1 /* Definitions for data structures and routines for the regular
3 Copyright (C) 1985,1989-93,1995-98,2000,2001,2002,2003,2005,2006,2008
4 Free Software Foundation, Inc.
5 This file is part of the GNU C Library.
7 The GNU C Library is free software; you can redistribute it and/or
8 modify it under the terms of the GNU Lesser General Public
9 License as published by the Free Software Foundation; either
10 version 2.1 of the License, or (at your option) any later version.
12 The GNU C Library 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 GNU
15 Lesser General Public License for more details.
17 You should have received a copy of the GNU Lesser General Public
18 License along with the GNU C Library; if not, write to the Free
19 Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA
25 #include <sys/types.h>
29 /* The following two types have to be signed and unsigned integer type
30 wide enough to hold a value of a pointer. For most ANSI compilers
31 ptrdiff_t and size_t should be likely OK. Still size of these two
32 types is 2 for Microsoft C. Ugh... */
33 typedef long int s_reg_t;
34 typedef unsigned long int active_reg_t;
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 long int reg_syntax_t;
44 /* If this bit is not set, then \ inside a bracket expression is literal.
45 If set, then such a \ quotes the following character. */
46 # define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1)
48 /* If this bit is not set, then + and ? are operators, and \+ and \? are
50 If set, then \+ and \? are operators and + and ? are literals. */
51 # define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
53 /* If this bit is set, then character classes are supported. They are:
54 [:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:],
55 [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
56 If not set, then character classes are not supported. */
57 # define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
59 /* If this bit is set, then ^ and $ are always anchors (outside bracket
60 expressions, of course).
61 If this bit is not set, then it depends:
62 ^ is an anchor if it is at the beginning of a regular
63 expression or after an open-group or an alternation operator;
64 $ is an anchor if it is at the end of a regular expression, or
65 before a close-group or an alternation operator.
67 This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
68 POSIX draft 11.2 says that * etc. in leading positions is undefined.
69 We already implemented a previous draft which made those constructs
70 invalid, though, so we haven't changed the code back. */
71 # define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
73 /* If this bit is set, then special characters are always special
74 regardless of where they are in the pattern.
75 If this bit is not set, then special characters are special only in
76 some contexts; otherwise they are ordinary. Specifically,
77 * + ? and intervals are only special when not after the beginning,
78 open-group, or alternation operator. */
79 # define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
81 /* If this bit is set, then *, +, ?, and { cannot be first in an re or
82 immediately after an alternation or begin-group operator. */
83 # define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
85 /* If this bit is set, then . matches newline.
86 If not set, then it doesn't. */
87 # define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
89 /* If this bit is set, then . doesn't match NUL.
90 If not set, then it does. */
91 # define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
93 /* If this bit is set, nonmatching lists [^...] do not match newline.
94 If not set, they do. */
95 # define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
97 /* If this bit is set, either \{...\} or {...} defines an
98 interval, depending on RE_NO_BK_BRACES.
99 If not set, \{, \}, {, and } are literals. */
100 # define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
102 /* If this bit is set, +, ? and | aren't recognized as operators.
103 If not set, they are. */
104 # define RE_LIMITED_OPS (RE_INTERVALS << 1)
106 /* If this bit is set, newline is an alternation operator.
107 If not set, newline is literal. */
108 # define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
110 /* If this bit is set, then `{...}' defines an interval, and \{ and \}
112 If not set, then `\{...\}' defines an interval. */
113 # define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
115 /* If this bit is set, (...) defines a group, and \( and \) are literals.
116 If not set, \(...\) defines a group, and ( and ) are literals. */
117 # define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
119 /* If this bit is set, then \<digit> matches <digit>.
120 If not set, then \<digit> is a back-reference. */
121 # define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
123 /* If this bit is set, then | is an alternation operator, and \| is literal.
124 If not set, then \| is an alternation operator, and | is literal. */
125 # define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
127 /* If this bit is set, then an ending range point collating higher
128 than the starting range point, as in [z-a], is invalid.
