/[pcre]/code/trunk/doc/pcre.txt
ViewVC logotype

Contents of /code/trunk/doc/pcre.txt

Parent Directory Parent Directory | Revision Log Revision Log


Revision 1194 - (show annotations)
Wed Oct 31 17:42:29 2012 UTC (6 years, 9 months ago) by ph10
File MIME type: text/plain
File size: 477802 byte(s)
More documentation updates
1 -----------------------------------------------------------------------------
2 This file contains a concatenation of the PCRE man pages, converted to plain
3 text format for ease of searching with a text editor, or for use on systems
4 that do not have a man page processor. The small individual files that give
5 synopses of each function in the library have not been included. Neither has
6 the pcredemo program. There are separate text files for the pcregrep and
7 pcretest commands.
8 -----------------------------------------------------------------------------
9
10
11 PCRE(3) PCRE(3)
12
13
14 NAME
15 PCRE - Perl-compatible regular expressions
16
17
18 INTRODUCTION
19
20 The PCRE library is a set of functions that implement regular expres-
21 sion pattern matching using the same syntax and semantics as Perl, with
22 just a few differences. Some features that appeared in Python and PCRE
23 before they appeared in Perl are also available using the Python syn-
24 tax, there is some support for one or two .NET and Oniguruma syntax
25 items, and there is an option for requesting some minor changes that
26 give better JavaScript compatibility.
27
28 Starting with release 8.30, it is possible to compile two separate PCRE
29 libraries: the original, which supports 8-bit character strings
30 (including UTF-8 strings), and a second library that supports 16-bit
31 character strings (including UTF-16 strings). The build process allows
32 either one or both to be built. The majority of the work to make this
33 possible was done by Zoltan Herczeg.
34
35 Starting with release 8.32 it is possible to compile a third separate
36 PCRE library, which supports 32-bit character strings (including UTF-32
37 strings). The build process allows any set of the 8-, 16- and 32-bit
38 libraries. The work to make this possible was done by Christian Persch.
39
40 The three libraries contain identical sets of functions, except that
41 the names in the 16-bit library start with pcre16_ instead of pcre_,
42 and the names in the 32-bit library start with pcre32_ instead of
43 pcre_. To avoid over-complication and reduce the documentation mainte-
44 nance load, most of the documentation describes the 8-bit library, with
45 the differences for the 16-bit and 32-bit libraries described sepa-
46 rately in the pcre16 and pcre32 pages. References to functions or
47 structures of the form pcre[16|32]_xxx should be read as meaning
48 "pcre_xxx when using the 8-bit library, pcre16_xxx when using the
49 16-bit library, or pcre32_xxx when using the 32-bit library".
50
51 The current implementation of PCRE corresponds approximately with Perl
52 5.12, including support for UTF-8/16/32 encoded strings and Unicode
53 general category properties. However, UTF-8/16/32 and Unicode support
54 has to be explicitly enabled; it is not the default. The Unicode tables
55 correspond to Unicode release 6.2.0.
56
57 In addition to the Perl-compatible matching function, PCRE contains an
58 alternative function that matches the same compiled patterns in a dif-
59 ferent way. In certain circumstances, the alternative function has some
60 advantages. For a discussion of the two matching algorithms, see the
61 pcrematching page.
62
63 PCRE is written in C and released as a C library. A number of people
64 have written wrappers and interfaces of various kinds. In particular,
65 Google Inc. have provided a comprehensive C++ wrapper for the 8-bit
66 library. This is now included as part of the PCRE distribution. The
67 pcrecpp page has details of this interface. Other people's contribu-
68 tions can be found in the Contrib directory at the primary FTP site,
69 which is:
70
71 ftp://ftp.csx.cam.ac.uk/pub/software/programming/pcre
72
73 Details of exactly which Perl regular expression features are and are
74 not supported by PCRE are given in separate documents. See the pcrepat-
75 tern and pcrecompat pages. There is a syntax summary in the pcresyntax
76 page.
77
78 Some features of PCRE can be included, excluded, or changed when the
79 library is built. The pcre_config() function makes it possible for a
80 client to discover which features are available. The features them-
81 selves are described in the pcrebuild page. Documentation about build-
82 ing PCRE for various operating systems can be found in the README and
83 NON-AUTOTOOLS_BUILD files in the source distribution.
84
85 The libraries contains a number of undocumented internal functions and
86 data tables that are used by more than one of the exported external
87 functions, but which are not intended for use by external callers.
88 Their names all begin with "_pcre_" or "_pcre16_" or "_pcre32_", which
89 hopefully will not provoke any name clashes. In some environments, it
90 is possible to control which external symbols are exported when a
91 shared library is built, and in these cases the undocumented symbols
92 are not exported.
93
94
95 SECURITY CONSIDERATIONS
96
97 If you are using PCRE in a non-UTF application that permits users to
98 supply arbitrary patterns for compilation, you should be aware of a
99 feature that allows users to turn on UTF support from within a pattern,
100 provided that PCRE was built with UTF support. For example, an 8-bit
101 pattern that begins with "(*UTF8)" turns on UTF-8 mode. This causes
102 both the pattern and any data against which it is matched to be checked
103 for UTF-8 validity. If the data string is very long, such a check might
104 use sufficiently many resources as to cause your application to lose
105 performance.
106
107 The best way of guarding against this possibility is to use the
108 pcre_fullinfo() function to check the compiled pattern's options for
109 UTF.
110
111 If your application is one that supports UTF, be aware that validity
112 checking can take time. If the same data string is to be matched many
113 times, you can use the PCRE_NO_UTF[8|16|32]_CHECK option for the second
114 and subsequent matches to save redundant checks.
115
116 Another way that performance can be hit is by running a pattern that
117 has a very large search tree against a string that will never match.
118 Nested unlimited repeats in a pattern are a common example. PCRE pro-
119 vides some protection against this: see the PCRE_EXTRA_MATCH_LIMIT fea-
120 ture in the pcreapi page.
121
122
123 USER DOCUMENTATION
124
125 The user documentation for PCRE comprises a number of different sec-
126 tions. In the "man" format, each of these is a separate "man page". In
127 the HTML format, each is a separate page, linked from the index page.
128 In the plain text format, all the sections, except the pcredemo sec-
129 tion, are concatenated, for ease of searching. The sections are as fol-
130 lows:
131
132 pcre this document
133 pcre16 details of the 16-bit library
134 pcre32 details of the 32-bit library
135 pcre-config show PCRE installation configuration information
136 pcreapi details of PCRE's native C API
137 pcrebuild options for building PCRE
138 pcrecallout details of the callout feature
139 pcrecompat discussion of Perl compatibility
140 pcrecpp details of the C++ wrapper for the 8-bit library
141 pcredemo a demonstration C program that uses PCRE
142 pcregrep description of the pcregrep command (8-bit only)
143 pcrejit discussion of the just-in-time optimization support
144 pcrelimits details of size and other limits
145 pcrematching discussion of the two matching algorithms
146 pcrepartial details of the partial matching facility
147 pcrepattern syntax and semantics of supported
148 regular expressions
149 pcreperform discussion of performance issues
150 pcreposix the POSIX-compatible C API for the 8-bit library
151 pcreprecompile details of saving and re-using precompiled patterns
152 pcresample discussion of the pcredemo program
153 pcrestack discussion of stack usage
154 pcresyntax quick syntax reference
155 pcretest description of the pcretest testing command
156 pcreunicode discussion of Unicode and UTF-8/16/32 support
157
158 In addition, in the "man" and HTML formats, there is a short page for
159 each C library function, listing its arguments and results.
160
161
162 AUTHOR
163
164 Philip Hazel
165 University Computing Service
166 Cambridge CB2 3QH, England.
167
168 Putting an actual email address here seems to have been a spam magnet,
169 so I've taken it away. If you want to email me, use my two initials,
170 followed by the two digits 10, at the domain cam.ac.uk.
171
172
173 REVISION
174
175 Last updated: 30 October 2012
176 Copyright (c) 1997-2012 University of Cambridge.
177 ------------------------------------------------------------------------------
178
179
180 PCRE(3) PCRE(3)
181
182
183 NAME
184 PCRE - Perl-compatible regular expressions
185
186 #include <pcre.h>
187
188
189 PCRE 16-BIT API BASIC FUNCTIONS
190
191 pcre16 *pcre16_compile(PCRE_SPTR16 pattern, int options,
192 const char **errptr, int *erroffset,
193 const unsigned char *tableptr);
194
195 pcre16 *pcre16_compile2(PCRE_SPTR16 pattern, int options,
196 int *errorcodeptr,
197 const char **errptr, int *erroffset,
198 const unsigned char *tableptr);
199
200 pcre16_extra *pcre16_study(const pcre16 *code, int options,
201 const char **errptr);
202
203 void pcre16_free_study(pcre16_extra *extra);
204
205 int pcre16_exec(const pcre16 *code, const pcre16_extra *extra,
206 PCRE_SPTR16 subject, int length, int startoffset,
207 int options, int *ovector, int ovecsize);
208
209 int pcre16_dfa_exec(const pcre16 *code, const pcre16_extra *extra,
210 PCRE_SPTR16 subject, int length, int startoffset,
211 int options, int *ovector, int ovecsize,
212 int *workspace, int wscount);
213
214
215 PCRE 16-BIT API STRING EXTRACTION FUNCTIONS
216
217 int pcre16_copy_named_substring(const pcre16 *code,
218 PCRE_SPTR16 subject, int *ovector,
219 int stringcount, PCRE_SPTR16 stringname,
220 PCRE_UCHAR16 *buffer, int buffersize);
221
222 int pcre16_copy_substring(PCRE_SPTR16 subject, int *ovector,
223 int stringcount, int stringnumber, PCRE_UCHAR16 *buffer,
224 int buffersize);
225
226 int pcre16_get_named_substring(const pcre16 *code,
227 PCRE_SPTR16 subject, int *ovector,
228 int stringcount, PCRE_SPTR16 stringname,
229 PCRE_SPTR16 *stringptr);
230
231 int pcre16_get_stringnumber(const pcre16 *code,
232 PCRE_SPTR16 name);
233
234 int pcre16_get_stringtable_entries(const pcre16 *code,
235 PCRE_SPTR16 name, PCRE_UCHAR16 **first, PCRE_UCHAR16 **last);
236
237 int pcre16_get_substring(PCRE_SPTR16 subject, int *ovector,
238 int stringcount, int stringnumber,
239 PCRE_SPTR16 *stringptr);
240
241 int pcre16_get_substring_list(PCRE_SPTR16 subject,
242 int *ovector, int stringcount, PCRE_SPTR16 **listptr);
243
244 void pcre16_free_substring(PCRE_SPTR16 stringptr);
245
246 void pcre16_free_substring_list(PCRE_SPTR16 *stringptr);
247
248
249 PCRE 16-BIT API AUXILIARY FUNCTIONS
250
251 pcre16_jit_stack *pcre16_jit_stack_alloc(int startsize, int maxsize);
252
253 void pcre16_jit_stack_free(pcre16_jit_stack *stack);
254
255 void pcre16_assign_jit_stack(pcre16_extra *extra,
256 pcre16_jit_callback callback, void *data);
257
258 const unsigned char *pcre16_maketables(void);
259
260 int pcre16_fullinfo(const pcre16 *code, const pcre16_extra *extra,
261 int what, void *where);
262
263 int pcre16_refcount(pcre16 *code, int adjust);
264
265 int pcre16_config(int what, void *where);
266
267 const char *pcre16_version(void);
268
269 int pcre16_pattern_to_host_byte_order(pcre16 *code,
270 pcre16_extra *extra, const unsigned char *tables);
271
272
273 PCRE 16-BIT API INDIRECTED FUNCTIONS
274
275 void *(*pcre16_malloc)(size_t);
276
277 void (*pcre16_free)(void *);
278
279 void *(*pcre16_stack_malloc)(size_t);
280
281 void (*pcre16_stack_free)(void *);
282
283 int (*pcre16_callout)(pcre16_callout_block *);
284
285
286 PCRE 16-BIT API 16-BIT-ONLY FUNCTION
287
288 int pcre16_utf16_to_host_byte_order(PCRE_UCHAR16 *output,
289 PCRE_SPTR16 input, int length, int *byte_order,
290 int keep_boms);
291
292
293 THE PCRE 16-BIT LIBRARY
294
295 Starting with release 8.30, it is possible to compile a PCRE library
296 that supports 16-bit character strings, including UTF-16 strings, as
297 well as or instead of the original 8-bit library. The majority of the
298 work to make this possible was done by Zoltan Herczeg. The two
299 libraries contain identical sets of functions, used in exactly the same
300 way. Only the names of the functions and the data types of their argu-
301 ments and results are different. To avoid over-complication and reduce
302 the documentation maintenance load, most of the PCRE documentation
303 describes the 8-bit library, with only occasional references to the
304 16-bit library. This page describes what is different when you use the
305 16-bit library.
306
307 WARNING: A single application can be linked with both libraries, but
308 you must take care when processing any particular pattern to use func-
309 tions from just one library. For example, if you want to study a pat-
310 tern that was compiled with pcre16_compile(), you must do so with
311 pcre16_study(), not pcre_study(), and you must free the study data with
312 pcre16_free_study().
313
314
315 THE HEADER FILE
316
317 There is only one header file, pcre.h. It contains prototypes for all
318 the functions in both libraries, as well as definitions of flags,
319 structures, error codes, etc.
320
321
322 THE LIBRARY NAME
323
324 In Unix-like systems, the 16-bit library is called libpcre16, and can
325 normally be accesss by adding -lpcre16 to the command for linking an
326 application that uses PCRE.
327
328
329 STRING TYPES
330
331 In the 8-bit library, strings are passed to PCRE library functions as
332 vectors of bytes with the C type "char *". In the 16-bit library,
333 strings are passed as vectors of unsigned 16-bit quantities. The macro
334 PCRE_UCHAR16 specifies an appropriate data type, and PCRE_SPTR16 is
335 defined as "const PCRE_UCHAR16 *". In very many environments, "short
336 int" is a 16-bit data type. When PCRE is built, it defines PCRE_UCHAR16
337 as "short int", but checks that it really is a 16-bit data type. If it
338 is not, the build fails with an error message telling the maintainer to
339 modify the definition appropriately.
340
341
342 STRUCTURE TYPES
343
344 The types of the opaque structures that are used for compiled 16-bit
345 patterns and JIT stacks are pcre16 and pcre16_jit_stack respectively.
346 The type of the user-accessible structure that is returned by
347 pcre16_study() is pcre16_extra, and the type of the structure that is
348 used for passing data to a callout function is pcre16_callout_block.
349 These structures contain the same fields, with the same names, as their
350 8-bit counterparts. The only difference is that pointers to character
351 strings are 16-bit instead of 8-bit types.
352
353
354 16-BIT FUNCTIONS
355
356 For every function in the 8-bit library there is a corresponding func-
357 tion in the 16-bit library with a name that starts with pcre16_ instead
358 of pcre_. The prototypes are listed above. In addition, there is one
359 extra function, pcre16_utf16_to_host_byte_order(). This is a utility
360 function that converts a UTF-16 character string to host byte order if
361 necessary. The other 16-bit functions expect the strings they are
362 passed to be in host byte order.
363
364 The input and output arguments of pcre16_utf16_to_host_byte_order() may
365 point to the same address, that is, conversion in place is supported.
366 The output buffer must be at least as long as the input.
367
368 The length argument specifies the number of 16-bit data units in the
369 input string; a negative value specifies a zero-terminated string.
370
371 If byte_order is NULL, it is assumed that the string starts off in host
372 byte order. This may be changed by byte-order marks (BOMs) anywhere in
373 the string (commonly as the first character).
374
375 If byte_order is not NULL, a non-zero value of the integer to which it
376 points means that the input starts off in host byte order, otherwise
377 the opposite order is assumed. Again, BOMs in the string can change
378 this. The final byte order is passed back at the end of processing.
379
380 If keep_boms is not zero, byte-order mark characters (0xfeff) are
381 copied into the output string. Otherwise they are discarded.
382
383 The result of the function is the number of 16-bit units placed into
384 the output buffer, including the zero terminator if the string was
385 zero-terminated.
386
387
388 SUBJECT STRING OFFSETS
389
390 The offsets within subject strings that are returned by the matching
391 functions are in 16-bit units rather than bytes.
392
393
394 NAMED SUBPATTERNS
395
396 The name-to-number translation table that is maintained for named sub-
397 patterns uses 16-bit characters. The pcre16_get_stringtable_entries()
398 function returns the length of each entry in the table as the number of
399 16-bit data units.
400
401
402 OPTION NAMES
403
404 There are two new general option names, PCRE_UTF16 and
405 PCRE_NO_UTF16_CHECK, which correspond to PCRE_UTF8 and
406 PCRE_NO_UTF8_CHECK in the 8-bit library. In fact, these new options
407 define the same bits in the options word. There is a discussion about
408 the validity of UTF-16 strings in the pcreunicode page.
409
410 For the pcre16_config() function there is an option PCRE_CONFIG_UTF16
411 that returns 1 if UTF-16 support is configured, otherwise 0. If this
412 option is given to pcre_config() or pcre32_config(), or if the
413 PCRE_CONFIG_UTF8 or PCRE_CONFIG_UTF32 option is given to pcre16_con-
414 fig(), the result is the PCRE_ERROR_BADOPTION error.
415
416
417 CHARACTER CODES
418
419 In 16-bit mode, when PCRE_UTF16 is not set, character values are
420 treated in the same way as in 8-bit, non UTF-8 mode, except, of course,
421 that they can range from 0 to 0xffff instead of 0 to 0xff. Character
422 types for characters less than 0xff can therefore be influenced by the
423 locale in the same way as before. Characters greater than 0xff have
424 only one case, and no "type" (such as letter or digit).
425
426 In UTF-16 mode, the character code is Unicode, in the range 0 to
427 0x10ffff, with the exception of values in the range 0xd800 to 0xdfff
428 because those are "surrogate" values that are used in pairs to encode
429 values greater than 0xffff.
430
431 A UTF-16 string can indicate its endianness by special code knows as a
432 byte-order mark (BOM). The PCRE functions do not handle this, expecting
433 strings to be in host byte order. A utility function called
434 pcre16_utf16_to_host_byte_order() is provided to help with this (see
435 above).
436
437
438 ERROR NAMES
439
440 The errors PCRE_ERROR_BADUTF16_OFFSET and PCRE_ERROR_SHORTUTF16 corre-
441 spond to their 8-bit counterparts. The error PCRE_ERROR_BADMODE is
442 given when a compiled pattern is passed to a function that processes
443 patterns in the other mode, for example, if a pattern compiled with
444 pcre_compile() is passed to pcre16_exec().
445
446 There are new error codes whose names begin with PCRE_UTF16_ERR for
447 invalid UTF-16 strings, corresponding to the PCRE_UTF8_ERR codes for
448 UTF-8 strings that are described in the section entitled "Reason codes
449 for invalid UTF-8 strings" in the main pcreapi page. The UTF-16 errors
450 are:
451
452 PCRE_UTF16_ERR1 Missing low surrogate at end of string
453 PCRE_UTF16_ERR2 Invalid low surrogate follows high surrogate
454 PCRE_UTF16_ERR3 Isolated low surrogate
455 PCRE_UTF16_ERR4 Non-character
456
457
458 ERROR TEXTS
459
460 If there is an error while compiling a pattern, the error text that is
461 passed back by pcre16_compile() or pcre16_compile2() is still an 8-bit
462 character string, zero-terminated.
463
464
465 CALLOUTS
466
467 The subject and mark fields in the callout block that is passed to a
468 callout function point to 16-bit vectors.
469
470
471 TESTING
472
473 The pcretest program continues to operate with 8-bit input and output
474 files, but it can be used for testing the 16-bit library. If it is run
475 with the command line option -16, patterns and subject strings are con-
476 verted from 8-bit to 16-bit before being passed to PCRE, and the 16-bit
477 library functions are used instead of the 8-bit ones. Returned 16-bit
478 strings are converted to 8-bit for output. If both the 8-bit and the
479 32-bit libraries were not compiled, pcretest defaults to 16-bit and the
480 -16 option is ignored.
481
482 When PCRE is being built, the RunTest script that is called by "make
483 check" uses the pcretest -C option to discover which of the 8-bit,
484 16-bit and 32-bit libraries has been built, and runs the tests appro-
485 priately.
486
487
488 NOT SUPPORTED IN 16-BIT MODE
489
490 Not all the features of the 8-bit library are available with the 16-bit
491 library. The C++ and POSIX wrapper functions support only the 8-bit
492 library, and the pcregrep program is at present 8-bit only.
493
494
495 AUTHOR
496
497 Philip Hazel
498 University Computing Service
499 Cambridge CB2 3QH, England.
500
501
502 REVISION
503
504 Last updated: 14 April 2012
505 Copyright (c) 1997-2012 University of Cambridge.
506 ------------------------------------------------------------------------------
507
508
509 PCRE(3) PCRE(3)
510
511
512 NAME
513 PCRE - Perl-compatible regular expressions
514
515 #include <pcre.h>
516
517
518 PCRE 32-BIT API BASIC FUNCTIONS
519
520 pcre32 *pcre32_compile(PCRE_SPTR32 pattern, int options,
521 const char **errptr, int *erroffset,
522 const unsigned char *tableptr);
523
524 pcre32 *pcre32_compile2(PCRE_SPTR32 pattern, int options,
525 int *errorcodeptr,
526 const char **errptr, int *erroffset,
527 const unsigned char *tableptr);
528
529 pcre32_extra *pcre32_study(const pcre32 *code, int options,
530 const char **errptr);
531
532 void pcre32_free_study(pcre32_extra *extra);
533
534 int pcre32_exec(const pcre32 *code, const pcre32_extra *extra,
535 PCRE_SPTR32 subject, int length, int startoffset,
536 int options, int *ovector, int ovecsize);
537
538 int pcre32_dfa_exec(const pcre32 *code, const pcre32_extra *extra,
539 PCRE_SPTR32 subject, int length, int startoffset,
540 int options, int *ovector, int ovecsize,
541 int *workspace, int wscount);
542
543
544 PCRE 32-BIT API STRING EXTRACTION FUNCTIONS
545
546 int pcre32_copy_named_substring(const pcre32 *code,
547 PCRE_SPTR32 subject, int *ovector,
548 int stringcount, PCRE_SPTR32 stringname,
549 PCRE_UCHAR32 *buffer, int buffersize);
550
551 int pcre32_copy_substring(PCRE_SPTR32 subject, int *ovector,
552 int stringcount, int stringnumber, PCRE_UCHAR32 *buffer,
553 int buffersize);
554
555 int pcre32_get_named_substring(const pcre32 *code,
556 PCRE_SPTR32 subject, int *ovector,
557 int stringcount, PCRE_SPTR32 stringname,
558 PCRE_SPTR32 *stringptr);
559
560 int pcre32_get_stringnumber(const pcre32 *code,
561 PCRE_SPTR32 name);
562
563 int pcre32_get_stringtable_entries(const pcre32 *code,
564 PCRE_SPTR32 name, PCRE_UCHAR32 **first, PCRE_UCHAR32 **last);
565
566 int pcre32_get_substring(PCRE_SPTR32 subject, int *ovector,
567 int stringcount, int stringnumber,
568 PCRE_SPTR32 *stringptr);
569
570 int pcre32_get_substring_list(PCRE_SPTR32 subject,
571 int *ovector, int stringcount, PCRE_SPTR32 **listptr);
572
573 void pcre32_free_substring(PCRE_SPTR32 stringptr);
574
575 void pcre32_free_substring_list(PCRE_SPTR32 *stringptr);
576
577
578 PCRE 32-BIT API AUXILIARY FUNCTIONS
579
580 pcre32_jit_stack *pcre32_jit_stack_alloc(int startsize, int maxsize);
581
582 void pcre32_jit_stack_free(pcre32_jit_stack *stack);
583
584 void pcre32_assign_jit_stack(pcre32_extra *extra,
585 pcre32_jit_callback callback, void *data);
586
587 const unsigned char *pcre32_maketables(void);
588
589 int pcre32_fullinfo(const pcre32 *code, const pcre32_extra *extra,
590 int what, void *where);
591
592 int pcre32_refcount(pcre32 *code, int adjust);
593
594 int pcre32_config(int what, void *where);
595
596 const char *pcre32_version(void);
597
598 int pcre32_pattern_to_host_byte_order(pcre32 *code,
599 pcre32_extra *extra, const unsigned char *tables);
600
601
602 PCRE 32-BIT API INDIRECTED FUNCTIONS
603
604 void *(*pcre32_malloc)(size_t);
605
606 void (*pcre32_free)(void *);
607
608 void *(*pcre32_stack_malloc)(size_t);
609
610 void (*pcre32_stack_free)(void *);
611
612 int (*pcre32_callout)(pcre32_callout_block *);
613
614
615 PCRE 32-BIT API 32-BIT-ONLY FUNCTION
616
617 int pcre32_utf32_to_host_byte_order(PCRE_UCHAR32 *output,
618 PCRE_SPTR32 input, int length, int *byte_order,
619 int keep_boms);
620
621
622 THE PCRE 32-BIT LIBRARY
623
624 Starting with release 8.32, it is possible to compile a PCRE library
625 that supports 32-bit character strings, including UTF-32 strings, as
626 well as or instead of the original 8-bit library. This work was done by
627 Christian Persch, based on the work done by Zoltan Herczeg for the
628 16-bit library. All three libraries contain identical sets of func-
629 tions, used in exactly the same way. Only the names of the functions
630 and the data types of their arguments and results are different. To
631 avoid over-complication and reduce the documentation maintenance load,
632 most of the PCRE documentation describes the 8-bit library, with only
633 occasional references to the 16-bit and 32-bit libraries. This page
634 describes what is different when you use the 32-bit library.
635
636 WARNING: A single application can be linked with all or any of the
637 three libraries, but you must take care when processing any particular
638 pattern to use functions from just one library. For example, if you
639 want to study a pattern that was compiled with pcre32_compile(), you
640 must do so with pcre32_study(), not pcre_study(), and you must free the
641 study data with pcre32_free_study().
642
643
644 THE HEADER FILE
645
646 There is only one header file, pcre.h. It contains prototypes for all
647 the functions in both libraries, as well as definitions of flags,
648 structures, error codes, etc.
649
650
651 THE LIBRARY NAME
652
653 In Unix-like systems, the 32-bit library is called libpcre32, and can
654 normally be accesss by adding -lpcre32 to the command for linking an
655 application that uses PCRE.
656
657
658 STRING TYPES
659
660 In the 8-bit library, strings are passed to PCRE library functions as
661 vectors of bytes with the C type "char *". In the 32-bit library,
662 strings are passed as vectors of unsigned 32-bit quantities. The macro
663 PCRE_UCHAR32 specifies an appropriate data type, and PCRE_SPTR32 is
664 defined as "const PCRE_UCHAR32 *". In very many environments, "unsigned
665 int" is a 32-bit data type. When PCRE is built, it defines PCRE_UCHAR32
666 as "unsigned int", but checks that it really is a 32-bit data type. If
667 it is not, the build fails with an error message telling the maintainer
668 to modify the definition appropriately.
669
670
671 STRUCTURE TYPES
672
673 The types of the opaque structures that are used for compiled 32-bit
674 patterns and JIT stacks are pcre32 and pcre32_jit_stack respectively.
675 The type of the user-accessible structure that is returned by
676 pcre32_study() is pcre32_extra, and the type of the structure that is
677 used for passing data to a callout function is pcre32_callout_block.
678 These structures contain the same fields, with the same names, as their
679 8-bit counterparts. The only difference is that pointers to character
680 strings are 32-bit instead of 8-bit types.
681
682
683 32-BIT FUNCTIONS
684
685 For every function in the 8-bit library there is a corresponding func-
686 tion in the 32-bit library with a name that starts with pcre32_ instead
687 of pcre_. The prototypes are listed above. In addition, there is one
688 extra function, pcre32_utf32_to_host_byte_order(). This is a utility
689 function that converts a UTF-32 character string to host byte order if
690 necessary. The other 32-bit functions expect the strings they are
691 passed to be in host byte order.
692
693 The input and output arguments of pcre32_utf32_to_host_byte_order() may
694 point to the same address, that is, conversion in place is supported.
695 The output buffer must be at least as long as the input.
696
697 The length argument specifies the number of 32-bit data units in the
698 input string; a negative value specifies a zero-terminated string.
699
700 If byte_order is NULL, it is assumed that the string starts off in host
701 byte order. This may be changed by byte-order marks (BOMs) anywhere in
702 the string (commonly as the first character).
703
704 If byte_order is not NULL, a non-zero value of the integer to which it
705 points means that the input starts off in host byte order, otherwise
706 the opposite order is assumed. Again, BOMs in the string can change
707 this. The final byte order is passed back at the end of processing.
708
709 If keep_boms is not zero, byte-order mark characters (0xfeff) are
710 copied into the output string. Otherwise they are discarded.
711
712 The result of the function is the number of 32-bit units placed into
713 the output buffer, including the zero terminator if the string was
714 zero-terminated.
715
716
717 SUBJECT STRING OFFSETS
718
719 The offsets within subject strings that are returned by the matching
720 functions are in 32-bit units rather than bytes.
721
722
723 NAMED SUBPATTERNS
724
725 The name-to-number translation table that is maintained for named sub-
726 patterns uses 32-bit characters. The pcre32_get_stringtable_entries()
727 function returns the length of each entry in the table as the number of
728 32-bit data units.
729
730
731 OPTION NAMES
732
733 There are two new general option names, PCRE_UTF32 and
734 PCRE_NO_UTF32_CHECK, which correspond to PCRE_UTF8 and
735 PCRE_NO_UTF8_CHECK in the 8-bit library. In fact, these new options
736 define the same bits in the options word. There is a discussion about
737 the validity of UTF-32 strings in the pcreunicode page.
738
739 For the pcre32_config() function there is an option PCRE_CONFIG_UTF32
740 that returns 1 if UTF-32 support is configured, otherwise 0. If this
741 option is given to pcre_config() or pcre16_config(), or if the
742 PCRE_CONFIG_UTF8 or PCRE_CONFIG_UTF16 option is given to pcre32_con-
743 fig(), the result is the PCRE_ERROR_BADOPTION error.
744
745
746 CHARACTER CODES
747
748 In 32-bit mode, when PCRE_UTF32 is not set, character values are
749 treated in the same way as in 8-bit, non UTF-8 mode, except, of course,
750 that they can range from 0 to 0x7fffffff instead of 0 to 0xff. Charac-
751 ter types for characters less than 0xff can therefore be influenced by
752 the locale in the same way as before. Characters greater than 0xff
753 have only one case, and no "type" (such as letter or digit).
754
755 In UTF-32 mode, the character code is Unicode, in the range 0 to
756 0x10ffff, with the exception of values in the range 0xd800 to 0xdfff
757 because those are "surrogate" values that are ill-formed in UTF-32.
758
759 A UTF-32 string can indicate its endianness by special code knows as a
760 byte-order mark (BOM). The PCRE functions do not handle this, expecting
761 strings to be in host byte order. A utility function called
762 pcre32_utf32_to_host_byte_order() is provided to help with this (see
763 above).
764
765
766 ERROR NAMES
767
768 The error PCRE_ERROR_BADUTF32 corresponds to its 8-bit counterpart.
769 The error PCRE_ERROR_BADMODE is given when a compiled pattern is passed
770 to a function that processes patterns in the other mode, for example,
771 if a pattern compiled with pcre_compile() is passed to pcre32_exec().
772
773 There are new error codes whose names begin with PCRE_UTF32_ERR for
774 invalid UTF-32 strings, corresponding to the PCRE_UTF8_ERR codes for
775 UTF-8 strings that are described in the section entitled "Reason codes
776 for invalid UTF-8 strings" in the main pcreapi page. The UTF-32 errors
777 are:
778
779 PCRE_UTF32_ERR1 Surrogate character (range from 0xd800 to 0xdfff)
780 PCRE_UTF32_ERR2 Non-character
781 PCRE_UTF32_ERR3 Character > 0x10ffff
782
783
784 ERROR TEXTS
785
786 If there is an error while compiling a pattern, the error text that is
787 passed back by pcre32_compile() or pcre32_compile2() is still an 8-bit
788 character string, zero-terminated.
789
790
791 CALLOUTS
792
793 The subject and mark fields in the callout block that is passed to a
794 callout function point to 32-bit vectors.
