1PCRE2API(3) Library Functions Manual PCRE2API(3)
2
3
4
6 PCRE2 - Perl-compatible regular expressions (revised API)
7
8 #include <pcre2.h>
9
10 PCRE2 is a new API for PCRE, starting at release 10.0. This document
11 contains a description of all its native functions. See the pcre2 docu‐
12 ment for an overview of all the PCRE2 documentation.
13
15
16 pcre2_code *pcre2_compile(PCRE2_SPTR pattern, PCRE2_SIZE length,
17 uint32_t options, int *errorcode, PCRE2_SIZE *erroroffset,
18 pcre2_compile_context *ccontext);
19
20 void pcre2_code_free(pcre2_code *code);
21
22 pcre2_match_data *pcre2_match_data_create(uint32_t ovecsize,
23 pcre2_general_context *gcontext);
24
25 pcre2_match_data *pcre2_match_data_create_from_pattern(
26 const pcre2_code *code, pcre2_general_context *gcontext);
27
28 int pcre2_match(const pcre2_code *code, PCRE2_SPTR subject,
29 PCRE2_SIZE length, PCRE2_SIZE startoffset,
30 uint32_t options, pcre2_match_data *match_data,
31 pcre2_match_context *mcontext);
32
33 int pcre2_dfa_match(const pcre2_code *code, PCRE2_SPTR subject,
34 PCRE2_SIZE length, PCRE2_SIZE startoffset,
35 uint32_t options, pcre2_match_data *match_data,
36 pcre2_match_context *mcontext,
37 int *workspace, PCRE2_SIZE wscount);
38
39 void pcre2_match_data_free(pcre2_match_data *match_data);
40
42
43 PCRE2_SPTR pcre2_get_mark(pcre2_match_data *match_data);
44
45 uint32_t pcre2_get_ovector_count(pcre2_match_data *match_data);
46
47 PCRE2_SIZE *pcre2_get_ovector_pointer(pcre2_match_data *match_data);
48
49 PCRE2_SIZE pcre2_get_startchar(pcre2_match_data *match_data);
50
52
53 pcre2_general_context *pcre2_general_context_create(
54 void *(*private_malloc)(PCRE2_SIZE, void *),
55 void (*private_free)(void *, void *), void *memory_data);
56
57 pcre2_general_context *pcre2_general_context_copy(
58 pcre2_general_context *gcontext);
59
60 void pcre2_general_context_free(pcre2_general_context *gcontext);
61
63
64 pcre2_compile_context *pcre2_compile_context_create(
65 pcre2_general_context *gcontext);
66
67 pcre2_compile_context *pcre2_compile_context_copy(
68 pcre2_compile_context *ccontext);
69
70 void pcre2_compile_context_free(pcre2_compile_context *ccontext);
71
72 int pcre2_set_bsr(pcre2_compile_context *ccontext,
73 uint32_t value);
74
75 int pcre2_set_character_tables(pcre2_compile_context *ccontext,
76 const unsigned char *tables);
77
78 int pcre2_set_compile_extra_options(pcre2_compile_context *ccontext,
79 uint32_t extra_options);
80
81 int pcre2_set_max_pattern_length(pcre2_compile_context *ccontext,
82 PCRE2_SIZE value);
83
84 int pcre2_set_newline(pcre2_compile_context *ccontext,
85 uint32_t value);
86
87 int pcre2_set_parens_nest_limit(pcre2_compile_context *ccontext,
88 uint32_t value);
89
90 int pcre2_set_compile_recursion_guard(pcre2_compile_context *ccontext,
91 int (*guard_function)(uint32_t, void *), void *user_data);
92
94
95 pcre2_match_context *pcre2_match_context_create(
96 pcre2_general_context *gcontext);
97
98 pcre2_match_context *pcre2_match_context_copy(
99 pcre2_match_context *mcontext);
100
101 void pcre2_match_context_free(pcre2_match_context *mcontext);
102
103 int pcre2_set_callout(pcre2_match_context *mcontext,
104 int (*callout_function)(pcre2_callout_block *, void *),
105 void *callout_data);
106
107 int pcre2_set_offset_limit(pcre2_match_context *mcontext,
108 PCRE2_SIZE value);
109
110 int pcre2_set_heap_limit(pcre2_match_context *mcontext,
111 uint32_t value);
112
113 int pcre2_set_match_limit(pcre2_match_context *mcontext,
114 uint32_t value);
115
116 int pcre2_set_depth_limit(pcre2_match_context *mcontext,
117 uint32_t value);
118
120
121 int pcre2_substring_copy_byname(pcre2_match_data *match_data,
122 PCRE2_SPTR name, PCRE2_UCHAR *buffer, PCRE2_SIZE *bufflen);
123
124 int pcre2_substring_copy_bynumber(pcre2_match_data *match_data,
125 uint32_t number, PCRE2_UCHAR *buffer,
126 PCRE2_SIZE *bufflen);
127
128 void pcre2_substring_free(PCRE2_UCHAR *buffer);
129
130 int pcre2_substring_get_byname(pcre2_match_data *match_data,
131 PCRE2_SPTR name, PCRE2_UCHAR **bufferptr, PCRE2_SIZE *bufflen);
132
133 int pcre2_substring_get_bynumber(pcre2_match_data *match_data,
134 uint32_t number, PCRE2_UCHAR **bufferptr,
135 PCRE2_SIZE *bufflen);
136
137 int pcre2_substring_length_byname(pcre2_match_data *match_data,
138 PCRE2_SPTR name, PCRE2_SIZE *length);
139
140 int pcre2_substring_length_bynumber(pcre2_match_data *match_data,
141 uint32_t number, PCRE2_SIZE *length);
142
143 int pcre2_substring_nametable_scan(const pcre2_code *code,
144 PCRE2_SPTR name, PCRE2_SPTR *first, PCRE2_SPTR *last);
145
146 int pcre2_substring_number_from_name(const pcre2_code *code,
147 PCRE2_SPTR name);
148
149 void pcre2_substring_list_free(PCRE2_SPTR *list);
150
151 int pcre2_substring_list_get(pcre2_match_data *match_data,
152 PCRE2_UCHAR ***listptr, PCRE2_SIZE **lengthsptr);
153
155
156 int pcre2_substitute(const pcre2_code *code, PCRE2_SPTR subject,
157 PCRE2_SIZE length, PCRE2_SIZE startoffset,
158 uint32_t options, pcre2_match_data *match_data,
159 pcre2_match_context *mcontext, PCRE2_SPTR replacementzfP,
160 PCRE2_SIZE rlength, PCRE2_UCHAR *outputbuffer,
161 PCRE2_SIZE *outlengthptr);
162
164
165 int pcre2_jit_compile(pcre2_code *code, uint32_t options);
166
167 int pcre2_jit_match(const pcre2_code *code, PCRE2_SPTR subject,
168 PCRE2_SIZE length, PCRE2_SIZE startoffset,
169 uint32_t options, pcre2_match_data *match_data,
170 pcre2_match_context *mcontext);
171
172 void pcre2_jit_free_unused_memory(pcre2_general_context *gcontext);
173
174 pcre2_jit_stack *pcre2_jit_stack_create(PCRE2_SIZE startsize,
175 PCRE2_SIZE maxsize, pcre2_general_context *gcontext);
176
177 void pcre2_jit_stack_assign(pcre2_match_context *mcontext,
178 pcre2_jit_callback callback_function, void *callback_data);
179
180 void pcre2_jit_stack_free(pcre2_jit_stack *jit_stack);
181
183
184 int32_t pcre2_serialize_decode(pcre2_code **codes,
185 int32_t number_of_codes, const uint8_t *bytes,
186 pcre2_general_context *gcontext);
187
188 int32_t pcre2_serialize_encode(const pcre2_code **codes,
189 int32_t number_of_codes, uint8_t **serialized_bytes,
190 PCRE2_SIZE *serialized_size, pcre2_general_context *gcontext);
191
192 void pcre2_serialize_free(uint8_t *bytes);
193
194 int32_t pcre2_serialize_get_number_of_codes(const uint8_t *bytes);
195
197
198 pcre2_code *pcre2_code_copy(const pcre2_code *code);
199
200 pcre2_code *pcre2_code_copy_with_tables(const pcre2_code *code);
201
202 int pcre2_get_error_message(int errorcode, PCRE2_UCHAR *buffer,
203 PCRE2_SIZE bufflen);
204
205 const unsigned char *pcre2_maketables(pcre2_general_context *gcontext);
206
207 int pcre2_pattern_info(const pcre2 *code, uint32_t what, void *where);
208
209 int pcre2_callout_enumerate(const pcre2_code *code,
210 int (*callback)(pcre2_callout_enumerate_block *, void *),
211 void *user_data);
212
213 int pcre2_config(uint32_t what, void *where);
214
216
217 int pcre2_set_recursion_limit(pcre2_match_context *mcontext,
218 uint32_t value);
219
220 int pcre2_set_recursion_memory_management(
221 pcre2_match_context *mcontext,
222 void *(*private_malloc)(PCRE2_SIZE, void *),
223 void (*private_free)(void *, void *), void *memory_data);
224
225 These functions became obsolete at release 10.30 and are retained only
226 for backward compatibility. They should not be used in new code. The
227 first is replaced by pcre2_set_depth_limit(); the second is no longer
228 needed and has no effect (it always returns zero).
229
231
232 pcre2_convert_context *pcre2_convert_context_create(
233 pcre2_general_context *gcontext);
234
235 pcre2_convert_context *pcre2_convert_context_copy(
236 pcre2_convert_context *cvcontext);
237
238 void pcre2_convert_context_free(pcre2_convert_context *cvcontext);
239
240 int pcre2_set_glob_escape(pcre2_convert_context *cvcontext,
241 uint32_t escape_char);
242
243 int pcre2_set_glob_separator(pcre2_convert_context *cvcontext,
244 uint32_t separator_char);
245
246 int pcre2_pattern_convert(PCRE2_SPTR pattern, PCRE2_SIZE length,
247 uint32_t options, PCRE2_UCHAR **buffer,
248 PCRE2_SIZE *blength, pcre2_convert_context *cvcontext);
249
250 void pcre2_converted_pattern_free(PCRE2_UCHAR *converted_pattern);
251
252 These functions provide a way of converting non-PCRE2 patterns into
253 patterns that can be processed by pcre2_compile(). This facility is
254 experimental and may be changed in future releases. At present, "globs"
255 and POSIX basic and extended patterns can be converted. Details are
256 given in the pcre2convert documentation.
257
259
260 There are three PCRE2 libraries, supporting 8-bit, 16-bit, and 32-bit
261 code units, respectively. However, there is just one header file,
262 pcre2.h. This contains the function prototypes and other definitions
263 for all three libraries. One, two, or all three can be installed simul‐
264 taneously. On Unix-like systems the libraries are called libpcre2-8,
265 libpcre2-16, and libpcre2-32, and they can also co-exist with the orig‐
266 inal PCRE libraries.
267
268 Character strings are passed to and from a PCRE2 library as a sequence
269 of unsigned integers in code units of the appropriate width. Every
270 PCRE2 function comes in three different forms, one for each library,
271 for example:
272
273 pcre2_compile_8()
274 pcre2_compile_16()
275 pcre2_compile_32()
276
277 There are also three different sets of data types:
278
279 PCRE2_UCHAR8, PCRE2_UCHAR16, PCRE2_UCHAR32
280 PCRE2_SPTR8, PCRE2_SPTR16, PCRE2_SPTR32
281
282 The UCHAR types define unsigned code units of the appropriate widths.
283 For example, PCRE2_UCHAR16 is usually defined as `uint16_t'. The SPTR
284 types are constant pointers to the equivalent UCHAR types, that is,
285 they are pointers to vectors of unsigned code units.
286
287 Many applications use only one code unit width. For their convenience,
288 macros are defined whose names are the generic forms such as pcre2_com‐
289 pile() and PCRE2_SPTR. These macros use the value of the macro
290 PCRE2_CODE_UNIT_WIDTH to generate the appropriate width-specific func‐
291 tion and macro names. PCRE2_CODE_UNIT_WIDTH is not defined by default.
292 An application must define it to be 8, 16, or 32 before including
293 pcre2.h in order to make use of the generic names.
294
295 Applications that use more than one code unit width can be linked with
296 more than one PCRE2 library, but must define PCRE2_CODE_UNIT_WIDTH to
297 be 0 before including pcre2.h, and then use the real function names.
298 Any code that is to be included in an environment where the value of
299 PCRE2_CODE_UNIT_WIDTH is unknown should also use the real function
300 names. (Unfortunately, it is not possible in C code to save and restore
301 the value of a macro.)
302
303 If PCRE2_CODE_UNIT_WIDTH is not defined before including pcre2.h, a
304 compiler error occurs.
305
306 When using multiple libraries in an application, you must take care
307 when processing any particular pattern to use only functions from a
308 single library. For example, if you want to run a match using a pat‐
309 tern that was compiled with pcre2_compile_16(), you must do so with
310 pcre2_match_16(), not pcre2_match_8() or pcre2_match_32().
311
312 In the function summaries above, and in the rest of this document and
313 other PCRE2 documents, functions and data types are described using
314 their generic names, without the _8, _16, or _32 suffix.
315
317
318 PCRE2 has its own native API, which is described in this document.
319 There are also some wrapper functions for the 8-bit library that corre‐
320 spond to the POSIX regular expression API, but they do not give access
321 to all the functionality of PCRE2. They are described in the pcre2posix
322 documentation. Both these APIs define a set of C function calls.
323
324 The native API C data types, function prototypes, option values, and
325 error codes are defined in the header file pcre2.h, which also contains
326 definitions of PCRE2_MAJOR and PCRE2_MINOR, the major and minor release
327 numbers for the library. Applications can use these to include support
328 for different releases of PCRE2.
329
330 In a Windows environment, if you want to statically link an application
331 program against a non-dll PCRE2 library, you must define PCRE2_STATIC
332 before including pcre2.h.
333
334 The functions pcre2_compile() and pcre2_match() are used for compiling
335 and matching regular expressions in a Perl-compatible manner. A sample
336 program that demonstrates the simplest way of using them is provided in
337 the file called pcre2demo.c in the PCRE2 source distribution. A listing
338 of this program is given in the pcre2demo documentation, and the
339 pcre2sample documentation describes how to compile and run it.
340
341 The compiling and matching functions recognize various options that are
342 passed as bits in an options argument. There are also some more compli‐
343 cated parameters such as custom memory management functions and
344 resource limits that are passed in "contexts" (which are just memory
345 blocks, described below). Simple applications do not need to make use
346 of contexts.
347
348 Just-in-time (JIT) compiler support is an optional feature of PCRE2
349 that can be built in appropriate hardware environments. It greatly
350 speeds up the matching performance of many patterns. Programs can
351 request that it be used if available by calling pcre2_jit_compile()
352 after a pattern has been successfully compiled by pcre2_compile(). This
353 does nothing if JIT support is not available.
354
355 More complicated programs might need to make use of the specialist
356 functions pcre2_jit_stack_create(), pcre2_jit_stack_free(), and
357 pcre2_jit_stack_assign() in order to control the JIT code's memory
358 usage.
359
360 JIT matching is automatically used by pcre2_match() if it is available,
361 unless the PCRE2_NO_JIT option is set. There is also a direct interface
362 for JIT matching, which gives improved performance at the expense of
363 less sanity checking. The JIT-specific functions are discussed in the
364 pcre2jit documentation.
365
366 A second matching function, pcre2_dfa_match(), which is not Perl-com‐
367 patible, is also provided. This uses a different algorithm for the
368 matching. The alternative algorithm finds all possible matches (at a
369 given point in the subject), and scans the subject just once (unless
370 there are lookaround assertions). However, this algorithm does not
371 return captured substrings. A description of the two matching algo‐
372 rithms and their advantages and disadvantages is given in the
373 pcre2matching documentation. There is no JIT support for
374 pcre2_dfa_match().
375
376 In addition to the main compiling and matching functions, there are
377 convenience functions for extracting captured substrings from a subject
378 string that has been matched by pcre2_match(). They are:
379
380 pcre2_substring_copy_byname()
381 pcre2_substring_copy_bynumber()
382 pcre2_substring_get_byname()
383 pcre2_substring_get_bynumber()
384 pcre2_substring_list_get()
385 pcre2_substring_length_byname()
386 pcre2_substring_length_bynumber()
387 pcre2_substring_nametable_scan()
388 pcre2_substring_number_from_name()
389
390 pcre2_substring_free() and pcre2_substring_list_free() are also pro‐
391 vided, to free memory used for extracted strings. If either of these
392 functions is called with a NULL argument, the function returns immedi‐
393 ately without doing anything.
394
395 The function pcre2_substitute() can be called to match a pattern and
396 return a copy of the subject string with substitutions for parts that
397 were matched.
398
399 Functions whose names begin with pcre2_serialize_ are used for saving
400 compiled patterns on disc or elsewhere, and reloading them later.
401
402 Finally, there are functions for finding out information about a com‐
403 piled pattern (pcre2_pattern_info()) and about the configuration with
404 which PCRE2 was built (pcre2_config()).
405
406 Functions with names ending with _free() are used for freeing memory
407 blocks of various sorts. In all cases, if one of these functions is
408 called with a NULL argument, it does nothing.
409
411
412 The PCRE2 API uses string lengths and offsets into strings of code
413 units in several places. These values are always of type PCRE2_SIZE,
414 which is an unsigned integer type, currently always defined as size_t.
415 The largest value that can be stored in such a type (that is
416 ~(PCRE2_SIZE)0) is reserved as a special indicator for zero-terminated
417 strings and unset offsets. Therefore, the longest string that can be
418 handled is one less than this maximum.
419
421
422 PCRE2 supports five different conventions for indicating line breaks in
423 strings: a single CR (carriage return) character, a single LF (line‐
424 feed) character, the two-character sequence CRLF, any of the three pre‐
425 ceding, or any Unicode newline sequence. The Unicode newline sequences
426 are the three just mentioned, plus the single characters VT (vertical
427 tab, U+000B), FF (form feed, U+000C), NEL (next line, U+0085), LS (line
428 separator, U+2028), and PS (paragraph separator, U+2029).
429
430 Each of the first three conventions is used by at least one operating
431 system as its standard newline sequence. When PCRE2 is built, a default
432 can be specified. If it is not, the default is set to LF, which is the
433 Unix standard. However, the newline convention can be changed by an
434 application when calling pcre2_compile(), or it can be specified by
435 special text at the start of the pattern itself; this overrides any
436 other settings. See the pcre2pattern page for details of the special
437 character sequences.
438
439 In the PCRE2 documentation the word "newline" is used to mean "the
440 character or pair of characters that indicate a line break". The choice
441 of newline convention affects the handling of the dot, circumflex, and
442 dollar metacharacters, the handling of #-comments in /x mode, and, when
443 CRLF is a recognized line ending sequence, the match position advance‐
444 ment for a non-anchored pattern. There is more detail about this in the
445 section on pcre2_match() options below.
446
447 The choice of newline convention does not affect the interpretation of
448 the \n or \r escape sequences, nor does it affect what \R matches; this
449 has its own separate convention.
450
452
453 In a multithreaded application it is important to keep thread-specific
454 data separate from data that can be shared between threads. The PCRE2
455 library code itself is thread-safe: it contains no static or global
456 variables. The API is designed to be fairly simple for non-threaded
457 applications while at the same time ensuring that multithreaded appli‐
458 cations can use it.
459
460 There are several different blocks of data that are used to pass infor‐
461 mation between the application and the PCRE2 libraries.
