1 /* Definitions for data structures and routines for the regular
2 expression library, version 0.12.
4 Copyright (C) 1985, 1989, 1990-1995 Free Software Foundation, Inc.
6 This program is free software; you can redistribute it and/or modify
7 it under the terms of the GNU General Public License as published by
8 the Free Software Foundation; either version 2, or (at your option)
11 This program is distributed in the hope that it will be useful,
12 but WITHOUT ANY WARRANTY; without even the implied warranty of
13 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 GNU General Public License for more details.
16 You should have received a copy of the GNU General Public License
17 along with this program; if not, write to the Free Software
18 Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */
20 #ifndef __REGEXP_LIBRARY_H__
21 #define __REGEXP_LIBRARY_H__
23 /* POSIX says that <sys/types.h> must be included (by the caller) before
27 /* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it
33 /* The following two types have to be signed and unsigned integer type
34 wide enough to hold a value of a pointer. For most ANSI compilers
35 ptrdiff_t and size_t should be likely OK. Still size of these two
36 types is 2 for Microsoft C. Ugh... */
38 typedef unsigned long active_reg_t;
40 /* The following bits are used to determine the regexp syntax we
41 recognize. The set/not-set meanings are chosen so that Emacs syntax
42 remains the value 0. The bits are given in alphabetical order, and
43 the definitions shifted by one from the previous bit; thus, when we
44 add or remove a bit, only one other definition need change. */
45 typedef unsigned long reg_syntax_t;
47 /* If this bit is not set, then \ inside a bracket expression is literal.
48 If set, then such a \ quotes the following character. */
49 #define RE_BACKSLASH_ESCAPE_IN_LISTS (1L)
51 /* If this bit is not set, then + and ? are operators, and \+ and \? are
53 If set, then \+ and \? are operators and + and ? are literals. */
54 #define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
56 /* If this bit is set, then character classes are supported. They are:
57 [:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:],
58 [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
59 If not set, then character classes are not supported. */
60 #define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
62 /* If this bit is set, then ^ and $ are always anchors (outside bracket
63 expressions, of course).
64 If this bit is not set, then it depends:
65 ^ is an anchor if it is at the beginning of a regular
66 expression or after an open-group or an alternation operator;
67 $ is an anchor if it is at the end of a regular expression, or
68 before a close-group or an alternation operator.
70 This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
71 POSIX draft 11.2 says that * etc. in leading positions is undefined.
72 We already implemented a previous draft which made those constructs
73 invalid, though, so we haven't changed the code back. */
74 #define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
76 /* If this bit is set, then special characters are always special
77 regardless of where they are in the pattern.
78 If this bit is not set, then special characters are special only in
79 some contexts; otherwise they are ordinary. Specifically,
80 * + ? and intervals are only special when not after the beginning,
81 open-group, or alternation operator. */
82 #define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
84 /* If this bit is set, then *, +, ?, and { cannot be first in an re or
85 immediately after an alternation or begin-group operator. */
86 #define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
88 /* If this bit is set, then . matches newline.
89 If not set, then it doesn't. */
90 #define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
92 /* If this bit is set, then . doesn't match NUL.
93 If not set, then it does. */
94 #define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
96 /* If this bit is set, nonmatching lists [^...] do not match newline.
97 If not set, they do. */
98 #define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
100 /* If this bit is set, either \{...\} or {...} defines an
101 interval, depending on RE_NO_BK_BRACES.
102 If not set, \{, \}, {, and } are literals. */
103 #define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
105 /* If this bit is set, +, ? and | aren't recognized as operators.
106 If not set, they are. */
107 #define RE_LIMITED_OPS (RE_INTERVALS << 1)
109 /* If this bit is set, newline is an alternation operator.
110 If not set, newline is literal. */
111 #define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
113 /* If this bit is set, then `{...}' defines an interval, and \{ and \}
115 If not set, then `\{...\}' defines an interval. */
116 #define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
118 /* If this bit is set, (...) defines a group, and \( and \) are literals.
119 If not set, \(...\) defines a group, and ( and ) are literals. */
120 #define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
122 /* If this bit is set, then \<digit> matches <digit>.
123 If not set, then \<digit> is a back-reference. */
124 #define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
126 /* If this bit is set, then | is an alternation operator, and \| is literal.
127 If not set, then \| is an alternation operator, and | is literal. */
128 #define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
130 /* If this bit is set, then an ending range point collating higher
131 than the starting range point, as in [z-a], is invalid.
