|
1
|
/* |
|
2
|
** Copyright (c) 2002 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
|
** SYNOPSIS: |
|
19
|
** |
|
20
|
** Input lines that begin with the "@" character are translated into |
|
21
|
** either cgi_printf() statements or string literals and the |
|
22
|
** translated code is written on standard output. |
|
23
|
** |
|
24
|
** The problem this program is attempt to solve is as follows: When |
|
25
|
** writing CGI programs in C, we typically want to output a lot of HTML |
|
26
|
** text to standard output. In pure C code, this involves doing a |
|
27
|
** printf() with a big string containing all that text. But we have |
|
28
|
** to insert special codes (ex: \n and \") for many common characters, |
|
29
|
** which interferes with the readability of the HTML. |
|
30
|
** |
|
31
|
** This tool allows us to put raw HTML, without the special codes, in |
|
32
|
** the middle of a C program. This program then translates the text |
|
33
|
** into standard C by inserting all necessary backslashes and other |
|
34
|
** punctuation. |
|
35
|
** |
|
36
|
** Enhancement #1: |
|
37
|
** |
|
38
|
** If the last non-whitespace character prior to the first "@" of a |
|
39
|
** @-block is "=" or "," then the @-block is a string literal initializer |
|
40
|
** rather than text that is to be output via cgi_printf(). Render it |
|
41
|
** as such. |
|
42
|
** |
|
43
|
** Enhancement #2: |
|
44
|
** |
|
45
|
** Comments of the form: "|* @-comment: CC" (where "|" is really "/") |
|
46
|
** cause CC to become a comment character for the @-substitution. |
|
47
|
** Typical values for CC are "--" (for SQL text) or "#" (for Tcl script) |
|
48
|
** or "//" (for C++ code). Lines of subsequent @-blocks that begin with |
|
49
|
** CC are omitted from the output. |
|
50
|
** |
|
51
|
** Enhancement #3: |
|
52
|
** |
|
53
|
** If a non-enhancement #1 line ends in backslash, the backslash and the |
|
54
|
** newline (\n) are not included in the argument to cgi_printf(). This |
|
55
|
** is used to split one long output line across multiple source lines. |
|
56
|
*/ |
|
57
|
#include <stdio.h> |
|
58
|
#include <ctype.h> |
|
59
|
#include <stdlib.h> |
|
60
|
#include <string.h> |
|
61
|
|
|
62
|
/* |
|
63
|
** Space to hold arguments at the end of the cgi_printf() |
|
64
|
*/ |
|
65
|
#define MX_ARG_SP 10000 |
|
66
|
static char zArg[MX_ARG_SP]; |
|
67
|
static int nArg = 0; |
|
68
|
|
|
69
|
/* |
|
70
|
** True if we are currently in a cgi_printf() |
|
71
|
*/ |
|
72
|
static int inPrint = 0; |
|
73
|
|
|
74
|
/* |
|
75
|
** True if we are currently doing a free string |
|
76
|
*/ |
|
77
|
static int inStr = 0; |
|
78
|
|
|
79
|
/* |
|
80
|
** Name of files being processed |
|
81
|
*/ |
|
82
|
static const char *zInFile = "(stdin)"; |
|
83
|
|
|
84
|
/* |
|
85
|
** The `fossil_isspace()' function copied from the Fossil source code. |
|
86
|
** Some MSVC runtime library versions of `isspace()' break with an `assert()' if |
|
87
|
** the input is smaller than -1 or greater than 255 in debug builds, due to sign |
|
88
|
** extension when promoting `signed char' to `int' for non-ASCII characters. Use |
|
89
|
** an `isspace()' replacement instead of explicit type casts to `unsigned char'. |
|
90
|
*/ |
|
91
|
int fossil_isspace(char c){ |
