1154
|
1 /* Definitions for data structures and routines for the regular
|
2454
|
2 expression library, version 0.12.
|
1154
|
3
|
28061
|
4 Copyright (C) 1985,89,90,91,92,93,95,2000 Free Software Foundation, Inc.
|
1154
|
5
|
|
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)
|
|
9 any later version.
|
|
10
|
|
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.
|
|
15
|
|
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
|
14414
|
18 Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307,
|
|
19 USA. */
|
1154
|
20
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
21 #ifndef _REGEX_H
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
22 #define _REGEX_H 1
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
23
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
24 /* Allow the use in C++ code. */
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
25 #ifdef __cplusplus
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
26 extern "C" {
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
27 #endif
|
1154
|
28
|
1637
|
29 /* POSIX says that <sys/types.h> must be included (by the caller) before
|
|
30 <regex.h>. */
|
|
31
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
32 #if !defined _POSIX_C_SOURCE && !defined _POSIX_SOURCE && defined VMS
|
1637
|
33 /* VMS doesn't have `size_t' in <sys/types.h>, even though POSIX says it
|
|
34 should be there. */
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
35 # include <stddef.h>
|
1579
|
36 #endif
|
|
37
|
1154
|
38 /* The following bits are used to determine the regexp syntax we
|
27692
bb0e45f6ca86
* regex.h (RE_SYNTAX_EMACS): Add RE_CHAR_CLASSES and RE_INTERVALS
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
39 recognize. The set/not-set meanings where historically chosen so
|
bb0e45f6ca86
* regex.h (RE_SYNTAX_EMACS): Add RE_CHAR_CLASSES and RE_INTERVALS
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
40 that Emacs syntax had the value 0.
|
bb0e45f6ca86
* regex.h (RE_SYNTAX_EMACS): Add RE_CHAR_CLASSES and RE_INTERVALS
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
41 The bits are given in alphabetical order, and
|
1154
|
42 the definitions shifted by one from the previous bit; thus, when we
|
|
43 add or remove a bit, only one other definition need change. */
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
44 typedef unsigned long int reg_syntax_t;
|
27450
|
45
|
1154
|
46 /* If this bit is not set, then \ inside a bracket expression is literal.
|
|
47 If set, then such a \ quotes the following character. */
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
48 #define RE_BACKSLASH_ESCAPE_IN_LISTS ((unsigned long int) 1)
|
1154
|
49
|
|
50 /* If this bit is not set, then + and ? are operators, and \+ and \? are
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
51 literals.
|
1154
|
52 If set, then \+ and \? are operators and + and ? are literals. */
|
|
53 #define RE_BK_PLUS_QM (RE_BACKSLASH_ESCAPE_IN_LISTS << 1)
|
|
54
|
|
55 /* If this bit is set, then character classes are supported. They are:
|
|
56 [:alpha:], [:upper:], [:lower:], [:digit:], [:alnum:], [:xdigit:],
|
|
57 [:space:], [:print:], [:punct:], [:graph:], and [:cntrl:].
|
|
58 If not set, then character classes are not supported. */
|
|
59 #define RE_CHAR_CLASSES (RE_BK_PLUS_QM << 1)
|
|
60
|
|
61 /* If this bit is set, then ^ and $ are always anchors (outside bracket
|
|
62 expressions, of course).
|
|
63 If this bit is not set, then it depends:
|
|
64 ^ is an anchor if it is at the beginning of a regular
|
|
65 expression or after an open-group or an alternation operator;
|
|
66 $ is an anchor if it is at the end of a regular expression, or
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
67 before a close-group or an alternation operator.
|
1154
|
68
|
|
69 This bit could be (re)combined with RE_CONTEXT_INDEP_OPS, because
|
|
70 POSIX draft 11.2 says that * etc. in leading positions is undefined.
|
|
71 We already implemented a previous draft which made those constructs
|
|
72 invalid, though, so we haven't changed the code back. */
|
|
73 #define RE_CONTEXT_INDEP_ANCHORS (RE_CHAR_CLASSES << 1)
|
|
74
|
|
75 /* If this bit is set, then special characters are always special
|
|
76 regardless of where they are in the pattern.
