Mercurial > emacs
annotate src/regex.h @ 19860:c17fd465ea95 libc-970911 libc-970912 libc-970913 libc-970914 libc-970915 libc-970916 libc-970917 libc-970918 libc-970919 libc-970920 libc-970921 libc-970922 libc-970923 libc-970924 libc-970925 libc-970926 libc-970927 libc-970928 libc-970929 libc-970930 libc-971001 libc-971018 libc-971019 libc-971020 libc-971021 libc-971022 libc-971023 libc-971024 libc-971025 libc-971026 libc-971027 libc-971028 libc-971029 libc-971030 libc-971031 libc-971101 libc-971102 libc-971103 libc-971104 libc-971105 libc-971106 libc-971107 libc-971108 libc-971109 libc-971110 libc-971111 libc-971112 libc-971113 libc-971114 libc-971115 libc-971116 libc-971117 libc-971118 libc-971120 libc-971121 libc-971122 libc-971123 libc-971124 libc-971125 libc-971126 libc-971127 libc-971128 libc-971129 libc-971130 libc-971201 libc-971203 libc-971204 libc-971205 libc-971206 libc-971207 libc-971208 libc-971209 libc-971210 libc-971211 libc-971212 libc-971213 libc-971214 libc-971217 libc-971218 libc-971219 libc-971220 libc-971221 libc-971222 libc-971223 libc-971224 libc-971225 libc-971226 libc-971227 libc-971228 libc-971229 libc-971230 libc-971231 libc-980103 libc-980104 libc-980105 libc-980106 libc-980107 libc-980108 libc-980109 libc-980110 libc-980111 libc-980112 libc-980114 libc-980115 libc-980116 libc-980117 libc-980118 libc-980119 libc-980120 libc-980121 libc-980122 libc-980123 libc-980124 libc-980125 libc-980126 libc-980127 libc-980128
typos.
author | Jeff Law <law@redhat.com> |
---|---|
date | Wed, 10 Sep 1997 21:16:20 +0000 |
parents | 083eba7e6a53 |
children | e6e2f267eb56 |
rev | line source |
---|---|
1154 | 1 /* Definitions for data structures and routines for the regular |
2454 | 2 expression library, version 0.12. |
1154 | 3 |
11840
8902ebe369bd
[VMS]: Don't #include <stddef.h> if _POSIX_SOURCE or _POSIX_C_SOURCE is
Roland McGrath <roland@gnu.org>
parents:
9982
diff
changeset
|
4 Copyright (C) 1985, 89, 90, 91, 92, 93, 95 Free Software Foundation, Inc. |
1154 | 5 |
6 This program is free software; you can redistribute it and/or modify | |
7 it under the terms of the GNU General Public License as published by | |
8 the Free Software Foundation; either version 2, or (at your option) | |
9 any later version. | |
10 | |
11 This program 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 | |
14 GNU General Public License for more details. | |
15 | |
16 You should have received a copy of the GNU General Public License | |
17 along with this program; if not, write to the Free Software | |
14414 | 18 Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, |
19 USA. */ | |
1154 | 20 |
21 #ifndef __REGEXP_LIBRARY_H__ | |
22 #define __REGEXP_LIBRARY_H__ | |
23 | |
1637 | 24 /* POSIX says that <sys/types.h> must be included (by the caller) before |
25 <regex.h>. */ | |
26 | |
11840
8902ebe369bd
[VMS]: Don't #include <stddef.h> if _POSIX_SOURCE or _POSIX_C_SOURCE is
Roland McGrath <roland@gnu.org>
parents:
9982
diff
changeset
|
27 #if !defined (_POSIX_C_SOURCE) && !defined (_POSIX_SOURCE) && defined (VMS) |
1637 | 28 /* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it |
29 should be there. */ | |
1579 | 30 #include <stddef.h> |
31 #endif | |
32 | |
1154 | 33 |
34 /* The following bits are used to determine the regexp syntax we | |
35 recognize. The set/not-set meanings are chosen so that Emacs syntax | |
36 remains the value 0. The bits are given in alphabetical order, and | |
37 the definitions shifted by one from the previous bit; thus, when we | |
38 add or remove a bit, only one other definition need change. */ | |
39 typedef unsigned reg_syntax_t; | |
40 | |
41 /* If this bit is not set, then \ inside a bracket expression is literal. | |
42 If set, then such a \ quotes the following character. */ | |
43 #define RE_BACKSLASH_ESCAPE_IN_LISTS (1) | |
44 | |
45 /* If this bit is not set, then + and ? are operators, and \+ and \? are | |
46 literals. | |
47 If set, then \+ and \? are operators and + and ? are literals. */ | |
48 #define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1) | |
49 | |
50 /* If this bit is set, then character classes are supported. They are: | |
51 [:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:], | |
52 [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:]. | |
