| 1 |
/*************************************************
|
| 2 |
* Perl-Compatible Regular Expressions *
|
| 3 |
*************************************************/
|
| 4 |
|
| 5 |
/*
|
| 6 |
This is a library of functions to support regular expressions whose syntax
|
| 7 |
and semantics are as close as possible to those of the Perl 5 language. See
|
| 8 |
the file Tech.Notes for some information on the internals.
|
| 9 |
|
| 10 |
This module is a wrapper that provides a POSIX API to the underlying PCRE
|
| 11 |
functions.
|
| 12 |
|
| 13 |
Written by: Philip Hazel <ph10@cam.ac.uk>
|
| 14 |
|
| 15 |
Copyright (c) 1997-2004 University of Cambridge
|
| 16 |
|
| 17 |
-----------------------------------------------------------------------------
|
| 18 |
Redistribution and use in source and binary forms, with or without
|
| 19 |
modification, are permitted provided that the following conditions are met:
|
| 20 |
|
| 21 |
* Redistributions of source code must retain the above copyright notice,
|
| 22 |
this list of conditions and the following disclaimer.
|
| 23 |
|
| 24 |
* Redistributions in binary form must reproduce the above copyright
|
| 25 |
notice, this list of conditions and the following disclaimer in the
|
| 26 |
documentation and/or other materials provided with the distribution.
|
| 27 |
|
| 28 |
* Neither the name of the University of Cambridge nor the names of its
|
| 29 |
contributors may be used to endorse or promote products derived from
|
| 30 |
this software without specific prior written permission.
|
| 31 |
|
| 32 |
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
|
| 33 |
AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
| 34 |
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
| 35 |
ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
|
| 36 |
LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
|
| 37 |
CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
|
| 38 |
SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
|
| 39 |
INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
|
| 40 |
CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
|
| 41 |
ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
|
| 42 |
POSSIBILITY OF SUCH DAMAGE.
|
| 43 |
-----------------------------------------------------------------------------
|
| 44 |
*/
|
| 45 |
|
| 46 |
#include "internal.h"
|
| 47 |
#include "pcreposix.h"
|
| 48 |
#include "stdlib.h"
|
| 49 |
|
| 50 |
|
| 51 |
|
| 52 |
/* Corresponding tables of PCRE error messages and POSIX error codes. */
|
| 53 |
|
| 54 |
static const char *const estring[] = {
|
| 55 |
ERR1, ERR2, ERR3, ERR4, ERR5, ERR6, ERR7, ERR8, ERR9, ERR10,
|
| 56 |
ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19, ERR20,
|
| 57 |
ERR21, ERR22, ERR23, ERR24, ERR25, ERR26, ERR27, ERR29, ERR29, ERR30,
|
| 58 |
ERR31, ERR32, ERR33, ERR34, ERR35, ERR36, ERR37, ERR38, ERR39, ERR40,
|
| 59 |
ERR41, ERR42, ERR43, ERR44, ERR45, ERR46, ERR47 };
|
| 60 |
|
| 61 |
static const int eint[] = {
|
| 62 |
REG_EESCAPE, /* "\\ at end of pattern" */
|
| 63 |
REG_EESCAPE, /* "\\c at end of pattern" */
|
| 64 |
REG_EESCAPE, /* "unrecognized character follows \\" */
|
| 65 |
REG_BADBR, /* "numbers out of order in {} quantifier" */
|
| 66 |
REG_BADBR, /* "number too big in {} quantifier" */
|
| 67 |
REG_EBRACK, /* "missing terminating ] for character class" */
|
| 68 |
REG_ECTYPE, /* "invalid escape sequence in character class" */
|
| 69 |
REG_ERANGE, /* "range out of order in character class" */
|
| 70 |
REG_BADRPT, /* "nothing to repeat" */
|
| 71 |
REG_BADRPT, /* "operand of unlimited repeat could match the empty string" */
|
| 72 |
REG_ASSERT, /* "internal error: unexpected repeat" */
|
| 73 |
REG_BADPAT, /* "unrecognized character after (?" */
|
| 74 |
REG_BADPAT, /* "POSIX named classes are supported only within a class" */
|
| 75 |
REG_EPAREN, /* "missing )" */
|
| 76 |
REG_ESUBREG, /* "reference to non-existent subpattern" */
|
| 77 |
REG_INVARG, /* "erroffset passed as NULL" */
|
| 78 |
