1 // class template regex -*- C++ -*-
2
3 // Copyright (C) 2010-2015 Free Software Foundation, Inc.
4 //
5 // This file is part of the GNU ISO C++ Library. This library is free
6 // software; you can redistribute it and/or modify it under the
7 // terms of the GNU General Public License as published by the
8 // Free Software Foundation; either version 3, or (at your option)
9 // any later version.
10
11 // This library 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 // Under Section 7 of GPL version 3, you are granted additional
17 // permissions described in the GCC Runtime Library Exception, version
18 // 3.1, as published by the Free Software Foundation.
19
20 // You should have received a copy of the GNU General Public License and
21 // a copy of the GCC Runtime Library Exception along with this program;
22 // see the files COPYING3 and COPYING.RUNTIME respectively. If not, see
23 // <http://www.gnu.org/licenses/>.
24
25 /**
26 * @file bits/regex_constants.h
27 * @brief Constant definitions for the std regex library.
28 *
29 * This is an internal header file, included by other library headers.
30 * Do not attempt to use it directly. @headername{regex}
31 */
32
_GLIBCXX_VISIBILITY(default)33 namespace std _GLIBCXX_VISIBILITY(default)
34 {
35 /**
36 * @defgroup regex Regular Expressions
37 *
38 * A facility for performing regular expression pattern matching.
39 * @{
40 */
41
42 /**
43 * @namespace std::regex_constants
44 * @brief ISO C++-0x entities sub namespace for regex.
45 */
46 namespace regex_constants
47 {
48 _GLIBCXX_BEGIN_NAMESPACE_VERSION
49
50 /**
51 * @name 5.1 Regular Expression Syntax Options
52 */
53 //@{
54 enum __syntax_option
55 {
56 _S_icase,
57 _S_nosubs,
58 _S_optimize,
59 _S_collate,
60 _S_ECMAScript,
61 _S_basic,
62 _S_extended,
63 _S_awk,
64 _S_grep,
65 _S_egrep,
66 _S_syntax_last
67 };
68
69 /**
70 * @brief This is a bitmask type indicating how to interpret the regex.
71 *
72 * The @c syntax_option_type is implementation defined but it is valid to
73 * perform bitwise operations on these values and expect the right thing to
74 * happen.
75 *
76 * A valid value of type syntax_option_type shall have exactly one of the
77 * elements @c ECMAScript, @c basic, @c extended, @c awk, @c grep, @c egrep
78 * %set.
79 */
80 enum syntax_option_type : unsigned int { };
81
82 /**
83 * Specifies that the matching of regular expressions against a character
84 * sequence shall be performed without regard to case.
85 */
86 constexpr syntax_option_type icase =
87 static_cast<syntax_option_type>(1 << _S_icase);
88
89 /**
90 * Specifies that when a regular expression is matched against a character
91 * container sequence, no sub-expression matches are to be stored in the
92 * supplied match_results structure.
93 */
94 constexpr syntax_option_type nosubs =
95 static_cast<syntax_option_type>(1 << _S_nosubs);
96
97 /**
98 * Specifies that the regular expression engine should pay more attention to
99 * the speed with which regular expressions are matched, and less to the
100 * speed with which regular expression objects are constructed. Otherwise
101 * it has no detectable effect on the program output.
102 */
103 constexpr syntax_option_type optimize =
104 static_cast<syntax_option_type>(1 << _S_optimize);
105
106 /**
107 * Specifies that character ranges of the form [a-b] should be locale
108 * sensitive.
109 */
110 constexpr syntax_option_type collate =
111 static_cast<syntax_option_type>(1 << _S_collate);
112
113 /**
114 * Specifies that the grammar recognized by the regular expression engine is
115 * that used by ECMAScript in ECMA-262 [Ecma International, ECMAScript
116 * Language Specification, Standard Ecma-262, third edition, 1999], as
117 * modified in section [28.13]. This grammar is similar to that defined
118 * in the PERL scripting language but extended with elements found in the
119 * POSIX regular expression grammar.
120 */
121 constexpr syntax_option_type ECMAScript =
122 static_cast<syntax_option_type>(1 << _S_ECMAScript);
123
124 /**
125 * Specifies that the grammar recognized by the regular expression engine is
126 * that used by POSIX basic regular expressions in IEEE Std 1003.1-2001,
127 * Portable Operating System Interface (POSIX), Base Definitions and
128 * Headers, Section 9, Regular Expressions [IEEE, Information Technology --
129 * Portable Operating System Interface (POSIX), IEEE Standard 1003.1-2001].
130 */
131 constexpr syntax_option_type basic =
132 static_cast<syntax_option_type>(1 << _S_basic);
133
134 /**
135 * Specifies that the grammar recognized by the regular expression engine is
136 * that used by POSIX extended regular expressions in IEEE Std 1003.1-2001,
137 * Portable Operating System Interface (POSIX), Base Definitions and
138 * Headers, Section 9, Regular Expressions.
