xref: /netbsd-src/external/gpl3/gcc/dist/libstdc++-v3/include/bits/regex_constants.h (revision 0a3071956a3a9fdebdbf7f338cf2d439b45fc728)
1 // Namespace std::regex_constants -*- 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 
_GLIBCXX_VISIBILITY(default)33 namespace 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  * @since C++11
43  *
44  * @{
45  */
46 
47 /**
48  * @namespace std::regex_constants
49  * @brief ISO C++ 2011 namespace for options and flags used with std::regex
50  */
51 namespace regex_constants
52 {
53   /**
54    * @name 5.1 Regular Expression Syntax Options
55    */
56   ///@{
57 
58   /**
59    * @brief This is a bitmask type indicating how to interpret the regex.
60    *
61    * The @c syntax_option_type is implementation defined but it is valid to
62    * perform bitwise operations on these values and expect the right thing to
63    * happen.
64    *
65    * A valid value of type syntax_option_type shall have exactly one of the
66    * elements @c ECMAScript, @c basic, @c extended, @c awk, @c grep, @c egrep
67    * %set.
68    */
69   enum syntax_option_type : unsigned int
70   {
71     _S_icase		= 1 << 0,
72     _S_nosubs		= 1 << 1,
73     _S_optimize		= 1 << 2,
74     _S_collate		= 1 << 3,
75     _S_ECMAScript	= 1 << 4,
76     _S_basic		= 1 << 5,
77     _S_extended		= 1 << 6,
78     _S_awk		= 1 << 7,
79     _S_grep		= 1 << 8,
80     _S_egrep		= 1 << 9,
81     _S_polynomial	= 1 << 10,
82     _S_multiline	= 1 << 11
83   };
84 
85   /**
86    * Specifies that the matching of regular expressions against a character
87    * sequence shall be performed without regard to case.
88    */
89   _GLIBCXX17_INLINE constexpr syntax_option_type icase = _S_icase;
90 
91   /**
92    * Specifies that when a regular expression is matched against a character
93    * container sequence, no sub-expression matches are to be stored in the
94    * supplied match_results structure.
95    */
96   _GLIBCXX17_INLINE constexpr syntax_option_type nosubs = _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 = _S_optimize;
105 
106   /**
107    * Specifies that character ranges of the form [a-b] should be locale
108    * sensitive.
109    */
110   _GLIBCXX17_INLINE constexpr syntax_option_type collate = _S_collate;
111 
112   /**
113    * Specifies that the grammar recognized by the regular expression engine is
114    * that used by ECMAScript in ECMA-262 [Ecma International, ECMAScript
115    * Language Specification, Standard Ecma-262, third edition, 1999], as
116    * modified in section [28.13].  This grammar is similar to that defined
117    * in the PERL scripting language but extended with elements found in the
118    * POSIX regular expression grammar.
119    */
120   _GLIBCXX17_INLINE constexpr syntax_option_type ECMAScript = _S_ECMAScript;
121 
122   /**
123    * Specifies that the grammar recognized by the regular expression engine is
124    * that used by POSIX basic regular expressions in IEEE Std 1003.1-2001,
125    * Portable Operating System Interface (POSIX), Base Definitions and
126    * Headers, Section 9, Regular Expressions [IEEE, Information Technology --
127    * Portable Operating System Interface (POSIX), IEEE Standard 1003.1-2001].
128    */
129   _GLIBCXX17_INLINE constexpr syntax_option_type basic = _S_basic;
130 
131   /**
132    * Specifies that the grammar recognized by the regular expression engine is
133    * that used by POSIX extended regular expressions in IEEE Std 1003.1-2001,
134    * Portable Operating System Interface (POSIX), Base Definitions and
135    * Headers, Section 9, Regular Expressions.
136    */
137   _GLIBCXX17_INLINE constexpr syntax_option_type extended = _S_extended;
138 
139   /**
140    * Specifies that the grammar recognized by the regular expression engine is
141    * that used by POSIX utility awk in IEEE Std 1003.1-2001.  This option is
142    * identical to syntax_option_type extended, except that C-style escape
143    * sequences are supported.  These sequences are:
144    * \\\\, \\a, \\b, \\f, \\n, \\r, \\t , \\v, \\&apos,, &apos,,
145    * and \\ddd (where ddd is one, two, or three octal digits).
146    */
147   _GLIBCXX17_INLINE constexpr syntax_option_type awk = _S_awk;
148 
149   /**
150    * Specifies that the grammar recognized by the regular expression engine is
151    * that used by POSIX utility grep in IEEE Std 1003.1-2001.  This option is
152    * identical to syntax_option_type basic, except that newlines are treated
153    * as whitespace.
154    */
155   _GLIBCXX17_INLINE constexpr syntax_option_type grep = _S_grep;
156 
157   /**
158    * Specifies that the grammar recognized by the regular expression engine is
159    * that used by POSIX utility grep when given the -E option in
160    * IEEE Std 1003.1-2001.  This option is identical to syntax_option_type
161    * extended, except that newlines are treated as whitespace.