53 If not set, then character classes are not supported. */ | |
54 #define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1) | |
55 | |
56 /* If this bit is set, then ^ and $ are always anchors (outside bracket | |
57 expressions, of course). | |
58 If this bit is not set, then it depends: | |
59 ^ is an anchor if it is at the beginning of a regular | |
60 expression or after an open-group or an alternation operator; | |
61 $ is an anchor if it is at the end of a regular expression, or | |
62 before a close-group or an alternation operator. | |
63 | |
64 This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because | |
65 POSIX draft 11.2 says that * etc. in leading positions is undefined. | |
66 We already implemented a previous draft which made those constructs | |
67 invalid, though, so we haven't changed the code back. */ | |
68 #define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1) | |
69 | |
70 /* If this bit is set, then special characters are always special | |
71 regardless of where they are in the pattern. | |
72 If this bit is not set, then special characters are special only in | |
73 some contexts; otherwise they are ordinary. Specifically, | |
74 * + ? and intervals are only special when not after the beginning, | |
75 open-group, or alternation operator. */ | |
76 #define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1) | |
77 | |
78 /* If this bit is set, then *, +, ?, and { cannot be first in an re or | |
79 immediately after an alternation or begin-group operator. */ | |
80 #define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1) | |
81 | |
82 /* If this bit is set, then . matches newline. | |
83 If not set, then it doesn't. */ | |
84 #define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1) | |
85 | |
86 /* If this bit is set, then . doesn't match NUL. | |
87 If not set, then it does. */ | |
88 #define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1) | |
89 | |
90 /* If this bit is set, nonmatching lists [^...] do not match newline. | |
91 If not set, they do. */ | |
92 #define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1) | |
93 | |
94 /* If this bit is set, either \{...\} or {...} defines an | |
95 interval, depending on RE_NO_BK_BRACES. | |
96 If not set, \{, \}, {, and } are literals. */ | |
97 #define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1) | |
98 | |
99 /* If this bit is set, +, ? and | aren't recognized as operators. | |
100 If not set, they are. */ | |
101 #define RE_LIMITED_OPS (RE_INTERVALS << 1) | |
102 | |
103 /* If this bit is set, newline is an alternation operator. | |
104 If not set, newline is literal. */ | |
105 #define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1) | |
106 | |
107 /* If this bit is set, then `{...}' defines an interval, and \{ and \} | |
108 are literals. | |
109 If not set, then `\{...\}' defines an interval. */ | |
110 #define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1) | |
111 | |
112 /* If this bit is set, (...) defines a group, and \( and \) are literals. | |
113 If not set, \(...\) defines a group, and ( and ) are literals. */ | |
114 #define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1) | |
115 | |
116 /* If this bit is set, then \<digit> matches <digit>. | |
117 If not set, then \<digit> is a back-reference. */ | |
118 #define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1) | |
119 | |
120 /* If this bit is set, then | is an alternation operator, and \| is literal. | |
121 If not set, then \| is an alternation operator, and | is literal. */ | |
122 #define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1) | |
123 | |
124 /* If this bit is set, then an ending range point collating higher | |
125 than the starting range point, as in [z-a], is invalid. | |
126 If not set, then when ending range point collates higher than the | |
127 starting range point, the range is ignored. */ | |
128 #define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1) | |
129 | |
130 /* If this bit is set, then an unmatched ) is ordinary. | |
131 If not set, then an unmatched ) is invalid. */ | |
132 #define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1) | |
133 | |
9982
65d2d56726ea
(RE_NO_POSIX_BACKTRACKING): New macro.
