1PCRE2API(3) Library Functions Manual PCRE2API(3)
2
6 PCRE2 - Perl-compatible regular expressions (revised API)
7 #include <pcre2.h>
9 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 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 void pcre2_code_free(pcre2_code *code);
21 pcre2_match_data *pcre2_match_data_create(uint32_t ovecsize,
23 pcre2_general_context *gcontext);
24 pcre2_match_data *pcre2_match_data_create_from_pattern(
26 const pcre2_code *code, pcre2_general_context *gcontext);
27 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 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 void pcre2_match_data_free(pcre2_match_data *match_data);
40
42 PCRE2_SPTR pcre2_get_mark(pcre2_match_data *match_data);
44 uint32_t pcre2_get_ovector_count(pcre2_match_data *match_data);
46 PCRE2_SIZE *pcre2_get_ovector_pointer(pcre2_match_data *match_data);
48 PCRE2_SIZE pcre2_get_startchar(pcre2_match_data *match_data);
50
52 pcre2_general_context *pcre2_general_context_create(
54 void *(*private_malloc)(PCRE2_SIZE, void *),
55 void (*private_free)(void *, void *), void *memory_data);
56 pcre2_general_context *pcre2_general_context_copy(
58 pcre2_general_context *gcontext);
59 void pcre2_general_context_free(pcre2_general_context *gcontext);
61
63 pcre2_compile_context *pcre2_compile_context_create(
65 pcre2_general_context *gcontext);
66 pcre2_compile_context *pcre2_compile_context_copy(
68 pcre2_compile_context *ccontext);
69 void pcre2_compile_context_free(pcre2_compile_context *ccontext);
71 int pcre2_set_bsr(pcre2_compile_context *ccontext,
73 uint32_t value);
74 int pcre2_set_character_tables(pcre2_compile_context *ccontext,
76 const uint8_t *tables);
77 int pcre2_set_compile_extra_options(pcre2_compile_context *ccontext,
79 uint32_t extra_options);
80 int pcre2_set_max_pattern_length(pcre2_compile_context *ccontext,
82 PCRE2_SIZE value);
83 int pcre2_set_newline(pcre2_compile_context *ccontext,
85 uint32_t value);
86 int pcre2_set_parens_nest_limit(pcre2_compile_context *ccontext,
88 uint32_t value);
89 int pcre2_set_compile_recursion_guard(pcre2_compile_context *ccontext,
91 int (*guard_function)(uint32_t, void *), void *user_data);
92
94 pcre2_match_context *pcre2_match_context_create(
96 pcre2_general_context *gcontext);
97 pcre2_match_context *pcre2_match_context_copy(
99 pcre2_match_context *mcontext);
100 void pcre2_match_context_free(pcre2_match_context *mcontext);
102 int pcre2_set_callout(pcre2_match_context *mcontext,
104 int (*callout_function)(pcre2_callout_block *, void *),
105 void *callout_data);
106 int pcre2_set_substitute_callout(pcre2_match_context *mcontext,
108 int (*callout_function)(pcre2_substitute_callout_block *, void *),
109 void *callout_data);
110 int pcre2_set_offset_limit(pcre2_match_context *mcontext,
112 PCRE2_SIZE value);
113 int pcre2_set_heap_limit(pcre2_match_context *mcontext,
115 uint32_t value);
116 int pcre2_set_match_limit(pcre2_match_context *mcontext,
118 uint32_t value);
119 int pcre2_set_depth_limit(pcre2_match_context *mcontext,
121 uint32_t value);
122
124 int pcre2_substring_copy_byname(pcre2_match_data *match_data,
126 PCRE2_SPTR name, PCRE2_UCHAR *buffer, PCRE2_SIZE *bufflen);
127 int pcre2_substring_copy_bynumber(pcre2_match_data *match_data,
129 uint32_t number, PCRE2_UCHAR *buffer,
130 PCRE2_SIZE *bufflen);
131 void pcre2_substring_free(PCRE2_UCHAR *buffer);
133 int pcre2_substring_get_byname(pcre2_match_data *match_data,
135 PCRE2_SPTR name, PCRE2_UCHAR **bufferptr, PCRE2_SIZE *bufflen);
136 int pcre2_substring_get_bynumber(pcre2_match_data *match_data,
138 uint32_t number, PCRE2_UCHAR **bufferptr,
139 PCRE2_SIZE *bufflen);
140 int pcre2_substring_length_byname(pcre2_match_data *match_data,
142 PCRE2_SPTR name, PCRE2_SIZE *length);
143 int pcre2_substring_length_bynumber(pcre2_match_data *match_data,
145 uint32_t number, PCRE2_SIZE *length);
146 int pcre2_substring_nametable_scan(const pcre2_code *code,
148 PCRE2_SPTR name, PCRE2_SPTR *first, PCRE2_SPTR *last);
149 int pcre2_substring_number_from_name(const pcre2_code *code,
151 PCRE2_SPTR name);
152 void pcre2_substring_list_free(PCRE2_SPTR *list);
154 int pcre2_substring_list_get(pcre2_match_data *match_data,
156 PCRE2_UCHAR ***listptr, PCRE2_SIZE **lengthsptr);
157
159 int pcre2_substitute(const pcre2_code *code, PCRE2_SPTR subject,
161 PCRE2_SIZE length, PCRE2_SIZE startoffset,
162 uint32_t options, pcre2_match_data *match_data,
163 pcre2_match_context *mcontext, PCRE2_SPTR replacementz,
164 PCRE2_SIZE rlength, PCRE2_UCHAR *outputbuffer,
165 PCRE2_SIZE *outlengthptr);
166
168 int pcre2_jit_compile(pcre2_code *code, uint32_t options);
170 int pcre2_jit_match(const pcre2_code *code, PCRE2_SPTR subject,
172 PCRE2_SIZE length, PCRE2_SIZE startoffset,
173 uint32_t options, pcre2_match_data *match_data,
174 pcre2_match_context *mcontext);
175 void pcre2_jit_free_unused_memory(pcre2_general_context *gcontext);
177 pcre2_jit_stack *pcre2_jit_stack_create(PCRE2_SIZE startsize,
179 PCRE2_SIZE maxsize, pcre2_general_context *gcontext);
180 void pcre2_jit_stack_assign(pcre2_match_context *mcontext,
182 pcre2_jit_callback callback_function, void *callback_data);
183 void pcre2_jit_stack_free(pcre2_jit_stack *jit_stack);
185
187 int32_t pcre2_serialize_decode(pcre2_code **codes,
189 int32_t number_of_codes, const uint8_t *bytes,
190 pcre2_general_context *gcontext);
191 int32_t pcre2_serialize_encode(const pcre2_code **codes,
193 int32_t number_of_codes, uint8_t **serialized_bytes,
194 PCRE2_SIZE *serialized_size, pcre2_general_context *gcontext);
195 void pcre2_serialize_free(uint8_t *bytes);
197 int32_t pcre2_serialize_get_number_of_codes(const uint8_t *bytes);
199
201 pcre2_code *pcre2_code_copy(const pcre2_code *code);
203 pcre2_code *pcre2_code_copy_with_tables(const pcre2_code *code);
205 int pcre2_get_error_message(int errorcode, PCRE2_UCHAR *buffer,
207 PCRE2_SIZE bufflen);
208 const uint8_t *pcre2_maketables(pcre2_general_context *gcontext);
210 void pcre2_maketables_free(pcre2_general_context *gcontext,
212 const uint8_t *tables);
213 int pcre2_pattern_info(const pcre2_code *code, uint32_t what,
215 void *where);
216 int pcre2_callout_enumerate(const pcre2_code *code,
218 int (*callback)(pcre2_callout_enumerate_block *, void *),
219 void *user_data);
220 int pcre2_config(uint32_t what, void *where);
222
224 int pcre2_set_recursion_limit(pcre2_match_context *mcontext,
226 uint32_t value);
227 int pcre2_set_recursion_memory_management(
229 pcre2_match_context *mcontext,
230 void *(*private_malloc)(PCRE2_SIZE, void *),
231 void (*private_free)(void *, void *), void *memory_data);
232 These functions became obsolete at release 10.30 and are retained only
234 for backward compatibility. They should not be used in new code. The
235 first is replaced by pcre2_set_depth_limit(); the second is no longer
236 needed and has no effect (it always returns zero).
237
239 pcre2_convert_context *pcre2_convert_context_create(
241 pcre2_general_context *gcontext);
242 pcre2_convert_context *pcre2_convert_context_copy(
244 pcre2_convert_context *cvcontext);
245 void pcre2_convert_context_free(pcre2_convert_context *cvcontext);
247 int pcre2_set_glob_escape(pcre2_convert_context *cvcontext,
249 uint32_t escape_char);
250 int pcre2_set_glob_separator(pcre2_convert_context *cvcontext,
252 uint32_t separator_char);
253 int pcre2_pattern_convert(PCRE2_SPTR pattern, PCRE2_SIZE length,
255 uint32_t options, PCRE2_UCHAR **buffer,
256 PCRE2_SIZE *blength, pcre2_convert_context *cvcontext);
257 void pcre2_converted_pattern_free(PCRE2_UCHAR *converted_pattern);
259 These functions provide a way of converting non-PCRE2 patterns into
261 patterns that can be processed by pcre2_compile(). This facility is
262 experimental and may be changed in future releases. At present, "globs"
263 and POSIX basic and extended patterns can be converted. Details are
264 given in the pcre2convert documentation.
265
267 There are three PCRE2 libraries, supporting 8-bit, 16-bit, and 32-bit
269 code units, respectively. However, there is just one header file,
270 pcre2.h. This contains the function prototypes and other definitions
271 for all three libraries. One, two, or all three can be installed simul‐
272 taneously. On Unix-like systems the libraries are called libpcre2-8,
273 libpcre2-16, and libpcre2-32, and they can also co-exist with the orig‐
274 inal PCRE libraries.
275 Character strings are passed to and from a PCRE2 library as a sequence
277 of unsigned integers in code units of the appropriate width. Every
278 PCRE2 function comes in three different forms, one for each library,
279 for example:
280 pcre2_compile_8()
282 pcre2_compile_16()
283 pcre2_compile_32()
284 There are also three different sets of data types:
286 PCRE2_UCHAR8, PCRE2_UCHAR16, PCRE2_UCHAR32
288 PCRE2_SPTR8, PCRE2_SPTR16, PCRE2_SPTR32
289 The UCHAR types define unsigned code units of the appropriate widths.
291 For example, PCRE2_UCHAR16 is usually defined as `uint16_t'. The SPTR
292 types are constant pointers to the equivalent UCHAR types, that is,
293 they are pointers to vectors of unsigned code units.
294 Many applications use only one code unit width. For their convenience,
296 macros are defined whose names are the generic forms such as pcre2_com‐
297 pile() and PCRE2_SPTR. These macros use the value of the macro
298 PCRE2_CODE_UNIT_WIDTH to generate the appropriate width-specific func‐
299 tion and macro names. PCRE2_CODE_UNIT_WIDTH is not defined by default.
300 An application must define it to be 8, 16, or 32 before including
301 pcre2.h in order to make use of the generic names.
302 Applications that use more than one code unit width can be linked with
304 more than one PCRE2 library, but must define PCRE2_CODE_UNIT_WIDTH to
305 be 0 before including pcre2.h, and then use the real function names.
306 Any code that is to be included in an environment where the value of
307 PCRE2_CODE_UNIT_WIDTH is unknown should also use the real function
308 names. (Unfortunately, it is not possible in C code to save and restore
309 the value of a macro.)
310 If PCRE2_CODE_UNIT_WIDTH is not defined before including pcre2.h, a
312 compiler error occurs.
313 When using multiple libraries in an application, you must take care
315 when processing any particular pattern to use only functions from a
316 single library. For example, if you want to run a match using a pat‐
317 tern that was compiled with pcre2_compile_16(), you must do so with
318 pcre2_match_16(), not pcre2_match_8() or pcre2_match_32().
319 In the function summaries above, and in the rest of this document and
321 other PCRE2 documents, functions and data types are described using
322 their generic names, without the _8, _16, or _32 suffix.
323
325 PCRE2 has its own native API, which is described in this document.
327 There are also some wrapper functions for the 8-bit library that corre‐
328 spond to the POSIX regular expression API, but they do not give access
329 to all the functionality of PCRE2. They are described in the pcre2posix
330 documentation. Both these APIs define a set of C function calls.
331 The native API C data types, function prototypes, option values, and
333 error codes are defined in the header file pcre2.h, which also contains
334 definitions of PCRE2_MAJOR and PCRE2_MINOR, the major and minor release
335 numbers for the library. Applications can use these to include support
336 for different releases of PCRE2.
337 In a Windows environment, if you want to statically link an application
339 program against a non-dll PCRE2 library, you must define PCRE2_STATIC
340 before including pcre2.h.
341 The functions pcre2_compile() and pcre2_match() are used for compiling
343 and matching regular expressions in a Perl-compatible manner. A sample
344 program that demonstrates the simplest way of using them is provided in
345 the file called pcre2demo.c in the PCRE2 source distribution. A listing
346 of this program is given in the pcre2demo documentation, and the
347 pcre2sample documentation describes how to compile and run it.
348 The compiling and matching functions recognize various options that are
350 passed as bits in an options argument. There are also some more compli‐
351 cated parameters such as custom memory management functions and
352 resource limits that are passed in "contexts" (which are just memory
353 blocks, described below). Simple applications do not need to make use
354 of contexts.
355 Just-in-time (JIT) compiler support is an optional feature of PCRE2
357 that can be built in appropriate hardware environments. It greatly
358 speeds up the matching performance of many patterns. Programs can
359 request that it be used if available by calling pcre2_jit_compile()
360 after a pattern has been successfully compiled by pcre2_compile(). This
361 does nothing if JIT support is not available.
362 More complicated programs might need to make use of the specialist
364 functions pcre2_jit_stack_create(), pcre2_jit_stack_free(), and
365 pcre2_jit_stack_assign() in order to control the JIT code's memory
366 usage.
367 JIT matching is automatically used by pcre2_match() if it is available,
369 unless the PCRE2_NO_JIT option is set. There is also a direct interface
370 for JIT matching, which gives improved performance at the expense of
371 less sanity checking. The JIT-specific functions are discussed in the
372 pcre2jit documentation.
373 A second matching function, pcre2_dfa_match(), which is not Perl-com‐
375 patible, is also provided. This uses a different algorithm for the
376 matching. The alternative algorithm finds all possible matches (at a
377 given point in the subject), and scans the subject just once (unless
378 there are lookaround assertions). However, this algorithm does not
379 return captured substrings. A description of the two matching algo‐
380 rithms and their advantages and disadvantages is given in the
381 pcre2matching documentation. There is no JIT support for
382 pcre2_dfa_match().
383 In addition to the main compiling and matching functions, there are
385 convenience functions for extracting captured substrings from a subject
386 string that has been matched by pcre2_match(). They are:
387 pcre2_substring_copy_byname()
389 pcre2_substring_copy_bynumber()
390 pcre2_substring_get_byname()
391 pcre2_substring_get_bynumber()
392 pcre2_substring_list_get()
393 pcre2_substring_length_byname()
394 pcre2_substring_length_bynumber()
395 pcre2_substring_nametable_scan()
396 pcre2_substring_number_from_name()
397 pcre2_substring_free() and pcre2_substring_list_free() are also pro‐
399 vided, to free memory used for extracted strings. If either of these
400 functions is called with a NULL argument, the function returns immedi‐
401 ately without doing anything.
402 The function pcre2_substitute() can be called to match a pattern and
404 return a copy of the subject string with substitutions for parts that
405 were matched.
406 Functions whose names begin with pcre2_serialize_ are used for saving
408 compiled patterns on disc or elsewhere, and reloading them later.
409 Finally, there are functions for finding out information about a com‐
411 piled pattern (pcre2_pattern_info()) and about the configuration with
412 which PCRE2 was built (pcre2_config()).
413 Functions with names ending with _free() are used for freeing memory
415 blocks of various sorts. In all cases, if one of these functions is
416 called with a NULL argument, it does nothing.
417
419 The PCRE2 API uses string lengths and offsets into strings of code
421 units in several places. These values are always of type PCRE2_SIZE,
422 which is an unsigned integer type, currently always defined as size_t.
423 The largest value that can be stored in such a type (that is
424 ~(PCRE2_SIZE)0) is reserved as a special indicator for zero-terminated
425 strings and unset offsets. Therefore, the longest string that can be
426 handled is one less than this maximum.
427
429 PCRE2 supports five different conventions for indicating line breaks in
431 strings: a single CR (carriage return) character, a single LF (line‐
432 feed) character, the two-character sequence CRLF, any of the three pre‐
433 ceding, or any Unicode newline sequence. The Unicode newline sequences
434 are the three just mentioned, plus the single characters VT (vertical
435 tab, U+000B), FF (form feed, U+000C), NEL (next line, U+0085), LS (line
436 separator, U+2028), and PS (paragraph separator, U+2029).
437 Each of the first three conventions is used by at least one operating
439 system as its standard newline sequence. When PCRE2 is built, a default
440 can be specified. If it is not, the default is set to LF, which is the
441 Unix standard. However, the newline convention can be changed by an
442 application when calling pcre2_compile(), or it can be specified by
443 special text at the start of the pattern itself; this overrides any
444 other settings. See the pcre2pattern page for details of the special
445 character sequences.
446 In the PCRE2 documentation the word "newline" is used to mean "the
448 character or pair of characters that indicate a line break". The choice
449 of newline convention affects the handling of the dot, circumflex, and
450 dollar metacharacters, the handling of #-comments in /x mode, and, when
451 CRLF is a recognized line ending sequence, the match position advance‐
452 ment for a non-anchored pattern. There is more detail about this in the
453 section on pcre2_match() options below.
454 The choice of newline convention does not affect the interpretation of
456 the \n or \r escape sequences, nor does it affect what \R matches; this
457 has its own separate convention.
458
460 In a multithreaded application it is important to keep thread-specific
462 data separate from data that can be shared between threads. The PCRE2
463 library code itself is thread-safe: it contains no static or global
464 variables. The API is designed to be fairly simple for non-threaded
465 applications while at the same time ensuring that multithreaded appli‐
466 cations can use it.
467 There are several different blocks of data that are used to pass infor‐
469 mation between the application and the PCRE2 libraries.
470 The compiled pattern
472 A pointer to the compiled form of a pattern is returned to the user
474 when pcre2_compile() is successful. The data in the compiled pattern is
475 fixed, and does not change when the pattern is matched. Therefore, it
476 is thread-safe, that is, the same compiled pattern can be used by more
477 than one thread simultaneously. For example, an application can compile
478 all its patterns at the start, before forking off multiple threads that
479 use them. However, if the just-in-time (JIT) optimization feature is
480 being used, it needs separate memory stack areas for each thread. See
481 the pcre2jit documentation for more details.
482 In a more complicated situation, where patterns are compiled only when
484 they are first needed, but are still shared between threads, pointers
485 to compiled patterns must be protected from simultaneous writing by
486 multiple threads. This is somewhat tricky to do correctly. If you know
487 that writing to a pointer is atomic in your environment, you can use
488 logic like this:
489 Get a read-only (shared) lock (mutex) for pointer
491 if (pointer == NULL)
492 {
493 Get a write (unique) lock for pointer
494 if (pointer == NULL) pointer = pcre2_compile(...
495 }
496 Release the lock
497 Use pointer in pcre2_match()
498 Of course, testing for compilation errors should also be included in
500 the code.
501 The reason for checking the pointer a second time is as follows: Sev‐
503 eral threads may have acquired the shared lock and tested the pointer
504 for being NULL, but only one of them will be given the write lock, with
505 the rest kept waiting. The winning thread will compile the pattern and
506 store the result. After this thread releases the write lock, another
507 thread will get it, and if it does not retest pointer for being NULL,
508 will recompile the pattern and overwrite the pointer, creating a memory
509 leak and possibly causing other issues.
510 In an environment where writing to a pointer may not be atomic, the
512 above logic is not sufficient. The thread that is doing the compiling
513 may be descheduled after writing only part of the pointer, which could
514 cause other threads to use an invalid value. Instead of checking the
515 pointer itself, a separate "pointer is valid" flag (that can be updated
516 atomically) must be used:
517 Get a read-only (shared) lock (mutex) for pointer
519 if (!pointer_is_valid)
520 {
521 Get a write (unique) lock for pointer
522 if (!pointer_is_valid)
523 {
524 pointer = pcre2_compile(...
