* lib/regex.c: New file, GNU regex implementation.
* lib/regex.h: New file. * lib/posix: New directory. * lib/posix/Makefile.am: New file. * lib/posix/regex.h: New file. * lib/Makefile.am: Add SUBDIRS=posix and regex.c dependencies.
Showing
6 changed files
with
622 additions
and
1 deletions
1 | AUTOMAKE_OPTIONS = ansi2knr | 1 | AUTOMAKE_OPTIONS = ansi2knr |
2 | noinst_LIBRARIES = libmailutils.a | 2 | noinst_LIBRARIES = libmailutils.a |
3 | 3 | ||
4 | SUBDIRS = posix | ||
5 | |||
4 | libmailutils_a_SOURCES = basename.c getopt.c getopt1.c md5.c getline.c \ | 6 | libmailutils_a_SOURCES = basename.c getopt.c getopt1.c md5.c getline.c \ |
5 | xstrdup.c xmalloc.c argcv.c | 7 | xstrdup.c xmalloc.c argcv.c |
6 | 8 | ||
... | @@ -8,7 +10,7 @@ EXTRA_DIST = alloca.c fnmatch.c setenv.c snprintf.c strchrnul.c strndup.c \ | ... | @@ -8,7 +10,7 @@ EXTRA_DIST = alloca.c fnmatch.c setenv.c snprintf.c strchrnul.c strndup.c \ |
8 | strnlen.c strtok_r.c strsignal.c xstrtol.c vasprintf.c malloc.c realloc.c | 10 | strnlen.c strtok_r.c strsignal.c xstrtol.c vasprintf.c malloc.c realloc.c |
9 | 11 | ||
10 | noinst_HEADERS = argcv.h error.h fnmatch.h getline.h getopt.h md5.h \ | 12 | noinst_HEADERS = argcv.h error.h fnmatch.h getline.h getopt.h md5.h \ |
11 | snprintf.h xalloc.h xstrtol.h | 13 | regex.h snprintf.h xalloc.h xstrtol.h |
12 | 14 | ||
13 | libmailutils_a_LIBADD = @LIBOBJS@ @ALLOCA@ | 15 | libmailutils_a_LIBADD = @LIBOBJS@ @ALLOCA@ |
14 | 16 | ... | ... |
lib/posix/.cvsignore
0 → 100644
lib/posix/Makefile.am
0 → 100644
lib/posix/regex.h
0 → 100644
1 | /* Definitions for data structures and routines for the regular | ||
2 | expression library, version 0.12. | ||
3 | Copyright (C) 1985,1989-1993,1995-1998, 2000 Free Software Foundation, Inc. | ||
4 | |||
5 | This file is part of the GNU C Library. Its master source is NOT part of | ||
6 | the C library, however. The master source lives in /gd/gnu/lib. | ||
7 | |||
8 | The GNU C Library is free software; you can redistribute it and/or | ||
9 | modify it under the terms of the GNU Library General Public License as | ||
10 | published by the Free Software Foundation; either version 2 of the | ||
11 | License, or (at your option) any later version. | ||
12 | |||
13 | The GNU C Library is distributed in the hope that it will be useful, | ||
14 | but WITHOUT ANY WARRANTY; without even the implied warranty of | ||
15 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU | ||
16 | Library General Public License for more details. | ||
17 | |||
18 | You should have received a copy of the GNU Library General Public | ||
19 | License along with the GNU C Library; see the file COPYING.LIB. If not, | ||
20 | write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330, | ||
21 | Boston, MA 02111-1307, USA. */ | ||
22 | |||
23 | #ifndef _REGEX_H | ||
24 | #define _REGEX_H 1 | ||
25 | |||
26 | /* Allow the use in C++ code. */ | ||
27 | #ifdef __cplusplus | ||
28 | extern "C" { | ||
29 | #endif | ||
30 | |||
31 | /* POSIX says that <sys/types.h> must be included (by the caller) before | ||
32 | <regex.h>. */ | ||
33 | |||
34 | #if !defined _POSIX_C_SOURCE && !defined _POSIX_SOURCE && defined VMS | ||
35 | /* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it | ||
36 | should be there. */ | ||
37 | # include <stddef.h> | ||
38 | #endif | ||
39 | |||
40 | /* The following two types have to be signed and unsigned integer type | ||
41 | wide enough to hold a value of a pointer. For most ANSI compilers | ||
42 | ptrdiff_t and size_t should be likely OK. Still size of these two | ||
43 | types is 2 for Microsoft C. Ugh... */ | ||
44 | typedef long int s_reg_t; | ||
45 | typedef unsigned long int active_reg_t; | ||
