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

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

Parent Directory Parent Directory | Revision Log Revision Log


Revision 148 - (hide annotations) (download)
Mon Apr 16 13:25:10 2007 UTC (7 years, 4 months ago) by ph10
File MIME type: text/plain
File size: 266301 byte(s)
Fix bugs in the documentation.

1 nigel 75 -----------------------------------------------------------------------------
2 nigel 63 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. There are
6     separate text files for the pcregrep and pcretest commands.
7     -----------------------------------------------------------------------------
8    
9 nigel 41
10 nigel 79 PCRE(3) PCRE(3)
11 nigel 41
12 nigel 79
13 nigel 73 NAME
14     PCRE - Perl-compatible regular expressions
15    
16 nigel 77
17 nigel 75 INTRODUCTION
18 nigel 41
19 nigel 73 The PCRE library is a set of functions that implement regular expres-
20     sion pattern matching using the same syntax and semantics as Perl, with
21 nigel 93 just a few differences. (Certain features that appeared in Python and
22     PCRE before they appeared in Perl are also available using the Python
23     syntax.)
24 nigel 63
25 nigel 93 The current implementation of PCRE (release 7.x) corresponds approxi-
26     mately with Perl 5.10, including support for UTF-8 encoded strings and
27     Unicode general category properties. However, UTF-8 and Unicode support
28     has to be explicitly enabled; it is not the default. The Unicode tables
29     correspond to Unicode release 5.0.0.
30 nigel 77
31 nigel 93 In addition to the Perl-compatible matching function, PCRE contains an
32     alternative matching function that matches the same compiled patterns
33     in a different way. In certain circumstances, the alternative function
34     has some advantages. For a discussion of the two matching algorithms,
35     see the pcrematching page.
36    
37     PCRE is written in C and released as a C library. A number of people
38     have written wrappers and interfaces of various kinds. In particular,
39     Google Inc. have provided a comprehensive C++ wrapper. This is now
40 nigel 77 included as part of the PCRE distribution. The pcrecpp page has details
41 nigel 93 of this interface. Other people's contributions can be found in the
42 nigel 77 Contrib directory at the primary FTP site, which is:
43 nigel 63
44 nigel 73 ftp://ftp.csx.cam.ac.uk/pub/software/programming/pcre
45 nigel 63
46 nigel 93 Details of exactly which Perl regular expression features are and are
47 nigel 73 not supported by PCRE are given in separate documents. See the pcrepat-
48     tern and pcrecompat pages.
49 nigel 63
50 nigel 93 Some features of PCRE can be included, excluded, or changed when the
51     library is built. The pcre_config() function makes it possible for a
52     client to discover which features are available. The features them-
53     selves are described in the pcrebuild page. Documentation about build-
54     ing PCRE for various operating systems can be found in the README file
55 nigel 75 in the source distribution.
56 nigel 63
57 nigel 93 The library contains a number of undocumented internal functions and
58     data tables that are used by more than one of the exported external
59     functions, but which are not intended for use by external callers.
60     Their names all begin with "_pcre_", which hopefully will not provoke
61 nigel 83 any name clashes. In some environments, it is possible to control which
62 nigel 93 external symbols are exported when a shared library is built, and in
63 nigel 83 these cases the undocumented symbols are not exported.
64 nigel 63
65 nigel 77
66 nigel 63 USER DOCUMENTATION
67    
68 nigel 93 The user documentation for PCRE comprises a number of different sec-
69     tions. In the "man" format, each of these is a separate "man page". In
70     the HTML format, each is a separate page, linked from the index page.
71     In the plain text format, all the sections are concatenated, for ease
72 nigel 75 of searching. The sections are as follows:
73 nigel 63
74 nigel 73 pcre this document
75 nigel 77 pcreapi details of PCRE's native C API
76 nigel 73 pcrebuild options for building PCRE
77     pcrecallout details of the callout feature
78     pcrecompat discussion of Perl compatibility
79 nigel 77 pcrecpp details of the C++ wrapper
80 nigel 73 pcregrep description of the pcregrep command
81 nigel 77 pcrematching discussion of the two matching algorithms
82 nigel 75 pcrepartial details of the partial matching facility
83 nigel 73 pcrepattern syntax and semantics of supported
84     regular expressions
85     pcreperform discussion of performance issues
86 nigel 77 pcreposix the POSIX-compatible C API
87 nigel 75 pcreprecompile details of saving and re-using precompiled patterns
88 nigel 73 pcresample discussion of the sample program
89 nigel 91 pcrestack discussion of stack usage
90 nigel 75 pcretest description of the pcretest testing command
91 nigel 63
92 nigel 93 In addition, in the "man" and HTML formats, there is a short page for
93 nigel 77 each C library function, listing its arguments and results.
94 nigel 63
95    
96     LIMITATIONS
97    
98 nigel 93 There are some size limitations in PCRE but it is hoped that they will
99 nigel 73 never in practice be relevant.
100 nigel 63
101 nigel 93 The maximum length of a compiled pattern is 65539 (sic) bytes if PCRE
102 nigel 73 is compiled with the default internal linkage size of 2. If you want to
103 nigel 93 process regular expressions that are truly enormous, you can compile
104     PCRE with an internal linkage size of 3 or 4 (see the README file in
105     the source distribution and the pcrebuild documentation for details).
106     In these cases the limit is substantially larger. However, the speed
107     of execution is slower.
108 nigel 63
109 nigel 93 All values in repeating quantifiers must be less than 65536. The maxi-
110     mum compiled length of subpattern with an explicit repeat count is
111 nigel 91 30000 bytes. The maximum number of capturing subpatterns is 65535.
112 nigel 63
113 nigel 93 There is no limit to the number of parenthesized subpatterns, but there
114     can be no more than 65535 capturing subpatterns.
115 nigel 63
116 nigel 93 The maximum length of name for a named subpattern is 32 characters, and
117     the maximum number of named subpatterns is 10000.
118 nigel 91
119 nigel 93 The maximum length of a subject string is the largest positive number
120     that an integer variable can hold. However, when using the traditional
121 nigel 77 matching function, PCRE uses recursion to handle subpatterns and indef-
122 nigel 93 inite repetition. This means that the available stack space may limit
123 nigel 77 the size of a subject string that can be processed by certain patterns.
124 nigel 91 For a discussion of stack issues, see the pcrestack documentation.
125 nigel 63
126    
127 nigel 75 UTF-8 AND UNICODE PROPERTY SUPPORT
128 nigel 63
129 nigel 93 From release 3.3, PCRE has had some support for character strings
130     encoded in the UTF-8 format. For release 4.0 this was greatly extended
131     to cover most common requirements, and in release 5.0 additional sup-
132 nigel 75 port for Unicode general category properties was added.
133 nigel 63
134 nigel 93 In order process UTF-8 strings, you must build PCRE to include UTF-8
135     support in the code, and, in addition, you must call pcre_compile()
136     with the PCRE_UTF8 option flag. When you do this, both the pattern and
137     any subject strings that are matched against it are treated as UTF-8
138 nigel 73 strings instead of just strings of bytes.
139 nigel 63
140 nigel 93 If you compile PCRE with UTF-8 support, but do not use it at run time,
141     the library will be a bit bigger, but the additional run time overhead
142     is limited to testing the PCRE_UTF8 flag occasionally, so should not be
143     very big.
144 nigel 63
145 nigel 75 If PCRE is built with Unicode character property support (which implies
146 nigel 93 UTF-8 support), the escape sequences \p{..}, \P{..}, and \X are sup-
147 nigel 75 ported. The available properties that can be tested are limited to the
148 nigel 93 general category properties such as Lu for an upper case letter or Nd
149     for a decimal number, the Unicode script names such as Arabic or Han,
150     and the derived properties Any and L&. A full list is given in the
151 nigel 87 pcrepattern documentation. Only the short names for properties are sup-
152 nigel 93 ported. For example, \p{L} matches a letter. Its Perl synonym, \p{Let-
153     ter}, is not supported. Furthermore, in Perl, many properties may
154     optionally be prefixed by "Is", for compatibility with Perl 5.6. PCRE
155 nigel 87 does not support this.
156 nigel 75
157 nigel 73 The following comments apply when PCRE is running in UTF-8 mode:
158 nigel 63
159 nigel 93 1. When you set the PCRE_UTF8 flag, the strings passed as patterns and
160     subjects are checked for validity on entry to the relevant functions.
161 nigel 73 If an invalid UTF-8 string is passed, an error return is given. In some
162 nigel 93 situations, you may already know that your strings are valid, and
163 nigel 73 therefore want to skip these checks in order to improve performance. If
164 nigel 93 you set the PCRE_NO_UTF8_CHECK flag at compile time or at run time,
165     PCRE assumes that the pattern or subject it is given (respectively)
166     contains only valid UTF-8 codes. In this case, it does not diagnose an
167     invalid UTF-8 string. If you pass an invalid UTF-8 string to PCRE when
168     PCRE_NO_UTF8_CHECK is set, the results are undefined. Your program may
169 nigel 73 crash.
170 nigel 63
171 nigel 93 2. An unbraced hexadecimal escape sequence (such as \xb3) matches a
172 nigel 87 two-byte UTF-8 character if the value is greater than 127.
173 nigel 63
174 nigel 93 3. Octal numbers up to \777 are recognized, and match two-byte UTF-8
175 nigel 91 characters for values greater than \177.
176    
177 nigel 93 4. Repeat quantifiers apply to complete UTF-8 characters, not to indi-
178 nigel 73 vidual bytes, for example: \x{100}{3}.
179 nigel 63
180 nigel 93 5. The dot metacharacter matches one UTF-8 character instead of a sin-
181 nigel 75 gle byte.
182 nigel 63
183 nigel 93 6. The escape sequence \C can be used to match a single byte in UTF-8
184     mode, but its use can lead to some strange effects. This facility is
185 nigel 77 not available in the alternative matching function, pcre_dfa_exec().
186 nigel 63
187 nigel 93 7. The character escapes \b, \B, \d, \D, \s, \S, \w, and \W correctly
188     test characters of any code value, but the characters that PCRE recog-
189     nizes as digits, spaces, or word characters remain the same set as
190 nigel 75 before, all with values less than 256. This remains true even when PCRE
191 nigel 93 includes Unicode property support, because to do otherwise would slow
192     down PCRE in many common cases. If you really want to test for a wider
193     sense of, say, "digit", you must use Unicode property tests such as
194 nigel 75 \p{Nd}.
