| 1 |
nigel |
63 |
/************************************************* |
| 2 |
|
|
* PCRE DEMONSTRATION PROGRAM * |
| 3 |
|
|
*************************************************/ |
| 4 |
nigel |
53 |
|
| 5 |
nigel |
63 |
/* This is a demonstration program to illustrate the most straightforward ways |
| 6 |
|
|
of calling the PCRE regular expression library from a C program. See the |
| 7 |
ph10 |
315 |
pcresample documentation for a short discussion ("man pcresample" if you have |
| 8 |
|
|
the PCRE man pages installed). |
| 9 |
nigel |
63 |
|
| 10 |
ph10 |
315 |
In Unix-like environments, compile this program thuswise: |
| 11 |
|
|
|
| 12 |
nigel |
75 |
gcc -Wall pcredemo.c -I/usr/local/include -L/usr/local/lib \ |
| 13 |
|
|
-R/usr/local/lib -lpcre |
| 14 |
nigel |
63 |
|
| 15 |
nigel |
75 |
Replace "/usr/local/include" and "/usr/local/lib" with wherever the include and |
| 16 |
ph10 |
200 |
library files for PCRE are installed on your system. You don't need -I and -L |
| 17 |
ph10 |
199 |
if PCRE is installed in the standard system libraries. Only some operating |
| 18 |
nigel |
63 |
systems (e.g. Solaris) use the -R option. |
| 19 |
nigel |
53 |
|
| 20 |
ph10 |
315 |
Building under Windows: |
| 21 |
nigel |
63 |
|
| 22 |
ph10 |
315 |
If you want to statically link this program against a non-dll .a file, you must |
| 23 |
|
|
define PCRE_STATIC before including pcre.h, otherwise the pcre_malloc() and |
| 24 |
|
|
pcre_free() exported functions will be declared __declspec(dllimport), with |
| 25 |
|
|
unwanted results. So in this environment, uncomment the following line. */ |
| 26 |
|
|
|
| 27 |
|
|
/* #define PCRE_STATIC */ |
| 28 |
|
|
|
| 29 |
nigel |
63 |
#include <stdio.h> |
| 30 |
|
|
#include <string.h> |
| 31 |
|
|
#include <pcre.h> |
| 32 |
|
|
|
| 33 |
nigel |
53 |
#define OVECCOUNT 30 /* should be a multiple of 3 */ |
| 34 |
|
|
|
| 35 |
nigel |
63 |
|
| 36 |
nigel |
53 |
int main(int argc, char **argv) |
| 37 |
|
|
{ |
| 38 |
|
|
pcre *re; |
| 39 |
|
|
const char *error; |
| 40 |
nigel |
63 |
char *pattern; |
| 41 |
|
|
char *subject; |
| 42 |
|
|
unsigned char *name_table; |
| 43 |
nigel |
53 |
int erroffset; |
| 44 |
nigel |
63 |
int find_all; |
| 45 |
|
|
int namecount; |
| 46 |
|
|
int name_entry_size; |
| 47 |
nigel |
53 |
int ovector[OVECCOUNT]; |
| 48 |
nigel |
63 |
int subject_length; |
| 49 |
nigel |
53 |
int rc, i; |
| 50 |
|
|
|
| 51 |
nigel |
63 |
|
| 52 |
nigel |
75 |
/************************************************************************** |
| 53 |
|
|
* First, sort out the command line. There is only one possible option at * |
| 54 |
|
|
* the moment, "-g" to request repeated matching to find all occurrences, * |
| 55 |
|
|
* like Perl's /g option. We set the variable find_all to a non-zero value * |
| 56 |
|
|
* if the -g option is present. Apart from that, there must be exactly two * |
| 57 |
|
|
* arguments. * |
| 58 |
|
|
**************************************************************************/ |
| 59 |
nigel |
63 |
|
| 60 |
|
|
find_all = 0; |
| 61 |
|
|
for (i = 1; i < argc; i++) |
| 62 |
nigel |
53 |
{ |
| 63 |
nigel |
63 |
if (strcmp(argv[i], "-g") == 0) find_all = 1; |
| 64 |
|
|
else break; |