46 | |||
47 | /* The following bits are used to determine the regexp syntax we | ||
48 | recognize. The set/not-set meanings are chosen so that Emacs syntax | ||
49 | remains the value 0. The bits are given in alphabetical order, and | ||
50 | the definitions shifted by one from the previous bit; thus, when we | ||
51 | add or remove a bit, only one other definition need change. */ | ||
52 | typedef unsigned long int reg_syntax_t; | ||
53 | |||
54 | /* If this bit is not set, then \ inside a bracket expression is literal. | ||
55 | If set, then such a \ quotes the following character. */ | ||
56 | #define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1) | ||
57 | |||
58 | /* If this bit is not set, then + and ? are operators, and \+ and \? are | ||
59 | literals. | ||
60 | If set, then \+ and \? are operators and + and ? are literals. */ | ||
61 | #define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1) | ||
62 | |||
63 | /* If this bit is set, then character classes are supported. They are: | ||
64 | [:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:], | ||
65 | [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:]. | ||
66 | If not set, then character classes are not supported. */ | ||
67 | #define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1) | ||
68 | |||
69 | /* If this bit is set, then ^ and $ are always anchors (outside bracket | ||
70 | expressions, of course). | ||
71 | If this bit is not set, then it depends: | ||
72 | ^ is an anchor if it is at the beginning of a regular | ||
73 | expression or after an open-group or an alternation operator; | ||
74 | $ is an anchor if it is at the end of a regular expression, or | ||
75 | before a close-group or an alternation operator. | ||
76 | |||
77 | This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because | ||
78 | POSIX draft 11.2 says that * etc. in leading positions is undefined. | ||
79 | We already implemented a previous draft which made those constructs | ||
80 | invalid, though, so we haven't changed the code back. */ | ||
81 | #define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1) | ||
82 | |||
83 | /* If this bit is set, then special characters are always special | ||
84 | regardless of where they are in the pattern. | ||
85 | If this bit is not set, then special characters are special only in | ||
86 | some contexts; otherwise they are ordinary. Specifically, | ||
87 | * + ? and intervals are only special when not after the beginning, | ||
88 | open-group, or alternation operator. */ | ||
89 | #define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1) | ||
90 | |||
91 | /* If this bit is set, then *, +, ?, and { cannot be first in an re or | ||
92 | immediately after an alternation or begin-group operator. */ | ||
93 | #define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1) | ||
94 | |||
95 | /* If this bit is set, then . matches newline. | ||
96 | If not set, then it doesn't. */ | ||
97 | #define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1) | ||
98 | |||
99 | /* If this bit is set, then . doesn't match NUL. | ||
100 | If not set, then it does. */ | ||
101 | #define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1) | ||
102 | |||
103 | /* If this bit is set, nonmatching lists [^...] do not match newline. | ||
104 | If not set, they do. */ | ||
105 | #define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1) | ||
106 | |||
107 | /* If this bit is set, either \{...\} or {...} defines an | ||
108 | interval, depending on RE_NO_BK_BRACES. | ||
109 | If not set, \{, \}, {, and } are literals. */ | ||
110 | #define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1) | ||
111 | |||
