1 /* Definitions for data structures and routines for the regular
2 expression library, version 0.12.
4 Copyright (C) 1985, 1989-1993, 1995, 2000-2016 Free Software
7 This program is free software; you can redistribute it and/or modify
8 it under the terms of the GNU General Public License as published by
9 the Free Software Foundation; either version 3, or (at your option)
12 This program is distributed in the hope that it will be useful,
13 but WITHOUT ANY WARRANTY; without even the implied warranty of
14 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 GNU General Public License for more details.
17 You should have received a copy of the GNU General Public License
18 along with this program. If not, see <http://www.gnu.org/licenses/>. */
23 /* Allow the use in C++ code. */
28 /* POSIX says that <sys/types.h> must be included (by the caller) before
31 #if !defined _POSIX_C_SOURCE && !defined _POSIX_SOURCE && defined VMS
32 /* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it
37 /* The following bits are used to determine the regexp syntax we
38 recognize. The set/not-set meanings where historically chosen so
39 that Emacs syntax had the value 0.
40 The bits are given in alphabetical order, and
41 the definitions shifted by one from the previous bit; thus, when we
42 add or remove a bit, only one other definition need change. */
43 typedef unsigned long reg_syntax_t
;
45 /* If this bit is not set, then \ inside a bracket expression is literal.
46 If set, then such a \ quotes the following character. */
47 #define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1)
49 /* If this bit is not set, then + and ? are operators, and \+ and \? are
51 If set, then \+ and \? are operators and + and ? are literals. */
52 #define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
54 /* If this bit is set, then character classes are supported. They are:
55 [:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:],
56 [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
57 If not set, then character classes are not supported. */
58 #define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
60 /* If this bit is set, then ^ and $ are always anchors (outside bracket
61 expressions, of course).
62 If this bit is not set, then it depends:
63 ^ is an anchor if it is at the beginning of a regular
64 expression or after an open-group or an alternation operator;
65 $ is an anchor if it is at the end of a regular expression, or
66 before a close-group or an alternation operator.
68 This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
69 POSIX draft 11.2 says that * etc. in leading positions is undefined.
70 We already implemented a previous draft which made those constructs
71 invalid, though, so we haven't changed the code back. */
72 #define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
74 /* If this bit is set, then special characters are always special
75 regardless of where they are in the pattern.
76 If this bit is not set, then special characters are special only in
77 some contexts; otherwise they are ordinary. Specifically,
78 * + ? and intervals are only special when not after the beginning,
79 open-group, or alternation operator. */
80 #define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
82 /* If this bit is set, then *, +, ?, and { cannot be first in an re or
83 immediately after an alternation or begin-group operator. */
84 #define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
86 /* If this bit is set, then . matches newline.
87 If not set, then it doesn't. */
88 #define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
90 /* If this bit is set, then . doesn't match NUL.
91 If not set, then it does. */
92 #define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
94 /* If this bit is set, nonmatching lists [^...] do not match newline.
95 If not set, they do. */
96 #define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
98 /* If this bit is set, either \{...\} or {...} defines an
99 interval, depending on RE_NO_BK_BRACES.
100 If not set, \{, \}, {, and } are literals. */
101 #define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
103 /* If this bit is set, +, ? and | aren't recognized as operators.
104 If not set, they are. */
105 #define RE_LIMITED_OPS (RE_INTERVALS << 1)
107 /* If this bit is set, newline is an alternation operator.
108 If not set, newline is literal. */
109 #define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
111 /* If this bit is set, then `{...}' defines an interval, and \{ and \}
113 If not set, then `\{...\}' defines an interval. */
114 #define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
116 /* If this bit is set, (...) defines a group, and \( and \) are literals.
117 If not set, \(...\) defines a group, and ( and ) are literals. */
118 #define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
120 /* If this bit is set, then \<digit> matches <digit>.
121 If not set, then \<digit> is a back-reference. */
122 #define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
124 /* If this bit is set, then | is an alternation operator, and \| is literal.
125 If not set, then \| is an alternation operator, and | is literal. */
126 #define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
128 /* If this bit is set, then an ending range point collating higher
129 than the starting range point, as in [z-a], is invalid.
130 If not set, then when ending range point collates higher than the
131 starting range point, the range is ignored. */
132 #define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
134 /* If this bit is set, then an unmatched ) is ordinary.
135 If not set, then an unmatched ) is invalid. */
136 #define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
138 /* If this bit is set, succeed as soon as we match the whole pattern,
139 without further backtracking. */
140 #define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1)
142 /* If this bit is set, do not process the GNU regex operators.