195 nigel 63
196 nigel 93 8. Similarly, characters that match the POSIX named character classes
197 nigel 75 are all low-valued characters.
198 nigel 63
199 nigel 93 9. Case-insensitive matching applies only to characters whose values
200     are less than 128, unless PCRE is built with Unicode property support.
201     Even when Unicode property support is available, PCRE still uses its
202     own character tables when checking the case of low-valued characters,
203     so as not to degrade performance. The Unicode property information is
204 nigel 87 used only for characters with higher values. Even when Unicode property
205     support is available, PCRE supports case-insensitive matching only when
206 nigel 93 there is a one-to-one mapping between a letter's cases. There are a
207     small number of many-to-one mappings in Unicode; these are not sup-
208 nigel 87 ported by PCRE.
209 nigel 63
210    
211     AUTHOR
212    
213 nigel 77 Philip Hazel
214 ph10 99 University Computing Service
215 nigel 93 Cambridge CB2 3QH, England.
216 nigel 63
217 nigel 93 Putting an actual email address here seems to have been a spam magnet,
218 nigel 77 so I've taken it away. If you want to email me, use my initial and sur-
219     name, separated by a dot, at the domain ucs.cam.ac.uk.
220    
221 nigel 63
222 ph10 99 REVISION
223 nigel 63
224 ph10 99 Last updated: 06 March 2007
225     Copyright (c) 1997-2007 University of Cambridge.
226     ------------------------------------------------------------------------------
227 ph10 111
228    
229 nigel 79 PCREBUILD(3) PCREBUILD(3)
230 nigel 63
231 nigel 79
232 nigel 73 NAME
233     PCRE - Perl-compatible regular expressions
234    
235 nigel 77
236 nigel 63 PCRE BUILD-TIME OPTIONS
237    
238 nigel 73 This document describes the optional features of PCRE that can be
239     selected when the library is compiled. They are all selected, or dese-
240 nigel 75 lected, by providing options to the configure script that is run before
241     the make command. The complete list of options for configure (which
242     includes the standard ones such as the selection of the installation
243     directory) can be obtained by running
244 nigel 63
245 nigel 73 ./configure --help
246 nigel 63
247 ph10 128 The following sections include descriptions of options whose names
248     begin with --enable or --disable. These settings specify changes to the
249     defaults for the configure command. Because of the way that configure
250     works, --enable and --disable always come in pairs, so the complemen-
251     tary option always exists as well, but as it specifies the default, it
252     is not described.
253 nigel 63
254    
255 nigel 83 C++ SUPPORT
256    
257     By default, the configure script will search for a C++ compiler and C++
258     header files. If it finds them, it automatically builds the C++ wrapper
259     library for PCRE. You can disable this by adding
260    
261     --disable-cpp
262    
263     to the configure command.
264    
265    
266 nigel 63 UTF-8 SUPPORT
267    
268 nigel 73 To build PCRE with support for UTF-8 character strings, add
269 nigel 63
270 nigel 73 --enable-utf8
271 nigel 63
272 nigel 73 to the configure command. Of itself, this does not make PCRE treat
273     strings as UTF-8. As well as compiling PCRE with this option, you also
274     have have to set the PCRE_UTF8 option when you call the pcre_compile()
275     function.
276 nigel 63
277    
278 nigel 75 UNICODE CHARACTER PROPERTY SUPPORT
279    
280     UTF-8 support allows PCRE to process character values greater than 255
281     in the strings that it handles. On its own, however, it does not pro-
282     vide any facilities for accessing the properties of such characters. If
283     you want to be able to use the pattern escapes \P, \p, and \X, which
284     refer to Unicode character properties, you must add
285    
286     --enable-unicode-properties
287    
288     to the configure command. This implies UTF-8 support, even if you have
289     not explicitly requested it.
290    
291 ph10 128 Including Unicode property support adds around 30K of tables to the
292     PCRE library. Only the general category properties such as Lu and Nd
293     are supported. Details are given in the pcrepattern documentation.
294 nigel 75
295    
296 nigel 63 CODE VALUE OF NEWLINE
297    
298 ph10 128 By default, PCRE interprets character 10 (linefeed, LF) as indicating
299     the end of a line. This is the normal newline character on Unix-like
300 nigel 91 systems. You can compile PCRE to use character 13 (carriage return, CR)
301     instead, by adding
302 nigel 63
303 nigel 73 --enable-newline-is-cr
304 nigel 63
305 ph10 128 to the configure command. There is also a --enable-newline-is-lf
306 nigel 91 option, which explicitly specifies linefeed as the newline character.
307 nigel 63
308 nigel 91 Alternatively, you can specify that line endings are to be indicated by
309     the two character sequence CRLF. If you want this, add
310 nigel 63
311 nigel 91 --enable-newline-is-crlf
312    
313 nigel 93 to the configure command. There is a fourth option, specified by
314 nigel 91
315 nigel 93 --enable-newline-is-any
316 nigel 91
317 nigel 93 which causes PCRE to recognize any Unicode newline sequence.
318    
319 ph10 128 Whatever line ending convention is selected when PCRE is built can be
320     overridden when the library functions are called. At build time it is
321 nigel 93 conventional to use the standard for your operating system.
322    
323    
324 nigel 63 BUILDING SHARED AND STATIC LIBRARIES
325    
326 ph10 128 The PCRE building process uses libtool to build both shared and static
327     Unix libraries by default. You can suppress one of these by adding one
328 nigel 73 of
329 nigel 63
330 nigel 73 --disable-shared
331     --disable-static
332 nigel 63
333 nigel 73 to the configure command, as required.
334 nigel 63
335    
336     POSIX MALLOC USAGE
337    
338 nigel 75 When PCRE is called through the POSIX interface (see the pcreposix doc-
339 ph10 128 umentation), additional working storage is required for holding the
340     pointers to capturing substrings, because PCRE requires three integers
341     per substring, whereas the POSIX interface provides only two. If the
342 nigel 73 number of expected substrings is small, the wrapper function uses space
343     on the stack, because this is faster than using malloc() for each call.
344     The default threshold above which the stack is no longer used is 10; it
345     can be changed by adding a setting such as
346 nigel 63
347 nigel 73 --with-posix-malloc-threshold=20
348 nigel 63
349 nigel 73 to the configure command.
350 nigel 63
351    
352     HANDLING VERY LARGE PATTERNS
353    
354 ph10 128 Within a compiled pattern, offset values are used to point from one
355     part to another (for example, from an opening parenthesis to an alter-
356     nation metacharacter). By default, two-byte values are used for these
357     offsets, leading to a maximum size for a compiled pattern of around
358     64K. This is sufficient to handle all but the most gigantic patterns.
359     Nevertheless, some people do want to process enormous patterns, so it
360     is possible to compile PCRE to use three-byte or four-byte offsets by
361 nigel 73 adding a setting such as
362 nigel 63
363 nigel 73 --with-link-size=3
364 nigel 63
365 ph10 128 to the configure command. The value given must be 2, 3, or 4. Using
366     longer offsets slows down the operation of PCRE because it has to load
367 nigel 73 additional bytes when handling them.
368 nigel 63
369    
370 nigel 73 AVOIDING EXCESSIVE STACK USAGE
371    
372 nigel 77 When matching with the pcre_exec() function, PCRE implements backtrack-
373 nigel 93 ing by making recursive calls to an internal function called match().
374     In environments where the size of the stack is limited, this can se-
375     verely limit PCRE's operation. (The Unix environment does not usually
376 nigel 91 suffer from this problem, but it may sometimes be necessary to increase
377 nigel 93 the maximum stack size. There is a discussion in the pcrestack docu-
378     mentation.) An alternative approach to recursion that uses memory from
379     the heap to remember data, instead of using recursive function calls,
380     has been implemented to work round the problem of limited stack size.
381 nigel 91 If you want to build a version of PCRE that works this way, add
382 nigel 73
383     --disable-stack-for-recursion
384    
385 nigel 93 to the configure command. With this configuration, PCRE will use the
386     pcre_stack_malloc and pcre_stack_free variables to call memory manage-
387     ment functions. Separate functions are provided because the usage is
388     very predictable: the block sizes requested are always the same, and
389     the blocks are always freed in reverse order. A calling program might
390     be able to implement optimized functions that perform better than the
391     standard malloc() and free() functions. PCRE runs noticeably more
392 nigel 77 slowly when built in this way. This option affects only the pcre_exec()
393     function; it is not relevant for the the pcre_dfa_exec() function.
394 nigel 73
395    
396 nigel 91 LIMITING PCRE RESOURCE USAGE
397    
398 nigel 93 Internally, PCRE has a function called match(), which it calls repeat-
399     edly (sometimes recursively) when matching a pattern with the
400     pcre_exec() function. By controlling the maximum number of times this
401     function may be called during a single matching operation, a limit can
402     be placed on the resources used by a single call to pcre_exec(). The
403     limit can be changed at run time, as described in the pcreapi documen-
404     tation. The default is 10 million, but this can be changed by adding a
405 nigel 91 setting such as
406    
407     --with-match-limit=500000
408    
409 nigel 93 to the configure command. This setting has no effect on the
410 nigel 91 pcre_dfa_exec() matching function.
411    
412 nigel 93 In some environments it is desirable to limit the depth of recursive
413 nigel 91 calls of match() more strictly than the total number of calls, in order
414 nigel 93 to restrict the maximum amount of stack (or heap, if --disable-stack-
415 nigel 91 for-recursion is specified) that is used. A second limit controls this;
416 nigel 93 it defaults to the value that is set for --with-match-limit, which
417     imposes no additional constraints. However, you can set a lower limit
418 nigel 91 by adding, for example,
419    
420     --with-match-limit-recursion=10000
421    
422 nigel 93 to the configure command. This value can also be overridden at run
423 nigel 91 time.
424    
425    
426 ph10 128 CREATING CHARACTER TABLES AT BUILD TIME
427    
428     PCRE uses fixed tables for processing characters whose code values are
429     less than 256. By default, PCRE is built with a set of tables that are
430     distributed in the file pcre_chartables.c.dist. These tables are for
431     ASCII codes only. If you add
432    
433     --enable-rebuild-chartables
434    
435     to the configure command, the distributed tables are no longer used.