| 65 |
|
|
} |
| 66 |
|
|
|
| 67 |
|
|
/* After the options, we require exactly two arguments, which are the pattern, |
| 68 |
|
|
and the subject string. */ |
| 69 |
|
|
|
| 70 |
|
|
if (argc - i != 2) |
| 71 |
|
|
{ |
| 72 |
nigel |
53 |
printf("Two arguments required: a regex and a subject string\n"); |
| 73 |
|
|
return 1; |
| 74 |
|
|
} |
| 75 |
|
|
|
| 76 |
nigel |
63 |
pattern = argv[i]; |
| 77 |
|
|
subject = argv[i+1]; |
| 78 |
|
|
subject_length = (int)strlen(subject); |
| 79 |
nigel |
53 |
|
| 80 |
nigel |
63 |
|
| 81 |
|
|
/************************************************************************* |
| 82 |
|
|
* Now we are going to compile the regular expression pattern, and handle * |
| 83 |
|
|
* and errors that are detected. * |
| 84 |
|
|
*************************************************************************/ |
| 85 |
|
|
|
| 86 |
nigel |
53 |
re = pcre_compile( |
| 87 |
nigel |
63 |
pattern, /* the pattern */ |
| 88 |
nigel |
53 |
0, /* default options */ |
| 89 |
|
|
&error, /* for error message */ |
| 90 |
|
|
&erroffset, /* for error offset */ |
| 91 |
|
|
NULL); /* use default character tables */ |
| 92 |
|
|
|
| 93 |
|
|
/* Compilation failed: print the error message and exit */ |
| 94 |
|
|
|
| 95 |
|
|
if (re == NULL) |
| 96 |
|
|
{ |
| 97 |
|
|
printf("PCRE compilation failed at offset %d: %s\n", erroffset, error); |
| 98 |
|
|
return 1; |
| 99 |
|
|
} |
| 100 |
|
|
|
| 101 |
|
|
|
| 102 |
nigel |
63 |
/************************************************************************* |
| 103 |
|
|
* If the compilation succeeded, we call PCRE again, in order to do a * |
| 104 |
nigel |
75 |
* pattern match against the subject string. This does just ONE match. If * |
| 105 |
nigel |
63 |
* further matching is needed, it will be done below. * |
| 106 |
|
|
*************************************************************************/ |
| 107 |
|
|
|
| 108 |
nigel |
53 |
rc = pcre_exec( |
| 109 |
|
|
re, /* the compiled pattern */ |
| 110 |
|
|
NULL, /* no extra data - we didn't study the pattern */ |
| 111 |
nigel |
63 |
subject, /* the subject string */ |
| 112 |
|
|
subject_length, /* the length of the subject */ |
| 113 |
nigel |
53 |
0, /* start at offset 0 in the subject */ |
| 114 |
|
|
0, /* default options */ |
| 115 |
|
|
ovector, /* output vector for substring information */ |
| 116 |
|
|
OVECCOUNT); /* number of elements in the output vector */ |
| 117 |
|
|
|
| 118 |
|
|
/* Matching failed: handle error cases */ |
| 119 |
|
|
|
| 120 |
|
|
if (rc < 0) |
| 121 |
|
|
{ |
| 122 |
|
|
switch(rc) |
| 123 |
|
|
{ |
| 124 |
|
|
case PCRE_ERROR_NOMATCH: printf("No match\n"); break; |
| 125 |
|
|
/* |
| 126 |
|
|
Handle other special cases if you like |
| 127 |
|
|
*/ |
| 128 |
|
|
default: printf("Matching error %d\n", rc); break; |
| 129 |
|
|
} |
| 130 |
nigel |
91 |
pcre_free(re); /* Release memory used for the compiled pattern */ |
| 131 |
nigel |
53 |
return 1; |
| 132 |
|
|
} |
| 133 |
|
|
|
| 134 |
|
|
/* Match succeded */ |
| 135 |
|
|
|
| 136 |
nigel |
63 |
printf("\nMatch succeeded at offset %d\n", ovector[0]); |
| 137 |
nigel |
53 |
|
| 138 |
nigel |
63 |
|
| 139 |
|
|
