]>
Commit | Line | Data |
---|---|---|
2b83a2a4 | 1 | /* Definitions for data structures and routines for the regular |
3b0bdc72 | 2 | expression library. |
94c24227 | 3 | Copyright (C) 1985,1989-93,1995-98,2000,2001,2002,2003 |
3b0bdc72 UD |
4 | Free Software Foundation, Inc. |
5 | This file is part of the GNU C Library. | |
2b83a2a4 | 6 | |
54d79e99 | 7 | The GNU C Library is free software; you can redistribute it and/or |
41bdb6e2 AJ |
8 | modify it under the terms of the GNU Lesser General Public |
9 | License as published by the Free Software Foundation; either | |
10 | version 2.1 of the License, or (at your option) any later version. | |
2b83a2a4 | 11 | |
54d79e99 UD |
12 | The GNU C Library is distributed in the hope that it will be useful, |
13 | but WITHOUT ANY WARRANTY; without even the implied warranty of | |
14 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU | |
41bdb6e2 | 15 | Lesser General Public License for more details. |
2b83a2a4 | 16 | |
41bdb6e2 AJ |
17 | You should have received a copy of the GNU Lesser General Public |
18 | License along with the GNU C Library; if not, write to the Free | |
19 | Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA | |
20 | 02111-1307 USA. */ | |
2b83a2a4 | 21 | |
5107cf1d UD |
22 | #ifndef _REGEX_H |
23 | #define _REGEX_H 1 | |
2b83a2a4 | 24 | |
94c24227 UD |
25 | #include <sys/types.h> |
26 | ||
1fb05e3d UD |
27 | /* Allow the use in C++ code. */ |
28 | #ifdef __cplusplus | |
29 | extern "C" { | |
30 | #endif | |
31 | ||
2b83a2a4 RM |
32 | /* POSIX says that <sys/types.h> must be included (by the caller) before |
33 | <regex.h>. */ | |
34 | ||
6973fc01 | 35 | #if !defined _POSIX_C_SOURCE && !defined _POSIX_SOURCE && defined VMS |
2b83a2a4 RM |
36 | /* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it |
37 | should be there. */ | |
6973fc01 | 38 | # include <stddef.h> |
2b83a2a4 RM |
39 | #endif |
40 | ||
4cca6b86 UD |
41 | /* The following two types have to be signed and unsigned integer type |
42 | wide enough to hold a value of a pointer. For most ANSI compilers | |
43 | ptrdiff_t and size_t should be likely OK. Still size of these two | |
44 | types is 2 for Microsoft C. Ugh... */ | |
45 | typedef long int s_reg_t; | |
46 | typedef unsigned long int active_reg_t; | |
47 | ||
2b83a2a4 RM |
48 | /* The following bits are used to determine the regexp syntax we |
49 | recognize. The set/not-set meanings are chosen so that Emacs syntax | |
50 | remains the value 0. The bits are given in alphabetical order, and | |
51 | the definitions shifted by one from the previous bit; thus, when we | |
52 | add or remove a bit, only one other definition need change. */ | |
4cca6b86 | 53 | typedef unsigned long int reg_syntax_t; |
2b83a2a4 RM |
54 | |
55 | /* If this bit is not set, then \ inside a bracket expression is literal. | |
56 | If set, then such a \ quotes the following character. */ | |
4cca6b86 | 57 | #define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1) |
2b83a2a4 RM |
58 | |
59 | /* If this bit is not set, then + and ? are operators, and \+ and \? are | |
54d79e99 | 60 | literals. |
2b83a2a4 RM |
61 | If set, then \+ and \? are operators and + and ? are literals. */ |
62 | #define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1) | |
63 | ||
64 | /* If this bit is set, then character classes are supported. They are: | |
65 | [:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:], | |
66 | [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:]. | |
67 | If not set, then character classes are not supported. */ | |
68 | #define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1) | |
69 | ||
70 | /* If this bit is set, then ^ and $ are always anchors (outside bracket | |
71 | expressions, of course). | |
