */ 98, 98, 98, 98, 301, 96, 96, 96, 96, 95,
134730	/* 220 */ 95, 94, 94, 94, 93, 351, 325, 938, 1326, 698,
134731	/* 230 */ 706, 1326, 242, 412, 99, 100, 90, 842, 842, 854,
134732	/* 240 */ 857, 846, 846, 97, 97, 98, 98, 98, 98, 347,
134733	/* 250 */ 96, 96, 96, 96, 95, 95, 94, 94, 94, 93,
134734	/* 260 */ 351, 325, 938, 1327, 384, 699, 1327, 381, 379, 99,
134735	/* 270 */ 100, 90, 842, 842, 854, 857, 846, 846, 97, 97,
134736	/* 280 */ 98, 98, 98, 98, 701, 96, 96, 96, 96, 95,
134737	/* 290 */ 95, 94, 94, 94, 93, 351, 325, 92, 89, 178,
134738	/* 300 */ 833, 936, 373, 700, 99, 100, 90, 842, 842, 854,
134739	/* 310 */ 857, 846, 846, 97, 97, 98, 98, 98, 98, 375,
134740	/* 320 */ 96, 96, 96, 96, 95, 95, 94, 94, 94, 93,
134741	/* 330 */ 351, 325, 1276, 947, 354, 818, 936, 739, 739, 99,
134742	/* 340 */ 100, 90, 842, 842, 854, 857, 846, 846, 97, 97,
134743	/* 350 */ 98, 98, 98, 98, 230, 96, 96, 96, 96, 95,
134744	/* 360 */ 95, 94, 94, 94, 93, 351, 325, 969, 227, 92,
134745	/* 370 */ 89, 178, 373, 300, 99, 100, 90, 842, 842, 854,
134746	/* 380 */ 857, 846, 846, 97, 97, 98, 98, 98, 98, 921,
134747	/* 390 */ 96, 96, 96, 96, 95, 95, 94, 94, 94, 93,
134748	/* 400 */ 351, 325, 449, 447, 447, 447, 147, 737, 737, 99,
134749	/* 410 */ 100, 90, 842, 842, 854, 857, 846, 846, 97, 97,
134750	/* 420 */ 98, 98, 98, 98, 296, 96, 96, 96, 96, 95,
134751	/* 430 */ 95, 94, 94, 94, 93, 351, 325, 419, 231, 958,
134752	/* 440 */ 958, 158, 25, 422, 99, 100, 90, 842, 842, 854,
134753	/* 450 */ 857, 846, 846, 97, 97, 98, 98, 98, 98, 450,
134754	/* 460 */ 96, 96, 96, 96, 95, 95, 94, 94, 94, 93,
134755	/* 470 */ 351, 443, 224, 224, 420, 958, 958, 962, 325, 52,
134756	/* 480 */ 52, 959, 960, 176, 415, 78, 99, 100, 90, 842,
134757	/* 490 */ 842, 854, 857, 846, 846, 97, 97, 98, 98, 98,
134758	/* 500 */ 98, 379, 96, 96, 96, 96, 95, 95, 94, 94,
134759	/* 510 */ 94, 93, 351, 325, 428, 418, 298, 959, 960, 962,
134760	/* 520 */ 81, 99, 88, 90, 842, 842, 854, 857, 846, 846,
134761	/* 530 */ 97, 97, 98, 98, 98, 98, 717, 96, 96, 96,
134762	/* 540 */ 96, 95, 95, 94, 94, 94, 93, 351, 325, 843,
134763	/* 550 */ 843, 855, 858, 996, 318, 343, 379, 100, 90, 842,
134764	/* 560 */ 842, 854, 857, 846, 846, 97, 97, 98, 98, 98,
134765	/* 570 */ 98, 450, 96, 96, 96, 96, 95, 95, 94, 94,
134766	/* 580 */ 94, 93, 351, 325, 350, 350, 350, 260, 377, 340,
134767	/* 590 */ 929, 52, 52, 90, 842, 842, 854, 857, 846, 846,
134768	/* 600 */ 97, 97, 98, 98, 98, 98, 361, 96, 96, 96,
134769	/* 610 */ 96, 95, 95, 94, 94, 94, 93, 351, 86, 445,
134770	/* 620 */ 847, 3, 1203, 361, 360, 378, 344, 813, 958, 958,
134771	/* 630 */ 1300, 86, 445, 729, 3, 212, 169, 287, 405, 282,
134772	/* 640 */ 404, 199, 232, 450, 300, 760, 83, 84, 280, 245,
134773	/* 650 */ 262, 365, 251, 85, 352, 352, 92, 89, 178, 83,
134774	/* 660 */ 84, 242, 412, 52, 52, 448, 85, 352, 352, 246,
134775	/* 670 */ 959, 960, 194, 455, 670, 402, 399, 398, 448, 243,
134776	/* 680 */ 221, 114, 434, 776, 361, 450, 397, 268, 747, 224,
134777	/* 690 */ 224, 132, 132, 198, 832, 434, 452, 451, 428, 427,
134778	/* 700 */ 819, 415, 734, 713, 132, 52, 52, 832, 268, 452,
134779	/* 710 */ 451, 734, 194, 819, 363, 402, 399, 398, 450, 1271,
134780	/* 720 */ 1271, 23, 958, 958, 86, 445, 397, 3, 228, 429,
134781	/* 730 */ 895, 824, 824, 826, 827, 19, 203, 720, 52, 52,
134782	/* 740 */ 428, 408, 439, 249, 824, 824, 826, 827, 19, 229,
134783	/* 750 */ 403, 153, 83, 84, 761, 177, 241, 450, 721, 85,
134784	/* 760 */ 352, 352, 120, 157, 959, 960, 58, 977, 409, 355,
134785	/* 770 */ 330, 448, 268, 428, 430, 320, 790, 32, 32, 86,
134786	/* 780 */ 445, 776, 3, 341, 98, 98, 98, 98, 434, 96,
134787	/* 790 */ 96, 96, 96, 95, 95, 94, 94, 94, 93, 351,
134788	/* 800 */ 832, 120, 452, 451, 813, 887, 819, 83, 84, 977,
134789	/* 810 */ 813, 132, 410, 920, 85, 352, 352, 132, 407, 789,
134790	/* 820 */ 958, 958, 92, 89, 178, 917, 448, 262, 370, 261,
134791	/* 830 */ 82, 914, 80, 262, 370, 261, 776, 824, 824, 826,
134792	/* 840 */ 827, 19, 934, 434, 96, 96, 96, 96, 95, 95,
134793	/* 850 */ 94, 94, 94, 93, 351, 832, 74, 452, 451, 958,
134794	/* 860 */ 958, 819, 959, 960, 120, 92, 89, 178, 945, 2,
134795	/* 870 */ 918, 965, 268, 1, 976, 76, 445, 762, 3, 708,
134796	/* 880 */ 901, 901, 387, 958, 958, 757, 919, 371, 740, 778,
134797	/* 890 */ 756, 257, 824, 824, 826, 827, 19, 417, 741, 450,
134798	/* 900 */ 24, 959, 960, 83, 84, 369, 958, 958, 177, 226,
134799	/* 910 */ 85, 352, 352, 885, 315, 314, 313, 215, 311, 10,
134800	/* 920 */ 10, 683, 448, 349, 348, 959, 960, 909, 777, 157,
134801	/* 930 */ 120, 958, 958, 337, 776, 416, 711, 310, 450, 434,
134802	/* 940 */ 450, 321, 450, 791, 103, 200, 175, 450, 959, 960,
134803	/* 950 */ 908, 832, 792, 452, 451, 9, 9, 819, 10, 10,
134804	/* 960 */ 52, 52, 51, 51, 180, 716, 248, 10, 10, 171,
134805	/* 970 */ 170, 167, 339, 959, 960, 247, 984, 702, 702, 450,
134806	/* 980 */ 715, 233, 686, 982, 889, 983, 182, 914, 824, 824,
134807	/* 990 */ 826, 827, 19, 183, 256, 423, 132, 181, 394, 10,
134808	/* 1000 */ 10, 889, 891, 749, 958, 958, 917, 268, 985, 198,
134809	/* 1010 */ 985, 349, 348, 425, 415, 299, 817, 832, 326, 825,
134810	/* 1020 */ 120, 332, 133, 819, 268, 98, 98, 98, 98, 91,
134811	/* 1030 */ 96, 96, 96, 96, 95, 95, 94, 94, 94, 93,
134812	/* 1040 */ 351, 157, 810, 371, 382, 359, 959, 960, 358, 268,
134813	/* 1050 */ 450, 918, 368, 324, 824, 824, 826, 450, 709, 450,
134814	/* 1060 */ 264, 380, 889, 450, 877, 746, 253, 919, 255, 433,
134815	/* 1070 */ 36, 36, 234, 450, 234, 120, 269, 37, 37, 12,
134816	/* 1080 */ 12, 334, 272, 27, 27, 450, 330, 118, 450, 162,
134817	/* 1090 */ 742, 280, 450, 38, 38, 450, 985, 356, 985, 450,
134818	/* 1100 */ 709, 1210, 450, 132, 450, 39, 39, 450, 40, 40,
134819	/* 1110 */ 450, 362, 41, 41, 450, 42, 42, 450, 254, 28,
134820	/* 1120 */ 28, 450, 29, 29, 31, 31, 450, 43, 43, 450,
134821	/* 1130 */ 44, 44, 450, 714, 45, 45, 450, 11, 11, 767,
134822	/* 1140 */ 450, 46, 46, 450, 268, 450, 105, 105, 450, 47,
134823	/* 1150 */ 47, 450, 48, 48, 450, 237, 33, 33, 450, 172,
134824	/* 1160 */ 49, 49, 450, 50, 50, 34, 34, 274, 122, 122,
134825	/* 1170 */ 450, 123, 123, 450, 124, 124, 450, 898, 56, 56,
134826	/* 1180 */ 450, 897, 35, 35, 450, 267, 450, 817, 450, 817,
134827	/* 1190 */ 106, 106, 450, 53, 53, 385, 107, 107, 450, 817,
134828	/* 1200 */ 108, 108, 817, 450, 104, 104, 121, 121, 119, 119,
134829	/* 1210 */ 450, 117, 112, 112, 450, 276, 450, 225, 111, 111,
134830	/* 1220 */ 450, 730, 450, 109, 109, 450, 673, 674, 675, 912,
134831	/* 1230 */ 110, 110, 317, 998, 55, 55, 57, 57, 692, 331,
134832	/* 1240 */ 54, 54, 26, 26, 696, 30, 30, 317, 937, 197,
134833	/* 1250 */ 196, 195, 335, 281, 336, 446, 331, 745, 689, 436,
134834	/* 1260 */ 440, 444, 120, 72, 386, 223, 175, 345, 757, 933,
134835	/* 1270 */ 20, 286, 319, 756, 815, 372, 374, 202, 202, 202,
134836	/* 1280 */ 263, 395, 285, 74, 208, 21, 696, 719, 718, 884,
134837	/* 1290 */ 120, 120, 120, 120, 120, 754, 278, 828, 77, 74,
134838	/* 1300 */ 726, 727, 785, 783, 880, 202, 999, 208, 894, 893,
134839	/* 1310 */ 894, 893, 694, 816, 763, 116, 774, 1290, 431, 432,
134840	/* 1320 */ 302, 999, 390, 303, 823, 697, 691, 680, 159, 289,
134841	/* 1330 */ 679, 884, 681, 952, 291, 218, 293, 7, 316, 828,
134842	/* 1340 */ 173, 805, 259, 364, 252, 911, 376, 713, 295, 435,
134843	/* 1350 */ 308, 168, 955, 993, 135, 400, 990, 284, 882, 881,
134844	/* 1360 */ 205, 928, 926, 59, 333, 62, 144, 156, 130, 72,
134845	/* 1370 */ 802, 366, 367, 393, 137, 185, 189, 160, 139, 383,
134846	/* 1380 */ 67, 896, 140, 141, 142, 148, 389, 812, 775, 266,
134847	/* 1390 */ 219, 190, 154, 391, 913, 876, 271, 406, 191, 322,
134848	/* 1400 */ 682, 733, 192, 342, 732, 724, 731, 711, 723, 421,
134849	/* 1410 */ 705, 71, 323, 6, 204, 771, 288, 79, 297, 346,
134850	/* 1420 */ 772, 704, 290, 283, 703, 770, 292, 294, 967, 239,
134851	/* 1430 */ 769, 102, 862, 438, 426, 240, 424, 442, 73, 213,
134852	/* 1440 */ 688, 238, 22, 453, 953, 214, 217, 216, 454, 677,
134853	/* 1450 */ 676, 671, 753, 125, 115, 235, 126, 669, 353, 166,
134854	/* 1460 */ 127, 244, 179, 357, 306, 304, 305, 307, 113, 892,
134855	/* 1470 */ 327, 890, 811, 328, 134, 128, 136, 138, 743, 258,
134856	/* 1480 */ 907, 184, 143, 129, 910, 186, 63, 64, 145, 187,
134857	/* 1490 */ 906, 65, 8, 66, 13, 188, 202, 899, 265, 149,
134858	/* 1500 */ 987, 388, 150, 685, 161, 392, 285, 193, 279, 396,
134859	/* 1510 */ 151, 401, 68, 14, 15, 722, 69, 236, 831, 131,
134860	/* 1520 */ 830, 860, 70, 751, 16, 414, 755, 4, 174, 220,
134861	/* 1530 */ 222, 784, 201, 152, 779, 77, 74, 17, 18, 875,
134862	/* 1540 */ 861, 859, 916, 864, 915, 207, 206, 942, 163, 437,
134863	/* 1550 */ 948, 943, 164, 209, 1002, 441, 863, 165, 210, 829,
134864	/* 1560 */ 695, 87, 312, 211, 1292, 1291, 309,
134865	};
134866	static const YYCODETYPE yy_lookahead[] = {
134867	/* 0 */ 19, 95, 53, 97, 22, 24, 24, 101, 27, 28,
134868	/* 10 */ 29, 30, 31, 32, 33, 34, 35, 36, 37, 38,
134869	/* 20 */ 39, 40, 41, 152, 43, 44, 45, 46, 47, 48,
	@@ -135112,55 +135568,55 @@
135112	/* 300 */ 1256, 1200, 1206, 1260, 1247, 1261, 1263, 1262, 1266, 1278,
135113	/* 310 */ 1282, 1292, 1294, 1297, 1298, 1299, 1300, 1221, 1224, 1228,
135114	/* 320 */ 1288, 1291, 1276, 1277, 1295,
135115	};
135116	static const YYACTIONTYPE yy_default[] = {
135117	/* 0 */ 1281, 1271, 1271, 1271, 1203, 1203, 1203, 1203, 1271, 1096,
135118	/* 10 */ 1125, 1125, 1255, 1332, 1332, 1332, 1332, 1332, 1332, 1202,
135119	/* 20 */ 1332, 1332, 1332, 1332, 1271, 1100, 1131, 1332, 1332, 1332,
135120	/* 30 */ 1332, 1204, 1205, 1332, 1332, 1332, 1254, 1256, 1141, 1140,
135121	/* 40 */ 1139, 1138, 1237, 1112, 1136, 1129, 1133, 1204, 1198, 1199,
135122	/* 50 */ 1197, 1201, 1205, 1332, 1132, 1167, 1182, 1166, 1332, 1332,
135123	/* 60 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135124	/* 70 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135125	/* 80 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135126	/* 90 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135127	/* 100 */ 1332, 1332, 1332, 1332, 1176, 1181, 1188, 1180, 1177, 1169,
135128	/* 110 */ 1168, 1170, 1171, 1332, 1019, 1067, 1332, 1332, 1332, 1172,
135129	/* 120 */ 1332, 1173, 1185, 1184, 1183, 1262, 1289, 1288, 1332, 1332,
135130	/* 130 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135131	/* 140 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135132	/* 150 */ 1332, 1332, 1332, 1332, 1332, 1281, 1271, 1025, 1025, 1332,
135133	/* 160 */ 1271, 1271, 1271, 1271, 1271, 1271, 1267, 1100, 1091, 1332,
135134	/* 170 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135135	/* 180 */ 1259, 1257, 1332, 1218, 1332, 1332, 1332, 1332, 1332, 1332,
135136	/* 190 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135137	/* 200 */ 1332, 1332, 1332, 1332, 1096, 1332, 1332, 1332, 1332, 1332,
135138	/* 210 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1283, 1332, 1232,
135139	/* 220 */ 1096, 1096, 1096, 1098, 1080, 1090, 1004, 1135, 1114, 1114,
135140	/* 230 */ 1321, 1135, 1321, 1042, 1303, 1039, 1125, 1114, 1200, 1125,
135141	/* 240 */ 1125, 1097, 1090, 1332, 1324, 1105, 1105, 1323, 1323, 1105,
135142	/* 250 */ 1146, 1070, 1135, 1076, 1076, 1076, 1076, 1105, 1016, 1135,
135143	/* 260 */ 1146, 1070, 1070, 1135, 1105, 1016, 1236, 1318, 1105, 1105,
135144	/* 270 */ 1016, 1211, 1105, 1016, 1105, 1016, 1211, 1068, 1068, 1068,
135145	/* 280 */ 1057, 1211, 1068, 1042, 1068, 1057, 1068, 1068, 1118, 1113,
135146	/* 290 */ 1118, 1113, 1118, 1113, 1118, 1113, 1105, 1206, 1105, 1332,
135147	/* 300 */ 1211, 1215, 1215, 1211, 1130, 1119, 1128, 1126, 1135, 1022,
135148	/* 310 */ 1060, 1286, 1286, 1282, 1282, 1282, 1282, 1329, 1329, 1267,
135149	/* 320 */ 1298, 1298, 1044, 1044, 1298, 1332, 1332, 1332, 1332, 1332,
135150	/* 330 */ 1332, 1293, 1332, 1220, 1332, 1332, 1332, 1332, 1332, 1332,
135151	/* 340 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135152	/* 350 */ 1332, 1332, 1152, 1332, 1000, 1264, 1332, 1332, 1263, 1332,
135153	/* 360 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135154	/* 370 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1320,
135155	/* 380 */ 1332, 1332, 1332, 1332, 1332, 1332, 1235, 1234, 1332, 1332,
135156	/* 390 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135157	/* 400 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135158	/* 410 */ 1332, 1082, 1332, 1332, 1332, 1307, 1332, 1332, 1332, 1332,
135159	/* 420 */ 1332, 1332, 1332, 1127, 1332, 1120, 1332, 1332, 1311, 1332,
135160	/* 430 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1273,
135161	/* 440 */ 1332, 1332, 1332, 1272, 1332, 1332, 1332, 1332, 1332, 1154,
135162	/* 450 */ 1332, 1153, 1157, 1332, 1010, 1332,
135163	};
135164	/******** End of lemon-generated parsing tables ***************************/
135165
135166	/* The next table maps tokens (terminal symbols) into fallback tokens.
	@@ -135602,147 +136058,147 @@
135602	/* 169 */ "expr ::= expr EQ\|NE expr",
135603	/* 170 */ "expr ::= expr BITAND\|BITOR\|LSHIFT\|RSHIFT expr",
135604	/* 171 */ "expr ::= expr PLUS\|MINUS expr",
135605	/* 172 */ "expr ::= expr STAR\|SLASH\|REM expr",
135606	/* 173 */ "expr ::= expr CONCAT expr",
135607	/* 174 */ "likeop ::= LIKE_KW\|MATCH",
135608	/* 175 */ "likeop ::= NOT LIKE_KW\|MATCH",
135609	/* 176 */ "expr ::= expr likeop expr",
135610	/* 177 */ "expr ::= expr likeop expr ESCAPE expr",
135611	/* 178 */ "expr ::= expr ISNULL\|NOTNULL",
135612	/* 179 */ "expr ::= expr NOT NULL",
135613	/* 180 */ "expr ::= expr IS expr",
135614	/* 181 */ "expr ::= expr IS NOT expr",
135615	/* 182 */ "expr ::= NOT expr",
135616	/* 183 */ "expr ::= BITNOT expr",
135617	/* 184 */ "expr ::= MINUS expr",
135618	/* 185 */ "expr ::= PLUS expr",
135619	/* 186 */ "between_op ::= BETWEEN",
135620	/* 187 */ "between_op ::= NOT BETWEEN",
135621	/* 188 */ "expr ::= expr between_op expr AND expr",
135622	/* 189 */ "in_op ::= IN",
135623	/* 190 */ "in_op ::= NOT IN",
135624	/* 191 */ "expr ::= expr in_op LP exprlist RP",
135625	/* 192 */ "expr ::= LP select RP",
135626	/* 193 */ "expr ::= expr in_op LP select RP",
135627	/* 194 */ "expr ::= expr in_op nm dbnm paren_exprlist",
135628	/* 195 */ "expr ::= EXISTS LP select RP",
135629	/* 196 */ "expr ::= CASE case_operand case_exprlist case_else END",
135630	/* 197 */ "case_exprlist ::= case_exprlist WHEN expr THEN expr",
135631	/* 198 */ "case_exprlist ::= WHEN expr THEN expr",
135632	/* 199 */ "case_else ::= ELSE expr",
135633	/* 200 */ "case_else ::=",
135634	/* 201 */ "case_operand ::= expr",
135635	/* 202 */ "case_operand ::=",
135636	/* 203 */ "exprlist ::=",
135637	/* 204 */ "nexprlist ::= nexprlist COMMA expr",
135638	/* 205 */ "nexprlist ::= expr",
135639	/* 206 */ "paren_exprlist ::=",
135640	/* 207 */ "paren_exprlist ::= LP exprlist RP",
135641	/* 208 */ "cmd ::= createkw uniqueflag INDEX ifnotexists nm dbnm ON nm LP sortlist RP where_opt",
135642	/* 209 */ "uniqueflag ::= UNIQUE",
135643	/* 210 */ "uniqueflag ::=",
135644	/* 211 */ "eidlist_opt ::=",
135645	/* 212 */ "eidlist_opt ::= LP eidlist RP",
135646	/* 213 */ "eidlist ::= eidlist COMMA nm collate sortorder",
135647	/* 214 */ "eidlist ::= nm collate sortorder",
135648	/* 215 */ "collate ::=",
135649	/* 216 */ "collate ::= COLLATE ID\|STRING",
135650	/* 217 */ "cmd ::= DROP INDEX ifexists fullname",
135651	/* 218 */ "cmd ::= VACUUM",
135652	/* 219 */ "cmd ::= VACUUM nm",
135653	/* 220 */ "cmd ::= PRAGMA nm dbnm",
135654	/* 221 */ "cmd ::= PRAGMA nm dbnm EQ nmnum",
135655	/* 222 */ "cmd ::= PRAGMA nm dbnm LP nmnum RP",
135656	/* 223 */ "cmd ::= PRAGMA nm dbnm EQ minus_num",
135657	/* 224 */ "cmd ::= PRAGMA nm dbnm LP minus_num RP",
135658	/* 225 */ "plus_num ::= PLUS INTEGER\|FLOAT",
135659	/* 226 */ "minus_num ::= MINUS INTEGER\|FLOAT",
135660	/* 227 */ "cmd ::= createkw trigger_decl BEGIN trigger_cmd_list END",
135661	/* 228 */ "trigger_decl ::= temp TRIGGER ifnotexists nm dbnm trigger_time trigger_event ON fullname foreach_clause when_clause",
135662	/* 229 */ "trigger_time ::= BEFORE",
135663	/* 230 */ "trigger_time ::= AFTER",
135664	/* 231 */ "trigger_time ::= INSTEAD OF",
135665	/* 232 */ "trigger_time ::=",
135666	/* 233 */ "trigger_event ::= DELETE\|INSERT",
135667	/* 234 */ "trigger_event ::= UPDATE",
135668	/* 235 */ "trigger_event ::= UPDATE OF idlist",
135669	/* 236 */ "when_clause ::=",
135670	/* 237 */ "when_clause ::= WHEN expr",
135671	/* 238 */ "trigger_cmd_list ::= trigger_cmd_list trigger_cmd SEMI",
135672	/* 239 */ "trigger_cmd_list ::= trigger_cmd SEMI",
135673	/* 240 */ "trnm ::= nm DOT nm",
135674	/* 241 */ "tridxby ::= INDEXED BY nm",
135675	/* 242 */ "tridxby ::= NOT INDEXED",
135676	/* 243 */ "trigger_cmd ::= UPDATE orconf trnm tridxby SET setlist where_opt",
135677	/* 244 */ "trigger_cmd ::= insert_cmd INTO trnm idlist_opt select",
135678	/* 245 */ "trigger_cmd ::= DELETE FROM trnm tridxby where_opt",
135679	/* 246 */ "trigger_cmd ::= select",
135680	/* 247 */ "expr ::= RAISE LP IGNORE RP",
135681	/* 248 */ "expr ::= RAISE LP raisetype COMMA nm RP",
135682	/* 249 */ "raisetype ::= ROLLBACK",
135683	/* 250 */ "raisetype ::= ABORT",
135684	/* 251 */ "raisetype ::= FAIL",
135685	/* 252 */ "cmd ::= DROP TRIGGER ifexists fullname",
135686	/* 253 */ "cmd ::= ATTACH database_kw_opt expr AS expr key_opt",
135687	/* 254 */ "cmd ::= DETACH database_kw_opt expr",
135688	/* 255 */ "key_opt ::=",
135689	/* 256 */ "key_opt ::= KEY expr",
135690	/* 257 */ "cmd ::= REINDEX",
135691	/* 258 */ "cmd ::= REINDEX nm dbnm",
135692	/* 259 */ "cmd ::= ANALYZE",
135693	/* 260 */ "cmd ::= ANALYZE nm dbnm",
135694	/* 261 */ "cmd ::= ALTER TABLE fullname RENAME TO nm",
135695	/* 262 */ "cmd ::= ALTER TABLE add_column_fullname ADD kwcolumn_opt columnname carglist",
135696	/* 263 */ "add_column_fullname ::= fullname",
135697	/* 264 */ "cmd ::= create_vtab",
135698	/* 265 */ "cmd ::= create_vtab LP vtabarglist RP",
135699	/* 266 */ "create_vtab ::= createkw VIRTUAL TABLE ifnotexists nm dbnm USING nm",
135700	/* 267 */ "vtabarg ::=",
135701	/* 268 */ "vtabargtoken ::= ANY",
135702	/* 269 */ "vtabargtoken ::= lp anylist RP",
135703	/* 270 */ "lp ::= LP",
135704	/* 271 */ "with ::=",
135705	/* 272 */ "with ::= WITH wqlist",
135706	/* 273 */ "with ::= WITH RECURSIVE wqlist",
135707	/* 274 */ "wqlist ::= nm eidlist_opt AS LP select RP",
135708	/* 275 */ "wqlist ::= wqlist COMMA nm eidlist_opt AS LP select RP",
135709	/* 276 */ "input ::= cmdlist",
135710	/* 277 */ "cmdlist ::= cmdlist ecmd",
135711	/* 278 */ "cmdlist ::= ecmd",
135712	/* 279 */ "ecmd ::= SEMI",
135713	/* 280 */ "ecmd ::= explain cmdx SEMI",
135714	/* 281 */ "explain ::=",
135715	/* 282 */ "trans_opt ::=",
135716	/* 283 */ "trans_opt ::= TRANSACTION",
135717	/* 284 */ "trans_opt ::= TRANSACTION nm",
135718	/* 285 */ "savepoint_opt ::= SAVEPOINT",
135719	/* 286 */ "savepoint_opt ::=",
135720	/* 287 */ "cmd ::= create_table create_table_args",
135721	/* 288 */ "columnlist ::= columnlist COMMA columnname carglist",
135722	/* 289 */ "columnlist ::= columnname carglist",
135723	/* 290 */ "nm ::= ID\|INDEXED",
135724	/* 291 */ "nm ::= STRING",
135725	/* 292 */ "nm ::= JOIN_KW",
135726	/* 293 */ "typetoken ::= typename",
135727	/* 294 */ "typename ::= ID\|STRING",
135728	/* 295 */ "signed ::= plus_num",
135729	/* 296 */ "signed ::= minus_num",
135730	/* 297 */ "carglist ::= carglist ccons",
135731	/* 298 */ "carglist ::=",
135732	/* 299 */ "ccons ::= NULL onconf",
135733	/* 300 */ "conslist_opt ::= COMMA conslist",
135734	/* 301 */ "conslist ::= conslist tconscomma tcons",
135735	/* 302 */ "conslist ::= tcons",
135736	/* 303 */ "tconscomma ::=",
135737	/* 304 */ "defer_subclause_opt ::= defer_subclause",
135738	/* 305 */ "resolvetype ::= raisetype",
135739	/* 306 */ "selectnowith ::= oneselect",
135740	/* 307 */ "oneselect ::= values",
135741	/* 308 */ "sclp ::= selcollist COMMA",
135742	/* 309 */ "as ::= ID\|STRING",
135743	/* 310 */ "expr ::= term",
135744	/* 311 */ "exprlist ::= nexprlist",
135745	/* 312 */ "nmnum ::= plus_num",
135746	/* 313 */ "nmnum ::= nm",
135747	/* 314 */ "nmnum ::= ON",
135748	/* 315 */ "nmnum ::= DELETE",
	@@ -136375,11 +136831,10 @@
136375	{ 173, 3 },
136376	{ 173, 3 },
136377	{ 173, 3 },
136378	{ 173, 3 },
136379	{ 173, 3 },
136380	{ 221, 1 },
136381	{ 221, 2 },
136382	{ 173, 3 },
136383	{ 173, 5 },
136384	{ 173, 2 },
136385	{ 173, 3 },
	@@ -136512,10 +136967,11 @@
136512	{ 194, 1 },
136513	{ 195, 1 },
136514	{ 209, 2 },
136515	{ 210, 1 },
136516	{ 173, 1 },

136517	{ 208, 1 },
136518	{ 230, 1 },
136519	{ 230, 1 },
136520	{ 230, 1 },
136521	{ 230, 1 },
	@@ -136654,11 +137110,11 @@
136654	case 42: /* autoinc ::= */ yytestcase(yyruleno==42);
136655	case 57: /* init_deferred_pred_opt ::= */ yytestcase(yyruleno==57);
136656	case 67: /* defer_subclause_opt ::= */ yytestcase(yyruleno==67);
136657	case 76: /* ifexists ::= */ yytestcase(yyruleno==76);
136658	case 90: /* distinct ::= */ yytestcase(yyruleno==90);
136659	case 215: /* collate ::= */ yytestcase(yyruleno==215);
136660	{yymsp[1].minor.yy194 = 0;}
136661	break;
136662	case 17: /* ifnotexists ::= IF NOT EXISTS */
136663	{yymsp[-2].minor.yy194 = 1;}
136664	break;
	@@ -136798,13 +137254,13 @@
136798	case 144: /* insert_cmd ::= INSERT orconf */ yytestcase(yyruleno==144);
136799	{yymsp[-1].minor.yy194 = yymsp[0].minor.yy194;}
136800	break;
136801	case 58: /* init_deferred_pred_opt ::= INITIALLY DEFERRED */
136802	case 75: /* ifexists ::= IF EXISTS */ yytestcase(yyruleno==75);
136803	case 187: /* between_op ::= NOT BETWEEN */ yytestcase(yyruleno==187);
136804	case 190: /* in_op ::= NOT IN */ yytestcase(yyruleno==190);
136805	case 216: /* collate ::= COLLATE ID\|STRING */ yytestcase(yyruleno==216);
136806	{yymsp[-1].minor.yy194 = 1;}
136807	break;
136808	case 59: /* init_deferred_pred_opt ::= INITIALLY IMMEDIATE */
136809	{yymsp[-1].minor.yy194 = 0;}
136810	break;
	@@ -136964,13 +137420,13 @@
136964	{yymsp[0].minor.yy194 = SF_All;}
136965	break;
136966	case 91: /* sclp ::= */
136967	case 119: /* orderby_opt ::= */ yytestcase(yyruleno==119);
136968	case 126: /* groupby_opt ::= */ yytestcase(yyruleno==126);
136969	case 203: /* exprlist ::= */ yytestcase(yyruleno==203);
136970	case 206: /* paren_exprlist ::= */ yytestcase(yyruleno==206);
136971	case 211: /* eidlist_opt ::= */ yytestcase(yyruleno==211);
136972	{yymsp[1].minor.yy148 = 0;}
136973	break;
136974	case 92: /* selcollist ::= sclp expr as */
136975	{
136976	yymsp[-2].minor.yy148 = sqlite3ExprListAppend(pParse, yymsp[-2].minor.yy148, yymsp[-1].minor.yy190.pExpr);
	@@ -136992,12 +137448,12 @@
136992	yymsp[-3].minor.yy148 = sqlite3ExprListAppend(pParse,yymsp[-3].minor.yy148, pDot);
136993	}
136994	break;
136995	case 95: /* as ::= AS nm */
136996	case 106: /* dbnm ::= DOT nm */ yytestcase(yyruleno==106);
136997	case 225: /* plus_num ::= PLUS INTEGER\|FLOAT */ yytestcase(yyruleno==225);
136998	case 226: /* minus_num ::= MINUS INTEGER\|FLOAT */ yytestcase(yyruleno==226);
136999	{yymsp[-1].minor.yy0 = yymsp[0].minor.yy0;}
137000	break;
137001	case 97: /* from ::= */
137002	{yymsp[1].minor.yy185 = sqlite3DbMallocZero(pParse->db, sizeof(*yymsp[1].minor.yy185));}
137003	break;
	@@ -137076,18 +137532,18 @@
137076	{yymsp[-3].minor.yy194 = sqlite3JoinType(pParse,&yymsp[-3].minor.yy0,&yymsp[-2].minor.yy0,&yymsp[-1].minor.yy0);/X-overwrites-A/}
137077	break;
137078	case 112: /* on_opt ::= ON expr */
137079	case 129: /* having_opt ::= HAVING expr */ yytestcase(yyruleno==129);
137080	case 136: /* where_opt ::= WHERE expr */ yytestcase(yyruleno==136);
137081	case 199: /* case_else ::= ELSE expr */ yytestcase(yyruleno==199);
137082	{yymsp[-1].minor.yy72 = yymsp[0].minor.yy190.pExpr;}
137083	break;
137084	case 113: /* on_opt ::= */
137085	case 128: /* having_opt ::= */ yytestcase(yyruleno==128);
137086	case 135: /* where_opt ::= */ yytestcase(yyruleno==135);
137087	case 200: /* case_else ::= */ yytestcase(yyruleno==200);
137088	case 202: /* case_operand ::= */ yytestcase(yyruleno==202);
137089	{yymsp[1].minor.yy72 = 0;}
137090	break;
137091	case 115: /* indexed_opt ::= INDEXED BY nm */
137092	{yymsp[-2].minor.yy0 = yymsp[0].minor.yy0;}
137093	break;
	@@ -137230,11 +137686,11 @@
137230	case 158: /* term ::= INTEGER */
137231	{
137232	yylhsminor.yy190.pExpr = sqlite3ExprAlloc(pParse->db, TK_INTEGER, &yymsp[0].minor.yy0, 1);
137233	yylhsminor.yy190.zStart = yymsp[0].minor.yy0.z;
137234	yylhsminor.yy190.zEnd = yymsp[0].minor.yy0.z + yymsp[0].minor.yy0.n;
137235	if( yylhsminor.yy190.pExpr ) yylhsminor.yy190.pExpr->flags \|= EP_Leaf;
137236	}
137237	yymsp[0].minor.yy190 = yylhsminor.yy190;
137238	break;
137239	case 159: /* expr ::= VARIABLE */
137240	{
	@@ -137320,17 +137776,14 @@
137320	case 171: /* expr ::= expr PLUS\|MINUS expr */ yytestcase(yyruleno==171);
137321	case 172: /* expr ::= expr STAR\|SLASH\|REM expr */ yytestcase(yyruleno==172);
137322	case 173: /* expr ::= expr CONCAT expr */ yytestcase(yyruleno==173);
137323	{spanBinaryExpr(pParse,yymsp[-1].major,&yymsp[-2].minor.yy190,&yymsp[0].minor.yy190);}
137324	break;
137325	case 174: /* likeop ::= LIKE_KW\|MATCH */
137326	{yymsp[0].minor.yy0=yymsp[0].minor.yy0;/A-overwrites-X/}
137327	break;
137328	case 175: /* likeop ::= NOT LIKE_KW\|MATCH */
137329	{yymsp[-1].minor.yy0=yymsp[0].minor.yy0; yymsp[-1].minor.yy0.n\|=0x80000000; /yymsp[-1].minor.yy0-overwrite-yymsp[0].minor.yy0/}
137330	break;
137331	case 176: /* expr ::= expr likeop expr */
137332	{
137333	ExprList *pList;
137334	int bNot = yymsp[-1].minor.yy0.n & 0x80000000;
137335	yymsp[-1].minor.yy0.n &= 0x7fffffff;
137336	pList = sqlite3ExprListAppend(pParse,0, yymsp[0].minor.yy190.pExpr);
	@@ -137339,11 +137792,11 @@
137339	exprNot(pParse, bNot, &yymsp[-2].minor.yy190);
137340	yymsp[-2].minor.yy190.zEnd = yymsp[0].minor.yy190.zEnd;
137341	if( yymsp[-2].minor.yy190.pExpr ) yymsp[-2].minor.yy190.pExpr->flags \|= EP_InfixFunc;
137342	}
137343	break;
137344	case 177: /* expr ::= expr likeop expr ESCAPE expr */
137345	{
137346	ExprList *pList;
137347	int bNot = yymsp[-3].minor.yy0.n & 0x80000000;
137348	yymsp[-3].minor.yy0.n &= 0x7fffffff;
137349	pList = sqlite3ExprListAppend(pParse,0, yymsp[-2].minor.yy190.pExpr);
	@@ -137353,43 +137806,43 @@
137353	exprNot(pParse, bNot, &yymsp[-4].minor.yy190);
137354	yymsp[-4].minor.yy190.zEnd = yymsp[0].minor.yy190.zEnd;
137355	if( yymsp[-4].minor.yy190.pExpr ) yymsp[-4].minor.yy190.pExpr->flags \|= EP_InfixFunc;
137356	}
137357	break;
137358	case 178: /* expr ::= expr ISNULL\|NOTNULL */
137359	{spanUnaryPostfix(pParse,yymsp[0].major,&yymsp[-1].minor.yy190,&yymsp[0].minor.yy0);}
137360	break;
137361	case 179: /* expr ::= expr NOT NULL */
137362	{spanUnaryPostfix(pParse,TK_NOTNULL,&yymsp[-2].minor.yy190,&yymsp[0].minor.yy0);}
137363	break;
137364	case 180: /* expr ::= expr IS expr */
137365	{
137366	spanBinaryExpr(pParse,TK_IS,&yymsp[-2].minor.yy190,&yymsp[0].minor.yy190);
137367	binaryToUnaryIfNull(pParse, yymsp[0].minor.yy190.pExpr, yymsp[-2].minor.yy190.pExpr, TK_ISNULL);
137368	}
137369	break;
137370	case 181: /* expr ::= expr IS NOT expr */
137371	{
137372	spanBinaryExpr(pParse,TK_ISNOT,&yymsp[-3].minor.yy190,&yymsp[0].minor.yy190);
137373	binaryToUnaryIfNull(pParse, yymsp[0].minor.yy190.pExpr, yymsp[-3].minor.yy190.pExpr, TK_NOTNULL);
137374	}
137375	break;
137376	case 182: /* expr ::= NOT expr */
137377	case 183: /* expr ::= BITNOT expr */ yytestcase(yyruleno==183);
137378	{spanUnaryPrefix(&yymsp[-1].minor.yy190,pParse,yymsp[-1].major,&yymsp[0].minor.yy190,&yymsp[-1].minor.yy0);/A-overwrites-B/}
137379	break;
137380	case 184: /* expr ::= MINUS expr */
137381	{spanUnaryPrefix(&yymsp[-1].minor.yy190,pParse,TK_UMINUS,&yymsp[0].minor.yy190,&yymsp[-1].minor.yy0);/A-overwrites-B/}
137382	break;
137383	case 185: /* expr ::= PLUS expr */
137384	{spanUnaryPrefix(&yymsp[-1].minor.yy190,pParse,TK_UPLUS,&yymsp[0].minor.yy190,&yymsp[-1].minor.yy0);/A-overwrites-B/}
137385	break;
137386	case 186: /* between_op ::= BETWEEN */
137387	case 189: /* in_op ::= IN */ yytestcase(yyruleno==189);
137388	{yymsp[0].minor.yy194 = 0;}
137389	break;
137390	case 188: /* expr ::= expr between_op expr AND expr */
137391	{
137392	ExprList *pList = sqlite3ExprListAppend(pParse,0, yymsp[-2].minor.yy190.pExpr);
137393	pList = sqlite3ExprListAppend(pParse,pList, yymsp[0].minor.yy190.pExpr);
137394	yymsp[-4].minor.yy190.pExpr = sqlite3PExpr(pParse, TK_BETWEEN, yymsp[-4].minor.yy190.pExpr, 0);
137395	if( yymsp[-4].minor.yy190.pExpr ){
	@@ -137399,11 +137852,11 @@
137399	}
137400	exprNot(pParse, yymsp[-3].minor.yy194, &yymsp[-4].minor.yy190);
137401	yymsp[-4].minor.yy190.zEnd = yymsp[0].minor.yy190.zEnd;
137402	}
137403	break;
137404	case 191: /* expr ::= expr in_op LP exprlist RP */
137405	{
137406	if( yymsp[-1].minor.yy148==0 ){
137407	/* Expressions of the form
137408	**
137409	** expr1 IN ()
	@@ -137452,26 +137905,26 @@
137452	exprNot(pParse, yymsp[-3].minor.yy194, &yymsp[-4].minor.yy190);
137453	}
137454	yymsp[-4].minor.yy190.zEnd = &yymsp[0].minor.yy0.z[yymsp[0].minor.yy0.n];
137455	}
137456	break;
137457	case 192: /* expr ::= LP select RP */
137458	{
137459	spanSet(&yymsp[-2].minor.yy190,&yymsp[-2].minor.yy0,&yymsp[0].minor.yy0); /A-overwrites-B/
137460	yymsp[-2].minor.yy190.pExpr = sqlite3PExpr(pParse, TK_SELECT, 0, 0);
137461	sqlite3PExprAddSelect(pParse, yymsp[-2].minor.yy190.pExpr, yymsp[-1].minor.yy243);
137462	}
137463	break;
137464	case 193: /* expr ::= expr in_op LP select RP */
137465	{
137466	yymsp[-4].minor.yy190.pExpr = sqlite3PExpr(pParse, TK_IN, yymsp[-4].minor.yy190.pExpr, 0);
137467	sqlite3PExprAddSelect(pParse, yymsp[-4].minor.yy190.pExpr, yymsp[-1].minor.yy243);
137468	exprNot(pParse, yymsp[-3].minor.yy194, &yymsp[-4].minor.yy190);
137469	yymsp[-4].minor.yy190.zEnd = &yymsp[0].minor.yy0.z[yymsp[0].minor.yy0.n];
137470	}
137471	break;
137472	case 194: /* expr ::= expr in_op nm dbnm paren_exprlist */
137473	{
137474	SrcList *pSrc = sqlite3SrcListAppend(pParse->db, 0,&yymsp[-2].minor.yy0,&yymsp[-1].minor.yy0);
137475	Select *pSelect = sqlite3SelectNew(pParse, 0,pSrc,0,0,0,0,0,0,0);
137476	if( yymsp[0].minor.yy148 ) sqlite3SrcListFuncArgs(pParse, pSelect ? pSrc : 0, yymsp[0].minor.yy148);
137477	yymsp[-4].minor.yy190.pExpr = sqlite3PExpr(pParse, TK_IN, yymsp[-4].minor.yy190.pExpr, 0);
	@@ -137478,19 +137931,19 @@
137478	sqlite3PExprAddSelect(pParse, yymsp[-4].minor.yy190.pExpr, pSelect);
137479	exprNot(pParse, yymsp[-3].minor.yy194, &yymsp[-4].minor.yy190);
137480	yymsp[-4].minor.yy190.zEnd = yymsp[-1].minor.yy0.z ? &yymsp[-1].minor.yy0.z[yymsp[-1].minor.yy0.n] : &yymsp[-2].minor.yy0.z[yymsp[-2].minor.yy0.n];
137481	}
137482	break;
137483	case 195: /* expr ::= EXISTS LP select RP */
137484	{
137485	Expr *p;
137486	spanSet(&yymsp[-3].minor.yy190,&yymsp[-3].minor.yy0,&yymsp[0].minor.yy0); /A-overwrites-B/
137487	p = yymsp[-3].minor.yy190.pExpr = sqlite3PExpr(pParse, TK_EXISTS, 0, 0);
137488	sqlite3PExprAddSelect(pParse, p, yymsp[-1].minor.yy243);
137489	}
137490	break;
137491	case 196: /* expr ::= CASE case_operand case_exprlist case_else END */
137492	{
137493	spanSet(&yymsp[-4].minor.yy190,&yymsp[-4].minor.yy0,&yymsp[0].minor.yy0); /A-overwrites-C/
137494	yymsp[-4].minor.yy190.pExpr = sqlite3PExpr(pParse, TK_CASE, yymsp[-3].minor.yy72, 0);
137495	if( yymsp[-4].minor.yy190.pExpr ){
137496	yymsp[-4].minor.yy190.pExpr->x.pList = yymsp[-1].minor.yy72 ? sqlite3ExprListAppend(pParse,yymsp[-2].minor.yy148,yymsp[-1].minor.yy72) : yymsp[-2].minor.yy148;
	@@ -137499,313 +137952,314 @@
137499	sqlite3ExprListDelete(pParse->db, yymsp[-2].minor.yy148);
137500	sqlite3ExprDelete(pParse->db, yymsp[-1].minor.yy72);
137501	}
137502	}
137503	break;
137504	case 197: /* case_exprlist ::= case_exprlist WHEN expr THEN expr */
137505	{
137506	yymsp[-4].minor.yy148 = sqlite3ExprListAppend(pParse,yymsp[-4].minor.yy148, yymsp[-2].minor.yy190.pExpr);
137507	yymsp[-4].minor.yy148 = sqlite3ExprListAppend(pParse,yymsp[-4].minor.yy148, yymsp[0].minor.yy190.pExpr);
137508	}
137509	break;
137510	case 198: /* case_exprlist ::= WHEN expr THEN expr */
137511	{
137512	yymsp[-3].minor.yy148 = sqlite3ExprListAppend(pParse,0, yymsp[-2].minor.yy190.pExpr);
137513	yymsp[-3].minor.yy148 = sqlite3ExprListAppend(pParse,yymsp[-3].minor.yy148, yymsp[0].minor.yy190.pExpr);
137514	}
137515	break;
137516	case 201: /* case_operand ::= expr */
137517	{yymsp[0].minor.yy72 = yymsp[0].minor.yy190.pExpr; /A-overwrites-X/}
137518	break;
137519	case 204: /* nexprlist ::= nexprlist COMMA expr */
137520	{yymsp[-2].minor.yy148 = sqlite3ExprListAppend(pParse,yymsp[-2].minor.yy148,yymsp[0].minor.yy190.pExpr);}
137521	break;
137522	case 205: /* nexprlist ::= expr */
137523	{yymsp[0].minor.yy148 = sqlite3ExprListAppend(pParse,0,yymsp[0].minor.yy190.pExpr); /A-overwrites-Y/}
137524	break;
137525	case 207: /* paren_exprlist ::= LP exprlist RP */
137526	case 212: /* eidlist_opt ::= LP eidlist RP */ yytestcase(yyruleno==212);
137527	{yymsp[-2].minor.yy148 = yymsp[-1].minor.yy148;}
137528	break;
137529	case 208: /* cmd ::= createkw uniqueflag INDEX ifnotexists nm dbnm ON nm LP sortlist RP where_opt */
137530	{
137531	sqlite3CreateIndex(pParse, &yymsp[-7].minor.yy0, &yymsp[-6].minor.yy0,
137532	sqlite3SrcListAppend(pParse->db,0,&yymsp[-4].minor.yy0,0), yymsp[-2].minor.yy148, yymsp[-10].minor.yy194,
137533	&yymsp[-11].minor.yy0, yymsp[0].minor.yy72, SQLITE_SO_ASC, yymsp[-8].minor.yy194, SQLITE_IDXTYPE_APPDEF);
137534	}
137535	break;
137536	case 209: /* uniqueflag ::= UNIQUE */
137537	case 250: /* raisetype ::= ABORT */ yytestcase(yyruleno==250);
137538	{yymsp[0].minor.yy194 = OE_Abort;}
137539	break;
137540	case 210: /* uniqueflag ::= */
137541	{yymsp[1].minor.yy194 = OE_None;}
137542	break;
137543	case 213: /* eidlist ::= eidlist COMMA nm collate sortorder */
137544	{
137545	yymsp[-4].minor.yy148 = parserAddExprIdListTerm(pParse, yymsp[-4].minor.yy148, &yymsp[-2].minor.yy0, yymsp[-1].minor.yy194, yymsp[0].minor.yy194);
137546	}
137547	break;
137548	case 214: /* eidlist ::= nm collate sortorder */
137549	{
137550	yymsp[-2].minor.yy148 = parserAddExprIdListTerm(pParse, 0, &yymsp[-2].minor.yy0, yymsp[-1].minor.yy194, yymsp[0].minor.yy194); /A-overwrites-Y/
137551	}
137552	break;
137553	case 217: /* cmd ::= DROP INDEX ifexists fullname */
137554	{sqlite3DropIndex(pParse, yymsp[0].minor.yy185, yymsp[-1].minor.yy194);}
137555	break;
137556	case 218: /* cmd ::= VACUUM */
137557	{sqlite3Vacuum(pParse,0);}
137558	break;
137559	case 219: /* cmd ::= VACUUM nm */
137560	{sqlite3Vacuum(pParse,&yymsp[0].minor.yy0);}
137561	break;
137562	case 220: /* cmd ::= PRAGMA nm dbnm */
137563	{sqlite3Pragma(pParse,&yymsp[-1].minor.yy0,&yymsp[0].minor.yy0,0,0);}
137564	break;
137565	case 221: /* cmd ::= PRAGMA nm dbnm EQ nmnum */
137566	{sqlite3Pragma(pParse,&yymsp[-3].minor.yy0,&yymsp[-2].minor.yy0,&yymsp[0].minor.yy0,0);}
137567	break;
137568	case 222: /* cmd ::= PRAGMA nm dbnm LP nmnum RP */
137569	{sqlite3Pragma(pParse,&yymsp[-4].minor.yy0,&yymsp[-3].minor.yy0,&yymsp[-1].minor.yy0,0);}
137570	break;
137571	case 223: /* cmd ::= PRAGMA nm dbnm EQ minus_num */
137572	{sqlite3Pragma(pParse,&yymsp[-3].minor.yy0,&yymsp[-2].minor.yy0,&yymsp[0].minor.yy0,1);}
137573	break;
137574	case 224: /* cmd ::= PRAGMA nm dbnm LP minus_num RP */
137575	{sqlite3Pragma(pParse,&yymsp[-4].minor.yy0,&yymsp[-3].minor.yy0,&yymsp[-1].minor.yy0,1);}
137576	break;
137577	case 227: /* cmd ::= createkw trigger_decl BEGIN trigger_cmd_list END */
137578	{
137579	Token all;
137580	all.z = yymsp[-3].minor.yy0.z;
137581	all.n = (int)(yymsp[0].minor.yy0.z - yymsp[-3].minor.yy0.z) + yymsp[0].minor.yy0.n;
137582	sqlite3FinishTrigger(pParse, yymsp[-1].minor.yy145, &all);
137583	}
137584	break;
137585	case 228: /* trigger_decl ::= temp TRIGGER ifnotexists nm dbnm trigger_time trigger_event ON fullname foreach_clause when_clause */
137586	{
137587	sqlite3BeginTrigger(pParse, &yymsp[-7].minor.yy0, &yymsp[-6].minor.yy0, yymsp[-5].minor.yy194, yymsp[-4].minor.yy332.a, yymsp[-4].minor.yy332.b, yymsp[-2].minor.yy185, yymsp[0].minor.yy72, yymsp[-10].minor.yy194, yymsp[-8].minor.yy194);
137588	yymsp[-10].minor.yy0 = (yymsp[-6].minor.yy0.n==0?yymsp[-7].minor.yy0:yymsp[-6].minor.yy0); /A-overwrites-T/
137589	}
137590	break;
137591	case 229: /* trigger_time ::= BEFORE */
137592	{ yymsp[0].minor.yy194 = TK_BEFORE; }
137593	break;
137594	case 230: /* trigger_time ::= AFTER */
137595	{ yymsp[0].minor.yy194 = TK_AFTER; }
137596	break;
137597	case 231: /* trigger_time ::= INSTEAD OF */
137598	{ yymsp[-1].minor.yy194 = TK_INSTEAD;}
137599	break;
137600	case 232: /* trigger_time ::= */
137601	{ yymsp[1].minor.yy194 = TK_BEFORE; }
137602	break;
137603	case 233: /* trigger_event ::= DELETE\|INSERT */
137604	case 234: /* trigger_event ::= UPDATE */ yytestcase(yyruleno==234);
137605	{yymsp[0].minor.yy332.a = yymsp[0].major; /A-overwrites-X/ yymsp[0].minor.yy332.b = 0;}
137606	break;
137607	case 235: /* trigger_event ::= UPDATE OF idlist */
137608	{yymsp[-2].minor.yy332.a = TK_UPDATE; yymsp[-2].minor.yy332.b = yymsp[0].minor.yy254;}
137609	break;
137610	case 236: /* when_clause ::= */
137611	case 255: /* key_opt ::= */ yytestcase(yyruleno==255);
137612	{ yymsp[1].minor.yy72 = 0; }
137613	break;
137614	case 237: /* when_clause ::= WHEN expr */
137615	case 256: /* key_opt ::= KEY expr */ yytestcase(yyruleno==256);
137616	{ yymsp[-1].minor.yy72 = yymsp[0].minor.yy190.pExpr; }
137617	break;
137618	case 238: /* trigger_cmd_list ::= trigger_cmd_list trigger_cmd SEMI */
137619	{
137620	assert( yymsp[-2].minor.yy145!=0 );
137621	yymsp[-2].minor.yy145->pLast->pNext = yymsp[-1].minor.yy145;
137622	yymsp[-2].minor.yy145->pLast = yymsp[-1].minor.yy145;
137623	}
137624	break;
137625	case 239: /* trigger_cmd_list ::= trigger_cmd SEMI */
137626	{
137627	assert( yymsp[-1].minor.yy145!=0 );
137628	yymsp[-1].minor.yy145->pLast = yymsp[-1].minor.yy145;
137629	}
137630	break;
137631	case 240: /* trnm ::= nm DOT nm */
137632	{
137633	yymsp[-2].minor.yy0 = yymsp[0].minor.yy0;
137634	sqlite3ErrorMsg(pParse,
137635	"qualified table names are not allowed on INSERT, UPDATE, and DELETE "
137636	"statements within triggers");
137637	}
137638	break;
137639	case 241: /* tridxby ::= INDEXED BY nm */
137640	{
137641	sqlite3ErrorMsg(pParse,
137642	"the INDEXED BY clause is not allowed on UPDATE or DELETE statements "
137643	"within triggers");
137644	}
137645	break;
137646	case 242: /* tridxby ::= NOT INDEXED */
137647	{
137648	sqlite3ErrorMsg(pParse,
137649	"the NOT INDEXED clause is not allowed on UPDATE or DELETE statements "
137650	"within triggers");
137651	}
137652	break;
137653	case 243: /* trigger_cmd ::= UPDATE orconf trnm tridxby SET setlist where_opt */
137654	{yymsp[-6].minor.yy145 = sqlite3TriggerUpdateStep(pParse->db, &yymsp[-4].minor.yy0, yymsp[-1].minor.yy148, yymsp[0].minor.yy72, yymsp[-5].minor.yy194);}
137655	break;
137656	case 244: /* trigger_cmd ::= insert_cmd INTO trnm idlist_opt select */
137657	{yymsp[-4].minor.yy145 = sqlite3TriggerInsertStep(pParse->db, &yymsp[-2].minor.yy0, yymsp[-1].minor.yy254, yymsp[0].minor.yy243, yymsp[-4].minor.yy194);/A-overwrites-R/}
137658	break;
137659	case 245: /* trigger_cmd ::= DELETE FROM trnm tridxby where_opt */
137660	{yymsp[-4].minor.yy145 = sqlite3TriggerDeleteStep(pParse->db, &yymsp[-2].minor.yy0, yymsp[0].minor.yy72);}
137661	break;
137662	case 246: /* trigger_cmd ::= select */
137663	{yymsp[0].minor.yy145 = sqlite3TriggerSelectStep(pParse->db, yymsp[0].minor.yy243); /A-overwrites-X/}
137664	break;
137665	case 247: /* expr ::= RAISE LP IGNORE RP */
137666	{
137667	spanSet(&yymsp[-3].minor.yy190,&yymsp[-3].minor.yy0,&yymsp[0].minor.yy0); /A-overwrites-X/
137668	yymsp[-3].minor.yy190.pExpr = sqlite3PExpr(pParse, TK_RAISE, 0, 0);
137669	if( yymsp[-3].minor.yy190.pExpr ){
137670	yymsp[-3].minor.yy190.pExpr->affinity = OE_Ignore;
137671	}
137672	}
137673	break;
137674	case 248: /* expr ::= RAISE LP raisetype COMMA nm RP */
137675	{
137676	spanSet(&yymsp[-5].minor.yy190,&yymsp[-5].minor.yy0,&yymsp[0].minor.yy0); /A-overwrites-X/
137677	yymsp[-5].minor.yy190.pExpr = sqlite3ExprAlloc(pParse->db, TK_RAISE, &yymsp[-1].minor.yy0, 1);
137678	if( yymsp[-5].minor.yy190.pExpr ) {
137679	yymsp[-5].minor.yy190.pExpr->affinity = (char)yymsp[-3].minor.yy194;
137680	}
137681	}
137682	break;
137683	case 249: /* raisetype ::= ROLLBACK */
137684	{yymsp[0].minor.yy194 = OE_Rollback;}
137685	break;
137686	case 251: /* raisetype ::= FAIL */
137687	{yymsp[0].minor.yy194 = OE_Fail;}
137688	break;
137689	case 252: /* cmd ::= DROP TRIGGER ifexists fullname */
137690	{
137691	sqlite3DropTrigger(pParse,yymsp[0].minor.yy185,yymsp[-1].minor.yy194);
137692	}
137693	break;
137694	case 253: /* cmd ::= ATTACH database_kw_opt expr AS expr key_opt */
137695	{
137696	sqlite3Attach(pParse, yymsp[-3].minor.yy190.pExpr, yymsp[-1].minor.yy190.pExpr, yymsp[0].minor.yy72);
137697	}
137698	break;
137699	case 254: /* cmd ::= DETACH database_kw_opt expr */
137700	{
137701	sqlite3Detach(pParse, yymsp[0].minor.yy190.pExpr);
137702	}
137703	break;
137704	case 257: /* cmd ::= REINDEX */
137705	{sqlite3Reindex(pParse, 0, 0);}
137706	break;
137707	case 258: /* cmd ::= REINDEX nm dbnm */
137708	{sqlite3Reindex(pParse, &yymsp[-1].minor.yy0, &yymsp[0].minor.yy0);}
137709	break;
137710	case 259: /* cmd ::= ANALYZE */
137711	{sqlite3Analyze(pParse, 0, 0);}
137712	break;
137713	case 260: /* cmd ::= ANALYZE nm dbnm */
137714	{sqlite3Analyze(pParse, &yymsp[-1].minor.yy0, &yymsp[0].minor.yy0);}
137715	break;
137716	case 261: /* cmd ::= ALTER TABLE fullname RENAME TO nm */
137717	{
137718	sqlite3AlterRenameTable(pParse,yymsp[-3].minor.yy185,&yymsp[0].minor.yy0);
137719	}
137720	break;
137721	case 262: /* cmd ::= ALTER TABLE add_column_fullname ADD kwcolumn_opt columnname carglist */
137722	{
137723	yymsp[-1].minor.yy0.n = (int)(pParse->sLastToken.z-yymsp[-1].minor.yy0.z) + pParse->sLastToken.n;
137724	sqlite3AlterFinishAddColumn(pParse, &yymsp[-1].minor.yy0);
137725	}
137726	break;
137727	case 263: /* add_column_fullname ::= fullname */
137728	{
137729	disableLookaside(pParse);
137730	sqlite3AlterBeginAddColumn(pParse, yymsp[0].minor.yy185);
137731	}
137732	break;
137733	case 264: /* cmd ::= create_vtab */
137734	{sqlite3VtabFinishParse(pParse,0);}
137735	break;
137736	case 265: /* cmd ::= create_vtab LP vtabarglist RP */
137737	{sqlite3VtabFinishParse(pParse,&yymsp[0].minor.yy0);}
137738	break;
137739	case 266: /* create_vtab ::= createkw VIRTUAL TABLE ifnotexists nm dbnm USING nm */
137740	{
137741	sqlite3VtabBeginParse(pParse, &yymsp[-3].minor.yy0, &yymsp[-2].minor.yy0, &yymsp[0].minor.yy0, yymsp[-4].minor.yy194);
137742	}
137743	break;
137744	case 267: /* vtabarg ::= */
137745	{sqlite3VtabArgInit(pParse);}
137746	break;
137747	case 268: /* vtabargtoken ::= ANY */
137748	case 269: /* vtabargtoken ::= lp anylist RP */ yytestcase(yyruleno==269);
137749	case 270: /* lp ::= LP */ yytestcase(yyruleno==270);
137750	{sqlite3VtabArgExtend(pParse,&yymsp[0].minor.yy0);}
137751	break;
137752	case 271: /* with ::= */
137753	{yymsp[1].minor.yy285 = 0;}
137754	break;
137755	case 272: /* with ::= WITH wqlist */
137756	{ yymsp[-1].minor.yy285 = yymsp[0].minor.yy285; }
137757	break;
137758	case 273: /* with ::= WITH RECURSIVE wqlist */
137759	{ yymsp[-2].minor.yy285 = yymsp[0].minor.yy285; }
137760	break;
137761	case 274: /* wqlist ::= nm eidlist_opt AS LP select RP */
137762	{
137763	yymsp[-5].minor.yy285 = sqlite3WithAdd(pParse, 0, &yymsp[-5].minor.yy0, yymsp[-4].minor.yy148, yymsp[-1].minor.yy243); /A-overwrites-X/
137764	}
137765	break;
137766	case 275: /* wqlist ::= wqlist COMMA nm eidlist_opt AS LP select RP */
137767	{
137768	yymsp[-7].minor.yy285 = sqlite3WithAdd(pParse, yymsp[-7].minor.yy285, &yymsp[-5].minor.yy0, yymsp[-4].minor.yy148, yymsp[-1].minor.yy243);
137769	}
137770	break;
137771	default:
137772	/* (276) input ::= cmdlist */ yytestcase(yyruleno==276);
137773	/* (277) cmdlist ::= cmdlist ecmd */ yytestcase(yyruleno==277);
137774	/* (278) cmdlist ::= ecmd (OPTIMIZED OUT) */ assert(yyruleno!=278);
137775	/* (279) ecmd ::= SEMI */ yytestcase(yyruleno==279);
137776	/* (280) ecmd ::= explain cmdx SEMI */ yytestcase(yyruleno==280);
137777	/* (281) explain ::= */ yytestcase(yyruleno==281);
137778	/* (282) trans_opt ::= */ yytestcase(yyruleno==282);
137779	/* (283) trans_opt ::= TRANSACTION */ yytestcase(yyruleno==283);
137780	/* (284) trans_opt ::= TRANSACTION nm */ yytestcase(yyruleno==284);
137781	/* (285) savepoint_opt ::= SAVEPOINT */ yytestcase(yyruleno==285);
137782	/* (286) savepoint_opt ::= */ yytestcase(yyruleno==286);
137783	/* (287) cmd ::= create_table create_table_args */ yytestcase(yyruleno==287);
137784	/* (288) columnlist ::= columnlist COMMA columnname carglist */ yytestcase(yyruleno==288);
137785	/* (289) columnlist ::= columnname carglist */ yytestcase(yyruleno==289);
137786	/* (290) nm ::= ID\|INDEXED */ yytestcase(yyruleno==290);
137787	/* (291) nm ::= STRING */ yytestcase(yyruleno==291);
137788	/* (292) nm ::= JOIN_KW */ yytestcase(yyruleno==292);
137789	/* (293) typetoken ::= typename */ yytestcase(yyruleno==293);
137790	/* (294) typename ::= ID\|STRING */ yytestcase(yyruleno==294);
137791	/* (295) signed ::= plus_num (OPTIMIZED OUT) */ assert(yyruleno!=295);
137792	/* (296) signed ::= minus_num (OPTIMIZED OUT) */ assert(yyruleno!=296);
137793	/* (297) carglist ::= carglist ccons */ yytestcase(yyruleno==297);
137794	/* (298) carglist ::= */ yytestcase(yyruleno==298);
137795	/* (299) ccons ::= NULL onconf */ yytestcase(yyruleno==299);
137796	/* (300) conslist_opt ::= COMMA conslist */ yytestcase(yyruleno==300);
137797	/* (301) conslist ::= conslist tconscomma tcons */ yytestcase(yyruleno==301);
137798	/* (302) conslist ::= tcons (OPTIMIZED OUT) */ assert(yyruleno!=302);
137799	/* (303) tconscomma ::= */ yytestcase(yyruleno==303);
137800	/* (304) defer_subclause_opt ::= defer_subclause (OPTIMIZED OUT) */ assert(yyruleno!=304);
137801	/* (305) resolvetype ::= raisetype (OPTIMIZED OUT) */ assert(yyruleno!=305);
137802	/* (306) selectnowith ::= oneselect (OPTIMIZED OUT) */ assert(yyruleno!=306);
137803	/* (307) oneselect ::= values */ yytestcase(yyruleno==307);
137804	/* (308) sclp ::= selcollist COMMA */ yytestcase(yyruleno==308);
137805	/* (309) as ::= ID\|STRING */ yytestcase(yyruleno==309);
137806	/* (310) expr ::= term (OPTIMIZED OUT) */ assert(yyruleno!=310);

137807	/* (311) exprlist ::= nexprlist */ yytestcase(yyruleno==311);
137808	/* (312) nmnum ::= plus_num (OPTIMIZED OUT) */ assert(yyruleno!=312);
137809	/* (313) nmnum ::= nm (OPTIMIZED OUT) */ assert(yyruleno!=313);
137810	/* (314) nmnum ::= ON */ yytestcase(yyruleno==314);
137811	/* (315) nmnum ::= DELETE */ yytestcase(yyruleno==315);
	@@ -138854,12 +139308,12 @@
138854	** memory obtained from sqlite3_malloc() and to make *pzErrMsg point to that
138855	** error message.
138856	*/
138857	SQLITE_PRIVATE int sqlite3RunParser(Parse pParse, const char zSql, char **pzErrMsg){
138858	int nErr = 0; /* Number of errors encountered */
138859	int i; /* Loop counter */
138860	void pEngine; / The LEMON-generated LALR(1) parser */

138861	int tokenType; /* type of the next token */
138862	int lastTokenParsed = -1; /* type of the previous token */
138863	sqlite3 db = pParse->db; / The database connection */
138864	int mxSqlLen; /* Max length of an SQL string */
138865	#ifdef sqlite3Parser_ENGINEALWAYSONSTACK
	@@ -138871,11 +139325,10 @@
138871	if( db->nVdbeActive==0 ){
138872	db->u1.isInterrupted = 0;
138873	}
138874	pParse->rc = SQLITE_OK;
138875	pParse->zTail = zSql;
138876	i = 0;
138877	assert( pzErrMsg!=0 );
138878	/* sqlite3ParserTrace(stdout, "parser: "); */
138879	#ifdef sqlite3Parser_ENGINEALWAYSONSTACK
138880	pEngine = zSpace;
138881	sqlite3ParserInit(pEngine);
	@@ -138889,16 +139342,14 @@
138889	assert( pParse->pNewTable==0 );
138890	assert( pParse->pNewTrigger==0 );
138891	assert( pParse->nVar==0 );
138892	assert( pParse->pVList==0 );
138893	while( 1 ){
138894	assert( i>=0 );
138895	if( zSql[i]!=0 ){
138896	pParse->sLastToken.z = &zSql[i];
138897	pParse->sLastToken.n = sqlite3GetToken((u8*)&zSql[i],&tokenType);
138898	i += pParse->sLastToken.n;
138899	if( i>mxSqlLen ){
138900	pParse->rc = SQLITE_TOOBIG;
138901	break;
138902	}
138903	}else{
138904	/* Upon reaching the end of input, call the parser two more times
	@@ -138908,30 +139359,34 @@
138908	}else if( lastTokenParsed==0 ){
138909	break;
138910	}else{
138911	tokenType = TK_SEMI;
138912	}

138913	}
138914	if( tokenType>=TK_SPACE ){
138915	assert( tokenType==TK_SPACE \|\| tokenType==TK_ILLEGAL );
138916	if( db->u1.isInterrupted ){
138917	pParse->rc = SQLITE_INTERRUPT;
138918	break;
138919	}
138920	if( tokenType==TK_ILLEGAL ){
138921	sqlite3ErrorMsg(pParse, "unrecognized token: \"%T\"",
138922	&pParse->sLastToken);
138923	break;
138924	}

138925	}else{


138926	sqlite3Parser(pEngine, tokenType, pParse->sLastToken, pParse);
138927	lastTokenParsed = tokenType;

138928	if( pParse->rc!=SQLITE_OK \|\| db->mallocFailed ) break;
138929	}
138930	}
138931	assert( nErr==0 );
138932	pParse->zTail = &zSql[i];
138933	#ifdef YYTRACKMAXSTACKDEPTH
138934	sqlite3_mutex_enter(sqlite3MallocMutex());
138935	sqlite3StatusHighwater(SQLITE_STATUS_PARSER_STACK,
138936	sqlite3ParserStackPeak(pEngine)
138937	);
	@@ -140301,10 +140756,25 @@
140301	return 0;
140302	}
140303	#endif
140304	return db->lastRowid;
140305	}















140306
140307	/*
140308	** Return the number of changes in the most recent call to sqlite3_exec().
140309	*/
140310	SQLITE_API int sqlite3_changes(sqlite3 *db){
	@@ -148111,12 +148581,14 @@
148111	** segments.
148112	*/
148113	const u32 nMinMerge = 64; /* Minimum amount of incr-merge work to do */
148114
148115	Fts3Table p = (Fts3Table)pVtab;
148116	int rc = sqlite3Fts3PendingTermsFlush(p);

148117

148118	if( rc==SQLITE_OK
148119	&& p->nLeafAdd>(nMinMerge/16)
148120	&& p->nAutoincrmerge && p->nAutoincrmerge!=0xff
148121	){
148122	int mxLevel = 0; /* Maximum relative level value in db */
	@@ -148127,10 +148599,11 @@
148127	A = p->nLeafAdd * mxLevel;
148128	A += (A/2);
148129	if( A>(int)nMinMerge ) rc = sqlite3Fts3Incrmerge(p, A, p->nAutoincrmerge);
148130	}
148131	sqlite3Fts3SegmentsClose(p);

148132	return rc;
148133	}
148134
148135	/*
148136	** If it is currently unknown whether or not the FTS table has an %_stat
	@@ -168462,10 +168935,11 @@
168462	int nStep; /* Rows processed for current object */
168463	int nProgress; /* Rows processed for all objects */
168464	RbuObjIter objiter; /* Iterator for skipping through tbl/idx */
168465	const char zVfsName; / Name of automatically created rbu vfs */
168466	rbu_file pTargetFd; / File handle open on target db */

168467	i64 iOalSz;
168468	i64 nPhaseOneStep;
168469
168470	/* The following state variables are used as part of the incremental
168471	** checkpoint stage (eStage==RBU_STAGE_CKPT). See comments surrounding
	@@ -170726,10 +171200,27 @@
170726
170727	if( p->rc==SQLITE_OK ){
170728	if( p->nFrame==0 \|\| (pState && pState->iWalCksum!=p->iWalCksum) ){
170729	p->rc = SQLITE_DONE;
170730	p->eStage = RBU_STAGE_DONE;

















170731	}
170732	}
170733	}
170734
170735	/*
	@@ -171381,13 +171872,30 @@
171381	if( p->rc==SQLITE_OK ){
171382	p->eStage = RBU_STAGE_DONE;
171383	p->rc = SQLITE_DONE;
171384	}
171385	}else{
171386	RbuFrame *pFrame = &p->aFrame[p->nStep];
171387	rbuCheckpointFrame(p, pFrame);
171388	p->nStep++;

















171389	}
171390	p->nProgress++;
171391	}
171392	break;
171393	}
	@@ -171823,10 +172331,16 @@
171823
171824	/* Commit the transaction to the -oal file. /
171825	if( p->rc==SQLITE_OK && p->eStage==RBU_STAGE_OAL ){
171826	p->rc = sqlite3_exec(p->dbMain, "COMMIT", 0, 0, &p->zErrmsg);
171827	}






171828
171829	rbuSaveState(p, p->eStage);
171830
171831	if( p->rc==SQLITE_OK && p->eStage==RBU_STAGE_OAL ){
171832	p->rc = sqlite3_exec(p->dbRbu, "COMMIT", 0, 0, &p->zErrmsg);
	@@ -171947,10 +172461,16 @@
171947	assert( p->eStage>=RBU_STAGE_OAL && p->eStage<=RBU_STAGE_DONE );
171948	if( p->eStage==RBU_STAGE_OAL ){
171949	assert( rc!=SQLITE_DONE );
171950	if( rc==SQLITE_OK ) rc = sqlite3_exec(p->dbMain, "COMMIT", 0, 0, 0);
171951	}






171952
171953	p->rc = rc;
171954	rbuSaveState(p, p->eStage);
171955	rc = p->rc;
171956
	@@ -181079,11 +181599,13 @@
181079	typedef unsigned short u16;
181080	typedef short i16;
181081	typedef sqlite3_int64 i64;
181082	typedef sqlite3_uint64 u64;
181083
181084	#define ArraySize(x) ((int)(sizeof(x) / sizeof(x[0])))


181085
181086	#define testcase(x)
181087	#define ALWAYS(x) 1
181088	#define NEVER(x) 0
181089
	@@ -186302,11 +186824,14 @@
186302	if( p1->bEof==0 ){
186303	if( (p1->iRowid==iLast)
186304	\|\| (bFromValid && fts5RowidCmp(pExpr, p1->iRowid, iFrom)<0)
186305	){
186306	int rc = fts5ExprNodeNext(pExpr, p1, bFromValid, iFrom);
186307	if( rc!=SQLITE_OK ) return rc;



186308	}
186309	}
186310	}
186311
186312	fts5ExprNodeTest_OR(pExpr, pNode);
	@@ -186333,11 +186858,14 @@
186333	Fts5ExprNode *pChild = pAnd->apChild[iChild];
186334	int cmp = fts5RowidCmp(pExpr, iLast, pChild->iRowid);
186335	if( cmp>0 ){
186336	/* Advance pChild until it points to iLast or laster */
186337	rc = fts5ExprNodeNext(pExpr, pChild, 1, iLast);
186338	if( rc!=SQLITE_OK ) return rc;



186339	}
186340
186341	/* If the child node is now at EOF, so is the parent AND node. Otherwise,
186342	** the child node is guaranteed to have advanced at least as far as
186343	** rowid iLast. So if it is not at exactly iLast, pChild->iRowid is the
	@@ -186372,10 +186900,12 @@
186372	i64 iFrom
186373	){
186374	int rc = fts5ExprNodeNext(pExpr, pNode->apChild[0], bFromValid, iFrom);
186375	if( rc==SQLITE_OK ){
186376	rc = fts5ExprNodeTest_AND(pExpr, pNode);


186377	}
186378	return rc;
186379	}
186380
186381	static int fts5ExprNodeTest_NOT(
	@@ -186414,10 +186944,13 @@
186414	){
186415	int rc = fts5ExprNodeNext(pExpr, pNode->apChild[0], bFromValid, iFrom);
186416	if( rc==SQLITE_OK ){
186417	rc = fts5ExprNodeTest_NOT(pExpr, pNode);
186418	}



186419	return rc;
186420	}
186421
186422	/*
186423	** If pNode currently points to a match, this function returns SQLITE_OK
	@@ -197534,11 +198067,11 @@
197534	int nArg, /* Number of args */
197535	sqlite3_value *apUnused / Function arguments */
197536	){
197537	assert( nArg==0 );
197538	UNUSED_PARAM2(nArg, apUnused);
197539	sqlite3_result_text(pCtx, "fts5: 2017-02-13 16:02:40 ada05cfa86ad7f5645450ac7a2a21c9aa6e57d2c", -1, SQLITE_TRANSIENT);
197540	}
197541
197542	static int fts5Init(sqlite3 *db){
197543	static const sqlite3_module fts5Mod = {
197544	/* iVersion */ 2,
	@@ -198197,15 +198730,10 @@
198197	sqlite3_step(pDel);
198198	rc = sqlite3_reset(pDel);
198199	}
198200	}
198201
198202	/* Write the averages record */
198203	if( rc==SQLITE_OK ){
198204	rc = fts5StorageSaveTotals(p);
198205	}
198206
198207	return rc;
198208	}
198209
198210	/*
198211	** Delete all entries in the FTS5 index.
	@@ -198405,15 +198933,10 @@
198405	if( rc==SQLITE_OK ){
198406	rc = fts5StorageInsertDocsize(p, iRowid, &buf);
198407	}
198408	sqlite3_free(buf.p);
198409
198410	/* Write the averages record */
198411	if( rc==SQLITE_OK ){
198412	rc = fts5StorageSaveTotals(p);
198413	}
198414
198415	return rc;
198416	}
198417
198418	static int fts5StorageCount(Fts5Storage p, const char zSuffix, i64 *pnRow){
198419	Fts5Config *pConfig = p->pConfig;
	@@ -198744,16 +199267,21 @@
198744
198745	/*
198746	** Flush any data currently held in-memory to disk.
198747	*/
198748	static int sqlite3Fts5StorageSync(Fts5Storage *p, int bCommit){
198749	if( bCommit && p->bTotalsValid ){
198750	int rc = fts5StorageSaveTotals(p);
198751	p->bTotalsValid = 0;
198752	if( rc!=SQLITE_OK ) return rc;

198753	}
198754	return sqlite3Fts5IndexSync(p->pIndex, bCommit);




198755	}
198756
198757	static int sqlite3Fts5StorageRollback(Fts5Storage *p){
198758	p->bTotalsValid = 0;
198759	return sqlite3Fts5IndexRollback(p->pIndex);
198760

	--- src/sqlite3.c
	+++ src/sqlite3.c
	@@ -1,8 +1,8 @@
1	/******************************************************************************
2	** This file is an amalgamation of many separate C source files from SQLite
3	** version 3.18.0. By combining all the individual C code files into this
4	** single large file, the entire code can be compiled as a single translation
5	** unit. This allows many compilers to do optimizations that would not be
6	** possible if the files were compiled separately. Performance improvements
7	** of 5% or more are commonly seen when SQLite is compiled as a single
8	** translation unit.
	@@ -396,13 +396,13 @@
396	**
397	** See also: [sqlite3_libversion()],
398	** [sqlite3_libversion_number()], [sqlite3_sourceid()],
399	** [sqlite_version()] and [sqlite_source_id()].
400	*/
401	#define SQLITE_VERSION "3.18.0"
402	#define SQLITE_VERSION_NUMBER 3018000
403	#define SQLITE_SOURCE_ID "2017-03-10 17:03:11 f8560c60d10c0365b33342ab05b5a953987b0471"
404
405	/*
406	** CAPI3REF: Run-Time Library Version Numbers
407	** KEYWORDS: sqlite3_version sqlite3_sourceid
408	**
	@@ -2315,24 +2315,34 @@
2315	** as an undeclared column named ROWID, OID, or _ROWID_ as long as those
2316	** names are not also used by explicitly declared columns. ^If
2317	** the table has a column of type [INTEGER PRIMARY KEY] then that column
2318	** is another alias for the rowid.
2319	**
2320	** ^The sqlite3_last_insert_rowid(D) interface usually returns the [rowid] of
2321	** the most recent successful [INSERT] into a rowid table or [virtual table]
2322	** on database connection D. ^Inserts into [WITHOUT ROWID] tables are not
2323	** recorded. ^If no successful [INSERT]s into rowid tables have ever occurred
2324	** on the database connection D, then sqlite3_last_insert_rowid(D) returns
2325	** zero.

2326	**
2327	** As well as being set automatically as rows are inserted into database
2328	** tables, the value returned by this function may be set explicitly by
2329	** [sqlite3_set_last_insert_rowid()]
2330	**
2331	** Some virtual table implementations may INSERT rows into rowid tables as
2332	** part of committing a transaction (e.g. to flush data accumulated in memory
2333	** to disk). In this case subsequent calls to this function return the rowid
2334	** associated with these internal INSERT operations, which leads to
2335	** unintuitive results. Virtual table implementations that do write to rowid
2336	** tables in this way can avoid this problem by restoring the original
2337	** rowid value using [sqlite3_set_last_insert_rowid()] before returning
2338	** control to the user.
2339	**
2340	** ^(If an [INSERT] occurs within a trigger then this routine will
2341	** return the [rowid] of the inserted row as long as the trigger is
2342	** running. Once the trigger program ends, the value returned
2343	** by this routine reverts to what it was before the trigger was fired.)^
2344	**
2345	** ^An [INSERT] that fails due to a constraint violation is not a
2346	** successful [INSERT] and does not change the value returned by this
2347	** routine. ^Thus INSERT OR FAIL, INSERT OR IGNORE, INSERT OR ROLLBACK,
2348	** and INSERT OR ABORT make no changes to the return value of this
	@@ -2355,10 +2365,20 @@
2365	** unpredictable and might not equal either the old or the new
2366	** last insert [rowid].
2367	*/
2368	SQLITE_API sqlite3_int64 sqlite3_last_insert_rowid(sqlite3*);
2369
2370	/*
2371	** CAPI3REF: Set the Last Insert Rowid value.
2372	** METHOD: sqlite3
2373	**
2374	** The sqlite3_set_last_insert_rowid(D, R) method allows the application to
2375	** set the value returned by calling sqlite3_last_insert_rowid(D) to R
2376	** without inserting a row into the database.
2377	*/
2378	SQLITE_API void sqlite3_set_last_insert_rowid(sqlite3*,sqlite3_int64);
2379
2380	/*
2381	** CAPI3REF: Count The Number Of Rows Modified
2382	** METHOD: sqlite3
2383	**
2384	** ^This function returns the number of rows modified, inserted or
	@@ -12457,10 +12477,11 @@
12477	SQLITE_PRIVATE const void sqlite3BtreePayloadFetch(BtCursor, u32 *pAmt);
12478	SQLITE_PRIVATE u32 sqlite3BtreePayloadSize(BtCursor*);
12479
12480	SQLITE_PRIVATE char sqlite3BtreeIntegrityCheck(Btree, int aRoot, int nRoot, int, int);
12481	SQLITE_PRIVATE struct Pager sqlite3BtreePager(Btree);
12482	SQLITE_PRIVATE i64 sqlite3BtreeRowCountEst(BtCursor*);
12483
12484	#ifndef SQLITE_OMIT_INCRBLOB
12485	SQLITE_PRIVATE int sqlite3BtreePayloadChecked(BtCursor, u32 offset, u32 amt, void);
12486	SQLITE_PRIVATE int sqlite3BtreePutData(BtCursor, u32 offset, u32 amt, void);
12487	SQLITE_PRIVATE void sqlite3BtreeIncrblobCursor(BtCursor *);
	@@ -12757,117 +12778,119 @@
12778	#define OP_Divide 50 /* same as TK_SLASH, synopsis: r[P3]=r[P2]/r[P1] */
12779	#define OP_Remainder 51 /* same as TK_REM, synopsis: r[P3]=r[P2]%r[P1] */
12780	#define OP_Concat 52 /* same as TK_CONCAT, synopsis: r[P3]=r[P2]+r[P1] */
12781	#define OP_Last 53
12782	#define OP_BitNot 54 /* same as TK_BITNOT, synopsis: r[P1]= ~r[P1] */
12783	#define OP_IfSmaller 55
12784	#define OP_SorterSort 56
12785	#define OP_Sort 57
12786	#define OP_Rewind 58
12787	#define OP_IdxLE 59 /* synopsis: key=r[P3@P4] */
12788	#define OP_IdxGT 60 /* synopsis: key=r[P3@P4] */
12789	#define OP_IdxLT 61 /* synopsis: key=r[P3@P4] */
12790	#define OP_IdxGE 62 /* synopsis: key=r[P3@P4] */
12791	#define OP_RowSetRead 63 /* synopsis: r[P3]=rowset(P1) */
12792	#define OP_RowSetTest 64 /* synopsis: if r[P3] in rowset(P1) goto P2 */
12793	#define OP_Program 65
12794	#define OP_FkIfZero 66 /* synopsis: if fkctr[P1]==0 goto P2 */
12795	#define OP_IfPos 67 /* synopsis: if r[P1]>0 then r[P1]-=P3, goto P2 */
12796	#define OP_IfNotZero 68 /* synopsis: if r[P1]!=0 then r[P1]--, goto P2 */
12797	#define OP_DecrJumpZero 69 /* synopsis: if (--r[P1])==0 goto P2 */
12798	#define OP_IncrVacuum 70
12799	#define OP_VNext 71
12800	#define OP_Init 72 /* synopsis: Start at P2 */
12801	#define OP_Return 73
12802	#define OP_EndCoroutine 74
12803	#define OP_HaltIfNull 75 /* synopsis: if r[P3]=null halt */
12804	#define OP_Halt 76
12805	#define OP_Integer 77 /* synopsis: r[P2]=P1 */
12806	#define OP_Int64 78 /* synopsis: r[P2]=P4 */
12807	#define OP_String 79 /* synopsis: r[P2]='P4' (len=P1) */
12808	#define OP_Null 80 /* synopsis: r[P2..P3]=NULL */
12809	#define OP_SoftNull 81 /* synopsis: r[P1]=NULL */
12810	#define OP_Blob 82 /* synopsis: r[P2]=P4 (len=P1) */
12811	#define OP_Variable 83 /* synopsis: r[P2]=parameter(P1,P4) */
12812	#define OP_Move 84 /* synopsis: r[P2@P3]=r[P1@P3] */
12813	#define OP_Copy 85 /* synopsis: r[P2@P3+1]=r[P1@P3+1] */
12814	#define OP_SCopy 86 /* synopsis: r[P2]=r[P1] */
12815	#define OP_IntCopy 87 /* synopsis: r[P2]=r[P1] */
12816	#define OP_ResultRow 88 /* synopsis: output=r[P1@P2] */
12817	#define OP_CollSeq 89
12818	#define OP_Function0 90 /* synopsis: r[P3]=func(r[P2@P5]) */
12819	#define OP_Function 91 /* synopsis: r[P3]=func(r[P2@P5]) */
12820	#define OP_AddImm 92 /* synopsis: r[P1]=r[P1]+P2 */
12821	#define OP_RealAffinity 93
12822	#define OP_Cast 94 /* synopsis: affinity(r[P1]) */
12823	#define OP_Permutation 95
12824	#define OP_Compare 96 /* synopsis: r[P1@P3] <-> r[P2@P3] */
12825	#define OP_String8 97 /* same as TK_STRING, synopsis: r[P2]='P4' */
12826	#define OP_Column 98 /* synopsis: r[P3]=PX */
12827	#define OP_Affinity 99 /* synopsis: affinity(r[P1@P2]) */
12828	#define OP_MakeRecord 100 /* synopsis: r[P3]=mkrec(r[P1@P2]) */
12829	#define OP_Count 101 /* synopsis: r[P2]=count() */
12830	#define OP_ReadCookie 102
12831	#define OP_SetCookie 103
12832	#define OP_ReopenIdx 104 /* synopsis: root=P2 iDb=P3 */
12833	#define OP_OpenRead 105 /* synopsis: root=P2 iDb=P3 */
12834	#define OP_OpenWrite 106 /* synopsis: root=P2 iDb=P3 */
12835	#define OP_OpenAutoindex 107 /* synopsis: nColumn=P2 */
12836	#define OP_OpenEphemeral 108 /* synopsis: nColumn=P2 */
12837	#define OP_SorterOpen 109
12838	#define OP_SequenceTest 110 /* synopsis: if( cursor[P1].ctr++ ) pc = P2 */
12839	#define OP_OpenPseudo 111 /* synopsis: P3 columns in r[P2] */
12840	#define OP_Close 112
12841	#define OP_ColumnsUsed 113
12842	#define OP_Sequence 114 /* synopsis: r[P2]=cursor[P1].ctr++ */
12843	#define OP_NewRowid 115 /* synopsis: r[P2]=rowid */
12844	#define OP_Insert 116 /* synopsis: intkey=r[P3] data=r[P2] */
12845	#define OP_InsertInt 117 /* synopsis: intkey=P3 data=r[P2] */
12846	#define OP_Delete 118
12847	#define OP_ResetCount 119
12848	#define OP_SorterCompare 120 /* synopsis: if key(P1)!=trim(r[P3],P4) goto P2 */
12849	#define OP_SorterData 121 /* synopsis: r[P2]=data */
12850	#define OP_RowData 122 /* synopsis: r[P2]=data */
12851	#define OP_Rowid 123 /* synopsis: r[P2]=rowid */
12852	#define OP_NullRow 124
12853	#define OP_SorterInsert 125 /* synopsis: key=r[P2] */
12854	#define OP_IdxInsert 126 /* synopsis: key=r[P2] */
12855	#define OP_IdxDelete 127 /* synopsis: key=r[P2@P3] */
12856	#define OP_Seek 128 /* synopsis: Move P3 to P1.rowid */
12857	#define OP_IdxRowid 129 /* synopsis: r[P2]=rowid */
12858	#define OP_Destroy 130
12859	#define OP_Clear 131
12860	#define OP_Real 132 /* same as TK_FLOAT, synopsis: r[P2]=P4 */
12861	#define OP_ResetSorter 133
12862	#define OP_CreateIndex 134 /* synopsis: r[P2]=root iDb=P1 */
12863	#define OP_CreateTable 135 /* synopsis: r[P2]=root iDb=P1 */
12864	#define OP_SqlExec 136
12865	#define OP_ParseSchema 137
12866	#define OP_LoadAnalysis 138
12867	#define OP_DropTable 139
12868	#define OP_DropIndex 140
12869	#define OP_DropTrigger 141
12870	#define OP_IntegrityCk 142
12871	#define OP_RowSetAdd 143 /* synopsis: rowset(P1)=r[P2] */
12872	#define OP_Param 144
12873	#define OP_FkCounter 145 /* synopsis: fkctr[P1]+=P2 */
12874	#define OP_MemMax 146 /* synopsis: r[P1]=max(r[P1],r[P2]) */
12875	#define OP_OffsetLimit 147 /* synopsis: if r[P1]>0 then r[P2]=r[P1]+max(0,r[P3]) else r[P2]=(-1) */
12876	#define OP_AggStep0 148 /* synopsis: accum=r[P3] step(r[P2@P5]) */
12877	#define OP_AggStep 149 /* synopsis: accum=r[P3] step(r[P2@P5]) */
12878	#define OP_AggFinal 150 /* synopsis: accum=r[P1] N=P2 */
12879	#define OP_Expire 151
12880	#define OP_TableLock 152 /* synopsis: iDb=P1 root=P2 write=P3 */
12881	#define OP_VBegin 153
12882	#define OP_VCreate 154
12883	#define OP_VDestroy 155
12884	#define OP_VOpen 156
12885	#define OP_VColumn 157 /* synopsis: r[P3]=vcolumn(P2) */
12886	#define OP_VRename 158
12887	#define OP_Pagecount 159
12888	#define OP_MaxPgcnt 160
12889	#define OP_CursorHint 161
12890	#define OP_Noop 162
12891	#define OP_Explain 163
12892
12893	/* Properties such as "out2" or "jump" that are specified in
12894	** comments following the "case" for each opcode in the vdbe.c
12895	** are encoded into bitvectors as follows:
12896	*/
	@@ -12883,32 +12906,32 @@
12906	/* 16 */ 0x03, 0x03, 0x01, 0x12, 0x01, 0x03, 0x03, 0x09,\
12907	/* 24 */ 0x09, 0x09, 0x09, 0x26, 0x26, 0x09, 0x09, 0x09,\
12908	/* 32 */ 0x09, 0x09, 0x03, 0x03, 0x0b, 0x0b, 0x0b, 0x0b,\
12909	/* 40 */ 0x0b, 0x0b, 0x01, 0x26, 0x26, 0x26, 0x26, 0x26,\
12910	/* 48 */ 0x26, 0x26, 0x26, 0x26, 0x26, 0x01, 0x12, 0x01,\
12911	/* 56 */ 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x23,\
12912	/* 64 */ 0x0b, 0x01, 0x01, 0x03, 0x03, 0x03, 0x01, 0x01,\
12913	/* 72 */ 0x01, 0x02, 0x02, 0x08, 0x00, 0x10, 0x10, 0x10,\
12914	/* 80 */ 0x10, 0x00, 0x10, 0x10, 0x00, 0x00, 0x10, 0x10,\
12915	/* 88 */ 0x00, 0x00, 0x00, 0x00, 0x02, 0x02, 0x02, 0x00,\
12916	/* 96 */ 0x00, 0x10, 0x00, 0x00, 0x00, 0x10, 0x10, 0x00,\
12917	/* 104 */ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,\
12918	/* 112 */ 0x00, 0x00, 0x10, 0x10, 0x00, 0x00, 0x00, 0x00,\
12919	/* 120 */ 0x00, 0x00, 0x00, 0x10, 0x00, 0x04, 0x04, 0x00,\
12920	/* 128 */ 0x00, 0x10, 0x10, 0x00, 0x10, 0x00, 0x10, 0x10,\
12921	/* 136 */ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x06,\
12922	/* 144 */ 0x10, 0x00, 0x04, 0x1a, 0x00, 0x00, 0x00, 0x00,\
12923	/* 152 */ 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x10,\
12924	/* 160 */ 0x10, 0x00, 0x00, 0x00,}
12925
12926	/* The sqlite3P2Values() routine is able to run faster if it knows
12927	** the value of the largest JUMP opcode. The smaller the maximum
12928	** JUMP opcode the better, so the mkopcodeh.tcl script that
12929	** generated this include file strives to group all JUMP opcodes
12930	** together near the beginning of the list.
12931	*/
12932	#define SQLITE_MX_JUMP_OPCODE 72 /* Maximum JUMP opcode */
12933
12934	/************ End of opcodes.h *******************************************/
12935	/************ Continuing where we left off in vdbe.h *********************/
12936
12937	/*
	@@ -14074,10 +14097,11 @@
14097	u8 suppressErr; /* Do not issue error messages if true */
14098	u8 vtabOnConflict; /* Value to return for s3_vtab_on_conflict() */
14099	u8 isTransactionSavepoint; /* True if the outermost savepoint is a TS */
14100	u8 mTrace; /* zero or more SQLITE_TRACE flags */
14101	u8 skipBtreeMutex; /* True if no shared-cache backends */
14102	u8 nSqlExec; /* Number of pending OP_SqlExec opcodes */
14103	int nextPagesize; /* Pagesize after VACUUM if >0 */
14104	u32 magic; /* Magic number for detect library misuse */
14105	int nChange; /* Value returned by sqlite3_changes() */
14106	int nTotalChange; /* Value returned by sqlite3_total_changes() */
14107	int aLimit[SQLITE_N_LIMIT]; /* Limits */
	@@ -14589,18 +14613,18 @@
14613	char zColAff; / String defining the affinity of each column */
14614	ExprList pCheck; / All CHECK constraints */
14615	/* ... also used as column name list in a VIEW */
14616	int tnum; /* Root BTree page for this table */
14617	u32 nTabRef; /* Number of pointers to this Table */
14618	u32 tabFlags; /* Mask of TF_* values */
14619	i16 iPKey; /* If not negative, use aCol[iPKey] as the rowid */
14620	i16 nCol; /* Number of columns in this table */
14621	LogEst nRowLogEst; /* Estimated rows in table - from sqlite_stat1 table */
14622	LogEst szTabRow; /* Estimated size of each table row in bytes */
14623	#ifdef SQLITE_ENABLE_COSTMULT
14624	LogEst costMult; /* Cost multiplier for using this table */
14625	#endif