525 pointer_is_valid = TRUE
526 }
527 }
528 Release the lock
529 Use pointer in pcre2_match()
530 If JIT is being used, but the JIT compilation is not being done immedi‐
532 ately (perhaps waiting to see if the pattern is used often enough),
533 similar logic is required. JIT compilation updates a value within the
534 compiled code block, so a thread must gain unique write access to the
535 pointer before calling pcre2_jit_compile(). Alternatively,
536 pcre2_code_copy() or pcre2_code_copy_with_tables() can be used to
537 obtain a private copy of the compiled code before calling the JIT com‐
538 piler.
539 Context blocks
541 The next main section below introduces the idea of "contexts" in which
543 PCRE2 functions are called. A context is nothing more than a collection
544 of parameters that control the way PCRE2 operates. Grouping a number of
545 parameters together in a context is a convenient way of passing them to
546 a PCRE2 function without using lots of arguments. The parameters that
547 are stored in contexts are in some sense "advanced features" of the
548 API. Many straightforward applications will not need to use contexts.
549 In a multithreaded application, if the parameters in a context are val‐
551 ues that are never changed, the same context can be used by all the
552 threads. However, if any thread needs to change any value in a context,
553 it must make its own thread-specific copy.
554 Match blocks
556 The matching functions need a block of memory for storing the results
558 of a match. This includes details of what was matched, as well as addi‐
559 tional information such as the name of a (*MARK) setting. Each thread
560 must provide its own copy of this memory.
561
563 Some PCRE2 functions have a lot of parameters, many of which are used
565 only by specialist applications, for example, those that use custom
566 memory management or non-standard character tables. To keep function
567 argument lists at a reasonable size, and at the same time to keep the
568 API extensible, "uncommon" parameters are passed to certain functions
569 in a context instead of directly. A context is just a block of memory
570 that holds the parameter values. Applications that do not need to
571 adjust any of the context parameters can pass NULL when a context
572 pointer is required.
573 There are three different types of context: a general context that is
575 relevant for several PCRE2 operations, a compile-time context, and a
576 match-time context.
577 The general context
579 At present, this context just contains pointers to (and data for)
581 external memory management functions that are called from several
582 places in the PCRE2 library. The context is named `general' rather than
583 specifically `memory' because in future other fields may be added. If
584 you do not want to supply your own custom memory management functions,
585 you do not need to bother with a general context. A general context is
586 created by:
587 pcre2_general_context *pcre2_general_context_create(
589 void *(*private_malloc)(PCRE2_SIZE, void *),
590 void (*private_free)(void *, void *), void *memory_data);
591 The two function pointers specify custom memory management functions,
593 whose prototypes are:
594 void *private_malloc(PCRE2_SIZE, void *);
596 void private_free(void *, void *);
597 Whenever code in PCRE2 calls these functions, the final argument is the
599 value of memory_data. Either of the first two arguments of the creation
600 function may be NULL, in which case the system memory management func‐
601 tions malloc() and free() are used. (This is not currently useful, as
602 there are no other fields in a general context, but in future there
603 might be.) The private_malloc() function is used (if supplied) to
604 obtain memory for storing the context, and all three values are saved
605 as part of the context.
606 Whenever PCRE2 creates a data block of any kind, the block contains a
608 pointer to the free() function that matches the malloc() function that
609 was used. When the time comes to free the block, this function is
610 called.
611 A general context can be copied by calling:
613 pcre2_general_context *pcre2_general_context_copy(
615 pcre2_general_context *gcontext);
616 The memory used for a general context should be freed by calling:
618 void pcre2_general_context_free(pcre2_general_context *gcontext);
620 If this function is passed a NULL argument, it returns immediately
622 without doing anything.
623 The compile context
625 A compile context is required if you want to provide an external func‐
627 tion for stack checking during compilation or to change the default
628 values of any of the following compile-time parameters:
629 What \R matches (Unicode newlines or CR, LF, CRLF only)
631 PCRE2's character tables
632 The newline character sequence
633 The compile time nested parentheses limit
634 The maximum length of the pattern string
635 The extra options bits (none set by default)
636 A compile context is also required if you are using custom memory man‐
638 agement. If none of these apply, just pass NULL as the context argu‐
639 ment of pcre2_compile().
640 A compile context is created, copied, and freed by the following func‐
642 tions:
643 pcre2_compile_context *pcre2_compile_context_create(
645 pcre2_general_context *gcontext);
646 pcre2_compile_context *pcre2_compile_context_copy(
648 pcre2_compile_context *ccontext);
649 void pcre2_compile_context_free(pcre2_compile_context *ccontext);
651 A compile context is created with default values for its parameters.
653 These can be changed by calling the following functions, which return 0
654 on success, or PCRE2_ERROR_BADDATA if invalid data is detected.
655 int pcre2_set_bsr(pcre2_compile_context *ccontext,
657 uint32_t value);
658 The value must be PCRE2_BSR_ANYCRLF, to specify that \R matches only
660 CR, LF, or CRLF, or PCRE2_BSR_UNICODE, to specify that \R matches any
661 Unicode line ending sequence. The value is used by the JIT compiler and
662 by the two interpreted matching functions, pcre2_match() and
663 pcre2_dfa_match().
664 int pcre2_set_character_tables(pcre2_compile_context *ccontext,
666 const uint8_t *tables);
667 The value must be the result of a call to pcre2_maketables(), whose
669 only argument is a general context. This function builds a set of char‐
670 acter tables in the current locale.
671 int pcre2_set_compile_extra_options(pcre2_compile_context *ccontext,
673 uint32_t extra_options);
674 As PCRE2 has developed, almost all the 32 option bits that are avail‐
676 able in the options argument of pcre2_compile() have been used up. To
677 avoid running out, the compile context contains a set of extra option
678 bits which are used for some newer, assumed rarer, options. This func‐
679 tion sets those bits. It always sets all the bits (either on or off).
680 It does not modify any existing setting. The available options are
681 defined in the section entitled "Extra compile options" below.
682 int pcre2_set_max_pattern_length(pcre2_compile_context *ccontext,
684 PCRE2_SIZE value);
685 This sets a maximum length, in code units, for any pattern string that
687 is compiled with this context. If the pattern is longer, an error is
688 generated. This facility is provided so that applications that accept
689 patterns from external sources can limit their size. The default is the
690 largest number that a PCRE2_SIZE variable can hold, which is effec‐
691 tively unlimited.
692 int pcre2_set_newline(pcre2_compile_context *ccontext,
694 uint32_t value);
695 This specifies which characters or character sequences are to be recog‐
697 nized as newlines. The value must be one of PCRE2_NEWLINE_CR (carriage
698 return only), PCRE2_NEWLINE_LF (linefeed only), PCRE2_NEWLINE_CRLF (the
699 two-character sequence CR followed by LF), PCRE2_NEWLINE_ANYCRLF (any
700 of the above), PCRE2_NEWLINE_ANY (any Unicode newline sequence), or
701 PCRE2_NEWLINE_NUL (the NUL character, that is a binary zero).
702 A pattern can override the value set in the compile context by starting
704 with a sequence such as (*CRLF). See the pcre2pattern page for details.
705 When a pattern is compiled with the PCRE2_EXTENDED or
707 PCRE2_EXTENDED_MORE option, the newline convention affects the recogni‐
708 tion of the end of internal comments starting with #. The value is
709 saved with the compiled pattern for subsequent use by the JIT compiler
710 and by the two interpreted matching functions, pcre2_match() and
711 pcre2_dfa_match().
712 int pcre2_set_parens_nest_limit(pcre2_compile_context *ccontext,
714 uint32_t value);
715 This parameter adjusts the limit, set when PCRE2 is built (default
717 250), on the depth of parenthesis nesting in a pattern. This limit
718 stops rogue patterns using up too much system stack when being com‐
719 piled. The limit applies to parentheses of all kinds, not just captur‐
720 ing parentheses.
721 int pcre2_set_compile_recursion_guard(pcre2_compile_context *ccontext,
723 int (*guard_function)(uint32_t, void *), void *user_data);
724 There is at least one application that runs PCRE2 in threads with very
726 limited system stack, where running out of stack is to be avoided at
727 all costs. The parenthesis limit above cannot take account of how much
728 stack is actually available during compilation. For a finer control,
729 you can supply a function that is called whenever pcre2_compile()
730 starts to compile a parenthesized part of a pattern. This function can
731 check the actual stack size (or anything else that it wants to, of
732 course).
733 The first argument to the callout function gives the current depth of
735 nesting, and the second is user data that is set up by the last argu‐
736 ment of pcre2_set_compile_recursion_guard(). The callout function
737 should return zero if all is well, or non-zero to force an error.
738 The match context
740 A match context is required if you want to:
742 Set up a callout function
744 Set an offset limit for matching an unanchored pattern
745 Change the limit on the amount of heap used when matching
746 Change the backtracking match limit
747 Change the backtracking depth limit
748 Set custom memory management specifically for the match
749 If none of these apply, just pass NULL as the context argument of
751 pcre2_match(), pcre2_dfa_match(), or pcre2_jit_match().
752 A match context is created, copied, and freed by the following func‐
754 tions:
755 pcre2_match_context *pcre2_match_context_create(
757 pcre2_general_context *gcontext);
758 pcre2_match_context *pcre2_match_context_copy(
760 pcre2_match_context *mcontext);
761 void pcre2_match_context_free(pcre2_match_context *mcontext);
763 A match context is created with default values for its parameters.
765 These can be changed by calling the following functions, which return 0
766 on success, or PCRE2_ERROR_BADDATA if invalid data is detected.
767 int pcre2_set_callout(pcre2_match_context *mcontext,
769 int (*callout_function)(pcre2_callout_block *, void *),
770 void *callout_data);
771 This sets up a callout function for PCRE2 to call at specified points
773 during a matching operation. Details are given in the pcre2callout doc‐
774 umentation.
775 int pcre2_set_substitute_callout(pcre2_match_context *mcontext,
777 int (*callout_function)(pcre2_substitute_callout_block *, void *),
778 void *callout_data);
779 This sets up a callout function for PCRE2 to call after each substitu‐
781 tion made by pcre2_substitute(). Details are given in the section enti‐
782 tled "Creating a new string with substitutions" below.
783 int pcre2_set_offset_limit(pcre2_match_context *mcontext,
785 PCRE2_SIZE value);
786 The offset_limit parameter limits how far an unanchored search can
788 advance in the subject string. The default value is PCRE2_UNSET. The
789 pcre2_match() and pcre2_dfa_match() functions return
790 PCRE2_ERROR_NOMATCH if a match with a starting point before or at the
791 given offset is not found. The pcre2_substitute() function makes no
792 more substitutions.
793 For example, if the pattern /abc/ is matched against "123abc" with an
795 offset limit less than 3, the result is PCRE2_ERROR_NOMATCH. A match
796 can never be found if the startoffset argument of pcre2_match(),
797 pcre2_dfa_match(), or pcre2_substitute() is greater than the offset
798 limit set in the match context.
799 When using this facility, you must set the PCRE2_USE_OFFSET_LIMIT
801 option when calling pcre2_compile() so that when JIT is in use, differ‐
802 ent code can be compiled. If a match is started with a non-default
803 match limit when PCRE2_USE_OFFSET_LIMIT is not set, an error is gener‐
804 ated.
805 The offset limit facility can be used to track progress when searching
807 large subject strings or to limit the extent of global substitutions.
808 See also the PCRE2_FIRSTLINE option, which requires a match to start
809 before or at the first newline that follows the start of matching in
810 the subject. If this is set with an offset limit, a match must occur in
811 the first line and also within the offset limit. In other words, which‐
812 ever limit comes first is used.
813 int pcre2_set_heap_limit(pcre2_match_context *mcontext,
815 uint32_t value);
816 The heap_limit parameter specifies, in units of kibibytes (1024 bytes),
818 the maximum amount of heap memory that pcre2_match() may use to hold
819 backtracking information when running an interpretive match. This limit
820 also applies to pcre2_dfa_match(), which may use the heap when process‐
821 ing patterns with a lot of nested pattern recursion or lookarounds or
822 atomic groups. This limit does not apply to matching with the JIT opti‐
823 mization, which has its own memory control arrangements (see the
824 pcre2jit documentation for more details). If the limit is reached, the
825 negative error code PCRE2_ERROR_HEAPLIMIT is returned. The default
826 limit can be set when PCRE2 is built; if it is not, the default is set
827 very large and is essentially "unlimited".
828 A value for the heap limit may also be supplied by an item at the start
830 of a pattern of the form
831 (*LIMIT_HEAP=ddd)
833 where ddd is a decimal number. However, such a setting is ignored
835 unless ddd is less than the limit set by the caller of pcre2_match()
836 or, if no such limit is set, less than the default.
837 The pcre2_match() function starts out using a 20KiB vector on the sys‐
839 tem stack for recording backtracking points. The more nested backtrack‐
840 ing points there are (that is, the deeper the search tree), the more
841 memory is needed. Heap memory is used only if the initial vector is
842 too small. If the heap limit is set to a value less than 21 (in partic‐
843 ular, zero) no heap memory will be used. In this case, only patterns
844 that do not have a lot of nested backtracking can be successfully pro‐
845 cessed.
846 Similarly, for pcre2_dfa_match(), a vector on the system stack is used
848 when processing pattern recursions, lookarounds, or atomic groups, and
849 only if this is not big enough is heap memory used. In this case, too,
850 setting a value of zero disables the use of the heap.
851 int pcre2_set_match_limit(pcre2_match_context *mcontext,
853 uint32_t value);
854 The match_limit parameter provides a means of preventing PCRE2 from
856 using up too many computing resources when processing patterns that are
857 not going to match, but which have a very large number of possibilities
858 in their search trees. The classic example is a pattern that uses
859 nested unlimited repeats.
860 There is an internal counter in pcre2_match() that is incremented each
862 time round its main matching loop. If this value reaches the match
863 limit, pcre2_match() returns the negative value PCRE2_ERROR_MATCHLIMIT.
864 This has the effect of limiting the amount of backtracking that can
865 take place. For patterns that are not anchored, the count restarts from
866 zero for each position in the subject string. This limit also applies
867 to pcre2_dfa_match(), though the counting is done in a different way.
868 When pcre2_match() is called with a pattern that was successfully pro‐
870 cessed by pcre2_jit_compile(), the way in which matching is executed is
871 entirely different. However, there is still the possibility of runaway
872 matching that goes on for a very long time, and so the match_limit
873 value is also used in this case (but in a different way) to limit how
874 long the matching can continue.
875 The default value for the limit can be set when PCRE2 is built; the
877 default default is 10 million, which handles all but the most extreme
878 cases. A value for the match limit may also be supplied by an item at
879 the start of a pattern of the form
880 (*LIMIT_MATCH=ddd)
882 where ddd is a decimal number. However, such a setting is ignored
884 unless ddd is less than the limit set by the caller of pcre2_match() or
885 pcre2_dfa_match() or, if no such limit is set, less than the default.
886 int pcre2_set_depth_limit(pcre2_match_context *mcontext,
888 uint32_t value);
889 This parameter limits the depth of nested backtracking in
891 pcre2_match(). Each time a nested backtracking point is passed, a new
892 memory "frame" is used to remember the state of matching at that point.
893 Thus, this parameter indirectly limits the amount of memory that is
894 used in a match. However, because the size of each memory "frame"
895 depends on the number of capturing parentheses, the actual memory limit
896 varies from pattern to pattern. This limit was more useful in versions
897 before 10.30, where function recursion was used for backtracking.
898 The depth limit is not relevant, and is ignored, when matching is done
900 using JIT compiled code. However, it is supported by pcre2_dfa_match(),
901 which uses it to limit the depth of nested internal recursive function
902 calls that implement atomic groups, lookaround assertions, and pattern
903 recursions. This limits, indirectly, the amount of system stack that is
904 used. It was more useful in versions before 10.32, when stack memory
905 was used for local workspace vectors for recursive function calls. From
906 version 10.32, only local variables are allocated on the stack and as
907 each call uses only a few hundred bytes, even a small stack can support
908 quite a lot of recursion.
909 If the depth of internal recursive function calls is great enough,
911 local workspace vectors are allocated on the heap from version 10.32
912 onwards, so the depth limit also indirectly limits the amount of heap
913 memory that is used. A recursive pattern such as /(.(?2))((?1)|)/, when
914 matched to a very long string using pcre2_dfa_match(), can use a great
915 deal of memory. However, it is probably better to limit heap usage
916 directly by calling pcre2_set_heap_limit().
917 The default value for the depth limit can be set when PCRE2 is built;
919 if it is not, the default is set to the same value as the default for
920 the match limit. If the limit is exceeded, pcre2_match() or
921 pcre2_dfa_match() returns PCRE2_ERROR_DEPTHLIMIT. A value for the depth
922 limit may also be supplied by an item at the start of a pattern of the
923 form
924 (*LIMIT_DEPTH=ddd)
926 where ddd is a decimal number. However, such a setting is ignored
928 unless ddd is less than the limit set by the caller of pcre2_match() or
929 pcre2_dfa_match() or, if no such limit is set, less than the default.
930
932 int pcre2_config(uint32_t what, void *where);
934 The function pcre2_config() makes it possible for a PCRE2 client to
936 find the value of certain configuration parameters and to discover
937 which optional features have been compiled into the PCRE2 library. The
938 pcre2build documentation has more details about these features.
939 The first argument for pcre2_config() specifies which information is
941 required. The second argument is a pointer to memory into which the
942 information is placed. If NULL is passed, the function returns the
943 amount of memory that is needed for the requested information. For
944 calls that return numerical values, the value is in bytes; when
945 requesting these values, where should point to appropriately aligned
946 memory. For calls that return strings, the required length is given in
947 code units, not counting the terminating zero.
948 When requesting information, the returned value from pcre2_config() is
950 non-negative on success, or the negative error code PCRE2_ERROR_BADOP‐
951 TION if the value in the first argument is not recognized. The follow‐
952 ing information is available:
953 PCRE2_CONFIG_BSR
955 The output is a uint32_t integer whose value indicates what character
957 sequences the \R escape sequence matches by default. A value of
958 PCRE2_BSR_UNICODE means that \R matches any Unicode line ending
959 sequence; a value of PCRE2_BSR_ANYCRLF means that \R matches only CR,
960 LF, or CRLF. The default can be overridden when a pattern is compiled.
961 PCRE2_CONFIG_COMPILED_WIDTHS
963 The output is a uint32_t integer whose lower bits indicate which code
965 unit widths were selected when PCRE2 was built. The 1-bit indicates
966 8-bit support, and the 2-bit and 4-bit indicate 16-bit and 32-bit sup‐
967 port, respectively.
968 PCRE2_CONFIG_DEPTHLIMIT
970 The output is a uint32_t integer that gives the default limit for the
972 depth of nested backtracking in pcre2_match() or the depth of nested
973 recursions, lookarounds, and atomic groups in pcre2_dfa_match(). Fur‐
974 ther details are given with pcre2_set_depth_limit() above.
975 PCRE2_CONFIG_HEAPLIMIT
977 The output is a uint32_t integer that gives, in kibibytes, the default
979 limit for the amount of heap memory used by pcre2_match() or
980 pcre2_dfa_match(). Further details are given with
981 pcre2_set_heap_limit() above.
982 PCRE2_CONFIG_JIT
984 The output is a uint32_t integer that is set to one if support for
986 just-in-time compiling is available; otherwise it is set to zero.
987 PCRE2_CONFIG_JITTARGET
989 The where argument should point to a buffer that is at least 48 code
991 units long. (The exact length required can be found by calling
992 pcre2_config() with where set to NULL.) The buffer is filled with a
993 string that contains the name of the architecture for which the JIT
994 compiler is configured, for example "x86 32bit (little endian +
995 unaligned)". If JIT support is not available, PCRE2_ERROR_BADOPTION is
996 returned, otherwise the number of code units used is returned. This is
997 the length of the string, plus one unit for the terminating zero.
998 PCRE2_CONFIG_LINKSIZE
1000 The output is a uint32_t integer that contains the number of bytes used
1002 for internal linkage in compiled regular expressions. When PCRE2 is
1003 configured, the value can be set to 2, 3, or 4, with the default being
1004 2. This is the value that is returned by pcre2_config(). However, when
1005 the 16-bit library is compiled, a value of 3 is rounded up to 4, and
1006 when the 32-bit library is compiled, internal linkages always use 4
1007 bytes, so the configured value is not relevant.
1008 The default value of 2 for the 8-bit and 16-bit libraries is sufficient
1010 for all but the most massive patterns, since it allows the size of the
1011 compiled pattern to be up to 65535 code units. Larger values allow
1012 larger regular expressions to be compiled by those two libraries, but
1013 at the expense of slower matching.