Richard M. Stallman <rms@gnu.org>
parents:
9717
diff
changeset
|
134 /* If this bit is set, succeed as soon as we match the whole pattern, |
65d2d56726ea
(RE_NO_POSIX_BACKTRACKING): New macro.
Richard M. Stallman <rms@gnu.org>
parents:
9717
diff
changeset
|
135 without further backtracking. */ |
65d2d56726ea
(RE_NO_POSIX_BACKTRACKING): New macro.
Richard M. Stallman <rms@gnu.org>
parents:
9717
diff
changeset
|
136 #define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1) |
65d2d56726ea
(RE_NO_POSIX_BACKTRACKING): New macro.
Richard M. Stallman <rms@gnu.org>
parents:
9717
diff
changeset
|
137 |
1154 | 138 /* This global variable defines the particular regexp syntax to use (for |
139 some interfaces). When a regexp is compiled, the syntax used is | |
140 stored in the pattern buffer, so changing this does not affect | |
141 already-compiled regexps. */ | |
142 extern reg_syntax_t re_syntax_options; | |
18261
2f0b00246056
(struct re_pattern_buffer): New member multibyte.
Richard M. Stallman <rms@gnu.org>
parents:
14414
diff
changeset
|
143 |
2f0b00246056
(struct re_pattern_buffer): New member multibyte.
Richard M. Stallman <rms@gnu.org>
parents:
14414
diff
changeset
|
144 #ifdef emacs |
2f0b00246056
(struct re_pattern_buffer): New member multibyte.
Richard M. Stallman <rms@gnu.org>
parents:
14414
diff
changeset
|
145 /* In Emacs, this is the string or buffer in which we |
2f0b00246056
(struct re_pattern_buffer): New member multibyte.
Richard M. Stallman <rms@gnu.org>
parents:
14414
diff
changeset
|
146 are matching. It is used for looking up syntax properties. */ |
2f0b00246056
(struct re_pattern_buffer): New member multibyte.
Richard M. Stallman <rms@gnu.org>
parents:
14414
diff
changeset
|
147 extern Lisp_Object re_match_object; |
2f0b00246056
(struct re_pattern_buffer): New member multibyte.
Richard M. Stallman <rms@gnu.org>
parents:
14414
diff
changeset
|
148 #endif |
2f0b00246056
(struct re_pattern_buffer): New member multibyte.
Richard M. Stallman <rms@gnu.org>
parents:
14414
diff
changeset
|
149 |
1154 | 150 |
151 /* Define combinations of the above bits for the standard possibilities. | |
152 (The [[[ comments delimit what gets put into the Texinfo file, so | |
153 don't delete them!) */ | |
154 /* [[[begin syntaxes]]] */ | |
155 #define RE_SYNTAX_EMACS 0 | |
156 | |
157 #define RE_SYNTAX_AWK \ | |
158 (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \ | |
159 | RE_NO_BK_PARENS | RE_NO_BK_REFS \ | |
1690
256b87a92511
*** empty log message ***
David J. MacKenzie <djm@gnu.org>
parents:
1637
diff
changeset
|
160 | RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \ |
1154 | 161 | RE_UNMATCHED_RIGHT_PAREN_ORD) |
162 | |
163 #define RE_SYNTAX_POSIX_AWK \ | |
164 (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS) | |
165 | |
166 #define RE_SYNTAX_GREP \ | |
167 (RE_BK_PLUS_QM | RE_CHAR_CLASSES \ | |
168 | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS \ | |
169 | RE_NEWLINE_ALT) | |
170 | |
171 #define RE_SYNTAX_EGREP \ | |
172 (RE_CHAR_CLASSES | RE_CONTEXT_INDEP_ANCHORS \ | |
173 | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE \ | |
174 | RE_NEWLINE_ALT | RE_NO_BK_PARENS \ | |
175 | RE_NO_BK_VBAR) | |
176 | |
177 #define RE_SYNTAX_POSIX_EGREP \ | |
178 (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES) | |
179 | |
1637 | 180 /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff. */ |
181 #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC | |
182 | |
1154 | 183 #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC |
184 | |
185 /* Syntax bits common to both basic and extended POSIX regex syntax. */ | |
186 #define _RE_SYNTAX_POSIX_COMMON \ | |
187 (RE_CHAR_CLASSES | RE_DOT_NEWLINE | RE_DOT_NOT_NULL \ | |
188 | RE_INTERVALS | RE_NO_EMPTY_RANGES) | |
189 | |
190 #define RE_SYNTAX_POSIX_BASIC \ | |
191 (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM) | |
192 | |
193 /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes | |
194 RE_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this | |
195 isn't minimal, since other operators, such as \`, aren't disabled. */ | |
196 #define RE_SYNTAX_POSIX_MINIMAL_BASIC \ | |