112 | /* If this bit is set, +, ? and | aren't recognized as operators. | ||
113 | If not set, they are. */ | ||
114 | #define RE_LIMITED_OPS (RE_INTERVALS << 1) | ||
115 | |||
116 | /* If this bit is set, newline is an alternation operator. | ||
117 | If not set, newline is literal. */ | ||
118 | #define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1) | ||
119 | |||
120 | /* If this bit is set, then `{...}' defines an interval, and \{ and \} | ||
121 | are literals. | ||
122 | If not set, then `\{...\}' defines an interval. */ | ||
123 | #define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1) | ||
124 | |||
125 | /* If this bit is set, (...) defines a group, and \( and \) are literals. | ||
126 | If not set, \(...\) defines a group, and ( and ) are literals. */ | ||
127 | #define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1) | ||
128 | |||
129 | /* If this bit is set, then \<digit> matches <digit>. | ||
130 | If not set, then \<digit> is a back-reference. */ | ||
131 | #define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1) | ||
132 | |||
133 | /* If this bit is set, then | is an alternation operator, and \| is literal. | ||
134 | If not set, then \| is an alternation operator, and | is literal. */ | ||
135 | #define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1) | ||
136 | |||
137 | /* If this bit is set, then an ending range point collating higher | ||
138 | than the starting range point, as in [z-a], is invalid. | ||
139 | If not set, then when ending range point collates higher than the | ||
140 | starting range point, the range is ignored. */ | ||
141 | #define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1) | ||
142 | |||
143 | /* If this bit is set, then an unmatched ) is ordinary. | ||
144 | If not set, then an unmatched ) is invalid. */ | ||
145 | #define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1) | ||
146 | |||
147 | /* If this bit is set, succeed as soon as we match the whole pattern, | ||
148 | without further backtracking. */ | ||
149 | #define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1) | ||
150 | |||
151 | /* If this bit is set, do not process the GNU regex operators. | ||
152 | If not set, then the GNU regex operators are recognized. */ | ||
153 | #define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1) | ||
154 | |||
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_GNU_OPS << 1) | ||
162 | |||
163 | /* If this bit is set, a syntactically invalid interval is treated as | ||
164 | a string of ordinary characters. For example, the ERE 'a{1' is | ||
165 | treated as 'a\{1'. */ | ||
166 | #define RE_INVALID_INTERVAL_ORD (RE_DEBUG << 1) | ||
167 | |||
168 | /* This global variable defines the particular regexp syntax to use (for | ||
169 | some interfaces). When a regexp is compiled, the syntax used is | ||
170 | stored in the pattern buffer, so changing this does not affect | ||
171 | already-compiled regexps. */ | ||
172 | extern reg_syntax_t re_syntax_options; | ||
173 | |||
174 | /* Define combinations of the above bits for the standard possibilities. | ||
175 | (The [[[ comments delimit what gets put into the Texinfo file, so | ||
176 | don't delete them!) */ | ||
177 | /* [[[begin syntaxes]]] */ | ||
178 | #define RE_SYNTAX_EMACS 0 | ||
179 | |||
180 | #define RE_SYNTAX_AWK \ | ||
181 | (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \ | ||
182 | | RE_NO_BK_PARENS | RE_NO_BK_REFS \ | ||
183 | | RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \ | ||
184 | | RE_DOT_NEWLINE | RE_CONTEXT_INDEP_ANCHORS \ | ||
185 | | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS) | ||
186 | |||
187 | #define RE_SYNTAX_GNU_AWK \ | ||
