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 replacementzfP,
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, at least until a pattern has been compiled. The logic
487 can be something like this:
488 Get a read-only (shared) lock (mutex) for pointer
490 if (pointer == NULL)
491 {
492 Get a write (unique) lock for pointer
493 pointer = pcre2_compile(...
494 }
495 Release the lock
496 Use pointer in pcre2_match()
497 Of course, testing for compilation errors should also be included in
499 the code.
500 If JIT is being used, but the JIT compilation is not being done immedi‐
502 ately, (perhaps waiting to see if the pattern is used often enough)
503 similar logic is required. JIT compilation updates a pointer within the
504 compiled code block, so a thread must gain unique write access to the
505 pointer before calling pcre2_jit_compile(). Alternatively,
506 pcre2_code_copy() or pcre2_code_copy_with_tables() can be used to
507 obtain a private copy of the compiled code before calling the JIT com‐
508 piler.
509 Context blocks
511 The next main section below introduces the idea of "contexts" in which
513 PCRE2 functions are called. A context is nothing more than a collection
514 of parameters that control the way PCRE2 operates. Grouping a number of
515 parameters together in a context is a convenient way of passing them to
516 a PCRE2 function without using lots of arguments. The parameters that
517 are stored in contexts are in some sense "advanced features" of the
518 API. Many straightforward applications will not need to use contexts.
519 In a multithreaded application, if the parameters in a context are val‐
521 ues that are never changed, the same context can be used by all the
522 threads. However, if any thread needs to change any value in a context,
523 it must make its own thread-specific copy.
524 Match blocks
526 The matching functions need a block of memory for storing the results
528 of a match. This includes details of what was matched, as well as addi‐
529 tional information such as the name of a (*MARK) setting. Each thread
530 must provide its own copy of this memory.
531
533 Some PCRE2 functions have a lot of parameters, many of which are used
535 only by specialist applications, for example, those that use custom
536 memory management or non-standard character tables. To keep function
537 argument lists at a reasonable size, and at the same time to keep the
538 API extensible, "uncommon" parameters are passed to certain functions
539 in a context instead of directly. A context is just a block of memory
540 that holds the parameter values. Applications that do not need to
541 adjust any of the context parameters can pass NULL when a context
542 pointer is required.
543 There are three different types of context: a general context that is
545 relevant for several PCRE2 operations, a compile-time context, and a
546 match-time context.
547 The general context
549 At present, this context just contains pointers to (and data for)
551 external memory management functions that are called from several
552 places in the PCRE2 library. The context is named `general' rather than
553 specifically `memory' because in future other fields may be added. If
554 you do not want to supply your own custom memory management functions,
555 you do not need to bother with a general context. A general context is
556 created by:
557 pcre2_general_context *pcre2_general_context_create(
559 void *(*private_malloc)(PCRE2_SIZE, void *),
560 void (*private_free)(void *, void *), void *memory_data);
561 The two function pointers specify custom memory management functions,
563 whose prototypes are:
564 void *private_malloc(PCRE2_SIZE, void *);
566 void private_free(void *, void *);
567 Whenever code in PCRE2 calls these functions, the final argument is the
569 value of memory_data. Either of the first two arguments of the creation
570 function may be NULL, in which case the system memory management func‐
571 tions malloc() and free() are used. (This is not currently useful, as
572 there are no other fields in a general context, but in future there
573 might be.) The private_malloc() function is used (if supplied) to
574 obtain memory for storing the context, and all three values are saved
575 as part of the context.
576 Whenever PCRE2 creates a data block of any kind, the block contains a
578 pointer to the free() function that matches the malloc() function that
579 was used. When the time comes to free the block, this function is
580 called.
581 A general context can be copied by calling:
583 pcre2_general_context *pcre2_general_context_copy(
585 pcre2_general_context *gcontext);
586 The memory used for a general context should be freed by calling:
588 void pcre2_general_context_free(pcre2_general_context *gcontext);
590 If this function is passed a NULL argument, it returns immediately
592 without doing anything.
593 The compile context
595 A compile context is required if you want to provide an external func‐
597 tion for stack checking during compilation or to change the default
598 values of any of the following compile-time parameters:
599 What \R matches (Unicode newlines or CR, LF, CRLF only)
601 PCRE2's character tables
602 The newline character sequence
603 The compile time nested parentheses limit
604 The maximum length of the pattern string
605 The extra options bits (none set by default)
606 A compile context is also required if you are using custom memory man‐
608 agement. If none of these apply, just pass NULL as the context argu‐
609 ment of pcre2_compile().
610 A compile context is created, copied, and freed by the following func‐
612 tions:
613 pcre2_compile_context *pcre2_compile_context_create(
615 pcre2_general_context *gcontext);
616 pcre2_compile_context *pcre2_compile_context_copy(
618 pcre2_compile_context *ccontext);
619 void pcre2_compile_context_free(pcre2_compile_context *ccontext);
621 A compile context is created with default values for its parameters.
623 These can be changed by calling the following functions, which return 0
624 on success, or PCRE2_ERROR_BADDATA if invalid data is detected.
625 int pcre2_set_bsr(pcre2_compile_context *ccontext,
627 uint32_t value);
628 The value must be PCRE2_BSR_ANYCRLF, to specify that \R matches only
630 CR, LF, or CRLF, or PCRE2_BSR_UNICODE, to specify that \R matches any
631 Unicode line ending sequence. The value is used by the JIT compiler and
632 by the two interpreted matching functions, pcre2_match() and
633 pcre2_dfa_match().
634 int pcre2_set_character_tables(pcre2_compile_context *ccontext,
636 const uint8_t *tables);
637 The value must be the result of a call to pcre2_maketables(), whose
639 only argument is a general context. This function builds a set of char‐
640 acter tables in the current locale.
641 int pcre2_set_compile_extra_options(pcre2_compile_context *ccontext,
643 uint32_t extra_options);
644 As PCRE2 has developed, almost all the 32 option bits that are avail‐
646 able in the options argument of pcre2_compile() have been used up. To
647 avoid running out, the compile context contains a set of extra option
648 bits which are used for some newer, assumed rarer, options. This func‐
649 tion sets those bits. It always sets all the bits (either on or off).
650 It does not modify any existing setting. The available options are
651 defined in the section entitled "Extra compile options" below.
652 int pcre2_set_max_pattern_length(pcre2_compile_context *ccontext,
654 PCRE2_SIZE value);
655 This sets a maximum length, in code units, for any pattern string that
657 is compiled with this context. If the pattern is longer, an error is
658 generated. This facility is provided so that applications that accept
659 patterns from external sources can limit their size. The default is the
660 largest number that a PCRE2_SIZE variable can hold, which is effec‐
661 tively unlimited.
662 int pcre2_set_newline(pcre2_compile_context *ccontext,
664 uint32_t value);
665 This specifies which characters or character sequences are to be recog‐
667 nized as newlines. The value must be one of PCRE2_NEWLINE_CR (carriage
668 return only), PCRE2_NEWLINE_LF (linefeed only), PCRE2_NEWLINE_CRLF (the
669 two-character sequence CR followed by LF), PCRE2_NEWLINE_ANYCRLF (any
670 of the above), PCRE2_NEWLINE_ANY (any Unicode newline sequence), or
671 PCRE2_NEWLINE_NUL (the NUL character, that is a binary zero).
672 A pattern can override the value set in the compile context by starting
674 with a sequence such as (*CRLF). See the pcre2pattern page for details.
675 When a pattern is compiled with the PCRE2_EXTENDED or
677 PCRE2_EXTENDED_MORE option, the newline convention affects the recogni‐
678 tion of the end of internal comments starting with #. The value is
679 saved with the compiled pattern for subsequent use by the JIT compiler
680 and by the two interpreted matching functions, pcre2_match() and
681 pcre2_dfa_match().
682 int pcre2_set_parens_nest_limit(pcre2_compile_context *ccontext,
684 uint32_t value);
685 This parameter adjusts the limit, set when PCRE2 is built (default
687 250), on the depth of parenthesis nesting in a pattern. This limit
688 stops rogue patterns using up too much system stack when being com‐
689 piled. The limit applies to parentheses of all kinds, not just captur‐
690 ing parentheses.
691 int pcre2_set_compile_recursion_guard(pcre2_compile_context *ccontext,
693 int (*guard_function)(uint32_t, void *), void *user_data);
694 There is at least one application that runs PCRE2 in threads with very
696 limited system stack, where running out of stack is to be avoided at
697 all costs. The parenthesis limit above cannot take account of how much
698 stack is actually available during compilation. For a finer control,
699 you can supply a function that is called whenever pcre2_compile()
700 starts to compile a parenthesized part of a pattern. This function can
701 check the actual stack size (or anything else that it wants to, of
702 course).
703 The first argument to the callout function gives the current depth of
705 nesting, and the second is user data that is set up by the last argu‐
706 ment of pcre2_set_compile_recursion_guard(). The callout function
707 should return zero if all is well, or non-zero to force an error.
708 The match context
710 A match context is required if you want to:
712 Set up a callout function
714 Set an offset limit for matching an unanchored pattern
715 Change the limit on the amount of heap used when matching
716 Change the backtracking match limit
717 Change the backtracking depth limit
718 Set custom memory management specifically for the match
719 If none of these apply, just pass NULL as the context argument of
721 pcre2_match(), pcre2_dfa_match(), or pcre2_jit_match().
722 A match context is created, copied, and freed by the following func‐
724 tions:
725 pcre2_match_context *pcre2_match_context_create(
727 pcre2_general_context *gcontext);
728 pcre2_match_context *pcre2_match_context_copy(
730 pcre2_match_context *mcontext);
731 void pcre2_match_context_free(pcre2_match_context *mcontext);
733 A match context is created with default values for its parameters.
735 These can be changed by calling the following functions, which return 0
736 on success, or PCRE2_ERROR_BADDATA if invalid data is detected.
737 int pcre2_set_callout(pcre2_match_context *mcontext,
739 int (*callout_function)(pcre2_callout_block *, void *),
740 void *callout_data);
741 This sets up a callout function for PCRE2 to call at specified points
743 during a matching operation. Details are given in the pcre2callout doc‐
744 umentation.
745 int pcre2_set_substitute_callout(pcre2_match_context *mcontext,
747 int (*callout_function)(pcre2_substitute_callout_block *, void *),
748 void *callout_data);
749 This sets up a callout function for PCRE2 to call after each substitu‐
751 tion made by pcre2_substitute(). Details are given in the section enti‐
752 tled "Creating a new string with substitutions" below.
753 int pcre2_set_offset_limit(pcre2_match_context *mcontext,
755 PCRE2_SIZE value);
756 The offset_limit parameter limits how far an unanchored search can
758 advance in the subject string. The default value is PCRE2_UNSET. The
759 pcre2_match() and pcre2_dfa_match() functions return
760 PCRE2_ERROR_NOMATCH if a match with a starting point before or at the
761 given offset is not found. The pcre2_substitute() function makes no
762 more substitutions.
763 For example, if the pattern /abc/ is matched against "123abc" with an
765 offset limit less than 3, the result is PCRE2_ERROR_NOMATCH. A match
766 can never be found if the startoffset argument of pcre2_match(),
767 pcre2_dfa_match(), or pcre2_substitute() is greater than the offset
768 limit set in the match context.
769 When using this facility, you must set the PCRE2_USE_OFFSET_LIMIT
771 option when calling pcre2_compile() so that when JIT is in use, differ‐
772 ent code can be compiled. If a match is started with a non-default
773 match limit when PCRE2_USE_OFFSET_LIMIT is not set, an error is gener‐
774 ated.
775 The offset limit facility can be used to track progress when searching
777 large subject strings or to limit the extent of global substitutions.
778 See also the PCRE2_FIRSTLINE option, which requires a match to start
779 before or at the first newline that follows the start of matching in
780 the subject. If this is set with an offset limit, a match must occur in
781 the first line and also within the offset limit. In other words, which‐
782 ever limit comes first is used.
783 int pcre2_set_heap_limit(pcre2_match_context *mcontext,
785 uint32_t value);
786 The heap_limit parameter specifies, in units of kibibytes (1024 bytes),
788 the maximum amount of heap memory that pcre2_match() may use to hold
789 backtracking information when running an interpretive match. This limit
790 also applies to pcre2_dfa_match(), which may use the heap when process‐
791 ing patterns with a lot of nested pattern recursion or lookarounds or
792 atomic groups. This limit does not apply to matching with the JIT opti‐
793 mization, which has its own memory control arrangements (see the
794 pcre2jit documentation for more details). If the limit is reached, the
795 negative error code PCRE2_ERROR_HEAPLIMIT is returned. The default
796 limit can be set when PCRE2 is built; if it is not, the default is set
797 very large and is essentially "unlimited".
798 A value for the heap limit may also be supplied by an item at the start
800 of a pattern of the form
801 (*LIMIT_HEAP=ddd)
803 where ddd is a decimal number. However, such a setting is ignored
805 unless ddd is less than the limit set by the caller of pcre2_match()
806 or, if no such limit is set, less than the default.
807 The pcre2_match() function starts out using a 20KiB vector on the sys‐
809 tem stack for recording backtracking points. The more nested backtrack‐
810 ing points there are (that is, the deeper the search tree), the more
811 memory is needed. Heap memory is used only if the initial vector is
812 too small. If the heap limit is set to a value less than 21 (in partic‐
813 ular, zero) no heap memory will be used. In this case, only patterns
814 that do not have a lot of nested backtracking can be successfully pro‐
815 cessed.
816 Similarly, for pcre2_dfa_match(), a vector on the system stack is used
818 when processing pattern recursions, lookarounds, or atomic groups, and
819 only if this is not big enough is heap memory used. In this case, too,
820 setting a value of zero disables the use of the heap.
821 int pcre2_set_match_limit(pcre2_match_context *mcontext,
823 uint32_t value);
824 The match_limit parameter provides a means of preventing PCRE2 from
826 using up too many computing resources when processing patterns that are
827 not going to match, but which have a very large number of possibilities
828 in their search trees. The classic example is a pattern that uses
829 nested unlimited repeats.
830 There is an internal counter in pcre2_match() that is incremented each
832 time round its main matching loop. If this value reaches the match
833 limit, pcre2_match() returns the negative value PCRE2_ERROR_MATCHLIMIT.
834 This has the effect of limiting the amount of backtracking that can
835 take place. For patterns that are not anchored, the count restarts from
836 zero for each position in the subject string. This limit also applies
837 to pcre2_dfa_match(), though the counting is done in a different way.
838 When pcre2_match() is called with a pattern that was successfully pro‐
840 cessed by pcre2_jit_compile(), the way in which matching is executed is
841 entirely different. However, there is still the possibility of runaway
842 matching that goes on for a very long time, and so the match_limit
843 value is also used in this case (but in a different way) to limit how
844 long the matching can continue.
845 The default value for the limit can be set when PCRE2 is built; the
847 default default is 10 million, which handles all but the most extreme
848 cases. A value for the match limit may also be supplied by an item at
849 the start of a pattern of the form
850 (*LIMIT_MATCH=ddd)
852 where ddd is a decimal number. However, such a setting is ignored
854 unless ddd is less than the limit set by the caller of pcre2_match() or
855 pcre2_dfa_match() or, if no such limit is set, less than the default.
856 int pcre2_set_depth_limit(pcre2_match_context *mcontext,
858 uint32_t value);
859 This parameter limits the depth of nested backtracking in
861 pcre2_match(). Each time a nested backtracking point is passed, a new
862 memory "frame" is used to remember the state of matching at that point.
863 Thus, this parameter indirectly limits the amount of memory that is
864 used in a match. However, because the size of each memory "frame"
865 depends on the number of capturing parentheses, the actual memory limit
866 varies from pattern to pattern. This limit was more useful in versions
867 before 10.30, where function recursion was used for backtracking.
868 The depth limit is not relevant, and is ignored, when matching is done
870 using JIT compiled code. However, it is supported by pcre2_dfa_match(),
871 which uses it to limit the depth of nested internal recursive function
872 calls that implement atomic groups, lookaround assertions, and pattern
873 recursions. This limits, indirectly, the amount of system stack that is
874 used. It was more useful in versions before 10.32, when stack memory
875 was used for local workspace vectors for recursive function calls. From
876 version 10.32, only local variables are allocated on the stack and as
877 each call uses only a few hundred bytes, even a small stack can support
878 quite a lot of recursion.
879 If the depth of internal recursive function calls is great enough,
881 local workspace vectors are allocated on the heap from version 10.32
882 onwards, so the depth limit also indirectly limits the amount of heap
883 memory that is used. A recursive pattern such as /(.(?2))((?1)|)/, when
884 matched to a very long string using pcre2_dfa_match(), can use a great
885 deal of memory. However, it is probably better to limit heap usage
886 directly by calling pcre2_set_heap_limit().
887 The default value for the depth limit can be set when PCRE2 is built;
889 if it is not, the default is set to the same value as the default for
890 the match limit. If the limit is exceeded, pcre2_match() or
891 pcre2_dfa_match() returns PCRE2_ERROR_DEPTHLIMIT. A value for the depth
892 limit may also be supplied by an item at the start of a pattern of the
893 form
894 (*LIMIT_DEPTH=ddd)
896 where ddd is a decimal number. However, such a setting is ignored
898 unless ddd is less than the limit set by the caller of pcre2_match() or
899 pcre2_dfa_match() or, if no such limit is set, less than the default.
900
902 int pcre2_config(uint32_t what, void *where);
904 The function pcre2_config() makes it possible for a PCRE2 client to
906 discover which optional features have been compiled into the PCRE2
907 library. The pcre2build documentation has more details about these
908 optional features.
