|
1 // Copyright (c) 2005, Google Inc. |
|
2 // All rights reserved. |
|
3 // |
|
4 // Redistribution and use in source and binary forms, with or without |
|
5 // modification, are permitted provided that the following conditions are |
|
6 // met: |
|
7 // |
|
8 // * Redistributions of source code must retain the above copyright |
|
9 // notice, this list of conditions and the following disclaimer. |
|
10 // * Redistributions in binary form must reproduce the above |
|
11 // copyright notice, this list of conditions and the following disclaimer |
|
12 // in the documentation and/or other materials provided with the |
|
13 // distribution. |
|
14 // * Neither the name of Google Inc. nor the names of its |
|
15 // contributors may be used to endorse or promote products derived from |
|
16 // this software without specific prior written permission. |
|
17 // |
|
18 // THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS |
|
19 // "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT |
|
20 // LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR |
|
21 // A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT |
|
22 // OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, |
|
23 // SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT |
|
24 // LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, |
|
25 // DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY |
|
26 // THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT |
|
27 // (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE |
|
28 // OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. |
|
29 // |
|
30 // Author: Sanjay Ghemawat |
|
31 // Support for PCRE_XXX modifiers added by Giuseppe Maxia, July 2005 |
|
32 |
|
33 #ifndef _PCRECPP_H |
|
34 #define _PCRECPP_H |
|
35 |
|
36 // C++ interface to the pcre regular-expression library. RE supports |
|
37 // Perl-style regular expressions (with extensions like \d, \w, \s, |
|
38 // ...). |
|
39 // |
|
40 // ----------------------------------------------------------------------- |
|
41 // REGEXP SYNTAX: |
|
42 // |
|
43 // This module is part of the pcre library and hence supports its syntax |
|
44 // for regular expressions. |
|
45 // |
|
46 // The syntax is pretty similar to Perl's. For those not familiar |
|
47 // with Perl's regular expressions, here are some examples of the most |
|
48 // commonly used extensions: |
|
49 // |
|
50 // "hello (\\w+) world" -- \w matches a "word" character |
|
51 // "version (\\d+)" -- \d matches a digit |
|
52 // "hello\\s+world" -- \s matches any whitespace character |
|
53 // "\\b(\\w+)\\b" -- \b matches empty string at a word boundary |
|
54 // "(?i)hello" -- (?i) turns on case-insensitive matching |
|
55 // "/\\*(.*?)\\*/" -- .*? matches . minimum no. of times possible |
|
56 // |
|
57 // ----------------------------------------------------------------------- |
|
58 // MATCHING INTERFACE: |
|
59 // |
|
60 // The "FullMatch" operation checks that supplied text matches a |
|
61 // supplied pattern exactly. |
|
62 // |
|
63 // Example: successful match |
|
64 // pcrecpp::RE re("h.*o"); |
|
65 // re.FullMatch("hello"); |
|
66 // |
|
67 // Example: unsuccessful match (requires full match): |
|
68 // pcrecpp::RE re("e"); |
|
69 // !re.FullMatch("hello"); |
|
70 // |
|
71 // Example: creating a temporary RE object: |
|
72 // pcrecpp::RE("h.*o").FullMatch("hello"); |
|
73 // |
|
74 // You can pass in a "const char*" or a "string" for "text". The |
|
75 // examples below tend to use a const char*. |
|
76 // |
|
77 // You can, as in the different examples above, store the RE object |
