Line | Count | Source (jump to first uncovered line) |
1 | | // Copyright 2003-2009 The RE2 Authors. All Rights Reserved. |
2 | | // Use of this source code is governed by a BSD-style |
3 | | // license that can be found in the LICENSE file. |
4 | | |
5 | | #ifndef RE2_RE2_H_ |
6 | | #define RE2_RE2_H_ |
7 | | |
8 | | // C++ interface to the re2 regular-expression library. |
9 | | // RE2 supports Perl-style regular expressions (with extensions like |
10 | | // \d, \w, \s, ...). |
11 | | // |
12 | | // ----------------------------------------------------------------------- |
13 | | // REGEXP SYNTAX: |
14 | | // |
15 | | // This module uses the re2 library and hence supports |
16 | | // its syntax for regular expressions, which is similar to Perl's with |
17 | | // some of the more complicated things thrown away. In particular, |
18 | | // backreferences and generalized assertions are not available, nor is \Z. |
19 | | // |
20 | | // See https://github.com/google/re2/wiki/Syntax for the syntax |
21 | | // supported by RE2, and a comparison with PCRE and PERL regexps. |
22 | | // |
23 | | // For those not familiar with Perl's regular expressions, |
24 | | // here are some examples of the most commonly used extensions: |
25 | | // |
26 | | // "hello (\\w+) world" -- \w matches a "word" character |
27 | | // "version (\\d+)" -- \d matches a digit |
28 | | // "hello\\s+world" -- \s matches any whitespace character |
29 | | // "\\b(\\w+)\\b" -- \b matches non-empty string at word boundary |
30 | | // "(?i)hello" -- (?i) turns on case-insensitive matching |
31 | | // "/\\*(.*?)\\*/" -- .*? matches . minimum no. of times possible |
32 | | // |
33 | | // The double backslashes are needed when writing C++ string literals. |
34 | | // However, they should NOT be used when writing C++11 raw string literals: |
35 | | // |
36 | | // R"(hello (\w+) world)" -- \w matches a "word" character |
37 | | // R"(version (\d+))" -- \d matches a digit |
38 | | // R"(hello\s+world)" -- \s matches any whitespace character |
39 | | // R"(\b(\w+)\b)" -- \b matches non-empty string at word boundary |
40 | | // R"((?i)hello)" -- (?i) turns on case-insensitive matching |
41 | | // R"(/\*(.*?)\*/)" -- .*? matches . minimum no. of times possible |
42 | | // |
43 | | // When using UTF-8 encoding, case-insensitive matching will perform |
44 | | // simple case folding, not full case folding. |
45 | | // |
46 | | // ----------------------------------------------------------------------- |
47 | | // MATCHING INTERFACE: |
48 | | // |
49 | | // The "FullMatch" operation checks that supplied text matches a |
50 | | // supplied pattern exactly. |
51 | | // |
52 | | // Example: successful match |
53 | | // CHECK(RE2::FullMatch("hello", "h.*o")); |
54 | | // |
55 | | // Example: unsuccessful match (requires full match): |
56 | | // CHECK(!RE2::FullMatch("hello", "e")); |
57 | | // |
58 | | // ----------------------------------------------------------------------- |
59 | | // UTF-8 AND THE MATCHING INTERFACE: |
60 | | // |
61 | | // By default, the pattern and input text are interpreted as UTF-8. |
62 | | // The RE2::Latin1 option causes them to be interpreted as Latin-1. |
63 | | // |
64 | | // Example: |
65 | | // CHECK(RE2::FullMatch(utf8_string, RE2(utf8_pattern))); |
66 | | // CHECK(RE2::FullMatch(latin1_string, RE2(latin1_pattern, RE2::Latin1))); |
67 | | // |
68 | | // ----------------------------------------------------------------------- |
69 | | // SUBMATCH EXTRACTION: |
70 | | // |
71 | | // You can supply extra pointer arguments to extract submatches. |
72 | | // On match failure, none of the pointees will have been modified. |
73 | | // On match success, the submatches will be converted (as necessary) and |
74 | | // their values will be assigned to their pointees until all conversions |
75 | | // have succeeded or one conversion has failed. |
76 | | // On conversion failure, the pointees will be in an indeterminate state |
77 | | // because the caller has no way of knowing which conversion failed. |
78 | | // However, conversion cannot fail for types like string and StringPiece |
79 | | // that do not inspect the submatch contents. Hence, in the common case |
80 | | // where all of the pointees are of such types, failure is always due to |
81 | | // match failure and thus none of the pointees will have been modified. |
82 | | // |
83 | | // Example: extracts "ruby" into "s" and 1234 into "i" |
84 | | // int i; |
85 | | // std::string s; |
86 | | // CHECK(RE2::FullMatch("ruby:1234", "(\\w+):(\\d+)", &s, &i)); |
87 | | // |
88 | | // Example: fails because string cannot be stored in integer |
89 | | // CHECK(!RE2::FullMatch("ruby", "(.*)", &i)); |
90 | | // |
91 | | // Example: fails because there aren't enough sub-patterns |
92 | | // CHECK(!RE2::FullMatch("ruby:1234", "\\w+:\\d+", &s)); |
93 | | // |
94 | | // Example: does not try to extract any extra sub-patterns |
95 | | // CHECK(RE2::FullMatch("ruby:1234", "(\\w+):(\\d+)", &s)); |
96 | | // |
97 | | // Example: does not try to extract into NULL |
98 | | // CHECK(RE2::FullMatch("ruby:1234", "(\\w+):(\\d+)", NULL, &i)); |
99 | | // |
100 | | // Example: integer overflow causes failure |
101 | | // CHECK(!RE2::FullMatch("ruby:1234567891234", "\\w+:(\\d+)", &i)); |
102 | | // |
103 | | // NOTE(rsc): Asking for submatches slows successful matches quite a bit. |
104 | | // This may get a little faster in the future, but right now is slower |
105 | | // than PCRE. On the other hand, failed matches run *very* fast (faster |
106 | | // than PCRE), as do matches without submatch extraction. |
107 | | // |
108 | | // ----------------------------------------------------------------------- |
109 | | // PARTIAL MATCHES |
110 | | // |
111 | | // You can use the "PartialMatch" operation when you want the pattern |
112 | | // to match any substring of the text. |
113 | | // |
114 | | // Example: simple search for a string: |
115 | | // CHECK(RE2::PartialMatch("hello", "ell")); |
116 | | // |
