Fossil SCM

Add notes from Ralf Engelshall as a text file: rse-notes.txt.

drh 2008-10-18 19:42 trunk
Commit f7a410eca92ce63f90aeb959c9670a5fea993fe2
Parent 9e80dc66cf7e678…
1 file changed +181
+ rse-notes.txt
A rse-notes.txt
+181
--- a/rse-notes.txt
+++ b/rse-notes.txt
@@ -0,0 +1,181 @@
1
+ with you:
2
+
3
+o No tarball
4
+
5
+ You currently still do not provide any tarba s of the Fossil sources.
6
+ Sure, you are still hacking wild on the c e and one can on-the-fly
7
+ fetch a ZIP archive, but that's a manual pr ess. For packagers (like
8
+ me in the OpenPKG project) this is very Fossil will usually still not be packaged. As a result it will be not
9
+ spreaded as much as possible and this way you still do not get as much
10
+ as possible feedback. Hence, I recommend that you let daily snapshot
11
+ tarballs (or ZIP files) be rolled. This would be great.
12
+
13
+o UUID
14
+
15
+ Under http://www.fossil-scm.org/index.html/doc/tip/www/concepts.wiki
16
+ you describe the concepts and you clearly name the artifact ID a
17
+ y name the artifact ID a
18
+ "UUID" and even say that this is a "Unive ally Unique Identifier".
19
+ Unfortunately, a UUID is a 128-bit enti  standardized by the ISO
20
+ (under ISO/IEC 11578:1996) and IETF (under FC-4122) and hence it is
21
+ *VERY MUCH* confusing and unclean that you 160-bit SHA1-based ID is
22
+  I *STRONGLY* recommend you to eith entities, where a UUID of version 5 even is SHA1-based!) or you name
23
+ your 160-bit SHA1 entities differently (perhaps AID for "Artifact
24
+ ll"
25
+Date: OctoFrCurrently we have only "fossil cgi <script>" where <script> contains
26
+ "repository: <file>". This is a neat idea and should be kept. But
27
+ this way one cannot easily host multiple repositories over this
28
+ CGI interface without creating such a script for every individual
29
+ repository.
30
+
31
+ ndividual
32
+ repository.
33
+
34
+ Perhaps a "fossil cgi --repository <file> would help here, as this
35
+ way one can use a generic CGI script w ch first figures out the
36
+ name <name> of the individual project nd then runs "fossil cgi
37
+ --repository /path/to/repositories/<name> overlooking something and this is already possible...
38
+
39
+o "fossil db <operation>"
40
+
41
+ In Monotone DVCS we have a "mtn db" command for the low-level SQLite
42
+ From: "Ralf S. Engelschall"
43
+Date: October 18, 2008 1:40:53 PM EDT
44
+To: d
45
+Df S. Engelschall"
46
+Date: October 18, 2008 1:40:53 PM EDT
47
+To: drh@hwaFrded as much as possible and this way you still do not get as much
48
+   possible feedback. Hence, I recommend that you let daily snapshot  tarballs (or ZIP files) be rolled. This would be great.
49
+
50
+o UUID
51
+
52
+ U er http://www.fossil-scm.org/index.html/doc/tip/www/concepts.wiki
53
+ you u clearly name the artifact ID a
54
+ "UUID" and even say that this is  "Universally Uniqu Identifier".
55
+ Unfortunately, a UUID is a 128-bit entity standard ed by the ISO
56
+ (under ISO/IEC 11578:1996) and IETF (under RFC-4122) a confusing and unclean that your 160-bit SHA1-based ID is
57
+ called "UU " in Fossil.
58
+
59
+ I *STRONGLY* recommend you to either use real UUIDs (128-bit
60
+ entities, where  28-bit
61
+ entities, where a UUID sed!) or you name
62
+ your 160-bit SHA1 entities differently (perhaps AID or "Artifact
63
+ Identifier"?).