|
78 // explicitly in a variable or use a temporary RE object. The |
|
79 // examples below use one mode or the other arbitrarily. Either |
|
80 // could correctly be used for any of these examples. |
|
81 // |
|
82 // ----------------------------------------------------------------------- |
|
83 // MATCHING WITH SUB-STRING EXTRACTION: |
|
84 // |
|
85 // You can supply extra pointer arguments to extract matched subpieces. |
|
86 // |
|
87 // Example: extracts "ruby" into "s" and 1234 into "i" |
|
88 // int i; |
|
89 // string s; |
|
90 // pcrecpp::RE re("(\\w+):(\\d+)"); |
|
91 // re.FullMatch("ruby:1234", &s, &i); |
|
92 // |
|
93 // Example: does not try to extract any extra sub-patterns |
|
94 // re.FullMatch("ruby:1234", &s); |
|
95 // |
|
96 // Example: does not try to extract into NULL |
|
97 // re.FullMatch("ruby:1234", NULL, &i); |
|
98 // |
|
99 // Example: integer overflow causes failure |
|
100 // !re.FullMatch("ruby:1234567891234", NULL, &i); |
|
101 // |
|
102 // Example: fails because there aren't enough sub-patterns: |
|
103 // !pcrecpp::RE("\\w+:\\d+").FullMatch("ruby:1234", &s); |
|
104 // |
|
105 // Example: fails because string cannot be stored in integer |
|
106 // !pcrecpp::RE("(.*)").FullMatch("ruby", &i); |
|
107 // |
|
108 // The provided pointer arguments can be pointers to any scalar numeric |
|
109 // type, or one of |
|
110 // string (matched piece is copied to string) |
|
111 // StringPiece (StringPiece is mutated to point to matched piece) |
|
112 // T (where "bool T::ParseFrom(const char*, int)" exists) |
|
113 // NULL (the corresponding matched sub-pattern is not copied) |
|
114 // |
|
115 // CAVEAT: An optional sub-pattern that does not exist in the matched |
|
116 // string is assigned the empty string. Therefore, the following will |
|
117 // return false (because the empty string is not a valid number): |
|
118 // int number; |
|
119 // pcrecpp::RE::FullMatch("abc", "[a-z]+(\\d+)?", &number); |
|
120 // |
|
121 // ----------------------------------------------------------------------- |
|
122 // DO_MATCH |
|
123 // |
|
124 // The matching interface supports at most 16 arguments per call. |
|
125 // If you need more, consider using the more general interface |
|
126 // pcrecpp::RE::DoMatch(). See pcrecpp.h for the signature for DoMatch. |
|
127 // |
|
128 // ----------------------------------------------------------------------- |
|
129 // PARTIAL MATCHES |
|
130 // |
|
131 // You can use the "PartialMatch" operation when you want the pattern |
|
132 // to match any substring of the text. |
|
133 // |
|
134 // Example: simple search for a string: |
|
135 // pcrecpp::RE("ell").PartialMatch("hello"); |
|
136 // |
|
137 // Example: find first number in a string: |
|
138 // int number; |
|
139 // pcrecpp::RE re("(\\d+)"); |
|
140 // re.PartialMatch("x*100 + 20", &number); |
|
141 // assert(number == 100); |
|
142 // |
|
143 // ----------------------------------------------------------------------- |
|
144 // UTF-8 AND THE MATCHING INTERFACE: |
|
145 // |
|
146 // By default, pattern and text are plain text, one byte per character. |
|
147 // The UTF8 flag, passed to the constructor, causes both pattern |
|
148 // and string to be treated as UTF-8 text, still a byte stream but |
|
149 // potentially multiple bytes per character. In practice, the text |
|
150 // is likelier to be UTF-8 than the pattern, but the match returned |
|
151 // may depend on the UTF8 flag, so always use it when matching |
|
152 // UTF8 text. E.g., "." will match one byte normally but with UTF8 |
|