72 | If this bit is not set, then it depends: | |
73 | ^ is an anchor if it is at the beginning of a regular | |
74 | expression or after an open-group or an alternation operator; | |
75 | $ is an anchor if it is at the end of a regular expression, or | |
54d79e99 | 76 | before a close-group or an alternation operator. |
2b83a2a4 RM |
77 | |
78 | This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because | |
79 | POSIX draft 11.2 says that * etc. in leading positions is undefined. | |
80 | We already implemented a previous draft which made those constructs | |
81 | invalid, though, so we haven't changed the code back. */ | |
82 | #define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1) | |
83 | ||
84 | /* If this bit is set, then special characters are always special | |
85 | regardless of where they are in the pattern. | |
86 | If this bit is not set, then special characters are special only in | |
54d79e99 | 87 | some contexts; otherwise they are ordinary. Specifically, |
2b83a2a4 RM |
88 | * + ? and intervals are only special when not after the beginning, |
89 | open-group, or alternation operator. */ | |
90 | #define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1) | |
91 | ||
92 | /* If this bit is set, then *, +, ?, and { cannot be first in an re or | |
93 | immediately after an alternation or begin-group operator. */ | |
94 | #define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1) | |
95 | ||
96 | /* If this bit is set, then . matches newline. | |
97 | If not set, then it doesn't. */ | |
98 | #define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1) | |
99 | ||
100 | /* If this bit is set, then . doesn't match NUL. | |
101 | If not set, then it does. */ | |
102 | #define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1) | |
103 | ||
104 | /* If this bit is set, nonmatching lists [^...] do not match newline. | |
105 | If not set, they do. */ | |
106 | #define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1) | |
107 | ||
108 | /* If this bit is set, either \{...\} or {...} defines an | |
54d79e99 | 109 | interval, depending on RE_NO_BK_BRACES. |
2b83a2a4 RM |
110 | If not set, \{, \}, {, and } are literals. */ |
111 | #define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1) | |
112 | ||
113 | /* If this bit is set, +, ? and | aren't recognized as operators. | |
114 | If not set, they are. */ | |
115 | #define RE_LIMITED_OPS (RE_INTERVALS << 1) | |
116 | ||
117 | /* If this bit is set, newline is an alternation operator. | |
118 | If not set, newline is literal. */ | |
119 | #define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1) | |
120 | ||
121 | /* If this bit is set, then `{...}' defines an interval, and \{ and \} | |
122 | are literals. | |
123 | If not set, then `\{...\}' defines an interval. */ | |
124 | #define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1) | |
125 | ||
126 | /* If this bit is set, (...) defines a group, and \( and \) are literals. | |
127 | If not set, \(...\) defines a group, and ( and ) are literals. */ | |
128 | #define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1) | |
129 | ||
130 | /* If this bit is set, then \<digit> matches <digit>. | |
131 | If not set, then \<digit> is a back-reference. */ | |
132 | #define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1) | |
133 | ||
54d79e99 | 134 | /* If this bit is set, then | is an alternation operator, and \| is literal. |
2b83a2a4 RM |
135 | If not set, then \| is an alternation operator, and | is literal. */ |
136 | #define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1) | |
137 | ||
138 | /* If this bit is set, then an ending range point collating higher | |
139 | than the starting range point, as in [z-a], is invalid. | |