REG_INVARG, /* "unknown option bit(s) set" */
|
| 79 |
REG_EPAREN, /* "missing ) after comment" */
|
| 80 |
REG_ESIZE, /* "parentheses nested too deeply" */
|
| 81 |
REG_ESIZE, /* "regular expression too large" */
|
| 82 |
REG_ESPACE, /* "failed to get memory" */
|
| 83 |
REG_EPAREN, /* "unmatched brackets" */
|
| 84 |
REG_ASSERT, /* "internal error: code overflow" */
|
| 85 |
REG_BADPAT, /* "unrecognized character after (?<" */
|
| 86 |
REG_BADPAT, /* "lookbehind assertion is not fixed length" */
|
| 87 |
REG_BADPAT, /* "malformed number after (?(" */
|
| 88 |
REG_BADPAT, /* "conditional group containe more than two branches" */
|
| 89 |
REG_BADPAT, /* "assertion expected after (?(" */
|
| 90 |
REG_BADPAT, /* "(?R or (?digits must be followed by )" */
|
| 91 |
REG_ECTYPE, /* "unknown POSIX class name" */
|
| 92 |
REG_BADPAT, /* "POSIX collating elements are not supported" */
|
| 93 |
REG_INVARG, /* "this version of PCRE is not compiled with PCRE_UTF8 support" */
|
| 94 |
REG_BADPAT, /* "spare error" */
|
| 95 |
REG_BADPAT, /* "character value in \x{...} sequence is too large" */
|
| 96 |
REG_BADPAT, /* "invalid condition (?(0)" */
|
| 97 |
REG_BADPAT, /* "\\C not allowed in lookbehind assertion" */
|
| 98 |
REG_EESCAPE, /* "PCRE does not support \\L, \\l, \\N, \\U, or \\u" */
|
| 99 |
REG_BADPAT, /* "number after (?C is > 255" */
|
| 100 |
REG_BADPAT, /* "closing ) for (?C expected" */
|
| 101 |
REG_BADPAT, /* "recursive call could loop indefinitely" */
|
| 102 |
REG_BADPAT, /* "unrecognized character after (?P" */
|
| 103 |
REG_BADPAT, /* "syntax error after (?P" */
|
| 104 |
REG_BADPAT, /* "two named groups have the same name" */
|
| 105 |
REG_BADPAT, /* "invalid UTF-8 string" */
|
| 106 |
REG_BADPAT, /* "support for \\P, \\p, and \\X has not been compiled" */
|
| 107 |
REG_BADPAT, /* "malformed \\P or \\p sequence" */
|
| 108 |
REG_BADPAT /* "unknown property name after \\P or \\p" */
|
| 109 |
};
|
| 110 |
|
| 111 |
/* Table of texts corresponding to POSIX error codes */
|
| 112 |
|
| 113 |
static const char *const pstring[] = {
|
| 114 |
"", /* Dummy for value 0 */
|
| 115 |
"internal error", /* REG_ASSERT */
|
| 116 |
"invalid repeat counts in {}", /* BADBR */
|
| 117 |
"pattern error", /* BADPAT */
|
| 118 |
"? * + invalid", /* BADRPT */
|
| 119 |
"unbalanced {}", /* EBRACE */
|
| 120 |
"unbalanced []", /* EBRACK */
|
| 121 |
"collation error - not relevant", /* ECOLLATE */
|
| 122 |
"bad class", /* ECTYPE */
|
| 123 |
"bad escape sequence", /* EESCAPE */
|
| 124 |
"empty expression", /* EMPTY */
|
| 125 |
"unbalanced ()", /* EPAREN */
|
| 126 |
"bad range inside []", /* ERANGE */
|
| 127 |
"expression too big", /* ESIZE */
|
| 128 |
"failed to get memory", /* ESPACE */
|
| 129 |
"bad back reference", /* ESUBREG */
|
| 130 |
"bad argument", /* INVARG */
|
| 131 |
"match failed" /* NOMATCH */
|
| 132 |
};
|
| 133 |
|
| 134 |
|
| 135 |
|
| 136 |
|
| 137 |
/*************************************************
|
| 138 |
* Translate PCRE text code to int *
|
| 139 |
*************************************************/
|
| 140 |
|
| 141 |
/* PCRE compile-time errors are given as strings defined as macros. We can just
|
| 142 |
look them up in a table to turn them into POSIX-style error codes. */
|
| 143 |
|
| 144 |
static int
|
| 145 |
pcre_posix_error_code(const char *s)
|
| 146 |
{
|
| 147 |
size_t i;
|
| 148 |
for (i = 0; i < sizeof(estring)/sizeof(char *); i++)
|
| 149 |
if (strcmp(s, estring[i]) == 0) return eint[i];
|
| 150 |
return REG_ASSERT;
|
| 151 |
}
|
| 152 |
|
| 153 |
|
| 154 |
|
| 155 |
/*************************************************
|
| 156 |
* Translate error code to string *
|
| 157 |
*************************************************/
|
| 158 |
|
| 159 |
EXPORT size_t
|
| 160 |
regerror(int errcode, const regex_t *preg, char *errbuf, size_t errbuf_size)
|
| 161 |
{
|
| 162 |
const char *message, *addmessage;
|
| 163 |
size_t length, addlength;
|
| 164 |
|
| 165 |
message = (errcode >= (int)(sizeof(pstring)/sizeof(char *)))?