795
796
797 TESTING
798
799 The pcretest program continues to operate with 8-bit input and output
800 files, but it can be used for testing the 32-bit library. If it is run
801 with the command line option -32, patterns and subject strings are con-
802 verted from 8-bit to 32-bit before being passed to PCRE, and the 32-bit
803 library functions are used instead of the 8-bit ones. Returned 32-bit
804 strings are converted to 8-bit for output. If both the 8-bit and the
805 16-bit libraries were not compiled, pcretest defaults to 32-bit and the
806 -32 option is ignored.
807
808 When PCRE is being built, the RunTest script that is called by "make
809 check" uses the pcretest -C option to discover which of the 8-bit,
810 16-bit and 32-bit libraries has been built, and runs the tests appro-
811 priately.
812
813
814 NOT SUPPORTED IN 32-BIT MODE
815
816 Not all the features of the 8-bit library are available with the 32-bit
817 library. The C++ and POSIX wrapper functions support only the 8-bit
818 library, and the pcregrep program is at present 8-bit only.
819
820
821 AUTHOR
822
823 Philip Hazel
824 University Computing Service
825 Cambridge CB2 3QH, England.
826
827
828 REVISION
829
830 Last updated: 24 June 2012
831 Copyright (c) 1997-2012 University of Cambridge.
832 ------------------------------------------------------------------------------
833
834
835 PCREBUILD(3) PCREBUILD(3)
836
837
838 NAME
839 PCRE - Perl-compatible regular expressions
840
841
842 PCRE BUILD-TIME OPTIONS
843
844 This document describes the optional features of PCRE that can be
845 selected when the library is compiled. It assumes use of the configure
846 script, where the optional features are selected or deselected by pro-
847 viding options to configure before running the make command. However,
848 the same options can be selected in both Unix-like and non-Unix-like
849 environments using the GUI facility of cmake-gui if you are using CMake
850 instead of configure to build PCRE.
851
852 There is a lot more information about building PCRE without using con-
853 figure (including information about using CMake or building "by hand")
854 in the file called NON-AUTOTOOLS-BUILD, which is part of the PCRE dis-
855 tribution. You should consult this file as well as the README file if
856 you are building in a non-Unix-like environment.
857
858 The complete list of options for configure (which includes the standard
859 ones such as the selection of the installation directory) can be
860 obtained by running
861
862 ./configure --help
863
864 The following sections include descriptions of options whose names
865 begin with --enable or --disable. These settings specify changes to the
866 defaults for the configure command. Because of the way that configure
867 works, --enable and --disable always come in pairs, so the complemen-
868 tary option always exists as well, but as it specifies the default, it
869 is not described.
870
871
872 BUILDING 8-BIT, 16-BIT AND 32-BIT LIBRARIES
873
874 By default, a library called libpcre is built, containing functions
875 that take string arguments contained in vectors of bytes, either as
876 single-byte characters, or interpreted as UTF-8 strings. You can also
877 build a separate library, called libpcre16, in which strings are con-
878 tained in vectors of 16-bit data units and interpreted either as sin-
879 gle-unit characters or UTF-16 strings, by adding
880
881 --enable-pcre16
882
883 to the configure command. You can also build a separate library, called
884 libpcre32, in which strings are contained in vectors of 32-bit data
885 units and interpreted either as single-unit characters or UTF-32
886 strings, by adding
887
888 --enable-pcre32
889
890 to the configure command. If you do not want the 8-bit library, add
891
892 --disable-pcre8
893
894 as well. At least one of the three libraries must be built. Note that
895 the C++ and POSIX wrappers are for the 8-bit library only, and that
896 pcregrep is an 8-bit program. None of these are built if you select
897 only the 16-bit or 32-bit libraries.
898
899
900 BUILDING SHARED AND STATIC LIBRARIES
901
902 The PCRE building process uses libtool to build both shared and static
903 Unix libraries by default. You can suppress one of these by adding one
904 of
905
906 --disable-shared
907 --disable-static
908
909 to the configure command, as required.
910
911
912 C++ SUPPORT
913
914 By default, if the 8-bit library is being built, the configure script
915 will search for a C++ compiler and C++ header files. If it finds them,
916 it automatically builds the C++ wrapper library (which supports only
917 8-bit strings). You can disable this by adding
918
919 --disable-cpp
920
921 to the configure command.
922
923
924 UTF-8, UTF-16 AND UTF-32 SUPPORT
925
926 To build PCRE with support for UTF Unicode character strings, add
927
928 --enable-utf
929
930 to the configure command. This setting applies to all three libraries,
931 adding support for UTF-8 to the 8-bit library, support for UTF-16 to
932 the 16-bit library, and support for UTF-32 to the to the 32-bit
933 library. There are no separate options for enabling UTF-8, UTF-16 and
934 UTF-32 independently because that would allow ridiculous settings such
935 as requesting UTF-16 support while building only the 8-bit library. It
936 is not possible to build one library with UTF support and another with-
937 out in the same configuration. (For backwards compatibility, --enable-
938 utf8 is a synonym of --enable-utf.)
939
940 Of itself, this setting does not make PCRE treat strings as UTF-8,
941 UTF-16 or UTF-32. As well as compiling PCRE with this option, you also
942 have have to set the PCRE_UTF8, PCRE_UTF16 or PCRE_UTF32 option (as
943 appropriate) when you call one of the pattern compiling functions.
944
945 If you set --enable-utf when compiling in an EBCDIC environment, PCRE
946 expects its input to be either ASCII or UTF-8 (depending on the run-
947 time option). It is not possible to support both EBCDIC and UTF-8 codes
948 in the same version of the library. Consequently, --enable-utf and
949 --enable-ebcdic are mutually exclusive.
950
951
952 UNICODE CHARACTER PROPERTY SUPPORT
953
954 UTF support allows the libraries to process character codepoints up to
955 0x10ffff in the strings that they handle. On its own, however, it does
956 not provide any facilities for accessing the properties of such charac-
957 ters. If you want to be able to use the pattern escapes \P, \p, and \X,
958 which refer to Unicode character properties, you must add
959
960 --enable-unicode-properties
961
962 to the configure command. This implies UTF support, even if you have
963 not explicitly requested it.
964
965 Including Unicode property support adds around 30K of tables to the
966 PCRE library. Only the general category properties such as Lu and Nd
967 are supported. Details are given in the pcrepattern documentation.
968
969
970 JUST-IN-TIME COMPILER SUPPORT
971
972 Just-in-time compiler support is included in the build by specifying
973
974 --enable-jit
975
976 This support is available only for certain hardware architectures. If
977 this option is set for an unsupported architecture, a compile time
978 error occurs. See the pcrejit documentation for a discussion of JIT
979 usage. When JIT support is enabled, pcregrep automatically makes use of
980 it, unless you add
981
982 --disable-pcregrep-jit
983
984 to the "configure" command.
985
986
987 CODE VALUE OF NEWLINE
988
989 By default, PCRE interprets the linefeed (LF) character as indicating
990 the end of a line. This is the normal newline character on Unix-like
991 systems. You can compile PCRE to use carriage return (CR) instead, by
992 adding
993
994 --enable-newline-is-cr
995
996 to the configure command. There is also a --enable-newline-is-lf
997 option, which explicitly specifies linefeed as the newline character.
998
999 Alternatively, you can specify that line endings are to be indicated by
1000 the two character sequence CRLF. If you want this, add
1001
1002 --enable-newline-is-crlf
1003
1004 to the configure command. There is a fourth option, specified by
1005
1006 --enable-newline-is-anycrlf
1007
1008 which causes PCRE to recognize any of the three sequences CR, LF, or
1009 CRLF as indicating a line ending. Finally, a fifth option, specified by
1010
1011 --enable-newline-is-any
1012
1013 causes PCRE to recognize any Unicode newline sequence.
1014
1015 Whatever line ending convention is selected when PCRE is built can be
1016 overridden when the library functions are called. At build time it is
1017 conventional to use the standard for your operating system.
1018
1019
1020 WHAT \R MATCHES
1021
1022 By default, the sequence \R in a pattern matches any Unicode newline
1023 sequence, whatever has been selected as the line ending sequence. If
1024 you specify
1025
1026 --enable-bsr-anycrlf
1027
1028 the default is changed so that \R matches only CR, LF, or CRLF. What-
1029 ever is selected when PCRE is built can be overridden when the library
1030 functions are called.
1031
1032
1033 POSIX MALLOC USAGE
1034
1035 When the 8-bit library is called through the POSIX interface (see the
1036 pcreposix documentation), additional working storage is required for
1037 holding the pointers to capturing substrings, because PCRE requires
1038 three integers per substring, whereas the POSIX interface provides only
1039 two. If the number of expected substrings is small, the wrapper func-
1040 tion uses space on the stack, because this is faster than using mal-
1041 loc() for each call. The default threshold above which the stack is no
1042 longer used is 10; it can be changed by adding a setting such as
1043
1044 --with-posix-malloc-threshold=20
1045
1046 to the configure command.
1047
1048
1049 HANDLING VERY LARGE PATTERNS
1050
1051 Within a compiled pattern, offset values are used to point from one
1052 part to another (for example, from an opening parenthesis to an alter-
1053 nation metacharacter). By default, in the 8-bit and 16-bit libraries,
1054 two-byte values are used for these offsets, leading to a maximum size
1055 for a compiled pattern of around 64K. This is sufficient to handle all
1056 but the most gigantic patterns. Nevertheless, some people do want to
1057 process truly enormous patterns, so it is possible to compile PCRE to
1058 use three-byte or four-byte offsets by adding a setting such as
1059
1060 --with-link-size=3
1061
1062 to the configure command. The value given must be 2, 3, or 4. For the
1063 16-bit library, a value of 3 is rounded up to 4. In these libraries,
1064 using longer offsets slows down the operation of PCRE because it has to
1065 load additional data when handling them. For the 32-bit library the
1066 value is always 4 and cannot be overridden; the value of --with-link-
1067 size is ignored.
1068
1069
1070 AVOIDING EXCESSIVE STACK USAGE
1071
1072 When matching with the pcre_exec() function, PCRE implements backtrack-
1073 ing by making recursive calls to an internal function called match().
1074 In environments where the size of the stack is limited, this can se-
1075 verely limit PCRE's operation. (The Unix environment does not usually
1076 suffer from this problem, but it may sometimes be necessary to increase
1077 the maximum stack size. There is a discussion in the pcrestack docu-
1078 mentation.) An alternative approach to recursion that uses memory from
1079 the heap to remember data, instead of using recursive function calls,
1080 has been implemented to work round the problem of limited stack size.
1081 If you want to build a version of PCRE that works this way, add
1082
1083 --disable-stack-for-recursion
1084
1085 to the configure command. With this configuration, PCRE will use the
1086 pcre_stack_malloc and pcre_stack_free variables to call memory manage-
1087 ment functions. By default these point to malloc() and free(), but you
1088 can replace the pointers so that your own functions are used instead.
1089
1090 Separate functions are provided rather than using pcre_malloc and
1091 pcre_free because the usage is very predictable: the block sizes
1092 requested are always the same, and the blocks are always freed in
1093 reverse order. A calling program might be able to implement optimized
1094 functions that perform better than malloc() and free(). PCRE runs
1095 noticeably more slowly when built in this way. This option affects only
1096 the pcre_exec() function; it is not relevant for pcre_dfa_exec().
1097
1098
1099 LIMITING PCRE RESOURCE USAGE
1100
1101 Internally, PCRE has a function called match(), which it calls repeat-
1102 edly (sometimes recursively) when matching a pattern with the
1103 pcre_exec() function. By controlling the maximum number of times this
1104 function may be called during a single matching operation, a limit can
1105 be placed on the resources used by a single call to pcre_exec(). The
1106 limit can be changed at run time, as described in the pcreapi documen-
1107 tation. The default is 10 million, but this can be changed by adding a
1108 setting such as
1109
1110 --with-match-limit=500000
1111
1112 to the configure command. This setting has no effect on the
1113 pcre_dfa_exec() matching function.
1114
1115 In some environments it is desirable to limit the depth of recursive
1116 calls of match() more strictly than the total number of calls, in order
1117 to restrict the maximum amount of stack (or heap, if --disable-stack-
1118 for-recursion is specified) that is used. A second limit controls this;
1119 it defaults to the value that is set for --with-match-limit, which
1120 imposes no additional constraints. However, you can set a lower limit
1121 by adding, for example,
1122
1123 --with-match-limit-recursion=10000
1124
1125 to the configure command. This value can also be overridden at run
1126 time.
1127
1128
1129 CREATING CHARACTER TABLES AT BUILD TIME
1130
1131 PCRE uses fixed tables for processing characters whose code values are
1132 less than 256. By default, PCRE is built with a set of tables that are
1133 distributed in the file pcre_chartables.c.dist. These tables are for
1134 ASCII codes only. If you add
1135
1136 --enable-rebuild-chartables
1137
1138 to the configure command, the distributed tables are no longer used.
1139 Instead, a program called dftables is compiled and run. This outputs
1140 the source for new set of tables, created in the default locale of your
1141 C run-time system. (This method of replacing the tables does not work
1142 if you are cross compiling, because dftables is run on the local host.
1143 If you need to create alternative tables when cross compiling, you will
1144 have to do so "by hand".)
1145
1146
1147 USING EBCDIC CODE
1148
1149 PCRE assumes by default that it will run in an environment where the
1150 character code is ASCII (or Unicode, which is a superset of ASCII).
1151 This is the case for most computer operating systems. PCRE can, how-
1152 ever, be compiled to run in an EBCDIC environment by adding
1153
1154 --enable-ebcdic
1155
1156 to the configure command. This setting implies --enable-rebuild-charta-
1157 bles. You should only use it if you know that you are in an EBCDIC
1158 environment (for example, an IBM mainframe operating system). The
1159 --enable-ebcdic option is incompatible with --enable-utf.
1160
1161 The EBCDIC character that corresponds to an ASCII LF is assumed to have
1162 the value 0x15 by default. However, in some EBCDIC environments, 0x25
1163 is used. In such an environment you should use
1164
1165 --enable-ebcdic-nl25
1166
1167 as well as, or instead of, --enable-ebcdic. The EBCDIC character for CR
1168 has the same value as in ASCII, namely, 0x0d. Whichever of 0x15 and
1169 0x25 is not chosen as LF is made to correspond to the Unicode NEL char-
1170 acter (which, in Unicode, is 0x85).
1171
1172 The options that select newline behaviour, such as --enable-newline-is-
1173 cr, and equivalent run-time options, refer to these character values in
1174 an EBCDIC environment.
1175
1176
1177 PCREGREP OPTIONS FOR COMPRESSED FILE SUPPORT
1178
1179 By default, pcregrep reads all files as plain text. You can build it so
1180 that it recognizes files whose names end in .gz or .bz2, and reads them
1181 with libz or libbz2, respectively, by adding one or both of
1182
1183 --enable-pcregrep-libz
1184 --enable-pcregrep-libbz2
1185
1186 to the configure command. These options naturally require that the rel-
1187 evant libraries are installed on your system. Configuration will fail
1188 if they are not.
1189
1190
1191 PCREGREP BUFFER SIZE
1192
1193 pcregrep uses an internal buffer to hold a "window" on the file it is
1194 scanning, in order to be able to output "before" and "after" lines when
1195 it finds a match. The size of the buffer is controlled by a parameter
1196 whose default value is 20K. The buffer itself is three times this size,
1197 but because of the way it is used for holding "before" lines, the long-
1198 est line that is guaranteed to be processable is the parameter size.
1199 You can change the default parameter value by adding, for example,
1200
1201 --with-pcregrep-bufsize=50K
1202
1203 to the configure command. The caller of pcregrep can, however, override
1204 this value by specifying a run-time option.
1205
1206
1207 PCRETEST OPTION FOR LIBREADLINE SUPPORT
1208
1209 If you add
1210
1211 --enable-pcretest-libreadline
1212
1213 to the configure command, pcretest is linked with the libreadline
1214 library, and when its input is from a terminal, it reads it using the
1215 readline() function. This provides line-editing and history facilities.
1216 Note that libreadline is GPL-licensed, so if you distribute a binary of
1217 pcretest linked in this way, there may be licensing issues.
1218
1219 Setting this option causes the -lreadline option to be added to the
1220 pcretest build. In many operating environments with a sytem-installed
1221 libreadline this is sufficient. However, in some environments (e.g. if
1222 an unmodified distribution version of readline is in use), some extra
1223 configuration may be necessary. The INSTALL file for libreadline says
1224 this:
1225
1226 "Readline uses the termcap functions, but does not link with the
1227 termcap or curses library itself, allowing applications which link
1228 with readline the to choose an appropriate library."
1229
1230 If your environment has not been set up so that an appropriate library
1231 is automatically included, you may need to add something like
1232
1233 LIBS="-ncurses"
1234
1235 immediately before the configure command.
1236
1237
1238 DEBUGGING WITH VALGRIND SUPPORT
1239
1240 By adding the
1241
1242 --enable-valgrind
1243
1244 option to to the configure command, PCRE will use valgrind annotations
1245 to mark certain memory regions as unaddressable. This allows it to
1246 detect invalid memory accesses, and is mostly useful for debugging PCRE
1247 itself.
1248
1249
1250 CODE COVERAGE REPORTING
1251
1252 If your C compiler is gcc, you can build a version of PCRE that can
1253 generate a code coverage report for its test suite. To enable this, you
1254 must install lcov version 1.6 or above. Then specify
1255
1256 --enable-coverage
1257
1258 to the configure command and build PCRE in the usual way.
1259
1260 Note that using ccache (a caching C compiler) is incompatible with code
1261 coverage reporting. If you have configured ccache to run automatically
1262 on your system, you must set the environment variable
1263
1264 CCACHE_DISABLE=1
1265
1266 before running make to build PCRE, so that ccache is not used.
1267
1268 When --enable-coverage is used, the following addition targets are
1269 added to the Makefile:
1270
1271 make coverage
1272
1273 This creates a fresh coverage report for the PCRE test suite. It is
1274 equivalent to running "make coverage-reset", "make coverage-baseline",
1275 "make check", and then "make coverage-report".
1276
1277 make coverage-reset
1278
1279 This zeroes the coverage counters, but does nothing else.
1280
1281 make coverage-baseline
1282
1283 This captures baseline coverage information.
1284
1285 make coverage-report
1286
1287 This creates the coverage report.
1288
1289 make coverage-clean-report
1290
1291 This removes the generated coverage report without cleaning the cover-
1292 age data itself.
1293
1294 make coverage-clean-data
1295
1296 This removes the captured coverage data without removing the coverage
1297 files created at compile time (*.gcno).
1298
1299 make coverage-clean
1300
1301 This cleans all coverage data including the generated coverage report.
1302 For more information about code coverage, see the gcov and lcov docu-
1303 mentation.
1304
1305
1306 SEE ALSO
1307
1308 pcreapi(3), pcre16, pcre32, pcre_config(3).
1309
1310
1311 AUTHOR
1312
1313 Philip Hazel
1314 University Computing Service
1315 Cambridge CB2 3QH, England.
1316
1317
1318 REVISION
1319
1320 Last updated: 30 October 2012
1321 Copyright (c) 1997-2012 University of Cambridge.
1322 ------------------------------------------------------------------------------
1323
1324
1325 PCREMATCHING(3) PCREMATCHING(3)
1326
1327
1328 NAME
1329 PCRE - Perl-compatible regular expressions
1330
1331
1332 PCRE MATCHING ALGORITHMS
1333
1334 This document describes the two different algorithms that are available
1335 in PCRE for matching a compiled regular expression against a given sub-
1336 ject string. The "standard" algorithm is the one provided by the
1337 pcre_exec(), pcre16_exec() and pcre32_exec() functions. These work in
1338 the same as as Perl's matching function, and provide a Perl-compatible
1339 matching operation. The just-in-time (JIT) optimization that is
1340 described in the pcrejit documentation is compatible with these func-
1341 tions.
1342
1343 An alternative algorithm is provided by the pcre_dfa_exec(),
1344 pcre16_dfa_exec() and pcre32_dfa_exec() functions; they operate in a
1345 different way, and are not Perl-compatible. This alternative has advan-
1346 tages and disadvantages compared with the standard algorithm, and these
1347 are described below.
1348
1349 When there is only one possible way in which a given subject string can
1350 match a pattern, the two algorithms give the same answer. A difference
1351 arises, however, when there are multiple possibilities. For example, if
1352 the pattern
1353
1354 ^<.*>
1355
1356 is matched against the string
1357
1358 <something> <something else> <something further>
1359
1360 there are three possible answers. The standard algorithm finds only one
1361 of them, whereas the alternative algorithm finds all three.
1362
1363
1364 REGULAR EXPRESSIONS AS TREES
1365
1366 The set of strings that are matched by a regular expression can be rep-
1367 resented as a tree structure. An unlimited repetition in the pattern
1368 makes the tree of infinite size, but it is still a tree. Matching the
1369 pattern to a given subject string (from a given starting point) can be
1370 thought of as a search of the tree. There are two ways to search a
1371 tree: depth-first and breadth-first, and these correspond to the two
1372 matching algorithms provided by PCRE.
1373
1374
1375 THE STANDARD MATCHING ALGORITHM
1376
1377 In the terminology of Jeffrey Friedl's book "Mastering Regular Expres-
1378 sions", the standard algorithm is an "NFA algorithm". It conducts a
1379 depth-first search of the pattern tree. That is, it proceeds along a
1380 single path through the tree, checking that the subject matches what is
1381 required. When there is a mismatch, the algorithm tries any alterna-
1382 tives at the current point, and if they all fail, it backs up to the
1383 previous branch point in the tree, and tries the next alternative
1384 branch at that level. This often involves backing up (moving to the
1385 left) in the subject string as well. The order in which repetition
1386 branches are tried is controlled by the greedy or ungreedy nature of
1387 the quantifier.
1388
1389 If a leaf node is reached, a matching string has been found, and at
1390 that point the algorithm stops. Thus, if there is more than one possi-
1391 ble match, this algorithm returns the first one that it finds. Whether
1392 this is the shortest, the longest, or some intermediate length depends
1393 on the way the greedy and ungreedy repetition quantifiers are specified
1394 in the pattern.
1395
1396 Because it ends up with a single path through the tree, it is rela-
1397 tively straightforward for this algorithm to keep track of the sub-
1398 strings that are matched by portions of the pattern in parentheses.
1399 This provides support for capturing parentheses and back references.
1400
1401
1402 THE ALTERNATIVE MATCHING ALGORITHM
1403
1404 This algorithm conducts a breadth-first search of the tree. Starting
1405 from the first matching point in the subject, it scans the subject
1406 string from left to right, once, character by character, and as it does
1407 this, it remembers all the paths through the tree that represent valid
1408 matches. In Friedl's terminology, this is a kind of "DFA algorithm",
1409 though it is not implemented as a traditional finite state machine (it
1410 keeps multiple states active simultaneously).
1411
1412 Although the general principle of this matching algorithm is that it
1413 scans the subject string only once, without backtracking, there is one
1414 exception: when a lookaround assertion is encountered, the characters
1415 following or preceding the current point have to be independently
1416 inspected.
1417
1418 The scan continues until either the end of the subject is reached, or
1419 there are no more unterminated paths. At this point, terminated paths
1420 represent the different matching possibilities (if there are none, the
1421 match has failed). Thus, if there is more than one possible match,
1422 this algorithm finds all of them, and in particular, it finds the long-
1423 est. The matches are returned in decreasing order of length. There is
1424 an option to stop the algorithm after the first match (which is neces-
1425 sarily the shortest) is found.
1426
1427 Note that all the matches that are found start at the same point in the
1428 subject. If the pattern
1429
1430 cat(er(pillar)?)?
1431
1432 is matched against the string "the caterpillar catchment", the result
1433 will be the three strings "caterpillar", "cater", and "cat" that start
1434 at the fifth character of the subject. The algorithm does not automati-
1435 cally move on to find matches that start at later positions.
1436
1437 There are a number of features of PCRE regular expressions that are not
1438 supported by the alternative matching algorithm. They are as follows:
1439
1440 1. Because the algorithm finds all possible matches, the greedy or
1441 ungreedy nature of repetition quantifiers is not relevant. Greedy and
1442 ungreedy quantifiers are treated in exactly the same way. However, pos-
1443 sessive quantifiers can make a difference when what follows could also
1444 match what is quantified, for example in a pattern like this:
1445
1446 ^a++\w!
1447
1448 This pattern matches "aaab!" but not "aaa!", which would be matched by
1449 a non-possessive quantifier. Similarly, if an atomic group is present,
1450 it is matched as if it were a standalone pattern at the current point,
1451 and the longest match is then "locked in" for the rest of the overall
1452 pattern.
1453
1454 2. When dealing with multiple paths through the tree simultaneously, it
1455 is not straightforward to keep track of captured substrings for the
1456 different matching possibilities, and PCRE's implementation of this
1457 algorithm does not attempt to do this. This means that no captured sub-
1458 strings are available.
1459
1460 3. Because no substrings are captured, back references within the pat-
1461 tern are not supported, and cause errors if encountered.
1462
1463 4. For the same reason, conditional expressions that use a backrefer-
1464 ence as the condition or test for a specific group recursion are not
1465 supported.
1466
1467 5. Because many paths through the tree may be active, the \K escape
1468 sequence, which resets the start of the match when encountered (but may
1469 be on some paths and not on others), is not supported. It causes an
1470 error if encountered.
1471
1472 6. Callouts are supported, but the value of the capture_top field is
1473 always 1, and the value of the capture_last field is always -1.
1474
1475 7. The \C escape sequence, which (in the standard algorithm) always
1476 matches a single data unit, even in UTF-8, UTF-16 or UTF-32 modes, is
1477 not supported in these modes, because the alternative algorithm moves
1478 through the subject string one character (not data unit) at a time, for
1479 all active paths through the tree.
1480
1481 8. Except for (*FAIL), the backtracking control verbs such as (*PRUNE)
1482 are not supported. (*FAIL) is supported, and behaves like a failing
1483 negative assertion.
1484
1485
1486 ADVANTAGES OF THE ALTERNATIVE ALGORITHM
1487
1488 Using the alternative matching algorithm provides the following advan-
1489 tages:
1490
1491 1. All possible matches (at a single point in the subject) are automat-
1492 ically found, and in particular, the longest match is found. To find
1493 more than one match using the standard algorithm, you have to do kludgy
1494 things with callouts.
1495
1496 2. Because the alternative algorithm scans the subject string just
1497 once, and never needs to backtrack (except for lookbehinds), it is pos-
1498 sible to pass very long subject strings to the matching function in
1499 several pieces, checking for partial matching each time. Although it is
1500 possible to do multi-segment matching using the standard algorithm by
1501 retaining partially matched substrings, it is more complicated. The
1502 pcrepartial documentation gives details of partial matching and dis-
1503 cusses multi-segment matching.
1504
1505
1506 DISADVANTAGES OF THE ALTERNATIVE ALGORITHM
1507
1508 The alternative algorithm suffers from a number of disadvantages:
1509
1510 1. It is substantially slower than the standard algorithm. This is
1511 partly because it has to search for all possible matches, but is also
1512 because it is less susceptible to optimization.
1513
1514 2. Capturing parentheses and back references are not supported.
1515
1516 3. Although atomic groups are supported, their use does not provide the
1517 performance advantage that it does for the standard algorithm.
1518
1519
1520 AUTHOR
1521
1522 Philip Hazel
1523 University Computing Service
1524 Cambridge CB2 3QH, England.
1525
1526
1527 REVISION
1528
1529 Last updated: 08 January 2012
1530 Copyright (c) 1997-2012 University of Cambridge.
1531 ------------------------------------------------------------------------------
1532
1533
1534 PCREAPI(3) PCREAPI(3)
1535
1536
1537 NAME
1538 PCRE - Perl-compatible regular expressions
1539
1540 #include <pcre.h>
1541
1542
1543 PCRE NATIVE API BASIC FUNCTIONS
1544
1545 pcre *pcre_compile(const char *pattern, int options,
1546 const char **errptr, int *erroffset,
1547 const unsigned char *tableptr);
1548
1549 pcre *pcre_compile2(const char *pattern, int options,
1550 int *errorcodeptr,
1551 const char **errptr, int *erroffset,
1552 const unsigned char *tableptr);
1553
1554 pcre_extra *pcre_study(const pcre *code, int options,
1555 const char **errptr);
1556
1557 void pcre_free_study(pcre_extra *extra);
1558
1559 int pcre_exec(const pcre *code, const pcre_extra *extra,
1560 const char *subject, int length, int startoffset,
1561 int options, int *ovector, int ovecsize);
1562
1563 int pcre_dfa_exec(const pcre *code, const pcre_extra *extra,
1564 const char *subject, int length, int startoffset,
1565 int options, int *ovector, int ovecsize,
1566 int *workspace, int wscount);
1567
1568
1569 PCRE NATIVE API STRING EXTRACTION FUNCTIONS
1570
1571 int pcre_copy_named_substring(const pcre *code,
1572 const char *subject, int *ovector,
1573 int stringcount, const char *stringname,
1574 char *buffer, int buffersize);
1575
1576 int pcre_copy_substring(const char *subject, int *ovector,
1577 int stringcount, int stringnumber, char *buffer,
1578 int buffersize);
1579
1580 int pcre_get_named_substring(const pcre *code,
1581 const char *subject, int *ovector,
1582 int stringcount, const char *stringname,
1583 const char **stringptr);
1584
1585 int pcre_get_stringnumber(const pcre *code,
1586 const char *name);
1587
1588 int pcre_get_stringtable_entries(const pcre *code,
1589 const char *name, char **first, char **last);
1590
1591 int pcre_get_substring(const char *subject, int *ovector,
1592 int stringcount, int stringnumber,
1593 const char **stringptr);
1594
1595 int pcre_get_substring_list(const char *subject,
1596 int *ovector, int stringcount, const char ***listptr);
1597
1598 void pcre_free_substring(const char *stringptr);
1599
1600 void pcre_free_substring_list(const char **stringptr);
1601
1602
1603 PCRE NATIVE API AUXILIARY FUNCTIONS
1604
1605 int pcre_jit_exec(const pcre *code, const pcre_extra *extra,
1606 const char *subject, int length, int startoffset,
1607 int options, int *ovector, int ovecsize,
1608 pcre_jit_stack *jstack);
1609
1610 pcre_jit_stack *pcre_jit_stack_alloc(int startsize, int maxsize);
1611
1612 void pcre_jit_stack_free(pcre_jit_stack *stack);
1613
1614 void pcre_assign_jit_stack(pcre_extra *extra,
1615 pcre_jit_callback callback, void *data);
1616
1617 const unsigned char *pcre_maketables(void);
1618
1619 int pcre_fullinfo(const pcre *code, const pcre_extra *extra,
1620 int what, void *where);
1621
1622 int pcre_refcount(pcre *code, int adjust);
1623
1624 int pcre_config(int what, void *where);
1625
1626 const char *pcre_version(void);
1627
1628 int pcre_pattern_to_host_byte_order(pcre *code,
1629 pcre_extra *extra, const unsigned char *tables);
1630
1631
1632 PCRE NATIVE API INDIRECTED FUNCTIONS
1633
1634 void *(*pcre_malloc)(size_t);
1635
1636 void (*pcre_free)(void *);
1637
1638 void *(*pcre_stack_malloc)(size_t);
1639
1640 void (*pcre_stack_free)(void *);
1641
1642 int (*pcre_callout)(pcre_callout_block *);
1643
1644
1645 PCRE 8-BIT, 16-BIT AND 32-BIT LIBRARIES
1646
1647 From release 8.30, PCRE can be compiled as a library for handling
1648 16-bit character strings as well as, or instead of, the original
1649 library that handles 8-bit character strings. From release 8.32, PCRE
1650 can also be compiled as a library for handling 32-bit character
1651 strings. To avoid too much complication, this document describes the
1652 8-bit versions of the functions, with only occasional references to the
1653 16-bit and 32-bit libraries.
1654
1655 The 16-bit and 32-bit functions operate in the same way as their 8-bit
1656 counterparts; they just use different data types for their arguments
1657 and results, and their names start with pcre16_ or pcre32_ instead of
1658 pcre_. For every option that has UTF8 in its name (for example,
1659 PCRE_UTF8), there are corresponding 16-bit and 32-bit names with UTF8
1660 replaced by UTF16 or UTF32, respectively. This facility is in fact just
1661 cosmetic; the 16-bit and 32-bit option names define the same bit val-
1662 ues.
1663
1664 References to bytes and UTF-8 in this document should be read as refer-
1665 ences to 16-bit data quantities and UTF-16 when using the 16-bit
1666 library, or 32-bit data quantities and UTF-32 when using the 32-bit
1667 library, unless specified otherwise. More details of the specific dif-
1668 ferences for the 16-bit and 32-bit libraries are given in the pcre16
1669 and pcre32 pages.
