regex.h - enscript - GNU Enscript
(HTM) git clone git://thinkerwim.org/enscript.git
(DIR) Log
(DIR) Files
(DIR) Refs
(DIR) README
(DIR) LICENSE
---
regex.h (20549B)
---
1 /* Definitions for data structures and routines for the regular
2 expression library, version 0.12.
3 Copyright (C) 1985,89,90,91,92,93,95,96,97 Free Software Foundation, Inc.
4
5 This program is free software; you can redistribute it and/or modify
6 it under the terms of the GNU General Public License as published by
7 the Free Software Foundation; either version 2, or (at your option)
8 any later version.
9
10 This program is distributed in the hope that it will be useful,
11 but WITHOUT ANY WARRANTY; without even the implied warranty of
12 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 GNU General Public License for more details.
14
15 You should have received a copy of the GNU General Public License
16 along with this program; if not, write to the Free Software Foundation,
17 Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */
18
19 #ifndef __REGEXP_LIBRARY_H__
20 #define __REGEXP_LIBRARY_H__
21
22 /* Allow the use in C++ code. */
23 #ifdef __cplusplus
24 extern "C" {
25 #endif
26
27 /* POSIX says that <sys/types.h> must be included (by the caller) before
28 <regex.h>. */
29
30 #if !defined (_POSIX_C_SOURCE) && !defined (_POSIX_SOURCE) && defined (VMS)
31 /* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it
32 should be there. */
33 #include <stddef.h>
34 #endif
35
36 /* The following two types have to be signed and unsigned integer type
37 wide enough to hold a value of a pointer. For most ANSI compilers
38 ptrdiff_t and size_t should be likely OK. Still size of these two
39 types is 2 for Microsoft C. Ugh... */
40 typedef long int s_reg_t;
41 typedef unsigned long int active_reg_t;
42
43 /* The following bits are used to determine the regexp syntax we
44 recognize. The set/not-set meanings are chosen so that Emacs syntax
45 remains the value 0. The bits are given in alphabetical order, and
46 the definitions shifted by one from the previous bit; thus, when we
47 add or remove a bit, only one other definition need change. */
48 typedef unsigned long int reg_syntax_t;
49
50 /* If this bit is not set, then \ inside a bracket expression is literal.
51 If set, then such a \ quotes the following character. */
52 #define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1)
53
54 /* If this bit is not set, then + and ? are operators, and \+ and \? are
55 literals.
56 If set, then \+ and \? are operators and + and ? are literals. */
57 #define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
58
59 /* If this bit is set, then character classes are supported. They are:
60 [:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:],
61 [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
62 If not set, then character classes are not supported. */
63 #define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
64
65 /* If this bit is set, then ^ and $ are always anchors (outside bracket
66 expressions, of course).
67 If this bit is not set, then it depends:
68 ^ is an anchor if it is at the beginning of a regular
69 expression or after an open-group or an alternation operator;
70 $ is an anchor if it is at the end of a regular expression, or
71 before a close-group or an alternation operator.
72
73 This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
74 POSIX draft 11.2 says that * etc. in leading positions is undefined.
75 We already implemented a previous draft which made those constructs
76 invalid, though, so we haven't changed the code back. */
77 #define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
78
79 /* If this bit is set, then special characters are always special
80 regardless of where they are in the pattern.
81 If this bit is not set, then special characters are special only in
82 some contexts; otherwise they are ordinary. Specifically,
83 * + ? and intervals are only special when not after the beginning,
84 open-group, or alternation operator. */
85 #define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
86
87 /* If this bit is set, then *, +, ?, and { cannot be first in an re or
88 immediately after an alternation or begin-group operator. */
89 #define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
90
91 /* If this bit is set, then . matches newline.
92 If not set, then it doesn't. */
93 #define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
94
95 /* If this bit is set, then . doesn't match NUL.
96 If not set, then it does. */
97 #define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
98
99 /* If this bit is set, nonmatching lists [^...] do not match newline.
100 If not set, they do. */
101 #define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
102
103 /* If this bit is set, either \{...\} or {...} defines an
104 interval, depending on RE_NO_BK_BRACES.