909 The first argument for pcre2_config() specifies which information is
911 required. The second argument is a pointer to memory into which the
912 information is placed. If NULL is passed, the function returns the
913 amount of memory that is needed for the requested information. For
914 calls that return numerical values, the value is in bytes; when
915 requesting these values, where should point to appropriately aligned
916 memory. For calls that return strings, the required length is given in
917 code units, not counting the terminating zero.
918 When requesting information, the returned value from pcre2_config() is
920 non-negative on success, or the negative error code PCRE2_ERROR_BADOP‐
921 TION if the value in the first argument is not recognized. The follow‐
922 ing information is available:
923 PCRE2_CONFIG_BSR
925 The output is a uint32_t integer whose value indicates what character
927 sequences the \R escape sequence matches by default. A value of
928 PCRE2_BSR_UNICODE means that \R matches any Unicode line ending
929 sequence; a value of PCRE2_BSR_ANYCRLF means that \R matches only CR,
930 LF, or CRLF. The default can be overridden when a pattern is compiled.
931 PCRE2_CONFIG_COMPILED_WIDTHS
933 The output is a uint32_t integer whose lower bits indicate which code
935 unit widths were selected when PCRE2 was built. The 1-bit indicates
936 8-bit support, and the 2-bit and 4-bit indicate 16-bit and 32-bit sup‐
937 port, respectively.
938 PCRE2_CONFIG_DEPTHLIMIT
940 The output is a uint32_t integer that gives the default limit for the
942 depth of nested backtracking in pcre2_match() or the depth of nested
943 recursions, lookarounds, and atomic groups in pcre2_dfa_match(). Fur‐
944 ther details are given with pcre2_set_depth_limit() above.
945 PCRE2_CONFIG_HEAPLIMIT
947 The output is a uint32_t integer that gives, in kibibytes, the default
949 limit for the amount of heap memory used by pcre2_match() or
950 pcre2_dfa_match(). Further details are given with
951 pcre2_set_heap_limit() above.
952 PCRE2_CONFIG_JIT
954 The output is a uint32_t integer that is set to one if support for
956 just-in-time compiling is available; otherwise it is set to zero.
957 PCRE2_CONFIG_JITTARGET
959 The where argument should point to a buffer that is at least 48 code
961 units long. (The exact length required can be found by calling
962 pcre2_config() with where set to NULL.) The buffer is filled with a
963 string that contains the name of the architecture for which the JIT
964 compiler is configured, for example "x86 32bit (little endian +
965 unaligned)". If JIT support is not available, PCRE2_ERROR_BADOPTION is
966 returned, otherwise the number of code units used is returned. This is
967 the length of the string, plus one unit for the terminating zero.
968 PCRE2_CONFIG_LINKSIZE
970 The output is a uint32_t integer that contains the number of bytes used
972 for internal linkage in compiled regular expressions. When PCRE2 is
973 configured, the value can be set to 2, 3, or 4, with the default being
974 2. This is the value that is returned by pcre2_config(). However, when
975 the 16-bit library is compiled, a value of 3 is rounded up to 4, and
976 when the 32-bit library is compiled, internal linkages always use 4
977 bytes, so the configured value is not relevant.
978 The default value of 2 for the 8-bit and 16-bit libraries is sufficient
980 for all but the most massive patterns, since it allows the size of the
981 compiled pattern to be up to 65535 code units. Larger values allow
982 larger regular expressions to be compiled by those two libraries, but
983 at the expense of slower matching.
984 PCRE2_CONFIG_MATCHLIMIT
986 The output is a uint32_t integer that gives the default match limit for
988 pcre2_match(). Further details are given with pcre2_set_match_limit()
989 above.
990 PCRE2_CONFIG_NEWLINE
992 The output is a uint32_t integer whose value specifies the default
994 character sequence that is recognized as meaning "newline". The values
995 are:
996 PCRE2_NEWLINE_CR Carriage return (CR)
998 PCRE2_NEWLINE_LF Linefeed (LF)
999 PCRE2_NEWLINE_CRLF Carriage return, linefeed (CRLF)
1000 PCRE2_NEWLINE_ANY Any Unicode line ending
1001 PCRE2_NEWLINE_ANYCRLF Any of CR, LF, or CRLF
1002 PCRE2_NEWLINE_NUL The NUL character (binary zero)
1003 The default should normally correspond to the standard sequence for
1005 your operating system.
1006 PCRE2_CONFIG_NEVER_BACKSLASH_C
1008 The output is a uint32_t integer that is set to one if the use of \C
1010 was permanently disabled when PCRE2 was built; otherwise it is set to
1011 zero.
1012 PCRE2_CONFIG_PARENSLIMIT
1014 The output is a uint32_t integer that gives the maximum depth of nest‐
1016 ing of parentheses (of any kind) in a pattern. This limit is imposed to
1017 cap the amount of system stack used when a pattern is compiled. It is
1018 specified when PCRE2 is built; the default is 250. This limit does not
1019 take into account the stack that may already be used by the calling
1020 application. For finer control over compilation stack usage, see
1021 pcre2_set_compile_recursion_guard().
1022 PCRE2_CONFIG_STACKRECURSE
1024 This parameter is obsolete and should not be used in new code. The out‐
1026 put is a uint32_t integer that is always set to zero.
1027 PCRE2_CONFIG_UNICODE_VERSION
1029 The where argument should point to a buffer that is at least 24 code
1031 units long. (The exact length required can be found by calling
1032 pcre2_config() with where set to NULL.) If PCRE2 has been compiled
1033 without Unicode support, the buffer is filled with the text "Unicode
1034 not supported". Otherwise, the Unicode version string (for example,
1035 "8.0.0") is inserted. The number of code units used is returned. This
1036 is the length of the string plus one unit for the terminating zero.
1037 PCRE2_CONFIG_UNICODE
1039 The output is a uint32_t integer that is set to one if Unicode support
1041 is available; otherwise it is set to zero. Unicode support implies UTF
1042 support.
1043 PCRE2_CONFIG_VERSION
1045 The where argument should point to a buffer that is at least 24 code
1047 units long. (The exact length required can be found by calling
1048 pcre2_config() with where set to NULL.) The buffer is filled with the
1049 PCRE2 version string, zero-terminated. The number of code units used is
1050 returned. This is the length of the string plus one unit for the termi‐
1051 nating zero.
1052
1054 pcre2_code *pcre2_compile(PCRE2_SPTR pattern, PCRE2_SIZE length,
1056 uint32_t options, int *errorcode, PCRE2_SIZE *erroroffset,
1057 pcre2_compile_context *ccontext);
1058 void pcre2_code_free(pcre2_code *code);
1060 pcre2_code *pcre2_code_copy(const pcre2_code *code);
1062 pcre2_code *pcre2_code_copy_with_tables(const pcre2_code *code);
1064 The pcre2_compile() function compiles a pattern into an internal form.
1066 The pattern is defined by a pointer to a string of code units and a
1067 length (in code units). If the pattern is zero-terminated, the length
1068 can be specified as PCRE2_ZERO_TERMINATED. The function returns a
1069 pointer to a block of memory that contains the compiled pattern and
1070 related data, or NULL if an error occurred.
1071 If the compile context argument ccontext is NULL, memory for the com‐
1073 piled pattern is obtained by calling malloc(). Otherwise, it is
1074 obtained from the same memory function that was used for the compile
1075 context. The caller must free the memory by calling pcre2_code_free()
1076 when it is no longer needed. If pcre2_code_free() is called with a
1077 NULL argument, it returns immediately, without doing anything.
1078 The function pcre2_code_copy() makes a copy of the compiled code in new
1080 memory, using the same memory allocator as was used for the original.
1081 However, if the code has been processed by the JIT compiler (see
1082 below), the JIT information cannot be copied (because it is position-
1083 dependent). The new copy can initially be used only for non-JIT match‐
1084 ing, though it can be passed to pcre2_jit_compile() if required. If
1085 pcre2_code_copy() is called with a NULL argument, it returns NULL.
1086 The pcre2_code_copy() function provides a way for individual threads in
1088 a multithreaded application to acquire a private copy of shared com‐
1089 piled code. However, it does not make a copy of the character tables
1090 used by the compiled pattern; the new pattern code points to the same
1091 tables as the original code. (See "Locale Support" below for details
1092 of these character tables.) In many applications the same tables are
1093 used throughout, so this behaviour is appropriate. Nevertheless, there
1094 are occasions when a copy of a compiled pattern and the relevant tables
1095 are needed. The pcre2_code_copy_with_tables() provides this facility.
1096 Copies of both the code and the tables are made, with the new code
1097 pointing to the new tables. The memory for the new tables is automati‐
1098 cally freed when pcre2_code_free() is called for the new copy of the
1099 compiled code. If pcre2_code_copy_with_tables() is called with a NULL
1100 argument, it returns NULL.
1101 NOTE: When one of the matching functions is called, pointers to the
1103 compiled pattern and the subject string are set in the match data block
1104 so that they can be referenced by the substring extraction functions
1105 after a successful match. After running a match, you must not free a
1106 compiled pattern or a subject string until after all operations on the
1107 match data block have taken place, unless, in the case of the subject
1108 string, you have used the PCRE2_COPY_MATCHED_SUBJECT option, which is
1109 described in the section entitled "Option bits for pcre2_match()"
1110 below.
1111 The options argument for pcre2_compile() contains various bit settings
1113 that affect the compilation. It should be zero if none of them are
1114 required. The available options are described below. Some of them (in
1115 particular, those that are compatible with Perl, but some others as
1116 well) can also be set and unset from within the pattern (see the
1117 detailed description in the pcre2pattern documentation).
1118 For those options that can be different in different parts of the pat‐
1120 tern, the contents of the options argument specifies their settings at
1121 the start of compilation. The PCRE2_ANCHORED, PCRE2_ENDANCHORED, and
1122 PCRE2_NO_UTF_CHECK options can be set at the time of matching as well
1123 as at compile time.
1124 Some additional options and less frequently required compile-time
1126 parameters (for example, the newline setting) can be provided in a com‐
1127 pile context (as described above).
1128 If errorcode or erroroffset is NULL, pcre2_compile() returns NULL imme‐
1130 diately. Otherwise, the variables to which these point are set to an
1131 error code and an offset (number of code units) within the pattern,
1132 respectively, when pcre2_compile() returns NULL because a compilation
1133 error has occurred. The values are not defined when compilation is suc‐
1134 cessful and pcre2_compile() returns a non-NULL value.
1135 There are nearly 100 positive error codes that pcre2_compile() may
1137 return if it finds an error in the pattern. There are also some nega‐
1138 tive error codes that are used for invalid UTF strings when validity
1139 checking is in force. These are the same as given by pcre2_match() and
1140 pcre2_dfa_match(), and are described in the pcre2unicode documentation.
1141 There is no separate documentation for the positive error codes,
1142 because the textual error messages that are obtained by calling the
1143 pcre2_get_error_message() function (see "Obtaining a textual error mes‐
1144 sage" below) should be self-explanatory. Macro names starting with
1145 PCRE2_ERROR_ are defined for both positive and negative error codes in
1146 pcre2.h.
1147 The value returned in erroroffset is an indication of where in the pat‐
1149 tern the error occurred. It is not necessarily the furthest point in
1150 the pattern that was read. For example, after the error "lookbehind
1151 assertion is not fixed length", the error offset points to the start of
1152 the failing assertion. For an invalid UTF-8 or UTF-16 string, the off‐
1153 set is that of the first code unit of the failing character.
1154 Some errors are not detected until the whole pattern has been scanned;
1156 in these cases, the offset passed back is the length of the pattern.
1157 Note that the offset is in code units, not characters, even in a UTF
1158 mode. It may sometimes point into the middle of a UTF-8 or UTF-16 char‐
1159 acter.
1160 This code fragment shows a typical straightforward call to pcre2_com‐
1162 pile():
1163 pcre2_code *re;
1165 PCRE2_SIZE erroffset;
1166 int errorcode;
1167 re = pcre2_compile(
1168 "^A.*Z", /* the pattern */
1169 PCRE2_ZERO_TERMINATED, /* the pattern is zero-terminated */
1170 0, /* default options */
1171 &errorcode, /* for error code */
1172 &erroffset, /* for error offset */
1173 NULL); /* no compile context */
1174 Main compile options
1177 The following names for option bits are defined in the pcre2.h header
1179 file:
1180 PCRE2_ANCHORED
1182 If this bit is set, the pattern is forced to be "anchored", that is, it
1184 is constrained to match only at the first matching point in the string
1185 that is being searched (the "subject string"). This effect can also be
1186 achieved by appropriate constructs in the pattern itself, which is the
1187 only way to do it in Perl.
1188 PCRE2_ALLOW_EMPTY_CLASS
1190 By default, for compatibility with Perl, a closing square bracket that
1192 immediately follows an opening one is treated as a data character for
1193 the class. When PCRE2_ALLOW_EMPTY_CLASS is set, it terminates the
1194 class, which therefore contains no characters and so can never match.
1195 PCRE2_ALT_BSUX
1197 This option request alternative handling of three escape sequences,
1199 which makes PCRE2's behaviour more like ECMAscript (aka JavaScript).
1200 When it is set:
1201 (1) \U matches an upper case "U" character; by default \U causes a com‐
1203 pile time error (Perl uses \U to upper case subsequent characters).
1204 (2) \u matches a lower case "u" character unless it is followed by four
1206 hexadecimal digits, in which case the hexadecimal number defines the
1207 code point to match. By default, \u causes a compile time error (Perl
1208 uses it to upper case the following character).
1209 (3) \x matches a lower case "x" character unless it is followed by two
1211 hexadecimal digits, in which case the hexadecimal number defines the
1212 code point to match. By default, as in Perl, a hexadecimal number is
1213 always expected after \x, but it may have zero, one, or two digits (so,
1214 for example, \xz matches a binary zero character followed by z).
1215 ECMAscript 6 added additional functionality to \u. This can be accessed
1217 using the PCRE2_EXTRA_ALT_BSUX extra option (see "Extra compile
1218 options" below). Note that this alternative escape handling applies
1219 only to patterns. Neither of these options affects the processing of
1220 replacement strings passed to pcre2_substitute().
1221 PCRE2_ALT_CIRCUMFLEX
1223 In multiline mode (when PCRE2_MULTILINE is set), the circumflex
1225 metacharacter matches at the start of the subject (unless PCRE2_NOTBOL
1226 is set), and also after any internal newline. However, it does not
1227 match after a newline at the end of the subject, for compatibility with
1228 Perl. If you want a multiline circumflex also to match after a termi‐
1229 nating newline, you must set PCRE2_ALT_CIRCUMFLEX.
1230 PCRE2_ALT_VERBNAMES
1232 By default, for compatibility with Perl, the name in any verb sequence
1234 such as (*MARK:NAME) is any sequence of characters that does not
1235 include a closing parenthesis. The name is not processed in any way,
1236 and it is not possible to include a closing parenthesis in the name.
1237 However, if the PCRE2_ALT_VERBNAMES option is set, normal backslash
1238 processing is applied to verb names and only an unescaped closing
1239 parenthesis terminates the name. A closing parenthesis can be included
1240 in a name either as \) or between \Q and \E. If the PCRE2_EXTENDED or
1241 PCRE2_EXTENDED_MORE option is set with PCRE2_ALT_VERBNAMES, unescaped
1242 whitespace in verb names is skipped and #-comments are recognized,
1243 exactly as in the rest of the pattern.
1244 PCRE2_AUTO_CALLOUT
1246 If this bit is set, pcre2_compile() automatically inserts callout
1248 items, all with number 255, before each pattern item, except immedi‐
1249 ately before or after an explicit callout in the pattern. For discus‐
1250 sion of the callout facility, see the pcre2callout documentation.
1251 PCRE2_CASELESS
1253 If this bit is set, letters in the pattern match both upper and lower
1255 case letters in the subject. It is equivalent to Perl's /i option, and
1256 it can be changed within a pattern by a (?i) option setting. If
1257 PCRE2_UTF is set, Unicode properties are used for all characters with
1258 more than one other case, and for all characters whose code points are
1259 greater than U+007F. For lower valued characters with only one other
1260 case, a lookup table is used for speed. When PCRE2_UTF is not set, a
1261 lookup table is used for all code points less than 256, and higher code
1262 points (available only in 16-bit or 32-bit mode) are treated as not
1263 having another case.
1264 PCRE2_DOLLAR_ENDONLY
1266 If this bit is set, a dollar metacharacter in the pattern matches only
1268 at the end of the subject string. Without this option, a dollar also
1269 matches immediately before a newline at the end of the string (but not
1270 before any other newlines). The PCRE2_DOLLAR_ENDONLY option is ignored
1271 if PCRE2_MULTILINE is set. There is no equivalent to this option in
1272 Perl, and no way to set it within a pattern.
1273 PCRE2_DOTALL
1275 If this bit is set, a dot metacharacter in the pattern matches any
1277 character, including one that indicates a newline. However, it only
1278 ever matches one character, even if newlines are coded as CRLF. Without
1279 this option, a dot does not match when the current position in the sub‐
1280 ject is at a newline. This option is equivalent to Perl's /s option,
1281 and it can be changed within a pattern by a (?s) option setting. A neg‐
1282 ative class such as [^a] always matches newline characters, and the \N
1283 escape sequence always matches a non-newline character, independent of
1284 the setting of PCRE2_DOTALL.
1285 PCRE2_DUPNAMES
1287 If this bit is set, names used to identify capture groups need not be
1289 unique. This can be helpful for certain types of pattern when it is
1290 known that only one instance of the named group can ever be matched.
