]>
Commit | Line | Data |
---|---|---|
2b83a2a4 | 1 | /* Definitions for data structures and routines for the regular |
3b0bdc72 | 2 | expression library. |
f4cf5f2d | 3 | Copyright (C) 1985, 1989-2012 Free Software Foundation, Inc. |
3b0bdc72 | 4 | This file is part of the GNU C Library. |
2b83a2a4 | 5 | |
54d79e99 | 6 | The GNU C Library is free software; you can redistribute it and/or |
41bdb6e2 AJ |
7 | modify it under the terms of the GNU Lesser General Public |
8 | License as published by the Free Software Foundation; either | |
9 | version 2.1 of the License, or (at your option) any later version. | |
2b83a2a4 | 10 | |
54d79e99 UD |
11 | The GNU C Library is distributed in the hope that it will be useful, |
12 | but WITHOUT ANY WARRANTY; without even the implied warranty of | |
13 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU | |
41bdb6e2 | 14 | Lesser General Public License for more details. |
2b83a2a4 | 15 | |
41bdb6e2 | 16 | You should have received a copy of the GNU Lesser General Public |
59ba27a6 PE |
17 | License along with the GNU C Library; if not, see |
18 | <http://www.gnu.org/licenses/>. */ | |
2b83a2a4 | 19 | |
5107cf1d UD |
20 | #ifndef _REGEX_H |
21 | #define _REGEX_H 1 | |
2b83a2a4 | 22 | |
94c24227 UD |
23 | #include <sys/types.h> |
24 | ||
1fb05e3d UD |
25 | /* Allow the use in C++ code. */ |
26 | #ifdef __cplusplus | |
27 | extern "C" { | |
28 | #endif | |
29 | ||
4cca6b86 UD |
30 | /* The following two types have to be signed and unsigned integer type |
31 | wide enough to hold a value of a pointer. For most ANSI compilers | |
32 | ptrdiff_t and size_t should be likely OK. Still size of these two | |
33 | types is 2 for Microsoft C. Ugh... */ | |
34 | typedef long int s_reg_t; | |
35 | typedef unsigned long int active_reg_t; | |
36 | ||
2b83a2a4 RM |
37 | /* The following bits are used to determine the regexp syntax we |
38 | recognize. The set/not-set meanings are chosen so that Emacs syntax | |
39 | remains the value 0. The bits are given in alphabetical order, and | |
40 | the definitions shifted by one from the previous bit; thus, when we | |
41 | add or remove a bit, only one other definition need change. */ | |
4cca6b86 | 42 | typedef unsigned long int reg_syntax_t; |
2b83a2a4 | 43 | |
a53d3f82 | 44 | #ifdef __USE_GNU |
2b83a2a4 RM |
45 | /* If this bit is not set, then \ inside a bracket expression is literal. |
46 | If set, then such a \ quotes the following character. */ | |
a53d3f82 | 47 | # define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1) |
2b83a2a4 RM |
48 | |
49 | /* If this bit is not set, then + and ? are operators, and \+ and \? are | |
54d79e99 | 50 | literals. |
2b83a2a4 | 51 | If set, then \+ and \? are operators and + and ? are literals. */ |
a53d3f82 | 52 | # define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1) |
2b83a2a4 RM |
53 | |
54 | /* If this bit is set, then character classes are supported. They are: | |
55 | [:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:], | |
56 | [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:]. | |
57 | If not set, then character classes are not supported. */ | |
a53d3f82 | 58 | # define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1) |
2b83a2a4 RM |
59 | |
60 | /* If this bit is set, then ^ and $ are always anchors (outside bracket | |
61 | expressions, of course). | |
62 | If this bit is not set, then it depends: | |
a4b89fd8 AR |
63 | ^ is an anchor if it is at the beginning of a regular |
64 | expression or after an open-group or an alternation operator; | |
65 | $ is an anchor if it is at the end of a regular expression, or | |