132 If not set, then when ending range point collates higher than the
133 starting range point, the range is ignored. */
134 #define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
136 /* If this bit is set, then an unmatched ) is ordinary.
137 If not set, then an unmatched ) is invalid. */
138 #define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
140 /* If this bit is set, do not process the GNU regex operators.
141 IF not set, then the GNU regex operators are recognized. */
142 #define RE_NO_GNU_OPS (RE_UNMATCHED_RIGHT_PAREN_ORD << 1)
144 /* This global variable defines the particular regexp syntax to use (for
145 some interfaces). When a regexp is compiled, the syntax used is
146 stored in the pattern buffer, so changing this does not affect
147 already-compiled regexps. */
148 extern reg_syntax_t re_syntax_options;
150 /* Define combinations of the above bits for the standard possibilities.
151 (The [[[ comments delimit what gets put into the Texinfo file, so
152 don't delete them!) */
153 /* [[[begin syntaxes]]] */
154 #define RE_SYNTAX_EMACS 0
156 #define RE_SYNTAX_AWK \
157 (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \
158 | RE_NO_BK_PARENS | RE_NO_BK_REFS \
159 | RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \
161 | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS)
163 #define RE_SYNTAX_GNU_AWK \
164 (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS)
166 #define RE_SYNTAX_POSIX_AWK \
167 (RE_SYNTAX_GNU_AWK | RE_NO_GNU_OPS)
169 #define RE_SYNTAX_GREP \
170 (RE_BK_PLUS_QM | RE_CHAR_CLASSES \
171 | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS \
174 #define RE_SYNTAX_EGREP \
175 (RE_CHAR_CLASSES | RE_CONTEXT_INDEP_ANCHORS \
176 | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE \
177 | RE_NEWLINE_ALT | RE_NO_BK_PARENS \
180 #define RE_SYNTAX_POSIX_EGREP \
181 (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES)
183 /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff. */
184 #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
186 #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
188 /* Syntax bits common to both basic and extended POSIX regex syntax. */
189 #define _RE_SYNTAX_POSIX_COMMON \
190 (RE_CHAR_CLASSES | RE_DOT_NEWLINE | RE_DOT_NOT_NULL \
191 | RE_INTERVALS | RE_NO_EMPTY_RANGES)
193 #define RE_SYNTAX_POSIX_BASIC \
194 (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM)
196 /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
197 RE_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this
198 isn't minimal, since other operators, such as \`, aren't disabled. */
199 #define RE_SYNTAX_POSIX_MINIMAL_BASIC \
200 (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
202 #define RE_SYNTAX_POSIX_EXTENDED \
203 (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
204 | RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \
205 | RE_NO_BK_PARENS | RE_NO_BK_VBAR \
206 | RE_UNMATCHED_RIGHT_PAREN_ORD)
208 /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INVALID_OPS
209 replaces RE_CONTEXT_INDEP_OPS and RE_NO_BK_REFS is added. */
210 #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED \
211 (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
212 | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES \
213 | RE_NO_BK_PARENS | RE_NO_BK_REFS \
214 | RE_NO_BK_VBAR | RE_UNMATCHED_RIGHT_PAREN_ORD)
216 #define RE_PERL_MULTILINE_SYNTAX \
217 (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_CONTEXT_INDEP_ANCHORS \
218 | RE_CONTEXT_INDEP_OPS \
219 | RE_INTERVALS | RE_NO_BK_BRACES \
220 | RE_NO_BK_PARENS | RE_NO_BK_VBAR)
222 #define RE_PERL_SINGLELINE_SYNTAX \
223 (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_CONTEXT_INDEP_ANCHORS \
224 | RE_CONTEXT_INDEP_OPS | RE_DOT_NEWLINE \
225 | RE_INTERVALS | RE_NO_BK_BRACES \
226 | RE_NO_BK_PARENS | RE_NO_BK_VBAR)
227 /* [[[end syntaxes]]] */
229 /* Maximum number of duplicates an interval can allow. Some systems
230 (erroneously) define this in other header files, but we want our
231 value, so remove any previous define. */
235 /* if sizeof(int) == 2, then ((1 << 15) - 1) overflows */
236 #define RE_DUP_MAX (0x7fff)
239 /* POSIX `cflags' bits (i.e., information for `regcomp'). */
241 /* If this bit is set, then use extended regular expression syntax.
242 If not set, then use basic regular expression syntax. */
243 #define REG_EXTENDED 1
245 /* If this bit is set, then ignore case when matching.
246 If not set, then case is significant. */
247 #define REG_ICASE (REG_EXTENDED << 1)
249 /* If this bit is set, then anchors do not match at newline
250 characters in the string.