117 | | // Example: find first number in a string |
118 | | // int number; |
119 | | // CHECK(RE2::PartialMatch("x*100 + 20", "(\\d+)", &number)); |
120 | | // CHECK_EQ(number, 100); |
121 | | // |
122 | | // ----------------------------------------------------------------------- |
123 | | // PRE-COMPILED REGULAR EXPRESSIONS |
124 | | // |
125 | | // RE2 makes it easy to use any string as a regular expression, without |
126 | | // requiring a separate compilation step. |
127 | | // |
128 | | // If speed is of the essence, you can create a pre-compiled "RE2" |
129 | | // object from the pattern and use it multiple times. If you do so, |
130 | | // you can typically parse text faster than with sscanf. |
131 | | // |
132 | | // Example: precompile pattern for faster matching: |
133 | | // RE2 pattern("h.*o"); |
134 | | // while (ReadLine(&str)) { |
135 | | // if (RE2::FullMatch(str, pattern)) ...; |
136 | | // } |
137 | | // |
138 | | // ----------------------------------------------------------------------- |
139 | | // SCANNING TEXT INCREMENTALLY |
140 | | // |
141 | | // The "Consume" operation may be useful if you want to repeatedly |
142 | | // match regular expressions at the front of a string and skip over |
143 | | // them as they match. This requires use of the "StringPiece" type, |
144 | | // which represents a sub-range of a real string. |
145 | | // |
146 | | // Example: read lines of the form "var = value" from a string. |
147 | | // std::string contents = ...; // Fill string somehow |
148 | | // StringPiece input(contents); // Wrap a StringPiece around it |
149 | | // |
150 | | // std::string var; |
151 | | // int value; |
152 | | // while (RE2::Consume(&input, "(\\w+) = (\\d+)\n", &var, &value)) { |
153 | | // ...; |
154 | | // } |
155 | | // |
156 | | // Each successful call to "Consume" will set "var/value", and also |
157 | | // advance "input" so it points past the matched text. Note that if the |
158 | | // regular expression matches an empty string, input will advance |
159 | | // by 0 bytes. If the regular expression being used might match |
160 | | // an empty string, the loop body must check for this case and either |
161 | | // advance the string or break out of the loop. |
162 | | // |
163 | | // The "FindAndConsume" operation is similar to "Consume" but does not |
164 | | // anchor your match at the beginning of the string. For example, you |
165 | | // could extract all words from a string by repeatedly calling |
166 | | // RE2::FindAndConsume(&input, "(\\w+)", &word) |
167 | | // |
168 | | // ----------------------------------------------------------------------- |
169 | | // USING VARIABLE NUMBER OF ARGUMENTS |
170 | | // |
171 | | // The above operations require you to know the number of arguments |
172 | | // when you write the code. This is not always possible or easy (for |
173 | | // example, the regular expression may be calculated at run time). |
174 | | // You can use the "N" version of the operations when the number of |
175 | | // match arguments are determined at run time. |
176 | | // |
177 | | // Example: |
178 | | // const RE2::Arg* args[10]; |
179 | | // int n; |
180 | | // // ... populate args with pointers to RE2::Arg values ... |
181 | | // // ... set n to the number of RE2::Arg objects ... |
182 | | // bool match = RE2::FullMatchN(input, pattern, args, n); |
183 | | // |
184 | | // The last statement is equivalent to |
185 | | // |
186 | | // bool match = RE2::FullMatch(input, pattern, |
187 | | // *args[0], *args[1], ..., *args[n - 1]); |
188 | | // |
189 | | // ----------------------------------------------------------------------- |
190 | | // PARSING HEX/OCTAL/C-RADIX NUMBERS |
191 | | // |
192 | | // By default, if you pass a pointer to a numeric value, the |
193 | | // corresponding text is interpreted as a base-10 number. You can |
194 | | // instead wrap the pointer with a call to one of the operators Hex(), |
195 | | // Octal(), or CRadix() to interpret the text in another base. The |
196 | | // CRadix operator interprets C-style "0" (base-8) and "0x" (base-16) |
197 | | // prefixes, but defaults to base-10. |
198 | | // |
199 | | // Example: |
200 | | // int a, b, c, d; |
201 | | // CHECK(RE2::FullMatch("100 40 0100 0x40", "(.*) (.*) (.*) (.*)", |
202 | | // RE2::Octal(&a), RE2::Hex(&b), RE2::CRadix(&c), RE2::CRadix(&d)); |
203 | | // will leave 64 in a, b, c, and d. |
204 | | |
205 | | #include <stddef.h> |
206 | | #include <stdint.h> |
207 | | #include <algorithm> |
208 | | #include <map> |
209 | | #include <mutex> |
210 | | #include <string> |
211 | | #include <type_traits> |
212 | | #include <vector> |
213 | | |
214 | | #if defined(__APPLE__) |
215 | | #include <TargetConditionals.h> |
216 | | #endif |
217 | | |
218 | | #include "re2/stringpiece.h" |
219 | | |
220 | | namespace re2 { |
221 | | class Prog; |
222 | | class Regexp; |
223 | | } // namespace re2 |
224 | | |
225 | | namespace re2 { |
226 | | |
227 | | // Interface for regular expression matching. Also corresponds to a |
228 | | // pre-compiled regular expression. An "RE2" object is safe for |
229 | | // concurrent use by multiple threads. |
230 | | class RE2 { |
231 | | public: |
232 | | // We convert user-passed pointers into special Arg objects |
233 | | class Arg; |
234 | | class Options; |
235 | | |
236 | | // Defined in set.h. |
237 | | class Set; |
238 | | |
239 | | enum ErrorCode { |
240 | | NoError = 0, |
241 | | |
242 | | // Unexpected error |
243 | | ErrorInternal, |
244 | | |
245 | | // Parse errors |
246 | | ErrorBadEscape, // bad escape sequence |
247 | | ErrorBadCharClass, // bad character class |
248 | | ErrorBadCharRange, // bad character class range |
249 | | ErrorMissingBracket, // missing closing ] |
250 | | ErrorMissingParen, // missing closing ) |
251 | | ErrorUnexpectedParen, // unexpected closing ) |
252 | | ErrorTrailingBackslash, // trailing \ at end of regexp |
253 | | ErrorRepeatArgument, // repeat argument missing, e.g. "*" |
254 | | ErrorRepeatSize, // bad repetition argument |