|
|
77 If this bit is not set, then special characters are special only in
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
78 some contexts; otherwise they are ordinary. Specifically,
|
1154
|
79 * + ? and intervals are only special when not after the beginning,
|
|
80 open-group, or alternation operator. */
|
|
81 #define RE_CONTEXT_INDEP_OPS (RE_CONTEXT_INDEP_ANCHORS << 1)
|
|
82
|
|
83 /* If this bit is set, then *, +, ?, and { cannot be first in an re or
|
|
84 immediately after an alternation or begin-group operator. */
|
|
85 #define RE_CONTEXT_INVALID_OPS (RE_CONTEXT_INDEP_OPS << 1)
|
|
86
|
|
87 /* If this bit is set, then . matches newline.
|
|
88 If not set, then it doesn't. */
|
|
89 #define RE_DOT_NEWLINE (RE_CONTEXT_INVALID_OPS << 1)
|
|
90
|
|
91 /* If this bit is set, then . doesn't match NUL.
|
|
92 If not set, then it does. */
|
|
93 #define RE_DOT_NOT_NULL (RE_DOT_NEWLINE << 1)
|
|
94
|
|
95 /* If this bit is set, nonmatching lists [^...] do not match newline.
|
|
96 If not set, they do. */
|
|
97 #define RE_HAT_LISTS_NOT_NEWLINE (RE_DOT_NOT_NULL << 1)
|
|
98
|
|
99 /* If this bit is set, either \{...\} or {...} defines an
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
100 interval, depending on RE_NO_BK_BRACES.
|
1154
|
101 If not set, \{, \}, {, and } are literals. */
|
|
102 #define RE_INTERVALS (RE_HAT_LISTS_NOT_NEWLINE << 1)
|
|
103
|
|
104 /* If this bit is set, +, ? and | aren't recognized as operators.
|
|
105 If not set, they are. */
|
|
106 #define RE_LIMITED_OPS (RE_INTERVALS << 1)
|
|
107
|
|
108 /* If this bit is set, newline is an alternation operator.
|
|
109 If not set, newline is literal. */
|
|
110 #define RE_NEWLINE_ALT (RE_LIMITED_OPS << 1)
|
|
111
|
|
112 /* If this bit is set, then `{...}' defines an interval, and \{ and \}
|
|
113 are literals.
|
|
114 If not set, then `\{...\}' defines an interval. */
|
|
115 #define RE_NO_BK_BRACES (RE_NEWLINE_ALT << 1)
|
|
116
|
|
117 /* If this bit is set, (...) defines a group, and \( and \) are literals.
|
|
118 If not set, \(...\) defines a group, and ( and ) are literals. */
|
|
119 #define RE_NO_BK_PARENS (RE_NO_BK_BRACES << 1)
|
|
120
|
|
121 /* If this bit is set, then \<digit> matches <digit>.
|
|
122 If not set, then \<digit> is a back-reference. */
|
|
123 #define RE_NO_BK_REFS (RE_NO_BK_PARENS << 1)
|
|
124
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
125 /* If this bit is set, then | is an alternation operator, and \| is literal.
|
1154
|
126 If not set, then \| is an alternation operator, and | is literal. */
|
|
127 #define RE_NO_BK_VBAR (RE_NO_BK_REFS << 1)
|
|
128
|
|
129 /* If this bit is set, then an ending range point collating higher
|
|
130 than the starting range point, as in [z-a], is invalid.
|
|
131 If not set, then when ending range point collates higher than the
|
|
132 starting range point, the range is ignored. */
|
|
133 #define RE_NO_EMPTY_RANGES (RE_NO_BK_VBAR << 1)
|
|
134
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
135 /* If this bit is set, then an unmatched ) is ordinary.
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
136 If not set, then an unmatched ) is invalid. */
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
137 #define RE_UNMATCHED_RIGHT_PAREN_ORD (RE_NO_EMPTY_RANGES << 1)
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
138
|
27450
|
139 /* If this bit is set, succeed as soon as we match the whole pattern,
|
|
140 without further backtracking. */
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
141 #define RE_NO_POSIX_BACKTRACKING (RE_UNMATCHED_RIGHT_PAREN_ORD << 1)
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
142
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
143 /* If this bit is set, do not process the GNU regex operators.
