| 1 |
/*************************************************
|
| 2 |
* Perl-Compatible Regular Expressions *
|
| 3 |
*************************************************/
|
| 4 |
|
| 5 |
/* PCRE is a library of functions to support regular expressions whose syntax
|
| 6 |
and semantics are as close as possible to those of the Perl 5 language.
|
| 7 |
|
| 8 |
Written by Philip Hazel
|
| 9 |
Copyright (c) 1997-2006 University of Cambridge
|
| 10 |
|
| 11 |
-----------------------------------------------------------------------------
|
| 12 |
Redistribution and use in source and binary forms, with or without
|
| 13 |
modification, are permitted provided that the following conditions are met:
|
| 14 |
|
| 15 |
* Redistributions of source code must retain the above copyright notice,
|
| 16 |
this list of conditions and the following disclaimer.
|
| 17 |
|
| 18 |
* Redistributions in binary form must reproduce the above copyright
|
| 19 |
notice, this list of conditions and the following disclaimer in the
|
| 20 |
documentation and/or other materials provided with the distribution.
|
| 21 |
|
| 22 |
* Neither the name of the University of Cambridge nor the names of its
|
| 23 |
contributors may be used to endorse or promote products derived from
|
| 24 |
this software without specific prior written permission.
|
| 25 |
|
| 26 |
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
|
| 27 |
AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
| 28 |
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
| 29 |
ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
|
| 30 |
LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
|
| 31 |
CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
|
| 32 |
SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
|
| 33 |
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
|
| 34 |
CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
|
| 35 |
ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
|
| 36 |
POSSIBILITY OF SUCH DAMAGE.
|
| 37 |
-----------------------------------------------------------------------------
|
| 38 |
*/
|
| 39 |
|
| 40 |
|
| 41 |
/* This module contains some convenience functions for extracting substrings
|
| 42 |
from the subject string after a regex match has succeeded. The original idea
|
| 43 |
for these functions came from Scott Wimer. */
|
| 44 |
|
| 45 |
|
| 46 |
#include "pcre_internal.h"
|
| 47 |
|
| 48 |
|
| 49 |
/*************************************************
|
| 50 |
* Find number for named string *
|
| 51 |
*************************************************/
|
| 52 |
|
| 53 |
/* This function is used by the get_first_set() function below, as well
|
| 54 |
as being generally available. It assumes that names are unique.
|
| 55 |
|
| 56 |
Arguments:
|
| 57 |
code the compiled regex
|
| 58 |
stringname the name whose number is required
|
| 59 |
|
| 60 |
Returns: the number of the named parentheses, or a negative number
|
| 61 |
(PCRE_ERROR_NOSUBSTRING) if not found
|
| 62 |
*/
|
| 63 |
|
| 64 |
int
|
| 65 |
pcre_get_stringnumber(const pcre *code, const char *stringname)
|
| 66 |
{
|
| 67 |
int rc;
|
| 68 |
int entrysize;
|
| 69 |
int top, bot;
|
| 70 |
uschar *nametable;
|
| 71 |
|
| 72 |
if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMECOUNT, &top)) != 0)
|
| 73 |
return rc;
|
| 74 |
if (top <= 0) return PCRE_ERROR_NOSUBSTRING;
|
| 75 |
|
| 76 |
if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMEENTRYSIZE, &entrysize)) != 0)
|
| 77 |
return rc;
|
| 78 |
if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMETABLE, &nametable)) != 0)
|
| 79 |
return rc;
|
| 80 |
|
| 81 |
bot = 0;
|
| 82 |
while (top > bot)
|
| 83 |
{
|
| 84 |
int mid = (top + bot) / 2;
|
| 85 |
uschar *entry = nametable + entrysize*mid;
|
| 86 |
int c = strcmp(stringname, (char *)(entry + 2));
|
| 87 |
if (c == 0) return (entry[0] << 8) + entry[1];
|
| 88 |
if (c > 0) bot = mid + 1; else top = mid;
|
| 89 |
}
|
| 90 |
|
| 91 |
return PCRE_ERROR_NOSUBSTRING;
|
| 92 |
}
|
| 93 |
|
| 94 |
|
| 95 |
|
| 96 |
/*************************************************
|
| 97 |
* Find (multiple) entries for named string *
|
| 98 |
*************************************************/
|
| 99 |
|
| 100 |
/* This is used by the get_first_set() function below, as well as being
|
| 101 |
generally available. It is used when duplicated names are permitted.
