1 // class template regex -*- C++ -*-
2
3 // Copyright (C) 2010-2016 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_polynomial,
67 _S_syntax_last
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 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 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 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 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 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 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 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 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 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 constexpr syntax_option_type egrep =
171 static_cast<syntax_option_type>(1 << _S_egrep);
172
173 /**
174 * Extension: Ensure both space complexity of compiled regex and
175 * time complexity execution are not exponential.
176 * If specified in a regex with back-references, the exception
177 * regex_constants::error_complexity will be thrown.
178 */
179 constexpr syntax_option_type __polynomial =
180 static_cast<syntax_option_type>(1 << _S_polynomial);
181
182 constexpr inline syntax_option_type
183 operator&(syntax_option_type __a, syntax_option_type __b)
184 {
185 return (syntax_option_type)(static_cast<unsigned int>(__a)
186 & static_cast<unsigned int>(__b));
187 }
188
189 constexpr inline syntax_option_type
190 operator|(syntax_option_type __a, syntax_option_type __b)
191 {
192 return (syntax_option_type)(static_cast<unsigned int>(__a)
193 | static_cast<unsigned int>(__b));
194 }
195
196 constexpr inline syntax_option_type
197 operator^(syntax_option_type __a, syntax_option_type __b)
198 {
199 return (syntax_option_type)(static_cast<unsigned int>(__a)
200 ^ static_cast<unsigned int>(__b));
201 }
202
203 constexpr inline syntax_option_type
204 operator~(syntax_option_type __a)
205 { return (syntax_option_type)(~static_cast<unsigned int>(__a)); }
206
207 inline syntax_option_type&
208 operator&=(syntax_option_type& __a, syntax_option_type __b)
209 { return __a = __a & __b; }
210
211 inline syntax_option_type&
212 operator|=(syntax_option_type& __a, syntax_option_type __b)
213 { return __a = __a | __b; }
214
215 inline syntax_option_type&
216 operator^=(syntax_option_type& __a, syntax_option_type __b)
217 { return __a = __a ^ __b; }
218
219 //@}
220
221 /**
222 * @name 5.2 Matching Rules
223 *
224 * Matching a regular expression against a sequence of characters [first,
225 * last) proceeds according to the rules of the grammar specified for the
226 * regular expression object, modified according to the effects listed
227 * below for any bitmask elements set.
228 *
229 */
230 //@{
231
232 enum __match_flag
233 {
234 _S_not_bol,
235 _S_not_eol,
236 _S_not_bow,
237 _S_not_eow,
238 _S_any,
239 _S_not_null,
240 _S_continuous,
241 _S_prev_avail,
242 _S_sed,
243 _S_no_copy,
244 _S_first_only,
245 _S_match_flag_last
246 };
247
248 /**
249 * @brief This is a bitmask type indicating regex matching rules.
250 *
251 * The @c match_flag_type is implementation defined but it is valid to
252 * perform bitwise operations on these values and expect the right thing to
253 * happen.
254 */
255 enum match_flag_type : unsigned int { };
256
257 /**
258 * The default matching rules.
259 */
260 constexpr match_flag_type match_default = static_cast<match_flag_type>(0);
261
262 /**
263 * The first character in the sequence [first, last) is treated as though it
264 * is not at the beginning of a line, so the character (^) in the regular
265 * expression shall not match [first, first).
266 */
267 constexpr match_flag_type match_not_bol =
268 static_cast<match_flag_type>(1 << _S_not_bol);
269
270 /**
271 * The last character in the sequence [first, last) is treated as though it
272 * is not at the end of a line, so the character ($) in the regular
273 * expression shall not match [last, last).
274 */
275 constexpr match_flag_type match_not_eol =
276 static_cast<match_flag_type>(1 << _S_not_eol);
277
278 /**
279 * The expression \\b is not matched against the sub-sequence
280 * [first,first).
281 */
282 constexpr match_flag_type match_not_bow =
283 static_cast<match_flag_type>(1 << _S_not_bow);
284
285 /**
286 * The expression \\b should not be matched against the sub-sequence
287 * [last,last).
288 */
289 constexpr match_flag_type match_not_eow =
290 static_cast<match_flag_type>(1 << _S_not_eow);
291
292 /**
293 * If more than one match is possible then any match is an acceptable
294 * result.
295 */
296 constexpr match_flag_type match_any =
297 static_cast<match_flag_type>(1 << _S_any);
298
299 /**
300 * The expression does not match an empty sequence.
