« back to all changes in this revision
Viewing changes to src/gnu/regex.h
- browse files at revision 6
- compare with another revision
- download diff
- download tarball
- view history from revision 6
- Committer: Package Import Robot
- Author(s): Enrico Tassi
- Date: 2012-04-16 17:48:20 UTC
- mfrom: (1.1.2)
- Revision ID: package-import@ubuntu.com-20120416174820-ed9fmpd97qwctkjr
Tags: 2.6.0-1
* New upstream release
* New backend: TRE, regex with approximated matching
* Switch to dh-lua
* Packages renamed according to the new Lua policy
* Update Standards-Version to 3.9.3, no changes
* debian/compat set to 9
* Update watch file
* Update copyright file
* New backend: TRE, regex with approximated matching
* Switch to dh-lua
* Packages renamed according to the new Lua policy
* Update Standards-Version to 3.9.3, no changes
* debian/compat set to 9
* Update watch file
* Update copyright file
- files added:
- README.rst
- debian/lua5.2.onig.dh-lua.conf
- debian/lua5.2.pcre.dh-lua.conf
- debian/lua5.2.posix.dh-lua.conf
- debian/lua5.2.tre.dh-lua.conf
- files removed:
- README
- files modified:
- .pc/test.patch/test/runtest.lua
added
removed
src/gnu/regex.h
1
/* Definitions for data structures and routines for the regular
2
expression library.
3
Copyright (C) 1985, 1989, 1990, 1991, 1992, 1993, 1995, 1996, 1997, 1998,
4
2000, 2001, 2002, 2003, 2005, 2006, 2009, 2010 Free Software Foundation,
5
Inc.
6
This file is part of the GNU C Library.
7
8
This program is free software; you can redistribute it and/or modify
9
it under the terms of the GNU General Public License as published by
10
the Free Software Foundation; either version 3, or (at your option)
11
any later version.
12
13
This program is distributed in the hope that it will be useful,
14
but WITHOUT ANY WARRANTY; without even the implied warranty of
15
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
16
GNU General Public License for more details.
17
18
You should have received a copy of the GNU General Public License along
19
with this program; if not, write to the Free Software Foundation,
20
Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */
21
22
#ifndef _REGEX_H
23
#define _REGEX_H 1
24
25
#include <sys/types.h>
26
27
/* Allow the use in C++ code. */
28
#ifdef __cplusplus
29
extern "C" {
30
#endif
31
32
/* Define __USE_GNU_REGEX to declare GNU extensions that violate the
33
POSIX name space rules. */
34
#undef __USE_GNU_REGEX
35
#if (defined _GNU_SOURCE \
36
|| (!defined _POSIX_C_SOURCE && !defined _POSIX_SOURCE \
37
&& !defined _XOPEN_SOURCE))
38
# define __USE_GNU_REGEX 1
39
#endif
40
41
#ifdef _REGEX_LARGE_OFFSETS
42
43
/* Use types and values that are wide enough to represent signed and
44
unsigned byte offsets in memory. This currently works only when
45
the regex code is used outside of the GNU C library; it is not yet
46
supported within glibc itself, and glibc users should not define
47
_REGEX_LARGE_OFFSETS. */
48
49
/* The type of the offset of a byte within a string.