1291 There are more details of named capture groups below; see also the
1292 pcre2pattern documentation.
1293 PCRE2_ENDANCHORED
1295 If this bit is set, the end of any pattern match must be right at the
1297 end of the string being searched (the "subject string"). If the pattern
1298 match succeeds by reaching (*ACCEPT), but does not reach the end of the
1299 subject, the match fails at the current starting point. For unanchored
1300 patterns, a new match is then tried at the next starting point. How‐
1301 ever, if the match succeeds by reaching the end of the pattern, but not
1302 the end of the subject, backtracking occurs and an alternative match
1303 may be found. Consider these two patterns:
1304 .(*ACCEPT)|..
1306 .|..
1307 If matched against "abc" with PCRE2_ENDANCHORED set, the first matches
1309 "c" whereas the second matches "bc". The effect of PCRE2_ENDANCHORED
1310 can also be achieved by appropriate constructs in the pattern itself,
1311 which is the only way to do it in Perl.
1312 For DFA matching with pcre2_dfa_match(), PCRE2_ENDANCHORED applies only
1314 to the first (that is, the longest) matched string. Other parallel
1315 matches, which are necessarily substrings of the first one, must obvi‐
1316 ously end before the end of the subject.
1317 PCRE2_EXTENDED
1319 If this bit is set, most white space characters in the pattern are
1321 totally ignored except when escaped or inside a character class. How‐
1322 ever, white space is not allowed within sequences such as (?> that
1323 introduce various parenthesized groups, nor within numerical quanti‐
1324 fiers such as {1,3}. Ignorable white space is permitted between an item
1325 and a following quantifier and between a quantifier and a following +
1326 that indicates possessiveness. PCRE2_EXTENDED is equivalent to Perl's
1327 /x option, and it can be changed within a pattern by a (?x) option set‐
1328 ting.
1329 When PCRE2 is compiled without Unicode support, PCRE2_EXTENDED recog‐
1331 nizes as white space only those characters with code points less than
1332 256 that are flagged as white space in its low-character table. The ta‐
1333 ble is normally created by pcre2_maketables(), which uses the isspace()
1334 function to identify space characters. In most ASCII environments, the
1335 relevant characters are those with code points 0x0009 (tab), 0x000A
1336 (linefeed), 0x000B (vertical tab), 0x000C (formfeed), 0x000D (carriage
1337 return), and 0x0020 (space).
1338 When PCRE2 is compiled with Unicode support, in addition to these char‐
1340 acters, five more Unicode "Pattern White Space" characters are recog‐
1341 nized by PCRE2_EXTENDED. These are U+0085 (next line), U+200E (left-to-
1342 right mark), U+200F (right-to-left mark), U+2028 (line separator), and
1343 U+2029 (paragraph separator). This set of characters is the same as
1344 recognized by Perl's /x option. Note that the horizontal and vertical
1345 space characters that are matched by the \h and \v escapes in patterns
1346 are a much bigger set.
1347 As well as ignoring most white space, PCRE2_EXTENDED also causes char‐
1349 acters between an unescaped # outside a character class and the next
1350 newline, inclusive, to be ignored, which makes it possible to include
1351 comments inside complicated patterns. Note that the end of this type of
1352 comment is a literal newline sequence in the pattern; escape sequences
1353 that happen to represent a newline do not count.
1354 Which characters are interpreted as newlines can be specified by a set‐
1356 ting in the compile context that is passed to pcre2_compile() or by a
1357 special sequence at the start of the pattern, as described in the sec‐
1358 tion entitled "Newline conventions" in the pcre2pattern documentation.
1359 A default is defined when PCRE2 is built.
1360 PCRE2_EXTENDED_MORE
1362 This option has the effect of PCRE2_EXTENDED, but, in addition,
1364 unescaped space and horizontal tab characters are ignored inside a
1365 character class. Note: only these two characters are ignored, not the
1366 full set of pattern white space characters that are ignored outside a
1367 character class. PCRE2_EXTENDED_MORE is equivalent to Perl's /xx
1368 option, and it can be changed within a pattern by a (?xx) option set‐
1369 ting.
1370 PCRE2_FIRSTLINE
1372 If this option is set, the start of an unanchored pattern match must be
1374 before or at the first newline in the subject string following the
1375 start of matching, though the matched text may continue over the new‐
1376 line. If startoffset is non-zero, the limiting newline is not necessar‐
1377 ily the first newline in the subject. For example, if the subject
1378 string is "abc\nxyz" (where \n represents a single-character newline) a
1379 pattern match for "yz" succeeds with PCRE2_FIRSTLINE if startoffset is
1380 greater than 3. See also PCRE2_USE_OFFSET_LIMIT, which provides a more
1381 general limiting facility. If PCRE2_FIRSTLINE is set with an offset
1382 limit, a match must occur in the first line and also within the offset
1383 limit. In other words, whichever limit comes first is used.
1384 PCRE2_LITERAL
1386 If this option is set, all meta-characters in the pattern are disabled,
1388 and it is treated as a literal string. Matching literal strings with a
1389 regular expression engine is not the most efficient way of doing it. If
1390 you are doing a lot of literal matching and are worried about effi‐
1391 ciency, you should consider using other approaches. The only other main
1392 options that are allowed with PCRE2_LITERAL are: PCRE2_ANCHORED,
1393 PCRE2_ENDANCHORED, PCRE2_AUTO_CALLOUT, PCRE2_CASELESS, PCRE2_FIRSTLINE,
1394 PCRE2_MATCH_INVALID_UTF, PCRE2_NO_START_OPTIMIZE, PCRE2_NO_UTF_CHECK,
1395 PCRE2_UTF, and PCRE2_USE_OFFSET_LIMIT. The extra options
1396 PCRE2_EXTRA_MATCH_LINE and PCRE2_EXTRA_MATCH_WORD are also supported.
1397 Any other options cause an error.
1398 PCRE2_MATCH_INVALID_UTF
1400 This option forces PCRE2_UTF (see below) and also enables support for
1402 matching by pcre2_match() in subject strings that contain invalid UTF
1403 sequences. This facility is not supported for DFA matching. For
1404 details, see the pcre2unicode documentation.
1405 PCRE2_MATCH_UNSET_BACKREF
1407 If this option is set, a backreference to an unset capture group
1409 matches an empty string (by default this causes the current matching
1410 alternative to fail). A pattern such as (\1)(a) succeeds when this
1411 option is set (assuming it can find an "a" in the subject), whereas it
1412 fails by default, for Perl compatibility. Setting this option makes
1413 PCRE2 behave more like ECMAscript (aka JavaScript).
1414 PCRE2_MULTILINE
1416 By default, for the purposes of matching "start of line" and "end of
1418 line", PCRE2 treats the subject string as consisting of a single line
1419 of characters, even if it actually contains newlines. The "start of
1420 line" metacharacter (^) matches only at the start of the string, and
1421 the "end of line" metacharacter ($) matches only at the end of the
1422 string, or before a terminating newline (except when PCRE2_DOL‐
1423 LAR_ENDONLY is set). Note, however, that unless PCRE2_DOTALL is set,
1424 the "any character" metacharacter (.) does not match at a newline. This
1425 behaviour (for ^, $, and dot) is the same as Perl.
1426 When PCRE2_MULTILINE it is set, the "start of line" and "end of line"
1428 constructs match immediately following or immediately before internal
1429 newlines in the subject string, respectively, as well as at the very
1430 start and end. This is equivalent to Perl's /m option, and it can be
1431 changed within a pattern by a (?m) option setting. Note that the "start
1432 of line" metacharacter does not match after a newline at the end of the
1433 subject, for compatibility with Perl. However, you can change this by
1434 setting the PCRE2_ALT_CIRCUMFLEX option. If there are no newlines in a
1435 subject string, or no occurrences of ^ or $ in a pattern, setting
1436 PCRE2_MULTILINE has no effect.
1437 PCRE2_NEVER_BACKSLASH_C
1439 This option locks out the use of \C in the pattern that is being com‐
1441 piled. This escape can cause unpredictable behaviour in UTF-8 or
1442 UTF-16 modes, because it may leave the current matching point in the
1443 middle of a multi-code-unit character. This option may be useful in
1444 applications that process patterns from external sources. Note that
1445 there is also a build-time option that permanently locks out the use of
1446 \C.
1447 PCRE2_NEVER_UCP
1449 This option locks out the use of Unicode properties for handling \B,
1451 \b, \D, \d, \S, \s, \W, \w, and some of the POSIX character classes, as
1452 described for the PCRE2_UCP option below. In particular, it prevents
1453 the creator of the pattern from enabling this facility by starting the
1454 pattern with (*UCP). This option may be useful in applications that
1455 process patterns from external sources. The option combination PCRE_UCP
1456 and PCRE_NEVER_UCP causes an error.
1457 PCRE2_NEVER_UTF
1459 This option locks out interpretation of the pattern as UTF-8, UTF-16,
1461 or UTF-32, depending on which library is in use. In particular, it pre‐
1462 vents the creator of the pattern from switching to UTF interpretation
1463 by starting the pattern with (*UTF). This option may be useful in
1464 applications that process patterns from external sources. The combina‐
1465 tion of PCRE2_UTF and PCRE2_NEVER_UTF causes an error.
1466 PCRE2_NO_AUTO_CAPTURE
1468 If this option is set, it disables the use of numbered capturing paren‐
1470 theses in the pattern. Any opening parenthesis that is not followed by
1471 ? behaves as if it were followed by ?: but named parentheses can still
1472 be used for capturing (and they acquire numbers in the usual way). This
1473 is the same as Perl's /n option. Note that, when this option is set,
1474 references to capture groups (backreferences or recursion/subroutine
1475 calls) may only refer to named groups, though the reference can be by
1476 name or by number.
1477 PCRE2_NO_AUTO_POSSESS
1479 If this option is set, it disables "auto-possessification", which is an
1481 optimization that, for example, turns a+b into a++b in order to avoid
1482 backtracks into a+ that can never be successful. However, if callouts
1483 are in use, auto-possessification means that some callouts are never
1484 taken. You can set this option if you want the matching functions to do
1485 a full unoptimized search and run all the callouts, but it is mainly
1486 provided for testing purposes.
1487 PCRE2_NO_DOTSTAR_ANCHOR
1489 If this option is set, it disables an optimization that is applied when
1491 .* is the first significant item in a top-level branch of a pattern,
1492 and all the other branches also start with .* or with \A or \G or ^.
1493 The optimization is automatically disabled for .* if it is inside an
1494 atomic group or a capture group that is the subject of a backreference,
1495 or if the pattern contains (*PRUNE) or (*SKIP). When the optimization
1496 is not disabled, such a pattern is automatically anchored if
1497 PCRE2_DOTALL is set for all the .* items and PCRE2_MULTILINE is not set
1498 for any ^ items. Otherwise, the fact that any match must start either
1499 at the start of the subject or following a newline is remembered. Like
1500 other optimizations, this can cause callouts to be skipped.
1501 PCRE2_NO_START_OPTIMIZE
1503 This is an option whose main effect is at matching time. It does not
1505 change what pcre2_compile() generates, but it does affect the output of
1506 the JIT compiler.
1507 There are a number of optimizations that may occur at the start of a
1509 match, in order to speed up the process. For example, if it is known
1510 that an unanchored match must start with a specific code unit value,
1511 the matching code searches the subject for that value, and fails imme‐
1512 diately if it cannot find it, without actually running the main match‐
1513 ing function. This means that a special item such as (*COMMIT) at the
1514 start of a pattern is not considered until after a suitable starting
1515 point for the match has been found. Also, when callouts or (*MARK)
1516 items are in use, these "start-up" optimizations can cause them to be
1517 skipped if the pattern is never actually used. The start-up optimiza‐
1518 tions are in effect a pre-scan of the subject that takes place before
1519 the pattern is run.
1520 The PCRE2_NO_START_OPTIMIZE option disables the start-up optimizations,
1522 possibly causing performance to suffer, but ensuring that in cases
1523 where the result is "no match", the callouts do occur, and that items
1524 such as (*COMMIT) and (*MARK) are considered at every possible starting
1525 position in the subject string.
1526 Setting PCRE2_NO_START_OPTIMIZE may change the outcome of a matching
1528 operation. Consider the pattern
1529 (*COMMIT)ABC
1531 When this is compiled, PCRE2 records the fact that a match must start
1533 with the character "A". Suppose the subject string is "DEFABC". The
1534 start-up optimization scans along the subject, finds "A" and runs the
1535 first match attempt from there. The (*COMMIT) item means that the pat‐
1536 tern must match the current starting position, which in this case, it
1537 does. However, if the same match is run with PCRE2_NO_START_OPTIMIZE
1538 set, the initial scan along the subject string does not happen. The
1539 first match attempt is run starting from "D" and when this fails,
1540 (*COMMIT) prevents any further matches being tried, so the overall
1541 result is "no match".
1542 As another start-up optimization makes use of a minimum length for a
1544 matching subject, which is recorded when possible. Consider the pattern
1545 (*MARK:1)B(*MARK:2)(X|Y)
1547 The minimum length for a match is two characters. If the subject is
1549 "XXBB", the "starting character" optimization skips "XX", then tries to
1550 match "BB", which is long enough. In the process, (*MARK:2) is encoun‐
1551 tered and remembered. When the match attempt fails, the next "B" is
1552 found, but there is only one character left, so there are no more
1553 attempts, and "no match" is returned with the "last mark seen" set to
1554 "2". If NO_START_OPTIMIZE is set, however, matches are tried at every
1555 possible starting position, including at the end of the subject, where
1556 (*MARK:1) is encountered, but there is no "B", so the "last mark seen"
1557 that is returned is "1". In this case, the optimizations do not affect
1558 the overall match result, which is still "no match", but they do affect
1559 the auxiliary information that is returned.
1560 PCRE2_NO_UTF_CHECK
1562 When PCRE2_UTF is set, the validity of the pattern as a UTF string is
1564 automatically checked. There are discussions about the validity of
1565 UTF-8 strings, UTF-16 strings, and UTF-32 strings in the pcre2unicode
1566 document. If an invalid UTF sequence is found, pcre2_compile() returns
1567 a negative error code.
1568 If you know that your pattern is a valid UTF string, and you want to
1570 skip this check for performance reasons, you can set the
1571 PCRE2_NO_UTF_CHECK option. When it is set, the effect of passing an
1572 invalid UTF string as a pattern is undefined. It may cause your program
1573 to crash or loop.
1574 Note that this option can also be passed to pcre2_match() and
1576 pcre_dfa_match(), to suppress UTF validity checking of the subject
1577 string.
1578 Note also that setting PCRE2_NO_UTF_CHECK at compile time does not dis‐
1580 able the error that is given if an escape sequence for an invalid Uni‐
1581 code code point is encountered in the pattern. In particular, the so-
1582 called "surrogate" code points (0xd800 to 0xdfff) are invalid. If you
1583 want to allow escape sequences such as \x{d800} you can set the
1584 PCRE2_EXTRA_ALLOW_SURROGATE_ESCAPES extra option, as described in the
1585 section entitled "Extra compile options" below. However, this is pos‐
1586 sible only in UTF-8 and UTF-32 modes, because these values are not rep‐
1587 resentable in UTF-16.
1588 PCRE2_UCP
1590 This option changes the way PCRE2 processes \B, \b, \D, \d, \S, \s, \W,
1592 \w, and some of the POSIX character classes. By default, only ASCII
1593 characters are recognized, but if PCRE2_UCP is set, Unicode properties
1594 are used instead to classify characters. More details are given in the
1595 section on generic character types in the pcre2pattern page. If you set
1596 PCRE2_UCP, matching one of the items it affects takes much longer. The
1597 option is available only if PCRE2 has been compiled with Unicode sup‐
1598 port (which is the default).
1599 PCRE2_UNGREEDY
1601 This option inverts the "greediness" of the quantifiers so that they
1603 are not greedy by default, but become greedy if followed by "?". It is
1604 not compatible with Perl. It can also be set by a (?U) option setting
1605 within the pattern.
1606 PCRE2_USE_OFFSET_LIMIT
1608 This option must be set for pcre2_compile() if pcre2_set_offset_limit()
1610 is going to be used to set a non-default offset limit in a match con‐
1611 text for matches that use this pattern. An error is generated if an
1612 offset limit is set without this option. For more details, see the
1613 description of pcre2_set_offset_limit() in the section that describes
1614 match contexts. See also the PCRE2_FIRSTLINE option above.
1615 PCRE2_UTF
1617 This option causes PCRE2 to regard both the pattern and the subject
1619 strings that are subsequently processed as strings of UTF characters
1620 instead of single-code-unit strings. It is available when PCRE2 is
1621 built to include Unicode support (which is the default). If Unicode
1622 support is not available, the use of this option provokes an error.
1623 Details of how PCRE2_UTF changes the behaviour of PCRE2 are given in
1624 the pcre2unicode page. In particular, note that it changes the way
1625 PCRE2_CASELESS handles characters with code points greater than 127.
1626 Extra compile options
1628 The option bits that can be set in a compile context by calling the
1630 pcre2_set_compile_extra_options() function are as follows:
1631 PCRE2_EXTRA_ALLOW_SURROGATE_ESCAPES
1633 This option applies when compiling a pattern in UTF-8 or UTF-32 mode.
1635 It is forbidden in UTF-16 mode, and ignored in non-UTF modes. Unicode
1636 "surrogate" code points in the range 0xd800 to 0xdfff are used in pairs
1637 in UTF-16 to encode code points with values in the range 0x10000 to
1638 0x10ffff. The surrogates cannot therefore be represented in UTF-16.
