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

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

Parent Directory Parent Directory | Revision Log Revision Log


Revision 982 - (show annotations)
Wed Jun 20 15:15:27 2012 UTC (7 years, 3 months ago) by ph10
File MIME type: text/plain
File size: 442919 byte(s)
Source tidies for 8.31-RC2.
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 The two libraries contain identical sets of functions, except that the
36 names in the 16-bit library start with pcre16_ instead of pcre_. To
37 avoid over-complication and reduce the documentation maintenance load,
38 most of the documentation describes the 8-bit library, with the differ-
39 ences for the 16-bit library described separately in the pcre16 page.
40 References to functions or structures of the form pcre[16]_xxx should
41 be read as meaning "pcre_xxx when using the 8-bit library and
42 pcre16_xxx when using the 16-bit library".
43
44 The current implementation of PCRE corresponds approximately with Perl
45 5.12, including support for UTF-8/16 encoded strings and Unicode gen-
46 eral category properties. However, UTF-8/16 and Unicode support has to
47 be explicitly enabled; it is not the default. The Unicode tables corre-
48 spond to Unicode release 6.0.0.
49
50 In addition to the Perl-compatible matching function, PCRE contains an
51 alternative function that matches the same compiled patterns in a dif-
52 ferent way. In certain circumstances, the alternative function has some
53 advantages. For a discussion of the two matching algorithms, see the
54 pcrematching page.
55
56 PCRE is written in C and released as a C library. A number of people
57 have written wrappers and interfaces of various kinds. In particular,
58 Google Inc. have provided a comprehensive C++ wrapper for the 8-bit
59 library. This is now included as part of the PCRE distribution. The
60 pcrecpp page has details of this interface. Other people's contribu-
61 tions can be found in the Contrib directory at the primary FTP site,
62 which is:
63
64 ftp://ftp.csx.cam.ac.uk/pub/software/programming/pcre
65
66 Details of exactly which Perl regular expression features are and are
67 not supported by PCRE are given in separate documents. See the pcrepat-
68 tern and pcrecompat pages. There is a syntax summary in the pcresyntax
69 page.
70
71 Some features of PCRE can be included, excluded, or changed when the
72 library is built. The pcre_config() function makes it possible for a
73 client to discover which features are available. The features them-
74 selves are described in the pcrebuild page. Documentation about build-
75 ing PCRE for various operating systems can be found in the README and
76 NON-UNIX-USE files in the source distribution.
77
78 The libraries contains a number of undocumented internal functions and
79 data tables that are used by more than one of the exported external
80 functions, but which are not intended for use by external callers.
81 Their names all begin with "_pcre_" or "_pcre16_", which hopefully will
82 not provoke any name clashes. In some environments, it is possible to
83 control which external symbols are exported when a shared library is
84 built, and in these cases the undocumented symbols are not exported.
85
86
87 USER DOCUMENTATION
88
89 The user documentation for PCRE comprises a number of different sec-
90 tions. In the "man" format, each of these is a separate "man page". In
91 the HTML format, each is a separate page, linked from the index page.
92 In the plain text format, all the sections, except the pcredemo sec-
93 tion, are concatenated, for ease of searching. The sections are as fol-
94 lows:
95
96 pcre this document
97 pcre16 details of the 16-bit library
98 pcre-config show PCRE installation configuration information
99 pcreapi details of PCRE's native C API
100 pcrebuild options for building PCRE
101 pcrecallout details of the callout feature
102 pcrecompat discussion of Perl compatibility
103 pcrecpp details of the C++ wrapper for the 8-bit library
104 pcredemo a demonstration C program that uses PCRE
105 pcregrep description of the pcregrep command (8-bit only)
106 pcrejit discussion of the just-in-time optimization support
107 pcrelimits details of size and other limits
108 pcrematching discussion of the two matching algorithms
109 pcrepartial details of the partial matching facility
110 pcrepattern syntax and semantics of supported
111 regular expressions
112 pcreperform discussion of performance issues
113 pcreposix the POSIX-compatible C API for the 8-bit library
114 pcreprecompile details of saving and re-using precompiled patterns
115 pcresample discussion of the pcredemo program
116 pcrestack discussion of stack usage
117 pcresyntax quick syntax reference
118 pcretest description of the pcretest testing command
119 pcreunicode discussion of Unicode and UTF-8/16 support
120
121 In addition, in the "man" and HTML formats, there is a short page for
122 each 8-bit C library function, listing its arguments and results.
123
124
125 AUTHOR
126
127 Philip Hazel
128 University Computing Service
129 Cambridge CB2 3QH, England.
130
131 Putting an actual email address here seems to have been a spam magnet,
132 so I've taken it away. If you want to email me, use my two initials,
133 followed by the two digits 10, at the domain cam.ac.uk.
134
135
136 REVISION
137
138 Last updated: 10 January 2012
139 Copyright (c) 1997-2012 University of Cambridge.
140 ------------------------------------------------------------------------------
141
142
143 PCRE(3) PCRE(3)
144
145
146 NAME
147 PCRE - Perl-compatible regular expressions
148
149 #include <pcre.h>
150
151
152 PCRE 16-BIT API BASIC FUNCTIONS
153
154 pcre16 *pcre16_compile(PCRE_SPTR16 pattern, int options,
155 const char **errptr, int *erroffset,
156 const unsigned char *tableptr);
157
158 pcre16 *pcre16_compile2(PCRE_SPTR16 pattern, int options,
159 int *errorcodeptr,
160 const char **errptr, int *erroffset,
161 const unsigned char *tableptr);
162
163 pcre16_extra *pcre16_study(const pcre16 *code, int options,
164 const char **errptr);
165
166 void pcre16_free_study(pcre16_extra *extra);
167
168 int pcre16_exec(const pcre16 *code, const pcre16_extra *extra,
169 PCRE_SPTR16 subject, int length, int startoffset,
170 int options, int *ovector, int ovecsize);
171
172 int pcre16_dfa_exec(const pcre16 *code, const pcre16_extra *extra,
173 PCRE_SPTR16 subject, int length, int startoffset,
174 int options, int *ovector, int ovecsize,
175 int *workspace, int wscount);
176
177
178 PCRE 16-BIT API STRING EXTRACTION FUNCTIONS
179
180 int pcre16_copy_named_substring(const pcre16 *code,
181 PCRE_SPTR16 subject, int *ovector,
182 int stringcount, PCRE_SPTR16 stringname,
183 PCRE_UCHAR16 *buffer, int buffersize);
184
185 int pcre16_copy_substring(PCRE_SPTR16 subject, int *ovector,
186 int stringcount, int stringnumber, PCRE_UCHAR16 *buffer,
187 int buffersize);
188
189 int pcre16_get_named_substring(const pcre16 *code,
190 PCRE_SPTR16 subject, int *ovector,
191 int stringcount, PCRE_SPTR16 stringname,
192 PCRE_SPTR16 *stringptr);
193
194 int pcre16_get_stringnumber(const pcre16 *code,
195 PCRE_SPTR16 name);
196
197 int pcre16_get_stringtable_entries(const pcre16 *code,
198 PCRE_SPTR16 name, PCRE_UCHAR16 **first, PCRE_UCHAR16 **last);
199
200 int pcre16_get_substring(PCRE_SPTR16 subject, int *ovector,
201 int stringcount, int stringnumber,
202 PCRE_SPTR16 *stringptr);
203
204 int pcre16_get_substring_list(PCRE_SPTR16 subject,
205 int *ovector, int stringcount, PCRE_SPTR16 **listptr);
206
207 void pcre16_free_substring(PCRE_SPTR16 stringptr);
208
209 void pcre16_free_substring_list(PCRE_SPTR16 *stringptr);
210
211
212 PCRE 16-BIT API AUXILIARY FUNCTIONS
213
214 pcre16_jit_stack *pcre16_jit_stack_alloc(int startsize, int maxsize);
215
216 void pcre16_jit_stack_free(pcre16_jit_stack *stack);
217
218 void pcre16_assign_jit_stack(pcre16_extra *extra,
219 pcre16_jit_callback callback, void *data);
220
221 const unsigned char *pcre16_maketables(void);
222
223 int pcre16_fullinfo(const pcre16 *code, const pcre16_extra *extra,
224 int what, void *where);
225
226 int pcre16_refcount(pcre16 *code, int adjust);
227
228 int pcre16_config(int what, void *where);
229
230 const char *pcre16_version(void);
231
232 int pcre16_pattern_to_host_byte_order(pcre16 *code,
233 pcre16_extra *extra, const unsigned char *tables);
234
235
236 PCRE 16-BIT API INDIRECTED FUNCTIONS
237
238 void *(*pcre16_malloc)(size_t);
239
240 void (*pcre16_free)(void *);
241
242 void *(*pcre16_stack_malloc)(size_t);
243
244 void (*pcre16_stack_free)(void *);
245
246 int (*pcre16_callout)(pcre16_callout_block *);
247
248
249 PCRE 16-BIT API 16-BIT-ONLY FUNCTION
250
251 int pcre16_utf16_to_host_byte_order(PCRE_UCHAR16 *output,
252 PCRE_SPTR16 input, int length, int *byte_order,
253 int keep_boms);
254
255
256 THE PCRE 16-BIT LIBRARY
257
258 Starting with release 8.30, it is possible to compile a PCRE library
259 that supports 16-bit character strings, including UTF-16 strings, as
260 well as or instead of the original 8-bit library. The majority of the
261 work to make this possible was done by Zoltan Herczeg. The two
262 libraries contain identical sets of functions, used in exactly the same
263 way. Only the names of the functions and the data types of their argu-
264 ments and results are different. To avoid over-complication and reduce
265 the documentation maintenance load, most of the PCRE documentation
266 describes the 8-bit library, with only occasional references to the
267 16-bit library. This page describes what is different when you use the
268 16-bit library.
269
270 WARNING: A single application can be linked with both libraries, but
271 you must take care when processing any particular pattern to use func-
272 tions from just one library. For example, if you want to study a pat-
273 tern that was compiled with pcre16_compile(), you must do so with
274 pcre16_study(), not pcre_study(), and you must free the study data with
275 pcre16_free_study().
276
277
278 THE HEADER FILE
279
280 There is only one header file, pcre.h. It contains prototypes for all
281 the functions in both libraries, as well as definitions of flags,
282 structures, error codes, etc.
283
284
285 THE LIBRARY NAME
286
287 In Unix-like systems, the 16-bit library is called libpcre16, and can
288 normally be accesss by adding -lpcre16 to the command for linking an
289 application that uses PCRE.
290
291
292 STRING TYPES
293
294 In the 8-bit library, strings are passed to PCRE library functions as
295 vectors of bytes with the C type "char *". In the 16-bit library,
296 strings are passed as vectors of unsigned 16-bit quantities. The macro
297 PCRE_UCHAR16 specifies an appropriate data type, and PCRE_SPTR16 is
298 defined as "const PCRE_UCHAR16 *". In very many environments, "short
299 int" is a 16-bit data type. When PCRE is built, it defines PCRE_UCHAR16
300 as "short int", but checks that it really is a 16-bit data type. If it
301 is not, the build fails with an error message telling the maintainer to
302 modify the definition appropriately.
303
304
305 STRUCTURE TYPES
306
307 The types of the opaque structures that are used for compiled 16-bit
308 patterns and JIT stacks are pcre16 and pcre16_jit_stack respectively.
309 The type of the user-accessible structure that is returned by
310 pcre16_study() is pcre16_extra, and the type of the structure that is
311 used for passing data to a callout function is pcre16_callout_block.
312 These structures contain the same fields, with the same names, as their
313 8-bit counterparts. The only difference is that pointers to character
314 strings are 16-bit instead of 8-bit types.
315
316
317 16-BIT FUNCTIONS
318
319 For every function in the 8-bit library there is a corresponding func-
320 tion in the 16-bit library with a name that starts with pcre16_ instead
321 of pcre_. The prototypes are listed above. In addition, there is one
322 extra function, pcre16_utf16_to_host_byte_order(). This is a utility
323 function that converts a UTF-16 character string to host byte order if
324 necessary. The other 16-bit functions expect the strings they are
325 passed to be in host byte order.
326
327 The input and output arguments of pcre16_utf16_to_host_byte_order() may
328 point to the same address, that is, conversion in place is supported.
329 The output buffer must be at least as long as the input.
330
331 The length argument specifies the number of 16-bit data units in the
332 input string; a negative value specifies a zero-terminated string.
333
334 If byte_order is NULL, it is assumed that the string starts off in host
335 byte order. This may be changed by byte-order marks (BOMs) anywhere in
336 the string (commonly as the first character).
337
338 If byte_order is not NULL, a non-zero value of the integer to which it
339 points means that the input starts off in host byte order, otherwise
340 the opposite order is assumed. Again, BOMs in the string can change
341 this. The final byte order is passed back at the end of processing.
342
343 If keep_boms is not zero, byte-order mark characters (0xfeff) are
344 copied into the output string. Otherwise they are discarded.
345
346 The result of the function is the number of 16-bit units placed into
347 the output buffer, including the zero terminator if the string was
348 zero-terminated.
349
350
351 SUBJECT STRING OFFSETS
352
353 The offsets within subject strings that are returned by the matching
354 functions are in 16-bit units rather than bytes.
355
356
357 NAMED SUBPATTERNS
358
359 The name-to-number translation table that is maintained for named sub-
360 patterns uses 16-bit characters. The pcre16_get_stringtable_entries()
361 function returns the length of each entry in the table as the number of
362 16-bit data units.
363
364
365 OPTION NAMES
366
367 There are two new general option names, PCRE_UTF16 and
368 PCRE_NO_UTF16_CHECK, which correspond to PCRE_UTF8 and
369 PCRE_NO_UTF8_CHECK in the 8-bit library. In fact, these new options
370 define the same bits in the options word. There is a discussion about
371 the validity of UTF-16 strings in the pcreunicode page.
372
373 For the pcre16_config() function there is an option PCRE_CONFIG_UTF16
374 that returns 1 if UTF-16 support is configured, otherwise 0. If this
375 option is given to pcre_config(), or if the PCRE_CONFIG_UTF8 option is
376 given to pcre16_config(), the result is the PCRE_ERROR_BADOPTION error.
377
378
379 CHARACTER CODES
380
381 In 16-bit mode, when PCRE_UTF16 is not set, character values are
382 treated in the same way as in 8-bit, non UTF-8 mode, except, of course,
383 that they can range from 0 to 0xffff instead of 0 to 0xff. Character
384 types for characters less than 0xff can therefore be influenced by the
385 locale in the same way as before. Characters greater than 0xff have
386 only one case, and no "type" (such as letter or digit).
387
388 In UTF-16 mode, the character code is Unicode, in the range 0 to
389 0x10ffff, with the exception of values in the range 0xd800 to 0xdfff
390 because those are "surrogate" values that are used in pairs to encode
391 values greater than 0xffff.
392
393 A UTF-16 string can indicate its endianness by special code knows as a
394 byte-order mark (BOM). The PCRE functions do not handle this, expecting
395 strings to be in host byte order. A utility function called
396 pcre16_utf16_to_host_byte_order() is provided to help with this (see
397 above).
398
399
400 ERROR NAMES
401
402 The errors PCRE_ERROR_BADUTF16_OFFSET and PCRE_ERROR_SHORTUTF16 corre-
403 spond to their 8-bit counterparts. The error PCRE_ERROR_BADMODE is
404 given when a compiled pattern is passed to a function that processes
405 patterns in the other mode, for example, if a pattern compiled with
406 pcre_compile() is passed to pcre16_exec().
407
408 There are new error codes whose names begin with PCRE_UTF16_ERR for
409 invalid UTF-16 strings, corresponding to the PCRE_UTF8_ERR codes for
410 UTF-8 strings that are described in the section entitled "Reason codes
411 for invalid UTF-8 strings" in the main pcreapi page. The UTF-16 errors
412 are:
413
414 PCRE_UTF16_ERR1 Missing low surrogate at end of string
415 PCRE_UTF16_ERR2 Invalid low surrogate follows high surrogate
416 PCRE_UTF16_ERR3 Isolated low surrogate
417 PCRE_UTF16_ERR4 Invalid character 0xfffe
418
419
420 ERROR TEXTS
421
422 If there is an error while compiling a pattern, the error text that is
423 passed back by pcre16_compile() or pcre16_compile2() is still an 8-bit
424 character string, zero-terminated.
425
426
427 CALLOUTS
428
429 The subject and mark fields in the callout block that is passed to a
430 callout function point to 16-bit vectors.
431
432
433 TESTING
434
435 The pcretest program continues to operate with 8-bit input and output
436 files, but it can be used for testing the 16-bit library. If it is run
437 with the command line option -16, patterns and subject strings are con-
438 verted from 8-bit to 16-bit before being passed to PCRE, and the 16-bit
439 library functions are used instead of the 8-bit ones. Returned 16-bit
440 strings are converted to 8-bit for output. If the 8-bit library was not
441 compiled, pcretest defaults to 16-bit and the -16 option is ignored.
442
443 When PCRE is being built, the RunTest script that is called by "make
444 check" uses the pcretest -C option to discover which of the 8-bit and
445 16-bit libraries has been built, and runs the tests appropriately.
446
447
448 NOT SUPPORTED IN 16-BIT MODE
449
450 Not all the features of the 8-bit library are available with the 16-bit
451 library. The C++ and POSIX wrapper functions support only the 8-bit
452 library, and the pcregrep program is at present 8-bit only.
453
454
455 AUTHOR
456
457 Philip Hazel
458 University Computing Service
459 Cambridge CB2 3QH, England.
460
461
462 REVISION
463
464 Last updated: 14 April 2012
465 Copyright (c) 1997-2012 University of Cambridge.
466 ------------------------------------------------------------------------------
467
468
469 PCREBUILD(3) PCREBUILD(3)
470
471
472 NAME
473 PCRE - Perl-compatible regular expressions
474
475
476 PCRE BUILD-TIME OPTIONS
477
478 This document describes the optional features of PCRE that can be
479 selected when the library is compiled. It assumes use of the configure
480 script, where the optional features are selected or deselected by pro-
481 viding options to configure before running the make command. However,
482 the same options can be selected in both Unix-like and non-Unix-like
483 environments using the GUI facility of cmake-gui if you are using CMake
484 instead of configure to build PCRE.
485
486 There is a lot more information about building PCRE in non-Unix-like
487 environments in the file called NON_UNIX_USE, which is part of the PCRE
488 distribution. You should consult this file as well as the README file
489 if you are building in a non-Unix-like environment.
490
491 The complete list of options for configure (which includes the standard
492 ones such as the selection of the installation directory) can be
493 obtained by running
494
495 ./configure --help
496
497 The following sections include descriptions of options whose names
498 begin with --enable or --disable. These settings specify changes to the
499 defaults for the configure command. Because of the way that configure
500 works, --enable and --disable always come in pairs, so the complemen-
501 tary option always exists as well, but as it specifies the default, it
502 is not described.
503
504
505 BUILDING 8-BIT and 16-BIT LIBRARIES
506
507 By default, a library called libpcre is built, containing functions
508 that take string arguments contained in vectors of bytes, either as
509 single-byte characters, or interpreted as UTF-8 strings. You can also
510 build a separate library, called libpcre16, in which strings are con-
511 tained in vectors of 16-bit data units and interpreted either as sin-
512 gle-unit characters or UTF-16 strings, by adding
513
514 --enable-pcre16
515
516 to the configure command. If you do not want the 8-bit library, add
517
518 --disable-pcre8
519
520 as well. At least one of the two libraries must be built. Note that the
521 C++ and POSIX wrappers are for the 8-bit library only, and that pcre-
522 grep is an 8-bit program. None of these are built if you select only
523 the 16-bit library.
524
525
526 BUILDING SHARED AND STATIC LIBRARIES
527
528 The PCRE building process uses libtool to build both shared and static
529 Unix libraries by default. You can suppress one of these by adding one
530 of
531
532 --disable-shared
533 --disable-static
534
535 to the configure command, as required.
536
537
538 C++ SUPPORT
539
540 By default, if the 8-bit library is being built, the configure script
541 will search for a C++ compiler and C++ header files. If it finds them,
542 it automatically builds the C++ wrapper library (which supports only
543 8-bit strings). You can disable this by adding
544
545 --disable-cpp
546
547 to the configure command.
548
549
550 UTF-8 and UTF-16 SUPPORT
551
552 To build PCRE with support for UTF Unicode character strings, add
553
554 --enable-utf
555
556 to the configure command. This setting applies to both libraries,
557 adding support for UTF-8 to the 8-bit library and support for UTF-16 to
558 the 16-bit library. There are no separate options for enabling UTF-8
559 and UTF-16 independently because that would allow ridiculous settings
560 such as requesting UTF-16 support while building only the 8-bit
561 library. It is not possible to build one library with UTF support and
562 the other without in the same configuration. (For backwards compatibil-
563 ity, --enable-utf8 is a synonym of --enable-utf.)
564
565 Of itself, this setting does not make PCRE treat strings as UTF-8 or
566 UTF-16. As well as compiling PCRE with this option, you also have have
567 to set the PCRE_UTF8 or PCRE_UTF16 option when you call one of the pat-
568 tern compiling functions.
569
570 If you set --enable-utf when compiling in an EBCDIC environment, PCRE
571 expects its input to be either ASCII or UTF-8 (depending on the run-
572 time option). It is not possible to support both EBCDIC and UTF-8 codes
573 in the same version of the library. Consequently, --enable-utf and
574 --enable-ebcdic are mutually exclusive.
575
576
577 UNICODE CHARACTER PROPERTY SUPPORT
578
579 UTF support allows the libraries to process character codepoints up to
580 0x10ffff in the strings that they handle. On its own, however, it does
581 not provide any facilities for accessing the properties of such charac-
582 ters. If you want to be able to use the pattern escapes \P, \p, and \X,
583 which refer to Unicode character properties, you must add
584
585 --enable-unicode-properties
586
587 to the configure command. This implies UTF support, even if you have
588 not explicitly requested it.
589
590 Including Unicode property support adds around 30K of tables to the
591 PCRE library. Only the general category properties such as Lu and Nd
592 are supported. Details are given in the pcrepattern documentation.
593
594
595 JUST-IN-TIME COMPILER SUPPORT
596
597 Just-in-time compiler support is included in the build by specifying
598
599 --enable-jit
600
601 This support is available only for certain hardware architectures. If
602 this option is set for an unsupported architecture, a compile time
603 error occurs. See the pcrejit documentation for a discussion of JIT
604 usage. When JIT support is enabled, pcregrep automatically makes use of
605 it, unless you add
606
607 --disable-pcregrep-jit
608
609 to the "configure" command.
610
611
612 CODE VALUE OF NEWLINE
613
614 By default, PCRE interprets the linefeed (LF) character as indicating
615 the end of a line. This is the normal newline character on Unix-like
616 systems. You can compile PCRE to use carriage return (CR) instead, by
617 adding
618
619 --enable-newline-is-cr
620
621 to the configure command. There is also a --enable-newline-is-lf
622 option, which explicitly specifies linefeed as the newline character.
623
624 Alternatively, you can specify that line endings are to be indicated by
625 the two character sequence CRLF. If you want this, add
626
627 --enable-newline-is-crlf
628
629 to the configure command. There is a fourth option, specified by
630
631 --enable-newline-is-anycrlf
632
633 which causes PCRE to recognize any of the three sequences CR, LF, or
634 CRLF as indicating a line ending. Finally, a fifth option, specified by
635
636 --enable-newline-is-any
637
638 causes PCRE to recognize any Unicode newline sequence.
639
640 Whatever line ending convention is selected when PCRE is built can be
641 overridden when the library functions are called. At build time it is
642 conventional to use the standard for your operating system.
643
644
645 WHAT \R MATCHES
646
647 By default, the sequence \R in a pattern matches any Unicode newline
648 sequence, whatever has been selected as the line ending sequence. If
649 you specify
650
651 --enable-bsr-anycrlf
652
653 the default is changed so that \R matches only CR, LF, or CRLF. What-
654 ever is selected when PCRE is built can be overridden when the library
655 functions are called.
656
657
658 POSIX MALLOC USAGE
659
660 When the 8-bit library is called through the POSIX interface (see the
661 pcreposix documentation), additional working storage is required for
662 holding the pointers to capturing substrings, because PCRE requires
663 three integers per substring, whereas the POSIX interface provides only
664 two. If the number of expected substrings is small, the wrapper func-
665 tion uses space on the stack, because this is faster than using mal-
666 loc() for each call. The default threshold above which the stack is no
667 longer used is 10; it can be changed by adding a setting such as
668
669 --with-posix-malloc-threshold=20
670
671 to the configure command.
672
673
674 HANDLING VERY LARGE PATTERNS
675
676 Within a compiled pattern, offset values are used to point from one
677 part to another (for example, from an opening parenthesis to an alter-
678 nation metacharacter). By default, two-byte values are used for these
679 offsets, leading to a maximum size for a compiled pattern of around
680 64K. This is sufficient to handle all but the most gigantic patterns.
681 Nevertheless, some people do want to process truly enormous patterns,
682 so it is possible to compile PCRE to use three-byte or four-byte off-
683 sets by adding a setting such as
684
685 --with-link-size=3
686
687 to the configure command. The value given must be 2, 3, or 4. For the
688 16-bit library, a value of 3 is rounded up to 4. Using longer offsets
689 slows down the operation of PCRE because it has to load additional data
690 when handling them.
691
692
693 AVOIDING EXCESSIVE STACK USAGE
694
695 When matching with the pcre_exec() function, PCRE implements backtrack-
696 ing by making recursive calls to an internal function called match().
697 In environments where the size of the stack is limited, this can se-
698 verely limit PCRE's operation. (The Unix environment does not usually
699 suffer from this problem, but it may sometimes be necessary to increase
700 the maximum stack size. There is a discussion in the pcrestack docu-
701 mentation.) An alternative approach to recursion that uses memory from
702 the heap to remember data, instead of using recursive function calls,
703 has been implemented to work round the problem of limited stack size.
704 If you want to build a version of PCRE that works this way, add
705
706 --disable-stack-for-recursion
707
708 to the configure command. With this configuration, PCRE will use the
709 pcre_stack_malloc and pcre_stack_free variables to call memory manage-
710 ment functions. By default these point to malloc() and free(), but you
711 can replace the pointers so that your own functions are used instead.
712
713 Separate functions are provided rather than using pcre_malloc and
714 pcre_free because the usage is very predictable: the block sizes
715 requested are always the same, and the blocks are always freed in
716 reverse order. A calling program might be able to implement optimized
717 functions that perform better than malloc() and free(). PCRE runs
718 noticeably more slowly when built in this way. This option affects only
719 the pcre_exec() function; it is not relevant for pcre_dfa_exec().
720
721
722 LIMITING PCRE RESOURCE USAGE
723
724 Internally, PCRE has a function called match(), which it calls repeat-
725 edly (sometimes recursively) when matching a pattern with the
726 pcre_exec() function. By controlling the maximum number of times this
727 function may be called during a single matching operation, a limit can
728 be placed on the resources used by a single call to pcre_exec(). The
729 limit can be changed at run time, as described in the pcreapi documen-
730 tation. The default is 10 million, but this can be changed by adding a
731 setting such as
732
733 --with-match-limit=500000
734
735 to the configure command. This setting has no effect on the
736 pcre_dfa_exec() matching function.
737
738 In some environments it is desirable to limit the depth of recursive
739 calls of match() more strictly than the total number of calls, in order
740 to restrict the maximum amount of stack (or heap, if --disable-stack-
741 for-recursion is specified) that is used. A second limit controls this;
742 it defaults to the value that is set for --with-match-limit, which
743 imposes no additional constraints. However, you can set a lower limit
744 by adding, for example,
745
746 --with-match-limit-recursion=10000
747
748 to the configure command. This value can also be overridden at run
749 time.
750
751
752 CREATING CHARACTER TABLES AT BUILD TIME
753
754 PCRE uses fixed tables for processing characters whose code values are
755 less than 256. By default, PCRE is built with a set of tables that are
756 distributed in the file pcre_chartables.c.dist. These tables are for
757 ASCII codes only. If you add
758
759 --enable-rebuild-chartables
760
761 to the configure command, the distributed tables are no longer used.
762 Instead, a program called dftables is compiled and run. This outputs
763 the source for new set of tables, created in the default locale of your
764 C run-time system. (This method of replacing the tables does not work
765 if you are cross compiling, because dftables is run on the local host.
766 If you need to create alternative tables when cross compiling, you will
767 have to do so "by hand".)
768
769
770 USING EBCDIC CODE
771
772 PCRE assumes by default that it will run in an environment where the
773 character code is ASCII (or Unicode, which is a superset of ASCII).
774 This is the case for most computer operating systems. PCRE can, how-
775 ever, be compiled to run in an EBCDIC environment by adding
776
777 --enable-ebcdic
778
779 to the configure command. This setting implies --enable-rebuild-charta-
780 bles. You should only use it if you know that you are in an EBCDIC
781 environment (for example, an IBM mainframe operating system). The
782 --enable-ebcdic option is incompatible with --enable-utf.
783
784
785 PCREGREP OPTIONS FOR COMPRESSED FILE SUPPORT
786
787 By default, pcregrep reads all files as plain text. You can build it so
788 that it recognizes files whose names end in .gz or .bz2, and reads them
789 with libz or libbz2, respectively, by adding one or both of
790
791 --enable-pcregrep-libz
792 --enable-pcregrep-libbz2
793
794 to the configure command. These options naturally require that the rel-
795 evant libraries are installed on your system. Configuration will fail
796 if they are not.
797
798
799 PCREGREP BUFFER SIZE
800
801 pcregrep uses an internal buffer to hold a "window" on the file it is
802 scanning, in order to be able to output "before" and "after" lines when
803 it finds a match. The size of the buffer is controlled by a parameter
804 whose default value is 20K. The buffer itself is three times this size,
805 but because of the way it is used for holding "before" lines, the long-
806 est line that is guaranteed to be processable is the parameter size.
807 You can change the default parameter value by adding, for example,
808
809 --with-pcregrep-bufsize=50K
810
811 to the configure command. The caller of pcregrep can, however, override
812 this value by specifying a run-time option.
813
814
815 PCRETEST OPTION FOR LIBREADLINE SUPPORT
816
817 If you add
818
819 --enable-pcretest-libreadline
820
821 to the configure command, pcretest is linked with the libreadline
822 library, and when its input is from a terminal, it reads it using the
823 readline() function. This provides line-editing and history facilities.
824 Note that libreadline is GPL-licensed, so if you distribute a binary of
825 pcretest linked in this way, there may be licensing issues.
826
827 Setting this option causes the -lreadline option to be added to the
828 pcretest build. In many operating environments with a sytem-installed
829 libreadline this is sufficient. However, in some environments (e.g. if
830 an unmodified distribution version of readline is in use), some extra
831 configuration may be necessary. The INSTALL file for libreadline says
832 this:
833
834 "Readline uses the termcap functions, but does not link with the
835 termcap or curses library itself, allowing applications which link
836 with readline the to choose an appropriate library."
837
838 If your environment has not been set up so that an appropriate library
839 is automatically included, you may need to add something like
840
841 LIBS="-ncurses"
842
843 immediately before the configure command.
844
845
846 SEE ALSO
847
848 pcreapi(3), pcre16, pcre_config(3).
849
850
851 AUTHOR
852
853 Philip Hazel
854 University Computing Service
855 Cambridge CB2 3QH, England.
856
857
858 REVISION
859
860 Last updated: 07 January 2012
861 Copyright (c) 1997-2012 University of Cambridge.
862 ------------------------------------------------------------------------------
863
864
865 PCREMATCHING(3) PCREMATCHING(3)
866
867
868 NAME
869 PCRE - Perl-compatible regular expressions
870
871
872 PCRE MATCHING ALGORITHMS
873
874 This document describes the two different algorithms that are available
875 in PCRE for matching a compiled regular expression against a given sub-
876 ject string. The "standard" algorithm is the one provided by the
877 pcre_exec() and pcre16_exec() functions. These work in the same was as
878 Perl's matching function, and provide a Perl-compatible matching opera-
879 tion. The just-in-time (JIT) optimization that is described in the
880 pcrejit documentation is compatible with these functions.
