|
0
|
1 /* Definitions for data structures and routines for the regular
|
|
|
2 expression library, version 0.12.
|
|
|
3
|
|
|
4 Copyright (C) 1985, 89, 90, 91, 92, 93, 95 Free Software Foundation, Inc.
|
|
|
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; see the file COPYING. If not, write to
|
|
|
18 the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
|
|
|
19 Boston, MA 02111-1307, USA. */
|
|
|
20
|
|
|
21 /* Synched up with: FSF 19.29. */
|
|
|
22
|
|
|
23 #ifndef __REGEXP_LIBRARY_H__
|
|
|
24 #define __REGEXP_LIBRARY_H__
|
|
|
25
|
|
|
26 /* POSIX says that <sys/types.h> must be included (by the caller) before
|
|
|
27 <regex.h>. */
|
|
|
28
|
|
|
29 #if !defined (_POSIX_C_SOURCE) && !defined (_POSIX_SOURCE) && defined (VMS)
|
|
|
30 /* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it
|
|
|
31 should be there. */
|
|
|
32 #include <stddef.h>
|
|
|
33 #endif
|
|
|
34
|
|
|
35
|
|
|
36 /* The following bits are used to determine the regexp syntax we
|
|
|
37 recognize. The set/not-set meanings are chosen so that Emacs syntax
|
|
|
38 remains the value 0. The bits are given in alphabetical order, and
|
|
|
39 the definitions shifted by one from the previous bit; thus, when we
|
|
|
40 add or remove a bit, only one other definition need change. */
|
|
|
41 typedef unsigned reg_syntax_t;
|
|
|
42
|
|
|
43 /* If this bit is not set, then \ inside a bracket expression is literal.
|
|
|
44 If set, then such a \ quotes the following character. */
|
|
|
45 #define RE_BACKSLASH_ESCAPE_IN_LISTS (1)
|
|
|
46
|
|
|
47 /* If this bit is not set, then + and ? are operators, and \+ and \? are
|
|
|
48 literals.
|
|
|
49 If set, then \+ and \? are operators and + and ? are literals. */
|
|
|
50 #define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
|
|
|
51
|
|
|
52 /* If this bit is set, then character classes are supported. They are:
|
|
|
53 [:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:],
|
|
|
54 [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
|
|
|
55 If not set, then character classes are not supported. */
|
|
|
56 #define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
|
|
|
57
|
|
|
58 /* If this bit is set, then ^ and $ are always anchors (outside bracket
|
|
|
59 expressions, of course).
|
|
|
60 If this bit is not set, then it depends:
|
|
|
61 ^ is an anchor if it is at the beginning of a regular
|
|
|
62 expression or after an open-group or an alternation operator;
|
|
|
63 $ is an anchor if it is at the end of a regular expression, or
|
|
|
64 before a close-group or an alternation operator.
|
|
|
65
|
|
|
66 This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
|
|
|
67 POSIX draft 11.2 says that * etc. in leading positions is undefined.
|
|
|
68 We already implemented a previous draft which made those constructs
|
|
|
69 invalid, though, so we haven't changed the code back. */
|
|
|
70 #define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
|
|
|
71
|
|
|
72 /* If this bit is set, then special characters are always special
|
|
|
73 regardless of where they are in the pattern.
|
|
|
74 If this bit is not set, then special characters are special only in
|
|
|
75 some contexts; otherwise they are ordinary. Specifically,
|
|
|
76 * + ? and intervals are only special when not after the beginning,
|
|
|
77 open-group, or alternation operator. */
|
|
|
78 #define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
|
|
|
79
|
|
|
80 /* If this bit is set, then *, +, ?, and { cannot be first in an re or
|
|
|
81 immediately after an alternation or begin-group operator. */
|
|
|
82 #define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
|
|
|
83
|
|
|
84 /* If this bit is set, then . matches newline.
|
|
|
85 If not set, then it doesn't. */
|
|
|
86 #define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
|
|
|
87
|
|
|
88 /* If this bit is set, then . doesn't match NUL.
|
|
|
89 If not set, then it does. */
|
|
|
90 #define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
|
|
|
91
|
|
|
92 /* If this bit is set, nonmatching lists [^...] do not match newline.
|
|
|
93 If not set, they do. */
|
|
|
94 #define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
|
|
|
95
|
|
|
96 /* If this bit is set, either \{...\} or {...} defines an
|
|
|
97 interval, depending on RE_NO_BK_BRACES.