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
144 If not set, then the GNU regex operators are recognized. */
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
145 #define RE_NO_GNU_OPS (RE_NO_POSIX_BACKTRACKING << 1)
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
146
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
147 /* If this bit is set, then *?, +? and ?? match non greedily. */
|
31298
820483a506d0
(struct re_pattern_buffer): Use size_t for used/allocated.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
148 #define RE_FRUGAL (RE_NO_GNU_OPS << 1)
|
27450
|
149
|
28061
|
150 /* If this bit is set, then (?:...) is treated as a shy group. */
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
151 #define RE_SHY_GROUPS (RE_FRUGAL << 1)
|
9982
|
152
|
31312
|
153 /* If this bit is set, ^ and $ only match at beg/end of buffer. */
|
|
154 #define RE_NO_NEWLINE_ANCHOR (RE_SHY_GROUPS << 1)
|
|
155
|
|
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_NEWLINE_ANCHOR << 1)
|
|
163
|
1154
|
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;
|
18261
|
169
|
|
170 #ifdef emacs
|
|
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;
|
|
174 #endif
|
|
175
|
1154
|
176
|
|
177 /* Define combinations of the above bits for the standard possibilities.
|
|
178 (The [[[ comments delimit what gets put into the Texinfo file, so
|
27692
bb0e45f6ca86
* regex.h (RE_SYNTAX_EMACS): Add RE_CHAR_CLASSES and RE_INTERVALS
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
179 don't delete them!) */
|
1154
|
180 /* [[[begin syntaxes]]] */
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
181 #define RE_SYNTAX_EMACS \
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
182 (RE_CHAR_CLASSES | RE_INTERVALS | RE_SHY_GROUPS | RE_FRUGAL)
|
1154
|
183
|
|
184 #define RE_SYNTAX_AWK \
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
185 (RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DOT_NOT_NULL \
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
186 | RE_NO_BK_PARENS | RE_NO_BK_REFS \
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
187 | RE_NO_BK_VBAR | RE_NO_EMPTY_RANGES \
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
188 | RE_DOT_NEWLINE | RE_CONTEXT_INDEP_ANCHORS \
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
189 | RE_UNMATCHED_RIGHT_PAREN_ORD | RE_NO_GNU_OPS)
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
190
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
191 #define RE_SYNTAX_GNU_AWK \
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
192 ((RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS | RE_DEBUG) \
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
193 & ~(RE_DOT_NOT_NULL | RE_INTERVALS | RE_CONTEXT_INDEP_OPS))
|
1154
|
194
|
|
195 #define RE_SYNTAX_POSIX_AWK \
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
196 (RE_SYNTAX_POSIX_EXTENDED | RE_BACKSLASH_ESCAPE_IN_LISTS \
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
197 | RE_INTERVALS | RE_NO_GNU_OPS)
|
1154
|
198
|
|
199 #define RE_SYNTAX_GREP \
|
|
200 (RE_BK_PLUS_QM | RE_CHAR_CLASSES \
|
|
201 | RE_HAT_LISTS_NOT_NEWLINE | RE_INTERVALS \
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
202 | RE_NEWLINE_ALT)
|
1154
|
203
|
|
204 #define RE_SYNTAX_EGREP \
|
|
205 (RE_CHAR_CLASSES | RE_CONTEXT_INDEP_ANCHORS \
|
|
206 | RE_CONTEXT_INDEP_OPS | RE_HAT_LISTS_NOT_NEWLINE \
|
|
207 | RE_NEWLINE_ALT | RE_NO_BK_PARENS \
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
208 | RE_NO_BK_VBAR)
|
1154
|
209
|
|
210 #define RE_SYNTAX_POSIX_EGREP \
|
|
211 (RE_SYNTAX_EGREP | RE_INTERVALS | RE_NO_BK_BRACES)