140 | If not set, then when ending range point collates higher than the | |
141 | starting range point, the range is ignored. */ | |
142 | #define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1) | |
143 | ||
144 | /* If this bit is set, then an unmatched ) is ordinary. | |
145 | If not set, then an unmatched ) is invalid. */ | |
146 | #define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1) | |
147 | ||
148 | /* If this bit is set, succeed as soon as we match the whole pattern, | |
149 | without further backtracking. */ | |
150 | #define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1) | |
151 | ||
4cca6b86 UD |
152 | /* If this bit is set, do not process the GNU regex operators. |
153 | If not set, then the GNU regex operators are recognized. */ | |
154 | #define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1) | |
155 | ||
51702635 UD |
156 | /* If this bit is set, turn on internal regex debugging. |
157 | If not set, and debugging was on, turn it off. | |
158 | This only works if regex.c is compiled -DDEBUG. | |
159 | We define this bit always, so that all that's needed to turn on | |
160 | debugging is to recompile regex.c; the calling code can always have | |
161 | this bit set, and it won't affect anything in the normal case. */ | |
162 | #define RE_DEBUG (RE_NO_GNU_OPS << 1) | |
163 | ||
0a45b76c UD |
164 | /* If this bit is set, a syntactically invalid interval is treated as |
165 | a string of ordinary characters. For example, the ERE 'a{1' is | |
166 | treated as 'a\{1'. */ | |
167 | #define RE_INVALID_INTERVAL_ORD (RE_DEBUG << 1) | |
168 | ||
3b0bdc72 UD |
169 | /* If this bit is set, then ignore case when matching. |
170 | If not set, then case is significant. */ | |
171 | #define RE_ICASE (RE_INVALID_INTERVAL_ORD << 1) | |
172 | ||
134abcb5 UD |
173 | /* This bit is used internally like RE_CONTEXT_INDEP_ANCHORS but only |
174 | for ^, because it is difficult to scan the regex backwards to find | |
175 | whether ^ should be special. */ | |
176 | #define RE_CARET_ANCHORS_HERE (RE_ICASE << 1) | |
177 | ||
06e8303a UD |
178 | /* If this bit is set, then \{ cannot be first in an bre or |
179 | immediately after an alternation or begin-group operator. */ | |
180 | #define RE_CONTEXT_INVALID_DUP (RE_CARET_ANCHORS_HERE << 1) | |
181 | ||
2b83a2a4 RM |
182 | /* This global variable defines the particular regexp syntax to use (for |
183 | some interfaces). When a regexp is compiled, the syntax used is | |
184 | stored in the pattern buffer, so changing this does not affect | |
185 | already-compiled regexps. */ | |
186 | extern reg_syntax_t re_syntax_options; | |
187 | \f | |
188 | /* Define combinations of the above bits for the standard possibilities. | |
189 | (The [[[ comments delimit what gets put into the Texinfo file, so | |
54d79e99 | 190 | don't delete them!) */ |
2b83a2a4 RM |
191 | /* [[[begin syntaxes]]] */ |
192 | #define RE_SYNTAX_EMACS 0 | |
193 | ||
194 | #define RE_SYNTAX_AWK \ | |
4cca6b86 UD |
195 | (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \ |
196 | | RE_NO_BK_PARENS | RE_NO_BK_REFS \ | |
197 | | RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \ | |
51702635 | 198 | | RE_DOT_NEWLINE | RE_CONTEXT_INDEP_ANCHORS \ |
4cca6b86 UD |
199 | | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS) |
200 | ||
201 | #define RE_SYNTAX_GNU_AWK \ | |
51702635 | 202 | ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DEBUG) \ |
c0a0f9a3 UD |
203 | & ~(RE_DOT_NOT_NULL | RE_INTERVALS | RE_CONTEXT_INDEP_OPS \ |
204 | | RE_CONTEXT_INVALID_OPS )) | |
2b83a2a4 RM |
205 | |
206 | #define RE_SYNTAX_POSIX_AWK \ | |
51702635 UD |
207 | (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS \ |
208 | | RE_INTERVALS | RE_NO_GNU_OPS) | |
2b83a2a4 RM |
209 | |