1014 PCRE2_CONFIG_MATCHLIMIT
1016 The output is a uint32_t integer that gives the default match limit for
1018 pcre2_match(). Further details are given with pcre2_set_match_limit()
1019 above.
1020 PCRE2_CONFIG_NEWLINE
1022 The output is a uint32_t integer whose value specifies the default
1024 character sequence that is recognized as meaning "newline". The values
1025 are:
1026 PCRE2_NEWLINE_CR Carriage return (CR)
1028 PCRE2_NEWLINE_LF Linefeed (LF)
1029 PCRE2_NEWLINE_CRLF Carriage return, linefeed (CRLF)
1030 PCRE2_NEWLINE_ANY Any Unicode line ending
1031 PCRE2_NEWLINE_ANYCRLF Any of CR, LF, or CRLF
1032 PCRE2_NEWLINE_NUL The NUL character (binary zero)
1033 The default should normally correspond to the standard sequence for
1035 your operating system.
1036 PCRE2_CONFIG_NEVER_BACKSLASH_C
1038 The output is a uint32_t integer that is set to one if the use of \C
1040 was permanently disabled when PCRE2 was built; otherwise it is set to
1041 zero.
1042 PCRE2_CONFIG_PARENSLIMIT
1044 The output is a uint32_t integer that gives the maximum depth of nest‐
1046 ing of parentheses (of any kind) in a pattern. This limit is imposed to
1047 cap the amount of system stack used when a pattern is compiled. It is
1048 specified when PCRE2 is built; the default is 250. This limit does not
1049 take into account the stack that may already be used by the calling
1050 application. For finer control over compilation stack usage, see
1051 pcre2_set_compile_recursion_guard().
1052 PCRE2_CONFIG_STACKRECURSE
1054 This parameter is obsolete and should not be used in new code. The out‐
1056 put is a uint32_t integer that is always set to zero.
1057 PCRE2_CONFIG_TABLES_LENGTH
1059 The output is a uint32_t integer that gives the length of PCRE2's char‐
1061 acter processing tables in bytes. For details of these tables see the
1062 section on locale support below.
1063 PCRE2_CONFIG_UNICODE_VERSION
1065 The where argument should point to a buffer that is at least 24 code
1067 units long. (The exact length required can be found by calling
1068 pcre2_config() with where set to NULL.) If PCRE2 has been compiled
1069 without Unicode support, the buffer is filled with the text "Unicode
1070 not supported". Otherwise, the Unicode version string (for example,
1071 "8.0.0") is inserted. The number of code units used is returned. This
1072 is the length of the string plus one unit for the terminating zero.
1073 PCRE2_CONFIG_UNICODE
1075 The output is a uint32_t integer that is set to one if Unicode support
1077 is available; otherwise it is set to zero. Unicode support implies UTF
1078 support.
1079 PCRE2_CONFIG_VERSION
1081 The where argument should point to a buffer that is at least 24 code
1083 units long. (The exact length required can be found by calling
1084 pcre2_config() with where set to NULL.) The buffer is filled with the
1085 PCRE2 version string, zero-terminated. The number of code units used is
1086 returned. This is the length of the string plus one unit for the termi‐
1087 nating zero.
1088
1090 pcre2_code *pcre2_compile(PCRE2_SPTR pattern, PCRE2_SIZE length,
1092 uint32_t options, int *errorcode, PCRE2_SIZE *erroroffset,
1093 pcre2_compile_context *ccontext);
1094 void pcre2_code_free(pcre2_code *code);
1096 pcre2_code *pcre2_code_copy(const pcre2_code *code);
1098 pcre2_code *pcre2_code_copy_with_tables(const pcre2_code *code);
1100 The pcre2_compile() function compiles a pattern into an internal form.
1102 The pattern is defined by a pointer to a string of code units and a
1103 length (in code units). If the pattern is zero-terminated, the length
1104 can be specified as PCRE2_ZERO_TERMINATED. The function returns a
1105 pointer to a block of memory that contains the compiled pattern and
1106 related data, or NULL if an error occurred.
1107 If the compile context argument ccontext is NULL, memory for the com‐
1109 piled pattern is obtained by calling malloc(). Otherwise, it is
1110 obtained from the same memory function that was used for the compile
1111 context. The caller must free the memory by calling pcre2_code_free()
1112 when it is no longer needed. If pcre2_code_free() is called with a
1113 NULL argument, it returns immediately, without doing anything.
1114 The function pcre2_code_copy() makes a copy of the compiled code in new
1116 memory, using the same memory allocator as was used for the original.
1117 However, if the code has been processed by the JIT compiler (see
1118 below), the JIT information cannot be copied (because it is position-
1119 dependent). The new copy can initially be used only for non-JIT match‐
1120 ing, though it can be passed to pcre2_jit_compile() if required. If
1121 pcre2_code_copy() is called with a NULL argument, it returns NULL.
1122 The pcre2_code_copy() function provides a way for individual threads in
1124 a multithreaded application to acquire a private copy of shared com‐
1125 piled code. However, it does not make a copy of the character tables
1126 used by the compiled pattern; the new pattern code points to the same
1127 tables as the original code. (See "Locale Support" below for details
1128 of these character tables.) In many applications the same tables are
1129 used throughout, so this behaviour is appropriate. Nevertheless, there
1130 are occasions when a copy of a compiled pattern and the relevant tables
1131 are needed. The pcre2_code_copy_with_tables() provides this facility.
1132 Copies of both the code and the tables are made, with the new code
1133 pointing to the new tables. The memory for the new tables is automati‐
1134 cally freed when pcre2_code_free() is called for the new copy of the
1135 compiled code. If pcre2_code_copy_with_tables() is called with a NULL
1136 argument, it returns NULL.
1137 NOTE: When one of the matching functions is called, pointers to the
1139 compiled pattern and the subject string are set in the match data block
1140 so that they can be referenced by the substring extraction functions
1141 after a successful match. After running a match, you must not free a
1142 compiled pattern or a subject string until after all operations on the
1143 match data block have taken place, unless, in the case of the subject
1144 string, you have used the PCRE2_COPY_MATCHED_SUBJECT option, which is
1145 described in the section entitled "Option bits for pcre2_match()"
1146 below.
1147 The options argument for pcre2_compile() contains various bit settings
1149 that affect the compilation. It should be zero if none of them are
1150 required. The available options are described below. Some of them (in
1151 particular, those that are compatible with Perl, but some others as
1152 well) can also be set and unset from within the pattern (see the
1153 detailed description in the pcre2pattern documentation).
1154 For those options that can be different in different parts of the pat‐
1156 tern, the contents of the options argument specifies their settings at
1157 the start of compilation. The PCRE2_ANCHORED, PCRE2_ENDANCHORED, and
1158 PCRE2_NO_UTF_CHECK options can be set at the time of matching as well
1159 as at compile time.
1160 Some additional options and less frequently required compile-time
1162 parameters (for example, the newline setting) can be provided in a com‐
1163 pile context (as described above).
1164 If errorcode or erroroffset is NULL, pcre2_compile() returns NULL imme‐
1166 diately. Otherwise, the variables to which these point are set to an
1167 error code and an offset (number of code units) within the pattern,
1168 respectively, when pcre2_compile() returns NULL because a compilation
1169 error has occurred. The values are not defined when compilation is suc‐
1170 cessful and pcre2_compile() returns a non-NULL value.
1171 There are nearly 100 positive error codes that pcre2_compile() may
1173 return if it finds an error in the pattern. There are also some nega‐
1174 tive error codes that are used for invalid UTF strings when validity
1175 checking is in force. These are the same as given by pcre2_match() and
1176 pcre2_dfa_match(), and are described in the pcre2unicode documentation.
1177 There is no separate documentation for the positive error codes,
1178 because the textual error messages that are obtained by calling the
1179 pcre2_get_error_message() function (see "Obtaining a textual error mes‐
1180 sage" below) should be self-explanatory. Macro names starting with
1181 PCRE2_ERROR_ are defined for both positive and negative error codes in
1182 pcre2.h.
1183 The value returned in erroroffset is an indication of where in the pat‐
1185 tern the error occurred. It is not necessarily the furthest point in
1186 the pattern that was read. For example, after the error "lookbehind
1187 assertion is not fixed length", the error offset points to the start of
1188 the failing assertion. For an invalid UTF-8 or UTF-16 string, the off‐
1189 set is that of the first code unit of the failing character.
1190 Some errors are not detected until the whole pattern has been scanned;
1192 in these cases, the offset passed back is the length of the pattern.
1193 Note that the offset is in code units, not characters, even in a UTF
1194 mode. It may sometimes point into the middle of a UTF-8 or UTF-16 char‐
1195 acter.
1196 This code fragment shows a typical straightforward call to pcre2_com‐
1198 pile():
1199 pcre2_code *re;
1201 PCRE2_SIZE erroffset;
1202 int errorcode;
1203 re = pcre2_compile(
1204 "^A.*Z", /* the pattern */
1205 PCRE2_ZERO_TERMINATED, /* the pattern is zero-terminated */
1206 0, /* default options */
1207 &errorcode, /* for error code */
1208 &erroffset, /* for error offset */
1209 NULL); /* no compile context */
1210 Main compile options
1213 The following names for option bits are defined in the pcre2.h header
1215 file:
1216 PCRE2_ANCHORED
1218 If this bit is set, the pattern is forced to be "anchored", that is, it
1220 is constrained to match only at the first matching point in the string
1221 that is being searched (the "subject string"). This effect can also be
1222 achieved by appropriate constructs in the pattern itself, which is the
1223 only way to do it in Perl.
1224 PCRE2_ALLOW_EMPTY_CLASS
1226 By default, for compatibility with Perl, a closing square bracket that
1228 immediately follows an opening one is treated as a data character for
1229 the class. When PCRE2_ALLOW_EMPTY_CLASS is set, it terminates the
1230 class, which therefore contains no characters and so can never match.
1231 PCRE2_ALT_BSUX
1233 This option request alternative handling of three escape sequences,
1235 which makes PCRE2's behaviour more like ECMAscript (aka JavaScript).
1236 When it is set:
1237 (1) \U matches an upper case "U" character; by default \U causes a com‐
1239 pile time error (Perl uses \U to upper case subsequent characters).
1240 (2) \u matches a lower case "u" character unless it is followed by four
1242 hexadecimal digits, in which case the hexadecimal number defines the
1243 code point to match. By default, \u causes a compile time error (Perl
1244 uses it to upper case the following character).
1245 (3) \x matches a lower case "x" character unless it is followed by two
1247 hexadecimal digits, in which case the hexadecimal number defines the
1248 code point to match. By default, as in Perl, a hexadecimal number is
1249 always expected after \x, but it may have zero, one, or two digits (so,
1250 for example, \xz matches a binary zero character followed by z).
1251 ECMAscript 6 added additional functionality to \u. This can be accessed
1253 using the PCRE2_EXTRA_ALT_BSUX extra option (see "Extra compile
1254 options" below). Note that this alternative escape handling applies
1255 only to patterns. Neither of these options affects the processing of
1256 replacement strings passed to pcre2_substitute().
1257 PCRE2_ALT_CIRCUMFLEX
1259 In multiline mode (when PCRE2_MULTILINE is set), the circumflex
1261 metacharacter matches at the start of the subject (unless PCRE2_NOTBOL
1262 is set), and also after any internal newline. However, it does not
1263 match after a newline at the end of the subject, for compatibility with
1264 Perl. If you want a multiline circumflex also to match after a termi‐
1265 nating newline, you must set PCRE2_ALT_CIRCUMFLEX.
1266 PCRE2_ALT_VERBNAMES
1268 By default, for compatibility with Perl, the name in any verb sequence
1270 such as (*MARK:NAME) is any sequence of characters that does not
1271 include a closing parenthesis. The name is not processed in any way,
1272 and it is not possible to include a closing parenthesis in the name.
1273 However, if the PCRE2_ALT_VERBNAMES option is set, normal backslash
1274 processing is applied to verb names and only an unescaped closing
1275 parenthesis terminates the name. A closing parenthesis can be included
1276 in a name either as \) or between \Q and \E. If the PCRE2_EXTENDED or
1277 PCRE2_EXTENDED_MORE option is set with PCRE2_ALT_VERBNAMES, unescaped
1278 whitespace in verb names is skipped and #-comments are recognized,
1279 exactly as in the rest of the pattern.
1280 PCRE2_AUTO_CALLOUT
1282 If this bit is set, pcre2_compile() automatically inserts callout
1284 items, all with number 255, before each pattern item, except immedi‐
1285 ately before or after an explicit callout in the pattern. For discus‐
1286 sion of the callout facility, see the pcre2callout documentation.
1287 PCRE2_CASELESS
1289 If this bit is set, letters in the pattern match both upper and lower
1291 case letters in the subject. It is equivalent to Perl's /i option, and
1292 it can be changed within a pattern by a (?i) option setting. If either
1293 PCRE2_UTF or PCRE2_UCP is set, Unicode properties are used for all
1294 characters with more than one other case, and for all characters whose
1295 code points are greater than U+007F. Note that there are two ASCII
1296 characters, K and S, that, in addition to their lower case ASCII equiv‐
1297 alents, are case-equivalent with U+212A (Kelvin sign) and U+017F (long
1298 S) respectively. For lower valued characters with only one other case,
1299 a lookup table is used for speed. When neither PCRE2_UTF nor PCRE2_UCP
1300 is set, a lookup table is used for all code points less than 256, and
1301 higher code points (available only in 16-bit or 32-bit mode) are
1302 treated as not having another case.
1303 PCRE2_DOLLAR_ENDONLY
1305 If this bit is set, a dollar metacharacter in the pattern matches only
1307 at the end of the subject string. Without this option, a dollar also
1308 matches immediately before a newline at the end of the string (but not
1309 before any other newlines). The PCRE2_DOLLAR_ENDONLY option is ignored
1310 if PCRE2_MULTILINE is set. There is no equivalent to this option in
1311 Perl, and no way to set it within a pattern.
1312 PCRE2_DOTALL
1314 If this bit is set, a dot metacharacter in the pattern matches any
1316 character, including one that indicates a newline. However, it only
1317 ever matches one character, even if newlines are coded as CRLF. Without
1318 this option, a dot does not match when the current position in the sub‐
1319 ject is at a newline. This option is equivalent to Perl's /s option,
1320 and it can be changed within a pattern by a (?s) option setting. A neg‐
1321 ative class such as [^a] always matches newline characters, and the \N
1322 escape sequence always matches a non-newline character, independent of
1323 the setting of PCRE2_DOTALL.
1324 PCRE2_DUPNAMES
1326 If this bit is set, names used to identify capture groups need not be
1328 unique. This can be helpful for certain types of pattern when it is
1329 known that only one instance of the named group can ever be matched.
1330 There are more details of named capture groups below; see also the
1331 pcre2pattern documentation.
1332 PCRE2_ENDANCHORED
1334 If this bit is set, the end of any pattern match must be right at the
1336 end of the string being searched (the "subject string"). If the pattern
1337 match succeeds by reaching (*ACCEPT), but does not reach the end of the
1338 subject, the match fails at the current starting point. For unanchored
1339 patterns, a new match is then tried at the next starting point. How‐
1340 ever, if the match succeeds by reaching the end of the pattern, but not
1341 the end of the subject, backtracking occurs and an alternative match
1342 may be found. Consider these two patterns:
1343 .(*ACCEPT)|..
1345 .|..
1346 If matched against "abc" with PCRE2_ENDANCHORED set, the first matches
1348 "c" whereas the second matches "bc". The effect of PCRE2_ENDANCHORED
1349 can also be achieved by appropriate constructs in the pattern itself,
1350 which is the only way to do it in Perl.
1351 For DFA matching with pcre2_dfa_match(), PCRE2_ENDANCHORED applies only
1353 to the first (that is, the longest) matched string. Other parallel
1354 matches, which are necessarily substrings of the first one, must obvi‐
1355 ously end before the end of the subject.
1356 PCRE2_EXTENDED
1358 If this bit is set, most white space characters in the pattern are
1360 totally ignored except when escaped or inside a character class. How‐
1361 ever, white space is not allowed within sequences such as (?> that
1362 introduce various parenthesized groups, nor within numerical quanti‐
1363 fiers such as {1,3}. Ignorable white space is permitted between an item
1364 and a following quantifier and between a quantifier and a following +
1365 that indicates possessiveness. PCRE2_EXTENDED is equivalent to Perl's
1366 /x option, and it can be changed within a pattern by a (?x) option set‐
1367 ting.
1368 When PCRE2 is compiled without Unicode support, PCRE2_EXTENDED recog‐
1370 nizes as white space only those characters with code points less than
1371 256 that are flagged as white space in its low-character table. The ta‐
1372 ble is normally created by pcre2_maketables(), which uses the isspace()
1373 function to identify space characters. In most ASCII environments, the
1374 relevant characters are those with code points 0x0009 (tab), 0x000A
1375 (linefeed), 0x000B (vertical tab), 0x000C (formfeed), 0x000D (carriage
1376 return), and 0x0020 (space).
1377 When PCRE2 is compiled with Unicode support, in addition to these char‐
1379 acters, five more Unicode "Pattern White Space" characters are recog‐
1380 nized by PCRE2_EXTENDED. These are U+0085 (next line), U+200E (left-to-
1381 right mark), U+200F (right-to-left mark), U+2028 (line separator), and
1382 U+2029 (paragraph separator). This set of characters is the same as
1383 recognized by Perl's /x option. Note that the horizontal and vertical
1384 space characters that are matched by the \h and \v escapes in patterns
1385 are a much bigger set.
1386 As well as ignoring most white space, PCRE2_EXTENDED also causes char‐
1388 acters between an unescaped # outside a character class and the next
1389 newline, inclusive, to be ignored, which makes it possible to include
1390 comments inside complicated patterns. Note that the end of this type of
1391 comment is a literal newline sequence in the pattern; escape sequences
1392 that happen to represent a newline do not count.
1393 Which characters are interpreted as newlines can be specified by a set‐
1395 ting in the compile context that is passed to pcre2_compile() or by a
1396 special sequence at the start of the pattern, as described in the sec‐
1397 tion entitled "Newline conventions" in the pcre2pattern documentation.
1398 A default is defined when PCRE2 is built.
1399 PCRE2_EXTENDED_MORE
1401 This option has the effect of PCRE2_EXTENDED, but, in addition,
1403 unescaped space and horizontal tab characters are ignored inside a
1404 character class. Note: only these two characters are ignored, not the
1405 full set of pattern white space characters that are ignored outside a
1406 character class. PCRE2_EXTENDED_MORE is equivalent to Perl's /xx
1407 option, and it can be changed within a pattern by a (?xx) option set‐
1408 ting.
1409 PCRE2_FIRSTLINE
1411 If this option is set, the start of an unanchored pattern match must be
1413 before or at the first newline in the subject string following the
1414 start of matching, though the matched text may continue over the new‐
1415 line. If startoffset is non-zero, the limiting newline is not necessar‐
1416 ily the first newline in the subject. For example, if the subject
1417 string is "abc\nxyz" (where \n represents a single-character newline) a
1418 pattern match for "yz" succeeds with PCRE2_FIRSTLINE if startoffset is
1419 greater than 3. See also PCRE2_USE_OFFSET_LIMIT, which provides a more
1420 general limiting facility. If PCRE2_FIRSTLINE is set with an offset
1421 limit, a match must occur in the first line and also within the offset
1422 limit. In other words, whichever limit comes first is used.
1423 PCRE2_LITERAL
1425 If this option is set, all meta-characters in the pattern are disabled,
1427 and it is treated as a literal string. Matching literal strings with a
1428 regular expression engine is not the most efficient way of doing it. If
1429 you are doing a lot of literal matching and are worried about effi‐
1430 ciency, you should consider using other approaches. The only other main
1431 options that are allowed with PCRE2_LITERAL are: PCRE2_ANCHORED,
1432 PCRE2_ENDANCHORED, PCRE2_AUTO_CALLOUT, PCRE2_CASELESS, PCRE2_FIRSTLINE,
1433 PCRE2_MATCH_INVALID_UTF, PCRE2_NO_START_OPTIMIZE, PCRE2_NO_UTF_CHECK,
1434 PCRE2_UTF, and PCRE2_USE_OFFSET_LIMIT. The extra options
1435 PCRE2_EXTRA_MATCH_LINE and PCRE2_EXTRA_MATCH_WORD are also supported.
1436 Any other options cause an error.
1437 PCRE2_MATCH_INVALID_UTF
1439 This option forces PCRE2_UTF (see below) and also enables support for
1441 matching by pcre2_match() in subject strings that contain invalid UTF
1442 sequences. This facility is not supported for DFA matching. For
1443 details, see the pcre2unicode documentation.