143 If not set, then the GNU regex operators are recognized. */
144 #define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1)
146 /* If this bit is set, then *?, +? and ?? match non greedily. */
147 #define RE_FRUGAL (RE_NO_GNU_OPS << 1)
149 /* If this bit is set, then (?:...) is treated as a shy group. */
150 #define RE_SHY_GROUPS (RE_FRUGAL << 1)
152 /* If this bit is set, ^ and $ only match at beg/end of buffer. */
153 #define RE_NO_NEWLINE_ANCHOR (RE_SHY_GROUPS << 1)
155 /* If this bit is set, turn on internal regex debugging.
156 If not set, and debugging was on, turn it off.
157 This only works if regex.c is compiled -DDEBUG.
158 We define this bit always, so that all that's needed to turn on
159 debugging is to recompile regex.c; the calling code can always have
160 this bit set, and it won't affect anything in the normal case. */
161 #define RE_DEBUG (RE_NO_NEWLINE_ANCHOR << 1)
163 /* This global variable defines the particular regexp syntax to use (for
164 some interfaces). When a regexp is compiled, the syntax used is
165 stored in the pattern buffer, so changing this does not affect
166 already-compiled regexps. */
167 extern reg_syntax_t re_syntax_options
;
171 /* In Emacs, this is the string or buffer in which we
172 are matching. It is used for looking up syntax properties. */
173 extern Lisp_Object re_match_object
;
176 /* Roughly the maximum number of failure points on the stack. */
177 extern size_t re_max_failures
;
180 /* Define combinations of the above bits for the standard possibilities.
181 (The [[[ comments delimit what gets put into the Texinfo file, so
182 don't delete them!) */
183 /* [[[begin syntaxes]]] */
184 #define RE_SYNTAX_EMACS \
185 (RE_CHAR_CLASSES | RE_INTERVALS | RE_SHY_GROUPS | RE_FRUGAL)
187 #define RE_SYNTAX_AWK \
188 (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \
189 | RE_NO_BK_PARENS | RE_NO_BK_REFS \
190 | RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \
191 | RE_DOT_NEWLINE | RE_CONTEXT_INDEP_ANCHORS \
192 | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS)
194 #define RE_SYNTAX_GNU_AWK \
195 ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DEBUG) \
196 & ~(RE_DOT_NOT_NULL | RE_INTERVALS | RE_CONTEXT_INDEP_OPS))
198 #define RE_SYNTAX_POSIX_AWK \
199 (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS \
200 | RE_INTERVALS | RE_NO_GNU_OPS)
202 #define RE_SYNTAX_GREP \
203 (RE_BK_PLUS_QM | RE_CHAR_CLASSES \
204 | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS \
207 #define RE_SYNTAX_EGREP \
208 (RE_CHAR_CLASSES | RE_CONTEXT_INDEP_ANCHORS \
209 | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE \
210 | RE_NEWLINE_ALT | RE_NO_BK_PARENS \
213 #define RE_SYNTAX_POSIX_EGREP \
214 (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES)
216 /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff. */
217 #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
219 #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
221 /* Syntax bits common to both basic and extended POSIX regex syntax. */
222 #define _RE_SYNTAX_POSIX_COMMON \
223 (RE_CHAR_CLASSES | RE_DOT_NEWLINE | RE_DOT_NOT_NULL \
224 | RE_INTERVALS | RE_NO_EMPTY_RANGES)
226 #define RE_SYNTAX_POSIX_BASIC \
227 (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM)
229 /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
230 RE_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this
231 isn't minimal, since other operators, such as \`, aren't disabled. */
232 #define RE_SYNTAX_POSIX_MINIMAL_BASIC \
233 (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
235 #define RE_SYNTAX_POSIX_EXTENDED \
236 (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
237 | RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \
238 | RE_NO_BK_PARENS | RE_NO_BK_VBAR \
239 | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD)
241 /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is
242 removed and RE_NO_BK_REFS is added. */
243 #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED \
244 (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
245 | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES \
246 | RE_NO_BK_PARENS | RE_NO_BK_REFS \
247 | RE_NO_BK_VBAR | RE_UNMATCHED_RIGHT_PAREN_ORD)
248 /* [[[end syntaxes]]] */
250 /* Maximum number of duplicates an interval can allow. Some systems
251 (erroneously) define this in other header files, but we want our
252 value, so remove any previous define. */
256 /* If sizeof(int) == 2, then ((1 << 15) - 1) overflows. */
257 #define RE_DUP_MAX (0x7fff)
260 /* POSIX `cflags' bits (i.e., information for `regcomp'). */
262 /* If this bit is set, then use extended regular expression syntax.