1670
1671
1672 PCRE API OVERVIEW
1673
1674 PCRE has its own native API, which is described in this document. There
1675 are also some wrapper functions (for the 8-bit library only) that cor-
1676 respond to the POSIX regular expression API, but they do not give
1677 access to all the functionality. They are described in the pcreposix
1678 documentation. Both of these APIs define a set of C function calls. A
1679 C++ wrapper (again for the 8-bit library only) is also distributed with
1680 PCRE. It is documented in the pcrecpp page.
1681
1682 The native API C function prototypes are defined in the header file
1683 pcre.h, and on Unix-like systems the (8-bit) library itself is called
1684 libpcre. It can normally be accessed by adding -lpcre to the command
1685 for linking an application that uses PCRE. The header file defines the
1686 macros PCRE_MAJOR and PCRE_MINOR to contain the major and minor release
1687 numbers for the library. Applications can use these to include support
1688 for different releases of PCRE.
1689
1690 In a Windows environment, if you want to statically link an application
1691 program against a non-dll pcre.a file, you must define PCRE_STATIC
1692 before including pcre.h or pcrecpp.h, because otherwise the pcre_mal-
1693 loc() and pcre_free() exported functions will be declared
1694 __declspec(dllimport), with unwanted results.
1695
1696 The functions pcre_compile(), pcre_compile2(), pcre_study(), and
1697 pcre_exec() are used for compiling and matching regular expressions in
1698 a Perl-compatible manner. A sample program that demonstrates the sim-
1699 plest way of using them is provided in the file called pcredemo.c in
1700 the PCRE source distribution. A listing of this program is given in the
1701 pcredemo documentation, and the pcresample documentation describes how
1702 to compile and run it.
1703
1704 Just-in-time compiler support is an optional feature of PCRE that can
1705 be built in appropriate hardware environments. It greatly speeds up the
1706 matching performance of many patterns. Simple programs can easily
1707 request that it be used if available, by setting an option that is
1708 ignored when it is not relevant. More complicated programs might need
1709 to make use of the functions pcre_jit_stack_alloc(),
1710 pcre_jit_stack_free(), and pcre_assign_jit_stack() in order to control
1711 the JIT code's memory usage.
1712
1713 From release 8.32 there is also a direct interface for JIT execution,
1714 which gives improved performance. The JIT-specific functions are dis-
1715 cussed in the pcrejit documentation.
1716
1717 A second matching function, pcre_dfa_exec(), which is not Perl-compati-
1718 ble, is also provided. This uses a different algorithm for the match-
1719 ing. The alternative algorithm finds all possible matches (at a given
1720 point in the subject), and scans the subject just once (unless there
1721 are lookbehind assertions). However, this algorithm does not return
1722 captured substrings. A description of the two matching algorithms and
1723 their advantages and disadvantages is given in the pcrematching docu-
1724 mentation.
1725
1726 In addition to the main compiling and matching functions, there are
1727 convenience functions for extracting captured substrings from a subject
1728 string that is matched by pcre_exec(). They are:
1729
1730 pcre_copy_substring()
1731 pcre_copy_named_substring()
1732 pcre_get_substring()
1733 pcre_get_named_substring()
1734 pcre_get_substring_list()
1735 pcre_get_stringnumber()
1736 pcre_get_stringtable_entries()
1737
1738 pcre_free_substring() and pcre_free_substring_list() are also provided,
1739 to free the memory used for extracted strings.
1740
1741 The function pcre_maketables() is used to build a set of character
1742 tables in the current locale for passing to pcre_compile(),
1743 pcre_exec(), or pcre_dfa_exec(). This is an optional facility that is
1744 provided for specialist use. Most commonly, no special tables are
1745 passed, in which case internal tables that are generated when PCRE is
1746 built are used.
1747
1748 The function pcre_fullinfo() is used to find out information about a
1749 compiled pattern. The function pcre_version() returns a pointer to a
1750 string containing the version of PCRE and its date of release.
1751
1752 The function pcre_refcount() maintains a reference count in a data
1753 block containing a compiled pattern. This is provided for the benefit
1754 of object-oriented applications.
1755
1756 The global variables pcre_malloc and pcre_free initially contain the
1757 entry points of the standard malloc() and free() functions, respec-
1758 tively. PCRE calls the memory management functions via these variables,
1759 so a calling program can replace them if it wishes to intercept the
1760 calls. This should be done before calling any PCRE functions.
1761
1762 The global variables pcre_stack_malloc and pcre_stack_free are also
1763 indirections to memory management functions. These special functions
1764 are used only when PCRE is compiled to use the heap for remembering
1765 data, instead of recursive function calls, when running the pcre_exec()
1766 function. See the pcrebuild documentation for details of how to do
1767 this. It is a non-standard way of building PCRE, for use in environ-
1768 ments that have limited stacks. Because of the greater use of memory
1769 management, it runs more slowly. Separate functions are provided so
1770 that special-purpose external code can be used for this case. When
1771 used, these functions are always called in a stack-like manner (last
1772 obtained, first freed), and always for memory blocks of the same size.
1773 There is a discussion about PCRE's stack usage in the pcrestack docu-
1774 mentation.
1775
1776 The global variable pcre_callout initially contains NULL. It can be set
1777 by the caller to a "callout" function, which PCRE will then call at
1778 specified points during a matching operation. Details are given in the
1779 pcrecallout documentation.
1780
1781
1782 NEWLINES
1783
1784 PCRE supports five different conventions for indicating line breaks in
1785 strings: a single CR (carriage return) character, a single LF (line-
1786 feed) character, the two-character sequence CRLF, any of the three pre-
1787 ceding, or any Unicode newline sequence. The Unicode newline sequences
1788 are the three just mentioned, plus the single characters VT (vertical
1789 tab, U+000B), FF (form feed, U+000C), NEL (next line, U+0085), LS (line
1790 separator, U+2028), and PS (paragraph separator, U+2029).
1791
1792 Each of the first three conventions is used by at least one operating
1793 system as its standard newline sequence. When PCRE is built, a default
1794 can be specified. The default default is LF, which is the Unix stan-
1795 dard. When PCRE is run, the default can be overridden, either when a
1796 pattern is compiled, or when it is matched.
1797
1798 At compile time, the newline convention can be specified by the options
1799 argument of pcre_compile(), or it can be specified by special text at
1800 the start of the pattern itself; this overrides any other settings. See
1801 the pcrepattern page for details of the special character sequences.
1802
1803 In the PCRE documentation the word "newline" is used to mean "the char-
1804 acter or pair of characters that indicate a line break". The choice of
1805 newline convention affects the handling of the dot, circumflex, and
1806 dollar metacharacters, the handling of #-comments in /x mode, and, when
1807 CRLF is a recognized line ending sequence, the match position advance-
1808 ment for a non-anchored pattern. There is more detail about this in the
1809 section on pcre_exec() options below.
1810
1811 The choice of newline convention does not affect the interpretation of
1812 the \n or \r escape sequences, nor does it affect what \R matches,
1813 which is controlled in a similar way, but by separate options.
1814
1815
1816 MULTITHREADING
1817
1818 The PCRE functions can be used in multi-threading applications, with
1819 the proviso that the memory management functions pointed to by
1820 pcre_malloc, pcre_free, pcre_stack_malloc, and pcre_stack_free, and the
1821 callout function pointed to by pcre_callout, are shared by all threads.
1822
1823 The compiled form of a regular expression is not altered during match-
1824 ing, so the same compiled pattern can safely be used by several threads
1825 at once.
1826
1827 If the just-in-time optimization feature is being used, it needs sepa-
1828 rate memory stack areas for each thread. See the pcrejit documentation
1829 for more details.
1830
1831
1832 SAVING PRECOMPILED PATTERNS FOR LATER USE
1833
1834 The compiled form of a regular expression can be saved and re-used at a
1835 later time, possibly by a different program, and even on a host other
1836 than the one on which it was compiled. Details are given in the
1837 pcreprecompile documentation, which includes a description of the
1838 pcre_pattern_to_host_byte_order() function. However, compiling a regu-
1839 lar expression with one version of PCRE for use with a different ver-
1840 sion is not guaranteed to work and may cause crashes.
1841
1842
1843 CHECKING BUILD-TIME OPTIONS
1844
1845 int pcre_config(int what, void *where);
1846
1847 The function pcre_config() makes it possible for a PCRE client to dis-
1848 cover which optional features have been compiled into the PCRE library.
1849 The pcrebuild documentation has more details about these optional fea-
1850 tures.
1851
1852 The first argument for pcre_config() is an integer, specifying which
1853 information is required; the second argument is a pointer to a variable
1854 into which the information is placed. The returned value is zero on
1855 success, or the negative error code PCRE_ERROR_BADOPTION if the value
1856 in the first argument is not recognized. The following information is
1857 available:
1858
1859 PCRE_CONFIG_UTF8
1860
1861 The output is an integer that is set to one if UTF-8 support is avail-
1862 able; otherwise it is set to zero. This value should normally be given
1863 to the 8-bit version of this function, pcre_config(). If it is given to
1864 the 16-bit or 32-bit version of this function, the result is
1865 PCRE_ERROR_BADOPTION.
1866
1867 PCRE_CONFIG_UTF16
1868
1869 The output is an integer that is set to one if UTF-16 support is avail-
1870 able; otherwise it is set to zero. This value should normally be given
1871 to the 16-bit version of this function, pcre16_config(). If it is given
1872 to the 8-bit or 32-bit version of this function, the result is
1873 PCRE_ERROR_BADOPTION.
1874
1875 PCRE_CONFIG_UTF32
1876
1877 The output is an integer that is set to one if UTF-32 support is avail-
1878 able; otherwise it is set to zero. This value should normally be given
1879 to the 32-bit version of this function, pcre32_config(). If it is given
1880 to the 8-bit or 16-bit version of this function, the result is
1881 PCRE_ERROR_BADOPTION.
1882
1883 PCRE_CONFIG_UNICODE_PROPERTIES
1884
1885 The output is an integer that is set to one if support for Unicode
1886 character properties is available; otherwise it is set to zero.
1887
1888 PCRE_CONFIG_JIT
1889
1890 The output is an integer that is set to one if support for just-in-time
1891 compiling is available; otherwise it is set to zero.
1892
1893 PCRE_CONFIG_JITTARGET
1894
1895 The output is a pointer to a zero-terminated "const char *" string. If
1896 JIT support is available, the string contains the name of the architec-
1897 ture for which the JIT compiler is configured, for example "x86 32bit
1898 (little endian + unaligned)". If JIT support is not available, the
1899 result is NULL.
1900
1901 PCRE_CONFIG_NEWLINE
1902
1903 The output is an integer whose value specifies the default character
1904 sequence that is recognized as meaning "newline". The values that are
1905 supported in ASCII/Unicode environments are: 10 for LF, 13 for CR, 3338
1906 for CRLF, -2 for ANYCRLF, and -1 for ANY. In EBCDIC environments, CR,
1907 ANYCRLF, and ANY yield the same values. However, the value for LF is
1908 normally 21, though some EBCDIC environments use 37. The corresponding
1909 values for CRLF are 3349 and 3365. The default should normally corre-
1910 spond to the standard sequence for your operating system.
1911
1912 PCRE_CONFIG_BSR
1913
1914 The output is an integer whose value indicates what character sequences
1915 the \R escape sequence matches by default. A value of 0 means that \R
1916 matches any Unicode line ending sequence; a value of 1 means that \R
1917 matches only CR, LF, or CRLF. The default can be overridden when a pat-
1918 tern is compiled or matched.
1919
1920 PCRE_CONFIG_LINK_SIZE
1921
1922 The output is an integer that contains the number of bytes used for
1923 internal linkage in compiled regular expressions. For the 8-bit
1924 library, the value can be 2, 3, or 4. For the 16-bit library, the value
1925 is either 2 or 4 and is still a number of bytes. For the 32-bit
1926 library, the value is either 2 or 4 and is still a number of bytes. The
1927 default value of 2 is sufficient for all but the most massive patterns,
1928 since it allows the compiled pattern to be up to 64K in size. Larger
1929 values allow larger regular expressions to be compiled, at the expense
1930 of slower matching.
1931
1932 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
1933
1934 The output is an integer that contains the threshold above which the
1935 POSIX interface uses malloc() for output vectors. Further details are
1936 given in the pcreposix documentation.
1937
1938 PCRE_CONFIG_MATCH_LIMIT
1939
1940 The output is a long integer that gives the default limit for the num-
1941 ber of internal matching function calls in a pcre_exec() execution.
1942 Further details are given with pcre_exec() below.
1943
1944 PCRE_CONFIG_MATCH_LIMIT_RECURSION
1945
1946 The output is a long integer that gives the default limit for the depth
1947 of recursion when calling the internal matching function in a
1948 pcre_exec() execution. Further details are given with pcre_exec()
1949 below.
1950
1951 PCRE_CONFIG_STACKRECURSE
1952
1953 The output is an integer that is set to one if internal recursion when
1954 running pcre_exec() is implemented by recursive function calls that use
1955 the stack to remember their state. This is the usual way that PCRE is
1956 compiled. The output is zero if PCRE was compiled to use blocks of data
1957 on the heap instead of recursive function calls. In this case,
1958 pcre_stack_malloc and pcre_stack_free are called to manage memory
1959 blocks on the heap, thus avoiding the use of the stack.
1960
1961
1962 COMPILING A PATTERN
1963
1964 pcre *pcre_compile(const char *pattern, int options,
1965 const char **errptr, int *erroffset,
1966 const unsigned char *tableptr);
1967
1968 pcre *pcre_compile2(const char *pattern, int options,
1969 int *errorcodeptr,
1970 const char **errptr, int *erroffset,
1971 const unsigned char *tableptr);
1972
1973 Either of the functions pcre_compile() or pcre_compile2() can be called
1974 to compile a pattern into an internal form. The only difference between
1975 the two interfaces is that pcre_compile2() has an additional argument,
1976 errorcodeptr, via which a numerical error code can be returned. To
1977 avoid too much repetition, we refer just to pcre_compile() below, but
1978 the information applies equally to pcre_compile2().
1979
1980 The pattern is a C string terminated by a binary zero, and is passed in
1981 the pattern argument. A pointer to a single block of memory that is
1982 obtained via pcre_malloc is returned. This contains the compiled code
1983 and related data. The pcre type is defined for the returned block; this
1984 is a typedef for a structure whose contents are not externally defined.
1985 It is up to the caller to free the memory (via pcre_free) when it is no
1986 longer required.
1987
1988 Although the compiled code of a PCRE regex is relocatable, that is, it
1989 does not depend on memory location, the complete pcre data block is not
1990 fully relocatable, because it may contain a copy of the tableptr argu-
1991 ment, which is an address (see below).
1992
1993 The options argument contains various bit settings that affect the com-
1994 pilation. It should be zero if no options are required. The available
1995 options are described below. Some of them (in particular, those that
1996 are compatible with Perl, but some others as well) can also be set and
1997 unset from within the pattern (see the detailed description in the
1998 pcrepattern documentation). For those options that can be different in
1999 different parts of the pattern, the contents of the options argument
2000 specifies their settings at the start of compilation and execution. The
2001 PCRE_ANCHORED, PCRE_BSR_xxx, PCRE_NEWLINE_xxx, PCRE_NO_UTF8_CHECK, and
2002 PCRE_NO_START_OPTIMIZE options can be set at the time of matching as
2003 well as at compile time.
2004
2005 If errptr is NULL, pcre_compile() returns NULL immediately. Otherwise,
2006 if compilation of a pattern fails, pcre_compile() returns NULL, and
2007 sets the variable pointed to by errptr to point to a textual error mes-
2008 sage. This is a static string that is part of the library. You must not
2009 try to free it. Normally, the offset from the start of the pattern to
2010 the byte that was being processed when the error was discovered is
2011 placed in the variable pointed to by erroffset, which must not be NULL
2012 (if it is, an immediate error is given). However, for an invalid UTF-8
2013 string, the offset is that of the first byte of the failing character.
2014
2015 Some errors are not detected until the whole pattern has been scanned;
2016 in these cases, the offset passed back is the length of the pattern.
2017 Note that the offset is in bytes, not characters, even in UTF-8 mode.
2018 It may sometimes point into the middle of a UTF-8 character.
2019
2020 If pcre_compile2() is used instead of pcre_compile(), and the error-
2021 codeptr argument is not NULL, a non-zero error code number is returned
2022 via this argument in the event of an error. This is in addition to the
2023 textual error message. Error codes and messages are listed below.
2024
2025 If the final argument, tableptr, is NULL, PCRE uses a default set of
2026 character tables that are built when PCRE is compiled, using the
2027 default C locale. Otherwise, tableptr must be an address that is the
2028 result of a call to pcre_maketables(). This value is stored with the
2029 compiled pattern, and used again by pcre_exec(), unless another table
2030 pointer is passed to it. For more discussion, see the section on locale
2031 support below.
2032
2033 This code fragment shows a typical straightforward call to pcre_com-
2034 pile():
2035
2036 pcre *re;
2037 const char *error;
2038 int erroffset;
2039 re = pcre_compile(
2040 "^A.*Z", /* the pattern */
2041 0, /* default options */
2042 &error, /* for error message */
2043 &erroffset, /* for error offset */
2044 NULL); /* use default character tables */
2045
2046 The following names for option bits are defined in the pcre.h header
2047 file:
2048
2049 PCRE_ANCHORED
2050
2051 If this bit is set, the pattern is forced to be "anchored", that is, it
2052 is constrained to match only at the first matching point in the string
2053 that is being searched (the "subject string"). This effect can also be
2054 achieved by appropriate constructs in the pattern itself, which is the
2055 only way to do it in Perl.
2056
2057 PCRE_AUTO_CALLOUT
2058
2059 If this bit is set, pcre_compile() automatically inserts callout items,
2060 all with number 255, before each pattern item. For discussion of the
2061 callout facility, see the pcrecallout documentation.
2062
2063 PCRE_BSR_ANYCRLF
2064 PCRE_BSR_UNICODE
2065
2066 These options (which are mutually exclusive) control what the \R escape
2067 sequence matches. The choice is either to match only CR, LF, or CRLF,
2068 or to match any Unicode newline sequence. The default is specified when
2069 PCRE is built. It can be overridden from within the pattern, or by set-
2070 ting an option when a compiled pattern is matched.
2071
2072 PCRE_CASELESS
2073
2074 If this bit is set, letters in the pattern match both upper and lower
2075 case letters. It is equivalent to Perl's /i option, and it can be
2076 changed within a pattern by a (?i) option setting. In UTF-8 mode, PCRE
2077 always understands the concept of case for characters whose values are
2078 less than 128, so caseless matching is always possible. For characters
2079 with higher values, the concept of case is supported if PCRE is com-
2080 piled with Unicode property support, but not otherwise. If you want to
2081 use caseless matching for characters 128 and above, you must ensure
2082 that PCRE is compiled with Unicode property support as well as with
2083 UTF-8 support.
2084
2085 PCRE_DOLLAR_ENDONLY
2086
2087 If this bit is set, a dollar metacharacter in the pattern matches only
2088 at the end of the subject string. Without this option, a dollar also
2089 matches immediately before a newline at the end of the string (but not
2090 before any other newlines). The PCRE_DOLLAR_ENDONLY option is ignored
2091 if PCRE_MULTILINE is set. There is no equivalent to this option in
2092 Perl, and no way to set it within a pattern.
2093
2094 PCRE_DOTALL
2095
2096 If this bit is set, a dot metacharacter in the pattern matches a char-
2097 acter of any value, including one that indicates a newline. However, it
2098 only ever matches one character, even if newlines are coded as CRLF.
2099 Without this option, a dot does not match when the current position is
2100 at a newline. This option is equivalent to Perl's /s option, and it can
2101 be changed within a pattern by a (?s) option setting. A negative class
2102 such as [^a] always matches newline characters, independent of the set-
2103 ting of this option.
2104
2105 PCRE_DUPNAMES
2106
2107 If this bit is set, names used to identify capturing subpatterns need
2108 not be unique. This can be helpful for certain types of pattern when it
2109 is known that only one instance of the named subpattern can ever be
2110 matched. There are more details of named subpatterns below; see also
2111 the pcrepattern documentation.
2112
2113 PCRE_EXTENDED
2114
2115 If this bit is set, white space data characters in the pattern are
2116 totally ignored except when escaped or inside a character class. White
2117 space does not include the VT character (code 11). In addition, charac-
2118 ters between an unescaped # outside a character class and the next new-
2119 line, inclusive, are also ignored. This is equivalent to Perl's /x
2120 option, and it can be changed within a pattern by a (?x) option set-
2121 ting.
2122
2123 Which characters are interpreted as newlines is controlled by the
2124 options passed to pcre_compile() or by a special sequence at the start
2125 of the pattern, as described in the section entitled "Newline conven-
2126 tions" in the pcrepattern documentation. Note that the end of this type
2127 of comment is a literal newline sequence in the pattern; escape
2128 sequences that happen to represent a newline do not count.
2129
2130 This option makes it possible to include comments inside complicated
2131 patterns. Note, however, that this applies only to data characters.
2132 White space characters may never appear within special character
2133 sequences in a pattern, for example within the sequence (?( that intro-
2134 duces a conditional subpattern.
2135
2136 PCRE_EXTRA
2137
2138 This option was invented in order to turn on additional functionality
2139 of PCRE that is incompatible with Perl, but it is currently of very
2140 little use. When set, any backslash in a pattern that is followed by a
2141 letter that has no special meaning causes an error, thus reserving
2142 these combinations for future expansion. By default, as in Perl, a
2143 backslash followed by a letter with no special meaning is treated as a
2144 literal. (Perl can, however, be persuaded to give an error for this, by
2145 running it with the -w option.) There are at present no other features
2146 controlled by this option. It can also be set by a (?X) option setting
2147 within a pattern.
2148
2149 PCRE_FIRSTLINE
2150
2151 If this option is set, an unanchored pattern is required to match
2152 before or at the first newline in the subject string, though the
2153 matched text may continue over the newline.
2154
2155 PCRE_JAVASCRIPT_COMPAT
2156
2157 If this option is set, PCRE's behaviour is changed in some ways so that
2158 it is compatible with JavaScript rather than Perl. The changes are as
2159 follows:
2160
2161 (1) A lone closing square bracket in a pattern causes a compile-time
2162 error, because this is illegal in JavaScript (by default it is treated
2163 as a data character). Thus, the pattern AB]CD becomes illegal when this
2164 option is set.
2165
2166 (2) At run time, a back reference to an unset subpattern group matches
2167 an empty string (by default this causes the current matching alterna-
2168 tive to fail). A pattern such as (\1)(a) succeeds when this option is
2169 set (assuming it can find an "a" in the subject), whereas it fails by
2170 default, for Perl compatibility.
2171
2172 (3) \U matches an upper case "U" character; by default \U causes a com-
2173 pile time error (Perl uses \U to upper case subsequent characters).
2174
2175 (4) \u matches a lower case "u" character unless it is followed by four
2176 hexadecimal digits, in which case the hexadecimal number defines the
2177 code point to match. By default, \u causes a compile time error (Perl
2178 uses it to upper case the following character).
2179
2180 (5) \x matches a lower case "x" character unless it is followed by two
2181 hexadecimal digits, in which case the hexadecimal number defines the
2182 code point to match. By default, as in Perl, a hexadecimal number is
2183 always expected after \x, but it may have zero, one, or two digits (so,
2184 for example, \xz matches a binary zero character followed by z).
2185
2186 PCRE_MULTILINE
2187
2188 By default, PCRE treats the subject string as consisting of a single
2189 line of characters (even if it actually contains newlines). The "start
2190 of line" metacharacter (^) matches only at the start of the string,
2191 while the "end of line" metacharacter ($) matches only at the end of
2192 the string, or before a terminating newline (unless PCRE_DOLLAR_ENDONLY
2193 is set). This is the same as Perl.
2194
2195 When PCRE_MULTILINE it is set, the "start of line" and "end of line"
2196 constructs match immediately following or immediately before internal
2197 newlines in the subject string, respectively, as well as at the very
2198 start and end. This is equivalent to Perl's /m option, and it can be
2199 changed within a pattern by a (?m) option setting. If there are no new-
2200 lines in a subject string, or no occurrences of ^ or $ in a pattern,
2201 setting PCRE_MULTILINE has no effect.
2202
2203 PCRE_NEWLINE_CR
2204 PCRE_NEWLINE_LF
2205 PCRE_NEWLINE_CRLF
2206 PCRE_NEWLINE_ANYCRLF
2207 PCRE_NEWLINE_ANY
2208
2209 These options override the default newline definition that was chosen
2210 when PCRE was built. Setting the first or the second specifies that a
2211 newline is indicated by a single character (CR or LF, respectively).
2212 Setting PCRE_NEWLINE_CRLF specifies that a newline is indicated by the
2213 two-character CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies
2214 that any of the three preceding sequences should be recognized. Setting
2215 PCRE_NEWLINE_ANY specifies that any Unicode newline sequence should be
2216 recognized.
2217
2218 In an ASCII/Unicode environment, the Unicode newline sequences are the
2219 three just mentioned, plus the single characters VT (vertical tab,
2220 U+000B), FF (form feed, U+000C), NEL (next line, U+0085), LS (line sep-
2221 arator, U+2028), and PS (paragraph separator, U+2029). For the 8-bit
2222 library, the last two are recognized only in UTF-8 mode.
2223
2224 When PCRE is compiled to run in an EBCDIC (mainframe) environment, the
2225 code for CR is 0x0d, the same as ASCII. However, the character code for
2226 LF is normally 0x15, though in some EBCDIC environments 0x25 is used.
2227 Whichever of these is not LF is made to correspond to Unicode's NEL
2228 character. EBCDIC codes are all less than 256. For more details, see
2229 the pcrebuild documentation.
2230
2231 The newline setting in the options word uses three bits that are
2232 treated as a number, giving eight possibilities. Currently only six are
2233 used (default plus the five values above). This means that if you set
2234 more than one newline option, the combination may or may not be sensi-
2235 ble. For example, PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to
2236 PCRE_NEWLINE_CRLF, but other combinations may yield unused numbers and
2237 cause an error.
2238
2239 The only time that a line break in a pattern is specially recognized
2240 when compiling is when PCRE_EXTENDED is set. CR and LF are white space
2241 characters, and so are ignored in this mode. Also, an unescaped # out-
2242 side a character class indicates a comment that lasts until after the
2243 next line break sequence. In other circumstances, line break sequences
2244 in patterns are treated as literal data.
2245
2246 The newline option that is set at compile time becomes the default that
2247 is used for pcre_exec() and pcre_dfa_exec(), but it can be overridden.
2248
2249 PCRE_NO_AUTO_CAPTURE
2250
2251 If this option is set, it disables the use of numbered capturing paren-
2252 theses in the pattern. Any opening parenthesis that is not followed by
2253 ? behaves as if it were followed by ?: but named parentheses can still
2254 be used for capturing (and they acquire numbers in the usual way).
2255 There is no equivalent of this option in Perl.
2256
2257 NO_START_OPTIMIZE
2258
2259 This is an option that acts at matching time; that is, it is really an
2260 option for pcre_exec() or pcre_dfa_exec(). If it is set at compile
2261 time, it is remembered with the compiled pattern and assumed at match-
2262 ing time. For details see the discussion of PCRE_NO_START_OPTIMIZE
2263 below.
2264
2265 PCRE_UCP
2266
2267 This option changes the way PCRE processes \B, \b, \D, \d, \S, \s, \W,
2268 \w, and some of the POSIX character classes. By default, only ASCII
2269 characters are recognized, but if PCRE_UCP is set, Unicode properties
2270 are used instead to classify characters. More details are given in the
2271 section on generic character types in the pcrepattern page. If you set
2272 PCRE_UCP, matching one of the items it affects takes much longer. The
2273 option is available only if PCRE has been compiled with Unicode prop-
2274 erty support.
2275
2276 PCRE_UNGREEDY
2277
2278 This option inverts the "greediness" of the quantifiers so that they
2279 are not greedy by default, but become greedy if followed by "?". It is
2280 not compatible with Perl. It can also be set by a (?U) option setting
2281 within the pattern.
2282
2283 PCRE_UTF8
2284
2285 This option causes PCRE to regard both the pattern and the subject as
2286 strings of UTF-8 characters instead of single-byte strings. However, it
2287 is available only when PCRE is built to include UTF support. If not,
2288 the use of this option provokes an error. Details of how this option
2289 changes the behaviour of PCRE are given in the pcreunicode page.
2290
2291 PCRE_NO_UTF8_CHECK
2292
2293 When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
2294 automatically checked. There is a discussion about the validity of
2295 UTF-8 strings in the pcreunicode page. If an invalid UTF-8 sequence is
2296 found, pcre_compile() returns an error. If you already know that your
2297 pattern is valid, and you want to skip this check for performance rea-
2298 sons, you can set the PCRE_NO_UTF8_CHECK option. When it is set, the
2299 effect of passing an invalid UTF-8 string as a pattern is undefined. It
2300 may cause your program to crash. Note that this option can also be
2301 passed to pcre_exec() and pcre_dfa_exec(), to suppress the validity
2302 checking of subject strings only. If the same string is being matched
2303 many times, the option can be safely set for the second and subsequent
2304 matchings to improve performance.
2305
2306
2307 COMPILATION ERROR CODES
2308
2309 The following table lists the error codes than may be returned by
2310 pcre_compile2(), along with the error messages that may be returned by
2311 both compiling functions. Note that error messages are always 8-bit
2312 ASCII strings, even in 16-bit or 32-bit mode. As PCRE has developed,
2313 some error codes have fallen out of use. To avoid confusion, they have
2314 not been re-used.
2315
2316 0 no error
2317 1 \ at end of pattern
2318 2 \c at end of pattern
2319 3 unrecognized character follows \
2320 4 numbers out of order in {} quantifier
2321 5 number too big in {} quantifier
2322 6 missing terminating ] for character class
2323 7 invalid escape sequence in character class
2324 8 range out of order in character class
2325 9 nothing to repeat
2326 10 [this code is not in use]
2327 11 internal error: unexpected repeat
2328 12 unrecognized character after (? or (?-
2329 13 POSIX named classes are supported only within a class
2330 14 missing )
2331 15 reference to non-existent subpattern
2332 16 erroffset passed as NULL
2333 17 unknown option bit(s) set
2334 18 missing ) after comment
2335 19 [this code is not in use]
2336 20 regular expression is too large
2337 21 failed to get memory
2338 22 unmatched parentheses
2339 23 internal error: code overflow
2340 24 unrecognized character after (?<
2341 25 lookbehind assertion is not fixed length
2342 26 malformed number or name after (?(
2343 27 conditional group contains more than two branches
2344 28 assertion expected after (?(
2345 29 (?R or (?[+-]digits must be followed by )
2346 30 unknown POSIX class name
2347 31 POSIX collating elements are not supported
2348 32 this version of PCRE is compiled without UTF support
2349 33 [this code is not in use]
2350 34 character value in \x{...} sequence is too large
2351 35 invalid condition (?(0)
2352 36 \C not allowed in lookbehind assertion
2353 37 PCRE does not support \L, \l, \N{name}, \U, or \u
2354 38 number after (?C is > 255
2355 39 closing ) for (?C expected
2356 40 recursive call could loop indefinitely
2357 41 unrecognized character after (?P
2358 42 syntax error in subpattern name (missing terminator)
2359 43 two named subpatterns have the same name
2360 44 invalid UTF-8 string (specifically UTF-8)
2361 45 support for \P, \p, and \X has not been compiled
2362 46 malformed \P or \p sequence
2363 47 unknown property name after \P or \p
2364 48 subpattern name is too long (maximum 32 characters)
2365 49 too many named subpatterns (maximum 10000)
2366 50 [this code is not in use]
2367 51 octal value is greater than \377 in 8-bit non-UTF-8 mode
2368 52 internal error: overran compiling workspace
2369 53 internal error: previously-checked referenced subpattern
2370 not found
2371 54 DEFINE group contains more than one branch
2372 55 repeating a DEFINE group is not allowed
2373 56 inconsistent NEWLINE options
2374 57 \g is not followed by a braced, angle-bracketed, or quoted
2375 name/number or by a plain number
2376 58 a numbered reference must not be zero
2377 59 an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
2378 60 (*VERB) not recognized
2379 61 number is too big
2380 62 subpattern name expected
2381 63 digit expected after (?+
2382 64 ] is an invalid data character in JavaScript compatibility mode
2383 65 different names for subpatterns of the same number are
2384 not allowed
2385 66 (*MARK) must have an argument
2386 67 this version of PCRE is not compiled with Unicode property
2387 support
2388 68 \c must be followed by an ASCII character
2389 69 \k is not followed by a braced, angle-bracketed, or quoted name
2390 70 internal error: unknown opcode in find_fixedlength()
2391 71 \N is not supported in a class
2392 72 too many forward references
2393 73 disallowed Unicode code point (>= 0xd800 && <= 0xdfff)
2394 74 invalid UTF-16 string (specifically UTF-16)
2395 75 name is too long in (*MARK), (*PRUNE), (*SKIP), or (*THEN)
2396 76 character value in \u.... sequence is too large
2397 77 invalid UTF-32 string (specifically UTF-32)
2398
2399 The numbers 32 and 10000 in errors 48 and 49 are defaults; different
2400 values may be used if the limits were changed when PCRE was built.