197 (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS) | |
198 | |
199 #define RE_SYNTAX_POSIX_EXTENDED \ | |
200 (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \ | |
201 | RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \ | |
202 | RE_NO_BK_PARENS | RE_NO_BK_VBAR \ | |
203 | RE_UNMATCHED_RIGHT_PAREN_ORD) | |
204 | |
205 /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INVALID_OPS | |
206 replaces RE_CONTEXT_INDEP_OPS and RE_NO_BK_REFS is added. */ | |
207 #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED \ | |
208 (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \ | |
209 | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES \ | |
210 | RE_NO_BK_PARENS | RE_NO_BK_REFS \ | |
211 | RE_NO_BK_VBAR | RE_UNMATCHED_RIGHT_PAREN_ORD) | |
212 /* [[[end syntaxes]]] */ | |
213 | |
214 /* Maximum number of duplicates an interval can allow. Some systems | |
215 (erroneously) define this in other header files, but we want our | |
216 value, so remove any previous define. */ | |
217 #ifdef RE_DUP_MAX | |
218 #undef RE_DUP_MAX | |
219 #endif | |
220 #define RE_DUP_MAX ((1 << 15) - 1) | |
221 | |
222 | |
223 /* POSIX `cflags' bits (i.e., information for `regcomp'). */ | |
224 | |
225 /* If this bit is set, then use extended regular expression syntax. | |
226 If not set, then use basic regular expression syntax. */ | |
227 #define REG_EXTENDED 1 | |
228 | |
229 /* If this bit is set, then ignore case when matching. | |
230 If not set, then case is significant. */ | |
231 #define REG_ICASE (REG_EXTENDED << 1) | |
232 | |
233 /* If this bit is set, then anchors do not match at newline | |
234 characters in the string. | |
235 If not set, then anchors do match at newlines. */ | |
236 #define REG_NEWLINE (REG_ICASE << 1) | |
237 | |
238 /* If this bit is set, then report only success or fail in regexec. | |
239 If not set, then returns differ between not matching and errors. */ | |
240 #define REG_NOSUB (REG_NEWLINE << 1) | |
241 | |
242 | |
243 /* POSIX `eflags' bits (i.e., information for regexec). */ | |
244 | |
245 /* If this bit is set, then the beginning-of-line operator doesn't match | |
246 the beginning of the string (presumably because it's not the | |
247 beginning of a line). | |
248 If not set, then the beginning-of-line operator does match the | |
249 beginning of the string. */ | |
250 #define REG_NOTBOL 1 | |
251 | |
252 /* Like REG_NOTBOL, except for the end-of-line. */ | |
253 #define REG_NOTEOL (1 << 1) | |
254 | |
255 | |
256 /* If any error codes are removed, changed, or added, update the | |
257 `re_error_msg' table in regex.c. */ | |
258 typedef enum | |
259 { | |
260 REG_NOERROR = 0, /* Success. */ | |
261 REG_NOMATCH, /* Didn't find a match (for regexec). */ | |
262 | |
263 /* POSIX regcomp return error codes. (In the order listed in the | |
264 standard.) */ | |
265 REG_BADPAT, /* Invalid pattern. */ | |
266 REG_ECOLLATE, /* Not implemented. */ | |
267 REG_ECTYPE, /* Invalid character class name. */ | |
268 REG_EESCAPE, /* Trailing backslash. */ | |
269 REG_ESUBREG, /* Invalid back reference. */ | |
270 REG_EBRACK, /* Unmatched left bracket. */ | |
271 REG_EPAREN, /* Parenthesis imbalance. */ | |
272 REG_EBRACE, /* Unmatched \{. */ | |
273 REG_BADBR, /* Invalid contents of \{\}. */ | |
274 REG_ERANGE, /* Invalid range end. */ | |
275 REG_ESPACE, /* Ran out of memory. */ | |
276 REG_BADRPT, /* No preceding re for repetition op. */ | |
277 | |
278 /* Error codes we've added. */ | |
279 REG_EEND, /* Premature end. */ | |
280 REG_ESIZE, /* Compiled pattern bigger than 2^16 bytes. */ | |
281 REG_ERPAREN /* Unmatched ) or \); not returned from regcomp. */ | |
282 } reg_errcode_t; | |
283 | |
284 /* This data structure represents a compiled pattern. Before calling | |
285 the pattern compiler, the fields `buffer', `allocated', `fastmap', | |
286 `translate', and `no_sub' can be set. After the pattern has been | |
287 compiled, the `re_nsub' field is available. All other fields are | |
288 private to the regex routines. */ | |
289 | |
13249
c7dea1ffceab
(RE_TRANSLATE_TYPE): Define, if not already defined.
