xref: /netbsd-src/external/gpl3/gcc/dist/libstdc++-v3/include/bits/regex_scanner.tcc (revision 0a3071956a3a9fdebdbf7f338cf2d439b45fc728)
1 // class template regex -*- C++ -*-
2 
3 // Copyright (C) 2013-2022 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_scanner.tcc
27  *  This is an internal header file, included by other library headers.
28  *  Do not attempt to use it directly. @headername{regex}
29  */
30 
31 // FIXME make comments doxygen format.
32 
33 // N3376 specified 6 regex styles: ECMAScript, basic, extended, grep, egrep
34 // and awk
35 // 1) grep is basic except '\n' is treated as '|'
36 // 2) egrep is extended except '\n' is treated as '|'
37 // 3) awk is extended except special escaping rules, and there's no
38 //    back-reference.
39 //
40 // References:
41 //
42 // ECMAScript: ECMA-262 15.10
43 //
44 // basic, extended:
45 // http://pubs.opengroup.org/onlinepubs/009695399/basedefs/xbd_chap09.html
46 //
47 // awk: http://pubs.opengroup.org/onlinepubs/000095399/utilities/awk.html
48 
49 namespace std _GLIBCXX_VISIBILITY(default)
50 {
51 _GLIBCXX_BEGIN_NAMESPACE_VERSION
52 
53 namespace __detail
54 {
55   template<typename _CharT>
56     _Scanner<_CharT>::
_Scanner(const _CharT * __begin,const _CharT * __end,_FlagT __flags,std::locale __loc)57     _Scanner(const _CharT* __begin, const _CharT* __end,
58 	     _FlagT __flags, std::locale __loc)
59     : _ScannerBase(__flags),
60       _M_current(__begin), _M_end(__end),
61       _M_ctype(std::use_facet<_CtypeT>(__loc)),
62       _M_eat_escape(_M_is_ecma()
63 		    ? &_Scanner::_M_eat_escape_ecma
64 		    : &_Scanner::_M_eat_escape_posix)
65     { _M_advance(); }
66 
67   template<typename _CharT>
68     void
69     _Scanner<_CharT>::
_M_advance()70     _M_advance()
71     {
72       if (_M_current == _M_end)
73 	{
74 	  _M_token = _S_token_eof;
75 	  return;
76 	}
77 
78       if (_M_state == _S_state_normal)
79 	_M_scan_normal();
80       else if (_M_state == _S_state_in_bracket)
81 	_M_scan_in_bracket();
82       else if (_M_state == _S_state_in_brace)
83 	_M_scan_in_brace();
84       else
85 	{
86 	  __glibcxx_assert(!"unexpected state while processing regex");
87 	}
88     }
89 
90   // Differences between styles:
91   // 1) "\(", "\)", "\{" in basic. It's not escaping.
92   // 2) "(?:", "(?=", "(?!" in ECMAScript.
93   template<typename _CharT>
94     void
95     _Scanner<_CharT>::
_M_scan_normal()96     _M_scan_normal()
97     {
98       auto __c = *_M_current++;
99 
100       if (__builtin_strchr(_M_spec_char, _M_ctype.narrow(__c, ' ')) == nullptr)
101 	{
102 	  _M_token = _S_token_ord_char;
103 	  _M_value.assign(1, __c);
104 	  return;
105 	}
106       if (__c == '\\')
107 	{
108 	  if (_M_current == _M_end)
109 	    __throw_regex_error(
110 	      regex_constants::error_escape,
111 	      "Invalid escape at end of regular expression");
112 
113 	  if (!_M_is_basic()
114 	      || (*_M_current != '('
115 		  && *_M_current != ')'
116 		  && *_M_current != '{'))
117 	    {
118 	      (this->*_M_eat_escape)();
119 	      return;
120 	    }
121 	  __c = *_M_current++;
122 	}
123       if (__c == '(')
124 	{
125 	  if (_M_is_ecma() && *_M_current == '?')