2401
2402
2403 STUDYING A PATTERN
2404
2405 pcre_extra *pcre_study(const pcre *code, int options
2406 const char **errptr);
2407
2408 If a compiled pattern is going to be used several times, it is worth
2409 spending more time analyzing it in order to speed up the time taken for
2410 matching. The function pcre_study() takes a pointer to a compiled pat-
2411 tern as its first argument. If studying the pattern produces additional
2412 information that will help speed up matching, pcre_study() returns a
2413 pointer to a pcre_extra block, in which the study_data field points to
2414 the results of the study.
2415
2416 The returned value from pcre_study() can be passed directly to
2417 pcre_exec() or pcre_dfa_exec(). However, a pcre_extra block also con-
2418 tains other fields that can be set by the caller before the block is
2419 passed; these are described below in the section on matching a pattern.
2420
2421 If studying the pattern does not produce any useful information,
2422 pcre_study() returns NULL by default. In that circumstance, if the
2423 calling program wants to pass any of the other fields to pcre_exec() or
2424 pcre_dfa_exec(), it must set up its own pcre_extra block. However, if
2425 pcre_study() is called with the PCRE_STUDY_EXTRA_NEEDED option, it
2426 returns a pcre_extra block even if studying did not find any additional
2427 information. It may still return NULL, however, if an error occurs in
2428 pcre_study().
2429
2430 The second argument of pcre_study() contains option bits. There are
2431 three further options in addition to PCRE_STUDY_EXTRA_NEEDED:
2432
2433 PCRE_STUDY_JIT_COMPILE
2434 PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE
2435 PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE
2436
2437 If any of these are set, and the just-in-time compiler is available,
2438 the pattern is further compiled into machine code that executes much
2439 faster than the pcre_exec() interpretive matching function. If the
2440 just-in-time compiler is not available, these options are ignored. All
2441 undefined bits in the options argument must be zero.
2442
2443 JIT compilation is a heavyweight optimization. It can take some time
2444 for patterns to be analyzed, and for one-off matches and simple pat-
2445 terns the benefit of faster execution might be offset by a much slower
2446 study time. Not all patterns can be optimized by the JIT compiler. For
2447 those that cannot be handled, matching automatically falls back to the
2448 pcre_exec() interpreter. For more details, see the pcrejit documenta-
2449 tion.
2450
2451 The third argument for pcre_study() is a pointer for an error message.
2452 If studying succeeds (even if no data is returned), the variable it
2453 points to is set to NULL. Otherwise it is set to point to a textual
2454 error message. This is a static string that is part of the library. You
2455 must not try to free it. You should test the error pointer for NULL
2456 after calling pcre_study(), to be sure that it has run successfully.
2457
2458 When you are finished with a pattern, you can free the memory used for
2459 the study data by calling pcre_free_study(). This function was added to
2460 the API for release 8.20. For earlier versions, the memory could be
2461 freed with pcre_free(), just like the pattern itself. This will still
2462 work in cases where JIT optimization is not used, but it is advisable
2463 to change to the new function when convenient.
2464
2465 This is a typical way in which pcre_study() is used (except that in a
2466 real application there should be tests for errors):
2467
2468 int rc;
2469 pcre *re;
2470 pcre_extra *sd;
2471 re = pcre_compile("pattern", 0, &error, &erroroffset, NULL);
2472 sd = pcre_study(
2473 re, /* result of pcre_compile() */
2474 0, /* no options */
2475 &error); /* set to NULL or points to a message */
2476 rc = pcre_exec( /* see below for details of pcre_exec() options */
2477 re, sd, "subject", 7, 0, 0, ovector, 30);
2478 ...
2479 pcre_free_study(sd);
2480 pcre_free(re);
2481
2482 Studying a pattern does two things: first, a lower bound for the length
2483 of subject string that is needed to match the pattern is computed. This
2484 does not mean that there are any strings of that length that match, but
2485 it does guarantee that no shorter strings match. The value is used to
2486 avoid wasting time by trying to match strings that are shorter than the
2487 lower bound. You can find out the value in a calling program via the
2488 pcre_fullinfo() function.
2489
2490 Studying a pattern is also useful for non-anchored patterns that do not
2491 have a single fixed starting character. A bitmap of possible starting
2492 bytes is created. This speeds up finding a position in the subject at
2493 which to start matching. (In 16-bit mode, the bitmap is used for 16-bit
2494 values less than 256. In 32-bit mode, the bitmap is used for 32-bit
2495 values less than 256.)
2496
2497 These two optimizations apply to both pcre_exec() and pcre_dfa_exec(),
2498 and the information is also used by the JIT compiler. The optimiza-
2499 tions can be disabled by setting the PCRE_NO_START_OPTIMIZE option when
2500 calling pcre_exec() or pcre_dfa_exec(), but if this is done, JIT execu-
2501 tion is also disabled. You might want to do this if your pattern con-
2502 tains callouts or (*MARK) and you want to make use of these facilities
2503 in cases where matching fails. See the discussion of
2504 PCRE_NO_START_OPTIMIZE below.
2505
2506
2507 LOCALE SUPPORT
2508
2509 PCRE handles caseless matching, and determines whether characters are
2510 letters, digits, or whatever, by reference to a set of tables, indexed
2511 by character value. When running in UTF-8 mode, this applies only to
2512 characters with codes less than 128. By default, higher-valued codes
2513 never match escapes such as \w or \d, but they can be tested with \p if
2514 PCRE is built with Unicode character property support. Alternatively,
2515 the PCRE_UCP option can be set at compile time; this causes \w and
2516 friends to use Unicode property support instead of built-in tables. The
2517 use of locales with Unicode is discouraged. If you are handling charac-
2518 ters with codes greater than 128, you should either use UTF-8 and Uni-
2519 code, or use locales, but not try to mix the two.
2520
2521 PCRE contains an internal set of tables that are used when the final
2522 argument of pcre_compile() is NULL. These are sufficient for many
2523 applications. Normally, the internal tables recognize only ASCII char-
2524 acters. However, when PCRE is built, it is possible to cause the inter-
2525 nal tables to be rebuilt in the default "C" locale of the local system,
2526 which may cause them to be different.
2527
2528 The internal tables can always be overridden by tables supplied by the
2529 application that calls PCRE. These may be created in a different locale
2530 from the default. As more and more applications change to using Uni-
2531 code, the need for this locale support is expected to die away.
2532
2533 External tables are built by calling the pcre_maketables() function,
2534 which has no arguments, in the relevant locale. The result can then be
2535 passed to pcre_compile() or pcre_exec() as often as necessary. For
2536 example, to build and use tables that are appropriate for the French
2537 locale (where accented characters with values greater than 128 are
2538 treated as letters), the following code could be used:
2539
2540 setlocale(LC_CTYPE, "fr_FR");
2541 tables = pcre_maketables();
2542 re = pcre_compile(..., tables);
2543
2544 The locale name "fr_FR" is used on Linux and other Unix-like systems;
2545 if you are using Windows, the name for the French locale is "french".
2546
2547 When pcre_maketables() runs, the tables are built in memory that is
2548 obtained via pcre_malloc. It is the caller's responsibility to ensure
2549 that the memory containing the tables remains available for as long as
2550 it is needed.
2551
2552 The pointer that is passed to pcre_compile() is saved with the compiled
2553 pattern, and the same tables are used via this pointer by pcre_study()
2554 and normally also by pcre_exec(). Thus, by default, for any single pat-
2555 tern, compilation, studying and matching all happen in the same locale,
2556 but different patterns can be compiled in different locales.
2557
2558 It is possible to pass a table pointer or NULL (indicating the use of
2559 the internal tables) to pcre_exec(). Although not intended for this
2560 purpose, this facility could be used to match a pattern in a different
2561 locale from the one in which it was compiled. Passing table pointers at
2562 run time is discussed below in the section on matching a pattern.
2563
2564
2565 INFORMATION ABOUT A PATTERN
2566
2567 int pcre_fullinfo(const pcre *code, const pcre_extra *extra,
2568 int what, void *where);
2569
2570 The pcre_fullinfo() function returns information about a compiled pat-
2571 tern. It replaces the pcre_info() function, which was removed from the
2572 library at version 8.30, after more than 10 years of obsolescence.
2573
2574 The first argument for pcre_fullinfo() is a pointer to the compiled
2575 pattern. The second argument is the result of pcre_study(), or NULL if
2576 the pattern was not studied. The third argument specifies which piece
2577 of information is required, and the fourth argument is a pointer to a
2578 variable to receive the data. The yield of the function is zero for
2579 success, or one of the following negative numbers:
2580
2581 PCRE_ERROR_NULL the argument code was NULL
2582 the argument where was NULL
2583 PCRE_ERROR_BADMAGIC the "magic number" was not found
2584 PCRE_ERROR_BADENDIANNESS the pattern was compiled with different
2585 endianness
2586 PCRE_ERROR_BADOPTION the value of what was invalid
2587
2588 The "magic number" is placed at the start of each compiled pattern as
2589 an simple check against passing an arbitrary memory pointer. The endi-
2590 anness error can occur if a compiled pattern is saved and reloaded on a
2591 different host. Here is a typical call of pcre_fullinfo(), to obtain
2592 the length of the compiled pattern:
2593
2594 int rc;
2595 size_t length;
2596 rc = pcre_fullinfo(
2597 re, /* result of pcre_compile() */
2598 sd, /* result of pcre_study(), or NULL */
2599 PCRE_INFO_SIZE, /* what is required */
2600 &length); /* where to put the data */
2601
2602 The possible values for the third argument are defined in pcre.h, and
2603 are as follows:
2604
2605 PCRE_INFO_BACKREFMAX
2606
2607 Return the number of the highest back reference in the pattern. The
2608 fourth argument should point to an int variable. Zero is returned if
2609 there are no back references.
2610
2611 PCRE_INFO_CAPTURECOUNT
2612
2613 Return the number of capturing subpatterns in the pattern. The fourth
2614 argument should point to an int variable.
2615
2616 PCRE_INFO_DEFAULT_TABLES
2617
2618 Return a pointer to the internal default character tables within PCRE.
2619 The fourth argument should point to an unsigned char * variable. This
2620 information call is provided for internal use by the pcre_study() func-
2621 tion. External callers can cause PCRE to use its internal tables by
2622 passing a NULL table pointer.
2623
2624 PCRE_INFO_FIRSTBYTE
2625
2626 Return information about the first data unit of any matched string, for
2627 a non-anchored pattern. (The name of this option refers to the 8-bit
2628 library, where data units are bytes.) The fourth argument should point
2629 to an int variable.
2630
2631 If there is a fixed first value, for example, the letter "c" from a
2632 pattern such as (cat|cow|coyote), its value is returned. In the 8-bit
2633 library, the value is always less than 256. In the 16-bit library the
2634 value can be up to 0xffff. In the 32-bit library the value can be up to
2635 0x10ffff.
2636
2637 If there is no fixed first value, and if either
2638
2639 (a) the pattern was compiled with the PCRE_MULTILINE option, and every
2640 branch starts with "^", or
2641
2642 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not
2643 set (if it were set, the pattern would be anchored),
2644
2645 -1 is returned, indicating that the pattern matches only at the start
2646 of a subject string or after any newline within the string. Otherwise
2647 -2 is returned. For anchored patterns, -2 is returned.
2648
2649 Since for the 32-bit library using the non-UTF-32 mode, this function
2650 is unable to return the full 32-bit range of the character, this value
2651 is deprecated; instead the PCRE_INFO_FIRSTCHARACTERFLAGS and
2652 PCRE_INFO_FIRSTCHARACTER values should be used.
2653
2654 PCRE_INFO_FIRSTTABLE
2655
2656 If the pattern was studied, and this resulted in the construction of a
2657 256-bit table indicating a fixed set of values for the first data unit
2658 in any matching string, a pointer to the table is returned. Otherwise
2659 NULL is returned. The fourth argument should point to an unsigned char
2660 * variable.
2661
2662 PCRE_INFO_HASCRORLF
2663
2664 Return 1 if the pattern contains any explicit matches for CR or LF
2665 characters, otherwise 0. The fourth argument should point to an int
2666 variable. An explicit match is either a literal CR or LF character, or
2667 \r or \n.
2668
2669 PCRE_INFO_JCHANGED
2670
2671 Return 1 if the (?J) or (?-J) option setting is used in the pattern,
2672 otherwise 0. The fourth argument should point to an int variable. (?J)
2673 and (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
2674
2675 PCRE_INFO_JIT
2676
2677 Return 1 if the pattern was studied with one of the JIT options, and
2678 just-in-time compiling was successful. The fourth argument should point
2679 to an int variable. A return value of 0 means that JIT support is not
2680 available in this version of PCRE, or that the pattern was not studied
2681 with a JIT option, or that the JIT compiler could not handle this par-
2682 ticular pattern. See the pcrejit documentation for details of what can
2683 and cannot be handled.
2684
2685 PCRE_INFO_JITSIZE
2686
2687 If the pattern was successfully studied with a JIT option, return the
2688 size of the JIT compiled code, otherwise return zero. The fourth argu-
2689 ment should point to a size_t variable.
2690
2691 PCRE_INFO_LASTLITERAL
2692
2693 Return the value of the rightmost literal data unit that must exist in
2694 any matched string, other than at its start, if such a value has been
2695 recorded. The fourth argument should point to an int variable. If there
2696 is no such value, -1 is returned. For anchored patterns, a last literal
2697 value is recorded only if it follows something of variable length. For
2698 example, for the pattern /^a\d+z\d+/ the returned value is "z", but for
2699 /^a\dz\d/ the returned value is -1.
2700
2701 Since for the 32-bit library using the non-UTF-32 mode, this function
2702 is unable to return the full 32-bit range of the character, this value
2703 is deprecated; instead the PCRE_INFO_REQUIREDCHARFLAGS and
2704 PCRE_INFO_REQUIREDCHAR values should be used.
2705
2706 PCRE_INFO_MAXLOOKBEHIND
2707
2708 Return the number of characters (NB not bytes) in the longest lookbe-
2709 hind assertion in the pattern. Note that the simple assertions \b and
2710 \B require a one-character lookbehind. This information is useful when
2711 doing multi-segment matching using the partial matching facilities.
2712
2713 PCRE_INFO_MINLENGTH
2714
2715 If the pattern was studied and a minimum length for matching subject
2716 strings was computed, its value is returned. Otherwise the returned
2717 value is -1. The value is a number of characters, which in UTF-8 mode
2718 may be different from the number of bytes. The fourth argument should
2719 point to an int variable. A non-negative value is a lower bound to the
2720 length of any matching string. There may not be any strings of that
2721 length that do actually match, but every string that does match is at
2722 least that long.
2723
2724 PCRE_INFO_NAMECOUNT
2725 PCRE_INFO_NAMEENTRYSIZE
2726 PCRE_INFO_NAMETABLE
2727
2728 PCRE supports the use of named as well as numbered capturing parenthe-
2729 ses. The names are just an additional way of identifying the parenthe-
2730 ses, which still acquire numbers. Several convenience functions such as
2731 pcre_get_named_substring() are provided for extracting captured sub-
2732 strings by name. It is also possible to extract the data directly, by
2733 first converting the name to a number in order to access the correct
2734 pointers in the output vector (described with pcre_exec() below). To do
2735 the conversion, you need to use the name-to-number map, which is
2736 described by these three values.
2737
2738 The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT
2739 gives the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size
2740 of each entry; both of these return an int value. The entry size
2741 depends on the length of the longest name. PCRE_INFO_NAMETABLE returns
2742 a pointer to the first entry of the table. This is a pointer to char in
2743 the 8-bit library, where the first two bytes of each entry are the num-
2744 ber of the capturing parenthesis, most significant byte first. In the
2745 16-bit library, the pointer points to 16-bit data units, the first of
2746 which contains the parenthesis number. In the 32-bit library, the
2747 pointer points to 32-bit data units, the first of which contains the
2748 parenthesis number. The rest of the entry is the corresponding name,
2749 zero terminated.
2750
2751 The names are in alphabetical order. Duplicate names may appear if (?|
2752 is used to create multiple groups with the same number, as described in
2753 the section on duplicate subpattern numbers in the pcrepattern page.
2754 Duplicate names for subpatterns with different numbers are permitted
2755 only if PCRE_DUPNAMES is set. In all cases of duplicate names, they
2756 appear in the table in the order in which they were found in the pat-
2757 tern. In the absence of (?| this is the order of increasing number;
2758 when (?| is used this is not necessarily the case because later subpat-
2759 terns may have lower numbers.
2760
2761 As a simple example of the name/number table, consider the following
2762 pattern after compilation by the 8-bit library (assume PCRE_EXTENDED is
2763 set, so white space - including newlines - is ignored):
2764
2765 (?<date> (?<year>(\d\d)?\d\d) -
2766 (?<month>\d\d) - (?<day>\d\d) )
2767
2768 There are four named subpatterns, so the table has four entries, and
2769 each entry in the table is eight bytes long. The table is as follows,
2770 with non-printing bytes shows in hexadecimal, and undefined bytes shown
2771 as ??:
2772
2773 00 01 d a t e 00 ??
2774 00 05 d a y 00 ?? ??
2775 00 04 m o n t h 00
2776 00 02 y e a r 00 ??
2777
2778 When writing code to extract data from named subpatterns using the
2779 name-to-number map, remember that the length of the entries is likely
2780 to be different for each compiled pattern.
2781
2782 PCRE_INFO_OKPARTIAL
2783
2784 Return 1 if the pattern can be used for partial matching with
2785 pcre_exec(), otherwise 0. The fourth argument should point to an int
2786 variable. From release 8.00, this always returns 1, because the
2787 restrictions that previously applied to partial matching have been
2788 lifted. The pcrepartial documentation gives details of partial match-
2789 ing.
2790
2791 PCRE_INFO_OPTIONS
2792
2793 Return a copy of the options with which the pattern was compiled. The
2794 fourth argument should point to an unsigned long int variable. These
2795 option bits are those specified in the call to pcre_compile(), modified
2796 by any top-level option settings at the start of the pattern itself. In
2797 other words, they are the options that will be in force when matching
2798 starts. For example, if the pattern /(?im)abc(?-i)d/ is compiled with
2799 the PCRE_EXTENDED option, the result is PCRE_CASELESS, PCRE_MULTILINE,
2800 and PCRE_EXTENDED.
2801
2802 A pattern is automatically anchored by PCRE if all of its top-level
2803 alternatives begin with one of the following:
2804
2805 ^ unless PCRE_MULTILINE is set
2806 \A always
2807 \G always
2808 .* if PCRE_DOTALL is set and there are no back
2809 references to the subpattern in which .* appears
2810
2811 For such patterns, the PCRE_ANCHORED bit is set in the options returned
2812 by pcre_fullinfo().
2813
2814 PCRE_INFO_SIZE
2815
2816 Return the size of the compiled pattern in bytes (for both libraries).
2817 The fourth argument should point to a size_t variable. This value does
2818 not include the size of the pcre structure that is returned by
2819 pcre_compile(). The value that is passed as the argument to pcre_mal-
2820 loc() when pcre_compile() is getting memory in which to place the com-
2821 piled data is the value returned by this option plus the size of the
2822 pcre structure. Studying a compiled pattern, with or without JIT, does
2823 not alter the value returned by this option.
2824
2825 PCRE_INFO_STUDYSIZE
2826
2827 Return the size in bytes of the data block pointed to by the study_data
2828 field in a pcre_extra block. If pcre_extra is NULL, or there is no
2829 study data, zero is returned. The fourth argument should point to a
2830 size_t variable. The study_data field is set by pcre_study() to record
2831 information that will speed up matching (see the section entitled
2832 "Studying a pattern" above). The format of the study_data block is pri-
2833 vate, but its length is made available via this option so that it can
2834 be saved and restored (see the pcreprecompile documentation for
2835 details).
2836
2837 PCRE_INFO_FIRSTCHARACTERFLAGS
2838
2839 Return information about the first data unit of any matched string, for
2840 a non-anchored pattern. The fourth argument should point to an int
2841 variable.
2842
2843 If there is a fixed first value, for example, the letter "c" from a
2844 pattern such as (cat|cow|coyote), 1 is returned, and the character
2845 value can be retrieved using PCRE_INFO_FIRSTCHARACTER.
2846
2847 If there is no fixed first value, and if either
2848
2849 (a) the pattern was compiled with the PCRE_MULTILINE option, and every
2850 branch starts with "^", or
2851
2852 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not
2853 set (if it were set, the pattern would be anchored),
2854
2855 2 is returned, indicating that the pattern matches only at the start of
2856 a subject string or after any newline within the string. Otherwise 0 is
2857 returned. For anchored patterns, 0 is returned.
2858
2859 PCRE_INFO_FIRSTCHARACTER
2860
2861 Return the fixed first character value, if PCRE_INFO_FIRSTCHARACTER-
2862 FLAGS returned 1; otherwise returns 0. The fourth argument should point
2863 to an uint_t variable.
2864
2865 In the 8-bit library, the value is always less than 256. In the 16-bit
2866 library the value can be up to 0xffff. In the 32-bit library in UTF-32
2867 mode the value can be up to 0x10ffff, and up to 0xffffffff when not
2868 using UTF-32 mode.
2869
2870 If there is no fixed first value, and if either
2871
2872 (a) the pattern was compiled with the PCRE_MULTILINE option, and every
2873 branch starts with "^", or
2874
2875 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not
2876 set (if it were set, the pattern would be anchored),
2877
2878 -1 is returned, indicating that the pattern matches only at the start
2879 of a subject string or after any newline within the string. Otherwise
2880 -2 is returned. For anchored patterns, -2 is returned.
2881
2882 PCRE_INFO_REQUIREDCHARFLAGS
2883
2884 Returns 1 if there is a rightmost literal data unit that must exist in
2885 any matched string, other than at its start. The fourth argument should
2886 point to an int variable. If there is no such value, 0 is returned. If
2887 returning 1, the character value itself can be retrieved using
2888 PCRE_INFO_REQUIREDCHAR.
2889
2890 For anchored patterns, a last literal value is recorded only if it fol-
2891 lows something of variable length. For example, for the pattern
2892 /^a\d+z\d+/ the returned value 1 (with "z" returned from
2893 PCRE_INFO_REQUIREDCHAR), but for /^a\dz\d/ the returned value is 0.
2894
2895 PCRE_INFO_REQUIREDCHAR
2896
2897 Return the value of the rightmost literal data unit that must exist in
2898 any matched string, other than at its start, if such a value has been
2899 recorded. The fourth argument should point to an uint32_t variable. If
2900 there is no such value, 0 is returned.
2901
2902
2903 REFERENCE COUNTS
2904
2905 int pcre_refcount(pcre *code, int adjust);
2906
2907 The pcre_refcount() function is used to maintain a reference count in
2908 the data block that contains a compiled pattern. It is provided for the
2909 benefit of applications that operate in an object-oriented manner,
2910 where different parts of the application may be using the same compiled
2911 pattern, but you want to free the block when they are all done.
2912
2913 When a pattern is compiled, the reference count field is initialized to
2914 zero. It is changed only by calling this function, whose action is to
2915 add the adjust value (which may be positive or negative) to it. The
2916 yield of the function is the new value. However, the value of the count
2917 is constrained to lie between 0 and 65535, inclusive. If the new value
2918 is outside these limits, it is forced to the appropriate limit value.
2919
2920 Except when it is zero, the reference count is not correctly preserved
2921 if a pattern is compiled on one host and then transferred to a host
2922 whose byte-order is different. (This seems a highly unlikely scenario.)
2923
2924
2925 MATCHING A PATTERN: THE TRADITIONAL FUNCTION
2926
2927 int pcre_exec(const pcre *code, const pcre_extra *extra,
2928 const char *subject, int length, int startoffset,
2929 int options, int *ovector, int ovecsize);
2930
2931 The function pcre_exec() is called to match a subject string against a
2932 compiled pattern, which is passed in the code argument. If the pattern
2933 was studied, the result of the study should be passed in the extra
2934 argument. You can call pcre_exec() with the same code and extra argu-
2935 ments as many times as you like, in order to match different subject
2936 strings with the same pattern.
2937
2938 This function is the main matching facility of the library, and it
2939 operates in a Perl-like manner. For specialist use there is also an
2940 alternative matching function, which is described below in the section
2941 about the pcre_dfa_exec() function.
2942
2943 In most applications, the pattern will have been compiled (and option-
2944 ally studied) in the same process that calls pcre_exec(). However, it
2945 is possible to save compiled patterns and study data, and then use them
2946 later in different processes, possibly even on different hosts. For a
2947 discussion about this, see the pcreprecompile documentation.
2948
2949 Here is an example of a simple call to pcre_exec():
2950
2951 int rc;
2952 int ovector[30];
2953 rc = pcre_exec(
2954 re, /* result of pcre_compile() */
2955 NULL, /* we didn't study the pattern */
2956 "some string", /* the subject string */
2957 11, /* the length of the subject string */
2958 0, /* start at offset 0 in the subject */
2959 0, /* default options */
2960 ovector, /* vector of integers for substring information */
2961 30); /* number of elements (NOT size in bytes) */
2962
2963 Extra data for pcre_exec()
2964
2965 If the extra argument is not NULL, it must point to a pcre_extra data
2966 block. The pcre_study() function returns such a block (when it doesn't
2967 return NULL), but you can also create one for yourself, and pass addi-
2968 tional information in it. The pcre_extra block contains the following
2969 fields (not necessarily in this order):
2970
2971 unsigned long int flags;
2972 void *study_data;
2973 void *executable_jit;
2974 unsigned long int match_limit;
2975 unsigned long int match_limit_recursion;
2976 void *callout_data;
2977 const unsigned char *tables;
2978 unsigned char **mark;
2979
2980 In the 16-bit version of this structure, the mark field has type
2981 "PCRE_UCHAR16 **".
2982
2983 In the 32-bit version of this structure, the mark field has type
2984 "PCRE_UCHAR32 **".
2985
2986 The flags field is used to specify which of the other fields are set.
2987 The flag bits are:
2988
2989 PCRE_EXTRA_CALLOUT_DATA
2990 PCRE_EXTRA_EXECUTABLE_JIT
2991 PCRE_EXTRA_MARK
2992 PCRE_EXTRA_MATCH_LIMIT
2993 PCRE_EXTRA_MATCH_LIMIT_RECURSION
2994 PCRE_EXTRA_STUDY_DATA
2995 PCRE_EXTRA_TABLES
2996
2997 Other flag bits should be set to zero. The study_data field and some-
2998 times the executable_jit field are set in the pcre_extra block that is
2999 returned by pcre_study(), together with the appropriate flag bits. You
3000 should not set these yourself, but you may add to the block by setting
3001 other fields and their corresponding flag bits.
3002
3003 The match_limit field provides a means of preventing PCRE from using up
3004 a vast amount of resources when running patterns that are not going to
3005 match, but which have a very large number of possibilities in their
3006 search trees. The classic example is a pattern that uses nested unlim-
3007 ited repeats.
3008
3009 Internally, pcre_exec() uses a function called match(), which it calls
3010 repeatedly (sometimes recursively). The limit set by match_limit is
3011 imposed on the number of times this function is called during a match,
3012 which has the effect of limiting the amount of backtracking that can
3013 take place. For patterns that are not anchored, the count restarts from
3014 zero for each position in the subject string.
3015
3016 When pcre_exec() is called with a pattern that was successfully studied
3017 with a JIT option, the way that the matching is executed is entirely
3018 different. However, there is still the possibility of runaway matching
3019 that goes on for a very long time, and so the match_limit value is also
3020 used in this case (but in a different way) to limit how long the match-
3021 ing can continue.
3022
3023 The default value for the limit can be set when PCRE is built; the
3024 default default is 10 million, which handles all but the most extreme
3025 cases. You can override the default by suppling pcre_exec() with a
3026 pcre_extra block in which match_limit is set, and
3027 PCRE_EXTRA_MATCH_LIMIT is set in the flags field. If the limit is
3028 exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3029
3030 The match_limit_recursion field is similar to match_limit, but instead
3031 of limiting the total number of times that match() is called, it limits
3032 the depth of recursion. The recursion depth is a smaller number than
3033 the total number of calls, because not all calls to match() are recur-
3034 sive. This limit is of use only if it is set smaller than match_limit.
3035
3036 Limiting the recursion depth limits the amount of machine stack that
3037 can be used, or, when PCRE has been compiled to use memory on the heap
3038 instead of the stack, the amount of heap memory that can be used. This
3039 limit is not relevant, and is ignored, when matching is done using JIT
3040 compiled code.
3041
3042 The default value for match_limit_recursion can be set when PCRE is
3043 built; the default default is the same value as the default for
3044 match_limit. You can override the default by suppling pcre_exec() with
3045 a pcre_extra block in which match_limit_recursion is set, and
3046 PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the flags field. If the
3047 limit is exceeded, pcre_exec() returns PCRE_ERROR_RECURSIONLIMIT.
3048
3049 The callout_data field is used in conjunction with the "callout" fea-
3050 ture, and is described in the pcrecallout documentation.
3051
3052 The tables field is used to pass a character tables pointer to
3053 pcre_exec(); this overrides the value that is stored with the compiled
3054 pattern. A non-NULL value is stored with the compiled pattern only if
3055 custom tables were supplied to pcre_compile() via its tableptr argu-
3056 ment. If NULL is passed to pcre_exec() using this mechanism, it forces
3057 PCRE's internal tables to be used. This facility is helpful when re-
3058 using patterns that have been saved after compiling with an external
3059 set of tables, because the external tables might be at a different
3060 address when pcre_exec() is called. See the pcreprecompile documenta-
3061 tion for a discussion of saving compiled patterns for later use.
3062
3063 If PCRE_EXTRA_MARK is set in the flags field, the mark field must be
3064 set to point to a suitable variable. If the pattern contains any back-
3065 tracking control verbs such as (*MARK:NAME), and the execution ends up
3066 with a name to pass back, a pointer to the name string (zero termi-
3067 nated) is placed in the variable pointed to by the mark field. The
3068 names are within the compiled pattern; if you wish to retain such a
3069 name you must copy it before freeing the memory of a compiled pattern.
3070 If there is no name to pass back, the variable pointed to by the mark
3071 field is set to NULL. For details of the backtracking control verbs,
3072 see the section entitled "Backtracking control" in the pcrepattern doc-
3073 umentation.
3074
3075 Option bits for pcre_exec()
3076
3077 The unused bits of the options argument for pcre_exec() must be zero.
3078 The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_xxx,
3079 PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
3080 PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_HARD, and
3081 PCRE_PARTIAL_SOFT.
3082
3083 If the pattern was successfully studied with one of the just-in-time
3084 (JIT) compile options, the only supported options for JIT execution are
3085 PCRE_NO_UTF8_CHECK, PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY,
3086 PCRE_NOTEMPTY_ATSTART, PCRE_PARTIAL_HARD, and PCRE_PARTIAL_SOFT. If an
3087 unsupported option is used, JIT execution is disabled and the normal
3088 interpretive code in pcre_exec() is run.
3089
3090 PCRE_ANCHORED
3091
3092 The PCRE_ANCHORED option limits pcre_exec() to matching at the first
3093 matching position. If a pattern was compiled with PCRE_ANCHORED, or
3094 turned out to be anchored by virtue of its contents, it cannot be made
3095 unachored at matching time.
3096
3097 PCRE_BSR_ANYCRLF
3098 PCRE_BSR_UNICODE
3099
3100 These options (which are mutually exclusive) control what the \R escape
3101 sequence matches. The choice is either to match only CR, LF, or CRLF,
3102 or to match any Unicode newline sequence. These options override the
3103 choice that was made or defaulted when the pattern was compiled.
