libstdc++
regex_constants.h
Go to the documentation of this file.
1// class template regex -*- C++ -*-
2
3// Copyright (C) 2010-2021 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
33namespace std _GLIBCXX_VISIBILITY(default)
34{
35_GLIBCXX_BEGIN_NAMESPACE_VERSION
36
37/**
38 * @defgroup regex Regular Expressions
39 *
40 * A facility for performing regular expression pattern matching.
41 * @{
42 */
43
44/**
45 * @namespace std::regex_constants
46 * @brief ISO C++ 2011 namespace for options and flags used with std::regex
47 */
48namespace regex_constants
49{
50 /**
51 * @name 5.1 Regular Expression Syntax Options
52 */
53 ///@{
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_polynomial,
67 _S_multiline
68 };
69
70 /**
71 * @brief This is a bitmask type indicating how to interpret the regex.
72 *
73 * The @c syntax_option_type is implementation defined but it is valid to
74 * perform bitwise operations on these values and expect the right thing to
75 * happen.
76 *
77 * A valid value of type syntax_option_type shall have exactly one of the
78 * elements @c ECMAScript, @c basic, @c extended, @c awk, @c grep, @c egrep
79 * %set.
80 */
81 enum syntax_option_type : unsigned int { };
82
83 /**
84 * Specifies that the matching of regular expressions against a character
85 * sequence shall be performed without regard to case.
86 */
87 _GLIBCXX17_INLINE constexpr syntax_option_type icase =
88 static_cast<syntax_option_type>(1 << _S_icase);
89
90 /**
91 * Specifies that when a regular expression is matched against a character
92 * container sequence, no sub-expression matches are to be stored in the
93 * supplied match_results structure.
94 */
95 _GLIBCXX17_INLINE constexpr syntax_option_type nosubs =
96 static_cast<syntax_option_type>(1 << _S_nosubs);
97
98 /**
99 * Specifies that the regular expression engine should pay more attention to
100 * the speed with which regular expressions are matched, and less to the
101 * speed with which regular expression objects are constructed. Otherwise
102 * it has no detectable effect on the program output.
103 */
104 _GLIBCXX17_INLINE constexpr syntax_option_type optimize =
105 static_cast<syntax_option_type>(1 << _S_optimize);
106
107 /**
108 * Specifies that character ranges of the form [a-b] should be locale
109 * sensitive.
110 */
111 _GLIBCXX17_INLINE constexpr syntax_option_type collate =
112 static_cast<syntax_option_type>(1 << _S_collate);
113
114 /**
115 * Specifies that the grammar recognized by the regular expression engine is
116 * that used by ECMAScript in ECMA-262 [Ecma International, ECMAScript
117 * Language Specification, Standard Ecma-262, third edition, 1999], as
118 * modified in section [28.13]. This grammar is similar to that defined
119 * in the PERL scripting language but extended with elements found in the
120 * POSIX regular expression grammar.
121 */
122 _GLIBCXX17_INLINE constexpr syntax_option_type ECMAScript =
123 static_cast<syntax_option_type>(1 << _S_ECMAScript);
124
125 /**
126 * Specifies that the grammar recognized by the regular expression engine is
127 * that used by POSIX basic regular expressions in IEEE Std 1003.1-2001,
128 * Portable Operating System Interface (POSIX), Base Definitions and
129 * Headers, Section 9, Regular Expressions [IEEE, Information Technology --
130 * Portable Operating System Interface (POSIX), IEEE Standard 1003.1-2001].
131 */
132 _GLIBCXX17_INLINE constexpr syntax_option_type basic =
133 static_cast<syntax_option_type>(1 << _S_basic);
134
135 /**
136 * Specifies that the grammar recognized by the regular expression engine is
137 * that used by POSIX extended regular expressions in IEEE Std 1003.1-2001,
138 * Portable Operating System Interface (POSIX), Base Definitions and
139 * Headers, Section 9, Regular Expressions.
140 */
141 _GLIBCXX17_INLINE constexpr syntax_option_type extended =
142 static_cast<syntax_option_type>(1 << _S_extended);
143
144 /**
145 * Specifies that the grammar recognized by the regular expression engine is
146 * that used by POSIX utility awk in IEEE Std 1003.1-2001. This option is
147 * identical to syntax_option_type extended, except that C-style escape
148 * sequences are supported. These sequences are:
149 * \\\\, \\a, \\b, \\f, \\n, \\r, \\t , \\v, \\&apos,, &apos,,
150 * and \\ddd (where ddd is one, two, or three octal digits).