|
|
212
|
1637
|
213 /* P1003.2/D11.2, section 4.20.7.1, lines 5078ff. */
|
|
214 #define RE_SYNTAX_ED RE_SYNTAX_POSIX_BASIC
|
|
215
|
1154
|
216 #define RE_SYNTAX_SED RE_SYNTAX_POSIX_BASIC
|
|
217
|
|
218 /* Syntax bits common to both basic and extended POSIX regex syntax. */
|
|
219 #define _RE_SYNTAX_POSIX_COMMON \
|
|
220 (RE_CHAR_CLASSES | RE_DOT_NEWLINE | RE_DOT_NOT_NULL \
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
221 | RE_INTERVALS | RE_NO_EMPTY_RANGES)
|
1154
|
222
|
|
223 #define RE_SYNTAX_POSIX_BASIC \
|
|
224 (_RE_SYNTAX_POSIX_COMMON | RE_BK_PLUS_QM)
|
|
225
|
|
226 /* Differs from ..._POSIX_BASIC only in that RE_BK_PLUS_QM becomes
|
|
227 RE_LIMITED_OPS, i.e., \? \+ \| are not recognized. Actually, this
|
|
228 isn't minimal, since other operators, such as \`, aren't disabled. */
|
|
229 #define RE_SYNTAX_POSIX_MINIMAL_BASIC \
|
|
230 (_RE_SYNTAX_POSIX_COMMON | RE_LIMITED_OPS)
|
|
231
|
|
232 #define RE_SYNTAX_POSIX_EXTENDED \
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
233 (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
234 | RE_CONTEXT_INDEP_OPS | RE_NO_BK_BRACES \
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
235 | RE_NO_BK_PARENS | RE_NO_BK_VBAR \
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
236 | RE_CONTEXT_INVALID_OPS | RE_UNMATCHED_RIGHT_PAREN_ORD)
|
1154
|
237
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
238 /* Differs from ..._POSIX_EXTENDED in that RE_CONTEXT_INDEP_OPS is
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
239 removed and RE_NO_BK_REFS is added. */
|
1154
|
240 #define RE_SYNTAX_POSIX_MINIMAL_EXTENDED \
|
|
241 (_RE_SYNTAX_POSIX_COMMON | RE_CONTEXT_INDEP_ANCHORS \
|
|
242 | RE_CONTEXT_INVALID_OPS | RE_NO_BK_BRACES \
|
|
243 | RE_NO_BK_PARENS | RE_NO_BK_REFS \
|
|
244 | RE_NO_BK_VBAR | RE_UNMATCHED_RIGHT_PAREN_ORD)
|
|
245 /* [[[end syntaxes]]] */
|
|
246
|
|
247 /* Maximum number of duplicates an interval can allow. Some systems
|
|
248 (erroneously) define this in other header files, but we want our
|
|
249 value, so remove any previous define. */
|
|
250 #ifdef RE_DUP_MAX
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
251 # undef RE_DUP_MAX
|
1154
|
252 #endif
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
253 /* If sizeof(int) == 2, then ((1 << 15) - 1) overflows. */
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
254 #define RE_DUP_MAX (0x7fff)
|
1154
|
255
|
|
256
|
|
257 /* POSIX `cflags' bits (i.e., information for `regcomp'). */
|
|
258
|
|
259 /* If this bit is set, then use extended regular expression syntax.
|
|
260 If not set, then use basic regular expression syntax. */
|
|
261 #define REG_EXTENDED 1
|
|
262
|
|
263 /* If this bit is set, then ignore case when matching.
|
|
264 If not set, then case is significant. */
|
|
265 #define REG_ICASE (REG_EXTENDED << 1)
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
266
|
1154
|
267 /* If this bit is set, then anchors do not match at newline
|
|
268 characters in the string.
|
|
269 If not set, then anchors do match at newlines. */
|
|
270 #define REG_NEWLINE (REG_ICASE << 1)
|
|
271
|
|
272 /* If this bit is set, then report only success or fail in regexec.
|
|
273 If not set, then returns differ between not matching and errors. */
|
|
274 #define REG_NOSUB (REG_NEWLINE << 1)
|
|
275
|
|
276
|
|
277 /* POSIX `eflags' bits (i.e., information for regexec). */
|
|
278
|
|
279 /* If this bit is set, then the beginning-of-line operator doesn't match
|
|
280 the beginning of the string (presumably because it's not the
|
|
281 beginning of a line).