64
+
65
+o "fossil cgi <script>"
66
+
67
+ Currently e have only "fossil cgi <script>" where <script> contains
68
+ "repository <file>". This is a neat idea and should be kept. But
69
+ this way one c positories over this
70
+  There is "fossil first figures out l
71
+ repository.
72
+
73
+ Perhaps a " epository.
74
+
75
+ Perhaps a "fossil cgi --repository <file>" would help use a generic CGI script which first figures out the
76
+ name <name of the individual project and then runs "fossil cgi
77
+ --repository /pa /to/repositories/<name>.db". But perhaps I'm just
78
+ overlooking some ing and this is already possible...
79
+
80
+o "fossil db <operation>"
81
+
82
+ In onotone DVCS we have a "mtn db" c From: "Ralf S. Engelschall"
83
+Date: October 18, 2008 1:40:53 PM EDT
84
+Thaa sub-command option and not a global option. And on "fossil ui" or
85
+ "fossil server" it even is an argument and not an option. This is
86
+ d not an option. This is
87
+ confusing. Every time one has to figure out ow to set the repository
88
+ on the CLI. Monotone simply uses a *global it. Soas possible feedbacFossil also u s a single global option
89
+ "--repository"/"-R" instead of a command- ecific option. Sure, not
90
+ all commands might require a repository ignore the global option there than having to figure out every time
91
+ how the repository has to be  When hitting "Apply changes" on any setup page, one keeps staying on
92
+ this page. Sure, that's what an "apply" button usually is about. But
93
+ I at least would have liked to see a button (either instead of the
94
+ "apply changes" or at least in addition) which applies the changes and
95
+ goes back to the setup main page (from where one usually come).
96
+
97
+o _FOSSIL_
98
+
99
+    abase for the _FOSSIL_
100
+ control file. BUT: Why "_FOSSIL_"? CVS's "C  directory was ugly for
101
+ decades. Today we have ".svn", ".git", ".hg  rid of those ugly control files or directories of the DVCS! Sure,
102
+  dot-files are disliked by Windows people. But that's no problem, one
103
+  can accept "_XXX" in addition to   So, I really would like to see the file "_FOSSIL_" to be renamed
104
+    upper-case is important to you) ".FOSSIL" and "_FOSSIL". But to see
105
+  ct: Some Fossil Feedback
106
+
107
+I've today looked at your Fossil project (From tree is really rather
108
+ boring for Unix people!
109
+
110
+o "fossil open", "fossil ch kout", "fossil update".
111
+
112
+ I'm personally confused that "fossil open"  other DVCS and that "checkout" is just a variant of "update". Hmmm...
113
+ for me it looks at least one should be eleminated to reduce confusion.
114
+ The difference between checkout and update could become an option of
115
+ a single option. And the remaining to perhaps better should be either
116
+  better should be either
117
+ "open" + "update" or "checkout" + "update  Well, perhaps I'm still
118
+ not understanding Fossil enough. But I can o y tell that I got rather
119
+ con There is "fossil rom: "Ralf S. Engelschf S. "fossil status" seems to be a super-set of "fossil changes".
120
+ Looks a little bit less orthogonal than necessary. I would remove
121
+ "fossil changes" or at least do not show the file details on "fossil
122
+ status".
123
+
124
+Yours,
125
+ Ralf S. Engelschall
126
+
127
+pply changes" on an at's what an "apply" button usually is about. But
128
+ I at least would ave liked to see a button (either instead of the
129
+ "apply changes" o f the
130
+ "apply changes" or at least in addition) which applies the cha es and
131
+ goes back to the setup main page (from where one usually com .
132
+
133
+o _FOSSIL_
134
+
135
+ Very nice idea to use another SQLite database fo the _FOSSIL_
136
+ control file. BUT: Why "_FOSSIL_"? CVS's "CVS" direct for us Unix hackers ;-)
137
+
138
+ t", ".hg" and "_MTN"/".mtn" to get
139
+ rid of those ugly control files  those ugly control files or directories of the DVCS! Sure,
140
+ dot-fi s are disliked by Windows people. But that's no problem, one
141
+ can a ept "_XXX" in addition to ".XXX" easily, of course.