153 // set may match up to three bytes of a multi-byte character. |
|
154 // |
|
155 // Example: |
|
156 // pcrecpp::RE_Options options; |
|
157 // options.set_utf8(); |
|
158 // pcrecpp::RE re(utf8_pattern, options); |
|
159 // re.FullMatch(utf8_string); |
|
160 // |
|
161 // Example: using the convenience function UTF8(): |
|
162 // pcrecpp::RE re(utf8_pattern, pcrecpp::UTF8()); |
|
163 // re.FullMatch(utf8_string); |
|
164 // |
|
165 // NOTE: The UTF8 option is ignored if pcre was not configured with the |
|
166 // --enable-utf8 flag. |
|
167 // |
|
168 // ----------------------------------------------------------------------- |
|
169 // PASSING MODIFIERS TO THE REGULAR EXPRESSION ENGINE |
|
170 // |
|
171 // PCRE defines some modifiers to change the behavior of the regular |
|
172 // expression engine. |
|
173 // The C++ wrapper defines an auxiliary class, RE_Options, as a vehicle |
|
174 // to pass such modifiers to a RE class. |
|
175 // |
|
176 // Currently, the following modifiers are supported |
|
177 // |
|
178 // modifier description Perl corresponding |
|
179 // |
|
180 // PCRE_CASELESS case insensitive match /i |
|
181 // PCRE_MULTILINE multiple lines match /m |
|
182 // PCRE_DOTALL dot matches newlines /s |
|
183 // PCRE_DOLLAR_ENDONLY $ matches only at end N/A |
|
184 // PCRE_EXTRA strict escape parsing N/A |
|
185 // PCRE_EXTENDED ignore whitespaces /x |
|
186 // PCRE_UTF8 handles UTF8 chars built-in |
|
187 // PCRE_UNGREEDY reverses * and *? N/A |
|
188 // PCRE_NO_AUTO_CAPTURE disables matching parens N/A (*) |
|
189 // |
|
190 // (For a full account on how each modifier works, please check the |
|
191 // PCRE API reference manual). |
|
192 // |
|
193 // (*) Both Perl and PCRE allow non matching parentheses by means of the |
|
194 // "?:" modifier within the pattern itself. e.g. (?:ab|cd) does not |
|
195 // capture, while (ab|cd) does. |
|
196 // |
|
197 // For each modifier, there are two member functions whose name is made |
|
198 // out of the modifier in lowercase, without the "PCRE_" prefix. For |
|
199 // instance, PCRE_CASELESS is handled by |
|
200 // bool caseless(), |
|
201 // which returns true if the modifier is set, and |
|
202 // RE_Options & set_caseless(bool), |
|
203 // which sets or unsets the modifier. |
|
204 // |
|
205 // Moreover, PCRE_EXTRA_MATCH_LIMIT can be accessed through the |
|
206 // set_match_limit() and match_limit() member functions. |
|
207 // Setting match_limit to a non-zero value will limit the executation of |
|
208 // pcre to keep it from doing bad things like blowing the stack or taking |
|
209 // an eternity to return a result. A value of 5000 is good enough to stop |
|
210 // stack blowup in a 2MB thread stack. Setting match_limit to zero will |
|
211 // disable match limiting. Alternately, you can set match_limit_recursion() |
|
212 // which uses PCRE_EXTRA_MATCH_LIMIT_RECURSION to limit how much pcre |
|
213 // recurses. match_limit() caps the number of matches pcre does; |
|
214 // match_limit_recrusion() caps the depth of recursion. |
|
215 // |
|
216 // Normally, to pass one or more modifiers to a RE class, you declare |
|
217 // a RE_Options object, set the appropriate options, and pass this |
|
218 // object to a RE constructor. Example: |
|
219 // |
|
220 // RE_options opt; |
|
221 // opt.set_caseless(true); |
|
222 // |
|
223 // if (RE("HELLO", opt).PartialMatch("hello world")) ... |
|
224 // |
|
225 // RE_options has two constructors. The default constructor takes no |
|
226 // arguments and creates a set of flags that are off by default. |
|
227 // |
|
228 // The optional parameter 'option_flags' is to facilitate transfer |
