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, 98 Free Software Foundation, Inc.
5
the C library, however. The master source lives in /gd/gnu/lib.
7
NOTE: The canonical source of this file is maintained with the
8
GNU C Library. Bugs can be reported to bug-glibc@prep.ai.mit.edu.
10
This program is free software; you can redistribute it and/or modify it
11
under the terms of the GNU General Public License as published by the
12
Free Software Foundation; either version 2, or (at your option) any
15
This program is distributed in the hope that it will be useful,
16
but WITHOUT ANY WARRANTY; without even the implied warranty of
17
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
18
GNU General Public License for more details.
20
You should have received a copy of the GNU General Public License
21
along with this program; if not, write to the Free Software Foundation,
22
Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */
27
/* Allow the use in C++ code. */
32
/* POSIX says that <sys/types.h> must be included (by the caller) before
35
#if !defined (_POSIX_C_SOURCE) && !defined (_POSIX_SOURCE) && defined (VMS)
36
/* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it
41
/* The following two types have to be signed and unsigned integer type
42
wide enough to hold a value of a pointer. For most ANSI compilers
43
ptrdiff_t and size_t should be likely OK. Still size of these two
44
types is 2 for Microsoft C. Ugh... */
45
typedef long int s_reg_t;
46
typedef unsigned long int active_reg_t;
48
/* The following bits are used to determine the regexp syntax we
49
recognize. The set/not-set meanings are chosen so that Emacs syntax
50
remains the value 0. The bits are given in alphabetical order, and
51
the definitions shifted by one from the previous bit; thus, when we
52
add or remove a bit, only one other definition need change. */
53
typedef unsigned long int reg_syntax_t;
55
/* If this bit is not set, then \ inside a bracket expression is literal.
56
If set, then such a \ quotes the following character. */
57
#define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1)
59
/* If this bit is not set, then + and ? are operators, and \+ and \? are
61
If set, then \+ and \? are operators and + and ? are literals. */
62
#define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
64
/* If this bit is set, then character classes are supported. They are:
65
[:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:],
66
[:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
67
If not set, then character classes are not supported. */
68
#define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
70
/* If this bit is set, then ^ and $ are always anchors (outside bracket
71
expressions, of course).
72
If this bit is not set, then it depends:
73
^ is an anchor if it is at the beginning of a regular
74
expression or after an open-group or an alternation operator;
75
$ is an anchor if it is at the end of a regular expression, or
76
before a close-group or an alternation operator.
78
This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
79
POSIX draft 11.2 says that * etc. in leading positions is undefined.
80
We already implemented a previous draft which made those constructs
81
invalid, though, so we haven't changed the code back. */
82
#define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
84
/* If this bit is set, then special characters are always special
85
regardless of where they are in the pattern.
86
If this bit is not set, then special characters are special only in
87
some contexts; otherwise they are ordinary. Specifically,
88
* + ? and intervals are only special when not after the beginning,
89
open-group, or alternation operator. */
90
#define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
92
/* If this bit is set, then *, +, ?, and { cannot be first in an re or
93
immediately after an alternation or begin-group operator. */
94
#define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
96
/* If this bit is set, then . matches newline.
97
If not set, then it doesn't. */
98
#define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
100
/* If this bit is set, then . doesn't match NUL.
101
If not set, then it does. */
102
#define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
104
/* If this bit is set, nonmatching lists [^...] do not match newline.
105
If not set, they do. */
106
#define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
108
/* If this bit is set, either \{...\} or {...} defines an
109
interval, depending on RE_NO_BK_BRACES.
110
If not set, \{, \}, {, and } are literals. */
111
#define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
113
/* If this bit is set, +, ? and | aren't recognized as operators.
114
If not set, they are. */
115
#define RE_LIMITED_OPS (RE_INTERVALS << 1)
117
/* If this bit is set, newline is an alternation operator.
118
If not set, newline is literal. */
119
#define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
121
/* If this bit is set, then `{...}' defines an interval, and \{ and \}
123
If not set, then `\{...\}' defines an interval. */
124
#define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
126
/* If this bit is set, (...) defines a group, and \( and \) are literals.
127
If not set, \(...\) defines a group, and ( and ) are literals. */
128
#define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
130
/* If this bit is set, then \<digit> matches <digit>.
131
If not set, then \<digit> is a back-reference. */
132
#define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
134
/* If this bit is set, then | is an alternation operator, and \| is literal.
135
If not set, then \| is an alternation operator, and | is literal. */
136
#define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
138
/* If this bit is set, then an ending range point collating higher
139
than the starting range point, as in [z-a], is invalid.