162    */
163   _GLIBCXX17_INLINE constexpr syntax_option_type egrep = _S_egrep;
164 
165 #if __cplusplus >= 201703L || !defined __STRICT_ANSI__
166   // _GLIBCXX_RESOLVE_LIB_DEFECTS
167   // 2503. multiline option should be added to syntax_option_type
168   /**
169    * Specifies that the `^` anchor matches at the beginning of a line,
170    * and the `$` anchor matches at the end of a line, not only at the
171    * beginning/end of the input.
172    * Valid for the ECMAScript syntax, ignored otherwise.
173    * @since C++17
174    */
175   _GLIBCXX17_INLINE constexpr syntax_option_type multiline = _S_multiline;
176 #endif
177 
178   /// Extension: Equivalent to regex_constants::multiline for C++11 and C++14.
179   _GLIBCXX17_INLINE constexpr syntax_option_type __multiline = _S_multiline;
180 
181   /**
182    * Extension: Ensure both space complexity of compiled regex and
183    * time complexity execution are not exponential.
184    * If specified in a regex with back-references, the exception
185    * regex_constants::error_complexity will be thrown.
186    */
187   _GLIBCXX17_INLINE constexpr syntax_option_type __polynomial = _S_polynomial;
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, syntax_option_type __b)
205   {
206     return (syntax_option_type)(static_cast<unsigned int>(__a)
207 				^ static_cast<unsigned int>(__b));
208   }
209 
210   constexpr inline syntax_option_type
211   operator~(syntax_option_type __a)
212   { return (syntax_option_type)(~static_cast<unsigned int>(__a)); }
213 
214   _GLIBCXX14_CONSTEXPR
215   inline syntax_option_type&
216   operator&=(syntax_option_type& __a, syntax_option_type __b)
217   { return __a = __a & __b; }
218 
219   _GLIBCXX14_CONSTEXPR
220   inline syntax_option_type&
221   operator|=(syntax_option_type& __a, syntax_option_type __b)
222   { return __a = __a | __b; }
223 
224   _GLIBCXX14_CONSTEXPR
225   inline syntax_option_type&
226   operator^=(syntax_option_type& __a, syntax_option_type __b)
227   { return __a = __a ^ __b; }
228 
229   ///@}
230 
231   /**
232    * @name 5.2 Matching Rules
233    *
234    * Matching a regular expression against a sequence of characters [first,
235    * last) proceeds according to the rules of the grammar specified for the
236    * regular expression object, modified according to the effects listed
237    * below for any bitmask elements set.
238    *
239    */
240   ///@{
241 
242   /**
243    * @brief This is a bitmask type indicating regex matching rules.
244    *
245    * The @c match_flag_type is implementation defined but it is valid to
246    * perform bitwise operations on these values and expect the right thing to
247    * happen.
248    */
249   enum match_flag_type : unsigned int
250   {
251     _S_default,
252     _S_not_bol		= 1 << 0,
253     _S_not_eol		= 1 << 1,
254     _S_not_bow		= 1 << 2,
255     _S_not_eow		= 1 << 3,
256     _S_any		= 1 << 4,
257     _S_not_null		= 1 << 5,
258     _S_continuous	= 1 << 6,
259     _S_prev_avail	= 1 << 7,
260     _S_sed		= 1 << 8,
261     _S_no_copy		= 1 << 9,
262     _S_first_only	= 1 << 10,
263     _S_match_flag_last	= 1 << 11
264   };
265 
266   /**
267    * The default matching rules.
268    */
269   _GLIBCXX17_INLINE constexpr match_flag_type match_default = _S_default;
270 
271   /**
272    * The first character in the sequence [first, last) is treated as though it
273    * is not at the beginning of a line, so the character (^) in the regular
274    * expression shall not match [first, first).
275    */
276   _GLIBCXX17_INLINE constexpr match_flag_type match_not_bol = _S_not_bol;
277 
278   /**
279    * The last character in the sequence [first, last) is treated as though it
280    * is not at the end of a line, so the character ($) in the regular
281    * expression shall not match [last, last).
282    */
283   _GLIBCXX17_INLINE constexpr match_flag_type match_not_eol = _S_not_eol;
284 
285   /**
286    * The expression \\b is not matched against the sub-sequence
287    * [first,first).
288    */
289   _GLIBCXX17_INLINE constexpr match_flag_type match_not_bow = _S_not_bow;
290 
291   /**
292    * The expression \\b should not be matched against the sub-sequence
293    * [last,last).
294    */
295   _GLIBCXX17_INLINE constexpr match_flag_type match_not_eow = _S_not_eow;
296 
297   /**
298    * If more than one match is possible then any match is an acceptable
299    * result.