3104
3105 PCRE_NEWLINE_CR
3106 PCRE_NEWLINE_LF
3107 PCRE_NEWLINE_CRLF
3108 PCRE_NEWLINE_ANYCRLF
3109 PCRE_NEWLINE_ANY
3110
3111 These options override the newline definition that was chosen or
3112 defaulted when the pattern was compiled. For details, see the descrip-
3113 tion of pcre_compile() above. During matching, the newline choice
3114 affects the behaviour of the dot, circumflex, and dollar metacharac-
3115 ters. It may also alter the way the match position is advanced after a
3116 match failure for an unanchored pattern.
3117
3118 When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is
3119 set, and a match attempt for an unanchored pattern fails when the cur-
3120 rent position is at a CRLF sequence, and the pattern contains no
3121 explicit matches for CR or LF characters, the match position is
3122 advanced by two characters instead of one, in other words, to after the
3123 CRLF.
3124
3125 The above rule is a compromise that makes the most common cases work as
3126 expected. For example, if the pattern is .+A (and the PCRE_DOTALL
3127 option is not set), it does not match the string "\r\nA" because, after
3128 failing at the start, it skips both the CR and the LF before retrying.
3129 However, the pattern [\r\n]A does match that string, because it con-
3130 tains an explicit CR or LF reference, and so advances only by one char-
3131 acter after the first failure.
3132
3133 An explicit match for CR of LF is either a literal appearance of one of
3134 those characters, or one of the \r or \n escape sequences. Implicit
3135 matches such as [^X] do not count, nor does \s (which includes CR and
3136 LF in the characters that it matches).
3137
3138 Notwithstanding the above, anomalous effects may still occur when CRLF
3139 is a valid newline sequence and explicit \r or \n escapes appear in the
3140 pattern.
3141
3142 PCRE_NOTBOL
3143
3144 This option specifies that first character of the subject string is not
3145 the beginning of a line, so the circumflex metacharacter should not
3146 match before it. Setting this without PCRE_MULTILINE (at compile time)
3147 causes circumflex never to match. This option affects only the behav-
3148 iour of the circumflex metacharacter. It does not affect \A.
3149
3150 PCRE_NOTEOL
3151
3152 This option specifies that the end of the subject string is not the end
3153 of a line, so the dollar metacharacter should not match it nor (except
3154 in multiline mode) a newline immediately before it. Setting this with-
3155 out PCRE_MULTILINE (at compile time) causes dollar never to match. This
3156 option affects only the behaviour of the dollar metacharacter. It does
3157 not affect \Z or \z.
3158
3159 PCRE_NOTEMPTY
3160
3161 An empty string is not considered to be a valid match if this option is
3162 set. If there are alternatives in the pattern, they are tried. If all
3163 the alternatives match the empty string, the entire match fails. For
3164 example, if the pattern
3165
3166 a?b?
3167
3168 is applied to a string not beginning with "a" or "b", it matches an
3169 empty string at the start of the subject. With PCRE_NOTEMPTY set, this
3170 match is not valid, so PCRE searches further into the string for occur-
3171 rences of "a" or "b".
3172
3173 PCRE_NOTEMPTY_ATSTART
3174
3175 This is like PCRE_NOTEMPTY, except that an empty string match that is
3176 not at the start of the subject is permitted. If the pattern is
3177 anchored, such a match can occur only if the pattern contains \K.
3178
3179 Perl has no direct equivalent of PCRE_NOTEMPTY or
3180 PCRE_NOTEMPTY_ATSTART, but it does make a special case of a pattern
3181 match of the empty string within its split() function, and when using
3182 the /g modifier. It is possible to emulate Perl's behaviour after
3183 matching a null string by first trying the match again at the same off-
3184 set with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then if that
3185 fails, by advancing the starting offset (see below) and trying an ordi-
3186 nary match again. There is some code that demonstrates how to do this
3187 in the pcredemo sample program. In the most general case, you have to
3188 check to see if the newline convention recognizes CRLF as a newline,
3189 and if so, and the current character is CR followed by LF, advance the
3190 starting offset by two characters instead of one.
3191
3192 PCRE_NO_START_OPTIMIZE
3193
3194 There are a number of optimizations that pcre_exec() uses at the start
3195 of a match, in order to speed up the process. For example, if it is
3196 known that an unanchored match must start with a specific character, it
3197 searches the subject for that character, and fails immediately if it
3198 cannot find it, without actually running the main matching function.
3199 This means that a special item such as (*COMMIT) at the start of a pat-
3200 tern is not considered until after a suitable starting point for the
3201 match has been found. When callouts or (*MARK) items are in use, these
3202 "start-up" optimizations can cause them to be skipped if the pattern is
3203 never actually used. The start-up optimizations are in effect a pre-
3204 scan of the subject that takes place before the pattern is run.
3205
3206 The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations,
3207 possibly causing performance to suffer, but ensuring that in cases
3208 where the result is "no match", the callouts do occur, and that items
3209 such as (*COMMIT) and (*MARK) are considered at every possible starting
3210 position in the subject string. If PCRE_NO_START_OPTIMIZE is set at
3211 compile time, it cannot be unset at matching time. The use of
3212 PCRE_NO_START_OPTIMIZE disables JIT execution; when it is set, matching
3213 is always done using interpretively.
3214
3215 Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching
3216 operation. Consider the pattern
3217
3218 (*COMMIT)ABC
3219
3220 When this is compiled, PCRE records the fact that a match must start
3221 with the character "A". Suppose the subject string is "DEFABC". The
3222 start-up optimization scans along the subject, finds "A" and runs the
3223 first match attempt from there. The (*COMMIT) item means that the pat-
3224 tern must match the current starting position, which in this case, it
3225 does. However, if the same match is run with PCRE_NO_START_OPTIMIZE
3226 set, the initial scan along the subject string does not happen. The
3227 first match attempt is run starting from "D" and when this fails,
3228 (*COMMIT) prevents any further matches being tried, so the overall
3229 result is "no match". If the pattern is studied, more start-up opti-
3230 mizations may be used. For example, a minimum length for the subject
3231 may be recorded. Consider the pattern
3232
3233 (*MARK:A)(X|Y)
3234
3235 The minimum length for a match is one character. If the subject is
3236 "ABC", there will be attempts to match "ABC", "BC", "C", and then
3237 finally an empty string. If the pattern is studied, the final attempt
3238 does not take place, because PCRE knows that the subject is too short,
3239 and so the (*MARK) is never encountered. In this case, studying the
3240 pattern does not affect the overall match result, which is still "no
3241 match", but it does affect the auxiliary information that is returned.
3242
3243 PCRE_NO_UTF8_CHECK
3244
3245 When PCRE_UTF8 is set at compile time, the validity of the subject as a
3246 UTF-8 string is automatically checked when pcre_exec() is subsequently
3247 called. The entire string is checked before any other processing takes
3248 place. The value of startoffset is also checked to ensure that it
3249 points to the start of a UTF-8 character. There is a discussion about
3250 the validity of UTF-8 strings in the pcreunicode page. If an invalid
3251 sequence of bytes is found, pcre_exec() returns the error
3252 PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is a
3253 truncated character at the end of the subject, PCRE_ERROR_SHORTUTF8. In
3254 both cases, information about the precise nature of the error may also
3255 be returned (see the descriptions of these errors in the section enti-
3256 tled Error return values from pcre_exec() below). If startoffset con-
3257 tains a value that does not point to the start of a UTF-8 character (or
3258 to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is returned.
3259
3260 If you already know that your subject is valid, and you want to skip
3261 these checks for performance reasons, you can set the
3262 PCRE_NO_UTF8_CHECK option when calling pcre_exec(). You might want to
3263 do this for the second and subsequent calls to pcre_exec() if you are
3264 making repeated calls to find all the matches in a single subject
3265 string. However, you should be sure that the value of startoffset
3266 points to the start of a character (or the end of the subject). When
3267 PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid string as a
3268 subject or an invalid value of startoffset is undefined. Your program
3269 may crash.
3270
3271 PCRE_PARTIAL_HARD
3272 PCRE_PARTIAL_SOFT
3273
3274 These options turn on the partial matching feature. For backwards com-
3275 patibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial
3276 match occurs if the end of the subject string is reached successfully,
3277 but there are not enough subject characters to complete the match. If
3278 this happens when PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set,
3279 matching continues by testing any remaining alternatives. Only if no
3280 complete match can be found is PCRE_ERROR_PARTIAL returned instead of
3281 PCRE_ERROR_NOMATCH. In other words, PCRE_PARTIAL_SOFT says that the
3282 caller is prepared to handle a partial match, but only if no complete
3283 match can be found.
3284
3285 If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this
3286 case, if a partial match is found, pcre_exec() immediately returns
3287 PCRE_ERROR_PARTIAL, without considering any other alternatives. In
3288 other words, when PCRE_PARTIAL_HARD is set, a partial match is consid-
3289 ered to be more important that an alternative complete match.
3290
3291 In both cases, the portion of the string that was inspected when the
3292 partial match was found is set as the first matching string. There is a
3293 more detailed discussion of partial and multi-segment matching, with
3294 examples, in the pcrepartial documentation.
3295
3296 The string to be matched by pcre_exec()
3297
3298 The subject string is passed to pcre_exec() as a pointer in subject, a
3299 length in bytes in length, and a starting byte offset in startoffset.
3300 If this is negative or greater than the length of the subject,
3301 pcre_exec() returns PCRE_ERROR_BADOFFSET. When the starting offset is
3302 zero, the search for a match starts at the beginning of the subject,
3303 and this is by far the most common case. In UTF-8 mode, the byte offset
3304 must point to the start of a UTF-8 character (or the end of the sub-
3305 ject). Unlike the pattern string, the subject may contain binary zero
3306 bytes.
3307
3308 A non-zero starting offset is useful when searching for another match
3309 in the same subject by calling pcre_exec() again after a previous suc-
3310 cess. Setting startoffset differs from just passing over a shortened
3311 string and setting PCRE_NOTBOL in the case of a pattern that begins
3312 with any kind of lookbehind. For example, consider the pattern
3313
3314 \Biss\B
3315
3316 which finds occurrences of "iss" in the middle of words. (\B matches
3317 only if the current position in the subject is not a word boundary.)
3318 When applied to the string "Mississipi" the first call to pcre_exec()
3319 finds the first occurrence. If pcre_exec() is called again with just
3320 the remainder of the subject, namely "issipi", it does not match,
3321 because \B is always false at the start of the subject, which is deemed
3322 to be a word boundary. However, if pcre_exec() is passed the entire
3323 string again, but with startoffset set to 4, it finds the second occur-
3324 rence of "iss" because it is able to look behind the starting point to
3325 discover that it is preceded by a letter.
3326
3327 Finding all the matches in a subject is tricky when the pattern can
3328 match an empty string. It is possible to emulate Perl's /g behaviour by
3329 first trying the match again at the same offset, with the
3330 PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED options, and then if that
3331 fails, advancing the starting offset and trying an ordinary match
3332 again. There is some code that demonstrates how to do this in the pcre-
3333 demo sample program. In the most general case, you have to check to see
3334 if the newline convention recognizes CRLF as a newline, and if so, and
3335 the current character is CR followed by LF, advance the starting offset
3336 by two characters instead of one.
3337
3338 If a non-zero starting offset is passed when the pattern is anchored,
3339 one attempt to match at the given offset is made. This can only succeed
3340 if the pattern does not require the match to be at the start of the
3341 subject.
3342
3343 How pcre_exec() returns captured substrings
3344
3345 In general, a pattern matches a certain portion of the subject, and in
3346 addition, further substrings from the subject may be picked out by
3347 parts of the pattern. Following the usage in Jeffrey Friedl's book,
3348 this is called "capturing" in what follows, and the phrase "capturing
3349 subpattern" is used for a fragment of a pattern that picks out a sub-
3350 string. PCRE supports several other kinds of parenthesized subpattern
3351 that do not cause substrings to be captured.
3352
3353 Captured substrings are returned to the caller via a vector of integers
3354 whose address is passed in ovector. The number of elements in the vec-
3355 tor is passed in ovecsize, which must be a non-negative number. Note:
3356 this argument is NOT the size of ovector in bytes.
3357
3358 The first two-thirds of the vector is used to pass back captured sub-
3359 strings, each substring using a pair of integers. The remaining third
3360 of the vector is used as workspace by pcre_exec() while matching cap-
3361 turing subpatterns, and is not available for passing back information.
3362 The number passed in ovecsize should always be a multiple of three. If
3363 it is not, it is rounded down.
3364
3365 When a match is successful, information about captured substrings is
3366 returned in pairs of integers, starting at the beginning of ovector,
3367 and continuing up to two-thirds of its length at the most. The first
3368 element of each pair is set to the byte offset of the first character
3369 in a substring, and the second is set to the byte offset of the first
3370 character after the end of a substring. Note: these values are always
3371 byte offsets, even in UTF-8 mode. They are not character counts.
3372
3373 The first pair of integers, ovector[0] and ovector[1], identify the
3374 portion of the subject string matched by the entire pattern. The next
3375 pair is used for the first capturing subpattern, and so on. The value
3376 returned by pcre_exec() is one more than the highest numbered pair that
3377 has been set. For example, if two substrings have been captured, the
3378 returned value is 3. If there are no capturing subpatterns, the return
3379 value from a successful match is 1, indicating that just the first pair
3380 of offsets has been set.
3381
3382 If a capturing subpattern is matched repeatedly, it is the last portion
3383 of the string that it matched that is returned.
3384
3385 If the vector is too small to hold all the captured substring offsets,
3386 it is used as far as possible (up to two-thirds of its length), and the
3387 function returns a value of zero. If neither the actual string matched
3388 nor any captured substrings are of interest, pcre_exec() may be called
3389 with ovector passed as NULL and ovecsize as zero. However, if the pat-
3390 tern contains back references and the ovector is not big enough to
3391 remember the related substrings, PCRE has to get additional memory for
3392 use during matching. Thus it is usually advisable to supply an ovector
3393 of reasonable size.
3394
3395 There are some cases where zero is returned (indicating vector over-
3396 flow) when in fact the vector is exactly the right size for the final
3397 match. For example, consider the pattern
3398
3399 (a)(?:(b)c|bd)
3400
3401 If a vector of 6 elements (allowing for only 1 captured substring) is
3402 given with subject string "abd", pcre_exec() will try to set the second
3403 captured string, thereby recording a vector overflow, before failing to
3404 match "c" and backing up to try the second alternative. The zero
3405 return, however, does correctly indicate that the maximum number of
3406 slots (namely 2) have been filled. In similar cases where there is tem-
3407 porary overflow, but the final number of used slots is actually less
3408 than the maximum, a non-zero value is returned.
3409
3410 The pcre_fullinfo() function can be used to find out how many capturing
3411 subpatterns there are in a compiled pattern. The smallest size for
3412 ovector that will allow for n captured substrings, in addition to the
3413 offsets of the substring matched by the whole pattern, is (n+1)*3.
3414
3415 It is possible for capturing subpattern number n+1 to match some part
3416 of the subject when subpattern n has not been used at all. For example,
3417 if the string "abc" is matched against the pattern (a|(z))(bc) the
3418 return from the function is 4, and subpatterns 1 and 3 are matched, but
3419 2 is not. When this happens, both values in the offset pairs corre-
3420 sponding to unused subpatterns are set to -1.
3421
3422 Offset values that correspond to unused subpatterns at the end of the
3423 expression are also set to -1. For example, if the string "abc" is
3424 matched against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not
3425 matched. The return from the function is 2, because the highest used
3426 capturing subpattern number is 1, and the offsets for for the second
3427 and third capturing subpatterns (assuming the vector is large enough,
3428 of course) are set to -1.
3429
3430 Note: Elements in the first two-thirds of ovector that do not corre-
3431 spond to capturing parentheses in the pattern are never changed. That
3432 is, if a pattern contains n capturing parentheses, no more than ovec-
3433 tor[0] to ovector[2n+1] are set by pcre_exec(). The other elements (in
3434 the first two-thirds) retain whatever values they previously had.
3435
3436 Some convenience functions are provided for extracting the captured
3437 substrings as separate strings. These are described below.
3438
3439 Error return values from pcre_exec()
3440
3441 If pcre_exec() fails, it returns a negative number. The following are
3442 defined in the header file:
3443
3444 PCRE_ERROR_NOMATCH (-1)
3445
3446 The subject string did not match the pattern.
3447
3448 PCRE_ERROR_NULL (-2)
3449
3450 Either code or subject was passed as NULL, or ovector was NULL and
3451 ovecsize was not zero.
3452
3453 PCRE_ERROR_BADOPTION (-3)
3454
3455 An unrecognized bit was set in the options argument.
3456
3457 PCRE_ERROR_BADMAGIC (-4)
3458
3459 PCRE stores a 4-byte "magic number" at the start of the compiled code,
3460 to catch the case when it is passed a junk pointer and to detect when a
3461 pattern that was compiled in an environment of one endianness is run in
3462 an environment with the other endianness. This is the error that PCRE
3463 gives when the magic number is not present.
3464
3465 PCRE_ERROR_UNKNOWN_OPCODE (-5)
3466
3467 While running the pattern match, an unknown item was encountered in the
3468 compiled pattern. This error could be caused by a bug in PCRE or by
3469 overwriting of the compiled pattern.
3470
3471 PCRE_ERROR_NOMEMORY (-6)
3472
3473 If a pattern contains back references, but the ovector that is passed
3474 to pcre_exec() is not big enough to remember the referenced substrings,
3475 PCRE gets a block of memory at the start of matching to use for this
3476 purpose. If the call via pcre_malloc() fails, this error is given. The
3477 memory is automatically freed at the end of matching.
3478
3479 This error is also given if pcre_stack_malloc() fails in pcre_exec().
3480 This can happen only when PCRE has been compiled with --disable-stack-
3481 for-recursion.
3482
3483 PCRE_ERROR_NOSUBSTRING (-7)
3484
3485 This error is used by the pcre_copy_substring(), pcre_get_substring(),
3486 and pcre_get_substring_list() functions (see below). It is never
3487 returned by pcre_exec().
3488
3489 PCRE_ERROR_MATCHLIMIT (-8)
3490
3491 The backtracking limit, as specified by the match_limit field in a
3492 pcre_extra structure (or defaulted) was reached. See the description
3493 above.
3494
3495 PCRE_ERROR_CALLOUT (-9)
3496
3497 This error is never generated by pcre_exec() itself. It is provided for
3498 use by callout functions that want to yield a distinctive error code.
3499 See the pcrecallout documentation for details.
3500
3501 PCRE_ERROR_BADUTF8 (-10)
3502
3503 A string that contains an invalid UTF-8 byte sequence was passed as a
3504 subject, and the PCRE_NO_UTF8_CHECK option was not set. If the size of
3505 the output vector (ovecsize) is at least 2, the byte offset to the
3506 start of the the invalid UTF-8 character is placed in the first ele-
3507 ment, and a reason code is placed in the second element. The reason
3508 codes are listed in the following section. For backward compatibility,
3509 if PCRE_PARTIAL_HARD is set and the problem is a truncated UTF-8 char-
3510 acter at the end of the subject (reason codes 1 to 5),
3511 PCRE_ERROR_SHORTUTF8 is returned instead of PCRE_ERROR_BADUTF8.
3512
3513 PCRE_ERROR_BADUTF8_OFFSET (-11)
3514
3515 The UTF-8 byte sequence that was passed as a subject was checked and
3516 found to be valid (the PCRE_NO_UTF8_CHECK option was not set), but the
3517 value of startoffset did not point to the beginning of a UTF-8 charac-
3518 ter or the end of the subject.
3519
3520 PCRE_ERROR_PARTIAL (-12)
3521
3522 The subject string did not match, but it did match partially. See the
3523 pcrepartial documentation for details of partial matching.
3524
3525 PCRE_ERROR_BADPARTIAL (-13)
3526
3527 This code is no longer in use. It was formerly returned when the
3528 PCRE_PARTIAL option was used with a compiled pattern containing items
3529 that were not supported for partial matching. From release 8.00
3530 onwards, there are no restrictions on partial matching.
3531
3532 PCRE_ERROR_INTERNAL (-14)
3533
3534 An unexpected internal error has occurred. This error could be caused
3535 by a bug in PCRE or by overwriting of the compiled pattern.
3536
3537 PCRE_ERROR_BADCOUNT (-15)
3538
3539 This error is given if the value of the ovecsize argument is negative.
3540
3541 PCRE_ERROR_RECURSIONLIMIT (-21)
3542
3543 The internal recursion limit, as specified by the match_limit_recursion
3544 field in a pcre_extra structure (or defaulted) was reached. See the
3545 description above.
3546
3547 PCRE_ERROR_BADNEWLINE (-23)
3548
3549 An invalid combination of PCRE_NEWLINE_xxx options was given.
3550
3551 PCRE_ERROR_BADOFFSET (-24)
3552
3553 The value of startoffset was negative or greater than the length of the
3554 subject, that is, the value in length.
3555
3556 PCRE_ERROR_SHORTUTF8 (-25)
3557
3558 This error is returned instead of PCRE_ERROR_BADUTF8 when the subject
3559 string ends with a truncated UTF-8 character and the PCRE_PARTIAL_HARD
3560 option is set. Information about the failure is returned as for
3561 PCRE_ERROR_BADUTF8. It is in fact sufficient to detect this case, but
3562 this special error code for PCRE_PARTIAL_HARD precedes the implementa-
3563 tion of returned information; it is retained for backwards compatibil-
3564 ity.
3565
3566 PCRE_ERROR_RECURSELOOP (-26)
3567
3568 This error is returned when pcre_exec() detects a recursion loop within
3569 the pattern. Specifically, it means that either the whole pattern or a
3570 subpattern has been called recursively for the second time at the same
3571 position in the subject string. Some simple patterns that might do this
3572 are detected and faulted at compile time, but more complicated cases,
3573 in particular mutual recursions between two different subpatterns, can-
3574 not be detected until run time.
3575
3576 PCRE_ERROR_JIT_STACKLIMIT (-27)
3577
3578 This error is returned when a pattern that was successfully studied
3579 using a JIT compile option is being matched, but the memory available
3580 for the just-in-time processing stack is not large enough. See the
3581 pcrejit documentation for more details.
3582
3583 PCRE_ERROR_BADMODE (-28)
3584
3585 This error is given if a pattern that was compiled by the 8-bit library
3586 is passed to a 16-bit or 32-bit library function, or vice versa.
3587
3588 PCRE_ERROR_BADENDIANNESS (-29)
3589
3590 This error is given if a pattern that was compiled and saved is
3591 reloaded on a host with different endianness. The utility function
3592 pcre_pattern_to_host_byte_order() can be used to convert such a pattern
3593 so that it runs on the new host.
3594
3595 PCRE_ERROR_JIT_BADOPTION
3596
3597 This error is returned when a pattern that was successfully studied
3598 using a JIT compile option is being matched, but the matching mode
3599 (partial or complete match) does not correspond to any JIT compilation
3600 mode. When the JIT fast path function is used, this error may be also
3601 given for invalid options. See the pcrejit documentation for more
3602 details.
3603
3604 PCRE_ERROR_BADLENGTH (-32)
3605
3606 This error is given if pcre_exec() is called with a negative value for
3607 the length argument.
3608
3609 Error numbers -16 to -20, -22, and 30 are not used by pcre_exec().
3610
3611 Reason codes for invalid UTF-8 strings
3612
3613 This section applies only to the 8-bit library. The corresponding
3614 information for the 16-bit library is given in the pcre16 page. The
3615 corresponding information for the 32-bit library is given in the pcre32
3616 page.
3617
3618 When pcre_exec() returns either PCRE_ERROR_BADUTF8 or PCRE_ERROR_SHORT-
3619 UTF8, and the size of the output vector (ovecsize) is at least 2, the
3620 offset of the start of the invalid UTF-8 character is placed in the
3621 first output vector element (ovector[0]) and a reason code is placed in
3622 the second element (ovector[1]). The reason codes are given names in
3623 the pcre.h header file:
3624
3625 PCRE_UTF8_ERR1
3626 PCRE_UTF8_ERR2
3627 PCRE_UTF8_ERR3
3628 PCRE_UTF8_ERR4
3629 PCRE_UTF8_ERR5
3630
3631 The string ends with a truncated UTF-8 character; the code specifies
3632 how many bytes are missing (1 to 5). Although RFC 3629 restricts UTF-8
3633 characters to be no longer than 4 bytes, the encoding scheme (origi-
3634 nally defined by RFC 2279) allows for up to 6 bytes, and this is
3635 checked first; hence the possibility of 4 or 5 missing bytes.
3636
3637 PCRE_UTF8_ERR6
3638 PCRE_UTF8_ERR7
3639 PCRE_UTF8_ERR8
3640 PCRE_UTF8_ERR9
3641 PCRE_UTF8_ERR10
3642
3643 The two most significant bits of the 2nd, 3rd, 4th, 5th, or 6th byte of
3644 the character do not have the binary value 0b10 (that is, either the
3645 most significant bit is 0, or the next bit is 1).
3646
3647 PCRE_UTF8_ERR11
3648 PCRE_UTF8_ERR12
3649
3650 A character that is valid by the RFC 2279 rules is either 5 or 6 bytes
3651 long; these code points are excluded by RFC 3629.
3652
3653 PCRE_UTF8_ERR13
3654
3655 A 4-byte character has a value greater than 0x10fff; these code points
3656 are excluded by RFC 3629.
3657
3658 PCRE_UTF8_ERR14
3659
3660 A 3-byte character has a value in the range 0xd800 to 0xdfff; this
3661 range of code points are reserved by RFC 3629 for use with UTF-16, and
3662 so are excluded from UTF-8.
3663
3664 PCRE_UTF8_ERR15
3665 PCRE_UTF8_ERR16
3666 PCRE_UTF8_ERR17
3667 PCRE_UTF8_ERR18
3668 PCRE_UTF8_ERR19
3669
3670 A 2-, 3-, 4-, 5-, or 6-byte character is "overlong", that is, it codes
3671 for a value that can be represented by fewer bytes, which is invalid.
3672 For example, the two bytes 0xc0, 0xae give the value 0x2e, whose cor-
3673 rect coding uses just one byte.
3674
3675 PCRE_UTF8_ERR20
3676
3677 The two most significant bits of the first byte of a character have the
3678 binary value 0b10 (that is, the most significant bit is 1 and the sec-
3679 ond is 0). Such a byte can only validly occur as the second or subse-
3680 quent byte of a multi-byte character.
3681
3682 PCRE_UTF8_ERR21
3683
3684 The first byte of a character has the value 0xfe or 0xff. These values
3685 can never occur in a valid UTF-8 string.
3686
3687 PCRE_UTF8_ERR2
3688
3689 Non-character. These are the last two characters in each plane (0xfffe,
3690 0xffff, 0x1fffe, 0x1ffff .. 0x10fffe, 0x10ffff), and the characters
3691 0xfdd0..0xfdef.
3692
3693
3694 EXTRACTING CAPTURED SUBSTRINGS BY NUMBER
3695
3696 int pcre_copy_substring(const char *subject, int *ovector,
3697 int stringcount, int stringnumber, char *buffer,
3698 int buffersize);
3699
3700 int pcre_get_substring(const char *subject, int *ovector,
3701 int stringcount, int stringnumber,
3702 const char **stringptr);
3703
3704 int pcre_get_substring_list(const char *subject,
3705 int *ovector, int stringcount, const char ***listptr);
3706
3707 Captured substrings can be accessed directly by using the offsets
3708 returned by pcre_exec() in ovector. For convenience, the functions
3709 pcre_copy_substring(), pcre_get_substring(), and pcre_get_sub-
3710 string_list() are provided for extracting captured substrings as new,
3711 separate, zero-terminated strings. These functions identify substrings
3712 by number. The next section describes functions for extracting named
3713 substrings.
3714
3715 A substring that contains a binary zero is correctly extracted and has
3716 a further zero added on the end, but the result is not, of course, a C
3717 string. However, you can process such a string by referring to the
3718 length that is returned by pcre_copy_substring() and pcre_get_sub-
3719 string(). Unfortunately, the interface to pcre_get_substring_list() is
3720 not adequate for handling strings containing binary zeros, because the
3721 end of the final string is not independently indicated.
3722
3723 The first three arguments are the same for all three of these func-
3724 tions: subject is the subject string that has just been successfully
3725 matched, ovector is a pointer to the vector of integer offsets that was
3726 passed to pcre_exec(), and stringcount is the number of substrings that
3727 were captured by the match, including the substring that matched the
3728 entire regular expression. This is the value returned by pcre_exec() if
3729 it is greater than zero. If pcre_exec() returned zero, indicating that
3730 it ran out of space in ovector, the value passed as stringcount should
3731 be the number of elements in the vector divided by three.
3732
3733 The functions pcre_copy_substring() and pcre_get_substring() extract a
3734 single substring, whose number is given as stringnumber. A value of
3735 zero extracts the substring that matched the entire pattern, whereas
3736 higher values extract the captured substrings. For pcre_copy_sub-
3737 string(), the string is placed in buffer, whose length is given by
3738 buffersize, while for pcre_get_substring() a new block of memory is
3739 obtained via pcre_malloc, and its address is returned via stringptr.
3740 The yield of the function is the length of the string, not including
3741 the terminating zero, or one of these error codes:
3742
3743 PCRE_ERROR_NOMEMORY (-6)
3744
3745 The buffer was too small for pcre_copy_substring(), or the attempt to
3746 get memory failed for pcre_get_substring().
3747
3748 PCRE_ERROR_NOSUBSTRING (-7)
3749
3750 There is no substring whose number is stringnumber.
3751
3752 The pcre_get_substring_list() function extracts all available sub-
3753 strings and builds a list of pointers to them. All this is done in a
3754 single block of memory that is obtained via pcre_malloc. The address of
3755 the memory block is returned via listptr, which is also the start of
3756 the list of string pointers. The end of the list is marked by a NULL
3757 pointer. The yield of the function is zero if all went well, or the
3758 error code
3759
3760 PCRE_ERROR_NOMEMORY (-6)
3761
3762 if the attempt to get the memory block failed.
3763
3764 When any of these functions encounter a substring that is unset, which
3765 can happen when capturing subpattern number n+1 matches some part of
3766 the subject, but subpattern n has not been used at all, they return an
3767 empty string. This can be distinguished from a genuine zero-length sub-
3768 string by inspecting the appropriate offset in ovector, which is nega-
3769 tive for unset substrings.
3770
3771 The two convenience functions pcre_free_substring() and pcre_free_sub-
3772 string_list() can be used to free the memory returned by a previous
3773 call of pcre_get_substring() or pcre_get_substring_list(), respec-
3774 tively. They do nothing more than call the function pointed to by
3775 pcre_free, which of course could be called directly from a C program.
3776 However, PCRE is used in some situations where it is linked via a spe-
3777 cial interface to another programming language that cannot use
3778 pcre_free directly; it is for these cases that the functions are pro-
3779 vided.
3780
3781
3782 EXTRACTING CAPTURED SUBSTRINGS BY NAME
3783
3784 int pcre_get_stringnumber(const pcre *code,
3785 const char *name);
3786
3787 int pcre_copy_named_substring(const pcre *code,
3788 const char *subject, int *ovector,
3789 int stringcount, const char *stringname,
3790 char *buffer, int buffersize);
3791
3792 int pcre_get_named_substring(const pcre *code,
3793 const char *subject, int *ovector,
3794 int stringcount, const char *stringname,
3795 const char **stringptr);
3796
3797 To extract a substring by name, you first have to find associated num-
3798 ber. For example, for this pattern
3799
3800 (a+)b(?<xxx>\d+)...
3801
3802 the number of the subpattern called "xxx" is 2. If the name is known to
3803 be unique (PCRE_DUPNAMES was not set), you can find the number from the
3804 name by calling pcre_get_stringnumber(). The first argument is the com-
3805 piled pattern, and the second is the name. The yield of the function is
3806 the subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no
3807 subpattern of that name.
3808
3809 Given the number, you can extract the substring directly, or use one of
3810 the functions described in the previous section. For convenience, there
3811 are also two functions that do the whole job.
3812
3813 Most of the arguments of pcre_copy_named_substring() and
3814 pcre_get_named_substring() are the same as those for the similarly
3815 named functions that extract by number. As these are described in the
3816 previous section, they are not re-described here. There are just two
3817 differences:
3818
3819 First, instead of a substring number, a substring name is given. Sec-
3820 ond, there is an extra argument, given at the start, which is a pointer
3821 to the compiled pattern. This is needed in order to gain access to the
3822 name-to-number translation table.