/************************************************************************* |
| 140 |
|
|
* We have found the first match within the subject string. If the output * |
| 141 |
ph10 |
315 |
* vector wasn't big enough, say so. Then output any substrings that were * |
| 142 |
|
|
* captured. * |
| 143 |
nigel |
63 |
*************************************************************************/ |
| 144 |
|
|
|
| 145 |
nigel |
53 |
/* The output vector wasn't big enough */ |
| 146 |
|
|
|
| 147 |
|
|
if (rc == 0) |
| 148 |
|
|
{ |
| 149 |
|
|
rc = OVECCOUNT/3; |
| 150 |
|
|
printf("ovector only has room for %d captured substrings\n", rc - 1); |
| 151 |
|
|
} |
| 152 |
|
|
|
| 153 |
nigel |
63 |
/* Show substrings stored in the output vector by number. Obviously, in a real |
| 154 |
|
|
application you might want to do things other than print them. */ |
| 155 |
nigel |
53 |
|
| 156 |
|
|
for (i = 0; i < rc; i++) |
| 157 |
|
|
{ |
| 158 |
nigel |
63 |
char *substring_start = subject + ovector[2*i]; |
| 159 |
nigel |
53 |
int substring_length = ovector[2*i+1] - ovector[2*i]; |
| 160 |
|
|
printf("%2d: %.*s\n", i, substring_length, substring_start); |
| 161 |
|
|
} |
| 162 |
|
|
|
| 163 |
nigel |
63 |
|
| 164 |
nigel |
75 |
/************************************************************************** |
| 165 |
|
|
* That concludes the basic part of this demonstration program. We have * |
| 166 |
|
|
* compiled a pattern, and performed a single match. The code that follows * |
| 167 |
ph10 |
315 |
* shows first how to access named substrings, and then how to code for * |
| 168 |
nigel |
75 |
* repeated matches on the same subject. * |
| 169 |
|
|
**************************************************************************/ |
| 170 |
nigel |
63 |
|
| 171 |
|
|
/* See if there are any named substrings, and if so, show them by name. First |
| 172 |
|
|
we have to extract the count of named parentheses from the pattern. */ |
| 173 |
|
|
|
| 174 |
|
|
(void)pcre_fullinfo( |
| 175 |
|
|
re, /* the compiled pattern */ |
| 176 |
|
|
NULL, /* no extra data - we didn't study the pattern */ |
| 177 |
|
|
PCRE_INFO_NAMECOUNT, /* number of named substrings */ |
| 178 |
|
|
&namecount); /* where to put the answer */ |
| 179 |
|
|
|
| 180 |
|
|
if (namecount <= 0) printf("No named substrings\n"); else |
| 181 |
|
|
{ |
| 182 |
|
|
unsigned char *tabptr; |
| 183 |
|
|
printf("Named substrings\n"); |
| 184 |
|
|
|
| 185 |
|
|
/* Before we can access the substrings, we must extract the table for |
| 186 |
|
|
translating names to numbers, and the size of each entry in the table. */ |
| 187 |
|
|
|
| 188 |
|
|
(void)pcre_fullinfo( |
| 189 |
|
|
re, /* the compiled pattern */ |
| 190 |
|
|
NULL, /* no extra data - we didn't study the pattern */ |
| 191 |
|
|
PCRE_INFO_NAMETABLE, /* address of the table */ |
| 192 |
|
|
&name_table); /* where to put the answer */ |
| 193 |
|
|
|
| 194 |
|
|
(void)pcre_fullinfo( |
| 195 |
|
|
re, /* the compiled pattern */ |
| 196 |
|
|
NULL, /* no extra data - we didn't study the pattern */ |
| 197 |
|
|
PCRE_INFO_NAMEENTRYSIZE, /* size of each entry in the table */ |
| 198 |
|
|
&name_entry_size); /* where to put the answer */ |
| 199 |
|
|
|
| 200 |
|
|