126 	    {
127 	      if (++_M_current == _M_end)
128 		__throw_regex_error(regex_constants::error_paren);
129 
130 	      if (*_M_current == ':')
131 		{
132 		  ++_M_current;
133 		  _M_token = _S_token_subexpr_no_group_begin;
134 		}
135 	      else if (*_M_current == '=')
136 		{
137 		  ++_M_current;
138 		  _M_token = _S_token_subexpr_lookahead_begin;
139 		  _M_value.assign(1, 'p');
140 		}
141 	      else if (*_M_current == '!')
142 		{
143 		  ++_M_current;
144 		  _M_token = _S_token_subexpr_lookahead_begin;
145 		  _M_value.assign(1, 'n');
146 		}
147 	      else
148 		__throw_regex_error(regex_constants::error_paren,
149 				    "Invalid '(?...)' zero-width assertion "
150 				    "in regular expression");
151 	    }
152 	  else if (_M_flags & regex_constants::nosubs)
153 	    _M_token = _S_token_subexpr_no_group_begin;
154 	  else
155 	    _M_token = _S_token_subexpr_begin;
156 	}
157       else if (__c == ')')
158 	_M_token = _S_token_subexpr_end;
159       else if (__c == '[')
160 	{
161 	  _M_state = _S_state_in_bracket;
162 	  _M_at_bracket_start = true;
163 	  if (_M_current != _M_end && *_M_current == '^')
164 	    {
165 	      _M_token = _S_token_bracket_neg_begin;
166 	      ++_M_current;
167 	    }
168 	  else
169 	    _M_token = _S_token_bracket_begin;
170 	}
171       else if (__c == '{')
172 	{
173 	  _M_state = _S_state_in_brace;
174 	  _M_token = _S_token_interval_begin;
175 	}
176       else if (__builtin_expect(__c == _CharT(0), false))
177 	{
178 	  if (!_M_is_ecma())
179 	    __throw_regex_error(regex_constants::_S_null);
180 	  _M_token = _S_token_ord_char;
181 	  _M_value.assign(1, __c);
182 	}
183       else if (__c != ']' && __c != '}')
184 	{
185 	  auto __it = _M_token_tbl;
186 	  auto __narrowc = _M_ctype.narrow(__c, '\0');
187 	  for (; __it->first != '\0'; ++__it)
188 	    if (__it->first == __narrowc)
189 	      {
190 		_M_token = __it->second;
191 		return;
192 	      }
193 	  __glibcxx_assert(!"unexpected special character in regex");
194 	}
195       else
196 	{
197 	  _M_token = _S_token_ord_char;
198 	  _M_value.assign(1, __c);
199 	}
200     }
201 
202   // Differences between styles:
203   // 1) different semantics of "[]" and "[^]".
204   // 2) Escaping in bracket expr.
205   template<typename _CharT>
206     void
207     _Scanner<_CharT>::
_M_scan_in_bracket()208     _M_scan_in_bracket()
209     {
210       if (_M_current == _M_end)
211 	__throw_regex_error(regex_constants::error_brack);
212 
213       auto __c = *_M_current++;
214 
215       if (__c == '-')
216 	_M_token = _S_token_bracket_dash;
217       else if (__c == '[')
218 	{
219 	  if (_M_current == _M_end)
220 	    __throw_regex_error(regex_constants::error_brack,
221 				"Incomplete '[[' character class in "
222 				"regular expression");
223 
224 	  if (*_M_current == '.')
225 	    {
226 	      _M_token = _S_token_collsymbol;
227 	      _M_eat_class(*_M_current++);
228 	    }
229 	  else if (*_M_current == ':')
230 	    {
231 	      _M_token = _S_token_char_class_name;
232 	      _M_eat_class(*_M_current++);
233 	    }
234 	  else if (*_M_current == '=')
235 	    {
236 	      _M_token = _S_token_equiv_class_name;
237 	      _M_eat_class(*_M_current++);
238 	    }
239 	  else
240 	    {
241 	      _M_token = _S_token_ord_char;
242 	      _M_value.assign(1, __c);
243 	    }
244 	}
245       // In POSIX, when encountering "[]" or "[^]", the ']' is interpreted
246       // literally. So "[]]" and "[^]]" are valid regexes. See the testcases
247       // `.../empty_range.cc`.