66 | before a close-group or an alternation operator. | |
2b83a2a4 RM |
67 | |
68 | This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because | |
69 | POSIX draft 11.2 says that * etc. in leading positions is undefined. | |
70 | We already implemented a previous draft which made those constructs | |
71 | invalid, though, so we haven't changed the code back. */ | |
a53d3f82 | 72 | # define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1) |
2b83a2a4 RM |
73 | |
74 | /* If this bit is set, then special characters are always special | |
75 | regardless of where they are in the pattern. | |
76 | If this bit is not set, then special characters are special only in | |
54d79e99 | 77 | some contexts; otherwise they are ordinary. Specifically, |
2b83a2a4 RM |
78 | * + ? and intervals are only special when not after the beginning, |
79 | open-group, or alternation operator. */ | |
a53d3f82 | 80 | # define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1) |
2b83a2a4 RM |
81 | |
82 | /* If this bit is set, then *, +, ?, and { cannot be first in an re or | |
83 | immediately after an alternation or begin-group operator. */ | |
a53d3f82 | 84 | # define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1) |
2b83a2a4 RM |
85 | |
86 | /* If this bit is set, then . matches newline. | |
87 | If not set, then it doesn't. */ | |
a53d3f82 | 88 | # define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1) |
2b83a2a4 RM |
89 | |
90 | /* If this bit is set, then . doesn't match NUL. | |
91 | If not set, then it does. */ | |
a53d3f82 | 92 | # define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1) |
2b83a2a4 RM |
93 | |
94 | /* If this bit is set, nonmatching lists [^...] do not match newline. | |
95 | If not set, they do. */ | |
a53d3f82 | 96 | # define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1) |
2b83a2a4 RM |
97 | |
98 | /* If this bit is set, either \{...\} or {...} defines an | |
54d79e99 | 99 | interval, depending on RE_NO_BK_BRACES. |
2b83a2a4 | 100 | If not set, \{, \}, {, and } are literals. */ |
a53d3f82 | 101 | # define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1) |
2b83a2a4 RM |
102 | |
103 | /* If this bit is set, +, ? and | aren't recognized as operators. | |
104 | If not set, they are. */ | |
a53d3f82 | 105 | # define RE_LIMITED_OPS (RE_INTERVALS << 1) |
2b83a2a4 RM |
106 | |
107 | /* If this bit is set, newline is an alternation operator. | |
108 | If not set, newline is literal. */ | |
a53d3f82 | 109 | # define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1) |
2b83a2a4 RM |
110 | |
111 | /* If this bit is set, then `{...}' defines an interval, and \{ and \} | |
112 | are literals. | |
113 | If not set, then `\{...\}' defines an interval. */ | |
a53d3f82 | 114 | # define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1) |
2b83a2a4 RM |
115 | |
116 | /* If this bit is set, (...) defines a group, and \( and \) are literals. | |
117 | If not set, \(...\) defines a group, and ( and ) are literals. */ | |
a53d3f82 | 118 | # define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1) |
2b83a2a4 RM |
119 | |
120 | /* If this bit is set, then \<digit> matches <digit>. | |
121 | If not set, then \<digit> is a back-reference. */ | |
a53d3f82 | 122 | # define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1) |
2b83a2a4 | 123 | |
54d79e99 | 124 | /* If this bit is set, then | is an alternation operator, and \| is literal. |
2b83a2a4 | 125 | If not set, then \| is an alternation operator, and | is literal. */ |
a53d3f82 | 126 | # define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1) |
2b83a2a4 RM |
127 | |