140
If not set, then when ending range point collates higher than the
141
starting range point, the range is ignored. */
142
#define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
144
/* If this bit is set, then an unmatched ) is ordinary.
145
If not set, then an unmatched ) is invalid. */
146
#define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
148
/* If this bit is set, succeed as soon as we match the whole pattern,
149
without further backtracking. */
150
#define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1)
152
/* If this bit is set, do not process the GNU regex operators.
153
If not set, then the GNU regex operators are recognized. */
154
#define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1)
156
/* If this bit is set, turn on internal regex debugging.
157
If not set, and debugging was on, turn it off.
158
This only works if regex.c is compiled -DDEBUG.
159
We define this bit always, so that all that's needed to turn on
160
debugging is to recompile regex.c; the calling code can always have
161
this bit set, and it won't affect anything in the normal case. */
162
#define RE_DEBUG (RE_NO_GNU_OPS << 1)
164
/* This global variable defines the particular regexp syntax to use (for
165
some interfaces). When a regexp is compiled, the syntax used is
166
stored in the pattern buffer, so changing this does not affect
167
already-compiled regexps. */
168
extern reg_syntax_t re_syntax_options;
170
/* Define combinations of the above bits for the standard possibilities.
171
(The [[[ comments delimit what gets put into the Texinfo file, so
172
don't delete them!) */
173
/* [[[begin syntaxes]]] */
174
#define RE_SYNTAX_EMACS 0
176
#define RE_SYNTAX_AWK \
177
(RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \
178
| RE_NO_BK_PARENS | RE_NO_BK_REFS \
179
| RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \
180
| RE_DOT_NEWLINE | RE_CONTEXT_INDEP_ANCHORS \
181
| RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS)
183
#define RE_SYNTAX_GNU_AWK \
184
((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DEBUG) \
185
& ~(RE_DOT_NOT_NULL | RE_INTERVALS | RE_CONTEXT_INDEP_OPS))
187
#define RE_SYNTAX_POSIX_AWK \
188
(RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS \
189
| RE_INTERVALS | RE_NO_GNU_OPS)
191
#define RE_SYNTAX_GREP \
192
(RE_BK_PLUS_QM | RE_CHAR_CLASSES \
193
| RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS \
196
#define RE_SYNTAX_EGREP \
197
(RE_CHAR_CLASSES | RE_CONTEXT_INDEP_ANCHORS \
198
| RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE \
199
| RE_NEWLINE_ALT | RE_NO_BK_PARENS \
202
#define RE_SYNTAX_POSIX_EGREP \
203
(RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES)
205
/* P1003.2/D11.2, section 4.20.7.1, lines 5078ff. */
206
#define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
208
#define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
210
/* Syntax bits common to both basic and extended POSIX regex syntax. */
211
#define _RE_SYNTAX_POSIX_COMMON \
212
(RE_CHAR_CLASSES | RE_DOT_NEWLINE | RE_DOT_NOT_NULL \
213
| RE_INTERVALS | RE_NO_EMPTY_RANGES)
215
#define RE_SYNTAX_POSIX_BASIC \
216
(_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM)
218
/* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
219
RE_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this
220
isn't minimal, since other operators, such as \`, aren't disabled. */
221
#define RE_SYNTAX_POSIX_MINIMAL_BASIC \
222
(_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
224
#define RE_SYNTAX_POSIX_EXTENDED \
225
(_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
226
| RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \
227
| RE_NO_BK_PARENS | RE_NO_BK_VBAR \
228
| RE_UNMATCHED_RIGHT_PAREN_ORD)
230
/* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INVALID_OPS
231
replaces RE_CONTEXT_INDEP_OPS and RE_NO_BK_REFS is added. */
232
#define RE_SYNTAX_POSIX_MINIMAL_EXTENDED \
233
(_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
234
| RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES \
235
| RE_NO_BK_PARENS | RE_NO_BK_REFS \
236
| RE_NO_BK_VBAR | RE_UNMATCHED_RIGHT_PAREN_ORD)
237
/* [[[end syntaxes]]] */
239
/* Maximum number of duplicates an interval can allow. Some systems
240
(erroneously) define this in other header files, but we want our
241
value, so remove any previous define. */
245
/* If sizeof(int) == 2, then ((1 << 15) - 1) overflows. */
246
#define RE_DUP_MAX (0x7fff)
249
/* POSIX `cflags' bits (i.e., information for `regcomp'). */
251
/* If this bit is set, then use extended regular expression syntax.
252
If not set, then use basic regular expression syntax. */
253
#define REG_EXTENDED 1
255
/* If this bit is set, then ignore case when matching.