1444 PCRE2_MATCH_UNSET_BACKREF
1446 If this option is set, a backreference to an unset capture group
1448 matches an empty string (by default this causes the current matching
1449 alternative to fail). A pattern such as (\1)(a) succeeds when this
1450 option is set (assuming it can find an "a" in the subject), whereas it
1451 fails by default, for Perl compatibility. Setting this option makes
1452 PCRE2 behave more like ECMAscript (aka JavaScript).
1453 PCRE2_MULTILINE
1455 By default, for the purposes of matching "start of line" and "end of
1457 line", PCRE2 treats the subject string as consisting of a single line
1458 of characters, even if it actually contains newlines. The "start of
1459 line" metacharacter (^) matches only at the start of the string, and
1460 the "end of line" metacharacter ($) matches only at the end of the
1461 string, or before a terminating newline (except when PCRE2_DOL‐
1462 LAR_ENDONLY is set). Note, however, that unless PCRE2_DOTALL is set,
1463 the "any character" metacharacter (.) does not match at a newline. This
1464 behaviour (for ^, $, and dot) is the same as Perl.
1465 When PCRE2_MULTILINE it is set, the "start of line" and "end of line"
1467 constructs match immediately following or immediately before internal
1468 newlines in the subject string, respectively, as well as at the very
1469 start and end. This is equivalent to Perl's /m option, and it can be
1470 changed within a pattern by a (?m) option setting. Note that the "start
1471 of line" metacharacter does not match after a newline at the end of the
1472 subject, for compatibility with Perl. However, you can change this by
1473 setting the PCRE2_ALT_CIRCUMFLEX option. If there are no newlines in a
1474 subject string, or no occurrences of ^ or $ in a pattern, setting
1475 PCRE2_MULTILINE has no effect.
1476 PCRE2_NEVER_BACKSLASH_C
1478 This option locks out the use of \C in the pattern that is being com‐
1480 piled. This escape can cause unpredictable behaviour in UTF-8 or
1481 UTF-16 modes, because it may leave the current matching point in the
1482 middle of a multi-code-unit character. This option may be useful in
1483 applications that process patterns from external sources. Note that
1484 there is also a build-time option that permanently locks out the use of
1485 \C.
1486 PCRE2_NEVER_UCP
1488 This option locks out the use of Unicode properties for handling \B,
1490 \b, \D, \d, \S, \s, \W, \w, and some of the POSIX character classes, as
1491 described for the PCRE2_UCP option below. In particular, it prevents
1492 the creator of the pattern from enabling this facility by starting the
1493 pattern with (*UCP). This option may be useful in applications that
1494 process patterns from external sources. The option combination PCRE_UCP
1495 and PCRE_NEVER_UCP causes an error.
1496 PCRE2_NEVER_UTF
1498 This option locks out interpretation of the pattern as UTF-8, UTF-16,
1500 or UTF-32, depending on which library is in use. In particular, it pre‐
1501 vents the creator of the pattern from switching to UTF interpretation
1502 by starting the pattern with (*UTF). This option may be useful in
1503 applications that process patterns from external sources. The combina‐
1504 tion of PCRE2_UTF and PCRE2_NEVER_UTF causes an error.
1505 PCRE2_NO_AUTO_CAPTURE
1507 If this option is set, it disables the use of numbered capturing paren‐
1509 theses in the pattern. Any opening parenthesis that is not followed by
1510 ? behaves as if it were followed by ?: but named parentheses can still
1511 be used for capturing (and they acquire numbers in the usual way). This
1512 is the same as Perl's /n option. Note that, when this option is set,
1513 references to capture groups (backreferences or recursion/subroutine
1514 calls) may only refer to named groups, though the reference can be by
1515 name or by number.
1516 PCRE2_NO_AUTO_POSSESS
1518 If this option is set, it disables "auto-possessification", which is an
1520 optimization that, for example, turns a+b into a++b in order to avoid
1521 backtracks into a+ that can never be successful. However, if callouts
1522 are in use, auto-possessification means that some callouts are never
1523 taken. You can set this option if you want the matching functions to do
1524 a full unoptimized search and run all the callouts, but it is mainly
1525 provided for testing purposes.
1526 PCRE2_NO_DOTSTAR_ANCHOR
1528 If this option is set, it disables an optimization that is applied when
1530 .* is the first significant item in a top-level branch of a pattern,
1531 and all the other branches also start with .* or with \A or \G or ^.
1532 The optimization is automatically disabled for .* if it is inside an
1533 atomic group or a capture group that is the subject of a backreference,
1534 or if the pattern contains (*PRUNE) or (*SKIP). When the optimization
1535 is not disabled, such a pattern is automatically anchored if
1536 PCRE2_DOTALL is set for all the .* items and PCRE2_MULTILINE is not set
1537 for any ^ items. Otherwise, the fact that any match must start either
1538 at the start of the subject or following a newline is remembered. Like
1539 other optimizations, this can cause callouts to be skipped.
1540 PCRE2_NO_START_OPTIMIZE
1542 This is an option whose main effect is at matching time. It does not
1544 change what pcre2_compile() generates, but it does affect the output of
1545 the JIT compiler.
1546 There are a number of optimizations that may occur at the start of a
1548 match, in order to speed up the process. For example, if it is known
1549 that an unanchored match must start with a specific code unit value,
1550 the matching code searches the subject for that value, and fails imme‐
1551 diately if it cannot find it, without actually running the main match‐
1552 ing function. This means that a special item such as (*COMMIT) at the
1553 start of a pattern is not considered until after a suitable starting
1554 point for the match has been found. Also, when callouts or (*MARK)
1555 items are in use, these "start-up" optimizations can cause them to be
1556 skipped if the pattern is never actually used. The start-up optimiza‐
1557 tions are in effect a pre-scan of the subject that takes place before
1558 the pattern is run.
1559 The PCRE2_NO_START_OPTIMIZE option disables the start-up optimizations,
1561 possibly causing performance to suffer, but ensuring that in cases
1562 where the result is "no match", the callouts do occur, and that items
1563 such as (*COMMIT) and (*MARK) are considered at every possible starting
1564 position in the subject string.
1565 Setting PCRE2_NO_START_OPTIMIZE may change the outcome of a matching
1567 operation. Consider the pattern
1568 (*COMMIT)ABC
1570 When this is compiled, PCRE2 records the fact that a match must start
1572 with the character "A". Suppose the subject string is "DEFABC". The
1573 start-up optimization scans along the subject, finds "A" and runs the
1574 first match attempt from there. The (*COMMIT) item means that the pat‐
1575 tern must match the current starting position, which in this case, it
1576 does. However, if the same match is run with PCRE2_NO_START_OPTIMIZE
1577 set, the initial scan along the subject string does not happen. The
1578 first match attempt is run starting from "D" and when this fails,
1579 (*COMMIT) prevents any further matches being tried, so the overall
1580 result is "no match".
1581 As another start-up optimization makes use of a minimum length for a
1583 matching subject, which is recorded when possible. Consider the pattern
1584 (*MARK:1)B(*MARK:2)(X|Y)
1586 The minimum length for a match is two characters. If the subject is
1588 "XXBB", the "starting character" optimization skips "XX", then tries to
1589 match "BB", which is long enough. In the process, (*MARK:2) is encoun‐
1590 tered and remembered. When the match attempt fails, the next "B" is
1591 found, but there is only one character left, so there are no more
1592 attempts, and "no match" is returned with the "last mark seen" set to
1593 "2". If NO_START_OPTIMIZE is set, however, matches are tried at every
1594 possible starting position, including at the end of the subject, where
1595 (*MARK:1) is encountered, but there is no "B", so the "last mark seen"
1596 that is returned is "1". In this case, the optimizations do not affect
1597 the overall match result, which is still "no match", but they do affect
1598 the auxiliary information that is returned.
1599 PCRE2_NO_UTF_CHECK
1601 When PCRE2_UTF is set, the validity of the pattern as a UTF string is
1603 automatically checked. There are discussions about the validity of
1604 UTF-8 strings, UTF-16 strings, and UTF-32 strings in the pcre2unicode
1605 document. If an invalid UTF sequence is found, pcre2_compile() returns
1606 a negative error code.
1607 If you know that your pattern is a valid UTF string, and you want to
1609 skip this check for performance reasons, you can set the
1610 PCRE2_NO_UTF_CHECK option. When it is set, the effect of passing an
1611 invalid UTF string as a pattern is undefined. It may cause your program
1612 to crash or loop.
1613 Note that this option can also be passed to pcre2_match() and
1615 pcre_dfa_match(), to suppress UTF validity checking of the subject
1616 string.
1617 Note also that setting PCRE2_NO_UTF_CHECK at compile time does not dis‐
1619 able the error that is given if an escape sequence for an invalid Uni‐
1620 code code point is encountered in the pattern. In particular, the so-
1621 called "surrogate" code points (0xd800 to 0xdfff) are invalid. If you
1622 want to allow escape sequences such as \x{d800} you can set the
1623 PCRE2_EXTRA_ALLOW_SURROGATE_ESCAPES extra option, as described in the
1624 section entitled "Extra compile options" below. However, this is pos‐
1625 sible only in UTF-8 and UTF-32 modes, because these values are not rep‐
1626 resentable in UTF-16.
1627 PCRE2_UCP
1629 This option has two effects. Firstly, it change the way PCRE2 processes
1631 \B, \b, \D, \d, \S, \s, \W, \w, and some of the POSIX character
1632 classes. By default, only ASCII characters are recognized, but if
1633 PCRE2_UCP is set, Unicode properties are used instead to classify char‐
1634 acters. More details are given in the section on generic character
1635 types in the pcre2pattern page. If you set PCRE2_UCP, matching one of
1636 the items it affects takes much longer.
1637 The second effect of PCRE2_UCP is to force the use of Unicode proper‐
1639 ties for upper/lower casing operations on characters with code points
1640 greater than 127, even when PCRE2_UTF is not set. This makes it possi‐
1641 ble, for example, to process strings in the 16-bit UCS-2 code. This
1642 option is available only if PCRE2 has been compiled with Unicode sup‐
1643 port (which is the default).
1644 PCRE2_UNGREEDY
1646 This option inverts the "greediness" of the quantifiers so that they
1648 are not greedy by default, but become greedy if followed by "?". It is
1649 not compatible with Perl. It can also be set by a (?U) option setting
1650 within the pattern.
1651 PCRE2_USE_OFFSET_LIMIT
1653 This option must be set for pcre2_compile() if pcre2_set_offset_limit()
1655 is going to be used to set a non-default offset limit in a match con‐
1656 text for matches that use this pattern. An error is generated if an
1657 offset limit is set without this option. For more details, see the
1658 description of pcre2_set_offset_limit() in the section that describes
1659 match contexts. See also the PCRE2_FIRSTLINE option above.
1660 PCRE2_UTF
1662 This option causes PCRE2 to regard both the pattern and the subject
1664 strings that are subsequently processed as strings of UTF characters
1665 instead of single-code-unit strings. It is available when PCRE2 is
1666 built to include Unicode support (which is the default). If Unicode
1667 support is not available, the use of this option provokes an error.
1668 Details of how PCRE2_UTF changes the behaviour of PCRE2 are given in
1669 the pcre2unicode page. In particular, note that it changes the way
1670 PCRE2_CASELESS handles characters with code points greater than 127.
1671 Extra compile options
1673 The option bits that can be set in a compile context by calling the
1675 pcre2_set_compile_extra_options() function are as follows:
1676 PCRE2_EXTRA_ALLOW_SURROGATE_ESCAPES
1678 This option applies when compiling a pattern in UTF-8 or UTF-32 mode.
1680 It is forbidden in UTF-16 mode, and ignored in non-UTF modes. Unicode
1681 "surrogate" code points in the range 0xd800 to 0xdfff are used in pairs
1682 in UTF-16 to encode code points with values in the range 0x10000 to
1683 0x10ffff. The surrogates cannot therefore be represented in UTF-16.
1684 They can be represented in UTF-8 and UTF-32, but are defined as invalid
1685 code points, and cause errors if encountered in a UTF-8 or UTF-32
1686 string that is being checked for validity by PCRE2.
1687 These values also cause errors if encountered in escape sequences such
1689 as \x{d912} within a pattern. However, it seems that some applications,
1690 when using PCRE2 to check for unwanted characters in UTF-8 strings,
1691 explicitly test for the surrogates using escape sequences. The
1692 PCRE2_NO_UTF_CHECK option does not disable the error that occurs,
1693 because it applies only to the testing of input strings for UTF valid‐
1694 ity.
1695 If the extra option PCRE2_EXTRA_ALLOW_SURROGATE_ESCAPES is set, surro‐
1697 gate code point values in UTF-8 and UTF-32 patterns no longer provoke
1698 errors and are incorporated in the compiled pattern. However, they can
1699 only match subject characters if the matching function is called with
1700 PCRE2_NO_UTF_CHECK set.
1701 PCRE2_EXTRA_ALT_BSUX
1703 The original option PCRE2_ALT_BSUX causes PCRE2 to process \U, \u, and
1705 \x in the way that ECMAscript (aka JavaScript) does. Additional func‐
1706 tionality was defined by ECMAscript 6; setting PCRE2_EXTRA_ALT_BSUX has
1707 the effect of PCRE2_ALT_BSUX, but in addition it recognizes \u{hhh..}
1708 as a hexadecimal character code, where hhh.. is any number of hexadeci‐
1709 mal digits.
1710 PCRE2_EXTRA_BAD_ESCAPE_IS_LITERAL
1712 This is a dangerous option. Use with care. By default, an unrecognized
1714 escape such as \j or a malformed one such as \x{2z} causes a compile-
1715 time error when detected by pcre2_compile(). Perl is somewhat inconsis‐
1716 tent in handling such items: for example, \j is treated as a literal
1717 "j", and non-hexadecimal digits in \x{} are just ignored, though warn‐
1718 ings are given in both cases if Perl's warning switch is enabled. How‐
1719 ever, a malformed octal number after \o{ always causes an error in
1720 Perl.
1721 If the PCRE2_EXTRA_BAD_ESCAPE_IS_LITERAL extra option is passed to
1723 pcre2_compile(), all unrecognized or malformed escape sequences are
1724 treated as single-character escapes. For example, \j is a literal "j"
1725 and \x{2z} is treated as the literal string "x{2z}". Setting this
1726 option means that typos in patterns may go undetected and have unex‐
1727 pected results. Also note that a sequence such as [\N{] is interpreted
1728 as a malformed attempt at [\N{...}] and so is treated as [N{] whereas
1729 [\N] gives an error because an unqualified \N is a valid escape
1730 sequence but is not supported in a character class. To reiterate: this
1731 is a dangerous option. Use with great care.
1732 PCRE2_EXTRA_ESCAPED_CR_IS_LF
1734 There are some legacy applications where the escape sequence \r in a
1736 pattern is expected to match a newline. If this option is set, \r in a
1737 pattern is converted to \n so that it matches a LF (linefeed) instead
1738 of a CR (carriage return) character. The option does not affect a lit‐
1739 eral CR in the pattern, nor does it affect CR specified as an explicit
1740 code point such as \x{0D}.
1741 PCRE2_EXTRA_MATCH_LINE
1743 This option is provided for use by the -x option of pcre2grep. It
1745 causes the pattern only to match complete lines. This is achieved by
1746 automatically inserting the code for "^(?:" at the start of the com‐
1747 piled pattern and ")$" at the end. Thus, when PCRE2_MULTILINE is set,
1748 the matched line may be in the middle of the subject string. This
1749 option can be used with PCRE2_LITERAL.
1750 PCRE2_EXTRA_MATCH_WORD
1752 This option is provided for use by the -w option of pcre2grep. It
1754 causes the pattern only to match strings that have a word boundary at
1755 the start and the end. This is achieved by automatically inserting the
1756 code for "\b(?:" at the start of the compiled pattern and ")\b" at the
1757 end. The option may be used with PCRE2_LITERAL. However, it is ignored
1758 if PCRE2_EXTRA_MATCH_LINE is also set.
1759
1761 int pcre2_jit_compile(pcre2_code *code, uint32_t options);
1763 int pcre2_jit_match(const pcre2_code *code, PCRE2_SPTR subject,
1765 PCRE2_SIZE length, PCRE2_SIZE startoffset,
1766 uint32_t options, pcre2_match_data *match_data,
1767 pcre2_match_context *mcontext);
1768 void pcre2_jit_free_unused_memory(pcre2_general_context *gcontext);
1770 pcre2_jit_stack *pcre2_jit_stack_create(PCRE2_SIZE startsize,
1772 PCRE2_SIZE maxsize, pcre2_general_context *gcontext);
1773 void pcre2_jit_stack_assign(pcre2_match_context *mcontext,
1775 pcre2_jit_callback callback_function, void *callback_data);
1776 void pcre2_jit_stack_free(pcre2_jit_stack *jit_stack);
1778 These functions provide support for JIT compilation, which, if the
1780 just-in-time compiler is available, further processes a compiled pat‐
1781 tern into machine code that executes much faster than the pcre2_match()
1782 interpretive matching function. Full details are given in the pcre2jit
1783 documentation.
1784 JIT compilation is a heavyweight optimization. It can take some time
1786 for patterns to be analyzed, and for one-off matches and simple pat‐
1787 terns the benefit of faster execution might be offset by a much slower
1788 compilation time. Most (but not all) patterns can be optimized by the
1789 JIT compiler.
1790
1792 const uint8_t *pcre2_maketables(pcre2_general_context *gcontext);
1794 void pcre2_maketables_free(pcre2_general_context *gcontext,
1796 const uint8_t *tables);
1797 PCRE2 handles caseless matching, and determines whether characters are
1799 letters, digits, or whatever, by reference to a set of tables, indexed
1800 by character code point. However, this applies only to characters whose
1801 code points are less than 256. By default, higher-valued code points
1802 never match escapes such as \w or \d.
1803 When PCRE2 is built with Unicode support (the default), the Unicode
1805 properties of all characters can be tested with \p and \P, or, alterna‐
1806 tively, the PCRE2_UCP option can be set when a pattern is compiled;
1807 this causes \w and friends to use Unicode property support instead of
1808 the built-in tables. PCRE2_UCP also causes upper/lower casing opera‐
1809 tions on characters with code points greater than 127 to use Unicode
1810 properties. These effects apply even when PCRE2_UTF is not set.
1811 The use of locales with Unicode is discouraged. If you are handling
1813 characters with code points greater than 127, you should either use
1814 Unicode support, or use locales, but not try to mix the two.
1815 PCRE2 contains a built-in set of character tables that are used by
1817 default. These are sufficient for many applications. Normally, the
1818 internal tables recognize only ASCII characters. However, when PCRE2 is
1819 built, it is possible to cause the internal tables to be rebuilt in the
1820 default "C" locale of the local system, which may cause them to be dif‐
1821 ferent.
1822 The built-in tables can be overridden by tables supplied by the appli‐
1824 cation that calls PCRE2. These may be created in a different locale
1825 from the default. As more and more applications change to using Uni‐
1826 code, the need for this locale support is expected to die away.
1827 External tables are built by calling the pcre2_maketables() function,
1829 in the relevant locale. The only argument to this function is a general
1830 context, which can be used to pass a custom memory allocator. If the
1831 argument is NULL, the system malloc() is used. The result can be passed
1832 to pcre2_compile() as often as necessary, by creating a compile context
1833 and calling pcre2_set_character_tables() to set the tables pointer
1834 therein.
1835 For example, to build and use tables that are appropriate for the
1837 French locale (where accented characters with values greater than 127
1838 are treated as letters), the following code could be used:
1839 setlocale(LC_CTYPE, "fr_FR");
1841 tables = pcre2_maketables(NULL);
1842 ccontext = pcre2_compile_context_create(NULL);
1843 pcre2_set_character_tables(ccontext, tables);
1844 re = pcre2_compile(..., ccontext);
1845 The locale name "fr_FR" is used on Linux and other Unix-like systems;
1847 if you are using Windows, the name for the French locale is "french".
1848 The pointer that is passed (via the compile context) to pcre2_compile()
1850 is saved with the compiled pattern, and the same tables are used by the
1851 matching functions. Thus, for any single pattern, compilation and
1852 matching both happen in the same locale, but different patterns can be
1853 processed in different locales.
1854 It is the caller's responsibility to ensure that the memory containing
1856 the tables remains available while they are still in use. When they are
1857 no longer needed, you can discard them using pcre2_maketables_free(),
1858 which should pass as its first parameter the same global context that
1859 was used to create the tables.