881
882 An alternative algorithm is provided by the pcre_dfa_exec() and
883 pcre16_dfa_exec() functions; they operate in a different way, and are
884 not Perl-compatible. This alternative has advantages and disadvantages
885 compared with the standard algorithm, and these are described below.
886
887 When there is only one possible way in which a given subject string can
888 match a pattern, the two algorithms give the same answer. A difference
889 arises, however, when there are multiple possibilities. For example, if
890 the pattern
891
892 ^<.*>
893
894 is matched against the string
895
896 <something> <something else> <something further>
897
898 there are three possible answers. The standard algorithm finds only one
899 of them, whereas the alternative algorithm finds all three.
900
901
902 REGULAR EXPRESSIONS AS TREES
903
904 The set of strings that are matched by a regular expression can be rep-
905 resented as a tree structure. An unlimited repetition in the pattern
906 makes the tree of infinite size, but it is still a tree. Matching the
907 pattern to a given subject string (from a given starting point) can be
908 thought of as a search of the tree. There are two ways to search a
909 tree: depth-first and breadth-first, and these correspond to the two
910 matching algorithms provided by PCRE.
911
912
913 THE STANDARD MATCHING ALGORITHM
914
915 In the terminology of Jeffrey Friedl's book "Mastering Regular Expres-
916 sions", the standard algorithm is an "NFA algorithm". It conducts a
917 depth-first search of the pattern tree. That is, it proceeds along a
918 single path through the tree, checking that the subject matches what is
919 required. When there is a mismatch, the algorithm tries any alterna-
920 tives at the current point, and if they all fail, it backs up to the
921 previous branch point in the tree, and tries the next alternative
922 branch at that level. This often involves backing up (moving to the
923 left) in the subject string as well. The order in which repetition
924 branches are tried is controlled by the greedy or ungreedy nature of
925 the quantifier.
926
927 If a leaf node is reached, a matching string has been found, and at
928 that point the algorithm stops. Thus, if there is more than one possi-
929 ble match, this algorithm returns the first one that it finds. Whether
930 this is the shortest, the longest, or some intermediate length depends
931 on the way the greedy and ungreedy repetition quantifiers are specified
932 in the pattern.
933
934 Because it ends up with a single path through the tree, it is rela-
935 tively straightforward for this algorithm to keep track of the sub-
936 strings that are matched by portions of the pattern in parentheses.
937 This provides support for capturing parentheses and back references.
938
939
940 THE ALTERNATIVE MATCHING ALGORITHM
941
942 This algorithm conducts a breadth-first search of the tree. Starting
943 from the first matching point in the subject, it scans the subject
944 string from left to right, once, character by character, and as it does
945 this, it remembers all the paths through the tree that represent valid
946 matches. In Friedl's terminology, this is a kind of "DFA algorithm",
947 though it is not implemented as a traditional finite state machine (it
948 keeps multiple states active simultaneously).
949
950 Although the general principle of this matching algorithm is that it
951 scans the subject string only once, without backtracking, there is one
952 exception: when a lookaround assertion is encountered, the characters
953 following or preceding the current point have to be independently
954 inspected.
955
956 The scan continues until either the end of the subject is reached, or
957 there are no more unterminated paths. At this point, terminated paths
958 represent the different matching possibilities (if there are none, the
959 match has failed). Thus, if there is more than one possible match,
960 this algorithm finds all of them, and in particular, it finds the long-
961 est. The matches are returned in decreasing order of length. There is
962 an option to stop the algorithm after the first match (which is neces-
963 sarily the shortest) is found.
964
965 Note that all the matches that are found start at the same point in the
966 subject. If the pattern
967
968 cat(er(pillar)?)?
969
970 is matched against the string "the caterpillar catchment", the result
971 will be the three strings "caterpillar", "cater", and "cat" that start
972 at the fifth character of the subject. The algorithm does not automati-
973 cally move on to find matches that start at later positions.
974
975 There are a number of features of PCRE regular expressions that are not
976 supported by the alternative matching algorithm. They are as follows:
977
978 1. Because the algorithm finds all possible matches, the greedy or
979 ungreedy nature of repetition quantifiers is not relevant. Greedy and
980 ungreedy quantifiers are treated in exactly the same way. However, pos-
981 sessive quantifiers can make a difference when what follows could also
982 match what is quantified, for example in a pattern like this:
983
984 ^a++\w!
985
986 This pattern matches "aaab!" but not "aaa!", which would be matched by
987 a non-possessive quantifier. Similarly, if an atomic group is present,
988 it is matched as if it were a standalone pattern at the current point,
989 and the longest match is then "locked in" for the rest of the overall
990 pattern.
991
992 2. When dealing with multiple paths through the tree simultaneously, it
993 is not straightforward to keep track of captured substrings for the
994 different matching possibilities, and PCRE's implementation of this
995 algorithm does not attempt to do this. This means that no captured sub-
996 strings are available.
997
998 3. Because no substrings are captured, back references within the pat-
999 tern are not supported, and cause errors if encountered.
1000
1001 4. For the same reason, conditional expressions that use a backrefer-
1002 ence as the condition or test for a specific group recursion are not
1003 supported.
1004
1005 5. Because many paths through the tree may be active, the \K escape
1006 sequence, which resets the start of the match when encountered (but may
1007 be on some paths and not on others), is not supported. It causes an
1008 error if encountered.
1009
1010 6. Callouts are supported, but the value of the capture_top field is
1011 always 1, and the value of the capture_last field is always -1.
1012
1013 7. The \C escape sequence, which (in the standard algorithm) always
1014 matches a single data unit, even in UTF-8 or UTF-16 modes, is not sup-
1015 ported in these modes, because the alternative algorithm moves through
1016 the subject string one character (not data unit) at a time, for all
1017 active paths through the tree.
1018
1019 8. Except for (*FAIL), the backtracking control verbs such as (*PRUNE)
1020 are not supported. (*FAIL) is supported, and behaves like a failing
1021 negative assertion.
1022
1023
1024 ADVANTAGES OF THE ALTERNATIVE ALGORITHM
1025
1026 Using the alternative matching algorithm provides the following advan-
1027 tages:
1028
1029 1. All possible matches (at a single point in the subject) are automat-
1030 ically found, and in particular, the longest match is found. To find
1031 more than one match using the standard algorithm, you have to do kludgy
1032 things with callouts.
1033
1034 2. Because the alternative algorithm scans the subject string just
1035 once, and never needs to backtrack (except for lookbehinds), it is pos-
1036 sible to pass very long subject strings to the matching function in
1037 several pieces, checking for partial matching each time. Although it is
1038 possible to do multi-segment matching using the standard algorithm by
1039 retaining partially matched substrings, it is more complicated. The
1040 pcrepartial documentation gives details of partial matching and dis-
1041 cusses multi-segment matching.
1042
1043
1044 DISADVANTAGES OF THE ALTERNATIVE ALGORITHM
1045
1046 The alternative algorithm suffers from a number of disadvantages:
1047
1048 1. It is substantially slower than the standard algorithm. This is
1049 partly because it has to search for all possible matches, but is also
1050 because it is less susceptible to optimization.
1051
1052 2. Capturing parentheses and back references are not supported.
1053
1054 3. Although atomic groups are supported, their use does not provide the
1055 performance advantage that it does for the standard algorithm.
1056
1057
1058 AUTHOR
1059
1060 Philip Hazel
1061 University Computing Service
1062 Cambridge CB2 3QH, England.
1063
1064
1065 REVISION
1066
1067 Last updated: 08 January 2012
1068 Copyright (c) 1997-2012 University of Cambridge.
1069 ------------------------------------------------------------------------------
1070
1071
1072 PCREAPI(3) PCREAPI(3)
1073
1074
1075 NAME
1076 PCRE - Perl-compatible regular expressions
1077
1078 #include <pcre.h>
1079
1080
1081 PCRE NATIVE API BASIC FUNCTIONS
1082
1083 pcre *pcre_compile(const char *pattern, int options,
1084 const char **errptr, int *erroffset,
1085 const unsigned char *tableptr);
1086
1087 pcre *pcre_compile2(const char *pattern, int options,
1088 int *errorcodeptr,
1089 const char **errptr, int *erroffset,
1090 const unsigned char *tableptr);
1091
1092 pcre_extra *pcre_study(const pcre *code, int options,
1093 const char **errptr);
1094
1095 void pcre_free_study(pcre_extra *extra);
1096
1097 int pcre_exec(const pcre *code, const pcre_extra *extra,
1098 const char *subject, int length, int startoffset,
1099 int options, int *ovector, int ovecsize);
1100
1101 int pcre_dfa_exec(const pcre *code, const pcre_extra *extra,
1102 const char *subject, int length, int startoffset,
1103 int options, int *ovector, int ovecsize,
1104 int *workspace, int wscount);
1105
1106
1107 PCRE NATIVE API STRING EXTRACTION FUNCTIONS
1108
1109 int pcre_copy_named_substring(const pcre *code,
1110 const char *subject, int *ovector,
1111 int stringcount, const char *stringname,
1112 char *buffer, int buffersize);
1113
1114 int pcre_copy_substring(const char *subject, int *ovector,
1115 int stringcount, int stringnumber, char *buffer,
1116 int buffersize);
1117
1118 int pcre_get_named_substring(const pcre *code,
1119 const char *subject, int *ovector,
1120 int stringcount, const char *stringname,
1121 const char **stringptr);
1122
1123 int pcre_get_stringnumber(const pcre *code,
1124 const char *name);
1125
1126 int pcre_get_stringtable_entries(const pcre *code,
1127 const char *name, char **first, char **last);
1128
1129 int pcre_get_substring(const char *subject, int *ovector,
1130 int stringcount, int stringnumber,
1131 const char **stringptr);
1132
1133 int pcre_get_substring_list(const char *subject,
1134 int *ovector, int stringcount, const char ***listptr);
1135
1136 void pcre_free_substring(const char *stringptr);
1137
1138 void pcre_free_substring_list(const char **stringptr);
1139
1140
1141 PCRE NATIVE API AUXILIARY FUNCTIONS
1142
1143 pcre_jit_stack *pcre_jit_stack_alloc(int startsize, int maxsize);
1144
1145 void pcre_jit_stack_free(pcre_jit_stack *stack);
1146
1147 void pcre_assign_jit_stack(pcre_extra *extra,
1148 pcre_jit_callback callback, void *data);
1149
1150 const unsigned char *pcre_maketables(void);
1151
1152 int pcre_fullinfo(const pcre *code, const pcre_extra *extra,
1153 int what, void *where);
1154
1155 int pcre_refcount(pcre *code, int adjust);
1156
1157 int pcre_config(int what, void *where);
1158
1159 const char *pcre_version(void);
1160
1161 int pcre_pattern_to_host_byte_order(pcre *code,
1162 pcre_extra *extra, const unsigned char *tables);
1163
1164
1165 PCRE NATIVE API INDIRECTED FUNCTIONS
1166
1167 void *(*pcre_malloc)(size_t);
1168
1169 void (*pcre_free)(void *);
1170
1171 void *(*pcre_stack_malloc)(size_t);
1172
1173 void (*pcre_stack_free)(void *);
1174
1175 int (*pcre_callout)(pcre_callout_block *);
1176
1177
1178 PCRE 8-BIT AND 16-BIT LIBRARIES
1179
1180 From release 8.30, PCRE can be compiled as a library for handling
1181 16-bit character strings as well as, or instead of, the original
1182 library that handles 8-bit character strings. To avoid too much compli-
1183 cation, this document describes the 8-bit versions of the functions,
1184 with only occasional references to the 16-bit library.
1185
1186 The 16-bit functions operate in the same way as their 8-bit counter-
1187 parts; they just use different data types for their arguments and
1188 results, and their names start with pcre16_ instead of pcre_. For every
1189 option that has UTF8 in its name (for example, PCRE_UTF8), there is a
1190 corresponding 16-bit name with UTF8 replaced by UTF16. This facility is
1191 in fact just cosmetic; the 16-bit option names define the same bit val-
1192 ues.
1193
1194 References to bytes and UTF-8 in this document should be read as refer-
1195 ences to 16-bit data quantities and UTF-16 when using the 16-bit
1196 library, unless specified otherwise. More details of the specific dif-
1197 ferences for the 16-bit library are given in the pcre16 page.
1198
1199
1200 PCRE API OVERVIEW
1201
1202 PCRE has its own native API, which is described in this document. There
1203 are also some wrapper functions (for the 8-bit library only) that cor-
1204 respond to the POSIX regular expression API, but they do not give
1205 access to all the functionality. They are described in the pcreposix
1206 documentation. Both of these APIs define a set of C function calls. A
1207 C++ wrapper (again for the 8-bit library only) is also distributed with
1208 PCRE. It is documented in the pcrecpp page.
1209
1210 The native API C function prototypes are defined in the header file
1211 pcre.h, and on Unix-like systems the (8-bit) library itself is called
1212 libpcre. It can normally be accessed by adding -lpcre to the command
1213 for linking an application that uses PCRE. The header file defines the
1214 macros PCRE_MAJOR and PCRE_MINOR to contain the major and minor release
1215 numbers for the library. Applications can use these to include support
1216 for different releases of PCRE.
1217
1218 In a Windows environment, if you want to statically link an application
1219 program against a non-dll pcre.a file, you must define PCRE_STATIC
1220 before including pcre.h or pcrecpp.h, because otherwise the pcre_mal-
1221 loc() and pcre_free() exported functions will be declared
1222 __declspec(dllimport), with unwanted results.
1223
1224 The functions pcre_compile(), pcre_compile2(), pcre_study(), and
1225 pcre_exec() are used for compiling and matching regular expressions in
1226 a Perl-compatible manner. A sample program that demonstrates the sim-
1227 plest way of using them is provided in the file called pcredemo.c in
1228 the PCRE source distribution. A listing of this program is given in the
1229 pcredemo documentation, and the pcresample documentation describes how
1230 to compile and run it.
1231
1232 Just-in-time compiler support is an optional feature of PCRE that can
1233 be built in appropriate hardware environments. It greatly speeds up the
1234 matching performance of many patterns. Simple programs can easily
1235 request that it be used if available, by setting an option that is
1236 ignored when it is not relevant. More complicated programs might need
1237 to make use of the functions pcre_jit_stack_alloc(),
1238 pcre_jit_stack_free(), and pcre_assign_jit_stack() in order to control
1239 the JIT code's memory usage. These functions are discussed in the
1240 pcrejit documentation.
1241
1242 A second matching function, pcre_dfa_exec(), which is not Perl-compati-
1243 ble, is also provided. This uses a different algorithm for the match-
1244 ing. The alternative algorithm finds all possible matches (at a given
1245 point in the subject), and scans the subject just once (unless there
1246 are lookbehind assertions). However, this algorithm does not return
1247 captured substrings. A description of the two matching algorithms and
1248 their advantages and disadvantages is given in the pcrematching docu-
1249 mentation.
1250
1251 In addition to the main compiling and matching functions, there are
1252 convenience functions for extracting captured substrings from a subject
1253 string that is matched by pcre_exec(). They are:
1254
1255 pcre_copy_substring()
1256 pcre_copy_named_substring()
1257 pcre_get_substring()
1258 pcre_get_named_substring()
1259 pcre_get_substring_list()
1260 pcre_get_stringnumber()
1261 pcre_get_stringtable_entries()
1262
1263 pcre_free_substring() and pcre_free_substring_list() are also provided,
1264 to free the memory used for extracted strings.
1265
1266 The function pcre_maketables() is used to build a set of character
1267 tables in the current locale for passing to pcre_compile(),
1268 pcre_exec(), or pcre_dfa_exec(). This is an optional facility that is
1269 provided for specialist use. Most commonly, no special tables are
1270 passed, in which case internal tables that are generated when PCRE is
1271 built are used.
1272
1273 The function pcre_fullinfo() is used to find out information about a
1274 compiled pattern. The function pcre_version() returns a pointer to a
1275 string containing the version of PCRE and its date of release.
1276
1277 The function pcre_refcount() maintains a reference count in a data
1278 block containing a compiled pattern. This is provided for the benefit
1279 of object-oriented applications.
1280
1281 The global variables pcre_malloc and pcre_free initially contain the
1282 entry points of the standard malloc() and free() functions, respec-
1283 tively. PCRE calls the memory management functions via these variables,
1284 so a calling program can replace them if it wishes to intercept the
1285 calls. This should be done before calling any PCRE functions.
1286
1287 The global variables pcre_stack_malloc and pcre_stack_free are also
1288 indirections to memory management functions. These special functions
1289 are used only when PCRE is compiled to use the heap for remembering
1290 data, instead of recursive function calls, when running the pcre_exec()
1291 function. See the pcrebuild documentation for details of how to do
1292 this. It is a non-standard way of building PCRE, for use in environ-
1293 ments that have limited stacks. Because of the greater use of memory
1294 management, it runs more slowly. Separate functions are provided so
1295 that special-purpose external code can be used for this case. When
1296 used, these functions are always called in a stack-like manner (last
1297 obtained, first freed), and always for memory blocks of the same size.
1298 There is a discussion about PCRE's stack usage in the pcrestack docu-
1299 mentation.
1300
1301 The global variable pcre_callout initially contains NULL. It can be set
1302 by the caller to a "callout" function, which PCRE will then call at
1303 specified points during a matching operation. Details are given in the
1304 pcrecallout documentation.
1305
1306
1307 NEWLINES
1308
1309 PCRE supports five different conventions for indicating line breaks in
1310 strings: a single CR (carriage return) character, a single LF (line-
1311 feed) character, the two-character sequence CRLF, any of the three pre-
1312 ceding, or any Unicode newline sequence. The Unicode newline sequences
1313 are the three just mentioned, plus the single characters VT (vertical
1314 tab, U+000B), FF (form feed, U+000C), NEL (next line, U+0085), LS (line
1315 separator, U+2028), and PS (paragraph separator, U+2029).
1316
1317 Each of the first three conventions is used by at least one operating
1318 system as its standard newline sequence. When PCRE is built, a default
1319 can be specified. The default default is LF, which is the Unix stan-
1320 dard. When PCRE is run, the default can be overridden, either when a
1321 pattern is compiled, or when it is matched.
1322
1323 At compile time, the newline convention can be specified by the options
1324 argument of pcre_compile(), or it can be specified by special text at
1325 the start of the pattern itself; this overrides any other settings. See
1326 the pcrepattern page for details of the special character sequences.
1327
1328 In the PCRE documentation the word "newline" is used to mean "the char-
1329 acter or pair of characters that indicate a line break". The choice of
1330 newline convention affects the handling of the dot, circumflex, and
1331 dollar metacharacters, the handling of #-comments in /x mode, and, when
1332 CRLF is a recognized line ending sequence, the match position advance-
1333 ment for a non-anchored pattern. There is more detail about this in the
1334 section on pcre_exec() options below.
1335
1336 The choice of newline convention does not affect the interpretation of
1337 the \n or \r escape sequences, nor does it affect what \R matches,
1338 which is controlled in a similar way, but by separate options.
1339
1340
1341 MULTITHREADING
1342
1343 The PCRE functions can be used in multi-threading applications, with
1344 the proviso that the memory management functions pointed to by
1345 pcre_malloc, pcre_free, pcre_stack_malloc, and pcre_stack_free, and the
1346 callout function pointed to by pcre_callout, are shared by all threads.
1347
1348 The compiled form of a regular expression is not altered during match-
1349 ing, so the same compiled pattern can safely be used by several threads
1350 at once.
1351
1352 If the just-in-time optimization feature is being used, it needs sepa-
1353 rate memory stack areas for each thread. See the pcrejit documentation
1354 for more details.
1355
1356
1357 SAVING PRECOMPILED PATTERNS FOR LATER USE
1358
1359 The compiled form of a regular expression can be saved and re-used at a
1360 later time, possibly by a different program, and even on a host other
1361 than the one on which it was compiled. Details are given in the
1362 pcreprecompile documentation, which includes a description of the
1363 pcre_pattern_to_host_byte_order() function. However, compiling a regu-
1364 lar expression with one version of PCRE for use with a different ver-
1365 sion is not guaranteed to work and may cause crashes.
1366
1367
1368 CHECKING BUILD-TIME OPTIONS
1369
1370 int pcre_config(int what, void *where);
1371
1372 The function pcre_config() makes it possible for a PCRE client to dis-
1373 cover which optional features have been compiled into the PCRE library.
1374 The pcrebuild documentation has more details about these optional fea-
1375 tures.
1376
1377 The first argument for pcre_config() is an integer, specifying which
1378 information is required; the second argument is a pointer to a variable
1379 into which the information is placed. The returned value is zero on
1380 success, or the negative error code PCRE_ERROR_BADOPTION if the value
1381 in the first argument is not recognized. The following information is
1382 available:
1383
1384 PCRE_CONFIG_UTF8
1385
1386 The output is an integer that is set to one if UTF-8 support is avail-
1387 able; otherwise it is set to zero. If this option is given to the
1388 16-bit version of this function, pcre16_config(), the result is
1389 PCRE_ERROR_BADOPTION.
1390
1391 PCRE_CONFIG_UTF16
1392
1393 The output is an integer that is set to one if UTF-16 support is avail-
1394 able; otherwise it is set to zero. This value should normally be given
1395 to the 16-bit version of this function, pcre16_config(). If it is given
1396 to the 8-bit version of this function, the result is PCRE_ERROR_BADOP-
1397 TION.
1398
1399 PCRE_CONFIG_UNICODE_PROPERTIES
1400
1401 The output is an integer that is set to one if support for Unicode
1402 character properties is available; otherwise it is set to zero.
1403
1404 PCRE_CONFIG_JIT
1405
1406 The output is an integer that is set to one if support for just-in-time
1407 compiling is available; otherwise it is set to zero.
1408
1409 PCRE_CONFIG_JITTARGET
1410
1411 The output is a pointer to a zero-terminated "const char *" string. If
1412 JIT support is available, the string contains the name of the architec-
1413 ture for which the JIT compiler is configured, for example "x86 32bit
1414 (little endian + unaligned)". If JIT support is not available, the
1415 result is NULL.
1416
1417 PCRE_CONFIG_NEWLINE
1418
1419 The output is an integer whose value specifies the default character
1420 sequence that is recognized as meaning "newline". The four values that
1421 are supported are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF,
1422 and -1 for ANY. Though they are derived from ASCII, the same values
1423 are returned in EBCDIC environments. The default should normally corre-
1424 spond to the standard sequence for your operating system.
1425
1426 PCRE_CONFIG_BSR
1427
1428 The output is an integer whose value indicates what character sequences
1429 the \R escape sequence matches by default. A value of 0 means that \R
1430 matches any Unicode line ending sequence; a value of 1 means that \R
1431 matches only CR, LF, or CRLF. The default can be overridden when a pat-
1432 tern is compiled or matched.
1433
1434 PCRE_CONFIG_LINK_SIZE
1435
1436 The output is an integer that contains the number of bytes used for
1437 internal linkage in compiled regular expressions. For the 8-bit
1438 library, the value can be 2, 3, or 4. For the 16-bit library, the value
1439 is either 2 or 4 and is still a number of bytes. The default value of 2
1440 is sufficient for all but the most massive patterns, since it allows
1441 the compiled pattern to be up to 64K in size. Larger values allow
1442 larger regular expressions to be compiled, at the expense of slower
1443 matching.
1444
1445 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
1446
1447 The output is an integer that contains the threshold above which the
1448 POSIX interface uses malloc() for output vectors. Further details are
1449 given in the pcreposix documentation.
1450
1451 PCRE_CONFIG_MATCH_LIMIT
1452
1453 The output is a long integer that gives the default limit for the num-
1454 ber of internal matching function calls in a pcre_exec() execution.
1455 Further details are given with pcre_exec() below.
1456
1457 PCRE_CONFIG_MATCH_LIMIT_RECURSION
1458
1459 The output is a long integer that gives the default limit for the depth
1460 of recursion when calling the internal matching function in a
1461 pcre_exec() execution. Further details are given with pcre_exec()
1462 below.
1463
1464 PCRE_CONFIG_STACKRECURSE
1465
1466 The output is an integer that is set to one if internal recursion when
1467 running pcre_exec() is implemented by recursive function calls that use
1468 the stack to remember their state. This is the usual way that PCRE is
1469 compiled. The output is zero if PCRE was compiled to use blocks of data
1470 on the heap instead of recursive function calls. In this case,
1471 pcre_stack_malloc and pcre_stack_free are called to manage memory
1472 blocks on the heap, thus avoiding the use of the stack.
1473
1474
1475 COMPILING A PATTERN
1476
1477 pcre *pcre_compile(const char *pattern, int options,
1478 const char **errptr, int *erroffset,
1479 const unsigned char *tableptr);
1480
1481 pcre *pcre_compile2(const char *pattern, int options,
1482 int *errorcodeptr,
1483 const char **errptr, int *erroffset,
1484 const unsigned char *tableptr);
1485
1486 Either of the functions pcre_compile() or pcre_compile2() can be called
1487 to compile a pattern into an internal form. The only difference between
1488 the two interfaces is that pcre_compile2() has an additional argument,
1489 errorcodeptr, via which a numerical error code can be returned. To
1490 avoid too much repetition, we refer just to pcre_compile() below, but
1491 the information applies equally to pcre_compile2().
1492
1493 The pattern is a C string terminated by a binary zero, and is passed in
1494 the pattern argument. A pointer to a single block of memory that is
1495 obtained via pcre_malloc is returned. This contains the compiled code
1496 and related data. The pcre type is defined for the returned block; this
1497 is a typedef for a structure whose contents are not externally defined.
1498 It is up to the caller to free the memory (via pcre_free) when it is no
1499 longer required.
1500
1501 Although the compiled code of a PCRE regex is relocatable, that is, it
1502 does not depend on memory location, the complete pcre data block is not
1503 fully relocatable, because it may contain a copy of the tableptr argu-
1504 ment, which is an address (see below).
1505
1506 The options argument contains various bit settings that affect the com-
1507 pilation. It should be zero if no options are required. The available
1508 options are described below. Some of them (in particular, those that
1509 are compatible with Perl, but some others as well) can also be set and
1510 unset from within the pattern (see the detailed description in the
1511 pcrepattern documentation). For those options that can be different in
1512 different parts of the pattern, the contents of the options argument
1513 specifies their settings at the start of compilation and execution. The
1514 PCRE_ANCHORED, PCRE_BSR_xxx, PCRE_NEWLINE_xxx, PCRE_NO_UTF8_CHECK, and
1515 PCRE_NO_START_OPTIMIZE options can be set at the time of matching as
1516 well as at compile time.
1517
1518 If errptr is NULL, pcre_compile() returns NULL immediately. Otherwise,
1519 if compilation of a pattern fails, pcre_compile() returns NULL, and
1520 sets the variable pointed to by errptr to point to a textual error mes-
1521 sage. This is a static string that is part of the library. You must not
1522 try to free it. Normally, the offset from the start of the pattern to
1523 the byte that was being processed when the error was discovered is
1524 placed in the variable pointed to by erroffset, which must not be NULL
1525 (if it is, an immediate error is given). However, for an invalid UTF-8
1526 string, the offset is that of the first byte of the failing character.
1527
1528 Some errors are not detected until the whole pattern has been scanned;
1529 in these cases, the offset passed back is the length of the pattern.
1530 Note that the offset is in bytes, not characters, even in UTF-8 mode.
1531 It may sometimes point into the middle of a UTF-8 character.
1532
1533 If pcre_compile2() is used instead of pcre_compile(), and the error-
1534 codeptr argument is not NULL, a non-zero error code number is returned
1535 via this argument in the event of an error. This is in addition to the
1536 textual error message. Error codes and messages are listed below.
1537
1538 If the final argument, tableptr, is NULL, PCRE uses a default set of
1539 character tables that are built when PCRE is compiled, using the
1540 default C locale. Otherwise, tableptr must be an address that is the
1541 result of a call to pcre_maketables(). This value is stored with the
1542 compiled pattern, and used again by pcre_exec(), unless another table
1543 pointer is passed to it. For more discussion, see the section on locale
1544 support below.
1545
1546 This code fragment shows a typical straightforward call to pcre_com-
1547 pile():
1548
1549 pcre *re;
1550 const char *error;
1551 int erroffset;
1552 re = pcre_compile(
1553 "^A.*Z", /* the pattern */
1554 0, /* default options */
1555 &error, /* for error message */
1556 &erroffset, /* for error offset */
1557 NULL); /* use default character tables */
1558
1559 The following names for option bits are defined in the pcre.h header
1560 file:
1561
1562 PCRE_ANCHORED
1563
1564 If this bit is set, the pattern is forced to be "anchored", that is, it
1565 is constrained to match only at the first matching point in the string
1566 that is being searched (the "subject string"). This effect can also be
1567 achieved by appropriate constructs in the pattern itself, which is the
1568 only way to do it in Perl.
1569
1570 PCRE_AUTO_CALLOUT
1571
1572 If this bit is set, pcre_compile() automatically inserts callout items,
1573 all with number 255, before each pattern item. For discussion of the
1574 callout facility, see the pcrecallout documentation.
1575
1576 PCRE_BSR_ANYCRLF
1577 PCRE_BSR_UNICODE
1578
1579 These options (which are mutually exclusive) control what the \R escape
1580 sequence matches. The choice is either to match only CR, LF, or CRLF,
1581 or to match any Unicode newline sequence. The default is specified when
1582 PCRE is built. It can be overridden from within the pattern, or by set-
1583 ting an option when a compiled pattern is matched.
1584
1585 PCRE_CASELESS
1586
1587 If this bit is set, letters in the pattern match both upper and lower
1588 case letters. It is equivalent to Perl's /i option, and it can be
1589 changed within a pattern by a (?i) option setting. In UTF-8 mode, PCRE
1590 always understands the concept of case for characters whose values are
1591 less than 128, so caseless matching is always possible. For characters
1592 with higher values, the concept of case is supported if PCRE is com-
1593 piled with Unicode property support, but not otherwise. If you want to
1594 use caseless matching for characters 128 and above, you must ensure
1595 that PCRE is compiled with Unicode property support as well as with
1596 UTF-8 support.
1597
1598 PCRE_DOLLAR_ENDONLY
1599
1600 If this bit is set, a dollar metacharacter in the pattern matches only
1601 at the end of the subject string. Without this option, a dollar also
1602 matches immediately before a newline at the end of the string (but not
1603 before any other newlines). The PCRE_DOLLAR_ENDONLY option is ignored
1604 if PCRE_MULTILINE is set. There is no equivalent to this option in
1605 Perl, and no way to set it within a pattern.
1606
1607 PCRE_DOTALL
1608
1609 If this bit is set, a dot metacharacter in the pattern matches a char-
1610 acter of any value, including one that indicates a newline. However, it
1611 only ever matches one character, even if newlines are coded as CRLF.
1612 Without this option, a dot does not match when the current position is
1613 at a newline. This option is equivalent to Perl's /s option, and it can
1614 be changed within a pattern by a (?s) option setting. A negative class
1615 such as [^a] always matches newline characters, independent of the set-
1616 ting of this option.
1617
1618 PCRE_DUPNAMES
1619
1620 If this bit is set, names used to identify capturing subpatterns need
1621 not be unique. This can be helpful for certain types of pattern when it
1622 is known that only one instance of the named subpattern can ever be
1623 matched. There are more details of named subpatterns below; see also
1624 the pcrepattern documentation.
1625
1626 PCRE_EXTENDED
1627
1628 If this bit is set, white space data characters in the pattern are
1629 totally ignored except when escaped or inside a character class. White
1630 space does not include the VT character (code 11). In addition, charac-
1631 ters between an unescaped # outside a character class and the next new-
1632 line, inclusive, are also ignored. This is equivalent to Perl's /x
1633 option, and it can be changed within a pattern by a (?x) option set-
1634 ting.
1635
1636 Which characters are interpreted as newlines is controlled by the
1637 options passed to pcre_compile() or by a special sequence at the start
1638 of the pattern, as described in the section entitled "Newline conven-
1639 tions" in the pcrepattern documentation. Note that the end of this type
1640 of comment is a literal newline sequence in the pattern; escape
1641 sequences that happen to represent a newline do not count.
1642
1643 This option makes it possible to include comments inside complicated
1644 patterns. Note, however, that this applies only to data characters.
1645 White space characters may never appear within special character
1646 sequences in a pattern, for example within the sequence (?( that intro-
1647 duces a conditional subpattern.