105 If not set, \{, \}, {, and } are literals. */
106 #define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
107
108 /* If this bit is set, +, ? and | aren't recognized as operators.
109 If not set, they are. */
110 #define RE_LIMITED_OPS (RE_INTERVALS << 1)
111
112 /* If this bit is set, newline is an alternation operator.
113 If not set, newline is literal. */
114 #define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
115
116 /* If this bit is set, then `{...}' defines an interval, and \{ and \}
117 are literals.
118 If not set, then `\{...\}' defines an interval. */
119 #define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
120
121 /* If this bit is set, (...) defines a group, and \( and \) are literals.
122 If not set, \(...\) defines a group, and ( and ) are literals. */
123 #define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
124
125 /* If this bit is set, then \<digit> matches <digit>.
126 If not set, then \<digit> is a back-reference. */
127 #define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
128
129 /* If this bit is set, then | is an alternation operator, and \| is literal.
130 If not set, then \| is an alternation operator, and | is literal. */
131 #define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
132
133 /* If this bit is set, then an ending range point collating higher
134 than the starting range point, as in [z-a], is invalid.
135 If not set, then when ending range point collates higher than the
136 starting range point, the range is ignored. */
137 #define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
138
139 /* If this bit is set, then an unmatched ) is ordinary.
140 If not set, then an unmatched ) is invalid. */
141 #define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
142
143 /* If this bit is set, succeed as soon as we match the whole pattern,
144 without further backtracking. */
145 #define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1)
146
147 /* If this bit is set, do not process the GNU regex operators.
148 If not set, then the GNU regex operators are recognized. */
149 #define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1)
150
151 /* If this bit is set, turn on internal regex debugging.
152 If not set, and debugging was on, turn it off.
153 This only works if regex.c is compiled -DDEBUG.
154 We define this bit always, so that all that's needed to turn on
155 debugging is to recompile regex.c; the calling code can always have
156 this bit set, and it won't affect anything in the normal case. */
157 #define RE_DEBUG (RE_NO_GNU_OPS << 1)
158
159 /* This global variable defines the particular regexp syntax to use (for
160 some interfaces). When a regexp is compiled, the syntax used is
161 stored in the pattern buffer, so changing this does not affect
162 already-compiled regexps. */
163 extern reg_syntax_t re_syntax_options;
164
165 /* Define combinations of the above bits for the standard possibilities.
166 (The [[[ comments delimit what gets put into the Texinfo file, so
167 don't delete them!) */
168 /* [[[begin syntaxes]]] */
169 #define RE_SYNTAX_EMACS 0
170
171 #define RE_SYNTAX_AWK \
172 (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \
173 | RE_NO_BK_PARENS | RE_NO_BK_REFS \
174 | RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \
175 | RE_DOT_NEWLINE | RE_CONTEXT_INDEP_ANCHORS \
176 | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS)
177
178 #define RE_SYNTAX_GNU_AWK \
179 ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DEBUG) \
180 & ~(RE_DOT_NOT_NULL | RE_INTERVALS | RE_CONTEXT_INDEP_OPS))
181
182 #define RE_SYNTAX_POSIX_AWK \
183 (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS \
184 | RE_INTERVALS | RE_NO_GNU_OPS)
185
186 #define RE_SYNTAX_GREP \
187 (RE_BK_PLUS_QM | RE_CHAR_CLASSES \
188 | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS \
189 | RE_NEWLINE_ALT)
190
191 #define RE_SYNTAX_EGREP \
192 (RE_CHAR_CLASSES | RE_CONTEXT_INDEP_ANCHORS \
193 | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE \
194 | RE_NEWLINE_ALT | RE_NO_BK_PARENS \
195 | RE_NO_BK_VBAR)
196
197 #define RE_SYNTAX_POSIX_EGREP \
198 (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES)
199
