Line data Source code
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) 1997-2000 University of Cambridge
16 :
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
20 : restrictions:
21 :
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.
25 :
26 : 2. The origin of this software must not be misrepresented, either by
27 : explicit claim or by omission.
28 :
29 : 3. Altered versions must be plainly marked as such, and must not be
30 : misrepresented as being the original software.
31 :
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 : -----------------------------------------------------------------------------
36 : */
37 :
38 : #include "internal.h"
39 : #include "pcreposix.h"
40 : #include "stdlib.h"
41 :
42 :
43 :
44 : /* Corresponding tables of PCRE error messages and POSIX error codes. */
45 :
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,
50 : ERR31 };
51 :
52 : static int eint[] = {
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)" */
88 : };
89 :
90 : /* Table of texts corresponding to POSIX error codes */
91 :
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 */
111 : };
112 :
113 :
114 :
115 :
116 : /*************************************************
117 : * Translate PCRE text code to int *
118 : *************************************************/
119 :
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. */
122 :
123 : static int
124 0 : pcre_posix_error_code(const char *s)
125 : {
126 : size_t i;
127 0 : for (i = 0; i < sizeof(estring)/sizeof(char *); i++)
128 0 : if (strcmp(s, estring[i]) == 0) return eint[i];
129 0 : return REG_ASSERT;
130 : }
131 :
132 :
133 :
134 : /*************************************************
135 : * Translate error code to string *
136 : *************************************************/
137 :
138 : size_t
139 0 : regerror(int errcode, const regex_t *preg, char *errbuf, size_t errbuf_size)
140 : {
141 : const char *message, *addmessage;
142 : size_t length, addlength;
143 :
144 0 : message = (errcode >= (int)(sizeof(pstring)/sizeof(char *)))?
145 0 : "unknown error code" : pstring[errcode];
146 0 : length = strlen(message) + 1;
147 :
148 0 : addmessage = " at offset ";
149 0 : addlength = (preg != NULL && (int)preg->re_erroffset != -1)?
150 0 : strlen(addmessage) + 6 : 0;
151 :
152 0 : if (errbuf_size > 0)
153 : {
154 0 : if (addlength > 0 && errbuf_size >= length + addlength)
155 0 : sprintf(errbuf, "%s%s%-6d", message, addmessage, (int)preg->re_erroffset);
156 : else
157 : {
158 0 : strncpy(errbuf, message, errbuf_size - 1);
159 0 : errbuf[errbuf_size-1] = 0;
160 : }
161 : }
162 :
163 0 : return length + addlength;
164 : }
165 :
166 :
167 :
168 :
169 : /*************************************************
170 : * Free store held by a regex *
171 : *************************************************/
172 :
173 : void
174 0 : regfree(regex_t *preg)
175 : {
176 0 : (pcre_free)(preg->re_pcre);
177 0 : }
178 :
179 :
180 :
181 :
182 : /*************************************************
183 : * Compile a regular expression *
184 : *************************************************/
185 :
186 : /*
187 : Arguments:
188 : preg points to a structure for recording the compiled expression
189 : pattern the pattern to compile
190 : cflags compilation flags
191 :
192 : Returns: 0 on success
193 : various non-zero codes on failure
194 : */
195 :
196 : int
197 0 : regcomp(regex_t *preg, const char *pattern, int cflags)
198 : {
199 : const char *errorptr;
200 : int erroffset;
201 0 : int options = 0;
202 :
203 0 : if ((cflags & REG_ICASE) != 0) options |= PCRE_CASELESS;
204 0 : if ((cflags & REG_NEWLINE) != 0) options |= PCRE_MULTILINE;
205 :
206 0 : preg->re_pcre = pcre_compile(pattern, options, &errorptr, &erroffset, NULL);
207 0 : preg->re_erroffset = erroffset;
208 :
209 0 : if (preg->re_pcre == NULL) return pcre_posix_error_code(errorptr);
210 :
211 0 : preg->re_nsub = pcre_info(preg->re_pcre, NULL, NULL);
212 0 : return 0;
213 : }
214 :
215 :
216 :
217 :
218 : /*************************************************
219 : * Match a regular expression *
220 : *************************************************/
221 :
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
225 : ints. */
226 :
227 : int
228 0 : regexec(regex_t *preg, const char *string, size_t nmatch,
229 : regmatch_t pmatch[], int eflags)
230 : {
231 : int rc;
232 0 : int options = 0;
233 0 : int *ovector = NULL;
234 :
235 0 : if ((eflags & REG_NOTBOL) != 0) options |= PCRE_NOTBOL;
236 0 : if ((eflags & REG_NOTEOL) != 0) options |= PCRE_NOTEOL;
237 :
238 0 : preg->re_erroffset = (size_t)(-1); /* Only has meaning after compile */
239 :
240 0 : if (nmatch > 0)
241 : {
242 0 : ovector = (int *)malloc(sizeof(int) * nmatch * 3);
243 0 : if (ovector == NULL) return REG_ESPACE;
244 : }
245 :
246 0 : rc = pcre_exec(preg->re_pcre, NULL, string, (int)strlen(string), 0, options,
247 : ovector, nmatch * 3);
248 :
249 0 : if (rc == 0) rc = nmatch; /* All captured slots were filled in */
250 :
251 0 : if (rc >= 0)
252 : {
253 : size_t i;
254 0 : for (i = 0; i < (size_t)rc; i++)
255 : {
256 0 : pmatch[i].rm_so = ovector[i*2];
257 0 : pmatch[i].rm_eo = ovector[i*2+1];
258 : }
259 0 : if (ovector != NULL) free(ovector);
260 0 : for (; i < nmatch; i++) pmatch[i].rm_so = pmatch[i].rm_eo = -1;
261 0 : return 0;
262 : }
263 :
264 : else
265 : {
266 0 : if (ovector != NULL) free(ovector);
267 0 : switch(rc)
268 : {
269 0 : case PCRE_ERROR_NOMATCH: return REG_NOMATCH;
270 0 : case PCRE_ERROR_NULL: return REG_INVARG;
271 0 : case PCRE_ERROR_BADOPTION: return REG_INVARG;
272 0 : case PCRE_ERROR_BADMAGIC: return REG_INVARG;
273 0 : case PCRE_ERROR_UNKNOWN_NODE: return REG_ASSERT;
274 0 : case PCRE_ERROR_NOMEMORY: return REG_ESPACE;
275 0 : default: return REG_ASSERT;
276 : }
277 : }
278 : }
279 :
280 : /* End of pcreposix.c */
|