436     Instead, a program called dftables is compiled and run. This outputs
437     the source for new set of tables, created in the default locale of your
438     C runtime system. (This method of replacing the tables does not work if
439     you are cross compiling, because dftables is run on the local host. If
440     you need to create alternative tables when cross compiling, you will
441     have to do so "by hand".)
442    
443    
444 nigel 73 USING EBCDIC CODE
445    
446 ph10 128 PCRE assumes by default that it will run in an environment where the
447     character code is ASCII (or Unicode, which is a superset of ASCII).
448     PCRE can, however, be compiled to run in an EBCDIC environment by
449 nigel 75 adding
450 nigel 73
451     --enable-ebcdic
452    
453 ph10 128 to the configure command. This setting implies --enable-rebuild-charta-
454     bles.
455 nigel 73
456 nigel 93
457     SEE ALSO
458    
459     pcreapi(3), pcre_config(3).
460    
461 nigel 63
462 ph10 99 AUTHOR
463 nigel 63
464 ph10 99 Philip Hazel
465     University Computing Service
466     Cambridge CB2 3QH, England.
467    
468    
469     REVISION
470    
471 ph10 128 Last updated: 20 March 2007
472 ph10 99 Copyright (c) 1997-2007 University of Cambridge.
473     ------------------------------------------------------------------------------
474 ph10 111
475    
476 nigel 79 PCREMATCHING(3) PCREMATCHING(3)
477 nigel 63
478 nigel 79
479 nigel 77 NAME
480     PCRE - Perl-compatible regular expressions
481 nigel 73
482 nigel 77
483     PCRE MATCHING ALGORITHMS
484    
485     This document describes the two different algorithms that are available
486     in PCRE for matching a compiled regular expression against a given sub-
487     ject string. The "standard" algorithm is the one provided by the
488     pcre_exec() function. This works in the same was as Perl's matching
489     function, and provides a Perl-compatible matching operation.
490    
491     An alternative algorithm is provided by the pcre_dfa_exec() function;
492     this operates in a different way, and is not Perl-compatible. It has
493     advantages and disadvantages compared with the standard algorithm, and
494     these are described below.
495    
496     When there is only one possible way in which a given subject string can
497     match a pattern, the two algorithms give the same answer. A difference
498     arises, however, when there are multiple possibilities. For example, if
499     the pattern
500    
501     ^<.*>
502    
503     is matched against the string
504    
505     <something> <something else> <something further>
506    
507     there are three possible answers. The standard algorithm finds only one
508 nigel 93 of them, whereas the alternative algorithm finds all three.
509 nigel 77
510    
511     REGULAR EXPRESSIONS AS TREES
512    
513     The set of strings that are matched by a regular expression can be rep-
514     resented as a tree structure. An unlimited repetition in the pattern
515     makes the tree of infinite size, but it is still a tree. Matching the
516     pattern to a given subject string (from a given starting point) can be
517 nigel 91 thought of as a search of the tree. There are two ways to search a
518     tree: depth-first and breadth-first, and these correspond to the two
519     matching algorithms provided by PCRE.
520 nigel 77
521    
522     THE STANDARD MATCHING ALGORITHM
523    
524 ph10 148 In the terminology of Jeffrey Friedl's book "Mastering Regular Expres-
525     sions", the standard algorithm is an "NFA algorithm". It conducts a
526 nigel 77 depth-first search of the pattern tree. That is, it proceeds along a
527     single path through the tree, checking that the subject matches what is
528     required. When there is a mismatch, the algorithm tries any alterna-
529     tives at the current point, and if they all fail, it backs up to the
530     previous branch point in the tree, and tries the next alternative
531     branch at that level. This often involves backing up (moving to the
532     left) in the subject string as well. The order in which repetition
533     branches are tried is controlled by the greedy or ungreedy nature of
534     the quantifier.
535    
536     If a leaf node is reached, a matching string has been found, and at
537     that point the algorithm stops. Thus, if there is more than one possi-
538     ble match, this algorithm returns the first one that it finds. Whether
539     this is the shortest, the longest, or some intermediate length depends
540     on the way the greedy and ungreedy repetition quantifiers are specified
541     in the pattern.
542    
543     Because it ends up with a single path through the tree, it is rela-
544     tively straightforward for this algorithm to keep track of the sub-
545     strings that are matched by portions of the pattern in parentheses.
546     This provides support for capturing parentheses and back references.
547    
548    
549 nigel 93 THE ALTERNATIVE MATCHING ALGORITHM
550 nigel 77
551 nigel 93 This algorithm conducts a breadth-first search of the tree. Starting
552     from the first matching point in the subject, it scans the subject
553     string from left to right, once, character by character, and as it does
554     this, it remembers all the paths through the tree that represent valid
555     matches. In Friedl's terminology, this is a kind of "DFA algorithm",
556     though it is not implemented as a traditional finite state machine (it
557     keeps multiple states active simultaneously).
558 nigel 77
559 nigel 93 The scan continues until either the end of the subject is reached, or
560     there are no more unterminated paths. At this point, terminated paths
561     represent the different matching possibilities (if there are none, the
562     match has failed). Thus, if there is more than one possible match,
563 nigel 77 this algorithm finds all of them, and in particular, it finds the long-
564 nigel 93 est. In PCRE, there is an option to stop the algorithm after the first
565 nigel 77 match (which is necessarily the shortest) has been found.
566    
567     Note that all the matches that are found start at the same point in the
568     subject. If the pattern
569    
570     cat(er(pillar)?)
571    
572 nigel 93 is matched against the string "the caterpillar catchment", the result
573     will be the three strings "cat", "cater", and "caterpillar" that start
574 nigel 77 at the fourth character of the subject. The algorithm does not automat-
575     ically move on to find matches that start at later positions.
576    
577     There are a number of features of PCRE regular expressions that are not
578 nigel 93 supported by the alternative matching algorithm. They are as follows:
579 nigel 77
580 nigel 93 1. Because the algorithm finds all possible matches, the greedy or
581     ungreedy nature of repetition quantifiers is not relevant. Greedy and
582     ungreedy quantifiers are treated in exactly the same way. However, pos-
583     sessive quantifiers can make a difference when what follows could also
584     match what is quantified, for example in a pattern like this:
585 nigel 77
586 nigel 93 ^a++\w!
587    
588     This pattern matches "aaab!" but not "aaa!", which would be matched by
589     a non-possessive quantifier. Similarly, if an atomic group is present,
590     it is matched as if it were a standalone pattern at the current point,
591     and the longest match is then "locked in" for the rest of the overall
592     pattern.
593    
594 nigel 77 2. When dealing with multiple paths through the tree simultaneously, it
595 nigel 93 is not straightforward to keep track of captured substrings for the
596     different matching possibilities, and PCRE's implementation of this
597 nigel 77 algorithm does not attempt to do this. This means that no captured sub-
598     strings are available.
599    
600 nigel 93 3. Because no substrings are captured, back references within the pat-
601 nigel 77 tern are not supported, and cause errors if encountered.
602    
603 nigel 93 4. For the same reason, conditional expressions that use a backrefer-
604     ence as the condition or test for a specific group recursion are not
605     supported.
606 nigel 77
607     5. Callouts are supported, but the value of the capture_top field is
608     always 1, and the value of the capture_last field is always -1.
609    
610     6. The \C escape sequence, which (in the standard algorithm) matches a
611 nigel 93 single byte, even in UTF-8 mode, is not supported because the alterna-
612     tive algorithm moves through the subject string one character at a
613     time, for all active paths through the tree.
614 nigel 77
615    
616 nigel 93 ADVANTAGES OF THE ALTERNATIVE ALGORITHM
617 nigel 77
618 nigel 93 Using the alternative matching algorithm provides the following advan-
619     tages:
620 nigel 77
621     1. All possible matches (at a single point in the subject) are automat-
622 nigel 93 ically found, and in particular, the longest match is found. To find
623 nigel 77 more than one match using the standard algorithm, you have to do kludgy
624     things with callouts.
625    
626 nigel 93 2. There is much better support for partial matching. The restrictions
627     on the content of the pattern that apply when using the standard algo-
628     rithm for partial matching do not apply to the alternative algorithm.
629     For non-anchored patterns, the starting position of a partial match is
630     available.
631 nigel 77
632 nigel 93 3. Because the alternative algorithm scans the subject string just
633     once, and never needs to backtrack, it is possible to pass very long
634     subject strings to the matching function in several pieces, checking
635     for partial matching each time.
636 nigel 77
637    
638 nigel 93 DISADVANTAGES OF THE ALTERNATIVE ALGORITHM
639 nigel 77
640 nigel 93 The alternative algorithm suffers from a number of disadvantages:
641 nigel 77
642 nigel 93 1. It is substantially slower than the standard algorithm. This is
643     partly because it has to search for all possible matches, but is also
644 nigel 77 because it is less susceptible to optimization.
645    
646     2. Capturing parentheses and back references are not supported.
647    
648 nigel 93 3. Although atomic groups are supported, their use does not provide the
649     performance advantage that it does for the standard algorithm.