3823
3824 These functions call pcre_get_stringnumber(), and if it succeeds, they
3825 then call pcre_copy_substring() or pcre_get_substring(), as appropri-
3826 ate. NOTE: If PCRE_DUPNAMES is set and there are duplicate names, the
3827 behaviour may not be what you want (see the next section).
3828
3829 Warning: If the pattern uses the (?| feature to set up multiple subpat-
3830 terns with the same number, as described in the section on duplicate
3831 subpattern numbers in the pcrepattern page, you cannot use names to
3832 distinguish the different subpatterns, because names are not included
3833 in the compiled code. The matching process uses only numbers. For this
3834 reason, the use of different names for subpatterns of the same number
3835 causes an error at compile time.
3836
3837
3838 DUPLICATE SUBPATTERN NAMES
3839
3840 int pcre_get_stringtable_entries(const pcre *code,
3841 const char *name, char **first, char **last);
3842
3843 When a pattern is compiled with the PCRE_DUPNAMES option, names for
3844 subpatterns are not required to be unique. (Duplicate names are always
3845 allowed for subpatterns with the same number, created by using the (?|
3846 feature. Indeed, if such subpatterns are named, they are required to
3847 use the same names.)
3848
3849 Normally, patterns with duplicate names are such that in any one match,
3850 only one of the named subpatterns participates. An example is shown in
3851 the pcrepattern documentation.
3852
3853 When duplicates are present, pcre_copy_named_substring() and
3854 pcre_get_named_substring() return the first substring corresponding to
3855 the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING
3856 (-7) is returned; no data is returned. The pcre_get_stringnumber()
3857 function returns one of the numbers that are associated with the name,
3858 but it is not defined which it is.
3859
3860 If you want to get full details of all captured substrings for a given
3861 name, you must use the pcre_get_stringtable_entries() function. The
3862 first argument is the compiled pattern, and the second is the name. The
3863 third and fourth are pointers to variables which are updated by the
3864 function. After it has run, they point to the first and last entries in
3865 the name-to-number table for the given name. The function itself
3866 returns the length of each entry, or PCRE_ERROR_NOSUBSTRING (-7) if
3867 there are none. The format of the table is described above in the sec-
3868 tion entitled Information about a pattern above. Given all the rele-
3869 vant entries for the name, you can extract each of their numbers, and
3870 hence the captured data, if any.
3871
3872
3873 FINDING ALL POSSIBLE MATCHES
3874
3875 The traditional matching function uses a similar algorithm to Perl,
3876 which stops when it finds the first match, starting at a given point in
3877 the subject. If you want to find all possible matches, or the longest
3878 possible match, consider using the alternative matching function (see
3879 below) instead. If you cannot use the alternative function, but still
3880 need to find all possible matches, you can kludge it up by making use
3881 of the callout facility, which is described in the pcrecallout documen-
3882 tation.
3883
3884 What you have to do is to insert a callout right at the end of the pat-
3885 tern. When your callout function is called, extract and save the cur-
3886 rent matched substring. Then return 1, which forces pcre_exec() to
3887 backtrack and try other alternatives. Ultimately, when it runs out of
3888 matches, pcre_exec() will yield PCRE_ERROR_NOMATCH.
3889
3890
3891 OBTAINING AN ESTIMATE OF STACK USAGE
3892
3893 Matching certain patterns using pcre_exec() can use a lot of process
3894 stack, which in certain environments can be rather limited in size.
3895 Some users find it helpful to have an estimate of the amount of stack
3896 that is used by pcre_exec(), to help them set recursion limits, as
3897 described in the pcrestack documentation. The estimate that is output
3898 by pcretest when called with the -m and -C options is obtained by call-
3899 ing pcre_exec with the values NULL, NULL, NULL, -999, and -999 for its
3900 first five arguments.
3901
3902 Normally, if its first argument is NULL, pcre_exec() immediately
3903 returns the negative error code PCRE_ERROR_NULL, but with this special
3904 combination of arguments, it returns instead a negative number whose
3905 absolute value is the approximate stack frame size in bytes. (A nega-
3906 tive number is used so that it is clear that no match has happened.)
3907 The value is approximate because in some cases, recursive calls to
3908 pcre_exec() occur when there are one or two additional variables on the
3909 stack.
3910
3911 If PCRE has been compiled to use the heap instead of the stack for
3912 recursion, the value returned is the size of each block that is
3913 obtained from the heap.
3914
3915
3916 MATCHING A PATTERN: THE ALTERNATIVE FUNCTION
3917
3918 int pcre_dfa_exec(const pcre *code, const pcre_extra *extra,
3919 const char *subject, int length, int startoffset,
3920 int options, int *ovector, int ovecsize,
3921 int *workspace, int wscount);
3922
3923 The function pcre_dfa_exec() is called to match a subject string
3924 against a compiled pattern, using a matching algorithm that scans the
3925 subject string just once, and does not backtrack. This has different
3926 characteristics to the normal algorithm, and is not compatible with
3927 Perl. Some of the features of PCRE patterns are not supported. Never-
3928 theless, there are times when this kind of matching can be useful. For
3929 a discussion of the two matching algorithms, and a list of features
3930 that pcre_dfa_exec() does not support, see the pcrematching documenta-
3931 tion.
3932
3933 The arguments for the pcre_dfa_exec() function are the same as for
3934 pcre_exec(), plus two extras. The ovector argument is used in a differ-
3935 ent way, and this is described below. The other common arguments are
3936 used in the same way as for pcre_exec(), so their description is not
3937 repeated here.
3938
3939 The two additional arguments provide workspace for the function. The
3940 workspace vector should contain at least 20 elements. It is used for
3941 keeping track of multiple paths through the pattern tree. More
3942 workspace will be needed for patterns and subjects where there are a
3943 lot of potential matches.
3944
3945 Here is an example of a simple call to pcre_dfa_exec():
3946
3947 int rc;
3948 int ovector[10];
3949 int wspace[20];
3950 rc = pcre_dfa_exec(
3951 re, /* result of pcre_compile() */
3952 NULL, /* we didn't study the pattern */
3953 "some string", /* the subject string */
3954 11, /* the length of the subject string */
3955 0, /* start at offset 0 in the subject */
3956 0, /* default options */
3957 ovector, /* vector of integers for substring information */
3958 10, /* number of elements (NOT size in bytes) */
3959 wspace, /* working space vector */
3960 20); /* number of elements (NOT size in bytes) */
3961
3962 Option bits for pcre_dfa_exec()
3963
3964 The unused bits of the options argument for pcre_dfa_exec() must be
3965 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEW-
3966 LINE_xxx, PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY,
3967 PCRE_NOTEMPTY_ATSTART, PCRE_NO_UTF8_CHECK, PCRE_BSR_ANYCRLF,
3968 PCRE_BSR_UNICODE, PCRE_NO_START_OPTIMIZE, PCRE_PARTIAL_HARD, PCRE_PAR-
3969 TIAL_SOFT, PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last
3970 four of these are exactly the same as for pcre_exec(), so their
3971 description is not repeated here.
3972
3973 PCRE_PARTIAL_HARD
3974 PCRE_PARTIAL_SOFT
3975
3976 These have the same general effect as they do for pcre_exec(), but the
3977 details are slightly different. When PCRE_PARTIAL_HARD is set for
3978 pcre_dfa_exec(), it returns PCRE_ERROR_PARTIAL if the end of the sub-
3979 ject is reached and there is still at least one matching possibility
3980 that requires additional characters. This happens even if some complete
3981 matches have also been found. When PCRE_PARTIAL_SOFT is set, the return
3982 code PCRE_ERROR_NOMATCH is converted into PCRE_ERROR_PARTIAL if the end
3983 of the subject is reached, there have been no complete matches, but
3984 there is still at least one matching possibility. The portion of the
3985 string that was inspected when the longest partial match was found is
3986 set as the first matching string in both cases. There is a more
3987 detailed discussion of partial and multi-segment matching, with exam-
3988 ples, in the pcrepartial documentation.
3989
3990 PCRE_DFA_SHORTEST
3991
3992 Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to
3993 stop as soon as it has found one match. Because of the way the alterna-
3994 tive algorithm works, this is necessarily the shortest possible match
3995 at the first possible matching point in the subject string.
3996
3997 PCRE_DFA_RESTART
3998
3999 When pcre_dfa_exec() returns a partial match, it is possible to call it
4000 again, with additional subject characters, and have it continue with
4001 the same match. The PCRE_DFA_RESTART option requests this action; when
4002 it is set, the workspace and wscount options must reference the same
4003 vector as before because data about the match so far is left in them
4004 after a partial match. There is more discussion of this facility in the
4005 pcrepartial documentation.
4006
4007 Successful returns from pcre_dfa_exec()
4008
4009 When pcre_dfa_exec() succeeds, it may have matched more than one sub-
4010 string in the subject. Note, however, that all the matches from one run
4011 of the function start at the same point in the subject. The shorter
4012 matches are all initial substrings of the longer matches. For example,
4013 if the pattern
4014
4015 <.*>
4016
4017 is matched against the string
4018
4019 This is <something> <something else> <something further> no more
4020
4021 the three matched strings are
4022
4023 <something>
4024 <something> <something else>
4025 <something> <something else> <something further>
4026
4027 On success, the yield of the function is a number greater than zero,
4028 which is the number of matched substrings. The substrings themselves
4029 are returned in ovector. Each string uses two elements; the first is
4030 the offset to the start, and the second is the offset to the end. In
4031 fact, all the strings have the same start offset. (Space could have
4032 been saved by giving this only once, but it was decided to retain some
4033 compatibility with the way pcre_exec() returns data, even though the
4034 meaning of the strings is different.)
4035
4036 The strings are returned in reverse order of length; that is, the long-
4037 est matching string is given first. If there were too many matches to
4038 fit into ovector, the yield of the function is zero, and the vector is
4039 filled with the longest matches. Unlike pcre_exec(), pcre_dfa_exec()
4040 can use the entire ovector for returning matched strings.
4041
4042 Error returns from pcre_dfa_exec()
4043
4044 The pcre_dfa_exec() function returns a negative number when it fails.
4045 Many of the errors are the same as for pcre_exec(), and these are
4046 described above. There are in addition the following errors that are
4047 specific to pcre_dfa_exec():
4048
4049 PCRE_ERROR_DFA_UITEM (-16)
4050
4051 This return is given if pcre_dfa_exec() encounters an item in the pat-
4052 tern that it does not support, for instance, the use of \C or a back
4053 reference.
4054
4055 PCRE_ERROR_DFA_UCOND (-17)
4056
4057 This return is given if pcre_dfa_exec() encounters a condition item
4058 that uses a back reference for the condition, or a test for recursion
4059 in a specific group. These are not supported.
4060
4061 PCRE_ERROR_DFA_UMLIMIT (-18)
4062
4063 This return is given if pcre_dfa_exec() is called with an extra block
4064 that contains a setting of the match_limit or match_limit_recursion
4065 fields. This is not supported (these fields are meaningless for DFA
4066 matching).
4067
4068 PCRE_ERROR_DFA_WSSIZE (-19)
4069
4070 This return is given if pcre_dfa_exec() runs out of space in the
4071 workspace vector.
4072
4073 PCRE_ERROR_DFA_RECURSE (-20)
4074
4075 When a recursive subpattern is processed, the matching function calls
4076 itself recursively, using private vectors for ovector and workspace.
4077 This error is given if the output vector is not large enough. This
4078 should be extremely rare, as a vector of size 1000 is used.
4079
4080 PCRE_ERROR_DFA_BADRESTART (-30)
4081
4082 When pcre_dfa_exec() is called with the PCRE_DFA_RESTART option, some
4083 plausibility checks are made on the contents of the workspace, which
4084 should contain data about the previous partial match. If any of these
4085 checks fail, this error is given.
4086
4087
4088 SEE ALSO
4089
4090 pcre16(3), pcre32(3), pcrebuild(3), pcrecallout(3), pcrecpp(3)(3),
4091 pcrematching(3), pcrepartial(3), pcreposix(3), pcreprecompile(3), pcre-
4092 sample(3), pcrestack(3).
4093
4094
4095 AUTHOR
4096
4097 Philip Hazel
4098 University Computing Service
4099 Cambridge CB2 3QH, England.
4100
4101
4102 REVISION
4103
4104 Last updated: 31 October 2012
4105 Copyright (c) 1997-2012 University of Cambridge.
4106 ------------------------------------------------------------------------------
4107
4108
4109 PCRECALLOUT(3) PCRECALLOUT(3)
4110
4111
4112 NAME
4113 PCRE - Perl-compatible regular expressions
4114
4115
4116 SYNOPSIS
4117
4118 #include <pcre.h>
4119
4120 int (*pcre_callout)(pcre_callout_block *);
4121
4122 int (*pcre16_callout)(pcre16_callout_block *);
4123
4124 int (*pcre32_callout)(pcre32_callout_block *);
4125
4126
4127 DESCRIPTION
4128
4129 PCRE provides a feature called "callout", which is a means of temporar-
4130 ily passing control to the caller of PCRE in the middle of pattern
4131 matching. The caller of PCRE provides an external function by putting
4132 its entry point in the global variable pcre_callout (pcre16_callout for
4133 the 16-bit library, pcre32_callout for the 32-bit library). By default,
4134 this variable contains NULL, which disables all calling out.
4135
4136 Within a regular expression, (?C) indicates the points at which the
4137 external function is to be called. Different callout points can be
4138 identified by putting a number less than 256 after the letter C. The
4139 default value is zero. For example, this pattern has two callout
4140 points:
4141
4142 (?C1)abc(?C2)def
4143
4144 If the PCRE_AUTO_CALLOUT option bit is set when a pattern is compiled,
4145 PCRE automatically inserts callouts, all with number 255, before each
4146 item in the pattern. For example, if PCRE_AUTO_CALLOUT is used with the
4147 pattern
4148
4149 A(\d{2}|--)
4150
4151 it is processed as if it were
4152
4153 (?C255)A(?C255)((?C255)\d{2}(?C255)|(?C255)-(?C255)-(?C255))(?C255)
4154
4155 Notice that there is a callout before and after each parenthesis and
4156 alternation bar. Automatic callouts can be used for tracking the
4157 progress of pattern matching. The pcretest command has an option that
4158 sets automatic callouts; when it is used, the output indicates how the
4159 pattern is matched. This is useful information when you are trying to
4160 optimize the performance of a particular pattern.
4161
4162 The use of callouts in a pattern makes it ineligible for optimization
4163 by the just-in-time compiler. Studying such a pattern with the
4164 PCRE_STUDY_JIT_COMPILE option always fails.
4165
4166
4167 MISSING CALLOUTS
4168
4169 You should be aware that, because of optimizations in the way PCRE
4170 matches patterns by default, callouts sometimes do not happen. For
4171 example, if the pattern is
4172
4173 ab(?C4)cd
4174
4175 PCRE knows that any matching string must contain the letter "d". If the
4176 subject string is "abyz", the lack of "d" means that matching doesn't
4177 ever start, and the callout is never reached. However, with "abyd",
4178 though the result is still no match, the callout is obeyed.
4179
4180 If the pattern is studied, PCRE knows the minimum length of a matching
4181 string, and will immediately give a "no match" return without actually
4182 running a match if the subject is not long enough, or, for unanchored
4183 patterns, if it has been scanned far enough.
4184
4185 You can disable these optimizations by passing the PCRE_NO_START_OPTI-
4186 MIZE option to the matching function, or by starting the pattern with
4187 (*NO_START_OPT). This slows down the matching process, but does ensure
4188 that callouts such as the example above are obeyed.
4189
4190
4191 THE CALLOUT INTERFACE
4192
4193 During matching, when PCRE reaches a callout point, the external func-
4194 tion defined by pcre_callout or pcre[16|32]_callout is called (if it is
4195 set). This applies to both normal and DFA matching. The only argument
4196 to the callout function is a pointer to a pcre_callout or
4197 pcre[16|32]_callout block. These structures contains the following
4198 fields:
4199
4200 int version;
4201 int callout_number;
4202 int *offset_vector;
4203 const char *subject; (8-bit version)
4204 PCRE_SPTR16 subject; (16-bit version)
4205 PCRE_SPTR32 subject; (32-bit version)
4206 int subject_length;
4207 int start_match;
4208 int current_position;
4209 int capture_top;
4210 int capture_last;
4211 void *callout_data;
4212 int pattern_position;
4213 int next_item_length;
4214 const unsigned char *mark; (8-bit version)
4215 const PCRE_UCHAR16 *mark; (16-bit version)
4216 const PCRE_UCHAR32 *mark; (32-bit version)
4217
4218 The version field is an integer containing the version number of the
4219 block format. The initial version was 0; the current version is 2. The
4220 version number will change again in future if additional fields are
4221 added, but the intention is never to remove any of the existing fields.
4222
4223 The callout_number field contains the number of the callout, as com-
4224 piled into the pattern (that is, the number after ?C for manual call-
4225 outs, and 255 for automatically generated callouts).
4226
4227 The offset_vector field is a pointer to the vector of offsets that was
4228 passed by the caller to the matching function. When pcre_exec() or
4229 pcre[16|32]_exec() is used, the contents can be inspected, in order to
4230 extract substrings that have been matched so far, in the same way as
4231 for extracting substrings after a match has completed. For the DFA
4232 matching functions, this field is not useful.
4233
4234 The subject and subject_length fields contain copies of the values that
4235 were passed to the matching function.
4236
4237 The start_match field normally contains the offset within the subject
4238 at which the current match attempt started. However, if the escape
4239 sequence \K has been encountered, this value is changed to reflect the
4240 modified starting point. If the pattern is not anchored, the callout
4241 function may be called several times from the same point in the pattern
4242 for different starting points in the subject.
4243
4244 The current_position field contains the offset within the subject of
4245 the current match pointer.
4246
4247 When the pcre_exec() or pcre[16|32]_exec() is used, the capture_top
4248 field contains one more than the number of the highest numbered cap-
4249 tured substring so far. If no substrings have been captured, the value
4250 of capture_top is one. This is always the case when the DFA functions
4251 are used, because they do not support captured substrings.
4252
4253 The capture_last field contains the number of the most recently cap-
4254 tured substring. If no substrings have been captured, its value is -1.
4255 This is always the case for the DFA matching functions.
4256
4257 The callout_data field contains a value that is passed to a matching
4258 function specifically so that it can be passed back in callouts. It is
4259 passed in the callout_data field of a pcre_extra or pcre[16|32]_extra
4260 data structure. If no such data was passed, the value of callout_data
4261 in a callout block is NULL. There is a description of the pcre_extra
4262 structure in the pcreapi documentation.
4263
4264 The pattern_position field is present from version 1 of the callout
4265 structure. It contains the offset to the next item to be matched in the
4266 pattern string.
4267
4268 The next_item_length field is present from version 1 of the callout
4269 structure. It contains the length of the next item to be matched in the
4270 pattern string. When the callout immediately precedes an alternation
4271 bar, a closing parenthesis, or the end of the pattern, the length is
4272 zero. When the callout precedes an opening parenthesis, the length is
4273 that of the entire subpattern.
4274
4275 The pattern_position and next_item_length fields are intended to help
4276 in distinguishing between different automatic callouts, which all have
4277 the same callout number. However, they are set for all callouts.
4278
4279 The mark field is present from version 2 of the callout structure. In
4280 callouts from pcre_exec() or pcre[16|32]_exec() it contains a pointer
4281 to the zero-terminated name of the most recently passed (*MARK),
4282 (*PRUNE), or (*THEN) item in the match, or NULL if no such items have
4283 been passed. Instances of (*PRUNE) or (*THEN) without a name do not
4284 obliterate a previous (*MARK). In callouts from the DFA matching func-
4285 tions this field always contains NULL.
4286
4287
4288 RETURN VALUES
4289
4290 The external callout function returns an integer to PCRE. If the value
4291 is zero, matching proceeds as normal. If the value is greater than
4292 zero, matching fails at the current point, but the testing of other
4293 matching possibilities goes ahead, just as if a lookahead assertion had
4294 failed. If the value is less than zero, the match is abandoned, the
4295 matching function returns the negative value.
4296
4297 Negative values should normally be chosen from the set of
4298 PCRE_ERROR_xxx values. In particular, PCRE_ERROR_NOMATCH forces a stan-
4299 dard "no match" failure. The error number PCRE_ERROR_CALLOUT is
4300 reserved for use by callout functions; it will never be used by PCRE
4301 itself.
4302
4303
4304 AUTHOR
4305
4306 Philip Hazel
4307 University Computing Service
4308 Cambridge CB2 3QH, England.
4309
4310
4311 REVISION
4312
4313 Last updated: 24 June 2012
4314 Copyright (c) 1997-2012 University of Cambridge.
4315 ------------------------------------------------------------------------------
4316
4317
4318 PCRECOMPAT(3) PCRECOMPAT(3)
4319
4320
4321 NAME
4322 PCRE - Perl-compatible regular expressions
4323
4324
4325 DIFFERENCES BETWEEN PCRE AND PERL
4326
4327 This document describes the differences in the ways that PCRE and Perl
4328 handle regular expressions. The differences described here are with
4329 respect to Perl versions 5.10 and above.
4330
4331 1. PCRE has only a subset of Perl's Unicode support. Details of what it
4332 does have are given in the pcreunicode page.
4333
4334 2. PCRE allows repeat quantifiers only on parenthesized assertions, but
4335 they do not mean what you might think. For example, (?!a){3} does not
4336 assert that the next three characters are not "a". It just asserts that
4337 the next character is not "a" three times (in principle: PCRE optimizes
4338 this to run the assertion just once). Perl allows repeat quantifiers on
4339 other assertions such as \b, but these do not seem to have any use.
4340
4341 3. Capturing subpatterns that occur inside negative lookahead asser-
4342 tions are counted, but their entries in the offsets vector are never
4343 set. Perl sets its numerical variables from any such patterns that are
4344 matched before the assertion fails to match something (thereby succeed-
4345 ing), but only if the negative lookahead assertion contains just one
4346 branch.
4347
4348 4. Though binary zero characters are supported in the subject string,
4349 they are not allowed in a pattern string because it is passed as a nor-
4350 mal C string, terminated by zero. The escape sequence \0 can be used in
4351 the pattern to represent a binary zero.
4352
4353 5. The following Perl escape sequences are not supported: \l, \u, \L,
4354 \U, and \N when followed by a character name or Unicode value. (\N on
4355 its own, matching a non-newline character, is supported.) In fact these
4356 are implemented by Perl's general string-handling and are not part of
4357 its pattern matching engine. If any of these are encountered by PCRE,
4358 an error is generated by default. However, if the PCRE_JAVASCRIPT_COM-
4359 PAT option is set, \U and \u are interpreted as JavaScript interprets
4360 them.
4361
4362 6. The Perl escape sequences \p, \P, and \X are supported only if PCRE
4363 is built with Unicode character property support. The properties that
4364 can be tested with \p and \P are limited to the general category prop-
4365 erties such as Lu and Nd, script names such as Greek or Han, and the
4366 derived properties Any and L&. PCRE does support the Cs (surrogate)
4367 property, which Perl does not; the Perl documentation says "Because
4368 Perl hides the need for the user to understand the internal representa-
4369 tion of Unicode characters, there is no need to implement the somewhat
4370 messy concept of surrogates."
4371
4372 7. PCRE does support the \Q...\E escape for quoting substrings. Charac-
4373 ters in between are treated as literals. This is slightly different
4374 from Perl in that $ and @ are also handled as literals inside the
4375 quotes. In Perl, they cause variable interpolation (but of course PCRE
4376 does not have variables). Note the following examples:
4377
4378 Pattern PCRE matches Perl matches
4379
4380 \Qabc$xyz\E abc$xyz abc followed by the
4381 contents of $xyz
4382 \Qabc\$xyz\E abc\$xyz abc\$xyz
4383 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
4384
4385 The \Q...\E sequence is recognized both inside and outside character
4386 classes.
4387
4388 8. Fairly obviously, PCRE does not support the (?{code}) and (??{code})
4389 constructions. However, there is support for recursive patterns. This
4390 is not available in Perl 5.8, but it is in Perl 5.10. Also, the PCRE
4391 "callout" feature allows an external function to be called during pat-
4392 tern matching. See the pcrecallout documentation for details.
4393
4394 9. Subpatterns that are called as subroutines (whether or not recur-
4395 sively) are always treated as atomic groups in PCRE. This is like
4396 Python, but unlike Perl. Captured values that are set outside a sub-
4397 routine call can be reference from inside in PCRE, but not in Perl.
4398 There is a discussion that explains these differences in more detail in
4399 the section on recursion differences from Perl in the pcrepattern page.
4400
4401 10. If any of the backtracking control verbs are used in an assertion
4402 or in a subpattern that is called as a subroutine (whether or not
4403 recursively), their effect is confined to that subpattern; it does not
4404 extend to the surrounding pattern. This is not always the case in Perl.
4405 In particular, if (*THEN) is present in a group that is called as a
4406 subroutine, its action is limited to that group, even if the group does
4407 not contain any | characters. There is one exception to this: the name
4408 from a *(MARK), (*PRUNE), or (*THEN) that is encountered in a success-
4409 ful positive assertion is passed back when a match succeeds (compare
4410 capturing parentheses in assertions). Note that such subpatterns are
4411 processed as anchored at the point where they are tested.
4412
4413 11. There are some differences that are concerned with the settings of
4414 captured strings when part of a pattern is repeated. For example,
4415 matching "aba" against the pattern /^(a(b)?)+$/ in Perl leaves $2
4416 unset, but in PCRE it is set to "b".
4417
4418 12. PCRE's handling of duplicate subpattern numbers and duplicate sub-
4419 pattern names is not as general as Perl's. This is a consequence of the
4420 fact the PCRE works internally just with numbers, using an external ta-
4421 ble to translate between numbers and names. In particular, a pattern
4422 such as (?|(?<a>A)|(?<b)B), where the two capturing parentheses have
4423 the same number but different names, is not supported, and causes an
4424 error at compile time. If it were allowed, it would not be possible to
4425 distinguish which parentheses matched, because both names map to cap-
4426 turing subpattern number 1. To avoid this confusing situation, an error
4427 is given at compile time.
4428
4429 13. Perl recognizes comments in some places that PCRE does not, for
4430 example, between the ( and ? at the start of a subpattern. If the /x
4431 modifier is set, Perl allows white space between ( and ? but PCRE never
4432 does, even if the PCRE_EXTENDED option is set.
4433
4434 14. PCRE provides some extensions to the Perl regular expression facil-
4435 ities. Perl 5.10 includes new features that are not in earlier ver-
4436 sions of Perl, some of which (such as named parentheses) have been in
4437 PCRE for some time. This list is with respect to Perl 5.10:
4438
4439 (a) Although lookbehind assertions in PCRE must match fixed length
4440 strings, each alternative branch of a lookbehind assertion can match a
4441 different length of string. Perl requires them all to have the same
4442 length.
4443
4444 (b) If PCRE_DOLLAR_ENDONLY is set and PCRE_MULTILINE is not set, the $
4445 meta-character matches only at the very end of the string.
4446
4447 (c) If PCRE_EXTRA is set, a backslash followed by a letter with no spe-
4448 cial meaning is faulted. Otherwise, like Perl, the backslash is quietly
4449 ignored. (Perl can be made to issue a warning.)
4450
4451 (d) If PCRE_UNGREEDY is set, the greediness of the repetition quanti-
4452 fiers is inverted, that is, by default they are not greedy, but if fol-
4453 lowed by a question mark they are.
4454
4455 (e) PCRE_ANCHORED can be used at matching time to force a pattern to be
4456 tried only at the first matching position in the subject string.
4457
4458 (f) The PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
4459 and PCRE_NO_AUTO_CAPTURE options for pcre_exec() have no Perl equiva-
4460 lents.
4461
4462 (g) The \R escape sequence can be restricted to match only CR, LF, or
4463 CRLF by the PCRE_BSR_ANYCRLF option.
4464
4465 (h) The callout facility is PCRE-specific.
4466
4467 (i) The partial matching facility is PCRE-specific.
4468
4469 (j) Patterns compiled by PCRE can be saved and re-used at a later time,
4470 even on different hosts that have the other endianness. However, this
4471 does not apply to optimized data created by the just-in-time compiler.
4472
4473 (k) The alternative matching functions (pcre_dfa_exec(),
4474 pcre16_dfa_exec() and pcre32_dfa_exec(),) match in a different way and
4475 are not Perl-compatible.
4476
4477 (l) PCRE recognizes some special sequences such as (*CR) at the start
4478 of a pattern that set overall options that cannot be changed within the
4479 pattern.
4480
4481
4482 AUTHOR
4483
4484 Philip Hazel
4485 University Computing Service
4486 Cambridge CB2 3QH, England.
4487
4488
4489 REVISION
4490
4491 Last updated: 25 August 2012
4492 Copyright (c) 1997-2012 University of Cambridge.
4493 ------------------------------------------------------------------------------
4494
4495
4496 PCREPATTERN(3) PCREPATTERN(3)
4497
4498
4499 NAME
4500 PCRE - Perl-compatible regular expressions
4501
4502
4503 PCRE REGULAR EXPRESSION DETAILS
4504
4505 The syntax and semantics of the regular expressions that are supported
4506 by PCRE are described in detail below. There is a quick-reference syn-
4507 tax summary in the pcresyntax page. PCRE tries to match Perl syntax and
4508 semantics as closely as it can. PCRE also supports some alternative
4509 regular expression syntax (which does not conflict with the Perl syn-
4510 tax) in order to provide some compatibility with regular expressions in
4511 Python, .NET, and Oniguruma.
4512
4513 Perl's regular expressions are described in its own documentation, and
4514 regular expressions in general are covered in a number of books, some
4515 of which have copious examples. Jeffrey Friedl's "Mastering Regular
4516 Expressions", published by O'Reilly, covers regular expressions in
4517 great detail. This description of PCRE's regular expressions is
4518 intended as reference material.
4519
4520 The original operation of PCRE was on strings of one-byte characters.
4521 However, there is now also support for UTF-8 strings in the original
4522 library, an extra library that supports 16-bit and UTF-16 character
4523 strings, and an extra library that supports 32-bit and UTF-32 character
4524 strings. To use these features, PCRE must be built to include appropri-
4525 ate support. When using UTF strings you must either call the compiling
4526 function with the PCRE_UTF8, PCRE_UTF16 or PCRE_UTF32 option, or the
4527 pattern must start with one of these special sequences:
4528
4529 (*UTF8)
4530 (*UTF16)
4531 (*UTF32)
4532
4533 Starting a pattern with such a sequence is equivalent to setting the
4534 relevant option. This feature is not Perl-compatible. How setting a UTF
4535 mode affects pattern matching is mentioned in several places below.
4536 There is also a summary of features in the pcreunicode page.
4537
4538 Another special sequence that may appear at the start of a pattern or
4539 in combination with (*UTF8) or (*UTF16) or (*UTF32) is:
4540
4541 (*UCP)
4542
4543 This has the same effect as setting the PCRE_UCP option: it causes
4544 sequences such as \d and \w to use Unicode properties to determine
4545 character types, instead of recognizing only characters with codes less
4546 than 128 via a lookup table.
4547
4548 If a pattern starts with (*NO_START_OPT), it has the same effect as
4549 setting the PCRE_NO_START_OPTIMIZE option either at compile or matching
4550 time. There are also some more of these special sequences that are con-
4551 cerned with the handling of newlines; they are described below.
4552
4553 The remainder of this document discusses the patterns that are sup-
4554 ported by PCRE when one its main matching functions, pcre_exec()
4555 (8-bit) or pcre[16|32]_exec() (16- or 32-bit), is used. PCRE also has
4556 alternative matching functions, pcre_dfa_exec() and
4557 pcre[16|32_dfa_exec(), which match using a different algorithm that is
4558 not Perl-compatible. Some of the features discussed below are not
4559 available when DFA matching is used. The advantages and disadvantages
4560 of the alternative functions, and how they differ from the normal func-
4561 tions, are discussed in the pcrematching page.
4562
4563
4564 EBCDIC CHARACTER CODES
4565
4566 PCRE can be compiled to run in an environment that uses EBCDIC as its
4567 character code rather than ASCII or Unicode (typically a mainframe sys-
4568 tem). In the sections below, character code values are ASCII or Uni-
4569 code; in an EBCDIC environment these characters may have different code
4570 values, and there are no code points greater than 255.
4571
4572
4573 NEWLINE CONVENTIONS
4574
4575 PCRE supports five different conventions for indicating line breaks in
4576 strings: a single CR (carriage return) character, a single LF (line-
4577 feed) character, the two-character sequence CRLF, any of the three pre-
4578 ceding, or any Unicode newline sequence. The pcreapi page has further
4579 discussion about newlines, and shows how to set the newline convention
4580 in the options arguments for the compiling and matching functions.
