1// class template regex -*- C++ -*-
2
3// Copyright (C) 2010-2022 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 ///@{
54
55 /**
56 * @brief This is a bitmask type indicating how to interpret the regex.
57 *
58 * The @c syntax_option_type is implementation defined but it is valid to
59 * perform bitwise operations on these values and expect the right thing to
60 * happen.
61 *
62 * A valid value of type syntax_option_type shall have exactly one of the
63 * elements @c ECMAScript, @c basic, @c extended, @c awk, @c grep, @c egrep
64 * %set.
65 */
66 enum syntax_option_type : unsigned int
67 {
68 _S_icase = 1 << 0,
69 _S_nosubs = 1 << 1,
70 _S_optimize = 1 << 2,
71 _S_collate = 1 << 3,
72 _S_ECMAScript = 1 << 4,
73 _S_basic = 1 << 5,
74 _S_extended = 1 << 6,
75 _S_awk = 1 << 7,
76 _S_grep = 1 << 8,
77 _S_egrep = 1 << 9,
78 _S_polynomial = 1 << 10,
79 _S_multiline = 1 << 11
80 };
81
82 /**
83 * Specifies that the matching of regular expressions against a character
84 * sequence shall be performed without regard to case.
85 */
86 _GLIBCXX17_INLINE constexpr syntax_option_type icase = _S_icase;
87
88 /**
89 * Specifies that when a regular expression is matched against a character
90 * container sequence, no sub-expression matches are to be stored in the
91 * supplied match_results structure.
92 */
93 _GLIBCXX17_INLINE constexpr syntax_option_type nosubs = _S_nosubs;
94
95 /**
96 * Specifies that the regular expression engine should pay more attention to
97 * the speed with which regular expressions are matched, and less to the
98 * speed with which regular expression objects are constructed. Otherwise
99 * it has no detectable effect on the program output.
100 */
101 _GLIBCXX17_INLINE constexpr syntax_option_type optimize = _S_optimize;
102
103 /**
104 * Specifies that character ranges of the form [a-b] should be locale
105 * sensitive.
106 */
107 _GLIBCXX17_INLINE constexpr syntax_option_type collate = _S_collate;
108
109 /**
110 * Specifies that the grammar recognized by the regular expression engine is
111 * that used by ECMAScript in ECMA-262 [Ecma International, ECMAScript
112 * Language Specification, Standard Ecma-262, third edition, 1999], as
113 * modified in section [28.13]. This grammar is similar to that defined
114 * in the PERL scripting language but extended with elements found in the
115 * POSIX regular expression grammar.
116 */
117 _GLIBCXX17_INLINE constexpr syntax_option_type ECMAScript = _S_ECMAScript;
118
119 /**
120 * Specifies that the grammar recognized by the regular expression engine is
121 * that used by POSIX basic regular expressions in IEEE Std 1003.1-2001,
122 * Portable Operating System Interface (POSIX), Base Definitions and
123 * Headers, Section 9, Regular Expressions [IEEE, Information Technology --
124 * Portable Operating System Interface (POSIX), IEEE Standard 1003.1-2001].
125 */
126 _GLIBCXX17_INLINE constexpr syntax_option_type basic = _S_basic;
127
128 /**
129 * Specifies that the grammar recognized by the regular expression engine is
130 * that used by POSIX extended regular expressions in IEEE Std 1003.1-2001,
131 * Portable Operating System Interface (POSIX), Base Definitions and
132 * Headers, Section 9, Regular Expressions.
133 */
134 _GLIBCXX17_INLINE constexpr syntax_option_type extended = _S_extended;
135
136 /**
137 * Specifies that the grammar recognized by the regular expression engine is
138 * that used by POSIX utility awk in IEEE Std 1003.1-2001. This option is
139 * identical to syntax_option_type extended, except that C-style escape
140 * sequences are supported. These sequences are:
141 * \\\\, \\a, \\b, \\f, \\n, \\r, \\t , \\v, \\&apos,, &apos,,
142 * and \\ddd (where ddd is one, two, or three octal digits).
143 */
144 _GLIBCXX17_INLINE constexpr syntax_option_type awk = _S_awk;
145
146 /**
147 * Specifies that the grammar recognized by the regular expression engine is
148 * that used by POSIX utility grep in IEEE Std 1003.1-2001. This option is
149 * identical to syntax_option_type basic, except that newlines are treated
150 * as whitespace.
151 */
152 _GLIBCXX17_INLINE constexpr syntax_option_type grep = _S_grep;
153
154 /**
155 * Specifies that the grammar recognized by the regular expression engine is
156 * that used by POSIX utility grep when given the -E option in
157 * IEEE Std 1003.1-2001. This option is identical to syntax_option_type
158 * extended, except that newlines are treated as whitespace.