650 nigel 77
651    
652 ph10 99 AUTHOR
653 nigel 77
654 ph10 99 Philip Hazel
655     University Computing Service
656     Cambridge CB2 3QH, England.
657    
658    
659     REVISION
660    
661     Last updated: 06 March 2007
662     Copyright (c) 1997-2007 University of Cambridge.
663     ------------------------------------------------------------------------------
664 ph10 111
665    
666 nigel 79 PCREAPI(3) PCREAPI(3)
667 nigel 77
668 nigel 79
669 nigel 73 NAME
670     PCRE - Perl-compatible regular expressions
671    
672 nigel 77
673 nigel 75 PCRE NATIVE API
674 nigel 63
675 nigel 73 #include <pcre.h>
676 nigel 41
677 nigel 73 pcre *pcre_compile(const char *pattern, int options,
678     const char **errptr, int *erroffset,
679     const unsigned char *tableptr);
680 nigel 41
681 nigel 77 pcre *pcre_compile2(const char *pattern, int options,
682     int *errorcodeptr,
683     const char **errptr, int *erroffset,
684     const unsigned char *tableptr);
685    
686 nigel 73 pcre_extra *pcre_study(const pcre *code, int options,
687     const char **errptr);
688 nigel 41
689 nigel 73 int pcre_exec(const pcre *code, const pcre_extra *extra,
690     const char *subject, int length, int startoffset,
691     int options, int *ovector, int ovecsize);
692 nigel 41
693 nigel 77 int pcre_dfa_exec(const pcre *code, const pcre_extra *extra,
694     const char *subject, int length, int startoffset,
695     int options, int *ovector, int ovecsize,
696     int *workspace, int wscount);
697    
698 nigel 73 int pcre_copy_named_substring(const pcre *code,
699     const char *subject, int *ovector,
700     int stringcount, const char *stringname,
701     char *buffer, int buffersize);
702 nigel 63
703 nigel 73 int pcre_copy_substring(const char *subject, int *ovector,
704     int stringcount, int stringnumber, char *buffer,
705     int buffersize);
706 nigel 41
707 nigel 73 int pcre_get_named_substring(const pcre *code,
708     const char *subject, int *ovector,
709     int stringcount, const char *stringname,
710     const char **stringptr);
711 nigel 63
712 nigel 73 int pcre_get_stringnumber(const pcre *code,
713     const char *name);
714 nigel 63
715 nigel 91 int pcre_get_stringtable_entries(const pcre *code,
716     const char *name, char **first, char **last);
717    
718 nigel 73 int pcre_get_substring(const char *subject, int *ovector,
719     int stringcount, int stringnumber,
720     const char **stringptr);
721 nigel 41
722 nigel 73 int pcre_get_substring_list(const char *subject,
723     int *ovector, int stringcount, const char ***listptr);
724 nigel 41
725 nigel 73 void pcre_free_substring(const char *stringptr);
726 nigel 49
727 nigel 73 void pcre_free_substring_list(const char **stringptr);
728 nigel 49
729 nigel 73 const unsigned char *pcre_maketables(void);
730 nigel 41
731 nigel 73 int pcre_fullinfo(const pcre *code, const pcre_extra *extra,
732     int what, void *where);
733 nigel 43
734 nigel 73 int pcre_info(const pcre *code, int *optptr, int *firstcharptr);
735 nigel 63
736 nigel 77 int pcre_refcount(pcre *code, int adjust);
737    
738 nigel 73 int pcre_config(int what, void *where);
739 nigel 41
740 nigel 73 char *pcre_version(void);
741 nigel 63
742 nigel 73 void *(*pcre_malloc)(size_t);
743 nigel 41
744 nigel 73 void (*pcre_free)(void *);
745 nigel 41
746 nigel 73 void *(*pcre_stack_malloc)(size_t);
747 nigel 41
748 nigel 73 void (*pcre_stack_free)(void *);
749 nigel 41
750 nigel 73 int (*pcre_callout)(pcre_callout_block *);
751 nigel 41
752 nigel 73
753 nigel 75 PCRE API OVERVIEW
754 nigel 41
755 nigel 73 PCRE has its own native API, which is described in this document. There
756 nigel 93 are also some wrapper functions that correspond to the POSIX regular
757 nigel 77 expression API. These are described in the pcreposix documentation.
758     Both of these APIs define a set of C function calls. A C++ wrapper is
759     distributed with PCRE. It is documented in the pcrecpp page.
760 nigel 43
761 nigel 77 The native API C function prototypes are defined in the header file
762     pcre.h, and on Unix systems the library itself is called libpcre. It
763 nigel 75 can normally be accessed by adding -lpcre to the command for linking an
764     application that uses PCRE. The header file defines the macros
765     PCRE_MAJOR and PCRE_MINOR to contain the major and minor release num-
766     bers for the library. Applications can use these to include support
767     for different releases of PCRE.
768 nigel 41
769 nigel 77 The functions pcre_compile(), pcre_compile2(), pcre_study(), and
770     pcre_exec() are used for compiling and matching regular expressions in
771     a Perl-compatible manner. A sample program that demonstrates the sim-
772     plest way of using them is provided in the file called pcredemo.c in
773     the source distribution. The pcresample documentation describes how to
774     run it.
775 nigel 49
776 nigel 77 A second matching function, pcre_dfa_exec(), which is not Perl-compati-
777     ble, is also provided. This uses a different algorithm for the match-
778 nigel 91 ing. The alternative algorithm finds all possible matches (at a given
779 nigel 93 point in the subject), and scans the subject just once. However, this
780     algorithm does not return captured substrings. A description of the two
781     matching algorithms and their advantages and disadvantages is given in
782     the pcrematching documentation.
783 nigel 63
784 nigel 77 In addition to the main compiling and matching functions, there are
785     convenience functions for extracting captured substrings from a subject
786     string that is matched by pcre_exec(). They are:
787    
788 nigel 73 pcre_copy_substring()
789     pcre_copy_named_substring()
790     pcre_get_substring()
791     pcre_get_named_substring()
792     pcre_get_substring_list()
793 nigel 75 pcre_get_stringnumber()
794 nigel 91 pcre_get_stringtable_entries()
795 nigel 63
796 nigel 73 pcre_free_substring() and pcre_free_substring_list() are also provided,
797     to free the memory used for extracted strings.
798 nigel 41
799 nigel 77 The function pcre_maketables() is used to build a set of character
800     tables in the current locale for passing to pcre_compile(),
801     pcre_exec(), or pcre_dfa_exec(). This is an optional facility that is
802     provided for specialist use. Most commonly, no special tables are
803     passed, in which case internal tables that are generated when PCRE is
804     built are used.
805 nigel 49
806 nigel 75 The function pcre_fullinfo() is used to find out information about a
807     compiled pattern; pcre_info() is an obsolete version that returns only
808     some of the available information, but is retained for backwards com-
809     patibility. The function pcre_version() returns a pointer to a string
810 nigel 73 containing the version of PCRE and its date of release.
811 nigel 41
812 nigel 77 The function pcre_refcount() maintains a reference count in a data
813     block containing a compiled pattern. This is provided for the benefit
814     of object-oriented applications.
815    
816 nigel 75 The global variables pcre_malloc and pcre_free initially contain the
817     entry points of the standard malloc() and free() functions, respec-
818 nigel 73 tively. PCRE calls the memory management functions via these variables,
819 nigel 75 so a calling program can replace them if it wishes to intercept the
820 nigel 73 calls. This should be done before calling any PCRE functions.
821 nigel 41
822 nigel 75 The global variables pcre_stack_malloc and pcre_stack_free are also
823     indirections to memory management functions. These special functions
824     are used only when PCRE is compiled to use the heap for remembering
825 nigel 77 data, instead of recursive function calls, when running the pcre_exec()
826 nigel 91 function. See the pcrebuild documentation for details of how to do
827     this. It is a non-standard way of building PCRE, for use in environ-
828     ments that have limited stacks. Because of the greater use of memory
829     management, it runs more slowly. Separate functions are provided so
830     that special-purpose external code can be used for this case. When
831     used, these functions are always called in a stack-like manner (last
832     obtained, first freed), and always for memory blocks of the same size.
833     There is a discussion about PCRE's stack usage in the pcrestack docu-
834     mentation.
835 nigel 41
836 nigel 73 The global variable pcre_callout initially contains NULL. It can be set
837 nigel 77 by the caller to a "callout" function, which PCRE will then call at
838     specified points during a matching operation. Details are given in the
839 nigel 73 pcrecallout documentation.
840 nigel 41
841 nigel 73
842 nigel 91 NEWLINES
843    
844 nigel 93 PCRE supports four different conventions for indicating line breaks in
845     strings: a single CR (carriage return) character, a single LF (line-
846     feed) character, the two-character sequence CRLF, or any Unicode new-
847     line sequence. The Unicode newline sequences are the three just men-
848     tioned, plus the single characters VT (vertical tab, U+000B), FF (form-
849     feed, U+000C), NEL (next line, U+0085), LS (line separator, U+2028),
850     and PS (paragraph separator, U+2029).
851    
852     Each of the first three conventions is used by at least one operating
853     system as its standard newline sequence. When PCRE is built, a default
854     can be specified. The default default is LF, which is the Unix stan-
855     dard. When PCRE is run, the default can be overridden, either when a
856     pattern is compiled, or when it is matched.
857    
858 nigel 91 In the PCRE documentation the word "newline" is used to mean "the char-
859 nigel 93 acter or pair of characters that indicate a line break". The choice of
860     newline convention affects the handling of the dot, circumflex, and
861     dollar metacharacters, the handling of #-comments in /x mode, and, when
862     CRLF is a recognized line ending sequence, the match position advance-
863     ment for a non-anchored pattern. The choice of newline convention does
864     not affect the interpretation of the \n or \r escape sequences.
865 nigel 91
866    
867 nigel 63 MULTITHREADING
868    
869 nigel 93 The PCRE functions can be used in multi-threading applications, with
870 nigel 73 the proviso that the memory management functions pointed to by
871     pcre_malloc, pcre_free, pcre_stack_malloc, and pcre_stack_free, and the
872     callout function pointed to by pcre_callout, are shared by all threads.
873 nigel 41
874 nigel 93 The compiled form of a regular expression is not altered during match-
875 nigel 73 ing, so the same compiled pattern can safely be used by several threads
876     at once.
877 nigel 41
878    
879 nigel 75 SAVING PRECOMPILED PATTERNS FOR LATER USE
880    
881     The compiled form of a regular expression can be saved and re-used at a
882 nigel 93 later time, possibly by a different program, and even on a host other
883     than the one on which it was compiled. Details are given in the
884 nigel 75 pcreprecompile documentation.
885    
886    
887 nigel 63 CHECKING BUILD-TIME OPTIONS
888 nigel 41
889 nigel 73 int pcre_config(int what, void *where);
890 nigel 63
891 nigel 93 The function pcre_config() makes it possible for a PCRE client to dis-
892 nigel 73 cover which optional features have been compiled into the PCRE library.
893 nigel 93 The pcrebuild documentation has more details about these optional fea-
894 nigel 73 tures.
895 nigel 63
896 nigel 93 The first argument for pcre_config() is an integer, specifying which
897 nigel 73 information is required; the second argument is a pointer to a variable
898 nigel 93 into which the information is placed. The following information is
899 nigel 73 available:
900 nigel 63
901 nigel 73 PCRE_CONFIG_UTF8
902 nigel 63
903 nigel 93 The output is an integer that is set to one if UTF-8 support is avail-
904 nigel 73 able; otherwise it is set to zero.