462
463 The compiled pattern
464
465 A pointer to the compiled form of a pattern is returned to the user
466 when pcre2_compile() is successful. The data in the compiled pattern is
467 fixed, and does not change when the pattern is matched. Therefore, it
468 is thread-safe, that is, the same compiled pattern can be used by more
469 than one thread simultaneously. For example, an application can compile
470 all its patterns at the start, before forking off multiple threads that
471 use them. However, if the just-in-time (JIT) optimization feature is
472 being used, it needs separate memory stack areas for each thread. See
473 the pcre2jit documentation for more details.
474
475 In a more complicated situation, where patterns are compiled only when
476 they are first needed, but are still shared between threads, pointers
477 to compiled patterns must be protected from simultaneous writing by
478 multiple threads, at least until a pattern has been compiled. The logic
479 can be something like this:
480
481 Get a read-only (shared) lock (mutex) for pointer
482 if (pointer == NULL)
483 {
484 Get a write (unique) lock for pointer
485 pointer = pcre2_compile(...
486 }
487 Release the lock
488 Use pointer in pcre2_match()
489
490 Of course, testing for compilation errors should also be included in
491 the code.
492
493 If JIT is being used, but the JIT compilation is not being done immedi‐
494 ately, (perhaps waiting to see if the pattern is used often enough)
495 similar logic is required. JIT compilation updates a pointer within the
496 compiled code block, so a thread must gain unique write access to the
497 pointer before calling pcre2_jit_compile(). Alternatively,
498 pcre2_code_copy() or pcre2_code_copy_with_tables() can be used to
499 obtain a private copy of the compiled code before calling the JIT com‐
500 piler.
501
502 Context blocks
503
504 The next main section below introduces the idea of "contexts" in which
505 PCRE2 functions are called. A context is nothing more than a collection
506 of parameters that control the way PCRE2 operates. Grouping a number of
507 parameters together in a context is a convenient way of passing them to
508 a PCRE2 function without using lots of arguments. The parameters that
509 are stored in contexts are in some sense "advanced features" of the
510 API. Many straightforward applications will not need to use contexts.
511
512 In a multithreaded application, if the parameters in a context are val‐
513 ues that are never changed, the same context can be used by all the
514 threads. However, if any thread needs to change any value in a context,
515 it must make its own thread-specific copy.
516
517 Match blocks
518
519 The matching functions need a block of memory for storing the results
520 of a match. This includes details of what was matched, as well as addi‐
521 tional information such as the name of a (*MARK) setting. Each thread
522 must provide its own copy of this memory.
523
525
526 Some PCRE2 functions have a lot of parameters, many of which are used
527 only by specialist applications, for example, those that use custom
528 memory management or non-standard character tables. To keep function
529 argument lists at a reasonable size, and at the same time to keep the
530 API extensible, "uncommon" parameters are passed to certain functions
531 in a context instead of directly. A context is just a block of memory
532 that holds the parameter values. Applications that do not need to
533 adjust any of the context parameters can pass NULL when a context
534 pointer is required.
535
536 There are three different types of context: a general context that is
537 relevant for several PCRE2 operations, a compile-time context, and a
538 match-time context.
539
540 The general context
541
542 At present, this context just contains pointers to (and data for)
543 external memory management functions that are called from several
544 places in the PCRE2 library. The context is named `general' rather than
545 specifically `memory' because in future other fields may be added. If
546 you do not want to supply your own custom memory management functions,
547 you do not need to bother with a general context. A general context is
548 created by:
549
550 pcre2_general_context *pcre2_general_context_create(
551 void *(*private_malloc)(PCRE2_SIZE, void *),
552 void (*private_free)(void *, void *), void *memory_data);
553
554 The two function pointers specify custom memory management functions,
555 whose prototypes are:
556
557 void *private_malloc(PCRE2_SIZE, void *);
558 void private_free(void *, void *);
559
560 Whenever code in PCRE2 calls these functions, the final argument is the
561 value of memory_data. Either of the first two arguments of the creation
562 function may be NULL, in which case the system memory management func‐
563 tions malloc() and free() are used. (This is not currently useful, as
564 there are no other fields in a general context, but in future there
565 might be.) The private_malloc() function is used (if supplied) to
566 obtain memory for storing the context, and all three values are saved
567 as part of the context.
568
569 Whenever PCRE2 creates a data block of any kind, the block contains a
570 pointer to the free() function that matches the malloc() function that
571 was used. When the time comes to free the block, this function is
572 called.
573
574 A general context can be copied by calling:
575
576 pcre2_general_context *pcre2_general_context_copy(
577 pcre2_general_context *gcontext);
578
579 The memory used for a general context should be freed by calling:
580
581 void pcre2_general_context_free(pcre2_general_context *gcontext);
582
583 If this function is passed a NULL argument, it returns immediately
584 without doing anything.
585
586 The compile context
587
588 A compile context is required if you want to provide an external func‐
589 tion for stack checking during compilation or to change the default
590 values of any of the following compile-time parameters:
591
592 What \R matches (Unicode newlines or CR, LF, CRLF only)
593 PCRE2's character tables
594 The newline character sequence
595 The compile time nested parentheses limit
596 The maximum length of the pattern string
597 The extra options bits (none set by default)
598
599 A compile context is also required if you are using custom memory man‐
600 agement. If none of these apply, just pass NULL as the context argu‐
601 ment of pcre2_compile().
602
603 A compile context is created, copied, and freed by the following func‐
604 tions:
605
606 pcre2_compile_context *pcre2_compile_context_create(
607 pcre2_general_context *gcontext);
608
609 pcre2_compile_context *pcre2_compile_context_copy(
610 pcre2_compile_context *ccontext);
611
612 void pcre2_compile_context_free(pcre2_compile_context *ccontext);
613
614 A compile context is created with default values for its parameters.
615 These can be changed by calling the following functions, which return 0
616 on success, or PCRE2_ERROR_BADDATA if invalid data is detected.
617
618 int pcre2_set_bsr(pcre2_compile_context *ccontext,
619 uint32_t value);
620
621 The value must be PCRE2_BSR_ANYCRLF, to specify that \R matches only
622 CR, LF, or CRLF, or PCRE2_BSR_UNICODE, to specify that \R matches any
623 Unicode line ending sequence. The value is used by the JIT compiler and
624 by the two interpreted matching functions, pcre2_match() and
625 pcre2_dfa_match().
626
627 int pcre2_set_character_tables(pcre2_compile_context *ccontext,
628 const unsigned char *tables);
629
630 The value must be the result of a call to pcre2_maketables(), whose
631 only argument is a general context. This function builds a set of char‐
632 acter tables in the current locale.
633
634 int pcre2_set_compile_extra_options(pcre2_compile_context *ccontext,
635 uint32_t extra_options);
636
637 As PCRE2 has developed, almost all the 32 option bits that are avail‐
638 able in the options argument of pcre2_compile() have been used up. To
639 avoid running out, the compile context contains a set of extra option
640 bits which are used for some newer, assumed rarer, options. This func‐
641 tion sets those bits. It always sets all the bits (either on or off).
642 It does not modify any existing setting. The available options are
643 defined in the section entitled "Extra compile options" below.
644
645 int pcre2_set_max_pattern_length(pcre2_compile_context *ccontext,
646 PCRE2_SIZE value);
647
648 This sets a maximum length, in code units, for any pattern string that
649 is compiled with this context. If the pattern is longer, an error is
650 generated. This facility is provided so that applications that accept
651 patterns from external sources can limit their size. The default is the
652 largest number that a PCRE2_SIZE variable can hold, which is effec‐
653 tively unlimited.
654
655 int pcre2_set_newline(pcre2_compile_context *ccontext,
656 uint32_t value);
657
658 This specifies which characters or character sequences are to be recog‐
659 nized as newlines. The value must be one of PCRE2_NEWLINE_CR (carriage
660 return only), PCRE2_NEWLINE_LF (linefeed only), PCRE2_NEWLINE_CRLF (the
661 two-character sequence CR followed by LF), PCRE2_NEWLINE_ANYCRLF (any
662 of the above), PCRE2_NEWLINE_ANY (any Unicode newline sequence), or
663 PCRE2_NEWLINE_NUL (the NUL character, that is a binary zero).
664
665 A pattern can override the value set in the compile context by starting
666 with a sequence such as (*CRLF). See the pcre2pattern page for details.
667
668 When a pattern is compiled with the PCRE2_EXTENDED or
669 PCRE2_EXTENDED_MORE option, the newline convention affects the recogni‐
670 tion of the end of internal comments starting with #. The value is
671 saved with the compiled pattern for subsequent use by the JIT compiler
672 and by the two interpreted matching functions, pcre2_match() and
673 pcre2_dfa_match().
674
675 int pcre2_set_parens_nest_limit(pcre2_compile_context *ccontext,
676 uint32_t value);
677
678 This parameter ajusts the limit, set when PCRE2 is built (default 250),
679 on the depth of parenthesis nesting in a pattern. This limit stops
680 rogue patterns using up too much system stack when being compiled. The
681 limit applies to parentheses of all kinds, not just capturing parenthe‐
682 ses.
683
684 int pcre2_set_compile_recursion_guard(pcre2_compile_context *ccontext,
685 int (*guard_function)(uint32_t, void *), void *user_data);
686
687 There is at least one application that runs PCRE2 in threads with very
688 limited system stack, where running out of stack is to be avoided at
689 all costs. The parenthesis limit above cannot take account of how much
690 stack is actually available during compilation. For a finer control,
691 you can supply a function that is called whenever pcre2_compile()
692 starts to compile a parenthesized part of a pattern. This function can
693 check the actual stack size (or anything else that it wants to, of
694 course).
695
696 The first argument to the callout function gives the current depth of
697 nesting, and the second is user data that is set up by the last argu‐
698 ment of pcre2_set_compile_recursion_guard(). The callout function
699 should return zero if all is well, or non-zero to force an error.
700
701 The match context
702
703 A match context is required if you want to:
704
705 Set up a callout function
706 Set an offset limit for matching an unanchored pattern
707 Change the limit on the amount of heap used when matching
708 Change the backtracking match limit
709 Change the backtracking depth limit
710 Set custom memory management specifically for the match
711
712 If none of these apply, just pass NULL as the context argument of
713 pcre2_match(), pcre2_dfa_match(), or pcre2_jit_match().
714
715 A match context is created, copied, and freed by the following func‐
716 tions:
717
718 pcre2_match_context *pcre2_match_context_create(
719 pcre2_general_context *gcontext);
720
721 pcre2_match_context *pcre2_match_context_copy(
722 pcre2_match_context *mcontext);
723
724 void pcre2_match_context_free(pcre2_match_context *mcontext);
725
726 A match context is created with default values for its parameters.
727 These can be changed by calling the following functions, which return 0
728 on success, or PCRE2_ERROR_BADDATA if invalid data is detected.
729
730 int pcre2_set_callout(pcre2_match_context *mcontext,
731 int (*callout_function)(pcre2_callout_block *, void *),
732 void *callout_data);
733
734 This sets up a "callout" function for PCRE2 to call at specified points
735 during a matching operation. Details are given in the pcre2callout doc‐
736 umentation.
737
738 int pcre2_set_offset_limit(pcre2_match_context *mcontext,
739 PCRE2_SIZE value);
740
741 The offset_limit parameter limits how far an unanchored search can
742 advance in the subject string. The default value is PCRE2_UNSET. The
743 pcre2_match() and pcre2_dfa_match() functions return
744 PCRE2_ERROR_NOMATCH if a match with a starting point before or at the
745 given offset is not found. The pcre2_substitute() function makes no
746 more substitutions.
747
748 For example, if the pattern /abc/ is matched against "123abc" with an
749 offset limit less than 3, the result is PCRE2_ERROR_NO_MATCH. A match
750 can never be found if the startoffset argument of pcre2_match(),
751 pcre2_dfa_match(), or pcre2_substitute() is greater than the offset
752 limit set in the match context.
753
754 When using this facility, you must set the PCRE2_USE_OFFSET_LIMIT
755 option when calling pcre2_compile() so that when JIT is in use, differ‐
756 ent code can be compiled. If a match is started with a non-default
757 match limit when PCRE2_USE_OFFSET_LIMIT is not set, an error is gener‐
758 ated.
759
760 The offset limit facility can be used to track progress when searching
761 large subject strings or to limit the extent of global substitutions.
762 See also the PCRE2_FIRSTLINE option, which requires a match to start
763 before or at the first newline that follows the start of matching in
764 the subject. If this is set with an offset limit, a match must occur in
765 the first line and also within the offset limit. In other words, which‐
766 ever limit comes first is used.
767
768 int pcre2_set_heap_limit(pcre2_match_context *mcontext,
769 uint32_t value);
770
771 The heap_limit parameter specifies, in units of kibibytes (1024 bytes),
772 the maximum amount of heap memory that pcre2_match() may use to hold
773 backtracking information when running an interpretive match. This limit
774 also applies to pcre2_dfa_match(), which may use the heap when process‐
775 ing patterns with a lot of nested pattern recursion or lookarounds or
776 atomic groups. This limit does not apply to matching with the JIT opti‐
777 mization, which has its own memory control arrangements (see the
778 pcre2jit documentation for more details). If the limit is reached, the
779 negative error code PCRE2_ERROR_HEAPLIMIT is returned. The default
780 limit can be set when PCRE2 is built; if it is not, the default is set
781 very large and is essentially "unlimited".
782
783 A value for the heap limit may also be supplied by an item at the start
784 of a pattern of the form
785
786 (*LIMIT_HEAP=ddd)
787
788 where ddd is a decimal number. However, such a setting is ignored
789 unless ddd is less than the limit set by the caller of pcre2_match()
790 or, if no such limit is set, less than the default.
791
792 The pcre2_match() function starts out using a 20KiB vector on the sys‐
793 tem stack for recording backtracking points. The more nested backtrack‐
794 ing points there are (that is, the deeper the search tree), the more
795 memory is needed. Heap memory is used only if the initial vector is
796 too small. If the heap limit is set to a value less than 21 (in partic‐
797 ular, zero) no heap memory will be used. In this case, only patterns
798 that do not have a lot of nested backtracking can be successfully pro‐
799 cessed.
800
801 Similarly, for pcre2_dfa_match(), a vector on the system stack is used
802 when processing pattern recursions, lookarounds, or atomic groups, and
803 only if this is not big enough is heap memory used. In this case, too,
804 setting a value of zero disables the use of the heap.
805
806 int pcre2_set_match_limit(pcre2_match_context *mcontext,
807 uint32_t value);
808
809 The match_limit parameter provides a means of preventing PCRE2 from
810 using up too many computing resources when processing patterns that are
811 not going to match, but which have a very large number of possibilities
812 in their search trees. The classic example is a pattern that uses
813 nested unlimited repeats.
814
815 There is an internal counter in pcre2_match() that is incremented each
816 time round its main matching loop. If this value reaches the match
817 limit, pcre2_match() returns the negative value PCRE2_ERROR_MATCHLIMIT.
818 This has the effect of limiting the amount of backtracking that can
819 take place. For patterns that are not anchored, the count restarts from
820 zero for each position in the subject string. This limit also applies
821 to pcre2_dfa_match(), though the counting is done in a different way.
822
823 When pcre2_match() is called with a pattern that was successfully pro‐
824 cessed by pcre2_jit_compile(), the way in which matching is executed is
825 entirely different. However, there is still the possibility of runaway
826 matching that goes on for a very long time, and so the match_limit
827 value is also used in this case (but in a different way) to limit how
828 long the matching can continue.
829
830 The default value for the limit can be set when PCRE2 is built; the
831 default default is 10 million, which handles all but the most extreme
832 cases. A value for the match limit may also be supplied by an item at
833 the start of a pattern of the form
834
835 (*LIMIT_MATCH=ddd)
836
837 where ddd is a decimal number. However, such a setting is ignored
838 unless ddd is less than the limit set by the caller of pcre2_match() or
839 pcre2_dfa_match() or, if no such limit is set, less than the default.
840
841 int pcre2_set_depth_limit(pcre2_match_context *mcontext,
842 uint32_t value);
843
844 This parameter limits the depth of nested backtracking in
845 pcre2_match(). Each time a nested backtracking point is passed, a new
846 memory "frame" is used to remember the state of matching at that point.
847 Thus, this parameter indirectly limits the amount of memory that is
848 used in a match. However, because the size of each memory "frame"
849 depends on the number of capturing parentheses, the actual memory limit
850 varies from pattern to pattern. This limit was more useful in versions
851 before 10.30, where function recursion was used for backtracking.
852
853 The depth limit is not relevant, and is ignored, when matching is done
854 using JIT compiled code. However, it is supported by pcre2_dfa_match(),
855 which uses it to limit the depth of nested internal recursive function
856 calls that implement atomic groups, lookaround assertions, and pattern
857 recursions. This limits, indirectly, the amount of system stack that is
858 used. It was more useful in versions before 10.32, when stack memory
859 was used for local workspace vectors for recursive function calls. From
860 version 10.32, only local variables are allocated on the stack and as
861 each call uses only a few hundred bytes, even a small stack can support
862 quite a lot of recursion.
863
864 If the depth of internal recursive function calls is great enough,
865 local workspace vectors are allocated on the heap from version 10.32
866 onwards, so the depth limit also indirectly limits the amount of heap
867 memory that is used. A recursive pattern such as /(.(?2))((?1)|)/, when
868 matched to a very long string using pcre2_dfa_match(), can use a great
869 deal of memory. However, it is probably better to limit heap usage
870 directly by calling pcre2_set_heap_limit().
871
872 The default value for the depth limit can be set when PCRE2 is built;
873 if it is not, the default is set to the same value as the default for
874 the match limit. If the limit is exceeded, pcre2_match() or
875 pcre2_dfa_match() returns PCRE2_ERROR_DEPTHLIMIT. A value for the depth
876 limit may also be supplied by an item at the start of a pattern of the
877 form
878
879 (*LIMIT_DEPTH=ddd)
880
881 where ddd is a decimal number. However, such a setting is ignored
882 unless ddd is less than the limit set by the caller of pcre2_match() or
883 pcre2_dfa_match() or, if no such limit is set, less than the default.
884
886
887 int pcre2_config(uint32_t what, void *where);
888
889 The function pcre2_config() makes it possible for a PCRE2 client to
890 discover which optional features have been compiled into the PCRE2
891 library. The pcre2build documentation has more details about these
892 optional features.
893
894 The first argument for pcre2_config() specifies which information is
895 required. The second argument is a pointer to memory into which the
896 information is placed. If NULL is passed, the function returns the
897 amount of memory that is needed for the requested information. For
898 calls that return numerical values, the value is in bytes; when
899 requesting these values, where should point to appropriately aligned
900 memory. For calls that return strings, the required length is given in
901 code units, not counting the terminating zero.
902
903 When requesting information, the returned value from pcre2_config() is
904 non-negative on success, or the negative error code PCRE2_ERROR_BADOP‐
905 TION if the value in the first argument is not recognized. The follow‐
906 ing information is available:
907
908 PCRE2_CONFIG_BSR
909
910 The output is a uint32_t integer whose value indicates what character
911 sequences the \R escape sequence matches by default. A value of
912 PCRE2_BSR_UNICODE means that \R matches any Unicode line ending
913 sequence; a value of PCRE2_BSR_ANYCRLF means that \R matches only CR,
914 LF, or CRLF. The default can be overridden when a pattern is compiled.
915
916 PCRE2_CONFIG_COMPILED_WIDTHS
917
918 The output is a uint32_t integer whose lower bits indicate which code
919 unit widths were selected when PCRE2 was built. The 1-bit indicates
920 8-bit support, and the 2-bit and 4-bit indicate 16-bit and 32-bit sup‐
921 port, respectively.