|
| 102 |
|
| 103 |
Arguments:
|
| 104 |
code the compiled regex
|
| 105 |
stringname the name whose entries required
|
| 106 |
firstptr where to put the pointer to the first entry
|
| 107 |
lastptr where to put the pointer to the last entry
|
| 108 |
|
| 109 |
Returns: the length of each entry, or a negative number
|
| 110 |
(PCRE_ERROR_NOSUBSTRING) if not found
|
| 111 |
*/
|
| 112 |
|
| 113 |
int
|
| 114 |
pcre_get_stringtable_entries(const pcre *code, const char *stringname,
|
| 115 |
char **firstptr, char **lastptr)
|
| 116 |
{
|
| 117 |
int rc;
|
| 118 |
int entrysize;
|
| 119 |
int top, bot;
|
| 120 |
uschar *nametable, *lastentry;
|
| 121 |
|
| 122 |
if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMECOUNT, &top)) != 0)
|
| 123 |
return rc;
|
| 124 |
if (top <= 0) return PCRE_ERROR_NOSUBSTRING;
|
| 125 |
|
| 126 |
if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMEENTRYSIZE, &entrysize)) != 0)
|
| 127 |
return rc;
|
| 128 |
if ((rc = pcre_fullinfo(code, NULL, PCRE_INFO_NAMETABLE, &nametable)) != 0)
|
| 129 |
return rc;
|
| 130 |
|
| 131 |
lastentry = nametable + entrysize * (top - 1);
|
| 132 |
bot = 0;
|
| 133 |
while (top > bot)
|
| 134 |
{
|
| 135 |
int mid = (top + bot) / 2;
|
| 136 |
uschar *entry = nametable + entrysize*mid;
|
| 137 |
int c = strcmp(stringname, (char *)(entry + 2));
|
| 138 |
if (c == 0)
|
| 139 |
{
|
| 140 |
uschar *first = entry;
|
| 141 |
uschar *last = entry;
|
| 142 |
while (first > nametable)
|
| 143 |
{
|
| 144 |
if (strcmp(stringname, (char *)(first - entrysize + 2)) != 0) break;
|
| 145 |
first -= entrysize;
|
| 146 |
}
|
| 147 |
while (last < lastentry)
|
| 148 |
{
|
| 149 |
if (strcmp(stringname, (char *)(last + entrysize + 2)) != 0) break;
|
| 150 |
last += entrysize;
|
| 151 |
}
|
| 152 |
*firstptr = (char *)first;
|
| 153 |
*lastptr = (char *)last;
|
| 154 |
return entrysize;
|
| 155 |
}
|
| 156 |
if (c > 0) bot = mid + 1; else top = mid;
|
| 157 |
}
|
| 158 |
|
| 159 |
return PCRE_ERROR_NOSUBSTRING;
|
| 160 |
}
|
| 161 |
|
| 162 |
|
| 163 |
|
| 164 |
/*************************************************
|
| 165 |
* Find first set of multiple named strings *
|
| 166 |
*************************************************/
|
| 167 |
|
| 168 |
/* This function allows for duplicate names in the table of named substrings.
|
| 169 |
It returns the number of the first one that was set in a pattern match.
|
| 170 |
|
| 171 |
Arguments:
|
| 172 |
code the compiled regex
|
| 173 |
stringname the name of the capturing substring
|
| 174 |
ovector the vector of matched substrings
|
| 175 |
|
| 176 |
Returns: the number of the first that is set,
|
| 177 |
or the number of the last one if none are set,
|
| 178 |
or a negative number on error
|
| 179 |
*/
|
| 180 |
|
| 181 |
static int
|
| 182 |
get_first_set(const pcre *code, const char *stringname, int *ovector)
|
| 183 |
{
|
| 184 |
const real_pcre *re = (const real_pcre *)code;
|
| 185 |
int entrysize;
|
| 186 |
char *first, *last;
|
| 187 |
uschar *entry;
|
| 188 |
if ((re->options & (PCRE_DUPNAMES | PCRE_JCHANGED)) == 0)
|
| 189 |
return pcre_get_stringnumber(code, stringname);
|
| 190 |
entrysize = pcre_get_stringtable_entries(code, stringname, &first, &last);
|
| 191 |
if (entrysize <= 0) return entrysize;
|
| 192 |
for (entry = (uschar *)first; entry <= (uschar *)last; entry += entrysize)
|
| 193 |
{
|
| 194 |
int n = (entry[0] << 8) + entry[1];
|
| 195 |
if (ovector[n*2] >= 0) return n;
|
| 196 |
}
|
| 197 |
return (first[0] << 8) + first[1];
|
| 198 |
}
|
| 199 |
|
| 200 |
|
| 201 |
|
| 202 |
|
| 203 |
/*************************************************
|
| 204 |
* Copy captured string to given buffer *
|
| 205 |
*************************************************/
|
| 206 |
|
| 207 |
/* This function copies a single captured substring into a given buffer.