151 */
152 _GLIBCXX17_INLINE constexpr syntax_option_type awk =
153 static_cast<syntax_option_type>(1 << _S_awk);
154
155 /**
156 * Specifies that the grammar recognized by the regular expression engine is
157 * that used by POSIX utility grep in IEEE Std 1003.1-2001. This option is
158 * identical to syntax_option_type basic, except that newlines are treated
159 * as whitespace.
160 */
161 _GLIBCXX17_INLINE constexpr syntax_option_type grep =
162 static_cast<syntax_option_type>(1 << _S_grep);
163
164 /**
165 * Specifies that the grammar recognized by the regular expression engine is
166 * that used by POSIX utility grep when given the -E option in
167 * IEEE Std 1003.1-2001. This option is identical to syntax_option_type
168 * extended, except that newlines are treated as whitespace.
169 */
170 _GLIBCXX17_INLINE constexpr syntax_option_type egrep =
171 static_cast<syntax_option_type>(1 << _S_egrep);
172
173#if __cplusplus >= 201703L || !defined __STRICT_ANSI__
174 // _GLIBCXX_RESOLVE_LIB_DEFECTS
175 // 2503. multiline option should be added to syntax_option_type
176 /**
177 * Specifies that the `^` anchor matches at the beginning of a line,
178 * and the `$` anchor matches at the end of a line, not only at the
179 * beginning/end of the input.
180 * Valid for the ECMAScript syntax, ignored otherwise.
181 * @since C++17
182 */
183 _GLIBCXX17_INLINE constexpr syntax_option_type multiline =
184 static_cast<syntax_option_type>(1 << _S_multiline);
185#endif
186
187 /// Extension: Equivalent to regex_constants::multiline for C++11 and C++14.
188 _GLIBCXX17_INLINE constexpr syntax_option_type __multiline =
189 static_cast<syntax_option_type>(1 << _S_multiline);
190
191 /**
192 * Extension: Ensure both space complexity of compiled regex and
193 * time complexity execution are not exponential.
194 * If specified in a regex with back-references, the exception
195 * regex_constants::error_complexity will be thrown.
196 */
197 _GLIBCXX17_INLINE constexpr syntax_option_type __polynomial =
198 static_cast<syntax_option_type>(1 << _S_polynomial);
199
200 constexpr inline syntax_option_type
202 {
203 return (syntax_option_type)(static_cast<unsigned int>(__a)
204 & static_cast<unsigned int>(__b));
205 }
206
207 constexpr inline syntax_option_type
209 {
210 return (syntax_option_type)(static_cast<unsigned int>(__a)
211 | static_cast<unsigned int>(__b));
212 }
213
214 constexpr inline syntax_option_type
216 {
217 return (syntax_option_type)(static_cast<unsigned int>(__a)
218 ^ static_cast<unsigned int>(__b));
219 }
220
221 constexpr inline syntax_option_type
223 { return (syntax_option_type)(~static_cast<unsigned int>(__a)); }
224
225 inline syntax_option_type&
227 { return __a = __a & __b; }
228
229 inline syntax_option_type&
231 { return __a = __a | __b; }
232
233 inline syntax_option_type&
235 { return __a = __a ^ __b; }
236
237 ///@}
238
239 /**
240 * @name 5.2 Matching Rules
241 *
242 * Matching a regular expression against a sequence of characters [first,
243 * last) proceeds according to the rules of the grammar specified for the
244 * regular expression object, modified according to the effects listed
245 * below for any bitmask elements set.
246 *
247 */
248 ///@{
249
251 {
252 _S_not_bol,
253 _S_not_eol,
254 _S_not_bow,
255 _S_not_eow,
256 _S_any,
257 _S_not_null,
258 _S_continuous,
259 _S_prev_avail,
260 _S_sed,
261 _S_no_copy,
262 _S_first_only,
263 _S_match_flag_last
264 };
265
266 /**
267 * @brief This is a bitmask type indicating regex matching rules.
268 *
269 * The @c match_flag_type is implementation defined but it is valid to
270 * perform bitwise operations on these values and expect the right thing to
271 * happen.