14626	u8 keyConf; /* What to do in case of uniqueness conflict on iPKey */
14627	#ifndef SQLITE_OMIT_ALTERTABLE
14628	int addColOffset; /* Offset in CREATE TABLE stmt to add a new column */
14629	#endif
14630	#ifndef SQLITE_OMIT_VIRTUALTABLE
	@@ -14620,27 +14644,29 @@
14644	** followed by non-hidden columns. Example: "CREATE VIRTUAL TABLE x USING
14645	** vtab1(a HIDDEN, b);". Since "b" is a non-hidden column but "a" is hidden,
14646	** the TF_OOOHidden attribute would apply in this case. Such tables require
14647	** special handling during INSERT processing.
14648	*/
14649	#define TF_Readonly 0x0001 /* Read-only system table */
14650	#define TF_Ephemeral 0x0002 /* An ephemeral table */
14651	#define TF_HasPrimaryKey 0x0004 /* Table has a primary key */
14652	#define TF_Autoincrement 0x0008 /* Integer primary key is autoincrement */
14653	#define TF_HasStat1 0x0010 /* nRowLogEst set from sqlite_stat1 */
14654	#define TF_WithoutRowid 0x0020 /* No rowid. PRIMARY KEY is the key */
14655	#define TF_NoVisibleRowid 0x0040 /* No user-visible "rowid" column */
14656	#define TF_OOOHidden 0x0080 /* Out-of-Order hidden columns */
14657	#define TF_StatsUsed 0x0100 /* Query planner decisions affected by
14658	** Index.aiRowLogEst[] values */
14659	#define TF_HasNotNull 0x0200 /* Contains NOT NULL constraints */
14660
14661	/*
14662	** Test to see whether or not a table is a virtual table. This is
14663	** done as a macro so that it will be optimized out when virtual
14664	** table support is omitted from the build.
14665	*/
14666	#ifndef SQLITE_OMIT_VIRTUALTABLE
14667	# define IsVirtual(X) ((X)->nModuleArg)
14668	#else
14669	# define IsVirtual(X) 0
14670	#endif
14671
14672	/*
	@@ -14871,10 +14897,11 @@
14897	unsigned bUnordered:1; /* Use this index for == or IN queries only */
14898	unsigned uniqNotNull:1; /* True if UNIQUE and NOT NULL for all columns */
14899	unsigned isResized:1; /* True if resizeIndexObject() has been called */
14900	unsigned isCovering:1; /* True if this is a covering index */
14901	unsigned noSkipScan:1; /* Do not try to use skip-scan if true */
14902	unsigned hasStat1:1; /* aiRowLogEst values come from sqlite_stat1 */
14903	#ifdef SQLITE_ENABLE_STAT3_OR_STAT4
14904	int nSample; /* Number of elements in aSample[] */
14905	int nSampleCol; /* Size of IndexSample.anEq[] and so on */
14906	tRowcnt aAvgEq; / Average nEq values for keys not in aSample */
14907	IndexSample aSample; / Samples of the left-most key */
	@@ -15181,11 +15208,11 @@
15208	** form is used for name resolution with nested FROM clauses.
15209	*/
15210	struct ExprList {
15211	int nExpr; /* Number of expressions on the list */
15212	struct ExprList_item { /* For each expression in the list */
15213	Expr pExpr; / The parse tree for this expression */
15214	char zName; / Token associated with this expression */
15215	char zSpan; / Original text of the expression */
15216	u8 sortOrder; /* 1 for DESC or 0 for ASC */
15217	unsigned done :1; /* A flag to indicate when processing is finished */
15218	unsigned bSpanIsTab :1; /* zSpan holds DB.TABLE.COLUMN */
	@@ -16504,10 +16531,11 @@
16531	SQLITE_PRIVATE void sqlite3UnlinkAndDeleteIndex(sqlite3,int,const char);
16532	SQLITE_PRIVATE void sqlite3Vacuum(Parse,Token);
16533	SQLITE_PRIVATE int sqlite3RunVacuum(char*, sqlite3, int);
16534	SQLITE_PRIVATE char sqlite3NameFromToken(sqlite3, Token*);
16535	SQLITE_PRIVATE int sqlite3ExprCompare(Expr, Expr, int);
16536	SQLITE_PRIVATE int sqlite3ExprCompareSkip(Expr, Expr, int);
16537	SQLITE_PRIVATE int sqlite3ExprListCompare(ExprList, ExprList, int);
16538	SQLITE_PRIVATE int sqlite3ExprImpliesExpr(Expr, Expr, int);
16539	SQLITE_PRIVATE void sqlite3ExprAnalyzeAggregates(NameContext, Expr);
16540	SQLITE_PRIVATE void sqlite3ExprAnalyzeAggList(NameContext,ExprList);
16541	SQLITE_PRIVATE int sqlite3ExprCoveredByIndex(Expr, int iCur, Index pIdx);
	@@ -17450,11 +17478,11 @@
17478	"COMPILER=gcc-" __VERSION__,
17479	#endif
17480	#if SQLITE_COVERAGE_TEST
17481	"COVERAGE_TEST",
17482	#endif
17483	#ifdef SQLITE_DEBUG
17484	"DEBUG",
17485	#endif
17486	#if SQLITE_DEFAULT_LOCKING_MODE
17487	"DEFAULT_LOCKING_MODE=" CTIMEOPT_VAL(SQLITE_DEFAULT_LOCKING_MODE),
17488	#endif
	@@ -19505,22 +19533,23 @@
19533	**
19534	** Move the date backwards to the beginning of the current day,
19535	** or month or year.
19536	*/
19537	if( sqlite3_strnicmp(z, "start of ", 9)!=0 ) break;
19538	if( !p->validJD && !p->validYMD && !p->validHMS ) break;
19539	z += 9;
19540	computeYMD(p);
19541	p->validHMS = 1;
19542	p->h = p->m = 0;
19543	p->s = 0.0;
19544	p->rawS = 0;
19545	p->validTZ = 0;
19546	p->validJD = 0;
19547	if( sqlite3_stricmp(z,"month")==0 ){
19548	p->D = 1;
19549	rc = 0;
19550	}else if( sqlite3_stricmp(z,"year")==0 ){

19551	p->M = 1;
19552	p->D = 1;
19553	rc = 0;
19554	}else if( sqlite3_stricmp(z,"day")==0 ){
19555	rc = 0;
	@@ -24050,12 +24079,12 @@
24079	#ifdef SQLITE_DEBUG
24080	assert( p->nRef>0 \|\| p->owner==0 );
24081	p->owner = tid;
24082	p->nRef++;
24083	if( p->trace ){
24084	OSTRACE(("ENTER-MUTEX tid=%lu, mutex(%d)=%p (%d), nRef=%d\n",
24085	tid, p->id, p, p->trace, p->nRef));
24086	}
24087	#endif
24088	}
24089
24090	static int winMutexTry(sqlite3_mutex *p){
	@@ -24093,12 +24122,12 @@
24122	#else
24123	UNUSED_PARAMETER(p);
24124	#endif
24125	#ifdef SQLITE_DEBUG
24126	if( p->trace ){
24127	OSTRACE(("TRY-MUTEX tid=%lu, mutex(%d)=%p (%d), owner=%lu, nRef=%d, rc=%s\n",
24128	tid, p->id, p, p->trace, p->owner, p->nRef, sqlite3ErrName(rc)));
24129	}
24130	#endif
24131	return rc;
24132	}
24133
	@@ -24122,12 +24151,12 @@
24151	#endif
24152	assert( winMutex_isInit==1 );
24153	LeaveCriticalSection(&p->mutex);
24154	#ifdef SQLITE_DEBUG
24155	if( p->trace ){
24156	OSTRACE(("LEAVE-MUTEX tid=%lu, mutex(%d)=%p (%d), nRef=%d\n",
24157	tid, p->id, p, p->trace, p->nRef));
24158	}
24159	#endif
24160	}
24161
24162	SQLITE_PRIVATE sqlite3_mutex_methods const *sqlite3DefaultMutex(void){
	@@ -24383,10 +24412,17 @@
24412	** implementation of malloc_good_size(), which must be called in debug
24413	** mode and specifically when the DMD "Dark Matter Detector" is enabled
24414	** or else a crash results. Hence, do not attempt to optimize out the
24415	** following xRoundup() call. */
24416	nFull = sqlite3GlobalConfig.m.xRoundup(n);
24417
24418	#ifdef SQLITE_MAX_MEMORY
24419	if( sqlite3StatusValue(SQLITE_STATUS_MEMORY_USED)+nFull>SQLITE_MAX_MEMORY ){
24420	*pp = 0;
24421	return;
24422	}
24423	#endif
24424
24425	sqlite3StatusHighwater(SQLITE_STATUS_MALLOC_SIZE, n);
24426	if( mem0.alarmThreshold>0 ){
24427	sqlite3_int64 nUsed = sqlite3StatusValue(SQLITE_STATUS_MEMORY_USED);
24428	if( nUsed >= mem0.alarmThreshold - nFull ){
	@@ -24572,11 +24608,11 @@
24608	return sqlite3GlobalConfig.m.xSize(p);
24609	}
24610	SQLITE_PRIVATE int sqlite3DbMallocSize(sqlite3 db, void p){
24611	assert( p!=0 );
24612	if( db==0 \|\| !isLookaside(db,p) ){
24613	#ifdef SQLITE_DEBUG
24614	if( db==0 ){
24615	assert( sqlite3MemdebugNoType(p, (u8)~MEMTYPE_HEAP) );
24616	assert( sqlite3MemdebugHasType(p, MEMTYPE_HEAP) );
24617	}else{
24618	assert( sqlite3MemdebugHasType(p, (MEMTYPE_LOOKASIDE\|MEMTYPE_HEAP)) );
	@@ -24633,11 +24669,11 @@
24669	measureAllocationSize(db, p);
24670	return;
24671	}
24672	if( isLookaside(db, p) ){
24673	LookasideSlot pBuf = (LookasideSlot)p;
24674	#ifdef SQLITE_DEBUG
24675	/* Trash all content in the buffer being freed */
24676	memset(p, 0xaa, db->lookaside.sz);
24677	#endif
24678	pBuf->pNext = db->lookaside.pFree;
24679	db->lookaside.pFree = pBuf;
	@@ -25002,11 +25038,11 @@
25038
25039	/*
25040	** Conversion types fall into various categories as defined by the
25041	** following enumeration.
25042	*/
25043	#define etRADIX 0 /* non-decimal integer types. %x %o */
25044	#define etFLOAT 1 /* Floating point. %f */
25045	#define etEXP 2 /* Exponentional notation. %e and %E */
25046	#define etGENERIC 3 /* Floating or exponential, depending on exponent. %g */
25047	#define etSIZE 4 /* Return number of characters processed so far. %n */
25048	#define etSTRING 5 /* Strings. %s */
	@@ -25020,12 +25056,13 @@
25056	#define etTOKEN 11 /* a pointer to a Token structure */
25057	#define etSRCLIST 12 /* a pointer to a SrcList */
25058	#define etPOINTER 13 /* The %p conversion */
25059	#define etSQLESCAPE3 14 /* %w -> Strings with '\"' doubled */
25060	#define etORDINAL 15 /* %r -> 1st, 2nd, 3rd, 4th, etc. English only */
25061	#define etDECIMAL 16 /* %d or %u, but not %x, %o */
25062
25063	#define etINVALID 17 /* Any unrecognized conversion type */
25064
25065
25066	/*
25067	** An "etByte" is an 8-bit unsigned value.
25068	*/
	@@ -25045,40 +25082,40 @@
25082	} et_info;
25083
25084	/*
25085	** Allowed values for et_info.flags
25086	*/
25087	#define FLAG_SIGNED 1 /* True if the value to convert is signed */
25088	#define FLAG_STRING 4 /* Allow infinite precision */
25089
25090
25091	/*
25092	** The following table is searched linearly, so it is good to put the
25093	** most frequently used conversion types first.
25094	*/
25095	static const char aDigits[] = "0123456789ABCDEF0123456789abcdef";
25096	static const char aPrefix[] = "-x0\000X0";
25097	static const et_info fmtinfo[] = {
25098	{ 'd', 10, 1, etDECIMAL, 0, 0 },
25099	{ 's', 0, 4, etSTRING, 0, 0 },
25100	{ 'g', 0, 1, etGENERIC, 30, 0 },
25101	{ 'z', 0, 4, etDYNSTRING, 0, 0 },
25102	{ 'q', 0, 4, etSQLESCAPE, 0, 0 },
25103	{ 'Q', 0, 4, etSQLESCAPE2, 0, 0 },
25104	{ 'w', 0, 4, etSQLESCAPE3, 0, 0 },
25105	{ 'c', 0, 0, etCHARX, 0, 0 },
25106	{ 'o', 8, 0, etRADIX, 0, 2 },
25107	{ 'u', 10, 0, etDECIMAL, 0, 0 },
25108	{ 'x', 16, 0, etRADIX, 16, 1 },
25109	{ 'X', 16, 0, etRADIX, 0, 4 },
25110	#ifndef SQLITE_OMIT_FLOATING_POINT
25111	{ 'f', 0, 1, etFLOAT, 0, 0 },
25112	{ 'e', 0, 1, etEXP, 30, 0 },
25113	{ 'E', 0, 1, etEXP, 14, 0 },
25114	{ 'G', 0, 1, etGENERIC, 14, 0 },
25115	#endif
25116	{ 'i', 10, 1, etDECIMAL, 0, 0 },
25117	{ 'n', 0, 0, etSIZE, 0, 0 },
25118	{ '%', 0, 0, etPERCENT, 0, 0 },
25119	{ 'p', 16, 0, etPOINTER, 0, 1 },
25120
25121	/* All the rest are undocumented and are for internal use only */
	@@ -25166,18 +25203,17 @@
25203	int precision; /* Precision of the current field */
25204	int length; /* Length of the field */
25205	int idx; /* A general purpose loop counter */
25206	int width; /* Width of the current field */
25207	etByte flag_leftjustify; /* True if "-" flag is present */
25208	etByte flag_prefix; /* '+' or ' ' or 0 for prefix */

25209	etByte flag_alternateform; /* True if "#" flag is present */
25210	etByte flag_altform2; /* True if "!" flag is present */
25211	etByte flag_zeropad; /* True if field width constant starts with zero */
25212	etByte flag_long; /* 1 for the "l" flag, 2 for "ll", 0 by default */

25213	etByte done; /* Loop termination flag */
25214	etByte cThousand; /* Thousands separator for %d and %u */
25215	etByte xtype = etINVALID; /* Conversion paradigm */
25216	u8 bArgList; /* True for SQLITE_PRINTF_SQLFUNC */
25217	char prefix; /* Prefix character. "+" or "-" or " " or '\0'. */
25218	sqlite_uint64 longvalue; /* Value for integer types */
25219	LONGDOUBLE_TYPE realvalue; /* Value for real types */
	@@ -25216,21 +25252,22 @@
25252	if( (c=(*++fmt))==0 ){
25253	sqlite3StrAccumAppend(pAccum, "%", 1);
25254	break;
25255	}
25256	/* Find out what flags are present */
25257	flag_leftjustify = flag_prefix = cThousand =
25258	flag_alternateform = flag_altform2 = flag_zeropad = 0;
25259	done = 0;
25260	do{
25261	switch( c ){
25262	case '-': flag_leftjustify = 1; break;
25263	case '+': flag_prefix = '+'; break;
25264	case ' ': flag_prefix = ' '; break;
25265	case '#': flag_alternateform = 1; break;
25266	case '!': flag_altform2 = 1; break;
25267	case '0': flag_zeropad = 1; break;
25268	case ',': cThousand = ','; break;
25269	default: done = 1; break;
25270	}
25271	}while( !done && (c=(*++fmt))!=0 );
25272	/* Get the field width */
25273	if( c=='*' ){
	@@ -25296,17 +25333,15 @@
25333	/* Get the conversion type modifier */
25334	if( c=='l' ){
25335	flag_long = 1;
25336	c = *++fmt;
25337	if( c=='l' ){
25338	flag_long = 2;
25339	c = *++fmt;


25340	}
25341	}else{
25342	flag_long = 0;
25343	}
25344	/* Fetch the info entry for the field */
25345	infop = &fmtinfo[0];
25346	xtype = etINVALID;
25347	for(idx=0; idx<ArraySize(fmtinfo); idx++){
	@@ -25320,41 +25355,42 @@
25355	/*
25356	** At this point, variables are initialized as follows:
25357	**
25358	** flag_alternateform TRUE if a '#' is present.
25359	** flag_altform2 TRUE if a '!' is present.
25360	** flag_prefix '+' or ' ' or zero
25361	** flag_leftjustify TRUE if a '-' is present or if the
25362	** field width was negative.
25363	** flag_zeropad TRUE if the width began with 0.
25364	** flag_long 1 for "l", 2 for "ll"




25365	** width The specified field width. This is
25366	** always non-negative. Zero is the default.
25367	** precision The specified precision. The default
25368	** is -1.
25369	** xtype The class of the conversion.
25370	** infop Pointer to the appropriate info struct.
25371	*/
25372	switch( xtype ){
25373	case etPOINTER:
25374	flag_long = sizeof(char*)==sizeof(i64) ? 2 :
25375	sizeof(char*)==sizeof(long int) ? 1 : 0;
25376	/* Fall through into the next case */
25377	case etORDINAL:
25378	case etRADIX:
25379	cThousand = 0;
25380	/* Fall through into the next case */
25381	case etDECIMAL:
25382	if( infop->flags & FLAG_SIGNED ){
25383	i64 v;
25384	if( bArgList ){
25385	v = getIntArg(pArgList);


25386	}else if( flag_long ){
25387	if( flag_long==2 ){
25388	v = va_arg(ap,i64) ;
25389	}else{
25390	v = va_arg(ap,long int);
25391	}
25392	}else{
25393	v = va_arg(ap,int);
25394	}
25395	if( v<0 ){
25396	if( v==SMALLEST_INT64 ){
	@@ -25363,35 +25399,35 @@
25399	longvalue = -v;
25400	}
25401	prefix = '-';
25402	}else{
25403	longvalue = v;
25404	prefix = flag_prefix;


25405	}
25406	}else{
25407	if( bArgList ){
25408	longvalue = (u64)getIntArg(pArgList);


25409	}else if( flag_long ){
25410	if( flag_long==2 ){
25411	longvalue = va_arg(ap,u64);
25412	}else{
25413	longvalue = va_arg(ap,unsigned long int);
25414	}
25415	}else{
25416	longvalue = va_arg(ap,unsigned int);
25417	}
25418	prefix = 0;
25419	}
25420	if( longvalue==0 ) flag_alternateform = 0;
25421	if( flag_zeropad && precision<width-(prefix!=0) ){
25422	precision = width-(prefix!=0);
25423	}
25424	if( precision<etBUFSIZE-10-etBUFSIZE/3 ){
25425	nOut = etBUFSIZE;
25426	zOut = buf;
25427	}else{
25428	nOut = precision + 10 + precision/3;
25429	zOut = zExtra = sqlite3Malloc( nOut );
25430	if( zOut==0 ){
25431	setStrAccumError(pAccum, STRACCUM_NOMEM);
25432	return;
25433	}
	@@ -25413,12 +25449,27 @@
25449	*(--bufpt) = cset[longvalue%base];
25450	longvalue = longvalue/base;
25451	}while( longvalue>0 );
25452	}
25453	length = (int)(&zOut[nOut-1]-bufpt);
25454	while( precision>length ){
25455	(--bufpt) = '0'; / Zero pad */
25456	length++;
25457	}
25458	if( cThousand ){
25459	int nn = (length - 1)/3; /* Number of "," to insert */
25460	int ix = (length - 1)%3 + 1;
25461	bufpt -= nn;
25462	for(idx=0; nn>0; idx++){
25463	bufpt[idx] = bufpt[idx+nn];
25464	ix--;
25465	if( ix==0 ){
25466	bufpt[++idx] = cThousand;
25467	nn--;
25468	ix = 3;
25469	}
25470	}
25471	}
25472	if( prefix ) (--bufpt) = prefix; / Add sign */
25473	if( flag_alternateform && infop->prefix ){ /* Add "0" or "0x" */
25474	const char *pre;
25475	char x;
	@@ -25441,13 +25492,11 @@
25492	if( precision<0 ) precision = 6; /* Set default precision */
25493	if( realvalue<0.0 ){
25494	realvalue = -realvalue;
25495	prefix = '-';
25496	}else{
25497	prefix = flag_prefix;


25498	}
25499	if( xtype==etGENERIC && precision>0 ) precision--;
25500	testcase( precision>0xfff );
25501	for(idx=precision&0xfff, rounder=0.5; idx>0; idx--, rounder*=0.1){}
25502	if( xtype==etFLOAT ) realvalue += rounder;
	@@ -29464,117 +29513,119 @@
29513	/* 50 */ "Divide" OpHelp("r[P3]=r[P2]/r[P1]"),
29514	/* 51 */ "Remainder" OpHelp("r[P3]=r[P2]%r[P1]"),
29515	/* 52 */ "Concat" OpHelp("r[P3]=r[P2]+r[P1]"),
29516	/* 53 */ "Last" OpHelp(""),
29517	/* 54 */ "BitNot" OpHelp("r[P1]= ~r[P1]"),
29518	/* 55 */ "IfSmaller" OpHelp(""),
29519	/* 56 */ "SorterSort" OpHelp(""),
29520	/* 57 */ "Sort" OpHelp(""),
29521	/* 58 */ "Rewind" OpHelp(""),
29522	/* 59 */ "IdxLE" OpHelp("key=r[P3@P4]"),
29523	/* 60 */ "IdxGT" OpHelp("key=r[P3@P4]"),
29524	/* 61 */ "IdxLT" OpHelp("key=r[P3@P4]"),
29525	/* 62 */ "IdxGE" OpHelp("key=r[P3@P4]"),
29526	/* 63 */ "RowSetRead" OpHelp("r[P3]=rowset(P1)"),
29527	/* 64 */ "RowSetTest" OpHelp("if r[P3] in rowset(P1) goto P2"),
29528	/* 65 */ "Program" OpHelp(""),
29529	/* 66 */ "FkIfZero" OpHelp("if fkctr[P1]==0 goto P2"),
29530	/* 67 */ "IfPos" OpHelp("if r[P1]>0 then r[P1]-=P3, goto P2"),
29531	/* 68 */ "IfNotZero" OpHelp("if r[P1]!=0 then r[P1]--, goto P2"),
29532	/* 69 */ "DecrJumpZero" OpHelp("if (--r[P1])==0 goto P2"),
29533	/* 70 */ "IncrVacuum" OpHelp(""),
29534	/* 71 */ "VNext" OpHelp(""),
29535	/* 72 */ "Init" OpHelp("Start at P2"),
29536	/* 73 */ "Return" OpHelp(""),
29537	/* 74 */ "EndCoroutine" OpHelp(""),
29538	/* 75 */ "HaltIfNull" OpHelp("if r[P3]=null halt"),
29539	/* 76 */ "Halt" OpHelp(""),
29540	/* 77 */ "Integer" OpHelp("r[P2]=P1"),
29541	/* 78 */ "Int64" OpHelp("r[P2]=P4"),
29542	/* 79 */ "String" OpHelp("r[P2]='P4' (len=P1)"),
29543	/* 80 */ "Null" OpHelp("r[P2..P3]=NULL"),
29544	/* 81 */ "SoftNull" OpHelp("r[P1]=NULL"),
29545	/* 82 */ "Blob" OpHelp("r[P2]=P4 (len=P1)"),
29546	/* 83 */ "Variable" OpHelp("r[P2]=parameter(P1,P4)"),
29547	/* 84 */ "Move" OpHelp("r[P2@P3]=r[P1@P3]"),
29548	/* 85 */ "Copy" OpHelp("r[P2@P3+1]=r[P1@P3+1]"),
29549	/* 86 */ "SCopy" OpHelp("r[P2]=r[P1]"),
29550	/* 87 */ "IntCopy" OpHelp("r[P2]=r[P1]"),
29551	/* 88 */ "ResultRow" OpHelp("output=r[P1@P2]"),
29552	/* 89 */ "CollSeq" OpHelp(""),
29553	/* 90 */ "Function0" OpHelp("r[P3]=func(r[P2@P5])"),
29554	/* 91 */ "Function" OpHelp("r[P3]=func(r[P2@P5])"),
29555	/* 92 */ "AddImm" OpHelp("r[P1]=r[P1]+P2"),
29556	/* 93 */ "RealAffinity" OpHelp(""),
29557	/* 94 */ "Cast" OpHelp("affinity(r[P1])"),
29558	/* 95 */ "Permutation" OpHelp(""),
29559	/* 96 */ "Compare" OpHelp("r[P1@P3] <-> r[P2@P3]"),
29560	/* 97 */ "String8" OpHelp("r[P2]='P4'"),
29561	/* 98 */ "Column" OpHelp("r[P3]=PX"),
29562	/* 99 */ "Affinity" OpHelp("affinity(r[P1@P2])"),
29563	/* 100 */ "MakeRecord" OpHelp("r[P3]=mkrec(r[P1@P2])"),
29564	/* 101 */ "Count" OpHelp("r[P2]=count()"),
29565	/* 102 */ "ReadCookie" OpHelp(""),
29566	/* 103 */ "SetCookie" OpHelp(""),
29567	/* 104 */ "ReopenIdx" OpHelp("root=P2 iDb=P3"),
29568	/* 105 */ "OpenRead" OpHelp("root=P2 iDb=P3"),
29569	/* 106 */ "OpenWrite" OpHelp("root=P2 iDb=P3"),
29570	/* 107 */ "OpenAutoindex" OpHelp("nColumn=P2"),
29571	/* 108 */ "OpenEphemeral" OpHelp("nColumn=P2"),
29572	/* 109 */ "SorterOpen" OpHelp(""),
29573	/* 110 */ "SequenceTest" OpHelp("if( cursor[P1].ctr++ ) pc = P2"),
29574	/* 111 */ "OpenPseudo" OpHelp("P3 columns in r[P2]"),
29575	/* 112 */ "Close" OpHelp(""),
29576	/* 113 */ "ColumnsUsed" OpHelp(""),
29577	/* 114 */ "Sequence" OpHelp("r[P2]=cursor[P1].ctr++"),
29578	/* 115 */ "NewRowid" OpHelp("r[P2]=rowid"),
29579	/* 116 */ "Insert" OpHelp("intkey=r[P3] data=r[P2]"),
29580	/* 117 */ "InsertInt" OpHelp("intkey=P3 data=r[P2]"),
29581	/* 118 */ "Delete" OpHelp(""),
29582	/* 119 */ "ResetCount" OpHelp(""),
29583	/* 120 */ "SorterCompare" OpHelp("if key(P1)!=trim(r[P3],P4) goto P2"),
29584	/* 121 */ "SorterData" OpHelp("r[P2]=data"),
29585	/* 122 */ "RowData" OpHelp("r[P2]=data"),
29586	/* 123 */ "Rowid" OpHelp("r[P2]=rowid"),
29587	/* 124 */ "NullRow" OpHelp(""),
29588	/* 125 */ "SorterInsert" OpHelp("key=r[P2]"),
29589	/* 126 */ "IdxInsert" OpHelp("key=r[P2]"),
29590	/* 127 */ "IdxDelete" OpHelp("key=r[P2@P3]"),
29591	/* 128 */ "Seek" OpHelp("Move P3 to P1.rowid"),
29592	/* 129 */ "IdxRowid" OpHelp("r[P2]=rowid"),
29593	/* 130 */ "Destroy" OpHelp(""),
29594	/* 131 */ "Clear" OpHelp(""),
29595	/* 132 */ "Real" OpHelp("r[P2]=P4"),
29596	/* 133 */ "ResetSorter" OpHelp(""),
29597	/* 134 */ "CreateIndex" OpHelp("r[P2]=root iDb=P1"),
29598	/* 135 */ "CreateTable" OpHelp("r[P2]=root iDb=P1"),
29599	/* 136 */ "SqlExec" OpHelp(""),
29600	/* 137 */ "ParseSchema" OpHelp(""),
29601	/* 138 */ "LoadAnalysis" OpHelp(""),
29602	/* 139 */ "DropTable" OpHelp(""),
29603	/* 140 */ "DropIndex" OpHelp(""),
29604	/* 141 */ "DropTrigger" OpHelp(""),
29605	/* 142 */ "IntegrityCk" OpHelp(""),
29606	/* 143 */ "RowSetAdd" OpHelp("rowset(P1)=r[P2]"),
29607	/* 144 */ "Param" OpHelp(""),
29608	/* 145 */ "FkCounter" OpHelp("fkctr[P1]+=P2"),
29609	/* 146 */ "MemMax" OpHelp("r[P1]=max(r[P1],r[P2])"),
29610	/* 147 */ "OffsetLimit" OpHelp("if r[P1]>0 then r[P2]=r[P1]+max(0,r[P3]) else r[P2]=(-1)"),
29611	/* 148 */ "AggStep0" OpHelp("accum=r[P3] step(r[P2@P5])"),
29612	/* 149 */ "AggStep" OpHelp("accum=r[P3] step(r[P2@P5])"),
29613	/* 150 */ "AggFinal" OpHelp("accum=r[P1] N=P2"),
29614	/* 151 */ "Expire" OpHelp(""),
29615	/* 152 */ "TableLock" OpHelp("iDb=P1 root=P2 write=P3"),
29616	/* 153 */ "VBegin" OpHelp(""),
29617	/* 154 */ "VCreate" OpHelp(""),
29618	/* 155 */ "VDestroy" OpHelp(""),
29619	/* 156 */ "VOpen" OpHelp(""),
29620	/* 157 */ "VColumn" OpHelp("r[P3]=vcolumn(P2)"),
29621	/* 158 */ "VRename" OpHelp(""),
29622	/* 159 */ "Pagecount" OpHelp(""),
29623	/* 160 */ "MaxPgcnt" OpHelp(""),
29624	/* 161 */ "CursorHint" OpHelp(""),
29625	/* 162 */ "Noop" OpHelp(""),
29626	/* 163 */ "Explain" OpHelp(""),
29627	};
29628	return azName[i];
29629	}
29630	#endif
29631
	@@ -37952,46 +38003,84 @@
38003	* be freed immediately and any attempt to access any of that freed
38004	* data will almost certainly result in an immediate access violation.
38005	******************************************************************************
38006	*/
38007	#ifndef SQLITE_WIN32_HEAP_CREATE
38008	# define SQLITE_WIN32_HEAP_CREATE (TRUE)
38009	#endif
38010
38011	/*
38012	* This is the maximum possible initial size of the Win32-specific heap, in
38013	* bytes.
38014	*/
38015	#ifndef SQLITE_WIN32_HEAP_MAX_INIT_SIZE
38016	# define SQLITE_WIN32_HEAP_MAX_INIT_SIZE (4294967295U)
38017	#endif
38018
38019	/*
38020	* This is the extra space for the initial size of the Win32-specific heap,
38021	* in bytes. This value may be zero.
38022	*/
38023	#ifndef SQLITE_WIN32_HEAP_INIT_EXTRA
38024	# define SQLITE_WIN32_HEAP_INIT_EXTRA (4194304)
38025	#endif
38026
38027	/*
38028	* Calculate the maximum legal cache size, in pages, based on the maximum
38029	* possible initial heap size and the default page size, setting aside the
38030	* needed extra space.
38031	*/
38032	#ifndef SQLITE_WIN32_MAX_CACHE_SIZE
38033	# define SQLITE_WIN32_MAX_CACHE_SIZE (((SQLITE_WIN32_HEAP_MAX_INIT_SIZE) - \
38034	(SQLITE_WIN32_HEAP_INIT_EXTRA)) / \
38035	(SQLITE_DEFAULT_PAGE_SIZE))
38036	#endif
38037
38038	/*
38039	* This is cache size used in the calculation of the initial size of the
38040	* Win32-specific heap. It cannot be negative.
38041	*/
38042	#ifndef SQLITE_WIN32_CACHE_SIZE
38043	# if SQLITE_DEFAULT_CACHE_SIZE>=0
38044	# define SQLITE_WIN32_CACHE_SIZE (SQLITE_DEFAULT_CACHE_SIZE)
38045	# else
38046	# define SQLITE_WIN32_CACHE_SIZE (-(SQLITE_DEFAULT_CACHE_SIZE))
38047	# endif
38048	#endif
38049
38050	/*
38051	* Make sure that the calculated cache size, in pages, cannot cause the
38052	* initial size of the Win32-specific heap to exceed the maximum amount
38053	* of memory that can be specified in the call to HeapCreate.
38054	*/
38055	#if SQLITE_WIN32_CACHE_SIZE>SQLITE_WIN32_MAX_CACHE_SIZE
38056	# undef SQLITE_WIN32_CACHE_SIZE
38057	# define SQLITE_WIN32_CACHE_SIZE (2000)
38058	#endif
38059
38060	/*
38061	* The initial size of the Win32-specific heap. This value may be zero.
38062	*/
38063	#ifndef SQLITE_WIN32_HEAP_INIT_SIZE
38064	# define SQLITE_WIN32_HEAP_INIT_SIZE ((SQLITE_WIN32_CACHE_SIZE) * \
38065	(SQLITE_DEFAULT_PAGE_SIZE) + \
38066	(SQLITE_WIN32_HEAP_INIT_EXTRA))
38067	#endif
38068
38069	/*
38070	* The maximum size of the Win32-specific heap. This value may be zero.
38071	*/
38072	#ifndef SQLITE_WIN32_HEAP_MAX_SIZE
38073	# define SQLITE_WIN32_HEAP_MAX_SIZE (0)
38074	#endif
38075
38076	/*
38077	* The extra flags to use in calls to the Win32 heap APIs. This value may be
38078	* zero for the default behavior.
38079	*/
38080	#ifndef SQLITE_WIN32_HEAP_FLAGS
38081	# define SQLITE_WIN32_HEAP_FLAGS (0)
38082	#endif
38083
38084
38085	/*
38086	** The winMemData structure stores information required by the Win32-specific
	@@ -44084,11 +44173,11 @@
44173	** This routine is for use inside of assert() statements only. For
44174	** example:
44175	**
44176	** assert( sqlite3PcachePageSanity(pPg) );
44177	*/
44178	#ifdef SQLITE_DEBUG
44179	SQLITE_PRIVATE int sqlite3PcachePageSanity(PgHdr *pPg){
44180	PCache *pCache;
44181	assert( pPg!=0 );
44182	assert( pPg->pgno>0 \|\| pPg->pPager==0 ); /* Page number is 1 or more */
44183	pCache = pPg->pCache;
	@@ -60177,21 +60266,22 @@
60266	}
60267	#endif
60268
60269
60270	/*
60271	** Defragment the page given. This routine reorganizes cells within the
60272	** page so that there are no free-blocks on the free-block list.
60273	**
60274	** Parameter nMaxFrag is the maximum amount of fragmented space that may be
60275	** present in the page after this routine returns.
60276	**
60277	** EVIDENCE-OF: R-44582-60138 SQLite may from time to time reorganize a
60278	** b-tree page so that there are no freeblocks or fragment bytes, all
60279	** unused bytes are contained in the unallocated space region, and all
60280	** cells are packed tightly at the end of the page.
60281	*/
60282	static int defragmentPage(MemPage *pPage, int nMaxFrag){
60283	int i; /* Loop counter */
60284	int pc; /* Address of the i-th cell */
60285	int hdr; /* Offset to the page header */
60286	int size; /* Size of a cell */
60287	int usableSize; /* Number of usable bytes on a page */
	@@ -60202,11 +60292,10 @@
60292	unsigned char temp; / Temp area for cell content */
60293	unsigned char src; / Source of content */
60294	int iCellFirst; /* First allowable cell index */
60295	int iCellLast; /* Last possible cell index */
60296

60297	assert( sqlite3PagerIswriteable(pPage->pDbPage) );
60298	assert( pPage->pBt!=0 );
60299	assert( pPage->pBt->usableSize <= SQLITE_MAX_PAGE_SIZE );
60300	assert( pPage->nOverflow==0 );
60301	assert( sqlite3_mutex_held(pPage->pBt->mutex) );
	@@ -60214,13 +60303,60 @@
60303	src = data = pPage->aData;
60304	hdr = pPage->hdrOffset;
60305	cellOffset = pPage->cellOffset;
60306	nCell = pPage->nCell;
60307	assert( nCell==get2byte(&data[hdr+3]) );
60308	iCellFirst = cellOffset + 2*nCell;
60309	usableSize = pPage->pBt->usableSize;
60310
60311	/* This block handles pages with two or fewer free blocks and nMaxFrag
60312	** or fewer fragmented bytes. In this case it is faster to move the
60313	** two (or one) blocks of cells using memmove() and add the required
60314	** offsets to each pointer in the cell-pointer array than it is to
60315	** reconstruct the entire page. */
60316	if( (int)data[hdr+7]<=nMaxFrag ){
60317	int iFree = get2byte(&data[hdr+1]);
60318	if( iFree ){
60319	int iFree2 = get2byte(&data[iFree]);
60320
60321	/* pageFindSlot() has already verified that free blocks are sorted
60322	** in order of offset within the page, and that no block extends
60323	** past the end of the page. Provided the two free slots do not
60324	** overlap, this guarantees that the memmove() calls below will not
60325	** overwrite the usableSize byte buffer, even if the database page
60326	** is corrupt. */
60327	assert( iFree2==0 \|\| iFree2>iFree );
60328	assert( iFree+get2byte(&data[iFree+2]) <= usableSize );
60329	assert( iFree2==0 \|\| iFree2+get2byte(&data[iFree2+2]) <= usableSize );
60330
60331	if( 0==iFree2 \|\| (data[iFree2]==0 && data[iFree2+1]==0) ){
60332	u8 pEnd = &data[cellOffset + nCell2];
60333	u8 *pAddr;
60334	int sz2 = 0;
60335	int sz = get2byte(&data[iFree+2]);
60336	int top = get2byte(&data[hdr+5]);
60337	if( iFree2 ){
60338	if( iFree+sz>iFree2 ) return SQLITE_CORRUPT_BKPT;
60339	sz2 = get2byte(&data[iFree2+2]);
60340	assert( iFree+sz+sz2+iFree2-(iFree+sz) <= usableSize );
60341	memmove(&data[iFree+sz+sz2], &data[iFree+sz], iFree2-(iFree+sz));
60342	sz += sz2;
60343	}
60344	cbrk = top+sz;
60345	assert( cbrk+(iFree-top) <= usableSize );
60346	memmove(&data[cbrk], &data[top], iFree-top);
60347	for(pAddr=&data[cellOffset]; pAddr<pEnd; pAddr+=2){
60348	pc = get2byte(pAddr);
60349	if( pc<iFree ){ put2byte(pAddr, pc+sz); }
60350	else if( pc<iFree2 ){ put2byte(pAddr, pc+sz2); }
60351	}
60352	goto defragment_out;
60353	}
60354	}
60355	}
60356
60357	cbrk = usableSize;

60358	iCellLast = usableSize - 4;
60359	for(i=0; i<nCell; i++){
60360	u8 pAddr; / The i-th cell pointer */
60361	pAddr = &data[cellOffset + i*2];
60362	pc = get2byte(pAddr);
	@@ -60250,20 +60386,22 @@
60386	memcpy(&temp[x], &data[x], (cbrk+size) - x);
60387	src = temp;
60388	}
60389	memcpy(&data[cbrk], &src[pc], size);
60390	}
60391	data[hdr+7] = 0;
60392
60393	defragment_out:
60394	if( data[hdr+7]+cbrk-iCellFirst!=pPage->nFree ){
60395	return SQLITE_CORRUPT_BKPT;
60396	}
60397	assert( cbrk>=iCellFirst );
60398	put2byte(&data[hdr+5], cbrk);
60399	data[hdr+1] = 0;
60400	data[hdr+2] = 0;

60401	memset(&data[iCellFirst], 0, cbrk-iCellFirst);
60402	assert( sqlite3PagerIswriteable(pPage->pDbPage) );



60403	return SQLITE_OK;
60404	}
60405
60406	/*
60407	** Search the free-list on page pPg for space to store a cell nByte bytes in
	@@ -60397,14 +60535,14 @@
60535	** to see if defragmentation is necessary.
60536	*/
60537	testcase( gap+2+nByte==top );
60538	if( gap+2+nByte>top ){
60539	assert( pPage->nCell>0 \|\| CORRUPT_DB );
60540	rc = defragmentPage(pPage, MIN(4, pPage->nFree - (2+nByte)));
60541	if( rc ) return rc;
60542	top = get2byteNotZero(&data[hdr+5]);
60543	assert( gap+2+nByte<=top );
60544	}
60545
60546
60547	/* Allocate memory from the gap in between the cell pointer array
60548	** and the cell content area. The btreeInitPage() call has already
	@@ -63636,11 +63774,11 @@
63774	pCur->aiIdx[pCur->iPage] = 0;
63775	return getAndInitPage(pBt, newPgno, &pCur->apPage[pCur->iPage],
63776	pCur, pCur->curPagerFlags);
63777	}
63778
63779	#ifdef SQLITE_DEBUG
63780	/*
63781	** Page pParent is an internal (non-leaf) tree page. This function
63782	** asserts that page number iChild is the left-child if the iIdx'th
63783	** cell in page pParent. Or, if iIdx is equal to the total number of
63784	** cells in pParent, that page number iChild is the right-child of
	@@ -64178,10 +64316,34 @@
64316	** have been deleted? This API will need to change to return an error code
64317	** as well as the boolean result value.
64318	*/
64319	return (CURSOR_VALID!=pCur->eState);
64320	}
64321
64322	/*
64323	** Return an estimate for the number of rows in the table that pCur is
64324	** pointing to. Return a negative number if no estimate is currently
64325	** available.
64326	*/
64327	SQLITE_PRIVATE i64 sqlite3BtreeRowCountEst(BtCursor *pCur){
64328	i64 n;
64329	u8 i;
64330
64331	assert( cursorOwnsBtShared(pCur) );
64332	assert( sqlite3_mutex_held(pCur->pBtree->db->mutex) );
64333
64334	/* Currently this interface is only called by the OP_IfSmaller
64335	** opcode, and it that case the cursor will always be valid and
64336	** will always point to a leaf node. */
64337	if( NEVER(pCur->eState!=CURSOR_VALID) ) return -1;
64338	if( NEVER(pCur->apPage[pCur->iPage]->leaf==0) ) return -1;
64339
64340	for(n=1, i=0; i<=pCur->iPage; i++){
64341	n *= pCur->apPage[i]->nCell;
64342	}
64343	return n;
64344	}
64345
64346	/*
64347	** Advance the cursor to the next entry in the database. If
64348	** successful then set *pRes=0. If the cursor
64349	** was already pointing to the last entry in the database before
	@@ -65023,11 +65185,11 @@
65185	** pPrior Where to write the pgno of the first overflow page
65186	**
65187	** Use a call to btreeParseCellPtr() to verify that the values above
65188	** were computed correctly.
65189	*/
65190	#ifdef SQLITE_DEBUG
65191	{
65192	CellInfo info;
65193	pPage->xParseCell(pPage, pCell, &info);
65194	assert( nHeader==(int)(info.pPayload - pCell) );
65195	assert( info.nKey==pX->nKey );
	@@ -66549,11 +66711,11 @@
66711	** copied into the parent, because if the parent is page 1 then it will
66712	** by smaller than the child due to the database header, and so all the
66713	** free space needs to be up front.
66714	*/
66715	assert( nNew==1 \|\| CORRUPT_DB );
66716	rc = defragmentPage(apNew[0], -1);
66717	testcase( rc!=SQLITE_OK );
66718	assert( apNew[0]->nFree ==
66719	(get2byte(&apNew[0]->aData[5])-apNew[0]->cellOffset-apNew[0]->nCell*2)
66720	\|\| rc!=SQLITE_OK
66721	);
	@@ -71201,10 +71363,11 @@
71363	** Remember the SQL string for a prepared statement.
71364	*/
71365	SQLITE_PRIVATE void sqlite3VdbeSetSql(Vdbe p, const char z, int n, int isPrepareV2){
71366	assert( isPrepareV2==1 \|\| isPrepareV2==0 );
71367	if( p==0 ) return;
71368	if( !isPrepareV2 ) p->expmask = 0;
71369	#if defined(SQLITE_OMIT_TRACE) && !defined(SQLITE_ENABLE_SQLLOG)
71370	if( !isPrepareV2 ) return;
71371	#endif
71372	assert( p->zSql==0 );
71373	p->zSql = sqlite3DbStrNDup(p->db, z, n);
	@@ -71229,10 +71392,11 @@
71392	pB->pPrev = pTmp;
71393	zTmp = pA->zSql;
71394	pA->zSql = pB->zSql;
71395	pB->zSql = zTmp;
71396	pB->isPrepareV2 = pA->isPrepareV2;
71397	pB->expmask = pA->expmask;
71398	}
71399
71400	/*
71401	** Resize the Vdbe.aOp array so that it is at least nOp elements larger
71402	** than its current size. nOp is guaranteed to be less than or equal
	@@ -73753,17 +73917,17 @@
73917	** Then the internal cache might have been left in an inconsistent
73918	** state. We need to rollback the statement transaction, if there is
73919	** one, or the complete transaction if there is no statement transaction.
73920	*/
73921
73922	if( p->magic!=VDBE_MAGIC_RUN ){
73923	return SQLITE_OK;
73924	}
73925	if( db->mallocFailed ){
73926	p->rc = SQLITE_NOMEM_BKPT;
73927	}
73928	closeAllCursors(p);



73929	checkActiveVdbeCnt(db);
73930
73931	/* No commit or rollback needed if the program never started or if the
73932	** SQL statement does not read or write a database file. */
73933	if( p->pc>=0 && p->bIsReader ){
	@@ -74706,11 +74870,11 @@
74870	}
74871	assert( u<=pKeyInfo->nField + 1 );
74872	p->nField = u;
74873	}
74874
74875	#ifdef SQLITE_DEBUG
74876	/*
74877	** This function compares two index or table record keys in the same way
74878	** as the sqlite3VdbeRecordCompare() routine. Unlike VdbeRecordCompare(),
74879	** this function deserializes and compares values using the
74880	** sqlite3VdbeSerialGet() and sqlite3MemCompare() functions. It is used
	@@ -74811,11 +74975,11 @@
74975	if( pKeyInfo->db->mallocFailed ) return 1;
74976	return 0;
74977	}
74978	#endif
74979
74980	#ifdef SQLITE_DEBUG
74981	/*
74982	** Count the number of fields (a.k.a. columns) in the record given by
74983	** pKey,nKey. The verify that this count is less than or equal to the
74984	** limit given by pKeyInfo->nField + pKeyInfo->nXField.
74985	**
	@@ -75694,12 +75858,12 @@
75858	** to sqlite3_reoptimize() that re-preparing the statement may result
75859	** in a better query plan.
75860	*/
75861	SQLITE_PRIVATE void sqlite3VdbeSetVarmask(Vdbe *v, int iVar){
75862	assert( iVar>0 );
75863	if( iVar>=32 ){
75864	v->expmask \|= 0x80000000;
75865	}else{
75866	v->expmask \|= ((u32)1 << (iVar-1));
75867	}
75868	}
75869
	@@ -75965,11 +76129,12 @@
76129	sqlite3_mutex_enter(mutex);
76130	for(i=0; i<p->nVar; i++){
76131	sqlite3VdbeMemRelease(&p->aVar[i]);
76132	p->aVar[i].flags = MEM_Null;
76133	}
76134	assert( p->isPrepareV2 \|\| p->expmask==0 );
76135	if( p->expmask ){
76136	p->expired = 1;
76137	}
76138	sqlite3_mutex_leave(mutex);
76139	return rc;
76140	}
	@@ -77069,13 +77234,12 @@
77234	** parameter in the WHERE clause might influence the choice of query plan
77235	** for a statement, then the statement will be automatically recompiled,
77236	** as if there had been a schema change, on the first sqlite3_step() call
77237	** following any change to the bindings of that parameter.
77238	*/
77239	assert( p->isPrepareV2 \|\| p->expmask==0 );
77240	if( p->expmask!=0 && (p->expmask & (i>=31 ? 0x80000000 : (u32)1<<i))!=0 ){

77241	p->expired = 1;
77242	}
77243	return SQLITE_OK;
77244	}
77245
	@@ -77334,14 +77498,16 @@
77498	Vdbe pFrom = (Vdbe)pFromStmt;
77499	Vdbe pTo = (Vdbe)pToStmt;
77500	if( pFrom->nVar!=pTo->nVar ){
77501	return SQLITE_ERROR;
77502	}
77503	assert( pTo->isPrepareV2 \|\| pTo->expmask==0 );
77504	if( pTo->expmask ){
77505	pTo->expired = 1;
77506	}
77507	assert( pFrom->isPrepareV2 \|\| pFrom->expmask==0 );
77508	if( pFrom->expmask ){
77509	pFrom->expired = 1;
77510	}
77511	return sqlite3TransferBindings(pFromStmt, pToStmt);
77512	}
77513	#endif
	@@ -79658,39 +79824,39 @@
79824	if( pCtx->pOut != pOut ){
79825	pCtx->pOut = pOut;
79826	for(i=pCtx->argc-1; i>=0; i--) pCtx->argv[i] = &aMem[pOp->p2+i];
79827	}
79828
79829	memAboutToChange(p, pOut);
79830	#ifdef SQLITE_DEBUG
79831	for(i=0; i<pCtx->argc; i++){
79832	assert( memIsValid(pCtx->argv[i]) );
79833	REGISTER_TRACE(pOp->p2+i, pCtx->argv[i]);
79834	}
79835	#endif
79836	MemSetTypeFlag(pOut, MEM_Null);
79837	pCtx->fErrorOrAux = 0;
79838	(pCtx->pFunc->xSFunc)(pCtx, pCtx->argc, pCtx->argv);/ IMP: R-24505-23230 */
79839
79840	/* If the function returned an error, throw an exception */
79841	if( pCtx->fErrorOrAux ){
79842	if( pCtx->isError ){
79843	sqlite3VdbeError(p, "%s", sqlite3_value_text(pOut));
79844	rc = pCtx->isError;
79845	}
79846	sqlite3VdbeDeleteAuxData(db, &p->pAuxData, pCtx->iOp, pOp->p1);
79847	if( rc ) goto abort_due_to_error;
79848	}
79849
79850	/* Copy the result of the function into register P3 */
79851	if( pOut->flags & (MEM_Str\|MEM_Blob) ){
79852	sqlite3VdbeChangeEncoding(pOut, encoding);
79853	if( sqlite3VdbeMemTooBig(pOut) ) goto too_big;
79854	}
79855
79856	REGISTER_TRACE(pOp->p3, pOut);
79857	UPDATE_MAX_BLOBSIZE(pOut);
79858	break;
79859	}
79860
79861	/* Opcode: BitAnd P1 P2 P3 * *
79862	** Synopsis: r[P3]=r[P1]&r[P2]
	@@ -80187,11 +80353,11 @@
80353	pKeyInfo = pOp->p4.pKeyInfo;
80354	assert( n>0 );
80355	assert( pKeyInfo!=0 );
80356	p1 = pOp->p1;
80357	p2 = pOp->p2;
80358	#ifdef SQLITE_DEBUG
80359	if( aPermute ){
80360	int k, mx = 0;
80361	for(k=0; k<n; k++) if( aPermute[k]>mx ) mx = aPermute[k];
80362	assert( p1>0 && p1+mx<=(p->nMem+1 - p->nCursor)+1 );
80363	assert( p2>0 && p2+mx<=(p->nMem+1 - p->nCursor)+1 );
	@@ -82836,10 +83002,37 @@
83002	assert( pOp->p2==0 );
83003	}
83004	break;
83005	}
83006
83007	/* Opcode: IfSmaller P1 P2 P3 * *
83008	**
83009	** Estimate the number of rows in the table P1. Jump to P2 if that
83010	estimate is less than approximately 2(0.1*P3).
83011	*/
83012	case OP_IfSmaller: { /* jump */
83013	VdbeCursor *pC;
83014	BtCursor *pCrsr;
83015	int res;
83016	i64 sz;
83017
83018	assert( pOp->p1>=0 && pOp->p1<p->nCursor );
83019	pC = p->apCsr[pOp->p1];
83020	assert( pC!=0 );
83021	pCrsr = pC->uc.pCursor;
83022	assert( pCrsr );
83023	rc = sqlite3BtreeFirst(pCrsr, &res);
83024	if( rc ) goto abort_due_to_error;
83025	if( res==0 ){
83026	sz = sqlite3BtreeRowCountEst(pCrsr);
83027	if( ALWAYS(sz>=0) && sqlite3LogEst((u64)sz)<pOp->p3 ) res = 1;
83028	}
83029	VdbeBranchTaken(res!=0,2);
83030	if( res ) goto jump_to_p2;
83031	break;
83032	}
83033
83034
83035	/* Opcode: SorterSort P1 P2 * * *
83036	**
83037	** After all records have been inserted into the Sorter object
83038	** identified by P1, invoke this opcode to actually do the sorting.
	@@ -83479,10 +83672,22 @@
83672	rc = sqlite3BtreeCreateTable(pDb->pBt, &pgno, flags);
83673	if( rc ) goto abort_due_to_error;
83674	pOut->u.i = pgno;
83675	break;
83676	}
83677
83678	/* Opcode: SqlExec * * * P4 *
83679	**
83680	** Run the SQL statement or statements specified in the P4 string.
83681	*/
83682	case OP_SqlExec: {
83683	db->nSqlExec++;
83684	rc = sqlite3_exec(db, pOp->p4.z, 0, 0, 0);
83685	db->nSqlExec--;
83686	if( rc ) goto abort_due_to_error;
83687	break;
83688	}
83689
83690	/* Opcode: ParseSchema P1 * * P4 *
83691	**
83692	** Read and parse all entries from the SQLITE_MASTER table of database P1
83693	** that match the WHERE clause P4.
	@@ -83600,11 +83805,11 @@
83805	**
83806	** Do an analysis of the currently open database. Store in
83807	** register P1 the text of an error message describing any problems.
83808	** If no problems are found, store a NULL in register P1.
83809	**
83810	** The register P3 contains one less than the maximum number of allowed errors.
83811	** At most reg(P3) errors will be reported.
83812	** In other words, the analysis stops as soon as reg(P1) errors are
83813	** seen. Reg(P1) is updated with the number of errors remaining.
83814	**
83815	** The root page numbers of all tables in the database are integers
	@@ -83633,18 +83838,18 @@
83838	assert( (pnErr->flags & (MEM_Str\|MEM_Blob))==0 );
83839	pIn1 = &aMem[pOp->p1];
83840	assert( pOp->p5<db->nDb );
83841	assert( DbMaskTest(p->btreeMask, pOp->p5) );
83842	z = sqlite3BtreeIntegrityCheck(db->aDb[pOp->p5].pBt, aRoot, nRoot,
83843	(int)pnErr->u.i+1, &nErr);

83844	sqlite3VdbeMemSetNull(pIn1);
83845	if( nErr==0 ){
83846	assert( z==0 );
83847	}else if( z==0 ){
83848	goto no_mem;
83849	}else{
83850	pnErr->u.i -= nErr-1;
83851	sqlite3VdbeMemSetStr(pIn1, z, -1, SQLITE_UTF8, sqlite3_free);
83852	}
83853	UPDATE_MAX_BLOBSIZE(pIn1);
83854	sqlite3VdbeChangeEncoding(pIn1, encoding);
83855	break;
	@@ -88787,15 +88992,15 @@
88992	** The return value from the callback should be one of the WRC_*
88993	** constants to specify how to proceed with the walk.
88994	**
88995	** WRC_Continue Continue descending down the tree.
88996	**
88997	** WRC_Prune Do not descend into child nodes, but allow
88998	** the walk to continue with sibling nodes.
88999	**
89000	** WRC_Abort Do no more callbacks. Unwind the stack and
89001	** return from the top-level walk call.
89002	**
89003	** The return value from this routine is WRC_Abort to abandon the tree walk
89004	** and WRC_Continue to continue.
89005	*/
89006	static SQLITE_NOINLINE int walkExpr(Walker pWalker, Expr pExpr){
	@@ -89153,11 +89358,12 @@
89358	}
89359	}
89360	}
89361
89362	/* Start at the inner-most context and move outward until a match is found */
89363	assert( pNC && cnt==0 );
89364	do{
89365	ExprList *pEList;
89366	SrcList *pSrcList = pNC->pSrcList;
89367
89368	if( pSrcList ){
89369	for(i=0, pItem=pSrcList->a; i<pSrcList->nSrc; i++, pItem++){
	@@ -89338,15 +89544,15 @@
89544	}
89545
89546	/* Advance to the next name context. The loop will exit when either
89547	** we have a match (cnt>0) or when we run out of name contexts.
89548	*/
89549	if( cnt ) break;
89550	pNC = pNC->pNext;
89551	nSubquery++;
89552	}while( pNC );
89553
89554
89555	/*
89556	** If X and Y are NULL (in other words if only the column name Z is
89557	** supplied) and the value of Z is enclosed in double-quotes, then
89558	** Z is a string literal if it doesn't match any column names. In that
	@@ -89532,37 +89738,42 @@
89738	break;
89739	}
89740	#endif /* defined(SQLITE_ENABLE_UPDATE_DELETE_LIMIT)
89741	&& !defined(SQLITE_OMIT_SUBQUERY) */
89742
89743	/* A column name: ID
89744	** Or table name and column name: ID.ID





89745	** Or a database, table and column: ID.ID.ID
89746	**
89747	** The TK_ID and TK_OUT cases are combined so that there will only
89748	** be one call to lookupName(). Then the compiler will in-line
89749	** lookupName() for a size reduction and performance increase.
89750	*/
89751	case TK_ID:
89752	case TK_DOT: {
89753	const char *zColumn;
89754	const char *zTable;
89755	const char *zDb;
89756	Expr *pRight;
89757
89758	if( pExpr->op==TK_ID ){



89759	zDb = 0;
89760	zTable = 0;
89761	zColumn = pExpr->u.zToken;
89762	}else{
89763	notValid(pParse, pNC, "the \".\" operator", NC_IdxExpr);
89764	pRight = pExpr->pRight;
89765	if( pRight->op==TK_ID ){
89766	zDb = 0;
89767	zTable = pExpr->pLeft->u.zToken;
89768	zColumn = pRight->u.zToken;
89769	}else{
89770	assert( pRight->op==TK_DOT );
89771	zDb = pExpr->pLeft->u.zToken;
89772	zTable = pRight->pLeft->u.zToken;
89773	zColumn = pRight->pRight->u.zToken;
89774	}
89775	}
89776	return lookupName(pParse, zDb, zTable, zColumn, pNC, pExpr);
89777	}
89778
89779	/* Resolve function names
	@@ -92833,11 +93044,11 @@
93044	int nVal = sqlite3ExprVectorSize(pLeft);
93045	Select *pSelect = (pExpr->flags & EP_xIsSelect) ? pExpr->x.pSelect : 0;
93046	char *zRet;
93047
93048	assert( pExpr->op==TK_IN );
93049	zRet = sqlite3DbMallocRaw(pParse->db, nVal+1);
93050	if( zRet ){
93051	int i;
93052	for(i=0; i<nVal; i++){
93053	Expr *pA = sqlite3VectorFieldSubexpr(pLeft, i);
93054	char a = sqlite3ExprAffinity(pA);
	@@ -93738,11 +93949,11 @@
93949	** Clear all column cache entries.
93950	*/
93951	SQLITE_PRIVATE void sqlite3ExprCacheClear(Parse *pParse){
93952	int i;
93953
93954	#ifdef SQLITE_DEBUG
93955	if( pParse->db->flags & SQLITE_VdbeAddopTrace ){
93956	printf("CLEAR\n");
93957	}
93958	#endif
93959	for(i=0; i<pParse->nColCache; i++){
	@@ -95145,10 +95356,21 @@
95356	if( pA->a[i].sortOrder!=pB->a[i].sortOrder ) return 1;
95357	if( sqlite3ExprCompare(pExprA, pExprB, iTab) ) return 1;
95358	}
95359	return 0;
95360	}
95361
95362	/*
95363	** Like sqlite3ExprCompare() except COLLATE operators at the top-level
95364	** are ignored.
95365	*/
95366	SQLITE_PRIVATE int sqlite3ExprCompareSkip(Expr pA, Expr pB, int iTab){
95367	return sqlite3ExprCompare(
95368	sqlite3ExprSkipCollate(pA),
95369	sqlite3ExprSkipCollate(pB),
95370	iTab);
95371	}
95372
95373	/*
95374	** Return true if we can prove the pE2 will always be true if pE1 is
95375	** true. Return false if we cannot complete the proof or if pE2 might
95376	** be false. Examples:
	@@ -96698,10 +96920,11 @@
96920	Stat4Sample current; /* Current row as a Stat4Sample */
96921	u32 iPrn; /* Pseudo-random number used for sampling */
96922	Stat4Sample aBest; / Array of nCol best samples */
96923	int iMin; /* Index in a[] of entry with minimum score */
96924	int nSample; /* Current number of samples */
96925	int nMaxEqZero; /* Max leading 0 in anEq[] for any a[] entry */
96926	int iGet; /* Index of current sample accessed by stat_get() */
96927	Stat4Sample a; / Array of mxSample Stat4Sample objects */
96928	sqlite3 db; / Database connection, for malloc() */
96929	};
96930
	@@ -96962,10 +97185,17 @@
97185	int i;
97186
97187	assert( IsStat4 \|\| nEqZero==0 );
97188
97189	#ifdef SQLITE_ENABLE_STAT4
97190	/* Stat4Accum.nMaxEqZero is set to the maximum number of leading 0
97191	** values in the anEq[] array of any sample in Stat4Accum.a[]. In
97192	** other words, if nMaxEqZero is n, then it is guaranteed that there
97193	** are no samples with Stat4Sample.anEq[m]==0 for (m>=n). */
97194	if( nEqZero>p->nMaxEqZero ){
97195	p->nMaxEqZero = nEqZero;
97196	}
97197	if( pNew->isPSample==0 ){
97198	Stat4Sample *pUpgrade = 0;
97199	assert( pNew->anEq[pNew->iCol]>0 );
97200
97201	/* This sample is being added because the prefix that ends in column
	@@ -97059,16 +97289,26 @@
97289	if( p->nSample<p->mxSample \|\| sampleIsBetter(p, pBest, &p->a[p->iMin]) ){
97290	sampleInsert(p, pBest, i);
97291	}
97292	}
97293
97294	/* Check that no sample contains an anEq[] entry with an index of
97295	** p->nMaxEqZero or greater set to zero. */
97296	for(i=p->nSample-1; i>=0; i--){
97297	int j;
97298	for(j=p->nMaxEqZero; j<p->nCol; j++) assert( p->a[i].anEq[j]>0 );
97299	}
97300
97301	/* Update the anEq[] fields of any samples already collected. */
97302	if( iChng<p->nMaxEqZero ){
97303	for(i=p->nSample-1; i>=0; i--){
97304	int j;
97305	for(j=iChng; j<p->nCol; j++){
97306	if( p->a[i].anEq[j]==0 ) p->a[i].anEq[j] = p->current.anEq[j];
97307	}
97308	}
97309	p->nMaxEqZero = iChng;
97310	}
97311	#endif
97312
97313	#if defined(SQLITE_ENABLE_STAT3) && !defined(SQLITE_ENABLE_STAT4)
97314	if( iChng==0 ){
	@@ -97778,44 +98018,32 @@
98018	/* Form 1: Analyze everything */
98019	for(i=0; i<db->nDb; i++){
98020	if( i==1 ) continue; /* Do not analyze the TEMP database */
98021	analyzeDatabase(pParse, i);
98022	}
98023	}else if( pName2->n==0 && (iDb = sqlite3FindDb(db, pName1))>=0 ){
98024	/* Analyze the schema named as the argument */
98025	analyzeDatabase(pParse, iDb);
98026	}else{
98027	/* Form 3: Analyze the table or index named as an argument */













98028	iDb = sqlite3TwoPartName(pParse, pName1, pName2, &pTableName);
98029	if( iDb>=0 ){
98030	zDb = pName2->n ? db->aDb[iDb].zDbSName : 0;
98031	z = sqlite3NameFromToken(db, pTableName);
98032	if( z ){
98033	if( (pIdx = sqlite3FindIndex(db, z, zDb))!=0 ){
98034	analyzeTable(pParse, pIdx->pTable, pIdx);
98035	}else if( (pTab = sqlite3LocateTable(pParse, 0, z, zDb))!=0 ){
98036	analyzeTable(pParse, pTab, 0);
98037	}
98038	sqlite3DbFree(db, z);
98039	}
98040	}
98041	}
98042	if( db->nSqlExec==0 && (v = sqlite3GetVdbe(pParse))!=0 ){
98043	sqlite3VdbeAddOp0(v, OP_Expire);
98044	}
98045	}
98046
98047	/*
98048	** Used to pass information from the analyzer reader through to the
98049	** callback routine.
	@@ -97940,19 +98168,24 @@
98168	}
98169	aiRowEst = pIndex->aiRowEst;
98170	#endif
98171	pIndex->bUnordered = 0;
98172	decodeIntArray((char*)z, nCol, aiRowEst, pIndex->aiRowLogEst, pIndex);
98173	pIndex->hasStat1 = 1;
98174	if( pIndex->pPartIdxWhere==0 ){
98175	pTable->nRowLogEst = pIndex->aiRowLogEst[0];
98176	pTable->tabFlags \|= TF_HasStat1;
98177	}
98178	}else{
98179	Index fakeIdx;
98180	fakeIdx.szIdxRow = pTable->szTabRow;
98181	#ifdef SQLITE_ENABLE_COSTMULT
98182	fakeIdx.pTable = pTable;
98183	#endif
98184	decodeIntArray((char*)z, 1, 0, &pTable->nRowLogEst, &fakeIdx);
98185	pTable->szTabRow = fakeIdx.szIdxRow;
98186	pTable->tabFlags \|= TF_HasStat1;
98187	}
98188
98189	return 0;
98190	}
98191
	@@ -98243,19 +98476,24 @@
98476	SQLITE_PRIVATE int sqlite3AnalysisLoad(sqlite3 *db, int iDb){
98477	analysisInfo sInfo;
98478	HashElem *i;
98479	char *zSql;
98480	int rc = SQLITE_OK;
98481	Schema *pSchema = db->aDb[iDb].pSchema;
98482
98483	assert( iDb>=0 && iDb<db->nDb );
98484	assert( db->aDb[iDb].pBt!=0 );
98485
98486	/* Clear any prior statistics */
98487	assert( sqlite3SchemaMutexHeld(db, iDb, 0) );
98488	for(i=sqliteHashFirst(&pSchema->tblHash); i; i=sqliteHashNext(i)){
98489	Table *pTab = sqliteHashData(i);
98490	pTab->tabFlags &= ~TF_HasStat1;
98491	}
98492	for(i=sqliteHashFirst(&pSchema->idxHash); i; i=sqliteHashNext(i)){
98493	Index *pIdx = sqliteHashData(i);
98494	pIdx->hasStat1 = 0;
98495	#ifdef SQLITE_ENABLE_STAT3_OR_STAT4
98496	sqlite3DeleteIndexSamples(db, pIdx);
98497	pIdx->aSample = 0;
98498	#endif
98499	}
	@@ -98274,23 +98512,23 @@
98512	}
98513	}
98514
98515	/* Set appropriate defaults on all indexes not in the sqlite_stat1 table */
98516	assert( sqlite3SchemaMutexHeld(db, iDb, 0) );
98517	for(i=sqliteHashFirst(&pSchema->idxHash); i; i=sqliteHashNext(i)){
98518	Index *pIdx = sqliteHashData(i);
98519	if( !pIdx->hasStat1 ) sqlite3DefaultRowEst(pIdx);
98520	}
98521
98522	/* Load the statistics from the sqlite_stat4 table. */
98523	#ifdef SQLITE_ENABLE_STAT3_OR_STAT4
98524	if( rc==SQLITE_OK && OptimizationEnabled(db, SQLITE_Stat34) ){
98525	db->lookaside.bDisable++;
98526	rc = loadStat4(db, sInfo.zDatabase);
98527	db->lookaside.bDisable--;
98528	}
98529	for(i=sqliteHashFirst(&pSchema->idxHash); i; i=sqliteHashNext(i)){
98530	Index *pIdx = sqliteHashData(i);
98531	sqlite3_free(pIdx->aiRowEst);
98532	pIdx->aiRowEst = 0;
98533	}
98534	#endif
	@@ -100271,10 +100509,11 @@
100509	SQLITE_PRIVATE void sqlite3AddNotNull(Parse *pParse, int onError){
100510	Table *p;
100511	p = pParse->pNewTable;
100512	if( p==0 \|\| NEVER(p->nCol<1) ) return;
100513	p->aCol[p->nCol-1].notNull = (u8)onError;
100514	p->tabFlags \|= TF_HasNotNull;
100515	}
100516
100517	/*
100518	** Scan the column type name zType (length nType) and return the
100519	** associated affinity type.
	@@ -102609,10 +102848,13 @@
102848	/* 10, 9, 8, 7, 6 */
102849	LogEst aVal[] = { 33, 32, 30, 28, 26 };
102850	LogEst *a = pIdx->aiRowLogEst;
102851	int nCopy = MIN(ArraySize(aVal), pIdx->nKeyCol);
102852	int i;
102853
102854	/* Indexes with default row estimates should not have stat1 data */
102855	assert( !pIdx->hasStat1 );
102856
102857	/* Set the first entry (number of rows in the index) to the estimated
102858	** number of rows in the table, or half the number of rows in the table
102859	** for a partial index. But do not let the estimate drop below 10. */
102860	a[0] = pIdx->pTable->nRowLogEst;
	@@ -109798,10 +110040,13 @@
110040	VdbeComment((v, "%s", iField<0 ? "rowid" : pTab->aCol[iField].zName));
110041	}
110042	}
110043	sqlite3VdbeAddOp3(v, OP_MakeRecord, regIdx, pIdx->nColumn, aRegIdx[ix]);
110044	VdbeComment((v, "for %s", pIdx->zName));
110045	#ifdef SQLITE_ENABLE_NULL_TRIM
110046	if( pIdx->idxType==2 ) sqlite3SetMakeRecordP5(v, pIdx->pTable);
110047	#endif
110048
110049	/* In an UPDATE operation, if this index is the PRIMARY KEY index
110050	** of a WITHOUT ROWID table and there has been no change the
110051	** primary key, then no collision is possible. The collision detection
110052	** logic below can all be skipped. */
	@@ -109953,12 +110198,15 @@
110198
110199	/* Records with omitted columns are only allowed for schema format
110200	** version 2 and later (SQLite version 3.1.4, 2005-02-20). */
110201	if( pTab->pSchema->file_format<2 ) return;
110202
110203	for(i=pTab->nCol-1; i>0; i--){
110204	if( pTab->aCol[i].pDflt!=0 ) break;
110205	if( pTab->aCol[i].colFlags & COLFLAG_PRIMKEY ) break;
110206	}
110207	sqlite3VdbeChangeP5(v, i+1);
110208	}
110209	#endif
110210
110211	/*
110212	** This routine generates code to finish the INSERT or UPDATE operation
	@@ -110244,11 +110492,11 @@
110492	}
110493	if( sqlite3TriggerList(pParse, pDest) ){
110494	return 0; /* tab1 must not have triggers */
110495	}
110496	#ifndef SQLITE_OMIT_VIRTUALTABLE
110497	if( IsVirtual(pDest) ){
110498	return 0; /* tab1 must not be a virtual table */
110499	}
110500	#endif
110501	if( onError==OE_Default ){
110502	if( pDest->iPKey>=0 ) onError = pDest->keyConf;
	@@ -110306,11 +110554,11 @@
110554	}
110555	if( HasRowid(pDest)!=HasRowid(pSrc) ){
110556	return 0; /* source and destination must both be WITHOUT ROWID or not */
110557	}
110558	#ifndef SQLITE_OMIT_VIRTUALTABLE
110559	if( IsVirtual(pSrc) ){
110560	return 0; /* tab2 must not be a virtual table */
110561	}
110562	#endif
110563	if( pSrc->pSelect ){
110564	return 0; /* tab2 may not be a view */
	@@ -110603,11 +110851,11 @@
110851	/* Invoke the callback function if required */
110852	if( xCallback && (SQLITE_ROW==rc \|\|
110853	(SQLITE_DONE==rc && !callbackIsInit
110854	&& db->flags&SQLITE_NullCallback)) ){
110855	if( !callbackIsInit ){
110856	azCols = sqlite3DbMallocRaw(db, (2nCol+1)sizeof(const char*));
110857	if( azCols==0 ){
110858	goto exec_out;
110859	}
110860	for(i=0; i<nCol; i++){
110861	azCols[i] = (char *)sqlite3_column_name(pStmt, i);
	@@ -110624,10 +110872,11 @@
110872	if( !azVals[i] && sqlite3_column_type(pStmt, i)!=SQLITE_NULL ){
110873	sqlite3OomFault(db);
110874	goto exec_out;
110875	}
110876	}
110877	azVals[i] = 0;
110878	}
110879	if( xCallback(pArg, nCol, azVals, azCols) ){
110880	/* EVIDENCE-OF: R-38229-40159 If the callback function to
110881	** sqlite3_exec() returns non-zero, then sqlite3_exec() will
110882	** return SQLITE_ABORT. */
	@@ -112094,15 +112343,15 @@
112343	#define PragTyp_JOURNAL_SIZE_LIMIT 20
112344	#define PragTyp_LOCK_PROXY_FILE 21
112345	#define PragTyp_LOCKING_MODE 22
112346	#define PragTyp_PAGE_COUNT 23
112347	#define PragTyp_MMAP_SIZE 24
112348	#define PragTyp_OPTIMIZE 25
112349	#define PragTyp_PAGE_SIZE 26
112350	#define PragTyp_SECURE_DELETE 27
112351	#define PragTyp_SHRINK_MEMORY 28
112352	#define PragTyp_SOFT_HEAP_LIMIT 29
112353	#define PragTyp_SYNCHRONOUS 30
112354	#define PragTyp_TABLE_INFO 31
112355	#define PragTyp_TEMP_STORE 32
112356	#define PragTyp_TEMP_STORE_DIRECTORY 33
112357	#define PragTyp_THREADS 34
	@@ -112112,10 +112361,11 @@
112361	#define PragTyp_HEXKEY 38
112362	#define PragTyp_KEY 39
112363	#define PragTyp_REKEY 40
112364	#define PragTyp_LOCK_STATUS 41
112365	#define PragTyp_PARSER_TRACE 42
112366	#define PragTyp_STATS 43
112367
112368	/* Property flags associated with various pragma. */
112369	#define PragFlg_NeedSchema 0x01 /* Force schema load before running */
112370	#define PragFlg_NoColumns 0x02 /* OP_ResultRow called with zero columns */
112371	#define PragFlg_NoColumns1 0x04 /* zero columns if RHS argument is present */
	@@ -112135,51 +112385,52 @@
112385	/* 2 */ "name",
112386	/* 3 */ "type",
112387	/* 4 */ "notnull",
112388	/* 5 */ "dflt_value",
112389	/* 6 */ "pk",
112390	/* 7 / "tbl", / Used by: stats */
112391	/* 8 */ "idx",
112392	/* 9 */ "wdth",
112393	/* 10 */ "hght",
112394	/* 11 */ "flgs",
112395	/* 12 / "seqno", / Used by: index_info */
112396	/* 13 */ "cid",
112397	/* 14 */ "name",
112398	/* 15 / "seqno", / Used by: index_xinfo */
112399	/* 16 */ "cid",
112400	/* 17 */ "name",
112401	/* 18 */ "desc",
112402	/* 19 */ "coll",
112403	/* 20 */ "key",
112404	/* 21 / "seq", / Used by: index_list */
112405	/* 22 */ "name",
112406	/* 23 */ "unique",
112407	/* 24 */ "origin",
112408	/* 25 */ "partial",
112409	/* 26 / "seq", / Used by: database_list */
112410	/* 27 */ "name",
112411	/* 28 */ "file",
112412	/* 29 / "seq", / Used by: collation_list */
112413	/* 30 */ "name",
112414	/* 31 / "id", / Used by: foreign_key_list */
112415	/* 32 */ "seq",
112416	/* 33 */ "table",
112417	/* 34 */ "from",
112418	/* 35 */ "to",
112419	/* 36 */ "on_update",
112420	/* 37 */ "on_delete",
112421	/* 38 */ "match",
112422	/* 39 / "table", / Used by: foreign_key_check */
112423	/* 40 */ "rowid",
112424	/* 41 */ "parent",
112425	/* 42 */ "fkid",
112426	/* 43 / "busy", / Used by: wal_checkpoint */
112427	/* 44 */ "log",
112428	/* 45 */ "checkpointed",
112429	/* 46 / "timeout", / Used by: busy_timeout */
112430	/* 47 / "database", / Used by: lock_status */
112431	/* 48 */ "status",
112432	};
112433
112434	/* Definitions of all built-in pragmas */
112435	typedef struct PragmaName {
112436	const char const zName; / Name of pragma */
	@@ -112221,11 +112472,11 @@
112472	#endif
112473	#endif
112474	{/* zName: */ "busy_timeout",
112475	/* ePragTyp: */ PragTyp_BUSY_TIMEOUT,
112476	/* ePragFlg: */ PragFlg_Result0,
112477	/* ColNames: */ 46, 1,
112478	/* iArg: */ 0 },
112479	#if !defined(SQLITE_OMIT_PAGER_PRAGMAS)
112480	{/* zName: */ "cache_size",
112481	/* ePragTyp: */ PragTyp_CACHE_SIZE,
112482	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result0\|PragFlg_SchemaReq\|PragFlg_NoColumns1,
	@@ -112258,11 +112509,11 @@
112509	#endif
112510	#if !defined(SQLITE_OMIT_SCHEMA_PRAGMAS)
112511	{/* zName: */ "collation_list",
112512	/* ePragTyp: */ PragTyp_COLLATION_LIST,
112513	/* ePragFlg: */ PragFlg_Result0,
112514	/* ColNames: */ 29, 2,
112515	/* iArg: */ 0 },
112516	#endif
112517	#if !defined(SQLITE_OMIT_COMPILEOPTION_DIAGS)
112518	{/* zName: */ "compile_options",
112519	/* ePragTyp: */ PragTyp_COMPILE_OPTIONS,
	@@ -112293,11 +112544,11 @@
112544	#endif
112545	#if !defined(SQLITE_OMIT_SCHEMA_PRAGMAS)
112546	{/* zName: */ "database_list",
112547	/* ePragTyp: */ PragTyp_DATABASE_LIST,
112548	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result0,
112549	/* ColNames: */ 26, 3,
112550	/* iArg: */ 0 },
112551	#endif
112552	#if !defined(SQLITE_OMIT_PAGER_PRAGMAS) && !defined(SQLITE_OMIT_DEPRECATED)
112553	{/* zName: */ "default_cache_size",
112554	/* ePragTyp: */ PragTyp_DEFAULT_CACHE_SIZE,
	@@ -112330,18 +112581,18 @@
112581	#endif
112582	#if !defined(SQLITE_OMIT_FOREIGN_KEY) && !defined(SQLITE_OMIT_TRIGGER)
112583	{/* zName: */ "foreign_key_check",
112584	/* ePragTyp: */ PragTyp_FOREIGN_KEY_CHECK,
112585	/* ePragFlg: */ PragFlg_NeedSchema,
112586	/* ColNames: */ 39, 4,
112587	/* iArg: */ 0 },
112588	#endif
112589	#if !defined(SQLITE_OMIT_FOREIGN_KEY)
112590	{/* zName: */ "foreign_key_list",
112591	/* ePragTyp: */ PragTyp_FOREIGN_KEY_LIST,
112592	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result1\|PragFlg_SchemaOpt,
112593	/* ColNames: */ 31, 8,
112594	/* iArg: */ 0 },
112595	#endif
112596	#if !defined(SQLITE_OMIT_FLAG_PRAGMAS)
112597	#if !defined(SQLITE_OMIT_FOREIGN_KEY) && !defined(SQLITE_OMIT_TRIGGER)
112598	{/* zName: */ "foreign_keys",
	@@ -112400,21 +112651,21 @@
112651	#endif
112652	#if !defined(SQLITE_OMIT_SCHEMA_PRAGMAS)
112653	{/* zName: */ "index_info",
112654	/* ePragTyp: */ PragTyp_INDEX_INFO,
112655	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result1\|PragFlg_SchemaOpt,
112656	/* ColNames: */ 12, 3,
112657	/* iArg: */ 0 },
112658	{/* zName: */ "index_list",
112659	/* ePragTyp: */ PragTyp_INDEX_LIST,
112660	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result1\|PragFlg_SchemaOpt,
112661	/* ColNames: */ 21, 5,
112662	/* iArg: */ 0 },
112663	{/* zName: */ "index_xinfo",
112664	/* ePragTyp: */ PragTyp_INDEX_INFO,
112665	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result1\|PragFlg_SchemaOpt,
112666	/* ColNames: */ 15, 6,
112667	/* iArg: */ 1 },
112668	#endif
112669	#if !defined(SQLITE_OMIT_INTEGRITY_CHECK)
112670	{/* zName: */ "integrity_check",
112671	/* ePragTyp: */ PragTyp_INTEGRITY_CHECK,
	@@ -112457,11 +112708,11 @@
112708	#endif
112709	#if defined(SQLITE_DEBUG) \|\| defined(SQLITE_TEST)
112710	{/* zName: */ "lock_status",
112711	/* ePragTyp: */ PragTyp_LOCK_STATUS,
112712	/* ePragFlg: */ PragFlg_Result0,
112713	/* ColNames: */ 47, 2,
112714	/* iArg: */ 0 },
112715	#endif
112716	#if !defined(SQLITE_OMIT_PAGER_PRAGMAS)
112717	{/* zName: */ "locking_mode",
112718	/* ePragTyp: */ PragTyp_LOCKING_MODE,
	@@ -112476,10 +112727,17 @@
112727	{/* zName: */ "mmap_size",
112728	/* ePragTyp: */ PragTyp_MMAP_SIZE,
112729	/* ePragFlg: */ 0,
112730	/* ColNames: */ 0, 0,
112731	/* iArg: */ 0 },
112732	#endif
112733	{/* zName: */ "optimize",
112734	/* ePragTyp: */ PragTyp_OPTIMIZE,
112735	/* ePragFlg: */ PragFlg_Result1,
112736	/* ColNames: */ 0, 0,
112737	/* iArg: */ 0 },
112738	#if !defined(SQLITE_OMIT_PAGER_PRAGMAS)
112739	{/* zName: */ "page_count",
112740	/* ePragTyp: */ PragTyp_PAGE_COUNT,
112741	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result0\|PragFlg_SchemaReq,
112742	/* ColNames: */ 0, 0,
112743	/* iArg: */ 0 },
	@@ -112574,15 +112832,15 @@
112832	/* ePragFlg: */ PragFlg_Result0\|PragFlg_NoColumns1,
112833	/* ColNames: */ 0, 0,
112834	/* iArg: */ SQLITE_SqlTrace },
112835	#endif
112836	#endif
112837	#if !defined(SQLITE_OMIT_SCHEMA_PRAGMAS) && defined(SQLITE_DEBUG)
112838	{/* zName: */ "stats",
112839	/* ePragTyp: */ PragTyp_STATS,
112840	/* ePragFlg: */ PragFlg_NeedSchema\|PragFlg_Result0\|PragFlg_SchemaReq,
112841	/* ColNames: */ 7, 5,
112842	/* iArg: */ 0 },
112843	#endif
112844	#if !defined(SQLITE_OMIT_PAGER_PRAGMAS)
112845	{/* zName: */ "synchronous",
112846	/* ePragTyp: */ PragTyp_SYNCHRONOUS,
	@@ -112657,11 +112915,11 @@
112915	/* ColNames: */ 0, 0,
112916	/* iArg: */ 0 },
112917	{/* zName: */ "wal_checkpoint",
112918	/* ePragTyp: */ PragTyp_WAL_CHECKPOINT,
112919	/* ePragFlg: */ PragFlg_NeedSchema,
112920	/* ColNames: */ 43, 3,
112921	/* iArg: */ 0 },
112922	#endif
112923	#if !defined(SQLITE_OMIT_FLAG_PRAGMAS)
112924	{/* zName: */ "writable_schema",
112925	/* ePragTyp: */ PragTyp_FLAG,
	@@ -112668,11 +112926,11 @@
112926	/* ePragFlg: */ PragFlg_Result0\|PragFlg_NoColumns1,
112927	/* ColNames: */ 0, 0,
112928	/* iArg: */ SQLITE_WriteSchema\|SQLITE_RecoveryMode },
112929	#endif
112930	};
112931	/* Number of pragmas: 60 on by default, 74 total. */
112932
112933	/************ End of pragma.h ********************************************/
112934	/************ Continuing where we left off in pragma.c *******************/
112935
112936	/*
	@@ -112937,10 +113195,26 @@
113195	lwr = mid + 1;
113196	}
113197	}
113198	return lwr>upr ? 0 : &aPragmaName[mid];
113199	}
113200
113201	/*
113202	** Helper subroutine for PRAGMA integrity_check:
113203	**
113204	** Generate code to output a single-column result row with the result
113205	** held in register regResult. Decrement the result count and halt if
113206	** the maximum number of result rows have been issued.
113207	*/
113208	static int integrityCheckResultRow(Vdbe *v, int regResult){
113209	int addr;
113210	sqlite3VdbeAddOp2(v, OP_ResultRow, regResult, 1);
113211	addr = sqlite3VdbeAddOp3(v, OP_IfPos, 1, sqlite3VdbeCurrentAddr(v)+2, 1);
113212	VdbeCoverage(v);
113213	sqlite3VdbeAddOp2(v, OP_Halt, 0, 0);
113214	return addr;
113215	}
113216
113217	/*
113218	** Process a pragma statement.
113219	**
113220	** Pragmas are of this form:
	@@ -113741,33 +114015,37 @@
114015	}
114016	}
114017	}
114018	break;
114019
114020	#ifdef SQLITE_DEBUG
114021	case PragTyp_STATS: {
114022	Index *pIdx;
114023	HashElem *i;
114024	pParse->nMem = 5;
114025	sqlite3CodeVerifySchema(pParse, iDb);
114026	for(i=sqliteHashFirst(&pDb->pSchema->tblHash); i; i=sqliteHashNext(i)){
114027	Table *pTab = sqliteHashData(i);
114028	sqlite3VdbeMultiLoad(v, 1, "ssiii",
114029	pTab->zName,
114030	0,
114031	pTab->szTabRow,
114032	pTab->nRowLogEst,
114033	pTab->tabFlags);
114034	sqlite3VdbeAddOp2(v, OP_ResultRow, 1, 5);
114035	for(pIdx=pTab->pIndex; pIdx; pIdx=pIdx->pNext){
114036	sqlite3VdbeMultiLoad(v, 2, "siii",
114037	pIdx->zName,
114038	pIdx->szIdxRow,
114039	pIdx->aiRowLogEst[0],
114040	pIdx->hasStat1);
114041	sqlite3VdbeAddOp2(v, OP_ResultRow, 1, 5);
114042	}
114043	}
114044	}
114045	break;
114046	#endif
114047
114048	case PragTyp_INDEX_INFO: if( zRight ){
114049	Index *pIdx;
114050	Table *pTab;
114051	pIdx = sqlite3FindIndex(db, zRight, zDb);
	@@ -114020,13 +114298,21 @@
114298	#ifndef SQLITE_INTEGRITY_CHECK_ERROR_MAX
114299	# define SQLITE_INTEGRITY_CHECK_ERROR_MAX 100
114300	#endif
114301
114302	#ifndef SQLITE_OMIT_INTEGRITY_CHECK
114303	/* PRAGMA integrity_check
114304	** PRAGMA integrity_check(N)
114305	** PRAGMA quick_check
114306	** PRAGMA quick_check(N)
114307	**
114308	** Verify the integrity of the database.
114309	**
114310	** The "quick_check" is reduced version of
114311	** integrity_check designed to detect most database corruption
114312	** without the overhead of cross-checking indexes. Quick_check
114313	** is linear time wherease integrity_check is O(NlogN).
114314	*/
114315	case PragTyp_INTEGRITY_CHECK: {
114316	int i, j, addr, mxErr;
114317
114318	int isQuick = (sqlite3Tolower(zLeft[0])=='q');
	@@ -114053,11 +114339,11 @@
114339	sqlite3GetInt32(zRight, &mxErr);
114340	if( mxErr<=0 ){
114341	mxErr = SQLITE_INTEGRITY_CHECK_ERROR_MAX;
114342	}
114343	}
114344	sqlite3VdbeAddOp2(v, OP_Integer, mxErr-1, 1); /* reg[1] holds errors left */
114345
114346	/* Do an integrity check on each database file */
114347	for(i=0; i<db->nDb; i++){
114348	HashElem *x;
114349	Hash *pTbls;
	@@ -114068,14 +114354,10 @@
114354
114355	if( OMIT_TEMPDB && i==1 ) continue;
114356	if( iDb>=0 && i!=iDb ) continue;
114357
114358	sqlite3CodeVerifySchema(pParse, i);




114359
114360	/* Do an integrity check of the B-Tree
114361	**
114362	** Begin by finding the root pages numbers
114363	** for all tables and indices in the database.
	@@ -114111,29 +114393,30 @@
114393	sqlite3VdbeAddOp4(v, OP_String8, 0, 3, 0,
114394	sqlite3MPrintf(db, "* in database %s *\n", db->aDb[i].zDbSName),
114395	P4_DYNAMIC);
114396	sqlite3VdbeAddOp3(v, OP_Move, 2, 4, 1);
114397	sqlite3VdbeAddOp3(v, OP_Concat, 4, 3, 2);
114398	integrityCheckResultRow(v, 2);
114399	sqlite3VdbeJumpHere(v, addr);
114400
114401	/* Make sure all the indices are constructed correctly.
114402	*/
114403	for(x=sqliteHashFirst(pTbls); x; x=sqliteHashNext(x)){
114404	Table *pTab = sqliteHashData(x);
114405	Index pIdx, pPk;
114406	Index *pPrior = 0;
114407	int loopTop;
114408	int iDataCur, iIdxCur;
114409	int r1 = -1;
114410
114411	if( pTab->pCheck==0
114412	&& (pTab->tabFlags & TF_HasNotNull)==0
114413	&& (pTab->pIndex==0 \|\| isQuick)
114414	){
114415	continue; /* No additional checks needed for this table */
114416	}
114417	pPk = HasRowid(pTab) ? 0 : sqlite3PrimaryKeyIndex(pTab);




114418	sqlite3ExprCacheClear(pParse);
114419	sqlite3OpenTableAndIndices(pParse, pTab, OP_OpenRead, 0,
114420	1, 0, &iDataCur, &iIdxCur);
114421	sqlite3VdbeAddOp2(v, OP_Integer, 0, 7);
114422	for(j=0, pIdx=pTab->pIndex; pIdx; pIdx=pIdx->pNext, j++){
	@@ -114144,28 +114427,46 @@
114427	sqlite3VdbeAddOp2(v, OP_Rewind, iDataCur, 0); VdbeCoverage(v);
114428	loopTop = sqlite3VdbeAddOp2(v, OP_AddImm, 7, 1);
114429	/* Verify that all NOT NULL columns really are NOT NULL */
114430	for(j=0; j<pTab->nCol; j++){
114431	char *zErr;
114432	int jmp2;
114433	if( j==pTab->iPKey ) continue;
114434	if( pTab->aCol[j].notNull==0 ) continue;
114435	sqlite3ExprCodeGetColumnOfTable(v, pTab, iDataCur, j, 3);
114436	sqlite3VdbeChangeP5(v, OPFLAG_TYPEOFARG);
114437	jmp2 = sqlite3VdbeAddOp1(v, OP_NotNull, 3); VdbeCoverage(v);

114438	zErr = sqlite3MPrintf(db, "NULL value in %s.%s", pTab->zName,
114439	pTab->aCol[j].zName);
114440	sqlite3VdbeAddOp4(v, OP_String8, 0, 3, 0, zErr, P4_DYNAMIC);
114441	integrityCheckResultRow(v, 3);


114442	sqlite3VdbeJumpHere(v, jmp2);
114443	}
114444	/* Verify CHECK constraints */
114445	if( pTab->pCheck && (db->flags & SQLITE_IgnoreChecks)==0 ){
114446	int addrCkFault = sqlite3VdbeMakeLabel(v);
114447	int addrCkOk = sqlite3VdbeMakeLabel(v);
114448	ExprList *pCheck = pTab->pCheck;
114449	char *zErr;
114450	int k;
114451	pParse->iSelfTab = iDataCur;
114452	sqlite3ExprCachePush(pParse);
114453	for(k=pCheck->nExpr-1; k>0; k--){
114454	sqlite3ExprIfFalse(pParse, pCheck->a[k].pExpr, addrCkFault, 0);
114455	}
114456	sqlite3ExprIfTrue(pParse, pCheck->a[0].pExpr, addrCkOk,
114457	SQLITE_JUMPIFNULL);
114458	sqlite3VdbeResolveLabel(v, addrCkFault);
114459	zErr = sqlite3MPrintf(db, "CHECK constraint failed in %s",
114460	pTab->zName);
114461	sqlite3VdbeAddOp4(v, OP_String8, 0, 3, 0, zErr, P4_DYNAMIC);
114462	integrityCheckResultRow(v, 3);
114463	sqlite3VdbeResolveLabel(v, addrCkOk);
114464	sqlite3ExprCachePop(pParse);
114465	}
114466	/* Validate index entries for the current row */
114467	for(j=0, pIdx=pTab->pIndex; pIdx && !isQuick; pIdx=pIdx->pNext, j++){
114468	int jmp2, jmp3, jmp4, jmp5;
114469	int ckUniq = sqlite3VdbeMakeLabel(v);
114470	if( pPk==pIdx ) continue;
114471	r1 = sqlite3GenerateIndexKey(pParse, pIdx, iDataCur, 0, 0, &jmp3,
114472	pPrior, r1);
	@@ -114172,20 +114473,17 @@
114473	pPrior = pIdx;
114474	sqlite3VdbeAddOp2(v, OP_AddImm, 8+j, 1); /* increment entry count */
114475	/* Verify that an index entry exists for the current table row */
114476	jmp2 = sqlite3VdbeAddOp4Int(v, OP_Found, iIdxCur+j, ckUniq, r1,
114477	pIdx->nColumn); VdbeCoverage(v);

114478	sqlite3VdbeLoadString(v, 3, "row ");
114479	sqlite3VdbeAddOp3(v, OP_Concat, 7, 3, 3);
114480	sqlite3VdbeLoadString(v, 4, " missing from index ");
114481	sqlite3VdbeAddOp3(v, OP_Concat, 4, 3, 3);
114482	jmp5 = sqlite3VdbeLoadString(v, 4, pIdx->zName);
114483	sqlite3VdbeAddOp3(v, OP_Concat, 4, 3, 3);
114484	jmp4 = integrityCheckResultRow(v, 3);


114485	sqlite3VdbeJumpHere(v, jmp2);
114486	/* For UNIQUE indexes, verify that only one entry exists with the
114487	** current key. The entry is unique if (1) any column is NULL
114488	** or (2) the next entry has a different key */
114489	if( IsUniqueIndex(pIdx) ){
	@@ -114202,11 +114500,10 @@
114500	jmp6 = sqlite3VdbeAddOp1(v, OP_Next, iIdxCur+j); VdbeCoverage(v);
114501	sqlite3VdbeGoto(v, uniqOk);
114502	sqlite3VdbeJumpHere(v, jmp6);
114503	sqlite3VdbeAddOp4Int(v, OP_IdxGT, iIdxCur+j, uniqOk, r1,
114504	pIdx->nKeyCol); VdbeCoverage(v);

114505	sqlite3VdbeLoadString(v, 3, "non-unique entry in index ");
114506	sqlite3VdbeGoto(v, jmp5);
114507	sqlite3VdbeResolveLabel(v, uniqOk);
114508	}
114509	sqlite3VdbeJumpHere(v, jmp4);
	@@ -114213,40 +114510,39 @@
114510	sqlite3ResolvePartIdxLabel(pParse, jmp3);
114511	}
114512	sqlite3VdbeAddOp2(v, OP_Next, iDataCur, loopTop); VdbeCoverage(v);
114513	sqlite3VdbeJumpHere(v, loopTop-1);
114514	#ifndef SQLITE_OMIT_BTREECOUNT
114515	if( !isQuick ){
114516	sqlite3VdbeLoadString(v, 2, "wrong # of entries in index ");
114517	for(j=0, pIdx=pTab->pIndex; pIdx; pIdx=pIdx->pNext, j++){
114518	if( pPk==pIdx ) continue;
114519	sqlite3VdbeAddOp2(v, OP_Count, iIdxCur+j, 3);
114520	addr = sqlite3VdbeAddOp3(v, OP_Eq, 8+j, 0, 3); VdbeCoverage(v);
114521	sqlite3VdbeChangeP5(v, SQLITE_NOTNULL);
114522	sqlite3VdbeLoadString(v, 3, pIdx->zName);
114523	sqlite3VdbeAddOp3(v, OP_Concat, 3, 2, 7);
114524	integrityCheckResultRow(v, 7);
114525	sqlite3VdbeJumpHere(v, addr);
114526	}

114527	}
114528	#endif /* SQLITE_OMIT_BTREECOUNT */
114529	}
114530	}
114531	{
114532	static const int iLn = VDBE_OFFSET_LINENO(2);
114533	static const VdbeOpList endCode[] = {
114534	{ OP_AddImm, 1, 0, 0}, /* 0 */
114535	{ OP_IfNotZero, 1, 4, 0}, /* 1 */
114536	{ OP_String8, 0, 3, 0}, /* 2 */
114537	{ OP_ResultRow, 3, 1, 0}, /* 3 */
114538	};
114539	VdbeOp *aOp;
114540
114541	aOp = sqlite3VdbeAddOpList(v, ArraySize(endCode), endCode, iLn);
114542	if( aOp ){
114543	aOp[0].p2 = 1-mxErr;
114544	aOp[2].p4type = P4_STATIC;
114545	aOp[2].p4.z = "ok";
114546	}
114547	}
114548	}
	@@ -114466,10 +114762,122 @@
114762	*/
114763	case PragTyp_SHRINK_MEMORY: {
114764	sqlite3_db_release_memory(db);
114765	break;
114766	}
114767
114768	/*
114769	** PRAGMA optimize
114770	** PRAGMA optimize(MASK)
114771	** PRAGMA schema.optimize
114772	** PRAGMA schema.optimize(MASK)
114773	**
114774	** Attempt to optimize the database. All schemas are optimized in the first
114775	** two forms, and only the specified schema is optimized in the latter two.
114776	**
114777	** The details of optimizations performed by this pragma are expected
114778	** to change and improve over time. Applications should anticipate that
114779	** this pragma will perform new optimizations in future releases.
114780	**
114781	** The optional argument is a bitmask of optimizations to perform:
114782	**
114783	** 0x0001 Debugging mode. Do not actually perform any optimizations
114784	** but instead return one line of text for each optimization
114785	** that would have been done. Off by default.
114786	**
114787	** 0x0002 Run ANALYZE on tables that might benefit. On by default.
114788	** See below for additional information.
114789	**
114790	** 0x0004 (Not yet implemented) Record usage and performance
114791	** information from the current session in the
114792	** database file so that it will be available to "optimize"
114793	** pragmas run by future database connections.
114794	**
114795	** 0x0008 (Not yet implemented) Create indexes that might have
114796	** been helpful to recent queries
114797	**
114798	The default MASK is and always shall be 0xfffe. 0xfffe means perform all of the optimizations listed above except Debug Mode, including new
114799	** optimizations that have not yet been invented. If new optimizations are
114800	** ever added that should be off by default, those off-by-default
114801	** optimizations will have bitmasks of 0x10000 or larger.
114802	**
114803	** DETERMINATION OF WHEN TO RUN ANALYZE
114804	**
114805	** In the current implementation, a table is analyzed if only if all of
114806	** the following are true:
114807	**
114808	** (1) MASK bit 0x02 is set.
114809	**
114810	** (2) The query planner used sqlite_stat1-style statistics for one or
114811	** more indexes of the table at some point during the lifetime of
114812	** the current connection.
114813	**
114814	** (3) One or more indexes of the table are currently unanalyzed OR
114815	** the number of rows in the table has increased by 25 times or more
114816	** since the last time ANALYZE was run.
114817	**
114818	** The rules for when tables are analyzed are likely to change in
114819	** future releases.
114820	*/
114821	case PragTyp_OPTIMIZE: {
114822	int iDbLast; /* Loop termination point for the schema loop */
114823	int iTabCur; /* Cursor for a table whose size needs checking */
114824	HashElem k; / Loop over tables of a schema */
114825	Schema pSchema; / The current schema */
114826	Table pTab; / A table in the schema */
114827	Index pIdx; / An index of the table */
114828	LogEst szThreshold; /* Size threshold above which reanalysis is needd */
114829	char zSubSql; / SQL statement for the OP_SqlExec opcode */
114830	u32 opMask; /* Mask of operations to perform */
114831
114832	if( zRight ){
114833	opMask = (u32)sqlite3Atoi(zRight);
114834	if( (opMask & 0x02)==0 ) break;
114835	}else{
114836	opMask = 0xfffe;
114837	}
114838	iTabCur = pParse->nTab++;
114839	for(iDbLast = zDb?iDb:db->nDb-1; iDb<=iDbLast; iDb++){
114840	if( iDb==1 ) continue;
114841	sqlite3CodeVerifySchema(pParse, iDb);
114842	pSchema = db->aDb[iDb].pSchema;
114843	for(k=sqliteHashFirst(&pSchema->tblHash); k; k=sqliteHashNext(k)){
114844	pTab = (Table*)sqliteHashData(k);
114845
114846	/* If table pTab has not been used in a way that would benefit from
114847	** having analysis statistics during the current session, then skip it.
114848	** This also has the effect of skipping virtual tables and views */
114849	if( (pTab->tabFlags & TF_StatsUsed)==0 ) continue;
114850
114851	/* Reanalyze if the table is 25 times larger than the last analysis */
114852	szThreshold = pTab->nRowLogEst + 46; assert( sqlite3LogEst(25)==46 );
114853	for(pIdx=pTab->pIndex; pIdx; pIdx=pIdx->pNext){
114854	if( !pIdx->hasStat1 ){
114855	szThreshold = 0; /* Always analyze if any index lacks statistics */
114856	break;
114857	}
114858	}
114859	if( szThreshold ){
114860	sqlite3OpenTable(pParse, iTabCur, iDb, pTab, OP_OpenRead);
114861	sqlite3VdbeAddOp3(v, OP_IfSmaller, iTabCur,
114862	sqlite3VdbeCurrentAddr(v)+2+(opMask&1), szThreshold);
114863	VdbeCoverage(v);
114864	}
114865	zSubSql = sqlite3MPrintf(db, "ANALYZE \"%w\".\"%w\"",
114866	db->aDb[iDb].zDbSName, pTab->zName);
114867	if( opMask & 0x01 ){
114868	int r1 = sqlite3GetTempReg(pParse);
114869	sqlite3VdbeAddOp4(v, OP_String8, 0, r1, 0, zSubSql, P4_DYNAMIC);
114870	sqlite3VdbeAddOp2(v, OP_ResultRow, r1, 1);
114871	}else{
114872	sqlite3VdbeAddOp4(v, OP_SqlExec, 0, 0, 0, zSubSql, P4_DYNAMIC);
114873	}
114874	}
114875	}
114876	sqlite3VdbeAddOp0(v, OP_Expire);
114877	break;
114878	}
114879
114880	/*
114881	** PRAGMA busy_timeout
114882	** PRAGMA busy_timeout = N
114883	**
	@@ -123782,12 +124190,29 @@
124190	** transient would cause the database file to appear to be deleted
124191	** following reboot.
124192	*/
124193	SQLITE_PRIVATE void sqlite3Vacuum(Parse pParse, Token pNm){
124194	Vdbe *v = sqlite3GetVdbe(pParse);
124195	int iDb = 0;
124196	if( v==0 ) return;
124197	if( pNm ){
124198	#ifndef SQLITE_BUG_COMPATIBLE_20160819
124199	/* Default behavior: Report an error if the argument to VACUUM is
124200	** not recognized */
124201	iDb = sqlite3TwoPartName(pParse, pNm, pNm, &pNm);
124202	if( iDb<0 ) return;
124203	#else
124204	/* When SQLITE_BUG_COMPATIBLE_20160819 is defined, unrecognized arguments
124205	** to VACUUM are silently ignored. This is a back-out of a bug fix that
124206	** occurred on 2016-08-19 (https://www.sqlite.org/src/info/083f9e6270).
124207	** The buggy behavior is required for binary compatibility with some
124208	** legacy applications. */
124209	iDb = sqlite3FindDb(pParse->db, pNm);
124210	if( iDb<0 ) iDb = 0;
124211	#endif
124212	}
124213	if( iDb!=1 ){
124214	sqlite3VdbeAddOp1(v, OP_Vacuum, iDb);
124215	sqlite3VdbeUsesBtree(v, iDb);
124216	}
124217	return;
124218	}
	@@ -124377,12 +124802,11 @@
124802
124803	db = pParse->db;
124804	iDb = sqlite3SchemaToIndex(db, pTable->pSchema);
124805	assert( iDb>=0 );
124806
124807	assert( pTable->nModuleArg==0 );