|
|
|
98 If not set, \{, \}, {, and } are literals. */
|
|
|
99 #define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
|
|
|
100
|
|
|
101 /* If this bit is set, +, ? and | aren't recognized as operators.
|
|
|
102 If not set, they are. */
|
|
|
103 #define RE_LIMITED_OPS (RE_INTERVALS << 1)
|
|
|
104
|
|
|
105 /* If this bit is set, newline is an alternation operator.
|
|
|
106 If not set, newline is literal. */
|
|
|
107 #define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
|
|
|
108
|
|
|
109 /* If this bit is set, then `{...}' defines an interval, and \{ and \}
|
|
|
110 are literals.
|
|
|
111 If not set, then `\{...\}' defines an interval. */
|
|
|
112 #define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
|
|
|
113
|
|
|
114 /* If this bit is set, (...) defines a group, and \( and \) are literals.
|
|
|
115 If not set, \(...\) defines a group, and ( and ) are literals. */
|
|
|
116 #define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
|
|
|
117
|
|
|
118 /* If this bit is set, then \<digit> matches <digit>.
|
|
|
119 If not set, then \<digit> is a back-reference. */
|
|
|
120 #define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
|
|
|
121
|
|
|
122 /* If this bit is set, then | is an alternation operator, and \| is literal.
|
|
|
123 If not set, then \| is an alternation operator, and | is literal. */
|
|
|
124 #define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
|
|
|
125
|
|
|
126 /* If this bit is set, then an ending range point collating higher
|
|
|
127 than the starting range point, as in [z-a], is invalid.
|
|
|
128 If not set, then when ending range point collates higher than the
|
|
|
129 starting range point, the range is ignored. */
|
|
|
130 #define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
|
|
|
131
|
|
|
132 /* If this bit is set, then an unmatched ) is ordinary.
|
|
|
133 If not set, then an unmatched ) is invalid. */
|
|
|
134 #define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
|
|
|
135
|
|
|
136 /* If this bit is set, succeed as soon as we match the whole pattern,
|
|
|
137 without further backtracking. */
|
|
|
138 #define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1)
|
|
|
139
|
|
|
140 /* This global variable defines the particular regexp syntax to use (for
|
|
|
141 some interfaces). When a regexp is compiled, the syntax used is
|
|
|
142 stored in the pattern buffer, so changing this does not affect
|
|
|
143 already-compiled regexps. */
|
|
|
144 extern reg_syntax_t re_syntax_options;
|
|
|
145
|
|
|
146 /* Define combinations of the above bits for the standard possibilities.
|
|
|
147 (The [[[ comments delimit what gets put into the Texinfo file, so
|
|
|
148 don't delete them!) */
|
|
|
149 /* [[[begin syntaxes]]] */
|
|
|
150 #define RE_SYNTAX_EMACS 0
|
|
|
151
|
|
|
152 #define RE_SYNTAX_AWK \
|
|
|
153 (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \
|
|
|
154 | RE_NO_BK_PARENS | RE_NO_BK_REFS \
|
|
|
155 | RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \
|
|
|
156 | RE_UNMATCHED_RIGHT_PAREN_ORD)
|
|
|
157
|
|
|
158 #define RE_SYNTAX_POSIX_AWK \
|
|
|
159 (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS)
|
|
|
160
|
|
|
161 #define RE_SYNTAX_GREP \
|
|
|
162 (RE_BK_PLUS_QM | RE_CHAR_CLASSES \
|
|
|
163 | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS \
|
|
|
164 | RE_NEWLINE_ALT)
|
|
|
165
|
|
|
166 #define RE_SYNTAX_EGREP \
|
|
|
167 (RE_CHAR_CLASSES | RE_CONTEXT_INDEP_ANCHORS \