1639 They can be represented in UTF-8 and UTF-32, but are defined as invalid
1640 code points, and cause errors if encountered in a UTF-8 or UTF-32
1641 string that is being checked for validity by PCRE2.
1642 These values also cause errors if encountered in escape sequences such
1644 as \x{d912} within a pattern. However, it seems that some applications,
1645 when using PCRE2 to check for unwanted characters in UTF-8 strings,
1646 explicitly test for the surrogates using escape sequences. The
1647 PCRE2_NO_UTF_CHECK option does not disable the error that occurs,
1648 because it applies only to the testing of input strings for UTF valid‐
1649 ity.
1650 If the extra option PCRE2_EXTRA_ALLOW_SURROGATE_ESCAPES is set, surro‐
1652 gate code point values in UTF-8 and UTF-32 patterns no longer provoke
1653 errors and are incorporated in the compiled pattern. However, they can
1654 only match subject characters if the matching function is called with
1655 PCRE2_NO_UTF_CHECK set.
1656 PCRE2_EXTRA_ALT_BSUX
1658 The original option PCRE2_ALT_BSUX causes PCRE2 to process \U, \u, and
1660 \x in the way that ECMAscript (aka JavaScript) does. Additional func‐
1661 tionality was defined by ECMAscript 6; setting PCRE2_EXTRA_ALT_BSUX has
1662 the effect of PCRE2_ALT_BSUX, but in addition it recognizes \u{hhh..}
1663 as a hexadecimal character code, where hhh.. is any number of hexadeci‐
1664 mal digits.
1665 PCRE2_EXTRA_BAD_ESCAPE_IS_LITERAL
1667 This is a dangerous option. Use with care. By default, an unrecognized
1669 escape such as \j or a malformed one such as \x{2z} causes a compile-
1670 time error when detected by pcre2_compile(). Perl is somewhat inconsis‐
1671 tent in handling such items: for example, \j is treated as a literal
1672 "j", and non-hexadecimal digits in \x{} are just ignored, though warn‐
1673 ings are given in both cases if Perl's warning switch is enabled. How‐
1674 ever, a malformed octal number after \o{ always causes an error in
1675 Perl.
1676 If the PCRE2_EXTRA_BAD_ESCAPE_IS_LITERAL extra option is passed to
1678 pcre2_compile(), all unrecognized or malformed escape sequences are
1679 treated as single-character escapes. For example, \j is a literal "j"
1680 and \x{2z} is treated as the literal string "x{2z}". Setting this
1681 option means that typos in patterns may go undetected and have unex‐
1682 pected results. Also note that a sequence such as [\N{] is interpreted
1683 as a malformed attempt at [\N{...}] and so is treated as [N{] whereas
1684 [\N] gives an error because an unqualified \N is a valid escape
1685 sequence but is not supported in a character class. To reiterate: this
1686 is a dangerous option. Use with great care.
1687 PCRE2_EXTRA_ESCAPED_CR_IS_LF
1689 There are some legacy applications where the escape sequence \r in a
1691 pattern is expected to match a newline. If this option is set, \r in a
1692 pattern is converted to \n so that it matches a LF (linefeed) instead
1693 of a CR (carriage return) character. The option does not affect a lit‐
1694 eral CR in the pattern, nor does it affect CR specified as an explicit
1695 code point such as \x{0D}.
1696 PCRE2_EXTRA_MATCH_LINE
1698 This option is provided for use by the -x option of pcre2grep. It
1700 causes the pattern only to match complete lines. This is achieved by
1701 automatically inserting the code for "^(?:" at the start of the com‐
1702 piled pattern and ")$" at the end. Thus, when PCRE2_MULTILINE is set,
1703 the matched line may be in the middle of the subject string. This
1704 option can be used with PCRE2_LITERAL.
1705 PCRE2_EXTRA_MATCH_WORD
1707 This option is provided for use by the -w option of pcre2grep. It
1709 causes the pattern only to match strings that have a word boundary at
1710 the start and the end. This is achieved by automatically inserting the
1711 code for "\b(?:" at the start of the compiled pattern and ")\b" at the
1712 end. The option may be used with PCRE2_LITERAL. However, it is ignored
1713 if PCRE2_EXTRA_MATCH_LINE is also set.
1714
1716 int pcre2_jit_compile(pcre2_code *code, uint32_t options);
1718 int pcre2_jit_match(const pcre2_code *code, PCRE2_SPTR subject,
1720 PCRE2_SIZE length, PCRE2_SIZE startoffset,
1721 uint32_t options, pcre2_match_data *match_data,
1722 pcre2_match_context *mcontext);
1723 void pcre2_jit_free_unused_memory(pcre2_general_context *gcontext);
1725 pcre2_jit_stack *pcre2_jit_stack_create(PCRE2_SIZE startsize,
1727 PCRE2_SIZE maxsize, pcre2_general_context *gcontext);
1728 void pcre2_jit_stack_assign(pcre2_match_context *mcontext,
1730 pcre2_jit_callback callback_function, void *callback_data);
1731 void pcre2_jit_stack_free(pcre2_jit_stack *jit_stack);
1733 These functions provide support for JIT compilation, which, if the
1735 just-in-time compiler is available, further processes a compiled pat‐
1736 tern into machine code that executes much faster than the pcre2_match()
1737 interpretive matching function. Full details are given in the pcre2jit
1738 documentation.
1739 JIT compilation is a heavyweight optimization. It can take some time
1741 for patterns to be analyzed, and for one-off matches and simple pat‐
1742 terns the benefit of faster execution might be offset by a much slower
1743 compilation time. Most (but not all) patterns can be optimized by the
1744 JIT compiler.
1745
1747 const uint8_t *pcre2_maketables(pcre2_general_context *gcontext);
1749 void pcre2_maketables_free(pcre2_general_context *gcontext,
1751 const uint8_t *tables);
1752 PCRE2 handles caseless matching, and determines whether characters are
1754 letters, digits, or whatever, by reference to a set of tables, indexed
1755 by character code point. However, this applies only to characters whose
1756 code points are less than 256. By default, higher-valued code points
1757 never match escapes such as \w or \d. When PCRE2 is built with Unicode
1758 support (the default), all characters can be tested with \p and \P, or,
1759 alternatively, the PCRE2_UCP option can be set when a pattern is com‐
1760 piled; this causes \w and friends to use Unicode property support
1761 instead of the built-in tables.
1762 The use of locales with Unicode is discouraged. If you are handling
1764 characters with code points greater than 128, you should either use
1765 Unicode support, or use locales, but not try to mix the two.
1766 PCRE2 contains a built-in set of character tables that are used by
1768 default. These are sufficient for many applications. Normally, the
1769 internal tables recognize only ASCII characters. However, when PCRE2 is
1770 built, it is possible to cause the internal tables to be rebuilt in the
1771 default "C" locale of the local system, which may cause them to be dif‐
1772 ferent.
1773 The built-in tables can be overridden by tables supplied by the appli‐
1775 cation that calls PCRE2. These may be created in a different locale
1776 from the default. As more and more applications change to using Uni‐
1777 code, the need for this locale support is expected to die away.
1778 External tables are built by calling the pcre2_maketables() function,
1780 in the relevant locale. The only argument to this function is a general
1781 context, which can be used to pass a custom memory allocator. If the
1782 argument is NULL, the system malloc() is used. The result can be passed
1783 to pcre2_compile() as often as necessary, by creating a compile context
1784 and calling pcre2_set_character_tables() to set the tables pointer
1785 therein.
1786 For example, to build and use tables that are appropriate for the
1788 French locale (where accented characters with values greater than 128
1789 are treated as letters), the following code could be used:
1790 setlocale(LC_CTYPE, "fr_FR");
1792 tables = pcre2_maketables(NULL);
1793 ccontext = pcre2_compile_context_create(NULL);
1794 pcre2_set_character_tables(ccontext, tables);
1795 re = pcre2_compile(..., ccontext);
1796 The locale name "fr_FR" is used on Linux and other Unix-like systems;
1798 if you are using Windows, the name for the French locale is "french".
1799 The pointer that is passed (via the compile context) to pcre2_compile()
1801 is saved with the compiled pattern, and the same tables are used by
1802 pcre2_match() and pcre_dfa_match(). Thus, for any single pattern, com‐
1803 pilation and matching both happen in the same locale, but different
1804 patterns can be processed in different locales.
1805 It is the caller's responsibility to ensure that the memory containing
1807 the tables remains available while they are still in use. When they are
1808 no longer needed, you can discard them using pcre2_maketables_free(),
1809 which should pass as its first parameter the same global context that
1810 was used to create the tables.
1811
1813 int pcre2_pattern_info(const pcre2 *code, uint32_t what, void *where);
1815 The pcre2_pattern_info() function returns general information about a
1817 compiled pattern. For information about callouts, see the next section.
1818 The first argument for pcre2_pattern_info() is a pointer to the com‐
1819 piled pattern. The second argument specifies which piece of information
1820 is required, and the third argument is a pointer to a variable to
1821 receive the data. If the third argument is NULL, the first argument is
1822 ignored, and the function returns the size in bytes of the variable
1823 that is required for the information requested. Otherwise, the yield of
1824 the function is zero for success, or one of the following negative num‐
1825 bers:
1826 PCRE2_ERROR_NULL the argument code was NULL
1828 PCRE2_ERROR_BADMAGIC the "magic number" was not found
1829 PCRE2_ERROR_BADOPTION the value of what was invalid
1830 PCRE2_ERROR_UNSET the requested field is not set
1831 The "magic number" is placed at the start of each compiled pattern as a
1833 simple check against passing an arbitrary memory pointer. Here is a
1834 typical call of pcre2_pattern_info(), to obtain the length of the com‐
1835 piled pattern:
1836 int rc;
1838 size_t length;
1839 rc = pcre2_pattern_info(
1840 re, /* result of pcre2_compile() */
1841 PCRE2_INFO_SIZE, /* what is required */
1842 &length); /* where to put the data */
1843 The possible values for the second argument are defined in pcre2.h, and
1845 are as follows:
1846 PCRE2_INFO_ALLOPTIONS
1848 PCRE2_INFO_ARGOPTIONS
1849 PCRE2_INFO_EXTRAOPTIONS
1850 Return copies of the pattern's options. The third argument should point
1852 to a uint32_t variable. PCRE2_INFO_ARGOPTIONS returns exactly the
1853 options that were passed to pcre2_compile(), whereas PCRE2_INFO_ALLOP‐
1854 TIONS returns the compile options as modified by any top-level (*XXX)
1855 option settings such as (*UTF) at the start of the pattern itself.
1856 PCRE2_INFO_EXTRAOPTIONS returns the extra options that were set in the
1857 compile context by calling the pcre2_set_compile_extra_options() func‐
1858 tion.
1859 For example, if the pattern /(*UTF)abc/ is compiled with the
1861 PCRE2_EXTENDED option, the result for PCRE2_INFO_ALLOPTIONS is
1862 PCRE2_EXTENDED and PCRE2_UTF. Option settings such as (?i) that can
1863 change within a pattern do not affect the result of PCRE2_INFO_ALLOP‐
1864 TIONS, even if they appear right at the start of the pattern. (This was
1865 different in some earlier releases.)
1866 A pattern compiled without PCRE2_ANCHORED is automatically anchored by
1868 PCRE2 if the first significant item in every top-level branch is one of
1869 the following:
1870 ^ unless PCRE2_MULTILINE is set
1872 \A always
1873 \G always
1874 .* sometimes - see below
1875 When .* is the first significant item, anchoring is possible only when
1877 all the following are true:
1878 .* is not in an atomic group
1880 .* is not in a capture group that is the subject
1881 of a backreference
1882 PCRE2_DOTALL is in force for .*
1883 Neither (*PRUNE) nor (*SKIP) appears in the pattern
1884 PCRE2_NO_DOTSTAR_ANCHOR is not set
1885 For patterns that are auto-anchored, the PCRE2_ANCHORED bit is set in
1887 the options returned for PCRE2_INFO_ALLOPTIONS.
1888 PCRE2_INFO_BACKREFMAX
1890 Return the number of the highest backreference in the pattern. The
1892 third argument should point to a uint32_t variable. Named capture
1893 groups acquire numbers as well as names, and these count towards the
1894 highest backreference. Backreferences such as \4 or \g{12} match the
1895 captured characters of the given group, but in addition, the check that
1896 a capture group is set in a conditional group such as (?(3)a|b) is also
1897 a backreference. Zero is returned if there are no backreferences.
1898 PCRE2_INFO_BSR
1900 The output is a uint32_t integer whose value indicates what character
1902 sequences the \R escape sequence matches. A value of PCRE2_BSR_UNICODE
1903 means that \R matches any Unicode line ending sequence; a value of
1904 PCRE2_BSR_ANYCRLF means that \R matches only CR, LF, or CRLF.
1905 PCRE2_INFO_CAPTURECOUNT
1907 Return the highest capture group number in the pattern. In patterns
1909 where (?| is not used, this is also the total number of capture groups.
1910 The third argument should point to a uint32_t variable.
1911 PCRE2_INFO_DEPTHLIMIT
1913 If the pattern set a backtracking depth limit by including an item of
1915 the form (*LIMIT_DEPTH=nnnn) at the start, the value is returned. The
1916 third argument should point to a uint32_t integer. If no such value has
1917 been set, the call to pcre2_pattern_info() returns the error
1918 PCRE2_ERROR_UNSET. Note that this limit will only be used during match‐
1919 ing if it is less than the limit set or defaulted by the caller of the
1920 match function.
1921 PCRE2_INFO_FIRSTBITMAP
1923 In the absence of a single first code unit for a non-anchored pattern,
1925 pcre2_compile() may construct a 256-bit table that defines a fixed set
1926 of values for the first code unit in any match. For example, a pattern
1927 that starts with [abc] results in a table with three bits set. When
1928 code unit values greater than 255 are supported, the flag bit for 255
1929 means "any code unit of value 255 or above". If such a table was con‐
1930 structed, a pointer to it is returned. Otherwise NULL is returned. The
1931 third argument should point to a const uint8_t * variable.
1932 PCRE2_INFO_FIRSTCODETYPE
1934 Return information about the first code unit of any matched string, for
1936 a non-anchored pattern. The third argument should point to a uint32_t
1937 variable. If there is a fixed first value, for example, the letter "c"
1938 from a pattern such as (cat|cow|coyote), 1 is returned, and the value
1939 can be retrieved using PCRE2_INFO_FIRSTCODEUNIT. If there is no fixed
1940 first value, but it is known that a match can occur only at the start
1941 of the subject or following a newline in the subject, 2 is returned.
1942 Otherwise, and for anchored patterns, 0 is returned.
1943 PCRE2_INFO_FIRSTCODEUNIT
1945 Return the value of the first code unit of any matched string for a
1947 pattern where PCRE2_INFO_FIRSTCODETYPE returns 1; otherwise return 0.
1948 The third argument should point to a uint32_t variable. In the 8-bit
1949 library, the value is always less than 256. In the 16-bit library the
1950 value can be up to 0xffff. In the 32-bit library in UTF-32 mode the
1951 value can be up to 0x10ffff, and up to 0xffffffff when not using UTF-32
1952 mode.
1953 PCRE2_INFO_FRAMESIZE
1955 Return the size (in bytes) of the data frames that are used to remember
1957 backtracking positions when the pattern is processed by pcre2_match()
1958 without the use of JIT. The third argument should point to a size_t
1959 variable. The frame size depends on the number of capturing parentheses
1960 in the pattern. Each additional capture group adds two PCRE2_SIZE vari‐
1961 ables.
1962 PCRE2_INFO_HASBACKSLASHC
1964 Return 1 if the pattern contains any instances of \C, otherwise 0. The
1966 third argument should point to a uint32_t variable.
1967 PCRE2_INFO_HASCRORLF
1969 Return 1 if the pattern contains any explicit matches for CR or LF
1971 characters, otherwise 0. The third argument should point to a uint32_t
1972 variable. An explicit match is either a literal CR or LF character, or
1973 \r or \n or one of the equivalent hexadecimal or octal escape
1974 sequences.
1975 PCRE2_INFO_HEAPLIMIT
1977 If the pattern set a heap memory limit by including an item of the form
1979 (*LIMIT_HEAP=nnnn) at the start, the value is returned. The third argu‐
1980 ment should point to a uint32_t integer. If no such value has been set,
1981 the call to pcre2_pattern_info() returns the error PCRE2_ERROR_UNSET.
1982 Note that this limit will only be used during matching if it is less
1983 than the limit set or defaulted by the caller of the match function.
1984 PCRE2_INFO_JCHANGED
1986 Return 1 if the (?J) or (?-J) option setting is used in the pattern,
1988 otherwise 0. The third argument should point to a uint32_t variable.
1989 (?J) and (?-J) set and unset the local PCRE2_DUPNAMES option, respec‐
1990 tively.
1991 PCRE2_INFO_JITSIZE
1993 If the compiled pattern was successfully processed by pcre2_jit_com‐
1995 pile(), return the size of the JIT compiled code, otherwise return
1996 zero. The third argument should point to a size_t variable.
1997 PCRE2_INFO_LASTCODETYPE
1999 Returns 1 if there is a rightmost literal code unit that must exist in
2001 any matched string, other than at its start. The third argument should
2002 point to a uint32_t variable. If there is no such value, 0 is returned.