263 If not set, then use basic regular expression syntax. */
264 #define REG_EXTENDED 1
266 /* If this bit is set, then ignore case when matching.
267 If not set, then case is significant. */
268 #define REG_ICASE (REG_EXTENDED << 1)
270 /* If this bit is set, then anchors do not match at newline
271 characters in the string.
272 If not set, then anchors do match at newlines. */
273 #define REG_NEWLINE (REG_ICASE << 1)
275 /* If this bit is set, then report only success or fail in regexec.
276 If not set, then returns differ between not matching and errors. */
277 #define REG_NOSUB (REG_NEWLINE << 1)
280 /* POSIX `eflags' bits (i.e., information for regexec). */
282 /* If this bit is set, then the beginning-of-line operator doesn't match
283 the beginning of the string (presumably because it's not the
284 beginning of a line).
285 If not set, then the beginning-of-line operator does match the
286 beginning of the string. */
289 /* Like REG_NOTBOL, except for the end-of-line. */
290 #define REG_NOTEOL (1 << 1)
293 /* If any error codes are removed, changed, or added, update the
294 `re_error_msg' table in regex.c. */
298 REG_ENOSYS
= -1, /* This will never happen for this implementation. */
301 REG_NOERROR
= 0, /* Success. */
302 REG_NOMATCH
, /* Didn't find a match (for regexec). */
304 /* POSIX regcomp return error codes. (In the order listed in the
306 REG_BADPAT
, /* Invalid pattern. */
307 REG_ECOLLATE
, /* Not implemented. */
308 REG_ECTYPE
, /* Invalid character class name. */
309 REG_EESCAPE
, /* Trailing backslash. */
310 REG_ESUBREG
, /* Invalid back reference. */
311 REG_EBRACK
, /* Unmatched left bracket. */
312 REG_EPAREN
, /* Parenthesis imbalance. */
313 REG_EBRACE
, /* Unmatched \{. */
314 REG_BADBR
, /* Invalid contents of \{\}. */
315 REG_ERANGE
, /* Invalid range end. */
316 REG_ESPACE
, /* Ran out of memory. */
317 REG_BADRPT
, /* No preceding re for repetition op. */
319 /* Error codes we've added. */
320 REG_EEND
, /* Premature end. */
321 REG_ESIZE
, /* Compiled pattern bigger than 2^16 bytes. */
322 REG_ERPAREN
, /* Unmatched ) or \); not returned from regcomp. */
323 REG_ERANGEX
/* Range striding over charsets. */
326 /* This data structure represents a compiled pattern. Before calling
327 the pattern compiler, the fields `buffer', `allocated', `fastmap',
328 `translate', and `no_sub' can be set. After the pattern has been
329 compiled, the `re_nsub' field is available. All other fields are
330 private to the regex routines. */
332 #ifndef RE_TRANSLATE_TYPE
333 # define RE_TRANSLATE_TYPE char *
336 struct re_pattern_buffer
338 /* [[[begin pattern_buffer]]] */
339 /* Space that holds the compiled pattern. It is declared as
340 `unsigned char *' because its elements are
341 sometimes used as array indexes. */
342 unsigned char *buffer
;
344 /* Number of bytes to which `buffer' points. */
347 /* Number of bytes actually used in `buffer'. */
350 /* Syntax setting with which the pattern was compiled. */
353 /* Pointer to a fastmap, if any, otherwise zero. re_search uses
354 the fastmap, if there is one, to skip over impossible
355 starting points for matches. */
358 /* Either a translate table to apply to all characters before
359 comparing them, or zero for no translation. The translation
360 is applied to a pattern when it is compiled and to a string
361 when it is matched. */
362 RE_TRANSLATE_TYPE translate
;
364 /* Number of subexpressions found by the compiler. */
367 /* Zero if this pattern cannot match the empty string, one else.
368 Well, in truth it's used only in `re_search_2', to see
369 whether or not we should use the fastmap, so we don't set
370 this absolutely perfectly; see `re_compile_fastmap'. */
371 unsigned can_be_null
: 1;
373 /* If REGS_UNALLOCATED, allocate space in the `regs' structure
374 for `max (RE_NREGS, re_nsub + 1)' groups.