128 | /* If this bit is set, then an ending range point collating higher | |
129 | than the starting range point, as in [z-a], is invalid. | |
130 | If not set, then when ending range point collates higher than the | |
131 | starting range point, the range is ignored. */ | |
a53d3f82 | 132 | # define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1) |
2b83a2a4 RM |
133 | |
134 | /* If this bit is set, then an unmatched ) is ordinary. | |
135 | If not set, then an unmatched ) is invalid. */ | |
a53d3f82 | 136 | # define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1) |
2b83a2a4 RM |
137 | |
138 | /* If this bit is set, succeed as soon as we match the whole pattern, | |
139 | without further backtracking. */ | |
a53d3f82 | 140 | # define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1) |
2b83a2a4 | 141 | |
4cca6b86 UD |
142 | /* If this bit is set, do not process the GNU regex operators. |
143 | If not set, then the GNU regex operators are recognized. */ | |
a53d3f82 | 144 | # define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1) |
4cca6b86 | 145 | |
51702635 UD |
146 | /* If this bit is set, turn on internal regex debugging. |
147 | If not set, and debugging was on, turn it off. | |
148 | This only works if regex.c is compiled -DDEBUG. | |
149 | We define this bit always, so that all that's needed to turn on | |
150 | debugging is to recompile regex.c; the calling code can always have | |
151 | this bit set, and it won't affect anything in the normal case. */ | |
a53d3f82 | 152 | # define RE_DEBUG (RE_NO_GNU_OPS << 1) |
51702635 | 153 | |
0a45b76c UD |
154 | /* If this bit is set, a syntactically invalid interval is treated as |
155 | a string of ordinary characters. For example, the ERE 'a{1' is | |
156 | treated as 'a\{1'. */ | |
a53d3f82 | 157 | # define RE_INVALID_INTERVAL_ORD (RE_DEBUG << 1) |
0a45b76c | 158 | |
3b0bdc72 UD |
159 | /* If this bit is set, then ignore case when matching. |
160 | If not set, then case is significant. */ | |
a53d3f82 | 161 | # define RE_ICASE (RE_INVALID_INTERVAL_ORD << 1) |
3b0bdc72 | 162 | |
134abcb5 UD |
163 | /* This bit is used internally like RE_CONTEXT_INDEP_ANCHORS but only |
164 | for ^, because it is difficult to scan the regex backwards to find | |
165 | whether ^ should be special. */ | |
a53d3f82 | 166 | # define RE_CARET_ANCHORS_HERE (RE_ICASE << 1) |
134abcb5 | 167 | |
06e8303a UD |
168 | /* If this bit is set, then \{ cannot be first in an bre or |
169 | immediately after an alternation or begin-group operator. */ | |
a53d3f82 | 170 | # define RE_CONTEXT_INVALID_DUP (RE_CARET_ANCHORS_HERE << 1) |
06e8303a | 171 | |
c06a6956 UD |
172 | /* If this bit is set, then no_sub will be set to 1 during |
173 | re_compile_pattern. */ | |
a53d3f82 UD |
174 | # define RE_NO_SUB (RE_CONTEXT_INVALID_DUP << 1) |
175 | #endif | |
c06a6956 | 176 | |
2b83a2a4 RM |
177 | /* This global variable defines the particular regexp syntax to use (for |
178 | some interfaces). When a regexp is compiled, the syntax used is | |
179 | stored in the pattern buffer, so changing this does not affect | |
180 | already-compiled regexps. */ | |
181 | extern reg_syntax_t re_syntax_options; | |
182 | \f | |
a53d3f82 | 183 | #ifdef __USE_GNU |
2b83a2a4 RM |
184 | /* Define combinations of the above bits for the standard possibilities. |
185 | (The [[[ comments delimit what gets put into the Texinfo file, so | |
54d79e99 | 186 | don't delete them!) */ |
2b83a2a4 RM |
187 | /* [[[begin syntaxes]]] */ |
188 | #define RE_SYNTAX_EMACS 0 | |
189 | ||
190 | #define RE_SYNTAX_AWK \ | |
4cca6b86 UD |
191 | (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \ |
192 | | RE_NO_BK_PARENS | RE_NO_BK_REFS \ | |
193 | | RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \ | |
51702635 | 194 | | RE_DOT_NEWLINE | RE_CONTEXT_INDEP_ANCHORS \ |
a4b89fd8 | 195 | | RE_CHAR_CLASSES \ |
4cca6b86 UD |
196 | | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS) |
197 | ||
198 | #define RE_SYNTAX_GNU_AWK \ | |
a4b89fd8 AR |
199 | ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS \ |
200 | | RE_INVALID_INTERVAL_ORD) \ | |
201 | & ~(RE_DOT_NOT_NULL | RE_CONTEXT_INDEP_OPS \ | |
202 | | RE_CONTEXT_INVALID_OPS )) | |
2b83a2a4 | 203 | |
92d892a8 | 204 | #define RE_SYNTAX_POSIX_AWK \ |
51702635 | 205 | (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS \ |
a4b89fd8 AR |
206 | | RE_INTERVALS | RE_NO_GNU_OPS \ |
207 | | RE_INVALID_INTERVAL_ORD) | |
2b83a2a4 RM |
208 | |
209 | #define RE_SYNTAX_GREP \ | |
210 | (RE_BK_PLUS_QM | RE_CHAR_CLASSES \ | |
211 | | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS \ | |
212 | | RE_NEWLINE_ALT) | |
213 | ||
214 | #define RE_SYNTAX_EGREP \ | |
215 | (RE_CHAR_CLASSES | RE_CONTEXT_INDEP_ANCHORS \ | |
216 | | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE \ | |
217 | | RE_NEWLINE_ALT | RE_NO_BK_PARENS \ | |
218 | | RE_NO_BK_VBAR) | |
219 | ||
220 | #define RE_SYNTAX_POSIX_EGREP \ | |
0a45b76c UD |
221 | (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES \ |
222 | | RE_INVALID_INTERVAL_ORD) | |
2b83a2a4 RM |
223 | |
224 | /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff. */ | |
225 | #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC | |
226 | ||
227 | #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC | |
228 | ||
229 | /* Syntax bits common to both basic and extended POSIX regex syntax. */ | |
230 | #define _RE_SYNTAX_POSIX_COMMON \ | |
231 | (RE_CHAR_CLASSES | RE_DOT_NEWLINE | RE_DOT_NOT_NULL \ | |
232 | | RE_INTERVALS | RE_NO_EMPTY_RANGES) | |
233 | ||
234 | #define RE_SYNTAX_POSIX_BASIC \ | |
06e8303a | 235 | (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM | RE_CONTEXT_INVALID_DUP) |
2b83a2a4 RM |
236 | |
237 | /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes | |
238 | RE_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this | |
239 | isn't minimal, since other operators, such as \`, aren't disabled. */ | |
240 | #define RE_SYNTAX_POSIX_MINIMAL_BASIC \ | |
241 | (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS) | |
242 | ||
243 | #define RE_SYNTAX_POSIX_EXTENDED \ | |
9281f45d UD |
244 | (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \ |
245 | | RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \ | |
246 | | RE_NO_BK_PARENS | RE_NO_BK_VBAR \ | |
247 | | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD) | |
2b83a2a4 | 248 | |
0e179ea6 UD |
249 | /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is |
250 | removed and RE_NO_BK_REFS is added. */ | |
2b83a2a4 RM |
251 | #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED \ |
252 | (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \ | |
253 | | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES \ | |
254 | | RE_NO_BK_PARENS | RE_NO_BK_REFS \ | |
255 | | RE_NO_BK_VBAR | RE_UNMATCHED_RIGHT_PAREN_ORD) | |
256 | /* [[[end syntaxes]]] */ | |
257 | \f | |
258 | /* Maximum number of duplicates an interval can allow. Some systems | |
259 | (erroneously) define this in other header files, but we want our | |
260 | value, so remove any previous define. */ | |
a53d3f82 UD |
261 | # ifdef RE_DUP_MAX |
262 | # undef RE_DUP_MAX | |
263 | # endif | |
4cca6b86 | 264 | /* If sizeof(int) == 2, then ((1 << 15) - 1) overflows. */ |
a53d3f82 UD |
265 | # define RE_DUP_MAX (0x7fff) |