188 | ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DEBUG) \ | ||
189 | & ~(RE_DOT_NOT_NULL | RE_INTERVALS | RE_CONTEXT_INDEP_OPS)) | ||
190 | |||
191 | #define RE_SYNTAX_POSIX_AWK \ | ||
192 | (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS \ | ||
193 | | RE_INTERVALS | RE_NO_GNU_OPS) | ||
194 | |||
195 | #define RE_SYNTAX_GREP \ | ||
196 | (RE_BK_PLUS_QM | RE_CHAR_CLASSES \ | ||
197 | | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS \ | ||
198 | | RE_NEWLINE_ALT) | ||
199 | |||
200 | #define RE_SYNTAX_EGREP \ | ||
201 | (RE_CHAR_CLASSES | RE_CONTEXT_INDEP_ANCHORS \ | ||
202 | | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE \ | ||
203 | | RE_NEWLINE_ALT | RE_NO_BK_PARENS \ | ||
204 | | RE_NO_BK_VBAR) | ||
205 | |||
206 | #define RE_SYNTAX_POSIX_EGREP \ | ||
207 | (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES \ | ||
208 | | RE_INVALID_INTERVAL_ORD) | ||
209 | |||
210 | /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff. */ | ||
211 | #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC | ||
212 | |||
213 | #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC | ||
214 | |||
215 | /* Syntax bits common to both basic and extended POSIX regex syntax. */ | ||
216 | #define _RE_SYNTAX_POSIX_COMMON \ | ||
217 | (RE_CHAR_CLASSES | RE_DOT_NEWLINE | RE_DOT_NOT_NULL \ | ||
218 | | RE_INTERVALS | RE_NO_EMPTY_RANGES) | ||
219 | |||
220 | #define RE_SYNTAX_POSIX_BASIC \ | ||
221 | (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM) | ||
222 | |||
223 | /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes | ||
224 | RE_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this | ||
225 | isn't minimal, since other operators, such as \`, aren't disabled. */ | ||
226 | #define RE_SYNTAX_POSIX_MINIMAL_BASIC \ | ||
227 | (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS) | ||
228 | |||
229 | #define RE_SYNTAX_POSIX_EXTENDED \ | ||
230 | (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \ | ||
231 | | RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \ | ||
232 | | RE_NO_BK_PARENS | RE_NO_BK_VBAR \ | ||
233 | | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD) | ||
234 | |||
235 | /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is | ||
236 | removed and RE_NO_BK_REFS is added. */ | ||
237 | #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED \ | ||
238 | (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \ | ||
239 | | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES \ | ||
240 | | RE_NO_BK_PARENS | RE_NO_BK_REFS \ | ||
241 | | RE_NO_BK_VBAR | RE_UNMATCHED_RIGHT_PAREN_ORD) | ||
242 | /* [[[end syntaxes]]] */ | ||
243 | |||
244 | /* Maximum number of duplicates an interval can allow. Some systems | ||
245 | (erroneously) define this in other header files, but we want our | ||
246 | value, so remove any previous define. */ | ||
247 | #ifdef RE_DUP_MAX | ||
248 | # undef RE_DUP_MAX | ||
249 | #endif | ||
250 | /* If sizeof(int) == 2, then ((1 << 15) - 1) overflows. */ | ||
251 | #define RE_DUP_MAX (0x7fff) | ||
252 | |||
253 | |||
254 | /* POSIX `cflags' bits (i.e., information for `regcomp'). */ | ||
255 | |||
256 | /* If this bit is set, then use extended regular expression syntax. | ||
257 | If not set, then use basic regular expression syntax. */ | ||
258 | #define REG_EXTENDED 1 | ||
259 | |||
260 | /* If this bit is set, then ignore case when matching. | ||
261 | If not set, then case is significant. */ | ||
262 | #define REG_ICASE (REG_EXTENDED << 1) | ||
263 | |||
264 | /* If this bit is set, then anchors do not match at newline | ||
265 | characters in the string. | ||
266 | If not set, then anchors do match at newlines. */ | ||
267 | #define REG_NEWLINE (REG_ICASE << 1) | ||