|
| 166 |
"unknown error code" : pstring[errcode];
|
| 167 |
length = strlen(message) + 1;
|
| 168 |
|
| 169 |
addmessage = " at offset ";
|
| 170 |
addlength = (preg != NULL && (int)preg->re_erroffset != -1)?
|
| 171 |
strlen(addmessage) + 6 : 0;
|
| 172 |
|
| 173 |
if (errbuf_size > 0)
|
| 174 |
{
|
| 175 |
if (addlength > 0 && errbuf_size >= length + addlength)
|
| 176 |
sprintf(errbuf, "%s%s%-6d", message, addmessage, (int)preg->re_erroffset);
|
| 177 |
else
|
| 178 |
{
|
| 179 |
strncpy(errbuf, message, errbuf_size - 1);
|
| 180 |
errbuf[errbuf_size-1] = 0;
|
| 181 |
}
|
| 182 |
}
|
| 183 |
|
| 184 |
return length + addlength;
|
| 185 |
}
|
| 186 |
|
| 187 |
|
| 188 |
|
| 189 |
|
| 190 |
/*************************************************
|
| 191 |
* Free store held by a regex *
|
| 192 |
*************************************************/
|
| 193 |
|
| 194 |
EXPORT void
|
| 195 |
regfree(regex_t *preg)
|
| 196 |
{
|
| 197 |
(pcre_free)(preg->re_pcre);
|
| 198 |
}
|
| 199 |
|
| 200 |
|
| 201 |
|
| 202 |
|
| 203 |
/*************************************************
|
| 204 |
* Compile a regular expression *
|
| 205 |
*************************************************/
|
| 206 |
|
| 207 |
/*
|
| 208 |
Arguments:
|
| 209 |
preg points to a structure for recording the compiled expression
|
| 210 |
pattern the pattern to compile
|
| 211 |
cflags compilation flags
|
| 212 |
|
| 213 |
Returns: 0 on success
|
| 214 |
various non-zero codes on failure
|
| 215 |
*/
|
| 216 |
|
| 217 |
EXPORT int
|
| 218 |
regcomp(regex_t *preg, const char *pattern, int cflags)
|
| 219 |
{
|
| 220 |
const char *errorptr;
|
| 221 |
int erroffset;
|
| 222 |
int options = 0;
|
| 223 |
|
| 224 |
if ((cflags & REG_ICASE) != 0) options |= PCRE_CASELESS;
|
| 225 |
if ((cflags & REG_NEWLINE) != 0) options |= PCRE_MULTILINE;
|
| 226 |
|
| 227 |
preg->re_pcre = pcre_compile(pattern, options, &errorptr, &erroffset, NULL);
|
| 228 |
preg->re_erroffset = erroffset;
|
| 229 |
|
| 230 |
if (preg->re_pcre == NULL) return pcre_posix_error_code(errorptr);
|
| 231 |
|
| 232 |
preg->re_nsub = pcre_info((const pcre *)preg->re_pcre, NULL, NULL);
|
| 233 |
return 0;
|
| 234 |
}
|
| 235 |
|
| 236 |
|
| 237 |
|
| 238 |
|
| 239 |
/*************************************************
|
| 240 |
* Match a regular expression *
|
| 241 |
*************************************************/
|
| 242 |
|
| 243 |
/* Unfortunately, PCRE requires 3 ints of working space for each captured
|
| 244 |
substring, so we have to get and release working store instead of just using
|
| 245 |