266 | #endif | |
2b83a2a4 RM |
267 | |
268 | ||
269 | /* POSIX `cflags' bits (i.e., information for `regcomp'). */ | |
270 | ||
271 | /* If this bit is set, then use extended regular expression syntax. | |
272 | If not set, then use basic regular expression syntax. */ | |
273 | #define REG_EXTENDED 1 | |
274 | ||
275 | /* If this bit is set, then ignore case when matching. | |
276 | If not set, then case is significant. */ | |
277 | #define REG_ICASE (REG_EXTENDED << 1) | |
54d79e99 | 278 | |
2b83a2a4 RM |
279 | /* If this bit is set, then anchors do not match at newline |
280 | characters in the string. | |
281 | If not set, then anchors do match at newlines. */ | |
282 | #define REG_NEWLINE (REG_ICASE << 1) | |
283 | ||
284 | /* If this bit is set, then report only success or fail in regexec. | |
285 | If not set, then returns differ between not matching and errors. */ | |
286 | #define REG_NOSUB (REG_NEWLINE << 1) | |
287 | ||
288 | ||
289 | /* POSIX `eflags' bits (i.e., information for regexec). */ | |
290 | ||
291 | /* If this bit is set, then the beginning-of-line operator doesn't match | |
292 | the beginning of the string (presumably because it's not the | |
293 | beginning of a line). | |
294 | If not set, then the beginning-of-line operator does match the | |
295 | beginning of the string. */ | |
296 | #define REG_NOTBOL 1 | |
297 | ||
298 | /* Like REG_NOTBOL, except for the end-of-line. */ | |
299 | #define REG_NOTEOL (1 << 1) | |
300 | ||
6fefb4e0 UD |
301 | /* Use PMATCH[0] to delimit the start and end of the search in the |
302 | buffer. */ | |
303 | #define REG_STARTEND (1 << 2) | |
304 | ||
2b83a2a4 RM |
305 | |
306 | /* If any error codes are removed, changed, or added, update the | |
307 | `re_error_msg' table in regex.c. */ | |
308 | typedef enum | |
309 | { | |
6701afc7 | 310 | #if defined _XOPEN_SOURCE || defined __USE_XOPEN2K |
4bae5567 | 311 | REG_ENOSYS = -1, /* This will never happen for this implementation. */ |
9756dfe1 UD |
312 | #endif |
313 | ||
2b83a2a4 RM |
314 | REG_NOERROR = 0, /* Success. */ |
315 | REG_NOMATCH, /* Didn't find a match (for regexec). */ | |
316 | ||
317 | /* POSIX regcomp return error codes. (In the order listed in the | |
318 | standard.) */ | |
319 | REG_BADPAT, /* Invalid pattern. */ | |
df759c2a | 320 | REG_ECOLLATE, /* Inalid collating element. */ |
2b83a2a4 RM |
321 | REG_ECTYPE, /* Invalid character class name. */ |
322 | REG_EESCAPE, /* Trailing backslash. */ | |
323 | REG_ESUBREG, /* Invalid back reference. */ | |
324 | REG_EBRACK, /* Unmatched left bracket. */ | |
54d79e99 | 325 | REG_EPAREN, /* Parenthesis imbalance. */ |
2b83a2a4 RM |
326 | REG_EBRACE, /* Unmatched \{. */ |
327 | REG_BADBR, /* Invalid contents of \{\}. */ | |
328 | REG_ERANGE, /* Invalid range end. */ | |
329 | REG_ESPACE, /* Ran out of memory. */ | |
330 | REG_BADRPT, /* No preceding re for repetition op. */ | |
331 | ||
332 | /* Error codes we've added. */ | |
333 | REG_EEND, /* Premature end. */ | |
334 | REG_ESIZE, /* Compiled pattern bigger than 2^16 bytes. */ | |
335 | REG_ERPAREN /* Unmatched ) or \); not returned from regcomp. */ | |
336 | } reg_errcode_t; | |
337 | \f | |
338 | /* This data structure represents a compiled pattern. Before calling | |
339 | the pattern compiler, the fields `buffer', `allocated', `fastmap', | |
78e64fdc RT |
340 | and `translate' can be set. After the pattern has been compiled, |
341 | the fields `re_nsub', `not_bol' and `not_eol' are available. All | |
342 | other fields are private to the regex routines. */ | |
2b83a2a4 | 343 | |
54d79e99 | 344 | #ifndef RE_TRANSLATE_TYPE |
a53d3f82 UD |