2003 When 1 is returned, the code unit value itself can be retrieved using
2004 PCRE2_INFO_LASTCODEUNIT. For anchored patterns, a last literal value is
2005 recorded only if it follows something of variable length. For example,
2006 for the pattern /^a\d+z\d+/ the returned value is 1 (with "z" returned
2007 from PCRE2_INFO_LASTCODEUNIT), but for /^a\dz\d/ the returned value is
2008 0.
2009 PCRE2_INFO_LASTCODEUNIT
2011 Return the value of the rightmost literal code unit that must exist in
2013 any matched string, other than at its start, for a pattern where
2014 PCRE2_INFO_LASTCODETYPE returns 1. Otherwise, return 0. The third argu‐
2015 ment should point to a uint32_t variable.
2016 PCRE2_INFO_MATCHEMPTY
2018 Return 1 if the pattern might match an empty string, otherwise 0. The
2020 third argument should point to a uint32_t variable. When a pattern con‐
2021 tains recursive subroutine calls it is not always possible to determine
2022 whether or not it can match an empty string. PCRE2 takes a cautious
2023 approach and returns 1 in such cases.
2024 PCRE2_INFO_MATCHLIMIT
2026 If the pattern set a match limit by including an item of the form
2028 (*LIMIT_MATCH=nnnn) at the start, the value is returned. The third
2029 argument should point to a uint32_t integer. If no such value has been
2030 set, the call to pcre2_pattern_info() returns the error
2031 PCRE2_ERROR_UNSET. Note that this limit will only be used during match‐
2032 ing if it is less than the limit set or defaulted by the caller of the
2033 match function.
2034 PCRE2_INFO_MAXLOOKBEHIND
2036 A lookbehind assertion moves back a certain number of characters (not
2038 code units) when it starts to process each of its branches. This
2039 request returns the largest of these backward moves. The third argument
2040 should point to a uint32_t integer. The simple assertions \b and \B
2041 require a one-character lookbehind and cause PCRE2_INFO_MAXLOOKBEHIND
2042 to return 1 in the absence of anything longer. \A also registers a one-
2043 character lookbehind, though it does not actually inspect the previous
2044 character.
2045 Note that this information is useful for multi-segment matching only if
2047 the pattern contains no nested lookbehinds. For example, the pattern
2048 (?<=a(?<=ba)c) returns a maximum lookbehind of 2, but when it is pro‐
2049 cessed, the first lookbehind moves back by two characters, matches one
2050 character, then the nested lookbehind also moves back by two charac‐
2051 ters. This puts the matching point three characters earlier than it was
2052 at the start. PCRE2_INFO_MAXLOOKBEHIND is really only useful as a
2053 debugging tool. See the pcre2partial documentation for a discussion of
2054 multi-segment matching.
2055 PCRE2_INFO_MINLENGTH
2057 If a minimum length for matching subject strings was computed, its
2059 value is returned. Otherwise the returned value is 0. This value is not
2060 computed when PCRE2_NO_START_OPTIMIZE is set. The value is a number of
2061 characters, which in UTF mode may be different from the number of code
2062 units. The third argument should point to a uint32_t variable. The
2063 value is a lower bound to the length of any matching string. There may
2064 not be any strings of that length that do actually match, but every
2065 string that does match is at least that long.
2066 PCRE2_INFO_NAMECOUNT
2068 PCRE2_INFO_NAMEENTRYSIZE
2069 PCRE2_INFO_NAMETABLE
2070 PCRE2 supports the use of named as well as numbered capturing parenthe‐
2072 ses. The names are just an additional way of identifying the parenthe‐
2073 ses, which still acquire numbers. Several convenience functions such as
2074 pcre2_substring_get_byname() are provided for extracting captured sub‐
2075 strings by name. It is also possible to extract the data directly, by
2076 first converting the name to a number in order to access the correct
2077 pointers in the output vector (described with pcre2_match() below). To
2078 do the conversion, you need to use the name-to-number map, which is
2079 described by these three values.
2080 The map consists of a number of fixed-size entries. PCRE2_INFO_NAME‐
2082 COUNT gives the number of entries, and PCRE2_INFO_NAMEENTRYSIZE gives
2083 the size of each entry in code units; both of these return a uint32_t
2084 value. The entry size depends on the length of the longest name.
2085 PCRE2_INFO_NAMETABLE returns a pointer to the first entry of the table.
2087 This is a PCRE2_SPTR pointer to a block of code units. In the 8-bit
2088 library, the first two bytes of each entry are the number of the cap‐
2089 turing parenthesis, most significant byte first. In the 16-bit library,
2090 the pointer points to 16-bit code units, the first of which contains
2091 the parenthesis number. In the 32-bit library, the pointer points to
2092 32-bit code units, the first of which contains the parenthesis number.
2093 The rest of the entry is the corresponding name, zero terminated.
2094 The names are in alphabetical order. If (?| is used to create multiple
2096 capture groups with the same number, as described in the section on
2097 duplicate group numbers in the pcre2pattern page, the groups may be
2098 given the same name, but there is only one entry in the table. Differ‐
2099 ent names for groups of the same number are not permitted.
2100 Duplicate names for capture groups with different numbers are permit‐
2102 ted, but only if PCRE2_DUPNAMES is set. They appear in the table in the
2103 order in which they were found in the pattern. In the absence of (?|
2104 this is the order of increasing number; when (?| is used this is not
2105 necessarily the case because later capture groups may have lower num‐
2106 bers.
2107 As a simple example of the name/number table, consider the following
2109 pattern after compilation by the 8-bit library (assume PCRE2_EXTENDED
2110 is set, so white space - including newlines - is ignored):
2111 (?<date> (?<year>(\d\d)?\d\d) -
2113 (?<month>\d\d) - (?<day>\d\d) )
2114 There are four named capture groups, so the table has four entries, and
2116 each entry in the table is eight bytes long. The table is as follows,
2117 with non-printing bytes shows in hexadecimal, and undefined bytes shown
2118 as ??:
2119 00 01 d a t e 00 ??
2121 00 05 d a y 00 ?? ??
2122 00 04 m o n t h 00
2123 00 02 y e a r 00 ??
2124 When writing code to extract data from named capture groups using the
2126 name-to-number map, remember that the length of the entries is likely
2127 to be different for each compiled pattern.
2128 PCRE2_INFO_NEWLINE
2130 The output is one of the following uint32_t values:
2132 PCRE2_NEWLINE_CR Carriage return (CR)
2134 PCRE2_NEWLINE_LF Linefeed (LF)
2135 PCRE2_NEWLINE_CRLF Carriage return, linefeed (CRLF)
2136 PCRE2_NEWLINE_ANY Any Unicode line ending
2137 PCRE2_NEWLINE_ANYCRLF Any of CR, LF, or CRLF
2138 PCRE2_NEWLINE_NUL The NUL character (binary zero)
2139 This identifies the character sequence that will be recognized as mean‐
2141 ing "newline" while matching.
2142 PCRE2_INFO_SIZE
2144 Return the size of the compiled pattern in bytes (for all three
2146 libraries). The third argument should point to a size_t variable. This
2147 value includes the size of the general data block that precedes the
2148 code units of the compiled pattern itself. The value that is used when
2149 pcre2_compile() is getting memory in which to place the compiled pat‐
2150 tern may be slightly larger than the value returned by this option,
2151 because there are cases where the code that calculates the size has to
2152 over-estimate. Processing a pattern with the JIT compiler does not
2153 alter the value returned by this option.
2154
2156 int pcre2_callout_enumerate(const pcre2_code *code,
2158 int (*callback)(pcre2_callout_enumerate_block *, void *),
2159 void *user_data);
2160 A script language that supports the use of string arguments in callouts
2162 might like to scan all the callouts in a pattern before running the
2163 match. This can be done by calling pcre2_callout_enumerate(). The first
2164 argument is a pointer to a compiled pattern, the second points to a
2165 callback function, and the third is arbitrary user data. The callback
2166 function is called for every callout in the pattern in the order in
2167 which they appear. Its first argument is a pointer to a callout enumer‐
2168 ation block, and its second argument is the user_data value that was
2169 passed to pcre2_callout_enumerate(). The contents of the callout enu‐
2170 meration block are described in the pcre2callout documentation, which
2171 also gives further details about callouts.
2172
2174 It is possible to save compiled patterns on disc or elsewhere, and
2176 reload them later, subject to a number of restrictions. The host on
2177 which the patterns are reloaded must be running the same version of
2178 PCRE2, with the same code unit width, and must also have the same endi‐
2179 anness, pointer width, and PCRE2_SIZE type. Before compiled patterns
2180 can be saved, they must be converted to a "serialized" form, which in
2181 the case of PCRE2 is really just a bytecode dump. The functions whose
2182 names begin with pcre2_serialize_ are used for converting to and from
2183 the serialized form. They are described in the pcre2serialize documen‐
2184 tation. Note that PCRE2 serialization does not convert compiled pat‐
2185 terns to an abstract format like Java or .NET serialization.
2186
2188 pcre2_match_data *pcre2_match_data_create(uint32_t ovecsize,
2190 pcre2_general_context *gcontext);
2191 pcre2_match_data *pcre2_match_data_create_from_pattern(
2193 const pcre2_code *code, pcre2_general_context *gcontext);
2194 void pcre2_match_data_free(pcre2_match_data *match_data);
2196 Information about a successful or unsuccessful match is placed in a
2198 match data block, which is an opaque structure that is accessed by
2199 function calls. In particular, the match data block contains a vector
2200 of offsets into the subject string that define the matched part of the
2201 subject and any substrings that were captured. This is known as the
2202 ovector.
2203 Before calling pcre2_match(), pcre2_dfa_match(), or pcre2_jit_match()
2205 you must create a match data block by calling one of the creation func‐
2206 tions above. For pcre2_match_data_create(), the first argument is the
2207 number of pairs of offsets in the ovector. One pair of offsets is
2208 required to identify the string that matched the whole pattern, with an
2209 additional pair for each captured substring. For example, a value of 4
2210 creates enough space to record the matched portion of the subject plus
2211 three captured substrings. A minimum of at least 1 pair is imposed by
2212 pcre2_match_data_create(), so it is always possible to return the over‐
2213 all matched string.
2214 The second argument of pcre2_match_data_create() is a pointer to a gen‐
2216 eral context, which can specify custom memory management for obtaining
2217 the memory for the match data block. If you are not using custom memory
2218 management, pass NULL, which causes malloc() to be used.
2219 For pcre2_match_data_create_from_pattern(), the first argument is a
2221 pointer to a compiled pattern. The ovector is created to be exactly the
2222 right size to hold all the substrings a pattern might capture. The sec‐
2223 ond argument is again a pointer to a general context, but in this case
2224 if NULL is passed, the memory is obtained using the same allocator that
2225 was used for the compiled pattern (custom or default).
2226 A match data block can be used many times, with the same or different
2228 compiled patterns. You can extract information from a match data block
2229 after a match operation has finished, using functions that are
2230 described in the sections on matched strings and other match data
2231 below.
2232 When a call of pcre2_match() fails, valid data is available in the
2234 match block only when the error is PCRE2_ERROR_NOMATCH,
2235 PCRE2_ERROR_PARTIAL, or one of the error codes for an invalid UTF
2236 string. Exactly what is available depends on the error, and is detailed
2237 below.
2238 When one of the matching functions is called, pointers to the compiled
2240 pattern and the subject string are set in the match data block so that
2241 they can be referenced by the extraction functions after a successful
2242 match. After running a match, you must not free a compiled pattern or a
2243 subject string until after all operations on the match data block (for
2244 that match) have taken place, unless, in the case of the subject
2245 string, you have used the PCRE2_COPY_MATCHED_SUBJECT option, which is
2246 described in the section entitled "Option bits for pcre2_match()"
2247 below.
2248 When a match data block itself is no longer needed, it should be freed
2250 by calling pcre2_match_data_free(). If this function is called with a
2251 NULL argument, it returns immediately, without doing anything.
2252
2254 int pcre2_match(const pcre2_code *code, PCRE2_SPTR subject,
2256 PCRE2_SIZE length, PCRE2_SIZE startoffset,
2257 uint32_t options, pcre2_match_data *match_data,
2258 pcre2_match_context *mcontext);
2259 The function pcre2_match() is called to match a subject string against
2261 a compiled pattern, which is passed in the code argument. You can call
2262 pcre2_match() with the same code argument as many times as you like, in
2263 order to find multiple matches in the subject string or to match dif‐
2264 ferent subject strings with the same pattern.
2265 This function is the main matching facility of the library, and it
2267 operates in a Perl-like manner. For specialist use there is also an
2268 alternative matching function, which is described below in the section
2269 about the pcre2_dfa_match() function.
2270 Here is an example of a simple call to pcre2_match():
2272 pcre2_match_data *md = pcre2_match_data_create(4, NULL);
2274 int rc = pcre2_match(
2275 re, /* result of pcre2_compile() */
2276 "some string", /* the subject string */
2277 11, /* the length of the subject string */
2278 0, /* start at offset 0 in the subject */
2279 0, /* default options */
2280 md, /* the match data block */
2281 NULL); /* a match context; NULL means use defaults */
2282 If the subject string is zero-terminated, the length can be given as
2284 PCRE2_ZERO_TERMINATED. A match context must be provided if certain less
2285 common matching parameters are to be changed. For details, see the sec‐
2286 tion on the match context above.
2287 The string to be matched by pcre2_match()
2289 The subject string is passed to pcre2_match() as a pointer in subject,
2291 a length in length, and a starting offset in startoffset. The length
2292 and offset are in code units, not characters. That is, they are in
2293 bytes for the 8-bit library, 16-bit code units for the 16-bit library,
2294 and 32-bit code units for the 32-bit library, whether or not UTF pro‐
2295 cessing is enabled.
2296 If startoffset is greater than the length of the subject, pcre2_match()
2298 returns PCRE2_ERROR_BADOFFSET. When the starting offset is zero, the
2299 search for a match starts at the beginning of the subject, and this is
2300 by far the most common case. In UTF-8 or UTF-16 mode, the starting off‐
2301 set must point to the start of a character, or to the end of the sub‐
2302 ject (in UTF-32 mode, one code unit equals one character, so all off‐
2303 sets are valid). Like the pattern string, the subject may contain
2304 binary zeros.
2305 A non-zero starting offset is useful when searching for another match
2307 in the same subject by calling pcre2_match() again after a previous
2308 success. Setting startoffset differs from passing over a shortened
2309 string and setting PCRE2_NOTBOL in the case of a pattern that begins
2310 with any kind of lookbehind. For example, consider the pattern
2311 \Biss\B
2313 which finds occurrences of "iss" in the middle of words. (\B matches
2315 only if the current position in the subject is not a word boundary.)
2316 When applied to the string "Mississipi" the first call to pcre2_match()
2317 finds the first occurrence. If pcre2_match() is called again with just
2318 the remainder of the subject, namely "issipi", it does not match,
2319 because \B is always false at the start of the subject, which is deemed
2320 to be a word boundary. However, if pcre2_match() is passed the entire
2321 string again, but with startoffset set to 4, it finds the second occur‐
2322 rence of "iss" because it is able to look behind the starting point to
2323 discover that it is preceded by a letter.
2324 Finding all the matches in a subject is tricky when the pattern can
2326 match an empty string. It is possible to emulate Perl's /g behaviour by
2327 first trying the match again at the same offset, with the
2328 PCRE2_NOTEMPTY_ATSTART and PCRE2_ANCHORED options, and then if that
2329 fails, advancing the starting offset and trying an ordinary match
2330 again. There is some code that demonstrates how to do this in the
2331 pcre2demo sample program. In the most general case, you have to check
2332 to see if the newline convention recognizes CRLF as a newline, and if
2333 so, and the current character is CR followed by LF, advance the start‐
2334 ing offset by two characters instead of one.
2335 If a non-zero starting offset is passed when the pattern is anchored, a
2337 single attempt to match at the given offset is made. This can only suc‐
2338 ceed if the pattern does not require the match to be at the start of
2339 the subject. In other words, the anchoring must be the result of set‐
2340 ting the PCRE2_ANCHORED option or the use of .* with PCRE2_DOTALL, not
2341 by starting the pattern with ^ or \A.
2342 Option bits for pcre2_match()
2344 The unused bits of the options argument for pcre2_match() must be zero.
2346 The only bits that may be set are PCRE2_ANCHORED,
2347 PCRE2_COPY_MATCHED_SUBJECT, PCRE2_ENDANCHORED, PCRE2_NOTBOL,
2348 PCRE2_NOTEOL, PCRE2_NOTEMPTY, PCRE2_NOTEMPTY_ATSTART, PCRE2_NO_JIT,
2349 PCRE2_NO_UTF_CHECK, PCRE2_PARTIAL_HARD, and PCRE2_PARTIAL_SOFT. Their
2350 action is described below.
2351 Setting PCRE2_ANCHORED or PCRE2_ENDANCHORED at match time is not sup‐
2353 ported by the just-in-time (JIT) compiler. If it is set, JIT matching
2354 is disabled and the interpretive code in pcre2_match() is run. Apart
2355 from PCRE2_NO_JIT (obviously), the remaining options are supported for
2356 JIT matching.
2357 PCRE2_ANCHORED
2359 The PCRE2_ANCHORED option limits pcre2_match() to matching at the first
2361 matching position. If a pattern was compiled with PCRE2_ANCHORED, or
2362 turned out to be anchored by virtue of its contents, it cannot be made
2363 unachored at matching time. Note that setting the option at match time
2364 disables JIT matching.