Richard M. Stallman <rms@gnu.org>
parents:
11840
diff
changeset
|
290 #ifndef RE_TRANSLATE_TYPE |
c7dea1ffceab
(RE_TRANSLATE_TYPE): Define, if not already defined.
Richard M. Stallman <rms@gnu.org>
parents:
11840
diff
changeset
|
291 #define RE_TRANSLATE_TYPE char * |
18615
083eba7e6a53
(RE_TRANSLATE): Define default assuming simple integers.
Richard M. Stallman <rms@gnu.org>
parents:
18261
diff
changeset
|
292 #define RE_TRANSLATE(TBL, C) ((TBL)[C]) |
13249
c7dea1ffceab
(RE_TRANSLATE_TYPE): Define, if not already defined.
Richard M. Stallman <rms@gnu.org>
parents:
11840
diff
changeset
|
293 #endif |
c7dea1ffceab
(RE_TRANSLATE_TYPE): Define, if not already defined.
Richard M. Stallman <rms@gnu.org>
parents:
11840
diff
changeset
|
294 |
1154 | 295 struct re_pattern_buffer |
296 { | |
297 /* [[[begin pattern_buffer]]] */ | |
298 /* Space that holds the compiled pattern. It is declared as | |
299 `unsigned char *' because its elements are | |
300 sometimes used as array indexes. */ | |
301 unsigned char *buffer; | |
302 | |
303 /* Number of bytes to which `buffer' points. */ | |
304 unsigned long allocated; | |
305 | |
306 /* Number of bytes actually used in `buffer'. */ | |
307 unsigned long used; | |
308 | |
309 /* Syntax setting with which the pattern was compiled. */ | |
310 reg_syntax_t syntax; | |
311 | |
312 /* Pointer to a fastmap, if any, otherwise zero. re_search uses | |
313 the fastmap, if there is one, to skip over impossible | |
314 starting points for matches. */ | |
315 char *fastmap; | |
316 | |
317 /* Either a translate table to apply to all characters before | |
318 comparing them, or zero for no translation. The translation | |
319 is applied to a pattern when it is compiled and to a string | |
320 when it is matched. */ | |
13249
c7dea1ffceab
(RE_TRANSLATE_TYPE): Define, if not already defined.
Richard M. Stallman <rms@gnu.org>
parents:
11840
diff
changeset
|
321 RE_TRANSLATE_TYPE translate; |
1154 | 322 |
323 /* Number of subexpressions found by the compiler. */ | |
324 size_t re_nsub; | |
325 | |
326 /* Zero if this pattern cannot match the empty string, one else. | |
327 Well, in truth it's used only in `re_search_2', to see | |
328 whether or not we should use the fastmap, so we don't set | |
329 this absolutely perfectly; see `re_compile_fastmap' (the | |
330 `duplicate' case). */ | |
331 unsigned can_be_null : 1; | |
332 | |
333 /* If REGS_UNALLOCATED, allocate space in the `regs' structure | |
334 for `max (RE_NREGS, re_nsub + 1)' groups. | |
335 If REGS_REALLOCATE, reallocate space if necessary. | |
336 If REGS_FIXED, use what's there. */ | |
337 #define REGS_UNALLOCATED 0 | |
338 #define REGS_REALLOCATE 1 | |
339 #define REGS_FIXED 2 | |
340 unsigned regs_allocated : 2; | |
341 | |
1637 | 342 /* Set to zero when `regex_compile' compiles a pattern; set to one |
343 by `re_compile_fastmap' if it updates the fastmap. */ | |
1154 | 344 unsigned fastmap_accurate : 1; |
345 | |
1637 | 346 /* If set, `re_match_2' does not return information about |
347 subexpressions. */ | |
1154 | 348 unsigned no_sub : 1; |
349 | |
350 /* If set, a beginning-of-line anchor doesn't match at the | |
351 beginning of the string. */ | |
352 unsigned not_bol : 1; | |
353 | |
354 /* Similarly for an end-of-line anchor. */ | |
355 unsigned not_eol : 1; | |
356 | |
357 /* If true, an anchor at a newline matches. */ | |
358 unsigned newline_anchor : 1; | |
359 | |
18261
2f0b00246056
(struct re_pattern_buffer): New member multibyte.