210 | #define RE_SYNTAX_GREP \ | |
211 | (RE_BK_PLUS_QM | RE_CHAR_CLASSES \ | |
212 | | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS \ | |
213 | | RE_NEWLINE_ALT) | |
214 | ||
215 | #define RE_SYNTAX_EGREP \ | |
216 | (RE_CHAR_CLASSES | RE_CONTEXT_INDEP_ANCHORS \ | |
217 | | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE \ | |
218 | | RE_NEWLINE_ALT | RE_NO_BK_PARENS \ | |
219 | | RE_NO_BK_VBAR) | |
220 | ||
221 | #define RE_SYNTAX_POSIX_EGREP \ | |
0a45b76c UD |
222 | (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES \ |
223 | | RE_INVALID_INTERVAL_ORD) | |
2b83a2a4 RM |
224 | |
225 | /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff. */ | |
226 | #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC | |
227 | ||
228 | #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC | |
229 | ||
230 | /* Syntax bits common to both basic and extended POSIX regex syntax. */ | |
231 | #define _RE_SYNTAX_POSIX_COMMON \ | |
232 | (RE_CHAR_CLASSES | RE_DOT_NEWLINE | RE_DOT_NOT_NULL \ | |
233 | | RE_INTERVALS | RE_NO_EMPTY_RANGES) | |
234 | ||
235 | #define RE_SYNTAX_POSIX_BASIC \ | |
06e8303a | 236 | (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM | RE_CONTEXT_INVALID_DUP) |
2b83a2a4 RM |
237 | |
238 | /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes | |
239 | RE_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this | |
240 | isn't minimal, since other operators, such as \`, aren't disabled. */ | |
241 | #define RE_SYNTAX_POSIX_MINIMAL_BASIC \ | |
242 | (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS) | |
243 | ||
244 | #define RE_SYNTAX_POSIX_EXTENDED \ | |
9281f45d UD |
245 | (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \ |
246 | | RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \ | |
247 | | RE_NO_BK_PARENS | RE_NO_BK_VBAR \ | |
248 | | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD) | |
2b83a2a4 | 249 | |
0e179ea6 UD |
250 | /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is |
251 | removed and RE_NO_BK_REFS is added. */ | |
2b83a2a4 RM |
252 | #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED \ |
253 | (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \ | |
254 | | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES \ | |
255 | | RE_NO_BK_PARENS | RE_NO_BK_REFS \ | |
256 | | RE_NO_BK_VBAR | RE_UNMATCHED_RIGHT_PAREN_ORD) | |
257 | /* [[[end syntaxes]]] */ | |
258 | \f | |
259 | /* Maximum number of duplicates an interval can allow. Some systems | |
260 | (erroneously) define this in other header files, but we want our | |
261 | value, so remove any previous define. */ | |
262 | #ifdef RE_DUP_MAX | |
6973fc01 | 263 | # undef RE_DUP_MAX |
2b83a2a4 | 264 | #endif |
4cca6b86 UD |
265 | /* If sizeof(int) == 2, then ((1 << 15) - 1) overflows. */ |
266 | #define RE_DUP_MAX (0x7fff) | |
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 | ||
301 | ||
302 | /* If any error codes are removed, changed, or added, update the | |
303 | `re_error_msg' table in regex.c. */ | |
304 | typedef enum | |
305 | { | |
e918a7fe | 306 | #ifdef _XOPEN_SOURCE |
4bae5567 | 307 | REG_ENOSYS = -1, /* This will never happen for this implementation. */ |
9756dfe1 UD |
308 | #endif |
309 | ||
2b83a2a4 RM |
310 | REG_NOERROR = 0, /* Success. */ |
311 | REG_NOMATCH, /* Didn't find a match (for regexec). */ | |
312 | ||
313 | /* POSIX regcomp return error codes. (In the order listed in the | |
314 | standard.) */ | |
315 | REG_BADPAT, /* Invalid pattern. */ | |
316 | REG_ECOLLATE, /* Not implemented. */ | |
317 | REG_ECTYPE, /* Invalid character class name. */ | |
318 | REG_EESCAPE, /* Trailing backslash. */ | |
319 | REG_ESUBREG, /* Invalid back reference. */ | |
320 | REG_EBRACK, /* Unmatched left bracket. */ | |
54d79e99 | 321 | REG_EPAREN, /* Parenthesis imbalance. */ |