|
|
282 If not set, then the beginning-of-line operator does match the
|
|
283 beginning of the string. */
|
|
284 #define REG_NOTBOL 1
|
|
285
|
|
286 /* Like REG_NOTBOL, except for the end-of-line. */
|
|
287 #define REG_NOTEOL (1 << 1)
|
|
288
|
|
289
|
|
290 /* If any error codes are removed, changed, or added, update the
|
|
291 `re_error_msg' table in regex.c. */
|
|
292 typedef enum
|
|
293 {
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
294 #ifdef _XOPEN_SOURCE
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
295 REG_ENOSYS = -1, /* This will never happen for this implementation. */
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
296 #endif
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
297
|
1154
|
298 REG_NOERROR = 0, /* Success. */
|
|
299 REG_NOMATCH, /* Didn't find a match (for regexec). */
|
|
300
|
|
301 /* POSIX regcomp return error codes. (In the order listed in the
|
|
302 standard.) */
|
|
303 REG_BADPAT, /* Invalid pattern. */
|
|
304 REG_ECOLLATE, /* Not implemented. */
|
|
305 REG_ECTYPE, /* Invalid character class name. */
|
|
306 REG_EESCAPE, /* Trailing backslash. */
|
|
307 REG_ESUBREG, /* Invalid back reference. */
|
|
308 REG_EBRACK, /* Unmatched left bracket. */
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
309 REG_EPAREN, /* Parenthesis imbalance. */
|
1154
|
310 REG_EBRACE, /* Unmatched \{. */
|
|
311 REG_BADBR, /* Invalid contents of \{\}. */
|
|
312 REG_ERANGE, /* Invalid range end. */
|
|
313 REG_ESPACE, /* Ran out of memory. */
|
|
314 REG_BADRPT, /* No preceding re for repetition op. */
|
|
315
|
|
316 /* Error codes we've added. */
|
|
317 REG_EEND, /* Premature end. */
|
|
318 REG_ESIZE, /* Compiled pattern bigger than 2^16 bytes. */
|
57666
|
319 REG_ERPAREN, /* Unmatched ) or \); not returned from regcomp. */
|
|
320 REG_ERANGEX /* Range striding over charsets. */
|
1154
|
321 } reg_errcode_t;
|
|
322
|
|
323 /* This data structure represents a compiled pattern. Before calling
|
|
324 the pattern compiler, the fields `buffer', `allocated', `fastmap',
|
|
325 `translate', and `no_sub' can be set. After the pattern has been
|
|
326 compiled, the `re_nsub' field is available. All other fields are
|
|
327 private to the regex routines. */
|
|
328
|
28202
|
329 #ifndef RE_TRANSLATE_TYPE
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
330 # define RE_TRANSLATE_TYPE char *
|
13249
|
331 #endif
|
|
332
|
1154
|
333 struct re_pattern_buffer
|
|
334 {
|
|
335 /* [[[begin pattern_buffer]]] */
|
|
336 /* Space that holds the compiled pattern. It is declared as
|
|
337 `unsigned char *' because its elements are
|
|
338 sometimes used as array indexes. */
|
|
339 unsigned char *buffer;
|
|
340
|
|
341 /* Number of bytes to which `buffer' points. */
|
31298
820483a506d0
(struct re_pattern_buffer): Use size_t for used/allocated.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
342 size_t allocated;
|
1154
|
343
|
|
344 /* Number of bytes actually used in `buffer'. */
|
31298
820483a506d0
(struct re_pattern_buffer): Use size_t for used/allocated.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
345 size_t used;
|
1154
|
346
|
|
347 /* Syntax setting with which the pattern was compiled. */
|
|
348 reg_syntax_t syntax;
|
|
349
|
|
350 /* Pointer to a fastmap, if any, otherwise zero. re_search uses
|
|
351 the fastmap, if there is one, to skip over impossible
|
|
352 starting points for matches. */
|
|
353 char *fastmap;
|
|
354
|
|
355 /* Either a translate table to apply to all characters before
|
|
356 comparing them, or zero for no translation. The translation
|
|
357 is applied to a pattern when it is compiled and to a string
|
|
358 when it is matched. */
|
13249
|
359 RE_TRANSLATE_TYPE translate;
|
1154
|
360
|
|
361 /* Number of subexpressions found by the compiler. */
|
|
362 size_t re_nsub;
|
|
363
|
|
364 /* Zero if this pattern cannot match the empty string, one else.