251 If not set, then anchors do match at newlines. */
252 #define REG_NEWLINE (REG_ICASE << 1)
254 /* If this bit is set, then report only success or fail in regexec.
255 If not set, then returns differ between not matching and errors. */
256 #define REG_NOSUB (REG_NEWLINE << 1)
259 /* POSIX `eflags' bits (i.e., information for regexec). */
261 /* If this bit is set, then the beginning-of-line operator doesn't match
262 the beginning of the string (presumably because it's not the
263 beginning of a line).
264 If not set, then the beginning-of-line operator does match the
265 beginning of the string. */
268 /* Like REG_NOTBOL, except for the end-of-line. */
269 #define REG_NOTEOL (1 << 1)
272 /* If any error codes are removed, changed, or added, update the
273 `re_error_msg' table in regex.c. */
276 REG_NOERROR = 0, /* Success. */
277 REG_NOMATCH, /* Didn't find a match (for regexec). */
279 /* POSIX regcomp return error codes. (In the order listed in the
281 REG_BADPAT, /* Invalid pattern. */
282 REG_ECOLLATE, /* Not implemented. */
283 REG_ECTYPE, /* Invalid character class name. */
284 REG_EESCAPE, /* Trailing backslash. */
285 REG_ESUBREG, /* Invalid back reference. */
286 REG_EBRACK, /* Unmatched left bracket. */
287 REG_EPAREN, /* Parenthesis imbalance. */
288 REG_EBRACE, /* Unmatched \{. */
289 REG_BADBR, /* Invalid contents of \{\}. */
290 REG_ERANGE, /* Invalid range end. */
291 REG_ESPACE, /* Ran out of memory. */
292 REG_BADRPT, /* No preceding re for repetition op. */
294 /* Error codes we've added. */
295 REG_EEND, /* Premature end. */
296 REG_ESIZE, /* Compiled pattern bigger than 2^16 bytes. */
297 REG_ERPAREN /* Unmatched ) or \); not returned from regcomp. */
300 /* This data structure represents a compiled pattern. Before calling
301 the pattern compiler, the fields `buffer', `allocated', `fastmap',
302 `translate', and `no_sub' can be set. After the pattern has been
303 compiled, the `re_nsub' field is available. All other fields are
304 private to the regex routines. */
306 struct re_pattern_buffer
308 /* [[[begin pattern_buffer]]] */
309 /* Space that holds the compiled pattern. It is declared as
310 `unsigned char *' because its elements are
311 sometimes used as array indexes. */
312 unsigned char *buffer;
314 /* Number of bytes to which `buffer' points. */
315 unsigned long allocated;
317 /* Number of bytes actually used in `buffer'. */
320 /* Syntax setting with which the pattern was compiled. */
323 /* Pointer to a fastmap, if any, otherwise zero. re_search uses
324 the fastmap, if there is one, to skip over impossible
325 starting points for matches. */
328 /* Either a translate table to apply to all characters before
329 comparing them, or zero for no translation. The translation
330 is applied to a pattern when it is compiled and to a string
331 when it is matched. */
334 /* Number of subexpressions found by the compiler. */
337 /* Zero if this pattern cannot match the empty string, one else.
338 Well, in truth it's used only in `re_search_2', to see
339 whether or not we should use the fastmap, so we don't set
340 this absolutely perfectly; see `re_compile_fastmap' (the
341 `duplicate' case). */
342 unsigned can_be_null : 1;
344 /* If REGS_UNALLOCATED, allocate space in the `regs' structure
345 for `max (RE_NREGS, re_nsub + 1)' groups.