1648
1649 PCRE_EXTRA
1650
1651 This option was invented in order to turn on additional functionality
1652 of PCRE that is incompatible with Perl, but it is currently of very
1653 little use. When set, any backslash in a pattern that is followed by a
1654 letter that has no special meaning causes an error, thus reserving
1655 these combinations for future expansion. By default, as in Perl, a
1656 backslash followed by a letter with no special meaning is treated as a
1657 literal. (Perl can, however, be persuaded to give an error for this, by
1658 running it with the -w option.) There are at present no other features
1659 controlled by this option. It can also be set by a (?X) option setting
1660 within a pattern.
1661
1662 PCRE_FIRSTLINE
1663
1664 If this option is set, an unanchored pattern is required to match
1665 before or at the first newline in the subject string, though the
1666 matched text may continue over the newline.
1667
1668 PCRE_JAVASCRIPT_COMPAT
1669
1670 If this option is set, PCRE's behaviour is changed in some ways so that
1671 it is compatible with JavaScript rather than Perl. The changes are as
1672 follows:
1673
1674 (1) A lone closing square bracket in a pattern causes a compile-time
1675 error, because this is illegal in JavaScript (by default it is treated
1676 as a data character). Thus, the pattern AB]CD becomes illegal when this
1677 option is set.
1678
1679 (2) At run time, a back reference to an unset subpattern group matches
1680 an empty string (by default this causes the current matching alterna-
1681 tive to fail). A pattern such as (\1)(a) succeeds when this option is
1682 set (assuming it can find an "a" in the subject), whereas it fails by
1683 default, for Perl compatibility.
1684
1685 (3) \U matches an upper case "U" character; by default \U causes a com-
1686 pile time error (Perl uses \U to upper case subsequent characters).
1687
1688 (4) \u matches a lower case "u" character unless it is followed by four
1689 hexadecimal digits, in which case the hexadecimal number defines the
1690 code point to match. By default, \u causes a compile time error (Perl
1691 uses it to upper case the following character).
1692
1693 (5) \x matches a lower case "x" character unless it is followed by two
1694 hexadecimal digits, in which case the hexadecimal number defines the
1695 code point to match. By default, as in Perl, a hexadecimal number is
1696 always expected after \x, but it may have zero, one, or two digits (so,
1697 for example, \xz matches a binary zero character followed by z).
1698
1699 PCRE_MULTILINE
1700
1701 By default, PCRE treats the subject string as consisting of a single
1702 line of characters (even if it actually contains newlines). The "start
1703 of line" metacharacter (^) matches only at the start of the string,
1704 while the "end of line" metacharacter ($) matches only at the end of
1705 the string, or before a terminating newline (unless PCRE_DOLLAR_ENDONLY
1706 is set). This is the same as Perl.
1707
1708 When PCRE_MULTILINE it is set, the "start of line" and "end of line"
1709 constructs match immediately following or immediately before internal
1710 newlines in the subject string, respectively, as well as at the very
1711 start and end. This is equivalent to Perl's /m option, and it can be
1712 changed within a pattern by a (?m) option setting. If there are no new-
1713 lines in a subject string, or no occurrences of ^ or $ in a pattern,
1714 setting PCRE_MULTILINE has no effect.
1715
1716 PCRE_NEWLINE_CR
1717 PCRE_NEWLINE_LF
1718 PCRE_NEWLINE_CRLF
1719 PCRE_NEWLINE_ANYCRLF
1720 PCRE_NEWLINE_ANY
1721
1722 These options override the default newline definition that was chosen
1723 when PCRE was built. Setting the first or the second specifies that a
1724 newline is indicated by a single character (CR or LF, respectively).
1725 Setting PCRE_NEWLINE_CRLF specifies that a newline is indicated by the
1726 two-character CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies
1727 that any of the three preceding sequences should be recognized. Setting
1728 PCRE_NEWLINE_ANY specifies that any Unicode newline sequence should be
1729 recognized. The Unicode newline sequences are the three just mentioned,
1730 plus the single characters VT (vertical tab, U+000B), FF (form feed,
1731 U+000C), NEL (next line, U+0085), LS (line separator, U+2028), and PS
1732 (paragraph separator, U+2029). For the 8-bit library, the last two are
1733 recognized only in UTF-8 mode.
1734
1735 The newline setting in the options word uses three bits that are
1736 treated as a number, giving eight possibilities. Currently only six are
1737 used (default plus the five values above). This means that if you set
1738 more than one newline option, the combination may or may not be sensi-
1739 ble. For example, PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to
1740 PCRE_NEWLINE_CRLF, but other combinations may yield unused numbers and
1741 cause an error.
1742
1743 The only time that a line break in a pattern is specially recognized
1744 when compiling is when PCRE_EXTENDED is set. CR and LF are white space
1745 characters, and so are ignored in this mode. Also, an unescaped # out-
1746 side a character class indicates a comment that lasts until after the
1747 next line break sequence. In other circumstances, line break sequences
1748 in patterns are treated as literal data.
1749
1750 The newline option that is set at compile time becomes the default that
1751 is used for pcre_exec() and pcre_dfa_exec(), but it can be overridden.
1752
1753 PCRE_NO_AUTO_CAPTURE
1754
1755 If this option is set, it disables the use of numbered capturing paren-
1756 theses in the pattern. Any opening parenthesis that is not followed by
1757 ? behaves as if it were followed by ?: but named parentheses can still
1758 be used for capturing (and they acquire numbers in the usual way).
1759 There is no equivalent of this option in Perl.
1760
1761 NO_START_OPTIMIZE
1762
1763 This is an option that acts at matching time; that is, it is really an
1764 option for pcre_exec() or pcre_dfa_exec(). If it is set at compile
1765 time, it is remembered with the compiled pattern and assumed at match-
1766 ing time. For details see the discussion of PCRE_NO_START_OPTIMIZE
1767 below.
1768
1769 PCRE_UCP
1770
1771 This option changes the way PCRE processes \B, \b, \D, \d, \S, \s, \W,
1772 \w, and some of the POSIX character classes. By default, only ASCII
1773 characters are recognized, but if PCRE_UCP is set, Unicode properties
1774 are used instead to classify characters. More details are given in the
1775 section on generic character types in the pcrepattern page. If you set
1776 PCRE_UCP, matching one of the items it affects takes much longer. The
1777 option is available only if PCRE has been compiled with Unicode prop-
1778 erty support.
1779
1780 PCRE_UNGREEDY
1781
1782 This option inverts the "greediness" of the quantifiers so that they
1783 are not greedy by default, but become greedy if followed by "?". It is
1784 not compatible with Perl. It can also be set by a (?U) option setting
1785 within the pattern.
1786
1787 PCRE_UTF8
1788
1789 This option causes PCRE to regard both the pattern and the subject as
1790 strings of UTF-8 characters instead of single-byte strings. However, it
1791 is available only when PCRE is built to include UTF support. If not,
1792 the use of this option provokes an error. Details of how this option
1793 changes the behaviour of PCRE are given in the pcreunicode page.
1794
1795 PCRE_NO_UTF8_CHECK
1796
1797 When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
1798 automatically checked. There is a discussion about the validity of
1799 UTF-8 strings in the pcreunicode page. If an invalid UTF-8 sequence is
1800 found, pcre_compile() returns an error. If you already know that your
1801 pattern is valid, and you want to skip this check for performance rea-
1802 sons, you can set the PCRE_NO_UTF8_CHECK option. When it is set, the
1803 effect of passing an invalid UTF-8 string as a pattern is undefined. It
1804 may cause your program to crash. Note that this option can also be
1805 passed to pcre_exec() and pcre_dfa_exec(), to suppress the validity
1806 checking of subject strings.
1807
1808
1809 COMPILATION ERROR CODES
1810
1811 The following table lists the error codes than may be returned by
1812 pcre_compile2(), along with the error messages that may be returned by
1813 both compiling functions. Note that error messages are always 8-bit
1814 ASCII strings, even in 16-bit mode. As PCRE has developed, some error
1815 codes have fallen out of use. To avoid confusion, they have not been
1816 re-used.
1817
1818 0 no error
1819 1 \ at end of pattern
1820 2 \c at end of pattern
1821 3 unrecognized character follows \
1822 4 numbers out of order in {} quantifier
1823 5 number too big in {} quantifier
1824 6 missing terminating ] for character class
1825 7 invalid escape sequence in character class
1826 8 range out of order in character class
1827 9 nothing to repeat
1828 10 [this code is not in use]
1829 11 internal error: unexpected repeat
1830 12 unrecognized character after (? or (?-
1831 13 POSIX named classes are supported only within a class
1832 14 missing )
1833 15 reference to non-existent subpattern
1834 16 erroffset passed as NULL
1835 17 unknown option bit(s) set
1836 18 missing ) after comment
1837 19 [this code is not in use]
1838 20 regular expression is too large
1839 21 failed to get memory
1840 22 unmatched parentheses
1841 23 internal error: code overflow
1842 24 unrecognized character after (?<
1843 25 lookbehind assertion is not fixed length
1844 26 malformed number or name after (?(
1845 27 conditional group contains more than two branches
1846 28 assertion expected after (?(
1847 29 (?R or (?[+-]digits must be followed by )
1848 30 unknown POSIX class name
1849 31 POSIX collating elements are not supported
1850 32 this version of PCRE is compiled without UTF support
1851 33 [this code is not in use]
1852 34 character value in \x{...} sequence is too large
1853 35 invalid condition (?(0)
1854 36 \C not allowed in lookbehind assertion
1855 37 PCRE does not support \L, \l, \N{name}, \U, or \u
1856 38 number after (?C is > 255
1857 39 closing ) for (?C expected
1858 40 recursive call could loop indefinitely
1859 41 unrecognized character after (?P
1860 42 syntax error in subpattern name (missing terminator)
1861 43 two named subpatterns have the same name
1862 44 invalid UTF-8 string (specifically UTF-8)
1863 45 support for \P, \p, and \X has not been compiled
1864 46 malformed \P or \p sequence
1865 47 unknown property name after \P or \p
1866 48 subpattern name is too long (maximum 32 characters)
1867 49 too many named subpatterns (maximum 10000)
1868 50 [this code is not in use]
1869 51 octal value is greater than \377 in 8-bit non-UTF-8 mode
1870 52 internal error: overran compiling workspace
1871 53 internal error: previously-checked referenced subpattern
1872 not found
1873 54 DEFINE group contains more than one branch
1874 55 repeating a DEFINE group is not allowed
1875 56 inconsistent NEWLINE options
1876 57 \g is not followed by a braced, angle-bracketed, or quoted
1877 name/number or by a plain number
1878 58 a numbered reference must not be zero
1879 59 an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
1880 60 (*VERB) not recognized
1881 61 number is too big
1882 62 subpattern name expected
1883 63 digit expected after (?+
1884 64 ] is an invalid data character in JavaScript compatibility mode
1885 65 different names for subpatterns of the same number are
1886 not allowed
1887 66 (*MARK) must have an argument
1888 67 this version of PCRE is not compiled with Unicode property
1889 support
1890 68 \c must be followed by an ASCII character
1891 69 \k is not followed by a braced, angle-bracketed, or quoted name
1892 70 internal error: unknown opcode in find_fixedlength()
1893 71 \N is not supported in a class
1894 72 too many forward references
1895 73 disallowed Unicode code point (>= 0xd800 && <= 0xdfff)
1896 74 invalid UTF-16 string (specifically UTF-16)
1897 75 name is too long in (*MARK), (*PRUNE), (*SKIP), or (*THEN)
1898 76 character value in \u.... sequence is too large
1899
1900 The numbers 32 and 10000 in errors 48 and 49 are defaults; different
1901 values may be used if the limits were changed when PCRE was built.
1902
1903
1904 STUDYING A PATTERN
1905
1906 pcre_extra *pcre_study(const pcre *code, int options
1907 const char **errptr);
1908
1909 If a compiled pattern is going to be used several times, it is worth
1910 spending more time analyzing it in order to speed up the time taken for
1911 matching. The function pcre_study() takes a pointer to a compiled pat-
1912 tern as its first argument. If studying the pattern produces additional
1913 information that will help speed up matching, pcre_study() returns a
1914 pointer to a pcre_extra block, in which the study_data field points to
1915 the results of the study.
1916
1917 The returned value from pcre_study() can be passed directly to
1918 pcre_exec() or pcre_dfa_exec(). However, a pcre_extra block also con-
1919 tains other fields that can be set by the caller before the block is
1920 passed; these are described below in the section on matching a pattern.
1921
1922 If studying the pattern does not produce any useful information,
1923 pcre_study() returns NULL. In that circumstance, if the calling program
1924 wants to pass any of the other fields to pcre_exec() or
1925 pcre_dfa_exec(), it must set up its own pcre_extra block.
1926
1927 The second argument of pcre_study() contains option bits. There are
1928 three options:
1929
1930 PCRE_STUDY_JIT_COMPILE
1931 PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE
1932 PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE
1933
1934 If any of these are set, and the just-in-time compiler is available,
1935 the pattern is further compiled into machine code that executes much
1936 faster than the pcre_exec() interpretive matching function. If the
1937 just-in-time compiler is not available, these options are ignored. All
1938 other bits in the options argument must be zero.
1939
1940 JIT compilation is a heavyweight optimization. It can take some time
1941 for patterns to be analyzed, and for one-off matches and simple pat-
1942 terns the benefit of faster execution might be offset by a much slower
1943 study time. Not all patterns can be optimized by the JIT compiler. For
1944 those that cannot be handled, matching automatically falls back to the
1945 pcre_exec() interpreter. For more details, see the pcrejit documenta-
1946 tion.
1947
1948 The third argument for pcre_study() is a pointer for an error message.
1949 If studying succeeds (even if no data is returned), the variable it
1950 points to is set to NULL. Otherwise it is set to point to a textual
1951 error message. This is a static string that is part of the library. You
1952 must not try to free it. You should test the error pointer for NULL
1953 after calling pcre_study(), to be sure that it has run successfully.
1954
1955 When you are finished with a pattern, you can free the memory used for
1956 the study data by calling pcre_free_study(). This function was added to
1957 the API for release 8.20. For earlier versions, the memory could be
1958 freed with pcre_free(), just like the pattern itself. This will still
1959 work in cases where JIT optimization is not used, but it is advisable
1960 to change to the new function when convenient.
1961
1962 This is a typical way in which pcre_study() is used (except that in a
1963 real application there should be tests for errors):
1964
1965 int rc;
1966 pcre *re;
1967 pcre_extra *sd;
1968 re = pcre_compile("pattern", 0, &error, &erroroffset, NULL);
1969 sd = pcre_study(
1970 re, /* result of pcre_compile() */
1971 0, /* no options */
1972 &error); /* set to NULL or points to a message */
1973 rc = pcre_exec( /* see below for details of pcre_exec() options */
1974 re, sd, "subject", 7, 0, 0, ovector, 30);
1975 ...
1976 pcre_free_study(sd);
1977 pcre_free(re);
1978
1979 Studying a pattern does two things: first, a lower bound for the length
1980 of subject string that is needed to match the pattern is computed. This
1981 does not mean that there are any strings of that length that match, but
1982 it does guarantee that no shorter strings match. The value is used by
1983 pcre_exec() and pcre_dfa_exec() to avoid wasting time by trying to
1984 match strings that are shorter than the lower bound. You can find out
1985 the value in a calling program via the pcre_fullinfo() function.
1986
1987 Studying a pattern is also useful for non-anchored patterns that do not
1988 have a single fixed starting character. A bitmap of possible starting
1989 bytes is created. This speeds up finding a position in the subject at
1990 which to start matching. (In 16-bit mode, the bitmap is used for 16-bit
1991 values less than 256.)
1992
1993 These two optimizations apply to both pcre_exec() and pcre_dfa_exec(),
1994 and the information is also used by the JIT compiler. The optimiza-
1995 tions can be disabled by setting the PCRE_NO_START_OPTIMIZE option when
1996 calling pcre_exec() or pcre_dfa_exec(), but if this is done, JIT execu-
1997 tion is also disabled. You might want to do this if your pattern con-
1998 tains callouts or (*MARK) and you want to make use of these facilities
1999 in cases where matching fails. See the discussion of
2000 PCRE_NO_START_OPTIMIZE below.
2001
2002
2003 LOCALE SUPPORT
2004
2005 PCRE handles caseless matching, and determines whether characters are
2006 letters, digits, or whatever, by reference to a set of tables, indexed
2007 by character value. When running in UTF-8 mode, this applies only to
2008 characters with codes less than 128. By default, higher-valued codes
2009 never match escapes such as \w or \d, but they can be tested with \p if
2010 PCRE is built with Unicode character property support. Alternatively,
2011 the PCRE_UCP option can be set at compile time; this causes \w and
2012 friends to use Unicode property support instead of built-in tables. The
2013 use of locales with Unicode is discouraged. If you are handling charac-
2014 ters with codes greater than 128, you should either use UTF-8 and Uni-
2015 code, or use locales, but not try to mix the two.
2016
2017 PCRE contains an internal set of tables that are used when the final
2018 argument of pcre_compile() is NULL. These are sufficient for many
2019 applications. Normally, the internal tables recognize only ASCII char-
2020 acters. However, when PCRE is built, it is possible to cause the inter-
2021 nal tables to be rebuilt in the default "C" locale of the local system,
2022 which may cause them to be different.
2023
2024 The internal tables can always be overridden by tables supplied by the
2025 application that calls PCRE. These may be created in a different locale
2026 from the default. As more and more applications change to using Uni-
2027 code, the need for this locale support is expected to die away.
2028
2029 External tables are built by calling the pcre_maketables() function,
2030 which has no arguments, in the relevant locale. The result can then be
2031 passed to pcre_compile() or pcre_exec() as often as necessary. For
2032 example, to build and use tables that are appropriate for the French
2033 locale (where accented characters with values greater than 128 are
2034 treated as letters), the following code could be used:
2035
2036 setlocale(LC_CTYPE, "fr_FR");
2037 tables = pcre_maketables();
2038 re = pcre_compile(..., tables);
2039
2040 The locale name "fr_FR" is used on Linux and other Unix-like systems;
2041 if you are using Windows, the name for the French locale is "french".
2042
2043 When pcre_maketables() runs, the tables are built in memory that is
2044 obtained via pcre_malloc. It is the caller's responsibility to ensure
2045 that the memory containing the tables remains available for as long as
2046 it is needed.
2047
2048 The pointer that is passed to pcre_compile() is saved with the compiled
2049 pattern, and the same tables are used via this pointer by pcre_study()
2050 and normally also by pcre_exec(). Thus, by default, for any single pat-
2051 tern, compilation, studying and matching all happen in the same locale,
2052 but different patterns can be compiled in different locales.
2053
2054 It is possible to pass a table pointer or NULL (indicating the use of
2055 the internal tables) to pcre_exec(). Although not intended for this
2056 purpose, this facility could be used to match a pattern in a different
2057 locale from the one in which it was compiled. Passing table pointers at
2058 run time is discussed below in the section on matching a pattern.
2059
2060
2061 INFORMATION ABOUT A PATTERN
2062
2063 int pcre_fullinfo(const pcre *code, const pcre_extra *extra,
2064 int what, void *where);
2065
2066 The pcre_fullinfo() function returns information about a compiled pat-
2067 tern. It replaces the pcre_info() function, which was removed from the
2068 library at version 8.30, after more than 10 years of obsolescence.
2069
2070 The first argument for pcre_fullinfo() is a pointer to the compiled
2071 pattern. The second argument is the result of pcre_study(), or NULL if
2072 the pattern was not studied. The third argument specifies which piece
2073 of information is required, and the fourth argument is a pointer to a
2074 variable to receive the data. The yield of the function is zero for
2075 success, or one of the following negative numbers:
2076
2077 PCRE_ERROR_NULL the argument code was NULL
2078 the argument where was NULL
2079 PCRE_ERROR_BADMAGIC the "magic number" was not found
2080 PCRE_ERROR_BADENDIANNESS the pattern was compiled with different
2081 endianness
2082 PCRE_ERROR_BADOPTION the value of what was invalid
2083
2084 The "magic number" is placed at the start of each compiled pattern as
2085 an simple check against passing an arbitrary memory pointer. The endi-
2086 anness error can occur if a compiled pattern is saved and reloaded on a
2087 different host. Here is a typical call of pcre_fullinfo(), to obtain
2088 the length of the compiled pattern:
2089
2090 int rc;
2091 size_t length;
2092 rc = pcre_fullinfo(
2093 re, /* result of pcre_compile() */
2094 sd, /* result of pcre_study(), or NULL */
2095 PCRE_INFO_SIZE, /* what is required */
2096 &length); /* where to put the data */
2097
2098 The possible values for the third argument are defined in pcre.h, and
2099 are as follows:
2100
2101 PCRE_INFO_BACKREFMAX
2102
2103 Return the number of the highest back reference in the pattern. The
2104 fourth argument should point to an int variable. Zero is returned if
2105 there are no back references.
2106
2107 PCRE_INFO_CAPTURECOUNT
2108
2109 Return the number of capturing subpatterns in the pattern. The fourth
2110 argument should point to an int variable.
2111
2112 PCRE_INFO_DEFAULT_TABLES
2113
2114 Return a pointer to the internal default character tables within PCRE.
2115 The fourth argument should point to an unsigned char * variable. This
2116 information call is provided for internal use by the pcre_study() func-
2117 tion. External callers can cause PCRE to use its internal tables by
2118 passing a NULL table pointer.
2119
2120 PCRE_INFO_FIRSTBYTE
2121
2122 Return information about the first data unit of any matched string, for
2123 a non-anchored pattern. (The name of this option refers to the 8-bit
2124 library, where data units are bytes.) The fourth argument should point
2125 to an int variable.
2126
2127 If there is a fixed first value, for example, the letter "c" from a
2128 pattern such as (cat|cow|coyote), its value is returned. In the 8-bit
2129 library, the value is always less than 256; in the 16-bit library the
2130 value can be up to 0xffff.
2131
2132 If there is no fixed first value, and if either
2133
2134 (a) the pattern was compiled with the PCRE_MULTILINE option, and every
2135 branch starts with "^", or
2136
2137 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not
2138 set (if it were set, the pattern would be anchored),
2139
2140 -1 is returned, indicating that the pattern matches only at the start
2141 of a subject string or after any newline within the string. Otherwise
2142 -2 is returned. For anchored patterns, -2 is returned.
2143
2144 PCRE_INFO_FIRSTTABLE
2145
2146 If the pattern was studied, and this resulted in the construction of a
2147 256-bit table indicating a fixed set of values for the first data unit
2148 in any matching string, a pointer to the table is returned. Otherwise
2149 NULL is returned. The fourth argument should point to an unsigned char
2150 * variable.
2151
2152 PCRE_INFO_HASCRORLF
2153
2154 Return 1 if the pattern contains any explicit matches for CR or LF
2155 characters, otherwise 0. The fourth argument should point to an int
2156 variable. An explicit match is either a literal CR or LF character, or
2157 \r or \n.
2158
2159 PCRE_INFO_JCHANGED
2160
2161 Return 1 if the (?J) or (?-J) option setting is used in the pattern,
2162 otherwise 0. The fourth argument should point to an int variable. (?J)
2163 and (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
2164
2165 PCRE_INFO_JIT
2166
2167 Return 1 if the pattern was studied with one of the JIT options, and
2168 just-in-time compiling was successful. The fourth argument should point
2169 to an int variable. A return value of 0 means that JIT support is not
2170 available in this version of PCRE, or that the pattern was not studied
2171 with a JIT option, or that the JIT compiler could not handle this par-
2172 ticular pattern. See the pcrejit documentation for details of what can
2173 and cannot be handled.
2174
2175 PCRE_INFO_JITSIZE
2176
2177 If the pattern was successfully studied with a JIT option, return the
2178 size of the JIT compiled code, otherwise return zero. The fourth argu-
2179 ment should point to a size_t variable.
2180
2181 PCRE_INFO_LASTLITERAL
2182
2183 Return the value of the rightmost literal data unit that must exist in
2184 any matched string, other than at its start, if such a value has been
2185 recorded. The fourth argument should point to an int variable. If there
2186 is no such value, -1 is returned. For anchored patterns, a last literal
2187 value is recorded only if it follows something of variable length. For
2188 example, for the pattern /^a\d+z\d+/ the returned value is "z", but for
2189 /^a\dz\d/ the returned value is -1.
2190
2191 PCRE_INFO_MAXLOOKBEHIND
2192
2193 Return the number of characters (NB not bytes) in the longest lookbe-
2194 hind assertion in the pattern. Note that the simple assertions \b and
2195 \B require a one-character lookbehind. This information is useful when
2196 doing multi-segment matching using the partial matching facilities.
2197
2198 PCRE_INFO_MINLENGTH
2199
2200 If the pattern was studied and a minimum length for matching subject
2201 strings was computed, its value is returned. Otherwise the returned
2202 value is -1. The value is a number of characters, which in UTF-8 mode
2203 may be different from the number of bytes. The fourth argument should
2204 point to an int variable. A non-negative value is a lower bound to the
2205 length of any matching string. There may not be any strings of that
2206 length that do actually match, but every string that does match is at
2207 least that long.
2208
2209 PCRE_INFO_NAMECOUNT
2210 PCRE_INFO_NAMEENTRYSIZE
2211 PCRE_INFO_NAMETABLE
2212
2213 PCRE supports the use of named as well as numbered capturing parenthe-
2214 ses. The names are just an additional way of identifying the parenthe-
2215 ses, which still acquire numbers. Several convenience functions such as
2216 pcre_get_named_substring() are provided for extracting captured sub-
2217 strings by name. It is also possible to extract the data directly, by
2218 first converting the name to a number in order to access the correct
2219 pointers in the output vector (described with pcre_exec() below). To do
2220 the conversion, you need to use the name-to-number map, which is
2221 described by these three values.
2222
2223 The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT
2224 gives the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size
2225 of each entry; both of these return an int value. The entry size
2226 depends on the length of the longest name. PCRE_INFO_NAMETABLE returns
2227 a pointer to the first entry of the table. This is a pointer to char in
2228 the 8-bit library, where the first two bytes of each entry are the num-
2229 ber of the capturing parenthesis, most significant byte first. In the
2230 16-bit library, the pointer points to 16-bit data units, the first of
2231 which contains the parenthesis number. The rest of the entry is the
2232 corresponding name, zero terminated.
2233
2234 The names are in alphabetical order. Duplicate names may appear if (?|
2235 is used to create multiple groups with the same number, as described in
2236 the section on duplicate subpattern numbers in the pcrepattern page.
2237 Duplicate names for subpatterns with different numbers are permitted
2238 only if PCRE_DUPNAMES is set. In all cases of duplicate names, they
2239 appear in the table in the order in which they were found in the pat-
2240 tern. In the absence of (?| this is the order of increasing number;
2241 when (?| is used this is not necessarily the case because later subpat-
2242 terns may have lower numbers.
2243
2244 As a simple example of the name/number table, consider the following
2245 pattern after compilation by the 8-bit library (assume PCRE_EXTENDED is
2246 set, so white space - including newlines - is ignored):
2247
2248 (?<date> (?<year>(\d\d)?\d\d) -
2249 (?<month>\d\d) - (?<day>\d\d) )
2250
2251 There are four named subpatterns, so the table has four entries, and
2252 each entry in the table is eight bytes long. The table is as follows,
2253 with non-printing bytes shows in hexadecimal, and undefined bytes shown
2254 as ??:
2255
2256 00 01 d a t e 00 ??
2257 00 05 d a y 00 ?? ??
2258 00 04 m o n t h 00
2259 00 02 y e a r 00 ??
2260
2261 When writing code to extract data from named subpatterns using the
2262 name-to-number map, remember that the length of the entries is likely
2263 to be different for each compiled pattern.
2264
2265 PCRE_INFO_OKPARTIAL
2266
2267 Return 1 if the pattern can be used for partial matching with
2268 pcre_exec(), otherwise 0. The fourth argument should point to an int
2269 variable. From release 8.00, this always returns 1, because the
2270 restrictions that previously applied to partial matching have been
2271 lifted. The pcrepartial documentation gives details of partial match-
2272 ing.
2273
2274 PCRE_INFO_OPTIONS
2275
2276 Return a copy of the options with which the pattern was compiled. The
2277 fourth argument should point to an unsigned long int variable. These
2278 option bits are those specified in the call to pcre_compile(), modified
2279 by any top-level option settings at the start of the pattern itself. In
2280 other words, they are the options that will be in force when matching
2281 starts. For example, if the pattern /(?im)abc(?-i)d/ is compiled with
2282 the PCRE_EXTENDED option, the result is PCRE_CASELESS, PCRE_MULTILINE,
2283 and PCRE_EXTENDED.
2284
2285 A pattern is automatically anchored by PCRE if all of its top-level
2286 alternatives begin with one of the following:
2287
2288 ^ unless PCRE_MULTILINE is set
2289 \A always
2290 \G always
2291 .* if PCRE_DOTALL is set and there are no back
2292 references to the subpattern in which .* appears
2293
2294 For such patterns, the PCRE_ANCHORED bit is set in the options returned
2295 by pcre_fullinfo().
2296
2297 PCRE_INFO_SIZE
2298
2299 Return the size of the compiled pattern in bytes (for both libraries).
2300 The fourth argument should point to a size_t variable. This value does
2301 not include the size of the pcre structure that is returned by
2302 pcre_compile(). The value that is passed as the argument to pcre_mal-
2303 loc() when pcre_compile() is getting memory in which to place the com-
2304 piled data is the value returned by this option plus the size of the
2305 pcre structure. Studying a compiled pattern, with or without JIT, does
2306 not alter the value returned by this option.
2307
2308 PCRE_INFO_STUDYSIZE
2309
2310 Return the size in bytes of the data block pointed to by the study_data
2311 field in a pcre_extra block. If pcre_extra is NULL, or there is no
2312 study data, zero is returned. The fourth argument should point to a
2313 size_t variable. The study_data field is set by pcre_study() to record
2314 information that will speed up matching (see the section entitled
2315 "Studying a pattern" above). The format of the study_data block is pri-
2316 vate, but its length is made available via this option so that it can
2317 be saved and restored (see the pcreprecompile documentation for
2318 details).
2319
2320
2321 REFERENCE COUNTS
2322
2323 int pcre_refcount(pcre *code, int adjust);
2324
2325 The pcre_refcount() function is used to maintain a reference count in
2326 the data block that contains a compiled pattern. It is provided for the
2327 benefit of applications that operate in an object-oriented manner,
2328 where different parts of the application may be using the same compiled
2329 pattern, but you want to free the block when they are all done.
2330
2331 When a pattern is compiled, the reference count field is initialized to
2332 zero. It is changed only by calling this function, whose action is to
2333 add the adjust value (which may be positive or negative) to it. The
2334 yield of the function is the new value. However, the value of the count
2335 is constrained to lie between 0 and 65535, inclusive. If the new value
2336 is outside these limits, it is forced to the appropriate limit value.
2337
2338 Except when it is zero, the reference count is not correctly preserved
2339 if a pattern is compiled on one host and then transferred to a host
2340 whose byte-order is different. (This seems a highly unlikely scenario.)
2341
2342
2343 MATCHING A PATTERN: THE TRADITIONAL FUNCTION
2344
2345 int pcre_exec(const pcre *code, const pcre_extra *extra,
2346 const char *subject, int length, int startoffset,
2347 int options, int *ovector, int ovecsize);
2348
2349 The function pcre_exec() is called to match a subject string against a
2350 compiled pattern, which is passed in the code argument. If the pattern
2351 was studied, the result of the study should be passed in the extra
2352 argument. You can call pcre_exec() with the same code and extra argu-
2353 ments as many times as you like, in order to match different subject
2354 strings with the same pattern.
2355
2356 This function is the main matching facility of the library, and it
2357 operates in a Perl-like manner. For specialist use there is also an
2358 alternative matching function, which is described below in the section
2359 about the pcre_dfa_exec() function.
2360
2361 In most applications, the pattern will have been compiled (and option-
2362 ally studied) in the same process that calls pcre_exec(). However, it
2363 is possible to save compiled patterns and study data, and then use them
2364 later in different processes, possibly even on different hosts. For a
2365 discussion about this, see the pcreprecompile documentation.