124808	addModuleArgument(db, pTable, sqlite3NameFromToken(db, pModuleName));
124809	addModuleArgument(db, pTable, 0);
124810	addModuleArgument(db, pTable, sqlite3DbStrDup(db, pTable->zName));
124811	assert( (pParse->sNameToken.z==pName2->z && pName2->z!=0)
124812	\|\| (pParse->sNameToken.z==pName1->z && pName2->z==0)
	@@ -124666,11 +125090,11 @@
125090	const char *zMod;
125091	Module *pMod;
125092	int rc;
125093
125094	assert( pTab );
125095	if( !IsVirtual(pTab) \|\| sqlite3GetVTable(db, pTab) ){
125096	return SQLITE_OK;
125097	}
125098
125099	/* Locate the required virtual table module */
125100	zMod = pTab->azModuleArg[0];
	@@ -124736,11 +125160,11 @@
125160	Table *pTab;
125161	Module *pMod;
125162	const char *zMod;
125163
125164	pTab = sqlite3FindTable(db, zTab, db->aDb[iDb].zDbSName);
125165	assert( pTab && IsVirtual(pTab) && !pTab->pVTable );
125166
125167	/* Locate the required virtual table module */
125168	zMod = pTab->azModuleArg[0];
125169	pMod = (Module*)sqlite3HashFind(&db->aModule, zMod);
125170
	@@ -124790,11 +125214,11 @@
125214	sqlite3Error(db, SQLITE_MISUSE);
125215	sqlite3_mutex_leave(db->mutex);
125216	return SQLITE_MISUSE_BKPT;
125217	}
125218	pTab = pCtx->pTab;
125219	assert( IsVirtual(pTab) );
125220
125221	pParse = sqlite3StackAllocZero(db, sizeof(*pParse));
125222	if( pParse==0 ){
125223	rc = SQLITE_NOMEM_BKPT;
125224	}else{
	@@ -124804,11 +125228,11 @@
125228
125229	if( SQLITE_OK==sqlite3RunParser(pParse, zCreateTable, &zErr)
125230	&& pParse->pNewTable
125231	&& !db->mallocFailed
125232	&& !pParse->pNewTable->pSelect
125233	&& !IsVirtual(pParse->pNewTable)
125234	){
125235	if( !pTab->aCol ){
125236	Table *pNew = pParse->pNewTable;
125237	Index *pIdx;
125238	pTab->aCol = pNew->aCol;
	@@ -125093,11 +125517,11 @@
125517	/* Check to see the left operand is a column in a virtual table */
125518	if( NEVER(pExpr==0) ) return pDef;
125519	if( pExpr->op!=TK_COLUMN ) return pDef;
125520	pTab = pExpr->pTab;
125521	if( NEVER(pTab==0) ) return pDef;
125522	if( !IsVirtual(pTab) ) return pDef;
125523	pVtab = sqlite3GetVTable(db, pTab)->pVtab;
125524	assert( pVtab!=0 );
125525	assert( pVtab->pModule!=0 );
125526	pMod = (sqlite3_module *)pVtab->pModule;
125527	if( pMod->xFindFunction==0 ) return pDef;
	@@ -125188,12 +125612,11 @@
125612	return 0;
125613	}
125614	pMod->pEpoTab = pTab;
125615	pTab->nTabRef = 1;
125616	pTab->pSchema = db->aDb[0].pSchema;
125617	assert( pTab->nModuleArg==0 );

125618	pTab->iPKey = -1;
125619	addModuleArgument(db, pTab, sqlite3DbStrDup(db, pTab->zName));
125620	addModuleArgument(db, pTab, 0);
125621	addModuleArgument(db, pTab, sqlite3DbStrDup(db, pTab->zName));
125622	rc = vtabCallConstructor(db, pTab, pMod, pModule->xConnect, &zErr);
	@@ -125260,11 +125683,11 @@
125683	case SQLITE_VTAB_CONSTRAINT_SUPPORT: {
125684	VtabCtx *p = db->pVtabCtx;
125685	if( !p ){
125686	rc = SQLITE_MISUSE_BKPT;
125687	}else{
125688	assert( p->pTab==0 \|\| IsVirtual(p->pTab) );
125689	p->pVTable->bConstraint = (u8)va_arg(ap, int);
125690	}
125691	break;
125692	}
125693	default:
	@@ -125699,12 +126122,17 @@
126122	WhereOrSet pOrSet; / Record best loops here, if not NULL */
126123	#ifdef SQLITE_ENABLE_STAT3_OR_STAT4
126124	UnpackedRecord pRec; / Probe for stat4 (if required) */
126125	int nRecValid; /* Number of valid fields currently in pRec */
126126	#endif
126127	unsigned int bldFlags; /* SQLITE_BLDF_* flags */
126128	};
126129
126130	/* Allowed values for WhereLoopBuider.bldFlags */
126131	#define SQLITE_BLDF_INDEXED 0x0001 /* An index is used */
126132	#define SQLITE_BLDF_UNIQUE 0x0002 /* All keys of a UNIQUE index used */
126133
126134	/*
126135	** The WHERE clause processing routine has two halves. The
126136	** first part does the start of the WHERE loop and the second
126137	** half does the tail of the WHERE loop. An instance of
126138	** this structure is returned by the first half and passed
	@@ -125715,11 +126143,11 @@
126143	*/
126144	struct WhereInfo {
126145	Parse pParse; / Parsing and code generating context */
126146	SrcList pTabList; / List of tables in the join */
126147	ExprList pOrderBy; / The ORDER BY clause or NULL */
126148	ExprList pResultSet; / Result set of the query */
126149	LogEst iLimit; /* LIMIT if wctrlFlags has WHERE_USE_LIMIT */
126150	int aiCurOnePass[2]; /* OP_OpenWrite cursors for the ONEPASS opt */
126151	int iContinue; /* Jump here to continue with next record */
126152	int iBreak; /* Jump here to break out of the loop */
126153	int savedNQueryLoop; /* pParse->nQueryLoop outside the WHERE loop */
	@@ -126899,10 +127327,11 @@
127327	Parse pParse; / Parsing context */
127328	sqlite3 db; / Database connection */
127329	Vdbe v; / The prepared stmt under constructions */
127330	struct SrcList_item pTabItem; / FROM clause term being coded */
127331	int addrBrk; /* Jump here to break out of the loop */
127332	int addrHalt; /* addrBrk for the outermost loop */
127333	int addrCont; /* Jump here to continue with next cycle */
127334	int iRowidReg = 0; /* Rowid is stored in this register, if not zero */
127335	int iReleaseReg = 0; /* Temp register to free before returning */
127336
127337	pParse = pWInfo->pParse;
	@@ -126939,10 +127368,15 @@
127368	if( pLevel->iFrom>0 && (pTabItem[0].fg.jointype & JT_LEFT)!=0 ){
127369	pLevel->iLeftJoin = ++pParse->nMem;
127370	sqlite3VdbeAddOp2(v, OP_Integer, 0, pLevel->iLeftJoin);
127371	VdbeComment((v, "init LEFT JOIN no-match flag"));
127372	}
127373
127374	/* Compute a safe address to jump to if we discover that the table for
127375	** this loop is empty and can never contribute content. */
127376	for(j=iLevel; j>0 && pWInfo->a[j].iLeftJoin==0; j--){}
127377	addrHalt = pWInfo->a[j].addrBrk;
127378
127379	/* Special case of a FROM clause subquery implemented as a co-routine */
127380	if( pTabItem->fg.viaCoroutine ){
127381	int regYield = pTabItem->regReturn;
127382	sqlite3VdbeAddOp3(v, OP_InitCoroutine, regYield, 0, pTabItem->addrFillSub);
	@@ -127124,11 +127558,11 @@
127558	VdbeCoverageIf(v, pX->op==TK_LT);
127559	VdbeCoverageIf(v, pX->op==TK_GE);
127560	sqlite3ExprCacheAffinityChange(pParse, r1, 1);
127561	sqlite3ReleaseTempReg(pParse, rTemp);
127562	}else{
127563	sqlite3VdbeAddOp2(v, bRev ? OP_Last : OP_Rewind, iCur, addrHalt);
127564	VdbeCoverageIf(v, bRev==0);
127565	VdbeCoverageIf(v, bRev!=0);
127566	}
127567	if( pEnd ){
127568	Expr *pX;
	@@ -127770,11 +128204,11 @@
128204	pLevel->op = OP_Noop;
128205	}else{
128206	codeCursorHint(pTabItem, pWInfo, pLevel, 0);
128207	pLevel->op = aStep[bRev];
128208	pLevel->p1 = iCur;
128209	pLevel->p2 = 1 + sqlite3VdbeAddOp2(v, aStart[bRev], iCur, addrHalt);
128210	VdbeCoverageIf(v, bRev==0);
128211	VdbeCoverageIf(v, bRev!=0);
128212	pLevel->p5 = SQLITE_STMTSTATUS_FULLSCAN_STEP;
128213	}
128214	}
	@@ -128095,19 +128529,10 @@
128529	#ifdef SQLITE_EBCDIC
128530	if( *pnoCase ) return 0;
128531	#endif
128532	pList = pExpr->x.pList;
128533	pLeft = pList->a[1].pExpr;









128534
128535	pRight = sqlite3ExprSkipCollate(pList->a[0].pExpr);
128536	op = pRight->op;
128537	if( op==TK_VARIABLE ){
128538	Vdbe *pReprepare = pParse->pReprepare;
	@@ -128120,10 +128545,27 @@
128545	assert( pRight->op==TK_VARIABLE \|\| pRight->op==TK_REGISTER );
128546	}else if( op==TK_STRING ){
128547	z = pRight->u.zToken;
128548	}
128549	if( z ){
128550
128551	/* If the RHS begins with a digit or a minus sign, then the LHS must
128552	** be an ordinary column (not a virtual table column) with TEXT affinity.
128553	** Otherwise the LHS might be numeric and "lhs >= rhs" would be false
128554	** even though "lhs LIKE rhs" is true. But if the RHS does not start
128555	** with a digit or '-', then "lhs LIKE rhs" will always be false if
128556	** the LHS is numeric and so the optimization still works.
128557	*/
128558	if( sqlite3Isdigit(z[0]) \|\| z[0]=='-' ){
128559	if( pLeft->op!=TK_COLUMN
128560	\|\| sqlite3ExprAffinity(pLeft)!=SQLITE_AFF_TEXT
128561	\|\| IsVirtual(pLeft->pTab) /* Value might be numeric */
128562	){
128563	sqlite3ValueFree(pVal);
128564	return 0;
128565	}
128566	}
128567	cnt = 0;
128568	while( (c=z[cnt])!=0 && c!=wc[0] && c!=wc[1] && c!=wc[2] ){
128569	cnt++;
128570	}
128571	if( cnt!=0 && 255!=(u8)z[cnt-1] ){
	@@ -128748,11 +129190,11 @@
129190	iCur = pFrom->a[i].iCursor;
129191	for(pIdx=pFrom->a[i].pTab->pIndex; pIdx; pIdx=pIdx->pNext){
129192	if( pIdx->aColExpr==0 ) continue;
129193	for(i=0; i<pIdx->nKeyCol; i++){
129194	if( pIdx->aiColumn[i]!=XN_EXPR ) continue;
129195	if( sqlite3ExprCompareSkip(pExpr, pIdx->aColExpr->a[i].pExpr, iCur)==0 ){
129196	*piCur = iCur;
129197	*piColumn = XN_EXPR;
129198	return 1;
129199	}
129200	}
	@@ -129539,11 +129981,12 @@
129981	do{
129982	for(pTerm=pWC->a+k; k<pWC->nTerm; k++, pTerm++){
129983	if( pTerm->leftCursor==iCur
129984	&& pTerm->u.leftColumn==iColumn
129985	&& (iColumn!=XN_EXPR
129986	\|\| sqlite3ExprCompareSkip(pTerm->pExpr->pLeft,
129987	pScan->pIdxExpr,iCur)==0)
129988	&& (pScan->iEquiv<=1 \|\| !ExprHasProperty(pTerm->pExpr, EP_FromJoin))
129989	){
129990	if( (pTerm->eOperator & WO_EQUIV)!=0
129991	&& pScan->nEquiv<ArraySize(pScan->aiCur)
129992	&& (pX = sqlite3ExprSkipCollate(pTerm->pExpr->pRight))->op==TK_COLUMN
	@@ -131716,10 +132159,15 @@
132159	testcase( eOp & WO_IS );
132160	testcase( eOp & WO_ISNULL );
132161	continue;
132162	}
132163
132164	if( IsUniqueIndex(pProbe) && saved_nEq==pProbe->nKeyCol-1 ){
132165	pBuilder->bldFlags \|= SQLITE_BLDF_UNIQUE;
132166	}else{
132167	pBuilder->bldFlags \|= SQLITE_BLDF_INDEXED;
132168	}
132169	pNew->wsFlags = saved_wsFlags;
132170	pNew->u.btree.nEq = saved_nEq;
132171	pNew->u.btree.nBtm = saved_nBtm;
132172	pNew->u.btree.nTop = saved_nTop;
132173	pNew->nLTerm = saved_nLTerm;
	@@ -132263,11 +132711,19 @@
132711	pNew->nOut = rSize;
132712	if( rc ) break;
132713	}
132714	}
132715
132716	pBuilder->bldFlags = 0;
132717	rc = whereLoopAddBtreeIndex(pBuilder, pSrc, pProbe, 0);
132718	if( pBuilder->bldFlags==SQLITE_BLDF_INDEXED ){
132719	/* If a non-unique index is used, or if a prefix of the key for
132720	** unique index is used (making the index functionally non-unique)
132721	** then the sqlite_stat1 data becomes important for scoring the
132722	** plan */
132723	pTab->tabFlags \|= TF_StatsUsed;
132724	}
132725	#ifdef SQLITE_ENABLE_STAT3_OR_STAT4
132726	sqlite3Stat4ProbeFree(pBuilder->pRec);
132727	pBuilder->nRecValid = 0;
132728	pBuilder->pRec = 0;
132729	#endif
	@@ -133443,13 +133899,13 @@
133899	&& (pWInfo->wctrlFlags & WHERE_DISTINCTBY)==0
133900	&& pWInfo->eDistinct==WHERE_DISTINCT_NOOP
133901	&& nRowEst
133902	){
133903	Bitmask notUsed;
133904	int rc = wherePathSatisfiesOrderBy(pWInfo, pWInfo->pResultSet, pFrom,
133905	WHERE_DISTINCTBY, nLoop-1, pFrom->aLoop[nLoop-1], &notUsed);
133906	if( rc==pWInfo->pResultSet->nExpr ){
133907	pWInfo->eDistinct = WHERE_DISTINCT_ORDERED;
133908	}
133909	}
133910	if( pWInfo->pOrderBy ){
133911	if( pWInfo->wctrlFlags & WHERE_DISTINCTBY ){
	@@ -133682,11 +134138,11 @@
134138	SQLITE_PRIVATE WhereInfo *sqlite3WhereBegin(
134139	Parse pParse, / The parser context */
134140	SrcList pTabList, / FROM clause: A list of all tables to be scanned */
134141	Expr pWhere, / The WHERE clause */
134142	ExprList pOrderBy, / An ORDER BY (or GROUP BY) clause, or NULL */
134143	ExprList pResultSet, / Query result set. Req'd for DISTINCT */
134144	u16 wctrlFlags, /* The WHERE_* flags defined in sqliteInt.h */
134145	int iAuxArg /* If WHERE_OR_SUBCLAUSE is set, index cursor number
134146	** If WHERE_USE_LIMIT, then the limit amount */
134147	){
134148	int nByteWInfo; /* Num. bytes allocated for WhereInfo struct */
	@@ -133758,11 +134214,11 @@
134214	goto whereBeginError;
134215	}
134216	pWInfo->pParse = pParse;
134217	pWInfo->pTabList = pTabList;
134218	pWInfo->pOrderBy = pOrderBy;
134219	pWInfo->pResultSet = pResultSet;
134220	pWInfo->aiCurOnePass[0] = pWInfo->aiCurOnePass[1] = -1;
134221	pWInfo->nLevel = nTabList;
134222	pWInfo->iBreak = pWInfo->iContinue = sqlite3VdbeMakeLabel(v);
134223	pWInfo->wctrlFlags = wctrlFlags;
134224	pWInfo->iLimit = iAuxArg;
	@@ -133836,17 +134292,17 @@
134292	/* Analyze all of the subexpressions. */
134293	sqlite3WhereExprAnalyze(pTabList, &pWInfo->sWC);
134294	if( db->mallocFailed ) goto whereBeginError;
134295
134296	if( wctrlFlags & WHERE_WANT_DISTINCT ){
134297	if( isDistinctRedundant(pParse, pTabList, &pWInfo->sWC, pResultSet) ){
134298	/* The DISTINCT marking is pointless. Ignore it. */
134299	pWInfo->eDistinct = WHERE_DISTINCT_UNIQUE;
134300	}else if( pOrderBy==0 ){
134301	/* Try to ORDER BY the result set to make distinct processing easier */
134302	pWInfo->wctrlFlags \|= WHERE_DISTINCTBY;
134303	pWInfo->pOrderBy = pResultSet;
134304	}
134305	}
134306
134307	/* Construct the WhereLoop objects */
134308	#if defined(WHERETRACE_ENABLED)
	@@ -133918,14 +134374,14 @@
134374	}
134375	}
134376	#endif
134377	/* Attempt to omit tables from the join that do not effect the result */
134378	if( pWInfo->nLevel>=2
134379	&& pResultSet!=0
134380	&& OptimizationEnabled(db, SQLITE_OmitNoopJoin)
134381	){
134382	Bitmask tabUsed = sqlite3WhereExprListUsage(pMaskSet, pResultSet);
134383	if( sWLB.pOrderBy ){
134384	tabUsed \|= sqlite3WhereExprListUsage(pMaskSet, sWLB.pOrderBy);
134385	}
134386	while( pWInfo->nLevel>=2 ){
134387	WhereTerm pTerm, pEnd;
	@@ -134704,166 +135160,166 @@
135160	**
135161	********* Begin parsing tables ********************************************/
135162	#define YY_ACTTAB_COUNT (1567)
135163	static const YYACTIONTYPE yy_action[] = {
135164	/* 0 */ 325, 832, 351, 825, 5, 203, 203, 819, 99, 100,
135165	/* 10 */ 90, 978, 978, 853, 856, 845, 845, 97, 97, 98,
135166	/* 20 */ 98, 98, 98, 301, 96, 96, 96, 96, 95, 95,
135167	/* 30 */ 94, 94, 94, 93, 351, 325, 976, 976, 824, 824,
135168	/* 40 */ 826, 946, 354, 99, 100, 90, 978, 978, 853, 856,
135169	/* 50 */ 845, 845, 97, 97, 98, 98, 98, 98, 338, 96,
135170	/* 60 */ 96, 96, 96, 95, 95, 94, 94, 94, 93, 351,
135171	/* 70 */ 95, 95, 94, 94, 94, 93, 351, 791, 976, 976,
135172	/* 80 */ 325, 94, 94, 94, 93, 351, 792, 75, 99, 100,
135173	/* 90 */ 90, 978, 978, 853, 856, 845, 845, 97, 97, 98,
135174	/* 100 */ 98, 98, 98, 450, 96, 96, 96, 96, 95, 95,
135175	/* 110 */ 94, 94, 94, 93, 351, 1333, 155, 155, 2, 325,
135176	/* 120 */ 275, 146, 132, 52, 52, 93, 351, 99, 100, 90,
135177	/* 130 */ 978, 978, 853, 856, 845, 845, 97, 97, 98, 98,
135178	/* 140 */ 98, 98, 101, 96, 96, 96, 96, 95, 95, 94,
135179	/* 150 */ 94, 94, 93, 351, 957, 957, 325, 268, 428, 413,
135180	/* 160 */ 411, 61, 752, 752, 99, 100, 90, 978, 978, 853,
135181	/* 170 */ 856, 845, 845, 97, 97, 98, 98, 98, 98, 60,
135182	/* 180 */ 96, 96, 96, 96, 95, 95, 94, 94, 94, 93,
135183	/* 190 */ 351, 325, 270, 329, 273, 277, 958, 959, 250, 99,
135184	/* 200 */ 100, 90, 978, 978, 853, 856, 845, 845, 97, 97,
135185	/* 210 */ 98, 98, 98, 98, 301, 96, 96, 96, 96, 95,
135186	/* 220 */ 95, 94, 94, 94, 93, 351, 325, 937, 1326, 698,
135187	/* 230 */ 706, 1326, 242, 412, 99, 100, 90, 978, 978, 853,
135188	/* 240 */ 856, 845, 845, 97, 97, 98, 98, 98, 98, 347,
135189	/* 250 */ 96, 96, 96, 96, 95, 95, 94, 94, 94, 93,
135190	/* 260 */ 351, 325, 937, 1327, 384, 699, 1327, 381, 379, 99,
135191	/* 270 */ 100, 90, 978, 978, 853, 856, 845, 845, 97, 97,
135192	/* 280 */ 98, 98, 98, 98, 701, 96, 96, 96, 96, 95,
135193	/* 290 */ 95, 94, 94, 94, 93, 351, 325, 92, 89, 178,
135194	/* 300 */ 833, 935, 373, 700, 99, 100, 90, 978, 978, 853,
135195	/* 310 */ 856, 845, 845, 97, 97, 98, 98, 98, 98, 375,
135196	/* 320 */ 96, 96, 96, 96, 95, 95, 94, 94, 94, 93,
135197	/* 330 */ 351, 325, 1275, 946, 354, 818, 935, 739, 739, 99,
135198	/* 340 */ 100, 90, 978, 978, 853, 856, 845, 845, 97, 97,
135199	/* 350 */ 98, 98, 98, 98, 230, 96, 96, 96, 96, 95,
135200	/* 360 */ 95, 94, 94, 94, 93, 351, 325, 968, 227, 92,
135201	/* 370 */ 89, 178, 373, 300, 99, 100, 90, 978, 978, 853,
135202	/* 380 */ 856, 845, 845, 97, 97, 98, 98, 98, 98, 920,
135203	/* 390 */ 96, 96, 96, 96, 95, 95, 94, 94, 94, 93,
135204	/* 400 */ 351, 325, 449, 447, 447, 447, 147, 737, 737, 99,
135205	/* 410 */ 100, 90, 978, 978, 853, 856, 845, 845, 97, 97,
135206	/* 420 */ 98, 98, 98, 98, 296, 96, 96, 96, 96, 95,
135207	/* 430 */ 95, 94, 94, 94, 93, 351, 325, 419, 231, 957,
135208	/* 440 */ 957, 158, 25, 422, 99, 100, 90, 978, 978, 853,
135209	/* 450 */ 856, 845, 845, 97, 97, 98, 98, 98, 98, 450,
135210	/* 460 */ 96, 96, 96, 96, 95, 95, 94, 94, 94, 93,
135211	/* 470 */ 351, 443, 224, 224, 420, 957, 957, 961, 325, 52,
135212	/* 480 */ 52, 958, 959, 176, 415, 78, 99, 100, 90, 978,
135213	/* 490 */ 978, 853, 856, 845, 845, 97, 97, 98, 98, 98,
135214	/* 500 */ 98, 379, 96, 96, 96, 96, 95, 95, 94, 94,
135215	/* 510 */ 94, 93, 351, 325, 428, 418, 298, 958, 959, 961,
135216	/* 520 */ 81, 99, 88, 90, 978, 978, 853, 856, 845, 845,
135217	/* 530 */ 97, 97, 98, 98, 98, 98, 717, 96, 96, 96,
135218	/* 540 */ 96, 95, 95, 94, 94, 94, 93, 351, 325, 842,
135219	/* 550 */ 842, 854, 857, 996, 318, 343, 379, 100, 90, 978,
135220	/* 560 */ 978, 853, 856, 845, 845, 97, 97, 98, 98, 98,
135221	/* 570 */ 98, 450, 96, 96, 96, 96, 95, 95, 94, 94,
135222	/* 580 */ 94, 93, 351, 325, 350, 350, 350, 260, 377, 340,
135223	/* 590 */ 928, 52, 52, 90, 978, 978, 853, 856, 845, 845,
135224	/* 600 */ 97, 97, 98, 98, 98, 98, 361, 96, 96, 96,
135225	/* 610 */ 96, 95, 95, 94, 94, 94, 93, 351, 86, 445,
135226	/* 620 */ 846, 3, 1202, 361, 360, 378, 344, 813, 957, 957,
135227	/* 630 */ 1299, 86, 445, 729, 3, 212, 169, 287, 405, 282,
135228	/* 640 */ 404, 199, 232, 450, 300, 760, 83, 84, 280, 245,
135229	/* 650 */ 262, 365, 251, 85, 352, 352, 92, 89, 178, 83,
135230	/* 660 */ 84, 242, 412, 52, 52, 448, 85, 352, 352, 246,
135231	/* 670 */ 958, 959, 194, 455, 670, 402, 399, 398, 448, 243,
135232	/* 680 */ 221, 114, 434, 776, 361, 450, 397, 268, 747, 224,
135233	/* 690 */ 224, 132, 132, 198, 832, 434, 452, 451, 428, 427,
135234	/* 700 */ 819, 415, 734, 713, 132, 52, 52, 832, 268, 452,
135235	/* 710 */ 451, 734, 194, 819, 363, 402, 399, 398, 450, 1270,
135236	/* 720 */ 1270, 23, 957, 957, 86, 445, 397, 3, 228, 429,
135237	/* 730 */ 894, 824, 824, 826, 827, 19, 203, 720, 52, 52,
135238	/* 740 */ 428, 408, 439, 249, 824, 824, 826, 827, 19, 229,
135239	/* 750 */ 403, 153, 83, 84, 761, 177, 241, 450, 721, 85,
135240	/* 760 */ 352, 352, 120, 157, 958, 959, 58, 976, 409, 355,
135241	/* 770 */ 330, 448, 268, 428, 430, 320, 790, 32, 32, 86,
135242	/* 780 */ 445, 776, 3, 341, 98, 98, 98, 98, 434, 96,
135243	/* 790 */ 96, 96, 96, 95, 95, 94, 94, 94, 93, 351,
135244	/* 800 */ 832, 120, 452, 451, 813, 886, 819, 83, 84, 976,
135245	/* 810 */ 813, 132, 410, 919, 85, 352, 352, 132, 407, 789,
135246	/* 820 */ 957, 957, 92, 89, 178, 916, 448, 262, 370, 261,
135247	/* 830 */ 82, 913, 80, 262, 370, 261, 776, 824, 824, 826,
135248	/* 840 */ 827, 19, 933, 434, 96, 96, 96, 96, 95, 95,
135249	/* 850 */ 94, 94, 94, 93, 351, 832, 74, 452, 451, 957,
135250	/* 860 */ 957, 819, 958, 959, 120, 92, 89, 178, 944, 2,
135251	/* 870 */ 917, 964, 268, 1, 975, 76, 445, 762, 3, 708,
135252	/* 880 */ 900, 900, 387, 957, 957, 757, 918, 371, 740, 778,
135253	/* 890 */ 756, 257, 824, 824, 826, 827, 19, 417, 741, 450,
135254	/* 900 */ 24, 958, 959, 83, 84, 369, 957, 957, 177, 226,
135255	/* 910 */ 85, 352, 352, 884, 315, 314, 313, 215, 311, 10,
135256	/* 920 */ 10, 683, 448, 349, 348, 958, 959, 908, 777, 157,
135257	/* 930 */ 120, 957, 957, 337, 776, 416, 711, 310, 450, 434,
135258	/* 940 */ 450, 321, 450, 791, 103, 200, 175, 450, 958, 959,
135259	/* 950 */ 907, 832, 792, 452, 451, 9, 9, 819, 10, 10,
135260	/* 960 */ 52, 52, 51, 51, 180, 716, 248, 10, 10, 171,
135261	/* 970 */ 170, 167, 339, 958, 959, 247, 984, 702, 702, 450,
135262	/* 980 */ 715, 233, 686, 982, 888, 983, 182, 913, 824, 824,
135263	/* 990 */ 826, 827, 19, 183, 256, 423, 132, 181, 394, 10,
135264	/* 1000 */ 10, 888, 890, 749, 957, 957, 916, 268, 985, 198,
135265	/* 1010 */ 985, 349, 348, 425, 415, 299, 817, 832, 326, 825,
135266	/* 1020 */ 120, 332, 133, 819, 268, 98, 98, 98, 98, 91,
135267	/* 1030 */ 96, 96, 96, 96, 95, 95, 94, 94, 94, 93,
135268	/* 1040 */ 351, 157, 810, 371, 382, 359, 958, 959, 358, 268,
135269	/* 1050 */ 450, 917, 368, 324, 824, 824, 826, 450, 709, 450,
135270	/* 1060 */ 264, 380, 888, 450, 876, 746, 253, 918, 255, 433,
135271	/* 1070 */ 36, 36, 234, 450, 234, 120, 269, 37, 37, 12,
135272	/* 1080 */ 12, 334, 272, 27, 27, 450, 330, 118, 450, 162,
135273	/* 1090 */ 742, 280, 450, 38, 38, 450, 985, 356, 985, 450,
135274	/* 1100 */ 709, 1209, 450, 132, 450, 39, 39, 450, 40, 40,
135275	/* 1110 */ 450, 362, 41, 41, 450, 42, 42, 450, 254, 28,
135276	/* 1120 */ 28, 450, 29, 29, 31, 31, 450, 43, 43, 450,
135277	/* 1130 */ 44, 44, 450, 714, 45, 45, 450, 11, 11, 767,
135278	/* 1140 */ 450, 46, 46, 450, 268, 450, 105, 105, 450, 47,
135279	/* 1150 */ 47, 450, 48, 48, 450, 237, 33, 33, 450, 172,
135280	/* 1160 */ 49, 49, 450, 50, 50, 34, 34, 274, 122, 122,
135281	/* 1170 */ 450, 123, 123, 450, 124, 124, 450, 897, 56, 56,
135282	/* 1180 */ 450, 896, 35, 35, 450, 267, 450, 817, 450, 817,
135283	/* 1190 */ 106, 106, 450, 53, 53, 385, 107, 107, 450, 817,
135284	/* 1200 */ 108, 108, 817, 450, 104, 104, 121, 121, 119, 119,
135285	/* 1210 */ 450, 117, 112, 112, 450, 276, 450, 225, 111, 111,
135286	/* 1220 */ 450, 730, 450, 109, 109, 450, 673, 674, 675, 911,
135287	/* 1230 */ 110, 110, 317, 998, 55, 55, 57, 57, 692, 331,
135288	/* 1240 */ 54, 54, 26, 26, 696, 30, 30, 317, 936, 197,
135289	/* 1250 */ 196, 195, 335, 281, 336, 446, 331, 745, 689, 436,
135290	/* 1260 */ 440, 444, 120, 72, 386, 223, 175, 345, 757, 932,
135291	/* 1270 */ 20, 286, 319, 756, 815, 372, 374, 202, 202, 202,
135292	/* 1280 */ 263, 395, 285, 74, 208, 21, 696, 719, 718, 883,
135293	/* 1290 */ 120, 120, 120, 120, 120, 754, 278, 828, 77, 74,
135294	/* 1300 */ 726, 727, 785, 783, 879, 202, 999, 208, 893, 892,
135295	/* 1310 */ 893, 892, 694, 816, 763, 116, 774, 1289, 431, 432,
135296	/* 1320 */ 302, 999, 390, 303, 823, 697, 691, 680, 159, 289,
135297	/* 1330 */ 679, 883, 681, 951, 291, 218, 293, 7, 316, 828,
135298	/* 1340 */ 173, 805, 259, 364, 252, 910, 376, 713, 295, 435,
135299	/* 1350 */ 308, 168, 954, 993, 135, 400, 990, 284, 881, 880,
135300	/* 1360 */ 205, 927, 925, 59, 333, 62, 144, 156, 130, 72,
135301	/* 1370 */ 802, 366, 367, 393, 137, 185, 189, 160, 139, 383,
135302	/* 1380 */ 67, 895, 140, 141, 142, 148, 389, 812, 775, 266,
135303	/* 1390 */ 219, 190, 154, 391, 912, 875, 271, 406, 191, 322,
135304	/* 1400 */ 682, 733, 192, 342, 732, 724, 731, 711, 723, 421,
135305	/* 1410 */ 705, 71, 323, 6, 204, 771, 288, 79, 297, 346,
135306	/* 1420 */ 772, 704, 290, 283, 703, 770, 292, 294, 966, 239,
135307	/* 1430 */ 769, 102, 861, 438, 426, 240, 424, 442, 73, 213,
135308	/* 1440 */ 688, 238, 22, 453, 952, 214, 217, 216, 454, 677,
135309	/* 1450 */ 676, 671, 753, 125, 115, 235, 126, 669, 353, 166,
135310	/* 1460 */ 127, 244, 179, 357, 306, 304, 305, 307, 113, 891,
135311	/* 1470 */ 327, 889, 811, 328, 134, 128, 136, 138, 743, 258,
135312	/* 1480 */ 906, 184, 143, 129, 909, 186, 63, 64, 145, 187,
135313	/* 1490 */ 905, 65, 8, 66, 13, 188, 202, 898, 265, 149,
135314	/* 1500 */ 987, 388, 150, 685, 161, 392, 285, 193, 279, 396,
135315	/* 1510 */ 151, 401, 68, 14, 15, 722, 69, 236, 831, 131,
135316	/* 1520 */ 830, 859, 70, 751, 16, 414, 755, 4, 174, 220,
135317	/* 1530 */ 222, 784, 201, 152, 779, 77, 74, 17, 18, 874,
135318	/* 1540 */ 860, 858, 915, 863, 914, 207, 206, 941, 163, 437,
135319	/* 1550 */ 947, 942, 164, 209, 1002, 441, 862, 165, 210, 829,
135320	/* 1560 */ 695, 87, 312, 211, 1291, 1290, 309,
135321	};
135322	static const YYCODETYPE yy_lookahead[] = {
135323	/* 0 */ 19, 95, 53, 97, 22, 24, 24, 101, 27, 28,
135324	/* 10 */ 29, 30, 31, 32, 33, 34, 35, 36, 37, 38,
135325	/* 20 */ 39, 40, 41, 152, 43, 44, 45, 46, 47, 48,
	@@ -135112,55 +135568,55 @@
135568	/* 300 */ 1256, 1200, 1206, 1260, 1247, 1261, 1263, 1262, 1266, 1278,
135569	/* 310 */ 1282, 1292, 1294, 1297, 1298, 1299, 1300, 1221, 1224, 1228,
135570	/* 320 */ 1288, 1291, 1276, 1277, 1295,
135571	};
135572	static const YYACTIONTYPE yy_default[] = {
135573	/* 0 */ 1280, 1270, 1270, 1270, 1202, 1202, 1202, 1202, 1270, 1096,
135574	/* 10 */ 1125, 1125, 1254, 1332, 1332, 1332, 1332, 1332, 1332, 1201,
135575	/* 20 */ 1332, 1332, 1332, 1332, 1270, 1100, 1131, 1332, 1332, 1332,
135576	/* 30 */ 1332, 1203, 1204, 1332, 1332, 1332, 1253, 1255, 1141, 1140,
135577	/* 40 */ 1139, 1138, 1236, 1112, 1136, 1129, 1133, 1203, 1197, 1198,
135578	/* 50 */ 1196, 1200, 1204, 1332, 1132, 1167, 1181, 1166, 1332, 1332,
135579	/* 60 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135580	/* 70 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135581	/* 80 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135582	/* 90 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135583	/* 100 */ 1332, 1332, 1332, 1332, 1175, 1180, 1187, 1179, 1176, 1169,
135584	/* 110 */ 1168, 1170, 1171, 1332, 1019, 1067, 1332, 1332, 1332, 1172,
135585	/* 120 */ 1332, 1173, 1184, 1183, 1182, 1261, 1288, 1287, 1332, 1332,
135586	/* 130 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135587	/* 140 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135588	/* 150 */ 1332, 1332, 1332, 1332, 1332, 1280, 1270, 1025, 1025, 1332,
135589	/* 160 */ 1270, 1270, 1270, 1270, 1270, 1270, 1266, 1100, 1091, 1332,
135590	/* 170 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135591	/* 180 */ 1258, 1256, 1332, 1217, 1332, 1332, 1332, 1332, 1332, 1332,
135592	/* 190 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135593	/* 200 */ 1332, 1332, 1332, 1332, 1096, 1332, 1332, 1332, 1332, 1332,
135594	/* 210 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1282, 1332, 1231,
135595	/* 220 */ 1096, 1096, 1096, 1098, 1080, 1090, 1004, 1135, 1114, 1114,
135596	/* 230 */ 1321, 1135, 1321, 1042, 1302, 1039, 1125, 1114, 1199, 1125,
135597	/* 240 */ 1125, 1097, 1090, 1332, 1324, 1105, 1105, 1323, 1323, 1105,
135598	/* 250 */ 1146, 1070, 1135, 1076, 1076, 1076, 1076, 1105, 1016, 1135,
135599	/* 260 */ 1146, 1070, 1070, 1135, 1105, 1016, 1235, 1318, 1105, 1105,
135600	/* 270 */ 1016, 1210, 1105, 1016, 1105, 1016, 1210, 1068, 1068, 1068,
135601	/* 280 */ 1057, 1210, 1068, 1042, 1068, 1057, 1068, 1068, 1118, 1113,
135602	/* 290 */ 1118, 1113, 1118, 1113, 1118, 1113, 1105, 1205, 1105, 1332,
135603	/* 300 */ 1210, 1214, 1214, 1210, 1130, 1119, 1128, 1126, 1135, 1022,
135604	/* 310 */ 1060, 1285, 1285, 1281, 1281, 1281, 1281, 1329, 1329, 1266,
135605	/* 320 */ 1297, 1297, 1044, 1044, 1297, 1332, 1332, 1332, 1332, 1332,
135606	/* 330 */ 1332, 1292, 1332, 1219, 1332, 1332, 1332, 1332, 1332, 1332,
135607	/* 340 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135608	/* 350 */ 1332, 1332, 1152, 1332, 1000, 1263, 1332, 1332, 1262, 1332,
135609	/* 360 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135610	/* 370 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1320,
135611	/* 380 */ 1332, 1332, 1332, 1332, 1332, 1332, 1234, 1233, 1332, 1332,
135612	/* 390 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135613	/* 400 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332,
135614	/* 410 */ 1332, 1082, 1332, 1332, 1332, 1306, 1332, 1332, 1332, 1332,
135615	/* 420 */ 1332, 1332, 1332, 1127, 1332, 1120, 1332, 1332, 1311, 1332,
135616	/* 430 */ 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1332, 1272,
135617	/* 440 */ 1332, 1332, 1332, 1271, 1332, 1332, 1332, 1332, 1332, 1154,
135618	/* 450 */ 1332, 1153, 1157, 1332, 1010, 1332,
135619	};
135620	/******** End of lemon-generated parsing tables ***************************/
135621
135622	/* The next table maps tokens (terminal symbols) into fallback tokens.
	@@ -135602,147 +136058,147 @@
136058	/* 169 */ "expr ::= expr EQ\|NE expr",
136059	/* 170 */ "expr ::= expr BITAND\|BITOR\|LSHIFT\|RSHIFT expr",
136060	/* 171 */ "expr ::= expr PLUS\|MINUS expr",
136061	/* 172 */ "expr ::= expr STAR\|SLASH\|REM expr",
136062	/* 173 */ "expr ::= expr CONCAT expr",
136063	/* 174 */ "likeop ::= NOT LIKE_KW\|MATCH",
136064	/* 175 */ "expr ::= expr likeop expr",
136065	/* 176 */ "expr ::= expr likeop expr ESCAPE expr",
136066	/* 177 */ "expr ::= expr ISNULL\|NOTNULL",
136067	/* 178 */ "expr ::= expr NOT NULL",
136068	/* 179 */ "expr ::= expr IS expr",
136069	/* 180 */ "expr ::= expr IS NOT expr",
136070	/* 181 */ "expr ::= NOT expr",
136071	/* 182 */ "expr ::= BITNOT expr",
136072	/* 183 */ "expr ::= MINUS expr",
136073	/* 184 */ "expr ::= PLUS expr",
136074	/* 185 */ "between_op ::= BETWEEN",
136075	/* 186 */ "between_op ::= NOT BETWEEN",
136076	/* 187 */ "expr ::= expr between_op expr AND expr",
136077	/* 188 */ "in_op ::= IN",
136078	/* 189 */ "in_op ::= NOT IN",
136079	/* 190 */ "expr ::= expr in_op LP exprlist RP",
136080	/* 191 */ "expr ::= LP select RP",
136081	/* 192 */ "expr ::= expr in_op LP select RP",
136082	/* 193 */ "expr ::= expr in_op nm dbnm paren_exprlist",
136083	/* 194 */ "expr ::= EXISTS LP select RP",
136084	/* 195 */ "expr ::= CASE case_operand case_exprlist case_else END",
136085	/* 196 */ "case_exprlist ::= case_exprlist WHEN expr THEN expr",
136086	/* 197 */ "case_exprlist ::= WHEN expr THEN expr",
136087	/* 198 */ "case_else ::= ELSE expr",
136088	/* 199 */ "case_else ::=",
136089	/* 200 */ "case_operand ::= expr",
136090	/* 201 */ "case_operand ::=",
136091	/* 202 */ "exprlist ::=",
136092	/* 203 */ "nexprlist ::= nexprlist COMMA expr",
136093	/* 204 */ "nexprlist ::= expr",
136094	/* 205 */ "paren_exprlist ::=",
136095	/* 206 */ "paren_exprlist ::= LP exprlist RP",
136096	/* 207 */ "cmd ::= createkw uniqueflag INDEX ifnotexists nm dbnm ON nm LP sortlist RP where_opt",
136097	/* 208 */ "uniqueflag ::= UNIQUE",
136098	/* 209 */ "uniqueflag ::=",
136099	/* 210 */ "eidlist_opt ::=",
136100	/* 211 */ "eidlist_opt ::= LP eidlist RP",
136101	/* 212 */ "eidlist ::= eidlist COMMA nm collate sortorder",
136102	/* 213 */ "eidlist ::= nm collate sortorder",
136103	/* 214 */ "collate ::=",
136104	/* 215 */ "collate ::= COLLATE ID\|STRING",
136105	/* 216 */ "cmd ::= DROP INDEX ifexists fullname",
136106	/* 217 */ "cmd ::= VACUUM",
136107	/* 218 */ "cmd ::= VACUUM nm",
136108	/* 219 */ "cmd ::= PRAGMA nm dbnm",
136109	/* 220 */ "cmd ::= PRAGMA nm dbnm EQ nmnum",
136110	/* 221 */ "cmd ::= PRAGMA nm dbnm LP nmnum RP",
136111	/* 222 */ "cmd ::= PRAGMA nm dbnm EQ minus_num",
136112	/* 223 */ "cmd ::= PRAGMA nm dbnm LP minus_num RP",
136113	/* 224 */ "plus_num ::= PLUS INTEGER\|FLOAT",
136114	/* 225 */ "minus_num ::= MINUS INTEGER\|FLOAT",
136115	/* 226 */ "cmd ::= createkw trigger_decl BEGIN trigger_cmd_list END",
136116	/* 227 */ "trigger_decl ::= temp TRIGGER ifnotexists nm dbnm trigger_time trigger_event ON fullname foreach_clause when_clause",
136117	/* 228 */ "trigger_time ::= BEFORE",
136118	/* 229 */ "trigger_time ::= AFTER",
136119	/* 230 */ "trigger_time ::= INSTEAD OF",
136120	/* 231 */ "trigger_time ::=",
136121	/* 232 */ "trigger_event ::= DELETE\|INSERT",
136122	/* 233 */ "trigger_event ::= UPDATE",
136123	/* 234 */ "trigger_event ::= UPDATE OF idlist",
136124	/* 235 */ "when_clause ::=",
136125	/* 236 */ "when_clause ::= WHEN expr",
136126	/* 237 */ "trigger_cmd_list ::= trigger_cmd_list trigger_cmd SEMI",
136127	/* 238 */ "trigger_cmd_list ::= trigger_cmd SEMI",
136128	/* 239 */ "trnm ::= nm DOT nm",
136129	/* 240 */ "tridxby ::= INDEXED BY nm",
136130	/* 241 */ "tridxby ::= NOT INDEXED",
136131	/* 242 */ "trigger_cmd ::= UPDATE orconf trnm tridxby SET setlist where_opt",
136132	/* 243 */ "trigger_cmd ::= insert_cmd INTO trnm idlist_opt select",
136133	/* 244 */ "trigger_cmd ::= DELETE FROM trnm tridxby where_opt",
136134	/* 245 */ "trigger_cmd ::= select",
136135	/* 246 */ "expr ::= RAISE LP IGNORE RP",
136136	/* 247 */ "expr ::= RAISE LP raisetype COMMA nm RP",
136137	/* 248 */ "raisetype ::= ROLLBACK",
136138	/* 249 */ "raisetype ::= ABORT",
136139	/* 250 */ "raisetype ::= FAIL",
136140	/* 251 */ "cmd ::= DROP TRIGGER ifexists fullname",
136141	/* 252 */ "cmd ::= ATTACH database_kw_opt expr AS expr key_opt",
136142	/* 253 */ "cmd ::= DETACH database_kw_opt expr",
136143	/* 254 */ "key_opt ::=",
136144	/* 255 */ "key_opt ::= KEY expr",
136145	/* 256 */ "cmd ::= REINDEX",
136146	/* 257 */ "cmd ::= REINDEX nm dbnm",
136147	/* 258 */ "cmd ::= ANALYZE",
136148	/* 259 */ "cmd ::= ANALYZE nm dbnm",
136149	/* 260 */ "cmd ::= ALTER TABLE fullname RENAME TO nm",
136150	/* 261 */ "cmd ::= ALTER TABLE add_column_fullname ADD kwcolumn_opt columnname carglist",
136151	/* 262 */ "add_column_fullname ::= fullname",
136152	/* 263 */ "cmd ::= create_vtab",
136153	/* 264 */ "cmd ::= create_vtab LP vtabarglist RP",
136154	/* 265 */ "create_vtab ::= createkw VIRTUAL TABLE ifnotexists nm dbnm USING nm",
136155	/* 266 */ "vtabarg ::=",
136156	/* 267 */ "vtabargtoken ::= ANY",
136157	/* 268 */ "vtabargtoken ::= lp anylist RP",
136158	/* 269 */ "lp ::= LP",
136159	/* 270 */ "with ::=",
136160	/* 271 */ "with ::= WITH wqlist",
136161	/* 272 */ "with ::= WITH RECURSIVE wqlist",
136162	/* 273 */ "wqlist ::= nm eidlist_opt AS LP select RP",
136163	/* 274 */ "wqlist ::= wqlist COMMA nm eidlist_opt AS LP select RP",
136164	/* 275 */ "input ::= cmdlist",
136165	/* 276 */ "cmdlist ::= cmdlist ecmd",
136166	/* 277 */ "cmdlist ::= ecmd",
136167	/* 278 */ "ecmd ::= SEMI",
136168	/* 279 */ "ecmd ::= explain cmdx SEMI",
136169	/* 280 */ "explain ::=",
136170	/* 281 */ "trans_opt ::=",
136171	/* 282 */ "trans_opt ::= TRANSACTION",
136172	/* 283 */ "trans_opt ::= TRANSACTION nm",
136173	/* 284 */ "savepoint_opt ::= SAVEPOINT",
136174	/* 285 */ "savepoint_opt ::=",
136175	/* 286 */ "cmd ::= create_table create_table_args",
136176	/* 287 */ "columnlist ::= columnlist COMMA columnname carglist",
136177	/* 288 */ "columnlist ::= columnname carglist",
136178	/* 289 */ "nm ::= ID\|INDEXED",
136179	/* 290 */ "nm ::= STRING",
136180	/* 291 */ "nm ::= JOIN_KW",
136181	/* 292 */ "typetoken ::= typename",
136182	/* 293 */ "typename ::= ID\|STRING",
136183	/* 294 */ "signed ::= plus_num",
136184	/* 295 */ "signed ::= minus_num",
136185	/* 296 */ "carglist ::= carglist ccons",
136186	/* 297 */ "carglist ::=",
136187	/* 298 */ "ccons ::= NULL onconf",
136188	/* 299 */ "conslist_opt ::= COMMA conslist",
136189	/* 300 */ "conslist ::= conslist tconscomma tcons",
136190	/* 301 */ "conslist ::= tcons",
136191	/* 302 */ "tconscomma ::=",
136192	/* 303 */ "defer_subclause_opt ::= defer_subclause",
136193	/* 304 */ "resolvetype ::= raisetype",
136194	/* 305 */ "selectnowith ::= oneselect",
136195	/* 306 */ "oneselect ::= values",
136196	/* 307 */ "sclp ::= selcollist COMMA",
136197	/* 308 */ "as ::= ID\|STRING",
136198	/* 309 */ "expr ::= term",
136199	/* 310 */ "likeop ::= LIKE_KW\|MATCH",
136200	/* 311 */ "exprlist ::= nexprlist",
136201	/* 312 */ "nmnum ::= plus_num",
136202	/* 313 */ "nmnum ::= nm",
136203	/* 314 */ "nmnum ::= ON",
136204	/* 315 */ "nmnum ::= DELETE",
	@@ -136375,11 +136831,10 @@
136831	{ 173, 3 },
136832	{ 173, 3 },
136833	{ 173, 3 },
136834	{ 173, 3 },
136835	{ 173, 3 },

136836	{ 221, 2 },
136837	{ 173, 3 },
136838	{ 173, 5 },
136839	{ 173, 2 },
136840	{ 173, 3 },
	@@ -136512,10 +136967,11 @@
136967	{ 194, 1 },
136968	{ 195, 1 },
136969	{ 209, 2 },
136970	{ 210, 1 },
136971	{ 173, 1 },
136972	{ 221, 1 },
136973	{ 208, 1 },
136974	{ 230, 1 },
136975	{ 230, 1 },
136976	{ 230, 1 },
136977	{ 230, 1 },
	@@ -136654,11 +137110,11 @@
137110	case 42: /* autoinc ::= */ yytestcase(yyruleno==42);
137111	case 57: /* init_deferred_pred_opt ::= */ yytestcase(yyruleno==57);
137112	case 67: /* defer_subclause_opt ::= */ yytestcase(yyruleno==67);
137113	case 76: /* ifexists ::= */ yytestcase(yyruleno==76);
137114	case 90: /* distinct ::= */ yytestcase(yyruleno==90);
137115	case 214: /* collate ::= */ yytestcase(yyruleno==214);
137116	{yymsp[1].minor.yy194 = 0;}
137117	break;
137118	case 17: /* ifnotexists ::= IF NOT EXISTS */
137119	{yymsp[-2].minor.yy194 = 1;}
137120	break;
	@@ -136798,13 +137254,13 @@
137254	case 144: /* insert_cmd ::= INSERT orconf */ yytestcase(yyruleno==144);
137255	{yymsp[-1].minor.yy194 = yymsp[0].minor.yy194;}
137256	break;
137257	case 58: /* init_deferred_pred_opt ::= INITIALLY DEFERRED */
137258	case 75: /* ifexists ::= IF EXISTS */ yytestcase(yyruleno==75);
137259	case 186: /* between_op ::= NOT BETWEEN */ yytestcase(yyruleno==186);
137260	case 189: /* in_op ::= NOT IN */ yytestcase(yyruleno==189);
137261	case 215: /* collate ::= COLLATE ID\|STRING */ yytestcase(yyruleno==215);
137262	{yymsp[-1].minor.yy194 = 1;}
137263	break;
137264	case 59: /* init_deferred_pred_opt ::= INITIALLY IMMEDIATE */
137265	{yymsp[-1].minor.yy194 = 0;}
137266	break;
	@@ -136964,13 +137420,13 @@
137420	{yymsp[0].minor.yy194 = SF_All;}
137421	break;
137422	case 91: /* sclp ::= */
137423	case 119: /* orderby_opt ::= */ yytestcase(yyruleno==119);
137424	case 126: /* groupby_opt ::= */ yytestcase(yyruleno==126);
137425	case 202: /* exprlist ::= */ yytestcase(yyruleno==202);
137426	case 205: /* paren_exprlist ::= */ yytestcase(yyruleno==205);
137427	case 210: /* eidlist_opt ::= */ yytestcase(yyruleno==210);
137428	{yymsp[1].minor.yy148 = 0;}
137429	break;
137430	case 92: /* selcollist ::= sclp expr as */
137431	{
137432	yymsp[-2].minor.yy148 = sqlite3ExprListAppend(pParse, yymsp[-2].minor.yy148, yymsp[-1].minor.yy190.pExpr);
	@@ -136992,12 +137448,12 @@
137448	yymsp[-3].minor.yy148 = sqlite3ExprListAppend(pParse,yymsp[-3].minor.yy148, pDot);
137449	}
137450	break;
137451	case 95: /* as ::= AS nm */
137452	case 106: /* dbnm ::= DOT nm */ yytestcase(yyruleno==106);
137453	case 224: /* plus_num ::= PLUS INTEGER\|FLOAT */ yytestcase(yyruleno==224);
137454	case 225: /* minus_num ::= MINUS INTEGER\|FLOAT */ yytestcase(yyruleno==225);
137455	{yymsp[-1].minor.yy0 = yymsp[0].minor.yy0;}
137456	break;
137457	case 97: /* from ::= */
137458	{yymsp[1].minor.yy185 = sqlite3DbMallocZero(pParse->db, sizeof(*yymsp[1].minor.yy185));}
137459	break;
	@@ -137076,18 +137532,18 @@
137532	{yymsp[-3].minor.yy194 = sqlite3JoinType(pParse,&yymsp[-3].minor.yy0,&yymsp[-2].minor.yy0,&yymsp[-1].minor.yy0);/X-overwrites-A/}
137533	break;
137534	case 112: /* on_opt ::= ON expr */
137535	case 129: /* having_opt ::= HAVING expr */ yytestcase(yyruleno==129);
137536	case 136: /* where_opt ::= WHERE expr */ yytestcase(yyruleno==136);
137537	case 198: /* case_else ::= ELSE expr */ yytestcase(yyruleno==198);
137538	{yymsp[-1].minor.yy72 = yymsp[0].minor.yy190.pExpr;}
137539	break;
137540	case 113: /* on_opt ::= */
137541	case 128: /* having_opt ::= */ yytestcase(yyruleno==128);
137542	case 135: /* where_opt ::= */ yytestcase(yyruleno==135);
137543	case 199: /* case_else ::= */ yytestcase(yyruleno==199);
137544	case 201: /* case_operand ::= */ yytestcase(yyruleno==201);
137545	{yymsp[1].minor.yy72 = 0;}
137546	break;
137547	case 115: /* indexed_opt ::= INDEXED BY nm */
137548	{yymsp[-2].minor.yy0 = yymsp[0].minor.yy0;}
137549	break;
	@@ -137230,11 +137686,11 @@
137686	case 158: /* term ::= INTEGER */
137687	{
137688	yylhsminor.yy190.pExpr = sqlite3ExprAlloc(pParse->db, TK_INTEGER, &yymsp[0].minor.yy0, 1);
137689	yylhsminor.yy190.zStart = yymsp[0].minor.yy0.z;
137690	yylhsminor.yy190.zEnd = yymsp[0].minor.yy0.z + yymsp[0].minor.yy0.n;
137691	if( yylhsminor.yy190.pExpr ) yylhsminor.yy190.pExpr->flags \|= EP_Leaf\|EP_Resolved;
137692	}
137693	yymsp[0].minor.yy190 = yylhsminor.yy190;
137694	break;
137695	case 159: /* expr ::= VARIABLE */
137696	{
	@@ -137320,17 +137776,14 @@
137776	case 171: /* expr ::= expr PLUS\|MINUS expr */ yytestcase(yyruleno==171);
137777	case 172: /* expr ::= expr STAR\|SLASH\|REM expr */ yytestcase(yyruleno==172);
137778	case 173: /* expr ::= expr CONCAT expr */ yytestcase(yyruleno==173);
137779	{spanBinaryExpr(pParse,yymsp[-1].major,&yymsp[-2].minor.yy190,&yymsp[0].minor.yy190);}
137780	break;
137781	case 174: /* likeop ::= NOT LIKE_KW\|MATCH */



137782	{yymsp[-1].minor.yy0=yymsp[0].minor.yy0; yymsp[-1].minor.yy0.n\|=0x80000000; /yymsp[-1].minor.yy0-overwrite-yymsp[0].minor.yy0/}
137783	break;
137784	case 175: /* expr ::= expr likeop expr */
137785	{
137786	ExprList *pList;
137787	int bNot = yymsp[-1].minor.yy0.n & 0x80000000;
137788	yymsp[-1].minor.yy0.n &= 0x7fffffff;
137789	pList = sqlite3ExprListAppend(pParse,0, yymsp[0].minor.yy190.pExpr);
	@@ -137339,11 +137792,11 @@
137792	exprNot(pParse, bNot, &yymsp[-2].minor.yy190);
137793	yymsp[-2].minor.yy190.zEnd = yymsp[0].minor.yy190.zEnd;
137794	if( yymsp[-2].minor.yy190.pExpr ) yymsp[-2].minor.yy190.pExpr->flags \|= EP_InfixFunc;
137795	}
137796	break;
137797	case 176: /* expr ::= expr likeop expr ESCAPE expr */
137798	{
137799	ExprList *pList;
137800	int bNot = yymsp[-3].minor.yy0.n & 0x80000000;
137801	yymsp[-3].minor.yy0.n &= 0x7fffffff;
137802	pList = sqlite3ExprListAppend(pParse,0, yymsp[-2].minor.yy190.pExpr);
	@@ -137353,43 +137806,43 @@
137806	exprNot(pParse, bNot, &yymsp[-4].minor.yy190);
137807	yymsp[-4].minor.yy190.zEnd = yymsp[0].minor.yy190.zEnd;
137808	if( yymsp[-4].minor.yy190.pExpr ) yymsp[-4].minor.yy190.pExpr->flags \|= EP_InfixFunc;
137809	}
137810	break;
137811	case 177: /* expr ::= expr ISNULL\|NOTNULL */
137812	{spanUnaryPostfix(pParse,yymsp[0].major,&yymsp[-1].minor.yy190,&yymsp[0].minor.yy0);}
137813	break;
137814	case 178: /* expr ::= expr NOT NULL */
137815	{spanUnaryPostfix(pParse,TK_NOTNULL,&yymsp[-2].minor.yy190,&yymsp[0].minor.yy0);}
137816	break;
137817	case 179: /* expr ::= expr IS expr */
137818	{
137819	spanBinaryExpr(pParse,TK_IS,&yymsp[-2].minor.yy190,&yymsp[0].minor.yy190);
137820	binaryToUnaryIfNull(pParse, yymsp[0].minor.yy190.pExpr, yymsp[-2].minor.yy190.pExpr, TK_ISNULL);
137821	}
137822	break;
137823	case 180: /* expr ::= expr IS NOT expr */
137824	{
137825	spanBinaryExpr(pParse,TK_ISNOT,&yymsp[-3].minor.yy190,&yymsp[0].minor.yy190);
137826	binaryToUnaryIfNull(pParse, yymsp[0].minor.yy190.pExpr, yymsp[-3].minor.yy190.pExpr, TK_NOTNULL);
137827	}
137828	break;
137829	case 181: /* expr ::= NOT expr */
137830	case 182: /* expr ::= BITNOT expr */ yytestcase(yyruleno==182);
137831	{spanUnaryPrefix(&yymsp[-1].minor.yy190,pParse,yymsp[-1].major,&yymsp[0].minor.yy190,&yymsp[-1].minor.yy0);/A-overwrites-B/}
137832	break;
137833	case 183: /* expr ::= MINUS expr */
137834	{spanUnaryPrefix(&yymsp[-1].minor.yy190,pParse,TK_UMINUS,&yymsp[0].minor.yy190,&yymsp[-1].minor.yy0);/A-overwrites-B/}
137835	break;
137836	case 184: /* expr ::= PLUS expr */
137837	{spanUnaryPrefix(&yymsp[-1].minor.yy190,pParse,TK_UPLUS,&yymsp[0].minor.yy190,&yymsp[-1].minor.yy0);/A-overwrites-B/}
137838	break;
137839	case 185: /* between_op ::= BETWEEN */
137840	case 188: /* in_op ::= IN */ yytestcase(yyruleno==188);
137841	{yymsp[0].minor.yy194 = 0;}
137842	break;
137843	case 187: /* expr ::= expr between_op expr AND expr */
137844	{
137845	ExprList *pList = sqlite3ExprListAppend(pParse,0, yymsp[-2].minor.yy190.pExpr);
137846	pList = sqlite3ExprListAppend(pParse,pList, yymsp[0].minor.yy190.pExpr);
137847	yymsp[-4].minor.yy190.pExpr = sqlite3PExpr(pParse, TK_BETWEEN, yymsp[-4].minor.yy190.pExpr, 0);
137848	if( yymsp[-4].minor.yy190.pExpr ){
	@@ -137399,11 +137852,11 @@
137852	}
137853	exprNot(pParse, yymsp[-3].minor.yy194, &yymsp[-4].minor.yy190);
137854	yymsp[-4].minor.yy190.zEnd = yymsp[0].minor.yy190.zEnd;
137855	}
137856	break;
137857	case 190: /* expr ::= expr in_op LP exprlist RP */
137858	{
137859	if( yymsp[-1].minor.yy148==0 ){
137860	/* Expressions of the form
137861	**
137862	** expr1 IN ()
	@@ -137452,26 +137905,26 @@
137905	exprNot(pParse, yymsp[-3].minor.yy194, &yymsp[-4].minor.yy190);
137906	}
137907	yymsp[-4].minor.yy190.zEnd = &yymsp[0].minor.yy0.z[yymsp[0].minor.yy0.n];
137908	}
137909	break;
137910	case 191: /* expr ::= LP select RP */
137911	{
137912	spanSet(&yymsp[-2].minor.yy190,&yymsp[-2].minor.yy0,&yymsp[0].minor.yy0); /A-overwrites-B/
137913	yymsp[-2].minor.yy190.pExpr = sqlite3PExpr(pParse, TK_SELECT, 0, 0);
137914	sqlite3PExprAddSelect(pParse, yymsp[-2].minor.yy190.pExpr, yymsp[-1].minor.yy243);
137915	}
137916	break;
137917	case 192: /* expr ::= expr in_op LP select RP */
137918	{
137919	yymsp[-4].minor.yy190.pExpr = sqlite3PExpr(pParse, TK_IN, yymsp[-4].minor.yy190.pExpr, 0);
137920	sqlite3PExprAddSelect(pParse, yymsp[-4].minor.yy190.pExpr, yymsp[-1].minor.yy243);
137921	exprNot(pParse, yymsp[-3].minor.yy194, &yymsp[-4].minor.yy190);
137922	yymsp[-4].minor.yy190.zEnd = &yymsp[0].minor.yy0.z[yymsp[0].minor.yy0.n];
137923	}
137924	break;
137925	case 193: /* expr ::= expr in_op nm dbnm paren_exprlist */
137926	{
137927	SrcList *pSrc = sqlite3SrcListAppend(pParse->db, 0,&yymsp[-2].minor.yy0,&yymsp[-1].minor.yy0);
137928	Select *pSelect = sqlite3SelectNew(pParse, 0,pSrc,0,0,0,0,0,0,0);
137929	if( yymsp[0].minor.yy148 ) sqlite3SrcListFuncArgs(pParse, pSelect ? pSrc : 0, yymsp[0].minor.yy148);
137930	yymsp[-4].minor.yy190.pExpr = sqlite3PExpr(pParse, TK_IN, yymsp[-4].minor.yy190.pExpr, 0);
	@@ -137478,19 +137931,19 @@
137931	sqlite3PExprAddSelect(pParse, yymsp[-4].minor.yy190.pExpr, pSelect);
137932	exprNot(pParse, yymsp[-3].minor.yy194, &yymsp[-4].minor.yy190);
137933	yymsp[-4].minor.yy190.zEnd = yymsp[-1].minor.yy0.z ? &yymsp[-1].minor.yy0.z[yymsp[-1].minor.yy0.n] : &yymsp[-2].minor.yy0.z[yymsp[-2].minor.yy0.n];
137934	}
137935	break;
137936	case 194: /* expr ::= EXISTS LP select RP */
137937	{
137938	Expr *p;
137939	spanSet(&yymsp[-3].minor.yy190,&yymsp[-3].minor.yy0,&yymsp[0].minor.yy0); /A-overwrites-B/
137940	p = yymsp[-3].minor.yy190.pExpr = sqlite3PExpr(pParse, TK_EXISTS, 0, 0);
137941	sqlite3PExprAddSelect(pParse, p, yymsp[-1].minor.yy243);
137942	}
137943	break;
137944	case 195: /* expr ::= CASE case_operand case_exprlist case_else END */
137945	{
137946	spanSet(&yymsp[-4].minor.yy190,&yymsp[-4].minor.yy0,&yymsp[0].minor.yy0); /A-overwrites-C/
137947	yymsp[-4].minor.yy190.pExpr = sqlite3PExpr(pParse, TK_CASE, yymsp[-3].minor.yy72, 0);
137948	if( yymsp[-4].minor.yy190.pExpr ){
137949	yymsp[-4].minor.yy190.pExpr->x.pList = yymsp[-1].minor.yy72 ? sqlite3ExprListAppend(pParse,yymsp[-2].minor.yy148,yymsp[-1].minor.yy72) : yymsp[-2].minor.yy148;
	@@ -137499,313 +137952,314 @@
137952	sqlite3ExprListDelete(pParse->db, yymsp[-2].minor.yy148);
137953	sqlite3ExprDelete(pParse->db, yymsp[-1].minor.yy72);
137954	}
137955	}
137956	break;
137957	case 196: /* case_exprlist ::= case_exprlist WHEN expr THEN expr */
137958	{
137959	yymsp[-4].minor.yy148 = sqlite3ExprListAppend(pParse,yymsp[-4].minor.yy148, yymsp[-2].minor.yy190.pExpr);
137960	yymsp[-4].minor.yy148 = sqlite3ExprListAppend(pParse,yymsp[-4].minor.yy148, yymsp[0].minor.yy190.pExpr);
137961	}
137962	break;
137963	case 197: /* case_exprlist ::= WHEN expr THEN expr */
137964	{
137965	yymsp[-3].minor.yy148 = sqlite3ExprListAppend(pParse,0, yymsp[-2].minor.yy190.pExpr);
137966	yymsp[-3].minor.yy148 = sqlite3ExprListAppend(pParse,yymsp[-3].minor.yy148, yymsp[0].minor.yy190.pExpr);
137967	}
137968	break;
137969	case 200: /* case_operand ::= expr */
137970	{yymsp[0].minor.yy72 = yymsp[0].minor.yy190.pExpr; /A-overwrites-X/}
137971	break;
137972	case 203: /* nexprlist ::= nexprlist COMMA expr */
137973	{yymsp[-2].minor.yy148 = sqlite3ExprListAppend(pParse,yymsp[-2].minor.yy148,yymsp[0].minor.yy190.pExpr);}
137974	break;
137975	case 204: /* nexprlist ::= expr */
137976	{yymsp[0].minor.yy148 = sqlite3ExprListAppend(pParse,0,yymsp[0].minor.yy190.pExpr); /A-overwrites-Y/}
137977	break;
137978	case 206: /* paren_exprlist ::= LP exprlist RP */
137979	case 211: /* eidlist_opt ::= LP eidlist RP */ yytestcase(yyruleno==211);
137980	{yymsp[-2].minor.yy148 = yymsp[-1].minor.yy148;}
137981	break;
137982	case 207: /* cmd ::= createkw uniqueflag INDEX ifnotexists nm dbnm ON nm LP sortlist RP where_opt */
137983	{
137984	sqlite3CreateIndex(pParse, &yymsp[-7].minor.yy0, &yymsp[-6].minor.yy0,
137985	sqlite3SrcListAppend(pParse->db,0,&yymsp[-4].minor.yy0,0), yymsp[-2].minor.yy148, yymsp[-10].minor.yy194,
137986	&yymsp[-11].minor.yy0, yymsp[0].minor.yy72, SQLITE_SO_ASC, yymsp[-8].minor.yy194, SQLITE_IDXTYPE_APPDEF);
137987	}
137988	break;
137989	case 208: /* uniqueflag ::= UNIQUE */
137990	case 249: /* raisetype ::= ABORT */ yytestcase(yyruleno==249);
137991	{yymsp[0].minor.yy194 = OE_Abort;}
137992	break;
137993	case 209: /* uniqueflag ::= */
137994	{yymsp[1].minor.yy194 = OE_None;}
137995	break;
137996	case 212: /* eidlist ::= eidlist COMMA nm collate sortorder */
137997	{
137998	yymsp[-4].minor.yy148 = parserAddExprIdListTerm(pParse, yymsp[-4].minor.yy148, &yymsp[-2].minor.yy0, yymsp[-1].minor.yy194, yymsp[0].minor.yy194);
137999	}
138000	break;
138001	case 213: /* eidlist ::= nm collate sortorder */
138002	{
138003	yymsp[-2].minor.yy148 = parserAddExprIdListTerm(pParse, 0, &yymsp[-2].minor.yy0, yymsp[-1].minor.yy194, yymsp[0].minor.yy194); /A-overwrites-Y/
138004	}
138005	break;
138006	case 216: /* cmd ::= DROP INDEX ifexists fullname */
138007	{sqlite3DropIndex(pParse, yymsp[0].minor.yy185, yymsp[-1].minor.yy194);}
138008	break;
138009	case 217: /* cmd ::= VACUUM */
138010	{sqlite3Vacuum(pParse,0);}
138011	break;
138012	case 218: /* cmd ::= VACUUM nm */
138013	{sqlite3Vacuum(pParse,&yymsp[0].minor.yy0);}
138014	break;
138015	case 219: /* cmd ::= PRAGMA nm dbnm */
138016	{sqlite3Pragma(pParse,&yymsp[-1].minor.yy0,&yymsp[0].minor.yy0,0,0);}
138017	break;
138018	case 220: /* cmd ::= PRAGMA nm dbnm EQ nmnum */
138019	{sqlite3Pragma(pParse,&yymsp[-3].minor.yy0,&yymsp[-2].minor.yy0,&yymsp[0].minor.yy0,0);}
138020	break;
138021	case 221: /* cmd ::= PRAGMA nm dbnm LP nmnum RP */
138022	{sqlite3Pragma(pParse,&yymsp[-4].minor.yy0,&yymsp[-3].minor.yy0,&yymsp[-1].minor.yy0,0);}
138023	break;
138024	case 222: /* cmd ::= PRAGMA nm dbnm EQ minus_num */
138025	{sqlite3Pragma(pParse,&yymsp[-3].minor.yy0,&yymsp[-2].minor.yy0,&yymsp[0].minor.yy0,1);}
138026	break;
138027	case 223: /* cmd ::= PRAGMA nm dbnm LP minus_num RP */
138028	{sqlite3Pragma(pParse,&yymsp[-4].minor.yy0,&yymsp[-3].minor.yy0,&yymsp[-1].minor.yy0,1);}
138029	break;
138030	case 226: /* cmd ::= createkw trigger_decl BEGIN trigger_cmd_list END */
138031	{
138032	Token all;
138033	all.z = yymsp[-3].minor.yy0.z;
138034	all.n = (int)(yymsp[0].minor.yy0.z - yymsp[-3].minor.yy0.z) + yymsp[0].minor.yy0.n;
138035	sqlite3FinishTrigger(pParse, yymsp[-1].minor.yy145, &all);
138036	}
138037	break;
138038	case 227: /* trigger_decl ::= temp TRIGGER ifnotexists nm dbnm trigger_time trigger_event ON fullname foreach_clause when_clause */
138039	{
138040	sqlite3BeginTrigger(pParse, &yymsp[-7].minor.yy0, &yymsp[-6].minor.yy0, yymsp[-5].minor.yy194, yymsp[-4].minor.yy332.a, yymsp[-4].minor.yy332.b, yymsp[-2].minor.yy185, yymsp[0].minor.yy72, yymsp[-10].minor.yy194, yymsp[-8].minor.yy194);
138041	yymsp[-10].minor.yy0 = (yymsp[-6].minor.yy0.n==0?yymsp[-7].minor.yy0:yymsp[-6].minor.yy0); /A-overwrites-T/
138042	}
138043	break;
138044	case 228: /* trigger_time ::= BEFORE */
138045	{ yymsp[0].minor.yy194 = TK_BEFORE; }
138046	break;
138047	case 229: /* trigger_time ::= AFTER */
138048	{ yymsp[0].minor.yy194 = TK_AFTER; }
138049	break;
138050	case 230: /* trigger_time ::= INSTEAD OF */
138051	{ yymsp[-1].minor.yy194 = TK_INSTEAD;}
138052	break;
138053	case 231: /* trigger_time ::= */
138054	{ yymsp[1].minor.yy194 = TK_BEFORE; }
138055	break;
138056	case 232: /* trigger_event ::= DELETE\|INSERT */
138057	case 233: /* trigger_event ::= UPDATE */ yytestcase(yyruleno==233);
138058	{yymsp[0].minor.yy332.a = yymsp[0].major; /A-overwrites-X/ yymsp[0].minor.yy332.b = 0;}
138059	break;
138060	case 234: /* trigger_event ::= UPDATE OF idlist */
138061	{yymsp[-2].minor.yy332.a = TK_UPDATE; yymsp[-2].minor.yy332.b = yymsp[0].minor.yy254;}
138062	break;
138063	case 235: /* when_clause ::= */
138064	case 254: /* key_opt ::= */ yytestcase(yyruleno==254);
138065	{ yymsp[1].minor.yy72 = 0; }
138066	break;
138067	case 236: /* when_clause ::= WHEN expr */
138068	case 255: /* key_opt ::= KEY expr */ yytestcase(yyruleno==255);
138069	{ yymsp[-1].minor.yy72 = yymsp[0].minor.yy190.pExpr; }
138070	break;
138071	case 237: /* trigger_cmd_list ::= trigger_cmd_list trigger_cmd SEMI */
138072	{
138073	assert( yymsp[-2].minor.yy145!=0 );
138074	yymsp[-2].minor.yy145->pLast->pNext = yymsp[-1].minor.yy145;
138075	yymsp[-2].minor.yy145->pLast = yymsp[-1].minor.yy145;
138076	}
138077	break;
138078	case 238: /* trigger_cmd_list ::= trigger_cmd SEMI */
138079	{
138080	assert( yymsp[-1].minor.yy145!=0 );
138081	yymsp[-1].minor.yy145->pLast = yymsp[-1].minor.yy145;
138082	}
138083	break;
138084	case 239: /* trnm ::= nm DOT nm */
138085	{
138086	yymsp[-2].minor.yy0 = yymsp[0].minor.yy0;
138087	sqlite3ErrorMsg(pParse,
138088	"qualified table names are not allowed on INSERT, UPDATE, and DELETE "
138089	"statements within triggers");
138090	}
138091	break;
138092	case 240: /* tridxby ::= INDEXED BY nm */
138093	{
138094	sqlite3ErrorMsg(pParse,
138095	"the INDEXED BY clause is not allowed on UPDATE or DELETE statements "
138096	"within triggers");
138097	}
138098	break;
138099	case 241: /* tridxby ::= NOT INDEXED */
138100	{
138101	sqlite3ErrorMsg(pParse,
138102	"the NOT INDEXED clause is not allowed on UPDATE or DELETE statements "
138103	"within triggers");
138104	}
138105	break;
138106	case 242: /* trigger_cmd ::= UPDATE orconf trnm tridxby SET setlist where_opt */
138107	{yymsp[-6].minor.yy145 = sqlite3TriggerUpdateStep(pParse->db, &yymsp[-4].minor.yy0, yymsp[-1].minor.yy148, yymsp[0].minor.yy72, yymsp[-5].minor.yy194);}
138108	break;
138109	case 243: /* trigger_cmd ::= insert_cmd INTO trnm idlist_opt select */
138110	{yymsp[-4].minor.yy145 = sqlite3TriggerInsertStep(pParse->db, &yymsp[-2].minor.yy0, yymsp[-1].minor.yy254, yymsp[0].minor.yy243, yymsp[-4].minor.yy194);/A-overwrites-R/}
138111	break;
138112	case 244: /* trigger_cmd ::= DELETE FROM trnm tridxby where_opt */
138113	{yymsp[-4].minor.yy145 = sqlite3TriggerDeleteStep(pParse->db, &yymsp[-2].minor.yy0, yymsp[0].minor.yy72);}
138114	break;
138115	case 245: /* trigger_cmd ::= select */
138116	{yymsp[0].minor.yy145 = sqlite3TriggerSelectStep(pParse->db, yymsp[0].minor.yy243); /A-overwrites-X/}
138117	break;
138118	case 246: /* expr ::= RAISE LP IGNORE RP */
138119	{
138120	spanSet(&yymsp[-3].minor.yy190,&yymsp[-3].minor.yy0,&yymsp[0].minor.yy0); /A-overwrites-X/
138121	yymsp[-3].minor.yy190.pExpr = sqlite3PExpr(pParse, TK_RAISE, 0, 0);
138122	if( yymsp[-3].minor.yy190.pExpr ){
138123	yymsp[-3].minor.yy190.pExpr->affinity = OE_Ignore;
138124	}
138125	}
138126	break;
138127	case 247: /* expr ::= RAISE LP raisetype COMMA nm RP */
138128	{
138129	spanSet(&yymsp[-5].minor.yy190,&yymsp[-5].minor.yy0,&yymsp[0].minor.yy0); /A-overwrites-X/
138130	yymsp[-5].minor.yy190.pExpr = sqlite3ExprAlloc(pParse->db, TK_RAISE, &yymsp[-1].minor.yy0, 1);
138131	if( yymsp[-5].minor.yy190.pExpr ) {
138132	yymsp[-5].minor.yy190.pExpr->affinity = (char)yymsp[-3].minor.yy194;
138133	}
138134	}
138135	break;
138136	case 248: /* raisetype ::= ROLLBACK */
138137	{yymsp[0].minor.yy194 = OE_Rollback;}
138138	break;
138139	case 250: /* raisetype ::= FAIL */
138140	{yymsp[0].minor.yy194 = OE_Fail;}
138141	break;
138142	case 251: /* cmd ::= DROP TRIGGER ifexists fullname */
138143	{
138144	sqlite3DropTrigger(pParse,yymsp[0].minor.yy185,yymsp[-1].minor.yy194);
138145	}
138146	break;
138147	case 252: /* cmd ::= ATTACH database_kw_opt expr AS expr key_opt */
138148	{
138149	sqlite3Attach(pParse, yymsp[-3].minor.yy190.pExpr, yymsp[-1].minor.yy190.pExpr, yymsp[0].minor.yy72);
138150	}
138151	break;
138152	case 253: /* cmd ::= DETACH database_kw_opt expr */
138153	{
138154	sqlite3Detach(pParse, yymsp[0].minor.yy190.pExpr);
138155	}
138156	break;
138157	case 256: /* cmd ::= REINDEX */
138158	{sqlite3Reindex(pParse, 0, 0);}
138159	break;
138160	case 257: /* cmd ::= REINDEX nm dbnm */
138161	{sqlite3Reindex(pParse, &yymsp[-1].minor.yy0, &yymsp[0].minor.yy0);}
138162	break;
138163	case 258: /* cmd ::= ANALYZE */
138164	{sqlite3Analyze(pParse, 0, 0);}
138165	break;
138166	case 259: /* cmd ::= ANALYZE nm dbnm */
138167	{sqlite3Analyze(pParse, &yymsp[-1].minor.yy0, &yymsp[0].minor.yy0);}
138168	break;
138169	case 260: /* cmd ::= ALTER TABLE fullname RENAME TO nm */
138170	{
138171	sqlite3AlterRenameTable(pParse,yymsp[-3].minor.yy185,&yymsp[0].minor.yy0);
138172	}
138173	break;
138174	case 261: /* cmd ::= ALTER TABLE add_column_fullname ADD kwcolumn_opt columnname carglist */
138175	{
138176	yymsp[-1].minor.yy0.n = (int)(pParse->sLastToken.z-yymsp[-1].minor.yy0.z) + pParse->sLastToken.n;
138177	sqlite3AlterFinishAddColumn(pParse, &yymsp[-1].minor.yy0);
138178	}
138179	break;
138180	case 262: /* add_column_fullname ::= fullname */
138181	{
138182	disableLookaside(pParse);
138183	sqlite3AlterBeginAddColumn(pParse, yymsp[0].minor.yy185);
138184	}
138185	break;
138186	case 263: /* cmd ::= create_vtab */
138187	{sqlite3VtabFinishParse(pParse,0);}
138188	break;
138189	case 264: /* cmd ::= create_vtab LP vtabarglist RP */
138190	{sqlite3VtabFinishParse(pParse,&yymsp[0].minor.yy0);}
138191	break;
138192	case 265: /* create_vtab ::= createkw VIRTUAL TABLE ifnotexists nm dbnm USING nm */
138193	{
138194	sqlite3VtabBeginParse(pParse, &yymsp[-3].minor.yy0, &yymsp[-2].minor.yy0, &yymsp[0].minor.yy0, yymsp[-4].minor.yy194);
138195	}
138196	break;
138197	case 266: /* vtabarg ::= */
138198	{sqlite3VtabArgInit(pParse);}
138199	break;
138200	case 267: /* vtabargtoken ::= ANY */
138201	case 268: /* vtabargtoken ::= lp anylist RP */ yytestcase(yyruleno==268);
138202	case 269: /* lp ::= LP */ yytestcase(yyruleno==269);
138203	{sqlite3VtabArgExtend(pParse,&yymsp[0].minor.yy0);}
138204	break;
138205	case 270: /* with ::= */
138206	{yymsp[1].minor.yy285 = 0;}
138207	break;
138208	case 271: /* with ::= WITH wqlist */
138209	{ yymsp[-1].minor.yy285 = yymsp[0].minor.yy285; }
138210	break;
138211	case 272: /* with ::= WITH RECURSIVE wqlist */
138212	{ yymsp[-2].minor.yy285 = yymsp[0].minor.yy285; }
138213	break;
138214	case 273: /* wqlist ::= nm eidlist_opt AS LP select RP */
138215	{
138216	yymsp[-5].minor.yy285 = sqlite3WithAdd(pParse, 0, &yymsp[-5].minor.yy0, yymsp[-4].minor.yy148, yymsp[-1].minor.yy243); /A-overwrites-X/
138217	}
138218	break;
138219	case 274: /* wqlist ::= wqlist COMMA nm eidlist_opt AS LP select RP */
138220	{
138221	yymsp[-7].minor.yy285 = sqlite3WithAdd(pParse, yymsp[-7].minor.yy285, &yymsp[-5].minor.yy0, yymsp[-4].minor.yy148, yymsp[-1].minor.yy243);
138222	}
138223	break;
138224	default:
138225	/* (275) input ::= cmdlist */ yytestcase(yyruleno==275);
138226	/* (276) cmdlist ::= cmdlist ecmd */ yytestcase(yyruleno==276);
138227	/* (277) cmdlist ::= ecmd (OPTIMIZED OUT) */ assert(yyruleno!=277);
138228	/* (278) ecmd ::= SEMI */ yytestcase(yyruleno==278);
138229	/* (279) ecmd ::= explain cmdx SEMI */ yytestcase(yyruleno==279);
138230	/* (280) explain ::= */ yytestcase(yyruleno==280);
138231	/* (281) trans_opt ::= */ yytestcase(yyruleno==281);
138232	/* (282) trans_opt ::= TRANSACTION */ yytestcase(yyruleno==282);
138233	/* (283) trans_opt ::= TRANSACTION nm */ yytestcase(yyruleno==283);
138234	/* (284) savepoint_opt ::= SAVEPOINT */ yytestcase(yyruleno==284);
138235	/* (285) savepoint_opt ::= */ yytestcase(yyruleno==285);
138236	/* (286) cmd ::= create_table create_table_args */ yytestcase(yyruleno==286);
138237	/* (287) columnlist ::= columnlist COMMA columnname carglist */ yytestcase(yyruleno==287);
138238	/* (288) columnlist ::= columnname carglist */ yytestcase(yyruleno==288);
138239	/* (289) nm ::= ID\|INDEXED */ yytestcase(yyruleno==289);
138240	/* (290) nm ::= STRING */ yytestcase(yyruleno==290);
138241	/* (291) nm ::= JOIN_KW */ yytestcase(yyruleno==291);
138242	/* (292) typetoken ::= typename */ yytestcase(yyruleno==292);
138243	/* (293) typename ::= ID\|STRING */ yytestcase(yyruleno==293);
138244	/* (294) signed ::= plus_num (OPTIMIZED OUT) */ assert(yyruleno!=294);
138245	/* (295) signed ::= minus_num (OPTIMIZED OUT) */ assert(yyruleno!=295);
138246	/* (296) carglist ::= carglist ccons */ yytestcase(yyruleno==296);
138247	/* (297) carglist ::= */ yytestcase(yyruleno==297);
138248	/* (298) ccons ::= NULL onconf */ yytestcase(yyruleno==298);
138249	/* (299) conslist_opt ::= COMMA conslist */ yytestcase(yyruleno==299);
138250	/* (300) conslist ::= conslist tconscomma tcons */ yytestcase(yyruleno==300);
138251	/* (301) conslist ::= tcons (OPTIMIZED OUT) */ assert(yyruleno!=301);
138252	/* (302) tconscomma ::= */ yytestcase(yyruleno==302);
138253	/* (303) defer_subclause_opt ::= defer_subclause (OPTIMIZED OUT) */ assert(yyruleno!=303);
138254	/* (304) resolvetype ::= raisetype (OPTIMIZED OUT) */ assert(yyruleno!=304);
138255	/* (305) selectnowith ::= oneselect (OPTIMIZED OUT) */ assert(yyruleno!=305);
138256	/* (306) oneselect ::= values */ yytestcase(yyruleno==306);
138257	/* (307) sclp ::= selcollist COMMA */ yytestcase(yyruleno==307);
138258	/* (308) as ::= ID\|STRING */ yytestcase(yyruleno==308);
138259	/* (309) expr ::= term (OPTIMIZED OUT) */ assert(yyruleno!=309);
138260	/* (310) likeop ::= LIKE_KW\|MATCH */ yytestcase(yyruleno==310);
138261	/* (311) exprlist ::= nexprlist */ yytestcase(yyruleno==311);
138262	/* (312) nmnum ::= plus_num (OPTIMIZED OUT) */ assert(yyruleno!=312);
138263	/* (313) nmnum ::= nm (OPTIMIZED OUT) */ assert(yyruleno!=313);
138264	/* (314) nmnum ::= ON */ yytestcase(yyruleno==314);
138265	/* (315) nmnum ::= DELETE */ yytestcase(yyruleno==315);
	@@ -138854,12 +139308,12 @@
139308	** memory obtained from sqlite3_malloc() and to make *pzErrMsg point to that
139309	** error message.
139310	*/
139311	SQLITE_PRIVATE int sqlite3RunParser(Parse pParse, const char zSql, char **pzErrMsg){
139312	int nErr = 0; /* Number of errors encountered */

139313	void pEngine; / The LEMON-generated LALR(1) parser */
139314	int n = 0; /* Length of the next token token */
139315	int tokenType; /* type of the next token */
139316	int lastTokenParsed = -1; /* type of the previous token */
139317	sqlite3 db = pParse->db; / The database connection */
139318	int mxSqlLen; /* Max length of an SQL string */
139319	#ifdef sqlite3Parser_ENGINEALWAYSONSTACK
	@@ -138871,11 +139325,10 @@
139325	if( db->nVdbeActive==0 ){
139326	db->u1.isInterrupted = 0;
139327	}
139328	pParse->rc = SQLITE_OK;
139329	pParse->zTail = zSql;

139330	assert( pzErrMsg!=0 );
139331	/* sqlite3ParserTrace(stdout, "parser: "); */
139332	#ifdef sqlite3Parser_ENGINEALWAYSONSTACK
139333	pEngine = zSpace;
139334	sqlite3ParserInit(pEngine);
	@@ -138889,16 +139342,14 @@
139342	assert( pParse->pNewTable==0 );
139343	assert( pParse->pNewTrigger==0 );
139344	assert( pParse->nVar==0 );
139345	assert( pParse->pVList==0 );
139346	while( 1 ){
139347	if( zSql[0]!=0 ){
139348	n = sqlite3GetToken((u8*)zSql, &tokenType);
139349	mxSqlLen -= n;
139350	if( mxSqlLen<0 ){


139351	pParse->rc = SQLITE_TOOBIG;
139352	break;
139353	}
139354	}else{
139355	/* Upon reaching the end of input, call the parser two more times
	@@ -138908,30 +139359,34 @@
139359	}else if( lastTokenParsed==0 ){
139360	break;
139361	}else{
139362	tokenType = TK_SEMI;
139363	}
139364	zSql -= n;
139365	}
139366	if( tokenType>=TK_SPACE ){
139367	assert( tokenType==TK_SPACE \|\| tokenType==TK_ILLEGAL );
139368	if( db->u1.isInterrupted ){
139369	pParse->rc = SQLITE_INTERRUPT;
139370	break;
139371	}
139372	if( tokenType==TK_ILLEGAL ){
139373	sqlite3ErrorMsg(pParse, "unrecognized token: \"%.*s\"", n, zSql);

139374	break;
139375	}
139376	zSql += n;
139377	}else{
139378	pParse->sLastToken.z = zSql;
139379	pParse->sLastToken.n = n;
139380	sqlite3Parser(pEngine, tokenType, pParse->sLastToken, pParse);
139381	lastTokenParsed = tokenType;
139382	zSql += n;
139383	if( pParse->rc!=SQLITE_OK \|\| db->mallocFailed ) break;
139384	}
139385	}
139386	assert( nErr==0 );
139387	pParse->zTail = zSql;
139388	#ifdef YYTRACKMAXSTACKDEPTH
139389	sqlite3_mutex_enter(sqlite3MallocMutex());
139390	sqlite3StatusHighwater(SQLITE_STATUS_PARSER_STACK,
139391	sqlite3ParserStackPeak(pEngine)
139392	);
	@@ -140301,10 +140756,25 @@
140756	return 0;
140757	}
140758	#endif
140759	return db->lastRowid;
140760	}
140761
140762	/*
140763	** Set the value returned by the sqlite3_last_insert_rowid() API function.
140764	*/
140765	SQLITE_API void sqlite3_set_last_insert_rowid(sqlite3 *db, sqlite3_int64 iRowid){
140766	#ifdef SQLITE_ENABLE_API_ARMOR
140767	if( !sqlite3SafetyCheckOk(db) ){
140768	(void)SQLITE_MISUSE_BKPT;
140769	return;
140770	}
140771	#endif
140772	sqlite3_mutex_enter(db->mutex);
140773	db->lastRowid = iRowid;
140774	sqlite3_mutex_leave(db->mutex);
140775	}
140776
140777	/*
140778	** Return the number of changes in the most recent call to sqlite3_exec().
140779	*/
140780	SQLITE_API int sqlite3_changes(sqlite3 *db){
	@@ -148111,12 +148581,14 @@
148581	** segments.
148582	*/
148583	const u32 nMinMerge = 64; /* Minimum amount of incr-merge work to do */
148584
148585	Fts3Table p = (Fts3Table)pVtab;
148586	int rc;
148587	i64 iLastRowid = sqlite3_last_insert_rowid(p->db);
148588
148589	rc = sqlite3Fts3PendingTermsFlush(p);
148590	if( rc==SQLITE_OK
148591	&& p->nLeafAdd>(nMinMerge/16)
148592	&& p->nAutoincrmerge && p->nAutoincrmerge!=0xff
148593	){
148594	int mxLevel = 0; /* Maximum relative level value in db */
	@@ -148127,10 +148599,11 @@
148599	A = p->nLeafAdd * mxLevel;
148600	A += (A/2);
148601	if( A>(int)nMinMerge ) rc = sqlite3Fts3Incrmerge(p, A, p->nAutoincrmerge);
148602	}
148603	sqlite3Fts3SegmentsClose(p);
148604	sqlite3_set_last_insert_rowid(p->db, iLastRowid);
148605	return rc;
148606	}
148607
148608	/*
148609	** If it is currently unknown whether or not the FTS table has an %_stat
	@@ -168462,10 +168935,11 @@
168935	int nStep; /* Rows processed for current object */
168936	int nProgress; /* Rows processed for all objects */
168937	RbuObjIter objiter; /* Iterator for skipping through tbl/idx */
168938	const char zVfsName; / Name of automatically created rbu vfs */
168939	rbu_file pTargetFd; / File handle open on target db */
168940	int nPagePerSector; /* Pages per sector for pTargetFd */
168941	i64 iOalSz;
168942	i64 nPhaseOneStep;
168943
168944	/* The following state variables are used as part of the incremental
168945	** checkpoint stage (eStage==RBU_STAGE_CKPT). See comments surrounding
	@@ -170726,10 +171200,27 @@
171200
171201	if( p->rc==SQLITE_OK ){
171202	if( p->nFrame==0 \|\| (pState && pState->iWalCksum!=p->iWalCksum) ){
171203	p->rc = SQLITE_DONE;
171204	p->eStage = RBU_STAGE_DONE;
171205	}else{
171206	int nSectorSize;
171207	sqlite3_file *pDb = p->pTargetFd->pReal;
171208	sqlite3_file *pWal = p->pTargetFd->pWalFd->pReal;
171209	assert( p->nPagePerSector==0 );
171210	nSectorSize = pDb->pMethods->xSectorSize(pDb);
171211	if( nSectorSize>p->pgsz ){
171212	p->nPagePerSector = nSectorSize / p->pgsz;
171213	}else{
171214	p->nPagePerSector = 1;
171215	}
171216
171217	/* Call xSync() on the wal file. This causes SQLite to sync the
171218	** directory in which the target database and the wal file reside, in
171219	** case it has not been synced since the rename() call in
171220	** rbuMoveOalFile(). */
171221	p->rc = pWal->pMethods->xSync(pWal, SQLITE_SYNC_NORMAL);
171222	}
171223	}
171224	}
171225
171226	/*
	@@ -171381,13 +171872,30 @@
171872	if( p->rc==SQLITE_OK ){
171873	p->eStage = RBU_STAGE_DONE;
171874	p->rc = SQLITE_DONE;
171875	}
171876	}else{
171877	/* At one point the following block copied a single frame from the
171878	** wal file to the database file. So that one call to sqlite3rbu_step()
171879	** checkpointed a single frame.
171880	**
171881	** However, if the sector-size is larger than the page-size, and the
171882	** application calls sqlite3rbu_savestate() or close() immediately
171883	** after this step, then rbu_step() again, then a power failure occurs,
171884	** then the database page written here may be damaged. Work around
171885	** this by checkpointing frames until the next page in the aFrame[]
171886	** lies on a different disk sector to the current one. */
171887	u32 iSector;
171888	do{
171889	RbuFrame *pFrame = &p->aFrame[p->nStep];
171890	iSector = (pFrame->iDbPage-1) / p->nPagePerSector;
171891	rbuCheckpointFrame(p, pFrame);
171892	p->nStep++;
171893	}while( p->nStep<p->nFrame
171894	&& iSector==((p->aFrame[p->nStep].iDbPage-1) / p->nPagePerSector)
171895	&& p->rc==SQLITE_OK
171896	);
171897	}
171898	p->nProgress++;
171899	}
171900	break;
171901	}
	@@ -171823,10 +172331,16 @@
172331
172332	/* Commit the transaction to the -oal file. /
172333	if( p->rc==SQLITE_OK && p->eStage==RBU_STAGE_OAL ){
172334	p->rc = sqlite3_exec(p->dbMain, "COMMIT", 0, 0, &p->zErrmsg);
172335	}
172336
172337	/* Sync the db file if currently doing an incremental checkpoint */
172338	if( p->rc==SQLITE_OK && p->eStage==RBU_STAGE_CKPT ){
172339	sqlite3_file *pDb = p->pTargetFd->pReal;
172340	p->rc = pDb->pMethods->xSync(pDb, SQLITE_SYNC_NORMAL);
172341	}
172342
172343	rbuSaveState(p, p->eStage);
172344
172345	if( p->rc==SQLITE_OK && p->eStage==RBU_STAGE_OAL ){
172346	p->rc = sqlite3_exec(p->dbRbu, "COMMIT", 0, 0, &p->zErrmsg);
	@@ -171947,10 +172461,16 @@
172461	assert( p->eStage>=RBU_STAGE_OAL && p->eStage<=RBU_STAGE_DONE );
172462	if( p->eStage==RBU_STAGE_OAL ){
172463	assert( rc!=SQLITE_DONE );
172464	if( rc==SQLITE_OK ) rc = sqlite3_exec(p->dbMain, "COMMIT", 0, 0, 0);
172465	}
172466
172467	/* Sync the db file */
172468	if( rc==SQLITE_OK && p->eStage==RBU_STAGE_CKPT ){
172469	sqlite3_file *pDb = p->pTargetFd->pReal;
172470	rc = pDb->pMethods->xSync(pDb, SQLITE_SYNC_NORMAL);
172471	}
172472
172473	p->rc = rc;
172474	rbuSaveState(p, p->eStage);
172475	rc = p->rc;
172476
	@@ -181079,11 +181599,13 @@
181599	typedef unsigned short u16;
181600	typedef short i16;
181601	typedef sqlite3_int64 i64;
181602	typedef sqlite3_uint64 u64;
181603
181604	#ifndef ArraySize
181605	# define ArraySize(x) ((int)(sizeof(x) / sizeof(x[0])))
181606	#endif
181607
181608	#define testcase(x)
181609	#define ALWAYS(x) 1
181610	#define NEVER(x) 0
181611
	@@ -186302,11 +186824,14 @@
186824	if( p1->bEof==0 ){
186825	if( (p1->iRowid==iLast)
186826	\|\| (bFromValid && fts5RowidCmp(pExpr, p1->iRowid, iFrom)<0)
186827	){
186828	int rc = fts5ExprNodeNext(pExpr, p1, bFromValid, iFrom);
186829	if( rc!=SQLITE_OK ){
186830	pNode->bNomatch = 0;
186831	return rc;
186832	}
186833	}
186834	}
186835	}
186836
186837	fts5ExprNodeTest_OR(pExpr, pNode);
	@@ -186333,11 +186858,14 @@
186858	Fts5ExprNode *pChild = pAnd->apChild[iChild];
186859	int cmp = fts5RowidCmp(pExpr, iLast, pChild->iRowid);
186860	if( cmp>0 ){
186861	/* Advance pChild until it points to iLast or laster */
186862	rc = fts5ExprNodeNext(pExpr, pChild, 1, iLast);
186863	if( rc!=SQLITE_OK ){
186864	pAnd->bNomatch = 0;
186865	return rc;
186866	}
186867	}
186868
186869	/* If the child node is now at EOF, so is the parent AND node. Otherwise,
186870	** the child node is guaranteed to have advanced at least as far as
186871	** rowid iLast. So if it is not at exactly iLast, pChild->iRowid is the
	@@ -186372,10 +186900,12 @@
186900	i64 iFrom
186901	){
186902	int rc = fts5ExprNodeNext(pExpr, pNode->apChild[0], bFromValid, iFrom);
186903	if( rc==SQLITE_OK ){
186904	rc = fts5ExprNodeTest_AND(pExpr, pNode);
186905	}else{
186906	pNode->bNomatch = 0;
186907	}
186908	return rc;
186909	}
186910
186911	static int fts5ExprNodeTest_NOT(
	@@ -186414,10 +186944,13 @@
186944	){
186945	int rc = fts5ExprNodeNext(pExpr, pNode->apChild[0], bFromValid, iFrom);
186946	if( rc==SQLITE_OK ){
186947	rc = fts5ExprNodeTest_NOT(pExpr, pNode);
186948	}
186949	if( rc!=SQLITE_OK ){
186950	pNode->bNomatch = 0;
186951	}
186952	return rc;
186953	}
186954
186955	/*
186956	** If pNode currently points to a match, this function returns SQLITE_OK
	@@ -197534,11 +198067,11 @@
198067	int nArg, /* Number of args */
198068	sqlite3_value *apUnused / Function arguments */
198069	){
198070	assert( nArg==0 );
198071	UNUSED_PARAM2(nArg, apUnused);
198072	sqlite3_result_text(pCtx, "fts5: 2017-03-10 17:03:11 f8560c60d10c0365b33342ab05b5a953987b0471", -1, SQLITE_TRANSIENT);
198073	}
198074
198075	static int fts5Init(sqlite3 *db){
198076	static const sqlite3_module fts5Mod = {
198077	/* iVersion */ 2,
	@@ -198197,15 +198730,10 @@
198730	sqlite3_step(pDel);
198731	rc = sqlite3_reset(pDel);
198732	}
198733	}
198734