256
If not set, then case is significant. */
257
#define REG_ICASE (REG_EXTENDED << 1)
259
/* If this bit is set, then anchors do not match at newline
260
characters in the string.
261
If not set, then anchors do match at newlines. */
262
#define REG_NEWLINE (REG_ICASE << 1)
264
/* If this bit is set, then report only success or fail in regexec.
265
If not set, then returns differ between not matching and errors. */
266
#define REG_NOSUB (REG_NEWLINE << 1)
269
/* POSIX `eflags' bits (i.e., information for regexec). */
271
/* If this bit is set, then the beginning-of-line operator doesn't match
272
the beginning of the string (presumably because it's not the
273
beginning of a line).
274
If not set, then the beginning-of-line operator does match the
275
beginning of the string. */
278
/* Like REG_NOTBOL, except for the end-of-line. */
279
#define REG_NOTEOL (1 << 1)
282
/* If any error codes are removed, changed, or added, update the
283
`re_error_msg' table in regex.c. */
286
REG_NOERROR = 0, /* Success. */
287
REG_NOMATCH, /* Didn't find a match (for regexec). */
289
/* POSIX regcomp return error codes. (In the order listed in the
291
REG_BADPAT, /* Invalid pattern. */
292
REG_ECOLLATE, /* Not implemented. */
293
REG_ECTYPE, /* Invalid character class name. */
294
REG_EESCAPE, /* Trailing backslash. */
295
REG_ESUBREG, /* Invalid back reference. */
296
REG_EBRACK, /* Unmatched left bracket. */
297
REG_EPAREN, /* Parenthesis imbalance. */
298
REG_EBRACE, /* Unmatched \{. */
299
REG_BADBR, /* Invalid contents of \{\}. */
300
REG_ERANGE, /* Invalid range end. */
301
REG_ESPACE, /* Ran out of memory. */
302
REG_BADRPT, /* No preceding re for repetition op. */
304
/* Error codes we've added. */
305
REG_EEND, /* Premature end. */
306
REG_ESIZE, /* Compiled pattern bigger than 2^16 bytes. */
307
REG_ERPAREN /* Unmatched ) or \); not returned from regcomp. */
310
/* This data structure represents a compiled pattern. Before calling
311
the pattern compiler, the fields `buffer', `allocated', `fastmap',
312
`translate', and `no_sub' can be set. After the pattern has been
313
compiled, the `re_nsub' field is available. All other fields are
314
private to the regex routines. */
316
#ifndef RE_TRANSLATE_TYPE
317
#define RE_TRANSLATE_TYPE char *
320
struct re_pattern_buffer
322
/* [[[begin pattern_buffer]]] */
323
/* Space that holds the compiled pattern. It is declared as
324
`unsigned char *' because its elements are
325
sometimes used as array indexes. */
326
unsigned char *buffer;
328
/* Number of bytes to which `buffer' points. */
329
unsigned long int allocated;
331
/* Number of bytes actually used in `buffer'. */
332
unsigned long int used;
334
/* Syntax setting with which the pattern was compiled. */
337
/* Pointer to a fastmap, if any, otherwise zero. re_search uses
338
the fastmap, if there is one, to skip over impossible
339
starting points for matches. */
342
/* Either a translate table to apply to all characters before
343
comparing them, or zero for no translation. The translation
344
is applied to a pattern when it is compiled and to a string
345
when it is matched. */
346
RE_TRANSLATE_TYPE translate;
348
/* Number of subexpressions found by the compiler. */
351
/* Zero if this pattern cannot match the empty string, one else.
352
Well, in truth it's used only in `re_search_2', to see
353
whether or not we should use the fastmap, so we don't set
354
this absolutely perfectly; see `re_compile_fastmap' (the
355
`duplicate' case). */
356
unsigned can_be_null : 1;
358
/* If REGS_UNALLOCATED, allocate space in the `regs' structure
359
for `max (RE_NREGS, re_nsub + 1)' groups.
360
If REGS_REALLOCATE, reallocate space if necessary.