|
229 // of legacy code from C programs. This lets you do |
|
230 // RE(pattern, RE_Options(PCRE_CASELESS|PCRE_MULTILINE)).PartialMatch(str); |
|
231 // |
|
232 // But new code is better off doing |
|
233 // RE(pattern, |
|
234 // RE_Options().set_caseless(true).set_multiline(true)).PartialMatch(str); |
|
235 // (See below) |
|
236 // |
|
237 // If you are going to pass one of the most used modifiers, there are some |
|
238 // convenience functions that return a RE_Options class with the |
|
239 // appropriate modifier already set: |
|
240 // CASELESS(), UTF8(), MULTILINE(), DOTALL(), EXTENDED() |
|
241 // |
|
242 // If you need to set several options at once, and you don't want to go |
|
243 // through the pains of declaring a RE_Options object and setting several |
|
244 // options, there is a parallel method that give you such ability on the |
|
245 // fly. You can concatenate several set_xxxxx member functions, since each |
|
246 // of them returns a reference to its class object. e.g.: to pass |
|
247 // PCRE_CASELESS, PCRE_EXTENDED, and PCRE_MULTILINE to a RE with one |
|
248 // statement, you may write |
|
249 // |
|
250 // RE(" ^ xyz \\s+ .* blah$", RE_Options() |
|
251 // .set_caseless(true) |
|
252 // .set_extended(true) |
|
253 // .set_multiline(true)).PartialMatch(sometext); |
|
254 // |
|
255 // ----------------------------------------------------------------------- |
|
256 // SCANNING TEXT INCREMENTALLY |
|
257 // |
|
258 // The "Consume" operation may be useful if you want to repeatedly |
|
259 // match regular expressions at the front of a string and skip over |
|
260 // them as they match. This requires use of the "StringPiece" type, |
|
261 // which represents a sub-range of a real string. Like RE, StringPiece |
|
262 // is defined in the pcrecpp namespace. |
|
263 // |
|
264 // Example: read lines of the form "var = value" from a string. |
|
265 // string contents = ...; // Fill string somehow |
|
266 // pcrecpp::StringPiece input(contents); // Wrap in a StringPiece |
|
267 // |
|
268 // string var; |
|
269 // int value; |
|
270 // pcrecpp::RE re("(\\w+) = (\\d+)\n"); |
|
271 // while (re.Consume(&input, &var, &value)) { |
|
272 // ...; |
|
273 // } |
|
274 // |
|
275 // Each successful call to "Consume" will set "var/value", and also |
|
276 // advance "input" so it points past the matched text. |
|
277 // |
|
278 // The "FindAndConsume" operation is similar to "Consume" but does not |
|
279 // anchor your match at the beginning of the string. For example, you |
|
280 // could extract all words from a string by repeatedly calling |
|
281 // pcrecpp::RE("(\\w+)").FindAndConsume(&input, &word) |
|
282 // |
|
283 // ----------------------------------------------------------------------- |
|
284 // PARSING HEX/OCTAL/C-RADIX NUMBERS |
|
285 // |
|
286 // By default, if you pass a pointer to a numeric value, the |
|
287 // corresponding text is interpreted as a base-10 number. You can |
|
288 // instead wrap the pointer with a call to one of the operators Hex(), |
|
289 // Octal(), or CRadix() to interpret the text in another base. The |
|
290 // CRadix operator interprets C-style "0" (base-8) and "0x" (base-16) |
|
291 // prefixes, but defaults to base-10. |
|
292 // |
|
293 // Example: |
|
294 // int a, b, c, d; |
|
295 // pcrecpp::RE re("(.*) (.*) (.*) (.*)"); |
|
296 // re.FullMatch("100 40 0100 0x40", |
|
297 // pcrecpp::Octal(&a), pcrecpp::Hex(&b), |
|
298 // pcrecpp::CRadix(&c), pcrecpp::CRadix(&d)); |
|
299 // will leave 64 in a, b, c, and d. |
|
300 // |
|