|
| 208 |
Note that we use memcpy() rather than strncpy() in case there are binary zeros
|
| 209 |
in the string.
|
| 210 |
|
| 211 |
Arguments:
|
| 212 |
subject the subject string that was matched
|
| 213 |
ovector pointer to the offsets table
|
| 214 |
stringcount the number of substrings that were captured
|
| 215 |
(i.e. the yield of the pcre_exec call, unless
|
| 216 |
that was zero, in which case it should be 1/3
|
| 217 |
of the offset table size)
|
| 218 |
stringnumber the number of the required substring
|
| 219 |
buffer where to put the substring
|
| 220 |
size the size of the buffer
|
| 221 |
|
| 222 |
Returns: if successful:
|
| 223 |
the length of the copied string, not including the zero
|
| 224 |
that is put on the end; can be zero
|
| 225 |
if not successful:
|
| 226 |
PCRE_ERROR_NOMEMORY (-6) buffer too small
|
| 227 |
PCRE_ERROR_NOSUBSTRING (-7) no such captured substring
|
| 228 |
*/
|
| 229 |
|
| 230 |
int
|
| 231 |
pcre_copy_substring(const char *subject, int *ovector, int stringcount,
|
| 232 |
int stringnumber, char *buffer, int size)
|
| 233 |
{
|
| 234 |
int yield;
|
| 235 |
if (stringnumber < 0 || stringnumber >= stringcount)
|
| 236 |
return PCRE_ERROR_NOSUBSTRING;
|
| 237 |
stringnumber *= 2;
|
| 238 |
yield = ovector[stringnumber+1] - ovector[stringnumber];
|
| 239 |
if (size < yield + 1) return PCRE_ERROR_NOMEMORY;
|
| 240 |
memcpy(buffer, subject + ovector[stringnumber], yield);
|
| 241 |
buffer[yield] = 0;
|
| 242 |
return yield;
|
| 243 |
}
|
| 244 |
|
| 245 |
|
| 246 |
|
| 247 |
/*************************************************
|
| 248 |
* Copy named captured string to given buffer *
|
| 249 |
*************************************************/
|
| 250 |
|
| 251 |
/* This function copies a single captured substring into a given buffer,
|
| 252 |
identifying it by name. If the regex permits duplicate names, the first
|
| 253 |
substring that is set is chosen.
|
| 254 |
|
| 255 |
Arguments:
|
| 256 |
code the compiled regex
|
| 257 |
subject the subject string that was matched
|
| 258 |
ovector pointer to the offsets table
|
| 259 |
stringcount the number of substrings that were captured
|
| 260 |
(i.e. the yield of the pcre_exec call, unless
|
| 261 |
that was zero, in which case it should be 1/3
|
| 262 |
of the offset table size)
|
| 263 |
stringname the name of the required substring
|
| 264 |
buffer where to put the substring
|
| 265 |
size the size of the buffer
|
| 266 |
|
| 267 |
Returns: if successful:
|
| 268 |
the length of the copied string, not including the zero
|
| 269 |
that is put on the end; can be zero
|
| 270 |
if not successful:
|
| 271 |
PCRE_ERROR_NOMEMORY (-6) buffer too small
|
| 272 |
PCRE_ERROR_NOSUBSTRING (-7) no such captured substring
|
| 273 |
*/
|
| 274 |
|
| 275 |
int
|
| 276 |
pcre_copy_named_substring(const pcre *code, const char *subject, int *ovector,
|
| 277 |
int stringcount, const char *stringname, char *buffer, int size)
|
| 278 |
{
|
| 279 |
int n = get_first_set(code, stringname, ovector);
|
| 280 |
if (n <= 0) return n;
|
| 281 |
return pcre_copy_substring(subject, ovector, stringcount, n, buffer, size);
|
| 282 |
}
|
| 283 |
|
| 284 |
|
| 285 |
|
| 286 |
/*************************************************
|
| 287 |
* Copy all captured strings to new store *
|
| 288 |
*************************************************/
|
| 289 |
|
| 290 |
/* This function gets one chunk of store and builds a list of pointers and all
|
| 291 |
of the captured substrings in it. A NULL pointer is put on the end of the list.