375 If REGS_REALLOCATE, reallocate space if necessary.
376 If REGS_FIXED, use what's there. */
377 #define REGS_UNALLOCATED 0
378 #define REGS_REALLOCATE 1
380 unsigned regs_allocated
: 2;
382 /* Set to zero when `regex_compile' compiles a pattern; set to one
383 by `re_compile_fastmap' if it updates the fastmap. */
384 unsigned fastmap_accurate
: 1;
386 /* If set, `re_match_2' does not return information about
390 /* If set, a beginning-of-line anchor doesn't match at the
391 beginning of the string. */
392 unsigned not_bol
: 1;
394 /* Similarly for an end-of-line anchor. */
395 unsigned not_eol
: 1;
397 /* If true, the compilation of the pattern had to look up the syntax table,
398 so the compiled pattern is only valid for the current syntax table. */
399 unsigned used_syntax
: 1;
402 /* If true, multi-byte form in the regexp pattern should be
403 recognized as a multibyte character. */
404 unsigned multibyte
: 1;
406 /* If true, multi-byte form in the target of match should be
407 recognized as a multibyte character. */
408 unsigned target_multibyte
: 1;
410 /* Charset of unibyte characters at compiling time. */
414 /* [[[end pattern_buffer]]] */
417 typedef struct re_pattern_buffer regex_t
;
419 /* Type for byte offsets within the string. POSIX mandates this to be an int,
420 but the Open Group has signaled its intention to change the requirement to
421 be that regoff_t be at least as wide as ptrdiff_t and ssize_t. Current
422 gnulib sources also use ssize_t, and we need this for supporting buffers and
423 strings > 2GB on 64-bit hosts. */
424 typedef ssize_t regoff_t
;
427 /* This is the structure we store register match data in. See
428 regex.texinfo for a full description of what registers match. */
437 /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
438 `re_match_2' returns information about at least this many registers
439 the first time a `regs' structure is passed. */
445 /* POSIX specification for registers. Aside from the different names than
446 `re_registers', POSIX uses an array of structures, instead of a
447 structure of arrays. */
450 regoff_t rm_so
; /* Byte offset from string's start to substring's start. */
451 regoff_t rm_eo
; /* Byte offset from string's start to substring's end. */
454 /* Declarations for routines. */
456 /* Sets the current default syntax to SYNTAX, and return the old syntax.
457 You can also simply assign to the `re_syntax_options' variable. */
458 extern reg_syntax_t
re_set_syntax (reg_syntax_t __syntax
);
460 /* Compile the regular expression PATTERN, with length LENGTH
461 and syntax given by the global `re_syntax_options', into the buffer
462 BUFFER. Return NULL if successful, and an error string if not. */
463 extern const char *re_compile_pattern (const char *__pattern
, size_t __length
,
464 struct re_pattern_buffer
*__buffer
);
467 /* Compile a fastmap for the compiled pattern in BUFFER; used to
468 accelerate searches. Return 0 if successful and -2 if was an
470 extern int re_compile_fastmap (struct re_pattern_buffer
*__buffer
);
473 /* Search in the string STRING (with length LENGTH) for the pattern
474 compiled into BUFFER. Start searching at position START, for RANGE
475 characters. Return the starting position of the match, -1 for no
476 match, or -2 for an internal error. Also return register
477 information in REGS (if REGS and BUFFER->no_sub are nonzero). */
478 extern regoff_t
re_search (struct re_pattern_buffer
*__buffer
,
479 const char *__string
, size_t __length
,
480 ssize_t __start
, ssize_t __range
,
481 struct re_registers
*__regs
);
484 /* Like `re_search', but search in the concatenation of STRING1 and
485 STRING2. Also, stop searching at index START + STOP. */
486 extern regoff_t
re_search_2 (struct re_pattern_buffer
*__buffer
,
487 const char *__string1
, size_t __length1
,
488 const char *__string2
, size_t __length2
,
489 ssize_t __start
, ssize_t __range
,
490 struct re_registers
*__regs
,
494 /* Like `re_search', but return how many characters in STRING the regexp
495 in BUFFER matched, starting at position START. */
496 extern regoff_t
re_match (struct re_pattern_buffer
*__buffer
,
497 const char *__string
, size_t __length
,
498 ssize_t __start
, struct re_registers
*__regs
);
501 /* Relates to `re_match' as `re_search_2' relates to `re_search'. */
502 extern regoff_t
re_match_2 (struct re_pattern_buffer
*__buffer
,
503 const char *__string1
, size_t __length1
,
504 const char *__string2
, size_t __length2
,
505 ssize_t __start
, struct re_registers
*__regs
,
509 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and
510 ENDS. Subsequent matches using BUFFER and REGS will use this memory
511 for recording register information. STARTS and ENDS must be
512 allocated with malloc, and must each be at least `NUM_REGS * sizeof
513 (regoff_t)' bytes long.