50
For historical reasons POSIX 1003.1-2004 requires that regoff_t be
51
at least as wide as off_t. However, many common POSIX platforms set
52
regoff_t to the more-sensible ssize_t and the Open Group has
53
signalled its intention to change the requirement to be that
54
regoff_t be at least as wide as ptrdiff_t and ssize_t; see XBD ERN
55
60 (2005-08-25). We don't know of any hosts where ssize_t or
56
ptrdiff_t is wider than ssize_t, so ssize_t is safe. */
57
typedef ssize_t regoff_t;
58
59
/* The type of nonnegative object indexes. Traditionally, GNU regex
60
uses 'int' for these. Code that uses __re_idx_t should work
61
regardless of whether the type is signed. */
62
typedef size_t __re_idx_t;
63
64
/* The type of object sizes. */
65
typedef size_t __re_size_t;
66
67
/* The type of object sizes, in places where the traditional code
68
uses unsigned long int. */
69
typedef size_t __re_long_size_t;
70
71
#else
72
73
/* Use types that are binary-compatible with the traditional GNU regex
74
implementation, which mishandles strings longer than INT_MAX. */
75
76
typedef int regoff_t;
77
typedef int __re_idx_t;
78
typedef unsigned int __re_size_t;
79
typedef unsigned long int __re_long_size_t;
80
81
#endif
82
83
/* The following two types have to be signed and unsigned integer type
84
wide enough to hold a value of a pointer. For most ANSI compilers
85
ptrdiff_t and size_t should be likely OK. Still size of these two
86
types is 2 for Microsoft C. Ugh... */
87
typedef long int s_reg_t;
88
typedef unsigned long int active_reg_t;
89
90
/* The following bits are used to determine the regexp syntax we
91
recognize. The set/not-set meanings are chosen so that Emacs syntax
92
remains the value 0. The bits are given in alphabetical order, and
93
the definitions shifted by one from the previous bit; thus, when we
94
add or remove a bit, only one other definition need change. */
95
typedef unsigned long int reg_syntax_t;
96
97
#ifdef __USE_GNU_REGEX
98
99
/* If this bit is not set, then \ inside a bracket expression is literal.
100
If set, then such a \ quotes the following character. */
101
# define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1)
102
103
/* If this bit is not set, then + and ? are operators, and \+ and \? are
104
literals.
105
If set, then \+ and \? are operators and + and ? are literals. */
106
# define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
107
108
/* If this bit is set, then character classes are supported. They are:
109
[:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:],
110
[:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
111
If not set, then character classes are not supported. */
112
# define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
113
114
/* If this bit is set, then ^ and $ are always anchors (outside bracket
115
expressions, of course).
116
If this bit is not set, then it depends:
117
^ is an anchor if it is at the beginning of a regular
118
expression or after an open-group or an alternation operator;
119
$ is an anchor if it is at the end of a regular expression, or
120
before a close-group or an alternation operator.
121
122
This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
123
POSIX draft 11.2 says that * etc. in leading positions is undefined.
124
We already implemented a previous draft which made those constructs
125
invalid, though, so we haven't changed the code back. */
126
# define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
127
128
/* If this bit is set, then special characters are always special
129
regardless of where they are in the pattern.
130
If this bit is not set, then special characters are special only in
131
some contexts; otherwise they are ordinary. Specifically,
132
* + ? and intervals are only special when not after the beginning,
133
open-group, or alternation operator. */
134
# define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
135
136
/* If this bit is set, then *, +, ?, and { cannot be first in an re or
137
immediately after an alternation or begin-group operator. */
138
# define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
139
140
/* If this bit is set, then . matches newline.
141
If not set, then it doesn't. */
142
# define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
143
144
/* If this bit is set, then . doesn't match NUL.
145
If not set, then it does. */
146
# define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
147
148
/* If this bit is set, nonmatching lists [^...] do not match newline.
149
If not set, they do. */
150
# define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
151
152
/* If this bit is set, either \{...\} or {...} defines an
153
interval, depending on RE_NO_BK_BRACES.
154
If not set, \{, \}, {, and } are literals. */
155
# define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
156
157
/* If this bit is set, +, ? and | aren't recognized as operators.
158
If not set, they are. */
159
# define RE_LIMITED_OPS (RE_INTERVALS << 1)
160
161
/* If this bit is set, newline is an alternation operator.
162
If not set, newline is literal. */
163
# define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
164
165
/* If this bit is set, then `{...}' defines an interval, and \{ and \}
166
are literals.
167
If not set, then `\{...\}' defines an interval. */
168
# define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
169
170
/* If this bit is set, (...) defines a group, and \( and \) are literals.
171
If not set, \(...\) defines a group, and ( and ) are literals. */
172
# define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
173
174
/* If this bit is set, then \<digit> matches <digit>.
175
If not set, then \<digit> is a back-reference. */
176
# define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
177
178
/* If this bit is set, then | is an alternation operator, and \| is literal.
179
If not set, then \| is an alternation operator, and | is literal. */
180
# define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
181
182
/* If this bit is set, then an ending range point collating higher
183
than the starting range point, as in [z-a], is invalid.
184
If not set, then when ending range point collates higher than the
185
starting range point, the range is ignored. */
186
# define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
187
188
/* If this bit is set, then an unmatched ) is ordinary.