|
|
365 Well, in truth it's used only in `re_search_2', to see
|
|
366 whether or not we should use the fastmap, so we don't set
|
28061
|
367 this absolutely perfectly; see `re_compile_fastmap'. */
|
1154
|
368 unsigned can_be_null : 1;
|
|
369
|
|
370 /* If REGS_UNALLOCATED, allocate space in the `regs' structure
|
|
371 for `max (RE_NREGS, re_nsub + 1)' groups.
|
|
372 If REGS_REALLOCATE, reallocate space if necessary.
|
|
373 If REGS_FIXED, use what's there. */
|
|
374 #define REGS_UNALLOCATED 0
|
|
375 #define REGS_REALLOCATE 1
|
|
376 #define REGS_FIXED 2
|
|
377 unsigned regs_allocated : 2;
|
|
378
|
1637
|
379 /* Set to zero when `regex_compile' compiles a pattern; set to one
|
|
380 by `re_compile_fastmap' if it updates the fastmap. */
|
1154
|
381 unsigned fastmap_accurate : 1;
|
|
382
|
1637
|
383 /* If set, `re_match_2' does not return information about
|
|
384 subexpressions. */
|
1154
|
385 unsigned no_sub : 1;
|
|
386
|
|
387 /* If set, a beginning-of-line anchor doesn't match at the
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
388 beginning of the string. */
|
1154
|
389 unsigned not_bol : 1;
|
|
390
|
|
391 /* Similarly for an end-of-line anchor. */
|
|
392 unsigned not_eol : 1;
|
|
393
|
28473
|
394 #ifdef emacs
|
18261
|
395 /* If true, multi-byte form in the `buffer' should be recognized as a
|
|
396 multibyte character. */
|
|
397 unsigned multibyte : 1;
|
28473
|
398 #endif
|
18261
|
399
|
1154
|
400 /* [[[end pattern_buffer]]] */
|
|
401 };
|
|
402
|
|
403 typedef struct re_pattern_buffer regex_t;
|
|
404
|
|
405 /* Type for byte offsets within the string. POSIX mandates this. */
|
|
406 typedef int regoff_t;
|
|
407
|
|
408
|
|
409 /* This is the structure we store register match data in. See
|
|
410 regex.texinfo for a full description of what registers match. */
|
|
411 struct re_registers
|
|
412 {
|
|
413 unsigned num_regs;
|
|
414 regoff_t *start;
|
|
415 regoff_t *end;
|
|
416 };
|
|
417
|
|
418
|
|
419 /* If `regs_allocated' is REGS_UNALLOCATED in the pattern buffer,
|
|
420 `re_match_2' returns information about at least this many registers
|
|
421 the first time a `regs' structure is passed. */
|
|
422 #ifndef RE_NREGS
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
423 # define RE_NREGS 30
|
1154
|
424 #endif
|
|
425
|
|
426
|
|
427 /* POSIX specification for registers. Aside from the different names than
|
|
428 `re_registers', POSIX uses an array of structures, instead of a
|
|
429 structure of arrays. */
|
|
430 typedef struct
|
|
431 {
|
|
432 regoff_t rm_so; /* Byte offset from string's start to substring's start. */
|
|
433 regoff_t rm_eo; /* Byte offset from string's start to substring's end. */
|
|
434 } regmatch_t;
|
|
435
|
|
436 /* Declarations for routines. */
|
|
437
|
|
438 /* To avoid duplicating every routine declaration -- once with a
|
|
439 prototype (if we are ANSI), and once without (if we aren't) -- we
|
|
440 use the following macro to declare argument types. This
|
|
441 unfortunately clutters up the declarations a bit, but I think it's
|
1797
|
442 worth it. */
|
1154
|
443
|
29754
|
444 #if defined __STDC__ || defined PROTOTYPES
|
1722
|
445
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
446 # define _RE_ARGS(args) args
|
1722
|
447
|
29754
|
448 #else /* not __STDC__ || PROTOTYPES */
|
1722
|
449
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
450 # define _RE_ARGS(args) ()
|
1722
|
451
|
29754
|
452 #endif /* not __STDC__ || PROTOTYPES */
|
1154
|
453
|
|
454 /* Sets the current default syntax to SYNTAX, and return the old syntax.