200 /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff. */
201 #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
202
203 #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
204
205 /* Syntax bits common to both basic and extended POSIX regex syntax. */
206 #define _RE_SYNTAX_POSIX_COMMON \
207 (RE_CHAR_CLASSES | RE_DOT_NEWLINE | RE_DOT_NOT_NULL \
208 | RE_INTERVALS | RE_NO_EMPTY_RANGES)
209
210 #define RE_SYNTAX_POSIX_BASIC \
211 (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM)
212
213 /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
214 RE_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this
215 isn't minimal, since other operators, such as \`, aren't disabled. */
216 #define RE_SYNTAX_POSIX_MINIMAL_BASIC \
217 (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
218
219 #define RE_SYNTAX_POSIX_EXTENDED \
220 (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
221 | RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \
222 | RE_NO_BK_PARENS | RE_NO_BK_VBAR \
223 | RE_UNMATCHED_RIGHT_PAREN_ORD)
224
225 /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INVALID_OPS
226 replaces RE_CONTEXT_INDEP_OPS and RE_NO_BK_REFS is added. */
227 #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED \
228 (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
229 | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES \
230 | RE_NO_BK_PARENS | RE_NO_BK_REFS \
231 | RE_NO_BK_VBAR | RE_UNMATCHED_RIGHT_PAREN_ORD)
232 /* [[[end syntaxes]]] */
233
234 /* Maximum number of duplicates an interval can allow. Some systems
235 (erroneously) define this in other header files, but we want our
236 value, so remove any previous define. */
237 #ifdef RE_DUP_MAX
238 #undef RE_DUP_MAX
239 #endif
240 /* If sizeof(int) == 2, then ((1 << 15) - 1) overflows. */
241 #define RE_DUP_MAX (0x7fff)
242
243
244 /* POSIX `cflags' bits (i.e., information for `regcomp'). */
245
246 /* If this bit is set, then use extended regular expression syntax.
247 If not set, then use basic regular expression syntax. */
248 #define REG_EXTENDED 1
249
250 /* If this bit is set, then ignore case when matching.
251 If not set, then case is significant. */
252 #define REG_ICASE (REG_EXTENDED << 1)
253
254 /* If this bit is set, then anchors do not match at newline
255 characters in the string.
256 If not set, then anchors do match at newlines. */
257 #define REG_NEWLINE (REG_ICASE << 1)
258
259 /* If this bit is set, then report only success or fail in regexec.
260 If not set, then returns differ between not matching and errors. */
261 #define REG_NOSUB (REG_NEWLINE << 1)
262
263
264 /* POSIX `eflags' bits (i.e., information for regexec). */
265
266 /* If this bit is set, then the beginning-of-line operator doesn't match
267 the beginning of the string (presumably because it's not the
268 beginning of a line).
269 If not set, then the beginning-of-line operator does match the
270 beginning of the string. */
271 #define REG_NOTBOL 1
272
273 /* Like REG_NOTBOL, except for the end-of-line. */
274 #define REG_NOTEOL (1 << 1)
275
276
277 /* If any error codes are removed, changed, or added, update the
278 `re_error_msg' table in regex.c. */
279 typedef enum
280 {
281 REG_NOERROR = 0, /* Success. */
282 REG_NOMATCH, /* Didn't find a match (for regexec). */
283
284 /* POSIX regcomp return error codes. (In the order listed in the
285 standard.) */
286 REG_BADPAT, /* Invalid pattern. */
287 REG_ECOLLATE, /* Not implemented. */
288 REG_ECTYPE, /* Invalid character class name. */
289 REG_EESCAPE, /* Trailing backslash. */
290 REG_ESUBREG, /* Invalid back reference. */
291 REG_EBRACK, /* Unmatched left bracket. */
292 REG_EPAREN, /* Parenthesis imbalance. */
293 REG_EBRACE, /* Unmatched \{. */
294 REG_BADBR, /* Invalid contents of \{\}. */
295 REG_ERANGE, /* Invalid range end. */
296 REG_ESPACE, /* Ran out of memory. */
297 REG_BADRPT, /* No preceding re for repetition op. */
298
299 /* Error codes we've added. */
300 REG_EEND, /* Premature end. */
301 REG_ESIZE, /* Compiled pattern bigger than 2^16 bytes. */
302 REG_ERPAREN /* Unmatched ) or \); not returned from regcomp. */
303 } reg_errcode_t;
304
305 /* This data structure represents a compiled pattern. Before calling