189
If not set, then an unmatched ) is invalid. */
190
# define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
191
192
/* If this bit is set, succeed as soon as we match the whole pattern,
193
without further backtracking. */
194
# define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1)
195
196
/* If this bit is set, do not process the GNU regex operators.
197
If not set, then the GNU regex operators are recognized. */
198
# define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1)
199
200
/* If this bit is set, turn on internal regex debugging.
201
If not set, and debugging was on, turn it off.
202
This only works if regex.c is compiled -DDEBUG.
203
We define this bit always, so that all that's needed to turn on
204
debugging is to recompile regex.c; the calling code can always have
205
this bit set, and it won't affect anything in the normal case. */
206
# define RE_DEBUG (RE_NO_GNU_OPS << 1)
207
208
/* If this bit is set, a syntactically invalid interval is treated as
209
a string of ordinary characters. For example, the ERE 'a{1' is
210
treated as 'a\{1'. */
211
# define RE_INVALID_INTERVAL_ORD (RE_DEBUG << 1)
212
213
/* If this bit is set, then ignore case when matching.
214
If not set, then case is significant. */
215
# define RE_ICASE (RE_INVALID_INTERVAL_ORD << 1)
216
217
/* This bit is used internally like RE_CONTEXT_INDEP_ANCHORS but only
218
for ^, because it is difficult to scan the regex backwards to find
219
whether ^ should be special. */
220
# define RE_CARET_ANCHORS_HERE (RE_ICASE << 1)
221
222
/* If this bit is set, then \{ cannot be first in a regex or
223
immediately after an alternation, open-group or \} operator. */
224
# define RE_CONTEXT_INVALID_DUP (RE_CARET_ANCHORS_HERE << 1)
225
226
/* If this bit is set, then no_sub will be set to 1 during
227
re_compile_pattern. */
228
# define RE_NO_SUB (RE_CONTEXT_INVALID_DUP << 1)
229
230
/* If this bit is set, then all special characters are ignored
231
by re_compile_pattern. */
232
# define RE_PLAIN (RE_NO_SUB << 1)
233
234
#endif /* defined __USE_GNU_REGEX */
235
236
/* This global variable defines the particular regexp syntax to use (for
237
some interfaces). When a regexp is compiled, the syntax used is
238
stored in the pattern buffer, so changing this does not affect
239
already-compiled regexps. */
240
extern reg_syntax_t re_syntax_options;
241
242
#ifdef __USE_GNU_REGEX
243
/* Define combinations of the above bits for the standard possibilities.
244
(The [[[ comments delimit what gets put into the Texinfo file, so
245
don't delete them!) */
246
/* [[[begin syntaxes]]] */
247
# define RE_SYNTAX_EMACS 0
248
249
# define RE_SYNTAX_AWK \
250
(RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \
251
| RE_NO_BK_PARENS | RE_NO_BK_REFS \
252
| RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \
253
| RE_DOT_NEWLINE | RE_CONTEXT_INDEP_ANCHORS \
254
| RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS)
255
256
# define RE_SYNTAX_GNU_AWK \
257
((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DEBUG) \
258
& ~(RE_DOT_NOT_NULL | RE_INTERVALS | RE_CONTEXT_INDEP_OPS \
259
| RE_CONTEXT_INVALID_OPS ))
260
261
# define RE_SYNTAX_POSIX_AWK \
262
(RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS \
263
| RE_INTERVALS | RE_NO_GNU_OPS)
264
265
# define RE_SYNTAX_GREP \
266
(RE_BK_PLUS_QM | RE_CHAR_CLASSES \
267
| RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS \
268
| RE_NEWLINE_ALT)
269
270
# define RE_SYNTAX_EGREP \
271
(RE_CHAR_CLASSES | RE_CONTEXT_INDEP_ANCHORS \
272
| RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE \
273
| RE_NEWLINE_ALT | RE_NO_BK_PARENS \
274
| RE_NO_BK_VBAR)
275
276
# define RE_SYNTAX_POSIX_EGREP \
277
(RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES \
278
| RE_INVALID_INTERVAL_ORD)
279
280
/* P1003.2/D11.2, section 4.20.7.1, lines 5078ff. */
281
# define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
282
283
# define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
284
285
/* Syntax bits common to both basic and extended POSIX regex syntax. */
286
# define _RE_SYNTAX_POSIX_COMMON \
287
(RE_CHAR_CLASSES | RE_DOT_NEWLINE | RE_DOT_NOT_NULL \
288
| RE_INTERVALS | RE_NO_EMPTY_RANGES)
289
290
# define RE_SYNTAX_POSIX_BASIC \
291
(_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM | RE_CONTEXT_INVALID_DUP)
292
293
/* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
294
RE_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this
295
isn't minimal, since other operators, such as \`, aren't disabled. */
296
# define RE_SYNTAX_POSIX_MINIMAL_BASIC \
297
(_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
298
299
# define RE_SYNTAX_POSIX_EXTENDED \
300
(_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
301
| RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \
302
| RE_NO_BK_PARENS | RE_NO_BK_VBAR \
303
| RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD)
304
305
/* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is
306
removed and RE_NO_BK_REFS is added. */
307
# define RE_SYNTAX_POSIX_MINIMAL_EXTENDED \
308
(_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
309
| RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES \
310
| RE_NO_BK_PARENS | RE_NO_BK_REFS \
311
| RE_NO_BK_VBAR | RE_UNMATCHED_RIGHT_PAREN_ORD)
312
/* [[[end syntaxes]]] */
313
314
#endif /* defined __USE_GNU_REGEX */
315
316
#ifdef __USE_GNU_REGEX
317
318
/* Maximum number of duplicates an interval can allow. POSIX-conforming
319
systems might define this in <limits.h>, but we want our
320
value, so remove any previous define. */
321
# ifdef RE_DUP_MAX
322
# undef RE_DUP_MAX
323
# endif
324
325
/* RE_DUP_MAX is 2**15 - 1 because an earlier implementation stored
326
the counter as a 2-byte signed integer. This is no longer true, so
327
RE_DUP_MAX could be increased to (INT_MAX / 10 - 1), or to
328
((SIZE_MAX - 2) / 10 - 1) if _REGEX_LARGE_OFFSETS is defined.
329
However, there would be a huge performance problem if someone
330
actually used a pattern like a\{214748363\}, so RE_DUP_MAX retains
331
its historical value. */
332
# define RE_DUP_MAX (0x7fff)
333
334
#endif /* defined __USE_GNU_REGEX */
335
336
337
/* POSIX `cflags' bits (i.e., information for `regcomp'). */
338
339
/* If this bit is set, then use extended regular expression syntax.
340
If not set, then use basic regular expression syntax. */
341
#define REG_EXTENDED 1
342
343
/* If this bit is set, then ignore case when matching.
344
If not set, then case is significant. */
345
#define REG_ICASE (1 << 1)
346
347
/* If this bit is set, then anchors do not match at newline
348
characters in the string.
349
If not set, then anchors do match at newlines. */
350
#define REG_NEWLINE (1 << 2)
351
352
/* If this bit is set, then report only success or fail in regexec.
353
If not set, then returns differ between not matching and errors. */
354
#define REG_NOSUB (1 << 3)
355
356
357
/* POSIX `eflags' bits (i.e., information for regexec). */
358
359
/* If this bit is set, then the beginning-of-line operator doesn't match
360
the beginning of the string (presumably because it's not the
361
beginning of a line).