345 | # define __RE_TRANSLATE_TYPE unsigned char * |
346 | # ifdef __USE_GNU | |
347 | # define RE_TRANSLATE_TYPE __RE_TRANSLATE_TYPE | |
348 | # endif | |
349 | #endif | |
350 | ||
351 | #ifdef __USE_GNU | |
352 | # define __REPB_PREFIX(name) name | |
353 | #else | |
354 | # define __REPB_PREFIX(name) __##name | |
03a75825 RM |
355 | #endif |
356 | ||
2b83a2a4 RM |
357 | struct re_pattern_buffer |
358 | { | |
2ff89ea4 UD |
359 | /* Space that holds the compiled pattern. It is declared as |
360 | `unsigned char *' because its elements are sometimes used as | |
361 | array indexes. */ | |
a53d3f82 | 362 | unsigned char *__REPB_PREFIX(buffer); |
2b83a2a4 | 363 | |
2ff89ea4 | 364 | /* Number of bytes to which `buffer' points. */ |
a53d3f82 | 365 | unsigned long int __REPB_PREFIX(allocated); |
2b83a2a4 | 366 | |
2ff89ea4 | 367 | /* Number of bytes actually used in `buffer'. */ |
a53d3f82 | 368 | unsigned long int __REPB_PREFIX(used); |
2b83a2a4 | 369 | |
2ff89ea4 | 370 | /* Syntax setting with which the pattern was compiled. */ |
a53d3f82 | 371 | reg_syntax_t __REPB_PREFIX(syntax); |
2b83a2a4 | 372 | |
2ff89ea4 UD |
373 | /* Pointer to a fastmap, if any, otherwise zero. re_search uses the |
374 | fastmap, if there is one, to skip over impossible starting points | |
375 | for matches. */ | |
a53d3f82 | 376 | char *__REPB_PREFIX(fastmap); |
2b83a2a4 | 377 | |
2ff89ea4 UD |
378 | /* Either a translate table to apply to all characters before |
379 | comparing them, or zero for no translation. The translation is | |
380 | applied to a pattern when it is compiled and to a string when it | |
381 | is matched. */ | |
a53d3f82 | 382 | __RE_TRANSLATE_TYPE __REPB_PREFIX(translate); |
2b83a2a4 | 383 | |
2ff89ea4 | 384 | /* Number of subexpressions found by the compiler. */ |
2b83a2a4 RM |
385 | size_t re_nsub; |
386 | ||
2ff89ea4 UD |
387 | /* Zero if this pattern cannot match the empty string, one else. |
388 | Well, in truth it's used only in `re_search_2', to see whether or | |
389 | not we should use the fastmap, so we don't set this absolutely | |
390 | perfectly; see `re_compile_fastmap' (the `duplicate' case). */ | |
a53d3f82 | 391 | unsigned __REPB_PREFIX(can_be_null) : 1; |
2b83a2a4 | 392 | |
2ff89ea4 UD |
393 | /* If REGS_UNALLOCATED, allocate space in the `regs' structure |
394 | for `max (RE_NREGS, re_nsub + 1)' groups. | |
395 | If REGS_REALLOCATE, reallocate space if necessary. | |
396 | If REGS_FIXED, use what's there. */ | |
a53d3f82 UD |
397 | #ifdef __USE_GNU |
398 | # define REGS_UNALLOCATED 0 | |
399 | # define REGS_REALLOCATE 1 | |
400 | # define REGS_FIXED 2 | |
401 | #endif | |
402 | unsigned __REPB_PREFIX(regs_allocated) : 2; | |
2b83a2a4 | 403 | |
2ff89ea4 UD |
404 | /* Set to zero when `regex_compile' compiles a pattern; set to one |
405 | by `re_compile_fastmap' if it updates the fastmap. */ | |
a53d3f82 | 406 | unsigned __REPB_PREFIX(fastmap_accurate) : 1; |
2b83a2a4 | 407 | |
2ff89ea4 UD |
408 | /* If set, `re_match_2' does not return information about |
409 | subexpressions. */ | |
a53d3f82 | 410 | unsigned __REPB_PREFIX(no_sub) : 1; |
2b83a2a4 | 411 | |
2ff89ea4 UD |
412 | /* If set, a beginning-of-line anchor doesn't match at the beginning |
413 | of the string. */ | |
a53d3f82 | 414 | unsigned __REPB_PREFIX(not_bol) : 1; |
2b83a2a4 | 415 | |
2ff89ea4 | 416 | /* Similarly for an end-of-line anchor. */ |
a53d3f82 | 417 | unsigned __REPB_PREFIX(not_eol) : 1; |
2b83a2a4 | 418 | |
2ff89ea4 | 419 | /* If true, an anchor at a newline matches. */ |
a53d3f82 | 420 | unsigned __REPB_PREFIX(newline_anchor) : 1; |
2b83a2a4 RM |
421 | }; |
422 | ||