4581
4582 It is also possible to specify a newline convention by starting a pat-
4583 tern string with one of the following five sequences:
4584
4585 (*CR) carriage return
4586 (*LF) linefeed
4587 (*CRLF) carriage return, followed by linefeed
4588 (*ANYCRLF) any of the three above
4589 (*ANY) all Unicode newline sequences
4590
4591 These override the default and the options given to the compiling func-
4592 tion. For example, on a Unix system where LF is the default newline
4593 sequence, the pattern
4594
4595 (*CR)a.b
4596
4597 changes the convention to CR. That pattern matches "a\nb" because LF is
4598 no longer a newline. Note that these special settings, which are not
4599 Perl-compatible, are recognized only at the very start of a pattern,
4600 and that they must be in upper case. If more than one of them is
4601 present, the last one is used.
4602
4603 The newline convention affects the interpretation of the dot metachar-
4604 acter when PCRE_DOTALL is not set, and also the behaviour of \N. How-
4605 ever, it does not affect what the \R escape sequence matches. By
4606 default, this is any Unicode newline sequence, for Perl compatibility.
4607 However, this can be changed; see the description of \R in the section
4608 entitled "Newline sequences" below. A change of \R setting can be com-
4609 bined with a change of newline convention.
4610
4611
4612 CHARACTERS AND METACHARACTERS
4613
4614 A regular expression is a pattern that is matched against a subject
4615 string from left to right. Most characters stand for themselves in a
4616 pattern, and match the corresponding characters in the subject. As a
4617 trivial example, the pattern
4618
4619 The quick brown fox
4620
4621 matches a portion of a subject string that is identical to itself. When
4622 caseless matching is specified (the PCRE_CASELESS option), letters are
4623 matched independently of case. In a UTF mode, PCRE always understands
4624 the concept of case for characters whose values are less than 128, so
4625 caseless matching is always possible. For characters with higher val-
4626 ues, the concept of case is supported if PCRE is compiled with Unicode
4627 property support, but not otherwise. If you want to use caseless
4628 matching for characters 128 and above, you must ensure that PCRE is
4629 compiled with Unicode property support as well as with UTF support.
4630
4631 The power of regular expressions comes from the ability to include
4632 alternatives and repetitions in the pattern. These are encoded in the
4633 pattern by the use of metacharacters, which do not stand for themselves
4634 but instead are interpreted in some special way.
4635
4636 There are two different sets of metacharacters: those that are recog-
4637 nized anywhere in the pattern except within square brackets, and those
4638 that are recognized within square brackets. Outside square brackets,
4639 the metacharacters are as follows:
4640
4641 \ general escape character with several uses
4642 ^ assert start of string (or line, in multiline mode)
4643 $ assert end of string (or line, in multiline mode)
4644 . match any character except newline (by default)
4645 [ start character class definition
4646 | start of alternative branch
4647 ( start subpattern
4648 ) end subpattern
4649 ? extends the meaning of (
4650 also 0 or 1 quantifier
4651 also quantifier minimizer
4652 * 0 or more quantifier
4653 + 1 or more quantifier
4654 also "possessive quantifier"
4655 { start min/max quantifier
4656
4657 Part of a pattern that is in square brackets is called a "character
4658 class". In a character class the only metacharacters are:
4659
4660 \ general escape character
4661 ^ negate the class, but only if the first character
4662 - indicates character range
4663 [ POSIX character class (only if followed by POSIX
4664 syntax)
4665 ] terminates the character class
4666
4667 The following sections describe the use of each of the metacharacters.
4668
4669
4670 BACKSLASH
4671
4672 The backslash character has several uses. Firstly, if it is followed by
4673 a character that is not a number or a letter, it takes away any special
4674 meaning that character may have. This use of backslash as an escape
4675 character applies both inside and outside character classes.
4676
4677 For example, if you want to match a * character, you write \* in the
4678 pattern. This escaping action applies whether or not the following
4679 character would otherwise be interpreted as a metacharacter, so it is
4680 always safe to precede a non-alphanumeric with backslash to specify
4681 that it stands for itself. In particular, if you want to match a back-
4682 slash, you write \\.
4683
4684 In a UTF mode, only ASCII numbers and letters have any special meaning
4685 after a backslash. All other characters (in particular, those whose
4686 codepoints are greater than 127) are treated as literals.
4687
4688 If a pattern is compiled with the PCRE_EXTENDED option, white space in
4689 the pattern (other than in a character class) and characters between a
4690 # outside a character class and the next newline are ignored. An escap-
4691 ing backslash can be used to include a white space or # character as
4692 part of the pattern.
4693
4694 If you want to remove the special meaning from a sequence of charac-
4695 ters, you can do so by putting them between \Q and \E. This is differ-
4696 ent from Perl in that $ and @ are handled as literals in \Q...\E
4697 sequences in PCRE, whereas in Perl, $ and @ cause variable interpola-
4698 tion. Note the following examples:
4699
4700 Pattern PCRE matches Perl matches
4701
4702 \Qabc$xyz\E abc$xyz abc followed by the
4703 contents of $xyz
4704 \Qabc\$xyz\E abc\$xyz abc\$xyz
4705 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
4706
4707 The \Q...\E sequence is recognized both inside and outside character
4708 classes. An isolated \E that is not preceded by \Q is ignored. If \Q
4709 is not followed by \E later in the pattern, the literal interpretation
4710 continues to the end of the pattern (that is, \E is assumed at the
4711 end). If the isolated \Q is inside a character class, this causes an
4712 error, because the character class is not terminated.
4713
4714 Non-printing characters
4715
4716 A second use of backslash provides a way of encoding non-printing char-
4717 acters in patterns in a visible manner. There is no restriction on the
4718 appearance of non-printing characters, apart from the binary zero that
4719 terminates a pattern, but when a pattern is being prepared by text
4720 editing, it is often easier to use one of the following escape
4721 sequences than the binary character it represents:
4722
4723 \a alarm, that is, the BEL character (hex 07)
4724 \cx "control-x", where x is any ASCII character
4725 \e escape (hex 1B)
4726 \f form feed (hex 0C)
4727 \n linefeed (hex 0A)
4728 \r carriage return (hex 0D)
4729 \t tab (hex 09)
4730 \ddd character with octal code ddd, or back reference
4731 \xhh character with hex code hh
4732 \x{hhh..} character with hex code hhh.. (non-JavaScript mode)
4733 \uhhhh character with hex code hhhh (JavaScript mode only)
4734
4735 The precise effect of \cx on ASCII characters is as follows: if x is a
4736 lower case letter, it is converted to upper case. Then bit 6 of the
4737 character (hex 40) is inverted. Thus \cA to \cZ become hex 01 to hex 1A
4738 (A is 41, Z is 5A), but \c{ becomes hex 3B ({ is 7B), and \c; becomes
4739 hex 7B (; is 3B). If the data item (byte or 16-bit value) following \c
4740 has a value greater than 127, a compile-time error occurs. This locks
4741 out non-ASCII characters in all modes.
4742
4743 The \c facility was designed for use with ASCII characters, but with
4744 the extension to Unicode it is even less useful than it once was. It
4745 is, however, recognized when PCRE is compiled in EBCDIC mode, where
4746 data items are always bytes. In this mode, all values are valid after
4747 \c. If the next character is a lower case letter, it is converted to
4748 upper case. Then the 0xc0 bits of the byte are inverted. Thus \cA
4749 becomes hex 01, as in ASCII (A is C1), but because the EBCDIC letters
4750 are disjoint, \cZ becomes hex 29 (Z is E9), and other characters also
4751 generate different values.
4752
4753 By default, after \x, from zero to two hexadecimal digits are read
4754 (letters can be in upper or lower case). Any number of hexadecimal dig-
4755 its may appear between \x{ and }, but the character code is constrained
4756 as follows:
4757
4758 8-bit non-UTF mode less than 0x100
4759 8-bit UTF-8 mode less than 0x10ffff and a valid codepoint
4760 16-bit non-UTF mode less than 0x10000
4761 16-bit UTF-16 mode less than 0x10ffff and a valid codepoint
4762 32-bit non-UTF mode less than 0x80000000
4763 32-bit UTF-32 mode less than 0x10ffff and a valid codepoint
4764
4765 Invalid Unicode codepoints are the range 0xd800 to 0xdfff (the so-
4766 called "surrogate" codepoints), and 0xffef.
4767
4768 If characters other than hexadecimal digits appear between \x{ and },
4769 or if there is no terminating }, this form of escape is not recognized.
4770 Instead, the initial \x will be interpreted as a basic hexadecimal
4771 escape, with no following digits, giving a character whose value is
4772 zero.
4773
4774 If the PCRE_JAVASCRIPT_COMPAT option is set, the interpretation of \x
4775 is as just described only when it is followed by two hexadecimal dig-
4776 its. Otherwise, it matches a literal "x" character. In JavaScript
4777 mode, support for code points greater than 256 is provided by \u, which
4778 must be followed by four hexadecimal digits; otherwise it matches a
4779 literal "u" character. Character codes specified by \u in JavaScript
4780 mode are constrained in the same was as those specified by \x in non-
4781 JavaScript mode.
4782
4783 Characters whose value is less than 256 can be defined by either of the
4784 two syntaxes for \x (or by \u in JavaScript mode). There is no differ-
4785 ence in the way they are handled. For example, \xdc is exactly the same
4786 as \x{dc} (or \u00dc in JavaScript mode).
4787
4788 After \0 up to two further octal digits are read. If there are fewer
4789 than two digits, just those that are present are used. Thus the
4790 sequence \0\x\07 specifies two binary zeros followed by a BEL character
4791 (code value 7). Make sure you supply two digits after the initial zero
4792 if the pattern character that follows is itself an octal digit.
4793
4794 The handling of a backslash followed by a digit other than 0 is compli-
4795 cated. Outside a character class, PCRE reads it and any following dig-
4796 its as a decimal number. If the number is less than 10, or if there
4797 have been at least that many previous capturing left parentheses in the
4798 expression, the entire sequence is taken as a back reference. A
4799 description of how this works is given later, following the discussion
4800 of parenthesized subpatterns.
4801
4802 Inside a character class, or if the decimal number is greater than 9
4803 and there have not been that many capturing subpatterns, PCRE re-reads
4804 up to three octal digits following the backslash, and uses them to gen-
4805 erate a data character. Any subsequent digits stand for themselves. The
4806 value of the character is constrained in the same way as characters
4807 specified in hexadecimal. For example:
4808
4809 \040 is another way of writing an ASCII space
4810 \40 is the same, provided there are fewer than 40
4811 previous capturing subpatterns
4812 \7 is always a back reference
4813 \11 might be a back reference, or another way of
4814 writing a tab
4815 \011 is always a tab
4816 \0113 is a tab followed by the character "3"
4817 \113 might be a back reference, otherwise the
4818 character with octal code 113
4819 \377 might be a back reference, otherwise
4820 the value 255 (decimal)
4821 \81 is either a back reference, or a binary zero
4822 followed by the two characters "8" and "1"
4823
4824 Note that octal values of 100 or greater must not be introduced by a
4825 leading zero, because no more than three octal digits are ever read.
4826
4827 All the sequences that define a single character value can be used both
4828 inside and outside character classes. In addition, inside a character
4829 class, \b is interpreted as the backspace character (hex 08).
4830
4831 \N is not allowed in a character class. \B, \R, and \X are not special
4832 inside a character class. Like other unrecognized escape sequences,
4833 they are treated as the literal characters "B", "R", and "X" by
4834 default, but cause an error if the PCRE_EXTRA option is set. Outside a
4835 character class, these sequences have different meanings.
4836
4837 Unsupported escape sequences
4838
4839 In Perl, the sequences \l, \L, \u, and \U are recognized by its string
4840 handler and used to modify the case of following characters. By
4841 default, PCRE does not support these escape sequences. However, if the
4842 PCRE_JAVASCRIPT_COMPAT option is set, \U matches a "U" character, and
4843 \u can be used to define a character by code point, as described in the
4844 previous section.
4845
4846 Absolute and relative back references
4847
4848 The sequence \g followed by an unsigned or a negative number, option-
4849 ally enclosed in braces, is an absolute or relative back reference. A
4850 named back reference can be coded as \g{name}. Back references are dis-
4851 cussed later, following the discussion of parenthesized subpatterns.
4852
4853 Absolute and relative subroutine calls
4854
4855 For compatibility with Oniguruma, the non-Perl syntax \g followed by a
4856 name or a number enclosed either in angle brackets or single quotes, is
4857 an alternative syntax for referencing a subpattern as a "subroutine".
4858 Details are discussed later. Note that \g{...} (Perl syntax) and
4859 \g<...> (Oniguruma syntax) are not synonymous. The former is a back
4860 reference; the latter is a subroutine call.
4861
4862 Generic character types
4863
4864 Another use of backslash is for specifying generic character types:
4865
4866 \d any decimal digit
4867 \D any character that is not a decimal digit
4868 \h any horizontal white space character
4869 \H any character that is not a horizontal white space character
4870 \s any white space character
4871 \S any character that is not a white space character
4872 \v any vertical white space character
4873 \V any character that is not a vertical white space character
4874 \w any "word" character
4875 \W any "non-word" character
4876
4877 There is also the single sequence \N, which matches a non-newline char-
4878 acter. This is the same as the "." metacharacter when PCRE_DOTALL is
4879 not set. Perl also uses \N to match characters by name; PCRE does not
4880 support this.
4881
4882 Each pair of lower and upper case escape sequences partitions the com-
4883 plete set of characters into two disjoint sets. Any given character
4884 matches one, and only one, of each pair. The sequences can appear both
4885 inside and outside character classes. They each match one character of
4886 the appropriate type. If the current matching point is at the end of
4887 the subject string, all of them fail, because there is no character to
4888 match.
4889
4890 For compatibility with Perl, \s does not match the VT character (code
4891 11). This makes it different from the the POSIX "space" class. The \s
4892 characters are HT (9), LF (10), FF (12), CR (13), and space (32). If
4893 "use locale;" is included in a Perl script, \s may match the VT charac-
4894 ter. In PCRE, it never does.
4895
4896 A "word" character is an underscore or any character that is a letter
4897 or digit. By default, the definition of letters and digits is con-
4898 trolled by PCRE's low-valued character tables, and may vary if locale-
4899 specific matching is taking place (see "Locale support" in the pcreapi
4900 page). For example, in a French locale such as "fr_FR" in Unix-like
4901 systems, or "french" in Windows, some character codes greater than 128
4902 are used for accented letters, and these are then matched by \w. The
4903 use of locales with Unicode is discouraged.
4904
4905 By default, in a UTF mode, characters with values greater than 128
4906 never match \d, \s, or \w, and always match \D, \S, and \W. These
4907 sequences retain their original meanings from before UTF support was
4908 available, mainly for efficiency reasons. However, if PCRE is compiled
4909 with Unicode property support, and the PCRE_UCP option is set, the be-
4910 haviour is changed so that Unicode properties are used to determine
4911 character types, as follows:
4912
4913 \d any character that \p{Nd} matches (decimal digit)
4914 \s any character that \p{Z} matches, plus HT, LF, FF, CR
4915 \w any character that \p{L} or \p{N} matches, plus underscore
4916
4917 The upper case escapes match the inverse sets of characters. Note that
4918 \d matches only decimal digits, whereas \w matches any Unicode digit,
4919 as well as any Unicode letter, and underscore. Note also that PCRE_UCP
4920 affects \b, and \B because they are defined in terms of \w and \W.
4921 Matching these sequences is noticeably slower when PCRE_UCP is set.
4922
4923 The sequences \h, \H, \v, and \V are features that were added to Perl
4924 at release 5.10. In contrast to the other sequences, which match only
4925 ASCII characters by default, these always match certain high-valued
4926 codepoints, whether or not PCRE_UCP is set. The horizontal space char-
4927 acters are:
4928
4929 U+0009 Horizontal tab (HT)
4930 U+0020 Space
4931 U+00A0 Non-break space
4932 U+1680 Ogham space mark
4933 U+180E Mongolian vowel separator
4934 U+2000 En quad
4935 U+2001 Em quad
4936 U+2002 En space
4937 U+2003 Em space
4938 U+2004 Three-per-em space
4939 U+2005 Four-per-em space
4940 U+2006 Six-per-em space
4941 U+2007 Figure space
4942 U+2008 Punctuation space
4943 U+2009 Thin space
4944 U+200A Hair space
4945 U+202F Narrow no-break space
4946 U+205F Medium mathematical space
4947 U+3000 Ideographic space
4948
4949 The vertical space characters are:
4950
4951 U+000A Linefeed (LF)
4952 U+000B Vertical tab (VT)
4953 U+000C Form feed (FF)
4954 U+000D Carriage return (CR)
4955 U+0085 Next line (NEL)
4956 U+2028 Line separator
4957 U+2029 Paragraph separator
4958
4959 In 8-bit, non-UTF-8 mode, only the characters with codepoints less than
4960 256 are relevant.
4961
4962 Newline sequences
4963
4964 Outside a character class, by default, the escape sequence \R matches
4965 any Unicode newline sequence. In 8-bit non-UTF-8 mode \R is equivalent
4966 to the following:
4967
4968 (?>\r\n|\n|\x0b|\f|\r|\x85)
4969
4970 This is an example of an "atomic group", details of which are given
4971 below. This particular group matches either the two-character sequence
4972 CR followed by LF, or one of the single characters LF (linefeed,
4973 U+000A), VT (vertical tab, U+000B), FF (form feed, U+000C), CR (car-
4974 riage return, U+000D), or NEL (next line, U+0085). The two-character
4975 sequence is treated as a single unit that cannot be split.
4976
4977 In other modes, two additional characters whose codepoints are greater
4978 than 255 are added: LS (line separator, U+2028) and PS (paragraph sepa-
4979 rator, U+2029). Unicode character property support is not needed for
4980 these characters to be recognized.
4981
4982 It is possible to restrict \R to match only CR, LF, or CRLF (instead of
4983 the complete set of Unicode line endings) by setting the option
4984 PCRE_BSR_ANYCRLF either at compile time or when the pattern is matched.
4985 (BSR is an abbrevation for "backslash R".) This can be made the default
4986 when PCRE is built; if this is the case, the other behaviour can be
4987 requested via the PCRE_BSR_UNICODE option. It is also possible to
4988 specify these settings by starting a pattern string with one of the
4989 following sequences:
4990
4991 (*BSR_ANYCRLF) CR, LF, or CRLF only
4992 (*BSR_UNICODE) any Unicode newline sequence
4993
4994 These override the default and the options given to the compiling func-
4995 tion, but they can themselves be overridden by options given to a
4996 matching function. Note that these special settings, which are not
4997 Perl-compatible, are recognized only at the very start of a pattern,
4998 and that they must be in upper case. If more than one of them is
4999 present, the last one is used. They can be combined with a change of
5000 newline convention; for example, a pattern can start with:
5001
5002 (*ANY)(*BSR_ANYCRLF)
5003
5004 They can also be combined with the (*UTF8), (*UTF16), (*UTF32) or
5005 (*UCP) special sequences. Inside a character class, \R is treated as an
5006 unrecognized escape sequence, and so matches the letter "R" by default,
5007 but causes an error if PCRE_EXTRA is set.
5008
5009 Unicode character properties
5010
5011 When PCRE is built with Unicode character property support, three addi-
5012 tional escape sequences that match characters with specific properties
5013 are available. When in 8-bit non-UTF-8 mode, these sequences are of
5014 course limited to testing characters whose codepoints are less than
5015 256, but they do work in this mode. The extra escape sequences are:
5016
5017 \p{xx} a character with the xx property
5018 \P{xx} a character without the xx property
5019 \X a Unicode extended grapheme cluster
5020
5021 The property names represented by xx above are limited to the Unicode
5022 script names, the general category properties, "Any", which matches any
5023 character (including newline), and some special PCRE properties
5024 (described in the next section). Other Perl properties such as "InMu-
5025 sicalSymbols" are not currently supported by PCRE. Note that \P{Any}
5026 does not match any characters, so always causes a match failure.
5027
5028 Sets of Unicode characters are defined as belonging to certain scripts.
5029 A character from one of these sets can be matched using a script name.
5030 For example:
5031
5032 \p{Greek}
5033 \P{Han}
5034
5035 Those that are not part of an identified script are lumped together as
5036 "Common". The current list of scripts is:
5037
5038 Arabic, Armenian, Avestan, Balinese, Bamum, Batak, Bengali, Bopomofo,
5039 Brahmi, Braille, Buginese, Buhid, Canadian_Aboriginal, Carian, Chakma,
5040 Cham, Cherokee, Common, Coptic, Cuneiform, Cypriot, Cyrillic, Deseret,
5041 Devanagari, Egyptian_Hieroglyphs, Ethiopic, Georgian, Glagolitic,
5042 Gothic, Greek, Gujarati, Gurmukhi, Han, Hangul, Hanunoo, Hebrew, Hira-
5043 gana, Imperial_Aramaic, Inherited, Inscriptional_Pahlavi, Inscrip-
5044 tional_Parthian, Javanese, Kaithi, Kannada, Katakana, Kayah_Li,
5045 Kharoshthi, Khmer, Lao, Latin, Lepcha, Limbu, Linear_B, Lisu, Lycian,
5046 Lydian, Malayalam, Mandaic, Meetei_Mayek, Meroitic_Cursive,
5047 Meroitic_Hieroglyphs, Miao, Mongolian, Myanmar, New_Tai_Lue, Nko,
5048 Ogham, Old_Italic, Old_Persian, Old_South_Arabian, Old_Turkic,
5049 Ol_Chiki, Oriya, Osmanya, Phags_Pa, Phoenician, Rejang, Runic, Samari-
5050 tan, Saurashtra, Sharada, Shavian, Sinhala, Sora_Sompeng, Sundanese,
5051 Syloti_Nagri, Syriac, Tagalog, Tagbanwa, Tai_Le, Tai_Tham, Tai_Viet,
5052 Takri, Tamil, Telugu, Thaana, Thai, Tibetan, Tifinagh, Ugaritic, Vai,
5053 Yi.
5054
5055 Each character has exactly one Unicode general category property, spec-
5056 ified by a two-letter abbreviation. For compatibility with Perl, nega-
5057 tion can be specified by including a circumflex between the opening
5058 brace and the property name. For example, \p{^Lu} is the same as
5059 \P{Lu}.
5060
5061 If only one letter is specified with \p or \P, it includes all the gen-
5062 eral category properties that start with that letter. In this case, in
5063 the absence of negation, the curly brackets in the escape sequence are
5064 optional; these two examples have the same effect:
5065
5066 \p{L}
5067 \pL
5068
5069 The following general category property codes are supported:
5070
5071 C Other
5072 Cc Control
5073 Cf Format
5074 Cn Unassigned
5075 Co Private use
5076 Cs Surrogate
5077
5078 L Letter
5079 Ll Lower case letter
5080 Lm Modifier letter
5081 Lo Other letter
5082 Lt Title case letter
5083 Lu Upper case letter
5084
5085 M Mark
5086 Mc Spacing mark
5087 Me Enclosing mark
5088 Mn Non-spacing mark
5089
5090 N Number
5091 Nd Decimal number
5092 Nl Letter number
5093 No Other number
5094
5095 P Punctuation
5096 Pc Connector punctuation
5097 Pd Dash punctuation
5098 Pe Close punctuation
5099 Pf Final punctuation
5100 Pi Initial punctuation
5101 Po Other punctuation
5102 Ps Open punctuation
5103
5104 S Symbol
5105 Sc Currency symbol
5106 Sk Modifier symbol
5107 Sm Mathematical symbol
5108 So Other symbol
5109
5110 Z Separator
5111 Zl Line separator
5112 Zp Paragraph separator
5113 Zs Space separator
5114
5115 The special property L& is also supported: it matches a character that
5116 has the Lu, Ll, or Lt property, in other words, a letter that is not
5117 classified as a modifier or "other".
5118
5119 The Cs (Surrogate) property applies only to characters in the range
5120 U+D800 to U+DFFF. Such characters are not valid in Unicode strings and
5121 so cannot be tested by PCRE, unless UTF validity checking has been
5122 turned off (see the discussion of PCRE_NO_UTF8_CHECK,
5123 PCRE_NO_UTF16_CHECK and PCRE_NO_UTF32_CHECK in the pcreapi page). Perl
5124 does not support the Cs property.
5125
5126 The long synonyms for property names that Perl supports (such as
5127 \p{Letter}) are not supported by PCRE, nor is it permitted to prefix
5128 any of these properties with "Is".
5129
5130 No character that is in the Unicode table has the Cn (unassigned) prop-
5131 erty. Instead, this property is assumed for any code point that is not
5132 in the Unicode table.
5133
5134 Specifying caseless matching does not affect these escape sequences.
5135 For example, \p{Lu} always matches only upper case letters.
5136
5137 Matching characters by Unicode property is not fast, because PCRE has
5138 to do a multistage table lookup in order to find a character's prop-
5139 erty. That is why the traditional escape sequences such as \d and \w do
5140 not use Unicode properties in PCRE by default, though you can make them
5141 do so by setting the PCRE_UCP option or by starting the pattern with
5142 (*UCP).
5143
5144 Extended grapheme clusters
5145
5146 The \X escape matches any number of Unicode characters that form an
5147 "extended grapheme cluster", and treats the sequence as an atomic group
5148 (see below). Up to and including release 8.31, PCRE matched an ear-
5149 lier, simpler definition that was equivalent to
5150
5151 (?>\PM\pM*)
5152
5153 That is, it matched a character without the "mark" property, followed
5154 by zero or more characters with the "mark" property. Characters with
5155 the "mark" property are typically non-spacing accents that affect the
5156 preceding character.
5157
5158 This simple definition was extended in Unicode to include more compli-
5159 cated kinds of composite character by giving each character a grapheme
5160 breaking property, and creating rules that use these properties to
5161 define the boundaries of extended grapheme clusters. In releases of
5162 PCRE later than 8.31, \X matches one of these clusters.
5163
5164 \X always matches at least one character. Then it decides whether to
5165 add additional characters according to the following rules for ending a
5166 cluster:
5167
5168 1. End at the end of the subject string.
5169
5170 2. Do not end between CR and LF; otherwise end after any control char-
5171 acter.
5172
5173 3. Do not break Hangul (a Korean script) syllable sequences. Hangul
5174 characters are of five types: L, V, T, LV, and LVT. An L character may
5175 be followed by an L, V, LV, or LVT character; an LV or V character may
5176 be followed by a V or T character; an LVT or T character may be follwed
5177 only by a T character.
5178
5179 4. Do not end before extending characters or spacing marks. Characters
5180 with the "mark" property always have the "extend" grapheme breaking
5181 property.
5182
5183 5. Do not end after prepend characters.
5184
5185 6. Otherwise, end the cluster.
5186
5187 PCRE's additional properties
5188
5189 As well as the standard Unicode properties described above, PCRE sup-
5190 ports four more that make it possible to convert traditional escape
5191 sequences such as \w and \s and POSIX character classes to use Unicode
5192 properties. PCRE uses these non-standard, non-Perl properties inter-
5193 nally when PCRE_UCP is set. They are:
5194
5195 Xan Any alphanumeric character
5196 Xps Any POSIX space character
5197 Xsp Any Perl space character
5198 Xwd Any Perl "word" character
5199
5200 Xan matches characters that have either the L (letter) or the N (num-
5201 ber) property. Xps matches the characters tab, linefeed, vertical tab,
5202 form feed, or carriage return, and any other character that has the Z
5203 (separator) property. Xsp is the same as Xps, except that vertical tab
5204 is excluded. Xwd matches the same characters as Xan, plus underscore.
5205
5206 Resetting the match start
5207
5208 The escape sequence \K causes any previously matched characters not to
5209 be included in the final matched sequence. For example, the pattern:
5210
5211 foo\Kbar
5212
5213 matches "foobar", but reports that it has matched "bar". This feature
5214 is similar to a lookbehind assertion (described below). However, in
5215 this case, the part of the subject before the real match does not have
5216 to be of fixed length, as lookbehind assertions do. The use of \K does
5217 not interfere with the setting of captured substrings. For example,
5218 when the pattern
5219
5220 (foo)\Kbar
5221
5222 matches "foobar", the first substring is still set to "foo".
5223
5224 Perl documents that the use of \K within assertions is "not well
5225 defined". In PCRE, \K is acted upon when it occurs inside positive
5226 assertions, but is ignored in negative assertions.
5227
5228 Simple assertions
5229
5230 The final use of backslash is for certain simple assertions. An asser-
5231 tion specifies a condition that has to be met at a particular point in
5232 a match, without consuming any characters from the subject string. The
5233 use of subpatterns for more complicated assertions is described below.
5234 The backslashed assertions are:
5235
5236 \b matches at a word boundary
5237 \B matches when not at a word boundary
5238 \A matches at the start of the subject
5239 \Z matches at the end of the subject
5240 also matches before a newline at the end of the subject
5241 \z matches only at the end of the subject
5242 \G matches at the first matching position in the subject
5243
5244 Inside a character class, \b has a different meaning; it matches the
5245 backspace character. If any other of these assertions appears in a
5246 character class, by default it matches the corresponding literal char-
5247 acter (for example, \B matches the letter B). However, if the
5248 PCRE_EXTRA option is set, an "invalid escape sequence" error is gener-
5249 ated instead.
5250
5251 A word boundary is a position in the subject string where the current
5252 character and the previous character do not both match \w or \W (i.e.
5253 one matches \w and the other matches \W), or the start or end of the
5254 string if the first or last character matches \w, respectively. In a
5255 UTF mode, the meanings of \w and \W can be changed by setting the
5256 PCRE_UCP option. When this is done, it also affects \b and \B. Neither
5257 PCRE nor Perl has a separate "start of word" or "end of word" metase-
5258 quence. However, whatever follows \b normally determines which it is.
5259 For example, the fragment \ba matches "a" at the start of a word.
5260
5261 The \A, \Z, and \z assertions differ from the traditional circumflex
5262 and dollar (described in the next section) in that they only ever match
5263 at the very start and end of the subject string, whatever options are
5264 set. Thus, they are independent of multiline mode. These three asser-
5265 tions are not affected by the PCRE_NOTBOL or PCRE_NOTEOL options, which
5266 affect only the behaviour of the circumflex and dollar metacharacters.
5267 However, if the startoffset argument of pcre_exec() is non-zero, indi-
5268 cating that matching is to start at a point other than the beginning of
5269 the subject, \A can never match. The difference between \Z and \z is
5270 that \Z matches before a newline at the end of the string as well as at
5271 the very end, whereas \z matches only at the end.
5272
5273 The \G assertion is true only when the current matching position is at
5274 the start point of the match, as specified by the startoffset argument
5275 of pcre_exec(). It differs from \A when the value of startoffset is
5276 non-zero. By calling pcre_exec() multiple times with appropriate argu-
5277 ments, you can mimic Perl's /g option, and it is in this kind of imple-
5278 mentation where \G can be useful.
5279
5280 Note, however, that PCRE's interpretation of \G, as the start of the
5281 current match, is subtly different from Perl's, which defines it as the
5282 end of the previous match. In Perl, these can be different when the
5283 previously matched string was empty. Because PCRE does just one match
5284 at a time, it cannot reproduce this behaviour.
5285
5286 If all the alternatives of a pattern begin with \G, the expression is
5287 anchored to the starting match position, and the "anchored" flag is set
5288 in the compiled regular expression.
5289
5290
5291 CIRCUMFLEX AND DOLLAR
5292
5293 Outside a character class, in the default matching mode, the circumflex
5294 character is an assertion that is true only if the current matching
5295 point is at the start of the subject string. If the startoffset argu-
5296 ment of pcre_exec() is non-zero, circumflex can never match if the
5297 PCRE_MULTILINE option is unset. Inside a character class, circumflex
5298 has an entirely different meaning (see below).
5299
5300 Circumflex need not be the first character of the pattern if a number
5301 of alternatives are involved, but it should be the first thing in each
5302 alternative in which it appears if the pattern is ever to match that
5303 branch. If all possible alternatives start with a circumflex, that is,
5304 if the pattern is constrained to match only at the start of the sub-
5305 ject, it is said to be an "anchored" pattern. (There are also other
5306 constructs that can cause a pattern to be anchored.)
5307
5308 A dollar character is an assertion that is true only if the current
5309 matching point is at the end of the subject string, or immediately
5310 before a newline at the end of the string (by default). Dollar need not
5311 be the last character of the pattern if a number of alternatives are
5312 involved, but it should be the last item in any branch in which it
5313 appears. Dollar has no special meaning in a character class.