129 If not set, then when ending range point collates higher than the
130 starting range point, the range is ignored. */
131 # define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
133 /* If this bit is set, then an unmatched ) is ordinary.
134 If not set, then an unmatched ) is invalid. */
135 # define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
137 /* If this bit is set, succeed as soon as we match the whole pattern,
138 without further backtracking. */
139 # define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1)
141 /* If this bit is set, do not process the GNU regex operators.
142 If not set, then the GNU regex operators are recognized. */
143 # define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1)
145 /* If this bit is set, turn on internal regex debugging.
146 If not set, and debugging was on, turn it off.
147 This only works if regex.c is compiled -DDEBUG.
148 We define this bit always, so that all that's needed to turn on
149 debugging is to recompile regex.c; the calling code can always have
150 this bit set, and it won't affect anything in the normal case. */
151 # define RE_DEBUG (RE_NO_GNU_OPS << 1)
153 /* If this bit is set, a syntactically invalid interval is treated as
154 a string of ordinary characters. For example, the ERE 'a{1' is
155 treated as 'a\{1'. */
156 # define RE_INVALID_INTERVAL_ORD (RE_DEBUG << 1)
158 /* If this bit is set, then ignore case when matching.
159 If not set, then case is significant. */
160 # define RE_ICASE (RE_INVALID_INTERVAL_ORD << 1)
162 /* This bit is used internally like RE_CONTEXT_INDEP_ANCHORS but only
163 for ^, because it is difficult to scan the regex backwards to find
164 whether ^ should be special. */
165 # define RE_CARET_ANCHORS_HERE (RE_ICASE << 1)
167 /* If this bit is set, then \{ cannot be first in an bre or
168 immediately after an alternation or begin-group operator. */
169 # define RE_CONTEXT_INVALID_DUP (RE_CARET_ANCHORS_HERE << 1)
171 /* If this bit is set, then no_sub will be set to 1 during
172 re_compile_pattern. */
173 # define RE_NO_SUB (RE_CONTEXT_INVALID_DUP << 1)
176 /* This global variable defines the particular regexp syntax to use (for
177 some interfaces). When a regexp is compiled, the syntax used is
178 stored in the pattern buffer, so changing this does not affect
179 already-compiled regexps. */
180 extern reg_syntax_t re_syntax_options;
183 /* Define combinations of the above bits for the standard possibilities.
184 (The [[[ comments delimit what gets put into the Texinfo file, so
185 don't delete them!) */
186 /* [[[begin syntaxes]]] */
187 #define RE_SYNTAX_EMACS 0
189 #define RE_SYNTAX_AWK \
190 (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \
191 | RE_NO_BK_PARENS | RE_NO_BK_REFS \
192 | RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \
193 | RE_DOT_NEWLINE | RE_CONTEXT_INDEP_ANCHORS \
194 | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS)
196 #define RE_SYNTAX_GNU_AWK \
197 ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DEBUG) \
198 & ~(RE_DOT_NOT_NULL | RE_INTERVALS | RE_CONTEXT_INDEP_OPS \
199 | RE_CONTEXT_INVALID_OPS ))
201 #define RE_SYNTAX_POSIX_AWK \
202 (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS \
203 | RE_INTERVALS | RE_NO_GNU_OPS)
205 #define RE_SYNTAX_GREP \
206 (RE_BK_PLUS_QM | RE_CHAR_CLASSES \
207 | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS \
210 #define RE_SYNTAX_EGREP \
211 (RE_CHAR_CLASSES | RE_CONTEXT_INDEP_ANCHORS \
212 | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE \
213 | RE_NEWLINE_ALT | RE_NO_BK_PARENS \
216 #define RE_SYNTAX_POSIX_EGREP \
217 (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES \
218 | RE_INVALID_INTERVAL_ORD)
220 /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff. */