Richard M. Stallman <rms@gnu.org>
parents:
14414
diff
changeset
|
360 /* If true, multi-byte form in the `buffer' should be recognized as a |
2f0b00246056
(struct re_pattern_buffer): New member multibyte.
Richard M. Stallman <rms@gnu.org>
parents:
14414
diff
changeset
|
361 multibyte character. */ |
2f0b00246056
(struct re_pattern_buffer): New member multibyte.
Richard M. Stallman <rms@gnu.org>
parents:
14414
diff
changeset
|
362 unsigned multibyte : 1; |
2f0b00246056
(struct re_pattern_buffer): New member multibyte.
Richard M. Stallman <rms@gnu.org>
parents:
14414
diff
changeset
|
363 |
1154 | 364 /* [[[end pattern_buffer]]] */ |
365 }; | |
366 | |
367 typedef struct re_pattern_buffer regex_t; | |
368 | |
369 /* Type for byte offsets within the string. POSIX mandates this. */ | |
370 typedef int regoff_t; | |
371 | |
372 | |
373 /* This is the structure we store register match data in. See | |
374 regex.texinfo for a full description of what registers match. */ | |
375 struct re_registers | |
376 { | |
377 unsigned num_regs; | |
378 regoff_t *start; | |
379 regoff_t *end; | |
380 }; | |
381 | |
382 | |
383 /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer, | |
384 `re_match_2' returns information about at least this many registers | |
385 the first time a `regs' structure is passed. */ | |
386 #ifndef RE_NREGS | |
387 #define RE_NREGS 30 | |
388 #endif | |
389 | |
390 | |
391 /* POSIX specification for registers. Aside from the different names than | |
392 `re_registers', POSIX uses an array of structures, instead of a | |
393 structure of arrays. */ | |
394 typedef struct | |
395 { | |
396 regoff_t rm_so; /* Byte offset from string's start to substring's start. */ | |
397 regoff_t rm_eo; /* Byte offset from string's start to substring's end. */ | |
398 } regmatch_t; | |
399 | |
400 /* Declarations for routines. */ | |
401 | |
402 /* To avoid duplicating every routine declaration -- once with a | |
403 prototype (if we are ANSI), and once without (if we aren't) -- we | |
404 use the following macro to declare argument types. This | |
405 unfortunately clutters up the declarations a bit, but I think it's | |
1797
cc86c049e576
*** empty log message ***
David J. MacKenzie <djm@gnu.org>
parents:
1722
diff
changeset
|
406 worth it. */ |
1154 | 407 |
408 #if __STDC__ | |
1722 | 409 |
1154 | 410 #define _RE_ARGS(args) args |
1722 | 411 |
1637 | 412 #else /* not __STDC__ */ |
1722 | 413 |
1154 | 414 #define _RE_ARGS(args) () |
1722 | 415 |
1637 | 416 #endif /* not __STDC__ */ |
1154 | 417 |
418 /* Sets the current default syntax to SYNTAX, and return the old syntax. | |
419 You can also simply assign to the `re_syntax_options' variable. */ | |
420 extern reg_syntax_t re_set_syntax _RE_ARGS ((reg_syntax_t syntax)); | |
421 | |
422 /* Compile the regular expression PATTERN, with length LENGTH | |
423 and syntax given by the global `re_syntax_options', into the buffer | |
424 BUFFER. Return NULL if successful, and an error string if not. */ | |
425 extern const char *re_compile_pattern | |
426 _RE_ARGS ((const char *pattern, int length, | |
427 struct re_pattern_buffer *buffer)); | |
428 | |
429 | |
430 /* Compile a fastmap for the compiled pattern in BUFFER; used to | |
431 accelerate searches. Return 0 if successful and -2 if was an | |
432 internal error. */ | |
433 extern int re_compile_fastmap _RE_ARGS ((struct re_pattern_buffer *buffer)); | |
434 | |
435 | |