|
|
|
168 | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE \
|
|
|
169 | RE_NEWLINE_ALT | RE_NO_BK_PARENS \
|
|
|
170 | RE_NO_BK_VBAR)
|
|
|
171
|
|
|
172 #define RE_SYNTAX_POSIX_EGREP \
|
|
|
173 (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES)
|
|
|
174
|
|
|
175 /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff. */
|
|
|
176 #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
|
|
|
177
|
|
|
178 #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
|
|
|
179
|
|
|
180 /* Syntax bits common to both basic and extended POSIX regex syntax. */
|
|
|
181 #define _RE_SYNTAX_POSIX_COMMON \
|
|
|
182 (RE_CHAR_CLASSES | RE_DOT_NEWLINE | RE_DOT_NOT_NULL \
|
|
|
183 | RE_INTERVALS | RE_NO_EMPTY_RANGES)
|
|
|
184
|
|
|
185 #define RE_SYNTAX_POSIX_BASIC \
|
|
|
186 (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM)
|
|
|
187
|
|
|
188 /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
|
|
|
189 RE_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this
|
|
|
190 isn't minimal, since other operators, such as \`, aren't disabled. */
|
|
|
191 #define RE_SYNTAX_POSIX_MINIMAL_BASIC \
|
|
|
192 (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
|
|
|
193
|
|
|
194 #define RE_SYNTAX_POSIX_EXTENDED \
|
|
|
195 (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
|
|
|
196 | RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \
|
|
|
197 | RE_NO_BK_PARENS | RE_NO_BK_VBAR \
|
|
|
198 | RE_UNMATCHED_RIGHT_PAREN_ORD)
|
|
|
199
|
|
|
200 /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INVALID_OPS
|
|
|
201 replaces RE_CONTEXT_INDEP_OPS and RE_NO_BK_REFS is added. */
|
|
|
202 #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED \
|
|
|
203 (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
|
|
|
204 | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES \
|
|
|
205 | RE_NO_BK_PARENS | RE_NO_BK_REFS \
|
|
|
206 | RE_NO_BK_VBAR | RE_UNMATCHED_RIGHT_PAREN_ORD)
|
|
|
207 /* [[[end syntaxes]]] */
|
|
|
208
|
|
|
209 /* Maximum number of duplicates an interval can allow. Some systems
|
|
|
210 (erroneously) define this in other header files, but we want our
|
|
|
211 value, so remove any previous define. */
|
|
|
212 #ifdef RE_DUP_MAX
|
|
|
213 #undef RE_DUP_MAX
|
|
|
214 #endif
|
|
|
215 #define RE_DUP_MAX ((1 << 15) - 1)
|
|
|
216
|
|
|
217
|
|
|
218 /* POSIX `cflags' bits (i.e., information for `regcomp'). */
|
|
|
219
|
|
|
220 /* If this bit is set, then use extended regular expression syntax.
|
|
|
221 If not set, then use basic regular expression syntax. */
|
|
|
222 #define REG_EXTENDED 1
|
|
|
223
|
|
|
224 /* If this bit is set, then ignore case when matching.
|
|
|
225 If not set, then case is significant. */
|
|
|
226 #define REG_ICASE (REG_EXTENDED << 1)
|
|
|
227
|
|
|
228 /* If this bit is set, then anchors do not match at newline
|
|
|
229 characters in the string.
|
|
|
230 If not set, then anchors do match at newlines. */
|
|
|
231 #define REG_NEWLINE (REG_ICASE << 1)
|
|
|
232
|
|
|
233 /* If this bit is set, then report only success or fail in regexec.
|
|
|
234 If not set, then returns differ between not matching and errors. */
|
|
|
235 #define REG_NOSUB (REG_NEWLINE << 1)
|
|
|
236
|
|
|
237
|
|
|
238 /* POSIX `eflags' bits (i.e., information for regexec). */
|
|
|
239
|
|
|
240 /* If this bit is set, then the beginning-of-line operator doesn't match
|
|
|
241 the beginning of the string (presumably because it's not the
|
|
|
242 beginning of a line).