198735	return rc;
198736	}
198737
198738	/*
198739	** Delete all entries in the FTS5 index.
	@@ -198405,15 +198933,10 @@
198933	if( rc==SQLITE_OK ){
198934	rc = fts5StorageInsertDocsize(p, iRowid, &buf);
198935	}
198936	sqlite3_free(buf.p);
198937





198938	return rc;
198939	}
198940
198941	static int fts5StorageCount(Fts5Storage p, const char zSuffix, i64 *pnRow){
198942	Fts5Config *pConfig = p->pConfig;
	@@ -198744,16 +199267,21 @@
199267
199268	/*
199269	** Flush any data currently held in-memory to disk.
199270	*/
199271	static int sqlite3Fts5StorageSync(Fts5Storage *p, int bCommit){
199272	int rc = SQLITE_OK;
199273	i64 iLastRowid = sqlite3_last_insert_rowid(p->pConfig->db);
199274	if( p->bTotalsValid ){
199275	rc = fts5StorageSaveTotals(p);
199276	if( bCommit ) p->bTotalsValid = 0;
199277	}
199278	if( rc==SQLITE_OK ){
199279	rc = sqlite3Fts5IndexSync(p->pIndex, bCommit);
199280	}
199281	sqlite3_set_last_insert_rowid(p->pConfig->db, iLastRowid);
199282	return rc;
199283	}
199284
199285	static int sqlite3Fts5StorageRollback(Fts5Storage *p){
199286	p->bTotalsValid = 0;
199287	return sqlite3Fts5IndexRollback(p->pIndex);
199288

M src/sqlite3.h

+36 -16

		--- src/sqlite3.h
		+++ src/sqlite3.h
		@@ -119,13 +119,13 @@
119	119	**
120	120	** See also: [sqlite3_libversion()],
121	121	** [sqlite3_libversion_number()], [sqlite3_sourceid()],
122	122	** [sqlite_version()] and [sqlite_source_id()].
123	123	*/
124		-#define SQLITE_VERSION "3.17.0"
125		-#define SQLITE_VERSION_NUMBER 3017000
126		-#define SQLITE_SOURCE_ID "2017-02-13 16:02:40 ada05cfa86ad7f5645450ac7a2a21c9aa6e57d2c"
	124	+#define SQLITE_VERSION "3.18.0"
	125	+#define SQLITE_VERSION_NUMBER 3018000
	126	+#define SQLITE_SOURCE_ID "2017-03-10 17:03:11 f8560c60d10c0365b33342ab05b5a953987b0471"
127	127
128	128	/*
129	129	** CAPI3REF: Run-Time Library Version Numbers
130	130	** KEYWORDS: sqlite3_version sqlite3_sourceid
131	131	**
		@@ -2038,24 +2038,34 @@
2038	2038	** as an undeclared column named ROWID, OID, or _ROWID_ as long as those
2039	2039	** names are not also used by explicitly declared columns. ^If
2040	2040	** the table has a column of type [INTEGER PRIMARY KEY] then that column
2041	2041	** is another alias for the rowid.
2042	2042	**
2043		-** ^The sqlite3_last_insert_rowid(D) interface returns the [rowid] of the
2044		-** most recent successful [INSERT] into a rowid table or [virtual table]
2045		-** on database connection D.
2046		-** ^Inserts into [WITHOUT ROWID] tables are not recorded.
2047		-** ^If no successful [INSERT]s into rowid tables
2048		-** have ever occurred on the database connection D,
2049		-** then sqlite3_last_insert_rowid(D) returns zero.
	2043	+** ^The sqlite3_last_insert_rowid(D) interface usually returns the [rowid] of
	2044	+** the most recent successful [INSERT] into a rowid table or [virtual table]
	2045	+** on database connection D. ^Inserts into [WITHOUT ROWID] tables are not
	2046	+** recorded. ^If no successful [INSERT]s into rowid tables have ever occurred
	2047	+** on the database connection D, then sqlite3_last_insert_rowid(D) returns
	2048	+** zero.
2050	2049	**
2051		-** ^(If an [INSERT] occurs within a trigger or within a [virtual table]
2052		-** method, then this routine will return the [rowid] of the inserted
2053		-** row as long as the trigger or virtual table method is running.
2054		-** But once the trigger or virtual table method ends, the value returned
2055		-** by this routine reverts to what it was before the trigger or virtual
2056		-** table method began.)^
	2050	+** As well as being set automatically as rows are inserted into database
	2051	+** tables, the value returned by this function may be set explicitly by
	2052	+** [sqlite3_set_last_insert_rowid()]
	2053	+**
	2054	+** Some virtual table implementations may INSERT rows into rowid tables as
	2055	+** part of committing a transaction (e.g. to flush data accumulated in memory
	2056	+** to disk). In this case subsequent calls to this function return the rowid
	2057	+** associated with these internal INSERT operations, which leads to
	2058	+** unintuitive results. Virtual table implementations that do write to rowid
	2059	+** tables in this way can avoid this problem by restoring the original
	2060	+** rowid value using [sqlite3_set_last_insert_rowid()] before returning
	2061	+** control to the user.
	2062	+**
	2063	+** ^(If an [INSERT] occurs within a trigger then this routine will
	2064	+** return the [rowid] of the inserted row as long as the trigger is
	2065	+** running. Once the trigger program ends, the value returned
	2066	+** by this routine reverts to what it was before the trigger was fired.)^
2057	2067	**
2058	2068	** ^An [INSERT] that fails due to a constraint violation is not a
2059	2069	** successful [INSERT] and does not change the value returned by this
2060	2070	** routine. ^Thus INSERT OR FAIL, INSERT OR IGNORE, INSERT OR ROLLBACK,
2061	2071	** and INSERT OR ABORT make no changes to the return value of this
		@@ -2078,10 +2088,20 @@
2078	2088	** unpredictable and might not equal either the old or the new
2079	2089	** last insert [rowid].
2080	2090	*/
2081	2091	SQLITE_API sqlite3_int64 sqlite3_last_insert_rowid(sqlite3*);
2082	2092
	2093	+/*
	2094	+** CAPI3REF: Set the Last Insert Rowid value.
	2095	+** METHOD: sqlite3
	2096	+**
	2097	+** The sqlite3_set_last_insert_rowid(D, R) method allows the application to
	2098	+** set the value returned by calling sqlite3_last_insert_rowid(D) to R
	2099	+** without inserting a row into the database.
	2100	+*/
	2101	+SQLITE_API void sqlite3_set_last_insert_rowid(sqlite3*,sqlite3_int64);
	2102	+
2083	2103	/*
2084	2104	** CAPI3REF: Count The Number Of Rows Modified
2085	2105	** METHOD: sqlite3
2086	2106	**
2087	2107	** ^This function returns the number of rows modified, inserted or
2088	2108

	--- src/sqlite3.h
	+++ src/sqlite3.h
	@@ -119,13 +119,13 @@
119	**
120	** See also: [sqlite3_libversion()],
121	** [sqlite3_libversion_number()], [sqlite3_sourceid()],
122	** [sqlite_version()] and [sqlite_source_id()].
123	*/
124	#define SQLITE_VERSION "3.17.0"
125	#define SQLITE_VERSION_NUMBER 3017000
126	#define SQLITE_SOURCE_ID "2017-02-13 16:02:40 ada05cfa86ad7f5645450ac7a2a21c9aa6e57d2c"
127
128	/*
129	** CAPI3REF: Run-Time Library Version Numbers
130	** KEYWORDS: sqlite3_version sqlite3_sourceid
131	**
	@@ -2038,24 +2038,34 @@
2038	** as an undeclared column named ROWID, OID, or _ROWID_ as long as those
2039	** names are not also used by explicitly declared columns. ^If
2040	** the table has a column of type [INTEGER PRIMARY KEY] then that column
2041	** is another alias for the rowid.
2042	**
2043	** ^The sqlite3_last_insert_rowid(D) interface returns the [rowid] of the
2044	** most recent successful [INSERT] into a rowid table or [virtual table]
2045	** on database connection D.
2046	** ^Inserts into [WITHOUT ROWID] tables are not recorded.
2047	** ^If no successful [INSERT]s into rowid tables
2048	** have ever occurred on the database connection D,
2049	** then sqlite3_last_insert_rowid(D) returns zero.
2050	**
2051	** ^(If an [INSERT] occurs within a trigger or within a [virtual table]
2052	** method, then this routine will return the [rowid] of the inserted
2053	** row as long as the trigger or virtual table method is running.
2054	** But once the trigger or virtual table method ends, the value returned
2055	** by this routine reverts to what it was before the trigger or virtual
2056	** table method began.)^











2057	**
2058	** ^An [INSERT] that fails due to a constraint violation is not a
2059	** successful [INSERT] and does not change the value returned by this
2060	** routine. ^Thus INSERT OR FAIL, INSERT OR IGNORE, INSERT OR ROLLBACK,
2061	** and INSERT OR ABORT make no changes to the return value of this
	@@ -2078,10 +2088,20 @@
2078	** unpredictable and might not equal either the old or the new
2079	** last insert [rowid].
2080	*/
2081	SQLITE_API sqlite3_int64 sqlite3_last_insert_rowid(sqlite3*);
2082










2083	/*
2084	** CAPI3REF: Count The Number Of Rows Modified
2085	** METHOD: sqlite3
2086	**
2087	** ^This function returns the number of rows modified, inserted or
2088

	--- src/sqlite3.h
	+++ src/sqlite3.h
	@@ -119,13 +119,13 @@
119	**
120	** See also: [sqlite3_libversion()],
121	** [sqlite3_libversion_number()], [sqlite3_sourceid()],
122	** [sqlite_version()] and [sqlite_source_id()].
123	*/
124	#define SQLITE_VERSION "3.18.0"
125	#define SQLITE_VERSION_NUMBER 3018000
126	#define SQLITE_SOURCE_ID "2017-03-10 17:03:11 f8560c60d10c0365b33342ab05b5a953987b0471"
127
128	/*
129	** CAPI3REF: Run-Time Library Version Numbers
130	** KEYWORDS: sqlite3_version sqlite3_sourceid
131	**
	@@ -2038,24 +2038,34 @@
2038	** as an undeclared column named ROWID, OID, or _ROWID_ as long as those
2039	** names are not also used by explicitly declared columns. ^If
2040	** the table has a column of type [INTEGER PRIMARY KEY] then that column
2041	** is another alias for the rowid.
2042	**
2043	** ^The sqlite3_last_insert_rowid(D) interface usually returns the [rowid] of
2044	** the most recent successful [INSERT] into a rowid table or [virtual table]
2045	** on database connection D. ^Inserts into [WITHOUT ROWID] tables are not
2046	** recorded. ^If no successful [INSERT]s into rowid tables have ever occurred
2047	** on the database connection D, then sqlite3_last_insert_rowid(D) returns
2048	** zero.

2049	**
2050	** As well as being set automatically as rows are inserted into database
2051	** tables, the value returned by this function may be set explicitly by
2052	** [sqlite3_set_last_insert_rowid()]
2053	**
2054	** Some virtual table implementations may INSERT rows into rowid tables as
2055	** part of committing a transaction (e.g. to flush data accumulated in memory
2056	** to disk). In this case subsequent calls to this function return the rowid
2057	** associated with these internal INSERT operations, which leads to
2058	** unintuitive results. Virtual table implementations that do write to rowid
2059	** tables in this way can avoid this problem by restoring the original
2060	** rowid value using [sqlite3_set_last_insert_rowid()] before returning
2061	** control to the user.
2062	**
2063	** ^(If an [INSERT] occurs within a trigger then this routine will
2064	** return the [rowid] of the inserted row as long as the trigger is
2065	** running. Once the trigger program ends, the value returned
2066	** by this routine reverts to what it was before the trigger was fired.)^
2067	**
2068	** ^An [INSERT] that fails due to a constraint violation is not a
2069	** successful [INSERT] and does not change the value returned by this
2070	** routine. ^Thus INSERT OR FAIL, INSERT OR IGNORE, INSERT OR ROLLBACK,
2071	** and INSERT OR ABORT make no changes to the return value of this
	@@ -2078,10 +2088,20 @@
2088	** unpredictable and might not equal either the old or the new
2089	** last insert [rowid].
2090	*/
2091	SQLITE_API sqlite3_int64 sqlite3_last_insert_rowid(sqlite3*);
2092
2093	/*
2094	** CAPI3REF: Set the Last Insert Rowid value.
2095	** METHOD: sqlite3
2096	**
2097	** The sqlite3_set_last_insert_rowid(D, R) method allows the application to
2098	** set the value returned by calling sqlite3_last_insert_rowid(D) to R
2099	** without inserting a row into the database.
2100	*/
2101	SQLITE_API void sqlite3_set_last_insert_rowid(sqlite3*,sqlite3_int64);
2102
2103	/*
2104	** CAPI3REF: Count The Number Of Rows Modified
2105	** METHOD: sqlite3
2106	**
2107	** ^This function returns the number of rows modified, inserted or
2108

M src/stash.c

+35 -32

		--- src/stash.c
		+++ src/stash.c
		@@ -332,52 +332,45 @@
332	332	isBin2 = fIncludeBinary ? 0 : looks_like_binary(&a);
333	333	diff_file_mem(&empty, &a, isBin1, isBin2, zNew, zDiffCmd,
334	334	zBinGlob, fIncludeBinary, diffFlags);
335	335	}else if( isRemoved ){
336	336	fossil_print("DELETE %s\n", zOrig);
337		- if( fBaseline==0 ){
338		- if( file_wd_islink(zOPath) ){
339		- blob_read_link(&a, zOPath);
340		- }else{
341		- blob_read_from_file(&a, zOPath);
342		- }
343		- }else{
344		- content_get(rid, &a);
345		- }
346		- diff_print_index(zNew, diffFlags);
347		- isBin1 = fIncludeBinary ? 0 : looks_like_binary(&a);
348		- isBin2 = 0;
349		- diff_file_mem(&a, &empty, isBin1, isBin2, zOrig, zDiffCmd,
350		- zBinGlob, fIncludeBinary, diffFlags);
351		- }else{
352		- Blob delta, disk;
	337	+ diff_print_index(zNew, diffFlags);
	338	+ isBin2 = 0;
	339	+ if( fBaseline ){
	340	+ content_get(rid, &a);
	341	+ isBin1 = fIncludeBinary ? 0 : looks_like_binary(&a);
	342	+ diff_file_mem(&a, &empty, isBin1, isBin2, zOrig, zDiffCmd,
	343	+ zBinGlob, fIncludeBinary, diffFlags);
	344	+ }else{
	345	+ }
	346	+ }else{
	347	+ Blob delta;
353	348	int isOrigLink = file_wd_islink(zOPath);
354	349	db_ephemeral_blob(&q, 6, &delta);
355		- if( fBaseline==0 ){
356		- if( isOrigLink ){
357		- blob_read_link(&disk, zOPath);
358		- }else{
359		- blob_read_from_file(&disk, zOPath);
360		- }
361		- }
362	350	fossil_print("CHANGED %s\n", zNew);
363	351	if( !isOrigLink != !isLink ){
364	352	diff_print_index(zNew, diffFlags);
365	353	diff_print_filenames(zOrig, zNew, diffFlags);
366	354	printf(DIFF_CANNOT_COMPUTE_SYMLINK);
367	355	}else{
368		- Blob *pBase = fBaseline ? &a : &disk;
369	356	content_get(rid, &a);
370	357	blob_delta_apply(&a, &delta, &b);
371		- isBin1 = fIncludeBinary ? 0 : looks_like_binary(pBase);
	358	+ isBin1 = fIncludeBinary ? 0 : looks_like_binary(&a);
372	359	isBin2 = fIncludeBinary ? 0 : looks_like_binary(&b);
373		- diff_file_mem(fBaseline? &a : &disk, &b, isBin1, isBin2, zNew,
374		- zDiffCmd, zBinGlob, fIncludeBinary, diffFlags);
	360	+ if( fBaseline ){
	361	+ diff_file_mem(&a, &b, isBin1, isBin2, zNew,
	362	+ zDiffCmd, zBinGlob, fIncludeBinary, diffFlags);
	363	+ }else{
	364	+ /*Diff with file on disk using fSwapDiff=1 to show the diff in the
	365	+ same direction as if fBaseline=1.*/
	366	+ diff_file(&b, isBin2, zOPath, zNew, zDiffCmd,
	367	+ zBinGlob, fIncludeBinary, diffFlags, 1);
	368	+ }
375	369	blob_reset(&a);
376	370	blob_reset(&b);
377	371	}
378		- if( !fBaseline ) blob_reset(&disk);
379	372	blob_reset(&delta);
380	373	}
381	374	}
382	375	db_finalize(&q);
383	376	}
		@@ -433,12 +426,15 @@
433	426	**
434	427	** List all changes sets currently stashed. Show information about
435	428	** individual files in each changeset if -v or --verbose is used.
436	429	**
437	430	** fossil stash show\|cat ?STASHID? ?DIFF-OPTIONS?
	431	+** fossil stash gshow\|gcat ?STASHID? ?DIFF-OPTIONS?
438	432	**
439		-** Show the contents of a stash.
	433	+** Show the contents of a stash as a diff against it's baseline.
	434	+** With gshow and gcat, gdiff-command is used instead of internal
	435	+** diff logic.
440	436	**
441	437	** fossil stash pop
442	438	** fossil stash apply ?STASHID?
443	439	**
444	440	** Apply STASHID or the most recently create stash to the current
		@@ -460,18 +456,20 @@
460	456	**
461	457	** fossil stash diff ?STASHID? ?DIFF-OPTIONS?
462	458	** fossil stash gdiff ?STASHID? ?DIFF-OPTIONS?
463	459	**
464	460	** Show diffs of the current working directory and what that
465		-** directory would be if STASHID were applied.
	461	+** directory would be if STASHID were applied. With gdiff,
	462	+** gdiff-command is used instead of internal diff logic.
466	463	**
467	464	** SUMMARY:
468	465	** fossil stash
469	466	** fossil stash save ?-m\|--comment COMMENT? ?FILES...?
470	467	** fossil stash snapshot ?-m\|--comment COMMENT? ?FILES...?
471	468	** fossil stash list\|ls ?-v\|--verbose? ?-W\|--width <num>?
472	469	** fossil stash show\|cat ?STASHID? ?DIFF-OPTIONS?
	470	+** fossil stash gshow\|gcat ?STASHID? ?DIFF-OPTIONS?
473	471	** fossil stash pop
474	472	** fossil stash apply\|goto ?STASHID?
475	473	** fossil stash drop\|rm ?STASHID? ?-a\|--all?
476	474	** fossil stash diff ?STASHID? ?DIFF-OPTIONS?
477	475	** fossil stash gdiff ?STASHID? ?DIFF-OPTIONS?
		@@ -654,25 +652,30 @@
654	652	undo_finish();
655	653	}else
656	654	if( memcmp(zCmd, "diff", nCmd)==0
657	655	\|\| memcmp(zCmd, "gdiff", nCmd)==0
658	656	\|\| memcmp(zCmd, "show", nCmd)==0
	657	+ \|\| memcmp(zCmd, "gshow", nCmd)==0
659	658	\|\| memcmp(zCmd, "cat", nCmd)==0
	659	+ \|\| memcmp(zCmd, "gcat", nCmd)==0
660	660	){
661	661	const char *zDiffCmd = 0;
662	662	const char *zBinGlob = 0;
663	663	int fIncludeBinary = 0;
664		- int fBaseline = zCmd[0]=='s' \|\| zCmd[0]=='c';
	664	+ int fBaseline = 0;
665	665	u64 diffFlags;
666	666
	667	+ if( strstr(zCmd,"show")!=0 \|\| strstr(zCmd,"cat")!=0 ){
	668	+ fBaseline = 1;
	669	+ }
667	670	if( find_option("tk",0,0)!=0 ){
668	671	db_close(0);
669	672	diff_tk(fBaseline ? "stash show" : "stash diff", 3);
670	673	return;
671	674	}
672	675	if( find_option("internal","i",0)==0 ){
673		- zDiffCmd = diff_command_external(memcmp(zCmd, "gdiff", nCmd)==0);
	676	+ zDiffCmd = diff_command_external(zCmd[0]=='g');
674	677	}
675	678	diffFlags = diff_options();
676	679	if( find_option("verbose","v",0)!=0 ) diffFlags \|= DIFF_VERBOSE;
677	680	if( g.argc>4 ) usage(mprintf("%s ?STASHID? ?DIFF-OPTIONS?", zCmd));
678	681	if( zDiffCmd ){
679	682

	--- src/stash.c
	+++ src/stash.c
	@@ -332,52 +332,45 @@
332	isBin2 = fIncludeBinary ? 0 : looks_like_binary(&a);
333	diff_file_mem(&empty, &a, isBin1, isBin2, zNew, zDiffCmd,
334	zBinGlob, fIncludeBinary, diffFlags);
335	}else if( isRemoved ){
336	fossil_print("DELETE %s\n", zOrig);
337	if( fBaseline==0 ){
338	if( file_wd_islink(zOPath) ){
339	blob_read_link(&a, zOPath);
340	}else{
341	blob_read_from_file(&a, zOPath);
342	}
343	}else{
344	content_get(rid, &a);
345	}
346	diff_print_index(zNew, diffFlags);
347	isBin1 = fIncludeBinary ? 0 : looks_like_binary(&a);
348	isBin2 = 0;
349	diff_file_mem(&a, &empty, isBin1, isBin2, zOrig, zDiffCmd,
350	zBinGlob, fIncludeBinary, diffFlags);
351	}else{
352	Blob delta, disk;
353	int isOrigLink = file_wd_islink(zOPath);
354	db_ephemeral_blob(&q, 6, &delta);
355	if( fBaseline==0 ){
356	if( isOrigLink ){
357	blob_read_link(&disk, zOPath);
358	}else{
359	blob_read_from_file(&disk, zOPath);
360	}
361	}
362	fossil_print("CHANGED %s\n", zNew);
363	if( !isOrigLink != !isLink ){
364	diff_print_index(zNew, diffFlags);
365	diff_print_filenames(zOrig, zNew, diffFlags);
366	printf(DIFF_CANNOT_COMPUTE_SYMLINK);
367	}else{
368	Blob *pBase = fBaseline ? &a : &disk;
369	content_get(rid, &a);
370	blob_delta_apply(&a, &delta, &b);
371	isBin1 = fIncludeBinary ? 0 : looks_like_binary(pBase);
372	isBin2 = fIncludeBinary ? 0 : looks_like_binary(&b);
373	diff_file_mem(fBaseline? &a : &disk, &b, isBin1, isBin2, zNew,
374	zDiffCmd, zBinGlob, fIncludeBinary, diffFlags);







375	blob_reset(&a);
376	blob_reset(&b);
377	}
378	if( !fBaseline ) blob_reset(&disk);
379	blob_reset(&delta);
380	}
381	}
382	db_finalize(&q);
383	}
	@@ -433,12 +426,15 @@
433	**
434	** List all changes sets currently stashed. Show information about
435	** individual files in each changeset if -v or --verbose is used.
436	**
437	** fossil stash show\|cat ?STASHID? ?DIFF-OPTIONS?

438	**
439	** Show the contents of a stash.


440	**
441	** fossil stash pop
442	** fossil stash apply ?STASHID?
443	**
444	** Apply STASHID or the most recently create stash to the current
	@@ -460,18 +456,20 @@
460	**
461	** fossil stash diff ?STASHID? ?DIFF-OPTIONS?
462	** fossil stash gdiff ?STASHID? ?DIFF-OPTIONS?
463	**
464	** Show diffs of the current working directory and what that
465	** directory would be if STASHID were applied.

466	**
467	** SUMMARY:
468	** fossil stash
469	** fossil stash save ?-m\|--comment COMMENT? ?FILES...?
470	** fossil stash snapshot ?-m\|--comment COMMENT? ?FILES...?
471	** fossil stash list\|ls ?-v\|--verbose? ?-W\|--width <num>?
472	** fossil stash show\|cat ?STASHID? ?DIFF-OPTIONS?

473	** fossil stash pop
474	** fossil stash apply\|goto ?STASHID?
475	** fossil stash drop\|rm ?STASHID? ?-a\|--all?
476	** fossil stash diff ?STASHID? ?DIFF-OPTIONS?
477	** fossil stash gdiff ?STASHID? ?DIFF-OPTIONS?
	@@ -654,25 +652,30 @@
654	undo_finish();
655	}else
656	if( memcmp(zCmd, "diff", nCmd)==0
657	\|\| memcmp(zCmd, "gdiff", nCmd)==0
658	\|\| memcmp(zCmd, "show", nCmd)==0

659	\|\| memcmp(zCmd, "cat", nCmd)==0

660	){
661	const char *zDiffCmd = 0;
662	const char *zBinGlob = 0;
663	int fIncludeBinary = 0;
664	int fBaseline = zCmd[0]=='s' \|\| zCmd[0]=='c';
665	u64 diffFlags;
666



667	if( find_option("tk",0,0)!=0 ){
668	db_close(0);
669	diff_tk(fBaseline ? "stash show" : "stash diff", 3);
670	return;
671	}
672	if( find_option("internal","i",0)==0 ){
673	zDiffCmd = diff_command_external(memcmp(zCmd, "gdiff", nCmd)==0);
674	}
675	diffFlags = diff_options();
676	if( find_option("verbose","v",0)!=0 ) diffFlags \|= DIFF_VERBOSE;
677	if( g.argc>4 ) usage(mprintf("%s ?STASHID? ?DIFF-OPTIONS?", zCmd));
678	if( zDiffCmd ){
679

	--- src/stash.c
	+++ src/stash.c
	@@ -332,52 +332,45 @@
332	isBin2 = fIncludeBinary ? 0 : looks_like_binary(&a);
333	diff_file_mem(&empty, &a, isBin1, isBin2, zNew, zDiffCmd,
334	zBinGlob, fIncludeBinary, diffFlags);
335	}else if( isRemoved ){
336	fossil_print("DELETE %s\n", zOrig);
337	diff_print_index(zNew, diffFlags);
338	isBin2 = 0;
339	if( fBaseline ){
340	content_get(rid, &a);
341	isBin1 = fIncludeBinary ? 0 : looks_like_binary(&a);
342	diff_file_mem(&a, &empty, isBin1, isBin2, zOrig, zDiffCmd,
343	zBinGlob, fIncludeBinary, diffFlags);
344	}else{
345	}
346	}else{
347	Blob delta;





348	int isOrigLink = file_wd_islink(zOPath);
349	db_ephemeral_blob(&q, 6, &delta);







350	fossil_print("CHANGED %s\n", zNew);
351	if( !isOrigLink != !isLink ){
352	diff_print_index(zNew, diffFlags);
353	diff_print_filenames(zOrig, zNew, diffFlags);
354	printf(DIFF_CANNOT_COMPUTE_SYMLINK);
355	}else{

356	content_get(rid, &a);
357	blob_delta_apply(&a, &delta, &b);
358	isBin1 = fIncludeBinary ? 0 : looks_like_binary(&a);
359	isBin2 = fIncludeBinary ? 0 : looks_like_binary(&b);
360	if( fBaseline ){
361	diff_file_mem(&a, &b, isBin1, isBin2, zNew,
362	zDiffCmd, zBinGlob, fIncludeBinary, diffFlags);
363	}else{
364	/*Diff with file on disk using fSwapDiff=1 to show the diff in the
365	same direction as if fBaseline=1.*/
366	diff_file(&b, isBin2, zOPath, zNew, zDiffCmd,
367	zBinGlob, fIncludeBinary, diffFlags, 1);
368	}
369	blob_reset(&a);
370	blob_reset(&b);
371	}

372	blob_reset(&delta);
373	}
374	}
375	db_finalize(&q);
376	}
	@@ -433,12 +426,15 @@
426	**
427	** List all changes sets currently stashed. Show information about
428	** individual files in each changeset if -v or --verbose is used.
429	**
430	** fossil stash show\|cat ?STASHID? ?DIFF-OPTIONS?
431	** fossil stash gshow\|gcat ?STASHID? ?DIFF-OPTIONS?
432	**
433	** Show the contents of a stash as a diff against it's baseline.
434	** With gshow and gcat, gdiff-command is used instead of internal
435	** diff logic.
436	**
437	** fossil stash pop
438	** fossil stash apply ?STASHID?
439	**
440	** Apply STASHID or the most recently create stash to the current
	@@ -460,18 +456,20 @@
456	**
457	** fossil stash diff ?STASHID? ?DIFF-OPTIONS?
458	** fossil stash gdiff ?STASHID? ?DIFF-OPTIONS?
459	**
460	** Show diffs of the current working directory and what that
461	** directory would be if STASHID were applied. With gdiff,
462	** gdiff-command is used instead of internal diff logic.
463	**
464	** SUMMARY:
465	** fossil stash
466	** fossil stash save ?-m\|--comment COMMENT? ?FILES...?
467	** fossil stash snapshot ?-m\|--comment COMMENT? ?FILES...?
468	** fossil stash list\|ls ?-v\|--verbose? ?-W\|--width <num>?
469	** fossil stash show\|cat ?STASHID? ?DIFF-OPTIONS?
470	** fossil stash gshow\|gcat ?STASHID? ?DIFF-OPTIONS?
471	** fossil stash pop
472	** fossil stash apply\|goto ?STASHID?
473	** fossil stash drop\|rm ?STASHID? ?-a\|--all?
474	** fossil stash diff ?STASHID? ?DIFF-OPTIONS?
475	** fossil stash gdiff ?STASHID? ?DIFF-OPTIONS?
	@@ -654,25 +652,30 @@
652	undo_finish();
653	}else
654	if( memcmp(zCmd, "diff", nCmd)==0
655	\|\| memcmp(zCmd, "gdiff", nCmd)==0
656	\|\| memcmp(zCmd, "show", nCmd)==0
657	\|\| memcmp(zCmd, "gshow", nCmd)==0
658	\|\| memcmp(zCmd, "cat", nCmd)==0
659	\|\| memcmp(zCmd, "gcat", nCmd)==0
660	){
661	const char *zDiffCmd = 0;
662	const char *zBinGlob = 0;
663	int fIncludeBinary = 0;
664	int fBaseline = 0;
665	u64 diffFlags;
666
667	if( strstr(zCmd,"show")!=0 \|\| strstr(zCmd,"cat")!=0 ){
668	fBaseline = 1;
669	}
670	if( find_option("tk",0,0)!=0 ){
671	db_close(0);
672	diff_tk(fBaseline ? "stash show" : "stash diff", 3);
673	return;
674	}
675	if( find_option("internal","i",0)==0 ){
676	zDiffCmd = diff_command_external(zCmd[0]=='g');
677	}
678	diffFlags = diff_options();
679	if( find_option("verbose","v",0)!=0 ) diffFlags \|= DIFF_VERBOSE;
680	if( g.argc>4 ) usage(mprintf("%s ?STASHID? ?DIFF-OPTIONS?", zCmd));
681	if( zDiffCmd ){
682

M src/stat.c

+71 -4

		--- src/stat.c
		+++ src/stat.c
		@@ -76,11 +76,11 @@
76	76	style_submenu_element("URLs", "urllist");
77	77	style_submenu_element("Schema", "repo_schema");
78	78	style_submenu_element("Web-Cache", "cachestat");
79	79	}
80	80	style_submenu_element("Activity Reports", "reports");
81		- style_submenu_element("SHA1 Collisions", "hash-collisions");
	81	+ style_submenu_element("Hash Collisions", "hash-collisions");
82	82	if( sqlite3_compileoption_used("ENABLE_DBSTAT_VTAB") ){
83	83	style_submenu_element("Table Sizes", "repo-tabsize");
84	84	}
85	85	if( g.perm.Admin \|\| g.perm.Setup \|\| db_get_boolean("test_env_enable",0) ){
86	86	style_submenu_element("Environment", "test_env");
		@@ -183,11 +183,16 @@
183	183	@ (%h(RELEASE_VERSION)) <a href='version?verbose=1'>(details)</a>
184	184	@ </td></tr>
185	185	@ <tr><th>SQLite Version:</th><td>%.19s(sqlite3_sourceid())
186	186	@ [%.10s(&sqlite3_sourceid()[20])] (%s(sqlite3_libversion()))
187	187	@ <a href='version?verbose=2'>(details)</a></td></tr>
188		- @ <tr><th>Schema Version:</th><td>%h(g.zAuxSchema)</td></tr>
	188	+ if( g.eHashPolicy!=HPOLICY_AUTO ){
	189	+ @ <tr><th>Schema Version:</th><td>%h(g.zAuxSchema),
	190	+ @ %s(hpolicy_name())</td></tr>
	191	+ }else{
	192	+ @ <tr><th>Schema Version:</th><td>%h(g.zAuxSchema)</td></tr>
	193	+ }
189	194	@ <tr><th>Repository Rebuilt:</th><td>
190	195	@ %h(db_get_mtime("rebuilt","%Y-%m-%d %H:%M:%S","Never"))
191	196	@ By Fossil %h(db_get("rebuilt","Unknown"))</td></tr>
192	197	@ <tr><th>Database Stats:</th><td>
193	198	@ %d(db_int(0, "PRAGMA repository.page_count")) pages,
		@@ -383,10 +388,12 @@
383	388	**
384	389	** Show the repository schema
385	390	*/
386	391	void repo_schema_page(void){
387	392	Stmt q;
	393	+ Blob sql;
	394	+ const char *zArg = P("n");
388	395	login_check_credentials();
389	396	if( !g.perm.Admin ){ login_needed(0); return; }
390	397
391	398	style_header("Repository Schema");
392	399	style_adunit_config(ADUNIT_RIGHT_OK);
		@@ -393,18 +400,78 @@
393	400	style_submenu_element("Stat", "stat");
394	401	style_submenu_element("URLs", "urllist");
395	402	if( sqlite3_compileoption_used("ENABLE_DBSTAT_VTAB") ){
396	403	style_submenu_element("Table Sizes", "repo-tabsize");
397	404	}
398		- db_prepare(&q,
399		- "SELECT sql FROM repository.sqlite_master WHERE sql IS NOT NULL");
	405	+ blob_init(&sql,
	406	+ "SELECT sql FROM repository.sqlite_master WHERE sql IS NOT NULL", -1);
	407	+ if( zArg ){
	408	+ style_submenu_element("All", "repo_schema");
	409	+ blob_appendf(&sql, " AND (tbl_name=%Q OR name=%Q)", zArg, zArg);
	410	+ }
	411	+ blob_appendf(&sql, " ORDER BY tbl_name, type<>'table', name");
	412	+ db_prepare(&q, "%s", blob_str(&sql)/safe-for-%s/);
	413	+ blob_reset(&sql);
400	414	@ <pre>
401	415	while( db_step(&q)==SQLITE_ROW ){
402	416	@ %h(db_column_text(&q, 0));
403	417	}
404	418	@ </pre>
405	419	db_finalize(&q);
	420	+ if( db_table_exists("repository","sqlite_stat1") ){
	421	+ if( zArg ){
	422	+ db_prepare(&q,
	423	+ "SELECT tbl, idx, stat FROM repository.sqlite_stat1"
	424	+ " WHERE tbl LIKE %Q OR idx LIKE %Q"
	425	+ " ORDER BY tbl, idx", zArg, zArg);
	426	+
	427	+ @ <hr>
	428	+ @ <pre>
	429	+ while( db_step(&q)==SQLITE_ROW ){
	430	+ const char *zTab = db_column_text(&q,0);
	431	+ const char *zIdx = db_column_text(&q,1);
	432	+ const char *zStat = db_column_text(&q,2);
	433	+ @ INSERT INTO sqlite_stat1 VALUES('%h(zTab)','%h(zIdx)','%h(zStat)');
	434	+ }
	435	+ @ </pre>
	436	+ db_finalize(&q);
	437	+ }else{
	438	+ style_submenu_element("Stat1","repo_stat1");
	439	+ }
	440	+ }
	441	+ style_footer();
	442	+}
	443	+
	444	+/*
	445	+** WEBPAGE: repo_stat1
	446	+**
	447	+** Show the sqlite_stat1 table for the repository schema
	448	+*/
	449	+void repo_stat1_page(void){
	450	+ login_check_credentials();
	451	+ if( !g.perm.Admin ){ login_needed(0); return; }
	452	+
	453	+ style_header("Repository STAT1 Table");
	454	+ style_adunit_config(ADUNIT_RIGHT_OK);
	455	+ style_submenu_element("Stat", "stat");
	456	+ style_submenu_element("Schema", "repo_schema");
	457	+ if( db_table_exists("repository","sqlite_stat1") ){
	458	+ Stmt q;
	459	+ db_prepare(&q,
	460	+ "SELECT tbl, idx, stat FROM repository.sqlite_stat1"
	461	+ " ORDER BY tbl, idx");
	462	+ @ <pre>
	463	+ while( db_step(&q)==SQLITE_ROW ){
	464	+ const char *zTab = db_column_text(&q,0);
	465	+ const char *zIdx = db_column_text(&q,1);
	466	+ const char *zStat = db_column_text(&q,2);
	467	+ char *zUrl = href("%R/repo_schema?n=%t",zTab);
	468	+ @ INSERT INTO sqlite_stat1 VALUES('%z(zUrl)%h(zTab)</a>','%h(zIdx)','%h(zStat)');
	469	+ }
	470	+ @ </pre>
	471	+ db_finalize(&q);
	472	+ }
406	473	style_footer();
407	474	}
408	475
409	476	/*
410	477	** WEBPAGE: repo-tabsize
411	478

	--- src/stat.c
	+++ src/stat.c
	@@ -76,11 +76,11 @@
76	style_submenu_element("URLs", "urllist");
77	style_submenu_element("Schema", "repo_schema");
78	style_submenu_element("Web-Cache", "cachestat");
79	}
80	style_submenu_element("Activity Reports", "reports");
81	style_submenu_element("SHA1 Collisions", "hash-collisions");
82	if( sqlite3_compileoption_used("ENABLE_DBSTAT_VTAB") ){
83	style_submenu_element("Table Sizes", "repo-tabsize");
84	}
85	if( g.perm.Admin \|\| g.perm.Setup \|\| db_get_boolean("test_env_enable",0) ){
86	style_submenu_element("Environment", "test_env");
	@@ -183,11 +183,16 @@
183	@ (%h(RELEASE_VERSION)) <a href='version?verbose=1'>(details)</a>
184	@ </td></tr>
185	@ <tr><th>SQLite Version:</th><td>%.19s(sqlite3_sourceid())
186	@ [%.10s(&sqlite3_sourceid()[20])] (%s(sqlite3_libversion()))
187	@ <a href='version?verbose=2'>(details)</a></td></tr>
188	@ <tr><th>Schema Version:</th><td>%h(g.zAuxSchema)</td></tr>





189	@ <tr><th>Repository Rebuilt:</th><td>
190	@ %h(db_get_mtime("rebuilt","%Y-%m-%d %H:%M:%S","Never"))
191	@ By Fossil %h(db_get("rebuilt","Unknown"))</td></tr>
192	@ <tr><th>Database Stats:</th><td>
193	@ %d(db_int(0, "PRAGMA repository.page_count")) pages,
	@@ -383,10 +388,12 @@
383	**
384	** Show the repository schema
385	*/
386	void repo_schema_page(void){
387	Stmt q;


388	login_check_credentials();
389	if( !g.perm.Admin ){ login_needed(0); return; }
390
391	style_header("Repository Schema");
392	style_adunit_config(ADUNIT_RIGHT_OK);
	@@ -393,18 +400,78 @@
393	style_submenu_element("Stat", "stat");
394	style_submenu_element("URLs", "urllist");
395	if( sqlite3_compileoption_used("ENABLE_DBSTAT_VTAB") ){
396	style_submenu_element("Table Sizes", "repo-tabsize");
397	}
398	db_prepare(&q,
399	"SELECT sql FROM repository.sqlite_master WHERE sql IS NOT NULL");







400	@ <pre>
401	while( db_step(&q)==SQLITE_ROW ){
402	@ %h(db_column_text(&q, 0));
403	}
404	@ </pre>
405	db_finalize(&q);





















































406	style_footer();
407	}
408
409	/*
410	** WEBPAGE: repo-tabsize
411

	--- src/stat.c
	+++ src/stat.c
	@@ -76,11 +76,11 @@
76	style_submenu_element("URLs", "urllist");
77	style_submenu_element("Schema", "repo_schema");
78	style_submenu_element("Web-Cache", "cachestat");
79	}
80	style_submenu_element("Activity Reports", "reports");
81	style_submenu_element("Hash Collisions", "hash-collisions");
82	if( sqlite3_compileoption_used("ENABLE_DBSTAT_VTAB") ){
83	style_submenu_element("Table Sizes", "repo-tabsize");
84	}
85	if( g.perm.Admin \|\| g.perm.Setup \|\| db_get_boolean("test_env_enable",0) ){
86	style_submenu_element("Environment", "test_env");
	@@ -183,11 +183,16 @@
183	@ (%h(RELEASE_VERSION)) <a href='version?verbose=1'>(details)</a>
184	@ </td></tr>
185	@ <tr><th>SQLite Version:</th><td>%.19s(sqlite3_sourceid())
186	@ [%.10s(&sqlite3_sourceid()[20])] (%s(sqlite3_libversion()))
187	@ <a href='version?verbose=2'>(details)</a></td></tr>
188	if( g.eHashPolicy!=HPOLICY_AUTO ){
189	@ <tr><th>Schema Version:</th><td>%h(g.zAuxSchema),
190	@ %s(hpolicy_name())</td></tr>
191	}else{
192	@ <tr><th>Schema Version:</th><td>%h(g.zAuxSchema)</td></tr>
193	}
194	@ <tr><th>Repository Rebuilt:</th><td>
195	@ %h(db_get_mtime("rebuilt","%Y-%m-%d %H:%M:%S","Never"))
196	@ By Fossil %h(db_get("rebuilt","Unknown"))</td></tr>
197	@ <tr><th>Database Stats:</th><td>
198	@ %d(db_int(0, "PRAGMA repository.page_count")) pages,
	@@ -383,10 +388,12 @@
388	**
389	** Show the repository schema
390	*/
391	void repo_schema_page(void){
392	Stmt q;
393	Blob sql;
394	const char *zArg = P("n");
395	login_check_credentials();
396	if( !g.perm.Admin ){ login_needed(0); return; }
397
398	style_header("Repository Schema");
399	style_adunit_config(ADUNIT_RIGHT_OK);
	@@ -393,18 +400,78 @@
400	style_submenu_element("Stat", "stat");
401	style_submenu_element("URLs", "urllist");
402	if( sqlite3_compileoption_used("ENABLE_DBSTAT_VTAB") ){
403	style_submenu_element("Table Sizes", "repo-tabsize");
404	}
405	blob_init(&sql,
406	"SELECT sql FROM repository.sqlite_master WHERE sql IS NOT NULL", -1);
407	if( zArg ){
408	style_submenu_element("All", "repo_schema");
409	blob_appendf(&sql, " AND (tbl_name=%Q OR name=%Q)", zArg, zArg);
410	}
411	blob_appendf(&sql, " ORDER BY tbl_name, type<>'table', name");
412	db_prepare(&q, "%s", blob_str(&sql)/safe-for-%s/);
413	blob_reset(&sql);
414	@ <pre>
415	while( db_step(&q)==SQLITE_ROW ){
416	@ %h(db_column_text(&q, 0));
417	}
418	@ </pre>
419	db_finalize(&q);
420	if( db_table_exists("repository","sqlite_stat1") ){
421	if( zArg ){
422	db_prepare(&q,
423	"SELECT tbl, idx, stat FROM repository.sqlite_stat1"
424	" WHERE tbl LIKE %Q OR idx LIKE %Q"
425	" ORDER BY tbl, idx", zArg, zArg);
426
427	@ <hr>
428	@ <pre>
429	while( db_step(&q)==SQLITE_ROW ){
430	const char *zTab = db_column_text(&q,0);
431	const char *zIdx = db_column_text(&q,1);
432	const char *zStat = db_column_text(&q,2);
433	@ INSERT INTO sqlite_stat1 VALUES('%h(zTab)','%h(zIdx)','%h(zStat)');
434	}
435	@ </pre>
436	db_finalize(&q);
437	}else{
438	style_submenu_element("Stat1","repo_stat1");
439	}
440	}
441	style_footer();
442	}
443
444	/*
445	** WEBPAGE: repo_stat1
446	**
447	** Show the sqlite_stat1 table for the repository schema
448	*/
449	void repo_stat1_page(void){
450	login_check_credentials();
451	if( !g.perm.Admin ){ login_needed(0); return; }
452
453	style_header("Repository STAT1 Table");
454	style_adunit_config(ADUNIT_RIGHT_OK);
455	style_submenu_element("Stat", "stat");
456	style_submenu_element("Schema", "repo_schema");
457	if( db_table_exists("repository","sqlite_stat1") ){
458	Stmt q;
459	db_prepare(&q,
460	"SELECT tbl, idx, stat FROM repository.sqlite_stat1"
461	" ORDER BY tbl, idx");
462	@ <pre>
463	while( db_step(&q)==SQLITE_ROW ){
464	const char *zTab = db_column_text(&q,0);
465	const char *zIdx = db_column_text(&q,1);
466	const char *zStat = db_column_text(&q,2);
467	char *zUrl = href("%R/repo_schema?n=%t",zTab);
468	@ INSERT INTO sqlite_stat1 VALUES('%z(zUrl)%h(zTab)</a>','%h(zIdx)','%h(zStat)');
469	}
470	@ </pre>
471	db_finalize(&q);
472	}
473	style_footer();
474	}
475
476	/*
477	** WEBPAGE: repo-tabsize
478

M src/tag.c

+1 -1

		--- src/tag.c
		+++ src/tag.c
		@@ -627,11 +627,11 @@
627	627	blob_init(&value, 0, 0);
628	628	for(i=3; i<g.argc; i++){
629	629	int pid = name_to_typed_rid(g.argv[i], "ci");
630	630	if( i>3 ) blob_append(&value, " ", 1);
631	631	zUuid = rid_to_uuid(pid);
632		- blob_append(&value, zUuid, UUID_SIZE);
	632	+ blob_append(&value, zUuid, strlen(zUuid));
633	633	fossil_free(zUuid);
634	634	}
635	635	if( bTest && !dryRun ){
636	636	tag_insert("parent", 1, blob_str(&value), -1, 0.0, rid);
637	637	}else{
638	638

	--- src/tag.c
	+++ src/tag.c
	@@ -627,11 +627,11 @@
627	blob_init(&value, 0, 0);
628	for(i=3; i<g.argc; i++){
629	int pid = name_to_typed_rid(g.argv[i], "ci");
630	if( i>3 ) blob_append(&value, " ", 1);
631	zUuid = rid_to_uuid(pid);
632	blob_append(&value, zUuid, UUID_SIZE);
633	fossil_free(zUuid);
634	}
635	if( bTest && !dryRun ){
636	tag_insert("parent", 1, blob_str(&value), -1, 0.0, rid);
637	}else{
638

	--- src/tag.c
	+++ src/tag.c
	@@ -627,11 +627,11 @@
627	blob_init(&value, 0, 0);
628	for(i=3; i<g.argc; i++){
629	int pid = name_to_typed_rid(g.argv[i], "ci");
630	if( i>3 ) blob_append(&value, " ", 1);
631	zUuid = rid_to_uuid(pid);
632	blob_append(&value, zUuid, strlen(zUuid));
633	fossil_free(zUuid);
634	}
635	if( bTest && !dryRun ){
636	tag_insert("parent", 1, blob_str(&value), -1, 0.0, rid);
637	}else{
638

M src/tar.c

+2 -6

		--- src/tar.c
		+++ src/tar.c
		@@ -483,11 +483,11 @@
483	483	content_get(rid, &mfile);
484	484	if( blob_size(&mfile)==0 ){
485	485	blob_zero(pTar);
486	486	return;
487	487	}
488		- blob_zero(&hash);
	488	+ blob_set_dynamic(&hash, rid_to_uuid(rid));
489	489	blob_zero(&filename);
490	490
491	491	if( zDir && zDir[0] ){
492	492	blob_appendf(&filename, "%s/", zDir);
493	493	}
		@@ -520,13 +520,10 @@
520	520	if( eflg & (MFESTFLG_RAW\|MFESTFLG_UUID) ){
521	521	if( eflg & MFESTFLG_RAW ){
522	522	blob_append(&filename, "manifest", -1);
523	523	zName = blob_str(&filename);
524	524	}
525		- if( eflg & MFESTFLG_UUID ){
526		- sha1sum_blob(&mfile, &hash);
527		- }
528	525	if( eflg & MFESTFLG_RAW ) {
529	526	sterilize_manifest(&mfile);
530	527	tar_add_file(zName, &mfile, 0, mTime);
531	528	}
532	529	}
		@@ -535,11 +532,10 @@
535	532	blob_append(&hash, "\n", 1);
536	533	blob_resize(&filename, nPrefix);
537	534	blob_append(&filename, "manifest.uuid", -1);
538	535	zName = blob_str(&filename);
539	536	tar_add_file(zName, &hash, 0, mTime);
540		- blob_reset(&hash);
541	537	}
542	538	if( eflg & MFESTFLG_TAGS ){
543	539	Blob tagslist;
544	540	blob_zero(&tagslist);
545	541	get_checkin_taglist(rid, &tagslist);
		@@ -564,19 +560,19 @@
564	560	tar_add_file(zName, &file, manifest_file_mperm(pFile), mTime);
565	561	blob_reset(&file);
566	562	}
567	563	}
568	564	}else{
569		- sha1sum_blob(&mfile, &hash);
570	565	blob_append(&filename, blob_str(&hash), 16);
571	566	zName = blob_str(&filename);
572	567	mTime = db_int64(0, "SELECT (julianday('now') - 2440587.5)*86400.0;");
573	568	tar_begin(mTime);
574	569	tar_add_file(zName, &mfile, 0, mTime);
575	570	}
576	571	manifest_destroy(pManifest);
577	572	blob_reset(&mfile);
	573	+ blob_reset(&hash);
578	574	blob_reset(&filename);
579	575	tar_finish(pTar);
580	576	}
581	577
582	578	/*
583	579

	--- src/tar.c
	+++ src/tar.c
	@@ -483,11 +483,11 @@
483	content_get(rid, &mfile);
484	if( blob_size(&mfile)==0 ){
485	blob_zero(pTar);
486	return;
487	}
488	blob_zero(&hash);
489	blob_zero(&filename);
490
491	if( zDir && zDir[0] ){
492	blob_appendf(&filename, "%s/", zDir);
493	}
	@@ -520,13 +520,10 @@
520	if( eflg & (MFESTFLG_RAW\|MFESTFLG_UUID) ){
521	if( eflg & MFESTFLG_RAW ){
522	blob_append(&filename, "manifest", -1);
523	zName = blob_str(&filename);
524	}
525	if( eflg & MFESTFLG_UUID ){
526	sha1sum_blob(&mfile, &hash);
527	}
528	if( eflg & MFESTFLG_RAW ) {
529	sterilize_manifest(&mfile);
530	tar_add_file(zName, &mfile, 0, mTime);
531	}
532	}
	@@ -535,11 +532,10 @@
535	blob_append(&hash, "\n", 1);
536	blob_resize(&filename, nPrefix);
537	blob_append(&filename, "manifest.uuid", -1);
538	zName = blob_str(&filename);
539	tar_add_file(zName, &hash, 0, mTime);
540	blob_reset(&hash);
541	}
542	if( eflg & MFESTFLG_TAGS ){
543	Blob tagslist;
544	blob_zero(&tagslist);
545	get_checkin_taglist(rid, &tagslist);
	@@ -564,19 +560,19 @@
564	tar_add_file(zName, &file, manifest_file_mperm(pFile), mTime);
565	blob_reset(&file);
566	}
567	}
568	}else{
569	sha1sum_blob(&mfile, &hash);
570	blob_append(&filename, blob_str(&hash), 16);
571	zName = blob_str(&filename);
572	mTime = db_int64(0, "SELECT (julianday('now') - 2440587.5)*86400.0;");
573	tar_begin(mTime);
574	tar_add_file(zName, &mfile, 0, mTime);
575	}
576	manifest_destroy(pManifest);
577	blob_reset(&mfile);

578	blob_reset(&filename);
579	tar_finish(pTar);
580	}
581
582	/*
583

	--- src/tar.c
	+++ src/tar.c
	@@ -483,11 +483,11 @@
483	content_get(rid, &mfile);
484	if( blob_size(&mfile)==0 ){
485	blob_zero(pTar);
486	return;
487	}
488	blob_set_dynamic(&hash, rid_to_uuid(rid));
489	blob_zero(&filename);
490
491	if( zDir && zDir[0] ){
492	blob_appendf(&filename, "%s/", zDir);
493	}
	@@ -520,13 +520,10 @@
520	if( eflg & (MFESTFLG_RAW\|MFESTFLG_UUID) ){
521	if( eflg & MFESTFLG_RAW ){
522	blob_append(&filename, "manifest", -1);
523	zName = blob_str(&filename);
524	}



525	if( eflg & MFESTFLG_RAW ) {
526	sterilize_manifest(&mfile);
527	tar_add_file(zName, &mfile, 0, mTime);
528	}
529	}
	@@ -535,11 +532,10 @@
532	blob_append(&hash, "\n", 1);
533	blob_resize(&filename, nPrefix);
534	blob_append(&filename, "manifest.uuid", -1);
535	zName = blob_str(&filename);
536	tar_add_file(zName, &hash, 0, mTime);

537	}
538	if( eflg & MFESTFLG_TAGS ){
539	Blob tagslist;
540	blob_zero(&tagslist);
541	get_checkin_taglist(rid, &tagslist);
	@@ -564,19 +560,19 @@
560	tar_add_file(zName, &file, manifest_file_mperm(pFile), mTime);
561	blob_reset(&file);
562	}
563	}
564	}else{

565	blob_append(&filename, blob_str(&hash), 16);
566	zName = blob_str(&filename);
567	mTime = db_int64(0, "SELECT (julianday('now') - 2440587.5)*86400.0;");
568	tar_begin(mTime);
569	tar_add_file(zName, &mfile, 0, mTime);
570	}
571	manifest_destroy(pManifest);
572	blob_reset(&mfile);
573	blob_reset(&hash);
574	blob_reset(&filename);
575	tar_finish(pTar);
576	}
577
578	/*
579

M src/timeline.c

+2 -2

		--- src/timeline.c
		+++ src/timeline.c
		@@ -723,11 +723,11 @@
723	723	** mi: "merge-in". An array of integer rail positions from which
724	724	** merge arrows should be drawn into this node. If the value is
725	725	** negative, then the rail position is the absolute value of mi[]
726	726	** and a thin merge-arrow descender is drawn to the bottom of
727	727	** the screen.
728		- ** h: The SHA1 hash of the object being graphed
	728	+ ** h: The artifact hash of the object being graphed
729	729	*/
730	730	cgi_printf("var rowinfo = [\n");
731	731	for(pRow=pGraph->pFirst; pRow; pRow=pRow->pNext){
732	732	cgi_printf("{id:%d,bg:\"%s\",r:%d,d:%d,mo:%d,mu:%d,u:%d,f:%d,au:",
733	733	pRow->idx, /* id */
		@@ -1461,11 +1461,11 @@
1461	1461	** v Show details of files changed
1462	1462	** f=CHECKIN Show family (immediate parents and children) of CHECKIN
1463	1463	** from=CHECKIN Path from...
1464	1464	** to=CHECKIN ... to this
1465	1465	** shortest ... show only the shortest path
1466		-** uf=FILE_SHA1 Show only check-ins that contain the given file version
	1466	+** uf=FILE_HASH Show only check-ins that contain the given file version
1467	1467	** chng=GLOBLIST Show only check-ins that involve changes to a file whose
1468	1468	** name matches one of the comma-separate GLOBLIST
1469	1469	** brbg Background color from branch name
1470	1470	** ubg Background color from user
1471	1471	** namechng Show only check-ins that have filename changes
1472	1472

	--- src/timeline.c
	+++ src/timeline.c
	@@ -723,11 +723,11 @@
723	** mi: "merge-in". An array of integer rail positions from which
724	** merge arrows should be drawn into this node. If the value is
725	** negative, then the rail position is the absolute value of mi[]
726	** and a thin merge-arrow descender is drawn to the bottom of
727	** the screen.
728	** h: The SHA1 hash of the object being graphed
729	*/
730	cgi_printf("var rowinfo = [\n");
731	for(pRow=pGraph->pFirst; pRow; pRow=pRow->pNext){
732	cgi_printf("{id:%d,bg:\"%s\",r:%d,d:%d,mo:%d,mu:%d,u:%d,f:%d,au:",
733	pRow->idx, /* id */
	@@ -1461,11 +1461,11 @@
1461	** v Show details of files changed
1462	** f=CHECKIN Show family (immediate parents and children) of CHECKIN
1463	** from=CHECKIN Path from...
1464	** to=CHECKIN ... to this
1465	** shortest ... show only the shortest path
1466	** uf=FILE_SHA1 Show only check-ins that contain the given file version
1467	** chng=GLOBLIST Show only check-ins that involve changes to a file whose
1468	** name matches one of the comma-separate GLOBLIST
1469	** brbg Background color from branch name
1470	** ubg Background color from user
1471	** namechng Show only check-ins that have filename changes
1472

	--- src/timeline.c
	+++ src/timeline.c
	@@ -723,11 +723,11 @@
723	** mi: "merge-in". An array of integer rail positions from which
724	** merge arrows should be drawn into this node. If the value is
725	** negative, then the rail position is the absolute value of mi[]
726	** and a thin merge-arrow descender is drawn to the bottom of
727	** the screen.
728	** h: The artifact hash of the object being graphed
729	*/
730	cgi_printf("var rowinfo = [\n");
731	for(pRow=pGraph->pFirst; pRow; pRow=pRow->pNext){
732	cgi_printf("{id:%d,bg:\"%s\",r:%d,d:%d,mo:%d,mu:%d,u:%d,f:%d,au:",
733	pRow->idx, /* id */
	@@ -1461,11 +1461,11 @@
1461	** v Show details of files changed
1462	** f=CHECKIN Show family (immediate parents and children) of CHECKIN
1463	** from=CHECKIN Path from...
1464	** to=CHECKIN ... to this
1465	** shortest ... show only the shortest path
1466	** uf=FILE_HASH Show only check-ins that contain the given file version
1467	** chng=GLOBLIST Show only check-ins that involve changes to a file whose
1468	** name matches one of the comma-separate GLOBLIST
1469	** brbg Background color from branch name
1470	** ubg Background color from user
1471	** namechng Show only check-ins that have filename changes
1472

M src/unversioned.c

+58 -1

		--- src/unversioned.c
		+++ src/unversioned.c
		@@ -456,11 +456,11 @@
456	456	** Query parameters:
457	457	**
458	458	** byage=1 Order the initial display be decreasing age
459	459	** showdel=0 Show deleted files
460	460	*/
461		-void uvstat_page(void){
	461	+void uvlist_page(void){
462	462	Stmt q;
463	463	sqlite3_int64 iNow;
464	464	sqlite3_int64 iTotalSz = 0;
465	465	int cnt = 0;
466	466	int n = 0;
		@@ -554,5 +554,62 @@
554	554	}else{
555	555	@ No unversioned files on this server.
556	556	}
557	557	style_footer();
558	558	}
	559	+
	560	+/*
	561	+** WEBPAGE: juvlist
	562	+**
	563	+** Return a complete list of unversioned files as JSON. The JSON
	564	+** looks like this:
	565	+**
	566	+** [{"name":NAME,
	567	+** "mtime":MTIME,
	568	+** "hash":HASH,
	569	+** "size":SIZE,
	570	+** "user":USER}]
	571	+*/
	572	+void uvlist_json_page(void){
	573	+ Stmt q;
	574	+ char *zSep = "[";
	575	+ Blob json;
	576	+
	577	+ login_check_credentials();
	578	+ if( !g.perm.Read ){ login_needed(g.anon.Read); return; }
	579	+ cgi_set_content_type("text/json");
	580	+ if( !db_table_exists("repository","unversioned") ){
	581	+ blob_init(&json, "[]", -1);
	582	+ cgi_set_content(&json);
	583	+ return;
	584	+ }
	585	+ blob_init(&json, 0, 0);
	586	+ db_prepare(&q,
	587	+ "SELECT"
	588	+ " name,"
	589	+ " mtime,"
	590	+ " hash,"
	591	+ " sz,"
	592	+ " (SELECT login FROM rcvfrom, user"
	593	+ " WHERE user.uid=rcvfrom.uid AND rcvfrom.rcvid=unversioned.rcvid)"
	594	+ " FROM unversioned WHERE hash IS NOT NULL"
	595	+ );
	596	+ while( db_step(&q)==SQLITE_ROW ){
	597	+ const char *zName = db_column_text(&q, 0);
	598	+ sqlite3_int64 mtime = db_column_int(&q, 1);
	599	+ const char *zHash = db_column_text(&q, 2);
	600	+ int fullSize = db_column_int(&q, 3);
	601	+ const char *zLogin = db_column_text(&q, 4);
	602	+ if( zLogin==0 ) zLogin = "";
	603	+ blob_appendf(&json, "%s{\"name\":\"", zSep);
	604	+ zSep = ",\n ";
	605	+ blob_append_json_string(&json, zName);
	606	+ blob_appendf(&json, "\",\n \"mtime\":%lld,\n \"hash\":\"", mtime);
	607	+ blob_append_json_string(&json, zHash);
	608	+ blob_appendf(&json, "\",\n \"size\":%d,\n \"user\":\"", fullSize);
	609	+ blob_append_json_string(&json, zLogin);
	610	+ blob_appendf(&json, "\"}");
	611	+ }
	612	+ db_finalize(&q);
	613	+ blob_appendf(&json,"]\n");
	614	+ cgi_set_content(&json);
	615	+}
559	616

	--- src/unversioned.c
	+++ src/unversioned.c
	@@ -456,11 +456,11 @@
456	** Query parameters:
457	**
458	** byage=1 Order the initial display be decreasing age
459	** showdel=0 Show deleted files
460	*/
461	void uvstat_page(void){
462	Stmt q;
463	sqlite3_int64 iNow;
464	sqlite3_int64 iTotalSz = 0;
465	int cnt = 0;
466	int n = 0;
	@@ -554,5 +554,62 @@
554	}else{
555	@ No unversioned files on this server.
556	}
557	style_footer();
558	}

























































559

	--- src/unversioned.c
	+++ src/unversioned.c
	@@ -456,11 +456,11 @@
456	** Query parameters:
457	**
458	** byage=1 Order the initial display be decreasing age
459	** showdel=0 Show deleted files
460	*/
461	void uvlist_page(void){
462	Stmt q;
463	sqlite3_int64 iNow;
464	sqlite3_int64 iTotalSz = 0;
465	int cnt = 0;
466	int n = 0;
	@@ -554,5 +554,62 @@
554	}else{
555	@ No unversioned files on this server.
556	}
557	style_footer();
558	}
559
560	/*
561	** WEBPAGE: juvlist
562	**
563	** Return a complete list of unversioned files as JSON. The JSON
564	** looks like this:
565	**
566	** [{"name":NAME,
567	** "mtime":MTIME,
568	** "hash":HASH,
569	** "size":SIZE,
570	** "user":USER}]
571	*/
572	void uvlist_json_page(void){
573	Stmt q;
574	char *zSep = "[";
575	Blob json;
576
577	login_check_credentials();
578	if( !g.perm.Read ){ login_needed(g.anon.Read); return; }
579	cgi_set_content_type("text/json");
580	if( !db_table_exists("repository","unversioned") ){
581	blob_init(&json, "[]", -1);
582	cgi_set_content(&json);
583	return;
584	}
585	blob_init(&json, 0, 0);
586	db_prepare(&q,
587	"SELECT"
588	" name,"
589	" mtime,"
590	" hash,"
591	" sz,"
592	" (SELECT login FROM rcvfrom, user"
593	" WHERE user.uid=rcvfrom.uid AND rcvfrom.rcvid=unversioned.rcvid)"
594	" FROM unversioned WHERE hash IS NOT NULL"
595	);
596	while( db_step(&q)==SQLITE_ROW ){
597	const char *zName = db_column_text(&q, 0);
598	sqlite3_int64 mtime = db_column_int(&q, 1);
599	const char *zHash = db_column_text(&q, 2);
600	int fullSize = db_column_int(&q, 3);
601	const char *zLogin = db_column_text(&q, 4);
602	if( zLogin==0 ) zLogin = "";
603	blob_appendf(&json, "%s{\"name\":\"", zSep);
604	zSep = ",\n ";
605	blob_append_json_string(&json, zName);
606	blob_appendf(&json, "\",\n \"mtime\":%lld,\n \"hash\":\"", mtime);
607	blob_append_json_string(&json, zHash);
608	blob_appendf(&json, "\",\n \"size\":%d,\n \"user\":\"", fullSize);
609	blob_append_json_string(&json, zLogin);
610	blob_appendf(&json, "\"}");
611	}
612	db_finalize(&q);
613	blob_appendf(&json,"]\n");
614	cgi_set_content(&json);
615	}
616

M src/verify.c

+6 -8

		--- src/verify.c
		+++ src/verify.c
		@@ -34,27 +34,25 @@
34	34	**
35	35	** Panic if anything goes wrong. If this procedure returns it means
36	36	** that everything is OK.
37	37	*/
38	38	static void verify_rid(int rid){
39		- Blob uuid, hash, content;
	39	+ Blob uuid, content;
40	40	if( content_size(rid, 0)<0 ){
41	41	return; /* No way to verify phantoms */
42	42	}
43	43	blob_zero(&uuid);
44	44	db_blob(&uuid, "SELECT uuid FROM blob WHERE rid=%d", rid);
45		- if( blob_size(&uuid)!=UUID_SIZE ){
	45	+ if( !hname_validate(blob_buffer(&uuid), blob_size(&uuid)) ){
46	46	fossil_fatal("not a valid rid: %d", rid);
47	47	}
48	48	if( content_get(rid, &content) ){
49		- sha1sum_blob(&content, &hash);
50		- blob_reset(&content);
51		- if( blob_compare(&uuid, &hash) ){
52		- fossil_fatal("hash of rid %d (%b) does not match its uuid (%b)",
53		- rid, &hash, &uuid);
	49	+ if( !hname_verify_hash(&content, blob_buffer(&uuid), blob_size(&uuid)) ){
	50	+ fossil_fatal("hash of rid %d does not match its uuid (%b)",
	51	+ rid, &uuid);
54	52	}
55		- blob_reset(&hash);
	53	+ blob_reset(&content);
56	54	}
57	55	blob_reset(&uuid);
58	56	}
59	57
60	58	/*
61	59

	--- src/verify.c
	+++ src/verify.c
	@@ -34,27 +34,25 @@
34	**
35	** Panic if anything goes wrong. If this procedure returns it means
36	** that everything is OK.
37	*/
38	static void verify_rid(int rid){
39	Blob uuid, hash, content;
40	if( content_size(rid, 0)<0 ){
41	return; /* No way to verify phantoms */
42	}
43	blob_zero(&uuid);
44	db_blob(&uuid, "SELECT uuid FROM blob WHERE rid=%d", rid);
45	if( blob_size(&uuid)!=UUID_SIZE ){
46	fossil_fatal("not a valid rid: %d", rid);
47	}
48	if( content_get(rid, &content) ){
49	sha1sum_blob(&content, &hash);
50	blob_reset(&content);
51	if( blob_compare(&uuid, &hash) ){
52	fossil_fatal("hash of rid %d (%b) does not match its uuid (%b)",
53	rid, &hash, &uuid);
54	}
55	blob_reset(&hash);
56	}
57	blob_reset(&uuid);
58	}
59
60	/*
61

	--- src/verify.c
	+++ src/verify.c
	@@ -34,27 +34,25 @@
34	**
35	** Panic if anything goes wrong. If this procedure returns it means
36	** that everything is OK.
37	*/
38	static void verify_rid(int rid){
39	Blob uuid, content;
40	if( content_size(rid, 0)<0 ){
41	return; /* No way to verify phantoms */
42	}
43	blob_zero(&uuid);
44	db_blob(&uuid, "SELECT uuid FROM blob WHERE rid=%d", rid);
45	if( !hname_validate(blob_buffer(&uuid), blob_size(&uuid)) ){
46	fossil_fatal("not a valid rid: %d", rid);
47	}
48	if( content_get(rid, &content) ){
49	if( !hname_verify_hash(&content, blob_buffer(&uuid), blob_size(&uuid)) ){
50	fossil_fatal("hash of rid %d does not match its uuid (%b)",
51	rid, &uuid);


52	}
53	blob_reset(&content);
54	}
55	blob_reset(&uuid);
56	}
57
58	/*
59

M src/vfile.c

+19 -30

		--- src/vfile.c
		+++ src/vfile.c
		@@ -21,12 +21,12 @@
21	21	#include "vfile.h"
22	22	#include <assert.h>
23	23	#include <sys/types.h>
24	24
25	25	/*
26		-** The input is guaranteed to be a 40-character well-formed UUID.
27		-** Find its rid.
	26	+** The input is guaranteed to be a 40- or 64-character well-formed
	27	+** artifact hash. Find its rid.
28	28	*/
29	29	int fast_uuid_to_rid(const char *zUuid){
30	30	static Stmt q;
31	31	int rid;
32	32	db_static_prepare(&q, "SELECT rid FROM blob WHERE uuid=:uuid");
		@@ -51,17 +51,17 @@
51	51	** create a phantom record. A private phantom is created for 2 and
52	52	** a public phantom is created for 1.
53	53	*/
54	54	int uuid_to_rid(const char *zUuid, int phantomize){
55	55	int rid, sz;
56		- char z[UUID_SIZE+1];
	56	+ char z[HNAME_MAX+1];
57	57
58	58	sz = strlen(zUuid);
59		- if( sz!=UUID_SIZE \|\| !validate16(zUuid, sz) ){
60		- return 0;
	59	+ if( !hname_validate(zUuid, sz) ){
	60	+ return 0; /* Not a valid hash */
61	61	}
62		- memcpy(z, zUuid, UUID_SIZE+1);
	62	+ memcpy(z, zUuid, sz+1);
63	63	canonical16(z, sz);
64	64	rid = fast_uuid_to_rid(z);
65	65	if( rid==0 && phantomize ){
66	66	rid = content_new(zUuid, phantomize-1);
67	67	}
		@@ -130,11 +130,11 @@
130	130	/*
131	131	** The cksigFlags parameter to vfile_check_signature() is an OR-ed
132	132	** combination of the following bits:
133	133	*/
134	134	#define CKSIG_ENOTFILE 0x001 /* non-file FS objects throw an error */
135		-#define CKSIG_SHA1 0x002 /* Verify file content using sha1sum */
	135	+#define CKSIG_HASH 0x002 /* Verify file content using hashing */
136	136	#define CKSIG_SETMTIME 0x004 /* Set mtime to last check-out time */
137	137
138	138	#endif /* INTERFACE */
139	139
140	140	/*
		@@ -159,12 +159,12 @@
159	159	**
160	160	** If the size of the file has changed, then we always know that the file
161	161	** changed without having to look at the mtime or on-disk content.
162	162	**
163	163	** The mtime of the file is only a factor if the mtime-changes setting
164		-** is false and the useSha1sum flag is false. If the mtime-changes
165		-** setting is true (or undefined - it defaults to true) or if useSha1sum
	164	+** is false and the CKSIG_HASH flag is false. If the mtime-changes
	165	+** setting is true (or undefined - it defaults to true) or if CKSIG_HASH
166	166	** is true, then we do not trust the mtime and will examine the on-disk
167	167	** content to determine if a file really is the same.
168	168	**
169	169	** If the mtime is used, it is used only to determine if files are the same.
170	170	** If the mtime of a file has changed, we still examine the on-disk content
		@@ -171,12 +171,11 @@
171	171	** to see whether or not the edit was a null-edit.
172	172	*/
173	173	void vfile_check_signature(int vid, unsigned int cksigFlags){
174	174	int nErr = 0;
175	175	Stmt q;
176		- Blob fileCksum, origCksum;
177		- int useMtime = (cksigFlags & CKSIG_SHA1)==0
	176	+ int useMtime = (cksigFlags & CKSIG_HASH)==0
178	177	&& db_get_boolean("mtime-changes", 1);
179	178
180	179	db_begin_transaction();
181	180	db_prepare(&q, "SELECT id, %Q \|\| pathname,"
182	181	" vfile.mrid, deleted, chnged, uuid, size, mtime,"
		@@ -222,40 +221,30 @@
222	221	chnged = 1;
223	222	}
224	223	if( origSize!=currentSize ){
225	224	if( chnged!=1 ){
226	225	/* A file size change is definitive - the file has changed. No
227		- ** need to check the mtime or sha1sum */
	226	+ ** need to check the mtime or hash */
228	227	chnged = 1;
229	228	}
230	229	}else if( chnged==1 && rid!=0 && !isDeleted ){
231	230	/* File is believed to have changed but it is the same size.
232	231	** Double check that it really has changed by looking at content. */
	232	+ const char *zUuid = db_column_text(&q, 5);
	233	+ int nUuid = db_column_bytes(&q, 5);
233	234	assert( origSize==currentSize );
234		- db_ephemeral_blob(&q, 5, &origCksum);
235		- if( sha1sum_file(zName, &fileCksum) ){
236		- blob_zero(&fileCksum);
237		- }
238		- if( blob_compare(&fileCksum, &origCksum)==0 ) chnged = 0;
239		- blob_reset(&origCksum);
240		- blob_reset(&fileCksum);
	235	+ if( hname_verify_file_hash(zName, zUuid, nUuid) ) chnged = 0;
241	236	}else if( (chnged==0 \|\| chnged==2 \|\| chnged==4)
242	237	&& (useMtime==0 \|\| currentMtime!=oldMtime) ){
243	238	/* For files that were formerly believed to be unchanged or that were
244	239	** changed by merging, if their mtime changes, or unconditionally
245		- ** if --sha1sum is used, check to see if they have been edited by
246		- ** looking at their SHA1 sum */
	240	+ ** if --hash is used, check to see if they have been edited by
	241	+ ** looking at their artifact hashes */
	242	+ const char *zUuid = db_column_text(&q, 5);
	243	+ int nUuid = db_column_bytes(&q, 5);
247	244	assert( origSize==currentSize );
248		- db_ephemeral_blob(&q, 5, &origCksum);
249		- if( sha1sum_file(zName, &fileCksum) ){
250		- blob_zero(&fileCksum);
251		- }
252		- if( blob_compare(&fileCksum, &origCksum) ){
253		- chnged = 1;
254		- }
255		- blob_reset(&origCksum);
256		- blob_reset(&fileCksum);
	245	+ if( !hname_verify_file_hash(zName, zUuid, nUuid) ) chnged = 1;
257	246	}
258	247	if( (cksigFlags & CKSIG_SETMTIME) && (chnged==0 \|\| chnged==2 \|\| chnged==4) ){
259	248	i64 desiredMtime;
260	249	if( mtime_of_manifest_file(vid,rid,&desiredMtime)==0 ){
261	250	if( currentMtime!=desiredMtime ){
262	251

	--- src/vfile.c
	+++ src/vfile.c
	@@ -21,12 +21,12 @@
21	#include "vfile.h"
22	#include <assert.h>
23	#include <sys/types.h>
24
25	/*
26	** The input is guaranteed to be a 40-character well-formed UUID.
27	** Find its rid.
28	*/
29	int fast_uuid_to_rid(const char *zUuid){
30	static Stmt q;
31	int rid;
32	db_static_prepare(&q, "SELECT rid FROM blob WHERE uuid=:uuid");
	@@ -51,17 +51,17 @@
51	** create a phantom record. A private phantom is created for 2 and
52	** a public phantom is created for 1.
53	*/
54	int uuid_to_rid(const char *zUuid, int phantomize){
55	int rid, sz;
56	char z[UUID_SIZE+1];
57
58	sz = strlen(zUuid);
59	if( sz!=UUID_SIZE \|\| !validate16(zUuid, sz) ){
60	return 0;
61	}
62	memcpy(z, zUuid, UUID_SIZE+1);
63	canonical16(z, sz);
64	rid = fast_uuid_to_rid(z);
65	if( rid==0 && phantomize ){
66	rid = content_new(zUuid, phantomize-1);
67	}
	@@ -130,11 +130,11 @@
130	/*
131	** The cksigFlags parameter to vfile_check_signature() is an OR-ed
132	** combination of the following bits:
133	*/
134	#define CKSIG_ENOTFILE 0x001 /* non-file FS objects throw an error */
135	#define CKSIG_SHA1 0x002 /* Verify file content using sha1sum */
136	#define CKSIG_SETMTIME 0x004 /* Set mtime to last check-out time */
137
138	#endif /* INTERFACE */
139
140	/*
	@@ -159,12 +159,12 @@
159	**
160	** If the size of the file has changed, then we always know that the file
161	** changed without having to look at the mtime or on-disk content.
162	**
163	** The mtime of the file is only a factor if the mtime-changes setting
164	** is false and the useSha1sum flag is false. If the mtime-changes
165	** setting is true (or undefined - it defaults to true) or if useSha1sum
166	** is true, then we do not trust the mtime and will examine the on-disk
167	** content to determine if a file really is the same.
168	**
169	** If the mtime is used, it is used only to determine if files are the same.
170	** If the mtime of a file has changed, we still examine the on-disk content
	@@ -171,12 +171,11 @@
171	** to see whether or not the edit was a null-edit.
172	*/
173	void vfile_check_signature(int vid, unsigned int cksigFlags){
174	int nErr = 0;
175	Stmt q;
176	Blob fileCksum, origCksum;
177	int useMtime = (cksigFlags & CKSIG_SHA1)==0
178	&& db_get_boolean("mtime-changes", 1);
179
180	db_begin_transaction();
181	db_prepare(&q, "SELECT id, %Q \|\| pathname,"
182	" vfile.mrid, deleted, chnged, uuid, size, mtime,"
	@@ -222,40 +221,30 @@
222	chnged = 1;
223	}
224	if( origSize!=currentSize ){
225	if( chnged!=1 ){
226	/* A file size change is definitive - the file has changed. No
227	** need to check the mtime or sha1sum */
228	chnged = 1;
229	}
230	}else if( chnged==1 && rid!=0 && !isDeleted ){
231	/* File is believed to have changed but it is the same size.
232	** Double check that it really has changed by looking at content. */


233	assert( origSize==currentSize );
234	db_ephemeral_blob(&q, 5, &origCksum);
235	if( sha1sum_file(zName, &fileCksum) ){
236	blob_zero(&fileCksum);
237	}
238	if( blob_compare(&fileCksum, &origCksum)==0 ) chnged = 0;
239	blob_reset(&origCksum);
240	blob_reset(&fileCksum);
241	}else if( (chnged==0 \|\| chnged==2 \|\| chnged==4)
242	&& (useMtime==0 \|\| currentMtime!=oldMtime) ){
243	/* For files that were formerly believed to be unchanged or that were
244	** changed by merging, if their mtime changes, or unconditionally
245	** if --sha1sum is used, check to see if they have been edited by
246	** looking at their SHA1 sum */


247	assert( origSize==currentSize );
248	db_ephemeral_blob(&q, 5, &origCksum);
249	if( sha1sum_file(zName, &fileCksum) ){
250	blob_zero(&fileCksum);
251	}
252	if( blob_compare(&fileCksum, &origCksum) ){
253	chnged = 1;
254	}
255	blob_reset(&origCksum);
256	blob_reset(&fileCksum);
257	}
258	if( (cksigFlags & CKSIG_SETMTIME) && (chnged==0 \|\| chnged==2 \|\| chnged==4) ){
259	i64 desiredMtime;
260	if( mtime_of_manifest_file(vid,rid,&desiredMtime)==0 ){
261	if( currentMtime!=desiredMtime ){
262

	--- src/vfile.c
	+++ src/vfile.c
	@@ -21,12 +21,12 @@
21	#include "vfile.h"
22	#include <assert.h>
23	#include <sys/types.h>
24
25	/*
26	** The input is guaranteed to be a 40- or 64-character well-formed
27	** artifact hash. Find its rid.
28	*/
29	int fast_uuid_to_rid(const char *zUuid){
30	static Stmt q;
31	int rid;
32	db_static_prepare(&q, "SELECT rid FROM blob WHERE uuid=:uuid");
	@@ -51,17 +51,17 @@
51	** create a phantom record. A private phantom is created for 2 and
52	** a public phantom is created for 1.
53	*/
54	int uuid_to_rid(const char *zUuid, int phantomize){
55	int rid, sz;
56	char z[HNAME_MAX+1];
57
58	sz = strlen(zUuid);
59	if( !hname_validate(zUuid, sz) ){
60	return 0; /* Not a valid hash */
61	}
62	memcpy(z, zUuid, sz+1);
63	canonical16(z, sz);
64	rid = fast_uuid_to_rid(z);
65	if( rid==0 && phantomize ){
66	rid = content_new(zUuid, phantomize-1);
67	}
	@@ -130,11 +130,11 @@
130	/*
131	** The cksigFlags parameter to vfile_check_signature() is an OR-ed
132	** combination of the following bits:
133	*/
134	#define CKSIG_ENOTFILE 0x001 /* non-file FS objects throw an error */
135	#define CKSIG_HASH 0x002 /* Verify file content using hashing */
136	#define CKSIG_SETMTIME 0x004 /* Set mtime to last check-out time */
137
138	#endif /* INTERFACE */
139
140	/*
	@@ -159,12 +159,12 @@
159	**
160	** If the size of the file has changed, then we always know that the file
161	** changed without having to look at the mtime or on-disk content.
162	**
163	** The mtime of the file is only a factor if the mtime-changes setting
164	** is false and the CKSIG_HASH flag is false. If the mtime-changes
165	** setting is true (or undefined - it defaults to true) or if CKSIG_HASH
166	** is true, then we do not trust the mtime and will examine the on-disk
167	** content to determine if a file really is the same.
168	**
169	** If the mtime is used, it is used only to determine if files are the same.
170	** If the mtime of a file has changed, we still examine the on-disk content
	@@ -171,12 +171,11 @@
171	** to see whether or not the edit was a null-edit.
172	*/
173	void vfile_check_signature(int vid, unsigned int cksigFlags){
174	int nErr = 0;
175	Stmt q;
176	int useMtime = (cksigFlags & CKSIG_HASH)==0

177	&& db_get_boolean("mtime-changes", 1);
178
179	db_begin_transaction();
180	db_prepare(&q, "SELECT id, %Q \|\| pathname,"
181	" vfile.mrid, deleted, chnged, uuid, size, mtime,"
	@@ -222,40 +221,30 @@
221	chnged = 1;
222	}
223	if( origSize!=currentSize ){
224	if( chnged!=1 ){
225	/* A file size change is definitive - the file has changed. No
226	** need to check the mtime or hash */
227	chnged = 1;
228	}
229	}else if( chnged==1 && rid!=0 && !isDeleted ){
230	/* File is believed to have changed but it is the same size.
231	** Double check that it really has changed by looking at content. */
232	const char *zUuid = db_column_text(&q, 5);
233	int nUuid = db_column_bytes(&q, 5);
234	assert( origSize==currentSize );
235	if( hname_verify_file_hash(zName, zUuid, nUuid) ) chnged = 0;






236	}else if( (chnged==0 \|\| chnged==2 \|\| chnged==4)
237	&& (useMtime==0 \|\| currentMtime!=oldMtime) ){
238	/* For files that were formerly believed to be unchanged or that were
239	** changed by merging, if their mtime changes, or unconditionally
240	** if --hash is used, check to see if they have been edited by
241	** looking at their artifact hashes */
242	const char *zUuid = db_column_text(&q, 5);
243	int nUuid = db_column_bytes(&q, 5);
244	assert( origSize==currentSize );
245	if( !hname_verify_file_hash(zName, zUuid, nUuid) ) chnged = 1;








246	}
247	if( (cksigFlags & CKSIG_SETMTIME) && (chnged==0 \|\| chnged==2 \|\| chnged==4) ){
248	i64 desiredMtime;
249	if( mtime_of_manifest_file(vid,rid,&desiredMtime)==0 ){
250	if( currentMtime!=desiredMtime ){
251

M src/wiki.c

+2 -2

		--- src/wiki.c
		+++ src/wiki.c
		@@ -1122,12 +1122,12 @@
1122	1122	*/
1123	1123	int wiki_technote_to_rid(const char *zETime) {
1124	1124	int rid=0; /* Artifact ID of the tech note */
1125	1125	int nETime = strlen(zETime);
1126	1126	Stmt q;
1127		- if( nETime>=4 && nETime<=UUID_SIZE && validate16(zETime, nETime) ){
1128		- char zUuid[UUID_SIZE+1];
	1127	+ if( nETime>=4 && nETime<=HNAME_MAX && validate16(zETime, nETime) ){
	1128	+ char zUuid[HNAME_MAX+1];
1129	1129	memcpy(zUuid, zETime, nETime+1);
1130	1130	canonical16(zUuid, nETime);
1131	1131	db_prepare(&q,
1132	1132	"SELECT e.objid"
1133	1133	" FROM event e, tag t"
1134	1134

	--- src/wiki.c
	+++ src/wiki.c
	@@ -1122,12 +1122,12 @@
1122	*/
1123	int wiki_technote_to_rid(const char *zETime) {
1124	int rid=0; /* Artifact ID of the tech note */
1125	int nETime = strlen(zETime);
1126	Stmt q;
1127	if( nETime>=4 && nETime<=UUID_SIZE && validate16(zETime, nETime) ){
1128	char zUuid[UUID_SIZE+1];
1129	memcpy(zUuid, zETime, nETime+1);
1130	canonical16(zUuid, nETime);
1131	db_prepare(&q,
1132	"SELECT e.objid"
1133	" FROM event e, tag t"
1134

	--- src/wiki.c
	+++ src/wiki.c
	@@ -1122,12 +1122,12 @@
1122	*/
1123	int wiki_technote_to_rid(const char *zETime) {
1124	int rid=0; /* Artifact ID of the tech note */
1125	int nETime = strlen(zETime);
1126	Stmt q;
1127	if( nETime>=4 && nETime<=HNAME_MAX && validate16(zETime, nETime) ){
1128	char zUuid[HNAME_MAX+1];
1129	memcpy(zUuid, zETime, nETime+1);
1130	canonical16(zUuid, nETime);
1131	db_prepare(&q,
1132	"SELECT e.objid"
1133	" FROM event e, tag t"
1134

M src/wikiformat.c

+1 -1

		--- src/wikiformat.c
		+++ src/wikiformat.c
		@@ -1067,11 +1067,11 @@
1067	1067	** If the input string corresponds to an existing baseline,
1068	1068	** return true.
1069	1069	*/
1070	1070	static int is_valid_uuid(const char *z){
1071	1071	int n = strlen(z);
1072		- if( n<4 \|\| n>UUID_SIZE ) return 0;
	1072	+ if( n<4 \|\| n>HNAME_MAX ) return 0;
1073	1073	if( !validate16(z, n) ) return 0;
1074	1074	return 1;
1075	1075	}
1076	1076
1077	1077	/*
1078	1078

	--- src/wikiformat.c
	+++ src/wikiformat.c
	@@ -1067,11 +1067,11 @@
1067	** If the input string corresponds to an existing baseline,
1068	** return true.
1069	*/
1070	static int is_valid_uuid(const char *z){
1071	int n = strlen(z);
1072	if( n<4 \|\| n>UUID_SIZE ) return 0;
1073	if( !validate16(z, n) ) return 0;
1074	return 1;
1075	}
1076
1077	/*
1078

	--- src/wikiformat.c
	+++ src/wikiformat.c
	@@ -1067,11 +1067,11 @@
1067	** If the input string corresponds to an existing baseline,
1068	** return true.
1069	*/
1070	static int is_valid_uuid(const char *z){
1071	int n = strlen(z);
1072	if( n<4 \|\| n>HNAME_MAX ) return 0;
1073	if( !validate16(z, n) ) return 0;
1074	return 1;
1075	}
1076
1077	/*
1078