248       else if (__c == ']' && (_M_is_ecma() || !_M_at_bracket_start))
249 	{
250 	  _M_token = _S_token_bracket_end;
251 	  _M_state = _S_state_normal;
252 	}
253       // ECMAScript and awk permits escaping in bracket.
254       else if (__c == '\\' && (_M_is_ecma() || _M_is_awk()))
255 	(this->*_M_eat_escape)();
256       else
257 	{
258 	  _M_token = _S_token_ord_char;
259 	  _M_value.assign(1, __c);
260 	}
261       _M_at_bracket_start = false;
262     }
263 
264   // Differences between styles:
265   // 1) "\}" in basic style.
266   template<typename _CharT>
267     void
268     _Scanner<_CharT>::
_M_scan_in_brace()269     _M_scan_in_brace()
270     {
271       if (_M_current == _M_end)
272 	__throw_regex_error(regex_constants::error_brace);
273 
274       auto __c = *_M_current++;
275 
276       if (_M_ctype.is(_CtypeT::digit, __c))
277 	{
278 	  _M_token = _S_token_dup_count;
279 	  _M_value.assign(1, __c);
280 	  while (_M_current != _M_end
281 		 && _M_ctype.is(_CtypeT::digit, *_M_current))
282 	    _M_value += *_M_current++;
283 	}
284       else if (__c == ',')
285 	_M_token = _S_token_comma;
286       // basic use \}.
287       else if (_M_is_basic())
288 	{
289 	  if (__c == '\\' && _M_current != _M_end && *_M_current == '}')
290 	    {
291 	      _M_state = _S_state_normal;
292 	      _M_token = _S_token_interval_end;
293 	      ++_M_current;
294 	    }
295 	  else
296 	    __throw_regex_error(regex_constants::error_badbrace);
297 	}
298       else if (__c == '}')
299 	{
300 	  _M_state = _S_state_normal;
301 	  _M_token = _S_token_interval_end;
302 	}
303       else
304 	__throw_regex_error(regex_constants::error_badbrace);
305     }
306 
307   template<typename _CharT>
308     void
309     _Scanner<_CharT>::
_M_eat_escape_ecma()310     _M_eat_escape_ecma()
311     {
312       if (_M_current == _M_end)
313 	__throw_regex_error(regex_constants::error_escape);
314 
315       auto __c = *_M_current++;
316       auto __pos = _M_find_escape(_M_ctype.narrow(__c, '\0'));
317 
318       if (__pos != nullptr && (__c != 'b' || _M_state == _S_state_in_bracket))
319 	{
320 	  _M_token = _S_token_ord_char;
321 	  _M_value.assign(1, *__pos);
322 	}
323       else if (__c == 'b')
324 	{
325 	  _M_token = _S_token_word_bound;
326 	  _M_value.assign(1, 'p');
327 	}
328       else if (__c == 'B')
329 	{
330 	  _M_token = _S_token_word_bound;
331 	  _M_value.assign(1, 'n');
332 	}
333       // N3376 28.13
334       else if (__c == 'd'
335 	       || __c == 'D'
336 	       || __c == 's'
337 	       || __c == 'S'
338 	       || __c == 'w'
339 	       || __c == 'W')
340 	{
341 	  _M_token = _S_token_quoted_class;
342 	  _M_value.assign(1, __c);
343 	}
344       else if (__c == 'c')
345 	{
346 	  if (_M_current == _M_end)
347 	    __throw_regex_error(regex_constants::error_escape,
348 				"invalid '\\cX' control character in "
349 				"regular expression");
350 	  _M_token = _S_token_ord_char;
351 	  _M_value.assign(1, *_M_current++);
352 	}
353       else if (__c == 'x' || __c == 'u')
354 	{
355 	  _M_value.clear();
356 	  const int __n = __c == 'x' ? 2 : 4;
357 	  for (int __i = 0; __i < __n; __i++)
358 	    {
359 	      if (_M_current == _M_end
360 		  || !_M_ctype.is(_CtypeT::xdigit, *_M_current))
361 		__throw_regex_error(regex_constants::error_escape,
362 				    __n == 2
363 				    ? "Invalid '\\xNN' control character in "
364 				      "regular expression"
365 				    : "Invalid '\\uNNNN' control character in "
366 				      "regular expression");
367 	      _M_value += *_M_current++;
368 	    }
369 	  _M_token = _S_token_hex_num;
370 	}
371       // ECMAScript recognizes multi-digit back-references.