|
|
|
243 If not set, then the beginning-of-line operator does match the
|
|
|
244 beginning of the string. */
|
|
|
245 #define REG_NOTBOL 1
|
|
|
246
|
|
|
247 /* Like REG_NOTBOL, except for the end-of-line. */
|
|
|
248 #define REG_NOTEOL (1 << 1)
|
|
|
249
|
|
|
250
|
|
|
251 /* If any error codes are removed, changed, or added, update the
|
|
|
252 `re_error_msg' table in regex.c. */
|
|
|
253 typedef enum
|
|
|
254 {
|
|
|
255 REG_NOERROR = 0, /* Success. */
|
|
|
256 REG_NOMATCH, /* Didn't find a match (for regexec). */
|
|
|
257
|
|
|
258 /* POSIX regcomp return error codes. (In the order listed in the
|
|
|
259 standard.) */
|
|
|
260 REG_BADPAT, /* Invalid pattern. */
|
|
|
261 REG_ECOLLATE, /* Not implemented. */
|
|
|
262 REG_ECTYPE, /* Invalid character class name. */
|
|
|
263 REG_EESCAPE, /* Trailing backslash. */
|
|
|
264 REG_ESUBREG, /* Invalid back reference. */
|
|
|
265 REG_EBRACK, /* Unmatched left bracket. */
|
|
|
266 REG_EPAREN, /* Parenthesis imbalance. */
|
|
|
267 REG_EBRACE, /* Unmatched \{. */
|
|
|
268 REG_BADBR, /* Invalid contents of \{\}. */
|
|
|
269 REG_ERANGE, /* Invalid range end. */
|
|
|
270 REG_ESPACE, /* Ran out of memory. */
|
|
|
271 REG_BADRPT, /* No preceding re for repetition op. */
|
|
|
272
|
|
|
273 /* Error codes we've added. */
|
|
|
274 REG_EEND, /* Premature end. */
|
|
|
275 REG_ESIZE, /* Compiled pattern bigger than 2^16 bytes. */
|
|
|
276 REG_ERPAREN /* Unmatched ) or \); not returned from regcomp. */
|
|
|
277 #ifdef emacs
|
|
|
278 ,REG_ESYNTAX /* Invalid syntax designator. */
|
|
|
279 #endif
|
|
70
|
280 #ifdef MULE
|
|
|
281 ,REG_ERANGESPAN /* Ranges may not span charsets. */
|
|
104
|
282 ,REG_ECATEGORY /* Invalid category designator */
|
|
70
|
283 #endif
|
|
0
|
284 } reg_errcode_t;
|
|
|
285
|
|
|
286 /* This data structure represents a compiled pattern. Before calling
|
|
|
287 the pattern compiler, the fields `buffer', `allocated', `fastmap',
|
|
|
288 `translate', and `no_sub' can be set. After the pattern has been
|
|
|
289 compiled, the `re_nsub' field is available. All other fields are
|
|
|
290 private to the regex routines. */
|
|
|
291
|
|
|
292 struct re_pattern_buffer
|
|
|
293 {
|
|
|
294 /* [[[begin pattern_buffer]]] */
|
|
|
295 /* Space that holds the compiled pattern. It is declared as
|
|
|
296 `unsigned char *' because its elements are
|
|
|
297 sometimes used as array indexes. */
|
|
|
298 unsigned char *buffer;
|
|
|
299
|
|
|
300 /* Number of bytes to which `buffer' points. */
|
|
|
301 unsigned long allocated;
|
|
|
302
|
|
|
303 /* Number of bytes actually used in `buffer'. */
|
|
|
304 unsigned long used;
|
|
|
305
|
|
|
306 /* Syntax setting with which the pattern was compiled. */
|
|
|
307 reg_syntax_t syntax;
|
|
|
308
|
|
|
309 /* Pointer to a fastmap, if any, otherwise zero. re_search uses
|
|
|
310 the fastmap, if there is one, to skip over impossible
|
|
|
311 starting points for matches. */
|
|
|
312 char *fastmap;
|
|
|
313
|
|
|
314 /* Either a translate table to apply to all characters before
|
|
|
315 comparing them, or zero for no translation. The translation
|
|
|
316 is applied to a pattern when it is compiled and to a string
|
|
|
317 when it is matched. */
|
|
|
318 char *translate;
|
|
|
319
|
|
|
320 /* Number of subexpressions found by the compiler. */
|
|
|
321 size_t re_nsub;
|
|
|
322
|
|
|
323 /* Zero if this pattern cannot match the empty string, one else.
|
|
|
324 Well, in truth it's used only in `re_search_2', to see
|
|
|
325 whether or not we should use the fastmap, so we don't set
|
|
|
326 this absolutely perfectly; see `re_compile_fastmap' (the
|
|
|
327 `duplicate' case). */
|
|
|
328 unsigned can_be_null : 1;
|
|
|
329
|
|
|
330 /* If REGS_UNALLOCATED, allocate space in the `regs' structure
|
|
|
331 for `max (RE_NREGS, re_nsub + 1)' groups.