515 If NUM_REGS == 0, then subsequent matches should allocate their own
518 Unless this function is called, the first search or match using
519 PATTERN_BUFFER will allocate its own register data, without
520 freeing the old data. */
521 extern void re_set_registers (struct re_pattern_buffer
*__buffer
,
522 struct re_registers
*__regs
,
524 regoff_t
*__starts
, regoff_t
*__ends
);
526 #if defined _REGEX_RE_COMP || defined _LIBC
528 /* 4.2 bsd compatibility. */
529 extern char *re_comp (const char *);
530 extern int re_exec (const char *);
534 /* GCC 2.95 and later have "__restrict"; C99 compilers have
535 "restrict", and "configure" may have defined "restrict".
536 Other compilers use __restrict, __restrict__, and _Restrict, and
537 'configure' might #define 'restrict' to those words, so pick a
540 # if 199901L <= __STDC_VERSION__
541 # define _Restrict_ restrict
542 # elif 2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__)
543 # define _Restrict_ __restrict
548 /* gcc 3.1 and up support the [restrict] syntax. Don't trust
549 sys/cdefs.h's definition of __restrict_arr, though, as it
550 mishandles gcc -ansi -pedantic. */
551 #ifndef _Restrict_arr_
552 # if ((199901L <= __STDC_VERSION__ \
553 || ((3 < __GNUC__ || (3 == __GNUC__ && 1 <= __GNUC_MINOR__)) \
554 && !defined __STRICT_ANSI__)) \
555 && !defined __GNUG__)
556 # define _Restrict_arr_ _Restrict_
558 # define _Restrict_arr_
562 /* POSIX compatibility. */
563 extern reg_errcode_t
regcomp (regex_t
*_Restrict_ __preg
,
564 const char *_Restrict_ __pattern
,
567 extern reg_errcode_t
regexec (const regex_t
*_Restrict_ __preg
,
568 const char *_Restrict_ __string
, size_t __nmatch
,
569 regmatch_t __pmatch
[_Restrict_arr_
],
572 extern size_t regerror (int __errcode
, const regex_t
* __preg
,
573 char *__errbuf
, size_t __errbuf_size
);
575 extern void regfree (regex_t
*__preg
);
582 /* For platform which support the ISO C amendment 1 functionality we
583 support user defined character classes. */
584 #if WIDE_CHAR_SUPPORT
585 /* Solaris 2.5 has a bug: <wchar.h> must be included before <wctype.h>. */
590 #if WIDE_CHAR_SUPPORT
591 /* The GNU C library provides support for user-defined character classes
592 and the functions from ISO C amendment 1. */
593 # ifdef CHARCLASS_NAME_MAX
594 # define CHAR_CLASS_MAX_LENGTH CHARCLASS_NAME_MAX
596 /* This shouldn't happen but some implementation might still have this
597 problem. Use a reasonable default value. */
598 # define CHAR_CLASS_MAX_LENGTH 256
600 typedef wctype_t re_wctype_t
;
601 typedef wchar_t re_wchar_t
;
602 # define re_wctype wctype
603 # define re_iswctype iswctype
604 # define re_wctype_to_bit(cc) 0
606 # define CHAR_CLASS_MAX_LENGTH 9 /* Namely, `multibyte'. */
611 /* Character classes. */
612 typedef enum { RECC_ERROR
= 0,
613 RECC_ALNUM
, RECC_ALPHA
, RECC_WORD
,
614 RECC_GRAPH
, RECC_PRINT
,
615 RECC_LOWER
, RECC_UPPER
,
616 RECC_PUNCT
, RECC_CNTRL
,
617 RECC_DIGIT
, RECC_XDIGIT
,
618 RECC_BLANK
, RECC_SPACE
,
619 RECC_MULTIBYTE
, RECC_NONASCII
,
620 RECC_ASCII
, RECC_UNIBYTE
623 extern char re_iswctype (int ch
, re_wctype_t cc
);
624 extern re_wctype_t
re_wctype (const unsigned char* str
);
626 typedef int re_wchar_t
;
628 extern void re_set_whitespace_regexp (const char *regexp
);
630 #endif /* not WIDE_CHAR_SUPPORT */