255 | | ErrorRepeatOp, // bad repetition operator |
256 | | ErrorBadPerlOp, // bad perl operator |
257 | | ErrorBadUTF8, // invalid UTF-8 in regexp |
258 | | ErrorBadNamedCapture, // bad named capture group |
259 | | ErrorPatternTooLarge // pattern too large (compile failed) |
260 | | }; |
261 | | |
262 | | // Predefined common options. |
263 | | // If you need more complicated things, instantiate |
264 | | // an Option class, possibly passing one of these to |
265 | | // the Option constructor, change the settings, and pass that |
266 | | // Option class to the RE2 constructor. |
267 | | enum CannedOptions { |
268 | | DefaultOptions = 0, |
269 | | Latin1, // treat input as Latin-1 (default UTF-8) |
270 | | POSIX, // POSIX syntax, leftmost-longest match |
271 | | Quiet // do not log about regexp parse errors |
272 | | }; |
273 | | |
274 | | // Need to have the const char* and const std::string& forms for implicit |
275 | | // conversions when passing string literals to FullMatch and PartialMatch. |
276 | | // Otherwise the StringPiece form would be sufficient. |
277 | | RE2(const char* pattern); |
278 | | RE2(const std::string& pattern); |
279 | | RE2(const StringPiece& pattern); |
280 | | RE2(const StringPiece& pattern, const Options& options); |
281 | | ~RE2(); |
282 | | |
283 | | // Not copyable. |
284 | | // RE2 objects are expensive. You should probably use std::shared_ptr<RE2> |
285 | | // instead. If you really must copy, RE2(first.pattern(), first.options()) |
286 | | // effectively does so: it produces a second object that mimics the first. |
287 | | RE2(const RE2&) = delete; |
288 | | RE2& operator=(const RE2&) = delete; |
289 | | // Not movable. |
290 | | // RE2 objects are thread-safe and logically immutable. You should probably |
291 | | // use std::unique_ptr<RE2> instead. Otherwise, consider std::deque<RE2> if |
292 | | // direct emplacement into a container is desired. If you really must move, |
293 | | // be prepared to submit a design document along with your feature request. |
294 | | RE2(RE2&&) = delete; |
295 | | RE2& operator=(RE2&&) = delete; |
296 | | |
297 | | // Returns whether RE2 was created properly. |
298 | 140k | bool ok() const { return error_code() == NoError; } |
299 | | |
300 | | // The string specification for this RE2. E.g. |
301 | | // RE2 re("ab*c?d+"); |
302 | | // re.pattern(); // "ab*c?d+" |
303 | 0 | const std::string& pattern() const { return *pattern_; } |
304 | | |
305 | | // If RE2 could not be created properly, returns an error string. |
306 | | // Else returns the empty string. |
307 | 0 | const std::string& error() const { return *error_; } |
308 | | |
309 | | // If RE2 could not be created properly, returns an error code. |
310 | | // Else returns RE2::NoError (== 0). |
311 | 140k | ErrorCode error_code() const { return error_code_; } |
312 | | |
313 | | // If RE2 could not be created properly, returns the offending |
314 | | // portion of the regexp. |
315 | 0 | const std::string& error_arg() const { return *error_arg_; } |
316 | | |
317 | | // Returns the program size, a very approximate measure of a regexp's "cost". |
318 | | // Larger numbers are more expensive than smaller numbers. |
319 | | int ProgramSize() const; |
320 | | int ReverseProgramSize() const; |
321 | | |
322 | | // If histogram is not null, outputs the program fanout |
323 | | // as a histogram bucketed by powers of 2. |
324 | | // Returns the number of the largest non-empty bucket. |
325 | | int ProgramFanout(std::vector<int>* histogram) const; |
326 | | int ReverseProgramFanout(std::vector<int>* histogram) const; |
327 | | |
328 | | // Returns the underlying Regexp; not for general use. |
329 | | // Returns entire_regexp_ so that callers don't need |
330 | | // to know about prefix_ and prefix_foldcase_. |
331 | 0 | re2::Regexp* Regexp() const { return entire_regexp_; } |
332 | | |
333 | | /***** The array-based matching interface ******/ |
334 | | |
335 | | // The functions here have names ending in 'N' and are used to implement |
336 | | // the functions whose names are the prefix before the 'N'. It is sometimes |
337 | | // useful to invoke them directly, but the syntax is awkward, so the 'N'-less |
338 | | // versions should be preferred. |
339 | | static bool FullMatchN(const StringPiece& text, const RE2& re, |
340 | | const Arg* const args[], int n); |
341 | | static bool PartialMatchN(const StringPiece& text, const RE2& re, |
342 | | const Arg* const args[], int n); |
343 | | static bool ConsumeN(StringPiece* input, const RE2& re, |
344 | | const Arg* const args[], int n); |
345 | | static bool FindAndConsumeN(StringPiece* input, const RE2& re, |
346 | | const Arg* const args[], int n); |
347 | | |
348 | | private: |
349 | | template <typename F, typename SP> |
350 | 41.8k | static inline bool Apply(F f, SP sp, const RE2& re) { |
351 | 41.8k | return f(sp, re, NULL, 0); |
352 | 41.8k | } |
353 | | |
354 | | template <typename F, typename SP, typename... A> |
355 | | static inline bool Apply(F f, SP sp, const RE2& re, const A&... a) { |
356 | | const Arg* const args[] = {&a...}; |
357 | | const int n = sizeof...(a); |
358 | | return f(sp, re, args, n); |
359 | | } |
360 | | |
361 | | public: |
362 | | // In order to allow FullMatch() et al. to be called with a varying number |
363 | | // of arguments of varying types, we use two layers of variadic templates. |
364 | | // The first layer constructs the temporary Arg objects. The second layer |
365 | | // (above) constructs the array of pointers to the temporary Arg objects. |
366 | | |
367 | | /***** The useful part: the matching interface *****/ |
368 | | |
369 | | // Matches "text" against "re". If pointer arguments are |
370 | | // supplied, copies matched sub-patterns into them. |
371 | | // |
372 | | // You can pass in a "const char*" or a "std::string" for "text". |
373 | | // You can pass in a "const char*" or a "std::string" or a "RE2" for "re". |
374 | | // |
375 | | // The provided pointer arguments can be pointers to any scalar numeric |