905 nigel 63
906 nigel 75 PCRE_CONFIG_UNICODE_PROPERTIES
907    
908 nigel 93 The output is an integer that is set to one if support for Unicode
909 nigel 75 character properties is available; otherwise it is set to zero.
910    
911 nigel 73 PCRE_CONFIG_NEWLINE
912 nigel 63
913 nigel 93 The output is an integer whose value specifies the default character
914     sequence that is recognized as meaning "newline". The four values that
915     are supported are: 10 for LF, 13 for CR, 3338 for CRLF, and -1 for ANY.
916     The default should normally be the standard sequence for your operating
917     system.
918 nigel 63
919 nigel 73 PCRE_CONFIG_LINK_SIZE
920 nigel 63
921 nigel 91 The output is an integer that contains the number of bytes used for
922 nigel 73 internal linkage in compiled regular expressions. The value is 2, 3, or
923 nigel 91 4. Larger values allow larger regular expressions to be compiled, at
924     the expense of slower matching. The default value of 2 is sufficient
925     for all but the most massive patterns, since it allows the compiled
926 nigel 73 pattern to be up to 64K in size.
927 nigel 63
928 nigel 73 PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
929 nigel 63
930 nigel 91 The output is an integer that contains the threshold above which the
931     POSIX interface uses malloc() for output vectors. Further details are
932 nigel 73 given in the pcreposix documentation.
933 nigel 63
934 nigel 73 PCRE_CONFIG_MATCH_LIMIT
935 nigel 63
936 nigel 73 The output is an integer that gives the default limit for the number of
937 nigel 91 internal matching function calls in a pcre_exec() execution. Further
938 nigel 73 details are given with pcre_exec() below.
939 nigel 63
940 nigel 87 PCRE_CONFIG_MATCH_LIMIT_RECURSION
941    
942 nigel 91 The output is an integer that gives the default limit for the depth of
943     recursion when calling the internal matching function in a pcre_exec()
944 nigel 87 execution. Further details are given with pcre_exec() below.
945    
946 nigel 73 PCRE_CONFIG_STACKRECURSE
947 nigel 63
948 nigel 91 The output is an integer that is set to one if internal recursion when
949 nigel 77 running pcre_exec() is implemented by recursive function calls that use
950 nigel 91 the stack to remember their state. This is the usual way that PCRE is
951 nigel 77 compiled. The output is zero if PCRE was compiled to use blocks of data
952 nigel 91 on the heap instead of recursive function calls. In this case,
953     pcre_stack_malloc and pcre_stack_free are called to manage memory
954 nigel 77 blocks on the heap, thus avoiding the use of the stack.
955 nigel 73
956    
957 nigel 41 COMPILING A PATTERN
958 nigel 63
959 nigel 73 pcre *pcre_compile(const char *pattern, int options,
960     const char **errptr, int *erroffset,
961     const unsigned char *tableptr);
962 nigel 63
963 nigel 77 pcre *pcre_compile2(const char *pattern, int options,
964     int *errorcodeptr,
965     const char **errptr, int *erroffset,
966     const unsigned char *tableptr);
967 nigel 41
968 nigel 77 Either of the functions pcre_compile() or pcre_compile2() can be called
969     to compile a pattern into an internal form. The only difference between
970 nigel 91 the two interfaces is that pcre_compile2() has an additional argument,
971 nigel 77 errorcodeptr, via which a numerical error code can be returned.
972    
973     The pattern is a C string terminated by a binary zero, and is passed in
974 nigel 91 the pattern argument. A pointer to a single block of memory that is
975     obtained via pcre_malloc is returned. This contains the compiled code
976 nigel 77 and related data. The pcre type is defined for the returned block; this
977     is a typedef for a structure whose contents are not externally defined.
978 nigel 91 It is up to the caller to free the memory (via pcre_free) when it is no
979     longer required.
980 nigel 77
981 nigel 91 Although the compiled code of a PCRE regex is relocatable, that is, it
982 nigel 73 does not depend on memory location, the complete pcre data block is not
983 nigel 91 fully relocatable, because it may contain a copy of the tableptr argu-
984 nigel 75 ment, which is an address (see below).
985 nigel 41
986 nigel 93 The options argument contains various bit settings that affect the com-
987     pilation. It should be zero if no options are required. The available
988 nigel 91 options are described below. Some of them, in particular, those that
989     are compatible with Perl, can also be set and unset from within the
990     pattern (see the detailed description in the pcrepattern documenta-
991     tion). For these options, the contents of the options argument speci-
992     fies their initial settings at the start of compilation and execution.
993     The PCRE_ANCHORED and PCRE_NEWLINE_xxx options can be set at the time
994     of matching as well as at compile time.
995 nigel 41
996 nigel 73 If errptr is NULL, pcre_compile() returns NULL immediately. Otherwise,
997 nigel 91 if compilation of a pattern fails, pcre_compile() returns NULL, and
998 nigel 73 sets the variable pointed to by errptr to point to a textual error mes-
999 nigel 87 sage. This is a static string that is part of the library. You must not
1000     try to free it. The offset from the start of the pattern to the charac-
1001     ter where the error was discovered is placed in the variable pointed to
1002 nigel 91 by erroffset, which must not be NULL. If it is, an immediate error is
1003 nigel 73 given.
1004 nigel 53
1005 nigel 91 If pcre_compile2() is used instead of pcre_compile(), and the error-
1006     codeptr argument is not NULL, a non-zero error code number is returned
1007     via this argument in the event of an error. This is in addition to the
1008 nigel 77 textual error message. Error codes and messages are listed below.
1009    
1010 nigel 91 If the final argument, tableptr, is NULL, PCRE uses a default set of
1011     character tables that are built when PCRE is compiled, using the
1012     default C locale. Otherwise, tableptr must be an address that is the
1013     result of a call to pcre_maketables(). This value is stored with the
1014     compiled pattern, and used again by pcre_exec(), unless another table
1015 nigel 75 pointer is passed to it. For more discussion, see the section on locale
1016     support below.
1017 nigel 53
1018 nigel 91 This code fragment shows a typical straightforward call to pcre_com-
1019 nigel 73 pile():
1020 nigel 41
1021 nigel 73 pcre *re;
1022     const char *error;
1023     int erroffset;
1024     re = pcre_compile(
1025     "^A.*Z", /* the pattern */
1026     0, /* default options */
1027     &error, /* for error message */
1028     &erroffset, /* for error offset */
1029     NULL); /* use default character tables */
1030 nigel 41
1031 nigel 91 The following names for option bits are defined in the pcre.h header
1032 nigel 75 file:
1033 nigel 41
1034 nigel 73 PCRE_ANCHORED
1035 nigel 41
1036 nigel 73 If this bit is set, the pattern is forced to be "anchored", that is, it
1037 nigel 91 is constrained to match only at the first matching point in the string
1038     that is being searched (the "subject string"). This effect can also be
1039     achieved by appropriate constructs in the pattern itself, which is the
1040 nigel 73 only way to do it in Perl.
1041 nigel 41
1042 nigel 75 PCRE_AUTO_CALLOUT
1043    
1044     If this bit is set, pcre_compile() automatically inserts callout items,
1045 nigel 91 all with number 255, before each pattern item. For discussion of the
1046 nigel 75 callout facility, see the pcrecallout documentation.
1047    
1048 nigel 73 PCRE_CASELESS
1049 nigel 41
1050 nigel 91 If this bit is set, letters in the pattern match both upper and lower
1051     case letters. It is equivalent to Perl's /i option, and it can be
1052     changed within a pattern by a (?i) option setting. In UTF-8 mode, PCRE
1053     always understands the concept of case for characters whose values are
1054     less than 128, so caseless matching is always possible. For characters
1055     with higher values, the concept of case is supported if PCRE is com-
1056     piled with Unicode property support, but not otherwise. If you want to
1057     use caseless matching for characters 128 and above, you must ensure
1058     that PCRE is compiled with Unicode property support as well as with
1059 nigel 77 UTF-8 support.
1060 nigel 41
1061 nigel 73 PCRE_DOLLAR_ENDONLY
1062 nigel 41
1063 nigel 91 If this bit is set, a dollar metacharacter in the pattern matches only
1064     at the end of the subject string. Without this option, a dollar also
1065     matches immediately before a newline at the end of the string (but not
1066     before any other newlines). The PCRE_DOLLAR_ENDONLY option is ignored
1067     if PCRE_MULTILINE is set. There is no equivalent to this option in
1068     Perl, and no way to set it within a pattern.
1069 nigel 41
1070 nigel 73 PCRE_DOTALL
1071 nigel 41
1072 nigel 73 If this bit is set, a dot metacharater in the pattern matches all char-
1073 nigel 91 acters, including those that indicate newline. Without it, a dot does
1074     not match when the current position is at a newline. This option is
1075     equivalent to Perl's /s option, and it can be changed within a pattern
1076     by a (?s) option setting. A negative class such as [^a] always matches
1077 nigel 93 newline characters, independent of the setting of this option.
1078 nigel 63
1079 nigel 91 PCRE_DUPNAMES
1080    
1081     If this bit is set, names used to identify capturing subpatterns need
1082     not be unique. This can be helpful for certain types of pattern when it
1083     is known that only one instance of the named subpattern can ever be
1084     matched. There are more details of named subpatterns below; see also
1085     the pcrepattern documentation.
1086    
1087 nigel 73 PCRE_EXTENDED
1088 nigel 41
1089 nigel 91 If this bit is set, whitespace data characters in the pattern are
1090 nigel 77 totally ignored except when escaped or inside a character class. White-
1091     space does not include the VT character (code 11). In addition, charac-
1092     ters between an unescaped # outside a character class and the next new-
1093 nigel 91 line, inclusive, are also ignored. This is equivalent to Perl's /x
1094     option, and it can be changed within a pattern by a (?x) option set-
1095     ting.
1096 nigel 41
1097 nigel 91 This option makes it possible to include comments inside complicated
1098     patterns. Note, however, that this applies only to data characters.
1099     Whitespace characters may never appear within special character
1100     sequences in a pattern, for example within the sequence (?( which
1101 nigel 73 introduces a conditional subpattern.