M src/xfer.c

+96 -66

		--- src/xfer.c
		+++ src/xfer.c
		@@ -49,16 +49,17 @@
49	49	int nDanglingFile; /* Number of dangling deltas received */
50	50	int mxSend; /* Stop sending "file" when pOut reaches this size */
51	51	int resync; /* Send igot cards for all holdings */
52	52	u8 syncPrivate; /* True to enable syncing private content */
53	53	u8 nextIsPrivate; /* If true, next "file" received is a private */
	54	+ u32 clientVersion; /* Version of the client software */
54	55	time_t maxTime; /* Time when this transfer should be finished */
55	56	};
56	57
57	58
58	59	/*
59		-** The input blob contains a UUID. Convert it into a record ID.
	60	+** The input blob contains an artifact. Convert it into a record ID.
60	61	** Create a phantom record if no prior record exists and
61	62	** phantomize is true.
62	63	**
63	64	** Compare to uuid_to_rid(). This routine takes a blob argument
64	65	** and does less error checking.
		@@ -100,12 +101,12 @@
100	101	** message. This routine finishes parsing that message and does
101	102	** a record insert of the file.
102	103	**
103	104	** The file line is in one of the following two forms:
104	105	**
105		-** file UUID SIZE \n CONTENT
106		-** file UUID DELTASRC SIZE \n CONTENT
	106	+** file HASH SIZE \n CONTENT
	107	+** file HASH DELTASRC SIZE \n CONTENT
107	108	**
108	109	** The content is SIZE bytes immediately following the newline.
109	110	** If DELTASRC exists, then the CONTENT is a delta against the
110	111	** content of DELTASRC.
111	112	**
		@@ -122,29 +123,30 @@
122	123	int *pnUuidList
123	124	){
124	125	int n;
125	126	int rid;
126	127	int srcid = 0;
127		- Blob content, hash;
	128	+ Blob content;
128	129	int isPriv;
	130	+ Blob *pUuid;
129	131
130	132	isPriv = pXfer->nextIsPrivate;
131	133	pXfer->nextIsPrivate = 0;
132	134	if( pXfer->nToken<3
133	135	\|\| pXfer->nToken>4
134		- \|\| !blob_is_uuid(&pXfer->aToken[1])
	136	+ \|\| !blob_is_hname(&pXfer->aToken[1])
135	137	\|\| !blob_is_int(&pXfer->aToken[pXfer->nToken-1], &n)
136	138	\|\| n<0
137		- \|\| (pXfer->nToken==4 && !blob_is_uuid(&pXfer->aToken[2]))
	139	+ \|\| (pXfer->nToken==4 && !blob_is_hname(&pXfer->aToken[2]))
138	140	){
139	141	blob_appendf(&pXfer->err, "malformed file line");
140	142	return;
141	143	}
142	144	blob_zero(&content);
143		- blob_zero(&hash);
144	145	blob_extract(pXfer->pIn, n, &content);
145		- if( !cloneFlag && uuid_is_shunned(blob_str(&pXfer->aToken[1])) ){
	146	+ pUuid = &pXfer->aToken[1];
	147	+ if( !cloneFlag && uuid_is_shunned(blob_str(pUuid)) ){
146	148	/* Ignore files that have been shunned */
147	149	blob_reset(&content);
148	150	return;
149	151	}
150	152	if( isPriv && !g.perm.Private ){
		@@ -158,26 +160,26 @@
158	160	pXfer->nDeltaRcvd++;
159	161	}else{
160	162	srcid = 0;
161	163	pXfer->nFileRcvd++;
162	164	}
163		- rid = content_put_ex(&content, blob_str(&pXfer->aToken[1]), srcid,
	165	+ rid = content_put_ex(&content, blob_str(pUuid), srcid,
164	166	0, isPriv);
165		- Th_AppendToList(pzUuidList, pnUuidList, blob_str(&pXfer->aToken[1]),
166		- blob_size(&pXfer->aToken[1]));
	167	+ Th_AppendToList(pzUuidList, pnUuidList, blob_str(pUuid),
	168	+ blob_size(pUuid));
167	169	remote_has(rid);
168	170	blob_reset(&content);
169	171	return;
170	172	}
171	173	if( pXfer->nToken==4 ){
172	174	Blob src, next;
173	175	srcid = rid_from_uuid(&pXfer->aToken[2], 1, isPriv);
174	176	if( content_get(srcid, &src)==0 ){
175		- rid = content_put_ex(&content, blob_str(&pXfer->aToken[1]), srcid,
	177	+ rid = content_put_ex(&content, blob_str(pUuid), srcid,
176	178	0, isPriv);
177		- Th_AppendToList(pzUuidList, pnUuidList, blob_str(&pXfer->aToken[1]),
178		- blob_size(&pXfer->aToken[1]));
	179	+ Th_AppendToList(pzUuidList, pnUuidList, blob_str(pUuid),
	180	+ blob_size(pUuid));
179	181	pXfer->nDanglingFile++;
180	182	db_multi_exec("DELETE FROM phantom WHERE rid=%d", rid);
181	183	if( !isPriv ) content_make_public(rid);
182	184	blob_reset(&src);
183	185	blob_reset(&content);
		@@ -189,19 +191,15 @@
189	191	blob_reset(&content);
190	192	content = next;
191	193	}else{
192	194	pXfer->nFileRcvd++;
193	195	}
194		- sha1sum_blob(&content, &hash);
195		- if( !blob_eq_str(&pXfer->aToken[1], blob_str(&hash), -1) ){
196		- blob_appendf(&pXfer->err,
197		- "wrong hash on received artifact: expected %s but got %s",
198		- blob_str(&pXfer->aToken[1]), blob_str(&hash));
199		- }
200		- rid = content_put_ex(&content, blob_str(&hash), 0, 0, isPriv);
201		- Th_AppendToList(pzUuidList, pnUuidList, blob_str(&hash), blob_size(&hash));
202		- blob_reset(&hash);
	196	+ if( hname_verify_hash(&content, blob_buffer(pUuid), blob_size(pUuid))==0 ){
	197	+ blob_appendf(&pXfer->err, "wrong hash on received artifact: %b", pUuid);
	198	+ }
	199	+ rid = content_put_ex(&content, blob_str(pUuid), 0, 0, isPriv);
	200	+ Th_AppendToList(pzUuidList, pnUuidList, blob_str(pUuid), blob_size(pUuid));
203	201	if( rid==0 ){
204	202	blob_appendf(&pXfer->err, "%s", g.zErrMsg);
205	203	blob_reset(&content);
206	204	}else{
207	205	if( !isPriv ) content_make_public(rid);
		@@ -217,18 +215,18 @@
217	215	** a record insert of the file. The difference between "file" and
218	216	** "cfile" is that with "cfile" the content is already compressed.
219	217	**
220	218	** The file line is in one of the following two forms:
221	219	**
222		-** cfile UUID USIZE CSIZE \n CONTENT
223		-** cfile UUID DELTASRC USIZE CSIZE \n CONTENT
	220	+** cfile HASH USIZE CSIZE \n CONTENT
	221	+** cfile HASH DELTASRC USIZE CSIZE \n CONTENT
224	222	**
225	223	** The content is CSIZE bytes immediately following the newline.
226	224	** If DELTASRC exists, then the CONTENT is a delta against the
227	225	** content of DELTASRC.
228	226	**
229		-** The original size of the UUID artifact is USIZE.
	227	+** The original size of the HASH artifact is USIZE.
230	228	**
231	229	** If any error occurs, write a message into pErr which has already
232	230	** be initialized to an empty string.
233	231	**
234	232	** Any artifact successfully received by this routine is considered to
		@@ -248,15 +246,15 @@
248	246
249	247	isPriv = pXfer->nextIsPrivate;
250	248	pXfer->nextIsPrivate = 0;
251	249	if( pXfer->nToken<4
252	250	\|\| pXfer->nToken>5
253		- \|\| !blob_is_uuid(&pXfer->aToken[1])
	251	+ \|\| !blob_is_hname(&pXfer->aToken[1])
254	252	\|\| !blob_is_int(&pXfer->aToken[pXfer->nToken-2], &szU)
255	253	\|\| !blob_is_int(&pXfer->aToken[pXfer->nToken-1], &szC)
256	254	\|\| szC<0 \|\| szU<0
257		- \|\| (pXfer->nToken==5 && !blob_is_uuid(&pXfer->aToken[2]))
	255	+ \|\| (pXfer->nToken==5 && !blob_is_hname(&pXfer->aToken[2]))
258	256	){
259	257	blob_appendf(&pXfer->err, "malformed cfile line");
260	258	return;
261	259	}
262	260	if( isPriv && !g.perm.Private ){
		@@ -297,11 +295,11 @@
297	295	**
298	296	** If the 0x0001 bit of FLAGS is set, that means the file has been
299	297	** deleted, SIZE is zero, the HASH is "-", and the "\n CONTENT" is omitted.
300	298	**
301	299	** SIZE is the number of bytes of CONTENT. The CONTENT is uncompressed.
302		-** HASH is the SHA1 hash of CONTENT.
	300	+** HASH is the artifact hash of CONTENT.
303	301	**
304	302	** If the 0x0004 bit of FLAGS is set, that means the CONTENT is omitted.
305	303	** The sender might have omitted the content because it is too big to
306	304	** transmit, or because it is unchanged and this record exists purely
307	305	** to update the MTIME.
		@@ -310,11 +308,10 @@
310	308	sqlite3_int64 mtime; /* The MTIME */
311	309	Blob pHash; / The HASH value */
312	310	int sz; /* The SIZE */
313	311	int flags; /* The FLAGS */
314	312	Blob content; /* The CONTENT */
315		- Blob hash; /* Hash computed from CONTENT to compare with HASH */
316	313	Blob x; /* Compressed content */
317	314	Stmt q; /* SQL statements for comparison and insert */
318	315	int isDelete; /* HASH is "-" indicating this is a delete */
319	316	int nullContent; /* True of CONTENT is NULL */
320	317	int iStatus; /* Result from unversioned_status() */
		@@ -321,25 +318,23 @@
321	318
322	319	pHash = &pXfer->aToken[3];
323	320	if( pXfer->nToken==5
324	321	\|\| !blob_is_filename(&pXfer->aToken[1])
325	322	\|\| !blob_is_int64(&pXfer->aToken[2], &mtime)
326		- \|\| (!blob_eq(pHash,"-") && !blob_is_uuid(pHash))
	323	+ \|\| (!blob_eq(pHash,"-") && !blob_is_hname(pHash))
327	324	\|\| !blob_is_int(&pXfer->aToken[4], &sz)
328	325	\|\| !blob_is_int(&pXfer->aToken[5], &flags)
329	326	){
330	327	blob_appendf(&pXfer->err, "malformed uvfile line");
331	328	return;
332	329	}
333	330	blob_init(&content, 0, 0);
334		- blob_init(&hash, 0, 0);
335	331	blob_init(&x, 0, 0);
336	332	if( sz>0 && (flags & 0x0005)==0 ){
337	333	blob_extract(pXfer->pIn, sz, &content);
338	334	nullContent = 0;
339		- sha1sum_blob(&content, &hash);
340		- if( blob_compare(&hash, pHash)!=0 ){
	335	+ if( hname_verify_hash(&content, blob_buffer(pHash), blob_size(pHash))==0 ){
341	336	blob_appendf(&pXfer->err, "in uvfile line, HASH does not match CONTENT");
342	337	goto end_accept_unversioned_file;
343	338	}
344	339	}else{
345	340	nullContent = 1;
		@@ -399,11 +394,10 @@
399	394	db_unset("uv-hash", 0);
400	395
401	396	end_accept_unversioned_file:
402	397	blob_reset(&x);
403	398	blob_reset(&content);
404		- blob_reset(&hash);
405	399	}
406	400
407	401	/*
408	402	** Try to send a file as a delta against its parent.
409	403	** If successful, return the number of bytes in the delta.
		@@ -415,11 +409,11 @@
415	409	static int send_delta_parent(
416	410	Xfer pXfer, / The transfer context */
417	411	int rid, /* record id of the file to send */
418	412	int isPrivate, /* True if rid is a private artifact */
419	413	Blob pContent, / The content of the file to send */
420		- Blob pUuid / The UUID of the file to send */
	414	+ Blob pUuid / The HASH of the file to send */
421	415	){
422	416	static const char *const azQuery[] = {
423	417	"SELECT pid FROM plink x"
424	418	" WHERE cid=%d"
425	419	" AND NOT EXISTS(SELECT 1 FROM phantom WHERE rid=pid)",
		@@ -469,11 +463,11 @@
469	463	*/
470	464	static int send_delta_native(
471	465	Xfer pXfer, / The transfer context */
472	466	int rid, /* record id of the file to send */
473	467	int isPrivate, /* True if rid is a private artifact */
474		- Blob pUuid / The UUID of the file to send */
	468	+ Blob pUuid / The HASH of the file to send */
475	469	){
476	470	Blob src, delta;
477	471	int size = 0;
478	472	int srcId;
479	473
		@@ -500,15 +494,26 @@
500	494	}else{
501	495	size = 0;
502	496	}
503	497	return size;
504	498	}
	499	+
	500	+/*
	501	+** Push an error message to alert the older client that the repository
	502	+** has SHA3 content and cannot be synced or cloned.
	503	+*/
	504	+static void xfer_cannot_send_sha3_error(Xfer *pXfer){
	505	+ blob_appendf(pXfer->pOut,
	506	+ "error Fossil\\sversion\\s2.0\\sor\\slater\\srequired.\n"
	507	+ );
	508	+}
	509	+
505	510
506	511	/*
507	512	** Send the file identified by rid.
508	513	**
509		-** The pUuid can be NULL in which case the correct UUID is computed
	514	+** The pUuid can be NULL in which case the correct hash is computed
510	515	** from the rid.
511	516	**
512	517	** Try to send the file as a native delta if nativeDelta is true, or
513	518	** as a parent delta if nativeDelta is false.
514	519	**
		@@ -527,10 +532,14 @@
527	532	}
528	533	blob_zero(&uuid);
529	534	db_blob(&uuid, "SELECT uuid FROM blob WHERE rid=%d AND size>=0", rid);
530	535	if( blob_size(&uuid)==0 ){
531	536	return;
	537	+ }
	538	+ if( blob_size(&uuid)>HNAME_LEN_SHA1 && pXfer->clientVersion<20000 ){
	539	+ xfer_cannot_send_sha3_error(pXfer);
	540	+ return;
532	541	}
533	542	if( pUuid ){
534	543	if( blob_compare(pUuid, &uuid)!=0 ){
535	544	blob_reset(&uuid);
536	545	return;
		@@ -617,10 +626,15 @@
617	626	szC = db_column_bytes(&q1, 2);
618	627	zContent = db_column_raw(&q1, 2);
619	628	srcIsPrivate = db_column_int(&q1, 3);
620	629	zDelta = db_column_text(&q1, 4);
621	630	if( isPrivate ) blob_append(pXfer->pOut, "private\n", -1);
	631	+ if( pXfer->clientVersion<20000 && db_column_bytes(&q1,0)!=HNAME_LEN_SHA1 ){
	632	+ xfer_cannot_send_sha3_error(pXfer);
	633	+ db_reset(&q1);
	634	+ return;
	635	+ }
622	636	blob_appendf(pXfer->pOut, "cfile %s ", zUuid);
623	637	if( !isPrivate && srcIsPrivate ){
624	638	content_get(rid, &fullContent);
625	639	szU = blob_size(&fullContent);
626	640	blob_compress(&fullContent, &fullContent);
		@@ -676,10 +690,15 @@
676	690	);
677	691	}
678	692	if( db_step(&q1)==SQLITE_ROW ){
679	693	sqlite3_int64 mtime = db_column_int64(&q1, 0);
680	694	const char *zHash = db_column_text(&q1, 1);
	695	+ if( pXfer->clientVersion<20000 && db_column_bytes(&q1,1)>HNAME_LEN_SHA1 ){
	696	+ xfer_cannot_send_sha3_error(pXfer);
	697	+ db_reset(&q1);
	698	+ return;
	699	+ }
681	700	if( blob_size(pXfer->pOut)>=pXfer->mxSend ){
682	701	/* If we have already reached the send size limit, send a (short)
683	702	** uvigot card rather than a uvfile card. This only happens on the
684	703	** server side. The uvigot card will provoke the client to resend
685	704	** another uvgimme on the next cycle. */
		@@ -728,23 +747,23 @@
728	747	}
729	748	db_finalize(&q);
730	749	}
731	750
732	751	/*
733		-** Compute an SHA1 hash on the tail of pMsg. Verify that it matches the
	752	+** Compute an hash on the tail of pMsg. Verify that it matches the
734	753	** the hash given in pHash. Return non-zero for an error and 0 on success.
	754	+**
	755	+** The type of hash computed (SHA1, SHA3-256) is determined by
	756	+** the length of the input hash in pHash.
735	757	*/
736	758	static int check_tail_hash(Blob pHash, Blob pMsg){
737	759	Blob tail;
738		- Blob h2;
739	760	int rc;
740	761	blob_tail(pMsg, &tail);
741		- sha1sum_blob(&tail, &h2);
742		- rc = blob_compare(pHash, &h2);
743		- blob_reset(&h2);
	762	+ rc = hname_verify_hash(&tail, blob_buffer(pHash), blob_size(pHash));
744	763	blob_reset(&tail);
745		- return rc;
	764	+ return rc==HNAME_ERROR;
746	765	}
747	766
748	767	/*
749	768	** Check the signature on an application/x-fossil payload received by
750	769	** the HTTP server. The signature is a line of the following form:
		@@ -1198,12 +1217,12 @@
1198	1217	while( blob_line(xfer.pIn, &xfer.line) ){
1199	1218	if( blob_buffer(&xfer.line)[0]=='#' ) continue;
1200	1219	if( blob_size(&xfer.line)==0 ) continue;
1201	1220	xfer.nToken = blob_tokenize(&xfer.line, xfer.aToken, count(xfer.aToken));
1202	1221
1203		- /* file UUID SIZE \n CONTENT
1204		- ** file UUID DELTASRC SIZE \n CONTENT
	1222	+ /* file HASH SIZE \n CONTENT
	1223	+ ** file HASH DELTASRC SIZE \n CONTENT
1205	1224	**
1206	1225	** Accept a file from the client.
1207	1226	*/
1208	1227	if( blob_eq(&xfer.aToken[0], "file") ){
1209	1228	if( !isPush ){
		@@ -1219,12 +1238,12 @@
1219	1238	nErr++;
1220	1239	break;
1221	1240	}
1222	1241	}else
1223	1242
1224		- /* cfile UUID USIZE CSIZE \n CONTENT
1225		- ** cfile UUID DELTASRC USIZE CSIZE \n CONTENT
	1243	+ /* cfile HASH USIZE CSIZE \n CONTENT
	1244	+ ** cfile HASH DELTASRC USIZE CSIZE \n CONTENT
1226	1245	**
1227	1246	** Accept a file from the client.
1228	1247	*/
1229	1248	if( blob_eq(&xfer.aToken[0], "cfile") ){
1230	1249	if( !isPush ){
		@@ -1254,17 +1273,17 @@
1254	1273	nErr++;
1255	1274	break;
1256	1275	}
1257	1276	}else
1258	1277
1259		- /* gimme UUID
	1278	+ /* gimme HASH
1260	1279	**
1261	1280	** Client is requesting a file. Send it.
1262	1281	*/
1263	1282	if( blob_eq(&xfer.aToken[0], "gimme")
1264	1283	&& xfer.nToken==2
1265		- && blob_is_uuid(&xfer.aToken[1])
	1284	+ && blob_is_hname(&xfer.aToken[1])
1266	1285	){
1267	1286	nGimme++;
1268	1287	if( isPull ){
1269	1288	int rid = rid_from_uuid(&xfer.aToken[1], 0, 0);
1270	1289	if( rid ){
		@@ -1282,18 +1301,18 @@
1282	1301	&& blob_is_filename(&xfer.aToken[1])
1283	1302	){
1284	1303	send_unversioned_file(&xfer, blob_str(&xfer.aToken[1]), 0);
1285	1304	}else
1286	1305
1287		- /* igot UUID ?ISPRIVATE?
	1306	+ /* igot HASH ?ISPRIVATE?
1288	1307	**
1289	1308	** Client announces that it has a particular file. If the ISPRIVATE
1290	1309	** argument exists and is non-zero, then the file is a private file.
1291	1310	*/
1292	1311	if( xfer.nToken>=2
1293	1312	&& blob_eq(&xfer.aToken[0], "igot")
1294		- && blob_is_uuid(&xfer.aToken[1])
	1313	+ && blob_is_hname(&xfer.aToken[1])
1295	1314	){
1296	1315	if( isPush ){
1297	1316	if( xfer.nToken==2 \|\| blob_eq(&xfer.aToken[2],"1")==0 ){
1298	1317	rid_from_uuid(&xfer.aToken[1], 1, 0);
1299	1318	}else if( g.perm.Private ){
		@@ -1307,16 +1326,15 @@
1307	1326
1308	1327	/* pull SERVERCODE PROJECTCODE
1309	1328	** push SERVERCODE PROJECTCODE
1310	1329	**
1311	1330	** The client wants either send or receive. The server should
1312		- ** verify that the project code matches.
	1331	+ ** verify that the project code matches. The server code is ignored.
1313	1332	*/
1314	1333	if( xfer.nToken==3
1315	1334	&& (blob_eq(&xfer.aToken[0], "pull") \|\| blob_eq(&xfer.aToken[0], "push"))
1316		- && blob_is_uuid(&xfer.aToken[1])
1317		- && blob_is_uuid(&xfer.aToken[2])
	1335	+ && blob_is_hname(&xfer.aToken[2])
1318	1336	){
1319	1337	const char *zPCode;
1320	1338	zPCode = db_get("project-code", 0);
1321	1339	if( zPCode==0 ){
1322	1340	fossil_panic("missing project code");
		@@ -1534,19 +1552,27 @@
1534	1552	** Send igot cards for all known artifacts.
1535	1553	*/
1536	1554	if( blob_eq(&xfer.aToken[1], "send-catalog") ){
1537	1555	xfer.resync = 0x7fffffff;
1538	1556	}
	1557	+
	1558	+ /* pragma client-version VERSION
	1559	+ **
	1560	+ ** Let the server know what version of Fossil is running on the client.
	1561	+ */
	1562	+ if( xfer.nToken>=3 && blob_eq(&xfer.aToken[1], "client-version") ){
	1563	+ xfer.clientVersion = atoi(blob_str(&xfer.aToken[2]));
	1564	+ }
1539	1565
1540	1566	/* pragma uv-hash HASH
1541	1567	**
1542	1568	** The client wants to make sure that unversioned files are all synced.
1543	1569	** If the HASH does not match, send a complete catalog of
1544	1570	** "uvigot" cards.
1545	1571	*/
1546	1572	if( blob_eq(&xfer.aToken[1], "uv-hash")
1547		- && blob_is_uuid(&xfer.aToken[2])
	1573	+ && blob_is_hname(&xfer.aToken[2])
1548	1574	){
1549	1575	if( !uvCatalogSent ){
1550	1576	if( g.perm.Read && g.perm.WrUnver ){
1551	1577	@ pragma uv-push-ok
1552	1578	send_unversioned_catalog(&xfer);
		@@ -1742,10 +1768,11 @@
1742	1768	memset(&xfer, 0, sizeof(xfer));
1743	1769	xfer.pIn = &recv;
1744	1770	xfer.pOut = &send;
1745	1771	xfer.mxSend = db_get_int("max-upload", 250000);
1746	1772	xfer.maxTime = -1;
	1773	+ xfer.clientVersion = RELEASE_VERSION_NUMBER;
1747	1774	if( syncFlags & SYNC_PRIVATE ){
1748	1775	g.perm.Private = 1;
1749	1776	xfer.syncPrivate = 1;
1750	1777	}
1751	1778
		@@ -1783,10 +1810,11 @@
1783	1810	}
1784	1811
1785	1812	/*
1786	1813	** Always begin with a clone, pull, or push message
1787	1814	*/
	1815	+ blob_appendf(&send, "pragma client-version %d\n", RELEASE_VERSION_NUMBER);
1788	1816	if( syncFlags & SYNC_CLONE ){
1789	1817	blob_appendf(&send, "clone 3 %d\n", cloneSeqno);
1790	1818	syncFlags &= ~(SYNC_PUSH\|SYNC_PULL);
1791	1819	nCardSent++;
1792	1820	/* TBD: Request all transferable configuration values */
		@@ -1819,10 +1847,11 @@
1819	1847	db_record_repository_filename(0);
1820	1848	db_multi_exec(
1821	1849	"CREATE TEMP TABLE onremote(rid INTEGER PRIMARY KEY);"
1822	1850	);
1823	1851	manifest_crosslink_begin();
	1852	+
1824	1853
1825	1854	/* Send back the most recently received cookie. Let the server
1826	1855	** figure out if this is a cookie that it cares about.
1827	1856	*/
1828	1857	zCookie = db_get("cookie", 0);
		@@ -1976,10 +2005,11 @@
1976	2005	xfer.nGimmeSent = 0;
1977	2006	xfer.nIGotSent = 0;
1978	2007
1979	2008	lastPctDone = -1;
1980	2009	blob_reset(&send);
	2010	+ blob_appendf(&send, "pragma client-version %d\n", RELEASE_VERSION_NUMBER);
1981	2011	rArrivalTime = db_double(0.0, "SELECT julianday('now')");
1982	2012
1983	2013	/* Send the send-private pragma if we are trying to sync private data */
1984	2014	if( syncFlags & SYNC_PRIVATE ){
1985	2015	blob_append(&send, "pragma send-private\n", -1);
		@@ -2028,22 +2058,22 @@
2028	2058	lastPctDone = pctDone;
2029	2059	fflush(stdout);
2030	2060	}
2031	2061	}
2032	2062
2033		- /* file UUID SIZE \n CONTENT
2034		- ** file UUID DELTASRC SIZE \n CONTENT
	2063	+ /* file HASH SIZE \n CONTENT
	2064	+ ** file HASH DELTASRC SIZE \n CONTENT
2035	2065	**
2036	2066	** Receive a file transmitted from the server.
2037	2067	*/
2038	2068	if( blob_eq(&xfer.aToken[0],"file") ){
2039	2069	xfer_accept_file(&xfer, (syncFlags & SYNC_CLONE)!=0, 0, 0);
2040	2070	nArtifactRcvd++;
2041	2071	}else
2042	2072
2043		- /* cfile UUID USIZE CSIZE \n CONTENT
2044		- ** cfile UUID DELTASRC USIZE CSIZE \n CONTENT
	2073	+ /* cfile HASH USIZE CSIZE \n CONTENT
	2074	+ ** cfile HASH DELTASRC USIZE CSIZE \n CONTENT
2045	2075	**
2046	2076	** Receive a compressed file transmitted from the server.
2047	2077	*/
2048	2078	if( blob_eq(&xfer.aToken[0],"cfile") ){
2049	2079	xfer_accept_compressed_file(&xfer, 0, 0);
		@@ -2062,27 +2092,27 @@
2062	2092	fossil_print("\rUnversioned-file received: %s\n",
2063	2093	blob_str(&xfer.aToken[1]));
2064	2094	}
2065	2095	}else
2066	2096
2067		- /* gimme UUID
	2097	+ /* gimme HASH
2068	2098	**
2069	2099	** Server is requesting a file. If the file is a manifest, assume
2070	2100	** that the server will also want to know all of the content files
2071	2101	** associated with the manifest and send those too.
2072	2102	*/
2073	2103	if( blob_eq(&xfer.aToken[0], "gimme")
2074	2104	&& xfer.nToken==2
2075		- && blob_is_uuid(&xfer.aToken[1])
	2105	+ && blob_is_hname(&xfer.aToken[1])
2076	2106	){
2077	2107	if( syncFlags & SYNC_PUSH ){
2078	2108	int rid = rid_from_uuid(&xfer.aToken[1], 0, 0);
2079	2109	if( rid ) send_file(&xfer, rid, &xfer.aToken[1], 0);
2080	2110	}
2081	2111	}else
2082	2112
2083		- /* igot UUID ?PRIVATEFLAG?
	2113	+ /* igot HASH ?PRIVATEFLAG?
2084	2114	**
2085	2115	** Server announces that it has a particular file. If this is
2086	2116	** not a file that we have and we are pulling, then create a
2087	2117	** phantom to cause this file to be requested on the next cycle.
2088	2118	** Always remember that the server has this file so that we do
		@@ -2092,11 +2122,11 @@
2092	2122	** private. Pretend it does not exists if we are not pulling
2093	2123	** private files.
2094	2124	*/
2095	2125	if( xfer.nToken>=2
2096	2126	&& blob_eq(&xfer.aToken[0], "igot")
2097		- && blob_is_uuid(&xfer.aToken[1])
	2127	+ && blob_is_hname(&xfer.aToken[1])
2098	2128	){
2099	2129	int rid;
2100	2130	int isPriv = xfer.nToken>=3 && blob_eq(&xfer.aToken[2],"1");
2101	2131	rid = rid_from_uuid(&xfer.aToken[1], 0, 0);
2102	2132	if( rid>0 ){
		@@ -2126,11 +2156,11 @@
2126	2156	if( xfer.nToken==5
2127	2157	&& blob_eq(&xfer.aToken[0], "uvigot")
2128	2158	&& blob_is_filename(&xfer.aToken[1])
2129	2159	&& blob_is_int64(&xfer.aToken[2], &mtime)
2130	2160	&& blob_is_int(&xfer.aToken[4], &size)
2131		- && (blob_eq(&xfer.aToken[3],"-") \|\| blob_is_uuid(&xfer.aToken[3]))
	2161	+ && (blob_eq(&xfer.aToken[3],"-") \|\| blob_is_hname(&xfer.aToken[3]))
2132	2162	){
2133	2163	const char *zName = blob_str(&xfer.aToken[1]);
2134	2164	const char *zHash = blob_str(&xfer.aToken[3]);
2135	2165	int iStatus;
2136	2166	iStatus = unversioned_status(zName, mtime, zHash);
		@@ -2188,11 +2218,11 @@
2188	2218	** the client what product to use for the new database.
2189	2219	*/
2190	2220	if( blob_eq(&xfer.aToken[0],"push")
2191	2221	&& xfer.nToken==3
2192	2222	&& (syncFlags & SYNC_CLONE)!=0
2193		- && blob_is_uuid(&xfer.aToken[2])
	2223	+ && blob_is_hname(&xfer.aToken[2])
2194	2224	){
2195	2225	if( zPCode==0 ){
2196	2226	zPCode = mprintf("%b", &xfer.aToken[2]);
2197	2227	db_set("project-code", zPCode, 0);
2198	2228	}
2199	2229

	--- src/xfer.c
	+++ src/xfer.c
	@@ -49,16 +49,17 @@
49	int nDanglingFile; /* Number of dangling deltas received */
50	int mxSend; /* Stop sending "file" when pOut reaches this size */
51	int resync; /* Send igot cards for all holdings */
52	u8 syncPrivate; /* True to enable syncing private content */
53	u8 nextIsPrivate; /* If true, next "file" received is a private */

54	time_t maxTime; /* Time when this transfer should be finished */
55	};
56
57
58	/*
59	** The input blob contains a UUID. Convert it into a record ID.
60	** Create a phantom record if no prior record exists and
61	** phantomize is true.
62	**
63	** Compare to uuid_to_rid(). This routine takes a blob argument
64	** and does less error checking.
	@@ -100,12 +101,12 @@
100	** message. This routine finishes parsing that message and does
101	** a record insert of the file.
102	**
103	** The file line is in one of the following two forms:
104	**
105	** file UUID SIZE \n CONTENT
106	** file UUID DELTASRC SIZE \n CONTENT
107	**
108	** The content is SIZE bytes immediately following the newline.
109	** If DELTASRC exists, then the CONTENT is a delta against the
110	** content of DELTASRC.
111	**
	@@ -122,29 +123,30 @@
122	int *pnUuidList
123	){
124	int n;
125	int rid;
126	int srcid = 0;
127	Blob content, hash;
128	int isPriv;

129
130	isPriv = pXfer->nextIsPrivate;
131	pXfer->nextIsPrivate = 0;
132	if( pXfer->nToken<3
133	\|\| pXfer->nToken>4
134	\|\| !blob_is_uuid(&pXfer->aToken[1])
135	\|\| !blob_is_int(&pXfer->aToken[pXfer->nToken-1], &n)
136	\|\| n<0
137	\|\| (pXfer->nToken==4 && !blob_is_uuid(&pXfer->aToken[2]))
138	){
139	blob_appendf(&pXfer->err, "malformed file line");
140	return;
141	}
142	blob_zero(&content);
143	blob_zero(&hash);
144	blob_extract(pXfer->pIn, n, &content);
145	if( !cloneFlag && uuid_is_shunned(blob_str(&pXfer->aToken[1])) ){

146	/* Ignore files that have been shunned */
147	blob_reset(&content);
148	return;
149	}
150	if( isPriv && !g.perm.Private ){
	@@ -158,26 +160,26 @@
158	pXfer->nDeltaRcvd++;
159	}else{
160	srcid = 0;
161	pXfer->nFileRcvd++;
162	}
163	rid = content_put_ex(&content, blob_str(&pXfer->aToken[1]), srcid,
164	0, isPriv);
165	Th_AppendToList(pzUuidList, pnUuidList, blob_str(&pXfer->aToken[1]),
166	blob_size(&pXfer->aToken[1]));
167	remote_has(rid);
168	blob_reset(&content);
169	return;
170	}
171	if( pXfer->nToken==4 ){
172	Blob src, next;
173	srcid = rid_from_uuid(&pXfer->aToken[2], 1, isPriv);
174	if( content_get(srcid, &src)==0 ){
175	rid = content_put_ex(&content, blob_str(&pXfer->aToken[1]), srcid,
176	0, isPriv);
177	Th_AppendToList(pzUuidList, pnUuidList, blob_str(&pXfer->aToken[1]),
178	blob_size(&pXfer->aToken[1]));
179	pXfer->nDanglingFile++;
180	db_multi_exec("DELETE FROM phantom WHERE rid=%d", rid);
181	if( !isPriv ) content_make_public(rid);
182	blob_reset(&src);
183	blob_reset(&content);
	@@ -189,19 +191,15 @@
189	blob_reset(&content);
190	content = next;
191	}else{
192	pXfer->nFileRcvd++;
193	}
194	sha1sum_blob(&content, &hash);
195	if( !blob_eq_str(&pXfer->aToken[1], blob_str(&hash), -1) ){
196	blob_appendf(&pXfer->err,
197	"wrong hash on received artifact: expected %s but got %s",
198	blob_str(&pXfer->aToken[1]), blob_str(&hash));
199	}
200	rid = content_put_ex(&content, blob_str(&hash), 0, 0, isPriv);
201	Th_AppendToList(pzUuidList, pnUuidList, blob_str(&hash), blob_size(&hash));
202	blob_reset(&hash);
203	if( rid==0 ){
204	blob_appendf(&pXfer->err, "%s", g.zErrMsg);
205	blob_reset(&content);
206	}else{
207	if( !isPriv ) content_make_public(rid);
	@@ -217,18 +215,18 @@
217	** a record insert of the file. The difference between "file" and
218	** "cfile" is that with "cfile" the content is already compressed.
219	**
220	** The file line is in one of the following two forms:
221	**
222	** cfile UUID USIZE CSIZE \n CONTENT
223	** cfile UUID DELTASRC USIZE CSIZE \n CONTENT
224	**
225	** The content is CSIZE bytes immediately following the newline.
226	** If DELTASRC exists, then the CONTENT is a delta against the
227	** content of DELTASRC.
228	**
229	** The original size of the UUID artifact is USIZE.
230	**
231	** If any error occurs, write a message into pErr which has already
232	** be initialized to an empty string.
233	**
234	** Any artifact successfully received by this routine is considered to
	@@ -248,15 +246,15 @@
248
249	isPriv = pXfer->nextIsPrivate;
250	pXfer->nextIsPrivate = 0;
251	if( pXfer->nToken<4
252	\|\| pXfer->nToken>5
253	\|\| !blob_is_uuid(&pXfer->aToken[1])
254	\|\| !blob_is_int(&pXfer->aToken[pXfer->nToken-2], &szU)
255	\|\| !blob_is_int(&pXfer->aToken[pXfer->nToken-1], &szC)
256	\|\| szC<0 \|\| szU<0
257	\|\| (pXfer->nToken==5 && !blob_is_uuid(&pXfer->aToken[2]))
258	){
259	blob_appendf(&pXfer->err, "malformed cfile line");
260	return;
261	}
262	if( isPriv && !g.perm.Private ){
	@@ -297,11 +295,11 @@
297	**
298	** If the 0x0001 bit of FLAGS is set, that means the file has been
299	** deleted, SIZE is zero, the HASH is "-", and the "\n CONTENT" is omitted.
300	**
301	** SIZE is the number of bytes of CONTENT. The CONTENT is uncompressed.
302	** HASH is the SHA1 hash of CONTENT.
303	**
304	** If the 0x0004 bit of FLAGS is set, that means the CONTENT is omitted.
305	** The sender might have omitted the content because it is too big to
306	** transmit, or because it is unchanged and this record exists purely
307	** to update the MTIME.
	@@ -310,11 +308,10 @@
310	sqlite3_int64 mtime; /* The MTIME */
311	Blob pHash; / The HASH value */
312	int sz; /* The SIZE */
313	int flags; /* The FLAGS */
314	Blob content; /* The CONTENT */
315	Blob hash; /* Hash computed from CONTENT to compare with HASH */
316	Blob x; /* Compressed content */
317	Stmt q; /* SQL statements for comparison and insert */
318	int isDelete; /* HASH is "-" indicating this is a delete */
319	int nullContent; /* True of CONTENT is NULL */
320	int iStatus; /* Result from unversioned_status() */
	@@ -321,25 +318,23 @@
321
322	pHash = &pXfer->aToken[3];
323	if( pXfer->nToken==5
324	\|\| !blob_is_filename(&pXfer->aToken[1])
325	\|\| !blob_is_int64(&pXfer->aToken[2], &mtime)
326	\|\| (!blob_eq(pHash,"-") && !blob_is_uuid(pHash))
327	\|\| !blob_is_int(&pXfer->aToken[4], &sz)
328	\|\| !blob_is_int(&pXfer->aToken[5], &flags)
329	){
330	blob_appendf(&pXfer->err, "malformed uvfile line");
331	return;
332	}
333	blob_init(&content, 0, 0);
334	blob_init(&hash, 0, 0);
335	blob_init(&x, 0, 0);
336	if( sz>0 && (flags & 0x0005)==0 ){
337	blob_extract(pXfer->pIn, sz, &content);
338	nullContent = 0;
339	sha1sum_blob(&content, &hash);
340	if( blob_compare(&hash, pHash)!=0 ){
341	blob_appendf(&pXfer->err, "in uvfile line, HASH does not match CONTENT");
342	goto end_accept_unversioned_file;
343	}
344	}else{
345	nullContent = 1;
	@@ -399,11 +394,10 @@
399	db_unset("uv-hash", 0);
400
401	end_accept_unversioned_file:
402	blob_reset(&x);
403	blob_reset(&content);
404	blob_reset(&hash);
405	}
406
407	/*
408	** Try to send a file as a delta against its parent.
409	** If successful, return the number of bytes in the delta.
	@@ -415,11 +409,11 @@
415	static int send_delta_parent(
416	Xfer pXfer, / The transfer context */
417	int rid, /* record id of the file to send */
418	int isPrivate, /* True if rid is a private artifact */
419	Blob pContent, / The content of the file to send */
420	Blob pUuid / The UUID of the file to send */
421	){
422	static const char *const azQuery[] = {
423	"SELECT pid FROM plink x"
424	" WHERE cid=%d"
425	" AND NOT EXISTS(SELECT 1 FROM phantom WHERE rid=pid)",
	@@ -469,11 +463,11 @@
469	*/
470	static int send_delta_native(
471	Xfer pXfer, / The transfer context */
472	int rid, /* record id of the file to send */
473	int isPrivate, /* True if rid is a private artifact */
474	Blob pUuid / The UUID of the file to send */
475	){
476	Blob src, delta;
477	int size = 0;
478	int srcId;
479
	@@ -500,15 +494,26 @@
500	}else{
501	size = 0;
502	}
503	return size;
504	}











505
506	/*
507	** Send the file identified by rid.
508	**
509	** The pUuid can be NULL in which case the correct UUID is computed
510	** from the rid.
511	**
512	** Try to send the file as a native delta if nativeDelta is true, or
513	** as a parent delta if nativeDelta is false.
514	**
	@@ -527,10 +532,14 @@
527	}
528	blob_zero(&uuid);
529	db_blob(&uuid, "SELECT uuid FROM blob WHERE rid=%d AND size>=0", rid);
530	if( blob_size(&uuid)==0 ){
531	return;




532	}
533	if( pUuid ){
534	if( blob_compare(pUuid, &uuid)!=0 ){
535	blob_reset(&uuid);
536	return;
	@@ -617,10 +626,15 @@
617	szC = db_column_bytes(&q1, 2);
618	zContent = db_column_raw(&q1, 2);
619	srcIsPrivate = db_column_int(&q1, 3);
620	zDelta = db_column_text(&q1, 4);
621	if( isPrivate ) blob_append(pXfer->pOut, "private\n", -1);





622	blob_appendf(pXfer->pOut, "cfile %s ", zUuid);
623	if( !isPrivate && srcIsPrivate ){
624	content_get(rid, &fullContent);
625	szU = blob_size(&fullContent);
626	blob_compress(&fullContent, &fullContent);
	@@ -676,10 +690,15 @@
676	);
677	}
678	if( db_step(&q1)==SQLITE_ROW ){
679	sqlite3_int64 mtime = db_column_int64(&q1, 0);
680	const char *zHash = db_column_text(&q1, 1);





681	if( blob_size(pXfer->pOut)>=pXfer->mxSend ){
682	/* If we have already reached the send size limit, send a (short)
683	** uvigot card rather than a uvfile card. This only happens on the
684	** server side. The uvigot card will provoke the client to resend
685	** another uvgimme on the next cycle. */
	@@ -728,23 +747,23 @@
728	}
729	db_finalize(&q);
730	}
731
732	/*
733	** Compute an SHA1 hash on the tail of pMsg. Verify that it matches the
734	** the hash given in pHash. Return non-zero for an error and 0 on success.



735	*/
736	static int check_tail_hash(Blob pHash, Blob pMsg){
737	Blob tail;
738	Blob h2;
739	int rc;
740	blob_tail(pMsg, &tail);
741	sha1sum_blob(&tail, &h2);
742	rc = blob_compare(pHash, &h2);
743	blob_reset(&h2);
744	blob_reset(&tail);
745	return rc;
746	}
747
748	/*
749	** Check the signature on an application/x-fossil payload received by
750	** the HTTP server. The signature is a line of the following form:
	@@ -1198,12 +1217,12 @@
1198	while( blob_line(xfer.pIn, &xfer.line) ){
1199	if( blob_buffer(&xfer.line)[0]=='#' ) continue;
1200	if( blob_size(&xfer.line)==0 ) continue;
1201	xfer.nToken = blob_tokenize(&xfer.line, xfer.aToken, count(xfer.aToken));
1202
1203	/* file UUID SIZE \n CONTENT
1204	** file UUID DELTASRC SIZE \n CONTENT
1205	**
1206	** Accept a file from the client.
1207	*/
1208	if( blob_eq(&xfer.aToken[0], "file") ){
1209	if( !isPush ){
	@@ -1219,12 +1238,12 @@
1219	nErr++;
1220	break;
1221	}
1222	}else
1223
1224	/* cfile UUID USIZE CSIZE \n CONTENT
1225	** cfile UUID DELTASRC USIZE CSIZE \n CONTENT
1226	**
1227	** Accept a file from the client.
1228	*/
1229	if( blob_eq(&xfer.aToken[0], "cfile") ){
1230	if( !isPush ){
	@@ -1254,17 +1273,17 @@
1254	nErr++;
1255	break;
1256	}
1257	}else
1258
1259	/* gimme UUID
1260	**
1261	** Client is requesting a file. Send it.
1262	*/
1263	if( blob_eq(&xfer.aToken[0], "gimme")
1264	&& xfer.nToken==2
1265	&& blob_is_uuid(&xfer.aToken[1])
1266	){
1267	nGimme++;
1268	if( isPull ){
1269	int rid = rid_from_uuid(&xfer.aToken[1], 0, 0);
1270	if( rid ){
	@@ -1282,18 +1301,18 @@
1282	&& blob_is_filename(&xfer.aToken[1])
1283	){
1284	send_unversioned_file(&xfer, blob_str(&xfer.aToken[1]), 0);
1285	}else
1286
1287	/* igot UUID ?ISPRIVATE?
1288	**
1289	** Client announces that it has a particular file. If the ISPRIVATE
1290	** argument exists and is non-zero, then the file is a private file.
1291	*/
1292	if( xfer.nToken>=2
1293	&& blob_eq(&xfer.aToken[0], "igot")
1294	&& blob_is_uuid(&xfer.aToken[1])
1295	){
1296	if( isPush ){
1297	if( xfer.nToken==2 \|\| blob_eq(&xfer.aToken[2],"1")==0 ){
1298	rid_from_uuid(&xfer.aToken[1], 1, 0);
1299	}else if( g.perm.Private ){
	@@ -1307,16 +1326,15 @@
1307
1308	/* pull SERVERCODE PROJECTCODE
1309	** push SERVERCODE PROJECTCODE
1310	**
1311	** The client wants either send or receive. The server should
1312	** verify that the project code matches.
1313	*/
1314	if( xfer.nToken==3
1315	&& (blob_eq(&xfer.aToken[0], "pull") \|\| blob_eq(&xfer.aToken[0], "push"))
1316	&& blob_is_uuid(&xfer.aToken[1])
1317	&& blob_is_uuid(&xfer.aToken[2])
1318	){
1319	const char *zPCode;
1320	zPCode = db_get("project-code", 0);
1321	if( zPCode==0 ){
1322	fossil_panic("missing project code");
	@@ -1534,19 +1552,27 @@
1534	** Send igot cards for all known artifacts.
1535	*/
1536	if( blob_eq(&xfer.aToken[1], "send-catalog") ){
1537	xfer.resync = 0x7fffffff;
1538	}








1539
1540	/* pragma uv-hash HASH
1541	**
1542	** The client wants to make sure that unversioned files are all synced.
1543	** If the HASH does not match, send a complete catalog of
1544	** "uvigot" cards.
1545	*/
1546	if( blob_eq(&xfer.aToken[1], "uv-hash")
1547	&& blob_is_uuid(&xfer.aToken[2])
1548	){
1549	if( !uvCatalogSent ){
1550	if( g.perm.Read && g.perm.WrUnver ){
1551	@ pragma uv-push-ok
1552	send_unversioned_catalog(&xfer);
	@@ -1742,10 +1768,11 @@
1742	memset(&xfer, 0, sizeof(xfer));
1743	xfer.pIn = &recv;
1744	xfer.pOut = &send;
1745	xfer.mxSend = db_get_int("max-upload", 250000);
1746	xfer.maxTime = -1;

1747	if( syncFlags & SYNC_PRIVATE ){
1748	g.perm.Private = 1;
1749	xfer.syncPrivate = 1;
1750	}
1751
	@@ -1783,10 +1810,11 @@
1783	}
1784
1785	/*
1786	** Always begin with a clone, pull, or push message
1787	*/

1788	if( syncFlags & SYNC_CLONE ){
1789	blob_appendf(&send, "clone 3 %d\n", cloneSeqno);
1790	syncFlags &= ~(SYNC_PUSH\|SYNC_PULL);
1791	nCardSent++;
1792	/* TBD: Request all transferable configuration values */
	@@ -1819,10 +1847,11 @@
1819	db_record_repository_filename(0);
1820	db_multi_exec(
1821	"CREATE TEMP TABLE onremote(rid INTEGER PRIMARY KEY);"
1822	);
1823	manifest_crosslink_begin();

1824
1825	/* Send back the most recently received cookie. Let the server
1826	** figure out if this is a cookie that it cares about.
1827	*/
1828	zCookie = db_get("cookie", 0);
	@@ -1976,10 +2005,11 @@
1976	xfer.nGimmeSent = 0;
1977	xfer.nIGotSent = 0;
1978
1979	lastPctDone = -1;
1980	blob_reset(&send);

1981	rArrivalTime = db_double(0.0, "SELECT julianday('now')");
1982
1983	/* Send the send-private pragma if we are trying to sync private data */
1984	if( syncFlags & SYNC_PRIVATE ){
1985	blob_append(&send, "pragma send-private\n", -1);
	@@ -2028,22 +2058,22 @@
2028	lastPctDone = pctDone;
2029	fflush(stdout);
2030	}
2031	}
2032
2033	/* file UUID SIZE \n CONTENT
2034	** file UUID DELTASRC SIZE \n CONTENT
2035	**
2036	** Receive a file transmitted from the server.
2037	*/
2038	if( blob_eq(&xfer.aToken[0],"file") ){
2039	xfer_accept_file(&xfer, (syncFlags & SYNC_CLONE)!=0, 0, 0);
2040	nArtifactRcvd++;
2041	}else
2042
2043	/* cfile UUID USIZE CSIZE \n CONTENT
2044	** cfile UUID DELTASRC USIZE CSIZE \n CONTENT
2045	**
2046	** Receive a compressed file transmitted from the server.
2047	*/
2048	if( blob_eq(&xfer.aToken[0],"cfile") ){
2049	xfer_accept_compressed_file(&xfer, 0, 0);
	@@ -2062,27 +2092,27 @@
2062	fossil_print("\rUnversioned-file received: %s\n",
2063	blob_str(&xfer.aToken[1]));
2064	}
2065	}else
2066
2067	/* gimme UUID
2068	**
2069	** Server is requesting a file. If the file is a manifest, assume
2070	** that the server will also want to know all of the content files
2071	** associated with the manifest and send those too.
2072	*/
2073	if( blob_eq(&xfer.aToken[0], "gimme")
2074	&& xfer.nToken==2
2075	&& blob_is_uuid(&xfer.aToken[1])
2076	){
2077	if( syncFlags & SYNC_PUSH ){
2078	int rid = rid_from_uuid(&xfer.aToken[1], 0, 0);
2079	if( rid ) send_file(&xfer, rid, &xfer.aToken[1], 0);
2080	}
2081	}else
2082
2083	/* igot UUID ?PRIVATEFLAG?
2084	**
2085	** Server announces that it has a particular file. If this is
2086	** not a file that we have and we are pulling, then create a
2087	** phantom to cause this file to be requested on the next cycle.
2088	** Always remember that the server has this file so that we do
	@@ -2092,11 +2122,11 @@
2092	** private. Pretend it does not exists if we are not pulling
2093	** private files.
2094	*/
2095	if( xfer.nToken>=2
2096	&& blob_eq(&xfer.aToken[0], "igot")
2097	&& blob_is_uuid(&xfer.aToken[1])
2098	){
2099	int rid;
2100	int isPriv = xfer.nToken>=3 && blob_eq(&xfer.aToken[2],"1");
2101	rid = rid_from_uuid(&xfer.aToken[1], 0, 0);
2102	if( rid>0 ){
	@@ -2126,11 +2156,11 @@
2126	if( xfer.nToken==5
2127	&& blob_eq(&xfer.aToken[0], "uvigot")
2128	&& blob_is_filename(&xfer.aToken[1])
2129	&& blob_is_int64(&xfer.aToken[2], &mtime)
2130	&& blob_is_int(&xfer.aToken[4], &size)
2131	&& (blob_eq(&xfer.aToken[3],"-") \|\| blob_is_uuid(&xfer.aToken[3]))
2132	){
2133	const char *zName = blob_str(&xfer.aToken[1]);
2134	const char *zHash = blob_str(&xfer.aToken[3]);
2135	int iStatus;
2136	iStatus = unversioned_status(zName, mtime, zHash);
	@@ -2188,11 +2218,11 @@
2188	** the client what product to use for the new database.
2189	*/
2190	if( blob_eq(&xfer.aToken[0],"push")
2191	&& xfer.nToken==3
2192	&& (syncFlags & SYNC_CLONE)!=0
2193	&& blob_is_uuid(&xfer.aToken[2])
2194	){
2195	if( zPCode==0 ){
2196	zPCode = mprintf("%b", &xfer.aToken[2]);
2197	db_set("project-code", zPCode, 0);
2198	}
2199

	--- src/xfer.c
	+++ src/xfer.c
	@@ -49,16 +49,17 @@
49	int nDanglingFile; /* Number of dangling deltas received */
50	int mxSend; /* Stop sending "file" when pOut reaches this size */
51	int resync; /* Send igot cards for all holdings */
52	u8 syncPrivate; /* True to enable syncing private content */
53	u8 nextIsPrivate; /* If true, next "file" received is a private */
54	u32 clientVersion; /* Version of the client software */
55	time_t maxTime; /* Time when this transfer should be finished */
56	};
57
58
59	/*
60	** The input blob contains an artifact. Convert it into a record ID.
61	** Create a phantom record if no prior record exists and
62	** phantomize is true.
63	**
64	** Compare to uuid_to_rid(). This routine takes a blob argument
65	** and does less error checking.
	@@ -100,12 +101,12 @@
101	** message. This routine finishes parsing that message and does
102	** a record insert of the file.
103	**
104	** The file line is in one of the following two forms:
105	**
106	** file HASH SIZE \n CONTENT
107	** file HASH DELTASRC SIZE \n CONTENT
108	**
109	** The content is SIZE bytes immediately following the newline.
110	** If DELTASRC exists, then the CONTENT is a delta against the
111	** content of DELTASRC.
112	**
	@@ -122,29 +123,30 @@
123	int *pnUuidList
124	){
125	int n;
126	int rid;
127	int srcid = 0;
128	Blob content;
129	int isPriv;
130	Blob *pUuid;
131
132	isPriv = pXfer->nextIsPrivate;
133	pXfer->nextIsPrivate = 0;
134	if( pXfer->nToken<3
135	\|\| pXfer->nToken>4
136	\|\| !blob_is_hname(&pXfer->aToken[1])
137	\|\| !blob_is_int(&pXfer->aToken[pXfer->nToken-1], &n)
138	\|\| n<0
139	\|\| (pXfer->nToken==4 && !blob_is_hname(&pXfer->aToken[2]))
140	){
141	blob_appendf(&pXfer->err, "malformed file line");
142	return;
143	}
144	blob_zero(&content);

145	blob_extract(pXfer->pIn, n, &content);
146	pUuid = &pXfer->aToken[1];
147	if( !cloneFlag && uuid_is_shunned(blob_str(pUuid)) ){
148	/* Ignore files that have been shunned */
149	blob_reset(&content);
150	return;
151	}
152	if( isPriv && !g.perm.Private ){
	@@ -158,26 +160,26 @@
160	pXfer->nDeltaRcvd++;
161	}else{
162	srcid = 0;
163	pXfer->nFileRcvd++;
164	}
165	rid = content_put_ex(&content, blob_str(pUuid), srcid,
166	0, isPriv);
167	Th_AppendToList(pzUuidList, pnUuidList, blob_str(pUuid),
168	blob_size(pUuid));
169	remote_has(rid);
170	blob_reset(&content);
171	return;
172	}
173	if( pXfer->nToken==4 ){
174	Blob src, next;
175	srcid = rid_from_uuid(&pXfer->aToken[2], 1, isPriv);
176	if( content_get(srcid, &src)==0 ){
177	rid = content_put_ex(&content, blob_str(pUuid), srcid,
178	0, isPriv);
179	Th_AppendToList(pzUuidList, pnUuidList, blob_str(pUuid),
180	blob_size(pUuid));
181	pXfer->nDanglingFile++;
182	db_multi_exec("DELETE FROM phantom WHERE rid=%d", rid);
183	if( !isPriv ) content_make_public(rid);
184	blob_reset(&src);
185	blob_reset(&content);
	@@ -189,19 +191,15 @@
191	blob_reset(&content);
192	content = next;
193	}else{
194	pXfer->nFileRcvd++;
195	}
196	if( hname_verify_hash(&content, blob_buffer(pUuid), blob_size(pUuid))==0 ){
197	blob_appendf(&pXfer->err, "wrong hash on received artifact: %b", pUuid);
198	}
199	rid = content_put_ex(&content, blob_str(pUuid), 0, 0, isPriv);
200	Th_AppendToList(pzUuidList, pnUuidList, blob_str(pUuid), blob_size(pUuid));




201	if( rid==0 ){
202	blob_appendf(&pXfer->err, "%s", g.zErrMsg);
203	blob_reset(&content);
204	}else{
205	if( !isPriv ) content_make_public(rid);
	@@ -217,18 +215,18 @@
215	** a record insert of the file. The difference between "file" and
216	** "cfile" is that with "cfile" the content is already compressed.
217	**
218	** The file line is in one of the following two forms:
219	**
220	** cfile HASH USIZE CSIZE \n CONTENT
221	** cfile HASH DELTASRC USIZE CSIZE \n CONTENT
222	**
223	** The content is CSIZE bytes immediately following the newline.
224	** If DELTASRC exists, then the CONTENT is a delta against the
225	** content of DELTASRC.
226	**
227	** The original size of the HASH artifact is USIZE.
228	**
229	** If any error occurs, write a message into pErr which has already
230	** be initialized to an empty string.
231	**
232	** Any artifact successfully received by this routine is considered to
	@@ -248,15 +246,15 @@
246
247	isPriv = pXfer->nextIsPrivate;
248	pXfer->nextIsPrivate = 0;
249	if( pXfer->nToken<4
250	\|\| pXfer->nToken>5
251	\|\| !blob_is_hname(&pXfer->aToken[1])
252	\|\| !blob_is_int(&pXfer->aToken[pXfer->nToken-2], &szU)
253	\|\| !blob_is_int(&pXfer->aToken[pXfer->nToken-1], &szC)
254	\|\| szC<0 \|\| szU<0
255	\|\| (pXfer->nToken==5 && !blob_is_hname(&pXfer->aToken[2]))
256	){
257	blob_appendf(&pXfer->err, "malformed cfile line");
258	return;
259	}
260	if( isPriv && !g.perm.Private ){
	@@ -297,11 +295,11 @@
295	**
296	** If the 0x0001 bit of FLAGS is set, that means the file has been
297	** deleted, SIZE is zero, the HASH is "-", and the "\n CONTENT" is omitted.
298	**
299	** SIZE is the number of bytes of CONTENT. The CONTENT is uncompressed.
300	** HASH is the artifact hash of CONTENT.
301	**
302	** If the 0x0004 bit of FLAGS is set, that means the CONTENT is omitted.
303	** The sender might have omitted the content because it is too big to
304	** transmit, or because it is unchanged and this record exists purely
305	** to update the MTIME.
	@@ -310,11 +308,10 @@
308	sqlite3_int64 mtime; /* The MTIME */
309	Blob pHash; / The HASH value */
310	int sz; /* The SIZE */
311	int flags; /* The FLAGS */
312	Blob content; /* The CONTENT */

313	Blob x; /* Compressed content */
314	Stmt q; /* SQL statements for comparison and insert */
315	int isDelete; /* HASH is "-" indicating this is a delete */
316	int nullContent; /* True of CONTENT is NULL */
317	int iStatus; /* Result from unversioned_status() */
	@@ -321,25 +318,23 @@
318
319	pHash = &pXfer->aToken[3];
320	if( pXfer->nToken==5
321	\|\| !blob_is_filename(&pXfer->aToken[1])
322	\|\| !blob_is_int64(&pXfer->aToken[2], &mtime)
323	\|\| (!blob_eq(pHash,"-") && !blob_is_hname(pHash))
324	\|\| !blob_is_int(&pXfer->aToken[4], &sz)
325	\|\| !blob_is_int(&pXfer->aToken[5], &flags)
326	){
327	blob_appendf(&pXfer->err, "malformed uvfile line");
328	return;
329	}
330	blob_init(&content, 0, 0);

331	blob_init(&x, 0, 0);
332	if( sz>0 && (flags & 0x0005)==0 ){
333	blob_extract(pXfer->pIn, sz, &content);
334	nullContent = 0;
335	if( hname_verify_hash(&content, blob_buffer(pHash), blob_size(pHash))==0 ){

336	blob_appendf(&pXfer->err, "in uvfile line, HASH does not match CONTENT");
337	goto end_accept_unversioned_file;
338	}
339	}else{
340	nullContent = 1;
	@@ -399,11 +394,10 @@
394	db_unset("uv-hash", 0);
395
396	end_accept_unversioned_file:
397	blob_reset(&x);
398	blob_reset(&content);

399	}
400
401	/*
402	** Try to send a file as a delta against its parent.
403	** If successful, return the number of bytes in the delta.
	@@ -415,11 +409,11 @@
409	static int send_delta_parent(
410	Xfer pXfer, / The transfer context */
411	int rid, /* record id of the file to send */
412	int isPrivate, /* True if rid is a private artifact */
413	Blob pContent, / The content of the file to send */
414	Blob pUuid / The HASH of the file to send */
415	){
416	static const char *const azQuery[] = {
417	"SELECT pid FROM plink x"
418	" WHERE cid=%d"
419	" AND NOT EXISTS(SELECT 1 FROM phantom WHERE rid=pid)",
	@@ -469,11 +463,11 @@
463	*/
464	static int send_delta_native(
465	Xfer pXfer, / The transfer context */
466	int rid, /* record id of the file to send */
467	int isPrivate, /* True if rid is a private artifact */
468	Blob pUuid / The HASH of the file to send */
469	){
470	Blob src, delta;
471	int size = 0;
472	int srcId;
473
	@@ -500,15 +494,26 @@
494	}else{
495	size = 0;
496	}
497	return size;
498	}
499
500	/*
501	** Push an error message to alert the older client that the repository
502	** has SHA3 content and cannot be synced or cloned.
503	*/
504	static void xfer_cannot_send_sha3_error(Xfer *pXfer){
505	blob_appendf(pXfer->pOut,
506	"error Fossil\\sversion\\s2.0\\sor\\slater\\srequired.\n"
507	);
508	}
509
510
511	/*
512	** Send the file identified by rid.
513	**
514	** The pUuid can be NULL in which case the correct hash is computed
515	** from the rid.
516	**
517	** Try to send the file as a native delta if nativeDelta is true, or
518	** as a parent delta if nativeDelta is false.
519	**
	@@ -527,10 +532,14 @@
532	}
533	blob_zero(&uuid);
534	db_blob(&uuid, "SELECT uuid FROM blob WHERE rid=%d AND size>=0", rid);
535	if( blob_size(&uuid)==0 ){
536	return;
537	}
538	if( blob_size(&uuid)>HNAME_LEN_SHA1 && pXfer->clientVersion<20000 ){
539	xfer_cannot_send_sha3_error(pXfer);
540	return;
541	}
542	if( pUuid ){
543	if( blob_compare(pUuid, &uuid)!=0 ){
544	blob_reset(&uuid);
545	return;
	@@ -617,10 +626,15 @@
626	szC = db_column_bytes(&q1, 2);
627	zContent = db_column_raw(&q1, 2);
628	srcIsPrivate = db_column_int(&q1, 3);
629	zDelta = db_column_text(&q1, 4);
630	if( isPrivate ) blob_append(pXfer->pOut, "private\n", -1);
631	if( pXfer->clientVersion<20000 && db_column_bytes(&q1,0)!=HNAME_LEN_SHA1 ){
632	xfer_cannot_send_sha3_error(pXfer);
633	db_reset(&q1);
634	return;
635	}
636	blob_appendf(pXfer->pOut, "cfile %s ", zUuid);
637	if( !isPrivate && srcIsPrivate ){
638	content_get(rid, &fullContent);
639	szU = blob_size(&fullContent);
640	blob_compress(&fullContent, &fullContent);
	@@ -676,10 +690,15 @@
690	);
691	}
692	if( db_step(&q1)==SQLITE_ROW ){
693	sqlite3_int64 mtime = db_column_int64(&q1, 0);
694	const char *zHash = db_column_text(&q1, 1);
695	if( pXfer->clientVersion<20000 && db_column_bytes(&q1,1)>HNAME_LEN_SHA1 ){
696	xfer_cannot_send_sha3_error(pXfer);
697	db_reset(&q1);
698	return;
699	}
700	if( blob_size(pXfer->pOut)>=pXfer->mxSend ){
701	/* If we have already reached the send size limit, send a (short)
702	** uvigot card rather than a uvfile card. This only happens on the
703	** server side. The uvigot card will provoke the client to resend
704	** another uvgimme on the next cycle. */
	@@ -728,23 +747,23 @@
747	}
748	db_finalize(&q);
749	}
750
751	/*
752	** Compute an hash on the tail of pMsg. Verify that it matches the
753	** the hash given in pHash. Return non-zero for an error and 0 on success.
754	**
755	** The type of hash computed (SHA1, SHA3-256) is determined by
756	** the length of the input hash in pHash.
757	*/
758	static int check_tail_hash(Blob pHash, Blob pMsg){
759	Blob tail;

760	int rc;
761	blob_tail(pMsg, &tail);
762	rc = hname_verify_hash(&tail, blob_buffer(pHash), blob_size(pHash));


763	blob_reset(&tail);
764	return rc==HNAME_ERROR;
765	}
766
767	/*
768	** Check the signature on an application/x-fossil payload received by
769	** the HTTP server. The signature is a line of the following form:
	@@ -1198,12 +1217,12 @@
1217	while( blob_line(xfer.pIn, &xfer.line) ){
1218	if( blob_buffer(&xfer.line)[0]=='#' ) continue;
1219	if( blob_size(&xfer.line)==0 ) continue;
1220	xfer.nToken = blob_tokenize(&xfer.line, xfer.aToken, count(xfer.aToken));
1221
1222	/* file HASH SIZE \n CONTENT
1223	** file HASH DELTASRC SIZE \n CONTENT
1224	**
1225	** Accept a file from the client.
1226	*/
1227	if( blob_eq(&xfer.aToken[0], "file") ){
1228	if( !isPush ){
	@@ -1219,12 +1238,12 @@
1238	nErr++;
1239	break;
1240	}
1241	}else
1242
1243	/* cfile HASH USIZE CSIZE \n CONTENT
1244	** cfile HASH DELTASRC USIZE CSIZE \n CONTENT
1245	**
1246	** Accept a file from the client.
1247	*/
1248	if( blob_eq(&xfer.aToken[0], "cfile") ){
1249	if( !isPush ){
	@@ -1254,17 +1273,17 @@
1273	nErr++;
1274	break;
1275	}
1276	}else
1277
1278	/* gimme HASH
1279	**
1280	** Client is requesting a file. Send it.
1281	*/
1282	if( blob_eq(&xfer.aToken[0], "gimme")
1283	&& xfer.nToken==2
1284	&& blob_is_hname(&xfer.aToken[1])
1285	){
1286	nGimme++;
1287	if( isPull ){
1288	int rid = rid_from_uuid(&xfer.aToken[1], 0, 0);
1289	if( rid ){
	@@ -1282,18 +1301,18 @@
1301	&& blob_is_filename(&xfer.aToken[1])
1302	){
1303	send_unversioned_file(&xfer, blob_str(&xfer.aToken[1]), 0);
1304	}else
1305
1306	/* igot HASH ?ISPRIVATE?
1307	**
1308	** Client announces that it has a particular file. If the ISPRIVATE
1309	** argument exists and is non-zero, then the file is a private file.
1310	*/
1311	if( xfer.nToken>=2
1312	&& blob_eq(&xfer.aToken[0], "igot")
1313	&& blob_is_hname(&xfer.aToken[1])
1314	){
1315	if( isPush ){
1316	if( xfer.nToken==2 \|\| blob_eq(&xfer.aToken[2],"1")==0 ){
1317	rid_from_uuid(&xfer.aToken[1], 1, 0);
1318	}else if( g.perm.Private ){
	@@ -1307,16 +1326,15 @@
1326
1327	/* pull SERVERCODE PROJECTCODE
1328	** push SERVERCODE PROJECTCODE
1329	**
1330	** The client wants either send or receive. The server should
1331	** verify that the project code matches. The server code is ignored.
1332	*/
1333	if( xfer.nToken==3
1334	&& (blob_eq(&xfer.aToken[0], "pull") \|\| blob_eq(&xfer.aToken[0], "push"))
1335	&& blob_is_hname(&xfer.aToken[2])

1336	){
1337	const char *zPCode;
1338	zPCode = db_get("project-code", 0);
1339	if( zPCode==0 ){
1340	fossil_panic("missing project code");
	@@ -1534,19 +1552,27 @@
1552	** Send igot cards for all known artifacts.
1553	*/
1554	if( blob_eq(&xfer.aToken[1], "send-catalog") ){
1555	xfer.resync = 0x7fffffff;
1556	}
1557
1558	/* pragma client-version VERSION
1559	**
1560	** Let the server know what version of Fossil is running on the client.
1561	*/
1562	if( xfer.nToken>=3 && blob_eq(&xfer.aToken[1], "client-version") ){
1563	xfer.clientVersion = atoi(blob_str(&xfer.aToken[2]));
1564	}
1565
1566	/* pragma uv-hash HASH
1567	**
1568	** The client wants to make sure that unversioned files are all synced.
1569	** If the HASH does not match, send a complete catalog of
1570	** "uvigot" cards.
1571	*/
1572	if( blob_eq(&xfer.aToken[1], "uv-hash")
1573	&& blob_is_hname(&xfer.aToken[2])
1574	){
1575	if( !uvCatalogSent ){
1576	if( g.perm.Read && g.perm.WrUnver ){
1577	@ pragma uv-push-ok
1578	send_unversioned_catalog(&xfer);
	@@ -1742,10 +1768,11 @@
1768	memset(&xfer, 0, sizeof(xfer));
1769	xfer.pIn = &recv;
1770	xfer.pOut = &send;
1771	xfer.mxSend = db_get_int("max-upload", 250000);
1772	xfer.maxTime = -1;
1773	xfer.clientVersion = RELEASE_VERSION_NUMBER;
1774	if( syncFlags & SYNC_PRIVATE ){
1775	g.perm.Private = 1;
1776	xfer.syncPrivate = 1;
1777	}
1778
	@@ -1783,10 +1810,11 @@
1810	}
1811
1812	/*
1813	** Always begin with a clone, pull, or push message
1814	*/
1815	blob_appendf(&send, "pragma client-version %d\n", RELEASE_VERSION_NUMBER);
1816	if( syncFlags & SYNC_CLONE ){
1817	blob_appendf(&send, "clone 3 %d\n", cloneSeqno);
1818	syncFlags &= ~(SYNC_PUSH\|SYNC_PULL);
1819	nCardSent++;
1820	/* TBD: Request all transferable configuration values */
	@@ -1819,10 +1847,11 @@
1847	db_record_repository_filename(0);
1848	db_multi_exec(
1849	"CREATE TEMP TABLE onremote(rid INTEGER PRIMARY KEY);"
1850	);
1851	manifest_crosslink_begin();
1852
1853
1854	/* Send back the most recently received cookie. Let the server
1855	** figure out if this is a cookie that it cares about.
1856	*/
1857	zCookie = db_get("cookie", 0);
	@@ -1976,10 +2005,11 @@
2005	xfer.nGimmeSent = 0;
2006	xfer.nIGotSent = 0;
2007
2008	lastPctDone = -1;
2009	blob_reset(&send);
2010	blob_appendf(&send, "pragma client-version %d\n", RELEASE_VERSION_NUMBER);
2011	rArrivalTime = db_double(0.0, "SELECT julianday('now')");
2012
2013	/* Send the send-private pragma if we are trying to sync private data */
2014	if( syncFlags & SYNC_PRIVATE ){
2015	blob_append(&send, "pragma send-private\n", -1);
	@@ -2028,22 +2058,22 @@
2058	lastPctDone = pctDone;
2059	fflush(stdout);
2060	}
2061	}
2062
2063	/* file HASH SIZE \n CONTENT
2064	** file HASH DELTASRC SIZE \n CONTENT
2065	**
2066	** Receive a file transmitted from the server.
2067	*/
2068	if( blob_eq(&xfer.aToken[0],"file") ){
2069	xfer_accept_file(&xfer, (syncFlags & SYNC_CLONE)!=0, 0, 0);
2070	nArtifactRcvd++;
2071	}else
2072
2073	/* cfile HASH USIZE CSIZE \n CONTENT
2074	** cfile HASH DELTASRC USIZE CSIZE \n CONTENT
2075	**
2076	** Receive a compressed file transmitted from the server.
2077	*/
2078	if( blob_eq(&xfer.aToken[0],"cfile") ){
2079	xfer_accept_compressed_file(&xfer, 0, 0);
	@@ -2062,27 +2092,27 @@
2092	fossil_print("\rUnversioned-file received: %s\n",
2093	blob_str(&xfer.aToken[1]));
2094	}
2095	}else
2096
2097	/* gimme HASH
2098	**
2099	** Server is requesting a file. If the file is a manifest, assume
2100	** that the server will also want to know all of the content files
2101	** associated with the manifest and send those too.
2102	*/
2103	if( blob_eq(&xfer.aToken[0], "gimme")
2104	&& xfer.nToken==2
2105	&& blob_is_hname(&xfer.aToken[1])
2106	){
2107	if( syncFlags & SYNC_PUSH ){
2108	int rid = rid_from_uuid(&xfer.aToken[1], 0, 0);
2109	if( rid ) send_file(&xfer, rid, &xfer.aToken[1], 0);
2110	}
2111	}else
2112
2113	/* igot HASH ?PRIVATEFLAG?
2114	**
2115	** Server announces that it has a particular file. If this is
2116	** not a file that we have and we are pulling, then create a
2117	** phantom to cause this file to be requested on the next cycle.
2118	** Always remember that the server has this file so that we do
	@@ -2092,11 +2122,11 @@
2122	** private. Pretend it does not exists if we are not pulling
2123	** private files.
2124	*/
2125	if( xfer.nToken>=2
2126	&& blob_eq(&xfer.aToken[0], "igot")
2127	&& blob_is_hname(&xfer.aToken[1])
2128	){
2129	int rid;
2130	int isPriv = xfer.nToken>=3 && blob_eq(&xfer.aToken[2],"1");
2131	rid = rid_from_uuid(&xfer.aToken[1], 0, 0);
2132	if( rid>0 ){
	@@ -2126,11 +2156,11 @@
2156	if( xfer.nToken==5
2157	&& blob_eq(&xfer.aToken[0], "uvigot")
2158	&& blob_is_filename(&xfer.aToken[1])
2159	&& blob_is_int64(&xfer.aToken[2], &mtime)
2160	&& blob_is_int(&xfer.aToken[4], &size)
2161	&& (blob_eq(&xfer.aToken[3],"-") \|\| blob_is_hname(&xfer.aToken[3]))
2162	){
2163	const char *zName = blob_str(&xfer.aToken[1]);
2164	const char *zHash = blob_str(&xfer.aToken[3]);
2165	int iStatus;
2166	iStatus = unversioned_status(zName, mtime, zHash);
	@@ -2188,11 +2218,11 @@
2218	** the client what product to use for the new database.
2219	*/
2220	if( blob_eq(&xfer.aToken[0],"push")
2221	&& xfer.nToken==3
2222	&& (syncFlags & SYNC_CLONE)!=0
2223	&& blob_is_hname(&xfer.aToken[2])
2224	){
2225	if( zPCode==0 ){
2226	zPCode = mprintf("%b", &xfer.aToken[2]);
2227	db_set("project-code", zPCode, 0);
2228	}
2229

M src/zip.c

+2 -5

		--- src/zip.c
		+++ src/zip.c
		@@ -339,11 +339,11 @@
339	339	content_get(rid, &mfile);
340	340	if( blob_size(&mfile)==0 ){
341	341	blob_zero(pZip);
342	342	return;
343	343	}
344		- blob_zero(&hash);
	344	+ blob_set_dynamic(&hash, rid_to_uuid(rid));
345	345	blob_zero(&filename);
346	346	zip_open();
347	347
348	348	if( zDir && zDir[0] ){
349	349	blob_appendf(&filename, "%s/", zDir);
		@@ -378,13 +378,10 @@
378	378	if( eflg & MFESTFLG_RAW ){
379	379	blob_append(&filename, "manifest", -1);
380	380	zName = blob_str(&filename);
381	381	zip_add_folders(zName);
382	382	}
383		- if( eflg & MFESTFLG_UUID ){
384		- sha1sum_blob(&mfile, &hash);
385		- }
386	383	if( eflg & MFESTFLG_RAW ){
387	384	sterilize_manifest(&mfile);
388	385	zip_add_file(zName, &mfile, 0);
389	386	}
390	387	}
		@@ -394,11 +391,10 @@
394	391	blob_resize(&filename, nPrefix);
395	392	blob_append(&filename, "manifest.uuid", -1);
396	393	zName = blob_str(&filename);
397	394	zip_add_folders(zName);
398	395	zip_add_file(zName, &hash, 0);
399		- blob_reset(&hash);
400	396	}
401	397	if( eflg & MFESTFLG_TAGS ){
402	398	Blob tagslist;
403	399	blob_zero(&tagslist);
404	400	get_checkin_taglist(rid, &tagslist);
		@@ -429,10 +425,11 @@
429	425	}else{
430	426	blob_reset(&mfile);
431	427	}
432	428	manifest_destroy(pManifest);
433	429	blob_reset(&filename);
	430	+ blob_reset(&hash);
434	431	zip_close(pZip);
435	432	}
436	433
437	434	/*
438	435	** COMMAND: zip*
439	436

	--- src/zip.c
	+++ src/zip.c
	@@ -339,11 +339,11 @@
339	content_get(rid, &mfile);
340	if( blob_size(&mfile)==0 ){
341	blob_zero(pZip);
342	return;
343	}
344	blob_zero(&hash);
345	blob_zero(&filename);
346	zip_open();
347
348	if( zDir && zDir[0] ){
349	blob_appendf(&filename, "%s/", zDir);
	@@ -378,13 +378,10 @@
378	if( eflg & MFESTFLG_RAW ){
379	blob_append(&filename, "manifest", -1);
380	zName = blob_str(&filename);
381	zip_add_folders(zName);
382	}
383	if( eflg & MFESTFLG_UUID ){
384	sha1sum_blob(&mfile, &hash);
385	}
386	if( eflg & MFESTFLG_RAW ){
387	sterilize_manifest(&mfile);
388	zip_add_file(zName, &mfile, 0);
389	}
390	}
	@@ -394,11 +391,10 @@
394	blob_resize(&filename, nPrefix);
395	blob_append(&filename, "manifest.uuid", -1);
396	zName = blob_str(&filename);
397	zip_add_folders(zName);
398	zip_add_file(zName, &hash, 0);
399	blob_reset(&hash);
400	}
401	if( eflg & MFESTFLG_TAGS ){
402	Blob tagslist;
403	blob_zero(&tagslist);
404	get_checkin_taglist(rid, &tagslist);
	@@ -429,10 +425,11 @@
429	}else{
430	blob_reset(&mfile);
431	}
432	manifest_destroy(pManifest);
433	blob_reset(&filename);

434	zip_close(pZip);
435	}
436
437	/*
438	** COMMAND: zip*
439

	--- src/zip.c
	+++ src/zip.c
	@@ -339,11 +339,11 @@
339	content_get(rid, &mfile);
340	if( blob_size(&mfile)==0 ){
341	blob_zero(pZip);
342	return;
343	}
344	blob_set_dynamic(&hash, rid_to_uuid(rid));
345	blob_zero(&filename);
346	zip_open();
347
348	if( zDir && zDir[0] ){
349	blob_appendf(&filename, "%s/", zDir);
	@@ -378,13 +378,10 @@
378	if( eflg & MFESTFLG_RAW ){
379	blob_append(&filename, "manifest", -1);
380	zName = blob_str(&filename);
381	zip_add_folders(zName);
382	}



383	if( eflg & MFESTFLG_RAW ){
384	sterilize_manifest(&mfile);
385	zip_add_file(zName, &mfile, 0);
386	}
387	}
	@@ -394,11 +391,10 @@
391	blob_resize(&filename, nPrefix);
392	blob_append(&filename, "manifest.uuid", -1);
393	zName = blob_str(&filename);
394	zip_add_folders(zName);
395	zip_add_file(zName, &hash, 0);

396	}
397	if( eflg & MFESTFLG_TAGS ){
398	Blob tagslist;
399	blob_zero(&tagslist);
400	get_checkin_taglist(rid, &tagslist);
	@@ -429,10 +425,11 @@
425	}else{
426	blob_reset(&mfile);
427	}
428	manifest_destroy(pManifest);
429	blob_reset(&filename);
430	blob_reset(&hash);
431	zip_close(pZip);
432	}
433
434	/*
435	** COMMAND: zip*
436

M test/json.test

+4 -4

		--- test/json.test
		+++ test/json.test
		@@ -749,14 +749,14 @@
749	749
750	750
751	751	# FOSSIL-1103 FSL_JSON_E_UNKNOWN
752	752	# Unknown error
753	753
754		-write_file bad.sql {
755		-CREATE TABLE spam(a integer, b text);
756		-}
757		-exec $::fossilexe sqlite3 --no-repository bad.fossil <bad.sql
	754	+#write_file bad.sql {
	755	+#CREATE TABLE spam(a integer, b text);
	756	+#}
	757	+#exec $::fossilexe sqlite3 --no-repository bad.fossil <bad.sql
758	758	#fossil_json HAI -R bad.fossil -expectError
759	759
760	760	# FOSSIL-1104 FSL_JSON_E_TIMEOUT
761	761	# Timeout reached
762	762	# FOSSIL-1105 FSL_JSON_E_ASSERT
763	763

	--- test/json.test
	+++ test/json.test
	@@ -749,14 +749,14 @@
749
750
751	# FOSSIL-1103 FSL_JSON_E_UNKNOWN
752	# Unknown error
753
754	write_file bad.sql {
755	CREATE TABLE spam(a integer, b text);
756	}
757	exec $::fossilexe sqlite3 --no-repository bad.fossil <bad.sql
758	#fossil_json HAI -R bad.fossil -expectError
759
760	# FOSSIL-1104 FSL_JSON_E_TIMEOUT
761	# Timeout reached
762	# FOSSIL-1105 FSL_JSON_E_ASSERT
763

	--- test/json.test
	+++ test/json.test
	@@ -749,14 +749,14 @@
749
750
751	# FOSSIL-1103 FSL_JSON_E_UNKNOWN
752	# Unknown error
753
754	#write_file bad.sql {
755	#CREATE TABLE spam(a integer, b text);
756	#}
757	#exec $::fossilexe sqlite3 --no-repository bad.fossil <bad.sql
758	#fossil_json HAI -R bad.fossil -expectError
759
760	# FOSSIL-1104 FSL_JSON_E_TIMEOUT
761	# Timeout reached
762	# FOSSIL-1105 FSL_JSON_E_ASSERT
763

M test/wiki.test

+12 -6

		--- test/wiki.test
		+++ test/wiki.test
		@@ -21,16 +21,22 @@
21	21	test_setup
22	22
23	23	# Return true if two files are similar (i.e. not only compress trailing spaces
24	24	# from a line, but remove any final LF from the file as well)
25	25	proc similar_file {a b} {
26		- set x [read_file $a]
27		- regsub -all { +\n} $x \n x
28		- regsub -all {\n$} $x {} x
29		- set y [read_file $b]
30		- regsub -all { +\n} $y \n y
31		- regsub -all {\n$} $y {} y
	26	+ set x ""
	27	+ if {[file exists $a]} {
	28	+ set x [read_file $a]
	29	+ regsub -all { +\n} $x \n x
	30	+ regsub -all {\n$} $x {} x
	31	+ }
	32	+ set y ""
	33	+ if {[file exists $b]} {
	34	+ set y [read_file $b]
	35	+ regsub -all { +\n} $y \n y
	36	+ regsub -all {\n$} $y {} y
	37	+ }
32	38	return [expr {$x==$y}]
33	39	}
34	40
35	41	# Return the mime type in the manifest for a given wiki page
36	42	# Defaults to "error: some text" if the manifest can't be located and
37	43

	--- test/wiki.test
	+++ test/wiki.test
	@@ -21,16 +21,22 @@
21	test_setup
22
23	# Return true if two files are similar (i.e. not only compress trailing spaces
24	# from a line, but remove any final LF from the file as well)
25	proc similar_file {a b} {
26	set x [read_file $a]
27	regsub -all { +\n} $x \n x
28	regsub -all {\n$} $x {} x
29	set y [read_file $b]
30	regsub -all { +\n} $y \n y
31	regsub -all {\n$} $y {} y






32	return [expr {$x==$y}]
33	}
34
35	# Return the mime type in the manifest for a given wiki page
36	# Defaults to "error: some text" if the manifest can't be located and
37

	--- test/wiki.test
	+++ test/wiki.test
	@@ -21,16 +21,22 @@
21	test_setup
22
23	# Return true if two files are similar (i.e. not only compress trailing spaces
24	# from a line, but remove any final LF from the file as well)
25	proc similar_file {a b} {
26	set x ""
27	if {[file exists $a]} {
28	set x [read_file $a]
29	regsub -all { +\n} $x \n x
30	regsub -all {\n$} $x {} x
31	}
32	set y ""
33	if {[file exists $b]} {
34	set y [read_file $b]
35	regsub -all { +\n} $y \n y
36	regsub -all {\n$} $y {} y
37	}
38	return [expr {$x==$y}]
39	}
40
41	# Return the mime type in the manifest for a given wiki page
42	# Defaults to "error: some text" if the manifest can't be located and
43

M tools/man_page_command_list.tcl

+1 -1

		--- tools/man_page_command_list.tcl
		+++ tools/man_page_command_list.tcl
		@@ -1,7 +1,7 @@
1	1	#!/usr/bin/env tclsh
2		-# command_man_list.tcl - generates common command list for fossil.1
	2	+# man_page_command_list.tcl - generates common command list for fossil.1
3	3
4	4	# Tunable configuration.
5	5	set columns 5
6	6	set width 15
7	7
8	8

	--- tools/man_page_command_list.tcl
	+++ tools/man_page_command_list.tcl
	@@ -1,7 +1,7 @@
1	#!/usr/bin/env tclsh
2	# command_man_list.tcl - generates common command list for fossil.1
3
4	# Tunable configuration.
5	set columns 5
6	set width 15
7
8

	--- tools/man_page_command_list.tcl
	+++ tools/man_page_command_list.tcl
	@@ -1,7 +1,7 @@
1	#!/usr/bin/env tclsh
2	# man_page_command_list.tcl - generates common command list for fossil.1
3
4	# Tunable configuration.
5	set columns 5
6	set width 15
7
8

M win/Makefile.dmc

+22 -4

		--- win/Makefile.dmc
		+++ win/Makefile.dmc
		@@ -28,13 +28,13 @@
28	28
29	29	SQLITE_OPTIONS = -DNDEBUG=1 -DSQLITE_THREADSAFE=0 -DSQLITE_DEFAULT_MEMSTATUS=0 -DSQLITE_DEFAULT_WAL_SYNCHRONOUS=1 -DSQLITE_LIKE_DOESNT_MATCH_BLOBS -DSQLITE_OMIT_DECLTYPE -DSQLITE_OMIT_DEPRECATED -DSQLITE_OMIT_PROGRESS_CALLBACK -DSQLITE_OMIT_SHARED_CACHE -DSQLITE_OMIT_LOAD_EXTENSION -DSQLITE_MAX_EXPR_DEPTH=0 -DSQLITE_USE_ALLOCA -DSQLITE_ENABLE_LOCKING_STYLE=0 -DSQLITE_DEFAULT_FILE_FORMAT=4 -DSQLITE_ENABLE_EXPLAIN_COMMENTS -DSQLITE_ENABLE_FTS4 -DSQLITE_ENABLE_FTS3_PARENTHESIS -DSQLITE_ENABLE_DBSTAT_VTAB -DSQLITE_ENABLE_JSON1 -DSQLITE_ENABLE_FTS5
30	30
31	31	SHELL_OPTIONS = -Dmain=sqlite3_shell -DSQLITE_SHELL_IS_UTF8=1 -DSQLITE_OMIT_LOAD_EXTENSION=1 -DUSE_SYSTEM_SQLITE=$(USE_SYSTEM_SQLITE) -DSQLITE_SHELL_DBNAME_PROC=fossil_open -Daccess=file_access -Dsystem=fossil_system -Dgetenv=fossil_getenv -Dfopen=fossil_fopen
32	32
33		-SRC = add_.c allrepo_.c attach_.c bag_.c bisect_.c blob_.c branch_.c browse_.c builtin_.c bundle_.c cache_.c captcha_.c cgi_.c checkin_.c checkout_.c clearsign_.c clone_.c comformat_.c configure_.c content_.c db_.c delta_.c deltacmd_.c descendants_.c diff_.c diffcmd_.c dispatch_.c doc_.c encode_.c event_.c export_.c file_.c finfo_.c foci_.c fshell_.c fusefs_.c glob_.c graph_.c gzip_.c http_.c http_socket_.c http_ssl_.c http_transport_.c import_.c info_.c json_.c json_artifact_.c json_branch_.c json_config_.c json_diff_.c json_dir_.c json_finfo_.c json_login_.c json_query_.c json_report_.c json_status_.c json_tag_.c json_timeline_.c json_user_.c json_wiki_.c leaf_.c loadctrl_.c login_.c lookslike_.c main_.c manifest_.c markdown_.c markdown_html_.c md5_.c merge_.c merge3_.c moderate_.c name_.c path_.c piechart_.c pivot_.c popen_.c pqueue_.c printf_.c publish_.c purge_.c rebuild_.c regexp_.c report_.c rss_.c schema_.c search_.c setup_.c sha1_.c shun_.c sitemap_.c skins_.c sqlcmd_.c stash_.c stat_.c statrep_.c style_.c sync_.c tag_.c tar_.c th_main_.c timeline_.c tkt_.c tktsetup_.c undo_.c unicode_.c unversioned_.c update_.c url_.c user_.c utf8_.c util_.c verify_.c vfile_.c wiki_.c wikiformat_.c winfile_.c winhttp_.c wysiwyg_.c xfer_.c xfersetup_.c zip_.c
	33	+SRC = add_.c allrepo_.c attach_.c bag_.c bisect_.c blob_.c branch_.c browse_.c builtin_.c bundle_.c cache_.c captcha_.c cgi_.c checkin_.c checkout_.c clearsign_.c clone_.c comformat_.c configure_.c content_.c db_.c delta_.c deltacmd_.c descendants_.c diff_.c diffcmd_.c dispatch_.c doc_.c encode_.c event_.c export_.c file_.c finfo_.c foci_.c fshell_.c fusefs_.c glob_.c graph_.c gzip_.c hname_.c http_.c http_socket_.c http_ssl_.c http_transport_.c import_.c info_.c json_.c json_artifact_.c json_branch_.c json_config_.c json_diff_.c json_dir_.c json_finfo_.c json_login_.c json_query_.c json_report_.c json_status_.c json_tag_.c json_timeline_.c json_user_.c json_wiki_.c leaf_.c loadctrl_.c login_.c lookslike_.c main_.c manifest_.c markdown_.c markdown_html_.c md5_.c merge_.c merge3_.c moderate_.c name_.c path_.c piechart_.c pivot_.c popen_.c pqueue_.c printf_.c publish_.c purge_.c rebuild_.c regexp_.c report_.c rss_.c schema_.c search_.c setup_.c sha1_.c sha1hard_.c sha3_.c shun_.c sitemap_.c skins_.c sqlcmd_.c stash_.c stat_.c statrep_.c style_.c sync_.c tag_.c tar_.c th_main_.c timeline_.c tkt_.c tktsetup_.c undo_.c unicode_.c unversioned_.c update_.c url_.c user_.c utf8_.c util_.c verify_.c vfile_.c wiki_.c wikiformat_.c winfile_.c winhttp_.c wysiwyg_.c xfer_.c xfersetup_.c zip_.c
34	34
35		-OBJ = $(OBJDIR)\add$O $(OBJDIR)\allrepo$O $(OBJDIR)\attach$O $(OBJDIR)\bag$O $(OBJDIR)\bisect$O $(OBJDIR)\blob$O $(OBJDIR)\branch$O $(OBJDIR)\browse$O $(OBJDIR)\builtin$O $(OBJDIR)\bundle$O $(OBJDIR)\cache$O $(OBJDIR)\captcha$O $(OBJDIR)\cgi$O $(OBJDIR)\checkin$O $(OBJDIR)\checkout$O $(OBJDIR)\clearsign$O $(OBJDIR)\clone$O $(OBJDIR)\comformat$O $(OBJDIR)\configure$O $(OBJDIR)\content$O $(OBJDIR)\db$O $(OBJDIR)\delta$O $(OBJDIR)\deltacmd$O $(OBJDIR)\descendants$O $(OBJDIR)\diff$O $(OBJDIR)\diffcmd$O $(OBJDIR)\dispatch$O $(OBJDIR)\doc$O $(OBJDIR)\encode$O $(OBJDIR)\event$O $(OBJDIR)\export$O $(OBJDIR)\file$O $(OBJDIR)\finfo$O $(OBJDIR)\foci$O $(OBJDIR)\fshell$O $(OBJDIR)\fusefs$O $(OBJDIR)\glob$O $(OBJDIR)\graph$O $(OBJDIR)\gzip$O $(OBJDIR)\http$O $(OBJDIR)\http_socket$O $(OBJDIR)\http_ssl$O $(OBJDIR)\http_transport$O $(OBJDIR)\import$O $(OBJDIR)\info$O $(OBJDIR)\json$O $(OBJDIR)\json_artifact$O $(OBJDIR)\json_branch$O $(OBJDIR)\json_config$O $(OBJDIR)\json_diff$O $(OBJDIR)\json_dir$O $(OBJDIR)\json_finfo$O $(OBJDIR)\json_login$O $(OBJDIR)\json_query$O $(OBJDIR)\json_report$O $(OBJDIR)\json_status$O $(OBJDIR)\json_tag$O $(OBJDIR)\json_timeline$O $(OBJDIR)\json_user$O $(OBJDIR)\json_wiki$O $(OBJDIR)\leaf$O $(OBJDIR)\loadctrl$O $(OBJDIR)\login$O $(OBJDIR)\lookslike$O $(OBJDIR)\main$O $(OBJDIR)\manifest$O $(OBJDIR)\markdown$O $(OBJDIR)\markdown_html$O $(OBJDIR)\md5$O $(OBJDIR)\merge$O $(OBJDIR)\merge3$O $(OBJDIR)\moderate$O $(OBJDIR)\name$O $(OBJDIR)\path$O $(OBJDIR)\piechart$O $(OBJDIR)\pivot$O $(OBJDIR)\popen$O $(OBJDIR)\pqueue$O $(OBJDIR)\printf$O $(OBJDIR)\publish$O $(OBJDIR)\purge$O $(OBJDIR)\rebuild$O $(OBJDIR)\regexp$O $(OBJDIR)\report$O $(OBJDIR)\rss$O $(OBJDIR)\schema$O $(OBJDIR)\search$O $(OBJDIR)\setup$O $(OBJDIR)\sha1$O $(OBJDIR)\shun$O $(OBJDIR)\sitemap$O $(OBJDIR)\skins$O $(OBJDIR)\sqlcmd$O $(OBJDIR)\stash$O $(OBJDIR)\stat$O $(OBJDIR)\statrep$O $(OBJDIR)\style$O $(OBJDIR)\sync$O $(OBJDIR)\tag$O $(OBJDIR)\tar$O $(OBJDIR)\th_main$O $(OBJDIR)\timeline$O $(OBJDIR)\tkt$O $(OBJDIR)\tktsetup$O $(OBJDIR)\undo$O $(OBJDIR)\unicode$O $(OBJDIR)\unversioned$O $(OBJDIR)\update$O $(OBJDIR)\url$O $(OBJDIR)\user$O $(OBJDIR)\utf8$O $(OBJDIR)\util$O $(OBJDIR)\verify$O $(OBJDIR)\vfile$O $(OBJDIR)\wiki$O $(OBJDIR)\wikiformat$O $(OBJDIR)\winfile$O $(OBJDIR)\winhttp$O $(OBJDIR)\wysiwyg$O $(OBJDIR)\xfer$O $(OBJDIR)\xfersetup$O $(OBJDIR)\zip$O $(OBJDIR)\shell$O $(OBJDIR)\sqlite3$O $(OBJDIR)\th$O $(OBJDIR)\th_lang$O
	35	+OBJ = $(OBJDIR)\add$O $(OBJDIR)\allrepo$O $(OBJDIR)\attach$O $(OBJDIR)\bag$O $(OBJDIR)\bisect$O $(OBJDIR)\blob$O $(OBJDIR)\branch$O $(OBJDIR)\browse$O $(OBJDIR)\builtin$O $(OBJDIR)\bundle$O $(OBJDIR)\cache$O $(OBJDIR)\captcha$O $(OBJDIR)\cgi$O $(OBJDIR)\checkin$O $(OBJDIR)\checkout$O $(OBJDIR)\clearsign$O $(OBJDIR)\clone$O $(OBJDIR)\comformat$O $(OBJDIR)\configure$O $(OBJDIR)\content$O $(OBJDIR)\db$O $(OBJDIR)\delta$O $(OBJDIR)\deltacmd$O $(OBJDIR)\descendants$O $(OBJDIR)\diff$O $(OBJDIR)\diffcmd$O $(OBJDIR)\dispatch$O $(OBJDIR)\doc$O $(OBJDIR)\encode$O $(OBJDIR)\event$O $(OBJDIR)\export$O $(OBJDIR)\file$O $(OBJDIR)\finfo$O $(OBJDIR)\foci$O $(OBJDIR)\fshell$O $(OBJDIR)\fusefs$O $(OBJDIR)\glob$O $(OBJDIR)\graph$O $(OBJDIR)\gzip$O $(OBJDIR)\hname$O $(OBJDIR)\http$O $(OBJDIR)\http_socket$O $(OBJDIR)\http_ssl$O $(OBJDIR)\http_transport$O $(OBJDIR)\import$O $(OBJDIR)\info$O $(OBJDIR)\json$O $(OBJDIR)\json_artifact$O $(OBJDIR)\json_branch$O $(OBJDIR)\json_config$O $(OBJDIR)\json_diff$O $(OBJDIR)\json_dir$O $(OBJDIR)\json_finfo$O $(OBJDIR)\json_login$O $(OBJDIR)\json_query$O $(OBJDIR)\json_report$O $(OBJDIR)\json_status$O $(OBJDIR)\json_tag$O $(OBJDIR)\json_timeline$O $(OBJDIR)\json_user$O $(OBJDIR)\json_wiki$O $(OBJDIR)\leaf$O $(OBJDIR)\loadctrl$O $(OBJDIR)\login$O $(OBJDIR)\lookslike$O $(OBJDIR)\main$O $(OBJDIR)\manifest$O $(OBJDIR)\markdown$O $(OBJDIR)\markdown_html$O $(OBJDIR)\md5$O $(OBJDIR)\merge$O $(OBJDIR)\merge3$O $(OBJDIR)\moderate$O $(OBJDIR)\name$O $(OBJDIR)\path$O $(OBJDIR)\piechart$O $(OBJDIR)\pivot$O $(OBJDIR)\popen$O $(OBJDIR)\pqueue$O $(OBJDIR)\printf$O $(OBJDIR)\publish$O $(OBJDIR)\purge$O $(OBJDIR)\rebuild$O $(OBJDIR)\regexp$O $(OBJDIR)\report$O $(OBJDIR)\rss$O $(OBJDIR)\schema$O $(OBJDIR)\search$O $(OBJDIR)\setup$O $(OBJDIR)\sha1$O $(OBJDIR)\sha1hard$O $(OBJDIR)\sha3$O $(OBJDIR)\shun$O $(OBJDIR)\sitemap$O $(OBJDIR)\skins$O $(OBJDIR)\sqlcmd$O $(OBJDIR)\stash$O $(OBJDIR)\stat$O $(OBJDIR)\statrep$O $(OBJDIR)\style$O $(OBJDIR)\sync$O $(OBJDIR)\tag$O $(OBJDIR)\tar$O $(OBJDIR)\th_main$O $(OBJDIR)\timeline$O $(OBJDIR)\tkt$O $(OBJDIR)\tktsetup$O $(OBJDIR)\undo$O $(OBJDIR)\unicode$O $(OBJDIR)\unversioned$O $(OBJDIR)\update$O $(OBJDIR)\url$O $(OBJDIR)\user$O $(OBJDIR)\utf8$O $(OBJDIR)\util$O $(OBJDIR)\verify$O $(OBJDIR)\vfile$O $(OBJDIR)\wiki$O $(OBJDIR)\wikiformat$O $(OBJDIR)\winfile$O $(OBJDIR)\winhttp$O $(OBJDIR)\wysiwyg$O $(OBJDIR)\xfer$O $(OBJDIR)\xfersetup$O $(OBJDIR)\zip$O $(OBJDIR)\shell$O $(OBJDIR)\sqlite3$O $(OBJDIR)\th$O $(OBJDIR)\th_lang$O
36	36
37	37
38	38	RC=$(DMDIR)\bin\rcc
39	39	RCFLAGS=-32 -w1 -I$(SRCDIR) /D__DMC__
40	40
		@@ -49,11 +49,11 @@
49	49
50	50	$(OBJDIR)\fossil.res: $B\win\fossil.rc
51	51	$(RC) $(RCFLAGS) -o$@ $**
52	52
53	53	$(OBJDIR)\link: $B\win\Makefile.dmc $(OBJDIR)\fossil.res
54		- +echo add allrepo attach bag bisect blob branch browse builtin bundle cache captcha cgi checkin checkout clearsign clone comformat configure content db delta deltacmd descendants diff diffcmd dispatch doc encode event export file finfo foci fshell fusefs glob graph gzip http http_socket http_ssl http_transport import info json json_artifact json_branch json_config json_diff json_dir json_finfo json_login json_query json_report json_status json_tag json_timeline json_user json_wiki leaf loadctrl login lookslike main manifest markdown markdown_html md5 merge merge3 moderate name path piechart pivot popen pqueue printf publish purge rebuild regexp report rss schema search setup sha1 shun sitemap skins sqlcmd stash stat statrep style sync tag tar th_main timeline tkt tktsetup undo unicode unversioned update url user utf8 util verify vfile wiki wikiformat winfile winhttp wysiwyg xfer xfersetup zip shell sqlite3 th th_lang > $@
	54	+ +echo add allrepo attach bag bisect blob branch browse builtin bundle cache captcha cgi checkin checkout clearsign clone comformat configure content db delta deltacmd descendants diff diffcmd dispatch doc encode event export file finfo foci fshell fusefs glob graph gzip hname http http_socket http_ssl http_transport import info json json_artifact json_branch json_config json_diff json_dir json_finfo json_login json_query json_report json_status json_tag json_timeline json_user json_wiki leaf loadctrl login lookslike main manifest markdown markdown_html md5 merge merge3 moderate name path piechart pivot popen pqueue printf publish purge rebuild regexp report rss schema search setup sha1 sha1hard sha3 shun sitemap skins sqlcmd stash stat statrep style sync tag tar th_main timeline tkt tktsetup undo unicode unversioned update url user utf8 util verify vfile wiki wikiformat winfile winhttp wysiwyg xfer xfersetup zip shell sqlite3 th th_lang > $@
55	55	+echo fossil >> $@
56	56	+echo fossil >> $@
57	57	+echo $(LIBS) >> $@
58	58	+echo. >> $@
59	59	+echo fossil >> $@
		@@ -356,10 +356,16 @@
356	356	$(OBJDIR)\gzip$O : gzip_.c gzip.h
357	357	$(TCC) -o$@ -c gzip_.c
358	358
359	359	gzip_.c : $(SRCDIR)\gzip.c
360	360	+translate$E $** > $@
	361	+
	362	+$(OBJDIR)\hname$O : hname_.c hname.h
	363	+ $(TCC) -o$@ -c hname_.c
	364	+
	365	+hname_.c : $(SRCDIR)\hname.c
	366	+ +translate$E $** > $@
361	367
362	368	$(OBJDIR)\http$O : http_.c http.h
363	369	$(TCC) -o$@ -c http_.c
364	370
365	371	http_.c : $(SRCDIR)\http.c
		@@ -656,10 +662,22 @@
656	662	$(OBJDIR)\sha1$O : sha1_.c sha1.h
657	663	$(TCC) -o$@ -c sha1_.c
658	664
659	665	sha1_.c : $(SRCDIR)\sha1.c
660	666	+translate$E $** > $@
	667	+
	668	+$(OBJDIR)\sha1hard$O : sha1hard_.c sha1hard.h
	669	+ $(TCC) -o$@ -c sha1hard_.c
	670	+
	671	+sha1hard_.c : $(SRCDIR)\sha1hard.c
	672	+ +translate$E $** > $@
	673	+
	674	+$(OBJDIR)\sha3$O : sha3_.c sha3.h
	675	+ $(TCC) -o$@ -c sha3_.c
	676	+
	677	+sha3_.c : $(SRCDIR)\sha3.c
	678	+ +translate$E $** > $@
661	679
662	680	$(OBJDIR)\shun$O : shun_.c shun.h
663	681	$(TCC) -o$@ -c shun_.c
664	682
665	683	shun_.c : $(SRCDIR)\shun.c
		@@ -856,7 +874,7 @@
856	874
857	875	zip_.c : $(SRCDIR)\zip.c
858	876	+translate$E $** > $@
859	877
860	878	headers: makeheaders$E page_index.h builtin_data.h VERSION.h
861		- +makeheaders$E add_.c:add.h allrepo_.c:allrepo.h attach_.c:attach.h bag_.c:bag.h bisect_.c:bisect.h blob_.c:blob.h branch_.c:branch.h browse_.c:browse.h builtin_.c:builtin.h bundle_.c:bundle.h cache_.c:cache.h captcha_.c:captcha.h cgi_.c:cgi.h checkin_.c:checkin.h checkout_.c:checkout.h clearsign_.c:clearsign.h clone_.c:clone.h comformat_.c:comformat.h configure_.c:configure.h content_.c:content.h db_.c:db.h delta_.c:delta.h deltacmd_.c:deltacmd.h descendants_.c:descendants.h diff_.c:diff.h diffcmd_.c:diffcmd.h dispatch_.c:dispatch.h doc_.c:doc.h encode_.c:encode.h event_.c:event.h export_.c:export.h file_.c:file.h finfo_.c:finfo.h foci_.c:foci.h fshell_.c:fshell.h fusefs_.c:fusefs.h glob_.c:glob.h graph_.c:graph.h gzip_.c:gzip.h http_.c:http.h http_socket_.c:http_socket.h http_ssl_.c:http_ssl.h http_transport_.c:http_transport.h import_.c:import.h info_.c:info.h json_.c:json.h json_artifact_.c:json_artifact.h json_branch_.c:json_branch.h json_config_.c:json_config.h json_diff_.c:json_diff.h json_dir_.c:json_dir.h json_finfo_.c:json_finfo.h json_login_.c:json_login.h json_query_.c:json_query.h json_report_.c:json_report.h json_status_.c:json_status.h json_tag_.c:json_tag.h json_timeline_.c:json_timeline.h json_user_.c:json_user.h json_wiki_.c:json_wiki.h leaf_.c:leaf.h loadctrl_.c:loadctrl.h login_.c:login.h lookslike_.c:lookslike.h main_.c:main.h manifest_.c:manifest.h markdown_.c:markdown.h markdown_html_.c:markdown_html.h md5_.c:md5.h merge_.c:merge.h merge3_.c:merge3.h moderate_.c:moderate.h name_.c:name.h path_.c:path.h piechart_.c:piechart.h pivot_.c:pivot.h popen_.c:popen.h pqueue_.c:pqueue.h printf_.c:printf.h publish_.c:publish.h purge_.c:purge.h rebuild_.c:rebuild.h regexp_.c:regexp.h report_.c:report.h rss_.c:rss.h schema_.c:schema.h search_.c:search.h setup_.c:setup.h sha1_.c:sha1.h shun_.c:shun.h sitemap_.c:sitemap.h skins_.c:skins.h sqlcmd_.c:sqlcmd.h stash_.c:stash.h stat_.c:stat.h statrep_.c:statrep.h style_.c:style.h sync_.c:sync.h tag_.c:tag.h tar_.c:tar.h th_main_.c:th_main.h timeline_.c:timeline.h tkt_.c:tkt.h tktsetup_.c:tktsetup.h undo_.c:undo.h unicode_.c:unicode.h unversioned_.c:unversioned.h update_.c:update.h url_.c:url.h user_.c:user.h utf8_.c:utf8.h util_.c:util.h verify_.c:verify.h vfile_.c:vfile.h wiki_.c:wiki.h wikiformat_.c:wikiformat.h winfile_.c:winfile.h winhttp_.c:winhttp.h wysiwyg_.c:wysiwyg.h xfer_.c:xfer.h xfersetup_.c:xfersetup.h zip_.c:zip.h $(SRCDIR)\sqlite3.h $(SRCDIR)\th.h VERSION.h $(SRCDIR)\cson_amalgamation.h
	879	+ +makeheaders$E add_.c:add.h allrepo_.c:allrepo.h attach_.c:attach.h bag_.c:bag.h bisect_.c:bisect.h blob_.c:blob.h branch_.c:branch.h browse_.c:browse.h builtin_.c:builtin.h bundle_.c:bundle.h cache_.c:cache.h captcha_.c:captcha.h cgi_.c:cgi.h checkin_.c:checkin.h checkout_.c:checkout.h clearsign_.c:clearsign.h clone_.c:clone.h comformat_.c:comformat.h configure_.c:configure.h content_.c:content.h db_.c:db.h delta_.c:delta.h deltacmd_.c:deltacmd.h descendants_.c:descendants.h diff_.c:diff.h diffcmd_.c:diffcmd.h dispatch_.c:dispatch.h doc_.c:doc.h encode_.c:encode.h event_.c:event.h export_.c:export.h file_.c:file.h finfo_.c:finfo.h foci_.c:foci.h fshell_.c:fshell.h fusefs_.c:fusefs.h glob_.c:glob.h graph_.c:graph.h gzip_.c:gzip.h hname_.c:hname.h http_.c:http.h http_socket_.c:http_socket.h http_ssl_.c:http_ssl.h http_transport_.c:http_transport.h import_.c:import.h info_.c:info.h json_.c:json.h json_artifact_.c:json_artifact.h json_branch_.c:json_branch.h json_config_.c:json_config.h json_diff_.c:json_diff.h json_dir_.c:json_dir.h json_finfo_.c:json_finfo.h json_login_.c:json_login.h json_query_.c:json_query.h json_report_.c:json_report.h json_status_.c:json_status.h json_tag_.c:json_tag.h json_timeline_.c:json_timeline.h json_user_.c:json_user.h json_wiki_.c:json_wiki.h leaf_.c:leaf.h loadctrl_.c:loadctrl.h login_.c:login.h lookslike_.c:lookslike.h main_.c:main.h manifest_.c:manifest.h markdown_.c:markdown.h markdown_html_.c:markdown_html.h md5_.c:md5.h merge_.c:merge.h merge3_.c:merge3.h moderate_.c:moderate.h name_.c:name.h path_.c:path.h piechart_.c:piechart.h pivot_.c:pivot.h popen_.c:popen.h pqueue_.c:pqueue.h printf_.c:printf.h publish_.c:publish.h purge_.c:purge.h rebuild_.c:rebuild.h regexp_.c:regexp.h report_.c:report.h rss_.c:rss.h schema_.c:schema.h search_.c:search.h setup_.c:setup.h sha1_.c:sha1.h sha1hard_.c:sha1hard.h sha3_.c:sha3.h shun_.c:shun.h sitemap_.c:sitemap.h skins_.c:skins.h sqlcmd_.c:sqlcmd.h stash_.c:stash.h stat_.c:stat.h statrep_.c:statrep.h style_.c:style.h sync_.c:sync.h tag_.c:tag.h tar_.c:tar.h th_main_.c:th_main.h timeline_.c:timeline.h tkt_.c:tkt.h tktsetup_.c:tktsetup.h undo_.c:undo.h unicode_.c:unicode.h unversioned_.c:unversioned.h update_.c:update.h url_.c:url.h user_.c:user.h utf8_.c:utf8.h util_.c:util.h verify_.c:verify.h vfile_.c:vfile.h wiki_.c:wiki.h wikiformat_.c:wikiformat.h winfile_.c:winfile.h winhttp_.c:winhttp.h wysiwyg_.c:wysiwyg.h xfer_.c:xfer.h xfersetup_.c:xfersetup.h zip_.c:zip.h $(SRCDIR)\sqlite3.h $(SRCDIR)\th.h VERSION.h $(SRCDIR)\cson_amalgamation.h
862	880	@copy /Y nul: headers
863	881