922
923 PCRE2_CONFIG_DEPTHLIMIT
924
925 The output is a uint32_t integer that gives the default limit for the
926 depth of nested backtracking in pcre2_match() or the depth of nested
927 recursions, lookarounds, and atomic groups in pcre2_dfa_match(). Fur‐
928 ther details are given with pcre2_set_depth_limit() above.
929
930 PCRE2_CONFIG_HEAPLIMIT
931
932 The output is a uint32_t integer that gives, in kibibytes, the default
933 limit for the amount of heap memory used by pcre2_match() or
934 pcre2_dfa_match(). Further details are given with
935 pcre2_set_heap_limit() above.
936
937 PCRE2_CONFIG_JIT
938
939 The output is a uint32_t integer that is set to one if support for
940 just-in-time compiling is available; otherwise it is set to zero.
941
942 PCRE2_CONFIG_JITTARGET
943
944 The where argument should point to a buffer that is at least 48 code
945 units long. (The exact length required can be found by calling
946 pcre2_config() with where set to NULL.) The buffer is filled with a
947 string that contains the name of the architecture for which the JIT
948 compiler is configured, for example "x86 32bit (little endian +
949 unaligned)". If JIT support is not available, PCRE2_ERROR_BADOPTION is
950 returned, otherwise the number of code units used is returned. This is
951 the length of the string, plus one unit for the terminating zero.
952
953 PCRE2_CONFIG_LINKSIZE
954
955 The output is a uint32_t integer that contains the number of bytes used
956 for internal linkage in compiled regular expressions. When PCRE2 is
957 configured, the value can be set to 2, 3, or 4, with the default being
958 2. This is the value that is returned by pcre2_config(). However, when
959 the 16-bit library is compiled, a value of 3 is rounded up to 4, and
960 when the 32-bit library is compiled, internal linkages always use 4
961 bytes, so the configured value is not relevant.
962
963 The default value of 2 for the 8-bit and 16-bit libraries is sufficient
964 for all but the most massive patterns, since it allows the size of the
965 compiled pattern to be up to 65535 code units. Larger values allow
966 larger regular expressions to be compiled by those two libraries, but
967 at the expense of slower matching.
968
969 PCRE2_CONFIG_MATCHLIMIT
970
971 The output is a uint32_t integer that gives the default match limit for
972 pcre2_match(). Further details are given with pcre2_set_match_limit()
973 above.
974
975 PCRE2_CONFIG_NEWLINE
976
977 The output is a uint32_t integer whose value specifies the default
978 character sequence that is recognized as meaning "newline". The values
979 are:
980
981 PCRE2_NEWLINE_CR Carriage return (CR)
982 PCRE2_NEWLINE_LF Linefeed (LF)
983 PCRE2_NEWLINE_CRLF Carriage return, linefeed (CRLF)
984 PCRE2_NEWLINE_ANY Any Unicode line ending
985 PCRE2_NEWLINE_ANYCRLF Any of CR, LF, or CRLF
986 PCRE2_NEWLINE_NUL The NUL character (binary zero)
987
988 The default should normally correspond to the standard sequence for
989 your operating system.
990
991 PCRE2_CONFIG_NEVER_BACKSLASH_C
992
993 The output is a uint32_t integer that is set to one if the use of \C
994 was permanently disabled when PCRE2 was built; otherwise it is set to
995 zero.
996
997 PCRE2_CONFIG_PARENSLIMIT
998
999 The output is a uint32_t integer that gives the maximum depth of nest‐
1000 ing of parentheses (of any kind) in a pattern. This limit is imposed to
1001 cap the amount of system stack used when a pattern is compiled. It is
1002 specified when PCRE2 is built; the default is 250. This limit does not
1003 take into account the stack that may already be used by the calling
1004 application. For finer control over compilation stack usage, see
1005 pcre2_set_compile_recursion_guard().
1006
1007 PCRE2_CONFIG_STACKRECURSE
1008
1009 This parameter is obsolete and should not be used in new code. The out‐
1010 put is a uint32_t integer that is always set to zero.
1011
1012 PCRE2_CONFIG_UNICODE_VERSION
1013
1014 The where argument should point to a buffer that is at least 24 code
1015 units long. (The exact length required can be found by calling
1016 pcre2_config() with where set to NULL.) If PCRE2 has been compiled
1017 without Unicode support, the buffer is filled with the text "Unicode
1018 not supported". Otherwise, the Unicode version string (for example,
1019 "8.0.0") is inserted. The number of code units used is returned. This
1020 is the length of the string plus one unit for the terminating zero.
1021
1022 PCRE2_CONFIG_UNICODE
1023
1024 The output is a uint32_t integer that is set to one if Unicode support
1025 is available; otherwise it is set to zero. Unicode support implies UTF
1026 support.
1027
1028 PCRE2_CONFIG_VERSION
1029
1030 The where argument should point to a buffer that is at least 24 code
1031 units long. (The exact length required can be found by calling
1032 pcre2_config() with where set to NULL.) The buffer is filled with the
1033 PCRE2 version string, zero-terminated. The number of code units used is
1034 returned. This is the length of the string plus one unit for the termi‐
1035 nating zero.
1036
1038
1039 pcre2_code *pcre2_compile(PCRE2_SPTR pattern, PCRE2_SIZE length,
1040 uint32_t options, int *errorcode, PCRE2_SIZE *erroroffset,
1041 pcre2_compile_context *ccontext);
1042
1043 void pcre2_code_free(pcre2_code *code);
1044
1045 pcre2_code *pcre2_code_copy(const pcre2_code *code);
1046
1047 pcre2_code *pcre2_code_copy_with_tables(const pcre2_code *code);
1048
1049 The pcre2_compile() function compiles a pattern into an internal form.
1050 The pattern is defined by a pointer to a string of code units and a
1051 length (in code units). If the pattern is zero-terminated, the length
1052 can be specified as PCRE2_ZERO_TERMINATED. The function returns a
1053 pointer to a block of memory that contains the compiled pattern and
1054 related data, or NULL if an error occurred.
1055
1056 If the compile context argument ccontext is NULL, memory for the com‐
1057 piled pattern is obtained by calling malloc(). Otherwise, it is
1058 obtained from the same memory function that was used for the compile
1059 context. The caller must free the memory by calling pcre2_code_free()
1060 when it is no longer needed. If pcre2_code_free() is called with a
1061 NULL argument, it returns immediately, without doing anything.
1062
1063 The function pcre2_code_copy() makes a copy of the compiled code in new
1064 memory, using the same memory allocator as was used for the original.
1065 However, if the code has been processed by the JIT compiler (see
1066 below), the JIT information cannot be copied (because it is position-
1067 dependent). The new copy can initially be used only for non-JIT match‐
1068 ing, though it can be passed to pcre2_jit_compile() if required. If
1069 pcre2_code_copy() is called with a NULL argument, it returns NULL.
1070
1071 The pcre2_code_copy() function provides a way for individual threads in
1072 a multithreaded application to acquire a private copy of shared com‐
1073 piled code. However, it does not make a copy of the character tables
1074 used by the compiled pattern; the new pattern code points to the same
1075 tables as the original code. (See "Locale Support" below for details
1076 of these character tables.) In many applications the same tables are
1077 used throughout, so this behaviour is appropriate. Nevertheless, there
1078 are occasions when a copy of a compiled pattern and the relevant tables
1079 are needed. The pcre2_code_copy_with_tables() provides this facility.
1080 Copies of both the code and the tables are made, with the new code
1081 pointing to the new tables. The memory for the new tables is automati‐
1082 cally freed when pcre2_code_free() is called for the new copy of the
1083 compiled code. If pcre2_code_copy_withy_tables() is called with a NULL
1084 argument, it returns NULL.
1085
1086 NOTE: When one of the matching functions is called, pointers to the
1087 compiled pattern and the subject string are set in the match data block
1088 so that they can be referenced by the substring extraction functions.
1089 After running a match, you must not free a compiled pattern (or a sub‐
1090 ject string) until after all operations on the match data block have
1091 taken place.
1092
1093 The options argument for pcre2_compile() contains various bit settings
1094 that affect the compilation. It should be zero if no options are
1095 required. The available options are described below. Some of them (in
1096 particular, those that are compatible with Perl, but some others as
1097 well) can also be set and unset from within the pattern (see the
1098 detailed description in the pcre2pattern documentation).
1099
1100 For those options that can be different in different parts of the pat‐
1101 tern, the contents of the options argument specifies their settings at
1102 the start of compilation. The PCRE2_ANCHORED, PCRE2_ENDANCHORED, and
1103 PCRE2_NO_UTF_CHECK options can be set at the time of matching as well
1104 as at compile time.
1105
1106 Other, less frequently required compile-time parameters (for example,
1107 the newline setting) can be provided in a compile context (as described
1108 above).
1109
1110 If errorcode or erroroffset is NULL, pcre2_compile() returns NULL imme‐
1111 diately. Otherwise, the variables to which these point are set to an
1112 error code and an offset (number of code units) within the pattern,
1113 respectively, when pcre2_compile() returns NULL because a compilation
1114 error has occurred. The values are not defined when compilation is suc‐
1115 cessful and pcre2_compile() returns a non-NULL value.
1116
1117 There are nearly 100 positive error codes that pcre2_compile() may
1118 return if it finds an error in the pattern. There are also some nega‐
1119 tive error codes that are used for invalid UTF strings. These are the
1120 same as given by pcre2_match() and pcre2_dfa_match(), and are described
1121 in the pcre2unicode page. There is no separate documentation for the
1122 positive error codes, because the textual error messages that are
1123 obtained by calling the pcre2_get_error_message() function (see
1124 "Obtaining a textual error message" below) should be self-explanatory.
1125 Macro names starting with PCRE2_ERROR_ are defined for both positive
1126 and negative error codes in pcre2.h.
1127
1128 The value returned in erroroffset is an indication of where in the pat‐
1129 tern the error occurred. It is not necessarily the furthest point in
1130 the pattern that was read. For example, after the error "lookbehind
1131 assertion is not fixed length", the error offset points to the start of
1132 the failing assertion. For an invalid UTF-8 or UTF-16 string, the off‐
1133 set is that of the first code unit of the failing character.
1134
1135 Some errors are not detected until the whole pattern has been scanned;
1136 in these cases, the offset passed back is the length of the pattern.
1137 Note that the offset is in code units, not characters, even in a UTF
1138 mode. It may sometimes point into the middle of a UTF-8 or UTF-16 char‐
1139 acter.
1140
1141 This code fragment shows a typical straightforward call to pcre2_com‐
1142 pile():
1143
1144 pcre2_code *re;
1145 PCRE2_SIZE erroffset;
1146 int errorcode;
1147 re = pcre2_compile(
1148 "^A.*Z", /* the pattern */
1149 PCRE2_ZERO_TERMINATED, /* the pattern is zero-terminated */
1150 0, /* default options */
1151 &errorcode, /* for error code */
1152 &erroffset, /* for error offset */
1153 NULL); /* no compile context */
1154
1155 The following names for option bits are defined in the pcre2.h header
1156 file:
1157
1158 PCRE2_ANCHORED
1159
1160 If this bit is set, the pattern is forced to be "anchored", that is, it
1161 is constrained to match only at the first matching point in the string
1162 that is being searched (the "subject string"). This effect can also be
1163 achieved by appropriate constructs in the pattern itself, which is the
1164 only way to do it in Perl.
1165
1166 PCRE2_ALLOW_EMPTY_CLASS
1167
1168 By default, for compatibility with Perl, a closing square bracket that
1169 immediately follows an opening one is treated as a data character for
1170 the class. When PCRE2_ALLOW_EMPTY_CLASS is set, it terminates the
1171 class, which therefore contains no characters and so can never match.
1172
1173 PCRE2_ALT_BSUX
1174
1175 This option request alternative handling of three escape sequences,
1176 which makes PCRE2's behaviour more like ECMAscript (aka JavaScript).
1177 When it is set:
1178
1179 (1) \U matches an upper case "U" character; by default \U causes a com‐
1180 pile time error (Perl uses \U to upper case subsequent characters).
1181
1182 (2) \u matches a lower case "u" character unless it is followed by four
1183 hexadecimal digits, in which case the hexadecimal number defines the
1184 code point to match. By default, \u causes a compile time error (Perl
1185 uses it to upper case the following character).
1186
1187 (3) \x matches a lower case "x" character unless it is followed by two
1188 hexadecimal digits, in which case the hexadecimal number defines the
1189 code point to match. By default, as in Perl, a hexadecimal number is
1190 always expected after \x, but it may have zero, one, or two digits (so,
1191 for example, \xz matches a binary zero character followed by z).
1192
1193 PCRE2_ALT_CIRCUMFLEX
1194
1195 In multiline mode (when PCRE2_MULTILINE is set), the circumflex
1196 metacharacter matches at the start of the subject (unless PCRE2_NOTBOL
1197 is set), and also after any internal newline. However, it does not
1198 match after a newline at the end of the subject, for compatibility with
1199 Perl. If you want a multiline circumflex also to match after a termi‐
1200 nating newline, you must set PCRE2_ALT_CIRCUMFLEX.
1201
1202 PCRE2_ALT_VERBNAMES
1203
1204 By default, for compatibility with Perl, the name in any verb sequence
1205 such as (*MARK:NAME) is any sequence of characters that does not
1206 include a closing parenthesis. The name is not processed in any way,
1207 and it is not possible to include a closing parenthesis in the name.
1208 However, if the PCRE2_ALT_VERBNAMES option is set, normal backslash
1209 processing is applied to verb names and only an unescaped closing
1210 parenthesis terminates the name. A closing parenthesis can be included
1211 in a name either as \) or between \Q and \E. If the PCRE2_EXTENDED or
1212 PCRE2_EXTENDED_MORE option is set with PCRE2_ALT_VERBNAMES, unescaped
1213 whitespace in verb names is skipped and #-comments are recognized,
1214 exactly as in the rest of the pattern.
1215
1216 PCRE2_AUTO_CALLOUT
1217
1218 If this bit is set, pcre2_compile() automatically inserts callout
1219 items, all with number 255, before each pattern item, except immedi‐
1220 ately before or after an explicit callout in the pattern. For discus‐
1221 sion of the callout facility, see the pcre2callout documentation.
1222
1223 PCRE2_CASELESS
1224
1225 If this bit is set, letters in the pattern match both upper and lower
1226 case letters in the subject. It is equivalent to Perl's /i option, and
1227 it can be changed within a pattern by a (?i) option setting. If
1228 PCRE2_UTF is set, Unicode properties are used for all characters with
1229 more than one other case, and for all characters whose code points are
1230 greater than U+007F. For lower valued characters with only one other
1231 case, a lookup table is used for speed. When PCRE2_UTF is not set, a
1232 lookup table is used for all code points less than 256, and higher code
1233 points (available only in 16-bit or 32-bit mode) are treated as not
1234 having another case.
1235
1236 PCRE2_DOLLAR_ENDONLY
1237
1238 If this bit is set, a dollar metacharacter in the pattern matches only
1239 at the end of the subject string. Without this option, a dollar also
1240 matches immediately before a newline at the end of the string (but not
1241 before any other newlines). The PCRE2_DOLLAR_ENDONLY option is ignored
1242 if PCRE2_MULTILINE is set. There is no equivalent to this option in
1243 Perl, and no way to set it within a pattern.
1244
1245 PCRE2_DOTALL
1246
1247 If this bit is set, a dot metacharacter in the pattern matches any
1248 character, including one that indicates a newline. However, it only
1249 ever matches one character, even if newlines are coded as CRLF. Without
1250 this option, a dot does not match when the current position in the sub‐
1251 ject is at a newline. This option is equivalent to Perl's /s option,
1252 and it can be changed within a pattern by a (?s) option setting. A neg‐
1253 ative class such as [^a] always matches newline characters, and the \N
1254 escape sequence always matches a non-newline character, independent of
1255 the setting of PCRE2_DOTALL.
1256
1257 PCRE2_DUPNAMES
1258
1259 If this bit is set, names used to identify capturing subpatterns need
1260 not be unique. This can be helpful for certain types of pattern when it
1261 is known that only one instance of the named subpattern can ever be
1262 matched. There are more details of named subpatterns below; see also
1263 the pcre2pattern documentation.
1264
1265 PCRE2_ENDANCHORED
1266
1267 If this bit is set, the end of any pattern match must be right at the
1268 end of the string being searched (the "subject string"). If the pattern
1269 match succeeds by reaching (*ACCEPT), but does not reach the end of the
1270 subject, the match fails at the current starting point. For unanchored
1271 patterns, a new match is then tried at the next starting point. How‐
1272 ever, if the match succeeds by reaching the end of the pattern, but not
1273 the end of the subject, backtracking occurs and an alternative match
1274 may be found. Consider these two patterns:
1275
1276 .(*ACCEPT)|..
1277 .|..
1278
1279 If matched against "abc" with PCRE2_ENDANCHORED set, the first matches
1280 "c" whereas the second matches "bc". The effect of PCRE2_ENDANCHORED
1281 can also be achieved by appropriate constructs in the pattern itself,
1282 which is the only way to do it in Perl.
1283
1284 For DFA matching with pcre2_dfa_match(), PCRE2_ENDANCHORED applies only
1285 to the first (that is, the longest) matched string. Other parallel
1286 matches, which are necessarily substrings of the first one, must obvi‐
1287 ously end before the end of the subject.
1288
1289 PCRE2_EXTENDED
1290
1291 If this bit is set, most white space characters in the pattern are
1292 totally ignored except when escaped or inside a character class. How‐
1293 ever, white space is not allowed within sequences such as (?> that
1294 introduce various parenthesized subpatterns, nor within numerical quan‐
1295 tifiers such as {1,3}. Ignorable white space is permitted between an
1296 item and a following quantifier and between a quantifier and a follow‐
1297 ing + that indicates possessiveness. PCRE2_EXTENDED is equivalent to
1298 Perl's /x option, and it can be changed within a pattern by a (?x)
1299 option setting.
1300
1301 When PCRE2 is compiled without Unicode support, PCRE2_EXTENDED recog‐
1302 nizes as white space only those characters with code points less than
1303 256 that are flagged as white space in its low-character table. The ta‐
1304 ble is normally created by pcre2_maketables(), which uses the isspace()
1305 function to identify space characters. In most ASCII environments, the
1306 relevant characters are those with code points 0x0009 (tab), 0x000A
1307 (linefeed), 0x000B (vertical tab), 0x000C (formfeed), 0x000D (carriage
1308 return), and 0x0020 (space).
1309
1310 When PCRE2 is compiled with Unicode support, in addition to these char‐
1311 acters, five more Unicode "Pattern White Space" characters are recog‐
1312 nized by PCRE2_EXTENDED. These are U+0085 (next line), U+200E (left-to-
1313 right mark), U+200F (right-to-left mark), U+2028 (line separator), and
1314 U+2029 (paragraph separator). This set of characters is the same as
1315 recognized by Perl's /x option. Note that the horizontal and vertical
1316 space characters that are matched by the \h and \v escapes in patterns
1317 are a much bigger set.
1318
1319 As well as ignoring most white space, PCRE2_EXTENDED also causes char‐
1320 acters between an unescaped # outside a character class and the next
1321 newline, inclusive, to be ignored, which makes it possible to include
1322 comments inside complicated patterns. Note that the end of this type of
1323 comment is a literal newline sequence in the pattern; escape sequences
1324 that happen to represent a newline do not count.