1860 Saving locale tables
1862 The tables described above are just a sequence of binary bytes, which
1864 makes them independent of hardware characteristics such as endianness
1865 or whether the processor is 32-bit or 64-bit. A copy of the result of
1866 pcre2_maketables() can therefore be saved in a file or elsewhere and
1867 re-used later, even in a different program or on another computer. The
1868 size of the tables (number of bytes) must be obtained by calling
1869 pcre2_config() with the PCRE2_CONFIG_TABLES_LENGTH option because
1870 pcre2_maketables() does not return this value. Note that the
1871 pcre2_dftables program, which is part of the PCRE2 build system, can be
1872 used stand-alone to create a file that contains a set of binary tables.
1873 See the pcre2build documentation for details.
1874
1876 int pcre2_pattern_info(const pcre2 *code, uint32_t what, void *where);
1878 The pcre2_pattern_info() function returns general information about a
1880 compiled pattern. For information about callouts, see the next section.
1881 The first argument for pcre2_pattern_info() is a pointer to the com‐
1882 piled pattern. The second argument specifies which piece of information
1883 is required, and the third argument is a pointer to a variable to
1884 receive the data. If the third argument is NULL, the first argument is
1885 ignored, and the function returns the size in bytes of the variable
1886 that is required for the information requested. Otherwise, the yield of
1887 the function is zero for success, or one of the following negative num‐
1888 bers:
1889 PCRE2_ERROR_NULL the argument code was NULL
1891 PCRE2_ERROR_BADMAGIC the "magic number" was not found
1892 PCRE2_ERROR_BADOPTION the value of what was invalid
1893 PCRE2_ERROR_UNSET the requested field is not set
1894 The "magic number" is placed at the start of each compiled pattern as a
1896 simple check against passing an arbitrary memory pointer. Here is a
1897 typical call of pcre2_pattern_info(), to obtain the length of the com‐
1898 piled pattern:
1899 int rc;
1901 size_t length;
1902 rc = pcre2_pattern_info(
1903 re, /* result of pcre2_compile() */
1904 PCRE2_INFO_SIZE, /* what is required */
1905 &length); /* where to put the data */
1906 The possible values for the second argument are defined in pcre2.h, and
1908 are as follows:
1909 PCRE2_INFO_ALLOPTIONS
1911 PCRE2_INFO_ARGOPTIONS
1912 PCRE2_INFO_EXTRAOPTIONS
1913 Return copies of the pattern's options. The third argument should point
1915 to a uint32_t variable. PCRE2_INFO_ARGOPTIONS returns exactly the
1916 options that were passed to pcre2_compile(), whereas PCRE2_INFO_ALLOP‐
1917 TIONS returns the compile options as modified by any top-level (*XXX)
1918 option settings such as (*UTF) at the start of the pattern itself.
1919 PCRE2_INFO_EXTRAOPTIONS returns the extra options that were set in the
1920 compile context by calling the pcre2_set_compile_extra_options() func‐
1921 tion.
1922 For example, if the pattern /(*UTF)abc/ is compiled with the
1924 PCRE2_EXTENDED option, the result for PCRE2_INFO_ALLOPTIONS is
1925 PCRE2_EXTENDED and PCRE2_UTF. Option settings such as (?i) that can
1926 change within a pattern do not affect the result of PCRE2_INFO_ALLOP‐
1927 TIONS, even if they appear right at the start of the pattern. (This was
1928 different in some earlier releases.)
1929 A pattern compiled without PCRE2_ANCHORED is automatically anchored by
1931 PCRE2 if the first significant item in every top-level branch is one of
1932 the following:
1933 ^ unless PCRE2_MULTILINE is set
1935 \A always
1936 \G always
1937 .* sometimes - see below
1938 When .* is the first significant item, anchoring is possible only when
1940 all the following are true:
1941 .* is not in an atomic group
1943 .* is not in a capture group that is the subject
1944 of a backreference
1945 PCRE2_DOTALL is in force for .*
1946 Neither (*PRUNE) nor (*SKIP) appears in the pattern
1947 PCRE2_NO_DOTSTAR_ANCHOR is not set
1948 For patterns that are auto-anchored, the PCRE2_ANCHORED bit is set in
1950 the options returned for PCRE2_INFO_ALLOPTIONS.
1951 PCRE2_INFO_BACKREFMAX
1953 Return the number of the highest backreference in the pattern. The
1955 third argument should point to a uint32_t variable. Named capture
1956 groups acquire numbers as well as names, and these count towards the
1957 highest backreference. Backreferences such as \4 or \g{12} match the
1958 captured characters of the given group, but in addition, the check that
1959 a capture group is set in a conditional group such as (?(3)a|b) is also
1960 a backreference. Zero is returned if there are no backreferences.
1961 PCRE2_INFO_BSR
1963 The output is a uint32_t integer whose value indicates what character
1965 sequences the \R escape sequence matches. A value of PCRE2_BSR_UNICODE
1966 means that \R matches any Unicode line ending sequence; a value of
1967 PCRE2_BSR_ANYCRLF means that \R matches only CR, LF, or CRLF.
1968 PCRE2_INFO_CAPTURECOUNT
1970 Return the highest capture group number in the pattern. In patterns
1972 where (?| is not used, this is also the total number of capture groups.
1973 The third argument should point to a uint32_t variable.
1974 PCRE2_INFO_DEPTHLIMIT
1976 If the pattern set a backtracking depth limit by including an item of
1978 the form (*LIMIT_DEPTH=nnnn) at the start, the value is returned. The
1979 third argument should point to a uint32_t integer. If no such value has
1980 been set, the call to pcre2_pattern_info() returns the error
1981 PCRE2_ERROR_UNSET. Note that this limit will only be used during match‐
1982 ing if it is less than the limit set or defaulted by the caller of the
1983 match function.
1984 PCRE2_INFO_FIRSTBITMAP
1986 In the absence of a single first code unit for a non-anchored pattern,
1988 pcre2_compile() may construct a 256-bit table that defines a fixed set
1989 of values for the first code unit in any match. For example, a pattern
1990 that starts with [abc] results in a table with three bits set. When
1991 code unit values greater than 255 are supported, the flag bit for 255
1992 means "any code unit of value 255 or above". If such a table was con‐
1993 structed, a pointer to it is returned. Otherwise NULL is returned. The
1994 third argument should point to a const uint8_t * variable.
1995 PCRE2_INFO_FIRSTCODETYPE
1997 Return information about the first code unit of any matched string, for
1999 a non-anchored pattern. The third argument should point to a uint32_t
2000 variable. If there is a fixed first value, for example, the letter "c"
2001 from a pattern such as (cat|cow|coyote), 1 is returned, and the value
2002 can be retrieved using PCRE2_INFO_FIRSTCODEUNIT. If there is no fixed
2003 first value, but it is known that a match can occur only at the start
2004 of the subject or following a newline in the subject, 2 is returned.
2005 Otherwise, and for anchored patterns, 0 is returned.
2006 PCRE2_INFO_FIRSTCODEUNIT
2008 Return the value of the first code unit of any matched string for a
2010 pattern where PCRE2_INFO_FIRSTCODETYPE returns 1; otherwise return 0.
2011 The third argument should point to a uint32_t variable. In the 8-bit
2012 library, the value is always less than 256. In the 16-bit library the
2013 value can be up to 0xffff. In the 32-bit library in UTF-32 mode the
2014 value can be up to 0x10ffff, and up to 0xffffffff when not using UTF-32
2015 mode.
2016 PCRE2_INFO_FRAMESIZE
2018 Return the size (in bytes) of the data frames that are used to remember
2020 backtracking positions when the pattern is processed by pcre2_match()
2021 without the use of JIT. The third argument should point to a size_t
2022 variable. The frame size depends on the number of capturing parentheses
2023 in the pattern. Each additional capture group adds two PCRE2_SIZE vari‐
2024 ables.
2025 PCRE2_INFO_HASBACKSLASHC
2027 Return 1 if the pattern contains any instances of \C, otherwise 0. The
2029 third argument should point to a uint32_t variable.
2030 PCRE2_INFO_HASCRORLF
2032 Return 1 if the pattern contains any explicit matches for CR or LF
2034 characters, otherwise 0. The third argument should point to a uint32_t
2035 variable. An explicit match is either a literal CR or LF character, or
2036 \r or \n or one of the equivalent hexadecimal or octal escape
2037 sequences.
2038 PCRE2_INFO_HEAPLIMIT
2040 If the pattern set a heap memory limit by including an item of the form
2042 (*LIMIT_HEAP=nnnn) at the start, the value is returned. The third argu‐
2043 ment should point to a uint32_t integer. If no such value has been set,
2044 the call to pcre2_pattern_info() returns the error PCRE2_ERROR_UNSET.
2045 Note that this limit will only be used during matching if it is less
2046 than the limit set or defaulted by the caller of the match function.
2047 PCRE2_INFO_JCHANGED
2049 Return 1 if the (?J) or (?-J) option setting is used in the pattern,
2051 otherwise 0. The third argument should point to a uint32_t variable.
2052 (?J) and (?-J) set and unset the local PCRE2_DUPNAMES option, respec‐
2053 tively.
2054 PCRE2_INFO_JITSIZE
2056 If the compiled pattern was successfully processed by pcre2_jit_com‐
2058 pile(), return the size of the JIT compiled code, otherwise return
2059 zero. The third argument should point to a size_t variable.
2060 PCRE2_INFO_LASTCODETYPE
2062 Returns 1 if there is a rightmost literal code unit that must exist in
2064 any matched string, other than at its start. The third argument should
2065 point to a uint32_t variable. If there is no such value, 0 is returned.
2066 When 1 is returned, the code unit value itself can be retrieved using
2067 PCRE2_INFO_LASTCODEUNIT. For anchored patterns, a last literal value is
2068 recorded only if it follows something of variable length. For example,
2069 for the pattern /^a\d+z\d+/ the returned value is 1 (with "z" returned
2070 from PCRE2_INFO_LASTCODEUNIT), but for /^a\dz\d/ the returned value is
2071 0.
2072 PCRE2_INFO_LASTCODEUNIT
2074 Return the value of the rightmost literal code unit that must exist in
2076 any matched string, other than at its start, for a pattern where
2077 PCRE2_INFO_LASTCODETYPE returns 1. Otherwise, return 0. The third argu‐
2078 ment should point to a uint32_t variable.
2079 PCRE2_INFO_MATCHEMPTY
2081 Return 1 if the pattern might match an empty string, otherwise 0. The
2083 third argument should point to a uint32_t variable. When a pattern con‐
2084 tains recursive subroutine calls it is not always possible to determine
2085 whether or not it can match an empty string. PCRE2 takes a cautious
2086 approach and returns 1 in such cases.
2087 PCRE2_INFO_MATCHLIMIT
2089 If the pattern set a match limit by including an item of the form
2091 (*LIMIT_MATCH=nnnn) at the start, the value is returned. The third
2092 argument should point to a uint32_t integer. If no such value has been
2093 set, the call to pcre2_pattern_info() returns the error
2094 PCRE2_ERROR_UNSET. Note that this limit will only be used during match‐
2095 ing if it is less than the limit set or defaulted by the caller of the
2096 match function.
2097 PCRE2_INFO_MAXLOOKBEHIND
2099 A lookbehind assertion moves back a certain number of characters (not
2101 code units) when it starts to process each of its branches. This
2102 request returns the largest of these backward moves. The third argument
2103 should point to a uint32_t integer. The simple assertions \b and \B
2104 require a one-character lookbehind and cause PCRE2_INFO_MAXLOOKBEHIND
2105 to return 1 in the absence of anything longer. \A also registers a one-
2106 character lookbehind, though it does not actually inspect the previous
2107 character.
2108 Note that this information is useful for multi-segment matching only if
2110 the pattern contains no nested lookbehinds. For example, the pattern
2111 (?<=a(?<=ba)c) returns a maximum lookbehind of 2, but when it is pro‐
2112 cessed, the first lookbehind moves back by two characters, matches one
2113 character, then the nested lookbehind also moves back by two charac‐
2114 ters. This puts the matching point three characters earlier than it was
2115 at the start. PCRE2_INFO_MAXLOOKBEHIND is really only useful as a
2116 debugging tool. See the pcre2partial documentation for a discussion of
2117 multi-segment matching.
2118 PCRE2_INFO_MINLENGTH
2120 If a minimum length for matching subject strings was computed, its
2122 value is returned. Otherwise the returned value is 0. This value is not
2123 computed when PCRE2_NO_START_OPTIMIZE is set. The value is a number of
2124 characters, which in UTF mode may be different from the number of code
2125 units. The third argument should point to a uint32_t variable. The
2126 value is a lower bound to the length of any matching string. There may
2127 not be any strings of that length that do actually match, but every
2128 string that does match is at least that long.
2129 PCRE2_INFO_NAMECOUNT
2131 PCRE2_INFO_NAMEENTRYSIZE
2132 PCRE2_INFO_NAMETABLE
2133 PCRE2 supports the use of named as well as numbered capturing parenthe‐
2135 ses. The names are just an additional way of identifying the parenthe‐
2136 ses, which still acquire numbers. Several convenience functions such as
2137 pcre2_substring_get_byname() are provided for extracting captured sub‐
2138 strings by name. It is also possible to extract the data directly, by
2139 first converting the name to a number in order to access the correct
2140 pointers in the output vector (described with pcre2_match() below). To
2141 do the conversion, you need to use the name-to-number map, which is
2142 described by these three values.
2143 The map consists of a number of fixed-size entries. PCRE2_INFO_NAME‐
2145 COUNT gives the number of entries, and PCRE2_INFO_NAMEENTRYSIZE gives
2146 the size of each entry in code units; both of these return a uint32_t
2147 value. The entry size depends on the length of the longest name.
2148 PCRE2_INFO_NAMETABLE returns a pointer to the first entry of the table.
2150 This is a PCRE2_SPTR pointer to a block of code units. In the 8-bit
2151 library, the first two bytes of each entry are the number of the cap‐
2152 turing parenthesis, most significant byte first. In the 16-bit library,
2153 the pointer points to 16-bit code units, the first of which contains
2154 the parenthesis number. In the 32-bit library, the pointer points to
2155 32-bit code units, the first of which contains the parenthesis number.
2156 The rest of the entry is the corresponding name, zero terminated.
2157 The names are in alphabetical order. If (?| is used to create multiple
2159 capture groups with the same number, as described in the section on
2160 duplicate group numbers in the pcre2pattern page, the groups may be
2161 given the same name, but there is only one entry in the table. Differ‐
2162 ent names for groups of the same number are not permitted.
2163 Duplicate names for capture groups with different numbers are permit‐
2165 ted, but only if PCRE2_DUPNAMES is set. They appear in the table in the
2166 order in which they were found in the pattern. In the absence of (?|
2167 this is the order of increasing number; when (?| is used this is not
2168 necessarily the case because later capture groups may have lower num‐
2169 bers.
2170 As a simple example of the name/number table, consider the following
2172 pattern after compilation by the 8-bit library (assume PCRE2_EXTENDED
2173 is set, so white space - including newlines - is ignored):
2174 (?<date> (?<year>(\d\d)?\d\d) -
2176 (?<month>\d\d) - (?<day>\d\d) )
2177 There are four named capture groups, so the table has four entries, and
2179 each entry in the table is eight bytes long. The table is as follows,
2180 with non-printing bytes shows in hexadecimal, and undefined bytes shown
2181 as ??:
2182 00 01 d a t e 00 ??
2184 00 05 d a y 00 ?? ??
2185 00 04 m o n t h 00
2186 00 02 y e a r 00 ??
2187 When writing code to extract data from named capture groups using the
2189 name-to-number map, remember that the length of the entries is likely
2190 to be different for each compiled pattern.
2191 PCRE2_INFO_NEWLINE
2193 The output is one of the following uint32_t values:
2195 PCRE2_NEWLINE_CR Carriage return (CR)
2197 PCRE2_NEWLINE_LF Linefeed (LF)
2198 PCRE2_NEWLINE_CRLF Carriage return, linefeed (CRLF)
2199 PCRE2_NEWLINE_ANY Any Unicode line ending
2200 PCRE2_NEWLINE_ANYCRLF Any of CR, LF, or CRLF
2201 PCRE2_NEWLINE_NUL The NUL character (binary zero)
2202 This identifies the character sequence that will be recognized as mean‐
2204 ing "newline" while matching.
2205 PCRE2_INFO_SIZE
2207 Return the size of the compiled pattern in bytes (for all three
2209 libraries). The third argument should point to a size_t variable. This
2210 value includes the size of the general data block that precedes the
2211 code units of the compiled pattern itself. The value that is used when
2212 pcre2_compile() is getting memory in which to place the compiled pat‐
2213 tern may be slightly larger than the value returned by this option,
2214 because there are cases where the code that calculates the size has to
2215 over-estimate. Processing a pattern with the JIT compiler does not
2216 alter the value returned by this option.
2217
2219 int pcre2_callout_enumerate(const pcre2_code *code,
2221 int (*callback)(pcre2_callout_enumerate_block *, void *),
2222 void *user_data);
2223 A script language that supports the use of string arguments in callouts
2225 might like to scan all the callouts in a pattern before running the
2226 match. This can be done by calling pcre2_callout_enumerate(). The first
2227 argument is a pointer to a compiled pattern, the second points to a
2228 callback function, and the third is arbitrary user data. The callback
2229 function is called for every callout in the pattern in the order in
2230 which they appear. Its first argument is a pointer to a callout enumer‐
2231 ation block, and its second argument is the user_data value that was
2232 passed to pcre2_callout_enumerate(). The contents of the callout enu‐
2233 meration block are described in the pcre2callout documentation, which
2234 also gives further details about callouts.
2235
2237 It is possible to save compiled patterns on disc or elsewhere, and
2239 reload them later, subject to a number of restrictions. The host on
2240 which the patterns are reloaded must be running the same version of
2241 PCRE2, with the same code unit width, and must also have the same endi‐
2242 anness, pointer width, and PCRE2_SIZE type. Before compiled patterns
2243 can be saved, they must be converted to a "serialized" form, which in
2244 the case of PCRE2 is really just a bytecode dump. The functions whose
2245 names begin with pcre2_serialize_ are used for converting to and from
2246 the serialized form. They are described in the pcre2serialize documen‐
2247 tation. Note that PCRE2 serialization does not convert compiled pat‐
2248 terns to an abstract format like Java or .NET serialization.
2249
2251 pcre2_match_data *pcre2_match_data_create(uint32_t ovecsize,
2253 pcre2_general_context *gcontext);
2254 pcre2_match_data *pcre2_match_data_create_from_pattern(
2256 const pcre2_code *code, pcre2_general_context *gcontext);
2257 void pcre2_match_data_free(pcre2_match_data *match_data);
2259 Information about a successful or unsuccessful match is placed in a
2261 match data block, which is an opaque structure that is accessed by
2262 function calls. In particular, the match data block contains a vector
2263 of offsets into the subject string that define the matched part of the
2264 subject and any substrings that were captured. This is known as the
2265 ovector.
2266 Before calling pcre2_match(), pcre2_dfa_match(), or pcre2_jit_match()
2268 you must create a match data block by calling one of the creation func‐
2269 tions above. For pcre2_match_data_create(), the first argument is the
2270 number of pairs of offsets in the ovector. One pair of offsets is
2271 required to identify the string that matched the whole pattern, with an
2272 additional pair for each captured substring. For example, a value of 4
2273 creates enough space to record the matched portion of the subject plus
2274 three captured substrings. A minimum of at least 1 pair is imposed by
2275 pcre2_match_data_create(), so it is always possible to return the over‐
2276 all matched string.
2277 The second argument of pcre2_match_data_create() is a pointer to a gen‐
2279 eral context, which can specify custom memory management for obtaining
2280 the memory for the match data block. If you are not using custom memory
2281 management, pass NULL, which causes malloc() to be used.
2282 For pcre2_match_data_create_from_pattern(), the first argument is a
2284 pointer to a compiled pattern. The ovector is created to be exactly the
2285 right size to hold all the substrings a pattern might capture. The sec‐
2286 ond argument is again a pointer to a general context, but in this case
2287 if NULL is passed, the memory is obtained using the same allocator that
2288 was used for the compiled pattern (custom or default).
2289 A match data block can be used many times, with the same or different
2291 compiled patterns. You can extract information from a match data block
2292 after a match operation has finished, using functions that are
2293 described in the sections on matched strings and other match data
2294 below.