376 | | // type, or one of: |
377 | | // std::string (matched piece is copied to string) |
378 | | // StringPiece (StringPiece is mutated to point to matched piece) |
379 | | // T (where "bool T::ParseFrom(const char*, size_t)" exists) |
380 | | // (void*)NULL (the corresponding matched sub-pattern is not copied) |
381 | | // |
382 | | // Returns true iff all of the following conditions are satisfied: |
383 | | // a. "text" matches "re" fully - from the beginning to the end of "text". |
384 | | // b. The number of matched sub-patterns is >= number of supplied pointers. |
385 | | // c. The "i"th argument has a suitable type for holding the |
386 | | // string captured as the "i"th sub-pattern. If you pass in |
387 | | // NULL for the "i"th argument, or pass fewer arguments than |
388 | | // number of sub-patterns, the "i"th captured sub-pattern is |
389 | | // ignored. |
390 | | // |
391 | | // CAVEAT: An optional sub-pattern that does not exist in the |
392 | | // matched string is assigned the empty string. Therefore, the |
393 | | // following will return false (because the empty string is not a |
394 | | // valid number): |
395 | | // int number; |
396 | | // RE2::FullMatch("abc", "[a-z]+(\\d+)?", &number); |
397 | | template <typename... A> |
398 | 41.8k | static bool FullMatch(const StringPiece& text, const RE2& re, A&&... a) { |
399 | 41.8k | return Apply(FullMatchN, text, re, Arg(std::forward<A>(a))...); |
400 | 41.8k | } |
401 | | |
402 | | // Like FullMatch(), except that "re" is allowed to match a substring |
403 | | // of "text". |
404 | | // |
405 | | // Returns true iff all of the following conditions are satisfied: |
406 | | // a. "text" matches "re" partially - for some substring of "text". |
407 | | // b. The number of matched sub-patterns is >= number of supplied pointers. |
408 | | // c. The "i"th argument has a suitable type for holding the |
409 | | // string captured as the "i"th sub-pattern. If you pass in |
410 | | // NULL for the "i"th argument, or pass fewer arguments than |
411 | | // number of sub-patterns, the "i"th captured sub-pattern is |
412 | | // ignored. |
413 | | template <typename... A> |
414 | | static bool PartialMatch(const StringPiece& text, const RE2& re, A&&... a) { |
415 | | return Apply(PartialMatchN, text, re, Arg(std::forward<A>(a))...); |
416 | | } |
417 | | |
418 | | // Like FullMatch() and PartialMatch(), except that "re" has to match |
419 | | // a prefix of the text, and "input" is advanced past the matched |
420 | | // text. Note: "input" is modified iff this routine returns true |
421 | | // and "re" matched a non-empty substring of "input". |
422 | | // |
423 | | // Returns true iff all of the following conditions are satisfied: |
424 | | // a. "input" matches "re" partially - for some prefix of "input". |
425 | | // b. The number of matched sub-patterns is >= number of supplied pointers. |
426 | | // c. The "i"th argument has a suitable type for holding the |
427 | | // string captured as the "i"th sub-pattern. If you pass in |
428 | | // NULL for the "i"th argument, or pass fewer arguments than |
429 | | // number of sub-patterns, the "i"th captured sub-pattern is |
430 | | // ignored. |
431 | | template <typename... A> |
432 | | static bool Consume(StringPiece* input, const RE2& re, A&&... a) { |
433 | | return Apply(ConsumeN, input, re, Arg(std::forward<A>(a))...); |
434 | | } |
435 | | |
436 | | // Like Consume(), but does not anchor the match at the beginning of |
437 | | // the text. That is, "re" need not start its match at the beginning |
438 | | // of "input". For example, "FindAndConsume(s, "(\\w+)", &word)" finds |
439 | | // the next word in "s" and stores it in "word". |
440 | | // |
441 | | // Returns true iff all of the following conditions are satisfied: |
442 | | // a. "input" matches "re" partially - for some substring of "input". |
443 | | // b. The number of matched sub-patterns is >= number of supplied pointers. |
444 | | // c. The "i"th argument has a suitable type for holding the |
445 | | // string captured as the "i"th sub-pattern. If you pass in |
446 | | // NULL for the "i"th argument, or pass fewer arguments than |
447 | | // number of sub-patterns, the "i"th captured sub-pattern is |
448 | | // ignored. |
449 | | template <typename... A> |
450 | | static bool FindAndConsume(StringPiece* input, const RE2& re, A&&... a) { |
451 | | return Apply(FindAndConsumeN, input, re, Arg(std::forward<A>(a))...); |
452 | | } |
453 | | |
454 | | // Replace the first match of "re" in "str" with "rewrite". |
455 | | // Within "rewrite", backslash-escaped digits (\1 to \9) can be |
456 | | // used to insert text matching corresponding parenthesized group |
457 | | // from the pattern. \0 in "rewrite" refers to the entire matching |
458 | | // text. E.g., |
459 | | // |
460 | | // std::string s = "yabba dabba doo"; |
461 | | // CHECK(RE2::Replace(&s, "b+", "d")); |
462 | | // |
463 | | // will leave "s" containing "yada dabba doo" |
464 | | // |
465 | | // Returns true if the pattern matches and a replacement occurs, |
466 | | // false otherwise. |
467 | | static bool Replace(std::string* str, |
468 | | const RE2& re, |
469 | | const StringPiece& rewrite); |
470 | | |
471 | | // Like Replace(), except replaces successive non-overlapping occurrences |
472 | | // of the pattern in the string with the rewrite. E.g. |
473 | | // |
474 | | // std::string s = "yabba dabba doo"; |
475 | | // CHECK(RE2::GlobalReplace(&s, "b+", "d")); |
476 | | // |
477 | | // will leave "s" containing "yada dada doo" |
478 | | // Replacements are not subject to re-matching. |
479 | | // |
480 | | // Because GlobalReplace only replaces non-overlapping matches, |
481 | | // replacing "ana" within "banana" makes only one replacement, not two. |
482 | | // |
483 | | // Returns the number of replacements made. |
484 | | static int GlobalReplace(std::string* str, |
485 | | const RE2& re, |
486 | | const StringPiece& rewrite); |
487 | | |
488 | | // Like Replace, except that if the pattern matches, "rewrite" |