139 */
140 constexpr syntax_option_type extended =
141 static_cast<syntax_option_type>(1 << _S_extended);
142
143 /**
144 * Specifies that the grammar recognized by the regular expression engine is
145 * that used by POSIX utility awk in IEEE Std 1003.1-2001. This option is
146 * identical to syntax_option_type extended, except that C-style escape
147 * sequences are supported. These sequences are:
148 * \\\\, \\a, \\b, \\f, \\n, \\r, \\t , \\v, \\&apos,, &apos,,
149 * and \\ddd (where ddd is one, two, or three octal digits).
150 */
151 constexpr syntax_option_type awk =
152 static_cast<syntax_option_type>(1 << _S_awk);
153
154 /**
155 * Specifies that the grammar recognized by the regular expression engine is
156 * that used by POSIX utility grep in IEEE Std 1003.1-2001. This option is
157 * identical to syntax_option_type basic, except that newlines are treated
158 * as whitespace.
159 */
160 constexpr syntax_option_type grep =
161 static_cast<syntax_option_type>(1 << _S_grep);
162
163 /**
164 * Specifies that the grammar recognized by the regular expression engine is
165 * that used by POSIX utility grep when given the -E option in
166 * IEEE Std 1003.1-2001. This option is identical to syntax_option_type
167 * extended, except that newlines are treated as whitespace.
168 */
169 constexpr syntax_option_type egrep =
170 static_cast<syntax_option_type>(1 << _S_egrep);
171
172 constexpr inline syntax_option_type
173 operator&(syntax_option_type __a, syntax_option_type __b)
174 {
175 return (syntax_option_type)(static_cast<unsigned int>(__a)
176 & static_cast<unsigned int>(__b));
177 }
178
179 constexpr inline syntax_option_type
180 operator|(syntax_option_type __a, syntax_option_type __b)
181 {
182 return (syntax_option_type)(static_cast<unsigned int>(__a)
183 | static_cast<unsigned int>(__b));
184 }
185
186 constexpr inline syntax_option_type
187 operator^(syntax_option_type __a, syntax_option_type __b)
188 {
189 return (syntax_option_type)(static_cast<unsigned int>(__a)
190 ^ static_cast<unsigned int>(__b));
191 }
192
193 constexpr inline syntax_option_type
194 operator~(syntax_option_type __a)
195 { return (syntax_option_type)(~static_cast<unsigned int>(__a)); }
196
197 inline syntax_option_type&
198 operator&=(syntax_option_type& __a, syntax_option_type __b)
199 { return __a = __a & __b; }
200
201 inline syntax_option_type&
202 operator|=(syntax_option_type& __a, syntax_option_type __b)
203 { return __a = __a | __b; }
204
205 inline syntax_option_type&
206 operator^=(syntax_option_type& __a, syntax_option_type __b)
207 { return __a = __a ^ __b; }
208
209 //@}
210
211 /**
212 * @name 5.2 Matching Rules
213 *
214 * Matching a regular expression against a sequence of characters [first,
215 * last) proceeds according to the rules of the grammar specified for the
216 * regular expression object, modified according to the effects listed
217 * below for any bitmask elements set.
218 *
219 */
220 //@{
221
222 enum __match_flag
223 {
224 _S_not_bol,
225 _S_not_eol,
226 _S_not_bow,
227 _S_not_eow,
228 _S_any,
229 _S_not_null,
230 _S_continuous,
231 _S_prev_avail,
232 _S_sed,
233 _S_no_copy,
234 _S_first_only,
235 _S_match_flag_last
236 };
237
238 /**
239 * @brief This is a bitmask type indicating regex matching rules.
240 *
241 * The @c match_flag_type is implementation defined but it is valid to
242 * perform bitwise operations on these values and expect the right thing to
243 * happen.
244 */
245 enum match_flag_type : unsigned int { };
246
247 /**
248 * The default matching rules.
249 */
250 constexpr match_flag_type match_default = static_cast<match_flag_type>(0);
251
252 /**
253 * The first character in the sequence [first, last) is treated as though it
254 * is not at the beginning of a line, so the character (^) in the regular
255 * expression shall not match [first, first).
256 */
257 constexpr match_flag_type match_not_bol =
258 static_cast<match_flag_type>(1 << _S_not_bol);
259
260 /**
261 * The last character in the sequence [first, last) is treated as though it
262 * is not at the end of a line, so the character ($) in the regular
263 * expression shall not match [last, last).
264 */
265 constexpr match_flag_type match_not_eol =
266 static_cast<match_flag_type>(1 << _S_not_eol);
267
268 /**
269 * The expression \\b is not matched against the sub-sequence
270 * [first,first).
271 */
272 constexpr match_flag_type match_not_bow =
273 static_cast<match_flag_type>(1 << _S_not_bow);
274
275 /**
276 * The expression \\b should not be matched against the sub-sequence
277 * [last,last).