301 // ----------------------------------------------------------------------- |
|
302 // REPLACING PARTS OF STRINGS |
|
303 // |
|
304 // You can replace the first match of "pattern" in "str" with |
|
305 // "rewrite". Within "rewrite", backslash-escaped digits (\1 to \9) |
|
306 // can be used to insert text matching corresponding parenthesized |
|
307 // group from the pattern. \0 in "rewrite" refers to the entire |
|
308 // matching text. E.g., |
|
309 // |
|
310 // string s = "yabba dabba doo"; |
|
311 // pcrecpp::RE("b+").Replace("d", &s); |
|
312 // |
|
313 // will leave "s" containing "yada dabba doo". The result is true if |
|
314 // the pattern matches and a replacement occurs, or false otherwise. |
|
315 // |
|
316 // GlobalReplace() is like Replace(), except that it replaces all |
|
317 // occurrences of the pattern in the string with the rewrite. |
|
318 // Replacements are not subject to re-matching. E.g., |
|
319 // |
|
320 // string s = "yabba dabba doo"; |
|
321 // pcrecpp::RE("b+").GlobalReplace("d", &s); |
|
322 // |
|
323 // will leave "s" containing "yada dada doo". It returns the number |
|
324 // of replacements made. |
|
325 // |
|
326 // Extract() is like Replace(), except that if the pattern matches, |
|
327 // "rewrite" is copied into "out" (an additional argument) with |
|
328 // substitutions. The non-matching portions of "text" are ignored. |
|
329 // Returns true iff a match occurred and the extraction happened |
|
330 // successfully. If no match occurs, the string is left unaffected. |
|
331 |
|
332 |
|
333 #include <string> |
|
334 #include <pcre.h> |
|
335 #include <pcrecpparg.h> // defines the Arg class |
|
336 // This isn't technically needed here, but we include it |
|
337 // anyway so folks who include pcrecpp.h don't have to. |
|
338 #include <pcre_stringpiece.h> |
|
339 |
|
340 namespace pcrecpp { |
|
341 |
|
342 #define PCRE_SET_OR_CLEAR(b, o) \ |
|
343 if (b) all_options_ |= (o); else all_options_ &= ~(o); \ |
|
344 return *this |
|
345 |
|
346 #define PCRE_IS_SET(o) \ |
|
347 (all_options_ & o) == o |
|
348 |
|
349 /***** Compiling regular expressions: the RE class *****/ |
|
350 |
|
351 // RE_Options allow you to set options to be passed along to pcre, |
|
352 // along with other options we put on top of pcre. |
|
353 // Only 9 modifiers, plus match_limit and match_limit_recursion, |
|
354 // are supported now. |
|
355 class PCRECPP_EXP_DEFN RE_Options { |
|
356 public: |
|
357 // constructor |
|
358 RE_Options() : match_limit_(0), match_limit_recursion_(0), all_options_(0) {} |
|
359 |
|
360 // alternative constructor. |
|
361 // To facilitate transfer of legacy code from C programs |
|
362 // |
|
363 // This lets you do |
|
364 // RE(pattern, RE_Options(PCRE_CASELESS|PCRE_MULTILINE)).PartialMatch(str); |
|
365 // But new code is better off doing |
|
366 // RE(pattern, |
|
367 // RE_Options().set_caseless(true).set_multiline(true)).PartialMatch(str); |
|
368 RE_Options(int option_flags) : match_limit_(0), match_limit_recursion_(0), |
|
369 all_options_(option_flags) {} |
|
370 // we're fine with the default destructor, copy constructor, etc. |
|
371 |
|
372 // accessors and mutators |
|
373 int match_limit() const { return match_limit_; }; |
|
374 RE_Options &set_match_limit(int limit) { |
|
375 match_limit_ = limit; |
|
376 return *this; |
|
377 } |
|
378 |
|
379 int match_limit_recursion() const { return match_limit_recursion_; }; |
|
380 RE_Options &set_match_limit_recursion(int limit) { |
|
381 match_limit_recursion_ = limit; |
|
382 return *this; |
|
383 } |
|
384 |
|
385 bool caseless() const { |
|
386 return PCRE_IS_SET(PCRE_CASELESS); |