362
If not set, then the beginning-of-line operator does match the
363
beginning of the string. */
364
#define REG_NOTBOL 1
365
366
/* Like REG_NOTBOL, except for the end-of-line. */
367
#define REG_NOTEOL (1 << 1)
368
369
/* Use PMATCH[0] to delimit the start and end of the search in the
370
buffer. */
371
#define REG_STARTEND (1 << 2)
372
373
374
/* If any error codes are removed, changed, or added, update the
375
`__re_error_msgid' table in regcomp.c. */
376
377
typedef enum
378
{
379
_REG_ENOSYS = -1, /* This will never happen for this implementation. */
380
_REG_NOERROR = 0, /* Success. */
381
_REG_NOMATCH, /* Didn't find a match (for regexec). */
382
383
/* POSIX regcomp return error codes. (In the order listed in the
384
standard.) */
385
_REG_BADPAT, /* Invalid pattern. */
386
_REG_ECOLLATE, /* Invalid collating element. */
387
_REG_ECTYPE, /* Invalid character class name. */
388
_REG_EESCAPE, /* Trailing backslash. */
389
_REG_ESUBREG, /* Invalid back reference. */
390
_REG_EBRACK, /* Unmatched left bracket. */
391
_REG_EPAREN, /* Parenthesis imbalance. */
392
_REG_EBRACE, /* Unmatched \{. */
393
_REG_BADBR, /* Invalid contents of \{\}. */
394
_REG_ERANGE, /* Invalid range end. */
395
_REG_ESPACE, /* Ran out of memory. */
396
_REG_BADRPT, /* No preceding re for repetition op. */
397
398
/* Error codes we've added. */
399
_REG_EEND, /* Premature end. */
400
_REG_ESIZE, /* Compiled pattern bigger than 2^16 bytes. */
401
_REG_ERPAREN /* Unmatched ) or \); not returned from regcomp. */
402
} reg_errcode_t;
403
404
#ifdef _XOPEN_SOURCE
405
# define REG_ENOSYS _REG_ENOSYS
406
#endif
407
#define REG_NOERROR _REG_NOERROR
408
#define REG_NOMATCH _REG_NOMATCH
409
#define REG_BADPAT _REG_BADPAT
410
#define REG_ECOLLATE _REG_ECOLLATE
411
#define REG_ECTYPE _REG_ECTYPE
412
#define REG_EESCAPE _REG_EESCAPE
413
#define REG_ESUBREG _REG_ESUBREG
414
#define REG_EBRACK _REG_EBRACK
415
#define REG_EPAREN _REG_EPAREN
416
#define REG_EBRACE _REG_EBRACE
417
#define REG_BADBR _REG_BADBR
418
#define REG_ERANGE _REG_ERANGE
419
#define REG_ESPACE _REG_ESPACE
420
#define REG_BADRPT _REG_BADRPT
421
#define REG_EEND _REG_EEND
422
#define REG_ESIZE _REG_ESIZE
423
#define REG_ERPAREN _REG_ERPAREN
424
425
/* struct re_pattern_buffer normally uses member names like `buffer'
426
that POSIX does not allow. In POSIX mode these members have names
427
with leading `re_' (e.g., `re_buffer'). */
428
#ifdef __USE_GNU_REGEX
429
# define _REG_RE_NAME(id) id
430
# define _REG_RM_NAME(id) id
431
#else
432
# define _REG_RE_NAME(id) re_##id
433
# define _REG_RM_NAME(id) rm_##id
434
#endif
435
436
/* The user can specify the type of the re_translate member by
437
defining the macro RE_TRANSLATE_TYPE, which defaults to unsigned
438
char *. This pollutes the POSIX name space, so in POSIX mode just
439
use unsigned char *. */
440
#ifdef __USE_GNU_REGEX
441
# ifndef RE_TRANSLATE_TYPE
442
# define RE_TRANSLATE_TYPE unsigned char *
443
# endif
444
# define REG_TRANSLATE_TYPE RE_TRANSLATE_TYPE
445
#else
446
# define REG_TRANSLATE_TYPE unsigned char *
447
#endif
448
449
/* This data structure represents a compiled pattern. Before calling
450
the pattern compiler, the fields `buffer', `allocated', `fastmap',
451
`translate', and `no_sub' can be set. After the pattern has been
452
compiled, the `re_nsub' field is available. All other fields are
453
private to the regex routines. */
454
455
struct re_pattern_buffer
456
{
457
/* Space that holds the compiled pattern. It is declared as
458
`unsigned char *' because its elements are sometimes used as
459
array indexes. */
460
unsigned char *_REG_RE_NAME (buffer);
461
462
/* Number of bytes to which `buffer' points. */
463
__re_long_size_t _REG_RE_NAME (allocated);
464
465
/* Number of bytes actually used in `buffer'. */
466
__re_long_size_t _REG_RE_NAME (used);
467
468
/* Syntax setting with which the pattern was compiled. */
469
reg_syntax_t _REG_RE_NAME (syntax);
470
471
/* Pointer to a fastmap, if any, otherwise zero. re_search uses the
472
fastmap, if there is one, to skip over impossible starting points
473
for matches. */
474
char *_REG_RE_NAME (fastmap);
475
476
/* Either a translate table to apply to all characters before
477
comparing them, or zero for no translation. The translation is
478
applied to a pattern when it is compiled and to a string when it
479
is matched. */
480
REG_TRANSLATE_TYPE _REG_RE_NAME (translate);
481
482
/* Number of subexpressions found by the compiler. */
483
size_t re_nsub;
484
485
/* Zero if this pattern cannot match the empty string, one else.