372       else if (_M_ctype.is(_CtypeT::digit, __c))
373 	{
374 	  _M_value.assign(1, __c);
375 	  while (_M_current != _M_end
376 		 && _M_ctype.is(_CtypeT::digit, *_M_current))
377 	    _M_value += *_M_current++;
378 	  _M_token = _S_token_backref;
379 	}
380       else
381 	{
382 	  _M_token = _S_token_ord_char;
383 	  _M_value.assign(1, __c);
384 	}
385     }
386 
387   // Differences between styles:
388   // 1) Extended doesn't support backref, but basic does.
389   template<typename _CharT>
390     void
391     _Scanner<_CharT>::
_M_eat_escape_posix()392     _M_eat_escape_posix()
393     {
394       if (_M_current == _M_end)
395 	__throw_regex_error(regex_constants::error_escape);
396 
397       auto __c = *_M_current;
398       auto __pos = __builtin_strchr(_M_spec_char, _M_ctype.narrow(__c, '\0'));
399 
400       if (__pos != nullptr && *__pos != '\0')
401 	{
402 	  _M_token = _S_token_ord_char;
403 	  _M_value.assign(1, __c);
404 	}
405       // We MUST judge awk before handling backrefs. There's no backref in awk.
406       else if (_M_is_awk())
407 	{
408 	  _M_eat_escape_awk();
409 	  return;
410 	}
411       else if (_M_is_basic() && _M_ctype.is(_CtypeT::digit, __c) && __c != '0')
412 	{
413 	  _M_token = _S_token_backref;
414 	  _M_value.assign(1, __c);
415 	}
416       else
417 	{
418 #ifdef __STRICT_ANSI__
419 	  // POSIX says it is undefined to escape ordinary characters
420 	  __throw_regex_error(regex_constants::error_escape);
421 #else
422 	  _M_token = _S_token_ord_char;
423 	  _M_value.assign(1, __c);
424 #endif
425 	}
426       ++_M_current;
427     }
428 
429   template<typename _CharT>
430     void
431     _Scanner<_CharT>::
_M_eat_escape_awk()432     _M_eat_escape_awk()
433     {
434       auto __c = *_M_current++;
435       auto __pos = _M_find_escape(_M_ctype.narrow(__c, '\0'));
436 
437       if (__pos != nullptr)
438 	{
439 	  _M_token = _S_token_ord_char;
440 	  _M_value.assign(1, *__pos);
441 	}
442       // \ddd for oct representation
443       else if (_M_ctype.is(_CtypeT::digit, __c)
444 	       && __c != '8'
445 	       && __c != '9')
446 	{
447 	  _M_value.assign(1,  __c);
448 	  for (int __i = 0;
449 	       __i < 2
450 	       && _M_current != _M_end
451 	       && _M_ctype.is(_CtypeT::digit, *_M_current)
452 	       && *_M_current != '8'
453 	       && *_M_current != '9';
454 	       __i++)
455 	    _M_value += *_M_current++;
456 	  _M_token = _S_token_oct_num;
457 	  return;
458 	}
459       else
460 	__throw_regex_error(regex_constants::error_escape);
461     }
462 
463   // Eats a character class or throws an exception.
464   // __ch could be ':', '.' or '=', _M_current is the char after ']' when
465   // returning.