221 #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
223 #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
225 /* Syntax bits common to both basic and extended POSIX regex syntax. */
226 #define _RE_SYNTAX_POSIX_COMMON \
227 (RE_CHAR_CLASSES | RE_DOT_NEWLINE | RE_DOT_NOT_NULL \
228 | RE_INTERVALS | RE_NO_EMPTY_RANGES)
230 #define RE_SYNTAX_POSIX_BASIC \
231 (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM | RE_CONTEXT_INVALID_DUP)
233 /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
234 RE_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this
235 isn't minimal, since other operators, such as \`, aren't disabled. */
236 #define RE_SYNTAX_POSIX_MINIMAL_BASIC \
237 (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
239 #define RE_SYNTAX_POSIX_EXTENDED \
240 (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
241 | RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \
242 | RE_NO_BK_PARENS | RE_NO_BK_VBAR \
243 | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD)
245 /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is
246 removed and RE_NO_BK_REFS is added. */
247 #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED \
248 (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
249 | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES \
250 | RE_NO_BK_PARENS | RE_NO_BK_REFS \
251 | RE_NO_BK_VBAR | RE_UNMATCHED_RIGHT_PAREN_ORD)
252 /* [[[end syntaxes]]] */
254 /* Maximum number of duplicates an interval can allow. Some systems
255 (erroneously) define this in other header files, but we want our
256 value, so remove any previous define. */
260 /* If sizeof(int) == 2, then ((1 << 15) - 1) overflows. */
261 # define RE_DUP_MAX (0x7fff)
265 /* POSIX `cflags' bits (i.e., information for `regcomp'). */
267 /* If this bit is set, then use extended regular expression syntax.
268 If not set, then use basic regular expression syntax. */
269 #define REG_EXTENDED 1
271 /* If this bit is set, then ignore case when matching.
272 If not set, then case is significant. */
273 #define REG_ICASE (REG_EXTENDED << 1)
275 /* If this bit is set, then anchors do not match at newline
276 characters in the string.
277 If not set, then anchors do match at newlines. */
278 #define REG_NEWLINE (REG_ICASE << 1)
280 /* If this bit is set, then report only success or fail in regexec.
281 If not set, then returns differ between not matching and errors. */
282 #define REG_NOSUB (REG_NEWLINE << 1)
285 /* POSIX `eflags' bits (i.e., information for regexec). */
287 /* If this bit is set, then the beginning-of-line operator doesn't match
288 the beginning of the string (presumably because it's not the
289 beginning of a line).
290 If not set, then the beginning-of-line operator does match the
291 beginning of the string. */
294 /* Like REG_NOTBOL, except for the end-of-line. */
295 #define REG_NOTEOL (1 << 1)
297 /* Use PMATCH[0] to delimit the start and end of the search in the
299 #define REG_STARTEND (1 << 2)
302 /* If any error codes are removed, changed, or added, update the
303 `re_error_msg' table in regex.c. */
306 #if defined _XOPEN_SOURCE || defined __USE_XOPEN2K
307 REG_ENOSYS = -1, /* This will never happen for this implementation. */
310 REG_NOERROR = 0, /* Success. */
311 REG_NOMATCH, /* Didn't find a match (for regexec). */
313 /* POSIX regcomp return error codes. (In the order listed in the
315 REG_BADPAT, /* Invalid pattern. */
316 REG_ECOLLATE, /* Invalid collating element. */
317 REG_ECTYPE, /* Invalid character class name. */
318 REG_EESCAPE, /* Trailing backslash. */
319 REG_ESUBREG, /* Invalid back reference. */
320 REG_EBRACK, /* Unmatched left bracket. */
321 REG_EPAREN, /* Parenthesis imbalance. */