1102 nigel 41
1103 nigel 73 PCRE_EXTRA
1104 nigel 41
1105 nigel 91 This option was invented in order to turn on additional functionality
1106     of PCRE that is incompatible with Perl, but it is currently of very
1107     little use. When set, any backslash in a pattern that is followed by a
1108     letter that has no special meaning causes an error, thus reserving
1109     these combinations for future expansion. By default, as in Perl, a
1110     backslash followed by a letter with no special meaning is treated as a
1111     literal. (Perl can, however, be persuaded to give a warning for this.)
1112     There are at present no other features controlled by this option. It
1113     can also be set by a (?X) option setting within a pattern.
1114 nigel 41
1115 nigel 77 PCRE_FIRSTLINE
1116    
1117 nigel 87 If this option is set, an unanchored pattern is required to match
1118 nigel 91 before or at the first newline in the subject string, though the
1119     matched text may continue over the newline.
1120 nigel 77
1121 nigel 73 PCRE_MULTILINE
1122 nigel 41
1123 nigel 87 By default, PCRE treats the subject string as consisting of a single
1124     line of characters (even if it actually contains newlines). The "start
1125     of line" metacharacter (^) matches only at the start of the string,
1126     while the "end of line" metacharacter ($) matches only at the end of
1127 nigel 75 the string, or before a terminating newline (unless PCRE_DOLLAR_ENDONLY
1128     is set). This is the same as Perl.
1129 nigel 63
1130 nigel 87 When PCRE_MULTILINE it is set, the "start of line" and "end of line"
1131 nigel 91 constructs match immediately following or immediately before internal
1132     newlines in the subject string, respectively, as well as at the very
1133     start and end. This is equivalent to Perl's /m option, and it can be
1134     changed within a pattern by a (?m) option setting. If there are no new-
1135     lines in a subject string, or no occurrences of ^ or $ in a pattern,
1136 nigel 73 setting PCRE_MULTILINE has no effect.
1137 nigel 63
1138 nigel 91 PCRE_NEWLINE_CR
1139     PCRE_NEWLINE_LF
1140     PCRE_NEWLINE_CRLF
1141 nigel 93 PCRE_NEWLINE_ANY
1142 nigel 91
1143     These options override the default newline definition that was chosen
1144     when PCRE was built. Setting the first or the second specifies that a
1145     newline is indicated by a single character (CR or LF, respectively).
1146 nigel 93 Setting PCRE_NEWLINE_CRLF specifies that a newline is indicated by the
1147     two-character CRLF sequence. Setting PCRE_NEWLINE_ANY specifies that
1148     any Unicode newline sequence should be recognized. The Unicode newline
1149     sequences are the three just mentioned, plus the single characters VT
1150     (vertical tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085),
1151     LS (line separator, U+2028), and PS (paragraph separator, U+2029). The
1152     last two are recognized only in UTF-8 mode.
1153 nigel 91
1154 nigel 93 The newline setting in the options word uses three bits that are
1155     treated as a number, giving eight possibilities. Currently only five
1156     are used (default plus the four values above). This means that if you
1157     set more than one newline option, the combination may or may not be
1158     sensible. For example, PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equiva-
1159     lent to PCRE_NEWLINE_CRLF, but other combinations yield unused numbers
1160     and cause an error.
1161 nigel 91
1162 nigel 93 The only time that a line break is specially recognized when compiling
1163     a pattern is if PCRE_EXTENDED is set, and an unescaped # outside a
1164     character class is encountered. This indicates a comment that lasts
1165     until after the next line break sequence. In other circumstances, line
1166     break sequences are treated as literal data, except that in
1167     PCRE_EXTENDED mode, both CR and LF are treated as whitespace characters
1168     and are therefore ignored.
1169    
1170     The newline option that is set at compile time becomes the default that
1171     is used for pcre_exec() and pcre_dfa_exec(), but it can be overridden.
1172    
1173 nigel 73 PCRE_NO_AUTO_CAPTURE
1174 nigel 41
1175 nigel 73 If this option is set, it disables the use of numbered capturing paren-
1176 nigel 91 theses in the pattern. Any opening parenthesis that is not followed by
1177     ? behaves as if it were followed by ?: but named parentheses can still
1178     be used for capturing (and they acquire numbers in the usual way).
1179 nigel 73 There is no equivalent of this option in Perl.
1180 nigel 41
1181 nigel 73 PCRE_UNGREEDY
1182 nigel 41
1183 nigel 91 This option inverts the "greediness" of the quantifiers so that they
1184     are not greedy by default, but become greedy if followed by "?". It is
1185     not compatible with Perl. It can also be set by a (?U) option setting
1186 nigel 73 within the pattern.
1187 nigel 41
1188 nigel 73 PCRE_UTF8
1189 nigel 49
1190 nigel 91 This option causes PCRE to regard both the pattern and the subject as
1191     strings of UTF-8 characters instead of single-byte character strings.
1192     However, it is available only when PCRE is built to include UTF-8 sup-
1193     port. If not, the use of this option provokes an error. Details of how
1194     this option changes the behaviour of PCRE are given in the section on
1195 nigel 75 UTF-8 support in the main pcre page.
1196 nigel 71
1197 nigel 73 PCRE_NO_UTF8_CHECK
1198 nigel 71
1199 nigel 73 When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
1200 nigel 91 automatically checked. If an invalid UTF-8 sequence of bytes is found,
1201     pcre_compile() returns an error. If you already know that your pattern
1202     is valid, and you want to skip this check for performance reasons, you
1203     can set the PCRE_NO_UTF8_CHECK option. When it is set, the effect of
1204 nigel 73 passing an invalid UTF-8 string as a pattern is undefined. It may cause
1205 nigel 91 your program to crash. Note that this option can also be passed to
1206     pcre_exec() and pcre_dfa_exec(), to suppress the UTF-8 validity check-
1207 nigel 77 ing of subject strings.
1208 nigel 71
1209 nigel 73
1210 nigel 77 COMPILATION ERROR CODES
1211    
1212 nigel 91 The following table lists the error codes than may be returned by
1213     pcre_compile2(), along with the error messages that may be returned by
1214 nigel 93 both compiling functions. As PCRE has developed, some error codes have
1215     fallen out of use. To avoid confusion, they have not been re-used.
1216 nigel 77
1217     0 no error
1218     1 \ at end of pattern
1219     2 \c at end of pattern
1220     3 unrecognized character follows \
1221     4 numbers out of order in {} quantifier
1222     5 number too big in {} quantifier
1223     6 missing terminating ] for character class
1224     7 invalid escape sequence in character class
1225     8 range out of order in character class
1226     9 nothing to repeat
1227 nigel 93 10 [this code is not in use]
1228 nigel 77 11 internal error: unexpected repeat
1229     12 unrecognized character after (?
1230     13 POSIX named classes are supported only within a class
1231     14 missing )
1232     15 reference to non-existent subpattern
1233     16 erroffset passed as NULL
1234     17 unknown option bit(s) set
1235     18 missing ) after comment
1236 nigel 93 19 [this code is not in use]
1237 nigel 77 20 regular expression too large
1238     21 failed to get memory
1239     22 unmatched parentheses
1240     23 internal error: code overflow
1241     24 unrecognized character after (?<
1242     25 lookbehind assertion is not fixed length
1243 nigel 91 26 malformed number or name after (?(
1244 nigel 77 27 conditional group contains more than two branches
1245     28 assertion expected after (?(
1246     29 (?R or (?digits must be followed by )
1247     30 unknown POSIX class name
1248     31 POSIX collating elements are not supported
1249     32 this version of PCRE is not compiled with PCRE_UTF8 support
1250 nigel 93 33 [this code is not in use]
1251 nigel 77 34 character value in \x{...} sequence is too large
1252     35 invalid condition (?(0)
1253     36 \C not allowed in lookbehind assertion
1254     37 PCRE does not support \L, \l, \N, \U, or \u
1255     38 number after (?C is > 255
1256     39 closing ) for (?C expected
1257     40 recursive call could loop indefinitely
1258     41 unrecognized character after (?P
1259 nigel 93 42 syntax error in subpattern name (missing terminator)
1260 nigel 91 43 two named subpatterns have the same name
1261 nigel 77 44 invalid UTF-8 string
1262     45 support for \P, \p, and \X has not been compiled
1263     46 malformed \P or \p sequence
1264     47 unknown property name after \P or \p
1265 nigel 91 48 subpattern name is too long (maximum 32 characters)
1266     49 too many named subpatterns (maximum 10,000)
1267     50 repeated subpattern is too long
1268     51 octal value is greater than \377 (not in UTF-8 mode)
1269 nigel 93 52 internal error: overran compiling workspace
1270     53 internal error: previously-checked referenced subpattern not
1271     found
1272     54 DEFINE group contains more than one branch
1273     55 repeating a DEFINE group is not allowed
1274     56 inconsistent NEWLINE options"
1275 nigel 77
1276    
1277 nigel 63 STUDYING A PATTERN
1278 nigel 49
1279 nigel 77 pcre_extra *pcre_study(const pcre *code, int options
1280 nigel 73 const char **errptr);
1281 nigel 63
1282 nigel 91 If a compiled pattern is going to be used several times, it is worth
1283 nigel 75 spending more time analyzing it in order to speed up the time taken for
1284 nigel 91 matching. The function pcre_study() takes a pointer to a compiled pat-
1285 nigel 75 tern as its first argument. If studying the pattern produces additional
1286 nigel 91 information that will help speed up matching, pcre_study() returns a
1287     pointer to a pcre_extra block, in which the study_data field points to
1288 nigel 75 the results of the study.
1289 nigel 41
1290 nigel 75 The returned value from pcre_study() can be passed directly to
1291 nigel 91 pcre_exec(). However, a pcre_extra block also contains other fields
1292     that can be set by the caller before the block is passed; these are
1293 nigel 75 described below in the section on matching a pattern.
1294 nigel 63
1295 nigel 91 If studying the pattern does not produce any additional information
1296 nigel 75 pcre_study() returns NULL. In that circumstance, if the calling program
1297 nigel 91 wants to pass any of the other fields to pcre_exec(), it must set up
1298 nigel 75 its own pcre_extra block.
1299 nigel 41
1300 nigel 91 The second argument of pcre_study() contains option bits. At present,
1301 nigel 75 no options are defined, and this argument should always be zero.
1302    
1303 nigel 91 The third argument for pcre_study() is a pointer for an error message.