489 | | // is copied into "out" with substitutions. The non-matching |
490 | | // portions of "text" are ignored. |
491 | | // |
492 | | // Returns true iff a match occurred and the extraction happened |
493 | | // successfully; if no match occurs, the string is left unaffected. |
494 | | // |
495 | | // REQUIRES: "text" must not alias any part of "*out". |
496 | | static bool Extract(const StringPiece& text, |
497 | | const RE2& re, |
498 | | const StringPiece& rewrite, |
499 | | std::string* out); |
500 | | |
501 | | // Escapes all potentially meaningful regexp characters in |
502 | | // 'unquoted'. The returned string, used as a regular expression, |
503 | | // will match exactly the original string. For example, |
504 | | // 1.5-2.0? |
505 | | // may become: |
506 | | // 1\.5\-2\.0\? |
507 | | static std::string QuoteMeta(const StringPiece& unquoted); |
508 | | |
509 | | // Computes range for any strings matching regexp. The min and max can in |
510 | | // some cases be arbitrarily precise, so the caller gets to specify the |
511 | | // maximum desired length of string returned. |
512 | | // |
513 | | // Assuming PossibleMatchRange(&min, &max, N) returns successfully, any |
514 | | // string s that is an anchored match for this regexp satisfies |
515 | | // min <= s && s <= max. |
516 | | // |
517 | | // Note that PossibleMatchRange() will only consider the first copy of an |
518 | | // infinitely repeated element (i.e., any regexp element followed by a '*' or |
519 | | // '+' operator). Regexps with "{N}" constructions are not affected, as those |
520 | | // do not compile down to infinite repetitions. |
521 | | // |
522 | | // Returns true on success, false on error. |
523 | | bool PossibleMatchRange(std::string* min, std::string* max, |
524 | | int maxlen) const; |
525 | | |
526 | | // Generic matching interface |
527 | | |
528 | | // Type of match. |
529 | | enum Anchor { |
530 | | UNANCHORED, // No anchoring |
531 | | ANCHOR_START, // Anchor at start only |
532 | | ANCHOR_BOTH // Anchor at start and end |
533 | | }; |
534 | | |
535 | | // Return the number of capturing subpatterns, or -1 if the |
536 | | // regexp wasn't valid on construction. The overall match ($0) |
537 | | // does not count: if the regexp is "(a)(b)", returns 2. |
538 | 83.6k | int NumberOfCapturingGroups() const { return num_captures_; } |
539 | | |
540 | | // Return a map from names to capturing indices. |
541 | | // The map records the index of the leftmost group |
542 | | // with the given name. |
543 | | // Only valid until the re is deleted. |
544 | | const std::map<std::string, int>& NamedCapturingGroups() const; |
545 | | |
546 | | // Return a map from capturing indices to names. |
547 | | // The map has no entries for unnamed groups. |
548 | | // Only valid until the re is deleted. |
549 | | const std::map<int, std::string>& CapturingGroupNames() const; |
550 | | |
551 | | // General matching routine. |
552 | | // Match against text starting at offset startpos |
553 | | // and stopping the search at offset endpos. |
554 | | // Returns true if match found, false if not. |
555 | | // On a successful match, fills in submatch[] (up to nsubmatch entries) |
556 | | // with information about submatches. |
557 | | // I.e. matching RE2("(foo)|(bar)baz") on "barbazbla" will return true, with |
558 | | // submatch[0] = "barbaz", submatch[1].data() = NULL, submatch[2] = "bar", |
559 | | // submatch[3].data() = NULL, ..., up to submatch[nsubmatch-1].data() = NULL. |
560 | | // Caveat: submatch[] may be clobbered even on match failure. |
561 | | // |
562 | | // Don't ask for more match information than you will use: |
563 | | // runs much faster with nsubmatch == 1 than nsubmatch > 1, and |
564 | | // runs even faster if nsubmatch == 0. |
565 | | // Doesn't make sense to use nsubmatch > 1 + NumberOfCapturingGroups(), |
566 | | // but will be handled correctly. |
567 | | // |
568 | | // Passing text == StringPiece(NULL, 0) will be handled like any other |
569 | | // empty string, but note that on return, it will not be possible to tell |
570 | | // whether submatch i matched the empty string or did not match: |
571 | | // either way, submatch[i].data() == NULL. |
572 | | bool Match(const StringPiece& text, |
573 | | size_t startpos, |
574 | | size_t endpos, |
575 | | Anchor re_anchor, |
576 | | StringPiece* submatch, |
577 | | int nsubmatch) const; |
578 | | |
579 | | // Check that the given rewrite string is suitable for use with this |
580 | | // regular expression. It checks that: |
581 | | // * The regular expression has enough parenthesized subexpressions |
582 | | // to satisfy all of the \N tokens in rewrite |
583 | | // * The rewrite string doesn't have any syntax errors. E.g., |
584 | | // '\' followed by anything other than a digit or '\'. |
585 | | // A true return value guarantees that Replace() and Extract() won't |
586 | | // fail because of a bad rewrite string. |
587 | | bool CheckRewriteString(const StringPiece& rewrite, |
588 | | std::string* error) const; |
589 | | |
590 | | // Returns the maximum submatch needed for the rewrite to be done by |
591 | | // Replace(). E.g. if rewrite == "foo \\2,\\1", returns 2. |
592 | | static int MaxSubmatch(const StringPiece& rewrite); |
593 | | |
594 | | // Append the "rewrite" string, with backslash subsitutions from "vec", |
595 | | // to string "out". |
596 | | // Returns true on success. This method can fail because of a malformed |
597 | | // rewrite string. CheckRewriteString guarantees that the rewrite will |
598 | | // be sucessful. |
599 | | bool Rewrite(std::string* out, |
600 | | const StringPiece& rewrite, |
601 | | const StringPiece* vec, |
602 | | int veclen) const; |
603 | | |
604 | | // Constructor options |
605 | | class Options { |
606 | | public: |
607 | | // The options are (defaults in parentheses): |
608 | | // |
609 | | // utf8 (true) text and pattern are UTF-8; otherwise Latin-1 |
610 | | // posix_syntax (false) restrict regexps to POSIX egrep syntax |