|
|
|
332 If REGS_REALLOCATE, reallocate space if necessary.
|
|
|
333 If REGS_FIXED, use what's there. */
|
|
|
334 #define REGS_UNALLOCATED 0
|
|
|
335 #define REGS_REALLOCATE 1
|
|
|
336 #define REGS_FIXED 2
|
|
|
337 unsigned regs_allocated : 2;
|
|
|
338
|
|
|
339 /* Set to zero when `regex_compile' compiles a pattern; set to one
|
|
|
340 by `re_compile_fastmap' if it updates the fastmap. */
|
|
|
341 unsigned fastmap_accurate : 1;
|
|
|
342
|
|
|
343 /* If set, `re_match_2' does not return information about
|
|
|
344 subexpressions. */
|
|
|
345 unsigned no_sub : 1;
|
|
|
346
|
|
|
347 /* If set, a beginning-of-line anchor doesn't match at the
|
|
|
348 beginning of the string. */
|
|
|
349 unsigned not_bol : 1;
|
|
|
350
|
|
|
351 /* Similarly for an end-of-line anchor. */
|
|
|
352 unsigned not_eol : 1;
|
|
|
353
|
|
|
354 /* If true, an anchor at a newline matches. */
|
|
|
355 unsigned newline_anchor : 1;
|
|
|
356
|
|
|
357 /* [[[end pattern_buffer]]] */
|
|
|
358 };
|
|
|
359
|
|
|
360 typedef struct re_pattern_buffer regex_t;
|
|
|
361
|
|
|
362 /* Type for byte offsets within the string. POSIX mandates this. */
|
|
|
363 typedef int regoff_t;
|
|
|
364
|
|
|
365
|
|
|
366 /* This is the structure we store register match data in. See
|
|
|
367 regex.texinfo for a full description of what registers match. */
|
|
|
368 struct re_registers
|
|
|
369 {
|
|
|
370 unsigned num_regs;
|
|
|
371 regoff_t *start;
|
|
|
372 regoff_t *end;
|
|
|
373 };
|
|
|
374
|
|
|
375
|
|
|
376 /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
|
|
|
377 `re_match_2' returns information about at least this many registers
|
|
|
378 the first time a `regs' structure is passed. */
|
|
|
379 #ifndef RE_NREGS
|
|
|
380 #define RE_NREGS 30
|
|
|
381 #endif
|
|
|
382
|
|
|
383
|
|
|
384 /* POSIX specification for registers. Aside from the different names than
|
|
|
385 `re_registers', POSIX uses an array of structures, instead of a
|
|
|
386 structure of arrays. */
|
|
|
387 typedef struct
|
|
|
388 {
|
|
|
389 regoff_t rm_so; /* Byte offset from string's start to substring's start. */
|
|
|
390 regoff_t rm_eo; /* Byte offset from string's start to substring's end. */
|
|
|
391 } regmatch_t;
|
|
|
392
|
|
|
393 /* Declarations for routines. */
|
|
|
394
|
|
|
395 /* To avoid duplicating every routine declaration -- once with a
|
|
|
396 prototype (if we are ANSI), and once without (if we aren't) -- we
|
|
|
397 use the following macro to declare argument types. This
|
|
|
398 unfortunately clutters up the declarations a bit, but I think it's
|
|
|
399 worth it. */
|
|
|
400
|
|
|
401 #ifdef __STDC__
|
|
|
402
|
|
|
403 #define _RE_ARGS(args) args
|
|
|
404
|
|
|
405 #else /* not __STDC__ */
|
|
|
406
|
|
|
407 #define _RE_ARGS(args) ()
|
|
|
408
|
|
|
409 #endif /* not __STDC__ */
|
|
|
410
|
|
|
411 /* Sets the current default syntax to SYNTAX, and return the old syntax.