2366
2367 Here is an example of a simple call to pcre_exec():
2368
2369 int rc;
2370 int ovector[30];
2371 rc = pcre_exec(
2372 re, /* result of pcre_compile() */
2373 NULL, /* we didn't study the pattern */
2374 "some string", /* the subject string */
2375 11, /* the length of the subject string */
2376 0, /* start at offset 0 in the subject */
2377 0, /* default options */
2378 ovector, /* vector of integers for substring information */
2379 30); /* number of elements (NOT size in bytes) */
2380
2381 Extra data for pcre_exec()
2382
2383 If the extra argument is not NULL, it must point to a pcre_extra data
2384 block. The pcre_study() function returns such a block (when it doesn't
2385 return NULL), but you can also create one for yourself, and pass addi-
2386 tional information in it. The pcre_extra block contains the following
2387 fields (not necessarily in this order):
2388
2389 unsigned long int flags;
2390 void *study_data;
2391 void *executable_jit;
2392 unsigned long int match_limit;
2393 unsigned long int match_limit_recursion;
2394 void *callout_data;
2395 const unsigned char *tables;
2396 unsigned char **mark;
2397
2398 In the 16-bit version of this structure, the mark field has type
2399 "PCRE_UCHAR16 **".
2400
2401 The flags field is used to specify which of the other fields are set.
2402 The flag bits are:
2403
2404 PCRE_EXTRA_CALLOUT_DATA
2405 PCRE_EXTRA_EXECUTABLE_JIT
2406 PCRE_EXTRA_MARK
2407 PCRE_EXTRA_MATCH_LIMIT
2408 PCRE_EXTRA_MATCH_LIMIT_RECURSION
2409 PCRE_EXTRA_STUDY_DATA
2410 PCRE_EXTRA_TABLES
2411
2412 Other flag bits should be set to zero. The study_data field and some-
2413 times the executable_jit field are set in the pcre_extra block that is
2414 returned by pcre_study(), together with the appropriate flag bits. You
2415 should not set these yourself, but you may add to the block by setting
2416 other fields and their corresponding flag bits.
2417
2418 The match_limit field provides a means of preventing PCRE from using up
2419 a vast amount of resources when running patterns that are not going to
2420 match, but which have a very large number of possibilities in their
2421 search trees. The classic example is a pattern that uses nested unlim-
2422 ited repeats.
2423
2424 Internally, pcre_exec() uses a function called match(), which it calls
2425 repeatedly (sometimes recursively). The limit set by match_limit is
2426 imposed on the number of times this function is called during a match,
2427 which has the effect of limiting the amount of backtracking that can
2428 take place. For patterns that are not anchored, the count restarts from
2429 zero for each position in the subject string.
2430
2431 When pcre_exec() is called with a pattern that was successfully studied
2432 with a JIT option, the way that the matching is executed is entirely
2433 different. However, there is still the possibility of runaway matching
2434 that goes on for a very long time, and so the match_limit value is also
2435 used in this case (but in a different way) to limit how long the match-
2436 ing can continue.
2437
2438 The default value for the limit can be set when PCRE is built; the
2439 default default is 10 million, which handles all but the most extreme
2440 cases. You can override the default by suppling pcre_exec() with a
2441 pcre_extra block in which match_limit is set, and
2442 PCRE_EXTRA_MATCH_LIMIT is set in the flags field. If the limit is
2443 exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
2444
2445 The match_limit_recursion field is similar to match_limit, but instead
2446 of limiting the total number of times that match() is called, it limits
2447 the depth of recursion. The recursion depth is a smaller number than
2448 the total number of calls, because not all calls to match() are recur-
2449 sive. This limit is of use only if it is set smaller than match_limit.
2450
2451 Limiting the recursion depth limits the amount of machine stack that
2452 can be used, or, when PCRE has been compiled to use memory on the heap
2453 instead of the stack, the amount of heap memory that can be used. This
2454 limit is not relevant, and is ignored, when matching is done using JIT
2455 compiled code.
2456
2457 The default value for match_limit_recursion can be set when PCRE is
2458 built; the default default is the same value as the default for
2459 match_limit. You can override the default by suppling pcre_exec() with
2460 a pcre_extra block in which match_limit_recursion is set, and
2461 PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the flags field. If the
2462 limit is exceeded, pcre_exec() returns PCRE_ERROR_RECURSIONLIMIT.
2463
2464 The callout_data field is used in conjunction with the "callout" fea-
2465 ture, and is described in the pcrecallout documentation.
2466
2467 The tables field is used to pass a character tables pointer to
2468 pcre_exec(); this overrides the value that is stored with the compiled
2469 pattern. A non-NULL value is stored with the compiled pattern only if
2470 custom tables were supplied to pcre_compile() via its tableptr argu-
2471 ment. If NULL is passed to pcre_exec() using this mechanism, it forces
2472 PCRE's internal tables to be used. This facility is helpful when re-
2473 using patterns that have been saved after compiling with an external
2474 set of tables, because the external tables might be at a different
2475 address when pcre_exec() is called. See the pcreprecompile documenta-
2476 tion for a discussion of saving compiled patterns for later use.
2477
2478 If PCRE_EXTRA_MARK is set in the flags field, the mark field must be
2479 set to point to a suitable variable. If the pattern contains any back-
2480 tracking control verbs such as (*MARK:NAME), and the execution ends up
2481 with a name to pass back, a pointer to the name string (zero termi-
2482 nated) is placed in the variable pointed to by the mark field. The
2483 names are within the compiled pattern; if you wish to retain such a
2484 name you must copy it before freeing the memory of a compiled pattern.
2485 If there is no name to pass back, the variable pointed to by the mark
2486 field is set to NULL. For details of the backtracking control verbs,
2487 see the section entitled "Backtracking control" in the pcrepattern doc-
2488 umentation.
2489
2490 Option bits for pcre_exec()
2491
2492 The unused bits of the options argument for pcre_exec() must be zero.
2493 The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_xxx,
2494 PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
2495 PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_HARD, and
2496 PCRE_PARTIAL_SOFT.
2497
2498 If the pattern was successfully studied with one of the just-in-time
2499 (JIT) compile options, the only supported options for JIT execution are
2500 PCRE_NO_UTF8_CHECK, PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY,
2501 PCRE_NOTEMPTY_ATSTART, PCRE_PARTIAL_HARD, and PCRE_PARTIAL_SOFT. If an
2502 unsupported option is used, JIT execution is disabled and the normal
2503 interpretive code in pcre_exec() is run.
2504
2505 PCRE_ANCHORED
2506
2507 The PCRE_ANCHORED option limits pcre_exec() to matching at the first
2508 matching position. If a pattern was compiled with PCRE_ANCHORED, or
2509 turned out to be anchored by virtue of its contents, it cannot be made
2510 unachored at matching time.
2511
2512 PCRE_BSR_ANYCRLF
2513 PCRE_BSR_UNICODE
2514
2515 These options (which are mutually exclusive) control what the \R escape
2516 sequence matches. The choice is either to match only CR, LF, or CRLF,
2517 or to match any Unicode newline sequence. These options override the
2518 choice that was made or defaulted when the pattern was compiled.
2519
2520 PCRE_NEWLINE_CR
2521 PCRE_NEWLINE_LF
2522 PCRE_NEWLINE_CRLF
2523 PCRE_NEWLINE_ANYCRLF
2524 PCRE_NEWLINE_ANY
2525
2526 These options override the newline definition that was chosen or
2527 defaulted when the pattern was compiled. For details, see the descrip-
2528 tion of pcre_compile() above. During matching, the newline choice
2529 affects the behaviour of the dot, circumflex, and dollar metacharac-
2530 ters. It may also alter the way the match position is advanced after a
2531 match failure for an unanchored pattern.
2532
2533 When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is
2534 set, and a match attempt for an unanchored pattern fails when the cur-
2535 rent position is at a CRLF sequence, and the pattern contains no
2536 explicit matches for CR or LF characters, the match position is
2537 advanced by two characters instead of one, in other words, to after the
2538 CRLF.
2539
2540 The above rule is a compromise that makes the most common cases work as
2541 expected. For example, if the pattern is .+A (and the PCRE_DOTALL
2542 option is not set), it does not match the string "\r\nA" because, after
2543 failing at the start, it skips both the CR and the LF before retrying.
2544 However, the pattern [\r\n]A does match that string, because it con-
2545 tains an explicit CR or LF reference, and so advances only by one char-
2546 acter after the first failure.
2547
2548 An explicit match for CR of LF is either a literal appearance of one of
2549 those characters, or one of the \r or \n escape sequences. Implicit
2550 matches such as [^X] do not count, nor does \s (which includes CR and
2551 LF in the characters that it matches).
2552
2553 Notwithstanding the above, anomalous effects may still occur when CRLF
2554 is a valid newline sequence and explicit \r or \n escapes appear in the
2555 pattern.
2556
2557 PCRE_NOTBOL
2558
2559 This option specifies that first character of the subject string is not
2560 the beginning of a line, so the circumflex metacharacter should not
2561 match before it. Setting this without PCRE_MULTILINE (at compile time)
2562 causes circumflex never to match. This option affects only the behav-
2563 iour of the circumflex metacharacter. It does not affect \A.
2564
2565 PCRE_NOTEOL
2566
2567 This option specifies that the end of the subject string is not the end
2568 of a line, so the dollar metacharacter should not match it nor (except
2569 in multiline mode) a newline immediately before it. Setting this with-
2570 out PCRE_MULTILINE (at compile time) causes dollar never to match. This
2571 option affects only the behaviour of the dollar metacharacter. It does
2572 not affect \Z or \z.
2573
2574 PCRE_NOTEMPTY
2575
2576 An empty string is not considered to be a valid match if this option is
2577 set. If there are alternatives in the pattern, they are tried. If all
2578 the alternatives match the empty string, the entire match fails. For
2579 example, if the pattern
2580
2581 a?b?
2582
2583 is applied to a string not beginning with "a" or "b", it matches an
2584 empty string at the start of the subject. With PCRE_NOTEMPTY set, this
2585 match is not valid, so PCRE searches further into the string for occur-
2586 rences of "a" or "b".
2587
2588 PCRE_NOTEMPTY_ATSTART
2589
2590 This is like PCRE_NOTEMPTY, except that an empty string match that is
2591 not at the start of the subject is permitted. If the pattern is
2592 anchored, such a match can occur only if the pattern contains \K.
2593
2594 Perl has no direct equivalent of PCRE_NOTEMPTY or
2595 PCRE_NOTEMPTY_ATSTART, but it does make a special case of a pattern
2596 match of the empty string within its split() function, and when using
2597 the /g modifier. It is possible to emulate Perl's behaviour after
2598 matching a null string by first trying the match again at the same off-
2599 set with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then if that
2600 fails, by advancing the starting offset (see below) and trying an ordi-
2601 nary match again. There is some code that demonstrates how to do this
2602 in the pcredemo sample program. In the most general case, you have to
2603 check to see if the newline convention recognizes CRLF as a newline,
2604 and if so, and the current character is CR followed by LF, advance the
2605 starting offset by two characters instead of one.
2606
2607 PCRE_NO_START_OPTIMIZE
2608
2609 There are a number of optimizations that pcre_exec() uses at the start
2610 of a match, in order to speed up the process. For example, if it is
2611 known that an unanchored match must start with a specific character, it
2612 searches the subject for that character, and fails immediately if it
2613 cannot find it, without actually running the main matching function.
2614 This means that a special item such as (*COMMIT) at the start of a pat-
2615 tern is not considered until after a suitable starting point for the
2616 match has been found. When callouts or (*MARK) items are in use, these
2617 "start-up" optimizations can cause them to be skipped if the pattern is
2618 never actually used. The start-up optimizations are in effect a pre-
2619 scan of the subject that takes place before the pattern is run.
2620
2621 The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations,
2622 possibly causing performance to suffer, but ensuring that in cases
2623 where the result is "no match", the callouts do occur, and that items
2624 such as (*COMMIT) and (*MARK) are considered at every possible starting
2625 position in the subject string. If PCRE_NO_START_OPTIMIZE is set at
2626 compile time, it cannot be unset at matching time. The use of
2627 PCRE_NO_START_OPTIMIZE disables JIT execution; when it is set, matching
2628 is always done using interpretively.
2629
2630 Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching
2631 operation. Consider the pattern
2632
2633 (*COMMIT)ABC
2634
2635 When this is compiled, PCRE records the fact that a match must start
2636 with the character "A". Suppose the subject string is "DEFABC". The
2637 start-up optimization scans along the subject, finds "A" and runs the
2638 first match attempt from there. The (*COMMIT) item means that the pat-
2639 tern must match the current starting position, which in this case, it
2640 does. However, if the same match is run with PCRE_NO_START_OPTIMIZE
2641 set, the initial scan along the subject string does not happen. The
2642 first match attempt is run starting from "D" and when this fails,
2643 (*COMMIT) prevents any further matches being tried, so the overall
2644 result is "no match". If the pattern is studied, more start-up opti-
2645 mizations may be used. For example, a minimum length for the subject
2646 may be recorded. Consider the pattern
2647
2648 (*MARK:A)(X|Y)
2649
2650 The minimum length for a match is one character. If the subject is
2651 "ABC", there will be attempts to match "ABC", "BC", "C", and then
2652 finally an empty string. If the pattern is studied, the final attempt
2653 does not take place, because PCRE knows that the subject is too short,
2654 and so the (*MARK) is never encountered. In this case, studying the
2655 pattern does not affect the overall match result, which is still "no
2656 match", but it does affect the auxiliary information that is returned.
2657
2658 PCRE_NO_UTF8_CHECK
2659
2660 When PCRE_UTF8 is set at compile time, the validity of the subject as a
2661 UTF-8 string is automatically checked when pcre_exec() is subsequently
2662 called. The entire string is checked before any other processing takes
2663 place. The value of startoffset is also checked to ensure that it
2664 points to the start of a UTF-8 character. There is a discussion about
2665 the validity of UTF-8 strings in the pcreunicode page. If an invalid
2666 sequence of bytes is found, pcre_exec() returns the error
2667 PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is a
2668 truncated character at the end of the subject, PCRE_ERROR_SHORTUTF8. In
2669 both cases, information about the precise nature of the error may also
2670 be returned (see the descriptions of these errors in the section enti-
2671 tled Error return values from pcre_exec() below). If startoffset con-
2672 tains a value that does not point to the start of a UTF-8 character (or
2673 to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is returned.
2674
2675 If you already know that your subject is valid, and you want to skip
2676 these checks for performance reasons, you can set the
2677 PCRE_NO_UTF8_CHECK option when calling pcre_exec(). You might want to
2678 do this for the second and subsequent calls to pcre_exec() if you are
2679 making repeated calls to find all the matches in a single subject
2680 string. However, you should be sure that the value of startoffset
2681 points to the start of a character (or the end of the subject). When
2682 PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid string as a
2683 subject or an invalid value of startoffset is undefined. Your program
2684 may crash.
2685
2686 PCRE_PARTIAL_HARD
2687 PCRE_PARTIAL_SOFT
2688
2689 These options turn on the partial matching feature. For backwards com-
2690 patibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial
2691 match occurs if the end of the subject string is reached successfully,
2692 but there are not enough subject characters to complete the match. If
2693 this happens when PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set,
2694 matching continues by testing any remaining alternatives. Only if no
2695 complete match can be found is PCRE_ERROR_PARTIAL returned instead of
2696 PCRE_ERROR_NOMATCH. In other words, PCRE_PARTIAL_SOFT says that the
2697 caller is prepared to handle a partial match, but only if no complete
2698 match can be found.
2699
2700 If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this
2701 case, if a partial match is found, pcre_exec() immediately returns
2702 PCRE_ERROR_PARTIAL, without considering any other alternatives. In
2703 other words, when PCRE_PARTIAL_HARD is set, a partial match is consid-
2704 ered to be more important that an alternative complete match.
2705
2706 In both cases, the portion of the string that was inspected when the
2707 partial match was found is set as the first matching string. There is a
2708 more detailed discussion of partial and multi-segment matching, with
2709 examples, in the pcrepartial documentation.
2710
2711 The string to be matched by pcre_exec()
2712
2713 The subject string is passed to pcre_exec() as a pointer in subject, a
2714 length in bytes in length, and a starting byte offset in startoffset.
2715 If this is negative or greater than the length of the subject,
2716 pcre_exec() returns PCRE_ERROR_BADOFFSET. When the starting offset is
2717 zero, the search for a match starts at the beginning of the subject,
2718 and this is by far the most common case. In UTF-8 mode, the byte offset
2719 must point to the start of a UTF-8 character (or the end of the sub-
2720 ject). Unlike the pattern string, the subject may contain binary zero
2721 bytes.
2722
2723 A non-zero starting offset is useful when searching for another match
2724 in the same subject by calling pcre_exec() again after a previous suc-
2725 cess. Setting startoffset differs from just passing over a shortened
2726 string and setting PCRE_NOTBOL in the case of a pattern that begins
2727 with any kind of lookbehind. For example, consider the pattern
2728
2729 \Biss\B
2730
2731 which finds occurrences of "iss" in the middle of words. (\B matches
2732 only if the current position in the subject is not a word boundary.)
2733 When applied to the string "Mississipi" the first call to pcre_exec()
2734 finds the first occurrence. If pcre_exec() is called again with just
2735 the remainder of the subject, namely "issipi", it does not match,
2736 because \B is always false at the start of the subject, which is deemed
2737 to be a word boundary. However, if pcre_exec() is passed the entire
2738 string again, but with startoffset set to 4, it finds the second occur-
2739 rence of "iss" because it is able to look behind the starting point to
2740 discover that it is preceded by a letter.
2741
2742 Finding all the matches in a subject is tricky when the pattern can
2743 match an empty string. It is possible to emulate Perl's /g behaviour by
2744 first trying the match again at the same offset, with the
2745 PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED options, and then if that
2746 fails, advancing the starting offset and trying an ordinary match
2747 again. There is some code that demonstrates how to do this in the pcre-
2748 demo sample program. In the most general case, you have to check to see
2749 if the newline convention recognizes CRLF as a newline, and if so, and
2750 the current character is CR followed by LF, advance the starting offset
2751 by two characters instead of one.
2752
2753 If a non-zero starting offset is passed when the pattern is anchored,
2754 one attempt to match at the given offset is made. This can only succeed
2755 if the pattern does not require the match to be at the start of the
2756 subject.
2757
2758 How pcre_exec() returns captured substrings
2759
2760 In general, a pattern matches a certain portion of the subject, and in
2761 addition, further substrings from the subject may be picked out by
2762 parts of the pattern. Following the usage in Jeffrey Friedl's book,
2763 this is called "capturing" in what follows, and the phrase "capturing
2764 subpattern" is used for a fragment of a pattern that picks out a sub-
2765 string. PCRE supports several other kinds of parenthesized subpattern
2766 that do not cause substrings to be captured.
2767
2768 Captured substrings are returned to the caller via a vector of integers
2769 whose address is passed in ovector. The number of elements in the vec-
2770 tor is passed in ovecsize, which must be a non-negative number. Note:
2771 this argument is NOT the size of ovector in bytes.
2772
2773 The first two-thirds of the vector is used to pass back captured sub-
2774 strings, each substring using a pair of integers. The remaining third
2775 of the vector is used as workspace by pcre_exec() while matching cap-
2776 turing subpatterns, and is not available for passing back information.
2777 The number passed in ovecsize should always be a multiple of three. If
2778 it is not, it is rounded down.
2779
2780 When a match is successful, information about captured substrings is
2781 returned in pairs of integers, starting at the beginning of ovector,
2782 and continuing up to two-thirds of its length at the most. The first
2783 element of each pair is set to the byte offset of the first character
2784 in a substring, and the second is set to the byte offset of the first
2785 character after the end of a substring. Note: these values are always
2786 byte offsets, even in UTF-8 mode. They are not character counts.
2787
2788 The first pair of integers, ovector[0] and ovector[1], identify the
2789 portion of the subject string matched by the entire pattern. The next
2790 pair is used for the first capturing subpattern, and so on. The value
2791 returned by pcre_exec() is one more than the highest numbered pair that
2792 has been set. For example, if two substrings have been captured, the
2793 returned value is 3. If there are no capturing subpatterns, the return
2794 value from a successful match is 1, indicating that just the first pair
2795 of offsets has been set.
2796
2797 If a capturing subpattern is matched repeatedly, it is the last portion
2798 of the string that it matched that is returned.
2799
2800 If the vector is too small to hold all the captured substring offsets,
2801 it is used as far as possible (up to two-thirds of its length), and the
2802 function returns a value of zero. If neither the actual string matched
2803 nor any captured substrings are of interest, pcre_exec() may be called
2804 with ovector passed as NULL and ovecsize as zero. However, if the pat-
2805 tern contains back references and the ovector is not big enough to
2806 remember the related substrings, PCRE has to get additional memory for
2807 use during matching. Thus it is usually advisable to supply an ovector
2808 of reasonable size.
2809
2810 There are some cases where zero is returned (indicating vector over-
2811 flow) when in fact the vector is exactly the right size for the final
2812 match. For example, consider the pattern
2813
2814 (a)(?:(b)c|bd)
2815
2816 If a vector of 6 elements (allowing for only 1 captured substring) is
2817 given with subject string "abd", pcre_exec() will try to set the second
2818 captured string, thereby recording a vector overflow, before failing to
2819 match "c" and backing up to try the second alternative. The zero
2820 return, however, does correctly indicate that the maximum number of
2821 slots (namely 2) have been filled. In similar cases where there is tem-
2822 porary overflow, but the final number of used slots is actually less
2823 than the maximum, a non-zero value is returned.
2824
2825 The pcre_fullinfo() function can be used to find out how many capturing
2826 subpatterns there are in a compiled pattern. The smallest size for
2827 ovector that will allow for n captured substrings, in addition to the
2828 offsets of the substring matched by the whole pattern, is (n+1)*3.
2829
2830 It is possible for capturing subpattern number n+1 to match some part
2831 of the subject when subpattern n has not been used at all. For example,
2832 if the string "abc" is matched against the pattern (a|(z))(bc) the
2833 return from the function is 4, and subpatterns 1 and 3 are matched, but
2834 2 is not. When this happens, both values in the offset pairs corre-
2835 sponding to unused subpatterns are set to -1.
2836
2837 Offset values that correspond to unused subpatterns at the end of the
2838 expression are also set to -1. For example, if the string "abc" is
2839 matched against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not
2840 matched. The return from the function is 2, because the highest used
2841 capturing subpattern number is 1, and the offsets for for the second
2842 and third capturing subpatterns (assuming the vector is large enough,
2843 of course) are set to -1.
2844
2845 Note: Elements in the first two-thirds of ovector that do not corre-
2846 spond to capturing parentheses in the pattern are never changed. That
2847 is, if a pattern contains n capturing parentheses, no more than ovec-
2848 tor[0] to ovector[2n+1] are set by pcre_exec(). The other elements (in
2849 the first two-thirds) retain whatever values they previously had.
2850
2851 Some convenience functions are provided for extracting the captured
2852 substrings as separate strings. These are described below.
2853
2854 Error return values from pcre_exec()
2855
2856 If pcre_exec() fails, it returns a negative number. The following are
2857 defined in the header file:
2858
2859 PCRE_ERROR_NOMATCH (-1)
2860
2861 The subject string did not match the pattern.
2862
2863 PCRE_ERROR_NULL (-2)
2864
2865 Either code or subject was passed as NULL, or ovector was NULL and
2866 ovecsize was not zero.
2867
2868 PCRE_ERROR_BADOPTION (-3)
2869
2870 An unrecognized bit was set in the options argument.
2871
2872 PCRE_ERROR_BADMAGIC (-4)
2873
2874 PCRE stores a 4-byte "magic number" at the start of the compiled code,
2875 to catch the case when it is passed a junk pointer and to detect when a
2876 pattern that was compiled in an environment of one endianness is run in
2877 an environment with the other endianness. This is the error that PCRE
2878 gives when the magic number is not present.
2879
2880 PCRE_ERROR_UNKNOWN_OPCODE (-5)
2881
2882 While running the pattern match, an unknown item was encountered in the
2883 compiled pattern. This error could be caused by a bug in PCRE or by
2884 overwriting of the compiled pattern.
2885
2886 PCRE_ERROR_NOMEMORY (-6)
2887
2888 If a pattern contains back references, but the ovector that is passed
2889 to pcre_exec() is not big enough to remember the referenced substrings,
2890 PCRE gets a block of memory at the start of matching to use for this
2891 purpose. If the call via pcre_malloc() fails, this error is given. The
2892 memory is automatically freed at the end of matching.
2893
2894 This error is also given if pcre_stack_malloc() fails in pcre_exec().
2895 This can happen only when PCRE has been compiled with --disable-stack-
2896 for-recursion.
2897
2898 PCRE_ERROR_NOSUBSTRING (-7)
2899
2900 This error is used by the pcre_copy_substring(), pcre_get_substring(),
2901 and pcre_get_substring_list() functions (see below). It is never
2902 returned by pcre_exec().
2903
2904 PCRE_ERROR_MATCHLIMIT (-8)
2905
2906 The backtracking limit, as specified by the match_limit field in a
2907 pcre_extra structure (or defaulted) was reached. See the description
2908 above.
2909
2910 PCRE_ERROR_CALLOUT (-9)
2911
2912 This error is never generated by pcre_exec() itself. It is provided for
2913 use by callout functions that want to yield a distinctive error code.
2914 See the pcrecallout documentation for details.
2915
2916 PCRE_ERROR_BADUTF8 (-10)
2917
2918 A string that contains an invalid UTF-8 byte sequence was passed as a
2919 subject, and the PCRE_NO_UTF8_CHECK option was not set. If the size of
2920 the output vector (ovecsize) is at least 2, the byte offset to the
2921 start of the the invalid UTF-8 character is placed in the first ele-
2922 ment, and a reason code is placed in the second element. The reason
2923 codes are listed in the following section. For backward compatibility,
2924 if PCRE_PARTIAL_HARD is set and the problem is a truncated UTF-8 char-
2925 acter at the end of the subject (reason codes 1 to 5),
2926 PCRE_ERROR_SHORTUTF8 is returned instead of PCRE_ERROR_BADUTF8.
2927
2928 PCRE_ERROR_BADUTF8_OFFSET (-11)
2929
2930 The UTF-8 byte sequence that was passed as a subject was checked and
2931 found to be valid (the PCRE_NO_UTF8_CHECK option was not set), but the
2932 value of startoffset did not point to the beginning of a UTF-8 charac-
2933 ter or the end of the subject.
2934
2935 PCRE_ERROR_PARTIAL (-12)
2936
2937 The subject string did not match, but it did match partially. See the
2938 pcrepartial documentation for details of partial matching.
2939
2940 PCRE_ERROR_BADPARTIAL (-13)
2941
2942 This code is no longer in use. It was formerly returned when the
2943 PCRE_PARTIAL option was used with a compiled pattern containing items
2944 that were not supported for partial matching. From release 8.00
2945 onwards, there are no restrictions on partial matching.
2946
2947 PCRE_ERROR_INTERNAL (-14)
2948
2949 An unexpected internal error has occurred. This error could be caused
2950 by a bug in PCRE or by overwriting of the compiled pattern.
2951
2952 PCRE_ERROR_BADCOUNT (-15)
2953
2954 This error is given if the value of the ovecsize argument is negative.
2955
2956 PCRE_ERROR_RECURSIONLIMIT (-21)
2957
2958 The internal recursion limit, as specified by the match_limit_recursion
2959 field in a pcre_extra structure (or defaulted) was reached. See the
2960 description above.
2961
2962 PCRE_ERROR_BADNEWLINE (-23)
2963
2964 An invalid combination of PCRE_NEWLINE_xxx options was given.
2965
2966 PCRE_ERROR_BADOFFSET (-24)
2967
2968 The value of startoffset was negative or greater than the length of the
2969 subject, that is, the value in length.
2970
2971 PCRE_ERROR_SHORTUTF8 (-25)
2972
2973 This error is returned instead of PCRE_ERROR_BADUTF8 when the subject
2974 string ends with a truncated UTF-8 character and the PCRE_PARTIAL_HARD
2975 option is set. Information about the failure is returned as for
2976 PCRE_ERROR_BADUTF8. It is in fact sufficient to detect this case, but
2977 this special error code for PCRE_PARTIAL_HARD precedes the implementa-
2978 tion of returned information; it is retained for backwards compatibil-
2979 ity.
2980
2981 PCRE_ERROR_RECURSELOOP (-26)
2982
2983 This error is returned when pcre_exec() detects a recursion loop within
2984 the pattern. Specifically, it means that either the whole pattern or a
2985 subpattern has been called recursively for the second time at the same
2986 position in the subject string. Some simple patterns that might do this
2987 are detected and faulted at compile time, but more complicated cases,
2988 in particular mutual recursions between two different subpatterns, can-
2989 not be detected until run time.
2990
2991 PCRE_ERROR_JIT_STACKLIMIT (-27)
2992
2993 This error is returned when a pattern that was successfully studied
2994 using a JIT compile option is being matched, but the memory available
2995 for the just-in-time processing stack is not large enough. See the
2996 pcrejit documentation for more details.
2997
2998 PCRE_ERROR_BADMODE (-28)
2999
3000 This error is given if a pattern that was compiled by the 8-bit library
3001 is passed to a 16-bit library function, or vice versa.
3002
3003 PCRE_ERROR_BADENDIANNESS (-29)
3004
3005 This error is given if a pattern that was compiled and saved is
3006 reloaded on a host with different endianness. The utility function
3007 pcre_pattern_to_host_byte_order() can be used to convert such a pattern
3008 so that it runs on the new host.
3009
3010 Error numbers -16 to -20, -22, and -30 are not used by pcre_exec().
3011
3012 Reason codes for invalid UTF-8 strings
3013
3014 This section applies only to the 8-bit library. The corresponding
3015 information for the 16-bit library is given in the pcre16 page.
3016
3017 When pcre_exec() returns either PCRE_ERROR_BADUTF8 or PCRE_ERROR_SHORT-
3018 UTF8, and the size of the output vector (ovecsize) is at least 2, the
3019 offset of the start of the invalid UTF-8 character is placed in the
3020 first output vector element (ovector[0]) and a reason code is placed in
3021 the second element (ovector[1]). The reason codes are given names in
3022 the pcre.h header file:
3023
3024 PCRE_UTF8_ERR1
3025 PCRE_UTF8_ERR2
3026 PCRE_UTF8_ERR3
3027 PCRE_UTF8_ERR4
3028 PCRE_UTF8_ERR5
3029
3030 The string ends with a truncated UTF-8 character; the code specifies
3031 how many bytes are missing (1 to 5). Although RFC 3629 restricts UTF-8
3032 characters to be no longer than 4 bytes, the encoding scheme (origi-
3033 nally defined by RFC 2279) allows for up to 6 bytes, and this is
3034 checked first; hence the possibility of 4 or 5 missing bytes.
3035
3036 PCRE_UTF8_ERR6
3037 PCRE_UTF8_ERR7
3038 PCRE_UTF8_ERR8
3039 PCRE_UTF8_ERR9
3040 PCRE_UTF8_ERR10
3041
3042 The two most significant bits of the 2nd, 3rd, 4th, 5th, or 6th byte of
3043 the character do not have the binary value 0b10 (that is, either the
3044 most significant bit is 0, or the next bit is 1).
3045
3046 PCRE_UTF8_ERR11
3047 PCRE_UTF8_ERR12
3048
3049 A character that is valid by the RFC 2279 rules is either 5 or 6 bytes
3050 long; these code points are excluded by RFC 3629.
3051
3052 PCRE_UTF8_ERR13
3053
3054 A 4-byte character has a value greater than 0x10fff; these code points
3055 are excluded by RFC 3629.
3056
3057 PCRE_UTF8_ERR14
3058
3059 A 3-byte character has a value in the range 0xd800 to 0xdfff; this
3060 range of code points are reserved by RFC 3629 for use with UTF-16, and
3061 so are excluded from UTF-8.
3062
3063 PCRE_UTF8_ERR15
3064 PCRE_UTF8_ERR16
3065 PCRE_UTF8_ERR17
3066 PCRE_UTF8_ERR18
3067 PCRE_UTF8_ERR19
3068
3069 A 2-, 3-, 4-, 5-, or 6-byte character is "overlong", that is, it codes
3070 for a value that can be represented by fewer bytes, which is invalid.