|
|
455 You can also simply assign to the `re_syntax_options' variable. */
|
|
456 extern reg_syntax_t re_set_syntax _RE_ARGS ((reg_syntax_t syntax));
|
|
457
|
|
458 /* Compile the regular expression PATTERN, with length LENGTH
|
|
459 and syntax given by the global `re_syntax_options', into the buffer
|
|
460 BUFFER. Return NULL if successful, and an error string if not. */
|
|
461 extern const char *re_compile_pattern
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
462 _RE_ARGS ((const char *pattern, size_t length,
|
1154
|
463 struct re_pattern_buffer *buffer));
|
|
464
|
|
465
|
|
466 /* Compile a fastmap for the compiled pattern in BUFFER; used to
|
|
467 accelerate searches. Return 0 if successful and -2 if was an
|
|
468 internal error. */
|
|
469 extern int re_compile_fastmap _RE_ARGS ((struct re_pattern_buffer *buffer));
|
|
470
|
|
471
|
|
472 /* Search in the string STRING (with length LENGTH) for the pattern
|
|
473 compiled into BUFFER. Start searching at position START, for RANGE
|
|
474 characters. Return the starting position of the match, -1 for no
|
|
475 match, or -2 for an internal error. Also return register
|
|
476 information in REGS (if REGS and BUFFER->no_sub are nonzero). */
|
|
477 extern int re_search
|
|
478 _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
|
|
479 int length, int start, int range, struct re_registers *regs));
|
|
480
|
|
481
|
|
482 /* Like `re_search', but search in the concatenation of STRING1 and
|
|
483 STRING2. Also, stop searching at index START + STOP. */
|
|
484 extern int re_search_2
|
|
485 _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
|
|
486 int length1, const char *string2, int length2,
|
|
487 int start, int range, struct re_registers *regs, int stop));
|
|
488
|
|
489
|
|
490 /* Like `re_search', but return how many characters in STRING the regexp
|
|
491 in BUFFER matched, starting at position START. */
|
|
492 extern int re_match
|
|
493 _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string,
|
|
494 int length, int start, struct re_registers *regs));
|
|
495
|
|
496
|
|
497 /* Relates to `re_match' as `re_search_2' relates to `re_search'. */
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
498 extern int re_match_2
|
1154
|
499 _RE_ARGS ((struct re_pattern_buffer *buffer, const char *string1,
|
|
500 int length1, const char *string2, int length2,
|
|
501 int start, struct re_registers *regs, int stop));
|
|
502
|
|
503
|
|
504 /* Set REGS to hold NUM_REGS registers, storing them in STARTS and
|
|
505 ENDS. Subsequent matches using BUFFER and REGS will use this memory
|
|
506 for recording register information. STARTS and ENDS must be
|
|
507 allocated with malloc, and must each be at least `NUM_REGS * sizeof
|
|
508 (regoff_t)' bytes long.
|
|
509
|
|
510 If NUM_REGS == 0, then subsequent matches should allocate their own
|
|
511 register data.
|
|
512
|
|
513 Unless this function is called, the first search or match using
|
|
514 PATTERN_BUFFER will allocate its own register data, without
|
|
515 freeing the old data. */
|
|
516 extern void re_set_registers
|
|
517 _RE_ARGS ((struct re_pattern_buffer *buffer, struct re_registers *regs,
|
|
518 unsigned num_regs, regoff_t *starts, regoff_t *ends));
|
|
519
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
520 #if defined _REGEX_RE_COMP || defined _LIBC
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
521 # ifndef _CRAY
|
1154
|
522 /* 4.2 bsd compatibility. */
|
|
523 extern char *re_comp _RE_ARGS ((const char *));
|
|
524 extern int re_exec _RE_ARGS ((const char *));
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
525 # endif
|
9717
|
526 #endif
|
1154
|
527
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
528 /* GCC 2.95 and later have "__restrict"; C99 compilers have
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
529 "restrict", and "configure" may have defined "restrict". */
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
530 #ifndef __restrict
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
531 # if ! (2 < __GNUC__ || (2 == __GNUC__ && 95 <= __GNUC_MINOR__))
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
532 # if defined restrict || 199901L <= __STDC_VERSION__
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
533 # define __restrict restrict
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
534 # else
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
535 # define __restrict
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
536 # endif
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
537 # endif
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
538 #endif
|
47713
|
539 /* For now conditionally define __restrict_arr to expand to nothing.