|
387 } |
|
388 RE_Options &set_caseless(bool x) { |
|
389 PCRE_SET_OR_CLEAR(x, PCRE_CASELESS); |
|
390 } |
|
391 |
|
392 bool multiline() const { |
|
393 return PCRE_IS_SET(PCRE_MULTILINE); |
|
394 } |
|
395 RE_Options &set_multiline(bool x) { |
|
396 PCRE_SET_OR_CLEAR(x, PCRE_MULTILINE); |
|
397 } |
|
398 |
|
399 bool dotall() const { |
|
400 return PCRE_IS_SET(PCRE_DOTALL); |
|
401 } |
|
402 RE_Options &set_dotall(bool x) { |
|
403 PCRE_SET_OR_CLEAR(x, PCRE_DOTALL); |
|
404 } |
|
405 |
|
406 bool extended() const { |
|
407 return PCRE_IS_SET(PCRE_EXTENDED); |
|
408 } |
|
409 RE_Options &set_extended(bool x) { |
|
410 PCRE_SET_OR_CLEAR(x, PCRE_EXTENDED); |
|
411 } |
|
412 |
|
413 bool dollar_endonly() const { |
|
414 return PCRE_IS_SET(PCRE_DOLLAR_ENDONLY); |
|
415 } |
|
416 RE_Options &set_dollar_endonly(bool x) { |
|
417 PCRE_SET_OR_CLEAR(x, PCRE_DOLLAR_ENDONLY); |
|
418 } |
|
419 |
|
420 bool extra() const { |
|
421 return PCRE_IS_SET(PCRE_EXTRA); |
|
422 } |
|
423 RE_Options &set_extra(bool x) { |
|
424 PCRE_SET_OR_CLEAR(x, PCRE_EXTRA); |
|
425 } |
|
426 |
|
427 bool ungreedy() const { |
|
428 return PCRE_IS_SET(PCRE_UNGREEDY); |
|
429 } |
|
430 RE_Options &set_ungreedy(bool x) { |
|
431 PCRE_SET_OR_CLEAR(x, PCRE_UNGREEDY); |
|
432 } |
|
433 |
|
434 bool utf8() const { |
|
435 return PCRE_IS_SET(PCRE_UTF8); |
|
436 } |
|
437 RE_Options &set_utf8(bool x) { |
|
438 PCRE_SET_OR_CLEAR(x, PCRE_UTF8); |
|
439 } |
|
440 |
|
441 bool no_auto_capture() const { |
|
442 return PCRE_IS_SET(PCRE_NO_AUTO_CAPTURE); |
|
443 } |
|
444 RE_Options &set_no_auto_capture(bool x) { |
|
445 PCRE_SET_OR_CLEAR(x, PCRE_NO_AUTO_CAPTURE); |
|
446 } |
|
447 |
|
448 RE_Options &set_all_options(int opt) { |
|
449 all_options_ = opt; |
|
450 return *this; |
|
451 } |
|
452 int all_options() const { |
|
453 return all_options_ ; |
|
454 } |
|
455 |
|
456 // TODO: add other pcre flags |
|
457 |
|
458 private: |
|
459 int match_limit_; |
|
460 int match_limit_recursion_; |
|
461 int all_options_; |
|
462 }; |
|
463 |
|
464 // These functions return some common RE_Options |
|
465 static inline RE_Options UTF8() { |
|
466 return RE_Options().set_utf8(true); |
|
467 } |
|
468 |
|
469 static inline RE_Options CASELESS() { |
|
470 return RE_Options().set_caseless(true); |
|
471 } |
|
472 static inline RE_Options MULTILINE() { |
|
473 return RE_Options().set_multiline(true); |
|
474 } |
|
475 |
|
476 static inline RE_Options DOTALL() { |
|
477 return RE_Options().set_dotall(true); |
|
478 } |
|
479 |
|
480 static inline RE_Options EXTENDED() { |
|
481 return RE_Options().set_extended(true); |
|
482 } |
|
483 |
|
484 // Interface for regular expression matching. Also corresponds to a |
|
485 // pre-compiled regular expression. An "RE" object is safe for |
|
486 // concurrent use by multiple threads. |
|
487 class PCRECPP_EXP_DEFN RE { |
|
488 public: |
|
489 // We provide implicit conversions from strings so that users can |
|
490 // pass in a string or a "const char*" wherever an "RE" is expected. |
|
491 RE(const string& pat) { Init(pat, NULL); } |
|
492 RE(const string& pat, const RE_Options& option) { Init(pat, &option); } |
|
493 RE(const char* pat) { Init(pat, NULL); } |
|
494 RE(const char* pat, const RE_Options& option) { Init(pat, &option); } |
|
495 RE(const unsigned char* pat) { |
|
496 Init(reinterpret_cast<const char*>(pat), NULL); |
|
497 } |
|
498 RE(const unsigned char* pat, const RE_Options& option) { |
|
499 Init(reinterpret_cast<const char*>(pat), &option); |
|
500 } |
|
501 |
|
502 // Copy constructor & assignment - note that these are expensive |