|
92
|
return c==' ' || (c<='\r' && c>='\t'); |
|
93
|
} |
|
94
|
|
|
95
|
/* |
|
96
|
** Terminate an active cgi_printf() or free string |
|
97
|
*/ |
|
98
|
static void end_block(FILE *out){ |
|
99
|
if( inPrint ){ |
|
100
|
zArg[nArg] = 0; |
|
101
|
fprintf(out, "%s);\n", zArg); |
|
102
|
nArg = 0; |
|
103
|
inPrint = 0; |
|
104
|
} |
|
105
|
} |
|
106
|
|
|
107
|
/* |
|
108
|
** Translate the input stream into the output stream |
|
109
|
*/ |
|
110
|
static void trans(FILE *in, FILE *out){ |
|
111
|
int i, j, k; /* Loop counters */ |
|
112
|
char c1, c2; /* Characters used to start a comment */ |
|
113
|
int lastWasEq = 0; /* True if last non-whitespace character was "=" */ |
|
114
|
int lastWasComma = 0; /* True if last non-whitespace character was "," */ |
|
115
|
int lineNo = 0; /* Line number */ |
|
116
|
char zLine[2000]; /* A single line of input */ |
|
117
|
char zOut[4000]; /* The input line translated into appropriate output */ |
|
118
|
|
|
119
|
c1 = c2 = '-'; |
|
120
|
while( fgets(zLine, sizeof(zLine), in) ){ |
|
121
|
lineNo++; |
|
122
|
for(i=0; zLine[i] && fossil_isspace(zLine[i]); i++){} |
|
123
|
if( zLine[i]!='@' ){ |
|
124
|
if( inPrint || inStr ) end_block(out); |
|
125
|
fprintf(out,"%s",zLine); |
|
126
|
/* 0123456789 12345 */ |
|
127
|
if( strncmp(zLine, "/* @-comment: ", 14)==0 ){ |
|
128
|
c1 = zLine[14]; |
|
129
|
c2 = zLine[15]; |
|
130
|
} |
|
131
|
i += strlen(&zLine[i]); |
|
132
|
while( i>0 && fossil_isspace(zLine[i-1]) ){ i--; } |
|
133
|
lastWasEq = i>0 && zLine[i-1]=='='; |
|
134
|
lastWasComma = i>0 && zLine[i-1]==','; |
|
135
|
}else if( lastWasEq || lastWasComma){ |
|
136
|
/* If the last non-whitespace character before the first @ was |
|
137
|
** an "="(var init/set) or a ","(const definition in list) then |
|
138
|
** generate a string literal. But skip comments |
|
139
|
** consisting of all text between c1 and c2 (default "--") |
|
140
|
** and end of line. |
|
141
|
*/ |
|
142
|
int indent, omitline; |
|
143
|
char *zNewline = "\\n"; |
|
144
|
i++; |
|
145
|
if( fossil_isspace(zLine[i]) ){ i++; } |
|
146
|
indent = i - 2; |
|
147
|
if( indent<0 ) indent = 0; |
|
148
|
omitline = 0; |
|
149
|
for(j=0; zLine[i] && zLine[i]!='\r' && zLine[i]!='\n'; i++){ |
|
150
|
if( zLine[i]==c1 && (c2==' ' || zLine[i+1]==c2) ){ |
|
151
|
omitline = 1; break; |
|
152
|
} |
|
153
|
if( zLine[i]=='\\' && (zLine[i+1]==0 || zLine[i+1]=='\r' |
|
154
|
|| zLine[i+1]=='\n') ){ |
|
155
|
zLine[i] = 0; |
|
156
|
zNewline = ""; |
|
157
|
/* fprintf(stderr, "%s:%d: omit newline\n", zInFile, lineNo); */ |
|
158
|
break; |
|
159
|
} |
|
160
|
if( zLine[i]=='\\' || zLine[i]=='"' ){ zOut[j++] = '\\'; } |
|
161
|
zOut[j++] = zLine[i]; |
|
162
|
} |
|
163
|
if( zNewline[0] ) while( j>0 && fossil_isspace(zOut[j-1]) ){ j--; } |
|
164
|
zOut[j] = 0; |
|
165
|
if( j<=0 && omitline ){ |
|
166
|
fprintf(out,"\n"); |
|
167
|
}else{ |
|
168
|
fprintf(out,"%*s\"%s%s\"\n",indent, "", zOut, zNewline); |
|
169
|
} |
|
170
|
}else{ |
|
171
|
/* Otherwise (if the last non-whitespace was not '=') then generate a |
|
172
|
** cgi_printf() statement whose format is the text following the '@'. |
|
173
|
** Substrings of the form "%C(...)" (where C is any sequence of characters |
|
174
|