	--- win/Makefile.dmc
	+++ win/Makefile.dmc
	@@ -28,13 +28,13 @@
28
29	SQLITE_OPTIONS = -DNDEBUG=1 -DSQLITE_THREADSAFE=0 -DSQLITE_DEFAULT_MEMSTATUS=0 -DSQLITE_DEFAULT_WAL_SYNCHRONOUS=1 -DSQLITE_LIKE_DOESNT_MATCH_BLOBS -DSQLITE_OMIT_DECLTYPE -DSQLITE_OMIT_DEPRECATED -DSQLITE_OMIT_PROGRESS_CALLBACK -DSQLITE_OMIT_SHARED_CACHE -DSQLITE_OMIT_LOAD_EXTENSION -DSQLITE_MAX_EXPR_DEPTH=0 -DSQLITE_USE_ALLOCA -DSQLITE_ENABLE_LOCKING_STYLE=0 -DSQLITE_DEFAULT_FILE_FORMAT=4 -DSQLITE_ENABLE_EXPLAIN_COMMENTS -DSQLITE_ENABLE_FTS4 -DSQLITE_ENABLE_FTS3_PARENTHESIS -DSQLITE_ENABLE_DBSTAT_VTAB -DSQLITE_ENABLE_JSON1 -DSQLITE_ENABLE_FTS5
30
31	SHELL_OPTIONS = -Dmain=sqlite3_shell -DSQLITE_SHELL_IS_UTF8=1 -DSQLITE_OMIT_LOAD_EXTENSION=1 -DUSE_SYSTEM_SQLITE=$(USE_SYSTEM_SQLITE) -DSQLITE_SHELL_DBNAME_PROC=fossil_open -Daccess=file_access -Dsystem=fossil_system -Dgetenv=fossil_getenv -Dfopen=fossil_fopen
32
33	SRC = add_.c allrepo_.c attach_.c bag_.c bisect_.c blob_.c branch_.c browse_.c builtin_.c bundle_.c cache_.c captcha_.c cgi_.c checkin_.c checkout_.c clearsign_.c clone_.c comformat_.c configure_.c content_.c db_.c delta_.c deltacmd_.c descendants_.c diff_.c diffcmd_.c dispatch_.c doc_.c encode_.c event_.c export_.c file_.c finfo_.c foci_.c fshell_.c fusefs_.c glob_.c graph_.c gzip_.c http_.c http_socket_.c http_ssl_.c http_transport_.c import_.c info_.c json_.c json_artifact_.c json_branch_.c json_config_.c json_diff_.c json_dir_.c json_finfo_.c json_login_.c json_query_.c json_report_.c json_status_.c json_tag_.c json_timeline_.c json_user_.c json_wiki_.c leaf_.c loadctrl_.c login_.c lookslike_.c main_.c manifest_.c markdown_.c markdown_html_.c md5_.c merge_.c merge3_.c moderate_.c name_.c path_.c piechart_.c pivot_.c popen_.c pqueue_.c printf_.c publish_.c purge_.c rebuild_.c regexp_.c report_.c rss_.c schema_.c search_.c setup_.c sha1_.c shun_.c sitemap_.c skins_.c sqlcmd_.c stash_.c stat_.c statrep_.c style_.c sync_.c tag_.c tar_.c th_main_.c timeline_.c tkt_.c tktsetup_.c undo_.c unicode_.c unversioned_.c update_.c url_.c user_.c utf8_.c util_.c verify_.c vfile_.c wiki_.c wikiformat_.c winfile_.c winhttp_.c wysiwyg_.c xfer_.c xfersetup_.c zip_.c
34
35	OBJ = $(OBJDIR)\add$O $(OBJDIR)\allrepo$O $(OBJDIR)\attach$O $(OBJDIR)\bag$O $(OBJDIR)\bisect$O $(OBJDIR)\blob$O $(OBJDIR)\branch$O $(OBJDIR)\browse$O $(OBJDIR)\builtin$O $(OBJDIR)\bundle$O $(OBJDIR)\cache$O $(OBJDIR)\captcha$O $(OBJDIR)\cgi$O $(OBJDIR)\checkin$O $(OBJDIR)\checkout$O $(OBJDIR)\clearsign$O $(OBJDIR)\clone$O $(OBJDIR)\comformat$O $(OBJDIR)\configure$O $(OBJDIR)\content$O $(OBJDIR)\db$O $(OBJDIR)\delta$O $(OBJDIR)\deltacmd$O $(OBJDIR)\descendants$O $(OBJDIR)\diff$O $(OBJDIR)\diffcmd$O $(OBJDIR)\dispatch$O $(OBJDIR)\doc$O $(OBJDIR)\encode$O $(OBJDIR)\event$O $(OBJDIR)\export$O $(OBJDIR)\file$O $(OBJDIR)\finfo$O $(OBJDIR)\foci$O $(OBJDIR)\fshell$O $(OBJDIR)\fusefs$O $(OBJDIR)\glob$O $(OBJDIR)\graph$O $(OBJDIR)\gzip$O $(OBJDIR)\http$O $(OBJDIR)\http_socket$O $(OBJDIR)\http_ssl$O $(OBJDIR)\http_transport$O $(OBJDIR)\import$O $(OBJDIR)\info$O $(OBJDIR)\json$O $(OBJDIR)\json_artifact$O $(OBJDIR)\json_branch$O $(OBJDIR)\json_config$O $(OBJDIR)\json_diff$O $(OBJDIR)\json_dir$O $(OBJDIR)\json_finfo$O $(OBJDIR)\json_login$O $(OBJDIR)\json_query$O $(OBJDIR)\json_report$O $(OBJDIR)\json_status$O $(OBJDIR)\json_tag$O $(OBJDIR)\json_timeline$O $(OBJDIR)\json_user$O $(OBJDIR)\json_wiki$O $(OBJDIR)\leaf$O $(OBJDIR)\loadctrl$O $(OBJDIR)\login$O $(OBJDIR)\lookslike$O $(OBJDIR)\main$O $(OBJDIR)\manifest$O $(OBJDIR)\markdown$O $(OBJDIR)\markdown_html$O $(OBJDIR)\md5$O $(OBJDIR)\merge$O $(OBJDIR)\merge3$O $(OBJDIR)\moderate$O $(OBJDIR)\name$O $(OBJDIR)\path$O $(OBJDIR)\piechart$O $(OBJDIR)\pivot$O $(OBJDIR)\popen$O $(OBJDIR)\pqueue$O $(OBJDIR)\printf$O $(OBJDIR)\publish$O $(OBJDIR)\purge$O $(OBJDIR)\rebuild$O $(OBJDIR)\regexp$O $(OBJDIR)\report$O $(OBJDIR)\rss$O $(OBJDIR)\schema$O $(OBJDIR)\search$O $(OBJDIR)\setup$O $(OBJDIR)\sha1$O $(OBJDIR)\shun$O $(OBJDIR)\sitemap$O $(OBJDIR)\skins$O $(OBJDIR)\sqlcmd$O $(OBJDIR)\stash$O $(OBJDIR)\stat$O $(OBJDIR)\statrep$O $(OBJDIR)\style$O $(OBJDIR)\sync$O $(OBJDIR)\tag$O $(OBJDIR)\tar$O $(OBJDIR)\th_main$O $(OBJDIR)\timeline$O $(OBJDIR)\tkt$O $(OBJDIR)\tktsetup$O $(OBJDIR)\undo$O $(OBJDIR)\unicode$O $(OBJDIR)\unversioned$O $(OBJDIR)\update$O $(OBJDIR)\url$O $(OBJDIR)\user$O $(OBJDIR)\utf8$O $(OBJDIR)\util$O $(OBJDIR)\verify$O $(OBJDIR)\vfile$O $(OBJDIR)\wiki$O $(OBJDIR)\wikiformat$O $(OBJDIR)\winfile$O $(OBJDIR)\winhttp$O $(OBJDIR)\wysiwyg$O $(OBJDIR)\xfer$O $(OBJDIR)\xfersetup$O $(OBJDIR)\zip$O $(OBJDIR)\shell$O $(OBJDIR)\sqlite3$O $(OBJDIR)\th$O $(OBJDIR)\th_lang$O
36
37
38	RC=$(DMDIR)\bin\rcc
39	RCFLAGS=-32 -w1 -I$(SRCDIR) /D__DMC__
40
	@@ -49,11 +49,11 @@
49
50	$(OBJDIR)\fossil.res: $B\win\fossil.rc
51	$(RC) $(RCFLAGS) -o$@ $**
52
53	$(OBJDIR)\link: $B\win\Makefile.dmc $(OBJDIR)\fossil.res
54	+echo add allrepo attach bag bisect blob branch browse builtin bundle cache captcha cgi checkin checkout clearsign clone comformat configure content db delta deltacmd descendants diff diffcmd dispatch doc encode event export file finfo foci fshell fusefs glob graph gzip http http_socket http_ssl http_transport import info json json_artifact json_branch json_config json_diff json_dir json_finfo json_login json_query json_report json_status json_tag json_timeline json_user json_wiki leaf loadctrl login lookslike main manifest markdown markdown_html md5 merge merge3 moderate name path piechart pivot popen pqueue printf publish purge rebuild regexp report rss schema search setup sha1 shun sitemap skins sqlcmd stash stat statrep style sync tag tar th_main timeline tkt tktsetup undo unicode unversioned update url user utf8 util verify vfile wiki wikiformat winfile winhttp wysiwyg xfer xfersetup zip shell sqlite3 th th_lang > $@
55	+echo fossil >> $@
56	+echo fossil >> $@
57	+echo $(LIBS) >> $@
58	+echo. >> $@
59	+echo fossil >> $@
	@@ -356,10 +356,16 @@
356	$(OBJDIR)\gzip$O : gzip_.c gzip.h
357	$(TCC) -o$@ -c gzip_.c
358
359	gzip_.c : $(SRCDIR)\gzip.c
360	+translate$E $** > $@






361
362	$(OBJDIR)\http$O : http_.c http.h
363	$(TCC) -o$@ -c http_.c
364
365	http_.c : $(SRCDIR)\http.c
	@@ -656,10 +662,22 @@
656	$(OBJDIR)\sha1$O : sha1_.c sha1.h
657	$(TCC) -o$@ -c sha1_.c
658
659	sha1_.c : $(SRCDIR)\sha1.c
660	+translate$E $** > $@












661
662	$(OBJDIR)\shun$O : shun_.c shun.h
663	$(TCC) -o$@ -c shun_.c
664
665	shun_.c : $(SRCDIR)\shun.c
	@@ -856,7 +874,7 @@
856
857	zip_.c : $(SRCDIR)\zip.c
858	+translate$E $** > $@
859
860	headers: makeheaders$E page_index.h builtin_data.h VERSION.h
861	+makeheaders$E add_.c:add.h allrepo_.c:allrepo.h attach_.c:attach.h bag_.c:bag.h bisect_.c:bisect.h blob_.c:blob.h branch_.c:branch.h browse_.c:browse.h builtin_.c:builtin.h bundle_.c:bundle.h cache_.c:cache.h captcha_.c:captcha.h cgi_.c:cgi.h checkin_.c:checkin.h checkout_.c:checkout.h clearsign_.c:clearsign.h clone_.c:clone.h comformat_.c:comformat.h configure_.c:configure.h content_.c:content.h db_.c:db.h delta_.c:delta.h deltacmd_.c:deltacmd.h descendants_.c:descendants.h diff_.c:diff.h diffcmd_.c:diffcmd.h dispatch_.c:dispatch.h doc_.c:doc.h encode_.c:encode.h event_.c:event.h export_.c:export.h file_.c:file.h finfo_.c:finfo.h foci_.c:foci.h fshell_.c:fshell.h fusefs_.c:fusefs.h glob_.c:glob.h graph_.c:graph.h gzip_.c:gzip.h http_.c:http.h http_socket_.c:http_socket.h http_ssl_.c:http_ssl.h http_transport_.c:http_transport.h import_.c:import.h info_.c:info.h json_.c:json.h json_artifact_.c:json_artifact.h json_branch_.c:json_branch.h json_config_.c:json_config.h json_diff_.c:json_diff.h json_dir_.c:json_dir.h json_finfo_.c:json_finfo.h json_login_.c:json_login.h json_query_.c:json_query.h json_report_.c:json_report.h json_status_.c:json_status.h json_tag_.c:json_tag.h json_timeline_.c:json_timeline.h json_user_.c:json_user.h json_wiki_.c:json_wiki.h leaf_.c:leaf.h loadctrl_.c:loadctrl.h login_.c:login.h lookslike_.c:lookslike.h main_.c:main.h manifest_.c:manifest.h markdown_.c:markdown.h markdown_html_.c:markdown_html.h md5_.c:md5.h merge_.c:merge.h merge3_.c:merge3.h moderate_.c:moderate.h name_.c:name.h path_.c:path.h piechart_.c:piechart.h pivot_.c:pivot.h popen_.c:popen.h pqueue_.c:pqueue.h printf_.c:printf.h publish_.c:publish.h purge_.c:purge.h rebuild_.c:rebuild.h regexp_.c:regexp.h report_.c:report.h rss_.c:rss.h schema_.c:schema.h search_.c:search.h setup_.c:setup.h sha1_.c:sha1.h shun_.c:shun.h sitemap_.c:sitemap.h skins_.c:skins.h sqlcmd_.c:sqlcmd.h stash_.c:stash.h stat_.c:stat.h statrep_.c:statrep.h style_.c:style.h sync_.c:sync.h tag_.c:tag.h tar_.c:tar.h th_main_.c:th_main.h timeline_.c:timeline.h tkt_.c:tkt.h tktsetup_.c:tktsetup.h undo_.c:undo.h unicode_.c:unicode.h unversioned_.c:unversioned.h update_.c:update.h url_.c:url.h user_.c:user.h utf8_.c:utf8.h util_.c:util.h verify_.c:verify.h vfile_.c:vfile.h wiki_.c:wiki.h wikiformat_.c:wikiformat.h winfile_.c:winfile.h winhttp_.c:winhttp.h wysiwyg_.c:wysiwyg.h xfer_.c:xfer.h xfersetup_.c:xfersetup.h zip_.c:zip.h $(SRCDIR)\sqlite3.h $(SRCDIR)\th.h VERSION.h $(SRCDIR)\cson_amalgamation.h
862	@copy /Y nul: headers
863

	--- win/Makefile.dmc
	+++ win/Makefile.dmc
	@@ -28,13 +28,13 @@
28
29	SQLITE_OPTIONS = -DNDEBUG=1 -DSQLITE_THREADSAFE=0 -DSQLITE_DEFAULT_MEMSTATUS=0 -DSQLITE_DEFAULT_WAL_SYNCHRONOUS=1 -DSQLITE_LIKE_DOESNT_MATCH_BLOBS -DSQLITE_OMIT_DECLTYPE -DSQLITE_OMIT_DEPRECATED -DSQLITE_OMIT_PROGRESS_CALLBACK -DSQLITE_OMIT_SHARED_CACHE -DSQLITE_OMIT_LOAD_EXTENSION -DSQLITE_MAX_EXPR_DEPTH=0 -DSQLITE_USE_ALLOCA -DSQLITE_ENABLE_LOCKING_STYLE=0 -DSQLITE_DEFAULT_FILE_FORMAT=4 -DSQLITE_ENABLE_EXPLAIN_COMMENTS -DSQLITE_ENABLE_FTS4 -DSQLITE_ENABLE_FTS3_PARENTHESIS -DSQLITE_ENABLE_DBSTAT_VTAB -DSQLITE_ENABLE_JSON1 -DSQLITE_ENABLE_FTS5
30
31	SHELL_OPTIONS = -Dmain=sqlite3_shell -DSQLITE_SHELL_IS_UTF8=1 -DSQLITE_OMIT_LOAD_EXTENSION=1 -DUSE_SYSTEM_SQLITE=$(USE_SYSTEM_SQLITE) -DSQLITE_SHELL_DBNAME_PROC=fossil_open -Daccess=file_access -Dsystem=fossil_system -Dgetenv=fossil_getenv -Dfopen=fossil_fopen
32
33	SRC = add_.c allrepo_.c attach_.c bag_.c bisect_.c blob_.c branch_.c browse_.c builtin_.c bundle_.c cache_.c captcha_.c cgi_.c checkin_.c checkout_.c clearsign_.c clone_.c comformat_.c configure_.c content_.c db_.c delta_.c deltacmd_.c descendants_.c diff_.c diffcmd_.c dispatch_.c doc_.c encode_.c event_.c export_.c file_.c finfo_.c foci_.c fshell_.c fusefs_.c glob_.c graph_.c gzip_.c hname_.c http_.c http_socket_.c http_ssl_.c http_transport_.c import_.c info_.c json_.c json_artifact_.c json_branch_.c json_config_.c json_diff_.c json_dir_.c json_finfo_.c json_login_.c json_query_.c json_report_.c json_status_.c json_tag_.c json_timeline_.c json_user_.c json_wiki_.c leaf_.c loadctrl_.c login_.c lookslike_.c main_.c manifest_.c markdown_.c markdown_html_.c md5_.c merge_.c merge3_.c moderate_.c name_.c path_.c piechart_.c pivot_.c popen_.c pqueue_.c printf_.c publish_.c purge_.c rebuild_.c regexp_.c report_.c rss_.c schema_.c search_.c setup_.c sha1_.c sha1hard_.c sha3_.c shun_.c sitemap_.c skins_.c sqlcmd_.c stash_.c stat_.c statrep_.c style_.c sync_.c tag_.c tar_.c th_main_.c timeline_.c tkt_.c tktsetup_.c undo_.c unicode_.c unversioned_.c update_.c url_.c user_.c utf8_.c util_.c verify_.c vfile_.c wiki_.c wikiformat_.c winfile_.c winhttp_.c wysiwyg_.c xfer_.c xfersetup_.c zip_.c
34
35	OBJ = $(OBJDIR)\add$O $(OBJDIR)\allrepo$O $(OBJDIR)\attach$O $(OBJDIR)\bag$O $(OBJDIR)\bisect$O $(OBJDIR)\blob$O $(OBJDIR)\branch$O $(OBJDIR)\browse$O $(OBJDIR)\builtin$O $(OBJDIR)\bundle$O $(OBJDIR)\cache$O $(OBJDIR)\captcha$O $(OBJDIR)\cgi$O $(OBJDIR)\checkin$O $(OBJDIR)\checkout$O $(OBJDIR)\clearsign$O $(OBJDIR)\clone$O $(OBJDIR)\comformat$O $(OBJDIR)\configure$O $(OBJDIR)\content$O $(OBJDIR)\db$O $(OBJDIR)\delta$O $(OBJDIR)\deltacmd$O $(OBJDIR)\descendants$O $(OBJDIR)\diff$O $(OBJDIR)\diffcmd$O $(OBJDIR)\dispatch$O $(OBJDIR)\doc$O $(OBJDIR)\encode$O $(OBJDIR)\event$O $(OBJDIR)\export$O $(OBJDIR)\file$O $(OBJDIR)\finfo$O $(OBJDIR)\foci$O $(OBJDIR)\fshell$O $(OBJDIR)\fusefs$O $(OBJDIR)\glob$O $(OBJDIR)\graph$O $(OBJDIR)\gzip$O $(OBJDIR)\hname$O $(OBJDIR)\http$O $(OBJDIR)\http_socket$O $(OBJDIR)\http_ssl$O $(OBJDIR)\http_transport$O $(OBJDIR)\import$O $(OBJDIR)\info$O $(OBJDIR)\json$O $(OBJDIR)\json_artifact$O $(OBJDIR)\json_branch$O $(OBJDIR)\json_config$O $(OBJDIR)\json_diff$O $(OBJDIR)\json_dir$O $(OBJDIR)\json_finfo$O $(OBJDIR)\json_login$O $(OBJDIR)\json_query$O $(OBJDIR)\json_report$O $(OBJDIR)\json_status$O $(OBJDIR)\json_tag$O $(OBJDIR)\json_timeline$O $(OBJDIR)\json_user$O $(OBJDIR)\json_wiki$O $(OBJDIR)\leaf$O $(OBJDIR)\loadctrl$O $(OBJDIR)\login$O $(OBJDIR)\lookslike$O $(OBJDIR)\main$O $(OBJDIR)\manifest$O $(OBJDIR)\markdown$O $(OBJDIR)\markdown_html$O $(OBJDIR)\md5$O $(OBJDIR)\merge$O $(OBJDIR)\merge3$O $(OBJDIR)\moderate$O $(OBJDIR)\name$O $(OBJDIR)\path$O $(OBJDIR)\piechart$O $(OBJDIR)\pivot$O $(OBJDIR)\popen$O $(OBJDIR)\pqueue$O $(OBJDIR)\printf$O $(OBJDIR)\publish$O $(OBJDIR)\purge$O $(OBJDIR)\rebuild$O $(OBJDIR)\regexp$O $(OBJDIR)\report$O $(OBJDIR)\rss$O $(OBJDIR)\schema$O $(OBJDIR)\search$O $(OBJDIR)\setup$O $(OBJDIR)\sha1$O $(OBJDIR)\sha1hard$O $(OBJDIR)\sha3$O $(OBJDIR)\shun$O $(OBJDIR)\sitemap$O $(OBJDIR)\skins$O $(OBJDIR)\sqlcmd$O $(OBJDIR)\stash$O $(OBJDIR)\stat$O $(OBJDIR)\statrep$O $(OBJDIR)\style$O $(OBJDIR)\sync$O $(OBJDIR)\tag$O $(OBJDIR)\tar$O $(OBJDIR)\th_main$O $(OBJDIR)\timeline$O $(OBJDIR)\tkt$O $(OBJDIR)\tktsetup$O $(OBJDIR)\undo$O $(OBJDIR)\unicode$O $(OBJDIR)\unversioned$O $(OBJDIR)\update$O $(OBJDIR)\url$O $(OBJDIR)\user$O $(OBJDIR)\utf8$O $(OBJDIR)\util$O $(OBJDIR)\verify$O $(OBJDIR)\vfile$O $(OBJDIR)\wiki$O $(OBJDIR)\wikiformat$O $(OBJDIR)\winfile$O $(OBJDIR)\winhttp$O $(OBJDIR)\wysiwyg$O $(OBJDIR)\xfer$O $(OBJDIR)\xfersetup$O $(OBJDIR)\zip$O $(OBJDIR)\shell$O $(OBJDIR)\sqlite3$O $(OBJDIR)\th$O $(OBJDIR)\th_lang$O
36
37
38	RC=$(DMDIR)\bin\rcc
39	RCFLAGS=-32 -w1 -I$(SRCDIR) /D__DMC__
40
	@@ -49,11 +49,11 @@
49
50	$(OBJDIR)\fossil.res: $B\win\fossil.rc
51	$(RC) $(RCFLAGS) -o$@ $**
52
53	$(OBJDIR)\link: $B\win\Makefile.dmc $(OBJDIR)\fossil.res
54	+echo add allrepo attach bag bisect blob branch browse builtin bundle cache captcha cgi checkin checkout clearsign clone comformat configure content db delta deltacmd descendants diff diffcmd dispatch doc encode event export file finfo foci fshell fusefs glob graph gzip hname http http_socket http_ssl http_transport import info json json_artifact json_branch json_config json_diff json_dir json_finfo json_login json_query json_report json_status json_tag json_timeline json_user json_wiki leaf loadctrl login lookslike main manifest markdown markdown_html md5 merge merge3 moderate name path piechart pivot popen pqueue printf publish purge rebuild regexp report rss schema search setup sha1 sha1hard sha3 shun sitemap skins sqlcmd stash stat statrep style sync tag tar th_main timeline tkt tktsetup undo unicode unversioned update url user utf8 util verify vfile wiki wikiformat winfile winhttp wysiwyg xfer xfersetup zip shell sqlite3 th th_lang > $@
55	+echo fossil >> $@
56	+echo fossil >> $@
57	+echo $(LIBS) >> $@
58	+echo. >> $@
59	+echo fossil >> $@
	@@ -356,10 +356,16 @@
356	$(OBJDIR)\gzip$O : gzip_.c gzip.h
357	$(TCC) -o$@ -c gzip_.c
358
359	gzip_.c : $(SRCDIR)\gzip.c
360	+translate$E $** > $@
361
362	$(OBJDIR)\hname$O : hname_.c hname.h
363	$(TCC) -o$@ -c hname_.c
364
365	hname_.c : $(SRCDIR)\hname.c
366	+translate$E $** > $@
367
368	$(OBJDIR)\http$O : http_.c http.h
369	$(TCC) -o$@ -c http_.c
370
371	http_.c : $(SRCDIR)\http.c
	@@ -656,10 +662,22 @@
662	$(OBJDIR)\sha1$O : sha1_.c sha1.h
663	$(TCC) -o$@ -c sha1_.c
664
665	sha1_.c : $(SRCDIR)\sha1.c
666	+translate$E $** > $@
667
668	$(OBJDIR)\sha1hard$O : sha1hard_.c sha1hard.h
669	$(TCC) -o$@ -c sha1hard_.c
670
671	sha1hard_.c : $(SRCDIR)\sha1hard.c
672	+translate$E $** > $@
673
674	$(OBJDIR)\sha3$O : sha3_.c sha3.h
675	$(TCC) -o$@ -c sha3_.c
676
677	sha3_.c : $(SRCDIR)\sha3.c
678	+translate$E $** > $@
679
680	$(OBJDIR)\shun$O : shun_.c shun.h
681	$(TCC) -o$@ -c shun_.c
682
683	shun_.c : $(SRCDIR)\shun.c
	@@ -856,7 +874,7 @@
874
875	zip_.c : $(SRCDIR)\zip.c
876	+translate$E $** > $@
877
878	headers: makeheaders$E page_index.h builtin_data.h VERSION.h
879	+makeheaders$E add_.c:add.h allrepo_.c:allrepo.h attach_.c:attach.h bag_.c:bag.h bisect_.c:bisect.h blob_.c:blob.h branch_.c:branch.h browse_.c:browse.h builtin_.c:builtin.h bundle_.c:bundle.h cache_.c:cache.h captcha_.c:captcha.h cgi_.c:cgi.h checkin_.c:checkin.h checkout_.c:checkout.h clearsign_.c:clearsign.h clone_.c:clone.h comformat_.c:comformat.h configure_.c:configure.h content_.c:content.h db_.c:db.h delta_.c:delta.h deltacmd_.c:deltacmd.h descendants_.c:descendants.h diff_.c:diff.h diffcmd_.c:diffcmd.h dispatch_.c:dispatch.h doc_.c:doc.h encode_.c:encode.h event_.c:event.h export_.c:export.h file_.c:file.h finfo_.c:finfo.h foci_.c:foci.h fshell_.c:fshell.h fusefs_.c:fusefs.h glob_.c:glob.h graph_.c:graph.h gzip_.c:gzip.h hname_.c:hname.h http_.c:http.h http_socket_.c:http_socket.h http_ssl_.c:http_ssl.h http_transport_.c:http_transport.h import_.c:import.h info_.c:info.h json_.c:json.h json_artifact_.c:json_artifact.h json_branch_.c:json_branch.h json_config_.c:json_config.h json_diff_.c:json_diff.h json_dir_.c:json_dir.h json_finfo_.c:json_finfo.h json_login_.c:json_login.h json_query_.c:json_query.h json_report_.c:json_report.h json_status_.c:json_status.h json_tag_.c:json_tag.h json_timeline_.c:json_timeline.h json_user_.c:json_user.h json_wiki_.c:json_wiki.h leaf_.c:leaf.h loadctrl_.c:loadctrl.h login_.c:login.h lookslike_.c:lookslike.h main_.c:main.h manifest_.c:manifest.h markdown_.c:markdown.h markdown_html_.c:markdown_html.h md5_.c:md5.h merge_.c:merge.h merge3_.c:merge3.h moderate_.c:moderate.h name_.c:name.h path_.c:path.h piechart_.c:piechart.h pivot_.c:pivot.h popen_.c:popen.h pqueue_.c:pqueue.h printf_.c:printf.h publish_.c:publish.h purge_.c:purge.h rebuild_.c:rebuild.h regexp_.c:regexp.h report_.c:report.h rss_.c:rss.h schema_.c:schema.h search_.c:search.h setup_.c:setup.h sha1_.c:sha1.h sha1hard_.c:sha1hard.h sha3_.c:sha3.h shun_.c:shun.h sitemap_.c:sitemap.h skins_.c:skins.h sqlcmd_.c:sqlcmd.h stash_.c:stash.h stat_.c:stat.h statrep_.c:statrep.h style_.c:style.h sync_.c:sync.h tag_.c:tag.h tar_.c:tar.h th_main_.c:th_main.h timeline_.c:timeline.h tkt_.c:tkt.h tktsetup_.c:tktsetup.h undo_.c:undo.h unicode_.c:unicode.h unversioned_.c:unversioned.h update_.c:update.h url_.c:url.h user_.c:user.h utf8_.c:utf8.h util_.c:util.h verify_.c:verify.h vfile_.c:vfile.h wiki_.c:wiki.h wikiformat_.c:wikiformat.h winfile_.c:winfile.h winhttp_.c:winhttp.h wysiwyg_.c:wysiwyg.h xfer_.c:xfer.h xfersetup_.c:xfersetup.h zip_.c:zip.h $(SRCDIR)\sqlite3.h $(SRCDIR)\th.h VERSION.h $(SRCDIR)\cson_amalgamation.h
880	@copy /Y nul: headers
881

M win/Makefile.mingw

+36

		--- win/Makefile.mingw
		+++ win/Makefile.mingw
		@@ -461,10 +461,11 @@
461	461	$(SRCDIR)/fshell.c \
462	462	$(SRCDIR)/fusefs.c \
463	463	$(SRCDIR)/glob.c \
464	464	$(SRCDIR)/graph.c \
465	465	$(SRCDIR)/gzip.c \
	466	+ $(SRCDIR)/hname.c \
466	467	$(SRCDIR)/http.c \
467	468	$(SRCDIR)/http_socket.c \
468	469	$(SRCDIR)/http_ssl.c \
469	470	$(SRCDIR)/http_transport.c \
470	471	$(SRCDIR)/import.c \
		@@ -511,10 +512,12 @@
511	512	$(SRCDIR)/rss.c \
512	513	$(SRCDIR)/schema.c \
513	514	$(SRCDIR)/search.c \
514	515	$(SRCDIR)/setup.c \
515	516	$(SRCDIR)/sha1.c \
	517	+ $(SRCDIR)/sha1hard.c \
	518	+ $(SRCDIR)/sha3.c \
516	519	$(SRCDIR)/shun.c \
517	520	$(SRCDIR)/sitemap.c \
518	521	$(SRCDIR)/skins.c \
519	522	$(SRCDIR)/sqlcmd.c \
520	523	$(SRCDIR)/stash.c \
		@@ -636,10 +639,11 @@
636	639	$(OBJDIR)/fshell_.c \
637	640	$(OBJDIR)/fusefs_.c \
638	641	$(OBJDIR)/glob_.c \
639	642	$(OBJDIR)/graph_.c \
640	643	$(OBJDIR)/gzip_.c \
	644	+ $(OBJDIR)/hname_.c \
641	645	$(OBJDIR)/http_.c \
642	646	$(OBJDIR)/http_socket_.c \
643	647	$(OBJDIR)/http_ssl_.c \
644	648	$(OBJDIR)/http_transport_.c \
645	649	$(OBJDIR)/import_.c \
		@@ -686,10 +690,12 @@
686	690	$(OBJDIR)/rss_.c \
687	691	$(OBJDIR)/schema_.c \
688	692	$(OBJDIR)/search_.c \
689	693	$(OBJDIR)/setup_.c \
690	694	$(OBJDIR)/sha1_.c \
	695	+ $(OBJDIR)/sha1hard_.c \
	696	+ $(OBJDIR)/sha3_.c \
691	697	$(OBJDIR)/shun_.c \
692	698	$(OBJDIR)/sitemap_.c \
693	699	$(OBJDIR)/skins_.c \
694	700	$(OBJDIR)/sqlcmd_.c \
695	701	$(OBJDIR)/stash_.c \
		@@ -760,10 +766,11 @@
760	766	$(OBJDIR)/fshell.o \
761	767	$(OBJDIR)/fusefs.o \
762	768	$(OBJDIR)/glob.o \
763	769	$(OBJDIR)/graph.o \
764	770	$(OBJDIR)/gzip.o \
	771	+ $(OBJDIR)/hname.o \
765	772	$(OBJDIR)/http.o \
766	773	$(OBJDIR)/http_socket.o \
767	774	$(OBJDIR)/http_ssl.o \
768	775	$(OBJDIR)/http_transport.o \
769	776	$(OBJDIR)/import.o \
		@@ -810,10 +817,12 @@
810	817	$(OBJDIR)/rss.o \
811	818	$(OBJDIR)/schema.o \
812	819	$(OBJDIR)/search.o \
813	820	$(OBJDIR)/setup.o \
814	821	$(OBJDIR)/sha1.o \
	822	+ $(OBJDIR)/sha1hard.o \
	823	+ $(OBJDIR)/sha3.o \
815	824	$(OBJDIR)/shun.o \
816	825	$(OBJDIR)/sitemap.o \
817	826	$(OBJDIR)/skins.o \
818	827	$(OBJDIR)/sqlcmd.o \
819	828	$(OBJDIR)/stash.o \
		@@ -1095,10 +1104,11 @@
1095	1104	$(OBJDIR)/fshell_.c:$(OBJDIR)/fshell.h \
1096	1105	$(OBJDIR)/fusefs_.c:$(OBJDIR)/fusefs.h \
1097	1106	$(OBJDIR)/glob_.c:$(OBJDIR)/glob.h \
1098	1107	$(OBJDIR)/graph_.c:$(OBJDIR)/graph.h \
1099	1108	$(OBJDIR)/gzip_.c:$(OBJDIR)/gzip.h \
	1109	+ $(OBJDIR)/hname_.c:$(OBJDIR)/hname.h \
1100	1110	$(OBJDIR)/http_.c:$(OBJDIR)/http.h \
1101	1111	$(OBJDIR)/http_socket_.c:$(OBJDIR)/http_socket.h \
1102	1112	$(OBJDIR)/http_ssl_.c:$(OBJDIR)/http_ssl.h \
1103	1113	$(OBJDIR)/http_transport_.c:$(OBJDIR)/http_transport.h \
1104	1114	$(OBJDIR)/import_.c:$(OBJDIR)/import.h \
		@@ -1145,10 +1155,12 @@
1145	1155	$(OBJDIR)/rss_.c:$(OBJDIR)/rss.h \
1146	1156	$(OBJDIR)/schema_.c:$(OBJDIR)/schema.h \
1147	1157	$(OBJDIR)/search_.c:$(OBJDIR)/search.h \
1148	1158	$(OBJDIR)/setup_.c:$(OBJDIR)/setup.h \
1149	1159	$(OBJDIR)/sha1_.c:$(OBJDIR)/sha1.h \
	1160	+ $(OBJDIR)/sha1hard_.c:$(OBJDIR)/sha1hard.h \
	1161	+ $(OBJDIR)/sha3_.c:$(OBJDIR)/sha3.h \
1150	1162	$(OBJDIR)/shun_.c:$(OBJDIR)/shun.h \
1151	1163	$(OBJDIR)/sitemap_.c:$(OBJDIR)/sitemap.h \
1152	1164	$(OBJDIR)/skins_.c:$(OBJDIR)/skins.h \
1153	1165	$(OBJDIR)/sqlcmd_.c:$(OBJDIR)/sqlcmd.h \
1154	1166	$(OBJDIR)/stash_.c:$(OBJDIR)/stash.h \
		@@ -1498,10 +1510,18 @@
1498	1510
1499	1511	$(OBJDIR)/gzip.o: $(OBJDIR)/gzip_.c $(OBJDIR)/gzip.h $(SRCDIR)/config.h
1500	1512	$(XTCC) -o $(OBJDIR)/gzip.o -c $(OBJDIR)/gzip_.c
1501	1513
1502	1514	$(OBJDIR)/gzip.h: $(OBJDIR)/headers
	1515	+
	1516	+$(OBJDIR)/hname_.c: $(SRCDIR)/hname.c $(TRANSLATE)
	1517	+ $(TRANSLATE) $(SRCDIR)/hname.c >$@
	1518	+
	1519	+$(OBJDIR)/hname.o: $(OBJDIR)/hname_.c $(OBJDIR)/hname.h $(SRCDIR)/config.h
	1520	+ $(XTCC) -o $(OBJDIR)/hname.o -c $(OBJDIR)/hname_.c
	1521	+
	1522	+$(OBJDIR)/hname.h: $(OBJDIR)/headers
1503	1523
1504	1524	$(OBJDIR)/http_.c: $(SRCDIR)/http.c $(TRANSLATE)
1505	1525	$(TRANSLATE) $(SRCDIR)/http.c >$@
1506	1526
1507	1527	$(OBJDIR)/http.o: $(OBJDIR)/http_.c $(OBJDIR)/http.h $(SRCDIR)/config.h
		@@ -1898,10 +1918,26 @@
1898	1918
1899	1919	$(OBJDIR)/sha1.o: $(OBJDIR)/sha1_.c $(OBJDIR)/sha1.h $(SRCDIR)/config.h
1900	1920	$(XTCC) -o $(OBJDIR)/sha1.o -c $(OBJDIR)/sha1_.c
1901	1921
1902	1922	$(OBJDIR)/sha1.h: $(OBJDIR)/headers
	1923	+
	1924	+$(OBJDIR)/sha1hard_.c: $(SRCDIR)/sha1hard.c $(TRANSLATE)
	1925	+ $(TRANSLATE) $(SRCDIR)/sha1hard.c >$@
	1926	+
	1927	+$(OBJDIR)/sha1hard.o: $(OBJDIR)/sha1hard_.c $(OBJDIR)/sha1hard.h $(SRCDIR)/config.h
	1928	+ $(XTCC) -o $(OBJDIR)/sha1hard.o -c $(OBJDIR)/sha1hard_.c
	1929	+
	1930	+$(OBJDIR)/sha1hard.h: $(OBJDIR)/headers
	1931	+
	1932	+$(OBJDIR)/sha3_.c: $(SRCDIR)/sha3.c $(TRANSLATE)
	1933	+ $(TRANSLATE) $(SRCDIR)/sha3.c >$@
	1934	+
	1935	+$(OBJDIR)/sha3.o: $(OBJDIR)/sha3_.c $(OBJDIR)/sha3.h $(SRCDIR)/config.h
	1936	+ $(XTCC) -o $(OBJDIR)/sha3.o -c $(OBJDIR)/sha3_.c
	1937	+
	1938	+$(OBJDIR)/sha3.h: $(OBJDIR)/headers
1903	1939
1904	1940	$(OBJDIR)/shun_.c: $(SRCDIR)/shun.c $(TRANSLATE)
1905	1941	$(TRANSLATE) $(SRCDIR)/shun.c >$@
1906	1942
1907	1943	$(OBJDIR)/shun.o: $(OBJDIR)/shun_.c $(OBJDIR)/shun.h $(SRCDIR)/config.h
1908	1944

	--- win/Makefile.mingw
	+++ win/Makefile.mingw
	@@ -461,10 +461,11 @@
461	$(SRCDIR)/fshell.c \
462	$(SRCDIR)/fusefs.c \
463	$(SRCDIR)/glob.c \
464	$(SRCDIR)/graph.c \
465	$(SRCDIR)/gzip.c \

466	$(SRCDIR)/http.c \
467	$(SRCDIR)/http_socket.c \
468	$(SRCDIR)/http_ssl.c \
469	$(SRCDIR)/http_transport.c \
470	$(SRCDIR)/import.c \
	@@ -511,10 +512,12 @@
511	$(SRCDIR)/rss.c \
512	$(SRCDIR)/schema.c \
513	$(SRCDIR)/search.c \
514	$(SRCDIR)/setup.c \
515	$(SRCDIR)/sha1.c \


516	$(SRCDIR)/shun.c \
517	$(SRCDIR)/sitemap.c \
518	$(SRCDIR)/skins.c \
519	$(SRCDIR)/sqlcmd.c \
520	$(SRCDIR)/stash.c \
	@@ -636,10 +639,11 @@
636	$(OBJDIR)/fshell_.c \
637	$(OBJDIR)/fusefs_.c \
638	$(OBJDIR)/glob_.c \
639	$(OBJDIR)/graph_.c \
640	$(OBJDIR)/gzip_.c \

641	$(OBJDIR)/http_.c \
642	$(OBJDIR)/http_socket_.c \
643	$(OBJDIR)/http_ssl_.c \
644	$(OBJDIR)/http_transport_.c \
645	$(OBJDIR)/import_.c \
	@@ -686,10 +690,12 @@
686	$(OBJDIR)/rss_.c \
687	$(OBJDIR)/schema_.c \
688	$(OBJDIR)/search_.c \
689	$(OBJDIR)/setup_.c \
690	$(OBJDIR)/sha1_.c \


691	$(OBJDIR)/shun_.c \
692	$(OBJDIR)/sitemap_.c \
693	$(OBJDIR)/skins_.c \
694	$(OBJDIR)/sqlcmd_.c \
695	$(OBJDIR)/stash_.c \
	@@ -760,10 +766,11 @@
760	$(OBJDIR)/fshell.o \
761	$(OBJDIR)/fusefs.o \
762	$(OBJDIR)/glob.o \
763	$(OBJDIR)/graph.o \
764	$(OBJDIR)/gzip.o \

765	$(OBJDIR)/http.o \
766	$(OBJDIR)/http_socket.o \
767	$(OBJDIR)/http_ssl.o \
768	$(OBJDIR)/http_transport.o \
769	$(OBJDIR)/import.o \
	@@ -810,10 +817,12 @@
810	$(OBJDIR)/rss.o \
811	$(OBJDIR)/schema.o \
812	$(OBJDIR)/search.o \
813	$(OBJDIR)/setup.o \
814	$(OBJDIR)/sha1.o \


815	$(OBJDIR)/shun.o \
816	$(OBJDIR)/sitemap.o \
817	$(OBJDIR)/skins.o \
818	$(OBJDIR)/sqlcmd.o \
819	$(OBJDIR)/stash.o \
	@@ -1095,10 +1104,11 @@
1095	$(OBJDIR)/fshell_.c:$(OBJDIR)/fshell.h \
1096	$(OBJDIR)/fusefs_.c:$(OBJDIR)/fusefs.h \
1097	$(OBJDIR)/glob_.c:$(OBJDIR)/glob.h \
1098	$(OBJDIR)/graph_.c:$(OBJDIR)/graph.h \
1099	$(OBJDIR)/gzip_.c:$(OBJDIR)/gzip.h \

1100	$(OBJDIR)/http_.c:$(OBJDIR)/http.h \
1101	$(OBJDIR)/http_socket_.c:$(OBJDIR)/http_socket.h \
1102	$(OBJDIR)/http_ssl_.c:$(OBJDIR)/http_ssl.h \
1103	$(OBJDIR)/http_transport_.c:$(OBJDIR)/http_transport.h \
1104	$(OBJDIR)/import_.c:$(OBJDIR)/import.h \
	@@ -1145,10 +1155,12 @@
1145	$(OBJDIR)/rss_.c:$(OBJDIR)/rss.h \
1146	$(OBJDIR)/schema_.c:$(OBJDIR)/schema.h \
1147	$(OBJDIR)/search_.c:$(OBJDIR)/search.h \
1148	$(OBJDIR)/setup_.c:$(OBJDIR)/setup.h \
1149	$(OBJDIR)/sha1_.c:$(OBJDIR)/sha1.h \


1150	$(OBJDIR)/shun_.c:$(OBJDIR)/shun.h \
1151	$(OBJDIR)/sitemap_.c:$(OBJDIR)/sitemap.h \
1152	$(OBJDIR)/skins_.c:$(OBJDIR)/skins.h \
1153	$(OBJDIR)/sqlcmd_.c:$(OBJDIR)/sqlcmd.h \
1154	$(OBJDIR)/stash_.c:$(OBJDIR)/stash.h \
	@@ -1498,10 +1510,18 @@
1498
1499	$(OBJDIR)/gzip.o: $(OBJDIR)/gzip_.c $(OBJDIR)/gzip.h $(SRCDIR)/config.h
1500	$(XTCC) -o $(OBJDIR)/gzip.o -c $(OBJDIR)/gzip_.c
1501
1502	$(OBJDIR)/gzip.h: $(OBJDIR)/headers








1503
1504	$(OBJDIR)/http_.c: $(SRCDIR)/http.c $(TRANSLATE)
1505	$(TRANSLATE) $(SRCDIR)/http.c >$@
1506
1507	$(OBJDIR)/http.o: $(OBJDIR)/http_.c $(OBJDIR)/http.h $(SRCDIR)/config.h
	@@ -1898,10 +1918,26 @@
1898
1899	$(OBJDIR)/sha1.o: $(OBJDIR)/sha1_.c $(OBJDIR)/sha1.h $(SRCDIR)/config.h
1900	$(XTCC) -o $(OBJDIR)/sha1.o -c $(OBJDIR)/sha1_.c
1901
1902	$(OBJDIR)/sha1.h: $(OBJDIR)/headers
















1903
1904	$(OBJDIR)/shun_.c: $(SRCDIR)/shun.c $(TRANSLATE)
1905	$(TRANSLATE) $(SRCDIR)/shun.c >$@
1906
1907	$(OBJDIR)/shun.o: $(OBJDIR)/shun_.c $(OBJDIR)/shun.h $(SRCDIR)/config.h
1908

	--- win/Makefile.mingw
	+++ win/Makefile.mingw
	@@ -461,10 +461,11 @@
461	$(SRCDIR)/fshell.c \
462	$(SRCDIR)/fusefs.c \
463	$(SRCDIR)/glob.c \
464	$(SRCDIR)/graph.c \
465	$(SRCDIR)/gzip.c \
466	$(SRCDIR)/hname.c \
467	$(SRCDIR)/http.c \
468	$(SRCDIR)/http_socket.c \
469	$(SRCDIR)/http_ssl.c \
470	$(SRCDIR)/http_transport.c \
471	$(SRCDIR)/import.c \
	@@ -511,10 +512,12 @@
512	$(SRCDIR)/rss.c \
513	$(SRCDIR)/schema.c \
514	$(SRCDIR)/search.c \
515	$(SRCDIR)/setup.c \
516	$(SRCDIR)/sha1.c \
517	$(SRCDIR)/sha1hard.c \
518	$(SRCDIR)/sha3.c \
519	$(SRCDIR)/shun.c \
520	$(SRCDIR)/sitemap.c \
521	$(SRCDIR)/skins.c \
522	$(SRCDIR)/sqlcmd.c \
523	$(SRCDIR)/stash.c \
	@@ -636,10 +639,11 @@
639	$(OBJDIR)/fshell_.c \
640	$(OBJDIR)/fusefs_.c \
641	$(OBJDIR)/glob_.c \
642	$(OBJDIR)/graph_.c \
643	$(OBJDIR)/gzip_.c \
644	$(OBJDIR)/hname_.c \
645	$(OBJDIR)/http_.c \
646	$(OBJDIR)/http_socket_.c \
647	$(OBJDIR)/http_ssl_.c \
648	$(OBJDIR)/http_transport_.c \
649	$(OBJDIR)/import_.c \
	@@ -686,10 +690,12 @@
690	$(OBJDIR)/rss_.c \
691	$(OBJDIR)/schema_.c \
692	$(OBJDIR)/search_.c \
693	$(OBJDIR)/setup_.c \
694	$(OBJDIR)/sha1_.c \
695	$(OBJDIR)/sha1hard_.c \
696	$(OBJDIR)/sha3_.c \
697	$(OBJDIR)/shun_.c \
698	$(OBJDIR)/sitemap_.c \
699	$(OBJDIR)/skins_.c \
700	$(OBJDIR)/sqlcmd_.c \
701	$(OBJDIR)/stash_.c \
	@@ -760,10 +766,11 @@
766	$(OBJDIR)/fshell.o \
767	$(OBJDIR)/fusefs.o \
768	$(OBJDIR)/glob.o \
769	$(OBJDIR)/graph.o \
770	$(OBJDIR)/gzip.o \
771	$(OBJDIR)/hname.o \
772	$(OBJDIR)/http.o \
773	$(OBJDIR)/http_socket.o \
774	$(OBJDIR)/http_ssl.o \
775	$(OBJDIR)/http_transport.o \
776	$(OBJDIR)/import.o \
	@@ -810,10 +817,12 @@
817	$(OBJDIR)/rss.o \
818	$(OBJDIR)/schema.o \
819	$(OBJDIR)/search.o \
820	$(OBJDIR)/setup.o \
821	$(OBJDIR)/sha1.o \
822	$(OBJDIR)/sha1hard.o \
823	$(OBJDIR)/sha3.o \
824	$(OBJDIR)/shun.o \
825	$(OBJDIR)/sitemap.o \
826	$(OBJDIR)/skins.o \
827	$(OBJDIR)/sqlcmd.o \
828	$(OBJDIR)/stash.o \
	@@ -1095,10 +1104,11 @@
1104	$(OBJDIR)/fshell_.c:$(OBJDIR)/fshell.h \
1105	$(OBJDIR)/fusefs_.c:$(OBJDIR)/fusefs.h \
1106	$(OBJDIR)/glob_.c:$(OBJDIR)/glob.h \
1107	$(OBJDIR)/graph_.c:$(OBJDIR)/graph.h \
1108	$(OBJDIR)/gzip_.c:$(OBJDIR)/gzip.h \
1109	$(OBJDIR)/hname_.c:$(OBJDIR)/hname.h \
1110	$(OBJDIR)/http_.c:$(OBJDIR)/http.h \
1111	$(OBJDIR)/http_socket_.c:$(OBJDIR)/http_socket.h \
1112	$(OBJDIR)/http_ssl_.c:$(OBJDIR)/http_ssl.h \
1113	$(OBJDIR)/http_transport_.c:$(OBJDIR)/http_transport.h \
1114	$(OBJDIR)/import_.c:$(OBJDIR)/import.h \
	@@ -1145,10 +1155,12 @@
1155	$(OBJDIR)/rss_.c:$(OBJDIR)/rss.h \
1156	$(OBJDIR)/schema_.c:$(OBJDIR)/schema.h \
1157	$(OBJDIR)/search_.c:$(OBJDIR)/search.h \
1158	$(OBJDIR)/setup_.c:$(OBJDIR)/setup.h \
1159	$(OBJDIR)/sha1_.c:$(OBJDIR)/sha1.h \
1160	$(OBJDIR)/sha1hard_.c:$(OBJDIR)/sha1hard.h \
1161	$(OBJDIR)/sha3_.c:$(OBJDIR)/sha3.h \
1162	$(OBJDIR)/shun_.c:$(OBJDIR)/shun.h \
1163	$(OBJDIR)/sitemap_.c:$(OBJDIR)/sitemap.h \
1164	$(OBJDIR)/skins_.c:$(OBJDIR)/skins.h \
1165	$(OBJDIR)/sqlcmd_.c:$(OBJDIR)/sqlcmd.h \
1166	$(OBJDIR)/stash_.c:$(OBJDIR)/stash.h \
	@@ -1498,10 +1510,18 @@
1510
1511	$(OBJDIR)/gzip.o: $(OBJDIR)/gzip_.c $(OBJDIR)/gzip.h $(SRCDIR)/config.h
1512	$(XTCC) -o $(OBJDIR)/gzip.o -c $(OBJDIR)/gzip_.c
1513
1514	$(OBJDIR)/gzip.h: $(OBJDIR)/headers
1515
1516	$(OBJDIR)/hname_.c: $(SRCDIR)/hname.c $(TRANSLATE)
1517	$(TRANSLATE) $(SRCDIR)/hname.c >$@
1518
1519	$(OBJDIR)/hname.o: $(OBJDIR)/hname_.c $(OBJDIR)/hname.h $(SRCDIR)/config.h
1520	$(XTCC) -o $(OBJDIR)/hname.o -c $(OBJDIR)/hname_.c
1521
1522	$(OBJDIR)/hname.h: $(OBJDIR)/headers
1523
1524	$(OBJDIR)/http_.c: $(SRCDIR)/http.c $(TRANSLATE)
1525	$(TRANSLATE) $(SRCDIR)/http.c >$@
1526
1527	$(OBJDIR)/http.o: $(OBJDIR)/http_.c $(OBJDIR)/http.h $(SRCDIR)/config.h
	@@ -1898,10 +1918,26 @@
1918
1919	$(OBJDIR)/sha1.o: $(OBJDIR)/sha1_.c $(OBJDIR)/sha1.h $(SRCDIR)/config.h
1920	$(XTCC) -o $(OBJDIR)/sha1.o -c $(OBJDIR)/sha1_.c
1921
1922	$(OBJDIR)/sha1.h: $(OBJDIR)/headers
1923
1924	$(OBJDIR)/sha1hard_.c: $(SRCDIR)/sha1hard.c $(TRANSLATE)
1925	$(TRANSLATE) $(SRCDIR)/sha1hard.c >$@
1926
1927	$(OBJDIR)/sha1hard.o: $(OBJDIR)/sha1hard_.c $(OBJDIR)/sha1hard.h $(SRCDIR)/config.h
1928	$(XTCC) -o $(OBJDIR)/sha1hard.o -c $(OBJDIR)/sha1hard_.c
1929
1930	$(OBJDIR)/sha1hard.h: $(OBJDIR)/headers
1931
1932	$(OBJDIR)/sha3_.c: $(SRCDIR)/sha3.c $(TRANSLATE)
1933	$(TRANSLATE) $(SRCDIR)/sha3.c >$@
1934
1935	$(OBJDIR)/sha3.o: $(OBJDIR)/sha3_.c $(OBJDIR)/sha3.h $(SRCDIR)/config.h
1936	$(XTCC) -o $(OBJDIR)/sha3.o -c $(OBJDIR)/sha3_.c
1937
1938	$(OBJDIR)/sha3.h: $(OBJDIR)/headers
1939
1940	$(OBJDIR)/shun_.c: $(SRCDIR)/shun.c $(TRANSLATE)
1941	$(TRANSLATE) $(SRCDIR)/shun.c >$@
1942
1943	$(OBJDIR)/shun.o: $(OBJDIR)/shun_.c $(OBJDIR)/shun.h $(SRCDIR)/config.h
1944

M win/Makefile.mingw

+36

		--- win/Makefile.mingw
		+++ win/Makefile.mingw
		@@ -461,10 +461,11 @@
461	461	$(SRCDIR)/fshell.c \
462	462	$(SRCDIR)/fusefs.c \
463	463	$(SRCDIR)/glob.c \
464	464	$(SRCDIR)/graph.c \
465	465	$(SRCDIR)/gzip.c \
	466	+ $(SRCDIR)/hname.c \
466	467	$(SRCDIR)/http.c \
467	468	$(SRCDIR)/http_socket.c \
468	469	$(SRCDIR)/http_ssl.c \
469	470	$(SRCDIR)/http_transport.c \
470	471	$(SRCDIR)/import.c \
		@@ -511,10 +512,12 @@
511	512	$(SRCDIR)/rss.c \
512	513	$(SRCDIR)/schema.c \
513	514	$(SRCDIR)/search.c \
514	515	$(SRCDIR)/setup.c \
515	516	$(SRCDIR)/sha1.c \
	517	+ $(SRCDIR)/sha1hard.c \
	518	+ $(SRCDIR)/sha3.c \
516	519	$(SRCDIR)/shun.c \
517	520	$(SRCDIR)/sitemap.c \
518	521	$(SRCDIR)/skins.c \
519	522	$(SRCDIR)/sqlcmd.c \
520	523	$(SRCDIR)/stash.c \
		@@ -636,10 +639,11 @@
636	639	$(OBJDIR)/fshell_.c \
637	640	$(OBJDIR)/fusefs_.c \
638	641	$(OBJDIR)/glob_.c \
639	642	$(OBJDIR)/graph_.c \
640	643	$(OBJDIR)/gzip_.c \
	644	+ $(OBJDIR)/hname_.c \
641	645	$(OBJDIR)/http_.c \
642	646	$(OBJDIR)/http_socket_.c \
643	647	$(OBJDIR)/http_ssl_.c \
644	648	$(OBJDIR)/http_transport_.c \
645	649	$(OBJDIR)/import_.c \
		@@ -686,10 +690,12 @@
686	690	$(OBJDIR)/rss_.c \
687	691	$(OBJDIR)/schema_.c \
688	692	$(OBJDIR)/search_.c \
689	693	$(OBJDIR)/setup_.c \
690	694	$(OBJDIR)/sha1_.c \
	695	+ $(OBJDIR)/sha1hard_.c \
	696	+ $(OBJDIR)/sha3_.c \
691	697	$(OBJDIR)/shun_.c \
692	698	$(OBJDIR)/sitemap_.c \
693	699	$(OBJDIR)/skins_.c \
694	700	$(OBJDIR)/sqlcmd_.c \
695	701	$(OBJDIR)/stash_.c \
		@@ -760,10 +766,11 @@
760	766	$(OBJDIR)/fshell.o \
761	767	$(OBJDIR)/fusefs.o \
762	768	$(OBJDIR)/glob.o \
763	769	$(OBJDIR)/graph.o \
764	770	$(OBJDIR)/gzip.o \
	771	+ $(OBJDIR)/hname.o \
765	772	$(OBJDIR)/http.o \
766	773	$(OBJDIR)/http_socket.o \
767	774	$(OBJDIR)/http_ssl.o \
768	775	$(OBJDIR)/http_transport.o \
769	776	$(OBJDIR)/import.o \
		@@ -810,10 +817,12 @@
810	817	$(OBJDIR)/rss.o \
811	818	$(OBJDIR)/schema.o \
812	819	$(OBJDIR)/search.o \
813	820	$(OBJDIR)/setup.o \
814	821	$(OBJDIR)/sha1.o \
	822	+ $(OBJDIR)/sha1hard.o \
	823	+ $(OBJDIR)/sha3.o \
815	824	$(OBJDIR)/shun.o \
816	825	$(OBJDIR)/sitemap.o \
817	826	$(OBJDIR)/skins.o \
818	827	$(OBJDIR)/sqlcmd.o \
819	828	$(OBJDIR)/stash.o \
		@@ -1095,10 +1104,11 @@
1095	1104	$(OBJDIR)/fshell_.c:$(OBJDIR)/fshell.h \
1096	1105	$(OBJDIR)/fusefs_.c:$(OBJDIR)/fusefs.h \
1097	1106	$(OBJDIR)/glob_.c:$(OBJDIR)/glob.h \
1098	1107	$(OBJDIR)/graph_.c:$(OBJDIR)/graph.h \
1099	1108	$(OBJDIR)/gzip_.c:$(OBJDIR)/gzip.h \
	1109	+ $(OBJDIR)/hname_.c:$(OBJDIR)/hname.h \
1100	1110	$(OBJDIR)/http_.c:$(OBJDIR)/http.h \
1101	1111	$(OBJDIR)/http_socket_.c:$(OBJDIR)/http_socket.h \
1102	1112	$(OBJDIR)/http_ssl_.c:$(OBJDIR)/http_ssl.h \
1103	1113	$(OBJDIR)/http_transport_.c:$(OBJDIR)/http_transport.h \
1104	1114	$(OBJDIR)/import_.c:$(OBJDIR)/import.h \
		@@ -1145,10 +1155,12 @@
1145	1155	$(OBJDIR)/rss_.c:$(OBJDIR)/rss.h \
1146	1156	$(OBJDIR)/schema_.c:$(OBJDIR)/schema.h \
1147	1157	$(OBJDIR)/search_.c:$(OBJDIR)/search.h \
1148	1158	$(OBJDIR)/setup_.c:$(OBJDIR)/setup.h \
1149	1159	$(OBJDIR)/sha1_.c:$(OBJDIR)/sha1.h \
	1160	+ $(OBJDIR)/sha1hard_.c:$(OBJDIR)/sha1hard.h \
	1161	+ $(OBJDIR)/sha3_.c:$(OBJDIR)/sha3.h \
1150	1162	$(OBJDIR)/shun_.c:$(OBJDIR)/shun.h \
1151	1163	$(OBJDIR)/sitemap_.c:$(OBJDIR)/sitemap.h \
1152	1164	$(OBJDIR)/skins_.c:$(OBJDIR)/skins.h \
1153	1165	$(OBJDIR)/sqlcmd_.c:$(OBJDIR)/sqlcmd.h \
1154	1166	$(OBJDIR)/stash_.c:$(OBJDIR)/stash.h \
		@@ -1498,10 +1510,18 @@
1498	1510
1499	1511	$(OBJDIR)/gzip.o: $(OBJDIR)/gzip_.c $(OBJDIR)/gzip.h $(SRCDIR)/config.h
1500	1512	$(XTCC) -o $(OBJDIR)/gzip.o -c $(OBJDIR)/gzip_.c
1501	1513
1502	1514	$(OBJDIR)/gzip.h: $(OBJDIR)/headers
	1515	+
	1516	+$(OBJDIR)/hname_.c: $(SRCDIR)/hname.c $(TRANSLATE)
	1517	+ $(TRANSLATE) $(SRCDIR)/hname.c >$@
	1518	+
	1519	+$(OBJDIR)/hname.o: $(OBJDIR)/hname_.c $(OBJDIR)/hname.h $(SRCDIR)/config.h
	1520	+ $(XTCC) -o $(OBJDIR)/hname.o -c $(OBJDIR)/hname_.c
	1521	+
	1522	+$(OBJDIR)/hname.h: $(OBJDIR)/headers
1503	1523
1504	1524	$(OBJDIR)/http_.c: $(SRCDIR)/http.c $(TRANSLATE)
1505	1525	$(TRANSLATE) $(SRCDIR)/http.c >$@
1506	1526
1507	1527	$(OBJDIR)/http.o: $(OBJDIR)/http_.c $(OBJDIR)/http.h $(SRCDIR)/config.h
		@@ -1898,10 +1918,26 @@
1898	1918
1899	1919	$(OBJDIR)/sha1.o: $(OBJDIR)/sha1_.c $(OBJDIR)/sha1.h $(SRCDIR)/config.h
1900	1920	$(XTCC) -o $(OBJDIR)/sha1.o -c $(OBJDIR)/sha1_.c
1901	1921
1902	1922	$(OBJDIR)/sha1.h: $(OBJDIR)/headers
	1923	+
	1924	+$(OBJDIR)/sha1hard_.c: $(SRCDIR)/sha1hard.c $(TRANSLATE)
	1925	+ $(TRANSLATE) $(SRCDIR)/sha1hard.c >$@
	1926	+
	1927	+$(OBJDIR)/sha1hard.o: $(OBJDIR)/sha1hard_.c $(OBJDIR)/sha1hard.h $(SRCDIR)/config.h
	1928	+ $(XTCC) -o $(OBJDIR)/sha1hard.o -c $(OBJDIR)/sha1hard_.c
	1929	+
	1930	+$(OBJDIR)/sha1hard.h: $(OBJDIR)/headers
	1931	+
	1932	+$(OBJDIR)/sha3_.c: $(SRCDIR)/sha3.c $(TRANSLATE)
	1933	+ $(TRANSLATE) $(SRCDIR)/sha3.c >$@
	1934	+
	1935	+$(OBJDIR)/sha3.o: $(OBJDIR)/sha3_.c $(OBJDIR)/sha3.h $(SRCDIR)/config.h
	1936	+ $(XTCC) -o $(OBJDIR)/sha3.o -c $(OBJDIR)/sha3_.c
	1937	+
	1938	+$(OBJDIR)/sha3.h: $(OBJDIR)/headers
1903	1939
1904	1940	$(OBJDIR)/shun_.c: $(SRCDIR)/shun.c $(TRANSLATE)
1905	1941	$(TRANSLATE) $(SRCDIR)/shun.c >$@
1906	1942
1907	1943	$(OBJDIR)/shun.o: $(OBJDIR)/shun_.c $(OBJDIR)/shun.h $(SRCDIR)/config.h
1908	1944

	--- win/Makefile.mingw
	+++ win/Makefile.mingw
	@@ -461,10 +461,11 @@
461	$(SRCDIR)/fshell.c \
462	$(SRCDIR)/fusefs.c \
463	$(SRCDIR)/glob.c \
464	$(SRCDIR)/graph.c \
465	$(SRCDIR)/gzip.c \

466	$(SRCDIR)/http.c \
467	$(SRCDIR)/http_socket.c \
468	$(SRCDIR)/http_ssl.c \
469	$(SRCDIR)/http_transport.c \
470	$(SRCDIR)/import.c \
	@@ -511,10 +512,12 @@
511	$(SRCDIR)/rss.c \
512	$(SRCDIR)/schema.c \
513	$(SRCDIR)/search.c \
514	$(SRCDIR)/setup.c \
515	$(SRCDIR)/sha1.c \


516	$(SRCDIR)/shun.c \
517	$(SRCDIR)/sitemap.c \
518	$(SRCDIR)/skins.c \
519	$(SRCDIR)/sqlcmd.c \
520	$(SRCDIR)/stash.c \
	@@ -636,10 +639,11 @@
636	$(OBJDIR)/fshell_.c \
637	$(OBJDIR)/fusefs_.c \
638	$(OBJDIR)/glob_.c \
639	$(OBJDIR)/graph_.c \
640	$(OBJDIR)/gzip_.c \

641	$(OBJDIR)/http_.c \
642	$(OBJDIR)/http_socket_.c \
643	$(OBJDIR)/http_ssl_.c \
644	$(OBJDIR)/http_transport_.c \
645	$(OBJDIR)/import_.c \
	@@ -686,10 +690,12 @@
686	$(OBJDIR)/rss_.c \
687	$(OBJDIR)/schema_.c \
688	$(OBJDIR)/search_.c \
689	$(OBJDIR)/setup_.c \
690	$(OBJDIR)/sha1_.c \


691	$(OBJDIR)/shun_.c \
692	$(OBJDIR)/sitemap_.c \
693	$(OBJDIR)/skins_.c \
694	$(OBJDIR)/sqlcmd_.c \
695	$(OBJDIR)/stash_.c \
	@@ -760,10 +766,11 @@
760	$(OBJDIR)/fshell.o \
761	$(OBJDIR)/fusefs.o \
762	$(OBJDIR)/glob.o \
763	$(OBJDIR)/graph.o \
764	$(OBJDIR)/gzip.o \

765	$(OBJDIR)/http.o \
766	$(OBJDIR)/http_socket.o \
767	$(OBJDIR)/http_ssl.o \
768	$(OBJDIR)/http_transport.o \
769	$(OBJDIR)/import.o \
	@@ -810,10 +817,12 @@
810	$(OBJDIR)/rss.o \
811	$(OBJDIR)/schema.o \
812	$(OBJDIR)/search.o \
813	$(OBJDIR)/setup.o \
814	$(OBJDIR)/sha1.o \


815	$(OBJDIR)/shun.o \
816	$(OBJDIR)/sitemap.o \
817	$(OBJDIR)/skins.o \
818	$(OBJDIR)/sqlcmd.o \
819	$(OBJDIR)/stash.o \
	@@ -1095,10 +1104,11 @@
1095	$(OBJDIR)/fshell_.c:$(OBJDIR)/fshell.h \
1096	$(OBJDIR)/fusefs_.c:$(OBJDIR)/fusefs.h \
1097	$(OBJDIR)/glob_.c:$(OBJDIR)/glob.h \
1098	$(OBJDIR)/graph_.c:$(OBJDIR)/graph.h \
1099	$(OBJDIR)/gzip_.c:$(OBJDIR)/gzip.h \

1100	$(OBJDIR)/http_.c:$(OBJDIR)/http.h \
1101	$(OBJDIR)/http_socket_.c:$(OBJDIR)/http_socket.h \
1102	$(OBJDIR)/http_ssl_.c:$(OBJDIR)/http_ssl.h \
1103	$(OBJDIR)/http_transport_.c:$(OBJDIR)/http_transport.h \
1104	$(OBJDIR)/import_.c:$(OBJDIR)/import.h \
	@@ -1145,10 +1155,12 @@
1145	$(OBJDIR)/rss_.c:$(OBJDIR)/rss.h \
1146	$(OBJDIR)/schema_.c:$(OBJDIR)/schema.h \
1147	$(OBJDIR)/search_.c:$(OBJDIR)/search.h \
1148	$(OBJDIR)/setup_.c:$(OBJDIR)/setup.h \
1149	$(OBJDIR)/sha1_.c:$(OBJDIR)/sha1.h \


1150	$(OBJDIR)/shun_.c:$(OBJDIR)/shun.h \
1151	$(OBJDIR)/sitemap_.c:$(OBJDIR)/sitemap.h \
1152	$(OBJDIR)/skins_.c:$(OBJDIR)/skins.h \
1153	$(OBJDIR)/sqlcmd_.c:$(OBJDIR)/sqlcmd.h \
1154	$(OBJDIR)/stash_.c:$(OBJDIR)/stash.h \
	@@ -1498,10 +1510,18 @@
1498
1499	$(OBJDIR)/gzip.o: $(OBJDIR)/gzip_.c $(OBJDIR)/gzip.h $(SRCDIR)/config.h
1500	$(XTCC) -o $(OBJDIR)/gzip.o -c $(OBJDIR)/gzip_.c
1501
1502	$(OBJDIR)/gzip.h: $(OBJDIR)/headers








1503
1504	$(OBJDIR)/http_.c: $(SRCDIR)/http.c $(TRANSLATE)
1505	$(TRANSLATE) $(SRCDIR)/http.c >$@
1506
1507	$(OBJDIR)/http.o: $(OBJDIR)/http_.c $(OBJDIR)/http.h $(SRCDIR)/config.h
	@@ -1898,10 +1918,26 @@
1898
1899	$(OBJDIR)/sha1.o: $(OBJDIR)/sha1_.c $(OBJDIR)/sha1.h $(SRCDIR)/config.h
1900	$(XTCC) -o $(OBJDIR)/sha1.o -c $(OBJDIR)/sha1_.c
1901
1902	$(OBJDIR)/sha1.h: $(OBJDIR)/headers
















1903
1904	$(OBJDIR)/shun_.c: $(SRCDIR)/shun.c $(TRANSLATE)
1905	$(TRANSLATE) $(SRCDIR)/shun.c >$@
1906
1907	$(OBJDIR)/shun.o: $(OBJDIR)/shun_.c $(OBJDIR)/shun.h $(SRCDIR)/config.h
1908

	--- win/Makefile.mingw
	+++ win/Makefile.mingw
	@@ -461,10 +461,11 @@
461	$(SRCDIR)/fshell.c \
462	$(SRCDIR)/fusefs.c \
463	$(SRCDIR)/glob.c \
464	$(SRCDIR)/graph.c \
465	$(SRCDIR)/gzip.c \
466	$(SRCDIR)/hname.c \
467	$(SRCDIR)/http.c \
468	$(SRCDIR)/http_socket.c \
469	$(SRCDIR)/http_ssl.c \
470	$(SRCDIR)/http_transport.c \
471	$(SRCDIR)/import.c \
	@@ -511,10 +512,12 @@
512	$(SRCDIR)/rss.c \
513	$(SRCDIR)/schema.c \
514	$(SRCDIR)/search.c \
515	$(SRCDIR)/setup.c \
516	$(SRCDIR)/sha1.c \
517	$(SRCDIR)/sha1hard.c \
518	$(SRCDIR)/sha3.c \
519	$(SRCDIR)/shun.c \
520	$(SRCDIR)/sitemap.c \
521	$(SRCDIR)/skins.c \
522	$(SRCDIR)/sqlcmd.c \
523	$(SRCDIR)/stash.c \
	@@ -636,10 +639,11 @@
639	$(OBJDIR)/fshell_.c \
640	$(OBJDIR)/fusefs_.c \
641	$(OBJDIR)/glob_.c \
642	$(OBJDIR)/graph_.c \
643	$(OBJDIR)/gzip_.c \
644	$(OBJDIR)/hname_.c \
645	$(OBJDIR)/http_.c \
646	$(OBJDIR)/http_socket_.c \
647	$(OBJDIR)/http_ssl_.c \
648	$(OBJDIR)/http_transport_.c \
649	$(OBJDIR)/import_.c \
	@@ -686,10 +690,12 @@
690	$(OBJDIR)/rss_.c \
691	$(OBJDIR)/schema_.c \
692	$(OBJDIR)/search_.c \
693	$(OBJDIR)/setup_.c \
694	$(OBJDIR)/sha1_.c \
695	$(OBJDIR)/sha1hard_.c \
696	$(OBJDIR)/sha3_.c \
697	$(OBJDIR)/shun_.c \
698	$(OBJDIR)/sitemap_.c \
699	$(OBJDIR)/skins_.c \
700	$(OBJDIR)/sqlcmd_.c \
701	$(OBJDIR)/stash_.c \
	@@ -760,10 +766,11 @@
766	$(OBJDIR)/fshell.o \
767	$(OBJDIR)/fusefs.o \
768	$(OBJDIR)/glob.o \
769	$(OBJDIR)/graph.o \
770	$(OBJDIR)/gzip.o \
771	$(OBJDIR)/hname.o \
772	$(OBJDIR)/http.o \
773	$(OBJDIR)/http_socket.o \
774	$(OBJDIR)/http_ssl.o \
775	$(OBJDIR)/http_transport.o \
776	$(OBJDIR)/import.o \
	@@ -810,10 +817,12 @@
817	$(OBJDIR)/rss.o \
818	$(OBJDIR)/schema.o \
819	$(OBJDIR)/search.o \
820	$(OBJDIR)/setup.o \
821	$(OBJDIR)/sha1.o \
822	$(OBJDIR)/sha1hard.o \
823	$(OBJDIR)/sha3.o \
824	$(OBJDIR)/shun.o \
825	$(OBJDIR)/sitemap.o \
826	$(OBJDIR)/skins.o \
827	$(OBJDIR)/sqlcmd.o \
828	$(OBJDIR)/stash.o \
	@@ -1095,10 +1104,11 @@
1104	$(OBJDIR)/fshell_.c:$(OBJDIR)/fshell.h \
1105	$(OBJDIR)/fusefs_.c:$(OBJDIR)/fusefs.h \
1106	$(OBJDIR)/glob_.c:$(OBJDIR)/glob.h \
1107	$(OBJDIR)/graph_.c:$(OBJDIR)/graph.h \
1108	$(OBJDIR)/gzip_.c:$(OBJDIR)/gzip.h \
1109	$(OBJDIR)/hname_.c:$(OBJDIR)/hname.h \
1110	$(OBJDIR)/http_.c:$(OBJDIR)/http.h \
1111	$(OBJDIR)/http_socket_.c:$(OBJDIR)/http_socket.h \
1112	$(OBJDIR)/http_ssl_.c:$(OBJDIR)/http_ssl.h \
1113	$(OBJDIR)/http_transport_.c:$(OBJDIR)/http_transport.h \
1114	$(OBJDIR)/import_.c:$(OBJDIR)/import.h \
	@@ -1145,10 +1155,12 @@
1155	$(OBJDIR)/rss_.c:$(OBJDIR)/rss.h \
1156	$(OBJDIR)/schema_.c:$(OBJDIR)/schema.h \
1157	$(OBJDIR)/search_.c:$(OBJDIR)/search.h \
1158	$(OBJDIR)/setup_.c:$(OBJDIR)/setup.h \
1159	$(OBJDIR)/sha1_.c:$(OBJDIR)/sha1.h \
1160	$(OBJDIR)/sha1hard_.c:$(OBJDIR)/sha1hard.h \
1161	$(OBJDIR)/sha3_.c:$(OBJDIR)/sha3.h \
1162	$(OBJDIR)/shun_.c:$(OBJDIR)/shun.h \
1163	$(OBJDIR)/sitemap_.c:$(OBJDIR)/sitemap.h \
1164	$(OBJDIR)/skins_.c:$(OBJDIR)/skins.h \
1165	$(OBJDIR)/sqlcmd_.c:$(OBJDIR)/sqlcmd.h \
1166	$(OBJDIR)/stash_.c:$(OBJDIR)/stash.h \
	@@ -1498,10 +1510,18 @@
1510
1511	$(OBJDIR)/gzip.o: $(OBJDIR)/gzip_.c $(OBJDIR)/gzip.h $(SRCDIR)/config.h
1512	$(XTCC) -o $(OBJDIR)/gzip.o -c $(OBJDIR)/gzip_.c
1513
1514	$(OBJDIR)/gzip.h: $(OBJDIR)/headers
1515
1516	$(OBJDIR)/hname_.c: $(SRCDIR)/hname.c $(TRANSLATE)
1517	$(TRANSLATE) $(SRCDIR)/hname.c >$@
1518
1519	$(OBJDIR)/hname.o: $(OBJDIR)/hname_.c $(OBJDIR)/hname.h $(SRCDIR)/config.h
1520	$(XTCC) -o $(OBJDIR)/hname.o -c $(OBJDIR)/hname_.c
1521
1522	$(OBJDIR)/hname.h: $(OBJDIR)/headers
1523
1524	$(OBJDIR)/http_.c: $(SRCDIR)/http.c $(TRANSLATE)
1525	$(TRANSLATE) $(SRCDIR)/http.c >$@
1526
1527	$(OBJDIR)/http.o: $(OBJDIR)/http_.c $(OBJDIR)/http.h $(SRCDIR)/config.h
	@@ -1898,10 +1918,26 @@
1918
1919	$(OBJDIR)/sha1.o: $(OBJDIR)/sha1_.c $(OBJDIR)/sha1.h $(SRCDIR)/config.h
1920	$(XTCC) -o $(OBJDIR)/sha1.o -c $(OBJDIR)/sha1_.c
1921
1922	$(OBJDIR)/sha1.h: $(OBJDIR)/headers
1923
1924	$(OBJDIR)/sha1hard_.c: $(SRCDIR)/sha1hard.c $(TRANSLATE)
1925	$(TRANSLATE) $(SRCDIR)/sha1hard.c >$@
1926
1927	$(OBJDIR)/sha1hard.o: $(OBJDIR)/sha1hard_.c $(OBJDIR)/sha1hard.h $(SRCDIR)/config.h
1928	$(XTCC) -o $(OBJDIR)/sha1hard.o -c $(OBJDIR)/sha1hard_.c
1929
1930	$(OBJDIR)/sha1hard.h: $(OBJDIR)/headers
1931
1932	$(OBJDIR)/sha3_.c: $(SRCDIR)/sha3.c $(TRANSLATE)
1933	$(TRANSLATE) $(SRCDIR)/sha3.c >$@
1934
1935	$(OBJDIR)/sha3.o: $(OBJDIR)/sha3_.c $(OBJDIR)/sha3.h $(SRCDIR)/config.h
1936	$(XTCC) -o $(OBJDIR)/sha3.o -c $(OBJDIR)/sha3_.c
1937
1938	$(OBJDIR)/sha3.h: $(OBJDIR)/headers
1939
1940	$(OBJDIR)/shun_.c: $(SRCDIR)/shun.c $(TRANSLATE)
1941	$(TRANSLATE) $(SRCDIR)/shun.c >$@
1942
1943	$(OBJDIR)/shun.o: $(OBJDIR)/shun_.c $(OBJDIR)/shun.h $(SRCDIR)/config.h
1944

M win/Makefile.mingw.mistachkin

+36

		--- win/Makefile.mingw.mistachkin
		+++ win/Makefile.mingw.mistachkin
		@@ -461,10 +461,11 @@
461	461	$(SRCDIR)/fshell.c \
462	462	$(SRCDIR)/fusefs.c \
463	463	$(SRCDIR)/glob.c \
464	464	$(SRCDIR)/graph.c \
465	465	$(SRCDIR)/gzip.c \
	466	+ $(SRCDIR)/hname.c \
466	467	$(SRCDIR)/http.c \
467	468	$(SRCDIR)/http_socket.c \
468	469	$(SRCDIR)/http_ssl.c \
469	470	$(SRCDIR)/http_transport.c \
470	471	$(SRCDIR)/import.c \
		@@ -511,10 +512,12 @@
511	512	$(SRCDIR)/rss.c \
512	513	$(SRCDIR)/schema.c \
513	514	$(SRCDIR)/search.c \
514	515	$(SRCDIR)/setup.c \
515	516	$(SRCDIR)/sha1.c \
	517	+ $(SRCDIR)/sha1hard.c \
	518	+ $(SRCDIR)/sha3.c \
516	519	$(SRCDIR)/shun.c \
517	520	$(SRCDIR)/sitemap.c \
518	521	$(SRCDIR)/skins.c \
519	522	$(SRCDIR)/sqlcmd.c \
520	523	$(SRCDIR)/stash.c \
		@@ -636,10 +639,11 @@
636	639	$(OBJDIR)/fshell_.c \
637	640	$(OBJDIR)/fusefs_.c \
638	641	$(OBJDIR)/glob_.c \
639	642	$(OBJDIR)/graph_.c \
640	643	$(OBJDIR)/gzip_.c \
	644	+ $(OBJDIR)/hname_.c \
641	645	$(OBJDIR)/http_.c \
642	646	$(OBJDIR)/http_socket_.c \
643	647	$(OBJDIR)/http_ssl_.c \
644	648	$(OBJDIR)/http_transport_.c \
645	649	$(OBJDIR)/import_.c \
		@@ -686,10 +690,12 @@
686	690	$(OBJDIR)/rss_.c \
687	691	$(OBJDIR)/schema_.c \
688	692	$(OBJDIR)/search_.c \
689	693	$(OBJDIR)/setup_.c \
690	694	$(OBJDIR)/sha1_.c \
	695	+ $(OBJDIR)/sha1hard_.c \
	696	+ $(OBJDIR)/sha3_.c \
691	697	$(OBJDIR)/shun_.c \
692	698	$(OBJDIR)/sitemap_.c \
693	699	$(OBJDIR)/skins_.c \
694	700	$(OBJDIR)/sqlcmd_.c \
695	701	$(OBJDIR)/stash_.c \
		@@ -760,10 +766,11 @@
760	766	$(OBJDIR)/fshell.o \
761	767	$(OBJDIR)/fusefs.o \
762	768	$(OBJDIR)/glob.o \
763	769	$(OBJDIR)/graph.o \
764	770	$(OBJDIR)/gzip.o \
	771	+ $(OBJDIR)/hname.o \
765	772	$(OBJDIR)/http.o \
766	773	$(OBJDIR)/http_socket.o \
767	774	$(OBJDIR)/http_ssl.o \
768	775	$(OBJDIR)/http_transport.o \
769	776	$(OBJDIR)/import.o \
		@@ -810,10 +817,12 @@
810	817	$(OBJDIR)/rss.o \
811	818	$(OBJDIR)/schema.o \
812	819	$(OBJDIR)/search.o \
813	820	$(OBJDIR)/setup.o \
814	821	$(OBJDIR)/sha1.o \
	822	+ $(OBJDIR)/sha1hard.o \
	823	+ $(OBJDIR)/sha3.o \
815	824	$(OBJDIR)/shun.o \
816	825	$(OBJDIR)/sitemap.o \
817	826	$(OBJDIR)/skins.o \
818	827	$(OBJDIR)/sqlcmd.o \
819	828	$(OBJDIR)/stash.o \
		@@ -1095,10 +1104,11 @@
1095	1104	$(OBJDIR)/fshell_.c:$(OBJDIR)/fshell.h \
1096	1105	$(OBJDIR)/fusefs_.c:$(OBJDIR)/fusefs.h \
1097	1106	$(OBJDIR)/glob_.c:$(OBJDIR)/glob.h \
1098	1107	$(OBJDIR)/graph_.c:$(OBJDIR)/graph.h \
1099	1108	$(OBJDIR)/gzip_.c:$(OBJDIR)/gzip.h \
	1109	+ $(OBJDIR)/hname_.c:$(OBJDIR)/hname.h \
1100	1110	$(OBJDIR)/http_.c:$(OBJDIR)/http.h \
1101	1111	$(OBJDIR)/http_socket_.c:$(OBJDIR)/http_socket.h \
1102	1112	$(OBJDIR)/http_ssl_.c:$(OBJDIR)/http_ssl.h \
1103	1113	$(OBJDIR)/http_transport_.c:$(OBJDIR)/http_transport.h \
1104	1114	$(OBJDIR)/import_.c:$(OBJDIR)/import.h \
		@@ -1145,10 +1155,12 @@
1145	1155	$(OBJDIR)/rss_.c:$(OBJDIR)/rss.h \
1146	1156	$(OBJDIR)/schema_.c:$(OBJDIR)/schema.h \
1147	1157	$(OBJDIR)/search_.c:$(OBJDIR)/search.h \
1148	1158	$(OBJDIR)/setup_.c:$(OBJDIR)/setup.h \
1149	1159	$(OBJDIR)/sha1_.c:$(OBJDIR)/sha1.h \
	1160	+ $(OBJDIR)/sha1hard_.c:$(OBJDIR)/sha1hard.h \
	1161	+ $(OBJDIR)/sha3_.c:$(OBJDIR)/sha3.h \
1150	1162	$(OBJDIR)/shun_.c:$(OBJDIR)/shun.h \
1151	1163	$(OBJDIR)/sitemap_.c:$(OBJDIR)/sitemap.h \
1152	1164	$(OBJDIR)/skins_.c:$(OBJDIR)/skins.h \
1153	1165	$(OBJDIR)/sqlcmd_.c:$(OBJDIR)/sqlcmd.h \
1154	1166	$(OBJDIR)/stash_.c:$(OBJDIR)/stash.h \
		@@ -1498,10 +1510,18 @@
1498	1510
1499	1511	$(OBJDIR)/gzip.o: $(OBJDIR)/gzip_.c $(OBJDIR)/gzip.h $(SRCDIR)/config.h
1500	1512	$(XTCC) -o $(OBJDIR)/gzip.o -c $(OBJDIR)/gzip_.c
1501	1513
1502	1514	$(OBJDIR)/gzip.h: $(OBJDIR)/headers
	1515	+
	1516	+$(OBJDIR)/hname_.c: $(SRCDIR)/hname.c $(TRANSLATE)
	1517	+ $(TRANSLATE) $(SRCDIR)/hname.c >$@
	1518	+
	1519	+$(OBJDIR)/hname.o: $(OBJDIR)/hname_.c $(OBJDIR)/hname.h $(SRCDIR)/config.h
	1520	+ $(XTCC) -o $(OBJDIR)/hname.o -c $(OBJDIR)/hname_.c
	1521	+
	1522	+$(OBJDIR)/hname.h: $(OBJDIR)/headers
1503	1523
1504	1524	$(OBJDIR)/http_.c: $(SRCDIR)/http.c $(TRANSLATE)
1505	1525	$(TRANSLATE) $(SRCDIR)/http.c >$@
1506	1526
1507	1527	$(OBJDIR)/http.o: $(OBJDIR)/http_.c $(OBJDIR)/http.h $(SRCDIR)/config.h
		@@ -1898,10 +1918,26 @@
1898	1918
1899	1919	$(OBJDIR)/sha1.o: $(OBJDIR)/sha1_.c $(OBJDIR)/sha1.h $(SRCDIR)/config.h
1900	1920	$(XTCC) -o $(OBJDIR)/sha1.o -c $(OBJDIR)/sha1_.c
1901	1921
1902	1922	$(OBJDIR)/sha1.h: $(OBJDIR)/headers
	1923	+
	1924	+$(OBJDIR)/sha1hard_.c: $(SRCDIR)/sha1hard.c $(TRANSLATE)
	1925	+ $(TRANSLATE) $(SRCDIR)/sha1hard.c >$@
	1926	+
	1927	+$(OBJDIR)/sha1hard.o: $(OBJDIR)/sha1hard_.c $(OBJDIR)/sha1hard.h $(SRCDIR)/config.h
	1928	+ $(XTCC) -o $(OBJDIR)/sha1hard.o -c $(OBJDIR)/sha1hard_.c
	1929	+
	1930	+$(OBJDIR)/sha1hard.h: $(OBJDIR)/headers
	1931	+
	1932	+$(OBJDIR)/sha3_.c: $(SRCDIR)/sha3.c $(TRANSLATE)
	1933	+ $(TRANSLATE) $(SRCDIR)/sha3.c >$@
	1934	+
	1935	+$(OBJDIR)/sha3.o: $(OBJDIR)/sha3_.c $(OBJDIR)/sha3.h $(SRCDIR)/config.h
	1936	+ $(XTCC) -o $(OBJDIR)/sha3.o -c $(OBJDIR)/sha3_.c
	1937	+
	1938	+$(OBJDIR)/sha3.h: $(OBJDIR)/headers
1903	1939
1904	1940	$(OBJDIR)/shun_.c: $(SRCDIR)/shun.c $(TRANSLATE)
1905	1941	$(TRANSLATE) $(SRCDIR)/shun.c >$@
1906	1942
1907	1943	$(OBJDIR)/shun.o: $(OBJDIR)/shun_.c $(OBJDIR)/shun.h $(SRCDIR)/config.h
1908	1944

	--- win/Makefile.mingw.mistachkin
	+++ win/Makefile.mingw.mistachkin
	@@ -461,10 +461,11 @@
461	$(SRCDIR)/fshell.c \
462	$(SRCDIR)/fusefs.c \
463	$(SRCDIR)/glob.c \
464	$(SRCDIR)/graph.c \
465	$(SRCDIR)/gzip.c \

466	$(SRCDIR)/http.c \
467	$(SRCDIR)/http_socket.c \
468	$(SRCDIR)/http_ssl.c \
469	$(SRCDIR)/http_transport.c \
470	$(SRCDIR)/import.c \
	@@ -511,10 +512,12 @@
511	$(SRCDIR)/rss.c \
512	$(SRCDIR)/schema.c \
513	$(SRCDIR)/search.c \
514	$(SRCDIR)/setup.c \
515	$(SRCDIR)/sha1.c \


516	$(SRCDIR)/shun.c \
517	$(SRCDIR)/sitemap.c \
518	$(SRCDIR)/skins.c \
519	$(SRCDIR)/sqlcmd.c \
520	$(SRCDIR)/stash.c \
	@@ -636,10 +639,11 @@
636	$(OBJDIR)/fshell_.c \
637	$(OBJDIR)/fusefs_.c \
638	$(OBJDIR)/glob_.c \
639	$(OBJDIR)/graph_.c \
640	$(OBJDIR)/gzip_.c \

641	$(OBJDIR)/http_.c \
642	$(OBJDIR)/http_socket_.c \
643	$(OBJDIR)/http_ssl_.c \
644	$(OBJDIR)/http_transport_.c \
645	$(OBJDIR)/import_.c \
	@@ -686,10 +690,12 @@
686	$(OBJDIR)/rss_.c \
687	$(OBJDIR)/schema_.c \
688	$(OBJDIR)/search_.c \
689	$(OBJDIR)/setup_.c \
690	$(OBJDIR)/sha1_.c \


691	$(OBJDIR)/shun_.c \
692	$(OBJDIR)/sitemap_.c \
693	$(OBJDIR)/skins_.c \
694	$(OBJDIR)/sqlcmd_.c \
695	$(OBJDIR)/stash_.c \
	@@ -760,10 +766,11 @@
760	$(OBJDIR)/fshell.o \
761	$(OBJDIR)/fusefs.o \
762	$(OBJDIR)/glob.o \
763	$(OBJDIR)/graph.o \
764	$(OBJDIR)/gzip.o \

765	$(OBJDIR)/http.o \
766	$(OBJDIR)/http_socket.o \
767	$(OBJDIR)/http_ssl.o \
768	$(OBJDIR)/http_transport.o \
769	$(OBJDIR)/import.o \
	@@ -810,10 +817,12 @@
810	$(OBJDIR)/rss.o \
811	$(OBJDIR)/schema.o \
812	$(OBJDIR)/search.o \
813	$(OBJDIR)/setup.o \
814	$(OBJDIR)/sha1.o \


815	$(OBJDIR)/shun.o \
816	$(OBJDIR)/sitemap.o \
817	$(OBJDIR)/skins.o \
818	$(OBJDIR)/sqlcmd.o \
819	$(OBJDIR)/stash.o \
	@@ -1095,10 +1104,11 @@
1095	$(OBJDIR)/fshell_.c:$(OBJDIR)/fshell.h \
1096	$(OBJDIR)/fusefs_.c:$(OBJDIR)/fusefs.h \
1097	$(OBJDIR)/glob_.c:$(OBJDIR)/glob.h \
1098	$(OBJDIR)/graph_.c:$(OBJDIR)/graph.h \
1099	$(OBJDIR)/gzip_.c:$(OBJDIR)/gzip.h \

1100	$(OBJDIR)/http_.c:$(OBJDIR)/http.h \
1101	$(OBJDIR)/http_socket_.c:$(OBJDIR)/http_socket.h \
1102	$(OBJDIR)/http_ssl_.c:$(OBJDIR)/http_ssl.h \
1103	$(OBJDIR)/http_transport_.c:$(OBJDIR)/http_transport.h \
1104	$(OBJDIR)/import_.c:$(OBJDIR)/import.h \
	@@ -1145,10 +1155,12 @@
1145	$(OBJDIR)/rss_.c:$(OBJDIR)/rss.h \
1146	$(OBJDIR)/schema_.c:$(OBJDIR)/schema.h \
1147	$(OBJDIR)/search_.c:$(OBJDIR)/search.h \
1148	$(OBJDIR)/setup_.c:$(OBJDIR)/setup.h \
1149	$(OBJDIR)/sha1_.c:$(OBJDIR)/sha1.h \