268 | |||
269 | /* If this bit is set, then report only success or fail in regexec. | ||
270 | If not set, then returns differ between not matching and errors. */ | ||
271 | #define REG_NOSUB (REG_NEWLINE << 1) | ||
272 | |||
273 | |||
274 | /* POSIX `eflags' bits (i.e., information for regexec). */ | ||
275 | |||
276 | /* If this bit is set, then the beginning-of-line operator doesn't match | ||
277 | the beginning of the string (presumably because it's not the | ||
278 | beginning of a line). | ||
279 | If not set, then the beginning-of-line operator does match the | ||
280 | beginning of the string. */ | ||
281 | #define REG_NOTBOL 1 | ||
282 | |||
283 | /* Like REG_NOTBOL, except for the end-of-line. */ | ||
284 | #define REG_NOTEOL (1 << 1) | ||
285 | |||
286 | |||
287 | /* If any error codes are removed, changed, or added, update the | ||
288 | `re_error_msg' table in regex.c. */ | ||
289 | typedef enum | ||
290 | { | ||
291 | #ifdef _XOPEN_SOURCE | ||
292 | REG_ENOSYS = -1, /* This will never happen for this implementation. */ | ||
293 | #endif | ||
294 | |||
295 | REG_NOERROR = 0, /* Success. */ | ||
296 | REG_NOMATCH, /* Didn't find a match (for regexec). */ | ||
297 | |||
298 | /* POSIX regcomp return error codes. (In the order listed in the | ||
299 | standard.) */ | ||
300 | REG_BADPAT, /* Invalid pattern. */ | ||
301 | REG_ECOLLATE, /* Not implemented. */ | ||
302 | REG_ECTYPE, /* Invalid character class name. */ | ||
303 | REG_EESCAPE, /* Trailing backslash. */ | ||
304 | REG_ESUBREG, /* Invalid back reference. */ | ||
305 | REG_EBRACK, /* Unmatched left bracket. */ | ||
306 | REG_EPAREN, /* Parenthesis imbalance. */ | ||
307 | REG_EBRACE, /* Unmatched \{. */ | ||
308 | REG_BADBR, /* Invalid contents of \{\}. */ | ||
309 | REG_ERANGE, /* Invalid range end. */ | ||
310 | REG_ESPACE, /* Ran out of memory. */ | ||
311 | REG_BADRPT, /* No preceding re for repetition op. */ | ||
312 | |||
313 | /* Error codes we've added. */ | ||
314 | REG_EEND, /* Premature end. */ | ||
315 | REG_ESIZE, /* Compiled pattern bigger than 2^16 bytes. */ | ||
316 | REG_ERPAREN /* Unmatched ) or \); not returned from regcomp. */ | ||
317 | } reg_errcode_t; | ||
318 | |||
319 | /* This data structure represents a compiled pattern. Before calling | ||
320 | the pattern compiler, the fields `buffer', `allocated', `fastmap', | ||
321 | `translate', and `no_sub' can be set. After the pattern has been | ||
322 | compiled, the `re_nsub' field is available. All other fields are | ||
323 | private to the regex routines. */ | ||
324 | |||
325 | #ifndef RE_TRANSLATE_TYPE | ||
326 | # define RE_TRANSLATE_TYPE char * | ||
327 | #endif | ||
328 | |||
329 | struct re_pattern_buffer | ||
330 | { | ||
331 | /* [[[begin pattern_buffer]]] */ | ||
332 | /* Space that holds the compiled pattern. It is declared as | ||
333 | `unsigned char *' because its elements are | ||
334 | sometimes used as array indexes. */ | ||
335 | unsigned char *buffer; | ||
336 | |||
337 | /* Number of bytes to which `buffer' points. */ | ||
338 | unsigned long int allocated; | ||
339 | |||
340 | /* Number of bytes actually used in `buffer'. */ | ||
341 | unsigned long int used; | ||
342 | |||
343 | /* Syntax setting with which the pattern was compiled. */ | ||
344 | reg_syntax_t syntax; | ||
345 | |||
346 | /* Pointer to a fastmap, if any, otherwise zero. re_search uses | ||
347 | the fastmap, if there is one, to skip over impossible | ||
348 | starting points for matches. */ | ||
349 | char *fastmap; | ||
350 | |||
351 | /* Either a translate table to apply to all characters before | ||
352 | comparing them, or zero for no translation. The translation | ||
