1
/*************************************************
2
* Perl-Compatible Regular Expressions *
3
*************************************************/
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.
10
This module is a wrapper that provides a POSIX API to the underlying PCRE
13
Written by: Philip Hazel <ph10@cam.ac.uk>
15
Copyright (c) 1997-2000 University of Cambridge
17
-----------------------------------------------------------------------------
18
Permission is granted to anyone to use this software for any purpose on any
19
computer system, and to redistribute it freely, subject to the following
22
1. This software is distributed in the hope that it will be useful,
23
but WITHOUT ANY WARRANTY; without even the implied warranty of
24
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
26
2. The origin of this software must not be misrepresented, either by
27
explicit claim or by omission.
29
3. Altered versions must be plainly marked as such, and must not be
30
misrepresented as being the original software.
32
4. If PCRE is embedded in any software that is released under the GNU
33
General Purpose Licence (GPL), then the terms of that licence shall
34
supersede any condition above with which it is incompatible.
35
-----------------------------------------------------------------------------
39
#include "pcreposix.h"
44
/* Corresponding tables of PCRE error messages and POSIX error codes. */
46
static const char *estring[] = {
47
ERR1, ERR2, ERR3, ERR4, ERR5, ERR6, ERR7, ERR8, ERR9, ERR10,
48
ERR11, ERR12, ERR13, ERR14, ERR15, ERR16, ERR17, ERR18, ERR19, ERR20,
49
ERR21, ERR22, ERR23, ERR24, ERR25, ERR26, ERR27, ERR29, ERR29, ERR30,
53
REG_EESCAPE, /* "\\ at end of pattern" */
54
REG_EESCAPE, /* "\\c at end of pattern" */
55
REG_EESCAPE, /* "unrecognized character follows \\" */
56
REG_BADBR, /* "numbers out of order in {} quantifier" */
57
REG_BADBR, /* "number too big in {} quantifier" */
58
REG_EBRACK, /* "missing terminating ] for character class" */
59
REG_ECTYPE, /* "invalid escape sequence in character class" */
60
REG_ERANGE, /* "range out of order in character class" */
61
REG_BADRPT, /* "nothing to repeat" */
62
REG_BADRPT, /* "operand of unlimited repeat could match the empty string" */
63
REG_ASSERT, /* "internal error: unexpected repeat" */
64
REG_BADPAT, /* "unrecognized character after (?" */
65
REG_ESIZE, /* "too many capturing parenthesized sub-patterns" */
66
REG_EPAREN, /* "missing )" */
67
REG_ESUBREG, /* "back reference to non-existent subpattern" */
68
REG_INVARG, /* "erroffset passed as NULL" */
69
REG_INVARG, /* "unknown option bit(s) set" */
70
REG_EPAREN, /* "missing ) after comment" */
71
REG_ESIZE, /* "too many sets of parentheses" */
72
REG_ESIZE, /* "regular expression too large" */
73
REG_ESPACE, /* "failed to get memory" */
74
REG_EPAREN, /* "unmatched brackets" */
75
REG_ASSERT, /* "internal error: code overflow" */
76
REG_BADPAT, /* "unrecognized character after (?<" */
77
REG_BADPAT, /* "lookbehind assertion is not fixed length" */
78
REG_BADPAT, /* "malformed number after (?(" */
79
REG_BADPAT, /* "conditional group containe more than two branches" */
80
REG_BADPAT, /* "assertion expected after (?(" */
81
REG_BADPAT, /* "(?p must be followed by )" */
82
REG_ECTYPE, /* "unknown POSIX class name" */
83
REG_BADPAT, /* "POSIX collating elements are not supported" */
84
REG_INVARG, /* "this version of PCRE is not compiled with PCRE_UTF8 support" */
85
REG_BADPAT, /* "characters with values > 255 are not yet supported in classes" */
86
REG_BADPAT, /* "character value in \x{...} sequence is too large" */
87
REG_BADPAT /* "invalid condition (?(0)" */
90
/* Table of texts corresponding to POSIX error codes */
92
static const char *pstring[] = {
93
"", /* Dummy for value 0 */
94
"internal error", /* REG_ASSERT */
95
"invalid repeat counts in {}", /* BADBR */
96
"pattern error", /* BADPAT */
97
"? * + invalid", /* BADRPT */
98
"unbalanced {}", /* EBRACE */
99
"unbalanced []", /* EBRACK */
100
"collation error - not relevant", /* ECOLLATE */
101
"bad class", /* ECTYPE */
102
"bad escape sequence", /* EESCAPE */
103
"empty expression", /* EMPTY */
104
"unbalanced ()", /* EPAREN */
105
"bad range inside []", /* ERANGE */
106
"expression too big", /* ESIZE */
107
"failed to get memory", /* ESPACE */
108
"bad back reference", /* ESUBREG */
109
"bad argument", /* INVARG */
110
"match failed" /* NOMATCH */
116
/*************************************************
117
* Translate PCRE text code to int *
118
*************************************************/
120
/* PCRE compile-time errors are given as strings defined as macros. We can just
121
look them up in a table to turn them into POSIX-style error codes. */
124
pcre_posix_error_code(const char *s)
127
for (i = 0; i < sizeof(estring)/sizeof(char *); i++)
128
if (strcmp(s, estring[i]) == 0) return eint[i];
134
/*************************************************
135
* Translate error code to string *
136
*************************************************/
139
regerror(int errcode, const regex_t *preg, char *errbuf, size_t errbuf_size)
141
const char *message, *addmessage;
142
size_t length, addlength;
144
message = (errcode >= (int)(sizeof(pstring)/sizeof(char *)))?