3071 For example, the two bytes 0xc0, 0xae give the value 0x2e, whose cor-
3072 rect coding uses just one byte.
3073
3074 PCRE_UTF8_ERR20
3075
3076 The two most significant bits of the first byte of a character have the
3077 binary value 0b10 (that is, the most significant bit is 1 and the sec-
3078 ond is 0). Such a byte can only validly occur as the second or subse-
3079 quent byte of a multi-byte character.
3080
3081 PCRE_UTF8_ERR21
3082
3083 The first byte of a character has the value 0xfe or 0xff. These values
3084 can never occur in a valid UTF-8 string.
3085
3086
3087 EXTRACTING CAPTURED SUBSTRINGS BY NUMBER
3088
3089 int pcre_copy_substring(const char *subject, int *ovector,
3090 int stringcount, int stringnumber, char *buffer,
3091 int buffersize);
3092
3093 int pcre_get_substring(const char *subject, int *ovector,
3094 int stringcount, int stringnumber,
3095 const char **stringptr);
3096
3097 int pcre_get_substring_list(const char *subject,
3098 int *ovector, int stringcount, const char ***listptr);
3099
3100 Captured substrings can be accessed directly by using the offsets
3101 returned by pcre_exec() in ovector. For convenience, the functions
3102 pcre_copy_substring(), pcre_get_substring(), and pcre_get_sub-
3103 string_list() are provided for extracting captured substrings as new,
3104 separate, zero-terminated strings. These functions identify substrings
3105 by number. The next section describes functions for extracting named
3106 substrings.
3107
3108 A substring that contains a binary zero is correctly extracted and has
3109 a further zero added on the end, but the result is not, of course, a C
3110 string. However, you can process such a string by referring to the
3111 length that is returned by pcre_copy_substring() and pcre_get_sub-
3112 string(). Unfortunately, the interface to pcre_get_substring_list() is
3113 not adequate for handling strings containing binary zeros, because the
3114 end of the final string is not independently indicated.
3115
3116 The first three arguments are the same for all three of these func-
3117 tions: subject is the subject string that has just been successfully
3118 matched, ovector is a pointer to the vector of integer offsets that was
3119 passed to pcre_exec(), and stringcount is the number of substrings that
3120 were captured by the match, including the substring that matched the
3121 entire regular expression. This is the value returned by pcre_exec() if
3122 it is greater than zero. If pcre_exec() returned zero, indicating that
3123 it ran out of space in ovector, the value passed as stringcount should
3124 be the number of elements in the vector divided by three.
3125
3126 The functions pcre_copy_substring() and pcre_get_substring() extract a
3127 single substring, whose number is given as stringnumber. A value of
3128 zero extracts the substring that matched the entire pattern, whereas
3129 higher values extract the captured substrings. For pcre_copy_sub-
3130 string(), the string is placed in buffer, whose length is given by
3131 buffersize, while for pcre_get_substring() a new block of memory is
3132 obtained via pcre_malloc, and its address is returned via stringptr.
3133 The yield of the function is the length of the string, not including
3134 the terminating zero, or one of these error codes:
3135
3136 PCRE_ERROR_NOMEMORY (-6)
3137
3138 The buffer was too small for pcre_copy_substring(), or the attempt to
3139 get memory failed for pcre_get_substring().
3140
3141 PCRE_ERROR_NOSUBSTRING (-7)
3142
3143 There is no substring whose number is stringnumber.
3144
3145 The pcre_get_substring_list() function extracts all available sub-
3146 strings and builds a list of pointers to them. All this is done in a
3147 single block of memory that is obtained via pcre_malloc. The address of
3148 the memory block is returned via listptr, which is also the start of
3149 the list of string pointers. The end of the list is marked by a NULL
3150 pointer. The yield of the function is zero if all went well, or the
3151 error code
3152
3153 PCRE_ERROR_NOMEMORY (-6)
3154
3155 if the attempt to get the memory block failed.
3156
3157 When any of these functions encounter a substring that is unset, which
3158 can happen when capturing subpattern number n+1 matches some part of
3159 the subject, but subpattern n has not been used at all, they return an
3160 empty string. This can be distinguished from a genuine zero-length sub-
3161 string by inspecting the appropriate offset in ovector, which is nega-
3162 tive for unset substrings.
3163
3164 The two convenience functions pcre_free_substring() and pcre_free_sub-
3165 string_list() can be used to free the memory returned by a previous
3166 call of pcre_get_substring() or pcre_get_substring_list(), respec-
3167 tively. They do nothing more than call the function pointed to by
3168 pcre_free, which of course could be called directly from a C program.
3169 However, PCRE is used in some situations where it is linked via a spe-
3170 cial interface to another programming language that cannot use
3171 pcre_free directly; it is for these cases that the functions are pro-
3172 vided.
3173
3174
3175 EXTRACTING CAPTURED SUBSTRINGS BY NAME
3176
3177 int pcre_get_stringnumber(const pcre *code,
3178 const char *name);
3179
3180 int pcre_copy_named_substring(const pcre *code,
3181 const char *subject, int *ovector,
3182 int stringcount, const char *stringname,
3183 char *buffer, int buffersize);
3184
3185 int pcre_get_named_substring(const pcre *code,
3186 const char *subject, int *ovector,
3187 int stringcount, const char *stringname,
3188 const char **stringptr);
3189
3190 To extract a substring by name, you first have to find associated num-
3191 ber. For example, for this pattern
3192
3193 (a+)b(?<xxx>\d+)...
3194
3195 the number of the subpattern called "xxx" is 2. If the name is known to
3196 be unique (PCRE_DUPNAMES was not set), you can find the number from the
3197 name by calling pcre_get_stringnumber(). The first argument is the com-
3198 piled pattern, and the second is the name. The yield of the function is
3199 the subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no
3200 subpattern of that name.
3201
3202 Given the number, you can extract the substring directly, or use one of
3203 the functions described in the previous section. For convenience, there
3204 are also two functions that do the whole job.
3205
3206 Most of the arguments of pcre_copy_named_substring() and
3207 pcre_get_named_substring() are the same as those for the similarly
3208 named functions that extract by number. As these are described in the
3209 previous section, they are not re-described here. There are just two
3210 differences:
3211
3212 First, instead of a substring number, a substring name is given. Sec-
3213 ond, there is an extra argument, given at the start, which is a pointer
3214 to the compiled pattern. This is needed in order to gain access to the
3215 name-to-number translation table.
3216
3217 These functions call pcre_get_stringnumber(), and if it succeeds, they
3218 then call pcre_copy_substring() or pcre_get_substring(), as appropri-
3219 ate. NOTE: If PCRE_DUPNAMES is set and there are duplicate names, the
3220 behaviour may not be what you want (see the next section).
3221
3222 Warning: If the pattern uses the (?| feature to set up multiple subpat-
3223 terns with the same number, as described in the section on duplicate
3224 subpattern numbers in the pcrepattern page, you cannot use names to
3225 distinguish the different subpatterns, because names are not included
3226 in the compiled code. The matching process uses only numbers. For this
3227 reason, the use of different names for subpatterns of the same number
3228 causes an error at compile time.
3229
3230
3231 DUPLICATE SUBPATTERN NAMES
3232
3233 int pcre_get_stringtable_entries(const pcre *code,
3234 const char *name, char **first, char **last);
3235
3236 When a pattern is compiled with the PCRE_DUPNAMES option, names for
3237 subpatterns are not required to be unique. (Duplicate names are always
3238 allowed for subpatterns with the same number, created by using the (?|
3239 feature. Indeed, if such subpatterns are named, they are required to
3240 use the same names.)
3241
3242 Normally, patterns with duplicate names are such that in any one match,
3243 only one of the named subpatterns participates. An example is shown in
3244 the pcrepattern documentation.
3245
3246 When duplicates are present, pcre_copy_named_substring() and
3247 pcre_get_named_substring() return the first substring corresponding to
3248 the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING
3249 (-7) is returned; no data is returned. The pcre_get_stringnumber()
3250 function returns one of the numbers that are associated with the name,
3251 but it is not defined which it is.
3252
3253 If you want to get full details of all captured substrings for a given
3254 name, you must use the pcre_get_stringtable_entries() function. The
3255 first argument is the compiled pattern, and the second is the name. The
3256 third and fourth are pointers to variables which are updated by the
3257 function. After it has run, they point to the first and last entries in
3258 the name-to-number table for the given name. The function itself
3259 returns the length of each entry, or PCRE_ERROR_NOSUBSTRING (-7) if
3260 there are none. The format of the table is described above in the sec-
3261 tion entitled Information about a pattern above. Given all the rele-
3262 vant entries for the name, you can extract each of their numbers, and
3263 hence the captured data, if any.
3264
3265
3266 FINDING ALL POSSIBLE MATCHES
3267
3268 The traditional matching function uses a similar algorithm to Perl,
3269 which stops when it finds the first match, starting at a given point in
3270 the subject. If you want to find all possible matches, or the longest
3271 possible match, consider using the alternative matching function (see
3272 below) instead. If you cannot use the alternative function, but still
3273 need to find all possible matches, you can kludge it up by making use
3274 of the callout facility, which is described in the pcrecallout documen-
3275 tation.
3276
3277 What you have to do is to insert a callout right at the end of the pat-
3278 tern. When your callout function is called, extract and save the cur-
3279 rent matched substring. Then return 1, which forces pcre_exec() to
3280 backtrack and try other alternatives. Ultimately, when it runs out of
3281 matches, pcre_exec() will yield PCRE_ERROR_NOMATCH.
3282
3283
3284 OBTAINING AN ESTIMATE OF STACK USAGE
3285
3286 Matching certain patterns using pcre_exec() can use a lot of process
3287 stack, which in certain environments can be rather limited in size.
3288 Some users find it helpful to have an estimate of the amount of stack
3289 that is used by pcre_exec(), to help them set recursion limits, as
3290 described in the pcrestack documentation. The estimate that is output
3291 by pcretest when called with the -m and -C options is obtained by call-
3292 ing pcre_exec with the values NULL, NULL, NULL, -999, and -999 for its
3293 first five arguments.
3294
3295 Normally, if its first argument is NULL, pcre_exec() immediately
3296 returns the negative error code PCRE_ERROR_NULL, but with this special
3297 combination of arguments, it returns instead a negative number whose
3298 absolute value is the approximate stack frame size in bytes. (A nega-
3299 tive number is used so that it is clear that no match has happened.)
3300 The value is approximate because in some cases, recursive calls to
3301 pcre_exec() occur when there are one or two additional variables on the
3302 stack.
3303
3304 If PCRE has been compiled to use the heap instead of the stack for
3305 recursion, the value returned is the size of each block that is
3306 obtained from the heap.
3307
3308
3309 MATCHING A PATTERN: THE ALTERNATIVE FUNCTION
3310
3311 int pcre_dfa_exec(const pcre *code, const pcre_extra *extra,
3312 const char *subject, int length, int startoffset,
3313 int options, int *ovector, int ovecsize,
3314 int *workspace, int wscount);
3315
3316 The function pcre_dfa_exec() is called to match a subject string
3317 against a compiled pattern, using a matching algorithm that scans the
3318 subject string just once, and does not backtrack. This has different
3319 characteristics to the normal algorithm, and is not compatible with
3320 Perl. Some of the features of PCRE patterns are not supported. Never-
3321 theless, there are times when this kind of matching can be useful. For
3322 a discussion of the two matching algorithms, and a list of features
3323 that pcre_dfa_exec() does not support, see the pcrematching documenta-
3324 tion.
3325
3326 The arguments for the pcre_dfa_exec() function are the same as for
3327 pcre_exec(), plus two extras. The ovector argument is used in a differ-
3328 ent way, and this is described below. The other common arguments are
3329 used in the same way as for pcre_exec(), so their description is not
3330 repeated here.
3331
3332 The two additional arguments provide workspace for the function. The
3333 workspace vector should contain at least 20 elements. It is used for
3334 keeping track of multiple paths through the pattern tree. More
3335 workspace will be needed for patterns and subjects where there are a
3336 lot of potential matches.
3337
3338 Here is an example of a simple call to pcre_dfa_exec():
3339
3340 int rc;
3341 int ovector[10];
3342 int wspace[20];
3343 rc = pcre_dfa_exec(
3344 re, /* result of pcre_compile() */
3345 NULL, /* we didn't study the pattern */
3346 "some string", /* the subject string */
3347 11, /* the length of the subject string */
3348 0, /* start at offset 0 in the subject */
3349 0, /* default options */
3350 ovector, /* vector of integers for substring information */
3351 10, /* number of elements (NOT size in bytes) */
3352 wspace, /* working space vector */
3353 20); /* number of elements (NOT size in bytes) */
3354
3355 Option bits for pcre_dfa_exec()
3356
3357 The unused bits of the options argument for pcre_dfa_exec() must be
3358 zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEW-
3359 LINE_xxx, PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY,
3360 PCRE_NOTEMPTY_ATSTART, PCRE_NO_UTF8_CHECK, PCRE_BSR_ANYCRLF,
3361 PCRE_BSR_UNICODE, PCRE_NO_START_OPTIMIZE, PCRE_PARTIAL_HARD, PCRE_PAR-
3362 TIAL_SOFT, PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last
3363 four of these are exactly the same as for pcre_exec(), so their
3364 description is not repeated here.
3365
3366 PCRE_PARTIAL_HARD
3367 PCRE_PARTIAL_SOFT
3368
3369 These have the same general effect as they do for pcre_exec(), but the
3370 details are slightly different. When PCRE_PARTIAL_HARD is set for
3371 pcre_dfa_exec(), it returns PCRE_ERROR_PARTIAL if the end of the sub-
3372 ject is reached and there is still at least one matching possibility
3373 that requires additional characters. This happens even if some complete
3374 matches have also been found. When PCRE_PARTIAL_SOFT is set, the return
3375 code PCRE_ERROR_NOMATCH is converted into PCRE_ERROR_PARTIAL if the end
3376 of the subject is reached, there have been no complete matches, but
3377 there is still at least one matching possibility. The portion of the
3378 string that was inspected when the longest partial match was found is
3379 set as the first matching string in both cases. There is a more
3380 detailed discussion of partial and multi-segment matching, with exam-
3381 ples, in the pcrepartial documentation.
3382
3383 PCRE_DFA_SHORTEST
3384
3385 Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to
3386 stop as soon as it has found one match. Because of the way the alterna-
3387 tive algorithm works, this is necessarily the shortest possible match
3388 at the first possible matching point in the subject string.
3389
3390 PCRE_DFA_RESTART
3391
3392 When pcre_dfa_exec() returns a partial match, it is possible to call it
3393 again, with additional subject characters, and have it continue with
3394 the same match. The PCRE_DFA_RESTART option requests this action; when
3395 it is set, the workspace and wscount options must reference the same
3396 vector as before because data about the match so far is left in them
3397 after a partial match. There is more discussion of this facility in the
3398 pcrepartial documentation.
3399
3400 Successful returns from pcre_dfa_exec()
3401
3402 When pcre_dfa_exec() succeeds, it may have matched more than one sub-
3403 string in the subject. Note, however, that all the matches from one run
3404 of the function start at the same point in the subject. The shorter
3405 matches are all initial substrings of the longer matches. For example,
3406 if the pattern
3407
3408 <.*>
3409
3410 is matched against the string
3411
3412 This is <something> <something else> <something further> no more
3413
3414 the three matched strings are
3415
3416 <something>
3417 <something> <something else>
3418 <something> <something else> <something further>
3419
3420 On success, the yield of the function is a number greater than zero,
3421 which is the number of matched substrings. The substrings themselves
3422 are returned in ovector. Each string uses two elements; the first is
3423 the offset to the start, and the second is the offset to the end. In
3424 fact, all the strings have the same start offset. (Space could have
3425 been saved by giving this only once, but it was decided to retain some
3426 compatibility with the way pcre_exec() returns data, even though the
3427 meaning of the strings is different.)
3428
3429 The strings are returned in reverse order of length; that is, the long-
3430 est matching string is given first. If there were too many matches to
3431 fit into ovector, the yield of the function is zero, and the vector is
3432 filled with the longest matches. Unlike pcre_exec(), pcre_dfa_exec()
3433 can use the entire ovector for returning matched strings.
3434
3435 Error returns from pcre_dfa_exec()
3436
3437 The pcre_dfa_exec() function returns a negative number when it fails.
3438 Many of the errors are the same as for pcre_exec(), and these are
3439 described above. There are in addition the following errors that are
3440 specific to pcre_dfa_exec():
3441
3442 PCRE_ERROR_DFA_UITEM (-16)
3443
3444 This return is given if pcre_dfa_exec() encounters an item in the pat-
3445 tern that it does not support, for instance, the use of \C or a back
3446 reference.
3447
3448 PCRE_ERROR_DFA_UCOND (-17)
3449
3450 This return is given if pcre_dfa_exec() encounters a condition item
3451 that uses a back reference for the condition, or a test for recursion
3452 in a specific group. These are not supported.
3453
3454 PCRE_ERROR_DFA_UMLIMIT (-18)
3455
3456 This return is given if pcre_dfa_exec() is called with an extra block
3457 that contains a setting of the match_limit or match_limit_recursion
3458 fields. This is not supported (these fields are meaningless for DFA
3459 matching).
3460
3461 PCRE_ERROR_DFA_WSSIZE (-19)
3462
3463 This return is given if pcre_dfa_exec() runs out of space in the
3464 workspace vector.
3465
3466 PCRE_ERROR_DFA_RECURSE (-20)
3467
3468 When a recursive subpattern is processed, the matching function calls
3469 itself recursively, using private vectors for ovector and workspace.
3470 This error is given if the output vector is not large enough. This
3471 should be extremely rare, as a vector of size 1000 is used.
3472
3473 PCRE_ERROR_DFA_BADRESTART (-30)
3474
3475 When pcre_dfa_exec() is called with the PCRE_DFA_RESTART option, some
3476 plausibility checks are made on the contents of the workspace, which
3477 should contain data about the previous partial match. If any of these
3478 checks fail, this error is given.
3479
3480
3481 SEE ALSO
3482
3483 pcre16(3), pcrebuild(3), pcrecallout(3), pcrecpp(3)(3), pcrematch-
3484 ing(3), pcrepartial(3), pcreposix(3), pcreprecompile(3), pcresample(3),
3485 pcrestack(3).
3486
3487
3488 AUTHOR
3489
3490 Philip Hazel
3491 University Computing Service
3492 Cambridge CB2 3QH, England.
3493
3494
3495 REVISION
3496
3497 Last updated: 17 June 2012
3498 Copyright (c) 1997-2012 University of Cambridge.
3499 ------------------------------------------------------------------------------
3500
3501
3502 PCRECALLOUT(3) PCRECALLOUT(3)
3503
3504
3505 NAME
3506 PCRE - Perl-compatible regular expressions
3507
3508
3509 PCRE CALLOUTS
3510
3511 int (*pcre_callout)(pcre_callout_block *);
3512
3513 int (*pcre16_callout)(pcre16_callout_block *);
3514
3515 PCRE provides a feature called "callout", which is a means of temporar-
3516 ily passing control to the caller of PCRE in the middle of pattern
3517 matching. The caller of PCRE provides an external function by putting
3518 its entry point in the global variable pcre_callout (pcre16_callout for
3519 the 16-bit library). By default, this variable contains NULL, which
3520 disables all calling out.
3521
3522 Within a regular expression, (?C) indicates the points at which the
3523 external function is to be called. Different callout points can be
3524 identified by putting a number less than 256 after the letter C. The
3525 default value is zero. For example, this pattern has two callout
3526 points:
3527
3528 (?C1)abc(?C2)def
3529
3530 If the PCRE_AUTO_CALLOUT option bit is set when a pattern is compiled,
3531 PCRE automatically inserts callouts, all with number 255, before each
3532 item in the pattern. For example, if PCRE_AUTO_CALLOUT is used with the
3533 pattern
3534
3535 A(\d{2}|--)
3536
3537 it is processed as if it were
3538
3539 (?C255)A(?C255)((?C255)\d{2}(?C255)|(?C255)-(?C255)-(?C255))(?C255)
3540
3541 Notice that there is a callout before and after each parenthesis and
3542 alternation bar. Automatic callouts can be used for tracking the
3543 progress of pattern matching. The pcretest command has an option that
3544 sets automatic callouts; when it is used, the output indicates how the
3545 pattern is matched. This is useful information when you are trying to
3546 optimize the performance of a particular pattern.
3547
3548 The use of callouts in a pattern makes it ineligible for optimization
3549 by the just-in-time compiler. Studying such a pattern with the
3550 PCRE_STUDY_JIT_COMPILE option always fails.
3551
3552
3553 MISSING CALLOUTS
3554
3555 You should be aware that, because of optimizations in the way PCRE
3556 matches patterns by default, callouts sometimes do not happen. For
3557 example, if the pattern is
3558
3559 ab(?C4)cd
3560
3561 PCRE knows that any matching string must contain the letter "d". If the
3562 subject string is "abyz", the lack of "d" means that matching doesn't
3563 ever start, and the callout is never reached. However, with "abyd",
3564 though the result is still no match, the callout is obeyed.
3565
3566 If the pattern is studied, PCRE knows the minimum length of a matching
3567 string, and will immediately give a "no match" return without actually
3568 running a match if the subject is not long enough, or, for unanchored
3569 patterns, if it has been scanned far enough.
3570
3571 You can disable these optimizations by passing the PCRE_NO_START_OPTI-
3572 MIZE option to the matching function, or by starting the pattern with
3573 (*NO_START_OPT). This slows down the matching process, but does ensure
3574 that callouts such as the example above are obeyed.
3575
3576
3577 THE CALLOUT INTERFACE
3578
3579 During matching, when PCRE reaches a callout point, the external func-
3580 tion defined by pcre_callout or pcre16_callout is called (if it is
3581 set). This applies to both normal and DFA matching. The only argument
3582 to the callout function is a pointer to a pcre_callout or pcre16_call-
3583 out block. These structures contains the following fields:
3584
3585 int version;
3586 int callout_number;
3587 int *offset_vector;
3588 const char *subject; (8-bit version)
3589 PCRE_SPTR16 subject; (16-bit version)
3590 int subject_length;
3591 int start_match;
3592 int current_position;
3593 int capture_top;
3594 int capture_last;
3595 void *callout_data;
3596 int pattern_position;
3597 int next_item_length;
3598 const unsigned char *mark; (8-bit version)
3599 const PCRE_UCHAR16 *mark; (16-bit version)
3600
3601 The version field is an integer containing the version number of the
3602 block format. The initial version was 0; the current version is 2. The
3603 version number will change again in future if additional fields are
3604 added, but the intention is never to remove any of the existing fields.
3605
3606 The callout_number field contains the number of the callout, as com-
3607 piled into the pattern (that is, the number after ?C for manual call-
3608 outs, and 255 for automatically generated callouts).
3609
3610 The offset_vector field is a pointer to the vector of offsets that was
3611 passed by the caller to the matching function. When pcre_exec() or
3612 pcre16_exec() is used, the contents can be inspected, in order to
3613 extract substrings that have been matched so far, in the same way as
3614 for extracting substrings after a match has completed. For the DFA
3615 matching functions, this field is not useful.
3616
3617 The subject and subject_length fields contain copies of the values that
3618 were passed to the matching function.
3619
3620 The start_match field normally contains the offset within the subject
3621 at which the current match attempt started. However, if the escape
3622 sequence \K has been encountered, this value is changed to reflect the
3623 modified starting point. If the pattern is not anchored, the callout
3624 function may be called several times from the same point in the pattern
3625 for different starting points in the subject.
3626
3627 The current_position field contains the offset within the subject of
3628 the current match pointer.
3629
3630 When the pcre_exec() or pcre16_exec() is used, the capture_top field
3631 contains one more than the number of the highest numbered captured sub-
3632 string so far. If no substrings have been captured, the value of cap-
3633 ture_top is one. This is always the case when the DFA functions are
3634 used, because they do not support captured substrings.
3635
3636 The capture_last field contains the number of the most recently cap-
3637 tured substring. If no substrings have been captured, its value is -1.
3638 This is always the case for the DFA matching functions.
3639
3640 The callout_data field contains a value that is passed to a matching
3641 function specifically so that it can be passed back in callouts. It is
3642 passed in the callout_data field of a pcre_extra or pcre16_extra data
3643 structure. If no such data was passed, the value of callout_data in a
3644 callout block is NULL. There is a description of the pcre_extra struc-
3645 ture in the pcreapi documentation.
3646
3647 The pattern_position field is present from version 1 of the callout
3648 structure. It contains the offset to the next item to be matched in the
3649 pattern string.
3650
3651 The next_item_length field is present from version 1 of the callout
3652 structure. It contains the length of the next item to be matched in the
3653 pattern string. When the callout immediately precedes an alternation
3654 bar, a closing parenthesis, or the end of the pattern, the length is
3655 zero. When the callout precedes an opening parenthesis, the length is
3656 that of the entire subpattern.
3657
3658 The pattern_position and next_item_length fields are intended to help
3659 in distinguishing between different automatic callouts, which all have
3660 the same callout number. However, they are set for all callouts.
3661
3662 The mark field is present from version 2 of the callout structure. In
3663 callouts from pcre_exec() or pcre16_exec() it contains a pointer to the
3664 zero-terminated name of the most recently passed (*MARK), (*PRUNE), or
3665 (*THEN) item in the match, or NULL if no such items have been passed.
3666 Instances of (*PRUNE) or (*THEN) without a name do not obliterate a
3667 previous (*MARK). In callouts from the DFA matching functions this
3668 field always contains NULL.
3669
3670
3671 RETURN VALUES
3672
3673 The external callout function returns an integer to PCRE. If the value
3674 is zero, matching proceeds as normal. If the value is greater than
3675 zero, matching fails at the current point, but the testing of other
3676 matching possibilities goes ahead, just as if a lookahead assertion had
3677 failed. If the value is less than zero, the match is abandoned, the
3678 matching function returns the negative value.
3679
3680 Negative values should normally be chosen from the set of
3681 PCRE_ERROR_xxx values. In particular, PCRE_ERROR_NOMATCH forces a stan-
3682 dard "no match" failure. The error number PCRE_ERROR_CALLOUT is
3683 reserved for use by callout functions; it will never be used by PCRE
3684 itself.
3685
3686
3687 AUTHOR
3688
3689 Philip Hazel
3690 University Computing Service
3691 Cambridge CB2 3QH, England.
3692
3693
3694 REVISION
3695
3696 Last updated: 08 Janurary 2012
3697 Copyright (c) 1997-2012 University of Cambridge.
3698 ------------------------------------------------------------------------------
3699
3700
3701 PCRECOMPAT(3) PCRECOMPAT(3)
3702
3703
3704 NAME
3705 PCRE - Perl-compatible regular expressions
3706
3707
3708 DIFFERENCES BETWEEN PCRE AND PERL
3709
3710 This document describes the differences in the ways that PCRE and Perl
3711 handle regular expressions. The differences described here are with
3712 respect to Perl versions 5.10 and above.
3713
3714 1. PCRE has only a subset of Perl's Unicode support. Details of what it
3715 does have are given in the pcreunicode page.
3716
3717 2. PCRE allows repeat quantifiers only on parenthesized assertions, but
3718 they do not mean what you might think. For example, (?!a){3} does not
3719 assert that the next three characters are not "a". It just asserts that
3720 the next character is not "a" three times (in principle: PCRE optimizes
3721 this to run the assertion just once). Perl allows repeat quantifiers on
3722 other assertions such as \b, but these do not seem to have any use.
3723
3724 3. Capturing subpatterns that occur inside negative lookahead asser-
3725 tions are counted, but their entries in the offsets vector are never
3726 set. Perl sets its numerical variables from any such patterns that are
3727 matched before the assertion fails to match something (thereby succeed-
3728 ing), but only if the negative lookahead assertion contains just one
3729 branch.
3730
3731 4. Though binary zero characters are supported in the subject string,
3732 they are not allowed in a pattern string because it is passed as a nor-
3733 mal C string, terminated by zero. The escape sequence \0 can be used in
3734 the pattern to represent a binary zero.
3735
3736 5. The following Perl escape sequences are not supported: \l, \u, \L,
3737 \U, and \N when followed by a character name or Unicode value. (\N on
3738 its own, matching a non-newline character, is supported.) In fact these
3739 are implemented by Perl's general string-handling and are not part of
3740 its pattern matching engine. If any of these are encountered by PCRE,
3741 an error is generated by default. However, if the PCRE_JAVASCRIPT_COM-
3742 PAT option is set, \U and \u are interpreted as JavaScript interprets
3743 them.
3744
3745 6. The Perl escape sequences \p, \P, and \X are supported only if PCRE
3746 is built with Unicode character property support. The properties that
3747 can be tested with \p and \P are limited to the general category prop-
3748 erties such as Lu and Nd, script names such as Greek or Han, and the
3749 derived properties Any and L&. PCRE does support the Cs (surrogate)
3750 property, which Perl does not; the Perl documentation says "Because
3751 Perl hides the need for the user to understand the internal representa-
3752 tion of Unicode characters, there is no need to implement the somewhat
3753 messy concept of surrogates."
3754
3755 7. PCRE implements a simpler version of \X than Perl, which changed to
3756 make \X match what Unicode calls an "extended grapheme cluster". This
3757 is more complicated than an extended Unicode sequence, which is what
3758 PCRE matches.
3759
3760 8. PCRE does support the \Q...\E escape for quoting substrings. Charac-
3761 ters in between are treated as literals. This is slightly different
3762 from Perl in that $ and @ are also handled as literals inside the
3763 quotes. In Perl, they cause variable interpolation (but of course PCRE
3764 does not have variables). Note the following examples:
3765
3766 Pattern PCRE matches Perl matches
3767
3768 \Qabc$xyz\E abc$xyz abc followed by the
3769 contents of $xyz
3770 \Qabc\$xyz\E abc\$xyz abc\$xyz
3771 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
3772
3773 The \Q...\E sequence is recognized both inside and outside character
3774 classes.
3775
3776 9. Fairly obviously, PCRE does not support the (?{code}) and (??{code})
3777 constructions. However, there is support for recursive patterns. This
3778 is not available in Perl 5.8, but it is in Perl 5.10. Also, the PCRE
3779 "callout" feature allows an external function to be called during pat-
3780 tern matching. See the pcrecallout documentation for details.
3781
3782 10. Subpatterns that are called as subroutines (whether or not recur-
3783 sively) are always treated as atomic groups in PCRE. This is like
3784 Python, but unlike Perl. Captured values that are set outside a sub-
3785 routine call can be reference from inside in PCRE, but not in Perl.
3786 There is a discussion that explains these differences in more detail in
3787 the section on recursion differences from Perl in the pcrepattern page.
3788
3789 11. If any of the backtracking control verbs are used in an assertion
3790 or in a subpattern that is called as a subroutine (whether or not
3791 recursively), their effect is confined to that subpattern; it does not
3792 extend to the surrounding pattern. This is not always the case in Perl.
3793 In particular, if (*THEN) is present in a group that is called as a
3794 subroutine, its action is limited to that group, even if the group does
3795 not contain any | characters. There is one exception to this: the name
3796 from a *(MARK), (*PRUNE), or (*THEN) that is encountered in a success-
3797 ful positive assertion is passed back when a match succeeds (compare
3798 capturing parentheses in assertions). Note that such subpatterns are
3799 processed as anchored at the point where they are tested.
3800
3801 12. There are some differences that are concerned with the settings of
3802 captured strings when part of a pattern is repeated. For example,
3803 matching "aba" against the pattern /^(a(b)?)+$/ in Perl leaves $2
3804 unset, but in PCRE it is set to "b".
3805
3806 13. PCRE's handling of duplicate subpattern numbers and duplicate sub-
3807 pattern names is not as general as Perl's. This is a consequence of the
3808 fact the PCRE works internally just with numbers, using an external ta-
3809 ble to translate between numbers and names. In particular, a pattern
3810 such as (?|(?<a>A)|(?<b)B), where the two capturing parentheses have
3811 the same number but different names, is not supported, and causes an
3812 error at compile time. If it were allowed, it would not be possible to
3813 distinguish which parentheses matched, because both names map to cap-
3814 turing subpattern number 1. To avoid this confusing situation, an error
3815 is given at compile time.