353 | is applied to a pattern when it is compiled and to a string | ||
354 | when it is matched. */ | ||
355 | RE_TRANSLATE_TYPE translate; | ||
356 | |||
357 | /* Number of subexpressions found by the compiler. */ | ||
358 | size_t re_nsub; | ||
359 | |||
360 | /* Zero if this pattern cannot match the empty string, one else. | ||
361 | Well, in truth it's used only in `re_search_2', to see | ||
362 | whether or not we should use the fastmap, so we don't set | ||
363 | this absolutely perfectly; see `re_compile_fastmap' (the | ||
364 | `duplicate' case). */ | ||
365 | unsigned can_be_null : 1; | ||
366 | |||
367 | /* If REGS_UNALLOCATED, allocate space in the `regs' structure | ||
368 | for `max (RE_NREGS, re_nsub + 1)' groups. | ||
369 | If REGS_REALLOCATE, reallocate space if necessary. | ||
370 | If REGS_FIXED, use what's there. */ | ||
371 | #define REGS_UNALLOCATED 0 | ||
372 | #define REGS_REALLOCATE 1 | ||
373 | #define REGS_FIXED 2 | ||
374 | unsigned regs_allocated : 2; | ||
375 | |||
376 | /* Set to zero when `regex_compile' compiles a pattern; set to one | ||
377 | by `re_compile_fastmap' if it updates the fastmap. */ | ||
378 | unsigned fastmap_accurate : 1; | ||
379 | |||
380 | /* If set, `re_match_2' does not return information about | ||
381 | subexpressions. */ | ||
382 | unsigned no_sub : 1; | ||
383 | |||
384 | /* If set, a beginning-of-line anchor doesn't match at the | ||
385 | beginning of the string. */ | ||
386 | unsigned not_bol : 1; | ||
387 | |||
388 | /* Similarly for an end-of-line anchor. */ | ||
389 | unsigned not_eol : 1; | ||
390 | |||
391 | /* If true, an anchor at a newline matches. */ | ||
392 | unsigned newline_anchor : 1; | ||
393 | |||
394 | /* [[[end pattern_buffer]]] */ | ||
395 | }; | ||
396 | |||
397 | typedef struct re_pattern_buffer regex_t; | ||
398 | |||
399 | /* Type for byte offsets within the string. POSIX mandates this. */ | ||
400 | typedef int regoff_t; | ||
401 | |||
402 | |||
403 | /* This is the structure we store register match data in. See | ||
404 | regex.texinfo for a full description of what registers match. */ | ||
405 | struct re_registers | ||
406 | { | ||
407 | unsigned num_regs; | ||
408 | regoff_t *start; | ||
409 | regoff_t *end; | ||
410 | }; | ||
411 | |||
412 | |||
413 | /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer, | ||
414 | `re_match_2' returns information about at least this many registers | ||
415 | the first time a `regs' structure is passed. */ | ||
416 | #ifndef RE_NREGS | ||
417 | # define RE_NREGS 30 | ||
418 | #endif | ||
419 | |||
420 | |||
421 | /* POSIX specification for registers. Aside from the different names than | ||
422 | `re_registers', POSIX uses an array of structures, instead of a | ||
423 | structure of arrays. */ | ||
424 | typedef struct | ||
425 | { | ||
426 | regoff_t rm_so; /* Byte offset from string's start to substring's start. */ | ||
427 | regoff_t rm_eo; /* Byte offset from string's start to substring's end. */ | ||
428 | } regmatch_t; | ||
429 | |||
430 | /* Declarations for routines. */ | ||
431 | |||
432 | /* To avoid duplicating every routine declaration -- once with a | ||
433 | prototype (if we are ANSI), and once without (if we aren't) -- we | ||
434 | use the following macro to declare argument types. This | ||
435 | unfortunately clutters up the declarations a bit, but I think it's | ||
436 | worth it. */ | ||
437 | |||
438 | #if __STDC__ | ||
439 | |||
440 | # define _RE_ARGS(args) args | ||
441 | |||
442 | #else /* not __STDC__ */ | ||
443 | |||
444 | # define _RE_ARGS(args) () | ||
445 | |||
446 | #endif /* not __STDC__ */ | ||
447 | |||
448 | /* Sets the current default syntax to SYNTAX, and return the old syntax. | ||