/* Now we can scan the table and, for each entry, print the number, the name, |
| 201 |
|
|
and the substring itself. */ |
| 202 |
|
|
|
| 203 |
|
|
tabptr = name_table; |
| 204 |
|
|
for (i = 0; i < namecount; i++) |
| 205 |
|
|
{ |
| 206 |
|
|
int n = (tabptr[0] << 8) | tabptr[1]; |
| 207 |
|
|
printf("(%d) %*s: %.*s\n", n, name_entry_size - 3, tabptr + 2, |
| 208 |
|
|
ovector[2*n+1] - ovector[2*n], subject + ovector[2*n]); |
| 209 |
|
|
tabptr += name_entry_size; |
| 210 |
|
|
} |
| 211 |
|
|
} |
| 212 |
|
|
|
| 213 |
|
|
|
| 214 |
|
|
/************************************************************************* |
| 215 |
|
|
* If the "-g" option was given on the command line, we want to continue * |
| 216 |
|
|
* to search for additional matches in the subject string, in a similar * |
| 217 |
|
|
* way to the /g option in Perl. This turns out to be trickier than you * |
| 218 |
|
|
* might think because of the possibility of matching an empty string. * |
| 219 |
|
|
* What happens is as follows: * |
| 220 |
|
|
* * |
| 221 |
|
|
* If the previous match was NOT for an empty string, we can just start * |
| 222 |
|
|
* the next match at the end of the previous one. * |
| 223 |
|
|
* * |
| 224 |
|
|
* If the previous match WAS for an empty string, we can't do that, as it * |
| 225 |
|
|
* would lead to an infinite loop. Instead, a special call of pcre_exec() * |
| 226 |
|
|
* is made with the PCRE_NOTEMPTY and PCRE_ANCHORED flags set. The first * |
| 227 |
|
|
* of these tells PCRE that an empty string is not a valid match; other * |
| 228 |
|
|
* possibilities must be tried. The second flag restricts PCRE to one * |
| 229 |
|
|
* match attempt at the initial string position. If this match succeeds, * |
| 230 |
|
|
* an alternative to the empty string match has been found, and we can * |
| 231 |
|
|
* proceed round the loop. * |
| 232 |
|
|
*************************************************************************/ |
| 233 |
|
|
|
| 234 |
nigel |
75 |
if (!find_all) |
| 235 |
|
|
{ |
| 236 |
nigel |
91 |
pcre_free(re); /* Release the memory used for the compiled pattern */ |
| 237 |
|
|
return 0; /* Finish unless -g was given */ |
| 238 |
nigel |
75 |
} |
| 239 |
nigel |
63 |
|
| 240 |
|
|
/* Loop for second and subsequent matches */ |
| 241 |
|
|
|
| 242 |
|
|
for (;;) |
| 243 |
|
|
{ |
| 244 |
|
|
int options = 0; /* Normally no options */ |
| 245 |
|
|
int start_offset = ovector[1]; /* Start at end of previous match */ |
| 246 |
|
|
|
| 247 |
|
|
/* If the previous match was for an empty string, we are finished if we are |
| 248 |
|
|
at the end of the subject. Otherwise, arrange to run another match at the |
| 249 |
|
|
same point to see if a non-empty match can be found. */ |
| 250 |
|
|
|
| 251 |
|
|
if (ovector[0] == ovector[1]) |
| 252 |
|
|
{ |
| 253 |
|
|
if (ovector[0] == subject_length) break; |
| 254 |
|
|
options = PCRE_NOTEMPTY | PCRE_ANCHORED; |
| 255 |
|
|
} |
| 256 |
|
|
|
| 257 |
|
|
/* Run the next matching operation */ |
| 258 |
|
|
|
| 259 |
|
|
rc = pcre_exec( |
| 260 |
|
|
re, /* the compiled pattern */ |
| 261 |
|
|
NULL, /* no extra data - we didn't study the pattern */ |
| 262 |
|
|
subject, /* the subject string */ |