2365 PCRE2_COPY_MATCHED_SUBJECT
2367 By default, a pointer to the subject is remembered in the match data
2369 block so that, after a successful match, it can be referenced by the
2370 substring extraction functions. This means that the subject's memory
2371 must not be freed until all such operations are complete. For some
2372 applications where the lifetime of the subject string is not guaran‐
2373 teed, it may be necessary to make a copy of the subject string, but it
2374 is wasteful to do this unless the match is successful. After a success‐
2375 ful match, if PCRE2_COPY_MATCHED_SUBJECT is set, the subject is copied
2376 and the new pointer is remembered in the match data block instead of
2377 the original subject pointer. The memory allocator that was used for
2378 the match block itself is used. The copy is automatically freed when
2379 pcre2_match_data_free() is called to free the match data block. It is
2380 also automatically freed if the match data block is re-used for another
2381 match operation.
2382 PCRE2_ENDANCHORED
2384 If the PCRE2_ENDANCHORED option is set, any string that pcre2_match()
2386 matches must be right at the end of the subject string. Note that set‐
2387 ting the option at match time disables JIT matching.
2388 PCRE2_NOTBOL
2390 This option specifies that first character of the subject string is not
2392 the beginning of a line, so the circumflex metacharacter should not
2393 match before it. Setting this without having set PCRE2_MULTILINE at
2394 compile time causes circumflex never to match. This option affects only
2395 the behaviour of the circumflex metacharacter. It does not affect \A.
2396 PCRE2_NOTEOL
2398 This option specifies that the end of the subject string is not the end
2400 of a line, so the dollar metacharacter should not match it nor (except
2401 in multiline mode) a newline immediately before it. Setting this with‐
2402 out having set PCRE2_MULTILINE at compile time causes dollar never to
2403 match. This option affects only the behaviour of the dollar metacharac‐
2404 ter. It does not affect \Z or \z.
2405 PCRE2_NOTEMPTY
2407 An empty string is not considered to be a valid match if this option is
2409 set. If there are alternatives in the pattern, they are tried. If all
2410 the alternatives match the empty string, the entire match fails. For
2411 example, if the pattern
2412 a?b?
2414 is applied to a string not beginning with "a" or "b", it matches an
2416 empty string at the start of the subject. With PCRE2_NOTEMPTY set, this
2417 match is not valid, so pcre2_match() searches further into the string
2418 for occurrences of "a" or "b".
2419 PCRE2_NOTEMPTY_ATSTART
2421 This is like PCRE2_NOTEMPTY, except that it locks out an empty string
2423 match only at the first matching position, that is, at the start of the
2424 subject plus the starting offset. An empty string match later in the
2425 subject is permitted. If the pattern is anchored, such a match can
2426 occur only if the pattern contains \K.
2427 PCRE2_NO_JIT
2429 By default, if a pattern has been successfully processed by
2431 pcre2_jit_compile(), JIT is automatically used when pcre2_match() is
2432 called with options that JIT supports. Setting PCRE2_NO_JIT disables
2433 the use of JIT; it forces matching to be done by the interpreter.
2434 PCRE2_NO_UTF_CHECK
2436 When PCRE2_UTF is set at compile time, the validity of the subject as a
2438 UTF string is checked unless PCRE2_NO_UTF_CHECK is passed to
2439 pcre2_match() or PCRE2_MATCH_INVALID_UTF was passed to pcre2_compile().
2440 The latter special case is discussed in detail in the pcre2unicode doc‐
2441 umentation.
2442 In the default case, if a non-zero starting offset is given, the check
2444 is applied only to that part of the subject that could be inspected
2445 during matching, and there is a check that the starting offset points
2446 to the first code unit of a character or to the end of the subject. If
2447 there are no lookbehind assertions in the pattern, the check starts at
2448 the starting offset. Otherwise, it starts at the length of the longest
2449 lookbehind before the starting offset, or at the start of the subject
2450 if there are not that many characters before the starting offset. Note
2451 that the sequences \b and \B are one-character lookbehinds.
2452 The check is carried out before any other processing takes place, and a
2454 negative error code is returned if the check fails. There are several
2455 UTF error codes for each code unit width, corresponding to different
2456 problems with the code unit sequence. There are discussions about the
2457 validity of UTF-8 strings, UTF-16 strings, and UTF-32 strings in the
2458 pcre2unicode documentation.
2459 If you know that your subject is valid, and you want to skip this check
2461 for performance reasons, you can set the PCRE2_NO_UTF_CHECK option when
2462 calling pcre2_match(). You might want to do this for the second and
2463 subsequent calls to pcre2_match() if you are making repeated calls to
2464 find multiple matches in the same subject string.
2465 Warning: Unless PCRE2_MATCH_INVALID_UTF was set at compile time, when
2467 PCRE2_NO_UTF_CHECK is set at match time the effect of passing an
2468 invalid string as a subject, or an invalid value of startoffset, is
2469 undefined. Your program may crash or loop indefinitely or give wrong
2470 results.
2471 PCRE2_PARTIAL_HARD
2473 PCRE2_PARTIAL_SOFT
2474 These options turn on the partial matching feature. A partial match
2476 occurs if the end of the subject string is reached successfully, but
2477 there are not enough subject characters to complete the match. In addi‐
2478 tion, either at least one character must have been inspected or the
2479 pattern must contain a lookbehind, or the pattern must be one that
2480 could match an empty string.
2481 If this situation arises when PCRE2_PARTIAL_SOFT (but not PCRE2_PAR‐
2483 TIAL_HARD) is set, matching continues by testing any remaining alterna‐
2484 tives. Only if no complete match can be found is PCRE2_ERROR_PARTIAL
2485 returned instead of PCRE2_ERROR_NOMATCH. In other words, PCRE2_PAR‐
2486 TIAL_SOFT specifies that the caller is prepared to handle a partial
2487 match, but only if no complete match can be found.
2488 If PCRE2_PARTIAL_HARD is set, it overrides PCRE2_PARTIAL_SOFT. In this
2490 case, if a partial match is found, pcre2_match() immediately returns
2491 PCRE2_ERROR_PARTIAL, without considering any other alternatives. In
2492 other words, when PCRE2_PARTIAL_HARD is set, a partial match is consid‐
2493 ered to be more important that an alternative complete match.
2494 There is a more detailed discussion of partial and multi-segment match‐
2496 ing, with examples, in the pcre2partial documentation.
2497
2499 When PCRE2 is built, a default newline convention is set; this is usu‐
2501 ally the standard convention for the operating system. The default can
2502 be overridden in a compile context by calling pcre2_set_newline(). It
2503 can also be overridden by starting a pattern string with, for example,
2504 (*CRLF), as described in the section on newline conventions in the
2505 pcre2pattern page. During matching, the newline choice affects the be‐
2506 haviour of the dot, circumflex, and dollar metacharacters. It may also
2507 alter the way the match starting position is advanced after a match
2508 failure for an unanchored pattern.
2509 When PCRE2_NEWLINE_CRLF, PCRE2_NEWLINE_ANYCRLF, or PCRE2_NEWLINE_ANY is
2511 set as the newline convention, and a match attempt for an unanchored
2512 pattern fails when the current starting position is at a CRLF sequence,
2513 and the pattern contains no explicit matches for CR or LF characters,
2514 the match position is advanced by two characters instead of one, in
2515 other words, to after the CRLF.
2516 The above rule is a compromise that makes the most common cases work as
2518 expected. For example, if the pattern is .+A (and the PCRE2_DOTALL
2519 option is not set), it does not match the string "\r\nA" because, after
2520 failing at the start, it skips both the CR and the LF before retrying.
2521 However, the pattern [\r\n]A does match that string, because it con‐
2522 tains an explicit CR or LF reference, and so advances only by one char‐
2523 acter after the first failure.
2524 An explicit match for CR of LF is either a literal appearance of one of
2526 those characters in the pattern, or one of the \r or \n or equivalent
2527 octal or hexadecimal escape sequences. Implicit matches such as [^X] do
2528 not count, nor does \s, even though it includes CR and LF in the char‐
2529 acters that it matches.
2530 Notwithstanding the above, anomalous effects may still occur when CRLF
2532 is a valid newline sequence and explicit \r or \n escapes appear in the
2533 pattern.
2534
2536 uint32_t pcre2_get_ovector_count(pcre2_match_data *match_data);
2538 PCRE2_SIZE *pcre2_get_ovector_pointer(pcre2_match_data *match_data);
2540 In general, a pattern matches a certain portion of the subject, and in
2542 addition, further substrings from the subject may be picked out by
2543 parenthesized parts of the pattern. Following the usage in Jeffrey
2544 Friedl's book, this is called "capturing" in what follows, and the
2545 phrase "capture group" (Perl terminology) is used for a fragment of a
2546 pattern that picks out a substring. PCRE2 supports several other kinds
2547 of parenthesized group that do not cause substrings to be captured. The
2548 pcre2_pattern_info() function can be used to find out how many capture
2549 groups there are in a compiled pattern.
2550 You can use auxiliary functions for accessing captured substrings by
2552 number or by name, as described in sections below.
2553 Alternatively, you can make direct use of the vector of PCRE2_SIZE val‐
2555 ues, called the ovector, which contains the offsets of captured
2556 strings. It is part of the match data block. The function
2557 pcre2_get_ovector_pointer() returns the address of the ovector, and
2558 pcre2_get_ovector_count() returns the number of pairs of values it con‐
2559 tains.
2560 Within the ovector, the first in each pair of values is set to the off‐
2562 set of the first code unit of a substring, and the second is set to the
2563 offset of the first code unit after the end of a substring. These val‐
2564 ues are always code unit offsets, not character offsets. That is, they
2565 are byte offsets in the 8-bit library, 16-bit offsets in the 16-bit
2566 library, and 32-bit offsets in the 32-bit library.
2567 After a partial match (error return PCRE2_ERROR_PARTIAL), only the
2569 first pair of offsets (that is, ovector[0] and ovector[1]) are set.
2570 They identify the part of the subject that was partially matched. See
2571 the pcre2partial documentation for details of partial matching.
2572 After a fully successful match, the first pair of offsets identifies
2574 the portion of the subject string that was matched by the entire pat‐
2575 tern. The next pair is used for the first captured substring, and so
2576 on. The value returned by pcre2_match() is one more than the highest
2577 numbered pair that has been set. For example, if two substrings have
2578 been captured, the returned value is 3. If there are no captured sub‐
2579 strings, the return value from a successful match is 1, indicating that
2580 just the first pair of offsets has been set.
2581 If a pattern uses the \K escape sequence within a positive assertion,
2583 the reported start of a successful match can be greater than the end of
2584 the match. For example, if the pattern (?=ab\K) is matched against
2585 "ab", the start and end offset values for the match are 2 and 0.
2586 If a capture group is matched repeatedly within a single match opera‐
2588 tion, it is the last portion of the subject that it matched that is
2589 returned.
2590 If the ovector is too small to hold all the captured substring offsets,
2592 as much as possible is filled in, and the function returns a value of
2593 zero. If captured substrings are not of interest, pcre2_match() may be
2594 called with a match data block whose ovector is of minimum length (that
2595 is, one pair).
2596 It is possible for capture group number n+1 to match some part of the
2598 subject when group n has not been used at all. For example, if the
2599 string "abc" is matched against the pattern (a|(z))(bc) the return from
2600 the function is 4, and groups 1 and 3 are matched, but 2 is not. When
2601 this happens, both values in the offset pairs corresponding to unused
2602 groups are set to PCRE2_UNSET.
2603 Offset values that correspond to unused groups at the end of the
2605 expression are also set to PCRE2_UNSET. For example, if the string
2606 "abc" is matched against the pattern (abc)(x(yz)?)? groups 2 and 3 are
2607 not matched. The return from the function is 2, because the highest
2608 used capture group number is 1. The offsets for for the second and
2609 third capture groupss (assuming the vector is large enough, of course)
2610 are set to PCRE2_UNSET.
2611 Elements in the ovector that do not correspond to capturing parentheses
2613 in the pattern are never changed. That is, if a pattern contains n cap‐
2614 turing parentheses, no more than ovector[0] to ovector[2n+1] are set by
2615 pcre2_match(). The other elements retain whatever values they previ‐
2616 ously had. After a failed match attempt, the contents of the ovector
2617 are unchanged.
2618
2620 PCRE2_SPTR pcre2_get_mark(pcre2_match_data *match_data);
2622 PCRE2_SIZE pcre2_get_startchar(pcre2_match_data *match_data);
2624 As well as the offsets in the ovector, other information about a match
2626 is retained in the match data block and can be retrieved by the above
2627 functions in appropriate circumstances. If they are called at other
2628 times, the result is undefined.
2629 After a successful match, a partial match (PCRE2_ERROR_PARTIAL), or a
2631 failure to match (PCRE2_ERROR_NOMATCH), a mark name may be available.
2632 The function pcre2_get_mark() can be called to access this name, which
2633 can be specified in the pattern by any of the backtracking control
2634 verbs, not just (*MARK). The same function applies to all the verbs. It
2635 returns a pointer to the zero-terminated name, which is within the com‐
2636 piled pattern. If no name is available, NULL is returned. The length of
2637 the name (excluding the terminating zero) is stored in the code unit
2638 that precedes the name. You should use this length instead of relying
2639 on the terminating zero if the name might contain a binary zero.
2640 After a successful match, the name that is returned is the last mark
2642 name encountered on the matching path through the pattern. Instances of
2643 backtracking verbs without names do not count. Thus, for example, if
2644 the matching path contains (*MARK:A)(*PRUNE), the name "A" is returned.
2645 After a "no match" or a partial match, the last encountered name is
2646 returned. For example, consider this pattern:
2647 ^(*MARK:A)((*MARK:B)a|b)c
2649 When it matches "bc", the returned name is A. The B mark is "seen" in
2651 the first branch of the group, but it is not on the matching path. On
2652 the other hand, when this pattern fails to match "bx", the returned
2653 name is B.
2654 Warning: By default, certain start-of-match optimizations are used to
2656 give a fast "no match" result in some situations. For example, if the
2657 anchoring is removed from the pattern above, there is an initial check
2658 for the presence of "c" in the subject before running the matching
2659 engine. This check fails for "bx", causing a match failure without see‐
2660 ing any marks. You can disable the start-of-match optimizations by set‐
2661 ting the PCRE2_NO_START_OPTIMIZE option for pcre2_compile() or by
2662 starting the pattern with (*NO_START_OPT).
2663 After a successful match, a partial match, or one of the invalid UTF
2665 errors (for example, PCRE2_ERROR_UTF8_ERR5), pcre2_get_startchar() can
2666 be called. After a successful or partial match it returns the code unit
2667 offset of the character at which the match started. For a non-partial
2668 match, this can be different to the value of ovector[0] if the pattern
2669 contains the \K escape sequence. After a partial match, however, this
2670 value is always the same as ovector[0] because \K does not affect the
2671 result of a partial match.
2672 After a UTF check failure, pcre2_get_startchar() can be used to obtain
2674 the code unit offset of the invalid UTF character. Details are given in
2675 the pcre2unicode page.
2676
2678 If pcre2_match() fails, it returns a negative number. This can be con‐
2680 verted to a text string by calling the pcre2_get_error_message() func‐
2681 tion (see "Obtaining a textual error message" below). Negative error
2682 codes are also returned by other functions, and are documented with
2683 them. The codes are given names in the header file. If UTF checking is
2684 in force and an invalid UTF subject string is detected, one of a number
2685 of UTF-specific negative error codes is returned. Details are given in
2686 the pcre2unicode page. The following are the other errors that may be
2687 returned by pcre2_match():
2688 PCRE2_ERROR_NOMATCH
2690 The subject string did not match the pattern.
2692 PCRE2_ERROR_PARTIAL
2694 The subject string did not match, but it did match partially. See the
2696 pcre2partial documentation for details of partial matching.
2697 PCRE2_ERROR_BADMAGIC
2699 PCRE2 stores a 4-byte "magic number" at the start of the compiled code,
2701 to catch the case when it is passed a junk pointer. This is the error
2702 that is returned when the magic number is not present.
2703 PCRE2_ERROR_BADMODE
2705 This error is given when a compiled pattern is passed to a function in
2707 a library of a different code unit width, for example, a pattern com‐
2708 piled by the 8-bit library is passed to a 16-bit or 32-bit library
2709 function.
2710 PCRE2_ERROR_BADOFFSET
2712 The value of startoffset was greater than the length of the subject.
2714 PCRE2_ERROR_BADOPTION
2716 An unrecognized bit was set in the options argument.
2718 PCRE2_ERROR_BADUTFOFFSET
2720 The UTF code unit sequence that was passed as a subject was checked and
2722 found to be valid (the PCRE2_NO_UTF_CHECK option was not set), but the
2723 value of startoffset did not point to the beginning of a UTF character
2724 or the end of the subject.
2725 PCRE2_ERROR_CALLOUT
2727 This error is never generated by pcre2_match() itself. It is provided
2729 for use by callout functions that want to cause pcre2_match() or
2730 pcre2_callout_enumerate() to return a distinctive error code. See the
2731 pcre2callout documentation for details.
2732 PCRE2_ERROR_DEPTHLIMIT
2734 The nested backtracking depth limit was reached.
2736 PCRE2_ERROR_HEAPLIMIT
2738 The heap limit was reached.
2740 PCRE2_ERROR_INTERNAL
2742 An unexpected internal error has occurred. This error could be caused
2744 by a bug in PCRE2 or by overwriting of the compiled pattern.
2745 PCRE2_ERROR_JIT_STACKLIMIT
2747 This error is returned when a pattern that was successfully studied
2749 using JIT is being matched, but the memory available for the just-in-
2750 time processing stack is not large enough. See the pcre2jit documenta‐
2751 tion for more details.