1325
1326 Which characters are interpreted as newlines can be specified by a set‐
1327 ting in the compile context that is passed to pcre2_compile() or by a
1328 special sequence at the start of the pattern, as described in the sec‐
1329 tion entitled "Newline conventions" in the pcre2pattern documentation.
1330 A default is defined when PCRE2 is built.
1331
1332 PCRE2_EXTENDED_MORE
1333
1334 This option has the effect of PCRE2_EXTENDED, but, in addition,
1335 unescaped space and horizontal tab characters are ignored inside a
1336 character class. Note: only these two characters are ignored, not the
1337 full set of pattern white space characters that are ignored outside a
1338 character class. PCRE2_EXTENDED_MORE is equivalent to Perl's /xx
1339 option, and it can be changed within a pattern by a (?xx) option set‐
1340 ting.
1341
1342 PCRE2_FIRSTLINE
1343
1344 If this option is set, the start of an unanchored pattern match must be
1345 before or at the first newline in the subject string following the
1346 start of matching, though the matched text may continue over the new‐
1347 line. If startoffset is non-zero, the limiting newline is not necessar‐
1348 ily the first newline in the subject. For example, if the subject
1349 string is "abc\nxyz" (where \n represents a single-character newline) a
1350 pattern match for "yz" succeeds with PCRE2_FIRSTLINE if startoffset is
1351 greater than 3. See also PCRE2_USE_OFFSET_LIMIT, which provides a more
1352 general limiting facility. If PCRE2_FIRSTLINE is set with an offset
1353 limit, a match must occur in the first line and also within the offset
1354 limit. In other words, whichever limit comes first is used.
1355
1356 PCRE2_LITERAL
1357
1358 If this option is set, all meta-characters in the pattern are disabled,
1359 and it is treated as a literal string. Matching literal strings with a
1360 regular expression engine is not the most efficient way of doing it. If
1361 you are doing a lot of literal matching and are worried about effi‐
1362 ciency, you should consider using other approaches. The only other main
1363 options that are allowed with PCRE2_LITERAL are: PCRE2_ANCHORED,
1364 PCRE2_ENDANCHORED, PCRE2_AUTO_CALLOUT, PCRE2_CASELESS, PCRE2_FIRSTLINE,
1365 PCRE2_NO_START_OPTIMIZE, PCRE2_NO_UTF_CHECK, PCRE2_UTF, and
1366 PCRE2_USE_OFFSET_LIMIT. The extra options PCRE2_EXTRA_MATCH_LINE and
1367 PCRE2_EXTRA_MATCH_WORD are also supported. Any other options cause an
1368 error.
1369
1370 PCRE2_MATCH_UNSET_BACKREF
1371
1372 If this option is set, a backreference to an unset subpattern group
1373 matches an empty string (by default this causes the current matching
1374 alternative to fail). A pattern such as (\1)(a) succeeds when this
1375 option is set (assuming it can find an "a" in the subject), whereas it
1376 fails by default, for Perl compatibility. Setting this option makes
1377 PCRE2 behave more like ECMAscript (aka JavaScript).
1378
1379 PCRE2_MULTILINE
1380
1381 By default, for the purposes of matching "start of line" and "end of
1382 line", PCRE2 treats the subject string as consisting of a single line
1383 of characters, even if it actually contains newlines. The "start of
1384 line" metacharacter (^) matches only at the start of the string, and
1385 the "end of line" metacharacter ($) matches only at the end of the
1386 string, or before a terminating newline (except when PCRE2_DOL‐
1387 LAR_ENDONLY is set). Note, however, that unless PCRE2_DOTALL is set,
1388 the "any character" metacharacter (.) does not match at a newline. This
1389 behaviour (for ^, $, and dot) is the same as Perl.
1390
1391 When PCRE2_MULTILINE it is set, the "start of line" and "end of line"
1392 constructs match immediately following or immediately before internal
1393 newlines in the subject string, respectively, as well as at the very
1394 start and end. This is equivalent to Perl's /m option, and it can be
1395 changed within a pattern by a (?m) option setting. Note that the "start
1396 of line" metacharacter does not match after a newline at the end of the
1397 subject, for compatibility with Perl. However, you can change this by
1398 setting the PCRE2_ALT_CIRCUMFLEX option. If there are no newlines in a
1399 subject string, or no occurrences of ^ or $ in a pattern, setting
1400 PCRE2_MULTILINE has no effect.
1401
1402 PCRE2_NEVER_BACKSLASH_C
1403
1404 This option locks out the use of \C in the pattern that is being com‐
1405 piled. This escape can cause unpredictable behaviour in UTF-8 or
1406 UTF-16 modes, because it may leave the current matching point in the
1407 middle of a multi-code-unit character. This option may be useful in
1408 applications that process patterns from external sources. Note that
1409 there is also a build-time option that permanently locks out the use of
1410 \C.
1411
1412 PCRE2_NEVER_UCP
1413
1414 This option locks out the use of Unicode properties for handling \B,
1415 \b, \D, \d, \S, \s, \W, \w, and some of the POSIX character classes, as
1416 described for the PCRE2_UCP option below. In particular, it prevents
1417 the creator of the pattern from enabling this facility by starting the
1418 pattern with (*UCP). This option may be useful in applications that
1419 process patterns from external sources. The option combination PCRE_UCP
1420 and PCRE_NEVER_UCP causes an error.
1421
1422 PCRE2_NEVER_UTF
1423
1424 This option locks out interpretation of the pattern as UTF-8, UTF-16,
1425 or UTF-32, depending on which library is in use. In particular, it pre‐
1426 vents the creator of the pattern from switching to UTF interpretation
1427 by starting the pattern with (*UTF). This option may be useful in
1428 applications that process patterns from external sources. The combina‐
1429 tion of PCRE2_UTF and PCRE2_NEVER_UTF causes an error.
1430
1431 PCRE2_NO_AUTO_CAPTURE
1432
1433 If this option is set, it disables the use of numbered capturing paren‐
1434 theses in the pattern. Any opening parenthesis that is not followed by
1435 ? behaves as if it were followed by ?: but named parentheses can still
1436 be used for capturing (and they acquire numbers in the usual way). This
1437 is the same as Perl's /n option. Note that, when this option is set,
1438 references to capturing groups (backreferences or recursion/subroutine
1439 calls) may only refer to named groups, though the reference can be by
1440 name or by number.
1441
1442 PCRE2_NO_AUTO_POSSESS
1443
1444 If this option is set, it disables "auto-possessification", which is an
1445 optimization that, for example, turns a+b into a++b in order to avoid
1446 backtracks into a+ that can never be successful. However, if callouts
1447 are in use, auto-possessification means that some callouts are never
1448 taken. You can set this option if you want the matching functions to do
1449 a full unoptimized search and run all the callouts, but it is mainly
1450 provided for testing purposes.
1451
1452 PCRE2_NO_DOTSTAR_ANCHOR
1453
1454 If this option is set, it disables an optimization that is applied when
1455 .* is the first significant item in a top-level branch of a pattern,
1456 and all the other branches also start with .* or with \A or \G or ^.
1457 The optimization is automatically disabled for .* if it is inside an
1458 atomic group or a capturing group that is the subject of a backrefer‐
1459 ence, or if the pattern contains (*PRUNE) or (*SKIP). When the opti‐
1460 mization is not disabled, such a pattern is automatically anchored if
1461 PCRE2_DOTALL is set for all the .* items and PCRE2_MULTILINE is not set
1462 for any ^ items. Otherwise, the fact that any match must start either
1463 at the start of the subject or following a newline is remembered. Like
1464 other optimizations, this can cause callouts to be skipped.
1465
1466 PCRE2_NO_START_OPTIMIZE
1467
1468 This is an option whose main effect is at matching time. It does not
1469 change what pcre2_compile() generates, but it does affect the output of
1470 the JIT compiler.
1471
1472 There are a number of optimizations that may occur at the start of a
1473 match, in order to speed up the process. For example, if it is known
1474 that an unanchored match must start with a specific code unit value,
1475 the matching code searches the subject for that value, and fails imme‐
1476 diately if it cannot find it, without actually running the main match‐
1477 ing function. This means that a special item such as (*COMMIT) at the
1478 start of a pattern is not considered until after a suitable starting
1479 point for the match has been found. Also, when callouts or (*MARK)
1480 items are in use, these "start-up" optimizations can cause them to be
1481 skipped if the pattern is never actually used. The start-up optimiza‐
1482 tions are in effect a pre-scan of the subject that takes place before
1483 the pattern is run.
1484
1485 The PCRE2_NO_START_OPTIMIZE option disables the start-up optimizations,
1486 possibly causing performance to suffer, but ensuring that in cases
1487 where the result is "no match", the callouts do occur, and that items
1488 such as (*COMMIT) and (*MARK) are considered at every possible starting
1489 position in the subject string.
1490
1491 Setting PCRE2_NO_START_OPTIMIZE may change the outcome of a matching
1492 operation. Consider the pattern
1493
1494 (*COMMIT)ABC
1495
1496 When this is compiled, PCRE2 records the fact that a match must start
1497 with the character "A". Suppose the subject string is "DEFABC". The
1498 start-up optimization scans along the subject, finds "A" and runs the
1499 first match attempt from there. The (*COMMIT) item means that the pat‐
1500 tern must match the current starting position, which in this case, it
1501 does. However, if the same match is run with PCRE2_NO_START_OPTIMIZE
1502 set, the initial scan along the subject string does not happen. The
1503 first match attempt is run starting from "D" and when this fails,
1504 (*COMMIT) prevents any further matches being tried, so the overall
1505 result is "no match".
1506
1507 There are also other start-up optimizations. For example, a minimum
1508 length for the subject may be recorded. Consider the pattern
1509
1510 (*MARK:A)(X|Y)
1511
1512 The minimum length for a match is one character. If the subject is
1513 "ABC", there will be attempts to match "ABC", "BC", and "C". An attempt
1514 to match an empty string at the end of the subject does not take place,
1515 because PCRE2 knows that the subject is now too short, and so the
1516 (*MARK) is never encountered. In this case, the optimization does not
1517 affect the overall match result, which is still "no match", but it does
1518 affect the auxiliary information that is returned.
1519
1520 PCRE2_NO_UTF_CHECK
1521
1522 When PCRE2_UTF is set, the validity of the pattern as a UTF string is
1523 automatically checked. There are discussions about the validity of
1524 UTF-8 strings, UTF-16 strings, and UTF-32 strings in the pcre2unicode
1525 document. If an invalid UTF sequence is found, pcre2_compile() returns
1526 a negative error code.
1527
1528 If you know that your pattern is a valid UTF string, and you want to
1529 skip this check for performance reasons, you can set the
1530 PCRE2_NO_UTF_CHECK option. When it is set, the effect of passing an
1531 invalid UTF string as a pattern is undefined. It may cause your program
1532 to crash or loop.
1533
1534 Note that this option can also be passed to pcre2_match() and
1535 pcre_dfa_match(), to suppress UTF validity checking of the subject
1536 string.
1537
1538 Note also that setting PCRE2_NO_UTF_CHECK at compile time does not dis‐
1539 able the error that is given if an escape sequence for an invalid Uni‐
1540 code code point is encountered in the pattern. In particular, the so-
1541 called "surrogate" code points (0xd800 to 0xdfff) are invalid. If you
1542 want to allow escape sequences such as \x{d800} you can set the
1543 PCRE2_EXTRA_ALLOW_SURROGATE_ESCAPES extra option, as described in the
1544 section entitled "Extra compile options" below. However, this is pos‐
1545 sible only in UTF-8 and UTF-32 modes, because these values are not rep‐
1546 resentable in UTF-16.
1547
1548 PCRE2_UCP
1549
1550 This option changes the way PCRE2 processes \B, \b, \D, \d, \S, \s, \W,
1551 \w, and some of the POSIX character classes. By default, only ASCII
1552 characters are recognized, but if PCRE2_UCP is set, Unicode properties
1553 are used instead to classify characters. More details are given in the
1554 section on generic character types in the pcre2pattern page. If you set
1555 PCRE2_UCP, matching one of the items it affects takes much longer. The
1556 option is available only if PCRE2 has been compiled with Unicode sup‐
1557 port (which is the default).
1558
1559 PCRE2_UNGREEDY
1560
1561 This option inverts the "greediness" of the quantifiers so that they
1562 are not greedy by default, but become greedy if followed by "?". It is
1563 not compatible with Perl. It can also be set by a (?U) option setting
1564 within the pattern.
1565
1566 PCRE2_USE_OFFSET_LIMIT
1567
1568 This option must be set for pcre2_compile() if pcre2_set_offset_limit()
1569 is going to be used to set a non-default offset limit in a match con‐
1570 text for matches that use this pattern. An error is generated if an
1571 offset limit is set without this option. For more details, see the
1572 description of pcre2_set_offset_limit() in the section that describes
1573 match contexts. See also the PCRE2_FIRSTLINE option above.
1574
1575 PCRE2_UTF
1576
1577 This option causes PCRE2 to regard both the pattern and the subject
1578 strings that are subsequently processed as strings of UTF characters
1579 instead of single-code-unit strings. It is available when PCRE2 is
1580 built to include Unicode support (which is the default). If Unicode
1581 support is not available, the use of this option provokes an error.
1582 Details of how PCRE2_UTF changes the behaviour of PCRE2 are given in
1583 the pcre2unicode page. In particular, note that it changes the way
1584 PCRE2_CASELESS handles characters with code points greater than 127.
1585
1586 Extra compile options
1587
1588 Unlike the main compile-time options, the extra options are not saved
1589 with the compiled pattern. The option bits that can be set in a compile
1590 context by calling the pcre2_set_compile_extra_options() function are
1591 as follows:
1592
1593 PCRE2_EXTRA_ALLOW_SURROGATE_ESCAPES
1594
1595 This option applies when compiling a pattern in UTF-8 or UTF-32 mode.
1596 It is forbidden in UTF-16 mode, and ignored in non-UTF modes. Unicode
1597 "surrogate" code points in the range 0xd800 to 0xdfff are used in pairs
1598 in UTF-16 to encode code points with values in the range 0x10000 to
1599 0x10ffff. The surrogates cannot therefore be represented in UTF-16.
1600 They can be represented in UTF-8 and UTF-32, but are defined as invalid
1601 code points, and cause errors if encountered in a UTF-8 or UTF-32
1602 string that is being checked for validity by PCRE2.
1603
1604 These values also cause errors if encountered in escape sequences such
1605 as \x{d912} within a pattern. However, it seems that some applications,
1606 when using PCRE2 to check for unwanted characters in UTF-8 strings,
1607 explicitly test for the surrogates using escape sequences. The
1608 PCRE2_NO_UTF_CHECK option does not disable the error that occurs,
1609 because it applies only to the testing of input strings for UTF valid‐
1610 ity.
1611
1612 If the extra option PCRE2_EXTRA_ALLOW_SURROGATE_ESCAPES is set, surro‐
1613 gate code point values in UTF-8 and UTF-32 patterns no longer provoke
1614 errors and are incorporated in the compiled pattern. However, they can
1615 only match subject characters if the matching function is called with
1616 PCRE2_NO_UTF_CHECK set.
1617
1618 PCRE2_EXTRA_BAD_ESCAPE_IS_LITERAL
1619
1620 This is a dangerous option. Use with care. By default, an unrecognized
1621 escape such as \j or a malformed one such as \x{2z} causes a compile-
1622 time error when detected by pcre2_compile(). Perl is somewhat inconsis‐
1623 tent in handling such items: for example, \j is treated as a literal
1624 "j", and non-hexadecimal digits in \x{} are just ignored, though warn‐
1625 ings are given in both cases if Perl's warning switch is enabled. How‐
1626 ever, a malformed octal number after \o{ always causes an error in
1627 Perl.
1628
1629 If the PCRE2_EXTRA_BAD_ESCAPE_IS_LITERAL extra option is passed to
1630 pcre2_compile(), all unrecognized or erroneous escape sequences are
1631 treated as single-character escapes. For example, \j is a literal "j"
1632 and \x{2z} is treated as the literal string "x{2z}". Setting this
1633 option means that typos in patterns may go undetected and have unex‐
1634 pected results. This is a dangerous option. Use with care.
1635
1636 PCRE2_EXTRA_MATCH_LINE
1637
1638 This option is provided for use by the -x option of pcre2grep. It
1639 causes the pattern only to match complete lines. This is achieved by
1640 automatically inserting the code for "^(?:" at the start of the com‐
1641 piled pattern and ")$" at the end. Thus, when PCRE2_MULTILINE is set,
1642 the matched line may be in the middle of the subject string. This
1643 option can be used with PCRE2_LITERAL.
1644
1645 PCRE2_EXTRA_MATCH_WORD
1646
1647 This option is provided for use by the -w option of pcre2grep. It
1648 causes the pattern only to match strings that have a word boundary at
1649 the start and the end. This is achieved by automatically inserting the
1650 code for "\b(?:" at the start of the compiled pattern and ")\b" at the
1651 end. The option may be used with PCRE2_LITERAL. However, it is ignored
1652 if PCRE2_EXTRA_MATCH_LINE is also set.
1653
1655
1656 int pcre2_jit_compile(pcre2_code *code, uint32_t options);
1657
1658 int pcre2_jit_match(const pcre2_code *code, PCRE2_SPTR subject,
1659 PCRE2_SIZE length, PCRE2_SIZE startoffset,
1660 uint32_t options, pcre2_match_data *match_data,
1661 pcre2_match_context *mcontext);
1662
1663 void pcre2_jit_free_unused_memory(pcre2_general_context *gcontext);
1664
1665 pcre2_jit_stack *pcre2_jit_stack_create(PCRE2_SIZE startsize,
1666 PCRE2_SIZE maxsize, pcre2_general_context *gcontext);
1667
1668 void pcre2_jit_stack_assign(pcre2_match_context *mcontext,
1669 pcre2_jit_callback callback_function, void *callback_data);
1670
1671 void pcre2_jit_stack_free(pcre2_jit_stack *jit_stack);
1672
1673 These functions provide support for JIT compilation, which, if the
1674 just-in-time compiler is available, further processes a compiled pat‐
1675 tern into machine code that executes much faster than the pcre2_match()
1676 interpretive matching function. Full details are given in the pcre2jit
1677 documentation.
1678
1679 JIT compilation is a heavyweight optimization. It can take some time
1680 for patterns to be analyzed, and for one-off matches and simple pat‐
1681 terns the benefit of faster execution might be offset by a much slower
1682 compilation time. Most (but not all) patterns can be optimized by the
1683 JIT compiler.
1684
1686
1687 PCRE2 handles caseless matching, and determines whether characters are
1688 letters, digits, or whatever, by reference to a set of tables, indexed
1689 by character code point. This applies only to characters whose code
1690 points are less than 256. By default, higher-valued code points never
1691 match escapes such as \w or \d. However, if PCRE2 is built with Uni‐
1692 code support, all characters can be tested with \p and \P, or, alterna‐
1693 tively, the PCRE2_UCP option can be set when a pattern is compiled;
1694 this causes \w and friends to use Unicode property support instead of
1695 the built-in tables.
1696
1697 The use of locales with Unicode is discouraged. If you are handling
1698 characters with code points greater than 128, you should either use
1699 Unicode support, or use locales, but not try to mix the two.
1700
1701 PCRE2 contains an internal set of character tables that are used by
1702 default. These are sufficient for many applications. Normally, the
1703 internal tables recognize only ASCII characters. However, when PCRE2 is
1704 built, it is possible to cause the internal tables to be rebuilt in the
1705 default "C" locale of the local system, which may cause them to be dif‐
1706 ferent.