2b83a2a4 RM |
322 | REG_EBRACE, /* Unmatched \{. */ |
323 | REG_BADBR, /* Invalid contents of \{\}. */ | |
324 | REG_ERANGE, /* Invalid range end. */ | |
325 | REG_ESPACE, /* Ran out of memory. */ | |
326 | REG_BADRPT, /* No preceding re for repetition op. */ | |
327 | ||
328 | /* Error codes we've added. */ | |
329 | REG_EEND, /* Premature end. */ | |
330 | REG_ESIZE, /* Compiled pattern bigger than 2^16 bytes. */ | |
331 | REG_ERPAREN /* Unmatched ) or \); not returned from regcomp. */ | |
332 | } reg_errcode_t; | |
333 | \f | |
334 | /* This data structure represents a compiled pattern. Before calling | |
335 | the pattern compiler, the fields `buffer', `allocated', `fastmap', | |
336 | `translate', and `no_sub' can be set. After the pattern has been | |
337 | compiled, the `re_nsub' field is available. All other fields are | |
338 | private to the regex routines. */ | |
339 | ||
54d79e99 | 340 | #ifndef RE_TRANSLATE_TYPE |
6973fc01 | 341 | # define RE_TRANSLATE_TYPE char * |
03a75825 RM |
342 | #endif |
343 | ||
2b83a2a4 RM |
344 | struct re_pattern_buffer |
345 | { | |
346 | /* [[[begin pattern_buffer]]] */ | |
347 | /* Space that holds the compiled pattern. It is declared as | |
348 | `unsigned char *' because its elements are | |
349 | sometimes used as array indexes. */ | |
350 | unsigned char *buffer; | |
351 | ||
352 | /* Number of bytes to which `buffer' points. */ | |
51702635 | 353 | unsigned long int allocated; |
2b83a2a4 RM |
354 | |
355 | /* Number of bytes actually used in `buffer'. */ | |
51702635 | 356 | unsigned long int used; |
2b83a2a4 RM |
357 | |
358 | /* Syntax setting with which the pattern was compiled. */ | |
359 | reg_syntax_t syntax; | |
360 | ||
361 | /* Pointer to a fastmap, if any, otherwise zero. re_search uses | |
362 | the fastmap, if there is one, to skip over impossible | |
363 | starting points for matches. */ | |
364 | char *fastmap; | |
365 | ||
366 | /* Either a translate table to apply to all characters before | |
367 | comparing them, or zero for no translation. The translation | |
368 | is applied to a pattern when it is compiled and to a string | |
369 | when it is matched. */ | |
03a75825 | 370 | RE_TRANSLATE_TYPE translate; |
2b83a2a4 RM |
371 | |
372 | /* Number of subexpressions found by the compiler. */ | |
373 | size_t re_nsub; | |
374 | ||
375 | /* Zero if this pattern cannot match the empty string, one else. | |
376 | Well, in truth it's used only in `re_search_2', to see | |
377 | whether or not we should use the fastmap, so we don't set | |
378 | this absolutely perfectly; see `re_compile_fastmap' (the | |
379 | `duplicate' case). */ | |
380 | unsigned can_be_null : 1; | |
381 | ||
382 | /* If REGS_UNALLOCATED, allocate space in the `regs' structure | |
383 | for `max (RE_NREGS, re_nsub + 1)' groups. | |
384 | If REGS_REALLOCATE, reallocate space if necessary. | |
385 | If REGS_FIXED, use what's there. */ | |
386 | #define REGS_UNALLOCATED 0 | |
387 | #define REGS_REALLOCATE 1 | |
388 | #define REGS_FIXED 2 | |
389 | unsigned regs_allocated : 2; | |
390 | ||
391 | /* Set to zero when `regex_compile' compiles a pattern; set to one | |
392 | by `re_compile_fastmap' if it updates the fastmap. */ | |
393 | unsigned fastmap_accurate : 1; | |
394 | ||
395 | /* If set, `re_match_2' does not return information about | |
396 | subexpressions. */ | |
397 | unsigned no_sub : 1; | |
398 | ||
399 | /* If set, a beginning-of-line anchor doesn't match at the | |
54d79e99 | 400 | beginning of the string. */ |
2b83a2a4 RM |
401 | unsigned not_bol : 1; |
402 | ||
403 | /* Similarly for an end-of-line anchor. */ | |
404 | unsigned not_eol : 1; | |
405 | ||
406 | /* If true, an anchor at a newline matches. */ | |
407 | unsigned newline_anchor : 1; | |