3816
3817 14. Perl recognizes comments in some places that PCRE does not, for
3818 example, between the ( and ? at the start of a subpattern. If the /x
3819 modifier is set, Perl allows white space between ( and ? but PCRE never
3820 does, even if the PCRE_EXTENDED option is set.
3821
3822 15. PCRE provides some extensions to the Perl regular expression facil-
3823 ities. Perl 5.10 includes new features that are not in earlier ver-
3824 sions of Perl, some of which (such as named parentheses) have been in
3825 PCRE for some time. This list is with respect to Perl 5.10:
3826
3827 (a) Although lookbehind assertions in PCRE must match fixed length
3828 strings, each alternative branch of a lookbehind assertion can match a
3829 different length of string. Perl requires them all to have the same
3830 length.
3831
3832 (b) If PCRE_DOLLAR_ENDONLY is set and PCRE_MULTILINE is not set, the $
3833 meta-character matches only at the very end of the string.
3834
3835 (c) If PCRE_EXTRA is set, a backslash followed by a letter with no spe-
3836 cial meaning is faulted. Otherwise, like Perl, the backslash is quietly
3837 ignored. (Perl can be made to issue a warning.)
3838
3839 (d) If PCRE_UNGREEDY is set, the greediness of the repetition quanti-
3840 fiers is inverted, that is, by default they are not greedy, but if fol-
3841 lowed by a question mark they are.
3842
3843 (e) PCRE_ANCHORED can be used at matching time to force a pattern to be
3844 tried only at the first matching position in the subject string.
3845
3846 (f) The PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
3847 and PCRE_NO_AUTO_CAPTURE options for pcre_exec() have no Perl equiva-
3848 lents.
3849
3850 (g) The \R escape sequence can be restricted to match only CR, LF, or
3851 CRLF by the PCRE_BSR_ANYCRLF option.
3852
3853 (h) The callout facility is PCRE-specific.
3854
3855 (i) The partial matching facility is PCRE-specific.
3856
3857 (j) Patterns compiled by PCRE can be saved and re-used at a later time,
3858 even on different hosts that have the other endianness. However, this
3859 does not apply to optimized data created by the just-in-time compiler.
3860
3861 (k) The alternative matching functions (pcre_dfa_exec() and
3862 pcre16_dfa_exec()) match in a different way and are not Perl-compati-
3863 ble.
3864
3865 (l) PCRE recognizes some special sequences such as (*CR) at the start
3866 of a pattern that set overall options that cannot be changed within the
3867 pattern.
3868
3869
3870 AUTHOR
3871
3872 Philip Hazel
3873 University Computing Service
3874 Cambridge CB2 3QH, England.
3875
3876
3877 REVISION
3878
3879 Last updated: 01 June 2012
3880 Copyright (c) 1997-2012 University of Cambridge.
3881 ------------------------------------------------------------------------------
3882
3883
3884 PCREPATTERN(3) PCREPATTERN(3)
3885
3886
3887 NAME
3888 PCRE - Perl-compatible regular expressions
3889
3890
3891 PCRE REGULAR EXPRESSION DETAILS
3892
3893 The syntax and semantics of the regular expressions that are supported
3894 by PCRE are described in detail below. There is a quick-reference syn-
3895 tax summary in the pcresyntax page. PCRE tries to match Perl syntax and
3896 semantics as closely as it can. PCRE also supports some alternative
3897 regular expression syntax (which does not conflict with the Perl syn-
3898 tax) in order to provide some compatibility with regular expressions in
3899 Python, .NET, and Oniguruma.
3900
3901 Perl's regular expressions are described in its own documentation, and
3902 regular expressions in general are covered in a number of books, some
3903 of which have copious examples. Jeffrey Friedl's "Mastering Regular
3904 Expressions", published by O'Reilly, covers regular expressions in
3905 great detail. This description of PCRE's regular expressions is
3906 intended as reference material.
3907
3908 The original operation of PCRE was on strings of one-byte characters.
3909 However, there is now also support for UTF-8 strings in the original
3910 library, and a second library that supports 16-bit and UTF-16 character
3911 strings. To use these features, PCRE must be built to include appropri-
3912 ate support. When using UTF strings you must either call the compiling
3913 function with the PCRE_UTF8 or PCRE_UTF16 option, or the pattern must
3914 start with one of these special sequences:
3915
3916 (*UTF8)
3917 (*UTF16)
3918
3919 Starting a pattern with such a sequence is equivalent to setting the
3920 relevant option. This feature is not Perl-compatible. How setting a UTF
3921 mode affects pattern matching is mentioned in several places below.
3922 There is also a summary of features in the pcreunicode page.
3923
3924 Another special sequence that may appear at the start of a pattern or
3925 in combination with (*UTF8) or (*UTF16) is:
3926
3927 (*UCP)
3928
3929 This has the same effect as setting the PCRE_UCP option: it causes
3930 sequences such as \d and \w to use Unicode properties to determine
3931 character types, instead of recognizing only characters with codes less
3932 than 128 via a lookup table.
3933
3934 If a pattern starts with (*NO_START_OPT), it has the same effect as
3935 setting the PCRE_NO_START_OPTIMIZE option either at compile or matching
3936 time. There are also some more of these special sequences that are con-
3937 cerned with the handling of newlines; they are described below.
3938
3939 The remainder of this document discusses the patterns that are sup-
3940 ported by PCRE when one its main matching functions, pcre_exec()
3941 (8-bit) or pcre16_exec() (16-bit), is used. PCRE also has alternative
3942 matching functions, pcre_dfa_exec() and pcre16_dfa_exec(), which match
3943 using a different algorithm that is not Perl-compatible. Some of the
3944 features discussed below are not available when DFA matching is used.
3945 The advantages and disadvantages of the alternative functions, and how
3946 they differ from the normal functions, are discussed in the pcrematch-
3947 ing page.
3948
3949
3950 NEWLINE CONVENTIONS
3951
3952 PCRE supports five different conventions for indicating line breaks in
3953 strings: a single CR (carriage return) character, a single LF (line-
3954 feed) character, the two-character sequence CRLF, any of the three pre-
3955 ceding, or any Unicode newline sequence. The pcreapi page has further
3956 discussion about newlines, and shows how to set the newline convention
3957 in the options arguments for the compiling and matching functions.
3958
3959 It is also possible to specify a newline convention by starting a pat-
3960 tern string with one of the following five sequences:
3961
3962 (*CR) carriage return
3963 (*LF) linefeed
3964 (*CRLF) carriage return, followed by linefeed
3965 (*ANYCRLF) any of the three above
3966 (*ANY) all Unicode newline sequences
3967
3968 These override the default and the options given to the compiling func-
3969 tion. For example, on a Unix system where LF is the default newline
3970 sequence, the pattern
3971
3972 (*CR)a.b
3973
3974 changes the convention to CR. That pattern matches "a\nb" because LF is
3975 no longer a newline. Note that these special settings, which are not
3976 Perl-compatible, are recognized only at the very start of a pattern,
3977 and that they must be in upper case. If more than one of them is
3978 present, the last one is used.
3979
3980 The newline convention affects the interpretation of the dot metachar-
3981 acter when PCRE_DOTALL is not set, and also the behaviour of \N. How-
3982 ever, it does not affect what the \R escape sequence matches. By
3983 default, this is any Unicode newline sequence, for Perl compatibility.
3984 However, this can be changed; see the description of \R in the section
3985 entitled "Newline sequences" below. A change of \R setting can be com-
3986 bined with a change of newline convention.
3987
3988
3989 CHARACTERS AND METACHARACTERS
3990
3991 A regular expression is a pattern that is matched against a subject
3992 string from left to right. Most characters stand for themselves in a
3993 pattern, and match the corresponding characters in the subject. As a
3994 trivial example, the pattern
3995
3996 The quick brown fox
3997
3998 matches a portion of a subject string that is identical to itself. When
3999 caseless matching is specified (the PCRE_CASELESS option), letters are
4000 matched independently of case. In a UTF mode, PCRE always understands
4001 the concept of case for characters whose values are less than 128, so
4002 caseless matching is always possible. For characters with higher val-
4003 ues, the concept of case is supported if PCRE is compiled with Unicode
4004 property support, but not otherwise. If you want to use caseless
4005 matching for characters 128 and above, you must ensure that PCRE is
4006 compiled with Unicode property support as well as with UTF support.
4007
4008 The power of regular expressions comes from the ability to include
4009 alternatives and repetitions in the pattern. These are encoded in the
4010 pattern by the use of metacharacters, which do not stand for themselves
4011 but instead are interpreted in some special way.
4012
4013 There are two different sets of metacharacters: those that are recog-
4014 nized anywhere in the pattern except within square brackets, and those
4015 that are recognized within square brackets. Outside square brackets,
4016 the metacharacters are as follows:
4017
4018 \ general escape character with several uses
4019 ^ assert start of string (or line, in multiline mode)
4020 $ assert end of string (or line, in multiline mode)
4021 . match any character except newline (by default)
4022 [ start character class definition
4023 | start of alternative branch
4024 ( start subpattern
4025 ) end subpattern
4026 ? extends the meaning of (
4027 also 0 or 1 quantifier
4028 also quantifier minimizer
4029 * 0 or more quantifier
4030 + 1 or more quantifier
4031 also "possessive quantifier"
4032 { start min/max quantifier
4033
4034 Part of a pattern that is in square brackets is called a "character
4035 class". In a character class the only metacharacters are:
4036
4037 \ general escape character
4038 ^ negate the class, but only if the first character
4039 - indicates character range
4040 [ POSIX character class (only if followed by POSIX
4041 syntax)
4042 ] terminates the character class
4043
4044 The following sections describe the use of each of the metacharacters.
4045
4046
4047 BACKSLASH
4048
4049 The backslash character has several uses. Firstly, if it is followed by
4050 a character that is not a number or a letter, it takes away any special
4051 meaning that character may have. This use of backslash as an escape
4052 character applies both inside and outside character classes.
4053
4054 For example, if you want to match a * character, you write \* in the
4055 pattern. This escaping action applies whether or not the following
4056 character would otherwise be interpreted as a metacharacter, so it is
4057 always safe to precede a non-alphanumeric with backslash to specify
4058 that it stands for itself. In particular, if you want to match a back-
4059 slash, you write \\.
4060
4061 In a UTF mode, only ASCII numbers and letters have any special meaning
4062 after a backslash. All other characters (in particular, those whose
4063 codepoints are greater than 127) are treated as literals.
4064
4065 If a pattern is compiled with the PCRE_EXTENDED option, white space in
4066 the pattern (other than in a character class) and characters between a
4067 # outside a character class and the next newline are ignored. An escap-
4068 ing backslash can be used to include a white space or # character as
4069 part of the pattern.
4070
4071 If you want to remove the special meaning from a sequence of charac-
4072 ters, you can do so by putting them between \Q and \E. This is differ-
4073 ent from Perl in that $ and @ are handled as literals in \Q...\E
4074 sequences in PCRE, whereas in Perl, $ and @ cause variable interpola-
4075 tion. Note the following examples:
4076
4077 Pattern PCRE matches Perl matches
4078
4079 \Qabc$xyz\E abc$xyz abc followed by the
4080 contents of $xyz
4081 \Qabc\$xyz\E abc\$xyz abc\$xyz
4082 \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
4083
4084 The \Q...\E sequence is recognized both inside and outside character
4085 classes. An isolated \E that is not preceded by \Q is ignored. If \Q
4086 is not followed by \E later in the pattern, the literal interpretation
4087 continues to the end of the pattern (that is, \E is assumed at the
4088 end). If the isolated \Q is inside a character class, this causes an
4089 error, because the character class is not terminated.
4090
4091 Non-printing characters
4092
4093 A second use of backslash provides a way of encoding non-printing char-
4094 acters in patterns in a visible manner. There is no restriction on the
4095 appearance of non-printing characters, apart from the binary zero that
4096 terminates a pattern, but when a pattern is being prepared by text
4097 editing, it is often easier to use one of the following escape
4098 sequences than the binary character it represents:
4099
4100 \a alarm, that is, the BEL character (hex 07)
4101 \cx "control-x", where x is any ASCII character
4102 \e escape (hex 1B)
4103 \f form feed (hex 0C)
4104 \n linefeed (hex 0A)
4105 \r carriage return (hex 0D)
4106 \t tab (hex 09)
4107 \ddd character with octal code ddd, or back reference
4108 \xhh character with hex code hh
4109 \x{hhh..} character with hex code hhh.. (non-JavaScript mode)
4110 \uhhhh character with hex code hhhh (JavaScript mode only)
4111
4112 The precise effect of \cx is as follows: if x is a lower case letter,
4113 it is converted to upper case. Then bit 6 of the character (hex 40) is
4114 inverted. Thus \cz becomes hex 1A (z is 7A), but \c{ becomes hex 3B ({
4115 is 7B), while \c; becomes hex 7B (; is 3B). If the byte following \c
4116 has a value greater than 127, a compile-time error occurs. This locks
4117 out non-ASCII characters in all modes. (When PCRE is compiled in EBCDIC
4118 mode, all byte values are valid. A lower case letter is converted to
4119 upper case, and then the 0xc0 bits are flipped.)
4120
4121 By default, after \x, from zero to two hexadecimal digits are read
4122 (letters can be in upper or lower case). Any number of hexadecimal dig-
4123 its may appear between \x{ and }, but the character code is constrained
4124 as follows:
4125
4126 8-bit non-UTF mode less than 0x100
4127 8-bit UTF-8 mode less than 0x10ffff and a valid codepoint
4128 16-bit non-UTF mode less than 0x10000
4129 16-bit UTF-16 mode less than 0x10ffff and a valid codepoint
4130
4131 Invalid Unicode codepoints are the range 0xd800 to 0xdfff (the so-
4132 called "surrogate" codepoints).
4133
4134 If characters other than hexadecimal digits appear between \x{ and },
4135 or if there is no terminating }, this form of escape is not recognized.
4136 Instead, the initial \x will be interpreted as a basic hexadecimal
4137 escape, with no following digits, giving a character whose value is
4138 zero.
4139
4140 If the PCRE_JAVASCRIPT_COMPAT option is set, the interpretation of \x
4141 is as just described only when it is followed by two hexadecimal dig-
4142 its. Otherwise, it matches a literal "x" character. In JavaScript
4143 mode, support for code points greater than 256 is provided by \u, which
4144 must be followed by four hexadecimal digits; otherwise it matches a
4145 literal "u" character. Character codes specified by \u in JavaScript
4146 mode are constrained in the same was as those specified by \x in non-
4147 JavaScript mode.
4148
4149 Characters whose value is less than 256 can be defined by either of the
4150 two syntaxes for \x (or by \u in JavaScript mode). There is no differ-
4151 ence in the way they are handled. For example, \xdc is exactly the same
4152 as \x{dc} (or \u00dc in JavaScript mode).
4153
4154 After \0 up to two further octal digits are read. If there are fewer
4155 than two digits, just those that are present are used. Thus the
4156 sequence \0\x\07 specifies two binary zeros followed by a BEL character
4157 (code value 7). Make sure you supply two digits after the initial zero
4158 if the pattern character that follows is itself an octal digit.
4159
4160 The handling of a backslash followed by a digit other than 0 is compli-
4161 cated. Outside a character class, PCRE reads it and any following dig-
4162 its as a decimal number. If the number is less than 10, or if there
4163 have been at least that many previous capturing left parentheses in the
4164 expression, the entire sequence is taken as a back reference. A
4165 description of how this works is given later, following the discussion
4166 of parenthesized subpatterns.
4167
4168 Inside a character class, or if the decimal number is greater than 9
4169 and there have not been that many capturing subpatterns, PCRE re-reads
4170 up to three octal digits following the backslash, and uses them to gen-
4171 erate a data character. Any subsequent digits stand for themselves. The
4172 value of the character is constrained in the same way as characters
4173 specified in hexadecimal. For example:
4174
4175 \040 is another way of writing a space
4176 \40 is the same, provided there are fewer than 40
4177 previous capturing subpatterns
4178 \7 is always a back reference
4179 \11 might be a back reference, or another way of
4180 writing a tab
4181 \011 is always a tab
4182 \0113 is a tab followed by the character "3"
4183 \113 might be a back reference, otherwise the
4184 character with octal code 113
4185 \377 might be a back reference, otherwise
4186 the value 255 (decimal)
4187 \81 is either a back reference, or a binary zero
4188 followed by the two characters "8" and "1"
4189
4190 Note that octal values of 100 or greater must not be introduced by a
4191 leading zero, because no more than three octal digits are ever read.
4192
4193 All the sequences that define a single character value can be used both
4194 inside and outside character classes. In addition, inside a character
4195 class, \b is interpreted as the backspace character (hex 08).
4196
4197 \N is not allowed in a character class. \B, \R, and \X are not special
4198 inside a character class. Like other unrecognized escape sequences,
4199 they are treated as the literal characters "B", "R", and "X" by
4200 default, but cause an error if the PCRE_EXTRA option is set. Outside a
4201 character class, these sequences have different meanings.
4202
4203 Unsupported escape sequences
4204
4205 In Perl, the sequences \l, \L, \u, and \U are recognized by its string
4206 handler and used to modify the case of following characters. By
4207 default, PCRE does not support these escape sequences. However, if the
4208 PCRE_JAVASCRIPT_COMPAT option is set, \U matches a "U" character, and
4209 \u can be used to define a character by code point, as described in the
4210 previous section.
4211
4212 Absolute and relative back references
4213
4214 The sequence \g followed by an unsigned or a negative number, option-
4215 ally enclosed in braces, is an absolute or relative back reference. A
4216 named back reference can be coded as \g{name}. Back references are dis-
4217 cussed later, following the discussion of parenthesized subpatterns.
4218
4219 Absolute and relative subroutine calls
4220
4221 For compatibility with Oniguruma, the non-Perl syntax \g followed by a
4222 name or a number enclosed either in angle brackets or single quotes, is
4223 an alternative syntax for referencing a subpattern as a "subroutine".
4224 Details are discussed later. Note that \g{...} (Perl syntax) and
4225 \g<...> (Oniguruma syntax) are not synonymous. The former is a back
4226 reference; the latter is a subroutine call.
4227
4228 Generic character types
4229
4230 Another use of backslash is for specifying generic character types:
4231
4232 \d any decimal digit
4233 \D any character that is not a decimal digit
4234 \h any horizontal white space character
4235 \H any character that is not a horizontal white space character
4236 \s any white space character
4237 \S any character that is not a white space character
4238 \v any vertical white space character
4239 \V any character that is not a vertical white space character
4240 \w any "word" character
4241 \W any "non-word" character
4242
4243 There is also the single sequence \N, which matches a non-newline char-
4244 acter. This is the same as the "." metacharacter when PCRE_DOTALL is
4245 not set. Perl also uses \N to match characters by name; PCRE does not
4246 support this.
4247
4248 Each pair of lower and upper case escape sequences partitions the com-
4249 plete set of characters into two disjoint sets. Any given character
4250 matches one, and only one, of each pair. The sequences can appear both
4251 inside and outside character classes. They each match one character of
4252 the appropriate type. If the current matching point is at the end of
4253 the subject string, all of them fail, because there is no character to
4254 match.
4255
4256 For compatibility with Perl, \s does not match the VT character (code
4257 11). This makes it different from the the POSIX "space" class. The \s
4258 characters are HT (9), LF (10), FF (12), CR (13), and space (32). If
4259 "use locale;" is included in a Perl script, \s may match the VT charac-
4260 ter. In PCRE, it never does.
4261
4262 A "word" character is an underscore or any character that is a letter
4263 or digit. By default, the definition of letters and digits is con-
4264 trolled by PCRE's low-valued character tables, and may vary if locale-
4265 specific matching is taking place (see "Locale support" in the pcreapi
4266 page). For example, in a French locale such as "fr_FR" in Unix-like
4267 systems, or "french" in Windows, some character codes greater than 128
4268 are used for accented letters, and these are then matched by \w. The
4269 use of locales with Unicode is discouraged.
4270
4271 By default, in a UTF mode, characters with values greater than 128
4272 never match \d, \s, or \w, and always match \D, \S, and \W. These
4273 sequences retain their original meanings from before UTF support was
4274 available, mainly for efficiency reasons. However, if PCRE is compiled
4275 with Unicode property support, and the PCRE_UCP option is set, the be-
4276 haviour is changed so that Unicode properties are used to determine
4277 character types, as follows:
4278
4279 \d any character that \p{Nd} matches (decimal digit)
4280 \s any character that \p{Z} matches, plus HT, LF, FF, CR
4281 \w any character that \p{L} or \p{N} matches, plus underscore
4282
4283 The upper case escapes match the inverse sets of characters. Note that
4284 \d matches only decimal digits, whereas \w matches any Unicode digit,
4285 as well as any Unicode letter, and underscore. Note also that PCRE_UCP
4286 affects \b, and \B because they are defined in terms of \w and \W.
4287 Matching these sequences is noticeably slower when PCRE_UCP is set.
4288
4289 The sequences \h, \H, \v, and \V are features that were added to Perl
4290 at release 5.10. In contrast to the other sequences, which match only
4291 ASCII characters by default, these always match certain high-valued
4292 codepoints, whether or not PCRE_UCP is set. The horizontal space char-
4293 acters are:
4294
4295 U+0009 Horizontal tab
4296 U+0020 Space
4297 U+00A0 Non-break space
4298 U+1680 Ogham space mark
4299 U+180E Mongolian vowel separator
4300 U+2000 En quad
4301 U+2001 Em quad
4302 U+2002 En space
4303 U+2003 Em space
4304 U+2004 Three-per-em space
4305 U+2005 Four-per-em space
4306 U+2006 Six-per-em space
4307 U+2007 Figure space
4308 U+2008 Punctuation space
4309 U+2009 Thin space
4310 U+200A Hair space
4311 U+202F Narrow no-break space
4312 U+205F Medium mathematical space
4313 U+3000 Ideographic space
4314
4315 The vertical space characters are:
4316
4317 U+000A Linefeed
4318 U+000B Vertical tab
4319 U+000C Form feed
4320 U+000D Carriage return
4321 U+0085 Next line
4322 U+2028 Line separator
4323 U+2029 Paragraph separator
4324
4325 In 8-bit, non-UTF-8 mode, only the characters with codepoints less than
4326 256 are relevant.
4327
4328 Newline sequences
4329
4330 Outside a character class, by default, the escape sequence \R matches
4331 any Unicode newline sequence. In 8-bit non-UTF-8 mode \R is equivalent
4332 to the following:
4333
4334 (?>\r\n|\n|\x0b|\f|\r|\x85)
4335
4336 This is an example of an "atomic group", details of which are given
4337 below. This particular group matches either the two-character sequence
4338 CR followed by LF, or one of the single characters LF (linefeed,
4339 U+000A), VT (vertical tab, U+000B), FF (form feed, U+000C), CR (car-
4340 riage return, U+000D), or NEL (next line, U+0085). The two-character
4341 sequence is treated as a single unit that cannot be split.
4342
4343 In other modes, two additional characters whose codepoints are greater
4344 than 255 are added: LS (line separator, U+2028) and PS (paragraph sepa-
4345 rator, U+2029). Unicode character property support is not needed for
4346 these characters to be recognized.
4347
4348 It is possible to restrict \R to match only CR, LF, or CRLF (instead of
4349 the complete set of Unicode line endings) by setting the option
4350 PCRE_BSR_ANYCRLF either at compile time or when the pattern is matched.
4351 (BSR is an abbrevation for "backslash R".) This can be made the default
4352 when PCRE is built; if this is the case, the other behaviour can be
4353 requested via the PCRE_BSR_UNICODE option. It is also possible to
4354 specify these settings by starting a pattern string with one of the
4355 following sequences:
4356
4357 (*BSR_ANYCRLF) CR, LF, or CRLF only
4358 (*BSR_UNICODE) any Unicode newline sequence
4359
4360 These override the default and the options given to the compiling func-
4361 tion, but they can themselves be overridden by options given to a
4362 matching function. Note that these special settings, which are not
4363 Perl-compatible, are recognized only at the very start of a pattern,
4364 and that they must be in upper case. If more than one of them is
4365 present, the last one is used. They can be combined with a change of
4366 newline convention; for example, a pattern can start with:
4367
4368 (*ANY)(*BSR_ANYCRLF)
4369
4370 They can also be combined with the (*UTF8), (*UTF16), or (*UCP) special
4371 sequences. Inside a character class, \R is treated as an unrecognized
4372 escape sequence, and so matches the letter "R" by default, but causes
4373 an error if PCRE_EXTRA is set.
4374
4375 Unicode character properties
4376
4377 When PCRE is built with Unicode character property support, three addi-
4378 tional escape sequences that match characters with specific properties
4379 are available. When in 8-bit non-UTF-8 mode, these sequences are of
4380 course limited to testing characters whose codepoints are less than
4381 256, but they do work in this mode. The extra escape sequences are:
4382
4383 \p{xx} a character with the xx property
4384 \P{xx} a character without the xx property
4385 \X an extended Unicode sequence
4386
4387 The property names represented by xx above are limited to the Unicode
4388 script names, the general category properties, "Any", which matches any
4389 character (including newline), and some special PCRE properties
4390 (described in the next section). Other Perl properties such as "InMu-
4391 sicalSymbols" are not currently supported by PCRE. Note that \P{Any}
4392 does not match any characters, so always causes a match failure.
4393
4394 Sets of Unicode characters are defined as belonging to certain scripts.
4395 A character from one of these sets can be matched using a script name.
4396 For example:
4397
4398 \p{Greek}
4399 \P{Han}
4400
4401 Those that are not part of an identified script are lumped together as
4402 "Common". The current list of scripts is:
4403
4404 Arabic, Armenian, Avestan, Balinese, Bamum, Batak, Bengali, Bopomofo,
4405 Brahmi, Braille, Buginese, Buhid, Canadian_Aboriginal, Carian, Chakma,
4406 Cham, Cherokee, Common, Coptic, Cuneiform, Cypriot, Cyrillic, Deseret,
4407 Devanagari, Egyptian_Hieroglyphs, Ethiopic, Georgian, Glagolitic,
4408 Gothic, Greek, Gujarati, Gurmukhi, Han, Hangul, Hanunoo, Hebrew, Hira-
4409 gana, Imperial_Aramaic, Inherited, Inscriptional_Pahlavi, Inscrip-
4410 tional_Parthian, Javanese, Kaithi, Kannada, Katakana, Kayah_Li,
4411 Kharoshthi, Khmer, Lao, Latin, Lepcha, Limbu, Linear_B, Lisu, Lycian,
4412 Lydian, Malayalam, Mandaic, Meetei_Mayek, Meroitic_Cursive,
4413 Meroitic_Hieroglyphs, Miao, Mongolian, Myanmar, New_Tai_Lue, Nko,
4414 Ogham, Old_Italic, Old_Persian, Old_South_Arabian, Old_Turkic,
4415 Ol_Chiki, Oriya, Osmanya, Phags_Pa, Phoenician, Rejang, Runic, Samari-
4416 tan, Saurashtra, Sharada, Shavian, Sinhala, Sora_Sompeng, Sundanese,
4417 Syloti_Nagri, Syriac, Tagalog, Tagbanwa, Tai_Le, Tai_Tham, Tai_Viet,
4418 Takri, Tamil, Telugu, Thaana, Thai, Tibetan, Tifinagh, Ugaritic, Vai,
4419 Yi.
4420
4421 Each character has exactly one Unicode general category property, spec-
4422 ified by a two-letter abbreviation. For compatibility with Perl, nega-
4423 tion can be specified by including a circumflex between the opening
4424 brace and the property name. For example, \p{^Lu} is the same as
4425 \P{Lu}.
4426
4427 If only one letter is specified with \p or \P, it includes all the gen-
4428 eral category properties that start with that letter. In this case, in
4429 the absence of negation, the curly brackets in the escape sequence are
4430 optional; these two examples have the same effect:
4431
4432 \p{L}
4433 \pL
4434
4435 The following general category property codes are supported:
4436
4437 C Other
4438 Cc Control
4439 Cf Format
4440 Cn Unassigned
4441 Co Private use
4442 Cs Surrogate
4443
4444 L Letter
4445 Ll Lower case letter
4446 Lm Modifier letter
4447 Lo Other letter
4448 Lt Title case letter
4449 Lu Upper case letter
4450
4451 M Mark
4452 Mc Spacing mark
4453 Me Enclosing mark
4454 Mn Non-spacing mark
4455
4456 N Number
4457 Nd Decimal number
4458 Nl Letter number
4459 No Other number
4460
4461 P Punctuation
4462 Pc Connector punctuation
4463 Pd Dash punctuation
4464 Pe Close punctuation
4465 Pf Final punctuation
4466 Pi Initial punctuation
4467 Po Other punctuation
4468 Ps Open punctuation
4469
4470 S Symbol
4471 Sc Currency symbol
4472 Sk Modifier symbol
4473 Sm Mathematical symbol
4474 So Other symbol
4475
4476 Z Separator
4477 Zl Line separator
4478 Zp Paragraph separator
4479 Zs Space separator
4480
4481 The special property L& is also supported: it matches a character that
4482 has the Lu, Ll, or Lt property, in other words, a letter that is not
4483 classified as a modifier or "other".
4484
4485 The Cs (Surrogate) property applies only to characters in the range
4486 U+D800 to U+DFFF. Such characters are not valid in Unicode strings and
4487 so cannot be tested by PCRE, unless UTF validity checking has been
4488 turned off (see the discussion of PCRE_NO_UTF8_CHECK and
4489 PCRE_NO_UTF16_CHECK in the pcreapi page). Perl does not support the Cs
4490 property.
4491
4492 The long synonyms for property names that Perl supports (such as
4493 \p{Letter}) are not supported by PCRE, nor is it permitted to prefix
4494 any of these properties with "Is".
4495
4496 No character that is in the Unicode table has the Cn (unassigned) prop-
4497 erty. Instead, this property is assumed for any code point that is not
4498 in the Unicode table.
4499
4500 Specifying caseless matching does not affect these escape sequences.
4501 For example, \p{Lu} always matches only upper case letters.
4502
4503 The \X escape matches any number of Unicode characters that form an
4504 extended Unicode sequence. \X is equivalent to
4505
4506 (?>\PM\pM*)
4507
4508 That is, it matches a character without the "mark" property, followed
4509 by zero or more characters with the "mark" property, and treats the
4510 sequence as an atomic group (see below). Characters with the "mark"
4511 property are typically accents that affect the preceding character.
4512 None of them have codepoints less than 256, so in 8-bit non-UTF-8 mode
4513 \X matches any one character.
4514
4515 Note that recent versions of Perl have changed \X to match what Unicode
4516 calls an "extended grapheme cluster", which has a more complicated def-
4517 inition.
4518
4519 Matching characters by Unicode property is not fast, because PCRE has
4520 to search a structure that contains data for over fifteen thousand
4521 characters. That is why the traditional escape sequences such as \d and
4522 \w do not use Unicode properties in PCRE by default, though you can
4523 make them do so by setting the PCRE_UCP option or by starting the pat-
4524 tern with (*UCP).
4525
4526 PCRE's additional properties
4527
4528 As well as the standard Unicode properties described in the previous
4529 section, PCRE supports four more that make it possible to convert tra-
4530 ditional escape sequences such as \w and \s and POSIX character classes
4531 to use Unicode properties. PCRE uses these non-standard, non-Perl prop-
4532 erties internally when PCRE_UCP is set. They are:
4533
4534 Xan Any alphanumeric character
4535 Xps Any POSIX space character
4536 Xsp Any Perl space character
4537 Xwd Any Perl "word" character
4538
4539 Xan matches characters that have either the L (letter) or the N (num-
4540 ber) property. Xps matches the characters tab, linefeed, vertical tab,
4541 form feed, or carriage return, and any other character that has the Z
4542 (separator) property. Xsp is the same as Xps, except that vertical tab
4543 is excluded. Xwd matches the same characters as Xan, plus underscore.