2295 When a call of pcre2_match() fails, valid data is available in the
2297 match block only when the error is PCRE2_ERROR_NOMATCH,
2298 PCRE2_ERROR_PARTIAL, or one of the error codes for an invalid UTF
2299 string. Exactly what is available depends on the error, and is detailed
2300 below.
2301 When one of the matching functions is called, pointers to the compiled
2303 pattern and the subject string are set in the match data block so that
2304 they can be referenced by the extraction functions after a successful
2305 match. After running a match, you must not free a compiled pattern or a
2306 subject string until after all operations on the match data block (for
2307 that match) have taken place, unless, in the case of the subject
2308 string, you have used the PCRE2_COPY_MATCHED_SUBJECT option, which is
2309 described in the section entitled "Option bits for pcre2_match()"
2310 below.
2311 When a match data block itself is no longer needed, it should be freed
2313 by calling pcre2_match_data_free(). If this function is called with a
2314 NULL argument, it returns immediately, without doing anything.
2315
2317 int pcre2_match(const pcre2_code *code, PCRE2_SPTR subject,
2319 PCRE2_SIZE length, PCRE2_SIZE startoffset,
2320 uint32_t options, pcre2_match_data *match_data,
2321 pcre2_match_context *mcontext);
2322 The function pcre2_match() is called to match a subject string against
2324 a compiled pattern, which is passed in the code argument. You can call
2325 pcre2_match() with the same code argument as many times as you like, in
2326 order to find multiple matches in the subject string or to match dif‐
2327 ferent subject strings with the same pattern.
2328 This function is the main matching facility of the library, and it
2330 operates in a Perl-like manner. For specialist use there is also an
2331 alternative matching function, which is described below in the section
2332 about the pcre2_dfa_match() function.
2333 Here is an example of a simple call to pcre2_match():
2335 pcre2_match_data *md = pcre2_match_data_create(4, NULL);
2337 int rc = pcre2_match(
2338 re, /* result of pcre2_compile() */
2339 "some string", /* the subject string */
2340 11, /* the length of the subject string */
2341 0, /* start at offset 0 in the subject */
2342 0, /* default options */
2343 md, /* the match data block */
2344 NULL); /* a match context; NULL means use defaults */
2345 If the subject string is zero-terminated, the length can be given as
2347 PCRE2_ZERO_TERMINATED. A match context must be provided if certain less
2348 common matching parameters are to be changed. For details, see the sec‐
2349 tion on the match context above.
2350 The string to be matched by pcre2_match()
2352 The subject string is passed to pcre2_match() as a pointer in subject,
2354 a length in length, and a starting offset in startoffset. The length
2355 and offset are in code units, not characters. That is, they are in
2356 bytes for the 8-bit library, 16-bit code units for the 16-bit library,
2357 and 32-bit code units for the 32-bit library, whether or not UTF pro‐
2358 cessing is enabled.
2359 If startoffset is greater than the length of the subject, pcre2_match()
2361 returns PCRE2_ERROR_BADOFFSET. When the starting offset is zero, the
2362 search for a match starts at the beginning of the subject, and this is
2363 by far the most common case. In UTF-8 or UTF-16 mode, the starting off‐
2364 set must point to the start of a character, or to the end of the sub‐
2365 ject (in UTF-32 mode, one code unit equals one character, so all off‐
2366 sets are valid). Like the pattern string, the subject may contain
2367 binary zeros.
2368 A non-zero starting offset is useful when searching for another match
2370 in the same subject by calling pcre2_match() again after a previous
2371 success. Setting startoffset differs from passing over a shortened
2372 string and setting PCRE2_NOTBOL in the case of a pattern that begins
2373 with any kind of lookbehind. For example, consider the pattern
2374 \Biss\B
2376 which finds occurrences of "iss" in the middle of words. (\B matches
2378 only if the current position in the subject is not a word boundary.)
2379 When applied to the string "Mississipi" the first call to pcre2_match()
2380 finds the first occurrence. If pcre2_match() is called again with just
2381 the remainder of the subject, namely "issipi", it does not match,
2382 because \B is always false at the start of the subject, which is deemed
2383 to be a word boundary. However, if pcre2_match() is passed the entire
2384 string again, but with startoffset set to 4, it finds the second occur‐
2385 rence of "iss" because it is able to look behind the starting point to
2386 discover that it is preceded by a letter.
2387 Finding all the matches in a subject is tricky when the pattern can
2389 match an empty string. It is possible to emulate Perl's /g behaviour by
2390 first trying the match again at the same offset, with the
2391 PCRE2_NOTEMPTY_ATSTART and PCRE2_ANCHORED options, and then if that
2392 fails, advancing the starting offset and trying an ordinary match
2393 again. There is some code that demonstrates how to do this in the
2394 pcre2demo sample program. In the most general case, you have to check
2395 to see if the newline convention recognizes CRLF as a newline, and if
2396 so, and the current character is CR followed by LF, advance the start‐
2397 ing offset by two characters instead of one.
2398 If a non-zero starting offset is passed when the pattern is anchored, a
2400 single attempt to match at the given offset is made. This can only suc‐
2401 ceed if the pattern does not require the match to be at the start of
2402 the subject. In other words, the anchoring must be the result of set‐
2403 ting the PCRE2_ANCHORED option or the use of .* with PCRE2_DOTALL, not
2404 by starting the pattern with ^ or \A.
2405 Option bits for pcre2_match()
2407 The unused bits of the options argument for pcre2_match() must be zero.
2409 The only bits that may be set are PCRE2_ANCHORED,
2410 PCRE2_COPY_MATCHED_SUBJECT, PCRE2_ENDANCHORED, PCRE2_NOTBOL,
2411 PCRE2_NOTEOL, PCRE2_NOTEMPTY, PCRE2_NOTEMPTY_ATSTART, PCRE2_NO_JIT,
2412 PCRE2_NO_UTF_CHECK, PCRE2_PARTIAL_HARD, and PCRE2_PARTIAL_SOFT. Their
2413 action is described below.
2414 Setting PCRE2_ANCHORED or PCRE2_ENDANCHORED at match time is not sup‐
2416 ported by the just-in-time (JIT) compiler. If it is set, JIT matching
2417 is disabled and the interpretive code in pcre2_match() is run. Apart
2418 from PCRE2_NO_JIT (obviously), the remaining options are supported for
2419 JIT matching.
2420 PCRE2_ANCHORED
2422 The PCRE2_ANCHORED option limits pcre2_match() to matching at the first
2424 matching position. If a pattern was compiled with PCRE2_ANCHORED, or
2425 turned out to be anchored by virtue of its contents, it cannot be made
2426 unachored at matching time. Note that setting the option at match time
2427 disables JIT matching.
2428 PCRE2_COPY_MATCHED_SUBJECT
2430 By default, a pointer to the subject is remembered in the match data
2432 block so that, after a successful match, it can be referenced by the
2433 substring extraction functions. This means that the subject's memory
2434 must not be freed until all such operations are complete. For some
2435 applications where the lifetime of the subject string is not guaran‐
2436 teed, it may be necessary to make a copy of the subject string, but it
2437 is wasteful to do this unless the match is successful. After a success‐
2438 ful match, if PCRE2_COPY_MATCHED_SUBJECT is set, the subject is copied
2439 and the new pointer is remembered in the match data block instead of
2440 the original subject pointer. The memory allocator that was used for
2441 the match block itself is used. The copy is automatically freed when
2442 pcre2_match_data_free() is called to free the match data block. It is
2443 also automatically freed if the match data block is re-used for another
2444 match operation.
2445 PCRE2_ENDANCHORED
2447 If the PCRE2_ENDANCHORED option is set, any string that pcre2_match()
2449 matches must be right at the end of the subject string. Note that set‐
2450 ting the option at match time disables JIT matching.
2451 PCRE2_NOTBOL
2453 This option specifies that first character of the subject string is not
2455 the beginning of a line, so the circumflex metacharacter should not
2456 match before it. Setting this without having set PCRE2_MULTILINE at
2457 compile time causes circumflex never to match. This option affects only
2458 the behaviour of the circumflex metacharacter. It does not affect \A.
2459 PCRE2_NOTEOL
2461 This option specifies that the end of the subject string is not the end
2463 of a line, so the dollar metacharacter should not match it nor (except
2464 in multiline mode) a newline immediately before it. Setting this with‐
2465 out having set PCRE2_MULTILINE at compile time causes dollar never to
2466 match. This option affects only the behaviour of the dollar metacharac‐
2467 ter. It does not affect \Z or \z.
2468 PCRE2_NOTEMPTY
2470 An empty string is not considered to be a valid match if this option is
2472 set. If there are alternatives in the pattern, they are tried. If all
2473 the alternatives match the empty string, the entire match fails. For
2474 example, if the pattern
2475 a?b?
2477 is applied to a string not beginning with "a" or "b", it matches an
2479 empty string at the start of the subject. With PCRE2_NOTEMPTY set, this
2480 match is not valid, so pcre2_match() searches further into the string
2481 for occurrences of "a" or "b".
2482 PCRE2_NOTEMPTY_ATSTART
2484 This is like PCRE2_NOTEMPTY, except that it locks out an empty string
2486 match only at the first matching position, that is, at the start of the
2487 subject plus the starting offset. An empty string match later in the
2488 subject is permitted. If the pattern is anchored, such a match can
2489 occur only if the pattern contains \K.
2490 PCRE2_NO_JIT
2492 By default, if a pattern has been successfully processed by
2494 pcre2_jit_compile(), JIT is automatically used when pcre2_match() is
2495 called with options that JIT supports. Setting PCRE2_NO_JIT disables
2496 the use of JIT; it forces matching to be done by the interpreter.
2497 PCRE2_NO_UTF_CHECK
2499 When PCRE2_UTF is set at compile time, the validity of the subject as a
2501 UTF string is checked unless PCRE2_NO_UTF_CHECK is passed to
2502 pcre2_match() or PCRE2_MATCH_INVALID_UTF was passed to pcre2_compile().
2503 The latter special case is discussed in detail in the pcre2unicode doc‐
2504 umentation.
2505 In the default case, if a non-zero starting offset is given, the check
2507 is applied only to that part of the subject that could be inspected
2508 during matching, and there is a check that the starting offset points
2509 to the first code unit of a character or to the end of the subject. If
2510 there are no lookbehind assertions in the pattern, the check starts at
2511 the starting offset. Otherwise, it starts at the length of the longest
2512 lookbehind before the starting offset, or at the start of the subject
2513 if there are not that many characters before the starting offset. Note
2514 that the sequences \b and \B are one-character lookbehinds.
2515 The check is carried out before any other processing takes place, and a
2517 negative error code is returned if the check fails. There are several
2518 UTF error codes for each code unit width, corresponding to different
2519 problems with the code unit sequence. There are discussions about the
2520 validity of UTF-8 strings, UTF-16 strings, and UTF-32 strings in the
2521 pcre2unicode documentation.
2522 If you know that your subject is valid, and you want to skip this check
2524 for performance reasons, you can set the PCRE2_NO_UTF_CHECK option when
2525 calling pcre2_match(). You might want to do this for the second and
2526 subsequent calls to pcre2_match() if you are making repeated calls to
2527 find multiple matches in the same subject string.
2528 Warning: Unless PCRE2_MATCH_INVALID_UTF was set at compile time, when
2530 PCRE2_NO_UTF_CHECK is set at match time the effect of passing an
2531 invalid string as a subject, or an invalid value of startoffset, is
2532 undefined. Your program may crash or loop indefinitely or give wrong
2533 results.
2534 PCRE2_PARTIAL_HARD
2536 PCRE2_PARTIAL_SOFT
2537 These options turn on the partial matching feature. A partial match
2539 occurs if the end of the subject string is reached successfully, but
2540 there are not enough subject characters to complete the match. In addi‐
2541 tion, either at least one character must have been inspected or the
2542 pattern must contain a lookbehind, or the pattern must be one that
2543 could match an empty string.
2544 If this situation arises when PCRE2_PARTIAL_SOFT (but not PCRE2_PAR‐
2546 TIAL_HARD) is set, matching continues by testing any remaining alterna‐
2547 tives. Only if no complete match can be found is PCRE2_ERROR_PARTIAL
2548 returned instead of PCRE2_ERROR_NOMATCH. In other words, PCRE2_PAR‐
2549 TIAL_SOFT specifies that the caller is prepared to handle a partial
2550 match, but only if no complete match can be found.
2551 If PCRE2_PARTIAL_HARD is set, it overrides PCRE2_PARTIAL_SOFT. In this
2553 case, if a partial match is found, pcre2_match() immediately returns
2554 PCRE2_ERROR_PARTIAL, without considering any other alternatives. In
2555 other words, when PCRE2_PARTIAL_HARD is set, a partial match is consid‐
2556 ered to be more important that an alternative complete match.
2557 There is a more detailed discussion of partial and multi-segment match‐
2559 ing, with examples, in the pcre2partial documentation.
2560
2562 When PCRE2 is built, a default newline convention is set; this is usu‐
2564 ally the standard convention for the operating system. The default can
2565 be overridden in a compile context by calling pcre2_set_newline(). It
2566 can also be overridden by starting a pattern string with, for example,
2567 (*CRLF), as described in the section on newline conventions in the
2568 pcre2pattern page. During matching, the newline choice affects the be‐
2569 haviour of the dot, circumflex, and dollar metacharacters. It may also
2570 alter the way the match starting position is advanced after a match
2571 failure for an unanchored pattern.
2572 When PCRE2_NEWLINE_CRLF, PCRE2_NEWLINE_ANYCRLF, or PCRE2_NEWLINE_ANY is
2574 set as the newline convention, and a match attempt for an unanchored
2575 pattern fails when the current starting position is at a CRLF sequence,
2576 and the pattern contains no explicit matches for CR or LF characters,
2577 the match position is advanced by two characters instead of one, in
2578 other words, to after the CRLF.
2579 The above rule is a compromise that makes the most common cases work as
2581 expected. For example, if the pattern is .+A (and the PCRE2_DOTALL
2582 option is not set), it does not match the string "\r\nA" because, after
2583 failing at the start, it skips both the CR and the LF before retrying.
2584 However, the pattern [\r\n]A does match that string, because it con‐
2585 tains an explicit CR or LF reference, and so advances only by one char‐
2586 acter after the first failure.
2587 An explicit match for CR of LF is either a literal appearance of one of
2589 those characters in the pattern, or one of the \r or \n or equivalent
2590 octal or hexadecimal escape sequences. Implicit matches such as [^X] do
2591 not count, nor does \s, even though it includes CR and LF in the char‐
2592 acters that it matches.
2593 Notwithstanding the above, anomalous effects may still occur when CRLF
2595 is a valid newline sequence and explicit \r or \n escapes appear in the
2596 pattern.
2597
2599 uint32_t pcre2_get_ovector_count(pcre2_match_data *match_data);
2601 PCRE2_SIZE *pcre2_get_ovector_pointer(pcre2_match_data *match_data);
2603 In general, a pattern matches a certain portion of the subject, and in
2605 addition, further substrings from the subject may be picked out by
2606 parenthesized parts of the pattern. Following the usage in Jeffrey
2607 Friedl's book, this is called "capturing" in what follows, and the
2608 phrase "capture group" (Perl terminology) is used for a fragment of a
2609 pattern that picks out a substring. PCRE2 supports several other kinds
2610 of parenthesized group that do not cause substrings to be captured. The
2611 pcre2_pattern_info() function can be used to find out how many capture
2612 groups there are in a compiled pattern.
2613 You can use auxiliary functions for accessing captured substrings by
2615 number or by name, as described in sections below.
2616 Alternatively, you can make direct use of the vector of PCRE2_SIZE val‐
2618 ues, called the ovector, which contains the offsets of captured
2619 strings. It is part of the match data block. The function
2620 pcre2_get_ovector_pointer() returns the address of the ovector, and
2621 pcre2_get_ovector_count() returns the number of pairs of values it con‐
2622 tains.
2623 Within the ovector, the first in each pair of values is set to the off‐
2625 set of the first code unit of a substring, and the second is set to the
2626 offset of the first code unit after the end of a substring. These val‐
2627 ues are always code unit offsets, not character offsets. That is, they
2628 are byte offsets in the 8-bit library, 16-bit offsets in the 16-bit
2629 library, and 32-bit offsets in the 32-bit library.
2630 After a partial match (error return PCRE2_ERROR_PARTIAL), only the
2632 first pair of offsets (that is, ovector[0] and ovector[1]) are set.
2633 They identify the part of the subject that was partially matched. See
2634 the pcre2partial documentation for details of partial matching.
2635 After a fully successful match, the first pair of offsets identifies
2637 the portion of the subject string that was matched by the entire pat‐
2638 tern. The next pair is used for the first captured substring, and so
2639 on. The value returned by pcre2_match() is one more than the highest
2640 numbered pair that has been set. For example, if two substrings have
2641 been captured, the returned value is 3. If there are no captured sub‐
2642 strings, the return value from a successful match is 1, indicating that
2643 just the first pair of offsets has been set.
2644 If a pattern uses the \K escape sequence within a positive assertion,
2646 the reported start of a successful match can be greater than the end of
2647 the match. For example, if the pattern (?=ab\K) is matched against
2648 "ab", the start and end offset values for the match are 2 and 0.
2649 If a capture group is matched repeatedly within a single match opera‐
2651 tion, it is the last portion of the subject that it matched that is
2652 returned.
2653 If the ovector is too small to hold all the captured substring offsets,
2655 as much as possible is filled in, and the function returns a value of
2656 zero. If captured substrings are not of interest, pcre2_match() may be
2657 called with a match data block whose ovector is of minimum length (that
2658 is, one pair).
2659 It is possible for capture group number n+1 to match some part of the
2661 subject when group n has not been used at all. For example, if the
2662 string "abc" is matched against the pattern (a|(z))(bc) the return from
2663 the function is 4, and groups 1 and 3 are matched, but 2 is not. When
2664 this happens, both values in the offset pairs corresponding to unused
2665 groups are set to PCRE2_UNSET.
2666 Offset values that correspond to unused groups at the end of the
2668 expression are also set to PCRE2_UNSET. For example, if the string
2669 "abc" is matched against the pattern (abc)(x(yz)?)? groups 2 and 3 are
2670 not matched. The return from the function is 2, because the highest
2671 used capture group number is 1. The offsets for for the second and
2672 third capture groupss (assuming the vector is large enough, of course)
2673 are set to PCRE2_UNSET.
2674 Elements in the ovector that do not correspond to capturing parentheses
2676 in the pattern are never changed. That is, if a pattern contains n cap‐
2677 turing parentheses, no more than ovector[0] to ovector[2n+1] are set by
2678 pcre2_match(). The other elements retain whatever values they previ‐
2679 ously had. After a failed match attempt, the contents of the ovector
2680 are unchanged.
2681
2683 PCRE2_SPTR pcre2_get_mark(pcre2_match_data *match_data);
2685 PCRE2_SIZE pcre2_get_startchar(pcre2_match_data *match_data);
2687 As well as the offsets in the ovector, other information about a match
2689 is retained in the match data block and can be retrieved by the above
2690 functions in appropriate circumstances. If they are called at other
2691 times, the result is undefined.
2692 After a successful match, a partial match (PCRE2_ERROR_PARTIAL), or a
2694 failure to match (PCRE2_ERROR_NOMATCH), a mark name may be available.
2695 The function pcre2_get_mark() can be called to access this name, which
2696 can be specified in the pattern by any of the backtracking control
2697 verbs, not just (*MARK). The same function applies to all the verbs. It
2698 returns a pointer to the zero-terminated name, which is within the com‐
2699 piled pattern. If no name is available, NULL is returned. The length of
2700 the name (excluding the terminating zero) is stored in the code unit
2701 that precedes the name. You should use this length instead of relying
2702 on the terminating zero if the name might contain a binary zero.
2703 After a successful match, the name that is returned is the last mark
2705 name encountered on the matching path through the pattern. Instances of
2706 backtracking verbs without names do not count. Thus, for example, if
2707 the matching path contains (*MARK:A)(*PRUNE), the name "A" is returned.
2708 After a "no match" or a partial match, the last encountered name is
2709 returned. For example, consider this pattern:
2710 ^(*MARK:A)((*MARK:B)a|b)c
2712 When it matches "bc", the returned name is A. The B mark is "seen" in
2714 the first branch of the group, but it is not on the matching path. On
2715 the other hand, when this pattern fails to match "bx", the returned
2716 name is B.