408 | ||
409 | /* [[[end pattern_buffer]]] */ | |
410 | }; | |
411 | ||
412 | typedef struct re_pattern_buffer regex_t; | |
413 | \f | |
414 | /* Type for byte offsets within the string. POSIX mandates this. */ | |
415 | typedef int regoff_t; | |
416 | ||
417 | ||
418 | /* This is the structure we store register match data in. See | |
419 | regex.texinfo for a full description of what registers match. */ | |
420 | struct re_registers | |
421 | { | |
422 | unsigned num_regs; | |
423 | regoff_t *start; | |
424 | regoff_t *end; | |
425 | }; | |
426 | ||
427 | ||
428 | /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer, | |
429 | `re_match_2' returns information about at least this many registers | |
430 | the first time a `regs' structure is passed. */ | |
431 | #ifndef RE_NREGS | |
6973fc01 | 432 | # define RE_NREGS 30 |
2b83a2a4 RM |
433 | #endif |
434 | ||
435 | ||
436 | /* POSIX specification for registers. Aside from the different names than | |
437 | `re_registers', POSIX uses an array of structures, instead of a | |
438 | structure of arrays. */ | |
439 | typedef struct | |
440 | { | |
441 | regoff_t rm_so; /* Byte offset from string's start to substring's start. */ | |
442 | regoff_t rm_eo; /* Byte offset from string's start to substring's end. */ | |
443 | } regmatch_t; | |
444 | \f | |
445 | /* Declarations for routines. */ | |
446 | ||
447 | /* To avoid duplicating every routine declaration -- once with a | |
448 | prototype (if we are ANSI), and once without (if we aren't) -- we | |
449 | use the following macro to declare argument types. This | |
450 | unfortunately clutters up the declarations a bit, but I think it's | |
451 | worth it. */ | |
452 | ||
453 | #if __STDC__ | |
454 | ||
6973fc01 | 455 | # define _RE_ARGS(args) args |
2b83a2a4 RM |
456 | |
457 | #else /* not __STDC__ */ | |
458 | ||
6973fc01 | 459 | # define _RE_ARGS(args) () |
2b83a2a4 RM |
460 | |
461 | #endif /* not __STDC__ */ | |
462 | ||
463 | /* Sets the current default syntax to SYNTAX, and return the old syntax. | |
464 | You can also simply assign to the `re_syntax_options' variable. */ | |
465 | extern reg_syntax_t re_set_syntax _RE_ARGS ((reg_syntax_t syntax)); | |
466 | ||
467 | /* Compile the regular expression PATTERN, with length LENGTH | |
468 | and syntax given by the global `re_syntax_options', into the buffer | |
469 | BUFFER. Return NULL if successful, and an error string if not. */ | |
470 | extern const char *re_compile_pattern | |
4cca6b86 | 471 | _RE_ARGS ((const char *pattern, size_t length, |
2b83a2a4 RM |
472 | struct re_pattern_buffer *buffer)); |
473 | ||
474 | ||
475 | /* Compile a fastmap for the compiled pattern in BUFFER; used to | |
476 | accelerate searches. Return 0 if successful and -2 if was an | |
477 | internal error. */ | |
478 | extern int re_compile_fastmap _RE_ARGS ((struct re_pattern_buffer *buffer)); | |
479 | ||
480 | ||
481 | /* Search in the string STRING (with length LENGTH) for the pattern | |
482 | compiled into BUFFER. Start searching at position START, for RANGE | |
483 | characters. Return the starting position of the match, -1 for no | |
484 | match, or -2 for an internal error. Also return register | |
485 | information in REGS (if REGS and BUFFER->no_sub are nonzero). */ | |
486 | extern int re_search | |
487 | _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string, | |
488 | int length, int start, int range, struct re_registers *regs)); | |
489 | ||
490 | ||
491 | /* Like `re_search', but search in the concatenation of STRING1 and | |
492 | STRING2. Also, stop searching at index START + STOP. */ | |
493 | extern int re_search_2 | |
494 | _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1, | |
495 | int length1, const char *string2, int length2, | |