145
"unknown error code" : pstring[errcode];
146
length = strlen(message) + 1;
148
addmessage = " at offset ";
149
addlength = (preg != NULL && (int)preg->re_erroffset != -1)?
150
strlen(addmessage) + 6 : 0;
154
if (addlength > 0 && errbuf_size >= length + addlength)
155
sprintf(errbuf, "%s%s%-6d", message, addmessage, (int)preg->re_erroffset);
158
strncpy(errbuf, message, errbuf_size - 1);
159
errbuf[errbuf_size-1] = 0;
163
return length + addlength;
169
/*************************************************
170
* Free store held by a regex *
171
*************************************************/
174
regfree(regex_t *preg)
176
(pcre_free)(preg->re_pcre);
182
/*************************************************
183
* Compile a regular expression *
184
*************************************************/
188
preg points to a structure for recording the compiled expression
189
pattern the pattern to compile
190
cflags compilation flags
192
Returns: 0 on success
193
various non-zero codes on failure
197
regcomp(regex_t *preg, const char *pattern, int cflags)
199
const char *errorptr;
203
if ((cflags & REG_ICASE) != 0) options |= PCRE_CASELESS;
204
if ((cflags & REG_NEWLINE) != 0) options |= PCRE_MULTILINE;
206
preg->re_pcre = pcre_compile(pattern, options, &errorptr, &erroffset, NULL);
207
preg->re_erroffset = erroffset;
209
if (preg->re_pcre == NULL) return pcre_posix_error_code(errorptr);
211
preg->re_nsub = pcre_info(preg->re_pcre, NULL, NULL);
218
/*************************************************
219
* Match a regular expression *
220
*************************************************/
222
/* Unfortunately, PCRE requires 3 ints of working space for each captured
223
substring, so we have to get and release working store instead of just using
224
the POSIX structures as was done in earlier releases when PCRE needed only 2
228
regexec(regex_t *preg, const char *string, size_t nmatch,
229
regmatch_t pmatch[], int eflags)
235
if ((eflags & REG_NOTBOL) != 0) options |= PCRE_NOTBOL;
236
if ((eflags & REG_NOTEOL) != 0) options |= PCRE_NOTEOL;
238
preg->re_erroffset = (size_t)(-1); /* Only has meaning after compile */
242
ovector = (int *)malloc(sizeof(int) * nmatch * 3);
243
if (ovector == NULL) return REG_ESPACE;
246
rc = pcre_exec(preg->re_pcre, NULL, string, (int)strlen(string), 0, options,
247
ovector, nmatch * 3);
249
if (rc == 0) rc = nmatch; /* All captured slots were filled in */
254
for (i = 0; i < rc; i++)
256
pmatch[i].rm_so = ovector[i*2];
257
pmatch[i].rm_eo = ovector[i*2+1];
259
if (ovector != NULL) free(ovector);
260
for (; i < nmatch; i++) pmatch[i].rm_so = pmatch[i].rm_eo = -1;
266
if (ovector != NULL) free(ovector);
269
case PCRE_ERROR_NOMATCH: return REG_NOMATCH;
270
case PCRE_ERROR_NULL: return REG_INVARG;
271
case PCRE_ERROR_BADOPTION: return REG_INVARG;
272
case PCRE_ERROR_BADMAGIC: return REG_INVARG;
273
case PCRE_ERROR_UNKNOWN_NODE: return REG_ASSERT;
274
case PCRE_ERROR_NOMEMORY: return REG_ESPACE;
275
default: return REG_ASSERT;
280
/* End of pcreposix.c */