449 | You can also simply assign to the `re_syntax_options' variable. */ | ||
450 | extern reg_syntax_t re_set_syntax _RE_ARGS ((reg_syntax_t syntax)); | ||
451 | |||
452 | /* Compile the regular expression PATTERN, with length LENGTH | ||
453 | and syntax given by the global `re_syntax_options', into the buffer | ||
454 | BUFFER. Return NULL if successful, and an error string if not. */ | ||
455 | extern const char *re_compile_pattern | ||
456 | _RE_ARGS ((const char *pattern, size_t length, | ||
457 | struct re_pattern_buffer *buffer)); | ||
458 | |||
459 | |||
460 | /* Compile a fastmap for the compiled pattern in BUFFER; used to | ||
461 | accelerate searches. Return 0 if successful and -2 if was an | ||
462 | internal error. */ | ||
463 | extern int re_compile_fastmap _RE_ARGS ((struct re_pattern_buffer *buffer)); | ||
464 | |||
465 | |||
466 | /* Search in the string STRING (with length LENGTH) for the pattern | ||
467 | compiled into BUFFER. Start searching at position START, for RANGE | ||
468 | characters. Return the starting position of the match, -1 for no | ||
469 | match, or -2 for an internal error. Also return register | ||
470 | information in REGS (if REGS and BUFFER->no_sub are nonzero). */ | ||
471 | extern int re_search | ||
472 | _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string, | ||
473 | int length, int start, int range, struct re_registers *regs)); | ||
474 | |||
475 | |||
476 | /* Like `re_search', but search in the concatenation of STRING1 and | ||
477 | STRING2. Also, stop searching at index START + STOP. */ | ||
478 | extern int re_search_2 | ||
479 | _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1, | ||
480 | int length1, const char *string2, int length2, | ||
481 | int start, int range, struct re_registers *regs, int stop)); | ||
482 | |||
483 | |||
484 | /* Like `re_search', but return how many characters in STRING the regexp | ||
485 | in BUFFER matched, starting at position START. */ | ||
486 | extern int re_match | ||
487 | _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string, | ||
488 | int length, int start, struct re_registers *regs)); | ||
489 | |||
490 | |||
491 | /* Relates to `re_match' as `re_search_2' relates to `re_search'. */ | ||
492 | extern int re_match_2 | ||
493 | _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1, | ||
494 | int length1, const char *string2, int length2, | ||
495 | int start, struct re_registers *regs, int stop)); | ||
496 | |||
497 | |||
498 | /* Set REGS to hold NUM_REGS registers, storing them in STARTS and | ||
499 | ENDS. Subsequent matches using BUFFER and REGS will use this memory | ||
500 | for recording register information. STARTS and ENDS must be | ||
501 | allocated with malloc, and must each be at least `NUM_REGS * sizeof | ||
502 | (regoff_t)' bytes long. | ||
503 | |||
504 | If NUM_REGS == 0, then subsequent matches should allocate their own | ||
505 | register data. | ||
506 | |||
507 | Unless this function is called, the first search or match using | ||
508 | PATTERN_BUFFER will allocate its own register data, without | ||
509 | freeing the old data. */ | ||
510 | extern void re_set_registers | ||
511 | _RE_ARGS ((struct re_pattern_buffer *buffer, struct re_registers *regs, | ||
512 | unsigned num_regs, regoff_t *starts, regoff_t *ends)); | ||
513 | |||
514 | #if defined _REGEX_RE_COMP || defined _LIBC | ||
515 | # ifndef _CRAY | ||
516 | /* 4.2 bsd compatibility. */ | ||
517 | extern char *re_comp _RE_ARGS ((const char *)); | ||
518 | extern int re_exec _RE_ARGS ((const char *)); | ||
519 | # endif | ||
520 | #endif | ||
521 | |||
522 | /* GCC 2.95 and later have "__restrict"; C99 compilers have | ||