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
540 Ideally we would have a test for the compiler which allows defining
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
541 it to restrict. */
|
47633
|
542 #ifndef __restrict_arr
|
47713
|
543 # define __restrict_arr
|
47633
|
544 #endif
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
545
|
1154
|
546 /* POSIX compatibility. */
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
547 extern int regcomp _RE_ARGS ((regex_t *__restrict __preg,
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
548 const char *__restrict __pattern,
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
549 int __cflags));
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
550
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
551 extern int regexec _RE_ARGS ((const regex_t *__restrict __preg,
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
552 const char *__restrict __string, size_t __nmatch,
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
553 regmatch_t __pmatch[__restrict_arr],
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
554 int __eflags));
|
1154
|
555
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
556 extern size_t regerror _RE_ARGS ((int __errcode, const regex_t *__preg,
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
557 char *__errbuf, size_t __errbuf_size));
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
558
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
559 extern void regfree _RE_ARGS ((regex_t *__preg));
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
560
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
561
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
562 #ifdef __cplusplus
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
563 }
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
564 #endif /* C++ */
|
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
565
|
56090
|
566 /* For platform which support the ISO C amendement 1 functionality we
|
|
567 support user defined character classes. */
|
|
568 #if WIDE_CHAR_SUPPORT
|
|
569 /* Solaris 2.5 has a bug: <wchar.h> must be included before <wctype.h>. */
|
|
570 # include <wchar.h>
|
|
571 # include <wctype.h>
|
|
572 #endif
|
|
573
|
|
574 #if WIDE_CHAR_SUPPORT
|
|
575 /* The GNU C library provides support for user-defined character classes
|
|
576 and the functions from ISO C amendement 1. */
|
|
577 # ifdef CHARCLASS_NAME_MAX
|
|
578 # define CHAR_CLASS_MAX_LENGTH CHARCLASS_NAME_MAX
|
|
579 # else
|
|
580 /* This shouldn't happen but some implementation might still have this
|
|
581 problem. Use a reasonable default value. */
|
|
582 # define CHAR_CLASS_MAX_LENGTH 256
|
|
583 # endif
|
|
584 typedef wctype_t re_wctype_t;
|
|
585 typedef wchar_t re_wchar_t;
|
|
586 # define re_wctype wctype
|
|
587 # define re_iswctype iswctype
|
|
588 # define re_wctype_to_bit(cc) 0
|
|
589 #else
|
|
590 # define CHAR_CLASS_MAX_LENGTH 9 /* Namely, `multibyte'. */
|
|
591 # define btowc(c) c
|
|
592
|
|
593 /* Character classes. */
|
|
594 typedef enum { RECC_ERROR = 0,
|
|
595 RECC_ALNUM, RECC_ALPHA, RECC_WORD,
|
|
596 RECC_GRAPH, RECC_PRINT,
|
|
597 RECC_LOWER, RECC_UPPER,
|
|
598 RECC_PUNCT, RECC_CNTRL,
|
|
599 RECC_DIGIT, RECC_XDIGIT,
|
|
600 RECC_BLANK, RECC_SPACE,
|
|
601 RECC_MULTIBYTE, RECC_NONASCII,
|
|
602 RECC_ASCII, RECC_UNIBYTE
|
|
603 } re_wctype_t;
|
|
604
|
|
605 typedef int re_wchar_t;
|
|
606
|
|
607 #endif /* not WIDE_CHAR_SUPPORT */
|
|
608
|
31213
9efb8adfefa4
* regex.c: Indent cpp directives and remove parens after `defined'.
Stefan Monnier <monnier@iro.umontreal.ca>
diff
changeset
|
609 #endif /* regex.h */
|
1154
|
610
|
|
611 /*
|
|
612 Local variables:
|
|
613 make-backup-files: t
|
|
614 version-control: t
|
|
615 trim-versions-without-asking: nil
|
|
616 End:
|
|
617 */
|
52401
|
618
|
|
619 /* arch-tag: bda6e3ec-3c02-4237-a55a-01ad2e120083
|
|
620 (do not change this comment) */
|