1707
1708 The internal tables can be overridden by tables supplied by the appli‐
1709 cation that calls PCRE2. These may be created in a different locale
1710 from the default. As more and more applications change to using Uni‐
1711 code, the need for this locale support is expected to die away.
1712
1713 External tables are built by calling the pcre2_maketables() function,
1714 in the relevant locale. The result can be passed to pcre2_compile() as
1715 often as necessary, by creating a compile context and calling
1716 pcre2_set_character_tables() to set the tables pointer therein. For
1717 example, to build and use tables that are appropriate for the French
1718 locale (where accented characters with values greater than 128 are
1719 treated as letters), the following code could be used:
1720
1721 setlocale(LC_CTYPE, "fr_FR");
1722 tables = pcre2_maketables(NULL);
1723 ccontext = pcre2_compile_context_create(NULL);
1724 pcre2_set_character_tables(ccontext, tables);
1725 re = pcre2_compile(..., ccontext);
1726
1727 The locale name "fr_FR" is used on Linux and other Unix-like systems;
1728 if you are using Windows, the name for the French locale is "french".
1729 It is the caller's responsibility to ensure that the memory containing
1730 the tables remains available for as long as it is needed.
1731
1732 The pointer that is passed (via the compile context) to pcre2_compile()
1733 is saved with the compiled pattern, and the same tables are used by
1734 pcre2_match() and pcre_dfa_match(). Thus, for any single pattern, com‐
1735 pilation and matching both happen in the same locale, but different
1736 patterns can be processed in different locales.
1737
1739
1740 int pcre2_pattern_info(const pcre2 *code, uint32_t what, void *where);
1741
1742 The pcre2_pattern_info() function returns general information about a
1743 compiled pattern. For information about callouts, see the next section.
1744 The first argument for pcre2_pattern_info() is a pointer to the com‐
1745 piled pattern. The second argument specifies which piece of information
1746 is required, and the third argument is a pointer to a variable to
1747 receive the data. If the third argument is NULL, the first argument is
1748 ignored, and the function returns the size in bytes of the variable
1749 that is required for the information requested. Otherwise, the yield of
1750 the function is zero for success, or one of the following negative num‐
1751 bers:
1752
1753 PCRE2_ERROR_NULL the argument code was NULL
1754 PCRE2_ERROR_BADMAGIC the "magic number" was not found
1755 PCRE2_ERROR_BADOPTION the value of what was invalid
1756 PCRE2_ERROR_UNSET the requested field is not set
1757
1758 The "magic number" is placed at the start of each compiled pattern as
1759 an simple check against passing an arbitrary memory pointer. Here is a
1760 typical call of pcre2_pattern_info(), to obtain the length of the com‐
1761 piled pattern:
1762
1763 int rc;
1764 size_t length;
1765 rc = pcre2_pattern_info(
1766 re, /* result of pcre2_compile() */
1767 PCRE2_INFO_SIZE, /* what is required */
1768 &length); /* where to put the data */
1769
1770 The possible values for the second argument are defined in pcre2.h, and
1771 are as follows:
1772
1773 PCRE2_INFO_ALLOPTIONS
1774 PCRE2_INFO_ARGOPTIONS
1775 PCRE2_INFO_EXTRAOPTIONS
1776
1777 Return copies of the pattern's options. The third argument should point
1778 to a uint32_t variable. PCRE2_INFO_ARGOPTIONS returns exactly the
1779 options that were passed to pcre2_compile(), whereas PCRE2_INFO_ALLOP‐
1780 TIONS returns the compile options as modified by any top-level (*XXX)
1781 option settings such as (*UTF) at the start of the pattern itself.
1782 PCRE2_INFO_EXTRAOPTIONS returns the extra options that were set in the
1783 compile context by calling the pcre2_set_compile_extra_options() func‐
1784 tion.
1785
1786 For example, if the pattern /(*UTF)abc/ is compiled with the
1787 PCRE2_EXTENDED option, the result for PCRE2_INFO_ALLOPTIONS is
1788 PCRE2_EXTENDED and PCRE2_UTF. Option settings such as (?i) that can
1789 change within a pattern do not affect the result of PCRE2_INFO_ALLOP‐
1790 TIONS, even if they appear right at the start of the pattern. (This was
1791 different in some earlier releases.)
1792
1793 A pattern compiled without PCRE2_ANCHORED is automatically anchored by
1794 PCRE2 if the first significant item in every top-level branch is one of
1795 the following:
1796
1797 ^ unless PCRE2_MULTILINE is set
1798 \A always
1799 \G always
1800 .* sometimes - see below
1801
1802 When .* is the first significant item, anchoring is possible only when
1803 all the following are true:
1804
1805 .* is not in an atomic group
1806 .* is not in a capturing group that is the subject
1807 of a backreference
1808 PCRE2_DOTALL is in force for .*
1809 Neither (*PRUNE) nor (*SKIP) appears in the pattern
1810 PCRE2_NO_DOTSTAR_ANCHOR is not set
1811
1812 For patterns that are auto-anchored, the PCRE2_ANCHORED bit is set in
1813 the options returned for PCRE2_INFO_ALLOPTIONS.
1814
1815 PCRE2_INFO_BACKREFMAX
1816
1817 Return the number of the highest backreference in the pattern. The
1818 third argument should point to an uint32_t variable. Named subpatterns
1819 acquire numbers as well as names, and these count towards the highest
1820 backreference. Backreferences such as \4 or \g{12} match the captured
1821 characters of the given group, but in addition, the check that a cap‐
1822 turing group is set in a conditional subpattern such as (?(3)a|b) is
1823 also a backreference. Zero is returned if there are no backreferences.
1824
1825 PCRE2_INFO_BSR
1826
1827 The output is a uint32_t integer whose value indicates what character
1828 sequences the \R escape sequence matches. A value of PCRE2_BSR_UNICODE
1829 means that \R matches any Unicode line ending sequence; a value of
1830 PCRE2_BSR_ANYCRLF means that \R matches only CR, LF, or CRLF.
1831
1832 PCRE2_INFO_CAPTURECOUNT
1833
1834 Return the highest capturing subpattern number in the pattern. In pat‐
1835 terns where (?| is not used, this is also the total number of capturing
1836 subpatterns. The third argument should point to an uint32_t variable.
1837
1838 PCRE2_INFO_DEPTHLIMIT
1839
1840 If the pattern set a backtracking depth limit by including an item of
1841 the form (*LIMIT_DEPTH=nnnn) at the start, the value is returned. The
1842 third argument should point to a uint32_t integer. If no such value has
1843 been set, the call to pcre2_pattern_info() returns the error
1844 PCRE2_ERROR_UNSET. Note that this limit will only be used during match‐
1845 ing if it is less than the limit set or defaulted by the caller of the
1846 match function.
1847
1848 PCRE2_INFO_FIRSTBITMAP
1849
1850 In the absence of a single first code unit for a non-anchored pattern,
1851 pcre2_compile() may construct a 256-bit table that defines a fixed set
1852 of values for the first code unit in any match. For example, a pattern
1853 that starts with [abc] results in a table with three bits set. When
1854 code unit values greater than 255 are supported, the flag bit for 255
1855 means "any code unit of value 255 or above". If such a table was con‐
1856 structed, a pointer to it is returned. Otherwise NULL is returned. The
1857 third argument should point to a const uint8_t * variable.
1858
1859 PCRE2_INFO_FIRSTCODETYPE
1860
1861 Return information about the first code unit of any matched string, for
1862 a non-anchored pattern. The third argument should point to an uint32_t
1863 variable. If there is a fixed first value, for example, the letter "c"
1864 from a pattern such as (cat|cow|coyote), 1 is returned, and the value
1865 can be retrieved using PCRE2_INFO_FIRSTCODEUNIT. If there is no fixed
1866 first value, but it is known that a match can occur only at the start
1867 of the subject or following a newline in the subject, 2 is returned.
1868 Otherwise, and for anchored patterns, 0 is returned.
1869
1870 PCRE2_INFO_FIRSTCODEUNIT
1871
1872 Return the value of the first code unit of any matched string for a
1873 pattern where PCRE2_INFO_FIRSTCODETYPE returns 1; otherwise return 0.
1874 The third argument should point to an uint32_t variable. In the 8-bit
1875 library, the value is always less than 256. In the 16-bit library the
1876 value can be up to 0xffff. In the 32-bit library in UTF-32 mode the
1877 value can be up to 0x10ffff, and up to 0xffffffff when not using UTF-32
1878 mode.
1879
1880 PCRE2_INFO_FRAMESIZE
1881
1882 Return the size (in bytes) of the data frames that are used to remember
1883 backtracking positions when the pattern is processed by pcre2_match()
1884 without the use of JIT. The third argument should point to a size_t
1885 variable. The frame size depends on the number of capturing parentheses
1886 in the pattern. Each additional capturing group adds two PCRE2_SIZE
1887 variables.
1888
1889 PCRE2_INFO_HASBACKSLASHC
1890
1891 Return 1 if the pattern contains any instances of \C, otherwise 0. The
1892 third argument should point to an uint32_t variable.
1893
1894 PCRE2_INFO_HASCRORLF
1895
1896 Return 1 if the pattern contains any explicit matches for CR or LF
1897 characters, otherwise 0. The third argument should point to an uint32_t
1898 variable. An explicit match is either a literal CR or LF character, or
1899 \r or \n or one of the equivalent hexadecimal or octal escape
1900 sequences.
1901
1902 PCRE2_INFO_HEAPLIMIT
1903
1904 If the pattern set a heap memory limit by including an item of the form
1905 (*LIMIT_HEAP=nnnn) at the start, the value is returned. The third argu‐
1906 ment should point to a uint32_t integer. If no such value has been set,
1907 the call to pcre2_pattern_info() returns the error PCRE2_ERROR_UNSET.
1908 Note that this limit will only be used during matching if it is less
1909 than the limit set or defaulted by the caller of the match function.
1910
1911 PCRE2_INFO_JCHANGED
1912
1913 Return 1 if the (?J) or (?-J) option setting is used in the pattern,
1914 otherwise 0. The third argument should point to an uint32_t variable.
1915 (?J) and (?-J) set and unset the local PCRE2_DUPNAMES option, respec‐
1916 tively.
1917
1918 PCRE2_INFO_JITSIZE
1919
1920 If the compiled pattern was successfully processed by pcre2_jit_com‐
1921 pile(), return the size of the JIT compiled code, otherwise return
1922 zero. The third argument should point to a size_t variable.
1923
1924 PCRE2_INFO_LASTCODETYPE
1925
1926 Returns 1 if there is a rightmost literal code unit that must exist in
1927 any matched string, other than at its start. The third argument should
1928 point to an uint32_t variable. If there is no such value, 0 is
1929 returned. When 1 is returned, the code unit value itself can be
1930 retrieved using PCRE2_INFO_LASTCODEUNIT. For anchored patterns, a last
1931 literal value is recorded only if it follows something of variable
1932 length. For example, for the pattern /^a\d+z\d+/ the returned value is
1933 1 (with "z" returned from PCRE2_INFO_LASTCODEUNIT), but for /^a\dz\d/
1934 the returned value is 0.
1935
1936 PCRE2_INFO_LASTCODEUNIT
1937
1938 Return the value of the rightmost literal code unit that must exist in
1939 any matched string, other than at its start, for a pattern where
1940 PCRE2_INFO_LASTCODETYPE returns 1. Otherwise, return 0. The third argu‐
1941 ment should point to an uint32_t variable.
1942
1943 PCRE2_INFO_MATCHEMPTY
1944
1945 Return 1 if the pattern might match an empty string, otherwise 0. The
1946 third argument should point to an uint32_t variable. When a pattern
1947 contains recursive subroutine calls it is not always possible to deter‐
1948 mine whether or not it can match an empty string. PCRE2 takes a cau‐
1949 tious approach and returns 1 in such cases.
1950
1951 PCRE2_INFO_MATCHLIMIT
1952
1953 If the pattern set a match limit by including an item of the form
1954 (*LIMIT_MATCH=nnnn) at the start, the value is returned. The third
1955 argument should point to a uint32_t integer. If no such value has been
1956 set, the call to pcre2_pattern_info() returns the error
1957 PCRE2_ERROR_UNSET. Note that this limit will only be used during match‐
1958 ing if it is less than the limit set or defaulted by the caller of the
1959 match function.
1960
1961 PCRE2_INFO_MAXLOOKBEHIND
1962
1963 Return the number of characters (not code units) in the longest lookbe‐
1964 hind assertion in the pattern. The third argument should point to a
1965 uint32_t integer. This information is useful when doing multi-segment
1966 matching using the partial matching facilities. Note that the simple
1967 assertions \b and \B require a one-character lookbehind. \A also regis‐
1968 ters a one-character lookbehind, though it does not actually inspect
1969 the previous character. This is to ensure that at least one character
1970 from the old segment is retained when a new segment is processed. Oth‐
1971 erwise, if there are no lookbehinds in the pattern, \A might match
1972 incorrectly at the start of a second or subsequent segment.
1973
1974 PCRE2_INFO_MINLENGTH
1975
1976 If a minimum length for matching subject strings was computed, its
1977 value is returned. Otherwise the returned value is 0. The value is a
1978 number of characters, which in UTF mode may be different from the num‐
1979 ber of code units. The third argument should point to an uint32_t
1980 variable. The value is a lower bound to the length of any matching
1981 string. There may not be any strings of that length that do actually
1982 match, but every string that does match is at least that long.
1983
1984 PCRE2_INFO_NAMECOUNT
1985 PCRE2_INFO_NAMEENTRYSIZE
1986 PCRE2_INFO_NAMETABLE
1987
1988 PCRE2 supports the use of named as well as numbered capturing parenthe‐
1989 ses. The names are just an additional way of identifying the parenthe‐
1990 ses, which still acquire numbers. Several convenience functions such as
1991 pcre2_substring_get_byname() are provided for extracting captured sub‐
1992 strings by name. It is also possible to extract the data directly, by
1993 first converting the name to a number in order to access the correct
1994 pointers in the output vector (described with pcre2_match() below). To
1995 do the conversion, you need to use the name-to-number map, which is
1996 described by these three values.
1997
1998 The map consists of a number of fixed-size entries. PCRE2_INFO_NAME‐
1999 COUNT gives the number of entries, and PCRE2_INFO_NAMEENTRYSIZE gives
2000 the size of each entry in code units; both of these return a uint32_t
2001 value. The entry size depends on the length of the longest name.
2002
2003 PCRE2_INFO_NAMETABLE returns a pointer to the first entry of the table.
2004 This is a PCRE2_SPTR pointer to a block of code units. In the 8-bit
2005 library, the first two bytes of each entry are the number of the cap‐
2006 turing parenthesis, most significant byte first. In the 16-bit library,
2007 the pointer points to 16-bit code units, the first of which contains
2008 the parenthesis number. In the 32-bit library, the pointer points to
2009 32-bit code units, the first of which contains the parenthesis number.
2010 The rest of the entry is the corresponding name, zero terminated.
2011
2012 The names are in alphabetical order. If (?| is used to create multiple
2013 groups with the same number, as described in the section on duplicate
2014 subpattern numbers in the pcre2pattern page, the groups may be given
2015 the same name, but there is only one entry in the table. Different
2016 names for groups of the same number are not permitted.
2017
2018 Duplicate names for subpatterns with different numbers are permitted,
2019 but only if PCRE2_DUPNAMES is set. They appear in the table in the
2020 order in which they were found in the pattern. In the absence of (?|
2021 this is the order of increasing number; when (?| is used this is not
2022 necessarily the case because later subpatterns may have lower numbers.
2023
2024 As a simple example of the name/number table, consider the following
2025 pattern after compilation by the 8-bit library (assume PCRE2_EXTENDED
2026 is set, so white space - including newlines - is ignored):
2027
2028 (?<date> (?<year>(\d\d)?\d\d) -
2029 (?<month>\d\d) - (?<day>\d\d) )
2030
2031 There are four named subpatterns, so the table has four entries, and
2032 each entry in the table is eight bytes long. The table is as follows,
2033 with non-printing bytes shows in hexadecimal, and undefined bytes shown
2034 as ??:
2035
2036 00 01 d a t e 00 ??
2037 00 05 d a y 00 ?? ??
2038 00 04 m o n t h 00
2039 00 02 y e a r 00 ??
2040
2041 When writing code to extract data from named subpatterns using the
2042 name-to-number map, remember that the length of the entries is likely
2043 to be different for each compiled pattern.
2044
2045 PCRE2_INFO_NEWLINE
2046
2047 The output is one of the following uint32_t values:
2048
2049 PCRE2_NEWLINE_CR Carriage return (CR)
2050 PCRE2_NEWLINE_LF Linefeed (LF)
2051 PCRE2_NEWLINE_CRLF Carriage return, linefeed (CRLF)
2052 PCRE2_NEWLINE_ANY Any Unicode line ending
2053 PCRE2_NEWLINE_ANYCRLF Any of CR, LF, or CRLF
2054 PCRE2_NEWLINE_NUL The NUL character (binary zero)
2055
2056 This identifies the character sequence that will be recognized as mean‐
2057 ing "newline" while matching.
2058
2059 PCRE2_INFO_SIZE
2060
2061 Return the size of the compiled pattern in bytes (for all three
2062 libraries). The third argument should point to a size_t variable. This
2063 value includes the size of the general data block that precedes the
2064 code units of the compiled pattern itself. The value that is used when
2065 pcre2_compile() is getting memory in which to place the compiled pat‐
2066 tern may be slightly larger than the value returned by this option,
2067 because there are cases where the code that calculates the size has to
2068 over-estimate. Processing a pattern with the JIT compiler does not
2069 alter the value returned by this option.
2070
2072
2073 int pcre2_callout_enumerate(const pcre2_code *code,
2074 int (*callback)(pcre2_callout_enumerate_block *, void *),
2075 void *user_data);
2076
2077 A script language that supports the use of string arguments in callouts
2078 might like to scan all the callouts in a pattern before running the
2079 match. This can be done by calling pcre2_callout_enumerate(). The first
2080 argument is a pointer to a compiled pattern, the second points to a
2081 callback function, and the third is arbitrary user data. The callback
2082 function is called for every callout in the pattern in the order in
2083 which they appear. Its first argument is a pointer to a callout enumer‐
2084 ation block, and its second argument is the user_data value that was
2085 passed to pcre2_callout_enumerate(). The contents of the callout enu‐
2086 meration block are described in the pcre2callout documentation, which
2087 also gives further details about callouts.
2088
2090
2091 It is possible to save compiled patterns on disc or elsewhere, and
2092 reload them later, subject to a number of restrictions. The host on
2093 which the patterns are reloaded must be running the same version of
2094 PCRE2, with the same code unit width, and must also have the same endi‐
2095 anness, pointer width, and PCRE2_SIZE type. Before compiled patterns
2096 can be saved, they must be converted to a "serialized" form, which in
2097 the case of PCRE2 is really just a bytecode dump. The functions whose
2098 names begin with pcre2_serialize_ are used for converting to and from
2099 the serialized form. They are described in the pcre2serialize documen‐
2100 tation. Note that PCRE2 serialization does not convert compiled pat‐
2101 terns to an abstract format like Java or .NET serialization.