523 | "restrict", and "configure" may have defined "restrict". */ | ||
524 | #ifndef __restrict | ||
525 | # if ! (2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__)) | ||
526 | # if defined restrict || 199901L <= __STDC_VERSION__ | ||
527 | # define __restrict restrict | ||
528 | # else | ||
529 | # define __restrict | ||
530 | # endif | ||
531 | # endif | ||
532 | #endif | ||
533 | /* For now unconditionally define __restrict_arr to expand to nothing. | ||
534 | Ideally we would have a test for the compiler which allows defining | ||
535 | it to restrict. */ | ||
536 | #define __restrict_arr | ||
537 | |||
538 | /* POSIX compatibility. */ | ||
539 | extern int regcomp _RE_ARGS ((regex_t *__restrict __preg, | ||
540 | const char *__restrict __pattern, | ||
541 | int __cflags)); | ||
542 | |||
543 | extern int regexec _RE_ARGS ((const regex_t *__restrict __preg, | ||
544 | const char *__restrict __string, size_t __nmatch, | ||
545 | regmatch_t __pmatch[__restrict_arr], | ||
546 | int __eflags)); | ||
547 | |||
548 | extern size_t regerror _RE_ARGS ((int __errcode, const regex_t *__preg, | ||
549 | char *__errbuf, size_t __errbuf_size)); | ||
550 | |||
551 | extern void regfree _RE_ARGS ((regex_t *__preg)); | ||
552 | |||
553 | |||
554 | #ifdef __cplusplus | ||
555 | } | ||
556 | #endif /* C++ */ | ||
557 | |||
558 | #endif /* regex.h */ | ||
559 | |||
560 | /* | ||
561 | Local variables: | ||
562 | make-backup-files: t | ||
563 | version-control: t | ||
564 | trim-versions-without-asking: nil | ||
565 | End: | ||
566 | */ |
lib/regex.c
0 → 100644
This diff could not be displayed because it is too large.
lib/regex.h
0 → 100644
1 | #ifndef _REGEX_H | ||
2 | #include <posix/regex.h> | ||
3 | |||
4 | /* Document internal interfaces. */ | ||
5 | extern reg_syntax_t __re_set_syntax _RE_ARGS ((reg_syntax_t syntax)); | ||
6 | |||
7 | extern const char *__re_compile_pattern | ||
8 | _RE_ARGS ((const char *pattern, size_t length, | ||
9 | struct re_pattern_buffer *buffer)); | ||
10 | |||
11 | extern int __re_compile_fastmap _RE_ARGS ((struct re_pattern_buffer *buffer)); | ||
12 | |||
13 | extern int __re_search | ||
14 | _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string, | ||
15 | int length, int start, int range, struct re_registers *regs)); | ||
16 | |||
17 | extern int __re_search_2 | ||
18 | _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1, | ||
19 | int length1, const char *string2, int length2, | ||
20 | int start, int range, struct re_registers *regs, int stop)); | ||
21 | |||
22 | extern int __re_match | ||
23 | _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string, | ||
24 | int length, int start, struct re_registers *regs)); | ||
25 | |||
26 | extern int __re_match_2 | ||
27 | _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1, | ||
28 | int length1, const char *string2, int length2, | ||
29 | int start, struct re_registers *regs, int stop)); | ||
30 | |||
31 | extern void __re_set_registers | ||
32 | _RE_ARGS ((struct re_pattern_buffer *buffer, struct re_registers *regs, | ||
33 | unsigned num_regs, regoff_t *starts, regoff_t *ends)); | ||
34 | |||
35 | extern int __regcomp _RE_ARGS ((regex_t *__preg, const char *__pattern, | ||
36 | int __cflags)); | ||
37 | |||
38 | extern int __regexec _RE_ARGS ((const regex_t *__preg, | ||
39 | const char *__string, size_t __nmatch, | ||
40 | regmatch_t __pmatch[], int __eflags)); | ||
41 | |||
42 | extern size_t __regerror _RE_ARGS ((int __errcode, const regex_t *__preg, | ||
43 | char *__errbuf, size_t __errbuf_size)); | ||
44 | |||
45 | extern void __regfree _RE_ARGS ((regex_t *__preg)); | ||
46 | #endif |
-
Please register or sign in to post a comment