|
|
|
412 You can also simply assign to the `re_syntax_options' variable. */
|
|
|
413 extern reg_syntax_t re_set_syntax _RE_ARGS ((reg_syntax_t syntax));
|
|
|
414
|
|
|
415 /* Compile the regular expression PATTERN, with length LENGTH
|
|
|
416 and syntax given by the global `re_syntax_options', into the buffer
|
|
|
417 BUFFER. Return NULL if successful, and an error string if not. */
|
|
|
418 extern CONST char *re_compile_pattern
|
|
|
419 _RE_ARGS ((CONST char *pattern, int length,
|
|
|
420 struct re_pattern_buffer *buffer));
|
|
|
421
|
|
|
422
|
|
|
423 /* Compile a fastmap for the compiled pattern in BUFFER; used to
|
|
|
424 accelerate searches. Return 0 if successful and -2 if was an
|
|
|
425 internal error. */
|
|
|
426 extern int re_compile_fastmap _RE_ARGS ((struct re_pattern_buffer *buffer));
|
|
|
427
|
|
|
428
|
|
|
429 /* Search in the string STRING (with length LENGTH) for the pattern
|
|
|
430 compiled into BUFFER. Start searching at position START, for RANGE
|
|
|
431 characters. Return the starting position of the match, -1 for no
|
|
|
432 match, or -2 for an internal error. Also return register
|
|
|
433 information in REGS (if REGS and BUFFER->no_sub are nonzero). */
|
|
|
434 extern int re_search
|
|
|
435 _RE_ARGS ((struct re_pattern_buffer *buffer, CONST char *string,
|
|
|
436 int length, int start, int range, struct re_registers *regs));
|
|
|
437
|
|
|
438
|
|
|
439 /* Like `re_search', but search in the concatenation of STRING1 and
|
|
|
440 STRING2. Also, stop searching at index START + STOP. */
|
|
|
441 extern int re_search_2
|
|
|
442 _RE_ARGS ((struct re_pattern_buffer *buffer, CONST char *string1,
|
|
|
443 int length1, CONST char *string2, int length2,
|
|
|
444 int start, int range, struct re_registers *regs, int stop));
|
|
|
445
|
|
|
446
|
|
|
447 /* Like `re_search', but return how many characters in STRING the regexp
|
|
|
448 in BUFFER matched, starting at position START. */
|
|
|
449 extern int re_match
|
|
|
450 _RE_ARGS ((struct re_pattern_buffer *buffer, CONST char *string,
|
|
|
451 int length, int start, struct re_registers *regs));
|
|
|
452
|
|
|
453
|
|
|
454 /* Relates to `re_match' as `re_search_2' relates to `re_search'. */
|
|
|
455 extern int re_match_2
|
|
|
456 _RE_ARGS ((struct re_pattern_buffer *buffer, CONST char *string1,
|
|
|
457 int length1, CONST char *string2, int length2,
|
|
|
458 int start, struct re_registers *regs, int stop));
|
|
|
459
|
|
|
460
|
|
|
461 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and
|
|
|
462 ENDS. Subsequent matches using BUFFER and REGS will use this memory
|
|
|
463 for recording register information. STARTS and ENDS must be
|
|
|
464 allocated with malloc, and must each be at least `NUM_REGS * sizeof
|
|
|
465 (regoff_t)' bytes long.
|
|
|
466
|
|
|
467 If NUM_REGS == 0, then subsequent matches should allocate their own
|
|
|
468 register data.
|
|
|
469
|
|
|
470 Unless this function is called, the first search or match using
|
|
|
471 PATTERN_BUFFER will allocate its own register data, without
|
|
|
472 freeing the old data. */
|
|
|
473 extern void re_set_registers
|
|
|
474 _RE_ARGS ((struct re_pattern_buffer *buffer, struct re_registers *regs,
|
|
|
475 unsigned num_regs, regoff_t *starts, regoff_t *ends));
|
|
|
476
|
|
|
477 #ifdef _REGEX_RE_COMP
|
|
|
478 /* 4.2 bsd compatibility. */
|
|
|
479 extern char *re_comp _RE_ARGS ((CONST char *));
|
|
|
480 extern int re_exec _RE_ARGS ((CONST char *));
|
|
|
481 #endif
|
|
|
482
|
|
|
483 /* POSIX compatibility. */
|
|
|
484 extern int regcomp _RE_ARGS ((regex_t *preg, CONST char *pattern, int cflags));
|
|
|
485 extern int regexec
|
|
|
486 _RE_ARGS ((CONST regex_t *preg, CONST char *string, size_t nmatch,
|
|
|
487 regmatch_t pmatch[], int eflags));
|
|
|
488 extern size_t regerror
|
|
|
489 _RE_ARGS ((int errcode, CONST regex_t *preg, char *errbuf,
|
|
|
490 size_t errbuf_size));
|
|
|
491 extern void regfree _RE_ARGS ((regex_t *preg));
|
|
|
492
|
|
|
493 #endif /* not __REGEXP_LIBRARY_H__ */
|
|
|
494
|
|
|
495 /*
|
|
|
496 Local variables:
|
|
|
497 make-backup-files: t
|
|
|
498 version-control: t
|
|
|
499 trim-versions-without-asking: nil
|
|
|
500 End:
|
|
|
501 */
|