|
| 292 |
|
| 293 |
Arguments:
|
| 294 |
subject the subject string that was matched
|
| 295 |
ovector pointer to the offsets table
|
| 296 |
stringcount the number of substrings that were captured
|
| 297 |
(i.e. the yield of the pcre_exec call, unless
|
| 298 |
that was zero, in which case it should be 1/3
|
| 299 |
of the offset table size)
|
| 300 |
listptr set to point to the list of pointers
|
| 301 |
|
| 302 |
Returns: if successful: 0
|
| 303 |
if not successful:
|
| 304 |
PCRE_ERROR_NOMEMORY (-6) failed to get store
|
| 305 |
*/
|
| 306 |
|
| 307 |
int
|
| 308 |
pcre_get_substring_list(const char *subject, int *ovector, int stringcount,
|
| 309 |
const char ***listptr)
|
| 310 |
{
|
| 311 |
int i;
|
| 312 |
int size = sizeof(char *);
|
| 313 |
int double_count = stringcount * 2;
|
| 314 |
char **stringlist;
|
| 315 |
char *p;
|
| 316 |
|
| 317 |
for (i = 0; i < double_count; i += 2)
|
| 318 |
size += sizeof(char *) + ovector[i+1] - ovector[i] + 1;
|
| 319 |
|
| 320 |
stringlist = (char **)(pcre_malloc)(size);
|
| 321 |
if (stringlist == NULL) return PCRE_ERROR_NOMEMORY;
|
| 322 |
|
| 323 |
*listptr = (const char **)stringlist;
|
| 324 |
p = (char *)(stringlist + stringcount + 1);
|
| 325 |
|
| 326 |
for (i = 0; i < double_count; i += 2)
|
| 327 |
{
|
| 328 |
int len = ovector[i+1] - ovector[i];
|
| 329 |
memcpy(p, subject + ovector[i], len);
|
| 330 |
*stringlist++ = p;
|
| 331 |
p += len;
|
| 332 |
*p++ = 0;
|
| 333 |
}
|
| 334 |
|
| 335 |
*stringlist = NULL;
|
| 336 |
return 0;
|
| 337 |
}
|
| 338 |
|
| 339 |
|
| 340 |
|
| 341 |
/*************************************************
|
| 342 |
* Free store obtained by get_substring_list *
|
| 343 |
*************************************************/
|
| 344 |
|
| 345 |
/* This function exists for the benefit of people calling PCRE from non-C
|
| 346 |
programs that can call its functions, but not free() or (pcre_free)() directly.