2717 Warning: By default, certain start-of-match optimizations are used to
2719 give a fast "no match" result in some situations. For example, if the
2720 anchoring is removed from the pattern above, there is an initial check
2721 for the presence of "c" in the subject before running the matching
2722 engine. This check fails for "bx", causing a match failure without see‐
2723 ing any marks. You can disable the start-of-match optimizations by set‐
2724 ting the PCRE2_NO_START_OPTIMIZE option for pcre2_compile() or by
2725 starting the pattern with (*NO_START_OPT).
2726 After a successful match, a partial match, or one of the invalid UTF
2728 errors (for example, PCRE2_ERROR_UTF8_ERR5), pcre2_get_startchar() can
2729 be called. After a successful or partial match it returns the code unit
2730 offset of the character at which the match started. For a non-partial
2731 match, this can be different to the value of ovector[0] if the pattern
2732 contains the \K escape sequence. After a partial match, however, this
2733 value is always the same as ovector[0] because \K does not affect the
2734 result of a partial match.
2735 After a UTF check failure, pcre2_get_startchar() can be used to obtain
2737 the code unit offset of the invalid UTF character. Details are given in
2738 the pcre2unicode page.
2739
2741 If pcre2_match() fails, it returns a negative number. This can be con‐
2743 verted to a text string by calling the pcre2_get_error_message() func‐
2744 tion (see "Obtaining a textual error message" below). Negative error
2745 codes are also returned by other functions, and are documented with
2746 them. The codes are given names in the header file. If UTF checking is
2747 in force and an invalid UTF subject string is detected, one of a number
2748 of UTF-specific negative error codes is returned. Details are given in
2749 the pcre2unicode page. The following are the other errors that may be
2750 returned by pcre2_match():
2751 PCRE2_ERROR_NOMATCH
2753 The subject string did not match the pattern.
2755 PCRE2_ERROR_PARTIAL
2757 The subject string did not match, but it did match partially. See the
2759 pcre2partial documentation for details of partial matching.
2760 PCRE2_ERROR_BADMAGIC
2762 PCRE2 stores a 4-byte "magic number" at the start of the compiled code,
2764 to catch the case when it is passed a junk pointer. This is the error
2765 that is returned when the magic number is not present.
2766 PCRE2_ERROR_BADMODE
2768 This error is given when a compiled pattern is passed to a function in
2770 a library of a different code unit width, for example, a pattern com‐
2771 piled by the 8-bit library is passed to a 16-bit or 32-bit library
2772 function.
2773 PCRE2_ERROR_BADOFFSET
2775 The value of startoffset was greater than the length of the subject.
2777 PCRE2_ERROR_BADOPTION
2779 An unrecognized bit was set in the options argument.
2781 PCRE2_ERROR_BADUTFOFFSET
2783 The UTF code unit sequence that was passed as a subject was checked and
2785 found to be valid (the PCRE2_NO_UTF_CHECK option was not set), but the
2786 value of startoffset did not point to the beginning of a UTF character
2787 or the end of the subject.
2788 PCRE2_ERROR_CALLOUT
2790 This error is never generated by pcre2_match() itself. It is provided
2792 for use by callout functions that want to cause pcre2_match() or
2793 pcre2_callout_enumerate() to return a distinctive error code. See the
2794 pcre2callout documentation for details.
2795 PCRE2_ERROR_DEPTHLIMIT
2797 The nested backtracking depth limit was reached.
2799 PCRE2_ERROR_HEAPLIMIT
2801 The heap limit was reached.
2803 PCRE2_ERROR_INTERNAL
2805 An unexpected internal error has occurred. This error could be caused
2807 by a bug in PCRE2 or by overwriting of the compiled pattern.
2808 PCRE2_ERROR_JIT_STACKLIMIT
2810 This error is returned when a pattern that was successfully studied
2812 using JIT is being matched, but the memory available for the just-in-
2813 time processing stack is not large enough. See the pcre2jit documenta‐
2814 tion for more details.
2815 PCRE2_ERROR_MATCHLIMIT
2817 The backtracking match limit was reached.
2819 PCRE2_ERROR_NOMEMORY
2821 If a pattern contains many nested backtracking points, heap memory is
2823 used to remember them. This error is given when the memory allocation
2824 function (default or custom) fails. Note that a different error,
2825 PCRE2_ERROR_HEAPLIMIT, is given if the amount of memory needed exceeds
2826 the heap limit. PCRE2_ERROR_NOMEMORY is also returned if
2827 PCRE2_COPY_MATCHED_SUBJECT is set and memory allocation fails.
2828 PCRE2_ERROR_NULL
2830 Either the code, subject, or match_data argument was passed as NULL.
2832 PCRE2_ERROR_RECURSELOOP
2834 This error is returned when pcre2_match() detects a recursion loop
2836 within the pattern. Specifically, it means that either the whole pat‐
2837 tern or a capture group has been called recursively for the second time
2838 at the same position in the subject string. Some simple patterns that
2839 might do this are detected and faulted at compile time, but more com‐
2840 plicated cases, in particular mutual recursions between two different
2841 groups, cannot be detected until matching is attempted.
2842
2844 int pcre2_get_error_message(int errorcode, PCRE2_UCHAR *buffer,
2846 PCRE2_SIZE bufflen);
2847 A text message for an error code from any PCRE2 function (compile,
2849 match, or auxiliary) can be obtained by calling pcre2_get_error_mes‐
2850 sage(). The code is passed as the first argument, with the remaining
2851 two arguments specifying a code unit buffer and its length in code
2852 units, into which the text message is placed. The message is returned
2853 in code units of the appropriate width for the library that is being
2854 used.
2855 The returned message is terminated with a trailing zero, and the func‐
2857 tion returns the number of code units used, excluding the trailing
2858 zero. If the error number is unknown, the negative error code
2859 PCRE2_ERROR_BADDATA is returned. If the buffer is too small, the mes‐
2860 sage is truncated (but still with a trailing zero), and the negative
2861 error code PCRE2_ERROR_NOMEMORY is returned. None of the messages are
2862 very long; a buffer size of 120 code units is ample.
2863
2865 int pcre2_substring_length_bynumber(pcre2_match_data *match_data,
2867 uint32_t number, PCRE2_SIZE *length);
2868 int pcre2_substring_copy_bynumber(pcre2_match_data *match_data,
2870 uint32_t number, PCRE2_UCHAR *buffer,
2871 PCRE2_SIZE *bufflen);
2872 int pcre2_substring_get_bynumber(pcre2_match_data *match_data,
2874 uint32_t number, PCRE2_UCHAR **bufferptr,
2875 PCRE2_SIZE *bufflen);
2876 void pcre2_substring_free(PCRE2_UCHAR *buffer);
2878 Captured substrings can be accessed directly by using the ovector as
2880 described above. For convenience, auxiliary functions are provided for
2881 extracting captured substrings as new, separate, zero-terminated
2882 strings. A substring that contains a binary zero is correctly extracted
2883 and has a further zero added on the end, but the result is not, of
2884 course, a C string.
2885 The functions in this section identify substrings by number. The number
2887 zero refers to the entire matched substring, with higher numbers refer‐
2888 ring to substrings captured by parenthesized groups. After a partial
2889 match, only substring zero is available. An attempt to extract any
2890 other substring gives the error PCRE2_ERROR_PARTIAL. The next section
2891 describes similar functions for extracting captured substrings by name.
2892 If a pattern uses the \K escape sequence within a positive assertion,
2894 the reported start of a successful match can be greater than the end of
2895 the match. For example, if the pattern (?=ab\K) is matched against
2896 "ab", the start and end offset values for the match are 2 and 0. In
2897 this situation, calling these functions with a zero substring number
2898 extracts a zero-length empty string.
2899 You can find the length in code units of a captured substring without
2901 extracting it by calling pcre2_substring_length_bynumber(). The first
2902 argument is a pointer to the match data block, the second is the group
2903 number, and the third is a pointer to a variable into which the length
2904 is placed. If you just want to know whether or not the substring has
2905 been captured, you can pass the third argument as NULL.
2906 The pcre2_substring_copy_bynumber() function copies a captured sub‐
2908 string into a supplied buffer, whereas pcre2_substring_get_bynumber()
2909 copies it into new memory, obtained using the same memory allocation
2910 function that was used for the match data block. The first two argu‐
2911 ments of these functions are a pointer to the match data block and a
2912 capture group number.
2913 The final arguments of pcre2_substring_copy_bynumber() are a pointer to
2915 the buffer and a pointer to a variable that contains its length in code
2916 units. This is updated to contain the actual number of code units used
2917 for the extracted substring, excluding the terminating zero.
2918 For pcre2_substring_get_bynumber() the third and fourth arguments point
2920 to variables that are updated with a pointer to the new memory and the
2921 number of code units that comprise the substring, again excluding the
2922 terminating zero. When the substring is no longer needed, the memory
2923 should be freed by calling pcre2_substring_free().
2924 The return value from all these functions is zero for success, or a
2926 negative error code. If the pattern match failed, the match failure
2927 code is returned. If a substring number greater than zero is used
2928 after a partial match, PCRE2_ERROR_PARTIAL is returned. Other possible
2929 error codes are:
2930 PCRE2_ERROR_NOMEMORY
2932 The buffer was too small for pcre2_substring_copy_bynumber(), or the
2934 attempt to get memory failed for pcre2_substring_get_bynumber().
2935 PCRE2_ERROR_NOSUBSTRING
2937 There is no substring with that number in the pattern, that is, the
2939 number is greater than the number of capturing parentheses.
2940 PCRE2_ERROR_UNAVAILABLE
2942 The substring number, though not greater than the number of captures in
2944 the pattern, is greater than the number of slots in the ovector, so the
2945 substring could not be captured.
2946 PCRE2_ERROR_UNSET
2948 The substring did not participate in the match. For example, if the
2950 pattern is (abc)|(def) and the subject is "def", and the ovector con‐
2951 tains at least two capturing slots, substring number 1 is unset.
2952
2954 int pcre2_substring_list_get(pcre2_match_data *match_data,
2956 PCRE2_UCHAR ***listptr, PCRE2_SIZE **lengthsptr);
2957 void pcre2_substring_list_free(PCRE2_SPTR *list);
2959 The pcre2_substring_list_get() function extracts all available sub‐
2961 strings and builds a list of pointers to them. It also (optionally)
2962 builds a second list that contains their lengths (in code units),
2963 excluding a terminating zero that is added to each of them. All this is
2964 done in a single block of memory that is obtained using the same memory
2965 allocation function that was used to get the match data block.
2966 This function must be called only after a successful match. If called
2968 after a partial match, the error code PCRE2_ERROR_PARTIAL is returned.
2969 The address of the memory block is returned via listptr, which is also
2971 the start of the list of string pointers. The end of the list is marked
2972 by a NULL pointer. The address of the list of lengths is returned via
2973 lengthsptr. If your strings do not contain binary zeros and you do not
2974 therefore need the lengths, you may supply NULL as the lengthsptr argu‐
2975 ment to disable the creation of a list of lengths. The yield of the
2976 function is zero if all went well, or PCRE2_ERROR_NOMEMORY if the mem‐
2977 ory block could not be obtained. When the list is no longer needed, it
2978 should be freed by calling pcre2_substring_list_free().
2979 If this function encounters a substring that is unset, which can happen
2981 when capture group number n+1 matches some part of the subject, but
2982 group n has not been used at all, it returns an empty string. This can
2983 be distinguished from a genuine zero-length substring by inspecting the
2984 appropriate offset in the ovector, which contain PCRE2_UNSET for unset
2985 substrings, or by calling pcre2_substring_length_bynumber().
2986
2988 int pcre2_substring_number_from_name(const pcre2_code *code,
2990 PCRE2_SPTR name);
2991 int pcre2_substring_length_byname(pcre2_match_data *match_data,
2993 PCRE2_SPTR name, PCRE2_SIZE *length);
2994 int pcre2_substring_copy_byname(pcre2_match_data *match_data,
2996 PCRE2_SPTR name, PCRE2_UCHAR *buffer, PCRE2_SIZE *bufflen);
2997 int pcre2_substring_get_byname(pcre2_match_data *match_data,
2999 PCRE2_SPTR name, PCRE2_UCHAR **bufferptr, PCRE2_SIZE *bufflen);
3000 void pcre2_substring_free(PCRE2_UCHAR *buffer);
3002 To extract a substring by name, you first have to find associated num‐
3004 ber. For example, for this pattern:
3005 (a+)b(?<xxx>\d+)...
3007 the number of the capture group called "xxx" is 2. If the name is known
3009 to be unique (PCRE2_DUPNAMES was not set), you can find the number from
3010 the name by calling pcre2_substring_number_from_name(). The first argu‐
3011 ment is the compiled pattern, and the second is the name. The yield of
3012 the function is the group number, PCRE2_ERROR_NOSUBSTRING if there is
3013 no group with that name, or PCRE2_ERROR_NOUNIQUESUBSTRING if there is
3014 more than one group with that name. Given the number, you can extract
3015 the substring directly from the ovector, or use one of the "bynumber"
3016 functions described above.
3017 For convenience, there are also "byname" functions that correspond to
3019 the "bynumber" functions, the only difference being that the second
3020 argument is a name instead of a number. If PCRE2_DUPNAMES is set and
3021 there are duplicate names, these functions scan all the groups with the
3022 given name, and return the captured substring from the first named
3023 group that is set.
3024 If there are no groups with the given name, PCRE2_ERROR_NOSUBSTRING is
3026 returned. If all groups with the name have numbers that are greater
3027 than the number of slots in the ovector, PCRE2_ERROR_UNAVAILABLE is
3028 returned. If there is at least one group with a slot in the ovector,
3029 but no group is found to be set, PCRE2_ERROR_UNSET is returned.
3030 Warning: If the pattern uses the (?| feature to set up multiple capture
3032 groups with the same number, as described in the section on duplicate
3033 group numbers in the pcre2pattern page, you cannot use names to distin‐
3034 guish the different capture groups, because names are not included in
3035 the compiled code. The matching process uses only numbers. For this
3036 reason, the use of different names for groups with the same number
3037 causes an error at compile time.
3038
3040 int pcre2_substitute(const pcre2_code *code, PCRE2_SPTR subject,
3042 PCRE2_SIZE length, PCRE2_SIZE startoffset,
3043 uint32_t options, pcre2_match_data *match_data,
3044 pcre2_match_context *mcontext, PCRE2_SPTR replacement,
3045 PCRE2_SIZE rlength, PCRE2_UCHAR *outputbuffer,
3046 PCRE2_SIZE *outlengthptr);
3047 This function optionally calls pcre2_match() and then makes a copy of
3049 the subject string in outputbuffer, replacing parts that were matched
3050 with the replacement string, whose length is supplied in rlength. This
3051 can be given as PCRE2_ZERO_TERMINATED for a zero-terminated string.
3052 There is an option (see PCRE2_SUBSTITUTE_REPLACEMENT_ONLY below) to
3053 return just the replacement string(s). The default action is to perform
3054 just one replacement if the pattern matches, but there is an option
3055 that requests multiple replacements (see PCRE2_SUBSTITUTE_GLOBAL
3056 below).
3057 If successful, pcre2_substitute() returns the number of substitutions
3059 that were carried out. This may be zero if no match was found, and is
3060 never greater than one unless PCRE2_SUBSTITUTE_GLOBAL is set. A nega‐
3061 tive value is returned if an error is detected.
3062 Matches in which a \K item in a lookahead in the pattern causes the
3064 match to end before it starts are not supported, and give rise to an
3065 error return. For global replacements, matches in which \K in a lookbe‐
3066 hind causes the match to start earlier than the point that was reached
3067 in the previous iteration are also not supported.
3068 The first seven arguments of pcre2_substitute() are the same as for
3070 pcre2_match(), except that the partial matching options are not permit‐
3071 ted, and match_data may be passed as NULL, in which case a match data
3072 block is obtained and freed within this function, using memory manage‐
3073 ment functions from the match context, if provided, or else those that
3074 were used to allocate memory for the compiled code.
3075 If match_data is not NULL and PCRE2_SUBSTITUTE_MATCHED is not set, the
3077 provided block is used for all calls to pcre2_match(), and its contents
3078 afterwards are the result of the final call. For global changes, this
3079 will always be a no-match error. The contents of the ovector within the
3080 match data block may or may not have been changed.
3081 As well as the usual options for pcre2_match(), a number of additional
3083 options can be set in the options argument of pcre2_substitute(). One
3084 such option is PCRE2_SUBSTITUTE_MATCHED. When this is set, an external
3085 match_data block must be provided, and it must have been used for an
3086 external call to pcre2_match(). The data in the match_data block
3087 (return code, offset vector) is used for the first substitution instead
3088 of calling pcre2_match() from within pcre2_substitute(). This allows an
3089 application to check for a match before choosing to substitute, without
3090 having to repeat the match.
3091 The contents of the externally supplied match data block are not
3093 changed when PCRE2_SUBSTITUTE_MATCHED is set. If PCRE2_SUBSTI‐
3094 TUTE_GLOBAL is also set, pcre2_match() is called after the first sub‐
3095 stitution to check for further matches, but this is done using an
3096 internally obtained match data block, thus always leaving the external
3097 block unchanged.
3098 The code argument is not used for matching before the first substitu‐
3100 tion when PCRE2_SUBSTITUTE_MATCHED is set, but it must be provided,
3101 even when PCRE2_SUBSTITUTE_GLOBAL is not set, because it contains
3102 information such as the UTF setting and the number of capturing paren‐
3103 theses in the pattern.
3104 The default action of pcre2_substitute() is to return a copy of the
3106 subject string with matched substrings replaced. However, if PCRE2_SUB‐
3107 STITUTE_REPLACEMENT_ONLY is set, only the replacement substrings are
3108 returned. In the global case, multiple replacements are concatenated in
3109 the output buffer. Substitution callouts (see below) can be used to
3110 separate them if necessary.
3111 The outlengthptr argument of pcre2_substitute() must point to a vari‐
3113 able that contains the length, in code units, of the output buffer. If
3114 the function is successful, the value is updated to contain the length
3115 in code units of the new string, excluding the trailing zero that is
3116 automatically added.
3117 If the function is not successful, the value set via outlengthptr
3119 depends on the type of error. For syntax errors in the replacement
3120 string, the value is the offset in the replacement string where the
3121 error was detected. For other errors, the value is PCRE2_UNSET by
3122 default. This includes the case of the output buffer being too small,
3123 unless PCRE2_SUBSTITUTE_OVERFLOW_LENGTH is set.
3124 PCRE2_SUBSTITUTE_OVERFLOW_LENGTH changes what happens when the output
3126 buffer is too small. The default action is to return PCRE2_ERROR_NOMEM‐
3127 ORY immediately. If this option is set, however, pcre2_substitute()
3128 continues to go through the motions of matching and substituting (with‐
3129 out, of course, writing anything) in order to compute the size of buf‐
3130 fer that is needed. This value is passed back via the outlengthptr
3131 variable, with the result of the function still being
3132 PCRE2_ERROR_NOMEMORY.
3133 Passing a buffer size of zero is a permitted way of finding out how
3135 much memory is needed for given substitution. However, this does mean
3136 that the entire operation is carried out twice. Depending on the appli‐
3137 cation, it may be more efficient to allocate a large buffer and free
3138 the excess afterwards, instead of using PCRE2_SUBSTITUTE_OVER‐
3139 FLOW_LENGTH.
3140 The replacement string, which is interpreted as a UTF string in UTF
3142 mode, is checked for UTF validity unless PCRE2_NO_UTF_CHECK is set. An
3143 invalid UTF replacement string causes an immediate return with the rel‐
3144 evant UTF error code.
3145 If PCRE2_SUBSTITUTE_LITERAL is set, the replacement string is not
3147 interpreted in any way. By default, however, a dollar character is an
3148 escape character that can specify the insertion of characters from cap‐
3149 ture groups and names from (*MARK) or other control verbs in the pat‐
3150 tern. The following forms are always recognized:
3151 $$ insert a dollar character
3153 $<n> or ${<n>} insert the contents of group <n>
3154 $*MARK or ${*MARK} insert a control verb name
3155 Either a group number or a group name can be given for <n>. Curly
3157 brackets are required only if the following character would be inter‐
3158 preted as part of the number or name. The number may be zero to include
3159 the entire matched string. For example, if the pattern a(b)c is
3160 matched with "=abc=" and the replacement string "+$1$0$1+", the result
3161 is "=+babcb+=".