278 */
279 constexpr match_flag_type match_not_eow =
280 static_cast<match_flag_type>(1 << _S_not_eow);
281
282 /**
283 * If more than one match is possible then any match is an acceptable
284 * result.
285 */
286 constexpr match_flag_type match_any =
287 static_cast<match_flag_type>(1 << _S_any);
288
289 /**
290 * The expression does not match an empty sequence.
291 */
292 constexpr match_flag_type match_not_null =
293 static_cast<match_flag_type>(1 << _S_not_null);
294
295 /**
296 * The expression only matches a sub-sequence that begins at first .
297 */
298 constexpr match_flag_type match_continuous =
299 static_cast<match_flag_type>(1 << _S_continuous);
300
301 /**
302 * --first is a valid iterator position. When this flag is set then the
303 * flags match_not_bol and match_not_bow are ignored by the regular
304 * expression algorithms 28.11 and iterators 28.12.
305 */
306 constexpr match_flag_type match_prev_avail =
307 static_cast<match_flag_type>(1 << _S_prev_avail);
308
309 /**
310 * When a regular expression match is to be replaced by a new string, the
311 * new string is constructed using the rules used by the ECMAScript replace
312 * function in ECMA- 262 [Ecma International, ECMAScript Language
313 * Specification, Standard Ecma-262, third edition, 1999], part 15.5.4.11
314 * String.prototype.replace. In addition, during search and replace
315 * operations all non-overlapping occurrences of the regular expression
316 * are located and replaced, and sections of the input that did not match
317 * the expression are copied unchanged to the output string.
318 *
319 * Format strings (from ECMA-262 [15.5.4.11]):
320 * @li $$ The dollar-sign itself ($)
321 * @li $& The matched substring.
322 * @li $` The portion of @a string that precedes the matched substring.
323 * This would be match_results::prefix().
324 * @li $' The portion of @a string that follows the matched substring.
325 * This would be match_results::suffix().
326 * @li $n The nth capture, where n is in [1,9] and $n is not followed by a
327 * decimal digit. If n <= match_results::size() and the nth capture
328 * is undefined, use the empty string instead. If n >
329 * match_results::size(), the result is implementation-defined.
330 * @li $nn The nnth capture, where nn is a two-digit decimal number on
331 * [01, 99]. If nn <= match_results::size() and the nth capture is
332 * undefined, use the empty string instead. If
333 * nn > match_results::size(), the result is implementation-defined.
334 */
335 constexpr match_flag_type format_default = static_cast<match_flag_type>(0);
336
337 /**
338 * When a regular expression match is to be replaced by a new string, the
339 * new string is constructed using the rules used by the POSIX sed utility
340 * in IEEE Std 1003.1- 2001 [IEEE, Information Technology -- Portable
341 * Operating System Interface (POSIX), IEEE Standard 1003.1-2001].
342 */
343 constexpr match_flag_type format_sed =
344 static_cast<match_flag_type>(1 << _S_sed);
345
346 /**
347 * During a search and replace operation, sections of the character
348 * container sequence being searched that do not match the regular
349 * expression shall not be copied to the output string.
350 */
351 constexpr match_flag_type format_no_copy =
352 static_cast<match_flag_type>(1 << _S_no_copy);
353
354 /**
355 * When specified during a search and replace operation, only the first
356 * occurrence of the regular expression shall be replaced.
357 */
358 constexpr match_flag_type format_first_only =
359 static_cast<match_flag_type>(1 << _S_first_only);
360
361 constexpr inline match_flag_type
362 operator&(match_flag_type __a, match_flag_type __b)
363 {
364 return (match_flag_type)(static_cast<unsigned int>(__a)
365 & static_cast<unsigned int>(__b));
366 }
367
368 constexpr inline match_flag_type
369 operator|(match_flag_type __a, match_flag_type __b)
370 {
371 return (match_flag_type)(static_cast<unsigned int>(__a)
372 | static_cast<unsigned int>(__b));
373 }
374
375 constexpr inline match_flag_type
376 operator^(match_flag_type __a, match_flag_type __b)
377 {
378 return (match_flag_type)(static_cast<unsigned int>(__a)
379 ^ static_cast<unsigned int>(__b));
380 }
381
382 constexpr inline match_flag_type
383 operator~(match_flag_type __a)
384 { return (match_flag_type)(~static_cast<unsigned int>(__a)); }
385
386 inline match_flag_type&
387 operator&=(match_flag_type& __a, match_flag_type __b)
388 { return __a = __a & __b; }
389
390 inline match_flag_type&
391 operator|=(match_flag_type& __a, match_flag_type __b)
392 { return __a = __a | __b; }
393
394 inline match_flag_type&
395 operator^=(match_flag_type& __a, match_flag_type __b)
396 { return __a = __a ^ __b; }
397
398 //@}
399
400 _GLIBCXX_END_NAMESPACE_VERSION
401 } // namespace regex_constants
402
403 /* @} */ // group regex
404 } // namespace std
405
406