1150	$(OBJDIR)/shun_.c:$(OBJDIR)/shun.h \
1151	$(OBJDIR)/sitemap_.c:$(OBJDIR)/sitemap.h \
1152	$(OBJDIR)/skins_.c:$(OBJDIR)/skins.h \
1153	$(OBJDIR)/sqlcmd_.c:$(OBJDIR)/sqlcmd.h \
1154	$(OBJDIR)/stash_.c:$(OBJDIR)/stash.h \
	@@ -1498,10 +1510,18 @@
1498
1499	$(OBJDIR)/gzip.o: $(OBJDIR)/gzip_.c $(OBJDIR)/gzip.h $(SRCDIR)/config.h
1500	$(XTCC) -o $(OBJDIR)/gzip.o -c $(OBJDIR)/gzip_.c
1501
1502	$(OBJDIR)/gzip.h: $(OBJDIR)/headers








1503
1504	$(OBJDIR)/http_.c: $(SRCDIR)/http.c $(TRANSLATE)
1505	$(TRANSLATE) $(SRCDIR)/http.c >$@
1506
1507	$(OBJDIR)/http.o: $(OBJDIR)/http_.c $(OBJDIR)/http.h $(SRCDIR)/config.h
	@@ -1898,10 +1918,26 @@
1898
1899	$(OBJDIR)/sha1.o: $(OBJDIR)/sha1_.c $(OBJDIR)/sha1.h $(SRCDIR)/config.h
1900	$(XTCC) -o $(OBJDIR)/sha1.o -c $(OBJDIR)/sha1_.c
1901
1902	$(OBJDIR)/sha1.h: $(OBJDIR)/headers
















1903
1904	$(OBJDIR)/shun_.c: $(SRCDIR)/shun.c $(TRANSLATE)
1905	$(TRANSLATE) $(SRCDIR)/shun.c >$@
1906
1907	$(OBJDIR)/shun.o: $(OBJDIR)/shun_.c $(OBJDIR)/shun.h $(SRCDIR)/config.h
1908

	--- win/Makefile.mingw.mistachkin
	+++ win/Makefile.mingw.mistachkin
	@@ -461,10 +461,11 @@
461	$(SRCDIR)/fshell.c \
462	$(SRCDIR)/fusefs.c \
463	$(SRCDIR)/glob.c \
464	$(SRCDIR)/graph.c \
465	$(SRCDIR)/gzip.c \
466	$(SRCDIR)/hname.c \
467	$(SRCDIR)/http.c \
468	$(SRCDIR)/http_socket.c \
469	$(SRCDIR)/http_ssl.c \
470	$(SRCDIR)/http_transport.c \
471	$(SRCDIR)/import.c \
	@@ -511,10 +512,12 @@
512	$(SRCDIR)/rss.c \
513	$(SRCDIR)/schema.c \
514	$(SRCDIR)/search.c \
515	$(SRCDIR)/setup.c \
516	$(SRCDIR)/sha1.c \
517	$(SRCDIR)/sha1hard.c \
518	$(SRCDIR)/sha3.c \
519	$(SRCDIR)/shun.c \
520	$(SRCDIR)/sitemap.c \
521	$(SRCDIR)/skins.c \
522	$(SRCDIR)/sqlcmd.c \
523	$(SRCDIR)/stash.c \
	@@ -636,10 +639,11 @@
639	$(OBJDIR)/fshell_.c \
640	$(OBJDIR)/fusefs_.c \
641	$(OBJDIR)/glob_.c \
642	$(OBJDIR)/graph_.c \
643	$(OBJDIR)/gzip_.c \
644	$(OBJDIR)/hname_.c \
645	$(OBJDIR)/http_.c \
646	$(OBJDIR)/http_socket_.c \
647	$(OBJDIR)/http_ssl_.c \
648	$(OBJDIR)/http_transport_.c \
649	$(OBJDIR)/import_.c \
	@@ -686,10 +690,12 @@
690	$(OBJDIR)/rss_.c \
691	$(OBJDIR)/schema_.c \
692	$(OBJDIR)/search_.c \
693	$(OBJDIR)/setup_.c \
694	$(OBJDIR)/sha1_.c \
695	$(OBJDIR)/sha1hard_.c \
696	$(OBJDIR)/sha3_.c \
697	$(OBJDIR)/shun_.c \
698	$(OBJDIR)/sitemap_.c \
699	$(OBJDIR)/skins_.c \
700	$(OBJDIR)/sqlcmd_.c \
701	$(OBJDIR)/stash_.c \
	@@ -760,10 +766,11 @@
766	$(OBJDIR)/fshell.o \
767	$(OBJDIR)/fusefs.o \
768	$(OBJDIR)/glob.o \
769	$(OBJDIR)/graph.o \
770	$(OBJDIR)/gzip.o \
771	$(OBJDIR)/hname.o \
772	$(OBJDIR)/http.o \
773	$(OBJDIR)/http_socket.o \
774	$(OBJDIR)/http_ssl.o \
775	$(OBJDIR)/http_transport.o \
776	$(OBJDIR)/import.o \
	@@ -810,10 +817,12 @@
817	$(OBJDIR)/rss.o \
818	$(OBJDIR)/schema.o \
819	$(OBJDIR)/search.o \
820	$(OBJDIR)/setup.o \
821	$(OBJDIR)/sha1.o \
822	$(OBJDIR)/sha1hard.o \
823	$(OBJDIR)/sha3.o \
824	$(OBJDIR)/shun.o \
825	$(OBJDIR)/sitemap.o \
826	$(OBJDIR)/skins.o \
827	$(OBJDIR)/sqlcmd.o \
828	$(OBJDIR)/stash.o \
	@@ -1095,10 +1104,11 @@
1104	$(OBJDIR)/fshell_.c:$(OBJDIR)/fshell.h \
1105	$(OBJDIR)/fusefs_.c:$(OBJDIR)/fusefs.h \
1106	$(OBJDIR)/glob_.c:$(OBJDIR)/glob.h \
1107	$(OBJDIR)/graph_.c:$(OBJDIR)/graph.h \
1108	$(OBJDIR)/gzip_.c:$(OBJDIR)/gzip.h \
1109	$(OBJDIR)/hname_.c:$(OBJDIR)/hname.h \
1110	$(OBJDIR)/http_.c:$(OBJDIR)/http.h \
1111	$(OBJDIR)/http_socket_.c:$(OBJDIR)/http_socket.h \
1112	$(OBJDIR)/http_ssl_.c:$(OBJDIR)/http_ssl.h \
1113	$(OBJDIR)/http_transport_.c:$(OBJDIR)/http_transport.h \
1114	$(OBJDIR)/import_.c:$(OBJDIR)/import.h \
	@@ -1145,10 +1155,12 @@
1155	$(OBJDIR)/rss_.c:$(OBJDIR)/rss.h \
1156	$(OBJDIR)/schema_.c:$(OBJDIR)/schema.h \
1157	$(OBJDIR)/search_.c:$(OBJDIR)/search.h \
1158	$(OBJDIR)/setup_.c:$(OBJDIR)/setup.h \
1159	$(OBJDIR)/sha1_.c:$(OBJDIR)/sha1.h \
1160	$(OBJDIR)/sha1hard_.c:$(OBJDIR)/sha1hard.h \
1161	$(OBJDIR)/sha3_.c:$(OBJDIR)/sha3.h \
1162	$(OBJDIR)/shun_.c:$(OBJDIR)/shun.h \
1163	$(OBJDIR)/sitemap_.c:$(OBJDIR)/sitemap.h \
1164	$(OBJDIR)/skins_.c:$(OBJDIR)/skins.h \
1165	$(OBJDIR)/sqlcmd_.c:$(OBJDIR)/sqlcmd.h \
1166	$(OBJDIR)/stash_.c:$(OBJDIR)/stash.h \
	@@ -1498,10 +1510,18 @@
1510
1511	$(OBJDIR)/gzip.o: $(OBJDIR)/gzip_.c $(OBJDIR)/gzip.h $(SRCDIR)/config.h
1512	$(XTCC) -o $(OBJDIR)/gzip.o -c $(OBJDIR)/gzip_.c
1513
1514	$(OBJDIR)/gzip.h: $(OBJDIR)/headers
1515
1516	$(OBJDIR)/hname_.c: $(SRCDIR)/hname.c $(TRANSLATE)
1517	$(TRANSLATE) $(SRCDIR)/hname.c >$@
1518
1519	$(OBJDIR)/hname.o: $(OBJDIR)/hname_.c $(OBJDIR)/hname.h $(SRCDIR)/config.h
1520	$(XTCC) -o $(OBJDIR)/hname.o -c $(OBJDIR)/hname_.c
1521
1522	$(OBJDIR)/hname.h: $(OBJDIR)/headers
1523
1524	$(OBJDIR)/http_.c: $(SRCDIR)/http.c $(TRANSLATE)
1525	$(TRANSLATE) $(SRCDIR)/http.c >$@
1526
1527	$(OBJDIR)/http.o: $(OBJDIR)/http_.c $(OBJDIR)/http.h $(SRCDIR)/config.h
	@@ -1898,10 +1918,26 @@
1918
1919	$(OBJDIR)/sha1.o: $(OBJDIR)/sha1_.c $(OBJDIR)/sha1.h $(SRCDIR)/config.h
1920	$(XTCC) -o $(OBJDIR)/sha1.o -c $(OBJDIR)/sha1_.c
1921
1922	$(OBJDIR)/sha1.h: $(OBJDIR)/headers
1923
1924	$(OBJDIR)/sha1hard_.c: $(SRCDIR)/sha1hard.c $(TRANSLATE)
1925	$(TRANSLATE) $(SRCDIR)/sha1hard.c >$@
1926
1927	$(OBJDIR)/sha1hard.o: $(OBJDIR)/sha1hard_.c $(OBJDIR)/sha1hard.h $(SRCDIR)/config.h
1928	$(XTCC) -o $(OBJDIR)/sha1hard.o -c $(OBJDIR)/sha1hard_.c
1929
1930	$(OBJDIR)/sha1hard.h: $(OBJDIR)/headers
1931
1932	$(OBJDIR)/sha3_.c: $(SRCDIR)/sha3.c $(TRANSLATE)
1933	$(TRANSLATE) $(SRCDIR)/sha3.c >$@
1934
1935	$(OBJDIR)/sha3.o: $(OBJDIR)/sha3_.c $(OBJDIR)/sha3.h $(SRCDIR)/config.h
1936	$(XTCC) -o $(OBJDIR)/sha3.o -c $(OBJDIR)/sha3_.c
1937
1938	$(OBJDIR)/sha3.h: $(OBJDIR)/headers
1939
1940	$(OBJDIR)/shun_.c: $(SRCDIR)/shun.c $(TRANSLATE)
1941	$(TRANSLATE) $(SRCDIR)/shun.c >$@
1942
1943	$(OBJDIR)/shun.o: $(OBJDIR)/shun_.c $(OBJDIR)/shun.h $(SRCDIR)/config.h
1944

M win/Makefile.mingw.mistachkin

+36

		--- win/Makefile.mingw.mistachkin
		+++ win/Makefile.mingw.mistachkin
		@@ -461,10 +461,11 @@
461	461	$(SRCDIR)/fshell.c \
462	462	$(SRCDIR)/fusefs.c \
463	463	$(SRCDIR)/glob.c \
464	464	$(SRCDIR)/graph.c \
465	465	$(SRCDIR)/gzip.c \
	466	+ $(SRCDIR)/hname.c \
466	467	$(SRCDIR)/http.c \
467	468	$(SRCDIR)/http_socket.c \
468	469	$(SRCDIR)/http_ssl.c \
469	470	$(SRCDIR)/http_transport.c \
470	471	$(SRCDIR)/import.c \
		@@ -511,10 +512,12 @@
511	512	$(SRCDIR)/rss.c \
512	513	$(SRCDIR)/schema.c \
513	514	$(SRCDIR)/search.c \
514	515	$(SRCDIR)/setup.c \
515	516	$(SRCDIR)/sha1.c \
	517	+ $(SRCDIR)/sha1hard.c \
	518	+ $(SRCDIR)/sha3.c \
516	519	$(SRCDIR)/shun.c \
517	520	$(SRCDIR)/sitemap.c \
518	521	$(SRCDIR)/skins.c \
519	522	$(SRCDIR)/sqlcmd.c \
520	523	$(SRCDIR)/stash.c \
		@@ -636,10 +639,11 @@
636	639	$(OBJDIR)/fshell_.c \
637	640	$(OBJDIR)/fusefs_.c \
638	641	$(OBJDIR)/glob_.c \
639	642	$(OBJDIR)/graph_.c \
640	643	$(OBJDIR)/gzip_.c \
	644	+ $(OBJDIR)/hname_.c \
641	645	$(OBJDIR)/http_.c \
642	646	$(OBJDIR)/http_socket_.c \
643	647	$(OBJDIR)/http_ssl_.c \
644	648	$(OBJDIR)/http_transport_.c \
645	649	$(OBJDIR)/import_.c \
		@@ -686,10 +690,12 @@
686	690	$(OBJDIR)/rss_.c \
687	691	$(OBJDIR)/schema_.c \
688	692	$(OBJDIR)/search_.c \
689	693	$(OBJDIR)/setup_.c \
690	694	$(OBJDIR)/sha1_.c \
	695	+ $(OBJDIR)/sha1hard_.c \
	696	+ $(OBJDIR)/sha3_.c \
691	697	$(OBJDIR)/shun_.c \
692	698	$(OBJDIR)/sitemap_.c \
693	699	$(OBJDIR)/skins_.c \
694	700	$(OBJDIR)/sqlcmd_.c \
695	701	$(OBJDIR)/stash_.c \
		@@ -760,10 +766,11 @@
760	766	$(OBJDIR)/fshell.o \
761	767	$(OBJDIR)/fusefs.o \
762	768	$(OBJDIR)/glob.o \
763	769	$(OBJDIR)/graph.o \
764	770	$(OBJDIR)/gzip.o \
	771	+ $(OBJDIR)/hname.o \
765	772	$(OBJDIR)/http.o \
766	773	$(OBJDIR)/http_socket.o \
767	774	$(OBJDIR)/http_ssl.o \
768	775	$(OBJDIR)/http_transport.o \
769	776	$(OBJDIR)/import.o \
		@@ -810,10 +817,12 @@
810	817	$(OBJDIR)/rss.o \
811	818	$(OBJDIR)/schema.o \
812	819	$(OBJDIR)/search.o \
813	820	$(OBJDIR)/setup.o \
814	821	$(OBJDIR)/sha1.o \
	822	+ $(OBJDIR)/sha1hard.o \
	823	+ $(OBJDIR)/sha3.o \
815	824	$(OBJDIR)/shun.o \
816	825	$(OBJDIR)/sitemap.o \
817	826	$(OBJDIR)/skins.o \
818	827	$(OBJDIR)/sqlcmd.o \
819	828	$(OBJDIR)/stash.o \
		@@ -1095,10 +1104,11 @@
1095	1104	$(OBJDIR)/fshell_.c:$(OBJDIR)/fshell.h \
1096	1105	$(OBJDIR)/fusefs_.c:$(OBJDIR)/fusefs.h \
1097	1106	$(OBJDIR)/glob_.c:$(OBJDIR)/glob.h \
1098	1107	$(OBJDIR)/graph_.c:$(OBJDIR)/graph.h \
1099	1108	$(OBJDIR)/gzip_.c:$(OBJDIR)/gzip.h \
	1109	+ $(OBJDIR)/hname_.c:$(OBJDIR)/hname.h \
1100	1110	$(OBJDIR)/http_.c:$(OBJDIR)/http.h \
1101	1111	$(OBJDIR)/http_socket_.c:$(OBJDIR)/http_socket.h \
1102	1112	$(OBJDIR)/http_ssl_.c:$(OBJDIR)/http_ssl.h \
1103	1113	$(OBJDIR)/http_transport_.c:$(OBJDIR)/http_transport.h \
1104	1114	$(OBJDIR)/import_.c:$(OBJDIR)/import.h \
		@@ -1145,10 +1155,12 @@
1145	1155	$(OBJDIR)/rss_.c:$(OBJDIR)/rss.h \
1146	1156	$(OBJDIR)/schema_.c:$(OBJDIR)/schema.h \
1147	1157	$(OBJDIR)/search_.c:$(OBJDIR)/search.h \
1148	1158	$(OBJDIR)/setup_.c:$(OBJDIR)/setup.h \
1149	1159	$(OBJDIR)/sha1_.c:$(OBJDIR)/sha1.h \
	1160	+ $(OBJDIR)/sha1hard_.c:$(OBJDIR)/sha1hard.h \
	1161	+ $(OBJDIR)/sha3_.c:$(OBJDIR)/sha3.h \
1150	1162	$(OBJDIR)/shun_.c:$(OBJDIR)/shun.h \
1151	1163	$(OBJDIR)/sitemap_.c:$(OBJDIR)/sitemap.h \
1152	1164	$(OBJDIR)/skins_.c:$(OBJDIR)/skins.h \
1153	1165	$(OBJDIR)/sqlcmd_.c:$(OBJDIR)/sqlcmd.h \
1154	1166	$(OBJDIR)/stash_.c:$(OBJDIR)/stash.h \
		@@ -1498,10 +1510,18 @@
1498	1510
1499	1511	$(OBJDIR)/gzip.o: $(OBJDIR)/gzip_.c $(OBJDIR)/gzip.h $(SRCDIR)/config.h
1500	1512	$(XTCC) -o $(OBJDIR)/gzip.o -c $(OBJDIR)/gzip_.c
1501	1513
1502	1514	$(OBJDIR)/gzip.h: $(OBJDIR)/headers
	1515	+
	1516	+$(OBJDIR)/hname_.c: $(SRCDIR)/hname.c $(TRANSLATE)
	1517	+ $(TRANSLATE) $(SRCDIR)/hname.c >$@
	1518	+
	1519	+$(OBJDIR)/hname.o: $(OBJDIR)/hname_.c $(OBJDIR)/hname.h $(SRCDIR)/config.h
	1520	+ $(XTCC) -o $(OBJDIR)/hname.o -c $(OBJDIR)/hname_.c
	1521	+
	1522	+$(OBJDIR)/hname.h: $(OBJDIR)/headers
1503	1523
1504	1524	$(OBJDIR)/http_.c: $(SRCDIR)/http.c $(TRANSLATE)
1505	1525	$(TRANSLATE) $(SRCDIR)/http.c >$@
1506	1526
1507	1527	$(OBJDIR)/http.o: $(OBJDIR)/http_.c $(OBJDIR)/http.h $(SRCDIR)/config.h
		@@ -1898,10 +1918,26 @@
1898	1918
1899	1919	$(OBJDIR)/sha1.o: $(OBJDIR)/sha1_.c $(OBJDIR)/sha1.h $(SRCDIR)/config.h
1900	1920	$(XTCC) -o $(OBJDIR)/sha1.o -c $(OBJDIR)/sha1_.c
1901	1921
1902	1922	$(OBJDIR)/sha1.h: $(OBJDIR)/headers
	1923	+
	1924	+$(OBJDIR)/sha1hard_.c: $(SRCDIR)/sha1hard.c $(TRANSLATE)
	1925	+ $(TRANSLATE) $(SRCDIR)/sha1hard.c >$@
	1926	+
	1927	+$(OBJDIR)/sha1hard.o: $(OBJDIR)/sha1hard_.c $(OBJDIR)/sha1hard.h $(SRCDIR)/config.h
	1928	+ $(XTCC) -o $(OBJDIR)/sha1hard.o -c $(OBJDIR)/sha1hard_.c
	1929	+
	1930	+$(OBJDIR)/sha1hard.h: $(OBJDIR)/headers
	1931	+
	1932	+$(OBJDIR)/sha3_.c: $(SRCDIR)/sha3.c $(TRANSLATE)
	1933	+ $(TRANSLATE) $(SRCDIR)/sha3.c >$@
	1934	+
	1935	+$(OBJDIR)/sha3.o: $(OBJDIR)/sha3_.c $(OBJDIR)/sha3.h $(SRCDIR)/config.h
	1936	+ $(XTCC) -o $(OBJDIR)/sha3.o -c $(OBJDIR)/sha3_.c
	1937	+
	1938	+$(OBJDIR)/sha3.h: $(OBJDIR)/headers
1903	1939
1904	1940	$(OBJDIR)/shun_.c: $(SRCDIR)/shun.c $(TRANSLATE)
1905	1941	$(TRANSLATE) $(SRCDIR)/shun.c >$@
1906	1942
1907	1943	$(OBJDIR)/shun.o: $(OBJDIR)/shun_.c $(OBJDIR)/shun.h $(SRCDIR)/config.h
1908	1944

	--- win/Makefile.mingw.mistachkin
	+++ win/Makefile.mingw.mistachkin
	@@ -461,10 +461,11 @@
461	$(SRCDIR)/fshell.c \
462	$(SRCDIR)/fusefs.c \
463	$(SRCDIR)/glob.c \
464	$(SRCDIR)/graph.c \
465	$(SRCDIR)/gzip.c \

466	$(SRCDIR)/http.c \
467	$(SRCDIR)/http_socket.c \
468	$(SRCDIR)/http_ssl.c \
469	$(SRCDIR)/http_transport.c \
470	$(SRCDIR)/import.c \
	@@ -511,10 +512,12 @@
511	$(SRCDIR)/rss.c \
512	$(SRCDIR)/schema.c \
513	$(SRCDIR)/search.c \
514	$(SRCDIR)/setup.c \
515	$(SRCDIR)/sha1.c \


516	$(SRCDIR)/shun.c \
517	$(SRCDIR)/sitemap.c \
518	$(SRCDIR)/skins.c \
519	$(SRCDIR)/sqlcmd.c \
520	$(SRCDIR)/stash.c \
	@@ -636,10 +639,11 @@
636	$(OBJDIR)/fshell_.c \
637	$(OBJDIR)/fusefs_.c \
638	$(OBJDIR)/glob_.c \
639	$(OBJDIR)/graph_.c \
640	$(OBJDIR)/gzip_.c \

641	$(OBJDIR)/http_.c \
642	$(OBJDIR)/http_socket_.c \
643	$(OBJDIR)/http_ssl_.c \
644	$(OBJDIR)/http_transport_.c \
645	$(OBJDIR)/import_.c \
	@@ -686,10 +690,12 @@
686	$(OBJDIR)/rss_.c \
687	$(OBJDIR)/schema_.c \
688	$(OBJDIR)/search_.c \
689	$(OBJDIR)/setup_.c \
690	$(OBJDIR)/sha1_.c \


691	$(OBJDIR)/shun_.c \
692	$(OBJDIR)/sitemap_.c \
693	$(OBJDIR)/skins_.c \
694	$(OBJDIR)/sqlcmd_.c \
695	$(OBJDIR)/stash_.c \
	@@ -760,10 +766,11 @@
760	$(OBJDIR)/fshell.o \
761	$(OBJDIR)/fusefs.o \
762	$(OBJDIR)/glob.o \
763	$(OBJDIR)/graph.o \
764	$(OBJDIR)/gzip.o \

765	$(OBJDIR)/http.o \
766	$(OBJDIR)/http_socket.o \
767	$(OBJDIR)/http_ssl.o \
768	$(OBJDIR)/http_transport.o \
769	$(OBJDIR)/import.o \
	@@ -810,10 +817,12 @@
810	$(OBJDIR)/rss.o \
811	$(OBJDIR)/schema.o \
812	$(OBJDIR)/search.o \
813	$(OBJDIR)/setup.o \
814	$(OBJDIR)/sha1.o \


815	$(OBJDIR)/shun.o \
816	$(OBJDIR)/sitemap.o \
817	$(OBJDIR)/skins.o \
818	$(OBJDIR)/sqlcmd.o \
819	$(OBJDIR)/stash.o \
	@@ -1095,10 +1104,11 @@
1095	$(OBJDIR)/fshell_.c:$(OBJDIR)/fshell.h \
1096	$(OBJDIR)/fusefs_.c:$(OBJDIR)/fusefs.h \
1097	$(OBJDIR)/glob_.c:$(OBJDIR)/glob.h \
1098	$(OBJDIR)/graph_.c:$(OBJDIR)/graph.h \
1099	$(OBJDIR)/gzip_.c:$(OBJDIR)/gzip.h \

1100	$(OBJDIR)/http_.c:$(OBJDIR)/http.h \
1101	$(OBJDIR)/http_socket_.c:$(OBJDIR)/http_socket.h \
1102	$(OBJDIR)/http_ssl_.c:$(OBJDIR)/http_ssl.h \
1103	$(OBJDIR)/http_transport_.c:$(OBJDIR)/http_transport.h \
1104	$(OBJDIR)/import_.c:$(OBJDIR)/import.h \
	@@ -1145,10 +1155,12 @@
1145	$(OBJDIR)/rss_.c:$(OBJDIR)/rss.h \
1146	$(OBJDIR)/schema_.c:$(OBJDIR)/schema.h \
1147	$(OBJDIR)/search_.c:$(OBJDIR)/search.h \
1148	$(OBJDIR)/setup_.c:$(OBJDIR)/setup.h \
1149	$(OBJDIR)/sha1_.c:$(OBJDIR)/sha1.h \


1150	$(OBJDIR)/shun_.c:$(OBJDIR)/shun.h \
1151	$(OBJDIR)/sitemap_.c:$(OBJDIR)/sitemap.h \
1152	$(OBJDIR)/skins_.c:$(OBJDIR)/skins.h \
1153	$(OBJDIR)/sqlcmd_.c:$(OBJDIR)/sqlcmd.h \
1154	$(OBJDIR)/stash_.c:$(OBJDIR)/stash.h \
	@@ -1498,10 +1510,18 @@
1498
1499	$(OBJDIR)/gzip.o: $(OBJDIR)/gzip_.c $(OBJDIR)/gzip.h $(SRCDIR)/config.h
1500	$(XTCC) -o $(OBJDIR)/gzip.o -c $(OBJDIR)/gzip_.c
1501
1502	$(OBJDIR)/gzip.h: $(OBJDIR)/headers








1503
1504	$(OBJDIR)/http_.c: $(SRCDIR)/http.c $(TRANSLATE)
1505	$(TRANSLATE) $(SRCDIR)/http.c >$@
1506
1507	$(OBJDIR)/http.o: $(OBJDIR)/http_.c $(OBJDIR)/http.h $(SRCDIR)/config.h
	@@ -1898,10 +1918,26 @@
1898
1899	$(OBJDIR)/sha1.o: $(OBJDIR)/sha1_.c $(OBJDIR)/sha1.h $(SRCDIR)/config.h
1900	$(XTCC) -o $(OBJDIR)/sha1.o -c $(OBJDIR)/sha1_.c
1901
1902	$(OBJDIR)/sha1.h: $(OBJDIR)/headers
















1903
1904	$(OBJDIR)/shun_.c: $(SRCDIR)/shun.c $(TRANSLATE)
1905	$(TRANSLATE) $(SRCDIR)/shun.c >$@
1906
1907	$(OBJDIR)/shun.o: $(OBJDIR)/shun_.c $(OBJDIR)/shun.h $(SRCDIR)/config.h
1908

	--- win/Makefile.mingw.mistachkin
	+++ win/Makefile.mingw.mistachkin
	@@ -461,10 +461,11 @@
461	$(SRCDIR)/fshell.c \
462	$(SRCDIR)/fusefs.c \
463	$(SRCDIR)/glob.c \
464	$(SRCDIR)/graph.c \
465	$(SRCDIR)/gzip.c \
466	$(SRCDIR)/hname.c \
467	$(SRCDIR)/http.c \
468	$(SRCDIR)/http_socket.c \
469	$(SRCDIR)/http_ssl.c \
470	$(SRCDIR)/http_transport.c \
471	$(SRCDIR)/import.c \
	@@ -511,10 +512,12 @@
512	$(SRCDIR)/rss.c \
513	$(SRCDIR)/schema.c \
514	$(SRCDIR)/search.c \
515	$(SRCDIR)/setup.c \
516	$(SRCDIR)/sha1.c \
517	$(SRCDIR)/sha1hard.c \
518	$(SRCDIR)/sha3.c \
519	$(SRCDIR)/shun.c \
520	$(SRCDIR)/sitemap.c \
521	$(SRCDIR)/skins.c \
522	$(SRCDIR)/sqlcmd.c \
523	$(SRCDIR)/stash.c \
	@@ -636,10 +639,11 @@
639	$(OBJDIR)/fshell_.c \
640	$(OBJDIR)/fusefs_.c \
641	$(OBJDIR)/glob_.c \
642	$(OBJDIR)/graph_.c \
643	$(OBJDIR)/gzip_.c \
644	$(OBJDIR)/hname_.c \
645	$(OBJDIR)/http_.c \
646	$(OBJDIR)/http_socket_.c \
647	$(OBJDIR)/http_ssl_.c \
648	$(OBJDIR)/http_transport_.c \
649	$(OBJDIR)/import_.c \
	@@ -686,10 +690,12 @@
690	$(OBJDIR)/rss_.c \
691	$(OBJDIR)/schema_.c \
692	$(OBJDIR)/search_.c \
693	$(OBJDIR)/setup_.c \
694	$(OBJDIR)/sha1_.c \
695	$(OBJDIR)/sha1hard_.c \
696	$(OBJDIR)/sha3_.c \
697	$(OBJDIR)/shun_.c \
698	$(OBJDIR)/sitemap_.c \
699	$(OBJDIR)/skins_.c \
700	$(OBJDIR)/sqlcmd_.c \
701	$(OBJDIR)/stash_.c \
	@@ -760,10 +766,11 @@
766	$(OBJDIR)/fshell.o \
767	$(OBJDIR)/fusefs.o \
768	$(OBJDIR)/glob.o \
769	$(OBJDIR)/graph.o \
770	$(OBJDIR)/gzip.o \
771	$(OBJDIR)/hname.o \
772	$(OBJDIR)/http.o \
773	$(OBJDIR)/http_socket.o \
774	$(OBJDIR)/http_ssl.o \
775	$(OBJDIR)/http_transport.o \
776	$(OBJDIR)/import.o \
	@@ -810,10 +817,12 @@
817	$(OBJDIR)/rss.o \
818	$(OBJDIR)/schema.o \
819	$(OBJDIR)/search.o \
820	$(OBJDIR)/setup.o \
821	$(OBJDIR)/sha1.o \
822	$(OBJDIR)/sha1hard.o \
823	$(OBJDIR)/sha3.o \
824	$(OBJDIR)/shun.o \
825	$(OBJDIR)/sitemap.o \
826	$(OBJDIR)/skins.o \
827	$(OBJDIR)/sqlcmd.o \
828	$(OBJDIR)/stash.o \
	@@ -1095,10 +1104,11 @@
1104	$(OBJDIR)/fshell_.c:$(OBJDIR)/fshell.h \
1105	$(OBJDIR)/fusefs_.c:$(OBJDIR)/fusefs.h \
1106	$(OBJDIR)/glob_.c:$(OBJDIR)/glob.h \
1107	$(OBJDIR)/graph_.c:$(OBJDIR)/graph.h \
1108	$(OBJDIR)/gzip_.c:$(OBJDIR)/gzip.h \
1109	$(OBJDIR)/hname_.c:$(OBJDIR)/hname.h \
1110	$(OBJDIR)/http_.c:$(OBJDIR)/http.h \
1111	$(OBJDIR)/http_socket_.c:$(OBJDIR)/http_socket.h \
1112	$(OBJDIR)/http_ssl_.c:$(OBJDIR)/http_ssl.h \
1113	$(OBJDIR)/http_transport_.c:$(OBJDIR)/http_transport.h \
1114	$(OBJDIR)/import_.c:$(OBJDIR)/import.h \
	@@ -1145,10 +1155,12 @@
1155	$(OBJDIR)/rss_.c:$(OBJDIR)/rss.h \
1156	$(OBJDIR)/schema_.c:$(OBJDIR)/schema.h \
1157	$(OBJDIR)/search_.c:$(OBJDIR)/search.h \
1158	$(OBJDIR)/setup_.c:$(OBJDIR)/setup.h \
1159	$(OBJDIR)/sha1_.c:$(OBJDIR)/sha1.h \
1160	$(OBJDIR)/sha1hard_.c:$(OBJDIR)/sha1hard.h \
1161	$(OBJDIR)/sha3_.c:$(OBJDIR)/sha3.h \
1162	$(OBJDIR)/shun_.c:$(OBJDIR)/shun.h \
1163	$(OBJDIR)/sitemap_.c:$(OBJDIR)/sitemap.h \
1164	$(OBJDIR)/skins_.c:$(OBJDIR)/skins.h \
1165	$(OBJDIR)/sqlcmd_.c:$(OBJDIR)/sqlcmd.h \
1166	$(OBJDIR)/stash_.c:$(OBJDIR)/stash.h \
	@@ -1498,10 +1510,18 @@
1510
1511	$(OBJDIR)/gzip.o: $(OBJDIR)/gzip_.c $(OBJDIR)/gzip.h $(SRCDIR)/config.h
1512	$(XTCC) -o $(OBJDIR)/gzip.o -c $(OBJDIR)/gzip_.c
1513
1514	$(OBJDIR)/gzip.h: $(OBJDIR)/headers
1515
1516	$(OBJDIR)/hname_.c: $(SRCDIR)/hname.c $(TRANSLATE)
1517	$(TRANSLATE) $(SRCDIR)/hname.c >$@
1518
1519	$(OBJDIR)/hname.o: $(OBJDIR)/hname_.c $(OBJDIR)/hname.h $(SRCDIR)/config.h
1520	$(XTCC) -o $(OBJDIR)/hname.o -c $(OBJDIR)/hname_.c
1521
1522	$(OBJDIR)/hname.h: $(OBJDIR)/headers
1523
1524	$(OBJDIR)/http_.c: $(SRCDIR)/http.c $(TRANSLATE)
1525	$(TRANSLATE) $(SRCDIR)/http.c >$@
1526
1527	$(OBJDIR)/http.o: $(OBJDIR)/http_.c $(OBJDIR)/http.h $(SRCDIR)/config.h
	@@ -1898,10 +1918,26 @@
1918
1919	$(OBJDIR)/sha1.o: $(OBJDIR)/sha1_.c $(OBJDIR)/sha1.h $(SRCDIR)/config.h
1920	$(XTCC) -o $(OBJDIR)/sha1.o -c $(OBJDIR)/sha1_.c
1921
1922	$(OBJDIR)/sha1.h: $(OBJDIR)/headers
1923
1924	$(OBJDIR)/sha1hard_.c: $(SRCDIR)/sha1hard.c $(TRANSLATE)
1925	$(TRANSLATE) $(SRCDIR)/sha1hard.c >$@
1926
1927	$(OBJDIR)/sha1hard.o: $(OBJDIR)/sha1hard_.c $(OBJDIR)/sha1hard.h $(SRCDIR)/config.h
1928	$(XTCC) -o $(OBJDIR)/sha1hard.o -c $(OBJDIR)/sha1hard_.c
1929
1930	$(OBJDIR)/sha1hard.h: $(OBJDIR)/headers
1931
1932	$(OBJDIR)/sha3_.c: $(SRCDIR)/sha3.c $(TRANSLATE)
1933	$(TRANSLATE) $(SRCDIR)/sha3.c >$@
1934
1935	$(OBJDIR)/sha3.o: $(OBJDIR)/sha3_.c $(OBJDIR)/sha3.h $(SRCDIR)/config.h
1936	$(XTCC) -o $(OBJDIR)/sha3.o -c $(OBJDIR)/sha3_.c
1937
1938	$(OBJDIR)/sha3.h: $(OBJDIR)/headers
1939
1940	$(OBJDIR)/shun_.c: $(SRCDIR)/shun.c $(TRANSLATE)
1941	$(TRANSLATE) $(SRCDIR)/shun.c >$@
1942
1943	$(OBJDIR)/shun.o: $(OBJDIR)/shun_.c $(OBJDIR)/shun.h $(SRCDIR)/config.h
1944

M win/Makefile.msc

+30

		--- win/Makefile.msc
		+++ win/Makefile.msc
		@@ -386,10 +386,11 @@
386	386	fshell_.c \
387	387	fusefs_.c \
388	388	glob_.c \
389	389	graph_.c \
390	390	gzip_.c \
	391	+ hname_.c \
391	392	http_.c \
392	393	http_socket_.c \
393	394	http_ssl_.c \
394	395	http_transport_.c \
395	396	import_.c \
		@@ -436,10 +437,12 @@
436	437	rss_.c \
437	438	schema_.c \
438	439	search_.c \
439	440	setup_.c \
440	441	sha1_.c \
	442	+ sha1hard_.c \
	443	+ sha3_.c \
441	444	shun_.c \
442	445	sitemap_.c \
443	446	skins_.c \
444	447	sqlcmd_.c \
445	448	stash_.c \
		@@ -560,10 +563,11 @@
560	563	$(OX)\fshell$O \
561	564	$(OX)\fusefs$O \
562	565	$(OX)\glob$O \
563	566	$(OX)\graph$O \
564	567	$(OX)\gzip$O \
	568	+ $(OX)\hname$O \
565	569	$(OX)\http$O \
566	570	$(OX)\http_socket$O \
567	571	$(OX)\http_ssl$O \
568	572	$(OX)\http_transport$O \
569	573	$(OX)\import$O \
		@@ -610,10 +614,12 @@
610	614	$(OX)\rss$O \
611	615	$(OX)\schema$O \
612	616	$(OX)\search$O \
613	617	$(OX)\setup$O \
614	618	$(OX)\sha1$O \
	619	+ $(OX)\sha1hard$O \
	620	+ $(OX)\sha3$O \
615	621	$(OX)\shell$O \
616	622	$(OX)\shun$O \
617	623	$(OX)\sitemap$O \
618	624	$(OX)\skins$O \
619	625	$(OX)\sqlcmd$O \
		@@ -743,10 +749,11 @@
743	749	echo $(OX)\fshell.obj >> $@
744	750	echo $(OX)\fusefs.obj >> $@
745	751	echo $(OX)\glob.obj >> $@
746	752	echo $(OX)\graph.obj >> $@
747	753	echo $(OX)\gzip.obj >> $@
	754	+ echo $(OX)\hname.obj >> $@
748	755	echo $(OX)\http.obj >> $@
749	756	echo $(OX)\http_socket.obj >> $@
750	757	echo $(OX)\http_ssl.obj >> $@
751	758	echo $(OX)\http_transport.obj >> $@
752	759	echo $(OX)\import.obj >> $@
		@@ -793,10 +800,12 @@
793	800	echo $(OX)\rss.obj >> $@
794	801	echo $(OX)\schema.obj >> $@
795	802	echo $(OX)\search.obj >> $@
796	803	echo $(OX)\setup.obj >> $@
797	804	echo $(OX)\sha1.obj >> $@
	805	+ echo $(OX)\sha1hard.obj >> $@
	806	+ echo $(OX)\sha3.obj >> $@
798	807	echo $(OX)\shell.obj >> $@
799	808	echo $(OX)\shun.obj >> $@
800	809	echo $(OX)\sitemap.obj >> $@
801	810	echo $(OX)\skins.obj >> $@
802	811	echo $(OX)\sqlcmd.obj >> $@
		@@ -1175,10 +1184,16 @@
1175	1184	$(OX)\gzip$O : gzip_.c gzip.h
1176	1185	$(TCC) /Fo$@ -c gzip_.c
1177	1186
1178	1187	gzip_.c : $(SRCDIR)\gzip.c
1179	1188	translate$E $** > $@
	1189	+
	1190	+$(OX)\hname$O : hname_.c hname.h
	1191	+ $(TCC) /Fo$@ -c hname_.c
	1192	+
	1193	+hname_.c : $(SRCDIR)\hname.c
	1194	+ translate$E $** > $@
1180	1195
1181	1196	$(OX)\http$O : http_.c http.h
1182	1197	$(TCC) /Fo$@ -c http_.c
1183	1198
1184	1199	http_.c : $(SRCDIR)\http.c
		@@ -1475,10 +1490,22 @@
1475	1490	$(OX)\sha1$O : sha1_.c sha1.h
1476	1491	$(TCC) /Fo$@ -c sha1_.c
1477	1492
1478	1493	sha1_.c : $(SRCDIR)\sha1.c
1479	1494	translate$E $** > $@
	1495	+
	1496	+$(OX)\sha1hard$O : sha1hard_.c sha1hard.h
	1497	+ $(TCC) /Fo$@ -c sha1hard_.c
	1498	+
	1499	+sha1hard_.c : $(SRCDIR)\sha1hard.c
	1500	+ translate$E $** > $@
	1501	+
	1502	+$(OX)\sha3$O : sha3_.c sha3.h
	1503	+ $(TCC) /Fo$@ -c sha3_.c
	1504	+
	1505	+sha3_.c : $(SRCDIR)\sha3.c
	1506	+ translate$E $** > $@
1480	1507
1481	1508	$(OX)\shun$O : shun_.c shun.h
1482	1509	$(TCC) /Fo$@ -c shun_.c
1483	1510
1484	1511	shun_.c : $(SRCDIR)\shun.c
		@@ -1717,10 +1744,11 @@
1717	1744	fshell_.c:fshell.h \
1718	1745	fusefs_.c:fusefs.h \
1719	1746	glob_.c:glob.h \
1720	1747	graph_.c:graph.h \
1721	1748	gzip_.c:gzip.h \
	1749	+ hname_.c:hname.h \
1722	1750	http_.c:http.h \
1723	1751	http_socket_.c:http_socket.h \
1724	1752	http_ssl_.c:http_ssl.h \
1725	1753	http_transport_.c:http_transport.h \
1726	1754	import_.c:import.h \
		@@ -1767,10 +1795,12 @@
1767	1795	rss_.c:rss.h \
1768	1796	schema_.c:schema.h \
1769	1797	search_.c:search.h \
1770	1798	setup_.c:setup.h \
1771	1799	sha1_.c:sha1.h \
	1800	+ sha1hard_.c:sha1hard.h \
	1801	+ sha3_.c:sha3.h \
1772	1802	shun_.c:shun.h \
1773	1803	sitemap_.c:sitemap.h \
1774	1804	skins_.c:skins.h \
1775	1805	sqlcmd_.c:sqlcmd.h \
1776	1806	stash_.c:stash.h \
1777	1807

	--- win/Makefile.msc
	+++ win/Makefile.msc
	@@ -386,10 +386,11 @@
386	fshell_.c \
387	fusefs_.c \
388	glob_.c \
389	graph_.c \
390	gzip_.c \

391	http_.c \
392	http_socket_.c \
393	http_ssl_.c \
394	http_transport_.c \
395	import_.c \
	@@ -436,10 +437,12 @@
436	rss_.c \
437	schema_.c \
438	search_.c \
439	setup_.c \
440	sha1_.c \


441	shun_.c \
442	sitemap_.c \
443	skins_.c \
444	sqlcmd_.c \
445	stash_.c \
	@@ -560,10 +563,11 @@
560	$(OX)\fshell$O \
561	$(OX)\fusefs$O \
562	$(OX)\glob$O \
563	$(OX)\graph$O \
564	$(OX)\gzip$O \

565	$(OX)\http$O \
566	$(OX)\http_socket$O \
567	$(OX)\http_ssl$O \
568	$(OX)\http_transport$O \
569	$(OX)\import$O \
	@@ -610,10 +614,12 @@
610	$(OX)\rss$O \
611	$(OX)\schema$O \
612	$(OX)\search$O \
613	$(OX)\setup$O \
614	$(OX)\sha1$O \


615	$(OX)\shell$O \
616	$(OX)\shun$O \
617	$(OX)\sitemap$O \
618	$(OX)\skins$O \
619	$(OX)\sqlcmd$O \
	@@ -743,10 +749,11 @@
743	echo $(OX)\fshell.obj >> $@
744	echo $(OX)\fusefs.obj >> $@
745	echo $(OX)\glob.obj >> $@
746	echo $(OX)\graph.obj >> $@
747	echo $(OX)\gzip.obj >> $@

748	echo $(OX)\http.obj >> $@
749	echo $(OX)\http_socket.obj >> $@
750	echo $(OX)\http_ssl.obj >> $@
751	echo $(OX)\http_transport.obj >> $@
752	echo $(OX)\import.obj >> $@
	@@ -793,10 +800,12 @@
793	echo $(OX)\rss.obj >> $@
794	echo $(OX)\schema.obj >> $@
795	echo $(OX)\search.obj >> $@
796	echo $(OX)\setup.obj >> $@
797	echo $(OX)\sha1.obj >> $@


798	echo $(OX)\shell.obj >> $@
799	echo $(OX)\shun.obj >> $@
800	echo $(OX)\sitemap.obj >> $@
801	echo $(OX)\skins.obj >> $@
802	echo $(OX)\sqlcmd.obj >> $@
	@@ -1175,10 +1184,16 @@
1175	$(OX)\gzip$O : gzip_.c gzip.h
1176	$(TCC) /Fo$@ -c gzip_.c
1177
1178	gzip_.c : $(SRCDIR)\gzip.c
1179	translate$E $** > $@






1180
1181	$(OX)\http$O : http_.c http.h
1182	$(TCC) /Fo$@ -c http_.c
1183
1184	http_.c : $(SRCDIR)\http.c
	@@ -1475,10 +1490,22 @@
1475	$(OX)\sha1$O : sha1_.c sha1.h
1476	$(TCC) /Fo$@ -c sha1_.c
1477
1478	sha1_.c : $(SRCDIR)\sha1.c
1479	translate$E $** > $@












1480
1481	$(OX)\shun$O : shun_.c shun.h
1482	$(TCC) /Fo$@ -c shun_.c
1483
1484	shun_.c : $(SRCDIR)\shun.c
	@@ -1717,10 +1744,11 @@
1717	fshell_.c:fshell.h \
1718	fusefs_.c:fusefs.h \
1719	glob_.c:glob.h \
1720	graph_.c:graph.h \
1721	gzip_.c:gzip.h \

1722	http_.c:http.h \
1723	http_socket_.c:http_socket.h \
1724	http_ssl_.c:http_ssl.h \
1725	http_transport_.c:http_transport.h \
1726	import_.c:import.h \
	@@ -1767,10 +1795,12 @@
1767	rss_.c:rss.h \
1768	schema_.c:schema.h \
1769	search_.c:search.h \
1770	setup_.c:setup.h \
1771	sha1_.c:sha1.h \


1772	shun_.c:shun.h \
1773	sitemap_.c:sitemap.h \
1774	skins_.c:skins.h \
1775	sqlcmd_.c:sqlcmd.h \
1776	stash_.c:stash.h \
1777

	--- win/Makefile.msc
	+++ win/Makefile.msc
	@@ -386,10 +386,11 @@
386	fshell_.c \
387	fusefs_.c \
388	glob_.c \
389	graph_.c \
390	gzip_.c \
391	hname_.c \
392	http_.c \
393	http_socket_.c \
394	http_ssl_.c \
395	http_transport_.c \
396	import_.c \
	@@ -436,10 +437,12 @@
437	rss_.c \
438	schema_.c \
439	search_.c \
440	setup_.c \
441	sha1_.c \
442	sha1hard_.c \
443	sha3_.c \
444	shun_.c \
445	sitemap_.c \
446	skins_.c \
447	sqlcmd_.c \
448	stash_.c \
	@@ -560,10 +563,11 @@
563	$(OX)\fshell$O \
564	$(OX)\fusefs$O \
565	$(OX)\glob$O \
566	$(OX)\graph$O \
567	$(OX)\gzip$O \
568	$(OX)\hname$O \
569	$(OX)\http$O \
570	$(OX)\http_socket$O \
571	$(OX)\http_ssl$O \
572	$(OX)\http_transport$O \
573	$(OX)\import$O \
	@@ -610,10 +614,12 @@
614	$(OX)\rss$O \
615	$(OX)\schema$O \
616	$(OX)\search$O \
617	$(OX)\setup$O \
618	$(OX)\sha1$O \
619	$(OX)\sha1hard$O \
620	$(OX)\sha3$O \
621	$(OX)\shell$O \
622	$(OX)\shun$O \
623	$(OX)\sitemap$O \
624	$(OX)\skins$O \
625	$(OX)\sqlcmd$O \
	@@ -743,10 +749,11 @@
749	echo $(OX)\fshell.obj >> $@
750	echo $(OX)\fusefs.obj >> $@
751	echo $(OX)\glob.obj >> $@
752	echo $(OX)\graph.obj >> $@
753	echo $(OX)\gzip.obj >> $@
754	echo $(OX)\hname.obj >> $@
755	echo $(OX)\http.obj >> $@
756	echo $(OX)\http_socket.obj >> $@
757	echo $(OX)\http_ssl.obj >> $@
758	echo $(OX)\http_transport.obj >> $@
759	echo $(OX)\import.obj >> $@
	@@ -793,10 +800,12 @@
800	echo $(OX)\rss.obj >> $@
801	echo $(OX)\schema.obj >> $@
802	echo $(OX)\search.obj >> $@
803	echo $(OX)\setup.obj >> $@
804	echo $(OX)\sha1.obj >> $@
805	echo $(OX)\sha1hard.obj >> $@
806	echo $(OX)\sha3.obj >> $@
807	echo $(OX)\shell.obj >> $@
808	echo $(OX)\shun.obj >> $@
809	echo $(OX)\sitemap.obj >> $@
810	echo $(OX)\skins.obj >> $@
811	echo $(OX)\sqlcmd.obj >> $@
	@@ -1175,10 +1184,16 @@
1184	$(OX)\gzip$O : gzip_.c gzip.h
1185	$(TCC) /Fo$@ -c gzip_.c
1186
1187	gzip_.c : $(SRCDIR)\gzip.c
1188	translate$E $** > $@
1189
1190	$(OX)\hname$O : hname_.c hname.h
1191	$(TCC) /Fo$@ -c hname_.c
1192
1193	hname_.c : $(SRCDIR)\hname.c
1194	translate$E $** > $@
1195
1196	$(OX)\http$O : http_.c http.h
1197	$(TCC) /Fo$@ -c http_.c
1198
1199	http_.c : $(SRCDIR)\http.c
	@@ -1475,10 +1490,22 @@
1490	$(OX)\sha1$O : sha1_.c sha1.h
1491	$(TCC) /Fo$@ -c sha1_.c
1492
1493	sha1_.c : $(SRCDIR)\sha1.c
1494	translate$E $** > $@
1495
1496	$(OX)\sha1hard$O : sha1hard_.c sha1hard.h
1497	$(TCC) /Fo$@ -c sha1hard_.c
1498
1499	sha1hard_.c : $(SRCDIR)\sha1hard.c
1500	translate$E $** > $@
1501
1502	$(OX)\sha3$O : sha3_.c sha3.h
1503	$(TCC) /Fo$@ -c sha3_.c
1504
1505	sha3_.c : $(SRCDIR)\sha3.c
1506	translate$E $** > $@
1507
1508	$(OX)\shun$O : shun_.c shun.h
1509	$(TCC) /Fo$@ -c shun_.c
1510
1511	shun_.c : $(SRCDIR)\shun.c
	@@ -1717,10 +1744,11 @@
1744	fshell_.c:fshell.h \
1745	fusefs_.c:fusefs.h \
1746	glob_.c:glob.h \
1747	graph_.c:graph.h \
1748	gzip_.c:gzip.h \
1749	hname_.c:hname.h \
1750	http_.c:http.h \
1751	http_socket_.c:http_socket.h \
1752	http_ssl_.c:http_ssl.h \
1753	http_transport_.c:http_transport.h \
1754	import_.c:import.h \
	@@ -1767,10 +1795,12 @@
1795	rss_.c:rss.h \
1796	schema_.c:schema.h \
1797	search_.c:search.h \
1798	setup_.c:setup.h \
1799	sha1_.c:sha1.h \
1800	sha1hard_.c:sha1hard.h \
1801	sha3_.c:sha3.h \
1802	shun_.c:shun.h \
1803	sitemap_.c:sitemap.h \
1804	skins_.c:skins.h \
1805	sqlcmd_.c:sqlcmd.h \
1806	stash_.c:stash.h \
1807

M win/Makefile.msc

+30

		--- win/Makefile.msc
		+++ win/Makefile.msc
		@@ -386,10 +386,11 @@
386	386	fshell_.c \
387	387	fusefs_.c \
388	388	glob_.c \
389	389	graph_.c \
390	390	gzip_.c \
	391	+ hname_.c \
391	392	http_.c \
392	393	http_socket_.c \
393	394	http_ssl_.c \
394	395	http_transport_.c \
395	396	import_.c \
		@@ -436,10 +437,12 @@
436	437	rss_.c \
437	438	schema_.c \
438	439	search_.c \
439	440	setup_.c \
440	441	sha1_.c \
	442	+ sha1hard_.c \
	443	+ sha3_.c \
441	444	shun_.c \
442	445	sitemap_.c \
443	446	skins_.c \
444	447	sqlcmd_.c \
445	448	stash_.c \
		@@ -560,10 +563,11 @@
560	563	$(OX)\fshell$O \
561	564	$(OX)\fusefs$O \
562	565	$(OX)\glob$O \
563	566	$(OX)\graph$O \
564	567	$(OX)\gzip$O \
	568	+ $(OX)\hname$O \
565	569	$(OX)\http$O \
566	570	$(OX)\http_socket$O \
567	571	$(OX)\http_ssl$O \
568	572	$(OX)\http_transport$O \
569	573	$(OX)\import$O \
		@@ -610,10 +614,12 @@
610	614	$(OX)\rss$O \
611	615	$(OX)\schema$O \
612	616	$(OX)\search$O \
613	617	$(OX)\setup$O \
614	618	$(OX)\sha1$O \
	619	+ $(OX)\sha1hard$O \
	620	+ $(OX)\sha3$O \
615	621	$(OX)\shell$O \
616	622	$(OX)\shun$O \
617	623	$(OX)\sitemap$O \
618	624	$(OX)\skins$O \
619	625	$(OX)\sqlcmd$O \
		@@ -743,10 +749,11 @@
743	749	echo $(OX)\fshell.obj >> $@
744	750	echo $(OX)\fusefs.obj >> $@
745	751	echo $(OX)\glob.obj >> $@
746	752	echo $(OX)\graph.obj >> $@
747	753	echo $(OX)\gzip.obj >> $@
	754	+ echo $(OX)\hname.obj >> $@
748	755	echo $(OX)\http.obj >> $@
749	756	echo $(OX)\http_socket.obj >> $@
750	757	echo $(OX)\http_ssl.obj >> $@
751	758	echo $(OX)\http_transport.obj >> $@
752	759	echo $(OX)\import.obj >> $@
		@@ -793,10 +800,12 @@
793	800	echo $(OX)\rss.obj >> $@
794	801	echo $(OX)\schema.obj >> $@
795	802	echo $(OX)\search.obj >> $@
796	803	echo $(OX)\setup.obj >> $@
797	804	echo $(OX)\sha1.obj >> $@
	805	+ echo $(OX)\sha1hard.obj >> $@
	806	+ echo $(OX)\sha3.obj >> $@
798	807	echo $(OX)\shell.obj >> $@
799	808	echo $(OX)\shun.obj >> $@
800	809	echo $(OX)\sitemap.obj >> $@
801	810	echo $(OX)\skins.obj >> $@
802	811	echo $(OX)\sqlcmd.obj >> $@
		@@ -1175,10 +1184,16 @@
1175	1184	$(OX)\gzip$O : gzip_.c gzip.h
1176	1185	$(TCC) /Fo$@ -c gzip_.c
1177	1186
1178	1187	gzip_.c : $(SRCDIR)\gzip.c
1179	1188	translate$E $** > $@
	1189	+
	1190	+$(OX)\hname$O : hname_.c hname.h
	1191	+ $(TCC) /Fo$@ -c hname_.c
	1192	+
	1193	+hname_.c : $(SRCDIR)\hname.c
	1194	+ translate$E $** > $@
1180	1195
1181	1196	$(OX)\http$O : http_.c http.h
1182	1197	$(TCC) /Fo$@ -c http_.c
1183	1198
1184	1199	http_.c : $(SRCDIR)\http.c
		@@ -1475,10 +1490,22 @@
1475	1490	$(OX)\sha1$O : sha1_.c sha1.h
1476	1491	$(TCC) /Fo$@ -c sha1_.c
1477	1492
1478	1493	sha1_.c : $(SRCDIR)\sha1.c
1479	1494	translate$E $** > $@
	1495	+
	1496	+$(OX)\sha1hard$O : sha1hard_.c sha1hard.h
	1497	+ $(TCC) /Fo$@ -c sha1hard_.c
	1498	+
	1499	+sha1hard_.c : $(SRCDIR)\sha1hard.c
	1500	+ translate$E $** > $@
	1501	+
	1502	+$(OX)\sha3$O : sha3_.c sha3.h
	1503	+ $(TCC) /Fo$@ -c sha3_.c
	1504	+
	1505	+sha3_.c : $(SRCDIR)\sha3.c
	1506	+ translate$E $** > $@
1480	1507
1481	1508	$(OX)\shun$O : shun_.c shun.h
1482	1509	$(TCC) /Fo$@ -c shun_.c
1483	1510
1484	1511	shun_.c : $(SRCDIR)\shun.c
		@@ -1717,10 +1744,11 @@
1717	1744	fshell_.c:fshell.h \
1718	1745	fusefs_.c:fusefs.h \
1719	1746	glob_.c:glob.h \
1720	1747	graph_.c:graph.h \
1721	1748	gzip_.c:gzip.h \
	1749	+ hname_.c:hname.h \
1722	1750	http_.c:http.h \
1723	1751	http_socket_.c:http_socket.h \
1724	1752	http_ssl_.c:http_ssl.h \
1725	1753	http_transport_.c:http_transport.h \
1726	1754	import_.c:import.h \
		@@ -1767,10 +1795,12 @@
1767	1795	rss_.c:rss.h \
1768	1796	schema_.c:schema.h \
1769	1797	search_.c:search.h \
1770	1798	setup_.c:setup.h \
1771	1799	sha1_.c:sha1.h \
	1800	+ sha1hard_.c:sha1hard.h \
	1801	+ sha3_.c:sha3.h \
1772	1802	shun_.c:shun.h \
1773	1803	sitemap_.c:sitemap.h \
1774	1804	skins_.c:skins.h \
1775	1805	sqlcmd_.c:sqlcmd.h \
1776	1806	stash_.c:stash.h \
1777	1807

	--- win/Makefile.msc
	+++ win/Makefile.msc
	@@ -386,10 +386,11 @@
386	fshell_.c \
387	fusefs_.c \
388	glob_.c \
389	graph_.c \
390	gzip_.c \

391	http_.c \
392	http_socket_.c \
393	http_ssl_.c \
394	http_transport_.c \
395	import_.c \
	@@ -436,10 +437,12 @@
436	rss_.c \
437	schema_.c \
438	search_.c \
439	setup_.c \
440	sha1_.c \


441	shun_.c \
442	sitemap_.c \
443	skins_.c \
444	sqlcmd_.c \
445	stash_.c \
	@@ -560,10 +563,11 @@
560	$(OX)\fshell$O \
561	$(OX)\fusefs$O \
562	$(OX)\glob$O \
563	$(OX)\graph$O \
564	$(OX)\gzip$O \

565	$(OX)\http$O \
566	$(OX)\http_socket$O \
567	$(OX)\http_ssl$O \
568	$(OX)\http_transport$O \
569	$(OX)\import$O \
	@@ -610,10 +614,12 @@
610	$(OX)\rss$O \
611	$(OX)\schema$O \
612	$(OX)\search$O \
613	$(OX)\setup$O \
614	$(OX)\sha1$O \


615	$(OX)\shell$O \
616	$(OX)\shun$O \
617	$(OX)\sitemap$O \
618	$(OX)\skins$O \
619	$(OX)\sqlcmd$O \
	@@ -743,10 +749,11 @@
743	echo $(OX)\fshell.obj >> $@
744	echo $(OX)\fusefs.obj >> $@
745	echo $(OX)\glob.obj >> $@
746	echo $(OX)\graph.obj >> $@
747	echo $(OX)\gzip.obj >> $@

748	echo $(OX)\http.obj >> $@
749	echo $(OX)\http_socket.obj >> $@
750	echo $(OX)\http_ssl.obj >> $@
751	echo $(OX)\http_transport.obj >> $@
752	echo $(OX)\import.obj >> $@
	@@ -793,10 +800,12 @@
793	echo $(OX)\rss.obj >> $@
794	echo $(OX)\schema.obj >> $@
795	echo $(OX)\search.obj >> $@
796	echo $(OX)\setup.obj >> $@
797	echo $(OX)\sha1.obj >> $@


798	echo $(OX)\shell.obj >> $@
799	echo $(OX)\shun.obj >> $@
800	echo $(OX)\sitemap.obj >> $@
801	echo $(OX)\skins.obj >> $@
802	echo $(OX)\sqlcmd.obj >> $@
	@@ -1175,10 +1184,16 @@
1175	$(OX)\gzip$O : gzip_.c gzip.h
1176	$(TCC) /Fo$@ -c gzip_.c
1177
1178	gzip_.c : $(SRCDIR)\gzip.c
1179	translate$E $** > $@






1180
1181	$(OX)\http$O : http_.c http.h
1182	$(TCC) /Fo$@ -c http_.c
1183
1184	http_.c : $(SRCDIR)\http.c
	@@ -1475,10 +1490,22 @@
1475	$(OX)\sha1$O : sha1_.c sha1.h
1476	$(TCC) /Fo$@ -c sha1_.c
1477
1478	sha1_.c : $(SRCDIR)\sha1.c
1479	translate$E $** > $@












1480
1481	$(OX)\shun$O : shun_.c shun.h
1482	$(TCC) /Fo$@ -c shun_.c
1483
1484	shun_.c : $(SRCDIR)\shun.c
	@@ -1717,10 +1744,11 @@
1717	fshell_.c:fshell.h \
1718	fusefs_.c:fusefs.h \
1719	glob_.c:glob.h \
1720	graph_.c:graph.h \
1721	gzip_.c:gzip.h \

1722	http_.c:http.h \
1723	http_socket_.c:http_socket.h \
1724	http_ssl_.c:http_ssl.h \
1725	http_transport_.c:http_transport.h \
1726	import_.c:import.h \
	@@ -1767,10 +1795,12 @@
1767	rss_.c:rss.h \
1768	schema_.c:schema.h \
1769	search_.c:search.h \
1770	setup_.c:setup.h \
1771	sha1_.c:sha1.h \


1772	shun_.c:shun.h \
1773	sitemap_.c:sitemap.h \
1774	skins_.c:skins.h \
1775	sqlcmd_.c:sqlcmd.h \
1776	stash_.c:stash.h \
1777

	--- win/Makefile.msc
	+++ win/Makefile.msc
	@@ -386,10 +386,11 @@
386	fshell_.c \
387	fusefs_.c \
388	glob_.c \
389	graph_.c \
390	gzip_.c \
391	hname_.c \
392	http_.c \
393	http_socket_.c \
394	http_ssl_.c \
395	http_transport_.c \
396	import_.c \
	@@ -436,10 +437,12 @@
437	rss_.c \
438	schema_.c \
439	search_.c \
440	setup_.c \
441	sha1_.c \
442	sha1hard_.c \
443	sha3_.c \
444	shun_.c \
445	sitemap_.c \
446	skins_.c \
447	sqlcmd_.c \
448	stash_.c \
	@@ -560,10 +563,11 @@
563	$(OX)\fshell$O \
564	$(OX)\fusefs$O \
565	$(OX)\glob$O \
566	$(OX)\graph$O \
567	$(OX)\gzip$O \
568	$(OX)\hname$O \
569	$(OX)\http$O \
570	$(OX)\http_socket$O \
571	$(OX)\http_ssl$O \
572	$(OX)\http_transport$O \
573	$(OX)\import$O \
	@@ -610,10 +614,12 @@
614	$(OX)\rss$O \
615	$(OX)\schema$O \
616	$(OX)\search$O \
617	$(OX)\setup$O \
618	$(OX)\sha1$O \
619	$(OX)\sha1hard$O \
620	$(OX)\sha3$O \
621	$(OX)\shell$O \
622	$(OX)\shun$O \
623	$(OX)\sitemap$O \
624	$(OX)\skins$O \
625	$(OX)\sqlcmd$O \
	@@ -743,10 +749,11 @@
749	echo $(OX)\fshell.obj >> $@
750	echo $(OX)\fusefs.obj >> $@
751	echo $(OX)\glob.obj >> $@
752	echo $(OX)\graph.obj >> $@
753	echo $(OX)\gzip.obj >> $@
754	echo $(OX)\hname.obj >> $@
755	echo $(OX)\http.obj >> $@
756	echo $(OX)\http_socket.obj >> $@
757	echo $(OX)\http_ssl.obj >> $@
758	echo $(OX)\http_transport.obj >> $@
759	echo $(OX)\import.obj >> $@
	@@ -793,10 +800,12 @@
800	echo $(OX)\rss.obj >> $@
801	echo $(OX)\schema.obj >> $@
802	echo $(OX)\search.obj >> $@
803	echo $(OX)\setup.obj >> $@
804	echo $(OX)\sha1.obj >> $@
805	echo $(OX)\sha1hard.obj >> $@
806	echo $(OX)\sha3.obj >> $@
807	echo $(OX)\shell.obj >> $@
808	echo $(OX)\shun.obj >> $@
809	echo $(OX)\sitemap.obj >> $@
810	echo $(OX)\skins.obj >> $@
811	echo $(OX)\sqlcmd.obj >> $@
	@@ -1175,10 +1184,16 @@
1184	$(OX)\gzip$O : gzip_.c gzip.h
1185	$(TCC) /Fo$@ -c gzip_.c
1186
1187	gzip_.c : $(SRCDIR)\gzip.c
1188	translate$E $** > $@
1189
1190	$(OX)\hname$O : hname_.c hname.h
1191	$(TCC) /Fo$@ -c hname_.c
1192
1193	hname_.c : $(SRCDIR)\hname.c
1194	translate$E $** > $@
1195
1196	$(OX)\http$O : http_.c http.h
1197	$(TCC) /Fo$@ -c http_.c
1198
1199	http_.c : $(SRCDIR)\http.c
	@@ -1475,10 +1490,22 @@
1490	$(OX)\sha1$O : sha1_.c sha1.h
1491	$(TCC) /Fo$@ -c sha1_.c
1492
1493	sha1_.c : $(SRCDIR)\sha1.c
1494	translate$E $** > $@
1495
1496	$(OX)\sha1hard$O : sha1hard_.c sha1hard.h
1497	$(TCC) /Fo$@ -c sha1hard_.c
1498
1499	sha1hard_.c : $(SRCDIR)\sha1hard.c
1500	translate$E $** > $@
1501
1502	$(OX)\sha3$O : sha3_.c sha3.h
1503	$(TCC) /Fo$@ -c sha3_.c
1504
1505	sha3_.c : $(SRCDIR)\sha3.c
1506	translate$E $** > $@
1507
1508	$(OX)\shun$O : shun_.c shun.h
1509	$(TCC) /Fo$@ -c shun_.c
1510
1511	shun_.c : $(SRCDIR)\shun.c
	@@ -1717,10 +1744,11 @@
1744	fshell_.c:fshell.h \
1745	fusefs_.c:fusefs.h \
1746	glob_.c:glob.h \
1747	graph_.c:graph.h \
1748	gzip_.c:gzip.h \
1749	hname_.c:hname.h \
1750	http_.c:http.h \
1751	http_socket_.c:http_socket.h \
1752	http_ssl_.c:http_ssl.h \
1753	http_transport_.c:http_transport.h \
1754	import_.c:import.h \
	@@ -1767,10 +1795,12 @@
1795	rss_.c:rss.h \
1796	schema_.c:schema.h \
1797	search_.c:search.h \
1798	setup_.c:setup.h \
1799	sha1_.c:sha1.h \
1800	sha1hard_.c:sha1hard.h \
1801	sha3_.c:sha3.h \
1802	shun_.c:shun.h \
1803	sitemap_.c:sitemap.h \
1804	skins_.c:skins.h \
1805	sqlcmd_.c:sqlcmd.h \
1806	stash_.c:stash.h \
1807

M www/branching.wiki

+3 -3

		--- www/branching.wiki
		+++ www/branching.wiki
		@@ -10,14 +10,14 @@
10	10	Figure 1
11	11	</td></tr></table>
12	12
13	13	Each circle represents a check-in. For the sake of clarity, the check-ins
14	14	are given small consecutive numbers. In a real system, of course, the
15		-check-in numbers would be 40-character SHA1 hashes since it is not possible
	15	+check-in numbers would be long hexadecimal hashes since it is not possible
16	16	to allocate collision-free sequential numbers in a distributed system.
17	17	But as sequential numbers are easier to read, we will substitute them for
18		-the 40-character SHA1 hashes in this document.
	18	+the long hashes in this document.
19	19
20	20	The arrows in figure 1 show the evolution of a project. The initial
21	21	check-in is 1. Check-in 2 is derived from 1. In other words, check-in 2
22	22	was created by making edits to check-in 1 and then committing those edits.
23	23	We say that 2 is a <i>child</i> of 1
		@@ -193,11 +193,11 @@
193	193	figure 5, that initial check-in is check-in 1. The <b>branch</b> tag
194	194	tells (by its value) what branch the check-in is a member of.
195	195	The default branch is called "trunk." All tags that begin with "<b>sym-</b>"
196	196	are symbolic name tags. When a symbolic name tag is attached to a
197	197	check-in, that allows you to refer to that check-in by its symbolic
198		-name rather than by its 40-character SHA1 hash name. When a symbolic name
	198	+name rather than by its hexadecimal hash name. When a symbolic name
199	199	tag propagates (as does the <b>sym-trunk</b> tag) then referring to that
200	200	name is the same as referring to the most recent check-in with that name.
201	201	Thus the two tags on check-in 1 cause all descendants to be in the
202	202	"trunk" branch and to have the symbolic name "trunk."
203	203
204	204

	--- www/branching.wiki
	+++ www/branching.wiki
	@@ -10,14 +10,14 @@
10	Figure 1
11	</td></tr></table>
12
13	Each circle represents a check-in. For the sake of clarity, the check-ins
14	are given small consecutive numbers. In a real system, of course, the
15	check-in numbers would be 40-character SHA1 hashes since it is not possible
16	to allocate collision-free sequential numbers in a distributed system.
17	But as sequential numbers are easier to read, we will substitute them for
18	the 40-character SHA1 hashes in this document.
19
20	The arrows in figure 1 show the evolution of a project. The initial
21	check-in is 1. Check-in 2 is derived from 1. In other words, check-in 2
22	was created by making edits to check-in 1 and then committing those edits.
23	We say that 2 is a <i>child</i> of 1
	@@ -193,11 +193,11 @@
193	figure 5, that initial check-in is check-in 1. The <b>branch</b> tag
194	tells (by its value) what branch the check-in is a member of.
195	The default branch is called "trunk." All tags that begin with "<b>sym-</b>"
196	are symbolic name tags. When a symbolic name tag is attached to a
197	check-in, that allows you to refer to that check-in by its symbolic
198	name rather than by its 40-character SHA1 hash name. When a symbolic name
199	tag propagates (as does the <b>sym-trunk</b> tag) then referring to that
200	name is the same as referring to the most recent check-in with that name.
201	Thus the two tags on check-in 1 cause all descendants to be in the
202	"trunk" branch and to have the symbolic name "trunk."
203
204

	--- www/branching.wiki
	+++ www/branching.wiki
	@@ -10,14 +10,14 @@
10	Figure 1
11	</td></tr></table>
12
13	Each circle represents a check-in. For the sake of clarity, the check-ins
14	are given small consecutive numbers. In a real system, of course, the
15	check-in numbers would be long hexadecimal hashes since it is not possible
16	to allocate collision-free sequential numbers in a distributed system.
17	But as sequential numbers are easier to read, we will substitute them for
18	the long hashes in this document.
19
20	The arrows in figure 1 show the evolution of a project. The initial
21	check-in is 1. Check-in 2 is derived from 1. In other words, check-in 2
22	was created by making edits to check-in 1 and then committing those edits.
23	We say that 2 is a <i>child</i> of 1
	@@ -193,11 +193,11 @@
193	figure 5, that initial check-in is check-in 1. The <b>branch</b> tag
194	tells (by its value) what branch the check-in is a member of.
195	The default branch is called "trunk." All tags that begin with "<b>sym-</b>"
196	are symbolic name tags. When a symbolic name tag is attached to a
197	check-in, that allows you to refer to that check-in by its symbolic
198	name rather than by its hexadecimal hash name. When a symbolic name
199	tag propagates (as does the <b>sym-trunk</b> tag) then referring to that
200	name is the same as referring to the most recent check-in with that name.
201	Thus the two tags on check-in 1 cause all descendants to be in the
202	"trunk" branch and to have the symbolic name "trunk."
203
204

M www/build.wiki

+7 -12

		--- www/build.wiki
		+++ www/build.wiki
		@@ -1,16 +1,14 @@
1	1	<title>Compiling and Installing Fossil</title>
2	2
3	3	<h2>0.0 Using A Pre-compiled Binary</h2>
4	4
5		-<p>Released versions of fossil come with
6		-<a href="/uv/download.html">pre-compiled binaries and
7		-a source archive</a> for that release. You can thus skip the following if you
8		-want to run or build a release version of fossil. Just download
9		-the appropriate package from the <a href="/uv/download.html">downloads page</a>
	5	+<p>[/uv/download.html\|Pre-compiled binaries] are available for recent
	6	+releases. Just download
	7	+the appropriate executable for your platform
10	8	and put it on your $PATH.
11		-To uninstall, simply delete the binary.
	9	+To uninstall, simply delete the executable.
12	10	To upgrade from an older release, just overwrite the older binary with
13	11	the newer one.</p>
14	12
15	13	<h2>0.1 Executive Summary</h2>
16	14
		@@ -29,20 +27,17 @@
29	27
30	28	<p>Fossil is self-hosting, so you can obtain a ZIP archive or tarball
31	29	containing a snapshot of the <em>latest</em> version directly from
32	30	Fossil's own fossil repository. Additionally, source archives of
33	31	<em>released</em> versions of
34		-fossil are available from the <a href="/uv/download.html">downloads page</a>.
	32	+fossil are available from the [/uv/download.html\|downloads page].
35	33	To obtain a development version of fossil, follow these steps:</p>
36	34
37	35	<ol>
38		-<li><p>Point your web browser to
39		-<a href="http://www.fossil-scm.org/">
40		-http://www.fossil-scm.org/</a>.</p></li>
	36	+<li><p>Point your web browser to [https://www.fossil-scm.org/]</li>
41	37
42		-<li><p>Click on the
43		-<a href="http://www.fossil-scm.org/fossil/timeline">Timeline</a>
	38	+<li><p>Click on the [/timeline\|Timeline]
44	39	link at the top of the page.</p></li>
45	40
46	41	<li><p>Select a version of of Fossil you want to download. The latest
47	42	version on the trunk branch is usually a good choice. Click on its
48	43	link.</p></li>
49	44

	--- www/build.wiki
	+++ www/build.wiki
	@@ -1,16 +1,14 @@
1	<title>Compiling and Installing Fossil</title>
2
3	<h2>0.0 Using A Pre-compiled Binary</h2>
4
5	<p>Released versions of fossil come with
6	<a href="/uv/download.html">pre-compiled binaries and
7	a source archive</a> for that release. You can thus skip the following if you
8	want to run or build a release version of fossil. Just download
9	the appropriate package from the <a href="/uv/download.html">downloads page</a>
10	and put it on your $PATH.
11	To uninstall, simply delete the binary.
12	To upgrade from an older release, just overwrite the older binary with
13	the newer one.</p>
14
15	<h2>0.1 Executive Summary</h2>
16
	@@ -29,20 +27,17 @@
29
30	<p>Fossil is self-hosting, so you can obtain a ZIP archive or tarball
31	containing a snapshot of the <em>latest</em> version directly from
32	Fossil's own fossil repository. Additionally, source archives of
33	<em>released</em> versions of
34	fossil are available from the <a href="/uv/download.html">downloads page</a>.
35	To obtain a development version of fossil, follow these steps:</p>
36
37	<ol>
38	<li><p>Point your web browser to
39	<a href="http://www.fossil-scm.org/">
40	http://www.fossil-scm.org/</a>.</p></li>
41
42	<li><p>Click on the
43	<a href="http://www.fossil-scm.org/fossil/timeline">Timeline</a>
44	link at the top of the page.</p></li>
45
46	<li><p>Select a version of of Fossil you want to download. The latest
47	version on the trunk branch is usually a good choice. Click on its
48	link.</p></li>
49

	--- www/build.wiki
	+++ www/build.wiki
	@@ -1,16 +1,14 @@
1	<title>Compiling and Installing Fossil</title>
2
3	<h2>0.0 Using A Pre-compiled Binary</h2>
4
5	<p>[/uv/download.html\|Pre-compiled binaries] are available for recent
6	releases. Just download
7	the appropriate executable for your platform


8	and put it on your $PATH.
9	To uninstall, simply delete the executable.
10	To upgrade from an older release, just overwrite the older binary with
11	the newer one.</p>
12
13	<h2>0.1 Executive Summary</h2>
14
	@@ -29,20 +27,17 @@
27
28	<p>Fossil is self-hosting, so you can obtain a ZIP archive or tarball
29	containing a snapshot of the <em>latest</em> version directly from
30	Fossil's own fossil repository. Additionally, source archives of
31	<em>released</em> versions of
32	fossil are available from the [/uv/download.html\|downloads page].
33	To obtain a development version of fossil, follow these steps:</p>
34
35	<ol>
36	<li><p>Point your web browser to [https://www.fossil-scm.org/]</li>


37
38	<li><p>Click on the [/timeline\|Timeline]

39	link at the top of the page.</p></li>
40
41	<li><p>Select a version of of Fossil you want to download. The latest
42	version on the trunk branch is usually a good choice. Click on its
43	link.</p></li>
44

M www/build.wiki

+7 -12

		--- www/build.wiki
		+++ www/build.wiki
		@@ -1,16 +1,14 @@
1	1	<title>Compiling and Installing Fossil</title>
2	2
3	3	<h2>0.0 Using A Pre-compiled Binary</h2>
4	4
5		-<p>Released versions of fossil come with
6		-<a href="/uv/download.html">pre-compiled binaries and
7		-a source archive</a> for that release. You can thus skip the following if you
8		-want to run or build a release version of fossil. Just download
9		-the appropriate package from the <a href="/uv/download.html">downloads page</a>
	5	+<p>[/uv/download.html\|Pre-compiled binaries] are available for recent
	6	+releases. Just download
	7	+the appropriate executable for your platform
10	8	and put it on your $PATH.
11		-To uninstall, simply delete the binary.
	9	+To uninstall, simply delete the executable.
12	10	To upgrade from an older release, just overwrite the older binary with
13	11	the newer one.</p>
14	12
15	13	<h2>0.1 Executive Summary</h2>
16	14
		@@ -29,20 +27,17 @@
29	27
30	28	<p>Fossil is self-hosting, so you can obtain a ZIP archive or tarball
31	29	containing a snapshot of the <em>latest</em> version directly from
32	30	Fossil's own fossil repository. Additionally, source archives of
33	31	<em>released</em> versions of
34		-fossil are available from the <a href="/uv/download.html">downloads page</a>.
	32	+fossil are available from the [/uv/download.html\|downloads page].
35	33	To obtain a development version of fossil, follow these steps:</p>
36	34
37	35	<ol>
38		-<li><p>Point your web browser to
39		-<a href="http://www.fossil-scm.org/">
40		-http://www.fossil-scm.org/</a>.</p></li>
	36	+<li><p>Point your web browser to [https://www.fossil-scm.org/]</li>
41	37
42		-<li><p>Click on the
43		-<a href="http://www.fossil-scm.org/fossil/timeline">Timeline</a>
	38	+<li><p>Click on the [/timeline\|Timeline]
44	39	link at the top of the page.</p></li>
45	40
46	41	<li><p>Select a version of of Fossil you want to download. The latest
47	42	version on the trunk branch is usually a good choice. Click on its
48	43	link.</p></li>
49	44

	--- www/build.wiki
	+++ www/build.wiki
	@@ -1,16 +1,14 @@
1	<title>Compiling and Installing Fossil</title>
2
3	<h2>0.0 Using A Pre-compiled Binary</h2>
4
5	<p>Released versions of fossil come with
6	<a href="/uv/download.html">pre-compiled binaries and
7	a source archive</a> for that release. You can thus skip the following if you
8	want to run or build a release version of fossil. Just download
9	the appropriate package from the <a href="/uv/download.html">downloads page</a>
10	and put it on your $PATH.
11	To uninstall, simply delete the binary.
12	To upgrade from an older release, just overwrite the older binary with
13	the newer one.</p>
14
15	<h2>0.1 Executive Summary</h2>
16
	@@ -29,20 +27,17 @@
29
30	<p>Fossil is self-hosting, so you can obtain a ZIP archive or tarball
31	containing a snapshot of the <em>latest</em> version directly from
32	Fossil's own fossil repository. Additionally, source archives of
33	<em>released</em> versions of
34	fossil are available from the <a href="/uv/download.html">downloads page</a>.
35	To obtain a development version of fossil, follow these steps:</p>
36
37	<ol>
38	<li><p>Point your web browser to
39	<a href="http://www.fossil-scm.org/">
40	http://www.fossil-scm.org/</a>.</p></li>
41
42	<li><p>Click on the
43	<a href="http://www.fossil-scm.org/fossil/timeline">Timeline</a>
44	link at the top of the page.</p></li>
45
46	<li><p>Select a version of of Fossil you want to download. The latest
47	version on the trunk branch is usually a good choice. Click on its
48	link.</p></li>
49

	--- www/build.wiki
	+++ www/build.wiki
	@@ -1,16 +1,14 @@
1	<title>Compiling and Installing Fossil</title>
2
3	<h2>0.0 Using A Pre-compiled Binary</h2>
4
5	<p>[/uv/download.html\|Pre-compiled binaries] are available for recent
6	releases. Just download
7	the appropriate executable for your platform


8	and put it on your $PATH.
9	To uninstall, simply delete the executable.
10	To upgrade from an older release, just overwrite the older binary with
11	the newer one.</p>
12
13	<h2>0.1 Executive Summary</h2>
14
	@@ -29,20 +27,17 @@
27
28	<p>Fossil is self-hosting, so you can obtain a ZIP archive or tarball
29	containing a snapshot of the <em>latest</em> version directly from
30	Fossil's own fossil repository. Additionally, source archives of
31	<em>released</em> versions of
32	fossil are available from the [/uv/download.html\|downloads page].
33	To obtain a development version of fossil, follow these steps:</p>
34
35	<ol>
36	<li><p>Point your web browser to [https://www.fossil-scm.org/]</li>


37
38	<li><p>Click on the [/timeline\|Timeline]

39	link at the top of the page.</p></li>
40
41	<li><p>Select a version of of Fossil you want to download. The latest
42	version on the trunk branch is usually a good choice. Click on its
43	link.</p></li>
44

M www/changes.wiki

+20 -2

		--- www/changes.wiki
		+++ www/changes.wiki
		@@ -1,10 +1,28 @@
1	1	<title>Change Log</title>
2	2
3		-<a name='v1_38'></a>
4		-<h2>Changes for Version 1.38 (2017-??-??)</h2>
	3	+<a name='v2_1'></a>
	4	+<h2>Changes for Version 2.1 (2017-03-10)</h2>
	5	+
	6	+ * Add support for [./hashpolicy.wiki\|hash policies] that control which
	7	+ of the Hardened-SHA1 or SHA3-256 algorithms is used to name new
	8	+ artifacts.
	9	+ * Add the "gshow" and "gcat" subcommands to [/help?cmd=stash\|fossil stash].
	10	+ * Add the [/help?cmd=/juvlist\|/juvlist] web page and use it to construct
	11	+ the [/uv/download.html\|Download Page] of the Fossil self-hosting website
	12	+ using Ajax.
	13	+
	14	+<a name='v2_0'></a>
	15	+<h2>Changes for Version 2.0 (2017-03-03)</h2>
5	16
	17	+ * Use the
	18	+ [https://github.com/cr-marcstevens/sha1collisiondetection\|hardened SHA1]
	19	+ implemenation by Marc Stevens and Dan Shumow.
	20	+ * Add the ability to read and understand
	21	+ [./fileformat.wiki#names\|artifact names] that are based on SHA3-256
	22	+ rather than SHA1, but do not actually generate any such names.
	23	+ * Added the [/help?cmd=sha3sum\|sha3sum] command.
6	24	* Update the built-in SQLite to version 3.17.0.
7	25
8	26	<a name='v1_37'></a>
9	27	<h2>Changes for Version 1.37 (2017-01-16)</h2>
10	28
11	29

	--- www/changes.wiki
	+++ www/changes.wiki
	@@ -1,10 +1,28 @@
1	<title>Change Log</title>
2
3	<a name='v1_38'></a>
4	<h2>Changes for Version 1.38 (2017-??-??)</h2>











5







6	* Update the built-in SQLite to version 3.17.0.
7
8	<a name='v1_37'></a>
9	<h2>Changes for Version 1.37 (2017-01-16)</h2>
10
11

	--- www/changes.wiki
	+++ www/changes.wiki
	@@ -1,10 +1,28 @@
1	<title>Change Log</title>
2
3	<a name='v2_1'></a>
4	<h2>Changes for Version 2.1 (2017-03-10)</h2>
5
6	* Add support for [./hashpolicy.wiki\|hash policies] that control which
7	of the Hardened-SHA1 or SHA3-256 algorithms is used to name new
8	artifacts.
9	* Add the "gshow" and "gcat" subcommands to [/help?cmd=stash\|fossil stash].
10	* Add the [/help?cmd=/juvlist\|/juvlist] web page and use it to construct
11	the [/uv/download.html\|Download Page] of the Fossil self-hosting website
12	using Ajax.
13
14	<a name='v2_0'></a>
15	<h2>Changes for Version 2.0 (2017-03-03)</h2>
16
17	* Use the
18	[https://github.com/cr-marcstevens/sha1collisiondetection\|hardened SHA1]
19	implemenation by Marc Stevens and Dan Shumow.
20	* Add the ability to read and understand
21	[./fileformat.wiki#names\|artifact names] that are based on SHA3-256
22	rather than SHA1, but do not actually generate any such names.
23	* Added the [/help?cmd=sha3sum\|sha3sum] command.
24	* Update the built-in SQLite to version 3.17.0.
25
26	<a name='v1_37'></a>
27	<h2>Changes for Version 1.37 (2017-01-16)</h2>
28
29

M www/checkin_names.wiki

+4 -4

		--- www/checkin_names.wiki
		+++ www/checkin_names.wiki
		@@ -4,11 +4,11 @@
4	4	<tr><td>
5	5	<h3>Executive Summary</h3>
6	6	<p>A check-in can be identified using any of the following
7	7	names:
8	8	<ul>
9		-<li> SHA1 hash prefix
	9	+<li> Cryptographic hash prefix
10	10	<li> Tag or branchname
11	11	<li> Timestamp: <i>YYYY-MM-DD HH:MM:SS</i>
12	12	<li> <i>tag-name</i> <big><b>:</b></big> <i>timestamp</i>
13	13	<li> <b>root :</b> <i>branchname</i>
14	14	<li> Special names:
		@@ -44,19 +44,19 @@
44	44	Fossil provides a variety of ways to specify a check-in. This
45	45	document describes the various methods.
46	46
47	47	<h2>Canonical Check-in Name</h2>
48	48
49		-The canonical name of a check-in is the SHA1 hash of its
50		-[./fileformat.wiki#manifest \| manifest] expressed as a 40-character
	49	+The canonical name of a check-in is the hash of its
	50	+[./fileformat.wiki#manifest \| manifest] expressed as a 40-or-more character
51	51	lowercase hexadecimal number. For example:
52	52
53	53	<blockquote><pre>
54	54	fossil info e5a734a19a9826973e1d073b49dc2a16aa2308f9
55	55	</pre></blockquote>
56	56
57		-The full 40-character SHA1 hash is unwieldy to remember and type, though,
	57	+The full 40+ character hash is unwieldy to remember and type, though,
58	58	so Fossil also accepts a unique prefix of the hash, using any combination
59	59	of upper and lower case letters, as long as the prefix is at least 4
60	60	characters long. Hence the following commands all
61	61	accomplish the same thing as the above:
62	62
63	63

	--- www/checkin_names.wiki
	+++ www/checkin_names.wiki
	@@ -4,11 +4,11 @@
4	<tr><td>
5	<h3>Executive Summary</h3>
6	<p>A check-in can be identified using any of the following
7	names:
8	<ul>
9	<li> SHA1 hash prefix
10	<li> Tag or branchname
11	<li> Timestamp: <i>YYYY-MM-DD HH:MM:SS</i>
12	<li> <i>tag-name</i> <big><b>:</b></big> <i>timestamp</i>
13	<li> <b>root :</b> <i>branchname</i>
14	<li> Special names:
	@@ -44,19 +44,19 @@
44	Fossil provides a variety of ways to specify a check-in. This
45	document describes the various methods.
46
47	<h2>Canonical Check-in Name</h2>
48
49	The canonical name of a check-in is the SHA1 hash of its
50	[./fileformat.wiki#manifest \| manifest] expressed as a 40-character
51	lowercase hexadecimal number. For example:
52
53	<blockquote><pre>
54	fossil info e5a734a19a9826973e1d073b49dc2a16aa2308f9
55	</pre></blockquote>
56
57	The full 40-character SHA1 hash is unwieldy to remember and type, though,
58	so Fossil also accepts a unique prefix of the hash, using any combination
59	of upper and lower case letters, as long as the prefix is at least 4
60	characters long. Hence the following commands all
61	accomplish the same thing as the above:
62
63

	--- www/checkin_names.wiki
	+++ www/checkin_names.wiki
	@@ -4,11 +4,11 @@
4	<tr><td>
5	<h3>Executive Summary</h3>
6	<p>A check-in can be identified using any of the following
7	names:
8	<ul>
9	<li> Cryptographic hash prefix
10	<li> Tag or branchname
11	<li> Timestamp: <i>YYYY-MM-DD HH:MM:SS</i>
12	<li> <i>tag-name</i> <big><b>:</b></big> <i>timestamp</i>
13	<li> <b>root :</b> <i>branchname</i>
14	<li> Special names:
	@@ -44,19 +44,19 @@
44	Fossil provides a variety of ways to specify a check-in. This
45	document describes the various methods.
46
47	<h2>Canonical Check-in Name</h2>
48
49	The canonical name of a check-in is the hash of its
50	[./fileformat.wiki#manifest \| manifest] expressed as a 40-or-more character
51	lowercase hexadecimal number. For example:
52
53	<blockquote><pre>
54	fossil info e5a734a19a9826973e1d073b49dc2a16aa2308f9
55	</pre></blockquote>
56
57	The full 40+ character hash is unwieldy to remember and type, though,
58	so Fossil also accepts a unique prefix of the hash, using any combination
59	of upper and lower case letters, as long as the prefix is at least 4
60	characters long. Hence the following commands all
61	accomplish the same thing as the above:
62
63

M www/customskin.md

+1 -1

		--- www/customskin.md
		+++ www/customskin.md
		@@ -177,11 +177,11 @@
177	177
178	178	* csrf_token - A token used to prevent cross-site request forgery.
179	179
180	180	* release_version - The release version of Fossil. Ex: "1.31"
181	181
182		- * manifest_version - A prefix on the SHA1 check-in hash of the
	182	+ * manifest_version - A prefix on the check-in hash of the
183	183	specific version of fossil that is running. Ex: "\[47bb6432a1\]"
184	184
185	185	* manifest_date - The date of the source-code check-in for the
186	186	version of fossil that is running.
187	187
188	188

	--- www/customskin.md
	+++ www/customskin.md
	@@ -177,11 +177,11 @@
177
178	* csrf_token - A token used to prevent cross-site request forgery.
179
180	* release_version - The release version of Fossil. Ex: "1.31"
181
182	* manifest_version - A prefix on the SHA1 check-in hash of the
183	specific version of fossil that is running. Ex: "\[47bb6432a1\]"
184
185	* manifest_date - The date of the source-code check-in for the
186	version of fossil that is running.
187
188

	--- www/customskin.md
	+++ www/customskin.md
	@@ -177,11 +177,11 @@
177
178	* csrf_token - A token used to prevent cross-site request forgery.
179
180	* release_version - The release version of Fossil. Ex: "1.31"
181
182	* manifest_version - A prefix on the check-in hash of the
183	specific version of fossil that is running. Ex: "\[47bb6432a1\]"
184
185	* manifest_date - The date of the source-code check-in for the
186	version of fossil that is running.
187
188

M www/fileformat.wiki

+79 -84

		--- www/fileformat.wiki
		+++ www/fileformat.wiki
		@@ -9,14 +9,19 @@
9	9	searchable, and extensible by people not yet born.
10	10
11	11	The global state of a fossil repository is an unordered
12	12	set of <i>artifacts</i>.
13	13	An artifact might be a source code file, the text of a wiki page,
14		-part of a trouble ticket, or one of several special control artifacts
15		-used to show the relationships between other artifacts within the
16		-project. Each artifact is normally represented on disk as a separate
17		-file. Artifacts can be text or binary.
	14	+part of a trouble ticket, a description of a check-in including all
	15	+the files in that check-in with the check-in comment and so forth.
	16	+Artifacts are broadly grouped into two types: content artifacts and
	17	+structural artifacts. Content artifacts are the raw project source-code
	18	+files that are checked into the repository. Structural artifacts have
	19	+special formatting rules and are used to show the relationships between
	20	+other artifacts in the repository. It is possible for an artifact to
	21	+be both a structure artifact and a content artifact, though this is
	22	+rare. Artifacts can be text or binary.
18	23
19	24	In addition to the global state,
20	25	each fossil repository also contains local state.
21	26	The local state consists of web-page formatting
22	27	preferences, authorized users, ticket display and reporting formats,
		@@ -27,18 +32,40 @@
27	32	with the global state.
28	33	The local state is not composed of artifacts and is not intended to be enduring.
29	34	This document is concerned with global state only. Local state is only
30	35	mentioned here in order to distinguish it from global state.
31	36
32		-Each artifact in the repository is named by its SHA1 hash.
33		-No prefixes or meta information is added to an artifact before
34		-its hash is computed. The name of an artifact in the repository
35		-is exactly the same SHA1 hash that is computed by sha1sum
36		-on the file as it exists in your source tree.</p>
37		-
38		-Some artifacts have a particular format which gives them special
39		-meaning to fossil. Fossil recognizes:
	37	+<a name="names"></a>
	38	+<h2>1.0 Artifact Names</h2>
	39	+
	40	+Each artifact in the repository is named by a hash of its content.
	41	+No prefixes, suffixes, or other information is added to an artifact before
	42	+the hash is computed. The artifact name is just the (lower-case
	43	+hexadecimal) hash of the raw artifact.
	44	+
	45	+Fossil currently computes artifact names using either SHA1 or SHA3-256. It
	46	+is relatively easy to add new algorithms in the future, but there are no
	47	+plans to do so at this time.
	48	+
	49	+When referring to artifacts in using tty commands or webpage URLs, it is
	50	+sufficient to specify a unique prefix for the artifact name. If the input
	51	+prefix is not unique, Fossil will show an error. Within a structural
	52	+artifact, however, all references to other artifacts must be the complete
	53	+hash.
	54	+
	55	+Prior to Fossil version 2.0, all names were formed from the SHA1 hash of
	56	+the artifact. The key innovation in Fossil 2.0 was adding support for
	57	+alternative hash algorithms.
	58	+
	59	+<a name="structural"></a>
	60	+<h2>2.0 Structural Artifacts</h2>
	61	+
	62	+A structural artifact is an artifact with a particular format
	63	+that is used to define the relationships between other artifacts in the
	64	+repository.
	65	+Fossil recognizes the following kinds of structural
	66	+artifacts:
40	67
41	68	<ul>
42	69	<li> [#manifest \| Manifests] </li>
43	70	<li> [#cluster \| Clusters] </li>
44	71	<li> [#ctrl \| Control Artifacts] </li>
		@@ -46,66 +73,55 @@
46	73	<li> [#tktchng \| Ticket Changes] </li>
47	74	<li> [#attachment \| Attachments] </li>
48	75	<li> [#event \| TechNotes] </li>
49	76	</ul>
50	77
51		-These seven artifact types are described in the following sections.
	78	+These seven structural artifact types are described in subsections below.
52	79
53		-In the current implementation (as of 2009-01-25) the artifacts that
	80	+Structural artifacts are ASCII text. The artifact may be PGP clearsigned.
	81	+After removal of the PGP clearsign header and suffix (if any) a structural
	82	+artifact consists of one or more "cards" separated by a single newline
	83	+(ASCII: 0x0a) character. Each card begins with a single
	84	+character "card type". Zero or more arguments may follow
	85	+the card type. All arguments are separated from each other
	86	+and from the card-type character by a single space
	87	+character. There is no surplus white space between arguments
	88	+and no leading or trailing whitespace except for the newline
	89	+character that acts as the card separator. All cards must be in strict
	90	+lexicographical order. There may not be any duplicate cards.
	91	+
	92	+In the current implementation (as of 2017-02-27) the artifacts that
54	93	make up a fossil repository are stored as delta- and zlib-compressed
55	94	blobs in an <a href="http://www.sqlite.org/">SQLite</a> database. This
56	95	is an implementation detail and might change in a future release. For
57	96	the purpose of this article "file format" means the format of the artifacts,
58	97	not how the artifacts are stored on disk. It is the artifact format that
59	98	is intended to be enduring. The specifics of how artifacts are stored on
60	99	disk, though stable, is not intended to live as long as the
61	100	artifact format.
62	101
63		-All of the artifacts can be extracted from a Fossil repository using
64		-the "fossil deconstruct" command.
65		-
66	102	<a name="manifest"></a>
67		-<h2>1.0 The Manifest</h2>
	103	+<h3>2.1 The Manifest</h3>
68	104
69		-A manifest defines a check-in or version of the project
70		-source tree. The manifest contains a list of artifacts for
	105	+A manifest defines a check-in.
	106	+A manifest contains a list of artifacts for
71	107	each file in the project and the corresponding filenames, as
72		-well as information such as parent check-ins, the name of the
	108	+well as information such as parent check-ins, the username of the
73	109	programmer who created the check-in, the date and time when
74	110	the check-in was created, and any check-in comments associated
75	111	with the check-in.
76	112
77		-Any artifact in the repository that follows the syntactic rules
78		-of a manifest is a manifest. Note that a manifest can
79		-be both a real manifest and also a content file, though this
80		-is rare.
81		-
82		-A manifest is a text file. Newline characters
83		-(ASCII 0x0a) separate the file into "cards".
84		-Each card begins with a single
85		-character "card type". Zero or more arguments may follow
86		-the card type. All arguments are separated from each other
87		-and from the card-type character by a single space
88		-character. There is no surplus white space between arguments
89		-and no leading or trailing whitespace except for the newline
90		-character that acts as the card separator.
91		-
92		-All cards of the manifest occur in strict sorted lexicographical order.
93		-No card may be duplicated.
94		-The entire manifest may be PGP clear-signed, but otherwise it
95		-may contain no additional text or data beyond what is described here.
96		-
97	113	Allowed cards in the manifest are as follows:
98	114
99	115	<blockquote>
100	116	<b>B</b> <i>baseline-manifest</i><br>
101	117	<b>C</b> <i>checkin-comment</i><br>
102	118	<b>D</b> <i>time-and-date-stamp</i><br>
103		-<b>F</b> <i>filename</i> ?<i>SHA1-hash</i>? ?<i>permissions</i>? ?<i>old-name</i>?<br>
	119	+<b>F</b> <i>filename</i> ?<i>hash</i>? ?<i>permissions</i>? ?<i>old-name</i>?<br>
104	120	<b>N</b> <i>mimetype</i><br>
105		-<b>P</b> <i>SHA1-hash</i>+<br>
106		-<b>Q</b> (<b>+</b>\|<b>-</b>)<i>SHA1-hash</i> ?<i>SHA1-hash</i>?<br>
	121	+<b>P</b> <i>artifact-hash</i>+<br>
	122	+<b>Q</b> (<b>+</b>\|<b>-</b>)<i>artifact-hash</i> ?<i>artifact-hash</i>?<br>
107	123	<b>R</b> <i>repository-checksum</i><br>
108	124	<b>T</b> (<b>+</b>\|<b>-</b>\|<b></b>)<i>tag-name</i> <b></b> ?<i>value</i>?<br>
109	125	<b>U</b> <i>user-login</i><br>
110	126	<b>Z</b> <i>manifest-checksum</i>
111	127	</blockquote>
		@@ -145,11 +161,11 @@
145	161	check-in relative to the root of the project file hierarchy. No ".."
146	162	or "." directories are allowed within the filename. Space characters
147	163	are escaped as in C-card comment text. Backslash characters and
148	164	newlines are not allowed within filenames. The directory separator
149	165	character is a forward slash (ASCII 0x2F). The second argument to the
150		-F-card is the full 40-character lower-case hexadecimal SHA1 hash of
	166	+F-card is the lower-case hexadecimal artifact hash of
151	167	the content artifact. The second argument is required for baseline
152	168	manifests but is optional for delta manifests. When the second
153	169	argument to the F-card is omitted, it means that the file has been
154	170	deleted relative to the baseline (files removed in baseline manifests
155	171	versions are <em>not</em> added as F-cards). The optional 3rd argument
		@@ -167,14 +183,14 @@
167	183	is used.
168	184
169	185	A manifest has zero or one P-cards. Most manifests have one P-card.
170	186	The P-card has a varying number of arguments that
171	187	define other manifests from which the current manifest
172		-is derived. Each argument is a 40-character lowercase
173		-hexadecimal SHA1 of a predecessor manifest. All arguments
	188	+is derived. Each argument is a lowercase
	189	+hexadecimal artifact hash of a predecessor manifest. All arguments
174	190	to the P-card must be unique within that card.
175		-The first argument is the SHA1 of the direct ancestor of the manifest.
	191	+The first argument is the artifact hash of the direct ancestor of the manifest.
176	192	Other arguments define manifests with which the first was
177	193	merged to yield the current manifest. Most manifests have
178	194	a P-card with a single argument. The first manifest in the
179	195	project has no ancestors and thus has no P-card or (depending
180	196	on the Fossil version) an empty P-card (no arguments).
		@@ -229,41 +245,27 @@
229	245	check-in comment argument to the C-card.
230	246
231	247	A manifest must have a single Z-card as its last line. The argument
232	248	to the Z-card is a 32-character lowercase hexadecimal MD5 hash
233	249	of all prior lines of the manifest up to and including the newline
234		-character that immediately precedes the "Z". The Z-card is
	250	+character that immediately precedes the "Z", excluding any PGP
	251	+clear-signing prefix. The Z-card is
235	252	a sanity check to prove that the manifest is well-formed and
236	253	consistent.
237	254
238	255	A sample manifest from Fossil itself can be seen
239	256	[/artifact/28987096ac \| here].
240	257
241	258	<a name="cluster"></a>
242		-<h2>2.0 Clusters</h2>
	259	+<h3>2.2 Clusters</h3>
243	260
244	261	A cluster is an artifact that declares the existence of other artifacts.
245	262	Clusters are used during repository synchronization to help
246	263	reduce network traffic. As such, clusters are an optimization and
247	264	may be removed from a repository without loss or damage to the
248	265	underlying project code.
249	266
250		-Clusters follow a syntax that is very similar to manifests.
251		-A cluster is a line-oriented text file. Newline characters
252		-(ASCII 0x0a) separate the artifact into cards. Each card begins with a single
253		-character "card type". Zero or more arguments may follow
254		-the card type. All arguments are separated from each other
255		-and from the card-type character by a single space
256		-character. There is no surplus white space between arguments
257		-and no leading or trailing whitespace except for the newline
258		-character that acts as the card separator.
259		-All cards of a cluster occur in strict sorted lexicographical order.
260		-No card may be duplicated.
261		-The cluster may not contain additional text or data beyond
262		-what is described here.
263		-Unlike manifests, clusters are never PGP signed.
264		-
265	267	Allowed cards in the cluster are as follows:
266	268
267	269	<blockquote>
268	270	<b>M</b> <i>artifact-id</i><br />
269	271	<b>Z</b> <i>checksum</i>
		@@ -278,20 +280,14 @@
278	280
279	281	An example cluster from Fossil can be seen
280	282	[/artifact/d03dbdd73a2a8 \| here].
281	283
282	284	<a name="ctrl"></a>
283		-<h2>3.0 Control Artifacts</h2>
	285	+<h3>2.3 Control Artifacts</h3>
284	286
285	287	Control artifacts are used to assign properties to other artifacts
286		-within the repository. The basic format of a control artifact is
287		-the same as a manifest or cluster. A control artifact is a text
288		-file divided into cards by newline characters. Each card has a
289		-single-character card type followed by arguments. Spaces separate
290		-the card type and the arguments. No surplus whitespace is allowed.
291		-All cards must occur in strict lexicographical order.
292		-
	288	+within the repository.
293	289	Allowed cards in a control artifact are as follows:
294	290
295	291	<blockquote>
296	292	<b>D</b> <i>time-and-date-stamp</i><br />
297	293	<b>T</b> (<b>+</b>\|<b>-</b>\|<b>*</b>)<i>tag-name</i> <i>artifact-id</i> ?<i>value</i>?<br />
		@@ -338,16 +334,15 @@
338	334
339	335	An example control artifacts can be seen [/info/9d302ccda8 \| here].
340	336
341	337
342	338	<a name="wikichng"></a>
343		-<h2>4.0 Wiki Pages</h2>
	339	+<h3>2.4 Wiki Pages</h3>
344	340
345		-A wiki page is an artifact with a format similar to manifests,
346		-clusters, and control artifacts. The artifact is divided into
347		-cards by newline characters. The format of each card is as in
348		-manifests, clusters, and control artifacts. Wiki artifacts accept
	341	+A wiki artifact defines a single version of a
	342	+single wiki page.
	343	+Wiki artifacts accept
349	344	the following card types:
350	345
351	346	<blockquote>
352	347	<b>D</b> <i>time-and-date-stamp</i><br />
353	348	<b>L</b> <i>wiki-title</i><br />
		@@ -375,11 +370,11 @@
375	370
376	371	An example wiki artifact can be seen
377	372	[/artifact?name=7b2f5fd0e0&txt=1 \| here].
378	373
379	374	<a name="tktchng"></a>
380		-<h2>5.0 Ticket Changes</h2>
	375	+<h3>2.5 Ticket Changes</h3>
381	376
382	377	A ticket-change artifact represents a change to a trouble ticket.
383	378	The following cards are allowed on a ticket change artifact:
384	379
385	380	<blockquote>
		@@ -421,11 +416,11 @@
421	416
422	417	An example ticket-change artifact can be seen
423	418	[/artifact/91f1ec6af053 \| here].
424	419
425	420	<a name="attachment"></a>
426		-<h2>6.0 Attachments</h2>
	421	+<h3>2.6 Attachments</h3>
427	422
428	423	An attachment artifact associates some other artifact that is the
429	424	attachment (the source artifact) with a ticket or wiki page or
430	425	technical note to which
431	426	the attachment is connected (the target artifact).
		@@ -463,11 +458,11 @@
463	458	The Z card is the usual checksum over the rest of the attachment artifact.
464	459	The Z card is required.
465	460
466	461
467	462	<a name="event"></a>
468		-<h2>7.0 Technical Notes</h2>
	463	+<h3>2.7 Technical Notes</h3>
469	464
470	465	A technical note or "technote" artifact (formerly known as an "event" artifact)
471	466	associates a timeline comment and a page of text
472	467	(similar to a wiki page) with a point in time. Technotes can be used
473	468	to record project milestones, release notes, blog entries, process
		@@ -532,11 +527,11 @@
532	527
533	528	The Z card is the required checksum over the rest of the artifact.
534	529
535	530
536	531	<a name="summary"></a>
537		-<h2>8.0 Card Summary</h2>
	532	+<h2>3.0 Card Summary</h2>
538	533
539	534	The following table summarizes the various kinds of cards that appear
540	535	on Fossil artifacts. A blank entry means that combination of card and
541	536	artifact is not legal. A number or range of numbers indicates the number
542	537	of times a card may (or must) appear in the corresponding artifact type.
		@@ -739,16 +734,16 @@
739	734	</tr>
740	735	</table>
741	736
742	737
743	738	<a name="addenda"></a>
744		-<h2>9.0 Addenda</h2>
	739	+<h2>4.0 Addenda</h2>
745	740
746	741	This section contains additional information which may be useful when
747	742	implementing algorithms described above.
748	743
749		-<h3>R Card Hash Calculation</h3>
	744	+<h3>4.1 R-Card Hash Calculation</h3>
750	745
751	746	Given a manifest file named <tt>MF</tt>, the following Bash shell code
752	747	demonstrates how to compute the value of the R card in that manifest.
753	748	This example uses manifest [28987096ac]. Lines starting with <tt>#</tt> are
754	749	shell input and other lines are output. This demonstration assumes that the
755	750

	--- www/fileformat.wiki
	+++ www/fileformat.wiki
	@@ -9,14 +9,19 @@
9	searchable, and extensible by people not yet born.
10
11	The global state of a fossil repository is an unordered
12	set of <i>artifacts</i>.
13	An artifact might be a source code file, the text of a wiki page,
14	part of a trouble ticket, or one of several special control artifacts
15	used to show the relationships between other artifacts within the
16	project. Each artifact is normally represented on disk as a separate
17	file. Artifacts can be text or binary.