436 /* Search in the string STRING (with length LENGTH) for the pattern | |
437 compiled into BUFFER. Start searching at position START, for RANGE | |
438 characters. Return the starting position of the match, -1 for no | |
439 match, or -2 for an internal error. Also return register | |
440 information in REGS (if REGS and BUFFER->no_sub are nonzero). */ | |
441 extern int re_search | |
442 _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string, | |
443 int length, int start, int range, struct re_registers *regs)); | |
444 | |
445 | |
446 /* Like `re_search', but search in the concatenation of STRING1 and | |
447 STRING2. Also, stop searching at index START + STOP. */ | |
448 extern int re_search_2 | |
449 _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1, | |
450 int length1, const char *string2, int length2, | |
451 int start, int range, struct re_registers *regs, int stop)); | |
452 | |
453 | |
454 /* Like `re_search', but return how many characters in STRING the regexp | |
455 in BUFFER matched, starting at position START. */ | |
456 extern int re_match | |
457 _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string, | |
458 int length, int start, struct re_registers *regs)); | |
459 | |
460 | |
461 /* Relates to `re_match' as `re_search_2' relates to `re_search'. */ | |
462 extern int re_match_2 | |
463 _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1, | |
464 int length1, const char *string2, int length2, | |
465 int start, struct re_registers *regs, int stop)); | |
466 | |
467 | |
468 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and | |
469 ENDS. Subsequent matches using BUFFER and REGS will use this memory | |
470 for recording register information. STARTS and ENDS must be | |
471 allocated with malloc, and must each be at least `NUM_REGS * sizeof | |
472 (regoff_t)' bytes long. | |
473 | |
474 If NUM_REGS == 0, then subsequent matches should allocate their own | |
475 register data. | |
476 | |
477 Unless this function is called, the first search or match using | |
478 PATTERN_BUFFER will allocate its own register data, without | |
479 freeing the old data. */ | |
480 extern void re_set_registers | |
481 _RE_ARGS ((struct re_pattern_buffer *buffer, struct re_registers *regs, | |
482 unsigned num_regs, regoff_t *starts, regoff_t *ends)); | |
483 | |
9717
96a592ccb751
(re_comp, re_exec): Define this obsolete 4.2bsd
Paul Eggert <eggert@twinsun.com>
parents:
9584
diff
changeset
|
484 #ifdef _REGEX_RE_COMP |
1154 | 485 /* 4.2 bsd compatibility. */ |
486 extern char *re_comp _RE_ARGS ((const char *)); | |
487 extern int re_exec _RE_ARGS ((const char *)); | |
9717
96a592ccb751
(re_comp, re_exec): Define this obsolete 4.2bsd
Paul Eggert <eggert@twinsun.com>
parents:
9584
diff
changeset
|
488 #endif |
1154 | 489 |
490 /* POSIX compatibility. */ | |
491 extern int regcomp _RE_ARGS ((regex_t *preg, const char *pattern, int cflags)); | |
492 extern int regexec | |
493 _RE_ARGS ((const regex_t *preg, const char *string, size_t nmatch, | |
494 regmatch_t pmatch[], int eflags)); | |
495 extern size_t regerror | |
496 _RE_ARGS ((int errcode, const regex_t *preg, char *errbuf, | |
497 size_t errbuf_size)); | |
498 extern void regfree _RE_ARGS ((regex_t *preg)); | |
499 | |
500 #endif /* not __REGEXP_LIBRARY_H__ */ | |
501 | |
502 /* | |
503 Local variables: | |
504 make-backup-files: t | |
505 version-control: t | |
506 trim-versions-without-asking: nil | |
507 End: | |
508 */ |