486
Well, in truth it's used only in `re_search_2', to see whether or
487
not we should use the fastmap, so we don't set this absolutely
488
perfectly; see `re_compile_fastmap' (the `duplicate' case). */
489
unsigned int _REG_RE_NAME (can_be_null) : 1;
490
491
/* If REGS_UNALLOCATED, allocate space in the `regs' structure
492
for `max (RE_NREGS, re_nsub + 1)' groups.
493
If REGS_REALLOCATE, reallocate space if necessary.
494
If REGS_FIXED, use what's there. */
495
#ifdef __USE_GNU_REGEX
496
# define REGS_UNALLOCATED 0
497
# define REGS_REALLOCATE 1
498
# define REGS_FIXED 2
499
#endif
500
unsigned int _REG_RE_NAME (regs_allocated) : 2;
501
502
/* Set to zero when `re_compile_pattern' compiles a pattern; set to
503
one by `re_compile_fastmap' if it updates the fastmap. */
504
unsigned int _REG_RE_NAME (fastmap_accurate) : 1;
505
506
/* If set, `re_match_2' does not return information about
507
subexpressions. */
508
unsigned int _REG_RE_NAME (no_sub) : 1;
509
510
/* If set, a beginning-of-line anchor doesn't match at the beginning
511
of the string. */
512
unsigned int _REG_RE_NAME (not_bol) : 1;
513
514
/* Similarly for an end-of-line anchor. */
515
unsigned int _REG_RE_NAME (not_eol) : 1;
516
517
/* If true, an anchor at a newline matches. */
518
unsigned int _REG_RE_NAME (newline_anchor) : 1;
519
520
/* [[[end pattern_buffer]]] */
521
};
522
523
typedef struct re_pattern_buffer regex_t;
524
525
/* This is the structure we store register match data in. See
526
regex.texinfo for a full description of what registers match. */
527
struct re_registers
528
{
529
__re_size_t _REG_RM_NAME (num_regs);
530
regoff_t *_REG_RM_NAME (start);
531
regoff_t *_REG_RM_NAME (end);
532
};
533
534
535
/* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
536
`re_match_2' returns information about at least this many registers
537
the first time a `regs' structure is passed. */
538
#if !defined RE_NREGS && defined __USE_GNU_REGEX
539
# define RE_NREGS 30
540
#endif
541
542
543
/* POSIX specification for registers. Aside from the different names than
544
`re_registers', POSIX uses an array of structures, instead of a
545
structure of arrays. */
546
typedef struct
547
{
548
regoff_t rm_so; /* Byte offset from string's start to substring's start. */
549
regoff_t rm_eo; /* Byte offset from string's start to substring's end. */
550
} regmatch_t;
551
552
/* Declarations for routines. */
553
554
/* Sets the current default syntax to SYNTAX, and return the old syntax.