272 */
273 enum match_flag_type : unsigned int { };
274
275 /**
276 * The default matching rules.
277 */
278 _GLIBCXX17_INLINE constexpr match_flag_type match_default =
279 static_cast<match_flag_type>(0);
280
281 /**
282 * The first character in the sequence [first, last) is treated as though it
283 * is not at the beginning of a line, so the character (^) in the regular
284 * expression shall not match [first, first).
285 */
286 _GLIBCXX17_INLINE constexpr match_flag_type match_not_bol =
287 static_cast<match_flag_type>(1 << _S_not_bol);
288
289 /**
290 * The last character in the sequence [first, last) is treated as though it
291 * is not at the end of a line, so the character ($) in the regular
292 * expression shall not match [last, last).
293 */
294 _GLIBCXX17_INLINE constexpr match_flag_type match_not_eol =
295 static_cast<match_flag_type>(1 << _S_not_eol);
296
297 /**
298 * The expression \\b is not matched against the sub-sequence
299 * [first,first).
300 */
301 _GLIBCXX17_INLINE constexpr match_flag_type match_not_bow =
302 static_cast<match_flag_type>(1 << _S_not_bow);
303
304 /**
305 * The expression \\b should not be matched against the sub-sequence
306 * [last,last).
307 */
308 _GLIBCXX17_INLINE constexpr match_flag_type match_not_eow =
309 static_cast<match_flag_type>(1 << _S_not_eow);
310
311 /**
312 * If more than one match is possible then any match is an acceptable
313 * result.
314 */
315 _GLIBCXX17_INLINE constexpr match_flag_type match_any =
316 static_cast<match_flag_type>(1 << _S_any);
317
318 /**
319 * The expression does not match an empty sequence.
320 */
321 _GLIBCXX17_INLINE constexpr match_flag_type match_not_null =
322 static_cast<match_flag_type>(1 << _S_not_null);
323
324 /**
325 * The expression only matches a sub-sequence that begins at first .
326 */
327 _GLIBCXX17_INLINE constexpr match_flag_type match_continuous =
328 static_cast<match_flag_type>(1 << _S_continuous);
329
330 /**
331 * `--first` is a valid iterator position. When this flag is set then the
332 * flags `match_not_bol` and `match_not_bow` are ignored by the algorithms
333 * `regex_match`, `regex_search`, and `regex_replace`, and by the iterators
334 * `regex_iterator` and `regex_token_iterator`.
335 */
336 _GLIBCXX17_INLINE constexpr match_flag_type match_prev_avail =
337 static_cast<match_flag_type>(1 << _S_prev_avail);
338
339 /**
340 * When a regular expression match is to be replaced by a new string, the
341 * new string is constructed using the rules used by the ECMAScript replace
342 * function in ECMA- 262 [Ecma International, ECMAScript Language
343 * Specification, Standard Ecma-262, third edition, 1999], part 15.5.4.11
344 * String.prototype.replace. In addition, during search and replace
345 * operations all non-overlapping occurrences of the regular expression
346 * are located and replaced, and sections of the input that did not match
347 * the expression are copied unchanged to the output string.
348 *
349 * Format strings (from ECMA-262 [15.5.4.11]):
350 * @li $$ The dollar-sign itself ($)
351 * @li $& The matched substring.
352 * @li $` The portion of @a string that precedes the matched substring.
353 * This would be match_results::prefix().
354 * @li $' The portion of @a string that follows the matched substring.
355 * This would be match_results::suffix().
356 * @li $n The nth capture, where n is in [1,9] and $n is not followed by a
357 * decimal digit. If n <= match_results::size() and the nth capture
358 * is undefined, use the empty string instead. If n >
359 * match_results::size(), the result is implementation-defined.
360 * @li $nn The nnth capture, where nn is a two-digit decimal number on
361 * [01, 99]. If nn <= match_results::size() and the nth capture is
362 * undefined, use the empty string instead. If
363 * nn > match_results::size(), the result is implementation-defined.
364 */
365 _GLIBCXX17_INLINE constexpr match_flag_type format_default =
366 static_cast<match_flag_type>(0);
367
368 /**
369 * When a regular expression match is to be replaced by a new string, the
370 * new string is constructed using the rules used by the POSIX sed utility
371 * in IEEE Std 1003.1- 2001 [IEEE, Information Technology -- Portable
372 * Operating System Interface (POSIX), IEEE Standard 1003.1-2001].