4544
4545 Resetting the match start
4546
4547 The escape sequence \K causes any previously matched characters not to
4548 be included in the final matched sequence. For example, the pattern:
4549
4550 foo\Kbar
4551
4552 matches "foobar", but reports that it has matched "bar". This feature
4553 is similar to a lookbehind assertion (described below). However, in
4554 this case, the part of the subject before the real match does not have
4555 to be of fixed length, as lookbehind assertions do. The use of \K does
4556 not interfere with the setting of captured substrings. For example,
4557 when the pattern
4558
4559 (foo)\Kbar
4560
4561 matches "foobar", the first substring is still set to "foo".
4562
4563 Perl documents that the use of \K within assertions is "not well
4564 defined". In PCRE, \K is acted upon when it occurs inside positive
4565 assertions, but is ignored in negative assertions.
4566
4567 Simple assertions
4568
4569 The final use of backslash is for certain simple assertions. An asser-
4570 tion specifies a condition that has to be met at a particular point in
4571 a match, without consuming any characters from the subject string. The
4572 use of subpatterns for more complicated assertions is described below.
4573 The backslashed assertions are:
4574
4575 \b matches at a word boundary
4576 \B matches when not at a word boundary
4577 \A matches at the start of the subject
4578 \Z matches at the end of the subject
4579 also matches before a newline at the end of the subject
4580 \z matches only at the end of the subject
4581 \G matches at the first matching position in the subject
4582
4583 Inside a character class, \b has a different meaning; it matches the
4584 backspace character. If any other of these assertions appears in a
4585 character class, by default it matches the corresponding literal char-
4586 acter (for example, \B matches the letter B). However, if the
4587 PCRE_EXTRA option is set, an "invalid escape sequence" error is gener-
4588 ated instead.
4589
4590 A word boundary is a position in the subject string where the current
4591 character and the previous character do not both match \w or \W (i.e.
4592 one matches \w and the other matches \W), or the start or end of the
4593 string if the first or last character matches \w, respectively. In a
4594 UTF mode, the meanings of \w and \W can be changed by setting the
4595 PCRE_UCP option. When this is done, it also affects \b and \B. Neither
4596 PCRE nor Perl has a separate "start of word" or "end of word" metase-
4597 quence. However, whatever follows \b normally determines which it is.
4598 For example, the fragment \ba matches "a" at the start of a word.
4599
4600 The \A, \Z, and \z assertions differ from the traditional circumflex
4601 and dollar (described in the next section) in that they only ever match
4602 at the very start and end of the subject string, whatever options are
4603 set. Thus, they are independent of multiline mode. These three asser-
4604 tions are not affected by the PCRE_NOTBOL or PCRE_NOTEOL options, which
4605 affect only the behaviour of the circumflex and dollar metacharacters.
4606 However, if the startoffset argument of pcre_exec() is non-zero, indi-
4607 cating that matching is to start at a point other than the beginning of
4608 the subject, \A can never match. The difference between \Z and \z is
4609 that \Z matches before a newline at the end of the string as well as at
4610 the very end, whereas \z matches only at the end.
4611
4612 The \G assertion is true only when the current matching position is at
4613 the start point of the match, as specified by the startoffset argument
4614 of pcre_exec(). It differs from \A when the value of startoffset is
4615 non-zero. By calling pcre_exec() multiple times with appropriate argu-
4616 ments, you can mimic Perl's /g option, and it is in this kind of imple-
4617 mentation where \G can be useful.
4618
4619 Note, however, that PCRE's interpretation of \G, as the start of the
4620 current match, is subtly different from Perl's, which defines it as the
4621 end of the previous match. In Perl, these can be different when the
4622 previously matched string was empty. Because PCRE does just one match
4623 at a time, it cannot reproduce this behaviour.
4624
4625 If all the alternatives of a pattern begin with \G, the expression is
4626 anchored to the starting match position, and the "anchored" flag is set
4627 in the compiled regular expression.
4628
4629
4630 CIRCUMFLEX AND DOLLAR
4631
4632 Outside a character class, in the default matching mode, the circumflex
4633 character is an assertion that is true only if the current matching
4634 point is at the start of the subject string. If the startoffset argu-
4635 ment of pcre_exec() is non-zero, circumflex can never match if the
4636 PCRE_MULTILINE option is unset. Inside a character class, circumflex
4637 has an entirely different meaning (see below).
4638
4639 Circumflex need not be the first character of the pattern if a number
4640 of alternatives are involved, but it should be the first thing in each
4641 alternative in which it appears if the pattern is ever to match that
4642 branch. If all possible alternatives start with a circumflex, that is,
4643 if the pattern is constrained to match only at the start of the sub-
4644 ject, it is said to be an "anchored" pattern. (There are also other
4645 constructs that can cause a pattern to be anchored.)
4646
4647 A dollar character is an assertion that is true only if the current
4648 matching point is at the end of the subject string, or immediately
4649 before a newline at the end of the string (by default). Dollar need not
4650 be the last character of the pattern if a number of alternatives are
4651 involved, but it should be the last item in any branch in which it
4652 appears. Dollar has no special meaning in a character class.
4653
4654 The meaning of dollar can be changed so that it matches only at the
4655 very end of the string, by setting the PCRE_DOLLAR_ENDONLY option at
4656 compile time. This does not affect the \Z assertion.
4657
4658 The meanings of the circumflex and dollar characters are changed if the
4659 PCRE_MULTILINE option is set. When this is the case, a circumflex
4660 matches immediately after internal newlines as well as at the start of
4661 the subject string. It does not match after a newline that ends the
4662 string. A dollar matches before any newlines in the string, as well as
4663 at the very end, when PCRE_MULTILINE is set. When newline is specified
4664 as the two-character sequence CRLF, isolated CR and LF characters do
4665 not indicate newlines.
4666
4667 For example, the pattern /^abc$/ matches the subject string "def\nabc"
4668 (where \n represents a newline) in multiline mode, but not otherwise.
4669 Consequently, patterns that are anchored in single line mode because
4670 all branches start with ^ are not anchored in multiline mode, and a
4671 match for circumflex is possible when the startoffset argument of
4672 pcre_exec() is non-zero. The PCRE_DOLLAR_ENDONLY option is ignored if
4673 PCRE_MULTILINE is set.
4674
4675 Note that the sequences \A, \Z, and \z can be used to match the start
4676 and end of the subject in both modes, and if all branches of a pattern
4677 start with \A it is always anchored, whether or not PCRE_MULTILINE is
4678 set.
4679
4680
4681 FULL STOP (PERIOD, DOT) AND \N
4682
4683 Outside a character class, a dot in the pattern matches any one charac-
4684 ter in the subject string except (by default) a character that signi-
4685 fies the end of a line.
4686
4687 When a line ending is defined as a single character, dot never matches
4688 that character; when the two-character sequence CRLF is used, dot does
4689 not match CR if it is immediately followed by LF, but otherwise it
4690 matches all characters (including isolated CRs and LFs). When any Uni-
4691 code line endings are being recognized, dot does not match CR or LF or
4692 any of the other line ending characters.
4693
4694 The behaviour of dot with regard to newlines can be changed. If the
4695 PCRE_DOTALL option is set, a dot matches any one character, without
4696 exception. If the two-character sequence CRLF is present in the subject
4697 string, it takes two dots to match it.
4698
4699 The handling of dot is entirely independent of the handling of circum-
4700 flex and dollar, the only relationship being that they both involve
4701 newlines. Dot has no special meaning in a character class.
4702
4703 The escape sequence \N behaves like a dot, except that it is not
4704 affected by the PCRE_DOTALL option. In other words, it matches any
4705 character except one that signifies the end of a line. Perl also uses
4706 \N to match characters by name; PCRE does not support this.
4707
4708
4709 MATCHING A SINGLE DATA UNIT
4710
4711 Outside a character class, the escape sequence \C matches any one data
4712 unit, whether or not a UTF mode is set. In the 8-bit library, one data
4713 unit is one byte; in the 16-bit library it is a 16-bit unit. Unlike a
4714 dot, \C always matches line-ending characters. The feature is provided
4715 in Perl in order to match individual bytes in UTF-8 mode, but it is
4716 unclear how it can usefully be used. Because \C breaks up characters
4717 into individual data units, matching one unit with \C in a UTF mode
4718 means that the rest of the string may start with a malformed UTF char-
4719 acter. This has undefined results, because PCRE assumes that it is
4720 dealing with valid UTF strings (and by default it checks this at the
4721 start of processing unless the PCRE_NO_UTF8_CHECK or
4722 PCRE_NO_UTF16_CHECK option is used).
4723
4724 PCRE does not allow \C to appear in lookbehind assertions (described
4725 below) in a UTF mode, because this would make it impossible to calcu-
4726 late the length of the lookbehind.
4727
4728 In general, the \C escape sequence is best avoided. However, one way of
4729 using it that avoids the problem of malformed UTF characters is to use
4730 a lookahead to check the length of the next character, as in this pat-
4731 tern, which could be used with a UTF-8 string (ignore white space and
4732 line breaks):
4733
4734 (?| (?=[\x00-\x7f])(\C) |
4735 (?=[\x80-\x{7ff}])(\C)(\C) |
4736 (?=[\x{800}-\x{ffff}])(\C)(\C)(\C) |
4737 (?=[\x{10000}-\x{1fffff}])(\C)(\C)(\C)(\C))
4738
4739 A group that starts with (?| resets the capturing parentheses numbers
4740 in each alternative (see "Duplicate Subpattern Numbers" below). The
4741 assertions at the start of each branch check the next UTF-8 character
4742 for values whose encoding uses 1, 2, 3, or 4 bytes, respectively. The
4743 character's individual bytes are then captured by the appropriate num-
4744 ber of groups.
4745
4746
4747 SQUARE BRACKETS AND CHARACTER CLASSES
4748
4749 An opening square bracket introduces a character class, terminated by a
4750 closing square bracket. A closing square bracket on its own is not spe-
4751 cial by default. However, if the PCRE_JAVASCRIPT_COMPAT option is set,
4752 a lone closing square bracket causes a compile-time error. If a closing
4753 square bracket is required as a member of the class, it should be the
4754 first data character in the class (after an initial circumflex, if
4755 present) or escaped with a backslash.
4756
4757 A character class matches a single character in the subject. In a UTF
4758 mode, the character may be more than one data unit long. A matched
4759 character must be in the set of characters defined by the class, unless
4760 the first character in the class definition is a circumflex, in which
4761 case the subject character must not be in the set defined by the class.
4762 If a circumflex is actually required as a member of the class, ensure
4763 it is not the first character, or escape it with a backslash.
4764
4765 For example, the character class [aeiou] matches any lower case vowel,
4766 while [^aeiou] matches any character that is not a lower case vowel.
4767 Note that a circumflex is just a convenient notation for specifying the
4768 characters that are in the class by enumerating those that are not. A
4769 class that starts with a circumflex is not an assertion; it still con-
4770 sumes a character from the subject string, and therefore it fails if
4771 the current pointer is at the end of the string.
4772
4773 In UTF-8 (UTF-16) mode, characters with values greater than 255
4774 (0xffff) can be included in a class as a literal string of data units,
4775 or by using the \x{ escaping mechanism.
4776
4777 When caseless matching is set, any letters in a class represent both
4778 their upper case and lower case versions, so for example, a caseless
4779 [aeiou] matches "A" as well as "a", and a caseless [^aeiou] does not
4780 match "A", whereas a caseful version would. In a UTF mode, PCRE always
4781 understands the concept of case for characters whose values are less
4782 than 128, so caseless matching is always possible. For characters with
4783 higher values, the concept of case is supported if PCRE is compiled
4784 with Unicode property support, but not otherwise. If you want to use
4785 caseless matching in a UTF mode for characters 128 and above, you must
4786 ensure that PCRE is compiled with Unicode property support as well as
4787 with UTF support.
4788
4789 Characters that might indicate line breaks are never treated in any
4790 special way when matching character classes, whatever line-ending
4791 sequence is in use, and whatever setting of the PCRE_DOTALL and
4792 PCRE_MULTILINE options is used. A class such as [^a] always matches one
4793 of these characters.
4794
4795 The minus (hyphen) character can be used to specify a range of charac-
4796 ters in a character class. For example, [d-m] matches any letter
4797 between d and m, inclusive. If a minus character is required in a
4798 class, it must be escaped with a backslash or appear in a position
4799 where it cannot be interpreted as indicating a range, typically as the
4800 first or last character in the class.
4801
4802 It is not possible to have the literal character "]" as the end charac-
4803 ter of a range. A pattern such as [W-]46] is interpreted as a class of
4804 two characters ("W" and "-") followed by a literal string "46]", so it
4805 would match "W46]" or "-46]". However, if the "]" is escaped with a
4806 backslash it is interpreted as the end of range, so [W-\]46] is inter-
4807 preted as a class containing a range followed by two other characters.
4808 The octal or hexadecimal representation of "]" can also be used to end
4809 a range.
4810
4811 Ranges operate in the collating sequence of character values. They can
4812 also be used for characters specified numerically, for example
4813 [\000-\037]. Ranges can include any characters that are valid for the
4814 current mode.
4815
4816 If a range that includes letters is used when caseless matching is set,
4817 it matches the letters in either case. For example, [W-c] is equivalent
4818 to [][\\^_`wxyzabc], matched caselessly, and in a non-UTF mode, if
4819 character tables for a French locale are in use, [\xc8-\xcb] matches
4820 accented E characters in both cases. In UTF modes, PCRE supports the
4821 concept of case for characters with values greater than 128 only when
4822 it is compiled with Unicode property support.
4823
4824 The character escape sequences \d, \D, \h, \H, \p, \P, \s, \S, \v, \V,
4825 \w, and \W may appear in a character class, and add the characters that
4826 they match to the class. For example, [\dABCDEF] matches any hexadeci-
4827 mal digit. In UTF modes, the PCRE_UCP option affects the meanings of
4828 \d, \s, \w and their upper case partners, just as it does when they
4829 appear outside a character class, as described in the section entitled
4830 "Generic character types" above. The escape sequence \b has a different
4831 meaning inside a character class; it matches the backspace character.
4832 The sequences \B, \N, \R, and \X are not special inside a character
4833 class. Like any other unrecognized escape sequences, they are treated
4834 as the literal characters "B", "N", "R", and "X" by default, but cause
4835 an error if the PCRE_EXTRA option is set.
4836
4837 A circumflex can conveniently be used with the upper case character
4838 types to specify a more restricted set of characters than the matching
4839 lower case type. For example, the class [^\W_] matches any letter or
4840 digit, but not underscore, whereas [\w] includes underscore. A positive
4841 character class should be read as "something OR something OR ..." and a
4842 negative class as "NOT something AND NOT something AND NOT ...".
4843
4844 The only metacharacters that are recognized in character classes are
4845 backslash, hyphen (only where it can be interpreted as specifying a
4846 range), circumflex (only at the start), opening square bracket (only
4847 when it can be interpreted as introducing a POSIX class name - see the
4848 next section), and the terminating closing square bracket. However,
4849 escaping other non-alphanumeric characters does no harm.
4850
4851
4852 POSIX CHARACTER CLASSES
4853
4854 Perl supports the POSIX notation for character classes. This uses names
4855 enclosed by [: and :] within the enclosing square brackets. PCRE also
4856 supports this notation. For example,
4857
4858 [01[:alpha:]%]
4859
4860 matches "0", "1", any alphabetic character, or "%". The supported class
4861 names are:
4862
4863 alnum letters and digits
4864 alpha letters
4865 ascii character codes 0 - 127
4866 blank space or tab only
4867 cntrl control characters
4868 digit decimal digits (same as \d)
4869 graph printing characters, excluding space
4870 lower lower case letters
4871 print printing characters, including space
4872 punct printing characters, excluding letters and digits and space
4873 space white space (not quite the same as \s)
4874 upper upper case letters
4875 word "word" characters (same as \w)
4876 xdigit hexadecimal digits
4877
4878 The "space" characters are HT (9), LF (10), VT (11), FF (12), CR (13),
4879 and space (32). Notice that this list includes the VT character (code
4880 11). This makes "space" different to \s, which does not include VT (for
4881 Perl compatibility).
4882
4883 The name "word" is a Perl extension, and "blank" is a GNU extension
4884 from Perl 5.8. Another Perl extension is negation, which is indicated
4885 by a ^ character after the colon. For example,
4886
4887 [12[:^digit:]]
4888
4889 matches "1", "2", or any non-digit. PCRE (and Perl) also recognize the
4890 POSIX syntax [.ch.] and [=ch=] where "ch" is a "collating element", but
4891 these are not supported, and an error is given if they are encountered.
4892
4893 By default, in UTF modes, characters with values greater than 128 do
4894 not match any of the POSIX character classes. However, if the PCRE_UCP
4895 option is passed to pcre_compile(), some of the classes are changed so
4896 that Unicode character properties are used. This is achieved by replac-
4897 ing the POSIX classes by other sequences, as follows:
4898
4899 [:alnum:] becomes \p{Xan}
4900 [:alpha:] becomes \p{L}
4901 [:blank:] becomes \h
4902 [:digit:] becomes \p{Nd}
4903 [:lower:] becomes \p{Ll}
4904 [:space:] becomes \p{Xps}
4905 [:upper:] becomes \p{Lu}
4906 [:word:] becomes \p{Xwd}
4907
4908 Negated versions, such as [:^alpha:] use \P instead of \p. The other
4909 POSIX classes are unchanged, and match only characters with code points
4910 less than 128.
4911
4912
4913 VERTICAL BAR
4914
4915 Vertical bar characters are used to separate alternative patterns. For
4916 example, the pattern
4917
4918 gilbert|sullivan
4919
4920 matches either "gilbert" or "sullivan". Any number of alternatives may
4921 appear, and an empty alternative is permitted (matching the empty
4922 string). The matching process tries each alternative in turn, from left
4923 to right, and the first one that succeeds is used. If the alternatives
4924 are within a subpattern (defined below), "succeeds" means matching the
4925 rest of the main pattern as well as the alternative in the subpattern.
4926
4927
4928 INTERNAL OPTION SETTING
4929
4930 The settings of the PCRE_CASELESS, PCRE_MULTILINE, PCRE_DOTALL, and
4931 PCRE_EXTENDED options (which are Perl-compatible) can be changed from
4932 within the pattern by a sequence of Perl option letters enclosed
4933 between "(?" and ")". The option letters are
4934
4935 i for PCRE_CASELESS
4936 m for PCRE_MULTILINE
4937 s for PCRE_DOTALL
4938 x for PCRE_EXTENDED
4939
4940 For example, (?im) sets caseless, multiline matching. It is also possi-
4941 ble to unset these options by preceding the letter with a hyphen, and a
4942 combined setting and unsetting such as (?im-sx), which sets PCRE_CASE-
4943 LESS and PCRE_MULTILINE while unsetting PCRE_DOTALL and PCRE_EXTENDED,
4944 is also permitted. If a letter appears both before and after the
4945 hyphen, the option is unset.
4946
4947 The PCRE-specific options PCRE_DUPNAMES, PCRE_UNGREEDY, and PCRE_EXTRA
4948 can be changed in the same way as the Perl-compatible options by using
4949 the characters J, U and X respectively.
4950
4951 When one of these option changes occurs at top level (that is, not
4952 inside subpattern parentheses), the change applies to the remainder of
4953 the pattern that follows. If the change is placed right at the start of
4954 a pattern, PCRE extracts it into the global options (and it will there-
4955 fore show up in data extracted by the pcre_fullinfo() function).
4956
4957 An option change within a subpattern (see below for a description of
4958 subpatterns) affects only that part of the subpattern that follows it,
4959 so
4960
4961 (a(?i)b)c
4962
4963 matches abc and aBc and no other strings (assuming PCRE_CASELESS is not
4964 used). By this means, options can be made to have different settings
4965 in different parts of the pattern. Any changes made in one alternative
4966 do carry on into subsequent branches within the same subpattern. For
4967 example,
4968
4969 (a(?i)b|c)
4970
4971 matches "ab", "aB", "c", and "C", even though when matching "C" the
4972 first branch is abandoned before the option setting. This is because
4973 the effects of option settings happen at compile time. There would be
4974 some very weird behaviour otherwise.
4975
4976 Note: There are other PCRE-specific options that can be set by the
4977 application when the compiling or matching functions are called. In
4978 some cases the pattern can contain special leading sequences such as
4979 (*CRLF) to override what the application has set or what has been
4980 defaulted. Details are given in the section entitled "Newline
4981 sequences" above. There are also the (*UTF8), (*UTF16), and (*UCP)
4982 leading sequences that can be used to set UTF and Unicode property
4983 modes; they are equivalent to setting the PCRE_UTF8, PCRE_UTF16, and
4984 the PCRE_UCP options, respectively.
4985
4986
4987 SUBPATTERNS
4988
4989 Subpatterns are delimited by parentheses (round brackets), which can be
4990 nested. Turning part of a pattern into a subpattern does two things:
4991
4992 1. It localizes a set of alternatives. For example, the pattern
4993
4994 cat(aract|erpillar|)
4995
4996 matches "cataract", "caterpillar", or "cat". Without the parentheses,
4997 it would match "cataract", "erpillar" or an empty string.
4998
4999 2. It sets up the subpattern as a capturing subpattern. This means
5000 that, when the whole pattern matches, that portion of the subject
5001 string that matched the subpattern is passed back to the caller via the
5002 ovector argument of the matching function. (This applies only to the
5003 traditional matching functions; the DFA matching functions do not sup-
5004 port capturing.)
5005
5006 Opening parentheses are counted from left to right (starting from 1) to
5007 obtain numbers for the capturing subpatterns. For example, if the
5008 string "the red king" is matched against the pattern
5009
5010 the ((red|white) (king|queen))
5011
5012 the captured substrings are "red king", "red", and "king", and are num-
5013 bered 1, 2, and 3, respectively.
5014
5015 The fact that plain parentheses fulfil two functions is not always
5016 helpful. There are often times when a grouping subpattern is required
5017 without a capturing requirement. If an opening parenthesis is followed
5018 by a question mark and a colon, the subpattern does not do any captur-
5019 ing, and is not counted when computing the number of any subsequent
5020 capturing subpatterns. For example, if the string "the white queen" is
5021 matched against the pattern
5022
5023 the ((?:red|white) (king|queen))
5024
5025 the captured substrings are "white queen" and "queen", and are numbered
5026 1 and 2. The maximum number of capturing subpatterns is 65535.
5027
5028 As a convenient shorthand, if any option settings are required at the
5029 start of a non-capturing subpattern, the option letters may appear
5030 between the "?" and the ":". Thus the two patterns
5031
5032 (?i:saturday|sunday)
5033 (?:(?i)saturday|sunday)
5034
5035 match exactly the same set of strings. Because alternative branches are
5036 tried from left to right, and options are not reset until the end of
5037 the subpattern is reached, an option setting in one branch does affect
5038 subsequent branches, so the above patterns match "SUNDAY" as well as
5039 "Saturday".
5040
5041
5042 DUPLICATE SUBPATTERN NUMBERS
5043
5044 Perl 5.10 introduced a feature whereby each alternative in a subpattern
5045 uses the same numbers for its capturing parentheses. Such a subpattern
5046 starts with (?| and is itself a non-capturing subpattern. For example,
5047 consider this pattern:
5048
5049 (?|(Sat)ur|(Sun))day
5050
5051 Because the two alternatives are inside a (?| group, both sets of cap-
5052 turing parentheses are numbered one. Thus, when the pattern matches,
5053 you can look at captured substring number one, whichever alternative
5054 matched. This construct is useful when you want to capture part, but
5055 not all, of one of a number of alternatives. Inside a (?| group, paren-
5056 theses are numbered as usual, but the number is reset at the start of
5057 each branch. The numbers of any capturing parentheses that follow the
5058 subpattern start after the highest number used in any branch. The fol-
5059 lowing example is taken from the Perl documentation. The numbers under-
5060 neath show in which buffer the captured content will be stored.
5061
5062 # before ---------------branch-reset----------- after
5063 / ( a ) (?| x ( y ) z | (p (q) r) | (t) u (v) ) ( z ) /x
5064 # 1 2 2 3 2 3 4
5065
5066 A back reference to a numbered subpattern uses the most recent value
5067 that is set for that number by any subpattern. The following pattern
5068 matches "abcabc" or "defdef":
5069
5070 /(?|(abc)|(def))\1/
5071
5072 In contrast, a subroutine call to a numbered subpattern always refers
5073 to the first one in the pattern with the given number. The following
5074 pattern matches "abcabc" or "defabc":
5075
5076 /(?|(abc)|(def))(?1)/
5077
5078 If a condition test for a subpattern's having matched refers to a non-
5079 unique number, the test is true if any of the subpatterns of that num-
5080 ber have matched.
5081
5082 An alternative approach to using this "branch reset" feature is to use
5083 duplicate named subpatterns, as described in the next section.
5084
5085
5086 NAMED SUBPATTERNS
5087
5088 Identifying capturing parentheses by number is simple, but it can be
5089 very hard to keep track of the numbers in complicated regular expres-
5090 sions. Furthermore, if an expression is modified, the numbers may
5091 change. To help with this difficulty, PCRE supports the naming of sub-
5092 patterns. This feature was not added to Perl until release 5.10. Python
5093 had the feature earlier, and PCRE introduced it at release 4.0, using
5094 the Python syntax. PCRE now supports both the Perl and the Python syn-
5095 tax. Perl allows identically numbered subpatterns to have different
5096 names, but PCRE does not.
5097
5098 In PCRE, a subpattern can be named in one of three ways: (?<name>...)
5099 or (?'name'...) as in Perl, or (?P<name>...) as in Python. References
5100 to capturing parentheses from other parts of the pattern, such as back
5101 references, recursion, and conditions, can be made by name as well as
5102 by number.
5103
5104 Names consist of up to 32 alphanumeric characters and underscores.
5105 Named capturing parentheses are still allocated numbers as well as
5106 names, exactly as if the names were not present. The PCRE API provides
5107 function calls for extracting the name-to-number translation table from
5108 a compiled pattern. There is also a convenience function for extracting
5109 a captured substring by name.
5110
5111 By default, a name must be unique within a pattern, but it is possible
5112 to relax this constraint by setting the PCRE_DUPNAMES option at compile
5113 time. (Duplicate names are also always permitted for subpatterns with
5114 the same number, set up as described in the previous section.) Dupli-
5115 cate names can be useful for patterns where only one instance of the
5116 named parentheses can match. Suppose you want to match the name of a
5117 weekday, either as a 3-letter abbreviation or as the full name, and in
5118 both cases you want to extract the abbreviation. This pattern (ignoring
5119 the line breaks) does the job:
5120
5121 (?<DN>Mon|Fri|Sun)(?:day)?|
5122 (?<DN>Tue)(?:sday)?|
5123 (?<DN>Wed)(?:nesday)?|
5124 (?<DN>Thu)(?:rsday)?|
5125 (?<DN>Sat)(?:urday)?
5126
5127 There are five capturing substrings, but only one is ever set after a
5128 match. (An alternative way of solving this problem is to use a "branch
5129 reset" subpattern, as described in the previous section.)
5130
5131 The convenience function for extracting the data by name returns the
5132 substring for the first (and in this example, the only) subpattern of
5133 that name that matched. This saves searching to find which numbered
5134 subpattern it was.
5135
5136 If you make a back reference to a non-unique named subpattern from
5137 elsewhere in the pattern, the one that corresponds to the first occur-
5138 rence of the name is used. In the absence of duplicate numbers (see the
5139 previous section) this is the one with the lowest number. If you use a
5140 named reference in a condition test (see the section about conditions
5141 below), either to check whether a subpattern has matched, or to check
5142 for recursion, all subpatterns with the same name are tested. If the
5143 condition is true for any one of them, the overall condition is true.
5144 This is the same behaviour as testing by number. For further details of
5145 the interfaces for handling named subpatterns, see the pcreapi documen-
5146 tation.
5147
5148 Warning: You cannot use different names to distinguish between two sub-
5149 patterns with the same number because PCRE uses only the numbers when
5150 matching. For this reason, an error is given at compile time if differ-
5151 ent names are given to subpatterns with the same number. However, you
5152 can give the same name to subpatterns with the same number, even when
5153 PCRE_DUPNAMES is not set.
5154
5155
5156 REPETITION
5157
5158 Repetition is specified by quantifiers, which can follow any of the
5159 following items:
5160
5161 a literal data character
5162 the dot metacharacter
5163 the \C escape sequence
5164 the \X escape sequence
5165 the \R escape sequence
5166 an escape such as \d or \pL that matches a single character
5167 a character class
5168 a back reference (see next section)
5169 a parenthesized subpattern (including assertions)
5170 a subroutine call to a subpattern (recursive or otherwise)
5171
5172 The general repetition quantifier specifies a minimum and maximum num-
5173 ber of permitted matches, by giving the two numbers in curly brackets
5174 (braces), separated by a comma. The numbers must be less than 65536,
5175 and the first must be less than or equal to the second. For example:
5176
5177 z{2,4}
5178
5179 matches "zz", "zzz", or "zzzz". A closing brace on its own is not a
5180 special character. If the second number is omitted, but the comma is
5181 present, there is no upper limit; if the second number and the comma
5182 are both omitted, the quantifier specifies an exact number of required
5183 matches. Thus
5184
5185 [aeiou]{3,}
5186
5187 matches at least 3 successive vowels, but may match many more, while
5188
5189 \d{8}
5190
5191 matches exactly 8 digits. An opening curly bracket that appears in a
5192 position where a quantifier is not allowed, or one that does not match
5193 the syntax of a quantifier, is taken as a literal character. For exam-
5194 ple, {,6} is not a quantifier, but a literal string of four characters.
5195
5196 In UTF modes, quantifiers apply to characters rather than to individual
5197 data units. Thus, for example, \x{100}{2} matches two characters, each
5198 of which is represented by a two-byte sequence in a UTF-8 string. Simi-
5199 larly, \X{3} matches three Unicode extended sequences, each of which
5200 may be several data units long (and they may be of different lengths).
5201
5202 The quantifier {0} is permitted, causing the expression to behave as if
5203 the previous item and the quantifier were not present. This may be use-
5204 ful for subpatterns that are referenced as subroutines from elsewhere
5205 in the pattern (but see also the section entitled "Defining subpatterns
5206 for use by reference only" below). Items other than subpatterns that
5207 have a {0} quantifier are omitted from the compiled pattern.
5208
5209 For convenience, the three most common quantifiers have single-charac-
5210 ter abbreviations:
5211
5212 * is equivalent to {0,}
5213 + is equivalent to {1,}
5214 ? is equivalent to {0,1}
5215
5216 It is possible to construct infinite loops by following a subpattern
5217 that can match no characters with a quantifier that has no upper limit,
5218 for example:
5219
5220 (a?)*
5221
5222 Earlier versions of Perl and PCRE used to give an error at compile time
5223 for such patterns. However, because there are cases where this can be
5224 useful, such patterns are now accepted, but if any repetition of the
5225 subpattern does in fact match no characters, the loop is forcibly bro-
5226 ken.
5227
5228 By default, the quantifiers are "greedy", that is, they match as much
5229 as possible (up to the maximum number of permitted times), without
5230 causing the rest of the pattern to fail. The classic example of where
5231 this gives problems is in trying to match comments in C programs. These
5232 appear between /* and */ and within the comment, individual * and /
5233 characters may appear. An attempt to match C comments by applying the
5234 pattern
5235
5236 /\*.*\*/
5237
5238 to the string
5239
5240 /* first comment */ not comment /* second comment */
5241
5242 fails, because it matches the entire string owing to the greediness of
5243 the .* item.
5244
5245 However, if a quantifier is followed by a question mark, it ceases to
5246 be greedy, and instead matches the minimum number of times possible, so
5247 the pattern
5248
5249 /\*.*?\*/
5250
5251 does the right thing with the C comments. The meaning of the various
5252 quantifiers is not otherwise changed, just the preferred number of
5253 matches. Do not confuse this use of question mark with its use as a
5254 quantifier in its own right. Because it has two uses, it can sometimes
5255 appear doubled, as in
5256
5257 \d??\d
5258
5259 which matches one digit by preference, but can match two if that is the
5260 only way the rest of the pattern matches.
5261
5262 If the PCRE_UNGREEDY option is set (an option that is not available in
5263 Perl), the quantifiers are not greedy by default, but individual ones
5264 can be made greedy by following them with a question mark. In other
5265 words, it inverts the default behaviour.