|
| 347 |
|
| 348 |
Argument: the result of a previous pcre_get_substring_list()
|
| 349 |
Returns: nothing
|
| 350 |
*/
|
| 351 |
|
| 352 |
void
|
| 353 |
pcre_free_substring_list(const char **pointer)
|
| 354 |
{
|
| 355 |
(pcre_free)((void *)pointer);
|
| 356 |
}
|
| 357 |
|
| 358 |
|
| 359 |
|
| 360 |
/*************************************************
|
| 361 |
* Copy captured string to new store *
|
| 362 |
*************************************************/
|
| 363 |
|
| 364 |
/* This function copies a single captured substring into a piece of new
|
| 365 |
store
|
| 366 |
|
| 367 |
Arguments:
|
| 368 |
subject the subject string that was matched
|
| 369 |
ovector pointer to the offsets table
|
| 370 |
stringcount the number of substrings that were captured
|
| 371 |
(i.e. the yield of the pcre_exec call, unless
|
| 372 |
that was zero, in which case it should be 1/3
|
| 373 |
of the offset table size)
|
| 374 |
stringnumber the number of the required substring
|
| 375 |
stringptr where to put a pointer to the substring
|
| 376 |
|
| 377 |
Returns: if successful:
|
| 378 |
the length of the string, not including the zero that
|
| 379 |
is put on the end; can be zero
|
| 380 |
if not successful:
|
| 381 |
PCRE_ERROR_NOMEMORY (-6) failed to get store
|
| 382 |
PCRE_ERROR_NOSUBSTRING (-7) substring not present
|
| 383 |
*/
|
| 384 |
|
| 385 |
int
|
| 386 |
pcre_get_substring(const char *subject, int *ovector, int stringcount,
|
| 387 |
int stringnumber, const char **stringptr)
|
| 388 |
{
|
| 389 |
int yield;
|
| 390 |
char *substring;
|
| 391 |
if (stringnumber < 0 || stringnumber >= stringcount)
|
| 392 |
return PCRE_ERROR_NOSUBSTRING;
|
| 393 |
stringnumber *= 2;
|
| 394 |
yield = ovector[stringnumber+1] - ovector[stringnumber];
|
| 395 |
substring = (char *)(pcre_malloc)(yield + 1);
|
| 396 |
if (substring == NULL) return PCRE_ERROR_NOMEMORY;
|
| 397 |
memcpy(substring, subject + ovector[stringnumber], yield);
|
| 398 |
substring[yield] = 0;
|
| 399 |
*stringptr = substring;
|
| 400 |
return yield;
|
| 401 |
}
|
| 402 |
|
| 403 |
|
| 404 |
|
| 405 |
/*************************************************
|
| 406 |
* Copy named captured string to new store *
|
| 407 |
*************************************************/
|
| 408 |
|
| 409 |
/* This function copies a single captured substring, identified by name, into
|
| 410 |
new store. If the regex permits duplicate names, the first substring that is
|
| 411 |
set is chosen.
|
| 412 |
|
| 413 |
Arguments:
|
| 414 |
code the compiled regex
|
| 415 |
subject the subject string that was matched
|
| 416 |
ovector pointer to the offsets table
|
| 417 |
stringcount the number of substrings that were captured
|
| 418 |
(i.e. the yield of the pcre_exec call, unless
|
| 419 |
that was zero, in which case it should be 1/3
|
| 420 |
of the offset table size)
|
| 421 |
stringname the name of the required substring
|
| 422 |
stringptr where to put the pointer
|
| 423 |
|
| 424 |
Returns: if successful:
|
| 425 |
the length of the copied string, not including the zero
|
| 426 |
that is put on the end; can be zero
|
| 427 |
if not successful:
|
| 428 |
PCRE_ERROR_NOMEMORY (-6) couldn't get memory
|
| 429 |
PCRE_ERROR_NOSUBSTRING (-7) no such captured substring
|
| 430 |
*/
|
| 431 |
|
| 432 |
int
|
| 433 |
pcre_get_named_substring(const pcre *code, const char *subject, int *ovector,
|
| 434 |
int stringcount, const char *stringname, const char **stringptr)
|
| 435 |
{
|
| 436 |
int n = get_first_set(code, stringname, ovector);
|
| 437 |
if (n <= 0) return n;
|
| 438 |
return pcre_get_substring(subject, ovector, stringcount, n, stringptr);
|
| 439 |
|
| 440 |
}
|
| 441 |
|
| 442 |
|
| 443 |
|
| 444 |
|
| 445 |
/*************************************************
|
| 446 |
* Free store obtained by get_substring *
|
| 447 |
*************************************************/
|
| 448 |
|
| 449 |
/* This function exists for the benefit of people calling PCRE from non-C
|
| 450 |
programs that can call its functions, but not free() or (pcre_free)() directly.
|
| 451 |
|
| 452 |
Argument: the result of a previous pcre_get_substring()
|
| 453 |
Returns: nothing
|
| 454 |
*/
|
| 455 |
|
| 456 |
void
|
| 457 |
pcre_free_substring(const char *pointer)
|
| 458 |
{
|
| 459 |
(pcre_free)((void *)pointer);
|
| 460 |
}
|
| 461 |
|
| 462 |
/* End of pcre_get.c */
|