466   template<typename _CharT>
467     void
468     _Scanner<_CharT>::
_M_eat_class(char __ch)469     _M_eat_class(char __ch)
470     {
471       for (_M_value.clear(); _M_current != _M_end && *_M_current != __ch;)
472 	_M_value += *_M_current++;
473       if (_M_current == _M_end
474 	  || *_M_current++ != __ch
475 	  || _M_current == _M_end // skip __ch
476 	  || *_M_current++ != ']') // skip ']'
477 	{
478 	  __throw_regex_error(__ch == ':' ? regex_constants::error_ctype
479 					  : regex_constants::error_collate);
480 	}
481     }
482 
483 #ifdef _GLIBCXX_DEBUG
484   template<typename _CharT>
485     std::ostream&
486     _Scanner<_CharT>::
_M_print(std::ostream & __ostr)487     _M_print(std::ostream& __ostr)
488     {
489       switch (_M_token)
490       {
491       case _S_token_anychar:
492 	__ostr << "any-character\n";
493 	break;
494       case _S_token_backref:
495 	__ostr << "backref\n";
496 	break;
497       case _S_token_bracket_begin:
498 	__ostr << "bracket-begin\n";
499 	break;
500       case _S_token_bracket_neg_begin:
501 	__ostr << "bracket-neg-begin\n";
502 	break;
503       case _S_token_bracket_end:
504 	__ostr << "bracket-end\n";
505 	break;
506       case _S_token_char_class_name:
507 	__ostr << "char-class-name \"" << _M_value << "\"\n";
508 	break;
509       case _S_token_closure0:
510 	__ostr << "closure0\n";
511 	break;
512       case _S_token_closure1:
513 	__ostr << "closure1\n";
514 	break;
515       case _S_token_collsymbol:
516 	__ostr << "collsymbol \"" << _M_value << "\"\n";
517 	break;
518       case _S_token_comma:
519 	__ostr << "comma\n";
520 	break;
521       case _S_token_dup_count:
522 	__ostr << "dup count: " << _M_value << "\n";
523 	break;
524       case _S_token_eof:
525 	__ostr << "EOF\n";
526 	break;
527       case _S_token_equiv_class_name:
528 	__ostr << "equiv-class-name \"" << _M_value << "\"\n";
529 	break;
530       case _S_token_interval_begin:
531 	__ostr << "interval begin\n";
532 	break;
533       case _S_token_interval_end:
534 	__ostr << "interval end\n";
535 	break;
536       case _S_token_line_begin:
537 	__ostr << "line begin\n";
538 	break;
539       case _S_token_line_end:
540 	__ostr << "line end\n";
541 	break;
542       case _S_token_opt:
543 	__ostr << "opt\n";
544 	break;
545       case _S_token_or:
546 	__ostr << "or\n";
547 	break;
548       case _S_token_ord_char:
549 	__ostr << "ordinary character: \"" << _M_value << "\"\n";
550 	break;
551       case _S_token_subexpr_begin:
552 	__ostr << "subexpr begin\n";
553 	break;
554       case _S_token_subexpr_no_group_begin:
555 	__ostr << "no grouping subexpr begin\n";
556 	break;
557       case _S_token_subexpr_lookahead_begin:
558 	__ostr << "lookahead subexpr begin\n";
559 	break;
560       case _S_token_subexpr_end:
561 	__ostr << "subexpr end\n";
562 	break;
563       case _S_token_unknown:
564 	__ostr << "-- unknown token --\n";
565 	break;
566       case _S_token_oct_num:
567 	__ostr << "oct number " << _M_value << "\n";
568 	break;
569       case _S_token_hex_num:
570 	__ostr << "hex number " << _M_value << "\n";
571 	break;
572       case _S_token_quoted_class:
573 	__ostr << "quoted class " << "\\" << _M_value << "\n";
574 	break;
575       default:
576 	_GLIBCXX_DEBUG_ASSERT(false);
577       }
578       return __ostr;
579     }
580 #endif
581 
582 } // namespace __detail
583 _GLIBCXX_END_NAMESPACE_VERSION
584 } // namespace
585