322 REG_EBRACE, /* Unmatched \{. */
323 REG_BADBR, /* Invalid contents of \{\}. */
324 REG_ERANGE, /* Invalid range end. */
325 REG_ESPACE, /* Ran out of memory. */
326 REG_BADRPT, /* No preceding re for repetition op. */
328 /* Error codes we've added. */
329 REG_EEND, /* Premature end. */
330 REG_ESIZE, /* Compiled pattern bigger than 2^16 bytes. */
331 REG_ERPAREN /* Unmatched ) or \); not returned from regcomp. */
334 /* This data structure represents a compiled pattern. Before calling
335 the pattern compiler, the fields `buffer', `allocated', `fastmap',
336 `translate', and `no_sub' can be set. After the pattern has been
337 compiled, the `re_nsub' field is available. All other fields are
338 private to the regex routines. */
340 #ifndef RE_TRANSLATE_TYPE
341 # define __RE_TRANSLATE_TYPE unsigned char *
343 # define RE_TRANSLATE_TYPE __RE_TRANSLATE_TYPE
348 # define __REPB_PREFIX(name) name
350 # define __REPB_PREFIX(name) __##name
353 struct re_pattern_buffer
355 /* Space that holds the compiled pattern. It is declared as
356 `unsigned char *' because its elements are sometimes used as
358 unsigned char *__REPB_PREFIX(buffer);
360 /* Number of bytes to which `buffer' points. */
361 unsigned long int __REPB_PREFIX(allocated);
363 /* Number of bytes actually used in `buffer'. */
364 unsigned long int __REPB_PREFIX(used);
366 /* Syntax setting with which the pattern was compiled. */
367 reg_syntax_t __REPB_PREFIX(syntax);
369 /* Pointer to a fastmap, if any, otherwise zero. re_search uses the
370 fastmap, if there is one, to skip over impossible starting points
372 char *__REPB_PREFIX(fastmap);
374 /* Either a translate table to apply to all characters before
375 comparing them, or zero for no translation. The translation is
376 applied to a pattern when it is compiled and to a string when it
378 __RE_TRANSLATE_TYPE __REPB_PREFIX(translate);
380 /* Number of subexpressions found by the compiler. */
383 /* Zero if this pattern cannot match the empty string, one else.
384 Well, in truth it's used only in `re_search_2', to see whether or
385 not we should use the fastmap, so we don't set this absolutely
386 perfectly; see `re_compile_fastmap' (the `duplicate' case). */
387 unsigned __REPB_PREFIX(can_be_null) : 1;
389 /* If REGS_UNALLOCATED, allocate space in the `regs' structure
390 for `max (RE_NREGS, re_nsub + 1)' groups.
391 If REGS_REALLOCATE, reallocate space if necessary.
392 If REGS_FIXED, use what's there. */
394 # define REGS_UNALLOCATED 0
395 # define REGS_REALLOCATE 1
396 # define REGS_FIXED 2
398 unsigned __REPB_PREFIX(regs_allocated) : 2;
400 /* Set to zero when `regex_compile' compiles a pattern; set to one
401 by `re_compile_fastmap' if it updates the fastmap. */
402 unsigned __REPB_PREFIX(fastmap_accurate) : 1;
404 /* If set, `re_match_2' does not return information about
406 unsigned __REPB_PREFIX(no_sub) : 1;
408 /* If set, a beginning-of-line anchor doesn't match at the beginning
410 unsigned __REPB_PREFIX(not_bol) : 1;
412 /* Similarly for an end-of-line anchor. */
413 unsigned __REPB_PREFIX(not_eol) : 1;
415 /* If true, an anchor at a newline matches. */
416 unsigned __REPB_PREFIX(newline_anchor) : 1;
419 typedef struct re_pattern_buffer regex_t;
421 /* Type for byte offsets within the string. POSIX mandates this. */
422 typedef int regoff_t;
426 /* This is the structure we store register match data in. See
427 regex.texinfo for a full description of what registers match. */
436 /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
437 `re_match_2' returns information about at least this many registers
438 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. */
457 /* Sets the current default syntax to SYNTAX, and return the old syntax.