** other than \000 and '(') will put "%C" in the format and add the |
|
175
|
** "(...)" as an argument to the cgi_printf call. Each '*' character |
|
176
|
** present in C (max two) causes one more "(...)" sequence to be consumed. |
|
177
|
** For example, "%*.*d(4)(2)(1)" converts to "%*.*d" with arguments "4", |
|
178
|
** "2", and "1", which will be used as the field width, precision, and |
|
179
|
** value, respectively, producing a final formatted result of " 01". |
|
180
|
*/ |
|
181
|
const char *zNewline = "\\n"; |
|
182
|
int indent; |
|
183
|
int nC; |
|
184
|
int nParam; |
|
185
|
char c; |
|
186
|
i++; |
|
187
|
if( fossil_isspace(zLine[i]) ){ i++; } |
|
188
|
indent = i; |
|
189
|
for(j=0; zLine[i] && zLine[i]!='\r' && zLine[i]!='\n'; i++){ |
|
190
|
if( zLine[i]=='\\' && (!zLine[i+1] || zLine[i+1]=='\r' |
|
191
|
|| zLine[i+1]=='\n') ){ |
|
192
|
zNewline = ""; |
|
193
|
break; |
|
194
|
} |
|
195
|
if( zLine[i]=='"' || zLine[i]=='\\' ){ zOut[j++] = '\\'; } |
|
196
|
zOut[j++] = zLine[i]; |
|
197
|
if( zLine[i]!='%' || zLine[i+1]=='%' || zLine[i+1]==0 ) continue; |
|
198
|
nParam=1; |
|
199
|
for(nC=1; zLine[i+nC] && zLine[i+nC]!='('; nC++){ |
|
200
|
if( zLine[i+nC]=='*' && nParam < 3 ) nParam++; |
|
201
|
} |
|
202
|
if( zLine[i+nC]!='(' || !isalpha(zLine[i+nC-1]) ) continue; |
|
203
|
while( --nC ) zOut[j++] = zLine[++i]; |
|
204
|
do{ |
|
205
|
zArg[nArg++] = ','; |
|
206
|
k = 0; i++; |
|
207
|
if( zLine[i]!='(' ) break; |
|
208
|
while( (c = zLine[i])!=0 ){ |
|
209
|
zArg[nArg++] = c; |
|
210
|
if( c==')' ){ |
|
211
|
k--; |
|
212
|
if( k==0 ) break; |
|
213
|
}else if( c=='(' ){ |
|
214
|
k++; |
|
215
|
} |
|
216
|
i++; |
|
217
|
} |
|
218
|
}while( --nParam ); |
|
219
|
} |
|
220
|
zOut[j] = 0; |
|
221
|
if( !inPrint ){ |
|
222
|
fprintf(out,"%*scgi_printf(\"%s%s\"",indent-2,"", zOut, zNewline); |
|
223
|
inPrint = 1; |
|
224
|
}else{ |
|
225
|
fprintf(out,"\n%*s\"%s%s\"",indent+5, "", zOut, zNewline); |
|
226
|
} |
|
227
|
} |
|
228
|
} |
|
229
|
} |
|
230
|
|
|
231
|
static void print_source_ref(const char *zSrcFile, FILE *out){ |
|
232
|
/* Set source line reference to the original source file. |
|
233
|
* This makes compiler show the original file name in the compile error |
|
234
|
* messages, instead of referring to the translated file. |
|
235
|
* NOTE: This somewhat complicates stepping in debugger, as the resuling |
|
236
|
* code would not match the referenced sources. |
|
237
|
*/ |
|
238
|
#ifndef FOSSIL_DEBUG |
|
239
|
const char *arg; |
|
240
|
if( !*zSrcFile ){ |
|
241
|
return; |
|
242
|
} |
|
243
|
fprintf(out,"#line 1 \""); |
|
244
|
for(arg=zSrcFile; *arg; arg++){ |
|
245
|
if( *arg!='\\' ){ |
|
246
|
fprintf(out,"%c", *arg); |
|
247
|
}else{ |
|
248
|
fprintf(out,"\\\\"); |
|
249
|
} |
|
250
|
} |
|
251
|
fprintf(out,"\"\n"); |
|
252
|
#endif |
|
253
|
} |
|
254
|
|
|
255
|
int main(int argc, char **argv){ |
|
256
|
if( argc==2 ){ |
|
257
|
FILE *in = fopen(argv[1], "r"); |
|
258
|
if( in==0 ){ |
|
259
|
fprintf(stderr,"can not open %s\n", argv[1]); |
|
260
|
exit(1); |
|
261
|
} |
|
262
|
zInFile = argv[1]; |
|
263
|
print_source_ref(zInFile, stdout); |
|
264
|
trans(in, stdout); |
|
265
|
fclose(in); |
|
266
|
}else{ |
|
267
|
trans(stdin, stdout); |
|
268
|
} |
|
269
|
return 0; |
|
270
|
} |
|
271
|
|