5266
5267 When a parenthesized subpattern is quantified with a minimum repeat
5268 count that is greater than 1 or with a limited maximum, more memory is
5269 required for the compiled pattern, in proportion to the size of the
5270 minimum or maximum.
5271
5272 If a pattern starts with .* or .{0,} and the PCRE_DOTALL option (equiv-
5273 alent to Perl's /s) is set, thus allowing the dot to match newlines,
5274 the pattern is implicitly anchored, because whatever follows will be
5275 tried against every character position in the subject string, so there
5276 is no point in retrying the overall match at any position after the
5277 first. PCRE normally treats such a pattern as though it were preceded
5278 by \A.
5279
5280 In cases where it is known that the subject string contains no new-
5281 lines, it is worth setting PCRE_DOTALL in order to obtain this opti-
5282 mization, or alternatively using ^ to indicate anchoring explicitly.
5283
5284 However, there is one situation where the optimization cannot be used.
5285 When .* is inside capturing parentheses that are the subject of a back
5286 reference elsewhere in the pattern, a match at the start may fail where
5287 a later one succeeds. Consider, for example:
5288
5289 (.*)abc\1
5290
5291 If the subject is "xyz123abc123" the match point is the fourth charac-
5292 ter. For this reason, such a pattern is not implicitly anchored.
5293
5294 When a capturing subpattern is repeated, the value captured is the sub-
5295 string that matched the final iteration. For example, after
5296
5297 (tweedle[dume]{3}\s*)+
5298
5299 has matched "tweedledum tweedledee" the value of the captured substring
5300 is "tweedledee". However, if there are nested capturing subpatterns,
5301 the corresponding captured values may have been set in previous itera-
5302 tions. For example, after
5303
5304 /(a|(b))+/
5305
5306 matches "aba" the value of the second captured substring is "b".
5307
5308
5309 ATOMIC GROUPING AND POSSESSIVE QUANTIFIERS
5310
5311 With both maximizing ("greedy") and minimizing ("ungreedy" or "lazy")
5312 repetition, failure of what follows normally causes the repeated item
5313 to be re-evaluated to see if a different number of repeats allows the
5314 rest of the pattern to match. Sometimes it is useful to prevent this,
5315 either to change the nature of the match, or to cause it fail earlier
5316 than it otherwise might, when the author of the pattern knows there is
5317 no point in carrying on.
5318
5319 Consider, for example, the pattern \d+foo when applied to the subject
5320 line
5321
5322 123456bar
5323
5324 After matching all 6 digits and then failing to match "foo", the normal
5325 action of the matcher is to try again with only 5 digits matching the
5326 \d+ item, and then with 4, and so on, before ultimately failing.
5327 "Atomic grouping" (a term taken from Jeffrey Friedl's book) provides
5328 the means for specifying that once a subpattern has matched, it is not
5329 to be re-evaluated in this way.
5330
5331 If we use atomic grouping for the previous example, the matcher gives
5332 up immediately on failing to match "foo" the first time. The notation
5333 is a kind of special parenthesis, starting with (?> as in this example:
5334
5335 (?>\d+)foo
5336
5337 This kind of parenthesis "locks up" the part of the pattern it con-
5338 tains once it has matched, and a failure further into the pattern is
5339 prevented from backtracking into it. Backtracking past it to previous
5340 items, however, works as normal.
5341
5342 An alternative description is that a subpattern of this type matches
5343 the string of characters that an identical standalone pattern would
5344 match, if anchored at the current point in the subject string.
5345
5346 Atomic grouping subpatterns are not capturing subpatterns. Simple cases
5347 such as the above example can be thought of as a maximizing repeat that
5348 must swallow everything it can. So, while both \d+ and \d+? are pre-
5349 pared to adjust the number of digits they match in order to make the
5350 rest of the pattern match, (?>\d+) can only match an entire sequence of
5351 digits.
5352
5353 Atomic groups in general can of course contain arbitrarily complicated
5354 subpatterns, and can be nested. However, when the subpattern for an
5355 atomic group is just a single repeated item, as in the example above, a
5356 simpler notation, called a "possessive quantifier" can be used. This
5357 consists of an additional + character following a quantifier. Using
5358 this notation, the previous example can be rewritten as
5359
5360 \d++foo
5361
5362 Note that a possessive quantifier can be used with an entire group, for
5363 example:
5364
5365 (abc|xyz){2,3}+
5366
5367 Possessive quantifiers are always greedy; the setting of the
5368 PCRE_UNGREEDY option is ignored. They are a convenient notation for the
5369 simpler forms of atomic group. However, there is no difference in the
5370 meaning of a possessive quantifier and the equivalent atomic group,
5371 though there may be a performance difference; possessive quantifiers
5372 should be slightly faster.
5373
5374 The possessive quantifier syntax is an extension to the Perl 5.8 syn-
5375 tax. Jeffrey Friedl originated the idea (and the name) in the first
5376 edition of his book. Mike McCloskey liked it, so implemented it when he
5377 built Sun's Java package, and PCRE copied it from there. It ultimately
5378 found its way into Perl at release 5.10.
5379
5380 PCRE has an optimization that automatically "possessifies" certain sim-
5381 ple pattern constructs. For example, the sequence A+B is treated as
5382 A++B because there is no point in backtracking into a sequence of A's
5383 when B must follow.
5384
5385 When a pattern contains an unlimited repeat inside a subpattern that
5386 can itself be repeated an unlimited number of times, the use of an
5387 atomic group is the only way to avoid some failing matches taking a
5388 very long time indeed. The pattern
5389
5390 (\D+|<\d+>)*[!?]
5391
5392 matches an unlimited number of substrings that either consist of non-
5393 digits, or digits enclosed in <>, followed by either ! or ?. When it
5394 matches, it runs quickly. However, if it is applied to
5395
5396 aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
5397
5398 it takes a long time before reporting failure. This is because the
5399 string can be divided between the internal \D+ repeat and the external
5400 * repeat in a large number of ways, and all have to be tried. (The
5401 example uses [!?] rather than a single character at the end, because
5402 both PCRE and Perl have an optimization that allows for fast failure
5403 when a single character is used. They remember the last single charac-
5404 ter that is required for a match, and fail early if it is not present
5405 in the string.) If the pattern is changed so that it uses an atomic
5406 group, like this:
5407
5408 ((?>\D+)|<\d+>)*[!?]
5409
5410 sequences of non-digits cannot be broken, and failure happens quickly.
5411
5412
5413 BACK REFERENCES
5414
5415 Outside a character class, a backslash followed by a digit greater than
5416 0 (and possibly further digits) is a back reference to a capturing sub-
5417 pattern earlier (that is, to its left) in the pattern, provided there
5418 have been that many previous capturing left parentheses.
5419
5420 However, if the decimal number following the backslash is less than 10,
5421 it is always taken as a back reference, and causes an error only if
5422 there are not that many capturing left parentheses in the entire pat-
5423 tern. In other words, the parentheses that are referenced need not be
5424 to the left of the reference for numbers less than 10. A "forward back
5425 reference" of this type can make sense when a repetition is involved
5426 and the subpattern to the right has participated in an earlier itera-
5427 tion.
5428
5429 It is not possible to have a numerical "forward back reference" to a
5430 subpattern whose number is 10 or more using this syntax because a
5431 sequence such as \50 is interpreted as a character defined in octal.
5432 See the subsection entitled "Non-printing characters" above for further
5433 details of the handling of digits following a backslash. There is no
5434 such problem when named parentheses are used. A back reference to any
5435 subpattern is possible using named parentheses (see below).
5436
5437 Another way of avoiding the ambiguity inherent in the use of digits
5438 following a backslash is to use the \g escape sequence. This escape
5439 must be followed by an unsigned number or a negative number, optionally
5440 enclosed in braces. These examples are all identical:
5441
5442 (ring), \1
5443 (ring), \g1
5444 (ring), \g{1}
5445
5446 An unsigned number specifies an absolute reference without the ambigu-
5447 ity that is present in the older syntax. It is also useful when literal
5448 digits follow the reference. A negative number is a relative reference.
5449 Consider this example:
5450
5451 (abc(def)ghi)\g{-1}
5452
5453 The sequence \g{-1} is a reference to the most recently started captur-
5454 ing subpattern before \g, that is, is it equivalent to \2 in this exam-
5455 ple. Similarly, \g{-2} would be equivalent to \1. The use of relative
5456 references can be helpful in long patterns, and also in patterns that
5457 are created by joining together fragments that contain references
5458 within themselves.
5459
5460 A back reference matches whatever actually matched the capturing sub-
5461 pattern in the current subject string, rather than anything matching
5462 the subpattern itself (see "Subpatterns as subroutines" below for a way
5463 of doing that). So the pattern
5464
5465 (sens|respons)e and \1ibility
5466
5467 matches "sense and sensibility" and "response and responsibility", but
5468 not "sense and responsibility". If caseful matching is in force at the
5469 time of the back reference, the case of letters is relevant. For exam-
5470 ple,
5471
5472 ((?i)rah)\s+\1
5473
5474 matches "rah rah" and "RAH RAH", but not "RAH rah", even though the
5475 original capturing subpattern is matched caselessly.
5476
5477 There are several different ways of writing back references to named
5478 subpatterns. The .NET syntax \k{name} and the Perl syntax \k<name> or
5479 \k'name' are supported, as is the Python syntax (?P=name). Perl 5.10's
5480 unified back reference syntax, in which \g can be used for both numeric
5481 and named references, is also supported. We could rewrite the above
5482 example in any of the following ways:
5483
5484 (?<p1>(?i)rah)\s+\k<p1>
5485 (?'p1'(?i)rah)\s+\k{p1}
5486 (?P<p1>(?i)rah)\s+(?P=p1)
5487 (?<p1>(?i)rah)\s+\g{p1}
5488
5489 A subpattern that is referenced by name may appear in the pattern
5490 before or after the reference.
5491
5492 There may be more than one back reference to the same subpattern. If a
5493 subpattern has not actually been used in a particular match, any back
5494 references to it always fail by default. For example, the pattern
5495
5496 (a|(bc))\2
5497
5498 always fails if it starts to match "a" rather than "bc". However, if
5499 the PCRE_JAVASCRIPT_COMPAT option is set at compile time, a back refer-
5500 ence to an unset value matches an empty string.
5501
5502 Because there may be many capturing parentheses in a pattern, all dig-
5503 its following a backslash are taken as part of a potential back refer-
5504 ence number. If the pattern continues with a digit character, some
5505 delimiter must be used to terminate the back reference. If the
5506 PCRE_EXTENDED option is set, this can be white space. Otherwise, the
5507 \g{ syntax or an empty comment (see "Comments" below) can be used.
5508
5509 Recursive back references
5510
5511 A back reference that occurs inside the parentheses to which it refers
5512 fails when the subpattern is first used, so, for example, (a\1) never
5513 matches. However, such references can be useful inside repeated sub-
5514 patterns. For example, the pattern
5515
5516 (a|b\1)+
5517
5518 matches any number of "a"s and also "aba", "ababbaa" etc. At each iter-
5519 ation of the subpattern, the back reference matches the character
5520 string corresponding to the previous iteration. In order for this to
5521 work, the pattern must be such that the first iteration does not need
5522 to match the back reference. This can be done using alternation, as in
5523 the example above, or by a quantifier with a minimum of zero.
5524
5525 Back references of this type cause the group that they reference to be
5526 treated as an atomic group. Once the whole group has been matched, a
5527 subsequent matching failure cannot cause backtracking into the middle
5528 of the group.
5529
5530
5531 ASSERTIONS
5532
5533 An assertion is a test on the characters following or preceding the
5534 current matching point that does not actually consume any characters.
5535 The simple assertions coded as \b, \B, \A, \G, \Z, \z, ^ and $ are
5536 described above.
5537
5538 More complicated assertions are coded as subpatterns. There are two
5539 kinds: those that look ahead of the current position in the subject
5540 string, and those that look behind it. An assertion subpattern is
5541 matched in the normal way, except that it does not cause the current
5542 matching position to be changed.
5543
5544 Assertion subpatterns are not capturing subpatterns. If such an asser-
5545 tion contains capturing subpatterns within it, these are counted for
5546 the purposes of numbering the capturing subpatterns in the whole pat-
5547 tern. However, substring capturing is carried out only for positive
5548 assertions, because it does not make sense for negative assertions.
5549
5550 For compatibility with Perl, assertion subpatterns may be repeated;
5551 though it makes no sense to assert the same thing several times, the
5552 side effect of capturing parentheses may occasionally be useful. In
5553 practice, there only three cases:
5554
5555 (1) If the quantifier is {0}, the assertion is never obeyed during
5556 matching. However, it may contain internal capturing parenthesized
5557 groups that are called from elsewhere via the subroutine mechanism.
5558
5559 (2) If quantifier is {0,n} where n is greater than zero, it is treated
5560 as if it were {0,1}. At run time, the rest of the pattern match is
5561 tried with and without the assertion, the order depending on the greed-
5562 iness of the quantifier.
5563
5564 (3) If the minimum repetition is greater than zero, the quantifier is
5565 ignored. The assertion is obeyed just once when encountered during
5566 matching.
5567
5568 Lookahead assertions
5569
5570 Lookahead assertions start with (?= for positive assertions and (?! for
5571 negative assertions. For example,
5572
5573 \w+(?=;)
5574
5575 matches a word followed by a semicolon, but does not include the semi-
5576 colon in the match, and
5577
5578 foo(?!bar)
5579
5580 matches any occurrence of "foo" that is not followed by "bar". Note
5581 that the apparently similar pattern
5582
5583 (?!foo)bar
5584
5585 does not find an occurrence of "bar" that is preceded by something
5586 other than "foo"; it finds any occurrence of "bar" whatsoever, because
5587 the assertion (?!foo) is always true when the next three characters are
5588 "bar". A lookbehind assertion is needed to achieve the other effect.
5589
5590 If you want to force a matching failure at some point in a pattern, the
5591 most convenient way to do it is with (?!) because an empty string
5592 always matches, so an assertion that requires there not to be an empty
5593 string must always fail. The backtracking control verb (*FAIL) or (*F)
5594 is a synonym for (?!).
5595
5596 Lookbehind assertions
5597
5598 Lookbehind assertions start with (?<= for positive assertions and (?<!
5599 for negative assertions. For example,
5600
5601 (?<!foo)bar
5602
5603 does find an occurrence of "bar" that is not preceded by "foo". The
5604 contents of a lookbehind assertion are restricted such that all the
5605 strings it matches must have a fixed length. However, if there are sev-
5606 eral top-level alternatives, they do not all have to have the same
5607 fixed length. Thus
5608
5609 (?<=bullock|donkey)
5610
5611 is permitted, but
5612
5613 (?<!dogs?|cats?)
5614
5615 causes an error at compile time. Branches that match different length
5616 strings are permitted only at the top level of a lookbehind assertion.
5617 This is an extension compared with Perl, which requires all branches to
5618 match the same length of string. An assertion such as
5619
5620 (?<=ab(c|de))
5621
5622 is not permitted, because its single top-level branch can match two
5623 different lengths, but it is acceptable to PCRE if rewritten to use two
5624 top-level branches:
5625
5626 (?<=abc|abde)
5627
5628 In some cases, the escape sequence \K (see above) can be used instead
5629 of a lookbehind assertion to get round the fixed-length restriction.
5630
5631 The implementation of lookbehind assertions is, for each alternative,
5632 to temporarily move the current position back by the fixed length and
5633 then try to match. If there are insufficient characters before the cur-
5634 rent position, the assertion fails.
5635
5636 In a UTF mode, PCRE does not allow the \C escape (which matches a sin-
5637 gle data unit even in a UTF mode) to appear in lookbehind assertions,
5638 because it makes it impossible to calculate the length of the lookbe-
5639 hind. The \X and \R escapes, which can match different numbers of data
5640 units, are also not permitted.
5641
5642 "Subroutine" calls (see below) such as (?2) or (?&X) are permitted in
5643 lookbehinds, as long as the subpattern matches a fixed-length string.
5644 Recursion, however, is not supported.
5645
5646 Possessive quantifiers can be used in conjunction with lookbehind
5647 assertions to specify efficient matching of fixed-length strings at the
5648 end of subject strings. Consider a simple pattern such as
5649
5650 abcd$
5651
5652 when applied to a long string that does not match. Because matching
5653 proceeds from left to right, PCRE will look for each "a" in the subject
5654 and then see if what follows matches the rest of the pattern. If the
5655 pattern is specified as
5656
5657 ^.*abcd$
5658
5659 the initial .* matches the entire string at first, but when this fails
5660 (because there is no following "a"), it backtracks to match all but the
5661 last character, then all but the last two characters, and so on. Once
5662 again the search for "a" covers the entire string, from right to left,
5663 so we are no better off. However, if the pattern is written as
5664
5665 ^.*+(?<=abcd)
5666
5667 there can be no backtracking for the .*+ item; it can match only the
5668 entire string. The subsequent lookbehind assertion does a single test
5669 on the last four characters. If it fails, the match fails immediately.
5670 For long strings, this approach makes a significant difference to the
5671 processing time.
5672
5673 Using multiple assertions
5674
5675 Several assertions (of any sort) may occur in succession. For example,
5676
5677 (?<=\d{3})(?<!999)foo
5678
5679 matches "foo" preceded by three digits that are not "999". Notice that
5680 each of the assertions is applied independently at the same point in
5681 the subject string. First there is a check that the previous three
5682 characters are all digits, and then there is a check that the same
5683 three characters are not "999". This pattern does not match "foo" pre-
5684 ceded by six characters, the first of which are digits and the last
5685 three of which are not "999". For example, it doesn't match "123abc-
5686 foo". A pattern to do that is
5687
5688 (?<=\d{3}...)(?<!999)foo
5689
5690 This time the first assertion looks at the preceding six characters,
5691 checking that the first three are digits, and then the second assertion
5692 checks that the preceding three characters are not "999".
5693
5694 Assertions can be nested in any combination. For example,
5695
5696 (?<=(?<!foo)bar)baz
5697
5698 matches an occurrence of "baz" that is preceded by "bar" which in turn
5699 is not preceded by "foo", while
5700
5701 (?<=\d{3}(?!999)...)foo
5702
5703 is another pattern that matches "foo" preceded by three digits and any
5704 three characters that are not "999".
5705
5706
5707 CONDITIONAL SUBPATTERNS
5708
5709 It is possible to cause the matching process to obey a subpattern con-
5710 ditionally or to choose between two alternative subpatterns, depending
5711 on the result of an assertion, or whether a specific capturing subpat-
5712 tern has already been matched. The two possible forms of conditional
5713 subpattern are:
5714
5715 (?(condition)yes-pattern)
5716 (?(condition)yes-pattern|no-pattern)
5717
5718 If the condition is satisfied, the yes-pattern is used; otherwise the
5719 no-pattern (if present) is used. If there are more than two alterna-
5720 tives in the subpattern, a compile-time error occurs. Each of the two
5721 alternatives may itself contain nested subpatterns of any form, includ-
5722 ing conditional subpatterns; the restriction to two alternatives
5723 applies only at the level of the condition. This pattern fragment is an
5724 example where the alternatives are complex:
5725
5726 (?(1) (A|B|C) | (D | (?(2)E|F) | E) )
5727
5728
5729 There are four kinds of condition: references to subpatterns, refer-
5730 ences to recursion, a pseudo-condition called DEFINE, and assertions.
5731
5732 Checking for a used subpattern by number
5733
5734 If the text between the parentheses consists of a sequence of digits,
5735 the condition is true if a capturing subpattern of that number has pre-
5736 viously matched. If there is more than one capturing subpattern with
5737 the same number (see the earlier section about duplicate subpattern
5738 numbers), the condition is true if any of them have matched. An alter-
5739 native notation is to precede the digits with a plus or minus sign. In
5740 this case, the subpattern number is relative rather than absolute. The
5741 most recently opened parentheses can be referenced by (?(-1), the next
5742 most recent by (?(-2), and so on. Inside loops it can also make sense
5743 to refer to subsequent groups. The next parentheses to be opened can be
5744 referenced as (?(+1), and so on. (The value zero in any of these forms
5745 is not used; it provokes a compile-time error.)
5746
5747 Consider the following pattern, which contains non-significant white
5748 space to make it more readable (assume the PCRE_EXTENDED option) and to
5749 divide it into three parts for ease of discussion:
5750
5751 ( \( )? [^()]+ (?(1) \) )
5752
5753 The first part matches an optional opening parenthesis, and if that
5754 character is present, sets it as the first captured substring. The sec-
5755 ond part matches one or more characters that are not parentheses. The
5756 third part is a conditional subpattern that tests whether or not the
5757 first set of parentheses matched. If they did, that is, if subject
5758 started with an opening parenthesis, the condition is true, and so the
5759 yes-pattern is executed and a closing parenthesis is required. Other-
5760 wise, since no-pattern is not present, the subpattern matches nothing.
5761 In other words, this pattern matches a sequence of non-parentheses,
5762 optionally enclosed in parentheses.
5763
5764 If you were embedding this pattern in a larger one, you could use a
5765 relative reference:
5766
5767 ...other stuff... ( \( )? [^()]+ (?(-1) \) ) ...
5768
5769 This makes the fragment independent of the parentheses in the larger
5770 pattern.
5771
5772 Checking for a used subpattern by name
5773
5774 Perl uses the syntax (?(<name>)...) or (?('name')...) to test for a
5775 used subpattern by name. For compatibility with earlier versions of
5776 PCRE, which had this facility before Perl, the syntax (?(name)...) is
5777 also recognized. However, there is a possible ambiguity with this syn-
5778 tax, because subpattern names may consist entirely of digits. PCRE
5779 looks first for a named subpattern; if it cannot find one and the name
5780 consists entirely of digits, PCRE looks for a subpattern of that num-
5781 ber, which must be greater than zero. Using subpattern names that con-
5782 sist entirely of digits is not recommended.
5783
5784 Rewriting the above example to use a named subpattern gives this:
5785
5786 (?<OPEN> \( )? [^()]+ (?(<OPEN>) \) )
5787
5788 If the name used in a condition of this kind is a duplicate, the test
5789 is applied to all subpatterns of the same name, and is true if any one
5790 of them has matched.
5791
5792 Checking for pattern recursion
5793
5794 If the condition is the string (R), and there is no subpattern with the
5795 name R, the condition is true if a recursive call to the whole pattern
5796 or any subpattern has been made. If digits or a name preceded by amper-
5797 sand follow the letter R, for example:
5798
5799 (?(R3)...) or (?(R&name)...)
5800
5801 the condition is true if the most recent recursion is into a subpattern
5802 whose number or name is given. This condition does not check the entire
5803 recursion stack. If the name used in a condition of this kind is a
5804 duplicate, the test is applied to all subpatterns of the same name, and
5805 is true if any one of them is the most recent recursion.
5806
5807 At "top level", all these recursion test conditions are false. The
5808 syntax for recursive patterns is described below.
5809
5810 Defining subpatterns for use by reference only
5811
5812 If the condition is the string (DEFINE), and there is no subpattern
5813 with the name DEFINE, the condition is always false. In this case,
5814 there may be only one alternative in the subpattern. It is always
5815 skipped if control reaches this point in the pattern; the idea of
5816 DEFINE is that it can be used to define subroutines that can be refer-
5817 enced from elsewhere. (The use of subroutines is described below.) For
5818 example, a pattern to match an IPv4 address such as "192.168.23.245"
5819 could be written like this (ignore white space and line breaks):
5820
5821 (?(DEFINE) (?<byte> 2[0-4]\d | 25[0-5] | 1\d\d | [1-9]?\d) )
5822 \b (?&byte) (\.(?&byte)){3} \b
5823
5824 The first part of the pattern is a DEFINE group inside which a another
5825 group named "byte" is defined. This matches an individual component of
5826 an IPv4 address (a number less than 256). When matching takes place,
5827 this part of the pattern is skipped because DEFINE acts like a false
5828 condition. The rest of the pattern uses references to the named group
5829 to match the four dot-separated components of an IPv4 address, insist-
5830 ing on a word boundary at each end.
5831
5832 Assertion conditions
5833
5834 If the condition is not in any of the above formats, it must be an
5835 assertion. This may be a positive or negative lookahead or lookbehind
5836 assertion. Consider this pattern, again containing non-significant
5837 white space, and with the two alternatives on the second line:
5838
5839 (?(?=[^a-z]*[a-z])
5840 \d{2}-[a-z]{3}-\d{2} | \d{2}-\d{2}-\d{2} )
5841
5842 The condition is a positive lookahead assertion that matches an
5843 optional sequence of non-letters followed by a letter. In other words,
5844 it tests for the presence of at least one letter in the subject. If a
5845 letter is found, the subject is matched against the first alternative;
5846 otherwise it is matched against the second. This pattern matches
5847 strings in one of the two forms dd-aaa-dd or dd-dd-dd, where aaa are
5848 letters and dd are digits.
5849
5850
5851 COMMENTS
5852
5853 There are two ways of including comments in patterns that are processed
5854 by PCRE. In both cases, the start of the comment must not be in a char-
5855 acter class, nor in the middle of any other sequence of related charac-
5856 ters such as (?: or a subpattern name or number. The characters that
5857 make up a comment play no part in the pattern matching.
5858
5859 The sequence (?# marks the start of a comment that continues up to the
5860 next closing parenthesis. Nested parentheses are not permitted. If the
5861 PCRE_EXTENDED option is set, an unescaped # character also introduces a
5862 comment, which in this case continues to immediately after the next
5863 newline character or character sequence in the pattern. Which charac-
5864 ters are interpreted as newlines is controlled by the options passed to
5865 a compiling function or by a special sequence at the start of the pat-
5866 tern, as described in the section entitled "Newline conventions" above.
5867 Note that the end of this type of comment is a literal newline sequence
5868 in the pattern; escape sequences that happen to represent a newline do
5869 not count. For example, consider this pattern when PCRE_EXTENDED is
5870 set, and the default newline convention is in force:
5871
5872 abc #comment \n still comment
5873
5874 On encountering the # character, pcre_compile() skips along, looking
5875 for a newline in the pattern. The sequence \n is still literal at this
5876 stage, so it does not terminate the comment. Only an actual character
5877 with the code value 0x0a (the default newline) does so.
5878
5879
5880 RECURSIVE PATTERNS
5881
5882 Consider the problem of matching a string in parentheses, allowing for
5883 unlimited nested parentheses. Without the use of recursion, the best
5884 that can be done is to use a pattern that matches up to some fixed
5885 depth of nesting. It is not possible to handle an arbitrary nesting
5886 depth.
5887
5888 For some time, Perl has provided a facility that allows regular expres-
5889 sions to recurse (amongst other things). It does this by interpolating
5890 Perl code in the expression at run time, and the code can refer to the
5891 expression itself. A Perl pattern using code interpolation to solve the
5892 parentheses problem can be created like this:
5893
5894 $re = qr{\( (?: (?>[^()]+) | (?p{$re}) )* \)}x;
5895
5896 The (?p{...}) item interpolates Perl code at run time, and in this case
5897 refers recursively to the pattern in which it appears.
5898
5899 Obviously, PCRE cannot support the interpolation of Perl code. Instead,
5900 it supports special syntax for recursion of the entire pattern, and
5901 also for individual subpattern recursion. After its introduction in
5902 PCRE and Python, this kind of recursion was subsequently introduced
5903 into Perl at release 5.10.
5904
5905 A special item that consists of (? followed by a number greater than
5906 zero and a closing parenthesis is a recursive subroutine call of the
5907 subpattern of the given number, provided that it occurs inside that
5908 subpattern. (If not, it is a non-recursive subroutine call, which is
5909 described in the next section.) The special item (?R) or (?0) is a
5910 recursive call of the entire regular expression.
5911
5912 This PCRE pattern solves the nested parentheses problem (assume the
5913 PCRE_EXTENDED option is set so that white space is ignored):
5914
5915 \( ( [^()]++ | (?R) )* \)
5916
5917 First it matches an opening parenthesis. Then it matches any number of
5918 substrings which can either be a sequence of non-parentheses, or a
5919 recursive match of the pattern itself (that is, a correctly parenthe-
5920 sized substring). Finally there is a closing parenthesis. Note the use
5921 of a possessive quantifier to avoid backtracking into sequences of non-
5922 parentheses.
5923
5924 If this were part of a larger pattern, you would not want to recurse
5925 the entire pattern, so instead you could use this:
5926
5927 ( \( ( [^()]++ | (?1) )* \) )
5928
5929 We have put the pattern into parentheses, and caused the recursion to
5930 refer to them instead of the whole pattern.
5931
5932 In a larger pattern, keeping track of parenthesis numbers can be
5933 tricky. This is made easier by the use of relative references. Instead
5934 of (?1) in the pattern above you can write (?-2) to refer to the second
5935 most recently opened parentheses preceding the recursion. In other
5936 words, a negative number counts capturing parentheses leftwards from
5937 the point at which it is encountered.
5938
5939 It is also possible to refer to subsequently opened parentheses, by
5940 writing references such as (?+2). However, these cannot be recursive
5941 because the reference is not inside the parentheses that are refer-
5942 enced. They are always non-recursive subroutine calls, as described in
5943 the next section.
5944
5945 An alternative approach is to use named parentheses instead. The Perl
5946 syntax for this is (?&name); PCRE's earlier syntax (?P>name) is also
5947 supported. We could rewrite the above example as follows:
5948
5949 (?<pn> \( ( [^()]++ | (?&pn) )* \) )
5950
5951 If there is more than one subpattern with the same name, the earliest
5952 one is used.
5953
5954 This particular example pattern that we have been looking at contains
5955 nested unlimited repeats, and so the use of a possessive quantifier for
5956 matching strings of non-parentheses is important when applying the pat-
5957 tern to strings that do not match. For example, when this pattern is
5958 applied to
5959
5960 (aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa()
5961
5962 it yields "no match" quickly. However, if a possessive quantifier is
5963 not used, the match runs for a very long time indeed because there are
5964 so many different ways the + and * repeats can carve up the subject,
5965 and all have to be tested before failure can be reported.
5966
5967 At the end of a match, the values of capturing parentheses are those
5968 from the outermost level. If you want to obtain intermediate values, a
5969 callout function can be used (see below and the pcrecallout documenta-
5970 tion). If the pattern above is matched against
5971
5972 (ab(cd)ef)
5973
5974 the value for the inner capturing parentheses (numbered 2) is "ef",
5975 which is the last value taken on at the top level. If a capturing sub-
5976 pattern is not matched at the top level, its final captured value is
5977 unset, even if it was (temporarily) set at a deeper level during the
5978 matching process.
5979
5980 If there are more than 15 capturing parentheses in a pattern, PCRE has
5981 to obtain extra memory to store data during a recursion, which it does
5982 by using pcre_malloc, freeing it via pcre_free afterwards. If no memory
5983 can be obtained, the match fails with the PCRE_ERROR_NOMEMORY error.
5984
5985 Do not confuse the (?R) item with the condition (R), which tests for
5986 recursion. Consider this pattern, which matches text in angle brack-
5987 ets, allowing for arbitrary nesting. Only digits are allowed in nested
5988 brackets (that is, when recursing), whereas any characters are permit-
5989 ted at the outer level.
5990
5991 < (?: (?(R) \d++ | [^<>]*+) | (?R)) * >
5992
5993 In this pattern, (?(R) is the start of a conditional subpattern, with
5994 two different alternatives for the recursive and non-recursive cases.
5995 The (?R) item is the actual recursive call.
5996
5997 Differences in recursion processing between PCRE and Perl
5998
5999 Recursion processing in PCRE differs from Perl in two important ways.
6000 In PCRE (like Python, but unlike Perl), a recursive subpattern call is
6001 always treated as an atomic group. That is, once it has matched some of
6002 the subject string, it is never re-entered, even if it contains untried
6003 alternatives and there is a subsequent matching failure. This can be
6004 illustrated by the following pattern, which purports to match a palin-
6005 dromic string that contains an odd number of characters (for example,
6006 "a", "aba", "abcba", "abcdcba"):
6007
6008 ^(.|(.)(?1)\2)$
6009
6010 The idea is that it either matches a single character, or two identical
6011 characters surrounding a sub-palindrome. In Perl, this pattern works;
6012 in PCRE it does not if the pattern is longer than three characters.
6013 Consider the subject string "abcba":
6014
6015 At the top level, the first character is matched, but as it is not at