361
If REGS_FIXED, use what's there. */
362
#define REGS_UNALLOCATED 0
363
#define REGS_REALLOCATE 1
365
unsigned regs_allocated : 2;
367
/* Set to zero when `regex_compile' compiles a pattern; set to one
368
by `re_compile_fastmap' if it updates the fastmap. */
369
unsigned fastmap_accurate : 1;
371
/* If set, `re_match_2' does not return information about
375
/* If set, a beginning-of-line anchor doesn't match at the
376
beginning of the string. */
377
unsigned not_bol : 1;
379
/* Similarly for an end-of-line anchor. */
380
unsigned not_eol : 1;
382
/* If true, an anchor at a newline matches. */
383
unsigned newline_anchor : 1;
385
/* [[[end pattern_buffer]]] */
388
typedef struct re_pattern_buffer regex_t;
390
/* Type for byte offsets within the string. POSIX mandates this. */
391
typedef int regoff_t;
394
/* This is the structure we store register match data in. See
395
regex.texinfo for a full description of what registers match. */
404
/* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
405
`re_match_2' returns information about at least this many registers
406
the first time a `regs' structure is passed. */
412
/* POSIX specification for registers. Aside from the different names than
413
`re_registers', POSIX uses an array of structures, instead of a
414
structure of arrays. */
417
regoff_t rm_so; /* Byte offset from string's start to substring's start. */
418
regoff_t rm_eo; /* Byte offset from string's start to substring's end. */
421
/* Declarations for routines. */
423
/* To avoid duplicating every routine declaration -- once with a
424
prototype (if we are ANSI), and once without (if we aren't) -- we
425
use the following macro to declare argument types. This
426
unfortunately clutters up the declarations a bit, but I think it's
429
#if defined(__STDC__)
431
#define _RE_ARGS(args) args
433
#else /* not __STDC__ */
435
#define _RE_ARGS(args) ()
437
#endif /* not __STDC__ */
439
/* Sets the current default syntax to SYNTAX, and return the old syntax.
440
You can also simply assign to the `re_syntax_options' variable. */
441
extern reg_syntax_t re_set_syntax _RE_ARGS ((reg_syntax_t syntax));
443
/* Compile the regular expression PATTERN, with length LENGTH
444
and syntax given by the global `re_syntax_options', into the buffer
445
BUFFER. Return NULL if successful, and an error string if not. */
446
extern const char *re_compile_pattern
447
_RE_ARGS ((const char *pattern, size_t length,
448
struct re_pattern_buffer *buffer));
451
/* Compile a fastmap for the compiled pattern in BUFFER; used to
452
accelerate searches. Return 0 if successful and -2 if was an
454
extern int re_compile_fastmap _RE_ARGS ((struct re_pattern_buffer *buffer));
457
/* Search in the string STRING (with length LENGTH) for the pattern
458
compiled into BUFFER. Start searching at position START, for RANGE
459
characters. Return the starting position of the match, -1 for no
460
match, or -2 for an internal error. Also return register
461
information in REGS (if REGS and BUFFER->no_sub are nonzero). */
463
_RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
464
int length, int start, int range, struct re_registers *regs));
467
/* Like `re_search', but search in the concatenation of STRING1 and
468
STRING2. Also, stop searching at index START + STOP. */
469
extern int re_search_2
470
_RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
471
int length1, const char *string2, int length2,
472
int start, int range, struct re_registers *regs, int stop));
475
/* Like `re_search', but return how many characters in STRING the regexp
476
in BUFFER matched, starting at position START. */
478
_RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
479
int length, int start, struct re_registers *regs));
482
/* Relates to `re_match' as `re_search_2' relates to `re_search'. */
483
extern int re_match_2
484
_RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
485
int length1, const char *string2, int length2,
486
int start, struct re_registers *regs, int stop));
489
/* Set REGS to hold NUM_REGS registers, storing them in STARTS and
490
ENDS. Subsequent matches using BUFFER and REGS will use this memory
491
for recording register information. STARTS and ENDS must be
492
allocated with malloc, and must each be at least `NUM_REGS * sizeof
493
(regoff_t)' bytes long.
495
If NUM_REGS == 0, then subsequent matches should allocate their own
498
Unless this function is called, the first search or match using
499
PATTERN_BUFFER will allocate its own register data, without
500
freeing the old data. */
501
extern void re_set_registers
502
_RE_ARGS ((struct re_pattern_buffer *buffer, struct re_registers *regs,
503
unsigned num_regs, regoff_t *starts, regoff_t *ends));
505
#ifdef _REGEX_RE_COMP
507
/* 4.2 bsd compatibility. */
508
extern char *re_comp _RE_ARGS ((const char *));
509
extern int re_exec _RE_ARGS ((const char *));
513
/* POSIX compatibility. */
514
extern int regcomp _RE_ARGS ((regex_t *preg, const char *pattern, int cflags));
516
_RE_ARGS ((const regex_t *preg, const char *string, size_t nmatch,
517
regmatch_t pmatch[], int eflags));
518
extern size_t regerror
519
_RE_ARGS ((int errcode, const regex_t *preg, char *errbuf,
520
size_t errbuf_size));
521
extern void regfree _RE_ARGS ((regex_t *preg));
534
trim-versions-without-asking: nil
added
removed
regexrepl/regex.h