5314
5315 The meaning of dollar can be changed so that it matches only at the
5316 very end of the string, by setting the PCRE_DOLLAR_ENDONLY option at
5317 compile time. This does not affect the \Z assertion.
5318
5319 The meanings of the circumflex and dollar characters are changed if the
5320 PCRE_MULTILINE option is set. When this is the case, a circumflex
5321 matches immediately after internal newlines as well as at the start of
5322 the subject string. It does not match after a newline that ends the
5323 string. A dollar matches before any newlines in the string, as well as
5324 at the very end, when PCRE_MULTILINE is set. When newline is specified
5325 as the two-character sequence CRLF, isolated CR and LF characters do
5326 not indicate newlines.
5327
5328 For example, the pattern /^abc$/ matches the subject string "def\nabc"
5329 (where \n represents a newline) in multiline mode, but not otherwise.
5330 Consequently, patterns that are anchored in single line mode because
5331 all branches start with ^ are not anchored in multiline mode, and a
5332 match for circumflex is possible when the startoffset argument of
5333 pcre_exec() is non-zero. The PCRE_DOLLAR_ENDONLY option is ignored if
5334 PCRE_MULTILINE is set.
5335
5336 Note that the sequences \A, \Z, and \z can be used to match the start
5337 and end of the subject in both modes, and if all branches of a pattern
5338 start with \A it is always anchored, whether or not PCRE_MULTILINE is
5339 set.
5340
5341
5342 FULL STOP (PERIOD, DOT) AND \N
5343
5344 Outside a character class, a dot in the pattern matches any one charac-
5345 ter in the subject string except (by default) a character that signi-
5346 fies the end of a line.
5347
5348 When a line ending is defined as a single character, dot never matches
5349 that character; when the two-character sequence CRLF is used, dot does
5350 not match CR if it is immediately followed by LF, but otherwise it
5351 matches all characters (including isolated CRs and LFs). When any Uni-
5352 code line endings are being recognized, dot does not match CR or LF or
5353 any of the other line ending characters.
5354
5355 The behaviour of dot with regard to newlines can be changed. If the
5356 PCRE_DOTALL option is set, a dot matches any one character, without
5357 exception. If the two-character sequence CRLF is present in the subject
5358 string, it takes two dots to match it.
5359
5360 The handling of dot is entirely independent of the handling of circum-
5361 flex and dollar, the only relationship being that they both involve
5362 newlines. Dot has no special meaning in a character class.
5363
5364 The escape sequence \N behaves like a dot, except that it is not
5365 affected by the PCRE_DOTALL option. In other words, it matches any
5366 character except one that signifies the end of a line. Perl also uses
5367 \N to match characters by name; PCRE does not support this.
5368
5369
5370 MATCHING A SINGLE DATA UNIT
5371
5372 Outside a character class, the escape sequence \C matches any one data
5373 unit, whether or not a UTF mode is set. In the 8-bit library, one data
5374 unit is one byte; in the 16-bit library it is a 16-bit unit; in the
5375 32-bit library it is a 32-bit unit. Unlike a dot, \C always matches
5376 line-ending characters. The feature is provided in Perl in order to
5377 match individual bytes in UTF-8 mode, but it is unclear how it can use-
5378 fully be used. Because \C breaks up characters into individual data
5379 units, matching one unit with \C in a UTF mode means that the rest of
5380 the string may start with a malformed UTF character. This has undefined
5381 results, because PCRE assumes that it is dealing with valid UTF strings
5382 (and by default it checks this at the start of processing unless the
5383 PCRE_NO_UTF8_CHECK, PCRE_NO_UTF16_CHECK or PCRE_NO_UTF32_CHECK option
5384 is used).
5385
5386 PCRE does not allow \C to appear in lookbehind assertions (described
5387 below) in a UTF mode, because this would make it impossible to calcu-
5388 late the length of the lookbehind.
5389
5390 In general, the \C escape sequence is best avoided. However, one way of
5391 using it that avoids the problem of malformed UTF characters is to use
5392 a lookahead to check the length of the next character, as in this pat-
5393 tern, which could be used with a UTF-8 string (ignore white space and
5394 line breaks):
5395
5396 (?| (?=[\x00-\x7f])(\C) |
5397 (?=[\x80-\x{7ff}])(\C)(\C) |
5398 (?=[\x{800}-\x{ffff}])(\C)(\C)(\C) |
5399 (?=[\x{10000}-\x{1fffff}])(\C)(\C)(\C)(\C))
5400
5401 A group that starts with (?| resets the capturing parentheses numbers
5402 in each alternative (see "Duplicate Subpattern Numbers" below). The
5403 assertions at the start of each branch check the next UTF-8 character
5404 for values whose encoding uses 1, 2, 3, or 4 bytes, respectively. The
5405 character's individual bytes are then captured by the appropriate num-
5406 ber of groups.
5407
5408
5409 SQUARE BRACKETS AND CHARACTER CLASSES
5410
5411 An opening square bracket introduces a character class, terminated by a
5412 closing square bracket. A closing square bracket on its own is not spe-
5413 cial by default. However, if the PCRE_JAVASCRIPT_COMPAT option is set,
5414 a lone closing square bracket causes a compile-time error. If a closing
5415 square bracket is required as a member of the class, it should be the
5416 first data character in the class (after an initial circumflex, if
5417 present) or escaped with a backslash.
5418
5419 A character class matches a single character in the subject. In a UTF
5420 mode, the character may be more than one data unit long. A matched
5421 character must be in the set of characters defined by the class, unless
5422 the first character in the class definition is a circumflex, in which
5423 case the subject character must not be in the set defined by the class.
5424 If a circumflex is actually required as a member of the class, ensure
5425 it is not the first character, or escape it with a backslash.
5426
5427 For example, the character class [aeiou] matches any lower case vowel,
5428 while [^aeiou] matches any character that is not a lower case vowel.
5429 Note that a circumflex is just a convenient notation for specifying the
5430 characters that are in the class by enumerating those that are not. A
5431 class that starts with a circumflex is not an assertion; it still con-
5432 sumes a character from the subject string, and therefore it fails if
5433 the current pointer is at the end of the string.
5434
5435 In UTF-8 (UTF-16, UTF-32) mode, characters with values greater than 255
5436 (0xffff) can be included in a class as a literal string of data units,
5437 or by using the \x{ escaping mechanism.
5438
5439 When caseless matching is set, any letters in a class represent both
5440 their upper case and lower case versions, so for example, a caseless
5441 [aeiou] matches "A" as well as "a", and a caseless [^aeiou] does not
5442 match "A", whereas a caseful version would. In a UTF mode, PCRE always
5443 understands the concept of case for characters whose values are less
5444 than 128, so caseless matching is always possible. For characters with
5445 higher values, the concept of case is supported if PCRE is compiled
5446 with Unicode property support, but not otherwise. If you want to use
5447 caseless matching in a UTF mode for characters 128 and above, you must
5448 ensure that PCRE is compiled with Unicode property support as well as
5449 with UTF support.
5450
5451 Characters that might indicate line breaks are never treated in any
5452 special way when matching character classes, whatever line-ending
5453 sequence is in use, and whatever setting of the PCRE_DOTALL and
5454 PCRE_MULTILINE options is used. A class such as [^a] always matches one
5455 of these characters.
5456
5457 The minus (hyphen) character can be used to specify a range of charac-
5458 ters in a character class. For example, [d-m] matches any letter
5459 between d and m, inclusive. If a minus character is required in a
5460 class, it must be escaped with a backslash or appear in a position
5461 where it cannot be interpreted as indicating a range, typically as the
5462 first or last character in the class.
5463
5464 It is not possible to have the literal character "]" as the end charac-
5465 ter of a range. A pattern such as [W-]46] is interpreted as a class of
5466 two characters ("W" and "-") followed by a literal string "46]", so it
5467 would match "W46]" or "-46]". However, if the "]" is escaped with a
5468 backslash it is interpreted as the end of range, so [W-\]46] is inter-
5469 preted as a class containing a range followed by two other characters.
5470 The octal or hexadecimal representation of "]" can also be used to end
5471 a range.
5472
5473 Ranges operate in the collating sequence of character values. They can
5474 also be used for characters specified numerically, for example
5475 [\000-\037]. Ranges can include any characters that are valid for the
5476 current mode.
5477
5478 If a range that includes letters is used when caseless matching is set,
5479 it matches the letters in either case. For example, [W-c] is equivalent
5480 to [][\\^_`wxyzabc], matched caselessly, and in a non-UTF mode, if
5481 character tables for a French locale are in use, [\xc8-\xcb] matches
5482 accented E characters in both cases. In UTF modes, PCRE supports the
5483 concept of case for characters with values greater than 128 only when
5484 it is compiled with Unicode property support.
5485
5486 The character escape sequences \d, \D, \h, \H, \p, \P, \s, \S, \v, \V,
5487 \w, and \W may appear in a character class, and add the characters that
5488 they match to the class. For example, [\dABCDEF] matches any hexadeci-
5489 mal digit. In UTF modes, the PCRE_UCP option affects the meanings of
5490 \d, \s, \w and their upper case partners, just as it does when they
5491 appear outside a character class, as described in the section entitled
5492 "Generic character types" above. The escape sequence \b has a different
5493 meaning inside a character class; it matches the backspace character.
5494 The sequences \B, \N, \R, and \X are not special inside a character
5495 class. Like any other unrecognized escape sequences, they are treated
5496 as the literal characters "B", "N", "R", and "X" by default, but cause
5497 an error if the PCRE_EXTRA option is set.
5498
5499 A circumflex can conveniently be used with the upper case character
5500 types to specify a more restricted set of characters than the matching
5501 lower case type. For example, the class [^\W_] matches any letter or
5502 digit, but not underscore, whereas [\w] includes underscore. A positive
5503 character class should be read as "something OR something OR ..." and a
5504 negative class as "NOT something AND NOT something AND NOT ...".
5505
5506 The only metacharacters that are recognized in character classes are
5507 backslash, hyphen (only where it can be interpreted as specifying a
5508 range), circumflex (only at the start), opening square bracket (only
5509 when it can be interpreted as introducing a POSIX class name - see the
5510 next section), and the terminating closing square bracket. However,
5511 escaping other non-alphanumeric characters does no harm.
5512
5513
5514 POSIX CHARACTER CLASSES
5515
5516 Perl supports the POSIX notation for character classes. This uses names
5517 enclosed by [: and :] within the enclosing square brackets. PCRE also
5518 supports this notation. For example,
5519
5520 [01[:alpha:]%]
5521
5522 matches "0", "1", any alphabetic character, or "%". The supported class
5523 names are:
5524
5525 alnum letters and digits
5526 alpha letters
5527 ascii character codes 0 - 127
5528 blank space or tab only
5529 cntrl control characters
5530 digit decimal digits (same as \d)
5531 graph printing characters, excluding space
5532 lower lower case letters
5533 print printing characters, including space
5534 punct printing characters, excluding letters and digits and space
5535 space white space (not quite the same as \s)
5536 upper upper case letters
5537 word "word" characters (same as \w)
5538 xdigit hexadecimal digits
5539
5540 The "space" characters are HT (9), LF (10), VT (11), FF (12), CR (13),
5541 and space (32). Notice that this list includes the VT character (code
5542 11). This makes "space" different to \s, which does not include VT (for
5543 Perl compatibility).
5544
5545 The name "word" is a Perl extension, and "blank" is a GNU extension
5546 from Perl 5.8. Another Perl extension is negation, which is indicated
5547 by a ^ character after the colon. For example,
5548
5549 [12[:^digit:]]
5550
5551 matches "1", "2", or any non-digit. PCRE (and Perl) also recognize the
5552 POSIX syntax [.ch.] and [=ch=] where "ch" is a "collating element", but
5553 these are not supported, and an error is given if they are encountered.
5554
5555 By default, in UTF modes, characters with values greater than 128 do
5556 not match any of the POSIX character classes. However, if the PCRE_UCP
5557 option is passed to pcre_compile(), some of the classes are changed so
5558 that Unicode character properties are used. This is achieved by replac-
5559 ing the POSIX classes by other sequences, as follows:
5560
5561 [:alnum:] becomes \p{Xan}
5562 [:alpha:] becomes \p{L}
5563 [:blank:] becomes \h
5564 [:digit:] becomes \p{Nd}
5565 [:lower:] becomes \p{Ll}
5566 [:space:] becomes \p{Xps}
5567 [:upper:] becomes \p{Lu}
5568 [:word:] becomes \p{Xwd}
5569
5570 Negated versions, such as [:^alpha:] use \P instead of \p. The other
5571 POSIX classes are unchanged, and match only characters with code points
5572 less than 128.
5573
5574
5575 VERTICAL BAR
5576
5577 Vertical bar characters are used to separate alternative patterns. For
5578 example, the pattern
5579
5580 gilbert|sullivan
5581
5582 matches either "gilbert" or "sullivan". Any number of alternatives may
5583 appear, and an empty alternative is permitted (matching the empty
5584 string). The matching process tries each alternative in turn, from left
5585 to right, and the first one that succeeds is used. If the alternatives
5586 are within a subpattern (defined below), "succeeds" means matching the
5587 rest of the main pattern as well as the alternative in the subpattern.
5588
5589
5590 INTERNAL OPTION SETTING
5591
5592 The settings of the PCRE_CASELESS, PCRE_MULTILINE, PCRE_DOTALL, and
5593 PCRE_EXTENDED options (which are Perl-compatible) can be changed from
5594 within the pattern by a sequence of Perl option letters enclosed
5595 between "(?" and ")". The option letters are
5596
5597 i for PCRE_CASELESS
5598 m for PCRE_MULTILINE
5599 s for PCRE_DOTALL
5600 x for PCRE_EXTENDED
5601
5602 For example, (?im) sets caseless, multiline matching. It is also possi-
5603 ble to unset these options by preceding the letter with a hyphen, and a
5604 combined setting and unsetting such as (?im-sx), which sets PCRE_CASE-
5605 LESS and PCRE_MULTILINE while unsetting PCRE_DOTALL and PCRE_EXTENDED,
5606 is also permitted. If a letter appears both before and after the
5607 hyphen, the option is unset.
5608
5609 The PCRE-specific options PCRE_DUPNAMES, PCRE_UNGREEDY, and PCRE_EXTRA
5610 can be changed in the same way as the Perl-compatible options by using
5611 the characters J, U and X respectively.
5612
5613 When one of these option changes occurs at top level (that is, not
5614 inside subpattern parentheses), the change applies to the remainder of
5615 the pattern that follows. If the change is placed right at the start of
5616 a pattern, PCRE extracts it into the global options (and it will there-
5617 fore show up in data extracted by the pcre_fullinfo() function).
5618
5619 An option change within a subpattern (see below for a description of
5620 subpatterns) affects only that part of the subpattern that follows it,
5621 so
5622
5623 (a(?i)b)c
5624
5625 matches abc and aBc and no other strings (assuming PCRE_CASELESS is not
5626 used). By this means, options can be made to have different settings
5627 in different parts of the pattern. Any changes made in one alternative
5628 do carry on into subsequent branches within the same subpattern. For
5629 example,
5630
5631 (a(?i)b|c)
5632
5633 matches "ab", "aB", "c", and "C", even though when matching "C" the
5634 first branch is abandoned before the option setting. This is because
5635 the effects of option settings happen at compile time. There would be
5636 some very weird behaviour otherwise.
5637
5638 Note: There are other PCRE-specific options that can be set by the
5639 application when the compiling or matching functions are called. In
5640 some cases the pattern can contain special leading sequences such as
5641 (*CRLF) to override what the application has set or what has been
5642 defaulted. Details are given in the section entitled "Newline
5643 sequences" above. There are also the (*UTF8), (*UTF16),(*UTF32) and
5644 (*UCP) leading sequences that can be used to set UTF and Unicode prop-
5645 erty modes; they are equivalent to setting the PCRE_UTF8, PCRE_UTF16,
5646 PCRE_UTF32 and the PCRE_UCP options, respectively.
5647
5648
5649 SUBPATTERNS
5650
5651 Subpatterns are delimited by parentheses (round brackets), which can be
5652 nested. Turning part of a pattern into a subpattern does two things:
5653
5654 1. It localizes a set of alternatives. For example, the pattern
5655
5656 cat(aract|erpillar|)
5657
5658 matches "cataract", "caterpillar", or "cat". Without the parentheses,
5659 it would match "cataract", "erpillar" or an empty string.
5660
5661 2. It sets up the subpattern as a capturing subpattern. This means
5662 that, when the whole pattern matches, that portion of the subject
5663 string that matched the subpattern is passed back to the caller via the
5664 ovector argument of the matching function. (This applies only to the
5665 traditional matching functions; the DFA matching functions do not sup-
5666 port capturing.)
5667
5668 Opening parentheses are counted from left to right (starting from 1) to
5669 obtain numbers for the capturing subpatterns. For example, if the
5670 string "the red king" is matched against the pattern
5671
5672 the ((red|white) (king|queen))
5673
5674 the captured substrings are "red king", "red", and "king", and are num-
5675 bered 1, 2, and 3, respectively.
5676
5677 The fact that plain parentheses fulfil two functions is not always
5678 helpful. There are often times when a grouping subpattern is required
5679 without a capturing requirement. If an opening parenthesis is followed
5680 by a question mark and a colon, the subpattern does not do any captur-
5681 ing, and is not counted when computing the number of any subsequent
5682 capturing subpatterns. For example, if the string "the white queen" is
5683 matched against the pattern
5684
5685 the ((?:red|white) (king|queen))
5686
5687 the captured substrings are "white queen" and "queen", and are numbered
5688 1 and 2. The maximum number of capturing subpatterns is 65535.
5689
5690 As a convenient shorthand, if any option settings are required at the
5691 start of a non-capturing subpattern, the option letters may appear
5692 between the "?" and the ":". Thus the two patterns
5693
5694 (?i:saturday|sunday)
5695 (?:(?i)saturday|sunday)
5696
5697 match exactly the same set of strings. Because alternative branches are
5698 tried from left to right, and options are not reset until the end of
5699 the subpattern is reached, an option setting in one branch does affect
5700 subsequent branches, so the above patterns match "SUNDAY" as well as
5701 "Saturday".
5702
5703
5704 DUPLICATE SUBPATTERN NUMBERS
5705
5706 Perl 5.10 introduced a feature whereby each alternative in a subpattern
5707 uses the same numbers for its capturing parentheses. Such a subpattern
5708 starts with (?| and is itself a non-capturing subpattern. For example,
5709 consider this pattern:
5710
5711 (?|(Sat)ur|(Sun))day
5712
5713 Because the two alternatives are inside a (?| group, both sets of cap-
5714 turing parentheses are numbered one. Thus, when the pattern matches,
5715 you can look at captured substring number one, whichever alternative
5716 matched. This construct is useful when you want to capture part, but
5717 not all, of one of a number of alternatives. Inside a (?| group, paren-
5718 theses are numbered as usual, but the number is reset at the start of
5719 each branch. The numbers of any capturing parentheses that follow the
5720 subpattern start after the highest number used in any branch. The fol-
5721 lowing example is taken from the Perl documentation. The numbers under-
5722 neath show in which buffer the captured content will be stored.
5723
5724 # before ---------------branch-reset----------- after
5725 / ( a ) (?| x ( y ) z | (p (q) r) | (t) u (v) ) ( z ) /x
5726 # 1 2 2 3 2 3 4
5727
5728 A back reference to a numbered subpattern uses the most recent value
5729 that is set for that number by any subpattern. The following pattern
5730 matches "abcabc" or "defdef":
5731
5732 /(?|(abc)|(def))\1/
5733
5734 In contrast, a subroutine call to a numbered subpattern always refers
5735 to the first one in the pattern with the given number. The following
5736 pattern matches "abcabc" or "defabc":
5737
5738 /(?|(abc)|(def))(?1)/
5739
5740 If a condition test for a subpattern's having matched refers to a non-
5741 unique number, the test is true if any of the subpatterns of that num-
5742 ber have matched.
5743
5744 An alternative approach to using this "branch reset" feature is to use
5745 duplicate named subpatterns, as described in the next section.
5746
5747
5748 NAMED SUBPATTERNS
5749
5750 Identifying capturing parentheses by number is simple, but it can be
5751 very hard to keep track of the numbers in complicated regular expres-
5752 sions. Furthermore, if an expression is modified, the numbers may
5753 change. To help with this difficulty, PCRE supports the naming of sub-
5754 patterns. This feature was not added to Perl until release 5.10. Python
5755 had the feature earlier, and PCRE introduced it at release 4.0, using
5756 the Python syntax. PCRE now supports both the Perl and the Python syn-
5757 tax. Perl allows identically numbered subpatterns to have different
5758 names, but PCRE does not.
5759
5760 In PCRE, a subpattern can be named in one of three ways: (?<name>...)
5761 or (?'name'...) as in Perl, or (?P<name>...) as in Python. References
5762 to capturing parentheses from other parts of the pattern, such as back
5763 references, recursion, and conditions, can be made by name as well as
5764 by number.
5765
5766 Names consist of up to 32 alphanumeric characters and underscores.
5767 Named capturing parentheses are still allocated numbers as well as
5768 names, exactly as if the names were not present. The PCRE API provides
5769 function calls for extracting the name-to-number translation table from
5770 a compiled pattern. There is also a convenience function for extracting
5771 a captured substring by name.
5772
5773 By default, a name must be unique within a pattern, but it is possible
5774 to relax this constraint by setting the PCRE_DUPNAMES option at compile
5775 time. (Duplicate names are also always permitted for subpatterns with
5776 the same number, set up as described in the previous section.) Dupli-
5777 cate names can be useful for patterns where only one instance of the
5778 named parentheses can match. Suppose you want to match the name of a
5779 weekday, either as a 3-letter abbreviation or as the full name, and in
5780 both cases you want to extract the abbreviation. This pattern (ignoring
5781 the line breaks) does the job:
5782
5783 (?<DN>Mon|Fri|Sun)(?:day)?|
5784 (?<DN>Tue)(?:sday)?|
5785 (?<DN>Wed)(?:nesday)?|
5786 (?<DN>Thu)(?:rsday)?|
5787 (?<DN>Sat)(?:urday)?
5788
5789 There are five capturing substrings, but only one is ever set after a
5790 match. (An alternative way of solving this problem is to use a "branch
5791 reset" subpattern, as described in the previous section.)
5792
5793 The convenience function for extracting the data by name returns the
5794 substring for the first (and in this example, the only) subpattern of
5795 that name that matched. This saves searching to find which numbered
5796 subpattern it was.
5797
5798 If you make a back reference to a non-unique named subpattern from
5799 elsewhere in the pattern, the one that corresponds to the first occur-
5800 rence of the name is used. In the absence of duplicate numbers (see the
5801 previous section) this is the one with the lowest number. If you use a
5802 named reference in a condition test (see the section about conditions
5803 below), either to check whether a subpattern has matched, or to check
5804 for recursion, all subpatterns with the same name are tested. If the
5805 condition is true for any one of them, the overall condition is true.
5806 This is the same behaviour as testing by number. For further details of
5807 the interfaces for handling named subpatterns, see the pcreapi documen-
5808 tation.
5809
5810 Warning: You cannot use different names to distinguish between two sub-
5811 patterns with the same number because PCRE uses only the numbers when
5812 matching. For this reason, an error is given at compile time if differ-
5813 ent names are given to subpatterns with the same number. However, you
5814 can give the same name to subpatterns with the same number, even when
5815 PCRE_DUPNAMES is not set.
5816
5817
5818 REPETITION
5819
5820 Repetition is specified by quantifiers, which can follow any of the
5821 following items:
5822
5823 a literal data character
5824 the dot metacharacter
5825 the \C escape sequence
5826 the \X escape sequence
5827 the \R escape sequence
5828 an escape such as \d or \pL that matches a single character
5829 a character class
5830 a back reference (see next section)
5831 a parenthesized subpattern (including assertions)
5832 a subroutine call to a subpattern (recursive or otherwise)
5833
5834 The general repetition quantifier specifies a minimum and maximum num-
5835 ber of permitted matches, by giving the two numbers in curly brackets
5836 (braces), separated by a comma. The numbers must be less than 65536,
5837 and the first must be less than or equal to the second. For example:
5838
5839 z{2,4}
5840
5841 matches "zz", "zzz", or "zzzz". A closing brace on its own is not a
5842 special character. If the second number is omitted, but the comma is
5843 present, there is no upper limit; if the second number and the comma
5844 are both omitted, the quantifier specifies an exact number of required
5845 matches. Thus
5846
5847 [aeiou]{3,}
5848
5849 matches at least 3 successive vowels, but may match many more, while
5850
5851 \d{8}
5852
5853 matches exactly 8 digits. An opening curly bracket that appears in a
5854 position where a quantifier is not allowed, or one that does not match
5855 the syntax of a quantifier, is taken as a literal character. For exam-
5856 ple, {,6} is not a quantifier, but a literal string of four characters.
5857
5858 In UTF modes, quantifiers apply to characters rather than to individual
5859 data units. Thus, for example, \x{100}{2} matches two characters, each
5860 of which is represented by a two-byte sequence in a UTF-8 string. Simi-
5861 larly, \X{3} matches three Unicode extended grapheme clusters, each of
5862 which may be several data units long (and they may be of different
5863 lengths).
5864
5865 The quantifier {0} is permitted, causing the expression to behave as if
5866 the previous item and the quantifier were not present. This may be use-
5867 ful for subpatterns that are referenced as subroutines from elsewhere
5868 in the pattern (but see also the section entitled "Defining subpatterns
5869 for use by reference only" below). Items other than subpatterns that
5870 have a {0} quantifier are omitted from the compiled pattern.
5871
5872 For convenience, the three most common quantifiers have single-charac-
5873 ter abbreviations:
5874
5875 * is equivalent to {0,}
5876 + is equivalent to {1,}
5877 ? is equivalent to {0,1}
5878
5879 It is possible to construct infinite loops by following a subpattern
5880 that can match no characters with a quantifier that has no upper limit,
5881 for example:
5882
5883 (a?)*
5884
5885 Earlier versions of Perl and PCRE used to give an error at compile time
5886 for such patterns. However, because there are cases where this can be
5887 useful, such patterns are now accepted, but if any repetition of the
5888 subpattern does in fact match no characters, the loop is forcibly bro-
5889 ken.
5890
5891 By default, the quantifiers are "greedy", that is, they match as much
5892 as possible (up to the maximum number of permitted times), without
5893 causing the rest of the pattern to fail. The classic example of where
5894 this gives problems is in trying to match comments in C programs. These
5895 appear between /* and */ and within the comment, individual * and /
5896 characters may appear. An attempt to match C comments by applying the
5897 pattern
5898
5899 /\*.*\*/
5900
5901 to the string
5902
5903 /* first comment */ not comment /* second comment */
5904
5905 fails, because it matches the entire string owing to the greediness of
5906 the .* item.
5907
5908 However, if a quantifier is followed by a question mark, it ceases to
5909 be greedy, and instead matches the minimum number of times possible, so
5910 the pattern
5911
5912 /\*.*?\*/
5913
5914 does the right thing with the C comments. The meaning of the various
5915 quantifiers is not otherwise changed, just the preferred number of
5916 matches. Do not confuse this use of question mark with its use as a
5917 quantifier in its own right. Because it has two uses, it can sometimes
5918 appear doubled, as in
5919
5920 \d??\d
5921
5922 which matches one digit by preference, but can match two if that is the
5923 only way the rest of the pattern matches.
5924
5925 If the PCRE_UNGREEDY option is set (an option that is not available in
5926 Perl), the quantifiers are not greedy by default, but individual ones
5927 can be made greedy by following them with a question mark. In other
5928 words, it inverts the default behaviour.
5929
5930 When a parenthesized subpattern is quantified with a minimum repeat
5931 count that is greater than 1 or with a limited maximum, more memory is
5932 required for the compiled pattern, in proportion to the size of the
5933 minimum or maximum.
5934
5935 If a pattern starts with .* or .{0,} and the PCRE_DOTALL option (equiv-
5936 alent to Perl's /s) is set, thus allowing the dot to match newlines,
5937 the pattern is implicitly anchored, because whatever follows will be
5938 tried against every character position in the subject string, so there
5939 is no point in retrying the overall match at any position after the
5940 first. PCRE normally treats such a pattern as though it were preceded
5941 by \A.
5942
5943 In cases where it is known that the subject string contains no new-
5944 lines, it is worth setting PCRE_DOTALL in order to obtain this opti-
5945 mization, or alternatively using ^ to indicate anchoring explicitly.
5946
5947 However, there are some cases where the optimization cannot be used.
5948 When .* is inside capturing parentheses that are the subject of a back
5949 reference elsewhere in the pattern, a match at the start may fail where
5950 a later one succeeds. Consider, for example:
5951
5952 (.*)abc\1
5953
5954 If the subject is "xyz123abc123" the match point is the fourth charac-
5955 ter. For this reason, such a pattern is not implicitly anchored.
5956
5957 Another case where implicit anchoring is not applied is when the lead-
5958 ing .* is inside an atomic group. Once again, a match at the start may
5959 fail where a later one succeeds. Consider this pattern:
5960
5961 (?>.*?a)b
5962
5963 It matches "ab" in the subject "aab". The use of the backtracking con-
5964 trol verbs (*PRUNE) and (*SKIP) also disable this optimization.
5965
5966 When a capturing subpattern is repeated, the value captured is the sub-
5967 string that matched the final iteration. For example, after
5968
5969 (tweedle[dume]{3}\s*)+
5970
5971 has matched "tweedledum tweedledee" the value of the captured substring
5972 is "tweedledee". However, if there are nested capturing subpatterns,
5973 the corresponding captured values may have been set in previous itera-
5974 tions. For example, after
5975
5976 /(a|(b))+/
5977
5978 matches "aba" the value of the second captured substring is "b".
5979
5980
5981 ATOMIC GROUPING AND POSSESSIVE QUANTIFIERS
5982
5983 With both maximizing ("greedy") and minimizing ("ungreedy" or "lazy")
5984 repetition, failure of what follows normally causes the repeated item
5985 to be re-evaluated to see if a different number of repeats allows the
5986 rest of the pattern to match. Sometimes it is useful to prevent this,
5987 either to change the nature of the match, or to cause it fail earlier
5988 than it otherwise might, when the author of the pattern knows there is
5989 no point in carrying on.
5990
5991 Consider, for example, the pattern \d+foo when applied to the subject
5992 line
5993
5994 123456bar
5995
5996 After matching all 6 digits and then failing to match "foo", the normal
5997 action of the matcher is to try again with only 5 digits matching the
5998 \d+ item, and then with 4, and so on, before ultimately failing.
5999 "Atomic grouping" (a term taken from Jeffrey Friedl's book) provides
6000 the means for specifying that once a subpattern has matched, it is not
6001 to be re-evaluated in this way.
6002
6003 If we use atomic grouping for the previous example, the matcher gives
6004 up immediately on failing to match "foo" the first time. The notation
6005 is a kind of special parenthesis, starting with (?> as in this example:
6006
6007 (?>\d+)foo
6008
6009 This kind of parenthesis "locks up" the part of the pattern it con-
6010 tains once it has matched, and a failure further into the pattern is
6011 prevented from backtracking into it. Backtracking past it to previous
6012 items, however, works as normal.
6013
6014 An alternative description is that a subpattern of this type matches
6015 the string of characters that an identical standalone pattern would
6016 match, if anchored at the current point in the subject string.
6017
6018 Atomic grouping subpatterns are not capturing subpatterns. Simple cases
6019 such as the above example can be thought of as a maximizing repeat that
6020 must swallow everything it can. So, while both \d+ and \d+? are pre-
6021 pared to adjust the number of digits they match in order to make the
6022 rest of the pattern match, (?>\d+) can only match an entire sequence of
6023 digits.
6024
6025 Atomic groups in general can of course contain arbitrarily complicated
6026 subpatterns, and can be nested. However, when the subpattern for an
6027 atomic group is just a single repeated item, as in the example above, a
6028 simpler notation, called a "possessive quantifier" can be used. This
6029 consists of an additional + character following a quantifier. Using
6030 this notation, the previous example can be rewritten as
6031
6032 \d++foo
6033
6034 Note that a possessive quantifier can be used with an entire group, for
6035 example:
6036
6037 (abc|xyz){2,3}+
6038
6039 Possessive quantifiers are always greedy; the setting of the
6040 PCRE_UNGREEDY option is ignored. They are a convenient notation for the
6041 simpler forms of atomic group. However, there is no difference in the
6042 meaning of a possessive quantifier and the equivalent atomic group,
6043 though there may be a performance difference; possessive quantifiers