| 263 |
|
|
subject_length, /* the length of the subject */ |
| 264 |
|
|
start_offset, /* starting offset in the subject */ |
| 265 |
|
|
options, /* options */ |
| 266 |
|
|
ovector, /* output vector for substring information */ |
| 267 |
|
|
OVECCOUNT); /* number of elements in the output vector */ |
| 268 |
|
|
|
| 269 |
|
|
/* This time, a result of NOMATCH isn't an error. If the value in "options" |
| 270 |
|
|
is zero, it just means we have found all possible matches, so the loop ends. |
| 271 |
|
|
Otherwise, it means we have failed to find a non-empty-string match at a |
| 272 |
|
|
point where there was a previous empty-string match. In this case, we do what |
| 273 |
|
|
Perl does: advance the matching position by one, and continue. We do this by |
| 274 |
|
|
setting the "end of previous match" offset, because that is picked up at the |
| 275 |
|
|
top of the loop as the point at which to start again. */ |
| 276 |
|
|
|
| 277 |
|
|
if (rc == PCRE_ERROR_NOMATCH) |
| 278 |
|
|
{ |
| 279 |
|
|
if (options == 0) break; |
| 280 |
|
|
ovector[1] = start_offset + 1; |
| 281 |
|
|
continue; /* Go round the loop again */ |
| 282 |
|
|
} |
| 283 |
|
|
|
| 284 |
|
|
/* Other matching errors are not recoverable. */ |
| 285 |
|
|
|
| 286 |
|
|
if (rc < 0) |
| 287 |
|
|
{ |
| 288 |
|
|
printf("Matching error %d\n", rc); |
| 289 |
nigel |
91 |
pcre_free(re); /* Release memory used for the compiled pattern */ |
| 290 |
nigel |
63 |
return 1; |
| 291 |
|
|
} |
| 292 |
|
|
|
| 293 |
|
|
/* Match succeded */ |
| 294 |
|
|
|
| 295 |
|
|
printf("\nMatch succeeded again at offset %d\n", ovector[0]); |
| 296 |
|
|
|
| 297 |
|
|
/* The match succeeded, but the output vector wasn't big enough. */ |
| 298 |
|
|
|
| 299 |
|
|
if (rc == 0) |
| 300 |
|
|
{ |
| 301 |
|
|
rc = OVECCOUNT/3; |
| 302 |
|
|
printf("ovector only has room for %d captured substrings\n", rc - 1); |
| 303 |
|
|
} |
| 304 |
|
|
|
| 305 |
|
|
/* As before, show substrings stored in the output vector by number, and then |
| 306 |
|
|
also any named substrings. */ |
| 307 |
|
|
|
| 308 |
|
|
for (i = 0; i < rc; i++) |
| 309 |
|
|
{ |
| 310 |
|
|
char *substring_start = subject + ovector[2*i]; |
| 311 |
|
|
int substring_length = ovector[2*i+1] - ovector[2*i]; |
| 312 |
|
|
printf("%2d: %.*s\n", i, substring_length, substring_start); |
| 313 |
|
|
} |
| 314 |
|
|
|
| 315 |
|
|
if (namecount <= 0) printf("No named substrings\n"); else |
| 316 |
|
|
{ |
| 317 |
|
|
unsigned char *tabptr = name_table; |
| 318 |
|
|
printf("Named substrings\n"); |
| 319 |
|
|
for (i = 0; i < namecount; i++) |
| 320 |
|
|
{ |
| 321 |
|
|
int n = (tabptr[0] << 8) | tabptr[1]; |
| 322 |
|
|
printf("(%d) %*s: %.*s\n", n, name_entry_size - 3, tabptr + 2, |
| 323 |
|
|
ovector[2*n+1] - ovector[2*n], subject + ovector[2*n]); |
| 324 |
|
|
tabptr += name_entry_size; |
| 325 |
|
|
} |
| 326 |
|
|
} |
| 327 |
|
|
} /* End of loop to find second and subsequent matches */ |
| 328 |
|
|
|
| 329 |
|
|
printf("\n"); |
| 330 |
nigel |
91 |
pcre_free(re); /* Release memory used for the compiled pattern */ |
| 331 |
nigel |
53 |
return 0; |
| 332 |
|
|
} |
| 333 |
|
|
|
| 334 |
nigel |
63 |
/* End of pcredemo.c */ |