301 */
302 constexpr match_flag_type match_not_null =
303 static_cast<match_flag_type>(1 << _S_not_null);
304
305 /**
306 * The expression only matches a sub-sequence that begins at first .
307 */
308 constexpr match_flag_type match_continuous =
309 static_cast<match_flag_type>(1 << _S_continuous);
310
311 /**
312 * --first is a valid iterator position. When this flag is set then the
313 * flags match_not_bol and match_not_bow are ignored by the regular
314 * expression algorithms 28.11 and iterators 28.12.
315 */
316 constexpr match_flag_type match_prev_avail =
317 static_cast<match_flag_type>(1 << _S_prev_avail);
318
319 /**
320 * When a regular expression match is to be replaced by a new string, the
321 * new string is constructed using the rules used by the ECMAScript replace
322 * function in ECMA- 262 [Ecma International, ECMAScript Language
323 * Specification, Standard Ecma-262, third edition, 1999], part 15.5.4.11
324 * String.prototype.replace. In addition, during search and replace
325 * operations all non-overlapping occurrences of the regular expression
326 * are located and replaced, and sections of the input that did not match
327 * the expression are copied unchanged to the output string.
328 *
329 * Format strings (from ECMA-262 [15.5.4.11]):
330 * @li $$ The dollar-sign itself ($)
331 * @li $& The matched substring.
332 * @li $` The portion of @a string that precedes the matched substring.
333 * This would be match_results::prefix().
334 * @li $' The portion of @a string that follows the matched substring.
335 * This would be match_results::suffix().
336 * @li $n The nth capture, where n is in [1,9] and $n is not followed by a
337 * decimal digit. If n <= match_results::size() and the nth capture
338 * is undefined, use the empty string instead. If n >
339 * match_results::size(), the result is implementation-defined.
340 * @li $nn The nnth capture, where nn is a two-digit decimal number on
341 * [01, 99]. If nn <= match_results::size() and the nth capture is
342 * undefined, use the empty string instead. If
343 * nn > match_results::size(), the result is implementation-defined.
344 */
345 constexpr match_flag_type format_default = static_cast<match_flag_type>(0);
346
347 /**
348 * When a regular expression match is to be replaced by a new string, the
349 * new string is constructed using the rules used by the POSIX sed utility
350 * in IEEE Std 1003.1- 2001 [IEEE, Information Technology -- Portable
351 * Operating System Interface (POSIX), IEEE Standard 1003.1-2001].
352 */
353 constexpr match_flag_type format_sed =
354 static_cast<match_flag_type>(1 << _S_sed);
355
356 /**
357 * During a search and replace operation, sections of the character
358 * container sequence being searched that do not match the regular
359 * expression shall not be copied to the output string.
360 */
361 constexpr match_flag_type format_no_copy =
362 static_cast<match_flag_type>(1 << _S_no_copy);
363
364 /**
365 * When specified during a search and replace operation, only the first
366 * occurrence of the regular expression shall be replaced.
367 */
368 constexpr match_flag_type format_first_only =
369 static_cast<match_flag_type>(1 << _S_first_only);
370
371 constexpr inline match_flag_type
372 operator&(match_flag_type __a, match_flag_type __b)
373 {
374 return (match_flag_type)(static_cast<unsigned int>(__a)
375 & static_cast<unsigned int>(__b));
376 }
377
378 constexpr inline match_flag_type
379 operator|(match_flag_type __a, match_flag_type __b)
380 {
381 return (match_flag_type)(static_cast<unsigned int>(__a)
382 | static_cast<unsigned int>(__b));
383 }
384
385 constexpr inline match_flag_type
386 operator^(match_flag_type __a, match_flag_type __b)
387 {
388 return (match_flag_type)(static_cast<unsigned int>(__a)
389 ^ static_cast<unsigned int>(__b));
390 }
391
392 constexpr inline match_flag_type
393 operator~(match_flag_type __a)
394 { return (match_flag_type)(~static_cast<unsigned int>(__a)); }
395
396 inline match_flag_type&
397 operator&=(match_flag_type& __a, match_flag_type __b)
398 { return __a = __a & __b; }
399
400 inline match_flag_type&
401 operator|=(match_flag_type& __a, match_flag_type __b)
402 { return __a = __a | __b; }
403
404 inline match_flag_type&
405 operator^=(match_flag_type& __a, match_flag_type __b)
406 { return __a = __a ^ __b; }
407
408 //@}
409
410 _GLIBCXX_END_NAMESPACE_VERSION
411 } // namespace regex_constants
412
413 /* @} */ // group regex
414 } // namespace std
415
416