|
503 // because they recompile the expression. |
|
504 RE(const RE& re) { Init(re.pattern_, &re.options_); } |
|
505 const RE& operator=(const RE& re) { |
|
506 if (this != &re) { |
|
507 Cleanup(); |
|
508 |
|
509 // This is the code that originally came from Google |
|
510 // Init(re.pattern_.c_str(), &re.options_); |
|
511 |
|
512 // This is the replacement from Ari Pollak |
|
513 Init(re.pattern_, &re.options_); |
|
514 } |
|
515 return *this; |
|
516 } |
|
517 |
|
518 |
|
519 ~RE(); |
|
520 |
|
521 // The string specification for this RE. E.g. |
|
522 // RE re("ab*c?d+"); |
|
523 // re.pattern(); // "ab*c?d+" |
|
524 const string& pattern() const { return pattern_; } |
|
525 |
|
526 // If RE could not be created properly, returns an error string. |
|
527 // Else returns the empty string. |
|
528 const string& error() const { return *error_; } |
|
529 |
|
530 /***** The useful part: the matching interface *****/ |
|
531 |
|
532 // This is provided so one can do pattern.ReplaceAll() just as |
|
533 // easily as ReplaceAll(pattern-text, ....) |
|
534 |
|
535 bool FullMatch(const StringPiece& text, |
|
536 const Arg& ptr1 = no_arg, |
|
537 const Arg& ptr2 = no_arg, |
|
538 const Arg& ptr3 = no_arg, |
|
539 const Arg& ptr4 = no_arg, |
|
540 const Arg& ptr5 = no_arg, |
|
541 const Arg& ptr6 = no_arg, |
|
542 const Arg& ptr7 = no_arg, |
|
543 const Arg& ptr8 = no_arg, |
|
544 const Arg& ptr9 = no_arg, |
|
545 const Arg& ptr10 = no_arg, |
|
546 const Arg& ptr11 = no_arg, |
|
547 const Arg& ptr12 = no_arg, |
|
548 const Arg& ptr13 = no_arg, |
|
549 const Arg& ptr14 = no_arg, |
|
550 const Arg& ptr15 = no_arg, |
|
551 const Arg& ptr16 = no_arg) const; |
|
552 |
|
553 bool PartialMatch(const StringPiece& text, |
|
554 const Arg& ptr1 = no_arg, |
|
555 const Arg& ptr2 = no_arg, |
|
556 const Arg& ptr3 = no_arg, |
|
557 const Arg& ptr4 = no_arg, |
|
558 const Arg& ptr5 = no_arg, |
|
559 const Arg& ptr6 = no_arg, |
|
560 const Arg& ptr7 = no_arg, |
|
561 const Arg& ptr8 = no_arg, |
|
562 const Arg& ptr9 = no_arg, |
|
563 const Arg& ptr10 = no_arg, |
|
564 const Arg& ptr11 = no_arg, |
|
565 const Arg& ptr12 = no_arg, |
|
566 const Arg& ptr13 = no_arg, |
|
567 const Arg& ptr14 = no_arg, |
|
568 const Arg& ptr15 = no_arg, |
|
569 const Arg& ptr16 = no_arg) const; |
|
570 |
|
571 bool Consume(StringPiece* input, |
|
572 const Arg& ptr1 = no_arg, |
|
573 const Arg& ptr2 = no_arg, |
|
574 const Arg& ptr3 = no_arg, |
|
575 const Arg& ptr4 = no_arg, |
|
576 const Arg& ptr5 = no_arg, |
|
577 const Arg& ptr6 = no_arg, |
|
578 const Arg& ptr7 = no_arg, |
|
579 const Arg& ptr8 = no_arg, |
|
580 const Arg& ptr9 = no_arg, |
|
581 const Arg& ptr10 = no_arg, |
|
582 const Arg& ptr11 = no_arg, |
|
583 const Arg& ptr12 = no_arg, |
|
584 const Arg& ptr13 = no_arg, |
|
585 const Arg& ptr14 = no_arg, |
|
586 const Arg& ptr15 = no_arg, |
|
587 const Arg& ptr16 = no_arg) const; |
|
588 |
|
589 bool FindAndConsume(StringPiece* input, |
|
590 const Arg& ptr1 = no_arg, |
|
591 const Arg& ptr2 = no_arg, |
|
592 const Arg& ptr3 = no_arg, |
|
593 const Arg& ptr4 = no_arg, |
|
594 const Arg& ptr5 = no_arg, |
|
595 const Arg& ptr6 = no_arg, |
|
596 const Arg& ptr7 = no_arg, |
|
597 const Arg& ptr8 = no_arg, |
|
598 const Arg& ptr9 = no_arg, |
|
599 const Arg& ptr10 = no_arg, |
|
600 const Arg& ptr11 = no_arg, |
|
601 const Arg& ptr12 = no_arg, |
|
602 const Arg& ptr13 = no_arg, |
|
603 const Arg& ptr14 = no_arg, |
|
604 const Arg& ptr15 = no_arg, |
|
605 const Arg& ptr16 = no_arg) const; |
|
606 |
|
607 bool Replace(const StringPiece& rewrite, |
|
608 string *str) const; |