423 | typedef struct re_pattern_buffer regex_t; | |
424 | \f | |
425 | /* Type for byte offsets within the string. POSIX mandates this. */ | |
426 | typedef int regoff_t; | |
427 | ||
428 | ||
a53d3f82 | 429 | #ifdef __USE_GNU |
2b83a2a4 RM |
430 | /* This is the structure we store register match data in. See |
431 | regex.texinfo for a full description of what registers match. */ | |
432 | struct re_registers | |
433 | { | |
434 | unsigned num_regs; | |
435 | regoff_t *start; | |
436 | regoff_t *end; | |
437 | }; | |
438 | ||
439 | ||
440 | /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer, | |
441 | `re_match_2' returns information about at least this many registers | |
442 | the first time a `regs' structure is passed. */ | |
a53d3f82 UD |
443 | # ifndef RE_NREGS |
444 | # define RE_NREGS 30 | |
445 | # endif | |
2b83a2a4 RM |
446 | #endif |
447 | ||
448 | ||
449 | /* POSIX specification for registers. Aside from the different names than | |
450 | `re_registers', POSIX uses an array of structures, instead of a | |
451 | structure of arrays. */ | |
452 | typedef struct | |
453 | { | |
454 | regoff_t rm_so; /* Byte offset from string's start to substring's start. */ | |
455 | regoff_t rm_eo; /* Byte offset from string's start to substring's end. */ | |
456 | } regmatch_t; | |
457 | \f | |
458 | /* Declarations for routines. */ | |
459 | ||
a53d3f82 | 460 | #ifdef __USE_GNU |
2b83a2a4 RM |
461 | /* Sets the current default syntax to SYNTAX, and return the old syntax. |
462 | You can also simply assign to the `re_syntax_options' variable. */ | |
2ff89ea4 | 463 | extern reg_syntax_t re_set_syntax (reg_syntax_t __syntax); |
2b83a2a4 RM |
464 | |
465 | /* Compile the regular expression PATTERN, with length LENGTH | |
466 | and syntax given by the global `re_syntax_options', into the buffer | |
78e64fdc RT |
467 | BUFFER. Return NULL if successful, and an error string if not. |
468 | ||
469 | To free the allocated storage, you must call `regfree' on BUFFER. | |
470 | Note that the translate table must either have been initialised by | |
471 | `regcomp', with a malloc'ed value, or set to NULL before calling | |
472 | `regfree'. */ | |
2ff89ea4 UD |
473 | extern const char *re_compile_pattern (const char *__pattern, size_t __length, |
474 | struct re_pattern_buffer *__buffer); | |
2b83a2a4 RM |
475 | |
476 | ||
477 | /* Compile a fastmap for the compiled pattern in BUFFER; used to | |
478 | accelerate searches. Return 0 if successful and -2 if was an | |
479 | internal error. */ | |
2ff89ea4 | 480 | extern int re_compile_fastmap (struct re_pattern_buffer *__buffer); |
2b83a2a4 RM |
481 | |
482 | ||
483 | /* Search in the string STRING (with length LENGTH) for the pattern | |
484 | compiled into BUFFER. Start searching at position START, for RANGE | |
485 | characters. Return the starting position of the match, -1 for no | |
486 | match, or -2 for an internal error. Also return register | |
487 | information in REGS (if REGS and BUFFER->no_sub are nonzero). */ | |
2ff89ea4 UD |
488 | extern int re_search (struct re_pattern_buffer *__buffer, const char *__string, |
489 | int __length, int __start, int __range, | |
490 | struct re_registers *__regs); | |
2b83a2a4 RM |
491 | |
492 | ||
493 | /* Like `re_search', but search in the concatenation of STRING1 and | |
494 | STRING2. Also, stop searching at index START + STOP. */ | |
2ff89ea4 UD |
495 | extern int re_search_2 (struct re_pattern_buffer *__buffer, |
496 | const char *__string1, int __length1, | |
497 | const char *__string2, int __length2, int __start, | |
498 | int __range, struct re_registers *__regs, int __stop); | |
2b83a2a4 RM |
499 | |