3162 $*MARK inserts the name from the last encountered backtracking control
3164 verb on the matching path that has a name. (*MARK) must always include
3165 a name, but the other verbs need not. For example, in the case of
3166 (*MARK:A)(*PRUNE) the name inserted is "A", but for (*MARK:A)(*PRUNE:B)
3167 the relevant name is "B". This facility can be used to perform simple
3168 simultaneous substitutions, as this pcre2test example shows:
3169 /(*MARK:pear)apple|(*MARK:orange)lemon/g,replace=${*MARK}
3171 apple lemon
3172 2: pear orange
3173 PCRE2_SUBSTITUTE_GLOBAL causes the function to iterate over the subject
3175 string, replacing every matching substring. If this option is not set,
3176 only the first matching substring is replaced. The search for matches
3177 takes place in the original subject string (that is, previous replace‐
3178 ments do not affect it). Iteration is implemented by advancing the
3179 startoffset value for each search, which is always passed the entire
3180 subject string. If an offset limit is set in the match context, search‐
3181 ing stops when that limit is reached.
3182 You can restrict the effect of a global substitution to a portion of
3184 the subject string by setting either or both of startoffset and an off‐
3185 set limit. Here is a pcre2test example:
3186 /B/g,replace=!,use_offset_limit
3188 ABC ABC ABC ABC\=offset=3,offset_limit=12
3189 2: ABC A!C A!C ABC
3190 When continuing with global substitutions after matching a substring
3192 with zero length, an attempt to find a non-empty match at the same off‐
3193 set is performed. If this is not successful, the offset is advanced by
3194 one character except when CRLF is a valid newline sequence and the next
3195 two characters are CR, LF. In this case, the offset is advanced by two
3196 characters.
3197 PCRE2_SUBSTITUTE_UNKNOWN_UNSET causes references to capture groups that
3199 do not appear in the pattern to be treated as unset groups. This option
3200 should be used with care, because it means that a typo in a group name
3201 or number no longer causes the PCRE2_ERROR_NOSUBSTRING error.
3202 PCRE2_SUBSTITUTE_UNSET_EMPTY causes unset capture groups (including
3204 unknown groups when PCRE2_SUBSTITUTE_UNKNOWN_UNSET is set) to be
3205 treated as empty strings when inserted as described above. If this
3206 option is not set, an attempt to insert an unset group causes the
3207 PCRE2_ERROR_UNSET error. This option does not influence the extended
3208 substitution syntax described below.
3209 PCRE2_SUBSTITUTE_EXTENDED causes extra processing to be applied to the
3211 replacement string. Without this option, only the dollar character is
3212 special, and only the group insertion forms listed above are valid.
3213 When PCRE2_SUBSTITUTE_EXTENDED is set, two things change:
3214 Firstly, backslash in a replacement string is interpreted as an escape
3216 character. The usual forms such as \n or \x{ddd} can be used to specify
3217 particular character codes, and backslash followed by any non-alphanu‐
3218 meric character quotes that character. Extended quoting can be coded
3219 using \Q...\E, exactly as in pattern strings.
3220 There are also four escape sequences for forcing the case of inserted
3222 letters. The insertion mechanism has three states: no case forcing,
3223 force upper case, and force lower case. The escape sequences change the
3224 current state: \U and \L change to upper or lower case forcing, respec‐
3225 tively, and \E (when not terminating a \Q quoted sequence) reverts to
3226 no case forcing. The sequences \u and \l force the next character (if
3227 it is a letter) to upper or lower case, respectively, and then the
3228 state automatically reverts to no case forcing. Case forcing applies to
3229 all inserted characters, including those from capture groups and let‐
3230 ters within \Q...\E quoted sequences. If either PCRE2_UTF or PCRE2_UCP
3231 was set when the pattern was compiled, Unicode properties are used for
3232 case forcing characters whose code points are greater than 127.
3233 Note that case forcing sequences such as \U...\E do not nest. For exam‐
3235 ple, the result of processing "\Uaa\LBB\Ecc\E" is "AAbbcc"; the final
3236 \E has no effect. Note also that the PCRE2_ALT_BSUX and
3237 PCRE2_EXTRA_ALT_BSUX options do not apply to replacement strings.
3238 The second effect of setting PCRE2_SUBSTITUTE_EXTENDED is to add more
3240 flexibility to capture group substitution. The syntax is similar to
3241 that used by Bash:
3242 ${<n>:-<string>}
3244 ${<n>:+<string1>:<string2>}
3245 As before, <n> may be a group number or a name. The first form speci‐
3247 fies a default value. If group <n> is set, its value is inserted; if
3248 not, <string> is expanded and the result inserted. The second form
3249 specifies strings that are expanded and inserted when group <n> is set
3250 or unset, respectively. The first form is just a convenient shorthand
3251 for
3252 ${<n>:+${<n>}:<string>}
3254 Backslash can be used to escape colons and closing curly brackets in
3256 the replacement strings. A change of the case forcing state within a
3257 replacement string remains in force afterwards, as shown in this
3258 pcre2test example:
3259 /(some)?(body)/substitute_extended,replace=${1:+\U:\L}HeLLo
3261 body
3262 1: hello
3263 somebody
3264 1: HELLO
3265 The PCRE2_SUBSTITUTE_UNSET_EMPTY option does not affect these extended
3267 substitutions. However, PCRE2_SUBSTITUTE_UNKNOWN_UNSET does cause
3268 unknown groups in the extended syntax forms to be treated as unset.
3269 If PCRE2_SUBSTITUTE_LITERAL is set, PCRE2_SUBSTITUTE_UNKNOWN_UNSET,
3271 PCRE2_SUBSTITUTE_UNSET_EMPTY, and PCRE2_SUBSTITUTE_EXTENDED are irrele‐
3272 vant and are ignored.
3273 Substitution errors
3275 In the event of an error, pcre2_substitute() returns a negative error
3277 code. Except for PCRE2_ERROR_NOMATCH (which is never returned), errors
3278 from pcre2_match() are passed straight back.
3279 PCRE2_ERROR_NOSUBSTRING is returned for a non-existent substring inser‐
3281 tion, unless PCRE2_SUBSTITUTE_UNKNOWN_UNSET is set.
3282 PCRE2_ERROR_UNSET is returned for an unset substring insertion (includ‐
3284 ing an unknown substring when PCRE2_SUBSTITUTE_UNKNOWN_UNSET is set)
3285 when the simple (non-extended) syntax is used and PCRE2_SUBSTI‐
3286 TUTE_UNSET_EMPTY is not set.
3287 PCRE2_ERROR_NOMEMORY is returned if the output buffer is not big
3289 enough. If the PCRE2_SUBSTITUTE_OVERFLOW_LENGTH option is set, the size
3290 of buffer that is needed is returned via outlengthptr. Note that this
3291 does not happen by default.
3292 PCRE2_ERROR_NULL is returned if PCRE2_SUBSTITUTE_MATCHED is set but the
3294 match_data argument is NULL.
3295 PCRE2_ERROR_BADREPLACEMENT is used for miscellaneous syntax errors in
3297 the replacement string, with more particular errors being
3298 PCRE2_ERROR_BADREPESCAPE (invalid escape sequence), PCRE2_ERROR_REP‐
3299 MISSINGBRACE (closing curly bracket not found), PCRE2_ERROR_BADSUBSTI‐
3300 TUTION (syntax error in extended group substitution), and
3301 PCRE2_ERROR_BADSUBSPATTERN (the pattern match ended before it started
3302 or the match started earlier than the current position in the subject,
3303 which can happen if \K is used in an assertion).
3304 As for all PCRE2 errors, a text message that describes the error can be
3306 obtained by calling the pcre2_get_error_message() function (see
3307 "Obtaining a textual error message" above).
3308 Substitution callouts
3310 int pcre2_set_substitute_callout(pcre2_match_context *mcontext,
3312 int (*callout_function)(pcre2_substitute_callout_block *, void *),
3313 void *callout_data);
3314 The pcre2_set_substitution_callout() function can be used to specify a
3316 callout function for pcre2_substitute(). This information is passed in
3317 a match context. The callout function is called after each substitution
3318 has been processed, but it can cause the replacement not to happen. The
3319 callout function is not called for simulated substitutions that happen
3320 as a result of the PCRE2_SUBSTITUTE_OVERFLOW_LENGTH option.
3321 The first argument of the callout function is a pointer to a substitute
3323 callout block structure, which contains the following fields, not nec‐
3324 essarily in this order:
3325 uint32_t version;
3327 uint32_t subscount;
3328 PCRE2_SPTR input;
3329 PCRE2_SPTR output;
3330 PCRE2_SIZE *ovector;
3331 uint32_t oveccount;
3332 PCRE2_SIZE output_offsets[2];
3333 The version field contains the version number of the block format. The
3335 current version is 0. The version number will increase in future if
3336 more fields are added, but the intention is never to remove any of the
3337 existing fields.
3338 The subscount field is the number of the current match. It is 1 for the
3340 first callout, 2 for the second, and so on. The input and output point‐
3341 ers are copies of the values passed to pcre2_substitute().
3342 The ovector field points to the ovector, which contains the result of
3344 the most recent match. The oveccount field contains the number of pairs
3345 that are set in the ovector, and is always greater than zero.
3346 The output_offsets vector contains the offsets of the replacement in
3348 the output string. This has already been processed for dollar and (if
3349 requested) backslash substitutions as described above.
3350 The second argument of the callout function is the value passed as
3352 callout_data when the function was registered. The value returned by
3353 the callout function is interpreted as follows:
3354 If the value is zero, the replacement is accepted, and, if PCRE2_SUB‐
3356 STITUTE_GLOBAL is set, processing continues with a search for the next
3357 match. If the value is not zero, the current replacement is not
3358 accepted. If the value is greater than zero, processing continues when
3359 PCRE2_SUBSTITUTE_GLOBAL is set. Otherwise (the value is less than zero
3360 or PCRE2_SUBSTITUTE_GLOBAL is not set), the the rest of the input is
3361 copied to the output and the call to pcre2_substitute() exits, return‐
3362 ing the number of matches so far.
3363
3365 int pcre2_substring_nametable_scan(const pcre2_code *code,
3367 PCRE2_SPTR name, PCRE2_SPTR *first, PCRE2_SPTR *last);
3368 When a pattern is compiled with the PCRE2_DUPNAMES option, names for
3370 capture groups are not required to be unique. Duplicate names are
3371 always allowed for groups with the same number, created by using the
3372 (?| feature. Indeed, if such groups are named, they are required to use
3373 the same names.
3374 Normally, patterns that use duplicate names are such that in any one
3376 match, only one of each set of identically-named groups participates.
3377 An example is shown in the pcre2pattern documentation.
3378 When duplicates are present, pcre2_substring_copy_byname() and
3380 pcre2_substring_get_byname() return the first substring corresponding
3381 to the given name that is set. Only if none are set is
3382 PCRE2_ERROR_UNSET is returned. The pcre2_substring_number_from_name()
3383 function returns the error PCRE2_ERROR_NOUNIQUESUBSTRING when there are
3384 duplicate names.
3385 If you want to get full details of all captured substrings for a given
3387 name, you must use the pcre2_substring_nametable_scan() function. The
3388 first argument is the compiled pattern, and the second is the name. If
3389 the third and fourth arguments are NULL, the function returns a group
3390 number for a unique name, or PCRE2_ERROR_NOUNIQUESUBSTRING otherwise.
3391 When the third and fourth arguments are not NULL, they must be pointers
3393 to variables that are updated by the function. After it has run, they
3394 point to the first and last entries in the name-to-number table for the
3395 given name, and the function returns the length of each entry in code
3396 units. In both cases, PCRE2_ERROR_NOSUBSTRING is returned if there are
3397 no entries for the given name.
3398 The format of the name table is described above in the section entitled
3400 Information about a pattern. Given all the relevant entries for the
3401 name, you can extract each of their numbers, and hence the captured
3402 data.
3403
3405 The traditional matching function uses a similar algorithm to Perl,
3407 which stops when it finds the first match at a given point in the sub‐
3408 ject. If you want to find all possible matches, or the longest possible
3409 match at a given position, consider using the alternative matching
3410 function (see below) instead. If you cannot use the alternative func‐
3411 tion, you can kludge it up by making use of the callout facility, which
3412 is described in the pcre2callout documentation.
3413 What you have to do is to insert a callout right at the end of the pat‐
3415 tern. When your callout function is called, extract and save the cur‐
3416 rent matched substring. Then return 1, which forces pcre2_match() to
3417 backtrack and try other alternatives. Ultimately, when it runs out of
3418 matches, pcre2_match() will yield PCRE2_ERROR_NOMATCH.
3419
3421 int pcre2_dfa_match(const pcre2_code *code, PCRE2_SPTR subject,
3423 PCRE2_SIZE length, PCRE2_SIZE startoffset,
3424 uint32_t options, pcre2_match_data *match_data,
3425 pcre2_match_context *mcontext,
3426 int *workspace, PCRE2_SIZE wscount);
3427 The function pcre2_dfa_match() is called to match a subject string
3429 against a compiled pattern, using a matching algorithm that scans the
3430 subject string just once (not counting lookaround assertions), and does
3431 not backtrack. This has different characteristics to the normal algo‐
3432 rithm, and is not compatible with Perl. Some of the features of PCRE2
3433 patterns are not supported. Nevertheless, there are times when this
3434 kind of matching can be useful. For a discussion of the two matching
3435 algorithms, and a list of features that pcre2_dfa_match() does not sup‐
3436 port, see the pcre2matching documentation.
3437 The arguments for the pcre2_dfa_match() function are the same as for
3439 pcre2_match(), plus two extras. The ovector within the match data block
3440 is used in a different way, and this is described below. The other com‐
3441 mon arguments are used in the same way as for pcre2_match(), so their
3442 description is not repeated here.
3443 The two additional arguments provide workspace for the function. The
3445 workspace vector should contain at least 20 elements. It is used for
3446 keeping track of multiple paths through the pattern tree. More
3447 workspace is needed for patterns and subjects where there are a lot of
3448 potential matches.
3449 Here is an example of a simple call to pcre2_dfa_match():
3451 int wspace[20];
3453 pcre2_match_data *md = pcre2_match_data_create(4, NULL);
3454 int rc = pcre2_dfa_match(
3455 re, /* result of pcre2_compile() */
3456 "some string", /* the subject string */
3457 11, /* the length of the subject string */
3458 0, /* start at offset 0 in the subject */
3459 0, /* default options */
3460 md, /* the match data block */
3461 NULL, /* a match context; NULL means use defaults */
3462 wspace, /* working space vector */
3463 20); /* number of elements (NOT size in bytes) */
3464 Option bits for pcre_dfa_match()
3466 The unused bits of the options argument for pcre2_dfa_match() must be
3468 zero. The only bits that may be set are PCRE2_ANCHORED,
3469 PCRE2_COPY_MATCHED_SUBJECT, PCRE2_ENDANCHORED, PCRE2_NOTBOL,
3470 PCRE2_NOTEOL, PCRE2_NOTEMPTY, PCRE2_NOTEMPTY_ATSTART,
3471 PCRE2_NO_UTF_CHECK, PCRE2_PARTIAL_HARD, PCRE2_PARTIAL_SOFT,
3472 PCRE2_DFA_SHORTEST, and PCRE2_DFA_RESTART. All but the last four of
3473 these are exactly the same as for pcre2_match(), so their description
3474 is not repeated here.
3475 PCRE2_PARTIAL_HARD
3477 PCRE2_PARTIAL_SOFT
3478 These have the same general effect as they do for pcre2_match(), but
3480 the details are slightly different. When PCRE2_PARTIAL_HARD is set for
3481 pcre2_dfa_match(), it returns PCRE2_ERROR_PARTIAL if the end of the
3482 subject is reached and there is still at least one matching possibility
3483 that requires additional characters. This happens even if some complete
3484 matches have already been found. When PCRE2_PARTIAL_SOFT is set, the
3485 return code PCRE2_ERROR_NOMATCH is converted into PCRE2_ERROR_PARTIAL
3486 if the end of the subject is reached, there have been no complete
3487 matches, but there is still at least one matching possibility. The por‐
3488 tion of the string that was inspected when the longest partial match
3489 was found is set as the first matching string in both cases. There is a
3490 more detailed discussion of partial and multi-segment matching, with
3491 examples, in the pcre2partial documentation.
3492 PCRE2_DFA_SHORTEST
3494 Setting the PCRE2_DFA_SHORTEST option causes the matching algorithm to
3496 stop as soon as it has found one match. Because of the way the alterna‐
3497 tive algorithm works, this is necessarily the shortest possible match
3498 at the first possible matching point in the subject string.
3499 PCRE2_DFA_RESTART
3501 When pcre2_dfa_match() returns a partial match, it is possible to call
3503 it again, with additional subject characters, and have it continue with
3504 the same match. The PCRE2_DFA_RESTART option requests this action; when
3505 it is set, the workspace and wscount options must reference the same
3506 vector as before because data about the match so far is left in them
3507 after a partial match. There is more discussion of this facility in the
3508 pcre2partial documentation.
3509 Successful returns from pcre2_dfa_match()
3511 When pcre2_dfa_match() succeeds, it may have matched more than one sub‐
3513 string in the subject. Note, however, that all the matches from one run
3514 of the function start at the same point in the subject. The shorter
3515 matches are all initial substrings of the longer matches. For example,
3516 if the pattern
3517 <.*>
3519 is matched against the string
3521 This is <something> <something else> <something further> no more
3523 the three matched strings are
3525 <something> <something else> <something further>
3527 <something> <something else>
3528 <something>
3529 On success, the yield of the function is a number greater than zero,
3531 which is the number of matched substrings. The offsets of the sub‐
3532 strings are returned in the ovector, and can be extracted by number in
3533 the same way as for pcre2_match(), but the numbers bear no relation to
3534 any capture groups that may exist in the pattern, because DFA matching
3535 does not support capturing.
3536 Calls to the convenience functions that extract substrings by name
3538 return the error PCRE2_ERROR_DFA_UFUNC (unsupported function) if used
3539 after a DFA match. The convenience functions that extract substrings by
3540 number never return PCRE2_ERROR_NOSUBSTRING.
3541 The matched strings are stored in the ovector in reverse order of
3543 length; that is, the longest matching string is first. If there were
3544 too many matches to fit into the ovector, the yield of the function is
3545 zero, and the vector is filled with the longest matches.
3546 NOTE: PCRE2's "auto-possessification" optimization usually applies to
3548 character repeats at the end of a pattern (as well as internally). For
3549 example, the pattern "a\d+" is compiled as if it were "a\d++". For DFA
3550 matching, this means that only one possible match is found. If you
3551 really do want multiple matches in such cases, either use an ungreedy
3552 repeat such as "a\d+?" or set the PCRE2_NO_AUTO_POSSESS option when
3553 compiling.
3554 Error returns from pcre2_dfa_match()
3556 The pcre2_dfa_match() function returns a negative number when it fails.
3558 Many of the errors are the same as for pcre2_match(), as described
3559 above. There are in addition the following errors that are specific to
3560 pcre2_dfa_match():
3561 PCRE2_ERROR_DFA_UITEM
3563 This return is given if pcre2_dfa_match() encounters an item in the
3565 pattern that it does not support, for instance, the use of \C in a UTF
3566 mode or a backreference.
3567 PCRE2_ERROR_DFA_UCOND
3569 This return is given if pcre2_dfa_match() encounters a condition item
3571 that uses a backreference for the condition, or a test for recursion in
3572 a specific capture group. These are not supported.
3573 PCRE2_ERROR_DFA_UINVALID_UTF
3575 This return is given if pcre2_dfa_match() is called for a pattern that
3577 was compiled with PCRE2_MATCH_INVALID_UTF. This is not supported for
3578 DFA matching.
3579 PCRE2_ERROR_DFA_WSSIZE
3581 This return is given if pcre2_dfa_match() runs out of space in the
3583 workspace vector.
3584 PCRE2_ERROR_DFA_RECURSE
3586 When a recursion or subroutine call is processed, the matching function
3588 calls itself recursively, using private memory for the ovector and
3589 workspace. This error is given if the internal ovector is not large
3590 enough. This should be extremely rare, as a vector of size 1000 is
3591 used.
3592 PCRE2_ERROR_DFA_BADRESTART
3594 When pcre2_dfa_match() is called with the PCRE2_DFA_RESTART option,
3596 some plausibility checks are made on the contents of the workspace,
3597 which should contain data about the previous partial match. If any of
3598 these checks fail, this error is given.
3599
3601 pcre2build(3), pcre2callout(3), pcre2demo(3), pcre2matching(3),
3603 pcre2partial(3), pcre2posix(3), pcre2sample(3), pcre2unicode(3).
3604
3606 Philip Hazel
3608 University Computing Service
3609 Cambridge, England.
3610
3612 Last updated: 04 November 2020
3614 Copyright (c) 1997-2020 University of Cambridge.
3615PCRE2 10.36 04 November 2020 PCRE2API(3)