2102
2104
2105 pcre2_match_data *pcre2_match_data_create(uint32_t ovecsize,
2106 pcre2_general_context *gcontext);
2107
2108 pcre2_match_data *pcre2_match_data_create_from_pattern(
2109 const pcre2_code *code, pcre2_general_context *gcontext);
2110
2111 void pcre2_match_data_free(pcre2_match_data *match_data);
2112
2113 Information about a successful or unsuccessful match is placed in a
2114 match data block, which is an opaque structure that is accessed by
2115 function calls. In particular, the match data block contains a vector
2116 of offsets into the subject string that define the matched part of the
2117 subject and any substrings that were captured. This is known as the
2118 ovector.
2119
2120 Before calling pcre2_match(), pcre2_dfa_match(), or pcre2_jit_match()
2121 you must create a match data block by calling one of the creation func‐
2122 tions above. For pcre2_match_data_create(), the first argument is the
2123 number of pairs of offsets in the ovector. One pair of offsets is
2124 required to identify the string that matched the whole pattern, with an
2125 additional pair for each captured substring. For example, a value of 4
2126 creates enough space to record the matched portion of the subject plus
2127 three captured substrings. A minimum of at least 1 pair is imposed by
2128 pcre2_match_data_create(), so it is always possible to return the over‐
2129 all matched string.
2130
2131 The second argument of pcre2_match_data_create() is a pointer to a gen‐
2132 eral context, which can specify custom memory management for obtaining
2133 the memory for the match data block. If you are not using custom memory
2134 management, pass NULL, which causes malloc() to be used.
2135
2136 For pcre2_match_data_create_from_pattern(), the first argument is a
2137 pointer to a compiled pattern. The ovector is created to be exactly the
2138 right size to hold all the substrings a pattern might capture. The sec‐
2139 ond argument is again a pointer to a general context, but in this case
2140 if NULL is passed, the memory is obtained using the same allocator that
2141 was used for the compiled pattern (custom or default).
2142
2143 A match data block can be used many times, with the same or different
2144 compiled patterns. You can extract information from a match data block
2145 after a match operation has finished, using functions that are
2146 described in the sections on matched strings and other match data
2147 below.
2148
2149 When a call of pcre2_match() fails, valid data is available in the
2150 match block only when the error is PCRE2_ERROR_NOMATCH,
2151 PCRE2_ERROR_PARTIAL, or one of the error codes for an invalid UTF
2152 string. Exactly what is available depends on the error, and is detailed
2153 below.
2154
2155 When one of the matching functions is called, pointers to the compiled
2156 pattern and the subject string are set in the match data block so that
2157 they can be referenced by the extraction functions. After running a
2158 match, you must not free a compiled pattern or a subject string until
2159 after all operations on the match data block (for that match) have
2160 taken place.
2161
2162 When a match data block itself is no longer needed, it should be freed
2163 by calling pcre2_match_data_free(). If this function is called with a
2164 NULL argument, it returns immediately, without doing anything.
2165
2167
2168 int pcre2_match(const pcre2_code *code, PCRE2_SPTR subject,
2169 PCRE2_SIZE length, PCRE2_SIZE startoffset,
2170 uint32_t options, pcre2_match_data *match_data,
2171 pcre2_match_context *mcontext);
2172
2173 The function pcre2_match() is called to match a subject string against
2174 a compiled pattern, which is passed in the code argument. You can call
2175 pcre2_match() with the same code argument as many times as you like, in
2176 order to find multiple matches in the subject string or to match dif‐
2177 ferent subject strings with the same pattern.
2178
2179 This function is the main matching facility of the library, and it
2180 operates in a Perl-like manner. For specialist use there is also an
2181 alternative matching function, which is described below in the section
2182 about the pcre2_dfa_match() function.
2183
2184 Here is an example of a simple call to pcre2_match():
2185
2186 pcre2_match_data *md = pcre2_match_data_create(4, NULL);
2187 int rc = pcre2_match(
2188 re, /* result of pcre2_compile() */
2189 "some string", /* the subject string */
2190 11, /* the length of the subject string */
2191 0, /* start at offset 0 in the subject */
2192 0, /* default options */
2193 md, /* the match data block */
2194 NULL); /* a match context; NULL means use defaults */
2195
2196 If the subject string is zero-terminated, the length can be given as
2197 PCRE2_ZERO_TERMINATED. A match context must be provided if certain less
2198 common matching parameters are to be changed. For details, see the sec‐
2199 tion on the match context above.
2200
2201 The string to be matched by pcre2_match()
2202
2203 The subject string is passed to pcre2_match() as a pointer in subject,
2204 a length in length, and a starting offset in startoffset. The length
2205 and offset are in code units, not characters. That is, they are in
2206 bytes for the 8-bit library, 16-bit code units for the 16-bit library,
2207 and 32-bit code units for the 32-bit library, whether or not UTF pro‐
2208 cessing is enabled.
2209
2210 If startoffset is greater than the length of the subject, pcre2_match()
2211 returns PCRE2_ERROR_BADOFFSET. When the starting offset is zero, the
2212 search for a match starts at the beginning of the subject, and this is
2213 by far the most common case. In UTF-8 or UTF-16 mode, the starting off‐
2214 set must point to the start of a character, or to the end of the sub‐
2215 ject (in UTF-32 mode, one code unit equals one character, so all off‐
2216 sets are valid). Like the pattern string, the subject may contain
2217 binary zeros.
2218
2219 A non-zero starting offset is useful when searching for another match
2220 in the same subject by calling pcre2_match() again after a previous
2221 success. Setting startoffset differs from passing over a shortened
2222 string and setting PCRE2_NOTBOL in the case of a pattern that begins
2223 with any kind of lookbehind. For example, consider the pattern
2224
2225 \Biss\B
2226
2227 which finds occurrences of "iss" in the middle of words. (\B matches
2228 only if the current position in the subject is not a word boundary.)
2229 When applied to the string "Mississipi" the first call to pcre2_match()
2230 finds the first occurrence. If pcre2_match() is called again with just
2231 the remainder of the subject, namely "issipi", it does not match,
2232 because \B is always false at the start of the subject, which is deemed
2233 to be a word boundary. However, if pcre2_match() is passed the entire
2234 string again, but with startoffset set to 4, it finds the second occur‐
2235 rence of "iss" because it is able to look behind the starting point to
2236 discover that it is preceded by a letter.
2237
2238 Finding all the matches in a subject is tricky when the pattern can
2239 match an empty string. It is possible to emulate Perl's /g behaviour by
2240 first trying the match again at the same offset, with the
2241 PCRE2_NOTEMPTY_ATSTART and PCRE2_ANCHORED options, and then if that
2242 fails, advancing the starting offset and trying an ordinary match
2243 again. There is some code that demonstrates how to do this in the
2244 pcre2demo sample program. In the most general case, you have to check
2245 to see if the newline convention recognizes CRLF as a newline, and if
2246 so, and the current character is CR followed by LF, advance the start‐
2247 ing offset by two characters instead of one.
2248
2249 If a non-zero starting offset is passed when the pattern is anchored, a
2250 single attempt to match at the given offset is made. This can only suc‐
2251 ceed if the pattern does not require the match to be at the start of
2252 the subject. In other words, the anchoring must be the result of set‐
2253 ting the PCRE2_ANCHORED option or the use of .* with PCRE2_DOTALL, not
2254 by starting the pattern with ^ or \A.
2255
2256 Option bits for pcre2_match()
2257
2258 The unused bits of the options argument for pcre2_match() must be zero.
2259 The only bits that may be set are PCRE2_ANCHORED, PCRE2_ENDANCHORED,
2260 PCRE2_NOTBOL, PCRE2_NOTEOL, PCRE2_NOTEMPTY, PCRE2_NOTEMPTY_ATSTART,
2261 PCRE2_NO_JIT, PCRE2_NO_UTF_CHECK, PCRE2_PARTIAL_HARD, and PCRE2_PAR‐
2262 TIAL_SOFT. Their action is described below.
2263
2264 Setting PCRE2_ANCHORED or PCRE2_ENDANCHORED at match time is not sup‐
2265 ported by the just-in-time (JIT) compiler. If it is set, JIT matching
2266 is disabled and the interpretive code in pcre2_match() is run. Apart
2267 from PCRE2_NO_JIT (obviously), the remaining options are supported for
2268 JIT matching.
2269
2270 PCRE2_ANCHORED
2271
2272 The PCRE2_ANCHORED option limits pcre2_match() to matching at the first
2273 matching position. If a pattern was compiled with PCRE2_ANCHORED, or
2274 turned out to be anchored by virtue of its contents, it cannot be made
2275 unachored at matching time. Note that setting the option at match time
2276 disables JIT matching.
2277
2278 PCRE2_ENDANCHORED
2279
2280 If the PCRE2_ENDANCHORED option is set, any string that pcre2_match()
2281 matches must be right at the end of the subject string. Note that set‐
2282 ting the option at match time disables JIT matching.
2283
2284 PCRE2_NOTBOL
2285
2286 This option specifies that first character of the subject string is not
2287 the beginning of a line, so the circumflex metacharacter should not
2288 match before it. Setting this without having set PCRE2_MULTILINE at
2289 compile time causes circumflex never to match. This option affects only
2290 the behaviour of the circumflex metacharacter. It does not affect \A.
2291
2292 PCRE2_NOTEOL
2293
2294 This option specifies that the end of the subject string is not the end
2295 of a line, so the dollar metacharacter should not match it nor (except
2296 in multiline mode) a newline immediately before it. Setting this with‐
2297 out having set PCRE2_MULTILINE at compile time causes dollar never to
2298 match. This option affects only the behaviour of the dollar metacharac‐
2299 ter. It does not affect \Z or \z.
2300
2301 PCRE2_NOTEMPTY
2302
2303 An empty string is not considered to be a valid match if this option is
2304 set. If there are alternatives in the pattern, they are tried. If all
2305 the alternatives match the empty string, the entire match fails. For
2306 example, if the pattern
2307
2308 a?b?
2309
2310 is applied to a string not beginning with "a" or "b", it matches an
2311 empty string at the start of the subject. With PCRE2_NOTEMPTY set, this
2312 match is not valid, so pcre2_match() searches further into the string
2313 for occurrences of "a" or "b".
2314
2315 PCRE2_NOTEMPTY_ATSTART
2316
2317 This is like PCRE2_NOTEMPTY, except that it locks out an empty string
2318 match only at the first matching position, that is, at the start of the
2319 subject plus the starting offset. An empty string match later in the
2320 subject is permitted. If the pattern is anchored, such a match can
2321 occur only if the pattern contains \K.
2322
2323 PCRE2_NO_JIT
2324
2325 By default, if a pattern has been successfully processed by
2326 pcre2_jit_compile(), JIT is automatically used when pcre2_match() is
2327 called with options that JIT supports. Setting PCRE2_NO_JIT disables
2328 the use of JIT; it forces matching to be done by the interpreter.
2329
2330 PCRE2_NO_UTF_CHECK
2331
2332 When PCRE2_UTF is set at compile time, the validity of the subject as a
2333 UTF string is checked by default when pcre2_match() is subsequently
2334 called. If a non-zero starting offset is given, the check is applied
2335 only to that part of the subject that could be inspected during match‐
2336 ing, and there is a check that the starting offset points to the first
2337 code unit of a character or to the end of the subject. If there are no
2338 lookbehind assertions in the pattern, the check starts at the starting
2339 offset. Otherwise, it starts at the length of the longest lookbehind
2340 before the starting offset, or at the start of the subject if there are
2341 not that many characters before the starting offset. Note that the
2342 sequences \b and \B are one-character lookbehinds.
2343
2344 The check is carried out before any other processing takes place, and a
2345 negative error code is returned if the check fails. There are several
2346 UTF error codes for each code unit width, corresponding to different
2347 problems with the code unit sequence. There are discussions about the
2348 validity of UTF-8 strings, UTF-16 strings, and UTF-32 strings in the
2349 pcre2unicode page.
2350
2351 If you know that your subject is valid, and you want to skip these
2352 checks for performance reasons, you can set the PCRE2_NO_UTF_CHECK
2353 option when calling pcre2_match(). You might want to do this for the
2354 second and subsequent calls to pcre2_match() if you are making repeated
2355 calls to find other matches in the same subject string.
2356
2357 Warning: When PCRE2_NO_UTF_CHECK is set, the effect of passing an
2358 invalid string as a subject, or an invalid value of startoffset, is
2359 undefined. Your program may crash or loop indefinitely.
2360
2361 PCRE2_PARTIAL_HARD
2362 PCRE2_PARTIAL_SOFT
2363
2364 These options turn on the partial matching feature. A partial match
2365 occurs if the end of the subject string is reached successfully, but
2366 there are not enough subject characters to complete the match. If this
2367 happens when PCRE2_PARTIAL_SOFT (but not PCRE2_PARTIAL_HARD) is set,
2368 matching continues by testing any remaining alternatives. Only if no
2369 complete match can be found is PCRE2_ERROR_PARTIAL returned instead of
2370 PCRE2_ERROR_NOMATCH. In other words, PCRE2_PARTIAL_SOFT specifies that
2371 the caller is prepared to handle a partial match, but only if no com‐
2372 plete match can be found.
2373
2374 If PCRE2_PARTIAL_HARD is set, it overrides PCRE2_PARTIAL_SOFT. In this
2375 case, if a partial match is found, pcre2_match() immediately returns
2376 PCRE2_ERROR_PARTIAL, without considering any other alternatives. In
2377 other words, when PCRE2_PARTIAL_HARD is set, a partial match is consid‐
2378 ered to be more important that an alternative complete match.
2379
2380 There is a more detailed discussion of partial and multi-segment match‐
2381 ing, with examples, in the pcre2partial documentation.
2382
2384
2385 When PCRE2 is built, a default newline convention is set; this is usu‐
2386 ally the standard convention for the operating system. The default can
2387 be overridden in a compile context by calling pcre2_set_newline(). It
2388 can also be overridden by starting a pattern string with, for example,
2389 (*CRLF), as described in the section on newline conventions in the
2390 pcre2pattern page. During matching, the newline choice affects the be‐
2391 haviour of the dot, circumflex, and dollar metacharacters. It may also
2392 alter the way the match starting position is advanced after a match
2393 failure for an unanchored pattern.
2394
2395 When PCRE2_NEWLINE_CRLF, PCRE2_NEWLINE_ANYCRLF, or PCRE2_NEWLINE_ANY is
2396 set as the newline convention, and a match attempt for an unanchored
2397 pattern fails when the current starting position is at a CRLF sequence,
2398 and the pattern contains no explicit matches for CR or LF characters,
2399 the match position is advanced by two characters instead of one, in
2400 other words, to after the CRLF.
2401
2402 The above rule is a compromise that makes the most common cases work as
2403 expected. For example, if the pattern is .+A (and the PCRE2_DOTALL
2404 option is not set), it does not match the string "\r\nA" because, after
2405 failing at the start, it skips both the CR and the LF before retrying.
2406 However, the pattern [\r\n]A does match that string, because it con‐
2407 tains an explicit CR or LF reference, and so advances only by one char‐
2408 acter after the first failure.
2409
2410 An explicit match for CR of LF is either a literal appearance of one of
2411 those characters in the pattern, or one of the \r or \n or equivalent
2412 octal or hexadecimal escape sequences. Implicit matches such as [^X] do
2413 not count, nor does \s, even though it includes CR and LF in the char‐
2414 acters that it matches.
2415
2416 Notwithstanding the above, anomalous effects may still occur when CRLF
2417 is a valid newline sequence and explicit \r or \n escapes appear in the
2418 pattern.
2419
2421
2422 uint32_t pcre2_get_ovector_count(pcre2_match_data *match_data);
2423
2424 PCRE2_SIZE *pcre2_get_ovector_pointer(pcre2_match_data *match_data);
2425
2426 In general, a pattern matches a certain portion of the subject, and in
2427 addition, further substrings from the subject may be picked out by
2428 parenthesized parts of the pattern. Following the usage in Jeffrey
2429 Friedl's book, this is called "capturing" in what follows, and the
2430 phrase "capturing subpattern" or "capturing group" is used for a frag‐
2431 ment of a pattern that picks out a substring. PCRE2 supports several
2432 other kinds of parenthesized subpattern that do not cause substrings to
2433 be captured. The pcre2_pattern_info() function can be used to find out
2434 how many capturing subpatterns there are in a compiled pattern.
2435
2436 You can use auxiliary functions for accessing captured substrings by
2437 number or by name, as described in sections below.
2438
2439 Alternatively, you can make direct use of the vector of PCRE2_SIZE val‐
2440 ues, called the ovector, which contains the offsets of captured
2441 strings. It is part of the match data block. The function
2442 pcre2_get_ovector_pointer() returns the address of the ovector, and
2443 pcre2_get_ovector_count() returns the number of pairs of values it con‐
2444 tains.
2445
2446 Within the ovector, the first in each pair of values is set to the off‐
2447 set of the first code unit of a substring, and the second is set to the
2448 offset of the first code unit after the end of a substring. These val‐
2449 ues are always code unit offsets, not character offsets. That is, they
2450 are byte offsets in the 8-bit library, 16-bit offsets in the 16-bit
2451 library, and 32-bit offsets in the 32-bit library.
2452
2453 After a partial match (error return PCRE2_ERROR_PARTIAL), only the
2454 first pair of offsets (that is, ovector[0] and ovector[1]) are set.
2455 They identify the part of the subject that was partially matched. See
2456 the pcre2partial documentation for details of partial matching.
2457
2458 After a fully successful match, the first pair of offsets identifies
2459 the portion of the subject string that was matched by the entire pat‐
2460 tern. The next pair is used for the first captured substring, and so
2461 on. The value returned by pcre2_match() is one more than the highest
2462 numbered pair that has been set. For example, if two substrings have
2463 been captured, the returned value is 3. If there are no captured sub‐
2464 strings, the return value from a successful match is 1, indicating that
2465 just the first pair of offsets has been set.
2466
2467 If a pattern uses the \K escape sequence within a positive assertion,
2468 the reported start of a successful match can be greater than the end of
2469 the match. For example, if the pattern (?=ab\K) is matched against
2470 "ab", the start and end offset values for the match are 2 and 0.
2471
2472 If a capturing subpattern group is matched repeatedly within a single
2473 match operation, it is the last portion of the subject that it matched
2474 that is returned.
2475
2476 If the ovector is too small to hold all the captured substring offsets,
2477 as much as possible is filled in, and the function returns a value of
2478 zero. If captured substrings are not of interest, pcre2_match() may be
2479 called with a match data block whose ovector is of minimum length (that
2480 is, one pair).
2481
2482 It is possible for capturing subpattern number n+1 to match some part
2483 of the subject when subpattern n has not been used at all. For example,
2484 if the string "abc" is matched against the pattern (a|(z))(bc) the
2485 return from the function is 4, and subpatterns 1 and 3 are matched, but
2486 2 is not. When this happens, both values in the offset pairs corre‐
2487 sponding to unused subpatterns are set to PCRE2_UNSET.
2488
2489 Offset values that correspond to unused subpatterns at the end of the
2490 expression are also set to PCRE2_UNSET. For example, if the string
2491 "abc" is matched against the pattern (abc)(x(yz)?)? subpatterns 2 and 3
2492 are not matched. The return from the function is 2, because the high‐
2493 est used capturing subpattern number is 1. The offsets for for the sec‐
2494 ond and third capturing subpatterns (assuming the vector is large
2495 enough, of course) are set to PCRE2_UNSET.
2496
2497 Elements in the ovector that do not correspond to capturing parentheses
2498 in the pattern are never changed. That is, if a pattern contains n cap‐
2499 turing parentheses, no more than ovector[0] to ovector[2n+1] are set by
2500 pcre2_match(). The other elements retain whatever values they previ‐
2501 ously had. After a failed match attempt, the contents of the ovector
2502 are unchanged.