159 */
160 _GLIBCXX17_INLINE constexpr syntax_option_type egrep = _S_egrep;
161
162#if __cplusplus >= 201703L || !defined __STRICT_ANSI__
163 // _GLIBCXX_RESOLVE_LIB_DEFECTS
164 // 2503. multiline option should be added to syntax_option_type
165 /**
166 * Specifies that the `^` anchor matches at the beginning of a line,
167 * and the `$` anchor matches at the end of a line, not only at the
168 * beginning/end of the input.
169 * Valid for the ECMAScript syntax, ignored otherwise.
170 * @since C++17
171 */
172 _GLIBCXX17_INLINE constexpr syntax_option_type multiline = _S_multiline;
173#endif
174
175 /// Extension: Equivalent to regex_constants::multiline for C++11 and C++14.
176 _GLIBCXX17_INLINE constexpr syntax_option_type __multiline = _S_multiline;
177
178 /**
179 * Extension: Ensure both space complexity of compiled regex and
180 * time complexity execution are not exponential.
181 * If specified in a regex with back-references, the exception
182 * regex_constants::error_complexity will be thrown.
183 */
184 _GLIBCXX17_INLINE constexpr syntax_option_type __polynomial = _S_polynomial;
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, syntax_option_type __b)
195 {
196 return (syntax_option_type)(static_cast<unsigned int>(__a)
197 | static_cast<unsigned int>(__b));
198 }
199
200 constexpr inline syntax_option_type
201 operator^(syntax_option_type __a, syntax_option_type __b)
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
208 operator~(syntax_option_type __a)
209 { return (syntax_option_type)(~static_cast<unsigned int>(__a)); }
210
211 _GLIBCXX14_CONSTEXPR
212 inline syntax_option_type&
213 operator&=(syntax_option_type& __a, syntax_option_type __b)
214 { return __a = __a & __b; }
215
216 _GLIBCXX14_CONSTEXPR
217 inline syntax_option_type&
218 operator|=(syntax_option_type& __a, syntax_option_type __b)
219 { return __a = __a | __b; }
220
221 _GLIBCXX14_CONSTEXPR
222 inline syntax_option_type&
223 operator^=(syntax_option_type& __a, syntax_option_type __b)
224 { return __a = __a ^ __b; }
225
226 ///@}
227
228 /**
229 * @name 5.2 Matching Rules
230 *
231 * Matching a regular expression against a sequence of characters [first,
232 * last) proceeds according to the rules of the grammar specified for the
233 * regular expression object, modified according to the effects listed
234 * below for any bitmask elements set.
235 *
236 */
237 ///@{
238
239 /**
240 * @brief This is a bitmask type indicating regex matching rules.
241 *
242 * The @c match_flag_type is implementation defined but it is valid to
243 * perform bitwise operations on these values and expect the right thing to
244 * happen.
245 */
246 enum match_flag_type : unsigned int
247 {
248 _S_default,
249 _S_not_bol = 1 << 0,
250 _S_not_eol = 1 << 1,
251 _S_not_bow = 1 << 2,
252 _S_not_eow = 1 << 3,
253 _S_any = 1 << 4,
254 _S_not_null = 1 << 5,
255 _S_continuous = 1 << 6,
256 _S_prev_avail = 1 << 7,
257 _S_sed = 1 << 8,
258 _S_no_copy = 1 << 9,
259 _S_first_only = 1 << 10,
260 _S_match_flag_last = 1 << 11
261 };
262
263 /**
264 * The default matching rules.
265 */
266 _GLIBCXX17_INLINE constexpr match_flag_type match_default = _S_default;
267
268 /**
269 * The first character in the sequence [first, last) is treated as though it
270 * is not at the beginning of a line, so the character (^) in the regular
271 * expression shall not match [first, first).
272 */
273 _GLIBCXX17_INLINE constexpr match_flag_type match_not_bol = _S_not_bol;
274
275 /**
276 * The last character in the sequence [first, last) is treated as though it
277 * is not at the end of a line, so the character ($) in the regular
278 * expression shall not match [last, last).
279 */
280 _GLIBCXX17_INLINE constexpr match_flag_type match_not_eol = _S_not_eol;
281
282 /**
283 * The expression \\b is not matched against the sub-sequence
284 * [first,first).
285 */
286 _GLIBCXX17_INLINE constexpr match_flag_type match_not_bow = _S_not_bow;
287
288 /**
289 * The expression \\b should not be matched against the sub-sequence
290 * [last,last).
291 */
292 _GLIBCXX17_INLINE constexpr match_flag_type match_not_eow = _S_not_eow;
293
294 /**
295 * If more than one match is possible then any match is an acceptable
296 * result.
297 */
298 _GLIBCXX17_INLINE constexpr match_flag_type match_any = _S_any;
299
300 /**
301 * The expression does not match an empty sequence.