306 the pattern compiler, the fields `buffer', `allocated', `fastmap',
307 `translate', and `no_sub' can be set. After the pattern has been
308 compiled, the `re_nsub' field is available. All other fields are
309 private to the regex routines. */
310
311 #ifndef RE_TRANSLATE_TYPE
312 #define RE_TRANSLATE_TYPE char *
313 #endif
314
315 struct re_pattern_buffer
316 {
317 /* [[[begin pattern_buffer]]] */
318 /* Space that holds the compiled pattern. It is declared as
319 `unsigned char *' because its elements are
320 sometimes used as array indexes. */
321 unsigned char *buffer;
322
323 /* Number of bytes to which `buffer' points. */
324 unsigned long int allocated;
325
326 /* Number of bytes actually used in `buffer'. */
327 unsigned long int used;
328
329 /* Syntax setting with which the pattern was compiled. */
330 reg_syntax_t syntax;
331
332 /* Pointer to a fastmap, if any, otherwise zero. re_search uses
333 the fastmap, if there is one, to skip over impossible
334 starting points for matches. */
335 char *fastmap;
336
337 /* Either a translate table to apply to all characters before
338 comparing them, or zero for no translation. The translation
339 is applied to a pattern when it is compiled and to a string
340 when it is matched. */
341 RE_TRANSLATE_TYPE translate;
342
343 /* Number of subexpressions found by the compiler. */
344 size_t re_nsub;
345
346 /* Zero if this pattern cannot match the empty string, one else.
347 Well, in truth it's used only in `re_search_2', to see
348 whether or not we should use the fastmap, so we don't set
349 this absolutely perfectly; see `re_compile_fastmap' (the
350 `duplicate' case). */
351 unsigned can_be_null : 1;
352
353 /* If REGS_UNALLOCATED, allocate space in the `regs' structure
354 for `max (RE_NREGS, re_nsub + 1)' groups.
355 If REGS_REALLOCATE, reallocate space if necessary.
356 If REGS_FIXED, use what's there. */
357 #define REGS_UNALLOCATED 0
358 #define REGS_REALLOCATE 1
359 #define REGS_FIXED 2
360 unsigned regs_allocated : 2;
361
362 /* Set to zero when `regex_compile' compiles a pattern; set to one
363 by `re_compile_fastmap' if it updates the fastmap. */
364 unsigned fastmap_accurate : 1;
365
366 /* If set, `re_match_2' does not return information about
367 subexpressions. */
368 unsigned no_sub : 1;
369
370 /* If set, a beginning-of-line anchor doesn't match at the
371 beginning of the string. */
372 unsigned not_bol : 1;
373
374 /* Similarly for an end-of-line anchor. */
375 unsigned not_eol : 1;
376
377 /* If true, an anchor at a newline matches. */
378 unsigned newline_anchor : 1;
379
380 /* [[[end pattern_buffer]]] */
381 };
382
383 typedef struct re_pattern_buffer regex_t;
384
385 /* Type for byte offsets within the string. POSIX mandates this. */
386 typedef int regoff_t;
387
388
389 /* This is the structure we store register match data in. See
390 regex.texinfo for a full description of what registers match. */
391 struct re_registers
392 {
393 unsigned num_regs;
394 regoff_t *start;
395 regoff_t *end;
396 };
397
398
399 /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
400 `re_match_2' returns information about at least this many registers
401 the first time a `regs' structure is passed. */
402 #ifndef RE_NREGS
403 #define RE_NREGS 30
404 #endif
405
406
407 /* POSIX specification for registers. Aside from the different names than
408 `re_registers', POSIX uses an array of structures, instead of a
409 structure of arrays. */
410 typedef struct
411 {
412 regoff_t rm_so; /* Byte offset from string's start to substring's start. */
413 regoff_t rm_eo; /* Byte offset from string's start to substring's end. */
414 } regmatch_t;
415
416 /* Declarations for routines. */
417
418 /* To avoid duplicating every routine declaration -- once with a
419 prototype (if we are ANSI), and once without (if we aren't) -- we
420 use the following macro to declare argument types. This
421 unfortunately clutters up the declarations a bit, but I think it's
422 worth it. */
423
424 #if __STDC__
425
426 #define _RE_ARGS(args) args
427
428 #else /* not __STDC__ */
429
430 #define _RE_ARGS(args) ()
431
432 #endif /* not __STDC__ */
433
434 /* Sets the current default syntax to SYNTAX, and return the old syntax.