611 | | // longest_match (false) search for longest match, not first match |
612 | | // log_errors (true) log syntax and execution errors to ERROR |
613 | | // max_mem (see below) approx. max memory footprint of RE2 |
614 | | // literal (false) interpret string as literal, not regexp |
615 | | // never_nl (false) never match \n, even if it is in regexp |
616 | | // dot_nl (false) dot matches everything including new line |
617 | | // never_capture (false) parse all parens as non-capturing |
618 | | // case_sensitive (true) match is case-sensitive (regexp can override |
619 | | // with (?i) unless in posix_syntax mode) |
620 | | // |
621 | | // The following options are only consulted when posix_syntax == true. |
622 | | // When posix_syntax == false, these features are always enabled and |
623 | | // cannot be turned off; to perform multi-line matching in that case, |
624 | | // begin the regexp with (?m). |
625 | | // perl_classes (false) allow Perl's \d \s \w \D \S \W |
626 | | // word_boundary (false) allow Perl's \b \B (word boundary and not) |
627 | | // one_line (false) ^ and $ only match beginning and end of text |
628 | | // |
629 | | // The max_mem option controls how much memory can be used |
630 | | // to hold the compiled form of the regexp (the Prog) and |
631 | | // its cached DFA graphs. Code Search placed limits on the number |
632 | | // of Prog instructions and DFA states: 10,000 for both. |
633 | | // In RE2, those limits would translate to about 240 KB per Prog |
634 | | // and perhaps 2.5 MB per DFA (DFA state sizes vary by regexp; RE2 does a |
635 | | // better job of keeping them small than Code Search did). |
636 | | // Each RE2 has two Progs (one forward, one reverse), and each Prog |
637 | | // can have two DFAs (one first match, one longest match). |
638 | | // That makes 4 DFAs: |
639 | | // |
640 | | // forward, first-match - used for UNANCHORED or ANCHOR_START searches |
641 | | // if opt.longest_match() == false |
642 | | // forward, longest-match - used for all ANCHOR_BOTH searches, |
643 | | // and the other two kinds if |
644 | | // opt.longest_match() == true |
645 | | // reverse, first-match - never used |
646 | | // reverse, longest-match - used as second phase for unanchored searches |
647 | | // |
648 | | // The RE2 memory budget is statically divided between the two |
649 | | // Progs and then the DFAs: two thirds to the forward Prog |
650 | | // and one third to the reverse Prog. The forward Prog gives half |
651 | | // of what it has left over to each of its DFAs. The reverse Prog |
652 | | // gives it all to its longest-match DFA. |
653 | | // |
654 | | // Once a DFA fills its budget, it flushes its cache and starts over. |
655 | | // If this happens too often, RE2 falls back on the NFA implementation. |
656 | | |
657 | | // For now, make the default budget something close to Code Search. |
658 | | static const int kDefaultMaxMem = 8<<20; |
659 | | |
660 | | enum Encoding { |
661 | | EncodingUTF8 = 1, |
662 | | EncodingLatin1 |
663 | | }; |
664 | | |
665 | | Options() : |
666 | | max_mem_(kDefaultMaxMem), |
667 | | encoding_(EncodingUTF8), |
668 | | posix_syntax_(false), |
669 | | longest_match_(false), |
670 | | log_errors_(true), |
671 | | literal_(false), |
672 | | never_nl_(false), |
673 | | dot_nl_(false), |
674 | | never_capture_(false), |
675 | | case_sensitive_(true), |
676 | | perl_classes_(false), |
677 | | word_boundary_(false), |
678 | 114k | one_line_(false) { |
679 | 114k | } |
680 | | |
681 | | /*implicit*/ Options(CannedOptions); |
682 | | |
683 | 42.7k | int64_t max_mem() const { return max_mem_; } |
684 | 0 | void set_max_mem(int64_t m) { max_mem_ = m; } |
685 | | |
686 | 57.1k | Encoding encoding() const { return encoding_; } |
687 | 40.8k | void set_encoding(Encoding encoding) { encoding_ = encoding; } |
688 | | |
689 | 57.1k | bool posix_syntax() const { return posix_syntax_; } |
690 | 57.1k | void set_posix_syntax(bool b) { posix_syntax_ = b; } |
691 | | |
692 | 57.1k | bool longest_match() const { return longest_match_; } |
693 | 57.1k | void set_longest_match(bool b) { longest_match_ = b; } |
694 | | |
695 | 19.7k | bool log_errors() const { return log_errors_; } |
696 | 57.1k | void set_log_errors(bool b) { log_errors_ = b; } |
697 | | |
698 | 57.1k | bool literal() const { return literal_; } |
699 | 57.1k | void set_literal(bool b) { literal_ = b; } |
700 | | |
701 | 57.1k | bool never_nl() const { return never_nl_; } |
702 | 57.1k | void set_never_nl(bool b) { never_nl_ = b; } |
703 | | |
704 | 57.1k | bool dot_nl() const { return dot_nl_; } |
705 | 57.1k | void set_dot_nl(bool b) { dot_nl_ = b; } |
706 | | |
707 | 57.1k | bool never_capture() const { return never_capture_; } |
708 | 57.1k | void set_never_capture(bool b) { never_capture_ = b; } |
709 | | |
710 | 57.1k | bool case_sensitive() const { return case_sensitive_; } |
711 | 57.1k | void set_case_sensitive(bool b) { case_sensitive_ = b; } |
712 | | |
713 | 57.1k | bool perl_classes() const { return perl_classes_; } |
714 | 57.1k | void set_perl_classes(bool b) { perl_classes_ = b; } |
715 | | |
716 | 57.1k | bool word_boundary() const { return word_boundary_; } |
717 | 57.1k | void set_word_boundary(bool b) { word_boundary_ = b; } |
718 | | |
719 | 57.1k | bool one_line() const { return one_line_; } |
720 | 57.1k | void set_one_line(bool b) { one_line_ = b; } |
721 | | |
722 | 57.1k | void Copy(const Options& src) { |
723 | 57.1k | *this = src; |
724 | 57.1k | } |
725 | | |
726 | | int ParseFlags() const; |
727 | | |
728 | | private: |
729 | | int64_t max_mem_; |
730 | | Encoding encoding_; |
731 | | bool posix_syntax_; |
732 | | bool longest_match_; |
733 | | bool log_errors_; |
734 | | bool literal_; |
735 | | bool never_nl_; |
736 | | bool dot_nl_; |
737 | | bool never_capture_; |
738 | | bool case_sensitive_; |
739 | | bool perl_classes_; |
740 | | bool word_boundary_; |
741 | | bool one_line_; |
742 | | }; |
743 | | |
744 | | // Returns the options set in the constructor. |