302 */
303 _GLIBCXX17_INLINE constexpr match_flag_type match_not_null = _S_not_null;
304
305 /**
306 * The expression only matches a sub-sequence that begins at first .
307 */
308 _GLIBCXX17_INLINE constexpr match_flag_type match_continuous = _S_continuous;
309
310 /**
311 * `--first` is a valid iterator position. When this flag is set then the
312 * flags `match_not_bol` and `match_not_bow` are ignored by the algorithms
313 * `regex_match`, `regex_search`, and `regex_replace`, and by the iterators
314 * `regex_iterator` and `regex_token_iterator`.
315 */
316 _GLIBCXX17_INLINE constexpr match_flag_type match_prev_avail = _S_prev_avail;
317
318 /**
319 * When a regular expression match is to be replaced by a new string, the
320 * new string is constructed using the rules used by the ECMAScript replace
321 * function in ECMA- 262 [Ecma International, ECMAScript Language
322 * Specification, Standard Ecma-262, third edition, 1999], part 15.5.4.11
323 * String.prototype.replace. In addition, during search and replace
324 * operations all non-overlapping occurrences of the regular expression
325 * are located and replaced, and sections of the input that did not match
326 * the expression are copied unchanged to the output string.
327 *
328 * Format strings (from ECMA-262 [15.5.4.11]):
329 * @li $$ The dollar-sign itself ($)
330 * @li $& The matched substring.
331 * @li $` The portion of @a string that precedes the matched substring.
332 * This would be match_results::prefix().
333 * @li $' The portion of @a string that follows the matched substring.
334 * This would be match_results::suffix().
335 * @li $n The nth capture, where n is in [1,9] and $n is not followed by a
336 * decimal digit. If n <= match_results::size() and the nth capture
337 * is undefined, use the empty string instead. If n >
338 * match_results::size(), the result is implementation-defined.
339 * @li $nn The nnth capture, where nn is a two-digit decimal number on
340 * [01, 99]. If nn <= match_results::size() and the nth capture is
341 * undefined, use the empty string instead. If
342 * nn > match_results::size(), the result is implementation-defined.
343 */
344 _GLIBCXX17_INLINE constexpr match_flag_type format_default = _S_default;
345
346 /**
347 * When a regular expression match is to be replaced by a new string, the
348 * new string is constructed using the rules used by the POSIX sed utility
349 * in IEEE Std 1003.1- 2001 [IEEE, Information Technology -- Portable
350 * Operating System Interface (POSIX), IEEE Standard 1003.1-2001].
351 */
352 _GLIBCXX17_INLINE constexpr match_flag_type format_sed = _S_sed;
353
354 /**
355 * During a search and replace operation, sections of the character
356 * container sequence being searched that do not match the regular
357 * expression shall not be copied to the output string.
358 */
359 _GLIBCXX17_INLINE constexpr match_flag_type format_no_copy = _S_no_copy;
360
361 /**
362 * When specified during a search and replace operation, only the first
363 * occurrence of the regular expression shall be replaced.
364 */
365 _GLIBCXX17_INLINE constexpr match_flag_type format_first_only = _S_first_only;
366
367 constexpr inline match_flag_type
368 operator&(match_flag_type __a, match_flag_type __b)
369 {
370 return (match_flag_type)(static_cast<unsigned int>(__a)
371 & static_cast<unsigned int>(__b));
372 }
373
374 constexpr inline match_flag_type
375 operator|(match_flag_type __a, match_flag_type __b)
376 {
377 return (match_flag_type)(static_cast<unsigned int>(__a)
378 | static_cast<unsigned int>(__b));
379 }
380
381 constexpr inline match_flag_type
382 operator^(match_flag_type __a, match_flag_type __b)
383 {
384 return (match_flag_type)(static_cast<unsigned int>(__a)
385 ^ static_cast<unsigned int>(__b));
386 }
387
388 constexpr inline match_flag_type
389 operator~(match_flag_type __a)
390 { return (match_flag_type)(~static_cast<unsigned int>(__a)); }
391
392 _GLIBCXX14_CONSTEXPR
393 inline match_flag_type&
394 operator&=(match_flag_type& __a, match_flag_type __b)
395 { return __a = __a & __b; }
396
397 _GLIBCXX14_CONSTEXPR
398 inline match_flag_type&
399 operator|=(match_flag_type& __a, match_flag_type __b)
400 { return __a = __a | __b; }
401
402 _GLIBCXX14_CONSTEXPR
403 inline match_flag_type&
404 operator^=(match_flag_type& __a, match_flag_type __b)
405 { return __a = __a ^ __b; }
406
407 ///@}
408} // namespace regex_constants
409/// @} group regex
410
411_GLIBCXX_END_NAMESPACE_VERSION
412} // namespace std
413
414