555
You can also simply assign to the `re_syntax_options' variable. */
556
extern reg_syntax_t re_set_syntax (reg_syntax_t __syntax);
557
558
/* Compile the regular expression PATTERN, with length LENGTH
559
and syntax given by the global `re_syntax_options', into the buffer
560
BUFFER. Return NULL if successful, and an error string if not. */
561
extern const char *re_compile_pattern (const char *__pattern, size_t __length,
562
struct re_pattern_buffer *__buffer);
563
564
565
/* Compile a fastmap for the compiled pattern in BUFFER; used to
566
accelerate searches. Return 0 if successful and -2 if was an
567
internal error. */
568
extern int re_compile_fastmap (struct re_pattern_buffer *__buffer);
569
570
571
/* Search in the string STRING (with length LENGTH) for the pattern
572
compiled into BUFFER. Start searching at position START, for RANGE
573
characters. Return the starting position of the match, -1 for no
574
match, or -2 for an internal error. Also return register
575
information in REGS (if REGS and BUFFER->no_sub are nonzero). */
576
extern regoff_t re_search (struct re_pattern_buffer *__buffer,
577
const char *__string, __re_idx_t __length,
578
__re_idx_t __start, regoff_t __range,
579
struct re_registers *__regs);
580
581
582
/* Like `re_search', but search in the concatenation of STRING1 and
583
STRING2. Also, stop searching at index START + STOP. */
584
extern regoff_t re_search_2 (struct re_pattern_buffer *__buffer,
585
const char *__string1, __re_idx_t __length1,
586
const char *__string2, __re_idx_t __length2,
587
__re_idx_t __start, regoff_t __range,
588
struct re_registers *__regs,
589
__re_idx_t __stop);
590
591
592
/* Like `re_search', but return how many characters in STRING the regexp
593
in BUFFER matched, starting at position START. */
594
extern regoff_t re_match (struct re_pattern_buffer *__buffer,
595
const char *__string, __re_idx_t __length,
596
__re_idx_t __start, struct re_registers *__regs);
597
598
599
/* Relates to `re_match' as `re_search_2' relates to `re_search'. */
600
extern regoff_t re_match_2 (struct re_pattern_buffer *__buffer,
601
const char *__string1, __re_idx_t __length1,
602
const char *__string2, __re_idx_t __length2,
603
__re_idx_t __start, struct re_registers *__regs,
604
__re_idx_t __stop);
605
606
607
/* Set REGS to hold NUM_REGS registers, storing them in STARTS and
608
ENDS. Subsequent matches using BUFFER and REGS will use this memory
609
for recording register information. STARTS and ENDS must be
610
allocated with malloc, and must each be at least `NUM_REGS * sizeof
611
(regoff_t)' bytes long.
612
613
If NUM_REGS == 0, then subsequent matches should allocate their own
614
register data.
615
616
Unless this function is called, the first search or match using
617
BUFFER will allocate its own register data, without freeing the old
618
data. */
619
extern void re_set_registers (struct re_pattern_buffer *__buffer,
620
struct re_registers *__regs,
621
__re_size_t __num_regs,
622
regoff_t *__starts, regoff_t *__ends);
623
624
#if defined _REGEX_RE_COMP || defined _LIBC
625
# ifndef _CRAY
626
/* 4.2 bsd compatibility. */
627
extern char *re_comp (const char *);
628
extern int re_exec (const char *);
629
# endif
630
#endif
631
632
/* GCC 2.95 and later have "__restrict"; C99 compilers have
633
"restrict", and "configure" may have defined "restrict".
634
Other compilers use __restrict, __restrict__, and _Restrict, and
635
'configure' might #define 'restrict' to those words, so pick a
636
different name. */
637
#ifndef _Restrict_
638
# if 199901L <= __STDC_VERSION__
639
# define _Restrict_ restrict
640
# elif 2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__)
641
# define _Restrict_ __restrict
642
# else
643
# define _Restrict_
644
# endif
645
#endif
646
/* gcc 3.1 and up support the [restrict] syntax. Don't trust
647
sys/cdefs.h's definition of __restrict_arr, though, as it
648
mishandles gcc -ansi -pedantic. */
649
#ifndef _Restrict_arr_
650
# if ((199901L <= __STDC_VERSION__ \
651
|| ((3 < __GNUC__ || (3 == __GNUC__ && 1 <= __GNUC_MINOR__)) \
652
&& !__STRICT_ANSI__)) \
653
&& !defined __GNUG__)
654
# define _Restrict_arr_ _Restrict_
655
# else
656
# define _Restrict_arr_
657
# endif
658
#endif
659
660
/* POSIX compatibility. */
661
extern int regcomp (regex_t *_Restrict_ __preg,
662
const char *_Restrict_ __pattern,
663
int __cflags);
664
665
extern int regexec (const regex_t *_Restrict_ __preg,
666
const char *_Restrict_ __string, size_t __nmatch,
667
regmatch_t __pmatch[_Restrict_arr_],
668
int __eflags);
669
670
extern size_t regerror (int __errcode, const regex_t *_Restrict_ __preg,
671
char *_Restrict_ __errbuf, size_t __errbuf_size);
672
673
extern void regfree (regex_t *__preg);
674
675
676
#ifdef __cplusplus
677
}
678
#endif /* C++ */
679
680
#endif /* regex.h */
Loggerhead is a web-based interface for Breezy
Version: 2.0.1