745 | 0 | const Options& options() const { return options_; } |
746 | | |
747 | | // Argument converters; see below. |
748 | | template <typename T> |
749 | | static Arg CRadix(T* ptr); |
750 | | template <typename T> |
751 | | static Arg Hex(T* ptr); |
752 | | template <typename T> |
753 | | static Arg Octal(T* ptr); |
754 | | |
755 | | // Controls the maximum count permitted by GlobalReplace(); -1 is unlimited. |
756 | | // FOR FUZZING ONLY. |
757 | | static void FUZZING_ONLY_set_maximum_global_replace_count(int i); |
758 | | |
759 | | private: |
760 | | void Init(const StringPiece& pattern, const Options& options); |
761 | | |
762 | | bool DoMatch(const StringPiece& text, |
763 | | Anchor re_anchor, |
764 | | size_t* consumed, |
765 | | const Arg* const args[], |
766 | | int n) const; |
767 | | |
768 | | re2::Prog* ReverseProg() const; |
769 | | |
770 | | // First cache line is relatively cold fields. |
771 | | const std::string* pattern_; // string regular expression |
772 | | Options options_; // option flags |
773 | | re2::Regexp* entire_regexp_; // parsed regular expression |
774 | | re2::Regexp* suffix_regexp_; // parsed regular expression, prefix_ removed |
775 | | const std::string* error_; // error indicator (or points to empty string) |
776 | | const std::string* error_arg_; // fragment of regexp showing error (or ditto) |
777 | | |
778 | | // Second cache line is relatively hot fields. |
779 | | // These are ordered oddly to pack everything. |
780 | | int num_captures_; // number of capturing groups |
781 | | ErrorCode error_code_ : 29; // error code (29 bits is more than enough) |
782 | | bool longest_match_ : 1; // cached copy of options_.longest_match() |
783 | | bool is_one_pass_ : 1; // can use prog_->SearchOnePass? |
784 | | bool prefix_foldcase_ : 1; // prefix_ is ASCII case-insensitive |
785 | | std::string prefix_; // required prefix (before suffix_regexp_) |
786 | | re2::Prog* prog_; // compiled program for regexp |
787 | | |
788 | | // Reverse Prog for DFA execution only |
789 | | mutable re2::Prog* rprog_; |
790 | | // Map from capture names to indices |
791 | | mutable const std::map<std::string, int>* named_groups_; |
792 | | // Map from capture indices to names |
793 | | mutable const std::map<int, std::string>* group_names_; |
794 | | |
795 | | mutable std::once_flag rprog_once_; |
796 | | mutable std::once_flag named_groups_once_; |
797 | | mutable std::once_flag group_names_once_; |
798 | | }; |
799 | | |
800 | | /***** Implementation details *****/ |
801 | | |
802 | | namespace re2_internal { |
803 | | |
804 | | // Types for which the 3-ary Parse() function template has specializations. |
805 | | template <typename T> struct Parse3ary : public std::false_type {}; |
806 | | template <> struct Parse3ary<void> : public std::true_type {}; |
807 | | template <> struct Parse3ary<std::string> : public std::true_type {}; |
808 | | template <> struct Parse3ary<StringPiece> : public std::true_type {}; |
809 | | template <> struct Parse3ary<char> : public std::true_type {}; |
810 | | template <> struct Parse3ary<signed char> : public std::true_type {}; |
811 | | template <> struct Parse3ary<unsigned char> : public std::true_type {}; |
812 | | template <> struct Parse3ary<float> : public std::true_type {}; |
813 | | template <> struct Parse3ary<double> : public std::true_type {}; |
814 | | |
815 | | template <typename T> |
816 | | bool Parse(const char* str, size_t n, T* dest); |
817 | | |
818 | | // Types for which the 4-ary Parse() function template has specializations. |
819 | | template <typename T> struct Parse4ary : public std::false_type {}; |
820 | | template <> struct Parse4ary<long> : public std::true_type {}; |
821 | | template <> struct Parse4ary<unsigned long> : public std::true_type {}; |
822 | | template <> struct Parse4ary<short> : public std::true_type {}; |
823 | | template <> struct Parse4ary<unsigned short> : public std::true_type {}; |
824 | | template <> struct Parse4ary<int> : public std::true_type {}; |
825 | | template <> struct Parse4ary<unsigned int> : public std::true_type {}; |
826 | | template <> struct Parse4ary<long long> : public std::true_type {}; |
827 | | template <> struct Parse4ary<unsigned long long> : public std::true_type {}; |
828 | | |
829 | | template <typename T> |
830 | | bool Parse(const char* str, size_t n, T* dest, int radix); |
831 | | |
832 | | } // namespace re2_internal |
833 | | |
834 | | class RE2::Arg { |
835 | | private: |
836 | | template <typename T> |
837 | | using CanParse3ary = typename std::enable_if< |
838 | | re2_internal::Parse3ary<T>::value, |
839 | | int>::type; |
840 | | |
841 | | template <typename T> |
842 | | using CanParse4ary = typename std::enable_if< |
843 | | re2_internal::Parse4ary<T>::value, |
844 | | int>::type; |
845 | | |
846 | | #if !defined(_MSC_VER) |
847 | | template <typename T> |
848 | | using CanParseFrom = typename std::enable_if< |
849 | | std::is_member_function_pointer< |
850 | | decltype(static_cast<bool (T::*)(const char*, size_t)>( |
851 | | &T::ParseFrom))>::value, |
852 | | int>::type; |
853 | | #endif |
854 | | |
855 | | public: |
856 | 0 | Arg() : Arg(nullptr) {} |
857 | 0 | Arg(std::nullptr_t ptr) : arg_(ptr), parser_(DoNothing) {} |
858 | | |
859 | | template <typename T, CanParse3ary<T> = 0> |
860 | | Arg(T* ptr) : arg_(ptr), parser_(DoParse3ary<T>) {} |
861 | | |
862 | | template <typename T, CanParse4ary<T> = 0> |
863 | | Arg(T* ptr) : arg_(ptr), parser_(DoParse4ary<T>) {} |
864 | | |
865 | | #if !defined(_MSC_VER) |
866 | | template <typename T, CanParseFrom<T> = 0> |
867 | | Arg(T* ptr) : arg_(ptr), parser_(DoParseFrom<T>) {} |
868 | | #endif |
869 | | |
870 | | typedef bool (*Parser)(const char* str, size_t n, void* dest); |
871 | | |
872 | | template <typename T> |
873 | | Arg(T* ptr, Parser parser) : arg_(ptr), parser_(parser) {} |
874 | | |
875 | 0 | bool Parse(const char* str, size_t n) const { |
876 | 0 | return (*parser_)(str, n, arg_); |
877 | 0 | } |
878 | | |
879 | | private: |