142
+
143
+ So, I reall would like to see the file "_FOSSIL_" to be renamed
144
+ to ".fossil"  r Windows or (if the
145
+ upper-case is important to you) ".FOSSIL" a  "_FOSSIL". But to see
146
+ an ugly "_FOSSIL_" at the top of every sour  tree is really rather
147
+ boring for Unix people!
148
+
149
+o "fossil open", "f sil checkout", "fossil update".
150
+
151
+ I'm personally confused that "fos l open" is what "checkout" does in
152
+ other DVCS and that "checkout" is  S and that "checkout" i  Hmmm...
153
+ for me it looks at least one should be eleminated to redu  confusion.
154
+ The difference between checkout and update could beco  an option of
155
+ a single option. And the remaining to perhaps better s uld be either
156
+ "open" + "update" or "checkout" + "update". Well, perha  I'm still
157
+ not understanding Fossil enough. But I can only tell th ossil FeedbackFrrom: "Ralf S. Engelschall"
158
+Date: October 18, 2008 1:40:53 PM EDT
159
+TEnplayed with Fossil I see that "manifest" and "manifest.uuid" files
160
+  database and instead externally represented in the source tree.
161
+ Hmmm... then I recommend that you use a .fossil *DIRECTORY*, place
162
+ the control file into .fossil/control (or whatever) and the manifest
163
+ atever) and the manifest
164
+ stuff into .fossil/manifest and .fossil/ma fest.uuid. But please do
165
+ not clutter the top-level of the source thre DVCS.
166
+
167
+o "fossil mv"
168
+
169
+ There is "fossil add" and "fossil rm", but no "fossil mv". Hopefully
170
+ this doesn't mean a file move is an "add" and a "remove" bundle, as
171
+ this way Fossil would have trouble to know that the file was renamed
172
+ and hence the full history tracking of a file seems to be broken.
173
+
174
+o "fossil cha Finally, I got confused by the "fossil changes" and "fossil status"
175
+ commands. "fossil status" seems to be a super-set of "fossil changes".
176
+ Looks a little bit less orthogonal than necessary. I would remove
177
+ lly, I got confus remove
178
+ "fossil changes status".
179
+
180
+Yours,
181
+                                   
--- a/rse-notes.txt
+++ b/rse-notes.txt
@@ -0,0 +1,181 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
--- a/rse-notes.txt
+++ b/rse-notes.txt
@@ -0,0 +1,181 @@
1  with you:
2
3 o No tarball
4
5 You currently still do not provide any tarba s of the Fossil sources.
6 Sure, you are still hacking wild on the c e and one can on-the-fly
7 fetch a ZIP archive, but that's a manual pr ess. For packagers (like
8 me in the OpenPKG project) this is very Fossil will usually still not be packaged. As a result it will be not
9  spreaded as much as possible and this way you still do not get as much
10  as possible feedback. Hence, I recommend that you let daily snapshot
11  tarballs (or ZIP files) be rolled. This would be great.
12
13 o UUID
14
15  Under http://www.fossil-scm.org/index.html/doc/tip/www/concepts.wiki
16  you describe the concepts and you clearly name the artifact ID a
17  y name the artifact ID a
18 "UUID" and even say that this is a "Unive ally Unique Identifier".
19 Unfortunately, a UUID is a 128-bit enti  standardized by the ISO
20 (under ISO/IEC 11578:1996) and IETF (under FC-4122) and hence it is
21 *VERY MUCH* confusing and unclean that you 160-bit SHA1-based ID is
22  I *STRONGLY* recommend you to eith entities, where a UUID of version 5 even is SHA1-based!) or you name
23  your 160-bit SHA1 entities differently (perhaps AID for "Artifact
24  ll"
25 Date: OctoFrCurrently we have only "fossil cgi <script>" where <script> contains
26  "repository: <file>". This is a neat idea and should be kept. But
27  this way one cannot easily host multiple repositories over this
28  CGI interface without creating such a script for every individual
29  repository.