2752 PCRE2_ERROR_MATCHLIMIT
2754 The backtracking match limit was reached.
2756 PCRE2_ERROR_NOMEMORY
2758 If a pattern contains many nested backtracking points, heap memory is
2760 used to remember them. This error is given when the memory allocation
2761 function (default or custom) fails. Note that a different error,
2762 PCRE2_ERROR_HEAPLIMIT, is given if the amount of memory needed exceeds
2763 the heap limit. PCRE2_ERROR_NOMEMORY is also returned if
2764 PCRE2_COPY_MATCHED_SUBJECT is set and memory allocation fails.
2765 PCRE2_ERROR_NULL
2767 Either the code, subject, or match_data argument was passed as NULL.
2769 PCRE2_ERROR_RECURSELOOP
2771 This error is returned when pcre2_match() detects a recursion loop
2773 within the pattern. Specifically, it means that either the whole pat‐
2774 tern or a capture group has been called recursively for the second time
2775 at the same position in the subject string. Some simple patterns that
2776 might do this are detected and faulted at compile time, but more com‐
2777 plicated cases, in particular mutual recursions between two different
2778 groups, cannot be detected until matching is attempted.
2779
2781 int pcre2_get_error_message(int errorcode, PCRE2_UCHAR *buffer,
2783 PCRE2_SIZE bufflen);
2784 A text message for an error code from any PCRE2 function (compile,
2786 match, or auxiliary) can be obtained by calling pcre2_get_error_mes‐
2787 sage(). The code is passed as the first argument, with the remaining
2788 two arguments specifying a code unit buffer and its length in code
2789 units, into which the text message is placed. The message is returned
2790 in code units of the appropriate width for the library that is being
2791 used.
2792 The returned message is terminated with a trailing zero, and the func‐
2794 tion returns the number of code units used, excluding the trailing
2795 zero. If the error number is unknown, the negative error code
2796 PCRE2_ERROR_BADDATA is returned. If the buffer is too small, the mes‐
2797 sage is truncated (but still with a trailing zero), and the negative
2798 error code PCRE2_ERROR_NOMEMORY is returned. None of the messages are
2799 very long; a buffer size of 120 code units is ample.
2800
2802 int pcre2_substring_length_bynumber(pcre2_match_data *match_data,
2804 uint32_t number, PCRE2_SIZE *length);
2805 int pcre2_substring_copy_bynumber(pcre2_match_data *match_data,
2807 uint32_t number, PCRE2_UCHAR *buffer,
2808 PCRE2_SIZE *bufflen);
2809 int pcre2_substring_get_bynumber(pcre2_match_data *match_data,
2811 uint32_t number, PCRE2_UCHAR **bufferptr,
2812 PCRE2_SIZE *bufflen);
2813 void pcre2_substring_free(PCRE2_UCHAR *buffer);
2815 Captured substrings can be accessed directly by using the ovector as
2817 described above. For convenience, auxiliary functions are provided for
2818 extracting captured substrings as new, separate, zero-terminated
2819 strings. A substring that contains a binary zero is correctly extracted
2820 and has a further zero added on the end, but the result is not, of
2821 course, a C string.
2822 The functions in this section identify substrings by number. The number
2824 zero refers to the entire matched substring, with higher numbers refer‐
2825 ring to substrings captured by parenthesized groups. After a partial
2826 match, only substring zero is available. An attempt to extract any
2827 other substring gives the error PCRE2_ERROR_PARTIAL. The next section
2828 describes similar functions for extracting captured substrings by name.
2829 If a pattern uses the \K escape sequence within a positive assertion,
2831 the reported start of a successful match can be greater than the end of
2832 the match. For example, if the pattern (?=ab\K) is matched against
2833 "ab", the start and end offset values for the match are 2 and 0. In
2834 this situation, calling these functions with a zero substring number
2835 extracts a zero-length empty string.
2836 You can find the length in code units of a captured substring without
2838 extracting it by calling pcre2_substring_length_bynumber(). The first
2839 argument is a pointer to the match data block, the second is the group
2840 number, and the third is a pointer to a variable into which the length
2841 is placed. If you just want to know whether or not the substring has
2842 been captured, you can pass the third argument as NULL.
2843 The pcre2_substring_copy_bynumber() function copies a captured sub‐
2845 string into a supplied buffer, whereas pcre2_substring_get_bynumber()
2846 copies it into new memory, obtained using the same memory allocation
2847 function that was used for the match data block. The first two argu‐
2848 ments of these functions are a pointer to the match data block and a
2849 capture group number.
2850 The final arguments of pcre2_substring_copy_bynumber() are a pointer to
2852 the buffer and a pointer to a variable that contains its length in code
2853 units. This is updated to contain the actual number of code units used
2854 for the extracted substring, excluding the terminating zero.
2855 For pcre2_substring_get_bynumber() the third and fourth arguments point
2857 to variables that are updated with a pointer to the new memory and the
2858 number of code units that comprise the substring, again excluding the
2859 terminating zero. When the substring is no longer needed, the memory
2860 should be freed by calling pcre2_substring_free().
2861 The return value from all these functions is zero for success, or a
2863 negative error code. If the pattern match failed, the match failure
2864 code is returned. If a substring number greater than zero is used
2865 after a partial match, PCRE2_ERROR_PARTIAL is returned. Other possible
2866 error codes are:
2867 PCRE2_ERROR_NOMEMORY
2869 The buffer was too small for pcre2_substring_copy_bynumber(), or the
2871 attempt to get memory failed for pcre2_substring_get_bynumber().
2872 PCRE2_ERROR_NOSUBSTRING
2874 There is no substring with that number in the pattern, that is, the
2876 number is greater than the number of capturing parentheses.
2877 PCRE2_ERROR_UNAVAILABLE
2879 The substring number, though not greater than the number of captures in
2881 the pattern, is greater than the number of slots in the ovector, so the
2882 substring could not be captured.
2883 PCRE2_ERROR_UNSET
2885 The substring did not participate in the match. For example, if the
2887 pattern is (abc)|(def) and the subject is "def", and the ovector con‐
2888 tains at least two capturing slots, substring number 1 is unset.
2889
2891 int pcre2_substring_list_get(pcre2_match_data *match_data,
2893 PCRE2_UCHAR ***listptr, PCRE2_SIZE **lengthsptr);
2894 void pcre2_substring_list_free(PCRE2_SPTR *list);
2896 The pcre2_substring_list_get() function extracts all available sub‐
2898 strings and builds a list of pointers to them. It also (optionally)
2899 builds a second list that contains their lengths (in code units),
2900 excluding a terminating zero that is added to each of them. All this is
2901 done in a single block of memory that is obtained using the same memory
2902 allocation function that was used to get the match data block.
2903 This function must be called only after a successful match. If called
2905 after a partial match, the error code PCRE2_ERROR_PARTIAL is returned.
2906 The address of the memory block is returned via listptr, which is also
2908 the start of the list of string pointers. The end of the list is marked
2909 by a NULL pointer. The address of the list of lengths is returned via
2910 lengthsptr. If your strings do not contain binary zeros and you do not
2911 therefore need the lengths, you may supply NULL as the lengthsptr argu‐
2912 ment to disable the creation of a list of lengths. The yield of the
2913 function is zero if all went well, or PCRE2_ERROR_NOMEMORY if the mem‐
2914 ory block could not be obtained. When the list is no longer needed, it
2915 should be freed by calling pcre2_substring_list_free().
2916 If this function encounters a substring that is unset, which can happen
2918 when capture group number n+1 matches some part of the subject, but
2919 group n has not been used at all, it returns an empty string. This can
2920 be distinguished from a genuine zero-length substring by inspecting the
2921 appropriate offset in the ovector, which contain PCRE2_UNSET for unset
2922 substrings, or by calling pcre2_substring_length_bynumber().
2923
2925 int pcre2_substring_number_from_name(const pcre2_code *code,
2927 PCRE2_SPTR name);
2928 int pcre2_substring_length_byname(pcre2_match_data *match_data,
2930 PCRE2_SPTR name, PCRE2_SIZE *length);
2931 int pcre2_substring_copy_byname(pcre2_match_data *match_data,
2933 PCRE2_SPTR name, PCRE2_UCHAR *buffer, PCRE2_SIZE *bufflen);
2934 int pcre2_substring_get_byname(pcre2_match_data *match_data,
2936 PCRE2_SPTR name, PCRE2_UCHAR **bufferptr, PCRE2_SIZE *bufflen);
2937 void pcre2_substring_free(PCRE2_UCHAR *buffer);
2939 To extract a substring by name, you first have to find associated num‐
2941 ber. For example, for this pattern:
2942 (a+)b(?<xxx>\d+)...
2944 the number of the capture group called "xxx" is 2. If the name is known
2946 to be unique (PCRE2_DUPNAMES was not set), you can find the number from
2947 the name by calling pcre2_substring_number_from_name(). The first argu‐
2948 ment is the compiled pattern, and the second is the name. The yield of
2949 the function is the group number, PCRE2_ERROR_NOSUBSTRING if there is
2950 no group with that name, or PCRE2_ERROR_NOUNIQUESUBSTRING if there is
2951 more than one group with that name. Given the number, you can extract
2952 the substring directly from the ovector, or use one of the "bynumber"
2953 functions described above.
2954 For convenience, there are also "byname" functions that correspond to
2956 the "bynumber" functions, the only difference being that the second
2957 argument is a name instead of a number. If PCRE2_DUPNAMES is set and
2958 there are duplicate names, these functions scan all the groups with the
2959 given name, and return the captured substring from the first named
2960 group that is set.
2961 If there are no groups with the given name, PCRE2_ERROR_NOSUBSTRING is
2963 returned. If all groups with the name have numbers that are greater
2964 than the number of slots in the ovector, PCRE2_ERROR_UNAVAILABLE is
2965 returned. If there is at least one group with a slot in the ovector,
2966 but no group is found to be set, PCRE2_ERROR_UNSET is returned.
2967 Warning: If the pattern uses the (?| feature to set up multiple capture
2969 groups with the same number, as described in the section on duplicate
2970 group numbers in the pcre2pattern page, you cannot use names to distin‐
2971 guish the different capture groups, because names are not included in
2972 the compiled code. The matching process uses only numbers. For this
2973 reason, the use of different names for groups with the same number
2974 causes an error at compile time.
2975
2977 int pcre2_substitute(const pcre2_code *code, PCRE2_SPTR subject,
2979 PCRE2_SIZE length, PCRE2_SIZE startoffset,
2980 uint32_t options, pcre2_match_data *match_data,
2981 pcre2_match_context *mcontext, PCRE2_SPTR replacement,
2982 PCRE2_SIZE rlength, PCRE2_UCHAR *outputbuffer,
2983 PCRE2_SIZE *outlengthptr);
2984 This function calls pcre2_match() and then makes a copy of the subject
2986 string in outputbuffer, replacing one or more parts that were matched
2987 with the replacement string, whose length is supplied in rlength. This
2988 can be given as PCRE2_ZERO_TERMINATED for a zero-terminated string.
2989 The default is to perform just one replacement, but there is an option
2990 that requests multiple replacements (see PCRE2_SUBSTITUTE_GLOBAL below
2991 for details).
2992 Matches in which a \K item in a lookahead in the pattern causes the
2994 match to end before it starts are not supported, and give rise to an
2995 error return. For global replacements, matches in which \K in a lookbe‐
2996 hind causes the match to start earlier than the point that was reached
2997 in the previous iteration are also not supported.
2998 The first seven arguments of pcre2_substitute() are the same as for
3000 pcre2_match(), except that the partial matching options are not permit‐
3001 ted, and match_data may be passed as NULL, in which case a match data
3002 block is obtained and freed within this function, using memory manage‐
3003 ment functions from the match context, if provided, or else those that
3004 were used to allocate memory for the compiled code.
3005 If an external match_data block is provided, its contents afterwards
3007 are those set by the final call to pcre2_match(). For global changes,
3008 this will have ended in a matching error. The contents of the ovector
3009 within the match data block may or may not have been changed.
3010 The outlengthptr argument must point to a variable that contains the
3012 length, in code units, of the output buffer. If the function is suc‐
3013 cessful, the value is updated to contain the length of the new string,
3014 excluding the trailing zero that is automatically added.
3015 If the function is not successful, the value set via outlengthptr
3017 depends on the type of error. For syntax errors in the replacement
3018 string, the value is the offset in the replacement string where the
3019 error was detected. For other errors, the value is PCRE2_UNSET by
3020 default. This includes the case of the output buffer being too small,
3021 unless PCRE2_SUBSTITUTE_OVERFLOW_LENGTH is set (see below), in which
3022 case the value is the minimum length needed, including space for the
3023 trailing zero. Note that in order to compute the required length,
3024 pcre2_substitute() has to simulate all the matching and copying,
3025 instead of giving an error return as soon as the buffer overflows. Note
3026 also that the length is in code units, not bytes.
3027 In the replacement string, which is interpreted as a UTF string in UTF
3029 mode, and is checked for UTF validity unless the PCRE2_NO_UTF_CHECK
3030 option is set, a dollar character is an escape character that can spec‐
3031 ify the insertion of characters from capture groups or names from
3032 (*MARK) or other control verbs in the pattern. The following forms are
3033 always recognized:
3034 $$ insert a dollar character
3036 $<n> or ${<n>} insert the contents of group <n>
3037 $*MARK or ${*MARK} insert a control verb name
3038 Either a group number or a group name can be given for <n>. Curly
3040 brackets are required only if the following character would be inter‐
3041 preted as part of the number or name. The number may be zero to include
3042 the entire matched string. For example, if the pattern a(b)c is
3043 matched with "=abc=" and the replacement string "+$1$0$1+", the result
3044 is "=+babcb+=".
3045 $*MARK inserts the name from the last encountered backtracking control
3047 verb on the matching path that has a name. (*MARK) must always include
3048 a name, but the other verbs need not. For example, in the case of
3049 (*MARK:A)(*PRUNE) the name inserted is "A", but for (*MARK:A)(*PRUNE:B)
3050 the relevant name is "B". This facility can be used to perform simple
3051 simultaneous substitutions, as this pcre2test example shows:
3052 /(*MARK:pear)apple|(*MARK:orange)lemon/g,replace=${*MARK}
3054 apple lemon
3055 2: pear orange
3056 As well as the usual options for pcre2_match(), a number of additional
3058 options can be set in the options argument of pcre2_substitute().
3059 PCRE2_SUBSTITUTE_GLOBAL causes the function to iterate over the subject
3061 string, replacing every matching substring. If this option is not set,
3062 only the first matching substring is replaced. The search for matches
3063 takes place in the original subject string (that is, previous replace‐
3064 ments do not affect it). Iteration is implemented by advancing the
3065 startoffset value for each search, which is always passed the entire
3066 subject string. If an offset limit is set in the match context, search‐
3067 ing stops when that limit is reached.
3068 You can restrict the effect of a global substitution to a portion of
3070 the subject string by setting either or both of startoffset and an off‐
3071 set limit. Here is a pcre2test example:
3072 /B/g,replace=!,use_offset_limit
3074 ABC ABC ABC ABC\=offset=3,offset_limit=12
3075 2: ABC A!C A!C ABC
3076 When continuing with global substitutions after matching a substring
3078 with zero length, an attempt to find a non-empty match at the same off‐
3079 set is performed. If this is not successful, the offset is advanced by
3080 one character except when CRLF is a valid newline sequence and the next
3081 two characters are CR, LF. In this case, the offset is advanced by two
3082 characters.
3083 PCRE2_SUBSTITUTE_OVERFLOW_LENGTH changes what happens when the output
3085 buffer is too small. The default action is to return PCRE2_ERROR_NOMEM‐
3086 ORY immediately. If this option is set, however, pcre2_substitute()
3087 continues to go through the motions of matching and substituting (with‐
3088 out, of course, writing anything) in order to compute the size of buf‐
3089 fer that is needed. This value is passed back via the outlengthptr
3090 variable, with the result of the function still being
3091 PCRE2_ERROR_NOMEMORY.
3092 Passing a buffer size of zero is a permitted way of finding out how
3094 much memory is needed for given substitution. However, this does mean
3095 that the entire operation is carried out twice. Depending on the appli‐
3096 cation, it may be more efficient to allocate a large buffer and free
3097 the excess afterwards, instead of using PCRE2_SUBSTITUTE_OVER‐
3098 FLOW_LENGTH.
3099 PCRE2_SUBSTITUTE_UNKNOWN_UNSET causes references to capture groups that
3101 do not appear in the pattern to be treated as unset groups. This option
3102 should be used with care, because it means that a typo in a group name
3103 or number no longer causes the PCRE2_ERROR_NOSUBSTRING error.
3104 PCRE2_SUBSTITUTE_UNSET_EMPTY causes unset capture groups (including
3106 unknown groups when PCRE2_SUBSTITUTE_UNKNOWN_UNSET is set) to be
3107 treated as empty strings when inserted as described above. If this
3108 option is not set, an attempt to insert an unset group causes the
3109 PCRE2_ERROR_UNSET error. This option does not influence the extended
3110 substitution syntax described below.
3111 PCRE2_SUBSTITUTE_EXTENDED causes extra processing to be applied to the
3113 replacement string. Without this option, only the dollar character is
3114 special, and only the group insertion forms listed above are valid.
3115 When PCRE2_SUBSTITUTE_EXTENDED is set, two things change:
3116 Firstly, backslash in a replacement string is interpreted as an escape
3118 character. The usual forms such as \n or \x{ddd} can be used to specify
3119 particular character codes, and backslash followed by any non-alphanu‐
3120 meric character quotes that character. Extended quoting can be coded
3121 using \Q...\E, exactly as in pattern strings.