2503
2505
2506 PCRE2_SPTR pcre2_get_mark(pcre2_match_data *match_data);
2507
2508 PCRE2_SIZE pcre2_get_startchar(pcre2_match_data *match_data);
2509
2510 As well as the offsets in the ovector, other information about a match
2511 is retained in the match data block and can be retrieved by the above
2512 functions in appropriate circumstances. If they are called at other
2513 times, the result is undefined.
2514
2515 After a successful match, a partial match (PCRE2_ERROR_PARTIAL), or a
2516 failure to match (PCRE2_ERROR_NOMATCH), a (*MARK), (*PRUNE), or (*THEN)
2517 name may be available. The function pcre2_get_mark() can be called to
2518 access this name. The same function applies to all three verbs. It
2519 returns a pointer to the zero-terminated name, which is within the com‐
2520 piled pattern. If no name is available, NULL is returned. The length of
2521 the name (excluding the terminating zero) is stored in the code unit
2522 that precedes the name. You should use this length instead of relying
2523 on the terminating zero if the name might contain a binary zero.
2524
2525 After a successful match, the name that is returned is the last
2526 (*MARK), (*PRUNE), or (*THEN) name encountered on the matching path
2527 through the pattern. Instances of (*PRUNE) and (*THEN) without names
2528 are ignored. Thus, for example, if the matching path contains
2529 (*MARK:A)(*PRUNE), the name "A" is returned. After a "no match" or a
2530 partial match, the last encountered name is returned. For example,
2531 consider this pattern:
2532
2533 ^(*MARK:A)((*MARK:B)a|b)c
2534
2535 When it matches "bc", the returned name is A. The B mark is "seen" in
2536 the first branch of the group, but it is not on the matching path. On
2537 the other hand, when this pattern fails to match "bx", the returned
2538 name is B.
2539
2540 Warning: By default, certain start-of-match optimizations are used to
2541 give a fast "no match" result in some situations. For example, if the
2542 anchoring is removed from the pattern above, there is an initial check
2543 for the presence of "c" in the subject before running the matching
2544 engine. This check fails for "bx", causing a match failure without see‐
2545 ing any marks. You can disable the start-of-match optimizations by set‐
2546 ting the PCRE2_NO_START_OPTIMIZE option for pcre2_compile() or starting
2547 the pattern with (*NO_START_OPT).
2548
2549 After a successful match, a partial match, or one of the invalid UTF
2550 errors (for example, PCRE2_ERROR_UTF8_ERR5), pcre2_get_startchar() can
2551 be called. After a successful or partial match it returns the code unit
2552 offset of the character at which the match started. For a non-partial
2553 match, this can be different to the value of ovector[0] if the pattern
2554 contains the \K escape sequence. After a partial match, however, this
2555 value is always the same as ovector[0] because \K does not affect the
2556 result of a partial match.
2557
2558 After a UTF check failure, pcre2_get_startchar() can be used to obtain
2559 the code unit offset of the invalid UTF character. Details are given in
2560 the pcre2unicode page.
2561
2563
2564 If pcre2_match() fails, it returns a negative number. This can be con‐
2565 verted to a text string by calling the pcre2_get_error_message() func‐
2566 tion (see "Obtaining a textual error message" below). Negative error
2567 codes are also returned by other functions, and are documented with
2568 them. The codes are given names in the header file. If UTF checking is
2569 in force and an invalid UTF subject string is detected, one of a number
2570 of UTF-specific negative error codes is returned. Details are given in
2571 the pcre2unicode page. The following are the other errors that may be
2572 returned by pcre2_match():
2573
2574 PCRE2_ERROR_NOMATCH
2575
2576 The subject string did not match the pattern.
2577
2578 PCRE2_ERROR_PARTIAL
2579
2580 The subject string did not match, but it did match partially. See the
2581 pcre2partial documentation for details of partial matching.
2582
2583 PCRE2_ERROR_BADMAGIC
2584
2585 PCRE2 stores a 4-byte "magic number" at the start of the compiled code,
2586 to catch the case when it is passed a junk pointer. This is the error
2587 that is returned when the magic number is not present.
2588
2589 PCRE2_ERROR_BADMODE
2590
2591 This error is given when a compiled pattern is passed to a function in
2592 a library of a different code unit width, for example, a pattern com‐
2593 piled by the 8-bit library is passed to a 16-bit or 32-bit library
2594 function.
2595
2596 PCRE2_ERROR_BADOFFSET
2597
2598 The value of startoffset was greater than the length of the subject.
2599
2600 PCRE2_ERROR_BADOPTION
2601
2602 An unrecognized bit was set in the options argument.
2603
2604 PCRE2_ERROR_BADUTFOFFSET
2605
2606 The UTF code unit sequence that was passed as a subject was checked and
2607 found to be valid (the PCRE2_NO_UTF_CHECK option was not set), but the
2608 value of startoffset did not point to the beginning of a UTF character
2609 or the end of the subject.
2610
2611 PCRE2_ERROR_CALLOUT
2612
2613 This error is never generated by pcre2_match() itself. It is provided
2614 for use by callout functions that want to cause pcre2_match() or
2615 pcre2_callout_enumerate() to return a distinctive error code. See the
2616 pcre2callout documentation for details.
2617
2618 PCRE2_ERROR_DEPTHLIMIT
2619
2620 The nested backtracking depth limit was reached.
2621
2622 PCRE2_ERROR_HEAPLIMIT
2623
2624 The heap limit was reached.
2625
2626 PCRE2_ERROR_INTERNAL
2627
2628 An unexpected internal error has occurred. This error could be caused
2629 by a bug in PCRE2 or by overwriting of the compiled pattern.
2630
2631 PCRE2_ERROR_JIT_STACKLIMIT
2632
2633 This error is returned when a pattern that was successfully studied
2634 using JIT is being matched, but the memory available for the just-in-
2635 time processing stack is not large enough. See the pcre2jit documenta‐
2636 tion for more details.
2637
2638 PCRE2_ERROR_MATCHLIMIT
2639
2640 The backtracking match limit was reached.
2641
2642 PCRE2_ERROR_NOMEMORY
2643
2644 If a pattern contains many nested backtracking points, heap memory is
2645 used to remember them. This error is given when the memory allocation
2646 function (default or custom) fails. Note that a different error,
2647 PCRE2_ERROR_HEAPLIMIT, is given if the amount of memory needed exceeds
2648 the heap limit.
2649
2650 PCRE2_ERROR_NULL
2651
2652 Either the code, subject, or match_data argument was passed as NULL.
2653
2654 PCRE2_ERROR_RECURSELOOP
2655
2656 This error is returned when pcre2_match() detects a recursion loop
2657 within the pattern. Specifically, it means that either the whole pat‐
2658 tern or a subpattern has been called recursively for the second time at
2659 the same position in the subject string. Some simple patterns that
2660 might do this are detected and faulted at compile time, but more com‐
2661 plicated cases, in particular mutual recursions between two different
2662 subpatterns, cannot be detected until matching is attempted.
2663
2665
2666 int pcre2_get_error_message(int errorcode, PCRE2_UCHAR *buffer,
2667 PCRE2_SIZE bufflen);
2668
2669 A text message for an error code from any PCRE2 function (compile,
2670 match, or auxiliary) can be obtained by calling pcre2_get_error_mes‐
2671 sage(). The code is passed as the first argument, with the remaining
2672 two arguments specifying a code unit buffer and its length in code
2673 units, into which the text message is placed. The message is returned
2674 in code units of the appropriate width for the library that is being
2675 used.
2676
2677 The returned message is terminated with a trailing zero, and the func‐
2678 tion returns the number of code units used, excluding the trailing
2679 zero. If the error number is unknown, the negative error code
2680 PCRE2_ERROR_BADDATA is returned. If the buffer is too small, the mes‐
2681 sage is truncated (but still with a trailing zero), and the negative
2682 error code PCRE2_ERROR_NOMEMORY is returned. None of the messages are
2683 very long; a buffer size of 120 code units is ample.
2684
2686
2687 int pcre2_substring_length_bynumber(pcre2_match_data *match_data,
2688 uint32_t number, PCRE2_SIZE *length);
2689
2690 int pcre2_substring_copy_bynumber(pcre2_match_data *match_data,
2691 uint32_t number, PCRE2_UCHAR *buffer,
2692 PCRE2_SIZE *bufflen);
2693
2694 int pcre2_substring_get_bynumber(pcre2_match_data *match_data,
2695 uint32_t number, PCRE2_UCHAR **bufferptr,
2696 PCRE2_SIZE *bufflen);
2697
2698 void pcre2_substring_free(PCRE2_UCHAR *buffer);
2699
2700 Captured substrings can be accessed directly by using the ovector as
2701 described above. For convenience, auxiliary functions are provided for
2702 extracting captured substrings as new, separate, zero-terminated
2703 strings. A substring that contains a binary zero is correctly extracted
2704 and has a further zero added on the end, but the result is not, of
2705 course, a C string.
2706
2707 The functions in this section identify substrings by number. The number
2708 zero refers to the entire matched substring, with higher numbers refer‐
2709 ring to substrings captured by parenthesized groups. After a partial
2710 match, only substring zero is available. An attempt to extract any
2711 other substring gives the error PCRE2_ERROR_PARTIAL. The next section
2712 describes similar functions for extracting captured substrings by name.
2713
2714 If a pattern uses the \K escape sequence within a positive assertion,
2715 the reported start of a successful match can be greater than the end of
2716 the match. For example, if the pattern (?=ab\K) is matched against
2717 "ab", the start and end offset values for the match are 2 and 0. In
2718 this situation, calling these functions with a zero substring number
2719 extracts a zero-length empty string.
2720
2721 You can find the length in code units of a captured substring without
2722 extracting it by calling pcre2_substring_length_bynumber(). The first
2723 argument is a pointer to the match data block, the second is the group
2724 number, and the third is a pointer to a variable into which the length
2725 is placed. If you just want to know whether or not the substring has
2726 been captured, you can pass the third argument as NULL.
2727
2728 The pcre2_substring_copy_bynumber() function copies a captured sub‐
2729 string into a supplied buffer, whereas pcre2_substring_get_bynumber()
2730 copies it into new memory, obtained using the same memory allocation
2731 function that was used for the match data block. The first two argu‐
2732 ments of these functions are a pointer to the match data block and a
2733 capturing group number.
2734
2735 The final arguments of pcre2_substring_copy_bynumber() are a pointer to
2736 the buffer and a pointer to a variable that contains its length in code
2737 units. This is updated to contain the actual number of code units used
2738 for the extracted substring, excluding the terminating zero.
2739
2740 For pcre2_substring_get_bynumber() the third and fourth arguments point
2741 to variables that are updated with a pointer to the new memory and the
2742 number of code units that comprise the substring, again excluding the
2743 terminating zero. When the substring is no longer needed, the memory
2744 should be freed by calling pcre2_substring_free().
2745
2746 The return value from all these functions is zero for success, or a
2747 negative error code. If the pattern match failed, the match failure
2748 code is returned. If a substring number greater than zero is used
2749 after a partial match, PCRE2_ERROR_PARTIAL is returned. Other possible
2750 error codes are:
2751
2752 PCRE2_ERROR_NOMEMORY
2753
2754 The buffer was too small for pcre2_substring_copy_bynumber(), or the
2755 attempt to get memory failed for pcre2_substring_get_bynumber().
2756
2757 PCRE2_ERROR_NOSUBSTRING
2758
2759 There is no substring with that number in the pattern, that is, the
2760 number is greater than the number of capturing parentheses.
2761
2762 PCRE2_ERROR_UNAVAILABLE
2763
2764 The substring number, though not greater than the number of captures in
2765 the pattern, is greater than the number of slots in the ovector, so the
2766 substring could not be captured.
2767
2768 PCRE2_ERROR_UNSET
2769
2770 The substring did not participate in the match. For example, if the
2771 pattern is (abc)|(def) and the subject is "def", and the ovector con‐
2772 tains at least two capturing slots, substring number 1 is unset.
2773
2775
2776 int pcre2_substring_list_get(pcre2_match_data *match_data,
2777 PCRE2_UCHAR ***listptr, PCRE2_SIZE **lengthsptr);
2778
2779 void pcre2_substring_list_free(PCRE2_SPTR *list);
2780
2781 The pcre2_substring_list_get() function extracts all available sub‐
2782 strings and builds a list of pointers to them. It also (optionally)
2783 builds a second list that contains their lengths (in code units),
2784 excluding a terminating zero that is added to each of them. All this is
2785 done in a single block of memory that is obtained using the same memory
2786 allocation function that was used to get the match data block.
2787
2788 This function must be called only after a successful match. If called
2789 after a partial match, the error code PCRE2_ERROR_PARTIAL is returned.
2790
2791 The address of the memory block is returned via listptr, which is also
2792 the start of the list of string pointers. The end of the list is marked
2793 by a NULL pointer. The address of the list of lengths is returned via
2794 lengthsptr. If your strings do not contain binary zeros and you do not
2795 therefore need the lengths, you may supply NULL as the lengthsptr argu‐
2796 ment to disable the creation of a list of lengths. The yield of the
2797 function is zero if all went well, or PCRE2_ERROR_NOMEMORY if the mem‐
2798 ory block could not be obtained. When the list is no longer needed, it
2799 should be freed by calling pcre2_substring_list_free().
2800
2801 If this function encounters a substring that is unset, which can happen
2802 when capturing subpattern number n+1 matches some part of the subject,
2803 but subpattern n has not been used at all, it returns an empty string.
2804 This can be distinguished from a genuine zero-length substring by
2805 inspecting the appropriate offset in the ovector, which contain
2806 PCRE2_UNSET for unset substrings, or by calling pcre2_sub‐
2807 string_length_bynumber().
2808
2810
2811 int pcre2_substring_number_from_name(const pcre2_code *code,
2812 PCRE2_SPTR name);
2813
2814 int pcre2_substring_length_byname(pcre2_match_data *match_data,
2815 PCRE2_SPTR name, PCRE2_SIZE *length);
2816
2817 int pcre2_substring_copy_byname(pcre2_match_data *match_data,
2818 PCRE2_SPTR name, PCRE2_UCHAR *buffer, PCRE2_SIZE *bufflen);
2819
2820 int pcre2_substring_get_byname(pcre2_match_data *match_data,
2821 PCRE2_SPTR name, PCRE2_UCHAR **bufferptr, PCRE2_SIZE *bufflen);
2822
2823 void pcre2_substring_free(PCRE2_UCHAR *buffer);
2824
2825 To extract a substring by name, you first have to find associated num‐
2826 ber. For example, for this pattern:
2827
2828 (a+)b(?<xxx>\d+)...
2829
2830 the number of the subpattern called "xxx" is 2. If the name is known to
2831 be unique (PCRE2_DUPNAMES was not set), you can find the number from
2832 the name by calling pcre2_substring_number_from_name(). The first argu‐
2833 ment is the compiled pattern, and the second is the name. The yield of
2834 the function is the subpattern number, PCRE2_ERROR_NOSUBSTRING if there
2835 is no subpattern of that name, or PCRE2_ERROR_NOUNIQUESUBSTRING if
2836 there is more than one subpattern of that name. Given the number, you
2837 can extract the substring directly from the ovector, or use one of the
2838 "bynumber" functions described above.
2839
2840 For convenience, there are also "byname" functions that correspond to
2841 the "bynumber" functions, the only difference being that the second
2842 argument is a name instead of a number. If PCRE2_DUPNAMES is set and
2843 there are duplicate names, these functions scan all the groups with the
2844 given name, and return the first named string that is set.
2845
2846 If there are no groups with the given name, PCRE2_ERROR_NOSUBSTRING is
2847 returned. If all groups with the name have numbers that are greater
2848 than the number of slots in the ovector, PCRE2_ERROR_UNAVAILABLE is
2849 returned. If there is at least one group with a slot in the ovector,
2850 but no group is found to be set, PCRE2_ERROR_UNSET is returned.
2851
2852 Warning: If the pattern uses the (?| feature to set up multiple subpat‐
2853 terns with the same number, as described in the section on duplicate
2854 subpattern numbers in the pcre2pattern page, you cannot use names to
2855 distinguish the different subpatterns, because names are not included
2856 in the compiled code. The matching process uses only numbers. For this
2857 reason, the use of different names for subpatterns of the same number
2858 causes an error at compile time.
2859
2861
2862 int pcre2_substitute(const pcre2_code *code, PCRE2_SPTR subject,
2863 PCRE2_SIZE length, PCRE2_SIZE startoffset,
2864 uint32_t options, pcre2_match_data *match_data,
2865 pcre2_match_context *mcontext, PCRE2_SPTR replacement,
2866 PCRE2_SIZE rlength, PCRE2_UCHAR *outputbufferfP,
2867 PCRE2_SIZE *outlengthptr);
2868
2869 This function calls pcre2_match() and then makes a copy of the subject
2870 string in outputbuffer, replacing the part that was matched with the
2871 replacement string, whose length is supplied in rlength. This can be
2872 given as PCRE2_ZERO_TERMINATED for a zero-terminated string. Matches in
2873 which a \K item in a lookahead in the pattern causes the match to end
2874 before it starts are not supported, and give rise to an error return.
2875 For global replacements, matches in which \K in a lookbehind causes the
2876 match to start earlier than the point that was reached in the previous
2877 iteration are also not supported.
2878
2879 The first seven arguments of pcre2_substitute() are the same as for
2880 pcre2_match(), except that the partial matching options are not permit‐
2881 ted, and match_data may be passed as NULL, in which case a match data
2882 block is obtained and freed within this function, using memory manage‐
2883 ment functions from the match context, if provided, or else those that
2884 were used to allocate memory for the compiled code.
2885
2886 If an external match_data block is provided, its contents afterwards
2887 are those set by the final call to pcre2_match(), which will have ended
2888 in a matching error. The contents of the ovector within the match data
2889 block may or may not have been changed.
2890
2891 The outlengthptr argument must point to a variable that contains the
2892 length, in code units, of the output buffer. If the function is suc‐
2893 cessful, the value is updated to contain the length of the new string,
2894 excluding the trailing zero that is automatically added.
2895
2896 If the function is not successful, the value set via outlengthptr
2897 depends on the type of error. For syntax errors in the replacement
2898 string, the value is the offset in the replacement string where the
2899 error was detected. For other errors, the value is PCRE2_UNSET by
2900 default. This includes the case of the output buffer being too small,
2901 unless PCRE2_SUBSTITUTE_OVERFLOW_LENGTH is set (see below), in which
2902 case the value is the minimum length needed, including space for the
2903 trailing zero. Note that in order to compute the required length,
2904 pcre2_substitute() has to simulate all the matching and copying,
2905 instead of giving an error return as soon as the buffer overflows. Note
2906 also that the length is in code units, not bytes.
2907
2908 In the replacement string, which is interpreted as a UTF string in UTF
2909 mode, and is checked for UTF validity unless the PCRE2_NO_UTF_CHECK
2910 option is set, a dollar character is an escape character that can spec‐
2911 ify the insertion of characters from capturing groups or (*MARK),
2912 (*PRUNE), or (*THEN) items in the pattern. The following forms are
2913 always recognized:
2914
2915 $$ insert a dollar character
2916 $<n> or ${<n>} insert the contents of group <n>
2917 $*MARK or ${*MARK} insert a (*MARK), (*PRUNE), or (*THEN) name
2918
2919 Either a group number or a group name can be given for <n>. Curly
2920 brackets are required only if the following character would be inter‐
2921 preted as part of the number or name. The number may be zero to include
2922 the entire matched string. For example, if the pattern a(b)c is
2923 matched with "=abc=" and the replacement string "+$1$0$1+", the result
2924 is "=+babcb+=".