346 If REGS_REALLOCATE, reallocate space if necessary.
347 If REGS_FIXED, use what's there. */
348 #define REGS_UNALLOCATED 0
349 #define REGS_REALLOCATE 1
351 unsigned regs_allocated : 2;
353 /* Set to zero when `regex_compile' compiles a pattern; set to one
354 by `re_compile_fastmap' if it updates the fastmap. */
355 unsigned fastmap_accurate : 1;
357 /* If set, `re_match_2' does not return information about
361 /* If set, a beginning-of-line anchor doesn't match at the
362 beginning of the string. */
363 unsigned not_bol : 1;
365 /* Similarly for an end-of-line anchor. */
366 unsigned not_eol : 1;
368 /* If true, an anchor at a newline matches. */
369 unsigned newline_anchor : 1;
371 /* [[[end pattern_buffer]]] */
374 typedef struct re_pattern_buffer regex_t;
377 /* search.c (search_buffer) in Emacs needs this one opcode value. It is
378 defined both in `regex.c' and here. */
379 #define RE_EXACTN_VALUE 1
381 /* Type for byte offsets within the string. POSIX mandates this. */
382 typedef int regoff_t;
385 /* This is the structure we store register match data in. See
386 regex.texinfo for a full description of what registers match. */
395 /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
396 `re_match_2' returns information about at least this many registers
397 the first time a `regs' structure is passed. */
403 /* POSIX specification for registers. Aside from the different names than
404 `re_registers', POSIX uses an array of structures, instead of a
405 structure of arrays. */
408 regoff_t rm_so; /* Byte offset from string's start to substring's start. */
409 regoff_t rm_eo; /* Byte offset from string's start to substring's end. */
412 /* Declarations for routines. */
414 /* To avoid duplicating every routine declaration -- once with a
415 prototype (if we are ANSI), and once without (if we aren't) -- we
416 use the following macro to declare argument types. This
417 unfortunately clutters up the declarations a bit, but I think it's
422 #define _RE_ARGS(args) args
424 #else /* not __STDC__ */
426 #define _RE_ARGS(args) ()
428 #endif /* not __STDC__ */
430 /* Sets the current default syntax to SYNTAX, and return the old syntax.
431 You can also simply assign to the `re_syntax_options' variable. */
432 extern reg_syntax_t re_set_syntax _RE_ARGS ((reg_syntax_t syntax));
434 /* Compile the regular expression PATTERN, with length LENGTH
435 and syntax given by the global `re_syntax_options', into the buffer
436 BUFFER. Return NULL if successful, and an error string if not. */
437 extern const char *re_compile_pattern
438 _RE_ARGS ((const char *pattern, size_t length,
439 struct re_pattern_buffer *buffer));
442 /* Compile a fastmap for the compiled pattern in BUFFER; used to
443 accelerate searches. Return 0 if successful and -2 if was an
445 extern int re_compile_fastmap _RE_ARGS ((struct re_pattern_buffer *buffer));
448 /* Search in the string STRING (with length LENGTH) for the pattern
449 compiled into BUFFER. Start searching at position START, for RANGE
450 characters. Return the starting position of the match, -1 for no
451 match, or -2 for an internal error. Also return register
452 information in REGS (if REGS and BUFFER->no_sub are nonzero). */
454 _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
455 int length, int start, int range, struct re_registers *regs));
458 /* Like `re_search', but search in the concatenation of STRING1 and
459 STRING2. Also, stop searching at index START + STOP. */
460 extern int re_search_2
461 _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
462 int length1, const char *string2, int length2,
463 int start, int range, struct re_registers *regs, int stop));
466 /* Like `re_search', but return how many characters in STRING the regexp
467 in BUFFER matched, starting at position START. */
469 _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
470 int length, int start, struct re_registers *regs));
473 /* Relates to `re_match' as `re_search_2' relates to `re_search'. */
474 extern int re_match_2
475 _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
476 int length1, const char *string2, int length2,
477 int start, struct re_registers *regs, int stop));
480 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and
481 ENDS. Subsequent matches using BUFFER and REGS will use this memory
482 for recording register information. STARTS and ENDS must be
483 allocated with malloc, and must each be at least `NUM_REGS * sizeof
484 (regoff_t)' bytes long.
486 If NUM_REGS == 0, then subsequent matches should allocate their own
489 Unless this function is called, the first search or match using
490 PATTERN_BUFFER will allocate its own register data, without
491 freeing the old data. */
492 extern void re_set_registers
493 _RE_ARGS ((struct re_pattern_buffer *buffer, struct re_registers *regs,
494 unsigned num_regs, regoff_t *starts, regoff_t *ends));
497 /* 4.2 bsd compatibility. */
498 extern char *re_comp _RE_ARGS ((const char *));
499 extern int re_exec _RE_ARGS ((const char *));
502 /* POSIX compatibility. */
503 extern int regcomp _RE_ARGS ((regex_t *preg, const char *pattern, int cflags));
505 _RE_ARGS ((const regex_t *preg, const char *string, size_t nmatch,
506 regmatch_t pmatch[], int eflags));
507 extern size_t regerror
508 _RE_ARGS ((int errcode, const regex_t *preg, char *errbuf,
509 size_t errbuf_size));
510 extern void regfree _RE_ARGS ((regex_t *preg));
512 #endif /* not __REGEXP_LIBRARY_H__ */
518 trim-versions-without-asking: nil