880 | 0 | static bool DoNothing(const char* /*str*/, size_t /*n*/, void* /*dest*/) { |
881 | 0 | return true; |
882 | 0 | } |
883 | | |
884 | | template <typename T> |
885 | | static bool DoParse3ary(const char* str, size_t n, void* dest) { |
886 | | return re2_internal::Parse(str, n, reinterpret_cast<T*>(dest)); |
887 | | } |
888 | | |
889 | | template <typename T> |
890 | | static bool DoParse4ary(const char* str, size_t n, void* dest) { |
891 | | return re2_internal::Parse(str, n, reinterpret_cast<T*>(dest), 10); |
892 | | } |
893 | | |
894 | | #if !defined(_MSC_VER) |
895 | | template <typename T> |
896 | | static bool DoParseFrom(const char* str, size_t n, void* dest) { |
897 | | if (dest == NULL) return true; |
898 | | return reinterpret_cast<T*>(dest)->ParseFrom(str, n); |
899 | | } |
900 | | #endif |
901 | | |
902 | | void* arg_; |
903 | | Parser parser_; |
904 | | }; |
905 | | |
906 | | template <typename T> |
907 | | inline RE2::Arg RE2::CRadix(T* ptr) { |
908 | | return RE2::Arg(ptr, [](const char* str, size_t n, void* dest) -> bool { |
909 | | return re2_internal::Parse(str, n, reinterpret_cast<T*>(dest), 0); |
910 | | }); |
911 | | } |
912 | | |
913 | | template <typename T> |
914 | | inline RE2::Arg RE2::Hex(T* ptr) { |
915 | | return RE2::Arg(ptr, [](const char* str, size_t n, void* dest) -> bool { |
916 | | return re2_internal::Parse(str, n, reinterpret_cast<T*>(dest), 16); |
917 | | }); |
918 | | } |
919 | | |
920 | | template <typename T> |
921 | | inline RE2::Arg RE2::Octal(T* ptr) { |
922 | | return RE2::Arg(ptr, [](const char* str, size_t n, void* dest) -> bool { |
923 | | return re2_internal::Parse(str, n, reinterpret_cast<T*>(dest), 8); |
924 | | }); |
925 | | } |
926 | | |
927 | | // Silence warnings about missing initializers for members of LazyRE2. |
928 | | #if !defined(__clang__) && defined(__GNUC__) && __GNUC__ >= 6 |
929 | | #pragma GCC diagnostic ignored "-Wmissing-field-initializers" |
930 | | #endif |
931 | | |
932 | | // Helper for writing global or static RE2s safely. |
933 | | // Write |
934 | | // static LazyRE2 re = {".*"}; |
935 | | // and then use *re instead of writing |
936 | | // static RE2 re(".*"); |
937 | | // The former is more careful about multithreaded |
938 | | // situations than the latter. |
939 | | // |
940 | | // N.B. This class never deletes the RE2 object that |
941 | | // it constructs: that's a feature, so that it can be used |
942 | | // for global and function static variables. |
943 | | class LazyRE2 { |
944 | | private: |
945 | | struct NoArg {}; |
946 | | |
947 | | public: |
948 | | typedef RE2 element_type; // support std::pointer_traits |
949 | | |
950 | | // Constructor omitted to preserve braced initialization in C++98. |
951 | | |
952 | | // Pretend to be a pointer to Type (never NULL due to on-demand creation): |
953 | 0 | RE2& operator*() const { return *get(); } |
954 | 0 | RE2* operator->() const { return get(); } |
955 | | |
956 | | // Named accessor/initializer: |
957 | 0 | RE2* get() const { |
958 | 0 | std::call_once(once_, &LazyRE2::Init, this); |
959 | 0 | return ptr_; |
960 | 0 | } |
961 | | |
962 | | // All data fields must be public to support {"foo"} initialization. |
963 | | const char* pattern_; |
964 | | RE2::CannedOptions options_; |
965 | | NoArg barrier_against_excess_initializers_; |
966 | | |
967 | | mutable RE2* ptr_; |
968 | | mutable std::once_flag once_; |
969 | | |
970 | | private: |
971 | 0 | static void Init(const LazyRE2* lazy_re2) { |
972 | 0 | lazy_re2->ptr_ = new RE2(lazy_re2->pattern_, lazy_re2->options_); |
973 | 0 | } |
974 | | |
975 | | void operator=(const LazyRE2&); // disallowed |
976 | | }; |
977 | | |
978 | | namespace hooks { |
979 | | |
980 | | // Most platforms support thread_local. Older versions of iOS don't support |
981 | | // thread_local, but for the sake of brevity, we lump together all versions |
982 | | // of Apple platforms that aren't macOS. If an iOS application really needs |
983 | | // the context pointee someday, we can get more specific then... |
984 | | // |
985 | | // As per https://github.com/google/re2/issues/325, thread_local support in |
986 | | // MinGW seems to be buggy. (FWIW, Abseil folks also avoid it.) |
987 | | #define RE2_HAVE_THREAD_LOCAL |
988 | | #if (defined(__APPLE__) && !(defined(TARGET_OS_OSX) && TARGET_OS_OSX)) || defined(__MINGW32__) |
989 | | #undef RE2_HAVE_THREAD_LOCAL |
990 | | #endif |
991 | | |
992 | | // A hook must not make any assumptions regarding the lifetime of the context |
993 | | // pointee beyond the current invocation of the hook. Pointers and references |
994 | | // obtained via the context pointee should be considered invalidated when the |
995 | | // hook returns. Hence, any data about the context pointee (e.g. its pattern) |
996 | | // would have to be copied in order for it to be kept for an indefinite time. |
997 | | // |
998 | | // A hook must not use RE2 for matching. Control flow reentering RE2::Match() |
999 | | // could result in infinite mutual recursion. To discourage that possibility, |
1000 | | // RE2 will not maintain the context pointer correctly when used in that way. |
1001 | | #ifdef RE2_HAVE_THREAD_LOCAL |
1002 | | extern thread_local const RE2* context; |
1003 | | #endif |
1004 | | |
1005 | | struct DFAStateCacheReset { |
1006 | | int64_t state_budget; |
1007 | | size_t state_cache_size; |
1008 | | }; |
1009 | | |
1010 | | struct DFASearchFailure { |
1011 | | // Nothing yet... |
1012 | | }; |
1013 | | |
1014 | | #define DECLARE_HOOK(type) \ |
1015 | | using type##Callback = void(const type&); \ |
1016 | | void Set##type##Hook(type##Callback* cb); \ |
1017 | | type##Callback* Get##type##Hook(); |
1018 | | |
1019 | | DECLARE_HOOK(DFAStateCacheReset) |
1020 | | DECLARE_HOOK(DFASearchFailure) |
1021 | | |
1022 | | #undef DECLARE_HOOK |
1023 | | |
1024 | | } // namespace hooks |
1025 | | |
1026 | | } // namespace re2 |
1027 | | |
1028 | | using re2::RE2; |
1029 | | using re2::LazyRE2; |
1030 | | |
1031 | | #endif // RE2_RE2_H_ |