the POSIX structures as was done in earlier releases when PCRE needed only 2
|
| 246 |
ints. However, if the number of possible capturing brackets is small, use a
|
| 247 |
block of store on the stack, to reduce the use of malloc/free. The threshold is
|
| 248 |
in a macro that can be changed at configure time. */
|
| 249 |
|
| 250 |
EXPORT int
|
| 251 |
regexec(const regex_t *preg, const char *string, size_t nmatch,
|
| 252 |
regmatch_t pmatch[], int eflags)
|
| 253 |
{
|
| 254 |
int rc;
|
| 255 |
int options = 0;
|
| 256 |
int *ovector = NULL;
|
| 257 |
int small_ovector[POSIX_MALLOC_THRESHOLD * 3];
|
| 258 |
BOOL allocated_ovector = FALSE;
|
| 259 |
|
| 260 |
if ((eflags & REG_NOTBOL) != 0) options |= PCRE_NOTBOL;
|
| 261 |
if ((eflags & REG_NOTEOL) != 0) options |= PCRE_NOTEOL;
|
| 262 |
|
| 263 |
((regex_t *)preg)->re_erroffset = (size_t)(-1); /* Only has meaning after compile */
|
| 264 |
|
| 265 |
if (nmatch > 0)
|
| 266 |
{
|
| 267 |
if (nmatch <= POSIX_MALLOC_THRESHOLD)
|
| 268 |
{
|
| 269 |
ovector = &(small_ovector[0]);
|
| 270 |
}
|
| 271 |
else
|
| 272 |
{
|
| 273 |
ovector = (int *)malloc(sizeof(int) * nmatch * 3);
|
| 274 |
if (ovector == NULL) return REG_ESPACE;
|
| 275 |
allocated_ovector = TRUE;
|
| 276 |
}
|
| 277 |
}
|
| 278 |
|
| 279 |
rc = pcre_exec((const pcre *)preg->re_pcre, NULL, string, (int)strlen(string),
|
| 280 |
0, options, ovector, nmatch * 3);
|
| 281 |
|
| 282 |
if (rc == 0) rc = nmatch; /* All captured slots were filled in */
|
| 283 |
|
| 284 |
if (rc >= 0)
|
| 285 |
{
|
| 286 |
size_t i;
|
| 287 |
for (i = 0; i < (size_t)rc; i++)
|
| 288 |
{
|
| 289 |
pmatch[i].rm_so = ovector[i*2];
|
| 290 |
pmatch[i].rm_eo = ovector[i*2+1];
|
| 291 |
}
|
| 292 |
if (allocated_ovector) free(ovector);
|
| 293 |
for (; i < nmatch; i++) pmatch[i].rm_so = pmatch[i].rm_eo = -1;
|
| 294 |
return 0;
|
| 295 |
}
|
| 296 |
|
| 297 |
else
|
| 298 |
{
|
| 299 |
if (allocated_ovector) free(ovector);
|
| 300 |
switch(rc)
|
| 301 |
{
|
| 302 |
case PCRE_ERROR_NOMATCH: return REG_NOMATCH;
|
| 303 |
case PCRE_ERROR_NULL: return REG_INVARG;
|
| 304 |
case PCRE_ERROR_BADOPTION: return REG_INVARG;
|
| 305 |
case PCRE_ERROR_BADMAGIC: return REG_INVARG;
|
| 306 |
case PCRE_ERROR_UNKNOWN_NODE: return REG_ASSERT;
|
| 307 |
case PCRE_ERROR_NOMEMORY: return REG_ESPACE;
|
| 308 |
case PCRE_ERROR_MATCHLIMIT: return REG_ESPACE;
|
| 309 |
case PCRE_ERROR_BADUTF8: return REG_INVARG;
|
| 310 |
case PCRE_ERROR_BADUTF8_OFFSET: return REG_INVARG;
|
| 311 |
default: return REG_ASSERT;
|
| 312 |
}
|
| 313 |
}
|
| 314 |
}
|
| 315 |
|
| 316 |
/* End of pcreposix.c */
|
Parent Directory
| 
Revision Log