2925
2926 $*MARK inserts the name from the last encountered (*MARK), (*PRUNE), or
2927 (*THEN) on the matching path that has a name. (*MARK) must always
2928 include a name, but (*PRUNE) and (*THEN) need not. For example, in the
2929 case of (*MARK:A)(*PRUNE) the name inserted is "A", but for
2930 (*MARK:A)(*PRUNE:B) the relevant name is "B". This facility can be
2931 used to perform simple simultaneous substitutions, as this pcre2test
2932 example shows:
2933
2934 /(*MARK:pear)apple|(*MARK:orange)lemon/g,replace=${*MARK}
2935 apple lemon
2936 2: pear orange
2937
2938 As well as the usual options for pcre2_match(), a number of additional
2939 options can be set in the options argument of pcre2_substitute().
2940
2941 PCRE2_SUBSTITUTE_GLOBAL causes the function to iterate over the subject
2942 string, replacing every matching substring. If this option is not set,
2943 only the first matching substring is replaced. The search for matches
2944 takes place in the original subject string (that is, previous replace‐
2945 ments do not affect it). Iteration is implemented by advancing the
2946 startoffset value for each search, which is always passed the entire
2947 subject string. If an offset limit is set in the match context, search‐
2948 ing stops when that limit is reached.
2949
2950 You can restrict the effect of a global substitution to a portion of
2951 the subject string by setting either or both of startoffset and an off‐
2952 set limit. Here is a pcre2test example:
2953
2954 /B/g,replace=!,use_offset_limit
2955 ABC ABC ABC ABC\=offset=3,offset_limit=12
2956 2: ABC A!C A!C ABC
2957
2958 When continuing with global substitutions after matching a substring
2959 with zero length, an attempt to find a non-empty match at the same off‐
2960 set is performed. If this is not successful, the offset is advanced by
2961 one character except when CRLF is a valid newline sequence and the next
2962 two characters are CR, LF. In this case, the offset is advanced by two
2963 characters.
2964
2965 PCRE2_SUBSTITUTE_OVERFLOW_LENGTH changes what happens when the output
2966 buffer is too small. The default action is to return PCRE2_ERROR_NOMEM‐
2967 ORY immediately. If this option is set, however, pcre2_substitute()
2968 continues to go through the motions of matching and substituting (with‐
2969 out, of course, writing anything) in order to compute the size of buf‐
2970 fer that is needed. This value is passed back via the outlengthptr
2971 variable, with the result of the function still being
2972 PCRE2_ERROR_NOMEMORY.
2973
2974 Passing a buffer size of zero is a permitted way of finding out how
2975 much memory is needed for given substitution. However, this does mean
2976 that the entire operation is carried out twice. Depending on the appli‐
2977 cation, it may be more efficient to allocate a large buffer and free
2978 the excess afterwards, instead of using PCRE2_SUBSTITUTE_OVER‐
2979 FLOW_LENGTH.
2980
2981 PCRE2_SUBSTITUTE_UNKNOWN_UNSET causes references to capturing groups
2982 that do not appear in the pattern to be treated as unset groups. This
2983 option should be used with care, because it means that a typo in a
2984 group name or number no longer causes the PCRE2_ERROR_NOSUBSTRING
2985 error.
2986
2987 PCRE2_SUBSTITUTE_UNSET_EMPTY causes unset capturing groups (including
2988 unknown groups when PCRE2_SUBSTITUTE_UNKNOWN_UNSET is set) to be
2989 treated as empty strings when inserted as described above. If this
2990 option is not set, an attempt to insert an unset group causes the
2991 PCRE2_ERROR_UNSET error. This option does not influence the extended
2992 substitution syntax described below.
2993
2994 PCRE2_SUBSTITUTE_EXTENDED causes extra processing to be applied to the
2995 replacement string. Without this option, only the dollar character is
2996 special, and only the group insertion forms listed above are valid.
2997 When PCRE2_SUBSTITUTE_EXTENDED is set, two things change:
2998
2999 Firstly, backslash in a replacement string is interpreted as an escape
3000 character. The usual forms such as \n or \x{ddd} can be used to specify
3001 particular character codes, and backslash followed by any non-alphanu‐
3002 meric character quotes that character. Extended quoting can be coded
3003 using \Q...\E, exactly as in pattern strings.
3004
3005 There are also four escape sequences for forcing the case of inserted
3006 letters. The insertion mechanism has three states: no case forcing,
3007 force upper case, and force lower case. The escape sequences change the
3008 current state: \U and \L change to upper or lower case forcing, respec‐
3009 tively, and \E (when not terminating a \Q quoted sequence) reverts to
3010 no case forcing. The sequences \u and \l force the next character (if
3011 it is a letter) to upper or lower case, respectively, and then the
3012 state automatically reverts to no case forcing. Case forcing applies to
3013 all inserted characters, including those from captured groups and let‐
3014 ters within \Q...\E quoted sequences.
3015
3016 Note that case forcing sequences such as \U...\E do not nest. For exam‐
3017 ple, the result of processing "\Uaa\LBB\Ecc\E" is "AAbbcc"; the final
3018 \E has no effect.
3019
3020 The second effect of setting PCRE2_SUBSTITUTE_EXTENDED is to add more
3021 flexibility to group substitution. The syntax is similar to that used
3022 by Bash:
3023
3024 ${<n>:-<string>}
3025 ${<n>:+<string1>:<string2>}
3026
3027 As before, <n> may be a group number or a name. The first form speci‐
3028 fies a default value. If group <n> is set, its value is inserted; if
3029 not, <string> is expanded and the result inserted. The second form
3030 specifies strings that are expanded and inserted when group <n> is set
3031 or unset, respectively. The first form is just a convenient shorthand
3032 for
3033
3034 ${<n>:+${<n>}:<string>}
3035
3036 Backslash can be used to escape colons and closing curly brackets in
3037 the replacement strings. A change of the case forcing state within a
3038 replacement string remains in force afterwards, as shown in this
3039 pcre2test example:
3040
3041 /(some)?(body)/substitute_extended,replace=${1:+\U:\L}HeLLo
3042 body
3043 1: hello
3044 somebody
3045 1: HELLO
3046
3047 The PCRE2_SUBSTITUTE_UNSET_EMPTY option does not affect these extended
3048 substitutions. However, PCRE2_SUBSTITUTE_UNKNOWN_UNSET does cause
3049 unknown groups in the extended syntax forms to be treated as unset.
3050
3051 If successful, pcre2_substitute() returns the number of replacements
3052 that were made. This may be zero if no matches were found, and is never
3053 greater than 1 unless PCRE2_SUBSTITUTE_GLOBAL is set.
3054
3055 In the event of an error, a negative error code is returned. Except for
3056 PCRE2_ERROR_NOMATCH (which is never returned), errors from
3057 pcre2_match() are passed straight back.
3058
3059 PCRE2_ERROR_NOSUBSTRING is returned for a non-existent substring inser‐
3060 tion, unless PCRE2_SUBSTITUTE_UNKNOWN_UNSET is set.
3061
3062 PCRE2_ERROR_UNSET is returned for an unset substring insertion (includ‐
3063 ing an unknown substring when PCRE2_SUBSTITUTE_UNKNOWN_UNSET is set)
3064 when the simple (non-extended) syntax is used and PCRE2_SUBSTI‐
3065 TUTE_UNSET_EMPTY is not set.
3066
3067 PCRE2_ERROR_NOMEMORY is returned if the output buffer is not big
3068 enough. If the PCRE2_SUBSTITUTE_OVERFLOW_LENGTH option is set, the size
3069 of buffer that is needed is returned via outlengthptr. Note that this
3070 does not happen by default.
3071
3072 PCRE2_ERROR_BADREPLACEMENT is used for miscellaneous syntax errors in
3073 the replacement string, with more particular errors being
3074 PCRE2_ERROR_BADREPESCAPE (invalid escape sequence), PCRE2_ERROR_REP‐
3075 MISSINGBRACE (closing curly bracket not found), PCRE2_ERROR_BADSUBSTI‐
3076 TUTION (syntax error in extended group substitution), and
3077 PCRE2_ERROR_BADSUBSPATTERN (the pattern match ended before it started
3078 or the match started earlier than the current position in the subject,
3079 which can happen if \K is used in an assertion).
3080
3081 As for all PCRE2 errors, a text message that describes the error can be
3082 obtained by calling the pcre2_get_error_message() function (see
3083 "Obtaining a textual error message" above).
3084
3086
3087 int pcre2_substring_nametable_scan(const pcre2_code *code,
3088 PCRE2_SPTR name, PCRE2_SPTR *first, PCRE2_SPTR *last);
3089
3090 When a pattern is compiled with the PCRE2_DUPNAMES option, names for
3091 subpatterns are not required to be unique. Duplicate names are always
3092 allowed for subpatterns with the same number, created by using the (?|
3093 feature. Indeed, if such subpatterns are named, they are required to
3094 use the same names.
3095
3096 Normally, patterns with duplicate names are such that in any one match,
3097 only one of the named subpatterns participates. An example is shown in
3098 the pcre2pattern documentation.
3099
3100 When duplicates are present, pcre2_substring_copy_byname() and
3101 pcre2_substring_get_byname() return the first substring corresponding
3102 to the given name that is set. Only if none are set is
3103 PCRE2_ERROR_UNSET is returned. The pcre2_substring_number_from_name()
3104 function returns the error PCRE2_ERROR_NOUNIQUESUBSTRING when there are
3105 duplicate names.
3106
3107 If you want to get full details of all captured substrings for a given
3108 name, you must use the pcre2_substring_nametable_scan() function. The
3109 first argument is the compiled pattern, and the second is the name. If
3110 the third and fourth arguments are NULL, the function returns a group
3111 number for a unique name, or PCRE2_ERROR_NOUNIQUESUBSTRING otherwise.
3112
3113 When the third and fourth arguments are not NULL, they must be pointers
3114 to variables that are updated by the function. After it has run, they
3115 point to the first and last entries in the name-to-number table for the
3116 given name, and the function returns the length of each entry in code
3117 units. In both cases, PCRE2_ERROR_NOSUBSTRING is returned if there are
3118 no entries for the given name.
3119
3120 The format of the name table is described above in the section entitled
3121 Information about a pattern. Given all the relevant entries for the
3122 name, you can extract each of their numbers, and hence the captured
3123 data.
3124
3126
3127 The traditional matching function uses a similar algorithm to Perl,
3128 which stops when it finds the first match at a given point in the sub‐
3129 ject. If you want to find all possible matches, or the longest possible
3130 match at a given position, consider using the alternative matching
3131 function (see below) instead. If you cannot use the alternative func‐
3132 tion, you can kludge it up by making use of the callout facility, which
3133 is described in the pcre2callout documentation.
3134
3135 What you have to do is to insert a callout right at the end of the pat‐
3136 tern. When your callout function is called, extract and save the cur‐
3137 rent matched substring. Then return 1, which forces pcre2_match() to
3138 backtrack and try other alternatives. Ultimately, when it runs out of
3139 matches, pcre2_match() will yield PCRE2_ERROR_NOMATCH.
3140
3142
3143 int pcre2_dfa_match(const pcre2_code *code, PCRE2_SPTR subject,
3144 PCRE2_SIZE length, PCRE2_SIZE startoffset,
3145 uint32_t options, pcre2_match_data *match_data,
3146 pcre2_match_context *mcontext,
3147 int *workspace, PCRE2_SIZE wscount);
3148
3149 The function pcre2_dfa_match() is called to match a subject string
3150 against a compiled pattern, using a matching algorithm that scans the
3151 subject string just once (not counting lookaround assertions), and does
3152 not backtrack. This has different characteristics to the normal algo‐
3153 rithm, and is not compatible with Perl. Some of the features of PCRE2
3154 patterns are not supported. Nevertheless, there are times when this
3155 kind of matching can be useful. For a discussion of the two matching
3156 algorithms, and a list of features that pcre2_dfa_match() does not sup‐
3157 port, see the pcre2matching documentation.
3158
3159 The arguments for the pcre2_dfa_match() function are the same as for
3160 pcre2_match(), plus two extras. The ovector within the match data block
3161 is used in a different way, and this is described below. The other com‐
3162 mon arguments are used in the same way as for pcre2_match(), so their
3163 description is not repeated here.
3164
3165 The two additional arguments provide workspace for the function. The
3166 workspace vector should contain at least 20 elements. It is used for
3167 keeping track of multiple paths through the pattern tree. More
3168 workspace is needed for patterns and subjects where there are a lot of
3169 potential matches.
3170
3171 Here is an example of a simple call to pcre2_dfa_match():
3172
3173 int wspace[20];
3174 pcre2_match_data *md = pcre2_match_data_create(4, NULL);
3175 int rc = pcre2_dfa_match(
3176 re, /* result of pcre2_compile() */
3177 "some string", /* the subject string */
3178 11, /* the length of the subject string */
3179 0, /* start at offset 0 in the subject */
3180 0, /* default options */
3181 md, /* the match data block */
3182 NULL, /* a match context; NULL means use defaults */
3183 wspace, /* working space vector */
3184 20); /* number of elements (NOT size in bytes) */
3185
3186 Option bits for pcre_dfa_match()
3187
3188 The unused bits of the options argument for pcre2_dfa_match() must be
3189 zero. The only bits that may be set are PCRE2_ANCHORED, PCRE2_ENDAN‐
3190 CHORED, PCRE2_NOTBOL, PCRE2_NOTEOL, PCRE2_NOTEMPTY,
3191 PCRE2_NOTEMPTY_ATSTART, PCRE2_NO_UTF_CHECK, PCRE2_PARTIAL_HARD,
3192 PCRE2_PARTIAL_SOFT, PCRE2_DFA_SHORTEST, and PCRE2_DFA_RESTART. All but
3193 the last four of these are exactly the same as for pcre2_match(), so
3194 their description is not repeated here.
3195
3196 PCRE2_PARTIAL_HARD
3197 PCRE2_PARTIAL_SOFT
3198
3199 These have the same general effect as they do for pcre2_match(), but
3200 the details are slightly different. When PCRE2_PARTIAL_HARD is set for
3201 pcre2_dfa_match(), it returns PCRE2_ERROR_PARTIAL if the end of the
3202 subject is reached and there is still at least one matching possibility
3203 that requires additional characters. This happens even if some complete
3204 matches have already been found. When PCRE2_PARTIAL_SOFT is set, the
3205 return code PCRE2_ERROR_NOMATCH is converted into PCRE2_ERROR_PARTIAL
3206 if the end of the subject is reached, there have been no complete
3207 matches, but there is still at least one matching possibility. The por‐
3208 tion of the string that was inspected when the longest partial match
3209 was found is set as the first matching string in both cases. There is a
3210 more detailed discussion of partial and multi-segment matching, with
3211 examples, in the pcre2partial documentation.
3212
3213 PCRE2_DFA_SHORTEST
3214
3215 Setting the PCRE2_DFA_SHORTEST option causes the matching algorithm to
3216 stop as soon as it has found one match. Because of the way the alterna‐
3217 tive algorithm works, this is necessarily the shortest possible match
3218 at the first possible matching point in the subject string.
3219
3220 PCRE2_DFA_RESTART
3221
3222 When pcre2_dfa_match() returns a partial match, it is possible to call
3223 it again, with additional subject characters, and have it continue with
3224 the same match. The PCRE2_DFA_RESTART option requests this action; when
3225 it is set, the workspace and wscount options must reference the same
3226 vector as before because data about the match so far is left in them
3227 after a partial match. There is more discussion of this facility in the
3228 pcre2partial documentation.
3229
3230 Successful returns from pcre2_dfa_match()
3231
3232 When pcre2_dfa_match() succeeds, it may have matched more than one sub‐
3233 string in the subject. Note, however, that all the matches from one run
3234 of the function start at the same point in the subject. The shorter
3235 matches are all initial substrings of the longer matches. For example,
3236 if the pattern
3237
3238 <.*>
3239
3240 is matched against the string
3241
3242 This is <something> <something else> <something further> no more
3243
3244 the three matched strings are
3245
3246 <something> <something else> <something further>
3247 <something> <something else>
3248 <something>
3249
3250 On success, the yield of the function is a number greater than zero,
3251 which is the number of matched substrings. The offsets of the sub‐
3252 strings are returned in the ovector, and can be extracted by number in
3253 the same way as for pcre2_match(), but the numbers bear no relation to
3254 any capturing groups that may exist in the pattern, because DFA match‐
3255 ing does not support group capture.
3256
3257 Calls to the convenience functions that extract substrings by name
3258 return the error PCRE2_ERROR_DFA_UFUNC (unsupported function) if used
3259 after a DFA match. The convenience functions that extract substrings by
3260 number never return PCRE2_ERROR_NOSUBSTRING.
3261
3262 The matched strings are stored in the ovector in reverse order of
3263 length; that is, the longest matching string is first. If there were
3264 too many matches to fit into the ovector, the yield of the function is
3265 zero, and the vector is filled with the longest matches.
3266
3267 NOTE: PCRE2's "auto-possessification" optimization usually applies to
3268 character repeats at the end of a pattern (as well as internally). For
3269 example, the pattern "a\d+" is compiled as if it were "a\d++". For DFA
3270 matching, this means that only one possible match is found. If you
3271 really do want multiple matches in such cases, either use an ungreedy
3272 repeat such as "a\d+?" or set the PCRE2_NO_AUTO_POSSESS option when
3273 compiling.
3274
3275 Error returns from pcre2_dfa_match()
3276
3277 The pcre2_dfa_match() function returns a negative number when it fails.
3278 Many of the errors are the same as for pcre2_match(), as described
3279 above. There are in addition the following errors that are specific to
3280 pcre2_dfa_match():
3281
3282 PCRE2_ERROR_DFA_UITEM
3283
3284 This return is given if pcre2_dfa_match() encounters an item in the
3285 pattern that it does not support, for instance, the use of \C in a UTF
3286 mode or a backreference.
3287
3288 PCRE2_ERROR_DFA_UCOND
3289
3290 This return is given if pcre2_dfa_match() encounters a condition item
3291 that uses a backreference for the condition, or a test for recursion in
3292 a specific group. These are not supported.
3293
3294 PCRE2_ERROR_DFA_WSSIZE
3295
3296 This return is given if pcre2_dfa_match() runs out of space in the
3297 workspace vector.
3298
3299 PCRE2_ERROR_DFA_RECURSE
3300
3301 When a recursive subpattern is processed, the matching function calls
3302 itself recursively, using private memory for the ovector and workspace.
3303 This error is given if the internal ovector is not large enough. This
3304 should be extremely rare, as a vector of size 1000 is used.
3305
3306 PCRE2_ERROR_DFA_BADRESTART
3307
3308 When pcre2_dfa_match() is called with the PCRE2_DFA_RESTART option,
3309 some plausibility checks are made on the contents of the workspace,
3310 which should contain data about the previous partial match. If any of
3311 these checks fail, this error is given.
3312
3314
3315 pcre2build(3), pcre2callout(3), pcre2demo(3), pcre2matching(3),
3316 pcre2partial(3), pcre2posix(3), pcre2sample(3), pcre2unicode(3).
3317
3319
3320 Philip Hazel
3321 University Computing Service
3322 Cambridge, England.
3323
3325
3326 Last updated: 07 September 2018
3327 Copyright (c) 1997-2018 University of Cambridge.
3328
3329
3330
3331PCRE2 10.32 07 September 2018 PCRE2API(3)