373 */
374 _GLIBCXX17_INLINE constexpr match_flag_type format_sed =
375 static_cast<match_flag_type>(1 << _S_sed);
376
377 /**
378 * During a search and replace operation, sections of the character
379 * container sequence being searched that do not match the regular
380 * expression shall not be copied to the output string.
381 */
382 _GLIBCXX17_INLINE constexpr match_flag_type format_no_copy =
383 static_cast<match_flag_type>(1 << _S_no_copy);
384
385 /**
386 * When specified during a search and replace operation, only the first
387 * occurrence of the regular expression shall be replaced.
388 */
389 _GLIBCXX17_INLINE constexpr match_flag_type format_first_only =
390 static_cast<match_flag_type>(1 << _S_first_only);
391
392 constexpr inline match_flag_type
394 {
395 return (match_flag_type)(static_cast<unsigned int>(__a)
396 & static_cast<unsigned int>(__b));
397 }
398
399 constexpr inline match_flag_type
401 {
402 return (match_flag_type)(static_cast<unsigned int>(__a)
403 | static_cast<unsigned int>(__b));
404 }
405
406 constexpr inline match_flag_type
408 {
409 return (match_flag_type)(static_cast<unsigned int>(__a)
410 ^ static_cast<unsigned int>(__b));
411 }
412
413 constexpr inline match_flag_type
415 { return (match_flag_type)(~static_cast<unsigned int>(__a)); }
416
417 inline match_flag_type&
419 { return __a = __a & __b; }
420
421 inline match_flag_type&
423 { return __a = __a | __b; }
424
425 inline match_flag_type&
427 { return __a = __a ^ __b; }
428
429 ///@}
430} // namespace regex_constants
431/// @} group regex
432
433_GLIBCXX_END_NAMESPACE_VERSION
434} // namespace std
435
ISO C++ entities toplevel namespace is std.
constexpr syntax_option_type __polynomial
constexpr syntax_option_type operator&(syntax_option_type __a, syntax_option_type __b)
This is a bitmask type indicating how to interpret the regex.
syntax_option_type & operator&=(syntax_option_type &__a, syntax_option_type __b)
This is a bitmask type indicating how to interpret the regex.
__match_flag
This is a bitmask type indicating regex matching rules.
constexpr match_flag_type match_not_bow
constexpr match_flag_type match_not_bol
constexpr syntax_option_type ECMAScript
constexpr match_flag_type match_any
constexpr syntax_option_type egrep
syntax_option_type
This is a bitmask type indicating how to interpret the regex.
constexpr syntax_option_type multiline
__syntax_option
This is a bitmask type indicating how to interpret the regex.
constexpr match_flag_type match_default
constexpr match_flag_type format_first_only
syntax_option_type & operator^=(syntax_option_type &__a, syntax_option_type __b)
This is a bitmask type indicating how to interpret the regex.
constexpr syntax_option_type awk
constexpr syntax_option_type __multiline
Extension: Equivalent to regex_constants::multiline for C++11 and C++14.
constexpr match_flag_type match_continuous
constexpr syntax_option_type extended
constexpr syntax_option_type operator^(syntax_option_type __a, syntax_option_type __b)
This is a bitmask type indicating how to interpret the regex.
constexpr syntax_option_type basic
constexpr match_flag_type match_not_eol
match_flag_type
This is a bitmask type indicating regex matching rules.
syntax_option_type & operator|=(syntax_option_type &__a, syntax_option_type __b)
This is a bitmask type indicating how to interpret the regex.
constexpr syntax_option_type icase
constexpr syntax_option_type optimize
constexpr match_flag_type format_default
constexpr syntax_option_type nosubs
constexpr syntax_option_type operator~(syntax_option_type __a)
This is a bitmask type indicating how to interpret the regex.
constexpr match_flag_type match_prev_avail
constexpr match_flag_type format_sed
constexpr syntax_option_type operator|(syntax_option_type __a, syntax_option_type __b)
This is a bitmask type indicating how to interpret the regex.
constexpr match_flag_type match_not_eow
constexpr match_flag_type match_not_null
constexpr match_flag_type format_no_copy
constexpr syntax_option_type grep
Facet for localized string comparison.