435 You can also simply assign to the `re_syntax_options' variable. */
436 extern reg_syntax_t re_set_syntax _RE_ARGS ((reg_syntax_t syntax));
437
438 /* Sets the character syntax for character CH. Possible values for SYNTAX
439 are 'w' for word characters and ' ' for non-word characters. */
440 extern void re_set_character_syntax _RE_ARGS ((unsigned char ch, char syntax));
441
442 /* Compile the regular expression PATTERN, with length LENGTH
443 and syntax given by the global `re_syntax_options', into the buffer
444 BUFFER. Return NULL if successful, and an error string if not. */
445 extern const char *re_compile_pattern
446 _RE_ARGS ((const char *pattern, size_t length,
447 struct re_pattern_buffer *buffer));
448
449
450 /* Compile a fastmap for the compiled pattern in BUFFER; used to
451 accelerate searches. Return 0 if successful and -2 if was an
452 internal error. */
453 extern int re_compile_fastmap _RE_ARGS ((struct re_pattern_buffer *buffer));
454
455
456 /* Search in the string STRING (with length LENGTH) for the pattern
457 compiled into BUFFER. Start searching at position START, for RANGE
458 characters. Return the starting position of the match, -1 for no
459 match, or -2 for an internal error. Also return register
460 information in REGS (if REGS and BUFFER->no_sub are nonzero). */
461 extern int re_search
462 _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
463 int length, int start, int range, struct re_registers *regs));
464
465
466 /* Like `re_search', but search in the concatenation of STRING1 and
467 STRING2. Also, stop searching at index START + STOP. */
468 extern int re_search_2
469 _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
470 int length1, const char *string2, int length2,
471 int start, int range, struct re_registers *regs, int stop));
472
473
474 /* Like `re_search', but return how many characters in STRING the regexp
475 in BUFFER matched, starting at position START. */
476 extern int re_match
477 _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
478 int length, int start, struct re_registers *regs));
479
480
481 /* Relates to `re_match' as `re_search_2' relates to `re_search'. */
482 extern int re_match_2
483 _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
484 int length1, const char *string2, int length2,
485 int start, struct re_registers *regs, int stop));
486
487
488 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and
489 ENDS. Subsequent matches using BUFFER and REGS will use this memory
490 for recording register information. STARTS and ENDS must be
491 allocated with malloc, and must each be at least `NUM_REGS * sizeof
492 (regoff_t)' bytes long.
493
494 If NUM_REGS == 0, then subsequent matches should allocate their own
495 register data.
496
497 Unless this function is called, the first search or match using
498 PATTERN_BUFFER will allocate its own register data, without
499 freeing the old data. */
500 extern void re_set_registers
501 _RE_ARGS ((struct re_pattern_buffer *buffer, struct re_registers *regs,
502 unsigned num_regs, regoff_t *starts, regoff_t *ends));
503
504 #ifdef _REGEX_RE_COMP
505 #ifndef _CRAY
506 /* 4.2 bsd compatibility. */
507 extern char *re_comp _RE_ARGS ((const char *));
508 extern int re_exec _RE_ARGS ((const char *));
509 #endif
510 #endif
511
512 /* POSIX compatibility. */
513 extern int regcomp _RE_ARGS ((regex_t *preg, const char *pattern, int cflags));
514 extern int regexec
515 _RE_ARGS ((const regex_t *preg, const char *string, size_t nmatch,
516 regmatch_t pmatch[], int eflags));
517 extern size_t regerror
518 _RE_ARGS ((int errcode, const regex_t *preg, char *errbuf,
519 size_t errbuf_size));
520 extern void regfree _RE_ARGS ((regex_t *preg));
521
522
523 #ifdef __cplusplus
524 }
525 #endif /* C++ */
526
527 #endif /* not __REGEXP_LIBRARY_H__ */
528
529 /*
530 Local variables:
531 make-backup-files: t
532 version-control: t
533 trim-versions-without-asking: nil
534 End:
535 */