|
609 |
|
610 int GlobalReplace(const StringPiece& rewrite, |
|
611 string *str) const; |
|
612 |
|
613 bool Extract(const StringPiece &rewrite, |
|
614 const StringPiece &text, |
|
615 string *out) const; |
|
616 |
|
617 // Escapes all potentially meaningful regexp characters in |
|
618 // 'unquoted'. The returned string, used as a regular expression, |
|
619 // will exactly match the original string. For example, |
|
620 // 1.5-2.0? |
|
621 // may become: |
|
622 // 1\.5\-2\.0\? |
|
623 // Note QuoteMeta behaves the same as perl's QuoteMeta function, |
|
624 // *except* that it escapes the NUL character (\0) as backslash + 0, |
|
625 // rather than backslash + NUL. |
|
626 static string QuoteMeta(const StringPiece& unquoted); |
|
627 |
|
628 |
|
629 /***** Generic matching interface *****/ |
|
630 |
|
631 // Type of match (TODO: Should be restructured as part of RE_Options) |
|
632 enum Anchor { |
|
633 UNANCHORED, // No anchoring |
|
634 ANCHOR_START, // Anchor at start only |
|
635 ANCHOR_BOTH // Anchor at start and end |
|
636 }; |
|
637 |
|
638 // General matching routine. Stores the length of the match in |
|
639 // "*consumed" if successful. |
|
640 bool DoMatch(const StringPiece& text, |
|
641 Anchor anchor, |
|
642 int* consumed, |
|
643 const Arg* const* args, int n) const; |
|
644 |
|
645 // Return the number of capturing subpatterns, or -1 if the |
|
646 // regexp wasn't valid on construction. |
|
647 int NumberOfCapturingGroups() const; |
|
648 |
|
649 // The default value for an argument, to indicate no arg was passed in |
|
650 static Arg no_arg; |
|
651 |
|
652 private: |
|
653 |
|
654 void Init(const string& pattern, const RE_Options* options); |
|
655 void Cleanup(); |
|
656 |
|
657 // Match against "text", filling in "vec" (up to "vecsize" * 2/3) with |
|
658 // pairs of integers for the beginning and end positions of matched |
|
659 // text. The first pair corresponds to the entire matched text; |
|
660 // subsequent pairs correspond, in order, to parentheses-captured |
|
661 // matches. Returns the number of pairs (one more than the number of |
|
662 // the last subpattern with a match) if matching was successful |
|
663 // and zero if the match failed. |
|
664 // I.e. for RE("(foo)|(bar)|(baz)") it will return 2, 3, and 4 when matching |
|
665 // against "foo", "bar", and "baz" respectively. |
|
666 // When matching RE("(foo)|hello") against "hello", it will return 1. |
|
667 // But the values for all subpattern are filled in into "vec". |
|
668 int TryMatch(const StringPiece& text, |
|
669 int startpos, |
|
670 Anchor anchor, |
|
671 int *vec, |
|
672 int vecsize) const; |
|
673 |
|
674 // Append the "rewrite" string, with backslash subsitutions from "text" |
|
675 // and "vec", to string "out". |
|
676 bool Rewrite(string *out, |
|
677 const StringPiece& rewrite, |
|
678 const StringPiece& text, |
|
679 int *vec, |
|
680 int veclen) const; |
|
681 |
|
682 // internal implementation for DoMatch |
|
683 bool DoMatchImpl(const StringPiece& text, |
|
684 Anchor anchor, |
|
685 int* consumed, |
|
686 const Arg* const args[], |
|
687 int n, |
|
688 int* vec, |
|
689 int vecsize) const; |
|
690 |
|
691 // Compile the regexp for the specified anchoring mode |
|
692 pcre* Compile(Anchor anchor); |
|
693 |
|
694 string pattern_; |
|
695 RE_Options options_; |
|
696 pcre* re_full_; // For full matches |
|
697 pcre* re_partial_; // For partial matches |
|
698 const string* error_; // Error indicator (or points to empty string) |
|
699 }; |
|
700 |
|
701 } // namespace pcrecpp |
|
702 |
|
703 #endif /* _PCRECPP_H */ |