300    */
301   _GLIBCXX17_INLINE constexpr match_flag_type match_any = _S_any;
302 
303   /**
304    * The expression does not match an empty sequence.
305    */
306   _GLIBCXX17_INLINE constexpr match_flag_type match_not_null = _S_not_null;
307 
308   /**
309    * The expression only matches a sub-sequence that begins at first .
310    */
311   _GLIBCXX17_INLINE constexpr match_flag_type match_continuous = _S_continuous;
312 
313   /**
314    * `--first` is a valid iterator position.  When this flag is set then the
315    * flags `match_not_bol` and `match_not_bow` are ignored by the algorithms
316    * `regex_match`, `regex_search`, and `regex_replace`, and by the iterators
317    * `regex_iterator` and `regex_token_iterator`.
318    */
319   _GLIBCXX17_INLINE constexpr match_flag_type match_prev_avail = _S_prev_avail;
320 
321   /**
322    * When a regular expression match is to be replaced by a new string, the
323    * new string is constructed using the rules used by the ECMAScript replace
324    * function in ECMA- 262 [Ecma International, ECMAScript Language
325    * Specification, Standard Ecma-262, third edition, 1999], part 15.5.4.11
326    * String.prototype.replace. In addition, during search and replace
327    * operations all non-overlapping occurrences of the regular expression
328    * are located and replaced, and sections of the input that did not match
329    * the expression are copied unchanged to the output string.
330    *
331    * Format strings (from ECMA-262 [15.5.4.11]):
332    * @li $$  The dollar-sign itself ($)
333    * @li $&  The matched substring.
334    * @li $`  The portion of @a string that precedes the matched substring.
335    *         This would be match_results::prefix().
336    * @li $'  The portion of @a string that follows the matched substring.
337    *         This would be match_results::suffix().
338    * @li $n  The nth capture, where n is in [1,9] and $n is not followed by a
339    *         decimal digit.  If n <= match_results::size() and the nth capture
340    *         is undefined, use the empty string instead.  If n >
341    *         match_results::size(), the result is implementation-defined.
342    * @li $nn The nnth capture, where nn is a two-digit decimal number on
343    *         [01, 99].  If nn <= match_results::size() and the nth capture is
344    *         undefined, use the empty string instead. If
345    *         nn > match_results::size(), the result is implementation-defined.
346    */
347   _GLIBCXX17_INLINE constexpr match_flag_type format_default = _S_default;
348 
349   /**
350    * When a regular expression match is to be replaced by a new string, the
351    * new string is constructed using the rules used by the POSIX sed utility
352    * in IEEE Std 1003.1- 2001 [IEEE, Information Technology -- Portable
353    * Operating System Interface (POSIX), IEEE Standard 1003.1-2001].
354    */
355   _GLIBCXX17_INLINE constexpr match_flag_type format_sed = _S_sed;
356 
357   /**
358    * During a search and replace operation, sections of the character
359    * container sequence being searched that do not match the regular
360    * expression shall not be copied to the output string.
361    */
362   _GLIBCXX17_INLINE constexpr match_flag_type format_no_copy = _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   _GLIBCXX17_INLINE constexpr match_flag_type format_first_only = _S_first_only;
369 
370   constexpr inline match_flag_type
371   operator&(match_flag_type __a, match_flag_type __b)
372   {
373     return (match_flag_type)(static_cast<unsigned int>(__a)
374 				& static_cast<unsigned int>(__b));
375   }
376 
377   constexpr inline match_flag_type
378   operator|(match_flag_type __a, match_flag_type __b)
379   {
380     return (match_flag_type)(static_cast<unsigned int>(__a)
381 				| static_cast<unsigned int>(__b));
382   }
383 
384   constexpr inline match_flag_type
385   operator^(match_flag_type __a, match_flag_type __b)
386   {
387     return (match_flag_type)(static_cast<unsigned int>(__a)
388 				^ static_cast<unsigned int>(__b));
389   }
390 
391   constexpr inline match_flag_type
392   operator~(match_flag_type __a)
393   { return (match_flag_type)(~static_cast<unsigned int>(__a)); }
394 
395   _GLIBCXX14_CONSTEXPR
396   inline match_flag_type&
397   operator&=(match_flag_type& __a, match_flag_type __b)
398   { return __a = __a & __b; }
399 
400   _GLIBCXX14_CONSTEXPR
401   inline match_flag_type&
402   operator|=(match_flag_type& __a, match_flag_type __b)
403   { return __a = __a | __b; }
404 
405   _GLIBCXX14_CONSTEXPR
406   inline match_flag_type&
407   operator^=(match_flag_type& __a, match_flag_type __b)
408   { return __a = __a ^ __b; }
409 
410   ///@}
411 } // namespace regex_constants
412 /// @} group regex
413 
414 _GLIBCXX_END_NAMESPACE_VERSION
415 } // namespace std
416 
417