496 | int start, int range, struct re_registers *regs, int stop)); | |
497 | ||
498 | ||
499 | /* Like `re_search', but return how many characters in STRING the regexp | |
500 | in BUFFER matched, starting at position START. */ | |
501 | extern int re_match | |
502 | _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string, | |
503 | int length, int start, struct re_registers *regs)); | |
504 | ||
505 | ||
506 | /* Relates to `re_match' as `re_search_2' relates to `re_search'. */ | |
54d79e99 | 507 | extern int re_match_2 |
2b83a2a4 RM |
508 | _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1, |
509 | int length1, const char *string2, int length2, | |
510 | int start, struct re_registers *regs, int stop)); | |
511 | ||
512 | ||
513 | /* Set REGS to hold NUM_REGS registers, storing them in STARTS and | |
514 | ENDS. Subsequent matches using BUFFER and REGS will use this memory | |
515 | for recording register information. STARTS and ENDS must be | |
516 | allocated with malloc, and must each be at least `NUM_REGS * sizeof | |
517 | (regoff_t)' bytes long. | |
518 | ||
519 | If NUM_REGS == 0, then subsequent matches should allocate their own | |
520 | register data. | |
521 | ||
522 | Unless this function is called, the first search or match using | |
523 | PATTERN_BUFFER will allocate its own register data, without | |
524 | freeing the old data. */ | |
525 | extern void re_set_registers | |
526 | _RE_ARGS ((struct re_pattern_buffer *buffer, struct re_registers *regs, | |
527 | unsigned num_regs, regoff_t *starts, regoff_t *ends)); | |
528 | ||
b77e6cd6 | 529 | #if defined _REGEX_RE_COMP || defined _LIBC |
6973fc01 | 530 | # ifndef _CRAY |
2b83a2a4 RM |
531 | /* 4.2 bsd compatibility. */ |
532 | extern char *re_comp _RE_ARGS ((const char *)); | |
533 | extern int re_exec _RE_ARGS ((const char *)); | |
6973fc01 | 534 | # endif |
4cca6b86 | 535 | #endif |
2b83a2a4 | 536 | |
7ca404ad UD |
537 | /* GCC 2.95 and later have "__restrict"; C99 compilers have |
538 | "restrict", and "configure" may have defined "restrict". */ | |
539 | #ifndef __restrict | |
540 | # if ! (2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__)) | |
541 | # if defined restrict || 199901L <= __STDC_VERSION__ | |
542 | # define __restrict restrict | |
543 | # else | |
544 | # define __restrict | |
545 | # endif | |
546 | # endif | |
da2a3ca6 | 547 | #endif |
8343eaee UD |
548 | /* gcc 3.1 and up support the [restrict] syntax. */ |
549 | #ifndef __restrict_arr | |
550 | # if __GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ >= 1) | |
551 | # define __restrict_arr __restrict | |
552 | # else | |
553 | # define __restrict_arr | |
554 | # endif | |
555 | #endif | |
7ca404ad | 556 | |
2b83a2a4 | 557 | /* POSIX compatibility. */ |
98cbe360 UD |
558 | extern int regcomp _RE_ARGS ((regex_t *__restrict __preg, |
559 | const char *__restrict __pattern, | |
30baa360 UD |
560 | int __cflags)); |
561 | ||
98cbe360 UD |
562 | extern int regexec _RE_ARGS ((const regex_t *__restrict __preg, |
563 | const char *__restrict __string, size_t __nmatch, | |
564 | regmatch_t __pmatch[__restrict_arr], | |
565 | int __eflags)); | |
30baa360 | 566 | |
30baa360 UD |
567 | extern size_t regerror _RE_ARGS ((int __errcode, const regex_t *__preg, |
568 | char *__errbuf, size_t __errbuf_size)); | |
569 | ||
30baa360 | 570 | extern void regfree _RE_ARGS ((regex_t *__preg)); |
2b83a2a4 | 571 | |
1fb05e3d UD |
572 | |
573 | #ifdef __cplusplus | |
574 | } | |
575 | #endif /* C++ */ | |
576 | ||
5107cf1d | 577 | #endif /* regex.h */ |
2b83a2a4 RM |
578 | \f |
579 | /* | |
580 | Local variables: | |
581 | make-backup-files: t | |
582 | version-control: t | |
583 | trim-versions-without-asking: nil | |
584 | End: | |
585 | */ |