458 You can also simply assign to the `re_syntax_options' variable. */
459 extern reg_syntax_t re_set_syntax (reg_syntax_t __syntax);
461 /* Compile the regular expression PATTERN, with length LENGTH
462 and syntax given by the global `re_syntax_options', into the buffer
463 BUFFER. Return NULL if successful, and an error string if not. */
464 extern const char *re_compile_pattern (const char *__pattern, size_t __length,
465 struct re_pattern_buffer *__buffer);
468 /* Compile a fastmap for the compiled pattern in BUFFER; used to
469 accelerate searches. Return 0 if successful and -2 if was an
471 extern int re_compile_fastmap (struct re_pattern_buffer *__buffer);
472 libc_hidden_proto(re_compile_fastmap)
475 /* Search in the string STRING (with length LENGTH) for the pattern
476 compiled into BUFFER. Start searching at position START, for RANGE
477 characters. Return the starting position of the match, -1 for no
478 match, or -2 for an internal error. Also return register
479 information in REGS (if REGS and BUFFER->no_sub are nonzero). */
480 extern int re_search (struct re_pattern_buffer *__buffer, const char *__string,
481 int __length, int __start, int __range,
482 struct re_registers *__regs);
483 libc_hidden_proto(re_search)
486 /* Like `re_search', but search in the concatenation of STRING1 and
487 STRING2. Also, stop searching at index START + STOP. */
488 extern int re_search_2 (struct re_pattern_buffer *__buffer,
489 const char *__string1, int __length1,
490 const char *__string2, int __length2, int __start,
491 int __range, struct re_registers *__regs, int __stop);
492 libc_hidden_proto(re_search_2)
495 /* Like `re_search', but return how many characters in STRING the regexp
496 in BUFFER matched, starting at position START. */
497 extern int re_match (struct re_pattern_buffer *__buffer, const char *__string,
498 int __length, int __start, struct re_registers *__regs);
501 /* Relates to `re_match' as `re_search_2' relates to `re_search'. */
502 extern int re_match_2 (struct re_pattern_buffer *__buffer,
503 const char *__string1, int __length1,
504 const char *__string2, int __length2, int __start,
505 struct re_registers *__regs, int __stop);
508 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and
509 ENDS. Subsequent matches using BUFFER and REGS will use this memory
510 for recording register information. STARTS and ENDS must be
511 allocated with malloc, and must each be at least `NUM_REGS * sizeof
512 (regoff_t)' bytes long.
514 If NUM_REGS == 0, then subsequent matches should allocate their own
517 Unless this function is called, the first search or match using
518 PATTERN_BUFFER will allocate its own register data, without
519 freeing the old data. */
520 extern void re_set_registers (struct re_pattern_buffer *__buffer,
521 struct re_registers *__regs,
522 unsigned int __num_regs,
523 regoff_t *__starts, regoff_t *__ends);
526 #if defined _REGEX_RE_COMP || (defined _LIBC && defined __USE_BSD)
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". */
537 # if ! (2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__))
538 # if defined restrict || 199901L <= __STDC_VERSION__
539 # define __restrict restrict
545 /* gcc 3.1 and up support the [restrict] syntax. */
546 #ifndef __restrict_arr
547 # if (__GNUC__ > 3 || (__GNUC__ == 3 && __GNUC_MINOR__ >= 1)) \
549 # define __restrict_arr __restrict
551 # define __restrict_arr
555 /* POSIX compatibility. */
556 extern int regcomp (regex_t *__restrict __preg,
557 const char *__restrict __pattern,
560 extern int regexec (const regex_t *__restrict __preg,
561 const char *__restrict __string, size_t __nmatch,
562 regmatch_t __pmatch[__restrict_arr],
564 libc_hidden_proto(regexec)
566 extern size_t regerror (int __errcode, const regex_t *__restrict __preg,
567 char *__restrict __errbuf, size_t __errbuf_size);
569 extern void regfree (regex_t *__preg);
570 libc_hidden_proto(regfree)