500 | ||
501 | /* Like `re_search', but return how many characters in STRING the regexp | |
502 | in BUFFER matched, starting at position START. */ | |
2ff89ea4 UD |
503 | extern int re_match (struct re_pattern_buffer *__buffer, const char *__string, |
504 | int __length, int __start, struct re_registers *__regs); | |
2b83a2a4 RM |
505 | |
506 | ||
507 | /* Relates to `re_match' as `re_search_2' relates to `re_search'. */ | |
2ff89ea4 UD |
508 | extern int re_match_2 (struct re_pattern_buffer *__buffer, |
509 | const char *__string1, int __length1, | |
510 | const char *__string2, int __length2, int __start, | |
511 | struct re_registers *__regs, int __stop); | |
2b83a2a4 RM |
512 | |
513 | ||
514 | /* Set REGS to hold NUM_REGS registers, storing them in STARTS and | |
515 | ENDS. Subsequent matches using BUFFER and REGS will use this memory | |
516 | for recording register information. STARTS and ENDS must be | |
517 | allocated with malloc, and must each be at least `NUM_REGS * sizeof | |
518 | (regoff_t)' bytes long. | |
519 | ||
520 | If NUM_REGS == 0, then subsequent matches should allocate their own | |
521 | register data. | |
522 | ||
523 | Unless this function is called, the first search or match using | |
524 | PATTERN_BUFFER will allocate its own register data, without | |
525 | freeing the old data. */ | |
2ff89ea4 UD |
526 | extern void re_set_registers (struct re_pattern_buffer *__buffer, |
527 | struct re_registers *__regs, | |
528 | unsigned int __num_regs, | |
529 | regoff_t *__starts, regoff_t *__ends); | |
a53d3f82 | 530 | #endif /* Use GNU */ |
2b83a2a4 | 531 | |
a53d3f82 | 532 | #if defined _REGEX_RE_COMP || (defined _LIBC && defined __USE_BSD) |
6973fc01 | 533 | # ifndef _CRAY |
2b83a2a4 | 534 | /* 4.2 bsd compatibility. */ |
2d87db5b UD |
535 | extern char *re_comp (const char *); |
536 | extern int re_exec (const char *); | |
6973fc01 | 537 | # endif |
4cca6b86 | 538 | #endif |
2b83a2a4 | 539 | |
7ca404ad UD |
540 | /* GCC 2.95 and later have "__restrict"; C99 compilers have |
541 | "restrict", and "configure" may have defined "restrict". */ | |
542 | #ifndef __restrict | |
543 | # if ! (2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__)) | |
544 | # if defined restrict || 199901L <= __STDC_VERSION__ | |
545 | # define __restrict restrict | |
546 | # else | |
547 | # define __restrict | |
548 | # endif | |
549 | # endif | |
da2a3ca6 | 550 | #endif |
8343eaee UD |
551 | /* gcc 3.1 and up support the [restrict] syntax. */ |
552 | #ifndef __restrict_arr | |
5631e740 UD |
553 | # if (__GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ >= 1)) \ |
554 | && !defined __GNUG__ | |
8343eaee UD |
555 | # define __restrict_arr __restrict |
556 | # else | |
557 | # define __restrict_arr | |
558 | # endif | |
559 | #endif | |
7ca404ad | 560 | |
2b83a2a4 | 561 | /* POSIX compatibility. */ |
2d87db5b UD |
562 | extern int regcomp (regex_t *__restrict __preg, |
563 | const char *__restrict __pattern, | |
564 | int __cflags); | |
30baa360 | 565 | |
2d87db5b UD |
566 | extern int regexec (const regex_t *__restrict __preg, |
567 | const char *__restrict __string, size_t __nmatch, | |
568 | regmatch_t __pmatch[__restrict_arr], | |
569 | int __eflags); | |
30baa360 | 570 | |
2d87db5b UD |
571 | extern size_t regerror (int __errcode, const regex_t *__restrict __preg, |
572 | char *__restrict __errbuf, size_t __errbuf_size); | |
30baa360 | 573 | |
2d87db5b | 574 | extern void regfree (regex_t *__preg); |
2b83a2a4 | 575 | |
1fb05e3d UD |
576 | |
577 | #ifdef __cplusplus | |
578 | } | |
579 | #endif /* C++ */ | |
580 | ||
5107cf1d | 581 | #endif /* regex.h */ |