1304     If studying succeeds (even if no data is returned), the variable it
1305     points to is set to NULL. Otherwise it is set to point to a textual
1306 nigel 87 error message. This is a static string that is part of the library. You
1307 nigel 91 must not try to free it. You should test the error pointer for NULL
1308 nigel 87 after calling pcre_study(), to be sure that it has run successfully.
1309 nigel 41
1310 nigel 73 This is a typical call to pcre_study():
1311 nigel 53
1312 nigel 73 pcre_extra *pe;
1313     pe = pcre_study(
1314     re, /* result of pcre_compile() */
1315     0, /* no options exist */
1316     &error); /* set to NULL or points to a message */
1317 nigel 53
1318 nigel 73 At present, studying a pattern is useful only for non-anchored patterns
1319 nigel 91 that do not have a single fixed starting character. A bitmap of possi-
1320 nigel 75 ble starting bytes is created.
1321 nigel 41
1322    
1323 nigel 63 LOCALE SUPPORT
1324 nigel 41
1325 nigel 91 PCRE handles caseless matching, and determines whether characters are
1326 ph10 142 letters, digits, or whatever, by reference to a set of tables, indexed
1327 nigel 91 by character value. When running in UTF-8 mode, this applies only to
1328     characters with codes less than 128. Higher-valued codes never match
1329     escapes such as \w or \d, but can be tested with \p if PCRE is built
1330     with Unicode character property support. The use of locales with Uni-
1331 ph10 142 code is discouraged. If you are handling characters with codes greater
1332     than 128, you should either use UTF-8 and Unicode, or use locales, but
1333     not try to mix the two.
1334 nigel 41
1335 ph10 142 PCRE contains an internal set of tables that are used when the final
1336     argument of pcre_compile() is NULL. These are sufficient for many
1337     applications. Normally, the internal tables recognize only ASCII char-
1338     acters. However, when PCRE is built, it is possible to cause the inter-
1339     nal tables to be rebuilt in the default "C" locale of the local system,
1340     which may cause them to be different.
1341 nigel 41
1342 ph10 142 The internal tables can always be overridden by tables supplied by the
1343     application that calls PCRE. These may be created in a different locale
1344     from the default. As more and more applications change to using Uni-
1345     code, the need for this locale support is expected to die away.
1346    
1347     External tables are built by calling the pcre_maketables() function,
1348     which has no arguments, in the relevant locale. The result can then be
1349     passed to pcre_compile() or pcre_exec() as often as necessary. For
1350     example, to build and use tables that are appropriate for the French
1351     locale (where accented characters with values greater than 128 are
1352 nigel 75 treated as letters), the following code could be used:
1353    
1354     setlocale(LC_CTYPE, "fr_FR");
1355 nigel 73 tables = pcre_maketables();
1356     re = pcre_compile(..., tables);
1357 nigel 41
1358 ph10 142 The locale name "fr_FR" is used on Linux and other Unix-like systems;
1359     if you are using Windows, the name for the French locale is "french".
1360    
1361 nigel 91 When pcre_maketables() runs, the tables are built in memory that is
1362     obtained via pcre_malloc. It is the caller's responsibility to ensure
1363     that the memory containing the tables remains available for as long as
1364 nigel 75 it is needed.
1365 nigel 41
1366 nigel 75 The pointer that is passed to pcre_compile() is saved with the compiled
1367 nigel 91 pattern, and the same tables are used via this pointer by pcre_study()
1368 nigel 75 and normally also by pcre_exec(). Thus, by default, for any single pat-
1369     tern, compilation, studying and matching all happen in the same locale,
1370     but different patterns can be compiled in different locales.
1371 nigel 41
1372 nigel 91 It is possible to pass a table pointer or NULL (indicating the use of
1373     the internal tables) to pcre_exec(). Although not intended for this
1374     purpose, this facility could be used to match a pattern in a different
1375 nigel 75 locale from the one in which it was compiled. Passing table pointers at
1376     run time is discussed below in the section on matching a pattern.
1377    
1378    
1379 nigel 63 INFORMATION ABOUT A PATTERN
1380 nigel 41
1381 nigel 73 int pcre_fullinfo(const pcre *code, const pcre_extra *extra,
1382     int what, void *where);
1383 nigel 63
1384 nigel 91 The pcre_fullinfo() function returns information about a compiled pat-
1385 nigel 73 tern. It replaces the obsolete pcre_info() function, which is neverthe-
1386     less retained for backwards compability (and is documented below).
1387 nigel 43
1388 nigel 91 The first argument for pcre_fullinfo() is a pointer to the compiled
1389     pattern. The second argument is the result of pcre_study(), or NULL if
1390     the pattern was not studied. The third argument specifies which piece
1391     of information is required, and the fourth argument is a pointer to a
1392     variable to receive the data. The yield of the function is zero for
1393 nigel 73 success, or one of the following negative numbers:
1394 nigel 41
1395 nigel 73 PCRE_ERROR_NULL the argument code was NULL
1396     the argument where was NULL
1397     PCRE_ERROR_BADMAGIC the "magic number" was not found
1398     PCRE_ERROR_BADOPTION the value of what was invalid
1399 nigel 53
1400 nigel 91 The "magic number" is placed at the start of each compiled pattern as
1401     an simple check against passing an arbitrary memory pointer. Here is a
1402     typical call of pcre_fullinfo(), to obtain the length of the compiled
1403 nigel 75 pattern:
1404 nigel 53
1405 nigel 73 int rc;
1406 nigel 91 size_t length;
1407 nigel 73 rc = pcre_fullinfo(
1408     re, /* result of pcre_compile() */
1409     pe, /* result of pcre_study(), or NULL */
1410     PCRE_INFO_SIZE, /* what is required */
1411     &length); /* where to put the data */
1412 nigel 43
1413 nigel 91 The possible values for the third argument are defined in pcre.h, and
1414 nigel 73 are as follows:
1415 nigel 43
1416 nigel 73 PCRE_INFO_BACKREFMAX
1417 nigel 41
1418 nigel 91 Return the number of the highest back reference in the pattern. The
1419     fourth argument should point to an int variable. Zero is returned if
1420 nigel 73 there are no back references.
1421 nigel 43
1422 nigel 73 PCRE_INFO_CAPTURECOUNT
1423 nigel 43
1424 nigel 91 Return the number of capturing subpatterns in the pattern. The fourth
1425 nigel 73 argument should point to an int variable.
1426 nigel 43
1427 nigel 77 PCRE_INFO_DEFAULT_TABLES
1428 nigel 75
1429 nigel 91 Return a pointer to the internal default character tables within PCRE.
1430     The fourth argument should point to an unsigned char * variable. This
1431 nigel 75 information call is provided for internal use by the pcre_study() func-
1432 nigel 91 tion. External callers can cause PCRE to use its internal tables by
1433 nigel 75 passing a NULL table pointer.
1434    
1435 nigel 73 PCRE_INFO_FIRSTBYTE
1436 nigel 43
1437 nigel 91 Return information about the first byte of any matched string, for a
1438     non-anchored pattern. The fourth argument should point to an int vari-
1439     able. (This option used to be called PCRE_INFO_FIRSTCHAR; the old name
1440     is still recognized for backwards compatibility.)
1441 nigel 41
1442 nigel 91 If there is a fixed first byte, for example, from a pattern such as
1443 nigel 93 (cat|cow|coyote), its value is returned. Otherwise, if either
1444 nigel 41
1445 nigel 87 (a) the pattern was compiled with the PCRE_MULTILINE option, and every
1446 nigel 73 branch starts with "^", or
1447 nigel 43
1448 nigel 73 (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not
1449     set (if it were set, the pattern would be anchored),
1450 nigel 41
1451 nigel 87 -1 is returned, indicating that the pattern matches only at the start
1452     of a subject string or after any newline within the string. Otherwise
1453 nigel 73 -2 is returned. For anchored patterns, -2 is returned.
1454 nigel 41
1455 nigel 73 PCRE_INFO_FIRSTTABLE
1456 nigel 41
1457 nigel 87 If the pattern was studied, and this resulted in the construction of a
1458 nigel 73 256-bit table indicating a fixed set of bytes for the first byte in any
1459 nigel 87 matching string, a pointer to the table is returned. Otherwise NULL is
1460     returned. The fourth argument should point to an unsigned char * vari-
1461 nigel 73 able.
1462 nigel 43
1463 nigel 73 PCRE_INFO_LASTLITERAL
1464 nigel 43
1465 nigel 87 Return the value of the rightmost literal byte that must exist in any
1466     matched string, other than at its start, if such a byte has been
1467 nigel 73 recorded. The fourth argument should point to an int variable. If there
1468 nigel 87 is no such byte, -1 is returned. For anchored patterns, a last literal
1469     byte is recorded only if it follows something of variable length. For
1470 nigel 73 example, for the pattern /^a\d+z\d+/ the returned value is "z", but for
1471     /^a\dz\d/ the returned value is -1.
1472 nigel 63
1473 nigel 73 PCRE_INFO_NAMECOUNT
1474     PCRE_INFO_NAMEENTRYSIZE
1475     PCRE_INFO_NAMETABLE
1476 nigel 63
1477 nigel 87 PCRE supports the use of named as well as numbered capturing parenthe-
1478     ses. The names are just an additional way of identifying the parenthe-
1479 nigel 91 ses, which still acquire numbers. Several convenience functions such as
1480     pcre_get_named_substring() are provided for extracting captured sub-
1481     strings by name. It is also possible to extract the data directly, by
1482     first converting the name to a number in order to access the correct
1483     pointers in the output vector (described with pcre_exec() below). To do
1484     the conversion, you need to use the name-to-number map, which is
1485     described by these three values.
1486 nigel 63
1487 nigel 73 The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT
1488     gives the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size
1489 nigel 87 of each entry; both of these return an int value. The entry size
1490     depends on the length of the longest name. PCRE_INFO_NAMETABLE returns
1491     a pointer to the first entry of the table (a pointer to char). The
1492 nigel 73 first two bytes of each entry are the number of the capturing parenthe-
1493 nigel 87 sis, most significant byte first. The rest of the entry is the corre-
1494     sponding name, zero terminated. The names are in alphabetical order.