3122 There are also four escape sequences for forcing the case of inserted
3124 letters. The insertion mechanism has three states: no case forcing,
3125 force upper case, and force lower case. The escape sequences change the
3126 current state: \U and \L change to upper or lower case forcing, respec‐
3127 tively, and \E (when not terminating a \Q quoted sequence) reverts to
3128 no case forcing. The sequences \u and \l force the next character (if
3129 it is a letter) to upper or lower case, respectively, and then the
3130 state automatically reverts to no case forcing. Case forcing applies to
3131 all inserted characters, including those from capture groups and let‐
3132 ters within \Q...\E quoted sequences.
3133 Note that case forcing sequences such as \U...\E do not nest. For exam‐
3135 ple, the result of processing "\Uaa\LBB\Ecc\E" is "AAbbcc"; the final
3136 \E has no effect. Note also that the PCRE2_ALT_BSUX and
3137 PCRE2_EXTRA_ALT_BSUX options do not apply to replacement strings.
3138 The second effect of setting PCRE2_SUBSTITUTE_EXTENDED is to add more
3140 flexibility to capture group substitution. The syntax is similar to
3141 that used by Bash:
3142 ${<n>:-<string>}
3144 ${<n>:+<string1>:<string2>}
3145 As before, <n> may be a group number or a name. The first form speci‐
3147 fies a default value. If group <n> is set, its value is inserted; if
3148 not, <string> is expanded and the result inserted. The second form
3149 specifies strings that are expanded and inserted when group <n> is set
3150 or unset, respectively. The first form is just a convenient shorthand
3151 for
3152 ${<n>:+${<n>}:<string>}
3154 Backslash can be used to escape colons and closing curly brackets in
3156 the replacement strings. A change of the case forcing state within a
3157 replacement string remains in force afterwards, as shown in this
3158 pcre2test example:
3159 /(some)?(body)/substitute_extended,replace=${1:+\U:\L}HeLLo
3161 body
3162 1: hello
3163 somebody
3164 1: HELLO
3165 The PCRE2_SUBSTITUTE_UNSET_EMPTY option does not affect these extended
3167 substitutions. However, PCRE2_SUBSTITUTE_UNKNOWN_UNSET does cause
3168 unknown groups in the extended syntax forms to be treated as unset.
3169 If successful, pcre2_substitute() returns the number of successful
3171 matches. This may be zero if no matches were found, and is never
3172 greater than 1 unless PCRE2_SUBSTITUTE_GLOBAL is set.
3173 In the event of an error, a negative error code is returned. Except for
3175 PCRE2_ERROR_NOMATCH (which is never returned), errors from
3176 pcre2_match() are passed straight back.
3177 PCRE2_ERROR_NOSUBSTRING is returned for a non-existent substring inser‐
3179 tion, unless PCRE2_SUBSTITUTE_UNKNOWN_UNSET is set.
3180 PCRE2_ERROR_UNSET is returned for an unset substring insertion (includ‐
3182 ing an unknown substring when PCRE2_SUBSTITUTE_UNKNOWN_UNSET is set)
3183 when the simple (non-extended) syntax is used and PCRE2_SUBSTI‐
3184 TUTE_UNSET_EMPTY is not set.
3185 PCRE2_ERROR_NOMEMORY is returned if the output buffer is not big
3187 enough. If the PCRE2_SUBSTITUTE_OVERFLOW_LENGTH option is set, the size
3188 of buffer that is needed is returned via outlengthptr. Note that this
3189 does not happen by default.
3190 PCRE2_ERROR_BADREPLACEMENT is used for miscellaneous syntax errors in
3192 the replacement string, with more particular errors being
3193 PCRE2_ERROR_BADREPESCAPE (invalid escape sequence), PCRE2_ERROR_REP‐
3194 MISSINGBRACE (closing curly bracket not found), PCRE2_ERROR_BADSUBSTI‐
3195 TUTION (syntax error in extended group substitution), and
3196 PCRE2_ERROR_BADSUBSPATTERN (the pattern match ended before it started
3197 or the match started earlier than the current position in the subject,
3198 which can happen if \K is used in an assertion).
3199 As for all PCRE2 errors, a text message that describes the error can be
3201 obtained by calling the pcre2_get_error_message() function (see
3202 "Obtaining a textual error message" above).
3203 Substitution callouts
3205 int pcre2_set_substitute_callout(pcre2_match_context *mcontext,
3207 int (*callout_function)(pcre2_substitute_callout_block *, void *),
3208 void *callout_data);
3209 The pcre2_set_substitution_callout() function can be used to specify a
3211 callout function for pcre2_substitute(). This information is passed in
3212 a match context. The callout function is called after each substitution
3213 has been processed, but it can cause the replacement not to happen. The
3214 callout function is not called for simulated substitutions that happen
3215 as a result of the PCRE2_SUBSTITUTE_OVERFLOW_LENGTH option.
3216 The first argument of the callout function is a pointer to a substitute
3218 callout block structure, which contains the following fields, not nec‐
3219 essarily in this order:
3220 uint32_t version;
3222 uint32_t subscount;
3223 PCRE2_SPTR input;
3224 PCRE2_SPTR output;
3225 PCRE2_SIZE *ovector;
3226 uint32_t oveccount;
3227 PCRE2_SIZE output_offsets[2];
3228 The version field contains the version number of the block format. The
3230 current version is 0. The version number will increase in future if
3231 more fields are added, but the intention is never to remove any of the
3232 existing fields.
3233 The subscount field is the number of the current match. It is 1 for the
3235 first callout, 2 for the second, and so on. The input and output point‐
3236 ers are copies of the values passed to pcre2_substitute().
3237 The ovector field points to the ovector, which contains the result of
3239 the most recent match. The oveccount field contains the number of pairs
3240 that are set in the ovector, and is always greater than zero.
3241 The output_offsets vector contains the offsets of the replacement in
3243 the output string. This has already been processed for dollar and (if
3244 requested) backslash substitutions as described above.
3245 The second argument of the callout function is the value passed as
3247 callout_data when the function was registered. The value returned by
3248 the callout function is interpreted as follows:
3249 If the value is zero, the replacement is accepted, and, if PCRE2_SUB‐
3251 STITUTE_GLOBAL is set, processing continues with a search for the next
3252 match. If the value is not zero, the current replacement is not
3253 accepted. If the value is greater than zero, processing continues when
3254 PCRE2_SUBSTITUTE_GLOBAL is set. Otherwise (the value is less than zero
3255 or PCRE2_SUBSTITUTE_GLOBAL is not set), the the rest of the input is
3256 copied to the output and the call to pcre2_substitute() exits, return‐
3257 ing the number of matches so far.
3258
3260 int pcre2_substring_nametable_scan(const pcre2_code *code,
3262 PCRE2_SPTR name, PCRE2_SPTR *first, PCRE2_SPTR *last);
3263 When a pattern is compiled with the PCRE2_DUPNAMES option, names for
3265 capture groups are not required to be unique. Duplicate names are
3266 always allowed for groups with the same number, created by using the
3267 (?| feature. Indeed, if such groups are named, they are required to use
3268 the same names.
3269 Normally, patterns that use duplicate names are such that in any one
3271 match, only one of each set of identically-named groups participates.
3272 An example is shown in the pcre2pattern documentation.
3273 When duplicates are present, pcre2_substring_copy_byname() and
3275 pcre2_substring_get_byname() return the first substring corresponding
3276 to the given name that is set. Only if none are set is
3277 PCRE2_ERROR_UNSET is returned. The pcre2_substring_number_from_name()
3278 function returns the error PCRE2_ERROR_NOUNIQUESUBSTRING when there are
3279 duplicate names.
3280 If you want to get full details of all captured substrings for a given
3282 name, you must use the pcre2_substring_nametable_scan() function. The
3283 first argument is the compiled pattern, and the second is the name. If
3284 the third and fourth arguments are NULL, the function returns a group
3285 number for a unique name, or PCRE2_ERROR_NOUNIQUESUBSTRING otherwise.
3286 When the third and fourth arguments are not NULL, they must be pointers
3288 to variables that are updated by the function. After it has run, they
3289 point to the first and last entries in the name-to-number table for the
3290 given name, and the function returns the length of each entry in code
3291 units. In both cases, PCRE2_ERROR_NOSUBSTRING is returned if there are
3292 no entries for the given name.
3293 The format of the name table is described above in the section entitled
3295 Information about a pattern. Given all the relevant entries for the
3296 name, you can extract each of their numbers, and hence the captured
3297 data.
3298
3300 The traditional matching function uses a similar algorithm to Perl,
3302 which stops when it finds the first match at a given point in the sub‐
3303 ject. If you want to find all possible matches, or the longest possible
3304 match at a given position, consider using the alternative matching
3305 function (see below) instead. If you cannot use the alternative func‐
3306 tion, you can kludge it up by making use of the callout facility, which
3307 is described in the pcre2callout documentation.
3308 What you have to do is to insert a callout right at the end of the pat‐
3310 tern. When your callout function is called, extract and save the cur‐
3311 rent matched substring. Then return 1, which forces pcre2_match() to
3312 backtrack and try other alternatives. Ultimately, when it runs out of
3313 matches, pcre2_match() will yield PCRE2_ERROR_NOMATCH.
3314
3316 int pcre2_dfa_match(const pcre2_code *code, PCRE2_SPTR subject,
3318 PCRE2_SIZE length, PCRE2_SIZE startoffset,
3319 uint32_t options, pcre2_match_data *match_data,
3320 pcre2_match_context *mcontext,
3321 int *workspace, PCRE2_SIZE wscount);
3322 The function pcre2_dfa_match() is called to match a subject string
3324 against a compiled pattern, using a matching algorithm that scans the
3325 subject string just once (not counting lookaround assertions), and does
3326 not backtrack. This has different characteristics to the normal algo‐
3327 rithm, and is not compatible with Perl. Some of the features of PCRE2
3328 patterns are not supported. Nevertheless, there are times when this
3329 kind of matching can be useful. For a discussion of the two matching
3330 algorithms, and a list of features that pcre2_dfa_match() does not sup‐
3331 port, see the pcre2matching documentation.
3332 The arguments for the pcre2_dfa_match() function are the same as for
3334 pcre2_match(), plus two extras. The ovector within the match data block
3335 is used in a different way, and this is described below. The other com‐
3336 mon arguments are used in the same way as for pcre2_match(), so their
3337 description is not repeated here.
3338 The two additional arguments provide workspace for the function. The
3340 workspace vector should contain at least 20 elements. It is used for
3341 keeping track of multiple paths through the pattern tree. More
3342 workspace is needed for patterns and subjects where there are a lot of
3343 potential matches.
3344 Here is an example of a simple call to pcre2_dfa_match():
3346 int wspace[20];
3348 pcre2_match_data *md = pcre2_match_data_create(4, NULL);
3349 int rc = pcre2_dfa_match(
3350 re, /* result of pcre2_compile() */
3351 "some string", /* the subject string */
3352 11, /* the length of the subject string */
3353 0, /* start at offset 0 in the subject */
3354 0, /* default options */
3355 md, /* the match data block */
3356 NULL, /* a match context; NULL means use defaults */
3357 wspace, /* working space vector */
3358 20); /* number of elements (NOT size in bytes) */
3359 Option bits for pcre_dfa_match()
3361 The unused bits of the options argument for pcre2_dfa_match() must be
3363 zero. The only bits that may be set are PCRE2_ANCHORED,
3364 PCRE2_COPY_MATCHED_SUBJECT, PCRE2_ENDANCHORED, PCRE2_NOTBOL,
3365 PCRE2_NOTEOL, PCRE2_NOTEMPTY, PCRE2_NOTEMPTY_ATSTART,
3366 PCRE2_NO_UTF_CHECK, PCRE2_PARTIAL_HARD, PCRE2_PARTIAL_SOFT,
3367 PCRE2_DFA_SHORTEST, and PCRE2_DFA_RESTART. All but the last four of
3368 these are exactly the same as for pcre2_match(), so their description
3369 is not repeated here.
3370 PCRE2_PARTIAL_HARD
3372 PCRE2_PARTIAL_SOFT
3373 These have the same general effect as they do for pcre2_match(), but
3375 the details are slightly different. When PCRE2_PARTIAL_HARD is set for
3376 pcre2_dfa_match(), it returns PCRE2_ERROR_PARTIAL if the end of the
3377 subject is reached and there is still at least one matching possibility
3378 that requires additional characters. This happens even if some complete
3379 matches have already been found. When PCRE2_PARTIAL_SOFT is set, the
3380 return code PCRE2_ERROR_NOMATCH is converted into PCRE2_ERROR_PARTIAL
3381 if the end of the subject is reached, there have been no complete
3382 matches, but there is still at least one matching possibility. The por‐
3383 tion of the string that was inspected when the longest partial match
3384 was found is set as the first matching string in both cases. There is a
3385 more detailed discussion of partial and multi-segment matching, with
3386 examples, in the pcre2partial documentation.
3387 PCRE2_DFA_SHORTEST
3389 Setting the PCRE2_DFA_SHORTEST option causes the matching algorithm to
3391 stop as soon as it has found one match. Because of the way the alterna‐
3392 tive algorithm works, this is necessarily the shortest possible match
3393 at the first possible matching point in the subject string.
3394 PCRE2_DFA_RESTART
3396 When pcre2_dfa_match() returns a partial match, it is possible to call
3398 it again, with additional subject characters, and have it continue with
3399 the same match. The PCRE2_DFA_RESTART option requests this action; when
3400 it is set, the workspace and wscount options must reference the same
3401 vector as before because data about the match so far is left in them
3402 after a partial match. There is more discussion of this facility in the
3403 pcre2partial documentation.
3404 Successful returns from pcre2_dfa_match()
3406 When pcre2_dfa_match() succeeds, it may have matched more than one sub‐
3408 string in the subject. Note, however, that all the matches from one run
3409 of the function start at the same point in the subject. The shorter
3410 matches are all initial substrings of the longer matches. For example,
3411 if the pattern
3412 <.*>
3414 is matched against the string
3416 This is <something> <something else> <something further> no more
3418 the three matched strings are
3420 <something> <something else> <something further>
3422 <something> <something else>
3423 <something>
3424 On success, the yield of the function is a number greater than zero,
3426 which is the number of matched substrings. The offsets of the sub‐
3427 strings are returned in the ovector, and can be extracted by number in
3428 the same way as for pcre2_match(), but the numbers bear no relation to
3429 any capture groups that may exist in the pattern, because DFA matching
3430 does not support capturing.
3431 Calls to the convenience functions that extract substrings by name
3433 return the error PCRE2_ERROR_DFA_UFUNC (unsupported function) if used
3434 after a DFA match. The convenience functions that extract substrings by
3435 number never return PCRE2_ERROR_NOSUBSTRING.
3436 The matched strings are stored in the ovector in reverse order of
3438 length; that is, the longest matching string is first. If there were
3439 too many matches to fit into the ovector, the yield of the function is
3440 zero, and the vector is filled with the longest matches.
3441 NOTE: PCRE2's "auto-possessification" optimization usually applies to
3443 character repeats at the end of a pattern (as well as internally). For
3444 example, the pattern "a\d+" is compiled as if it were "a\d++". For DFA
3445 matching, this means that only one possible match is found. If you
3446 really do want multiple matches in such cases, either use an ungreedy
3447 repeat such as "a\d+?" or set the PCRE2_NO_AUTO_POSSESS option when
3448 compiling.
3449 Error returns from pcre2_dfa_match()
3451 The pcre2_dfa_match() function returns a negative number when it fails.
3453 Many of the errors are the same as for pcre2_match(), as described
3454 above. There are in addition the following errors that are specific to
3455 pcre2_dfa_match():
3456 PCRE2_ERROR_DFA_UITEM
3458 This return is given if pcre2_dfa_match() encounters an item in the
3460 pattern that it does not support, for instance, the use of \C in a UTF
3461 mode or a backreference.
3462 PCRE2_ERROR_DFA_UCOND
3464 This return is given if pcre2_dfa_match() encounters a condition item
3466 that uses a backreference for the condition, or a test for recursion in
3467 a specific capture group. These are not supported.
3468 PCRE2_ERROR_DFA_UINVALID_UTF
3470 This return is given if pcre2_dfa_match() is called for a pattern that
3472 was compiled with PCRE2_MATCH_INVALID_UTF. This is not supported for
3473 DFA matching.
3474 PCRE2_ERROR_DFA_WSSIZE
3476 This return is given if pcre2_dfa_match() runs out of space in the
3478 workspace vector.
3479 PCRE2_ERROR_DFA_RECURSE
3481 When a recursion or subroutine call is processed, the matching function
3483 calls itself recursively, using private memory for the ovector and
3484 workspace. This error is given if the internal ovector is not large
3485 enough. This should be extremely rare, as a vector of size 1000 is
3486 used.
3487 PCRE2_ERROR_DFA_BADRESTART
3489 When pcre2_dfa_match() is called with the PCRE2_DFA_RESTART option,
3491 some plausibility checks are made on the contents of the workspace,
3492 which should contain data about the previous partial match. If any of
3493 these checks fail, this error is given.
3494
3496 pcre2build(3), pcre2callout(3), pcre2demo(3), pcre2matching(3),
3498 pcre2partial(3), pcre2posix(3), pcre2sample(3), pcre2unicode(3).
3499
3501 Philip Hazel
3503 University Computing Service
3504 Cambridge, England.
3505
3507 Last updated: 02 September 2019
3509 Copyright (c) 1997-2019 University of Cambridge.
3510PCRE2 10.34 02 September 2019 PCRE2API(3)