18
19	In addition to the global state,
20	each fossil repository also contains local state.
21	The local state consists of web-page formatting
22	preferences, authorized users, ticket display and reporting formats,
	@@ -27,18 +32,40 @@
27	with the global state.
28	The local state is not composed of artifacts and is not intended to be enduring.
29	This document is concerned with global state only. Local state is only
30	mentioned here in order to distinguish it from global state.
31
32	Each artifact in the repository is named by its SHA1 hash.
33	No prefixes or meta information is added to an artifact before
34	its hash is computed. The name of an artifact in the repository
35	is exactly the same SHA1 hash that is computed by sha1sum
36	on the file as it exists in your source tree.</p>
37
38	Some artifacts have a particular format which gives them special
39	meaning to fossil. Fossil recognizes:






















40
41	<ul>
42	<li> [#manifest \| Manifests] </li>
43	<li> [#cluster \| Clusters] </li>
44	<li> [#ctrl \| Control Artifacts] </li>
	@@ -46,66 +73,55 @@
46	<li> [#tktchng \| Ticket Changes] </li>
47	<li> [#attachment \| Attachments] </li>
48	<li> [#event \| TechNotes] </li>
49	</ul>
50
51	These seven artifact types are described in the following sections.
52
53	In the current implementation (as of 2009-01-25) the artifacts that












54	make up a fossil repository are stored as delta- and zlib-compressed
55	blobs in an <a href="http://www.sqlite.org/">SQLite</a> database. This
56	is an implementation detail and might change in a future release. For
57	the purpose of this article "file format" means the format of the artifacts,
58	not how the artifacts are stored on disk. It is the artifact format that
59	is intended to be enduring. The specifics of how artifacts are stored on
60	disk, though stable, is not intended to live as long as the
61	artifact format.
62
63	All of the artifacts can be extracted from a Fossil repository using
64	the "fossil deconstruct" command.
65
66	<a name="manifest"></a>
67	<h2>1.0 The Manifest</h2>
68
69	A manifest defines a check-in or version of the project
70	source tree. The manifest contains a list of artifacts for
71	each file in the project and the corresponding filenames, as
72	well as information such as parent check-ins, the name of the
73	programmer who created the check-in, the date and time when
74	the check-in was created, and any check-in comments associated
75	with the check-in.
76
77	Any artifact in the repository that follows the syntactic rules
78	of a manifest is a manifest. Note that a manifest can
79	be both a real manifest and also a content file, though this
80	is rare.
81
82	A manifest is a text file. Newline characters
83	(ASCII 0x0a) separate the file into "cards".
84	Each card begins with a single
85	character "card type". Zero or more arguments may follow
86	the card type. All arguments are separated from each other
87	and from the card-type character by a single space
88	character. There is no surplus white space between arguments
89	and no leading or trailing whitespace except for the newline
90	character that acts as the card separator.
91
92	All cards of the manifest occur in strict sorted lexicographical order.
93	No card may be duplicated.
94	The entire manifest may be PGP clear-signed, but otherwise it
95	may contain no additional text or data beyond what is described here.
96
97	Allowed cards in the manifest are as follows:
98
99	<blockquote>
100	<b>B</b> <i>baseline-manifest</i><br>
101	<b>C</b> <i>checkin-comment</i><br>
102	<b>D</b> <i>time-and-date-stamp</i><br>
103	<b>F</b> <i>filename</i> ?<i>SHA1-hash</i>? ?<i>permissions</i>? ?<i>old-name</i>?<br>
104	<b>N</b> <i>mimetype</i><br>
105	<b>P</b> <i>SHA1-hash</i>+<br>
106	<b>Q</b> (<b>+</b>\|<b>-</b>)<i>SHA1-hash</i> ?<i>SHA1-hash</i>?<br>
107	<b>R</b> <i>repository-checksum</i><br>
108	<b>T</b> (<b>+</b>\|<b>-</b>\|<b></b>)<i>tag-name</i> <b></b> ?<i>value</i>?<br>
109	<b>U</b> <i>user-login</i><br>
110	<b>Z</b> <i>manifest-checksum</i>
111	</blockquote>
	@@ -145,11 +161,11 @@
145	check-in relative to the root of the project file hierarchy. No ".."
146	or "." directories are allowed within the filename. Space characters
147	are escaped as in C-card comment text. Backslash characters and
148	newlines are not allowed within filenames. The directory separator
149	character is a forward slash (ASCII 0x2F). The second argument to the
150	F-card is the full 40-character lower-case hexadecimal SHA1 hash of
151	the content artifact. The second argument is required for baseline
152	manifests but is optional for delta manifests. When the second
153	argument to the F-card is omitted, it means that the file has been
154	deleted relative to the baseline (files removed in baseline manifests
155	versions are <em>not</em> added as F-cards). The optional 3rd argument
	@@ -167,14 +183,14 @@
167	is used.
168
169	A manifest has zero or one P-cards. Most manifests have one P-card.
170	The P-card has a varying number of arguments that
171	define other manifests from which the current manifest
172	is derived. Each argument is a 40-character lowercase
173	hexadecimal SHA1 of a predecessor manifest. All arguments
174	to the P-card must be unique within that card.
175	The first argument is the SHA1 of the direct ancestor of the manifest.
176	Other arguments define manifests with which the first was
177	merged to yield the current manifest. Most manifests have
178	a P-card with a single argument. The first manifest in the
179	project has no ancestors and thus has no P-card or (depending
180	on the Fossil version) an empty P-card (no arguments).
	@@ -229,41 +245,27 @@
229	check-in comment argument to the C-card.
230
231	A manifest must have a single Z-card as its last line. The argument
232	to the Z-card is a 32-character lowercase hexadecimal MD5 hash
233	of all prior lines of the manifest up to and including the newline
234	character that immediately precedes the "Z". The Z-card is

235	a sanity check to prove that the manifest is well-formed and
236	consistent.
237
238	A sample manifest from Fossil itself can be seen
239	[/artifact/28987096ac \| here].
240
241	<a name="cluster"></a>
242	<h2>2.0 Clusters</h2>
243
244	A cluster is an artifact that declares the existence of other artifacts.
245	Clusters are used during repository synchronization to help
246	reduce network traffic. As such, clusters are an optimization and
247	may be removed from a repository without loss or damage to the
248	underlying project code.
249
250	Clusters follow a syntax that is very similar to manifests.
251	A cluster is a line-oriented text file. Newline characters
252	(ASCII 0x0a) separate the artifact into cards. Each card begins with a single
253	character "card type". Zero or more arguments may follow
254	the card type. All arguments are separated from each other
255	and from the card-type character by a single space
256	character. There is no surplus white space between arguments
257	and no leading or trailing whitespace except for the newline
258	character that acts as the card separator.
259	All cards of a cluster occur in strict sorted lexicographical order.
260	No card may be duplicated.
261	The cluster may not contain additional text or data beyond
262	what is described here.
263	Unlike manifests, clusters are never PGP signed.
264
265	Allowed cards in the cluster are as follows:
266
267	<blockquote>
268	<b>M</b> <i>artifact-id</i><br />
269	<b>Z</b> <i>checksum</i>
	@@ -278,20 +280,14 @@
278
279	An example cluster from Fossil can be seen
280	[/artifact/d03dbdd73a2a8 \| here].
281
282	<a name="ctrl"></a>
283	<h2>3.0 Control Artifacts</h2>
284
285	Control artifacts are used to assign properties to other artifacts
286	within the repository. The basic format of a control artifact is
287	the same as a manifest or cluster. A control artifact is a text
288	file divided into cards by newline characters. Each card has a
289	single-character card type followed by arguments. Spaces separate
290	the card type and the arguments. No surplus whitespace is allowed.
291	All cards must occur in strict lexicographical order.
292
293	Allowed cards in a control artifact are as follows:
294
295	<blockquote>
296	<b>D</b> <i>time-and-date-stamp</i><br />
297	<b>T</b> (<b>+</b>\|<b>-</b>\|<b>*</b>)<i>tag-name</i> <i>artifact-id</i> ?<i>value</i>?<br />
	@@ -338,16 +334,15 @@
338
339	An example control artifacts can be seen [/info/9d302ccda8 \| here].
340
341
342	<a name="wikichng"></a>
343	<h2>4.0 Wiki Pages</h2>
344
345	A wiki page is an artifact with a format similar to manifests,
346	clusters, and control artifacts. The artifact is divided into
347	cards by newline characters. The format of each card is as in
348	manifests, clusters, and control artifacts. Wiki artifacts accept
349	the following card types:
350
351	<blockquote>
352	<b>D</b> <i>time-and-date-stamp</i><br />
353	<b>L</b> <i>wiki-title</i><br />
	@@ -375,11 +370,11 @@
375
376	An example wiki artifact can be seen
377	[/artifact?name=7b2f5fd0e0&txt=1 \| here].
378
379	<a name="tktchng"></a>
380	<h2>5.0 Ticket Changes</h2>
381
382	A ticket-change artifact represents a change to a trouble ticket.
383	The following cards are allowed on a ticket change artifact:
384
385	<blockquote>
	@@ -421,11 +416,11 @@
421
422	An example ticket-change artifact can be seen
423	[/artifact/91f1ec6af053 \| here].
424
425	<a name="attachment"></a>
426	<h2>6.0 Attachments</h2>
427
428	An attachment artifact associates some other artifact that is the
429	attachment (the source artifact) with a ticket or wiki page or
430	technical note to which
431	the attachment is connected (the target artifact).
	@@ -463,11 +458,11 @@
463	The Z card is the usual checksum over the rest of the attachment artifact.
464	The Z card is required.
465
466
467	<a name="event"></a>
468	<h2>7.0 Technical Notes</h2>
469
470	A technical note or "technote" artifact (formerly known as an "event" artifact)
471	associates a timeline comment and a page of text
472	(similar to a wiki page) with a point in time. Technotes can be used
473	to record project milestones, release notes, blog entries, process
	@@ -532,11 +527,11 @@
532
533	The Z card is the required checksum over the rest of the artifact.
534
535
536	<a name="summary"></a>
537	<h2>8.0 Card Summary</h2>
538
539	The following table summarizes the various kinds of cards that appear
540	on Fossil artifacts. A blank entry means that combination of card and
541	artifact is not legal. A number or range of numbers indicates the number
542	of times a card may (or must) appear in the corresponding artifact type.
	@@ -739,16 +734,16 @@
739	</tr>
740	</table>
741
742
743	<a name="addenda"></a>
744	<h2>9.0 Addenda</h2>
745
746	This section contains additional information which may be useful when
747	implementing algorithms described above.
748
749	<h3>R Card Hash Calculation</h3>
750
751	Given a manifest file named <tt>MF</tt>, the following Bash shell code
752	demonstrates how to compute the value of the R card in that manifest.
753	This example uses manifest [28987096ac]. Lines starting with <tt>#</tt> are
754	shell input and other lines are output. This demonstration assumes that the
755

	--- www/fileformat.wiki
	+++ www/fileformat.wiki
	@@ -9,14 +9,19 @@
9	searchable, and extensible by people not yet born.
10
11	The global state of a fossil repository is an unordered
12	set of <i>artifacts</i>.
13	An artifact might be a source code file, the text of a wiki page,
14	part of a trouble ticket, a description of a check-in including all
15	the files in that check-in with the check-in comment and so forth.
16	Artifacts are broadly grouped into two types: content artifacts and
17	structural artifacts. Content artifacts are the raw project source-code
18	files that are checked into the repository. Structural artifacts have
19	special formatting rules and are used to show the relationships between
20	other artifacts in the repository. It is possible for an artifact to
21	be both a structure artifact and a content artifact, though this is
22	rare. Artifacts can be text or binary.
23
24	In addition to the global state,
25	each fossil repository also contains local state.
26	The local state consists of web-page formatting
27	preferences, authorized users, ticket display and reporting formats,
	@@ -27,18 +32,40 @@
32	with the global state.
33	The local state is not composed of artifacts and is not intended to be enduring.
34	This document is concerned with global state only. Local state is only
35	mentioned here in order to distinguish it from global state.
36
37	<a name="names"></a>
38	<h2>1.0 Artifact Names</h2>
39
40	Each artifact in the repository is named by a hash of its content.
41	No prefixes, suffixes, or other information is added to an artifact before
42	the hash is computed. The artifact name is just the (lower-case
43	hexadecimal) hash of the raw artifact.
44
45	Fossil currently computes artifact names using either SHA1 or SHA3-256. It
46	is relatively easy to add new algorithms in the future, but there are no
47	plans to do so at this time.
48
49	When referring to artifacts in using tty commands or webpage URLs, it is
50	sufficient to specify a unique prefix for the artifact name. If the input
51	prefix is not unique, Fossil will show an error. Within a structural
52	artifact, however, all references to other artifacts must be the complete
53	hash.
54
55	Prior to Fossil version 2.0, all names were formed from the SHA1 hash of
56	the artifact. The key innovation in Fossil 2.0 was adding support for
57	alternative hash algorithms.
58
59	<a name="structural"></a>
60	<h2>2.0 Structural Artifacts</h2>
61
62	A structural artifact is an artifact with a particular format
63	that is used to define the relationships between other artifacts in the
64	repository.
65	Fossil recognizes the following kinds of structural
66	artifacts:
67
68	<ul>
69	<li> [#manifest \| Manifests] </li>
70	<li> [#cluster \| Clusters] </li>
71	<li> [#ctrl \| Control Artifacts] </li>
	@@ -46,66 +73,55 @@
73	<li> [#tktchng \| Ticket Changes] </li>
74	<li> [#attachment \| Attachments] </li>
75	<li> [#event \| TechNotes] </li>
76	</ul>
77
78	These seven structural artifact types are described in subsections below.
79
80	Structural artifacts are ASCII text. The artifact may be PGP clearsigned.
81	After removal of the PGP clearsign header and suffix (if any) a structural
82	artifact consists of one or more "cards" separated by a single newline
83	(ASCII: 0x0a) character. Each card begins with a single
84	character "card type". Zero or more arguments may follow
85	the card type. All arguments are separated from each other
86	and from the card-type character by a single space
87	character. There is no surplus white space between arguments
88	and no leading or trailing whitespace except for the newline
89	character that acts as the card separator. All cards must be in strict
90	lexicographical order. There may not be any duplicate cards.
91
92	In the current implementation (as of 2017-02-27) the artifacts that
93	make up a fossil repository are stored as delta- and zlib-compressed
94	blobs in an <a href="http://www.sqlite.org/">SQLite</a> database. This
95	is an implementation detail and might change in a future release. For
96	the purpose of this article "file format" means the format of the artifacts,
97	not how the artifacts are stored on disk. It is the artifact format that
98	is intended to be enduring. The specifics of how artifacts are stored on
99	disk, though stable, is not intended to live as long as the
100	artifact format.
101



102	<a name="manifest"></a>
103	<h3>2.1 The Manifest</h3>
104
105	A manifest defines a check-in.
106	A manifest contains a list of artifacts for
107	each file in the project and the corresponding filenames, as
108	well as information such as parent check-ins, the username of the
109	programmer who created the check-in, the date and time when
110	the check-in was created, and any check-in comments associated
111	with the check-in.
112




















113	Allowed cards in the manifest are as follows:
114
115	<blockquote>
116	<b>B</b> <i>baseline-manifest</i><br>
117	<b>C</b> <i>checkin-comment</i><br>
118	<b>D</b> <i>time-and-date-stamp</i><br>
119	<b>F</b> <i>filename</i> ?<i>hash</i>? ?<i>permissions</i>? ?<i>old-name</i>?<br>
120	<b>N</b> <i>mimetype</i><br>
121	<b>P</b> <i>artifact-hash</i>+<br>
122	<b>Q</b> (<b>+</b>\|<b>-</b>)<i>artifact-hash</i> ?<i>artifact-hash</i>?<br>
123	<b>R</b> <i>repository-checksum</i><br>
124	<b>T</b> (<b>+</b>\|<b>-</b>\|<b></b>)<i>tag-name</i> <b></b> ?<i>value</i>?<br>
125	<b>U</b> <i>user-login</i><br>
126	<b>Z</b> <i>manifest-checksum</i>
127	</blockquote>
	@@ -145,11 +161,11 @@
161	check-in relative to the root of the project file hierarchy. No ".."
162	or "." directories are allowed within the filename. Space characters
163	are escaped as in C-card comment text. Backslash characters and
164	newlines are not allowed within filenames. The directory separator
165	character is a forward slash (ASCII 0x2F). The second argument to the
166	F-card is the lower-case hexadecimal artifact hash of
167	the content artifact. The second argument is required for baseline
168	manifests but is optional for delta manifests. When the second
169	argument to the F-card is omitted, it means that the file has been
170	deleted relative to the baseline (files removed in baseline manifests
171	versions are <em>not</em> added as F-cards). The optional 3rd argument
	@@ -167,14 +183,14 @@
183	is used.
184
185	A manifest has zero or one P-cards. Most manifests have one P-card.
186	The P-card has a varying number of arguments that
187	define other manifests from which the current manifest
188	is derived. Each argument is a lowercase
189	hexadecimal artifact hash of a predecessor manifest. All arguments
190	to the P-card must be unique within that card.
191	The first argument is the artifact hash of the direct ancestor of the manifest.
192	Other arguments define manifests with which the first was
193	merged to yield the current manifest. Most manifests have
194	a P-card with a single argument. The first manifest in the
195	project has no ancestors and thus has no P-card or (depending
196	on the Fossil version) an empty P-card (no arguments).
	@@ -229,41 +245,27 @@
245	check-in comment argument to the C-card.
246
247	A manifest must have a single Z-card as its last line. The argument
248	to the Z-card is a 32-character lowercase hexadecimal MD5 hash
249	of all prior lines of the manifest up to and including the newline
250	character that immediately precedes the "Z", excluding any PGP
251	clear-signing prefix. The Z-card is
252	a sanity check to prove that the manifest is well-formed and
253	consistent.
254
255	A sample manifest from Fossil itself can be seen
256	[/artifact/28987096ac \| here].
257
258	<a name="cluster"></a>
259	<h3>2.2 Clusters</h3>
260
261	A cluster is an artifact that declares the existence of other artifacts.
262	Clusters are used during repository synchronization to help
263	reduce network traffic. As such, clusters are an optimization and
264	may be removed from a repository without loss or damage to the
265	underlying project code.
266















267	Allowed cards in the cluster are as follows:
268
269	<blockquote>
270	<b>M</b> <i>artifact-id</i><br />
271	<b>Z</b> <i>checksum</i>
	@@ -278,20 +280,14 @@
280
281	An example cluster from Fossil can be seen
282	[/artifact/d03dbdd73a2a8 \| here].
283
284	<a name="ctrl"></a>
285	<h3>2.3 Control Artifacts</h3>
286
287	Control artifacts are used to assign properties to other artifacts
288	within the repository.






289	Allowed cards in a control artifact are as follows:
290
291	<blockquote>
292	<b>D</b> <i>time-and-date-stamp</i><br />
293	<b>T</b> (<b>+</b>\|<b>-</b>\|<b>*</b>)<i>tag-name</i> <i>artifact-id</i> ?<i>value</i>?<br />
	@@ -338,16 +334,15 @@
334
335	An example control artifacts can be seen [/info/9d302ccda8 \| here].
336
337
338	<a name="wikichng"></a>
339	<h3>2.4 Wiki Pages</h3>
340
341	A wiki artifact defines a single version of a
342	single wiki page.
343	Wiki artifacts accept

344	the following card types:
345
346	<blockquote>
347	<b>D</b> <i>time-and-date-stamp</i><br />
348	<b>L</b> <i>wiki-title</i><br />
	@@ -375,11 +370,11 @@
370
371	An example wiki artifact can be seen
372	[/artifact?name=7b2f5fd0e0&txt=1 \| here].
373
374	<a name="tktchng"></a>
375	<h3>2.5 Ticket Changes</h3>
376
377	A ticket-change artifact represents a change to a trouble ticket.
378	The following cards are allowed on a ticket change artifact:
379
380	<blockquote>
	@@ -421,11 +416,11 @@
416
417	An example ticket-change artifact can be seen
418	[/artifact/91f1ec6af053 \| here].
419
420	<a name="attachment"></a>
421	<h3>2.6 Attachments</h3>
422
423	An attachment artifact associates some other artifact that is the
424	attachment (the source artifact) with a ticket or wiki page or
425	technical note to which
426	the attachment is connected (the target artifact).
	@@ -463,11 +458,11 @@
458	The Z card is the usual checksum over the rest of the attachment artifact.
459	The Z card is required.
460
461
462	<a name="event"></a>
463	<h3>2.7 Technical Notes</h3>
464
465	A technical note or "technote" artifact (formerly known as an "event" artifact)
466	associates a timeline comment and a page of text
467	(similar to a wiki page) with a point in time. Technotes can be used
468	to record project milestones, release notes, blog entries, process
	@@ -532,11 +527,11 @@
527
528	The Z card is the required checksum over the rest of the artifact.
529
530
531	<a name="summary"></a>
532	<h2>3.0 Card Summary</h2>
533
534	The following table summarizes the various kinds of cards that appear
535	on Fossil artifacts. A blank entry means that combination of card and
536	artifact is not legal. A number or range of numbers indicates the number
537	of times a card may (or must) appear in the corresponding artifact type.
	@@ -739,16 +734,16 @@
734	</tr>
735	</table>
736
737
738	<a name="addenda"></a>
739	<h2>4.0 Addenda</h2>
740
741	This section contains additional information which may be useful when
742	implementing algorithms described above.
743
744	<h3>4.1 R-Card Hash Calculation</h3>
745
746	Given a manifest file named <tt>MF</tt>, the following Bash shell code
747	demonstrates how to compute the value of the R card in that manifest.
748	This example uses manifest [28987096ac]. Lines starting with <tt>#</tt> are
749	shell input and other lines are output. This demonstration assumes that the
750

M www/fossil-v-git.wiki

+4 -3

		--- www/fossil-v-git.wiki
		+++ www/fossil-v-git.wiki
		@@ -67,11 +67,12 @@
67	67
68	68	<h3>3.2 Database</h3>
69	69
70	70	The baseline data structures for Fossil and Git are the same (modulo
71	71	formatting details). Both systems store check-ins as immutable
72		-objects referencing their immediate ancestors and named by their SHA1 hash.
	72	+objects referencing their immediate ancestors and named by a
	73	+cryptographic hash of the check-in content.
73	74
74	75	The difference is that Git stores its objects as individual files
75	76	in the ".git" folder or compressed into
76	77	bespoke "pack-files", whereas Fossil stores its objects in a
77	78	relational ([https://www.sqlite.org/\|SQLite]) database file. To put it
		@@ -117,11 +118,11 @@
117	118	closely supervised by an highly engaged architect and implemented by
118	119	a clique of developers.
119	120
120	121	Nota Bene: This is not to say that Git cannot be used for cathedral-style
121	122	development or that Fossil cannot be used for bazaar-style development.
122		-They can be. But those modes are not their design intent nor the their
	123	+They can be. But those modes are not their design intent nor their
123	124	low-friction path.
124	125
125	126	Git encourages a style in which individual developers work in relative
126	127	isolation, maintaining their
127	128	own branches and occasionally rebasing and pushing selected changes up
		@@ -144,11 +145,11 @@
144	145	Fossil was specifically designed to support the development of SQLite.
145	146
146	147	Both SQLite and Linux are important pieces of software.
147	148	SQLite is found on far more systems than Linux. (Almost every Linux
148	149	system uses SQLite, but there are many non-Linux systems such as
149		-iPhones, PlayStations, and Windows PC that use SQLite.) On the other
	150	+iPhones, PlayStations, and Windows PCs that use SQLite.) On the other
150	151	hand, for those systems that do use Linux, Linux is a far more important
151	152	component.
152	153
153	154	Linux uses a bazaar-style development model. There are thousands and
154	155	thousands of contributors, most of whom do not know each others names.
155	156
156	157	ADDED www/hashpolicy.wiki

	--- www/fossil-v-git.wiki
	+++ www/fossil-v-git.wiki
	@@ -67,11 +67,12 @@
67
68	<h3>3.2 Database</h3>
69
70	The baseline data structures for Fossil and Git are the same (modulo
71	formatting details). Both systems store check-ins as immutable
72	objects referencing their immediate ancestors and named by their SHA1 hash.

73
74	The difference is that Git stores its objects as individual files
75	in the ".git" folder or compressed into
76	bespoke "pack-files", whereas Fossil stores its objects in a
77	relational ([https://www.sqlite.org/\|SQLite]) database file. To put it
	@@ -117,11 +118,11 @@
117	closely supervised by an highly engaged architect and implemented by
118	a clique of developers.
119
120	Nota Bene: This is not to say that Git cannot be used for cathedral-style
121	development or that Fossil cannot be used for bazaar-style development.
122	They can be. But those modes are not their design intent nor the their
123	low-friction path.
124
125	Git encourages a style in which individual developers work in relative
126	isolation, maintaining their
127	own branches and occasionally rebasing and pushing selected changes up
	@@ -144,11 +145,11 @@
144	Fossil was specifically designed to support the development of SQLite.
145
146	Both SQLite and Linux are important pieces of software.
147	SQLite is found on far more systems than Linux. (Almost every Linux
148	system uses SQLite, but there are many non-Linux systems such as
149	iPhones, PlayStations, and Windows PC that use SQLite.) On the other
150	hand, for those systems that do use Linux, Linux is a far more important
151	component.
152
153	Linux uses a bazaar-style development model. There are thousands and
154	thousands of contributors, most of whom do not know each others names.
155
156	DDED www/hashpolicy.wiki

	--- www/fossil-v-git.wiki
	+++ www/fossil-v-git.wiki
	@@ -67,11 +67,12 @@
67
68	<h3>3.2 Database</h3>
69
70	The baseline data structures for Fossil and Git are the same (modulo
71	formatting details). Both systems store check-ins as immutable
72	objects referencing their immediate ancestors and named by a
73	cryptographic hash of the check-in content.
74
75	The difference is that Git stores its objects as individual files
76	in the ".git" folder or compressed into
77	bespoke "pack-files", whereas Fossil stores its objects in a
78	relational ([https://www.sqlite.org/\|SQLite]) database file. To put it
	@@ -117,11 +118,11 @@
118	closely supervised by an highly engaged architect and implemented by
119	a clique of developers.
120
121	Nota Bene: This is not to say that Git cannot be used for cathedral-style
122	development or that Fossil cannot be used for bazaar-style development.
123	They can be. But those modes are not their design intent nor their
124	low-friction path.
125
126	Git encourages a style in which individual developers work in relative
127	isolation, maintaining their
128	own branches and occasionally rebasing and pushing selected changes up
	@@ -144,11 +145,11 @@
145	Fossil was specifically designed to support the development of SQLite.
146
147	Both SQLite and Linux are important pieces of software.
148	SQLite is found on far more systems than Linux. (Almost every Linux
149	system uses SQLite, but there are many non-Linux systems such as
150	iPhones, PlayStations, and Windows PCs that use SQLite.) On the other
151	hand, for those systems that do use Linux, Linux is a far more important
152	component.
153
154	Linux uses a bazaar-style development model. There are thousands and
155	thousands of contributors, most of whom do not know each others names.
156
157	DDED www/hashpolicy.wiki

M www/hashpolicy.wiki

+20

		--- a/www/hashpolicy.wiki
		+++ b/www/hashpolicy.wiki
		@@ -0,0 +1,20 @@
	1	+<title>Hash Policy</title>
	2	+
	3	+<h2> Executive Summary, Orcutive Summary</h2>
	4	+
	5	+<b>Or: How To </h2>
	6	+
	7	+There i This Article</b>
	8	+
	9	+Thham now
	10	+upgraded to
	11	+change texpected to be
	12	+replaced ot expected to be
	13	+replaced until Ma
	14	+out o
	15	+Debian 9 is implement0 or later
	16	+
	17	+work and
	18	+Hash Policy</title>
	19	+
	20	+<h2>< Introduction ha", not generic SHA1sequel

	--- a/www/hashpolicy.wiki
	+++ b/www/hashpolicy.wiki
	@@ -0,0 +1,20 @@

	--- a/www/hashpolicy.wiki
	+++ b/www/hashpolicy.wiki
	@@ -0,0 +1,20 @@
1	<title>Hash Policy</title>
2
3	<h2> Executive Summary, Orcutive Summary</h2>
4
5	<b>Or: How To </h2>
6
7	There i This Article</b>
8
9	Thham now
10	upgraded to
11	change texpected to be
12	replaced ot expected to be
13	replaced until Ma
14	out o
15	Debian 9 is implement0 or later
16
17	work and
18	Hash Policy</title>
19
20	<h2>< Introduction ha", not generic SHA1sequel

M www/mkdownload.tcl

+2 -2

		--- www/mkdownload.tcl
		+++ www/mkdownload.tcl
		@@ -37,12 +37,12 @@
37	37	set avers($version) 1
38	38	}
39	39	}
40	40	close $in
41	41
	42	+set vdate(2.0) 2017-03-03
42	43	set vdate(1.37) 2017-01-15
43		-set vdate(1.36) 2016-10-24
44	44
45	45	# Do all versions from newest to oldest
46	46	#
47	47	foreach vers [lsort -decr -real [array names avers]] {
48	48	# set hr "../timeline?c=version-$vers;y=ci"
		@@ -57,11 +57,11 @@
57	57	puts $out "</b></center>"
58	58	puts $out "</td></tr>"
59	59	puts $out "<tr>"
60	60
61	61	foreach {prefix img desc} {
62		- fossil-linux-x86 linux.gif {Linux 3.x x86}
	62	+ fossil-linux linux.gif {Linux 3.x x64}
63	63	fossil-macosx mac.gif {Mac 10.x x86}
64	64	fossil-openbsd-x86 openbsd.gif {OpenBSD 5.x x86}
65	65	fossil-w32 win32.gif {Windows}
66	66	fossil-src src.gif {Source Tarball}
67	67	} {
68	68

	--- www/mkdownload.tcl
	+++ www/mkdownload.tcl
	@@ -37,12 +37,12 @@
37	set avers($version) 1
38	}
39	}
40	close $in
41

42	set vdate(1.37) 2017-01-15
43	set vdate(1.36) 2016-10-24
44
45	# Do all versions from newest to oldest
46	#
47	foreach vers [lsort -decr -real [array names avers]] {
48	# set hr "../timeline?c=version-$vers;y=ci"
	@@ -57,11 +57,11 @@
57	puts $out "</b></center>"
58	puts $out "</td></tr>"
59	puts $out "<tr>"
60
61	foreach {prefix img desc} {
62	fossil-linux-x86 linux.gif {Linux 3.x x86}
63	fossil-macosx mac.gif {Mac 10.x x86}
64	fossil-openbsd-x86 openbsd.gif {OpenBSD 5.x x86}
65	fossil-w32 win32.gif {Windows}
66	fossil-src src.gif {Source Tarball}
67	} {
68

	--- www/mkdownload.tcl
	+++ www/mkdownload.tcl
	@@ -37,12 +37,12 @@
37	set avers($version) 1
38	}
39	}
40	close $in
41
42	set vdate(2.0) 2017-03-03
43	set vdate(1.37) 2017-01-15

44
45	# Do all versions from newest to oldest
46	#
47	foreach vers [lsort -decr -real [array names avers]] {
48	# set hr "../timeline?c=version-$vers;y=ci"
	@@ -57,11 +57,11 @@
57	puts $out "</b></center>"
58	puts $out "</td></tr>"
59	puts $out "<tr>"
60
61	foreach {prefix img desc} {
62	fossil-linux linux.gif {Linux 3.x x64}
63	fossil-macosx mac.gif {Mac 10.x x86}
64	fossil-openbsd-x86 openbsd.gif {OpenBSD 5.x x86}
65	fossil-w32 win32.gif {Windows}
66	fossil-src src.gif {Source Tarball}
67	} {
68

M www/mkindex.tcl

		--- www/mkindex.tcl
		+++ www/mkindex.tcl
		@@ -36,10 +36,11 @@
36	36	fiveminutes.wiki {Update and Running in 5 Minutes as a Single User}
37	37	foss-cklist.wiki {Checklist For Successful Open-Source Projects}
38	38	fossil-from-msvc.wiki {Integrating Fossil in the Microsoft Express 2010 IDE}
39	39	fossil-v-git.wiki {Fossil Versus Git}
40	40	hacker-howto.wiki {Hacker How-To}
	41	+ hashpolicy.wiki {Hash Policy: Choosing Between SHA1 and SHA3-256}
41	42	/help {Lists of Commands and Webpages}
42	43	hints.wiki {Fossil Tips And Usage Hints}
43	44	index.wiki {Home Page}
44	45	inout.wiki {Import And Export To And From Git}
45	46	makefile.wiki {The Fossil Build Process}
46	47

	--- www/mkindex.tcl
	+++ www/mkindex.tcl
	@@ -36,10 +36,11 @@
36	fiveminutes.wiki {Update and Running in 5 Minutes as a Single User}
37	foss-cklist.wiki {Checklist For Successful Open-Source Projects}
38	fossil-from-msvc.wiki {Integrating Fossil in the Microsoft Express 2010 IDE}
39	fossil-v-git.wiki {Fossil Versus Git}
40	hacker-howto.wiki {Hacker How-To}

41	/help {Lists of Commands and Webpages}
42	hints.wiki {Fossil Tips And Usage Hints}
43	index.wiki {Home Page}
44	inout.wiki {Import And Export To And From Git}
45	makefile.wiki {The Fossil Build Process}
46

	--- www/mkindex.tcl
	+++ www/mkindex.tcl
	@@ -36,10 +36,11 @@
36	fiveminutes.wiki {Update and Running in 5 Minutes as a Single User}
37	foss-cklist.wiki {Checklist For Successful Open-Source Projects}
38	fossil-from-msvc.wiki {Integrating Fossil in the Microsoft Express 2010 IDE}
39	fossil-v-git.wiki {Fossil Versus Git}
40	hacker-howto.wiki {Hacker How-To}
41	hashpolicy.wiki {Hash Policy: Choosing Between SHA1 and SHA3-256}
42	/help {Lists of Commands and Webpages}
43	hints.wiki {Fossil Tips And Usage Hints}
44	index.wiki {Home Page}
45	inout.wiki {Import And Export To And From Git}
46	makefile.wiki {The Fossil Build Process}
47

M www/permutedindex.html

		--- www/permutedindex.html
		+++ www/permutedindex.html
		@@ -29,10 +29,11 @@
29	29	<li><a href="blame.wiki">Annotate/Blame Algorithm Of Fossil — The</a></li>
30	30	<li><a href="customskin.md">Appearance of Web Pages — Theming: Customizing The</a></li>
31	31	<li><a href="faq.wiki">Asked Questions — Frequently</a></li>
32	32	<li><a href="password.wiki">Authentication — Password Management And</a></li>
33	33	<li><a href="whyusefossil.wiki"><b>Benefits Of Version Control</b></a></li>
	34	+<li><a href="hashpolicy.wiki">Between SHA1 and SHA3-256 — Hash Policy: Choosing</a></li>
34	35	<li><a href="antibot.wiki">Bots — Defense against Spiders and</a></li>
35	36	<li><a href="private.wiki">Branches — Creating, Syncing, and Deleting Private</a></li>
36	37	<li><a href="branching.wiki"><b>Branching, Forking, Merging, and Tagging</b></a></li>
37	38	<li><a href="bugtheory.wiki"><b>Bug Tracking In Fossil</b></a></li>
38	39	<li><a href="makefile.wiki">Build Process — The Fossil</a></li>
		@@ -43,10 +44,11 @@
43	44	<li><a href="checkin.wiki">Checklist — Check-in</a></li>
44	45	<li><a href="../test/release-checklist.wiki">Checklist — Pre-Release Testing</a></li>
45	46	<li><a href="foss-cklist.wiki"><b>Checklist For Successful Open-Source Projects</b></a></li>
46	47	<li><a href="selfcheck.wiki">Checks — Fossil Repository Integrity Self</a></li>
47	48	<li><a href="childprojects.wiki"><b>Child Projects</b></a></li>
	49	+<li><a href="hashpolicy.wiki">Choosing Between SHA1 and SHA3-256 — Hash Policy:</a></li>
48	50	<li><a href="contribute.wiki">Code or Documentation To The Fossil Project — Contributing</a></li>
49	51	<li><a href="style.wiki">Code Style Guidelines — Source</a></li>
50	52	<li><a href="../../../help">Commands and Webpages — Lists of</a></li>
51	53	<li><a href="build.wiki"><b>Compiling and Installing Fossil</b></a></li>
52	54	<li><a href="concepts.wiki">Concepts — Fossil Core</a></li>
		@@ -111,10 +113,11 @@
111	113	<li><a href="customgraph.md">Graph — Theming: Customizing the Timeline</a></li>
112	114	<li><a href="quickstart.wiki">Guide — Fossil Quick Start</a></li>
113	115	<li><a href="style.wiki">Guidelines — Source Code Style</a></li>
114	116	<li><a href="hacker-howto.wiki"><b>Hacker How-To</b></a></li>
115	117	<li><a href="adding_code.wiki"><b>Hacking Fossil</b></a></li>
	118	+<li><a href="hashpolicy.wiki"><b>Hash Policy: Choosing Between SHA1 and SHA3-256</b></a></li>
116	119	<li><a href="hints.wiki">Hints — Fossil Tips And Usage</a></li>
117	120	<li><a href="index.wiki"><b>Home Page</b></a></li>
118	121	<li><a href="selfhost.wiki">Hosting Repositories — Fossil Self</a></li>
119	122	<li><a href="aboutcgi.wiki"><b>How CGI Works In Fossil</b></a></li>
120	123	<li><a href="server.wiki"><b>How To Configure A Fossil Server</b></a></li>
		@@ -147,10 +150,11 @@
147	150	<li><a href="index.wiki">Page — Home</a></li>
148	151	<li><a href="customskin.md">Pages — Theming: Customizing The Appearance of Web</a></li>
149	152	<li><a href="password.wiki"><b>Password Management And Authentication</b></a></li>
150	153	<li><a href="quotes.wiki">People Are Saying About Fossil, Git, and DVCSes in General — Quotes: What</a></li>
151	154	<li><a href="stats.wiki"><b>Performance Statistics</b></a></li>
	155	+<li><a href="hashpolicy.wiki">Policy: Choosing Between SHA1 and SHA3-256 — Hash</a></li>
152	156	<li><a href="../test/release-checklist.wiki"><b>Pre-Release Testing Checklist</b></a></li>
153	157	<li><a href="pop.wiki"><b>Principles Of Operation</b></a></li>
154	158	<li><a href="private.wiki">Private Branches — Creating, Syncing, and Deleting</a></li>
155	159	<li><a href="makefile.wiki">Process — The Fossil Build</a></li>
156	160	<li><a href="contribute.wiki">Project — Contributing Code or Documentation To The Fossil</a></li>
		@@ -174,10 +178,12 @@
174	178	<li><a href="th1.md">Scripting Language — The TH1</a></li>
175	179	<li><a href="selfcheck.wiki">Self Checks — Fossil Repository Integrity</a></li>
176	180	<li><a href="selfhost.wiki">Self Hosting Repositories — Fossil</a></li>
177	181	<li><a href="server.wiki">Server — How To Configure A Fossil</a></li>
178	182	<li><a href="settings.wiki">Settings — Fossil</a></li>
	183	+<li><a href="hashpolicy.wiki">SHA1 and SHA3-256 — Hash Policy: Choosing Between</a></li>
	184	+<li><a href="hashpolicy.wiki">SHA3-256 — Hash Policy: Choosing Between SHA1 and</a></li>
179	185	<li><a href="shunning.wiki"><b>Shunning: Deleting Content From Fossil</b></a></li>
180	186	<li><a href="fiveminutes.wiki">Single User — Update and Running in 5 Minutes as a</a></li>
181	187	<li><a href="../../../sitemap"><b>Site Map</b></a></li>
182	188	<li><a href="style.wiki"><b>Source Code Style Guidelines</b></a></li>
183	189	<li><a href="antibot.wiki">Spiders and Bots — Defense against</a></li>
184	190

	--- www/permutedindex.html
	+++ www/permutedindex.html
	@@ -29,10 +29,11 @@
29	<li><a href="blame.wiki">Annotate/Blame Algorithm Of Fossil — The</a></li>
30	<li><a href="customskin.md">Appearance of Web Pages — Theming: Customizing The</a></li>
31	<li><a href="faq.wiki">Asked Questions — Frequently</a></li>
32	<li><a href="password.wiki">Authentication — Password Management And</a></li>
33	<li><a href="whyusefossil.wiki"><b>Benefits Of Version Control</b></a></li>

34	<li><a href="antibot.wiki">Bots — Defense against Spiders and</a></li>
35	<li><a href="private.wiki">Branches — Creating, Syncing, and Deleting Private</a></li>
36	<li><a href="branching.wiki"><b>Branching, Forking, Merging, and Tagging</b></a></li>
37	<li><a href="bugtheory.wiki"><b>Bug Tracking In Fossil</b></a></li>
38	<li><a href="makefile.wiki">Build Process — The Fossil</a></li>
	@@ -43,10 +44,11 @@
43	<li><a href="checkin.wiki">Checklist — Check-in</a></li>
44	<li><a href="../test/release-checklist.wiki">Checklist — Pre-Release Testing</a></li>
45	<li><a href="foss-cklist.wiki"><b>Checklist For Successful Open-Source Projects</b></a></li>
46	<li><a href="selfcheck.wiki">Checks — Fossil Repository Integrity Self</a></li>
47	<li><a href="childprojects.wiki"><b>Child Projects</b></a></li>

48	<li><a href="contribute.wiki">Code or Documentation To The Fossil Project — Contributing</a></li>
49	<li><a href="style.wiki">Code Style Guidelines — Source</a></li>
50	<li><a href="../../../help">Commands and Webpages — Lists of</a></li>
51	<li><a href="build.wiki"><b>Compiling and Installing Fossil</b></a></li>
52	<li><a href="concepts.wiki">Concepts — Fossil Core</a></li>
	@@ -111,10 +113,11 @@
111	<li><a href="customgraph.md">Graph — Theming: Customizing the Timeline</a></li>
112	<li><a href="quickstart.wiki">Guide — Fossil Quick Start</a></li>
113	<li><a href="style.wiki">Guidelines — Source Code Style</a></li>
114	<li><a href="hacker-howto.wiki"><b>Hacker How-To</b></a></li>
115	<li><a href="adding_code.wiki"><b>Hacking Fossil</b></a></li>

116	<li><a href="hints.wiki">Hints — Fossil Tips And Usage</a></li>
117	<li><a href="index.wiki"><b>Home Page</b></a></li>
118	<li><a href="selfhost.wiki">Hosting Repositories — Fossil Self</a></li>
119	<li><a href="aboutcgi.wiki"><b>How CGI Works In Fossil</b></a></li>
120	<li><a href="server.wiki"><b>How To Configure A Fossil Server</b></a></li>
	@@ -147,10 +150,11 @@
147	<li><a href="index.wiki">Page — Home</a></li>
148	<li><a href="customskin.md">Pages — Theming: Customizing The Appearance of Web</a></li>
149	<li><a href="password.wiki"><b>Password Management And Authentication</b></a></li>
150	<li><a href="quotes.wiki">People Are Saying About Fossil, Git, and DVCSes in General — Quotes: What</a></li>
151	<li><a href="stats.wiki"><b>Performance Statistics</b></a></li>

152	<li><a href="../test/release-checklist.wiki"><b>Pre-Release Testing Checklist</b></a></li>
153	<li><a href="pop.wiki"><b>Principles Of Operation</b></a></li>
154	<li><a href="private.wiki">Private Branches — Creating, Syncing, and Deleting</a></li>
155	<li><a href="makefile.wiki">Process — The Fossil Build</a></li>
156	<li><a href="contribute.wiki">Project — Contributing Code or Documentation To The Fossil</a></li>
	@@ -174,10 +178,12 @@
174	<li><a href="th1.md">Scripting Language — The TH1</a></li>
175	<li><a href="selfcheck.wiki">Self Checks — Fossil Repository Integrity</a></li>
176	<li><a href="selfhost.wiki">Self Hosting Repositories — Fossil</a></li>
177	<li><a href="server.wiki">Server — How To Configure A Fossil</a></li>
178	<li><a href="settings.wiki">Settings — Fossil</a></li>


179	<li><a href="shunning.wiki"><b>Shunning: Deleting Content From Fossil</b></a></li>
180	<li><a href="fiveminutes.wiki">Single User — Update and Running in 5 Minutes as a</a></li>
181	<li><a href="../../../sitemap"><b>Site Map</b></a></li>
182	<li><a href="style.wiki"><b>Source Code Style Guidelines</b></a></li>
183	<li><a href="antibot.wiki">Spiders and Bots — Defense against</a></li>
184

	--- www/permutedindex.html
	+++ www/permutedindex.html
	@@ -29,10 +29,11 @@
29	<li><a href="blame.wiki">Annotate/Blame Algorithm Of Fossil — The</a></li>
30	<li><a href="customskin.md">Appearance of Web Pages — Theming: Customizing The</a></li>
31	<li><a href="faq.wiki">Asked Questions — Frequently</a></li>
32	<li><a href="password.wiki">Authentication — Password Management And</a></li>
33	<li><a href="whyusefossil.wiki"><b>Benefits Of Version Control</b></a></li>
34	<li><a href="hashpolicy.wiki">Between SHA1 and SHA3-256 — Hash Policy: Choosing</a></li>
35	<li><a href="antibot.wiki">Bots — Defense against Spiders and</a></li>
36	<li><a href="private.wiki">Branches — Creating, Syncing, and Deleting Private</a></li>
37	<li><a href="branching.wiki"><b>Branching, Forking, Merging, and Tagging</b></a></li>
38	<li><a href="bugtheory.wiki"><b>Bug Tracking In Fossil</b></a></li>
39	<li><a href="makefile.wiki">Build Process — The Fossil</a></li>
	@@ -43,10 +44,11 @@
44	<li><a href="checkin.wiki">Checklist — Check-in</a></li>
45	<li><a href="../test/release-checklist.wiki">Checklist — Pre-Release Testing</a></li>
46	<li><a href="foss-cklist.wiki"><b>Checklist For Successful Open-Source Projects</b></a></li>
47	<li><a href="selfcheck.wiki">Checks — Fossil Repository Integrity Self</a></li>
48	<li><a href="childprojects.wiki"><b>Child Projects</b></a></li>
49	<li><a href="hashpolicy.wiki">Choosing Between SHA1 and SHA3-256 — Hash Policy:</a></li>
50	<li><a href="contribute.wiki">Code or Documentation To The Fossil Project — Contributing</a></li>
51	<li><a href="style.wiki">Code Style Guidelines — Source</a></li>
52	<li><a href="../../../help">Commands and Webpages — Lists of</a></li>
53	<li><a href="build.wiki"><b>Compiling and Installing Fossil</b></a></li>
54	<li><a href="concepts.wiki">Concepts — Fossil Core</a></li>
	@@ -111,10 +113,11 @@
113	<li><a href="customgraph.md">Graph — Theming: Customizing the Timeline</a></li>
114	<li><a href="quickstart.wiki">Guide — Fossil Quick Start</a></li>
115	<li><a href="style.wiki">Guidelines — Source Code Style</a></li>
116	<li><a href="hacker-howto.wiki"><b>Hacker How-To</b></a></li>
117	<li><a href="adding_code.wiki"><b>Hacking Fossil</b></a></li>
118	<li><a href="hashpolicy.wiki"><b>Hash Policy: Choosing Between SHA1 and SHA3-256</b></a></li>
119	<li><a href="hints.wiki">Hints — Fossil Tips And Usage</a></li>
120	<li><a href="index.wiki"><b>Home Page</b></a></li>
121	<li><a href="selfhost.wiki">Hosting Repositories — Fossil Self</a></li>
122	<li><a href="aboutcgi.wiki"><b>How CGI Works In Fossil</b></a></li>
123	<li><a href="server.wiki"><b>How To Configure A Fossil Server</b></a></li>
	@@ -147,10 +150,11 @@
150	<li><a href="index.wiki">Page — Home</a></li>
151	<li><a href="customskin.md">Pages — Theming: Customizing The Appearance of Web</a></li>
152	<li><a href="password.wiki"><b>Password Management And Authentication</b></a></li>
153	<li><a href="quotes.wiki">People Are Saying About Fossil, Git, and DVCSes in General — Quotes: What</a></li>
154	<li><a href="stats.wiki"><b>Performance Statistics</b></a></li>
155	<li><a href="hashpolicy.wiki">Policy: Choosing Between SHA1 and SHA3-256 — Hash</a></li>
156	<li><a href="../test/release-checklist.wiki"><b>Pre-Release Testing Checklist</b></a></li>
157	<li><a href="pop.wiki"><b>Principles Of Operation</b></a></li>
158	<li><a href="private.wiki">Private Branches — Creating, Syncing, and Deleting</a></li>
159	<li><a href="makefile.wiki">Process — The Fossil Build</a></li>
160	<li><a href="contribute.wiki">Project — Contributing Code or Documentation To The Fossil</a></li>
	@@ -174,10 +178,12 @@
178	<li><a href="th1.md">Scripting Language — The TH1</a></li>
179	<li><a href="selfcheck.wiki">Self Checks — Fossil Repository Integrity</a></li>
180	<li><a href="selfhost.wiki">Self Hosting Repositories — Fossil</a></li>
181	<li><a href="server.wiki">Server — How To Configure A Fossil</a></li>
182	<li><a href="settings.wiki">Settings — Fossil</a></li>
183	<li><a href="hashpolicy.wiki">SHA1 and SHA3-256 — Hash Policy: Choosing Between</a></li>
184	<li><a href="hashpolicy.wiki">SHA3-256 — Hash Policy: Choosing Between SHA1 and</a></li>
185	<li><a href="shunning.wiki"><b>Shunning: Deleting Content From Fossil</b></a></li>
186	<li><a href="fiveminutes.wiki">Single User — Update and Running in 5 Minutes as a</a></li>
187	<li><a href="../../../sitemap"><b>Site Map</b></a></li>
188	<li><a href="style.wiki"><b>Source Code Style Guidelines</b></a></li>
189	<li><a href="antibot.wiki">Spiders and Bots — Defense against</a></li>
190

M www/pop.wiki

+7 -6

		--- www/pop.wiki
		+++ www/pop.wiki
		@@ -25,17 +25,18 @@
25	25	The global state represents the content of the project.
26	26	The local state identifies the authorized users and
27	27	access policies for a particular repository.</p></li>
28	28
29	29	<li><p>The global state of a repository is an unordered
30		-collection of artifacts. Each artifact is named by
31		-its SHA1 hash encoded in lowercase hexadecimal.
	30	+collection of artifacts. Each artifact is named by a
	31	+cryptographic hash (SHA1 or SHA3-256) encoded in
	32	+lowercase hexadecimal.
32	33	In many contexts, the name can be
33	34	abbreviated to a unique prefix. A five- or six-character
34	35	prefix usually suffices to uniquely identify a file.</p></li>
35	36
36		-<li><p>Because artifacts are named by their SHA1 hash, all artifacts
	37	+<li><p>Because artifacts are named by a cryptographic hash, all artifacts
37	38	are immutable. Any change to the content of an artifact also
38	39	changes the hash that forms the artifacts name, thus
39	40	creating a new artifact. Both the old original version of the
40	41	artifact and the new change are preserved under different names.</p></li>
41	42
		@@ -42,16 +43,16 @@
42	43	<li><p>It is theoretically possible for two artifacts with different
43	44	content to share the same hash. But finding two such
44	45	artifacts is so incredibly difficult and unlikely that we
45	46	consider it to be an impossibility.</p></li>
46	47
47		-<li><p>The signature of an artifact is the SHA1 hash of the
	48	+<li><p>The signature of an artifact is the cryptographic hash of the
48	49	artifact itself, exactly as it would appear in a disk file. No prefix
49	50	or meta-information about the artifact is added before computing
50	51	the hash. So you can
51		-always find the SHA1 signature of a file by using the
52		-"sha1sum" command-line utility.</p></li>
	52	+always find the signature of a file by using the
	53	+"sha1sum" or "sha3sum" or similar command-line utilities.</p></li>
53	54
54	55	<li><p>The artifacts that comprise the global state of a repository
55	56	are the complete global state of that repository. The SQLite
56	57	database that holds the repository contains additional information
57	58	about linkages between artifacts, but all of that added information
58	59

	--- www/pop.wiki
	+++ www/pop.wiki
	@@ -25,17 +25,18 @@
25	The global state represents the content of the project.
26	The local state identifies the authorized users and
27	access policies for a particular repository.</p></li>
28
29	<li><p>The global state of a repository is an unordered
30	collection of artifacts. Each artifact is named by
31	its SHA1 hash encoded in lowercase hexadecimal.

32	In many contexts, the name can be
33	abbreviated to a unique prefix. A five- or six-character
34	prefix usually suffices to uniquely identify a file.</p></li>
35
36	<li><p>Because artifacts are named by their SHA1 hash, all artifacts
37	are immutable. Any change to the content of an artifact also
38	changes the hash that forms the artifacts name, thus
39	creating a new artifact. Both the old original version of the
40	artifact and the new change are preserved under different names.</p></li>
41
	@@ -42,16 +43,16 @@
42	<li><p>It is theoretically possible for two artifacts with different
43	content to share the same hash. But finding two such
44	artifacts is so incredibly difficult and unlikely that we
45	consider it to be an impossibility.</p></li>
46
47	<li><p>The signature of an artifact is the SHA1 hash of the
48	artifact itself, exactly as it would appear in a disk file. No prefix
49	or meta-information about the artifact is added before computing
50	the hash. So you can
51	always find the SHA1 signature of a file by using the
52	"sha1sum" command-line utility.</p></li>
53
54	<li><p>The artifacts that comprise the global state of a repository
55	are the complete global state of that repository. The SQLite
56	database that holds the repository contains additional information
57	about linkages between artifacts, but all of that added information
58

	--- www/pop.wiki
	+++ www/pop.wiki
	@@ -25,17 +25,18 @@
25	The global state represents the content of the project.
26	The local state identifies the authorized users and
27	access policies for a particular repository.</p></li>
28
29	<li><p>The global state of a repository is an unordered
30	collection of artifacts. Each artifact is named by a
31	cryptographic hash (SHA1 or SHA3-256) encoded in
32	lowercase hexadecimal.
33	In many contexts, the name can be
34	abbreviated to a unique prefix. A five- or six-character
35	prefix usually suffices to uniquely identify a file.</p></li>
36
37	<li><p>Because artifacts are named by a cryptographic hash, all artifacts
38	are immutable. Any change to the content of an artifact also
39	changes the hash that forms the artifacts name, thus
40	creating a new artifact. Both the old original version of the
41	artifact and the new change are preserved under different names.</p></li>
42
	@@ -42,16 +43,16 @@
43	<li><p>It is theoretically possible for two artifacts with different
44	content to share the same hash. But finding two such
45	artifacts is so incredibly difficult and unlikely that we
46	consider it to be an impossibility.</p></li>
47
48	<li><p>The signature of an artifact is the cryptographic hash of the
49	artifact itself, exactly as it would appear in a disk file. No prefix
50	or meta-information about the artifact is added before computing
51	the hash. So you can
52	always find the signature of a file by using the
53	"sha1sum" or "sha3sum" or similar command-line utilities.</p></li>
54
55	<li><p>The artifacts that comprise the global state of a repository
56	are the complete global state of that repository. The SQLite
57	database that holds the repository contains additional information
58	about linkages between artifacts, but all of that added information
59

M www/selfcheck.wiki

+10 -9

		--- www/selfcheck.wiki
		+++ www/selfcheck.wiki
		@@ -51,14 +51,14 @@
51	51	commit. So during the course of check-in (or other repository
52	52	operation) many different files
53	53	in the repository might be modified. Some files are simply
54	54	compressed. Other files are delta encoded and then compressed.
55	55	While all this is going on, fossil makes a record of every file
56		-that is encoded and the SHA1 hash of the original content of that
	56	+and the SHA1 or SHA3-256 hash of the original content of that
57	57	file. Then just before transaction commit, fossil re-extracts
58		-the original content of all files that were written, computes
59		-the SHA1 checksum again, and verifies that the checksums match.
	58	+the original content of all files that were written, recomputes
	59	+the hash, and verifies that the recomputed hash still matches.
60	60	If anything does not match up, an error
61	61	message is printed and the transaction rolls back.
62	62
63	63	So, in other words, fossil always checks to make sure it can
64	64	re-extract a file before it commits a change to that file.
		@@ -73,20 +73,20 @@
73	73	and of all other files in the manifest. Prior to any check-in
74	74	commit, these checksums are verified to ensure that the check-in
75	75	agrees exactly with what is on disk. Similarly,
76	76	the repository checksum is verified after a checkout to make
77	77	sure that the entire repository was checked out correctly.
78		-Note that these added checks use a different hash (MD5 instead
79		-of SHA1) in order to avoid common-mode failures in the hash
	78	+Note that these added checks use a different hash algorithm (MD5)
	79	+in order to avoid common-mode failures in the hash
80	80	algorithm implementation.
81	81
82	82
83		-<h2>Checksums On Control Artifacts And Deltas</h2>
	83	+<h2>Checksums On Structural Artifacts And Deltas</h2>
84	84
85		-Every [./fileformat.wiki \| control artifact] in a fossil repository
	85	+Every [./fileformat.wiki \| structural artifact] in a fossil repository
86	86	contains a "Z-card" bearing an MD5 checksum over the rest of the
87		-artifact. Any mismatch causes the control artifact to be ignored.
	87	+artifact. Any mismatch causes the structural artifact to be ignored.
88	88
89	89	The [./delta_format.wiki \| file delta format] includes a 32-bit
90	90	checksum of the target file. Whenever a file is reconstructed from
91	91	a delta, that checksum is verified to make sure the reconstruction
92	92	was done correctly.
		@@ -103,6 +103,7 @@
103	103	reliability is more important than raw speed. The developers of
104	104	fossil see no merit in getting the wrong answer quickly.
105	105
106	106	Fossil may not be the fastest versioning system, but it is "fast enough".
107	107	Fossil runs quickly enough to stay out of the developers way.
108		-Most operations complete in under a second.
	108	+Most operations complete in milliseconds, faster that you can press
	109	+the "Enter" key.
109	110

	--- www/selfcheck.wiki
	+++ www/selfcheck.wiki
	@@ -51,14 +51,14 @@
51	commit. So during the course of check-in (or other repository
52	operation) many different files
53	in the repository might be modified. Some files are simply
54	compressed. Other files are delta encoded and then compressed.
55	While all this is going on, fossil makes a record of every file
56	that is encoded and the SHA1 hash of the original content of that
57	file. Then just before transaction commit, fossil re-extracts
58	the original content of all files that were written, computes
59	the SHA1 checksum again, and verifies that the checksums match.
60	If anything does not match up, an error
61	message is printed and the transaction rolls back.
62
63	So, in other words, fossil always checks to make sure it can
64	re-extract a file before it commits a change to that file.
	@@ -73,20 +73,20 @@
73	and of all other files in the manifest. Prior to any check-in
74	commit, these checksums are verified to ensure that the check-in
75	agrees exactly with what is on disk. Similarly,
76	the repository checksum is verified after a checkout to make
77	sure that the entire repository was checked out correctly.
78	Note that these added checks use a different hash (MD5 instead
79	of SHA1) in order to avoid common-mode failures in the hash
80	algorithm implementation.
81
82
83	<h2>Checksums On Control Artifacts And Deltas</h2>
84
85	Every [./fileformat.wiki \| control artifact] in a fossil repository
86	contains a "Z-card" bearing an MD5 checksum over the rest of the
87	artifact. Any mismatch causes the control artifact to be ignored.
88
89	The [./delta_format.wiki \| file delta format] includes a 32-bit
90	checksum of the target file. Whenever a file is reconstructed from
91	a delta, that checksum is verified to make sure the reconstruction
92	was done correctly.
	@@ -103,6 +103,7 @@
103	reliability is more important than raw speed. The developers of
104	fossil see no merit in getting the wrong answer quickly.
105
106	Fossil may not be the fastest versioning system, but it is "fast enough".
107	Fossil runs quickly enough to stay out of the developers way.
108	Most operations complete in under a second.

109

	--- www/selfcheck.wiki
	+++ www/selfcheck.wiki
	@@ -51,14 +51,14 @@
51	commit. So during the course of check-in (or other repository
52	operation) many different files
53	in the repository might be modified. Some files are simply
54	compressed. Other files are delta encoded and then compressed.
55	While all this is going on, fossil makes a record of every file
56	and the SHA1 or SHA3-256 hash of the original content of that
57	file. Then just before transaction commit, fossil re-extracts
58	the original content of all files that were written, recomputes
59	the hash, and verifies that the recomputed hash still matches.
60	If anything does not match up, an error
61	message is printed and the transaction rolls back.
62
63	So, in other words, fossil always checks to make sure it can
64	re-extract a file before it commits a change to that file.
	@@ -73,20 +73,20 @@
73	and of all other files in the manifest. Prior to any check-in
74	commit, these checksums are verified to ensure that the check-in
75	agrees exactly with what is on disk. Similarly,
76	the repository checksum is verified after a checkout to make
77	sure that the entire repository was checked out correctly.
78	Note that these added checks use a different hash algorithm (MD5)
79	in order to avoid common-mode failures in the hash
80	algorithm implementation.
81
82
83	<h2>Checksums On Structural Artifacts And Deltas</h2>
84
85	Every [./fileformat.wiki \| structural artifact] in a fossil repository
86	contains a "Z-card" bearing an MD5 checksum over the rest of the
87	artifact. Any mismatch causes the structural artifact to be ignored.
88
89	The [./delta_format.wiki \| file delta format] includes a 32-bit
90	checksum of the target file. Whenever a file is reconstructed from
91	a delta, that checksum is verified to make sure the reconstruction
92	was done correctly.
	@@ -103,6 +103,7 @@
103	reliability is more important than raw speed. The developers of
104	fossil see no merit in getting the wrong answer quickly.
105
106	Fossil may not be the fastest versioning system, but it is "fast enough".
107	Fossil runs quickly enough to stay out of the developers way.
108	Most operations complete in milliseconds, faster that you can press
109	the "Enter" key.
110

M www/shunning.wiki

+1 -1

		--- www/shunning.wiki
		+++ www/shunning.wiki
		@@ -23,11 +23,11 @@
23	23	<h2>Shunning</h2>
24	24
25	25	Fossil provides a mechanism called "shunning" for removing content from
26	26	a repository.
27	27
28		-Every Fossil repository maintains a list of the SHA1 hash names of
	28	+Every Fossil repository maintains a list of the hash names of
29	29	"shunned" artifacts.
30	30	Fossil will refuse to push or pull any shunned artifact.
31	31	Furthermore, all shunned artifacts (but not the shunning list
32	32	itself) are removed from the
33	33	repository whenever the repository is reconstructed using the
34	34

	--- www/shunning.wiki
	+++ www/shunning.wiki
	@@ -23,11 +23,11 @@
23	<h2>Shunning</h2>
24
25	Fossil provides a mechanism called "shunning" for removing content from
26	a repository.
27
28	Every Fossil repository maintains a list of the SHA1 hash names of
29	"shunned" artifacts.
30	Fossil will refuse to push or pull any shunned artifact.
31	Furthermore, all shunned artifacts (but not the shunning list
32	itself) are removed from the
33	repository whenever the repository is reconstructed using the
34

	--- www/shunning.wiki
	+++ www/shunning.wiki
	@@ -23,11 +23,11 @@
23	<h2>Shunning</h2>
24
25	Fossil provides a mechanism called "shunning" for removing content from
26	a repository.
27
28	Every Fossil repository maintains a list of the hash names of
29	"shunned" artifacts.
30	Fossil will refuse to push or pull any shunned artifact.
31	Furthermore, all shunned artifacts (but not the shunning list
32	itself) are removed from the
33	repository whenever the repository is reconstructed using the
34

M www/sync.wiki

+14 -14

		--- www/sync.wiki
		+++ www/sync.wiki
		@@ -4,23 +4,23 @@
4	4	content between two Fossil repositories.</p>
5	5
6	6	<h2>1.0 Overview</h2>
7	7
8	8	<p>The global state of a fossil repository consists of an unordered
9		-collection of artifacts. Each artifact is identified by its SHA1 hash
10		-expressed as a 40-character lower-case hexadecimal string.
	9	+collection of artifacts. Each artifact is identified by a cryptographic
	10	+hash of its content, expressed as a lower-case hexadecimal string.
11	11	Synchronization is the process of sharing artifacts between
12	12	servers so that all servers have copies of all artifacts. Because
13	13	artifacts are unordered, the order in which artifacts are received
14		-at a server is inconsequential. It is assumed that the SHA1 hashes
15		-of artifacts are unique - that every artifact has a different SHA1 hash.
	14	+at a server is inconsequential. It is assumed that the hash names
	15	+of artifacts are unique - that every artifact has a different hash.
16	16	To a first approximation, synchronization proceeds by sharing lists
17		-SHA1 hashes of available artifacts, then sharing those artifacts that
18		-are not found on one side or the other of the connection. In practice,
19		-a repository might contain millions of artifacts. The list of
20		-SHA1 hashes for this many artifacts can be large. So optimizations are
21		-employed that usually reduce the number of SHA1 hashes that need to be
	17	+hash values for available artifacts, then sharing the content of artifacts
	18	+whose names are missing from one side or the other of the connection.
	19	+In practice, a repository might contain millions of artifacts. The list of
	20	+hash names for this many artifacts can be large. So optimizations are
	21	+employed that usually reduce the number of hashes that need to be
22	22	shared to a few hundred.</p>
23	23
24	24	<p>Each repository also has local state. The local state determines
25	25	the web-page formatting preferences, authorized users, ticket formats,
26	26	and similar information that varies from one repository to another.
		@@ -198,11 +198,11 @@
198	198	terminates the file card.
199	199	</p>
200	200
201	201	<p>The first argument of a file card is the ID of the artifact that
202	202	is being transferred. The artifact ID is the lower-case hexadecimal
203		-representation of the SHA1 hash of the artifact.
	203	+representation of the name hash for the artifact.
204	204	The last argument of the file card is the number of bytes of
205	205	payload that immediately follow the file card. If the file
206	206	card has only two arguments, that means the payload is the
207	207	complete content of the artifact. If the file card has three
208	208	arguments, then the payload is a delta and second argument is
		@@ -231,11 +231,11 @@
231	231	<b>cfile</b> <i>artifact-id delta-artifact-id usize csize</i> <b>\n</b> <i>content</i><br>
232	232	</blockquote>
233	233
234	234	<p>The first argument of the cfile card is the ID of the artifact that
235	235	is being transferred. The artifact ID is the lower-case hexadecimal
236		-representation of the SHA1 hash of the artifact. The second argument of
	236	+representation of the name hash for the artifact. The second argument of
237	237	the cfile card is the original size in bytes of the artifact. The last
238	238	argument of the cfile card is the number of compressed bytes of payload
239	239	that immediately follow the cfile card. If the cfile card has only
240	240	three arguments, that means the payload is the complete content of the
241	241	artifact. If the cfile card has four arguments, then the payload is a
		@@ -271,11 +271,11 @@
271	271	<b>uvfile</b> <i>name mtime hash size flags</i> <b>\n</b> <i>content</i>
272	272	</blockquote>
273	273
274	274	<p>The <i>name</i> field is the name of the unversioned file. The
275	275	<i>mtime</i> is the last modification time of the file in seconds
276		-since 1970. The <i>hash</i> field is the SHA1 hash of the content
	276	+since 1970. The <i>hash</i> field is the hash of the content
277	277	for the unversioned file, or "<b>-</b>" for deleted content.
278	278	The <i>size</i> field is the (uncompressed) size of the content
279	279	in bytes. The <i>flags</i> field is an integer which is interpreted
280	280	as an array of bits. The 0x0004 bit of <i>flags</i> indicates that
281	281	the <i>content</i> is to be omitted. The content might be omitted if
		@@ -409,12 +409,12 @@
409	409	</blockquote>
410	410
411	411	<p>The <i>name</i> argument is the name of an unversioned file.
412	412	The <i>mtime</i> is the last modification time of the unversioned file
413	413	in seconds since 1970.
414		-The <i>hash</i> is the SHA1 hash of the unversioned file content, or
415		-"<b>-</b>" if the file has been deleted.
	414	+The <i>hash</i> is the SHA1 or SHA3-256 hash of the unversioned file
	415	+content, or "<b>-</b>" if the file has been deleted.
416	416	The <i>size</i> is the uncompressed size of the file in bytes.
417	417
418	418	<p>When the server sees a "pragma uv-hash" card for which the hash
419	419	does not match, it sends uvigot cards for every unversioned file that it
420	420	holds. The client will use this information to figure out which
421	421

	--- www/sync.wiki
	+++ www/sync.wiki
	@@ -4,23 +4,23 @@
4	content between two Fossil repositories.</p>
5
6	<h2>1.0 Overview</h2>
7
8	<p>The global state of a fossil repository consists of an unordered
9	collection of artifacts. Each artifact is identified by its SHA1 hash
10	expressed as a 40-character lower-case hexadecimal string.
11	Synchronization is the process of sharing artifacts between
12	servers so that all servers have copies of all artifacts. Because
13	artifacts are unordered, the order in which artifacts are received
14	at a server is inconsequential. It is assumed that the SHA1 hashes
15	of artifacts are unique - that every artifact has a different SHA1 hash.
16	To a first approximation, synchronization proceeds by sharing lists
17	SHA1 hashes of available artifacts, then sharing those artifacts that
18	are not found on one side or the other of the connection. In practice,
19	a repository might contain millions of artifacts. The list of
20	SHA1 hashes for this many artifacts can be large. So optimizations are
21	employed that usually reduce the number of SHA1 hashes that need to be
22	shared to a few hundred.</p>
23
24	<p>Each repository also has local state. The local state determines
25	the web-page formatting preferences, authorized users, ticket formats,
26	and similar information that varies from one repository to another.
	@@ -198,11 +198,11 @@
198	terminates the file card.
199	</p>
200
201	<p>The first argument of a file card is the ID of the artifact that
202	is being transferred. The artifact ID is the lower-case hexadecimal
203	representation of the SHA1 hash of the artifact.
204	The last argument of the file card is the number of bytes of
205	payload that immediately follow the file card. If the file
206	card has only two arguments, that means the payload is the
207	complete content of the artifact. If the file card has three
208	arguments, then the payload is a delta and second argument is
	@@ -231,11 +231,11 @@
231	<b>cfile</b> <i>artifact-id delta-artifact-id usize csize</i> <b>\n</b> <i>content</i><br>
232	</blockquote>
233
234	<p>The first argument of the cfile card is the ID of the artifact that
235	is being transferred. The artifact ID is the lower-case hexadecimal
236	representation of the SHA1 hash of the artifact. The second argument of
237	the cfile card is the original size in bytes of the artifact. The last
238	argument of the cfile card is the number of compressed bytes of payload
239	that immediately follow the cfile card. If the cfile card has only
240	three arguments, that means the payload is the complete content of the
241	artifact. If the cfile card has four arguments, then the payload is a
	@@ -271,11 +271,11 @@
271	<b>uvfile</b> <i>name mtime hash size flags</i> <b>\n</b> <i>content</i>
272	</blockquote>
273
274	<p>The <i>name</i> field is the name of the unversioned file. The
275	<i>mtime</i> is the last modification time of the file in seconds
276	since 1970. The <i>hash</i> field is the SHA1 hash of the content
277	for the unversioned file, or "<b>-</b>" for deleted content.
278	The <i>size</i> field is the (uncompressed) size of the content
279	in bytes. The <i>flags</i> field is an integer which is interpreted
280	as an array of bits. The 0x0004 bit of <i>flags</i> indicates that
281	the <i>content</i> is to be omitted. The content might be omitted if
	@@ -409,12 +409,12 @@
409	</blockquote>
410
411	<p>The <i>name</i> argument is the name of an unversioned file.
412	The <i>mtime</i> is the last modification time of the unversioned file
413	in seconds since 1970.
414	The <i>hash</i> is the SHA1 hash of the unversioned file content, or
415	"<b>-</b>" if the file has been deleted.
416	The <i>size</i> is the uncompressed size of the file in bytes.
417
418	<p>When the server sees a "pragma uv-hash" card for which the hash
419	does not match, it sends uvigot cards for every unversioned file that it
420	holds. The client will use this information to figure out which
421

	--- www/sync.wiki
	+++ www/sync.wiki
	@@ -4,23 +4,23 @@
4	content between two Fossil repositories.</p>
5
6	<h2>1.0 Overview</h2>
7
8	<p>The global state of a fossil repository consists of an unordered
9	collection of artifacts. Each artifact is identified by a cryptographic
10	hash of its content, expressed as a lower-case hexadecimal string.
11	Synchronization is the process of sharing artifacts between
12	servers so that all servers have copies of all artifacts. Because
13	artifacts are unordered, the order in which artifacts are received
14	at a server is inconsequential. It is assumed that the hash names
15	of artifacts are unique - that every artifact has a different hash.
16	To a first approximation, synchronization proceeds by sharing lists
17	hash values for available artifacts, then sharing the content of artifacts
18	whose names are missing from one side or the other of the connection.
19	In practice, a repository might contain millions of artifacts. The list of
20	hash names for this many artifacts can be large. So optimizations are
21	employed that usually reduce the number of hashes that need to be
22	shared to a few hundred.</p>
23
24	<p>Each repository also has local state. The local state determines
25	the web-page formatting preferences, authorized users, ticket formats,
26	and similar information that varies from one repository to another.
	@@ -198,11 +198,11 @@
198	terminates the file card.
199	</p>
200
201	<p>The first argument of a file card is the ID of the artifact that
202	is being transferred. The artifact ID is the lower-case hexadecimal
203	representation of the name hash for the artifact.
204	The last argument of the file card is the number of bytes of
205	payload that immediately follow the file card. If the file
206	card has only two arguments, that means the payload is the
207	complete content of the artifact. If the file card has three
208	arguments, then the payload is a delta and second argument is
	@@ -231,11 +231,11 @@
231	<b>cfile</b> <i>artifact-id delta-artifact-id usize csize</i> <b>\n</b> <i>content</i><br>
232	</blockquote>
233
234	<p>The first argument of the cfile card is the ID of the artifact that
235	is being transferred. The artifact ID is the lower-case hexadecimal
236	representation of the name hash for the artifact. The second argument of
237	the cfile card is the original size in bytes of the artifact. The last
238	argument of the cfile card is the number of compressed bytes of payload
239	that immediately follow the cfile card. If the cfile card has only
240	three arguments, that means the payload is the complete content of the
241	artifact. If the cfile card has four arguments, then the payload is a
	@@ -271,11 +271,11 @@
271	<b>uvfile</b> <i>name mtime hash size flags</i> <b>\n</b> <i>content</i>
272	</blockquote>
273
274	<p>The <i>name</i> field is the name of the unversioned file. The
275	<i>mtime</i> is the last modification time of the file in seconds
276	since 1970. The <i>hash</i> field is the hash of the content
277	for the unversioned file, or "<b>-</b>" for deleted content.
278	The <i>size</i> field is the (uncompressed) size of the content
279	in bytes. The <i>flags</i> field is an integer which is interpreted
280	as an array of bits. The 0x0004 bit of <i>flags</i> indicates that
281	the <i>content</i> is to be omitted. The content might be omitted if
	@@ -409,12 +409,12 @@
409	</blockquote>
410
411	<p>The <i>name</i> argument is the name of an unversioned file.
412	The <i>mtime</i> is the last modification time of the unversioned file
413	in seconds since 1970.
414	The <i>hash</i> is the SHA1 or SHA3-256 hash of the unversioned file
415	content, or "<b>-</b>" if the file has been deleted.
416	The <i>size</i> is the uncompressed size of the file in bytes.
417
418	<p>When the server sees a "pragma uv-hash" card for which the hash
419	does not match, it sends uvigot cards for every unversioned file that it
420	holds. The client will use this information to figure out which
421

M www/tech_overview.wiki

+5 -4

		--- www/tech_overview.wiki
		+++ www/tech_overview.wiki
		@@ -6,12 +6,12 @@
6	6	<h2>1.0 Introduction</h2>
7	7
8	8	At its lowest level, a Fossil repository consists of an unordered set
9	9	of immutable "artifacts". You might think of these artifacts as "files",
10	10	since in many cases the artifacts are exactly that.
11		-But other "control artifacts"
12		-are also included in the mix. These control artifacts define the relationships
	11	+But other "structural artifacts" are also included in the mix.
	12	+These structural artifacts define the relationships
13	13	between artifacts - which files go together to form a particular
14	14	version of the project, who checked in that version and when, what was
15	15	the check-in comment, what wiki pages are included with the project, what
16	16	are the edit histories of each wiki page, what bug reports or tickets are
17	17	included, who contributed to the evolution of each ticket, and so forth.
		@@ -173,11 +173,12 @@
173	173	the [/help/deconstruct \| fossil deconstruct]
174	174	command. Individual artifacts can be extracted using the
175	175	[/help/artifact \| fossil artifact] command.
176	176	When accessing the repository database using raw SQL and the
177	177	[/help/sqlite3 \| fossil sql] command, the extension function
178		-"<tt>content()</tt>" with a single argument which is the SHA1 hash
	178	+"<tt>content()</tt>" with a single argument which is the SHA1 or
	179	+SHA3-256 hash
179	180	of an artifact will return the complete undeleted and uncompressed
180	181	content of that artifact.
181	182
182	183	Going the other way, the [/help/reconstruct \| fossil reconstruct]
183	184	command will scan a directory hierarchy and add all files found to
		@@ -278,11 +279,11 @@
278	279	project - is intended to be an append-only database. In other words,
279	280	new artifacts can be added but artifacts can never be removed. But
280	281	it sometimes happens that inappropriate content is mistakenly or
281	282	maliciously added to a repository. The only way to get rid of
282	283	the undesired content is to [./shunning.wiki \| "shun"] it.
283		-The "shun" table in the repository database records the SHA1 hash of
	284	+The "shun" table in the repository database records the hash values for
284	285	all shunned artifacts.
285	286
286	287	The shun table can be pushed or pulled using
287	288	the [/help/config \| fossil config] command with the "shun" AREA argument.
288	289	The shun table is also copied during a [/help/clone \| clone].
289	290

	--- www/tech_overview.wiki
	+++ www/tech_overview.wiki
	@@ -6,12 +6,12 @@
6	<h2>1.0 Introduction</h2>
7
8	At its lowest level, a Fossil repository consists of an unordered set
9	of immutable "artifacts". You might think of these artifacts as "files",
10	since in many cases the artifacts are exactly that.
11	But other "control artifacts"
12	are also included in the mix. These control artifacts define the relationships
13	between artifacts - which files go together to form a particular
14	version of the project, who checked in that version and when, what was
15	the check-in comment, what wiki pages are included with the project, what
16	are the edit histories of each wiki page, what bug reports or tickets are
17	included, who contributed to the evolution of each ticket, and so forth.
	@@ -173,11 +173,12 @@
173	the [/help/deconstruct \| fossil deconstruct]
174	command. Individual artifacts can be extracted using the
175	[/help/artifact \| fossil artifact] command.
176	When accessing the repository database using raw SQL and the
177	[/help/sqlite3 \| fossil sql] command, the extension function
178	"<tt>content()</tt>" with a single argument which is the SHA1 hash

179	of an artifact will return the complete undeleted and uncompressed
180	content of that artifact.
181
182	Going the other way, the [/help/reconstruct \| fossil reconstruct]
183	command will scan a directory hierarchy and add all files found to
	@@ -278,11 +279,11 @@
278	project - is intended to be an append-only database. In other words,
279	new artifacts can be added but artifacts can never be removed. But
280	it sometimes happens that inappropriate content is mistakenly or
281	maliciously added to a repository. The only way to get rid of
282	the undesired content is to [./shunning.wiki \| "shun"] it.
283	The "shun" table in the repository database records the SHA1 hash of
284	all shunned artifacts.
285
286	The shun table can be pushed or pulled using
287	the [/help/config \| fossil config] command with the "shun" AREA argument.
288	The shun table is also copied during a [/help/clone \| clone].
289

	--- www/tech_overview.wiki
	+++ www/tech_overview.wiki
	@@ -6,12 +6,12 @@
6	<h2>1.0 Introduction</h2>
7
8	At its lowest level, a Fossil repository consists of an unordered set
9	of immutable "artifacts". You might think of these artifacts as "files",
10	since in many cases the artifacts are exactly that.
11	But other "structural artifacts" are also included in the mix.
12	These structural artifacts define the relationships
13	between artifacts - which files go together to form a particular
14	version of the project, who checked in that version and when, what was
15	the check-in comment, what wiki pages are included with the project, what
16	are the edit histories of each wiki page, what bug reports or tickets are
17	included, who contributed to the evolution of each ticket, and so forth.
	@@ -173,11 +173,12 @@
173	the [/help/deconstruct \| fossil deconstruct]
174	command. Individual artifacts can be extracted using the
175	[/help/artifact \| fossil artifact] command.
176	When accessing the repository database using raw SQL and the
177	[/help/sqlite3 \| fossil sql] command, the extension function
178	"<tt>content()</tt>" with a single argument which is the SHA1 or
179	SHA3-256 hash
180	of an artifact will return the complete undeleted and uncompressed
181	content of that artifact.
182
183	Going the other way, the [/help/reconstruct \| fossil reconstruct]
184	command will scan a directory hierarchy and add all files found to
	@@ -278,11 +279,11 @@
279	project - is intended to be an append-only database. In other words,
280	new artifacts can be added but artifacts can never be removed. But
281	it sometimes happens that inappropriate content is mistakenly or
282	maliciously added to a repository. The only way to get rid of
283	the undesired content is to [./shunning.wiki \| "shun"] it.
284	The "shun" table in the repository database records the hash values for
285	all shunned artifacts.
286
287	The shun table can be pushed or pulled using
288	the [/help/config \| fossil config] command with the "shun" AREA argument.
289	The shun table is also copied during a [/help/clone \| clone].
290

M www/theory1.wiki

+2 -1

		--- www/theory1.wiki
		+++ www/theory1.wiki
		@@ -38,11 +38,12 @@
38	38	been checked into the Fossil repository. Call these "content artifacts".
39	39	Other artifacts, known as
40	40	"control artifacts", contain ASCII text in a particular format that
41	41	defines relationships between other artifacts, such as which
42	42	content artifacts that go together to form a particular version of the
43		-project. Each artifact is named by its SHA1 hash and is thus immutable.
	43	+project. Each artifact is named by its SHA1 or SHA3-256 hash and is
	44	+thus immutable.
44	45	Artifacts can be added to the database but not removed (if we ignore
45	46	the exceptional case of [./shunning.wiki \| shunning].) Repositories
46	47	synchronize by computing the union of their artifact sets. SQL and
47	48	relation theory play no role in any of this.
48	49
49	50

	--- www/theory1.wiki
	+++ www/theory1.wiki
	@@ -38,11 +38,12 @@
38	been checked into the Fossil repository. Call these "content artifacts".
39	Other artifacts, known as
40	"control artifacts", contain ASCII text in a particular format that
41	defines relationships between other artifacts, such as which
42	content artifacts that go together to form a particular version of the
43	project. Each artifact is named by its SHA1 hash and is thus immutable.

44	Artifacts can be added to the database but not removed (if we ignore
45	the exceptional case of [./shunning.wiki \| shunning].) Repositories
46	synchronize by computing the union of their artifact sets. SQL and
47	relation theory play no role in any of this.
48
49

	--- www/theory1.wiki
	+++ www/theory1.wiki
	@@ -38,11 +38,12 @@
38	been checked into the Fossil repository. Call these "content artifacts".
39	Other artifacts, known as
40	"control artifacts", contain ASCII text in a particular format that
41	defines relationships between other artifacts, such as which
42	content artifacts that go together to form a particular version of the
43	project. Each artifact is named by its SHA1 or SHA3-256 hash and is
44	thus immutable.
45	Artifacts can be added to the database but not removed (if we ignore
46	the exceptional case of [./shunning.wiki \| shunning].) Repositories
47	synchronize by computing the union of their artifact sets. SQL and
48	relation theory play no role in any of this.
49
50

M www/webpage-ex.md

+1 -1

		--- www/webpage-ex.md
		+++ www/webpage-ex.md
		@@ -123,10 +123,10 @@
123	123	href='$ROOT/bigbloblist'>Example</a>
124	124	The largest objects in the repository.
125	125
126	126	* <a target='_blank' class='exbtn'
127	127	href='$ROOT/hash-collisions'>Example</a>
128		- SHA1 prefix collisions
	128	+ Hash prefix collisions
129	129
130	130	* <a target='_blank' class='exbtn'
131	131	href='$ROOT/sitemap'>Example</a>
132	132	The "sitemap" containing links to many other pages
133	133

	--- www/webpage-ex.md
	+++ www/webpage-ex.md
	@@ -123,10 +123,10 @@
123	href='$ROOT/bigbloblist'>Example</a>
124	The largest objects in the repository.
125
126	* <a target='_blank' class='exbtn'
127	href='$ROOT/hash-collisions'>Example</a>
128	SHA1 prefix collisions
129
130	* <a target='_blank' class='exbtn'
131	href='$ROOT/sitemap'>Example</a>
132	The "sitemap" containing links to many other pages
133

	--- www/webpage-ex.md
	+++ www/webpage-ex.md
	@@ -123,10 +123,10 @@
123	href='$ROOT/bigbloblist'>Example</a>
124	The largest objects in the repository.
125
126	* <a target='_blank' class='exbtn'
127	href='$ROOT/hash-collisions'>Example</a>
128	Hash prefix collisions
129
130	* <a target='_blank' class='exbtn'
131	href='$ROOT/sitemap'>Example</a>
132	The "sitemap" containing links to many other pages
133

M www/whyusefossil.wiki

+4 -3

		--- www/whyusefossil.wiki
		+++ www/whyusefossil.wiki
		@@ -232,18 +232,19 @@
232	232	</ul>
233	233	</ul>
234	234	<li><p><b>Why version control is important (reprise)</b>
235	235	<ol type="A">
236	236	<li><p>Every check-in and every individual file has a unique name - its
237		- SHA1 hash. Team members can unambiguously identify any specific
	237	+ SHA1 or SHA3-256 hash. Team members can unambiguously identify
	238	+ any specific
238	239	version of the overall project or any specific version of an
239	240	individual file.
240	241	<li><p>Any historical version of the whole project or of any individual
241	242	file can be easily recreated at any time and by any team member.
242	243	<li><p>Accidental changes to files can be detected by recomputing their
243		- SHA1 hash.
244		- <li><p>Files of unknown origin can be identified using their SHA1 hash.
	244	+ cryptographic hash.
	245	+ <li><p>Files of unknown origin can be identified using their hash.
245	246	<li><p>Developers are able to work in parallel, review each others work,
246	247	and easily merge their changes together. External revisions to
247	248	the baseline can be easily incorporated into the latest changes.
248	249	<li><p>Developers can follow experimental lines of development, then
249	250	revert back to an earlier stable version if the experiment does
250	251

	--- www/whyusefossil.wiki
	+++ www/whyusefossil.wiki
	@@ -232,18 +232,19 @@
232	</ul>
233	</ul>
234	<li><p><b>Why version control is important (reprise)</b>
235	<ol type="A">
236	<li><p>Every check-in and every individual file has a unique name - its
237	SHA1 hash. Team members can unambiguously identify any specific

238	version of the overall project or any specific version of an
239	individual file.
240	<li><p>Any historical version of the whole project or of any individual
241	file can be easily recreated at any time and by any team member.
242	<li><p>Accidental changes to files can be detected by recomputing their
243	SHA1 hash.
244	<li><p>Files of unknown origin can be identified using their SHA1 hash.
245	<li><p>Developers are able to work in parallel, review each others work,
246	and easily merge their changes together. External revisions to
247	the baseline can be easily incorporated into the latest changes.
248	<li><p>Developers can follow experimental lines of development, then
249	revert back to an earlier stable version if the experiment does
250

	--- www/whyusefossil.wiki
	+++ www/whyusefossil.wiki
	@@ -232,18 +232,19 @@
232	</ul>
233	</ul>
234	<li><p><b>Why version control is important (reprise)</b>
235	<ol type="A">
236	<li><p>Every check-in and every individual file has a unique name - its
237	SHA1 or SHA3-256 hash. Team members can unambiguously identify
238	any specific
239	version of the overall project or any specific version of an
240	individual file.
241	<li><p>Any historical version of the whole project or of any individual
242	file can be easily recreated at any time and by any team member.
243	<li><p>Accidental changes to files can be detected by recomputing their
244	cryptographic hash.
245	<li><p>Files of unknown origin can be identified using their hash.
246	<li><p>Developers are able to work in parallel, review each others work,
247	and easily merge their changes together. External revisions to
248	the baseline can be easily incorporated into the latest changes.
249	<li><p>Developers can follow experimental lines of development, then
250	revert back to an earlier stable version if the experiment does
251

Fossil SCM

Keyboard Shortcuts