30
31  ndividual
32 repository.
33
34 Perhaps a "fossil cgi --repository <file> would help here, as this
35 way one can use a generic CGI script w ch first figures out the
36 name <name> of the individual project nd then runs "fossil cgi
37 --repository /path/to/repositories/<name> overlooking something and this is already possible...
38
39 o "fossil db <operation>"
40
41  In Monotone DVCS we have a "mtn db" command for the low-level SQLite
42  From: "Ralf S. Engelschall"
43 Date: October 18, 2008 1:40:53 PM EDT
44 To: d
45 Df S. Engelschall"
46 Date: October 18, 2008 1:40:53 PM EDT
47 To: drh@hwaFrded as much as possible and this way you still do not get as much
48   possible feedback. Hence, I recommend that you let daily snapshot  tarballs (or ZIP files) be rolled. This would be great.
49
50 o UUID
51
52 U er http://www.fossil-scm.org/index.html/doc/tip/www/concepts.wiki
53 you u clearly name the artifact ID a
54 "UUID" and even say that this is  "Universally Uniqu Identifier".
55 Unfortunately, a UUID is a 128-bit entity standard ed by the ISO
56 (under ISO/IEC 11578:1996) and IETF (under RFC-4122) a confusing and unclean that your 160-bit SHA1-based ID is
57 called "UU " in Fossil.
58
59 I *STRONGLY* recommend you to either use real UUIDs (128-bit
60 entities, where  28-bit
61 entities, where a UUID sed!) or you name
62 your 160-bit SHA1 entities differently (perhaps AID or "Artifact
63 Identifier"?).
64
65 o "fossil cgi <script>"
66
67 Currently e have only "fossil cgi <script>" where <script> contains
68 "repository <file>". This is a neat idea and should be kept. But
69 this way one c positories over this
70  There is "fossil first figures out l
71 repository.
72
73 Perhaps a " epository.
74
75 Perhaps a "fossil cgi --repository <file>" would help use a generic CGI script which first figures out the
76 name <name of the individual project and then runs "fossil cgi
77 --repository /pa /to/repositories/<name>.db". But perhaps I'm just
78 overlooking some ing and this is already possible...
79
80 o "fossil db <operation>"
81
82 In onotone DVCS we have a "mtn db" c From: "Ralf S. Engelschall"
83 Date: October 18, 2008 1:40:53 PM EDT
84 Thaa sub-command option and not a global option. And on "fossil ui" or
85  "fossil server" it even is an argument and not an option. This is
86  d not an option. This is
87 confusing. Every time one has to figure out ow to set the repository
88 on the CLI. Monotone simply uses a *global it. Soas possible feedbacFossil also u s a single global option
89 "--repository"/"-R" instead of a command- ecific option. Sure, not
90 all commands might require a repository ignore the global option there than having to figure out every time
91  how the repository has to be  When hitting "Apply changes" on any setup page, one keeps staying on
92  this page. Sure, that's what an "apply" button usually is about. But
93  I at least would have liked to see a button (either instead of the
94  "apply changes" or at least in addition) which applies the changes and
95  goes back to the setup main page (from where one usually come).
96
97 o _FOSSIL_
98
99     abase for the _FOSSIL_
100 control file. BUT: Why "_FOSSIL_"? CVS's "C  directory was ugly for
101 decades. Today we have ".svn", ".git", ".hg  rid of those ugly control files or directories of the DVCS! Sure,
102   dot-files are disliked by Windows people. But that's no problem, one
103   can accept "_XXX" in addition to   So, I really would like to see the file "_FOSSIL_" to be renamed
104     upper-case is important to you) ".FOSSIL" and "_FOSSIL". But to see
105   ct: Some Fossil Feedback
106
107 I've today looked at your Fossil project (From tree is really rather
108 boring for Unix people!
109
110 o "fossil open", "fossil ch kout", "fossil update".
111
112 I'm personally confused that "fossil open"  other DVCS and that "checkout" is just a variant of "update". Hmmm...
113  for me it looks at least one should be eleminated to reduce confusion.
114  The difference between checkout and update could become an option of
115  a single option. And the remaining to perhaps better should be either
116   better should be either
117 "open" + "update" or "checkout" + "update  Well, perhaps I'm still
118 not understanding Fossil enough. But I can o y tell that I got rather
119 con There is "fossil rom: "Ralf S. Engelschf S. "fossil status" seems to be a super-set of "fossil changes".
120 Looks a little bit less orthogonal than necessary. I would remove
121 "fossil changes" or at least do not show the file details on "fossil
122 status".
123
124 Yours,
125 Ralf S. Engelschall
126
127 pply changes" on an at's what an "apply" button usually is about. But
128 I at least would ave liked to see a button (either instead of the
129 "apply changes" o f the
130 "apply changes" or at least in addition) which applies the cha es and
131 goes back to the setup main page (from where one usually com .
132
133 o _FOSSIL_
134
135 Very nice idea to use another SQLite database fo the _FOSSIL_
136 control file. BUT: Why "_FOSSIL_"? CVS's "CVS" direct for us Unix hackers ;-)
137
138  t", ".hg" and "_MTN"/".mtn" to get
139 rid of those ugly control files  those ugly control files or directories of the DVCS! Sure,
140 dot-fi s are disliked by Windows people. But that's no problem, one
141 can a ept "_XXX" in addition to ".XXX" easily, of course.
142
143 So, I reall would like to see the file "_FOSSIL_" to be renamed
144 to ".fossil"  r Windows or (if the
145 upper-case is important to you) ".FOSSIL" a  "_FOSSIL". But to see
146 an ugly "_FOSSIL_" at the top of every sour  tree is really rather
147 boring for Unix people!
148
149 o "fossil open", "f sil checkout", "fossil update".
150
151 I'm personally confused that "fos l open" is what "checkout" does in
152 other DVCS and that "checkout" is  S and that "checkout" i  Hmmm...
153 for me it looks at least one should be eleminated to redu  confusion.
154 The difference between checkout and update could beco  an option of
155 a single option. And the remaining to perhaps better s uld be either
156 "open" + "update" or "checkout" + "update". Well, perha  I'm still
157 not understanding Fossil enough. But I can only tell th ossil FeedbackFrrom: "Ralf S. Engelschall"
158 Date: October 18, 2008 1:40:53 PM EDT
159 TEnplayed with Fossil I see that "manifest" and "manifest.uuid" files
160   database and instead externally represented in the source tree.
161  Hmmm... then I recommend that you use a .fossil *DIRECTORY*, place
162  the control file into .fossil/control (or whatever) and the manifest
163  atever) and the manifest
164 stuff into .fossil/manifest and .fossil/ma fest.uuid. But please do
165 not clutter the top-level of the source thre DVCS.
166
167 o "fossil mv"
168
169  There is "fossil add" and "fossil rm", but no "fossil mv". Hopefully
170  this doesn't mean a file move is an "add" and a "remove" bundle, as
171  this way Fossil would have trouble to know that the file was renamed
172  and hence the full history tracking of a file seems to be broken.
173
174 o "fossil cha Finally, I got confused by the "fossil changes" and "fossil status"
175  commands. "fossil status" seems to be a super-set of "fossil changes".
176  Looks a little bit less orthogonal than necessary. I would remove
177  lly, I got confus remove
178 "fossil changes status".
179
180 Yours,
181                                    

Keyboard Shortcuts

Open search /
Next entry (timeline) j
Previous entry (timeline) k
Open focused entry Enter
Show this help ?
Toggle theme Top nav button