1495 nigel 91 When PCRE_DUPNAMES is set, duplicate names are in order of their paren-
1496     theses numbers. For example, consider the following pattern (assume
1497     PCRE_EXTENDED is set, so white space - including newlines - is
1498     ignored):
1499 nigel 63
1500 nigel 93 (?<date> (?<year>(\d\d)?\d\d) -
1501     (?<month>\d\d) - (?<day>\d\d) )
1502 nigel 63
1503 nigel 87 There are four named subpatterns, so the table has four entries, and
1504     each entry in the table is eight bytes long. The table is as follows,
1505 nigel 75 with non-printing bytes shows in hexadecimal, and undefined bytes shown
1506     as ??:
1507 nigel 63
1508 nigel 73 00 01 d a t e 00 ??
1509     00 05 d a y 00 ?? ??
1510     00 04 m o n t h 00
1511     00 02 y e a r 00 ??
1512 nigel 63
1513 nigel 87 When writing code to extract data from named subpatterns using the
1514 nigel 91 name-to-number map, remember that the length of the entries is likely
1515     to be different for each compiled pattern.
1516 nigel 63
1517 nigel 73 PCRE_INFO_OPTIONS
1518 nigel 63
1519 nigel 87 Return a copy of the options with which the pattern was compiled. The
1520     fourth argument should point to an unsigned long int variable. These
1521 nigel 73 option bits are those specified in the call to pcre_compile(), modified
1522     by any top-level option settings within the pattern itself.
1523 nigel 63
1524 nigel 87 A pattern is automatically anchored by PCRE if all of its top-level
1525 nigel 73 alternatives begin with one of the following:
1526 nigel 63
1527 nigel 73 ^ unless PCRE_MULTILINE is set
1528     \A always
1529     \G always
1530     .* if PCRE_DOTALL is set and there are no back
1531     references to the subpattern in which .* appears
1532 nigel 63
1533 nigel 73 For such patterns, the PCRE_ANCHORED bit is set in the options returned
1534     by pcre_fullinfo().
1535 nigel 63
1536 nigel 73 PCRE_INFO_SIZE
1537 nigel 63
1538 nigel 87 Return the size of the compiled pattern, that is, the value that was
1539 nigel 73 passed as the argument to pcre_malloc() when PCRE was getting memory in
1540     which to place the compiled data. The fourth argument should point to a
1541     size_t variable.
1542 nigel 63
1543 nigel 73 PCRE_INFO_STUDYSIZE
1544 nigel 63
1545 nigel 75 Return the size of the data block pointed to by the study_data field in
1546 nigel 87 a pcre_extra block. That is, it is the value that was passed to
1547 nigel 73 pcre_malloc() when PCRE was getting memory into which to place the data
1548 nigel 87 created by pcre_study(). The fourth argument should point to a size_t
1549 nigel 73 variable.
1550 nigel 63
1551 nigel 73
1552 nigel 63 OBSOLETE INFO FUNCTION
1553    
1554 nigel 73 int pcre_info(const pcre *code, int *optptr, int *firstcharptr);
1555 nigel 63
1556 nigel 87 The pcre_info() function is now obsolete because its interface is too
1557     restrictive to return all the available data about a compiled pattern.
1558     New programs should use pcre_fullinfo() instead. The yield of
1559     pcre_info() is the number of capturing subpatterns, or one of the fol-
1560 nigel 73 lowing negative numbers:
1561 nigel 43
1562 nigel 73 PCRE_ERROR_NULL the argument code was NULL
1563     PCRE_ERROR_BADMAGIC the "magic number" was not found
1564 nigel 43
1565 nigel 87 If the optptr argument is not NULL, a copy of the options with which
1566     the pattern was compiled is placed in the integer it points to (see
1567 nigel 73 PCRE_INFO_OPTIONS above).
1568 nigel 43
1569 nigel 87 If the pattern is not anchored and the firstcharptr argument is not
1570     NULL, it is used to pass back information about the first character of
1571 nigel 73 any matched string (see PCRE_INFO_FIRSTBYTE above).
1572 nigel 43
1573    
1574 nigel 77 REFERENCE COUNTS
1575 nigel 53
1576 nigel 77 int pcre_refcount(pcre *code, int adjust);
1577    
1578 nigel 87 The pcre_refcount() function is used to maintain a reference count in
1579 nigel 77 the data block that contains a compiled pattern. It is provided for the
1580 nigel 87 benefit of applications that operate in an object-oriented manner,
1581 nigel 77 where different parts of the application may be using the same compiled
1582     pattern, but you want to free the block when they are all done.
1583    
1584     When a pattern is compiled, the reference count field is initialized to
1585 nigel 87 zero. It is changed only by calling this function, whose action is to
1586     add the adjust value (which may be positive or negative) to it. The
1587 nigel 77 yield of the function is the new value. However, the value of the count
1588 nigel 87 is constrained to lie between 0 and 65535, inclusive. If the new value
1589 nigel 77 is outside these limits, it is forced to the appropriate limit value.
1590    
1591 nigel 87 Except when it is zero, the reference count is not correctly preserved
1592     if a pattern is compiled on one host and then transferred to a host
1593 nigel 77 whose byte-order is different. (This seems a highly unlikely scenario.)
1594    
1595    
1596     MATCHING A PATTERN: THE TRADITIONAL FUNCTION
1597    
1598 nigel 73 int pcre_exec(const pcre *code, const pcre_extra *extra,
1599     const char *subject, int length, int startoffset,
1600     int options, int *ovector, int ovecsize);
1601 nigel 53
1602 nigel 87 The function pcre_exec() is called to match a subject string against a
1603     compiled pattern, which is passed in the code argument. If the pattern
1604 nigel 75 has been studied, the result of the study should be passed in the extra
1605 nigel 87 argument. This function is the main matching facility of the library,
1606 nigel 77 and it operates in a Perl-like manner. For specialist use there is also
1607 nigel 87 an alternative matching function, which is described below in the sec-
1608 nigel 77 tion about the pcre_dfa_exec() function.
1609 nigel 41
1610 nigel 87 In most applications, the pattern will have been compiled (and option-
1611     ally studied) in the same process that calls pcre_exec(). However, it
1612 nigel 75 is possible to save compiled patterns and study data, and then use them
1613 nigel 87 later in different processes, possibly even on different hosts. For a
1614 nigel 75 discussion about this, see the pcreprecompile documentation.
1615    
1616 nigel 73 Here is an example of a simple call to pcre_exec():
1617 nigel 53
1618 nigel 73 int rc;
1619     int ovector[30];
1620     rc = pcre_exec(
1621     re, /* result of pcre_compile() */
1622     NULL, /* we didn't study the pattern */
1623     "some string", /* the subject string */
1624     11, /* the length of the subject string */
1625     0, /* start at offset 0 in the subject */
1626     0, /* default options */
1627 nigel 75 ovector, /* vector of integers for substring information */
1628 nigel 77 30); /* number of elements (NOT size in bytes) */
1629 nigel 53
1630 nigel 75 Extra data for pcre_exec()
1631 nigel 63
1632 nigel 87 If the extra argument is not NULL, it must point to a pcre_extra data
1633     block. The pcre_study() function returns such a block (when it doesn't
1634     return NULL), but you can also create one for yourself, and pass addi-
1635     tional information in it. The pcre_extra block contains the following
1636     fields (not necessarily in this order):
1637 nigel 75
1638 nigel 73 unsigned long int flags;
1639     void *study_data;
1640     unsigned long int match_limit;
1641 nigel 87 unsigned long int match_limit_recursion;
1642 nigel 73 void *callout_data;
1643 nigel 75 const unsigned char *tables;
1644 nigel 63
1645 nigel 87 The flags field is a bitmap that specifies which of the other fields
1646 nigel 73 are set. The flag bits are:
1647 nigel 63
1648 nigel 73 PCRE_EXTRA_STUDY_DATA
1649     PCRE_EXTRA_MATCH_LIMIT
1650 nigel 87 PCRE_EXTRA_MATCH_LIMIT_RECURSION
1651 nigel 73 PCRE_EXTRA_CALLOUT_DATA
1652 nigel 75 PCRE_EXTRA_TABLES
1653 nigel 63
1654 nigel 87 Other flag bits should be set to zero. The study_data field is set in
1655     the pcre_extra block that is returned by pcre_study(), together with
1656 nigel 75 the appropriate flag bit. You should not set this yourself, but you may
1657 nigel 87 add to the block by setting the other fields and their corresponding
1658 nigel 75 flag bits.
1659 nigel 63
1660 nigel 73 The match_limit field provides a means of preventing PCRE from using up
1661 nigel 87 a vast amount of resources when running patterns that are not going to
1662     match, but which have a very large number of possibilities in their
1663     search trees. The classic example is the use of nested unlimited
1664 nigel 75 repeats.
1665 nigel 63
1666 nigel 87 Internally, PCRE uses a function called match() which it calls repeat-
1667     edly (sometimes recursively). The limit set by match_limit is imposed
1668     on the number of times this function is called during a match, which
1669     has the effect of limiting the amount of backtracking that can take
1670     place. For patterns that are not anchored, the count restarts from zero
1671     for each position in the subject string.
1672 nigel 75
1673 nigel 87 The default value for the limit can be set when PCRE is built; the
1674     default default is 10 million, which handles all but the most extreme
1675     cases. You can override the default by suppling pcre_exec() with a
1676     pcre_extra block in which match_limit is set, and
1677     PCRE_EXTRA_MATCH_LIMIT is set in the flags field. If the limit is
1678 nigel 73 exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1679 nigel 63
1680 nigel 87 The match_limit_recursion field is similar to match_limit, but instead
1681     of limiting the total number of times that match() is called, it limits
1682     the depth of recursion. The recursion depth is a smaller number than
1683     the total number of calls, because not all calls to match() are recur-
1684     sive. This limit is of use only if it is set smaller than match_limit.
1685    
1686     Limiting the recursion depth limits the amount of stack that can be
1687     used, or, when PCRE has been compiled to use memory on the heap instead
1688     of the stack, the amount of heap memory that can be used.
1689    
1690     The default value for match_limit_recursion can be set when PCRE is
1691     built; the default default is the same value as the default for
1692     match_limit. You can override the default by suppling pcre_exec() with
1693     a pcre_extra block in which match_limit_recursion is set, and
1694     PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the flags field. If the
1695     limit is exceeded, pcre_exec() returns PCRE_ERROR_RECURSIONLIMIT.
1696    
1697     The pcre_callout field is used in conjunction with the "callout" fea-
1698 nigel 73 ture, which is described in the pcrecallout documentation.
1699 nigel 63
1700 nigel 87 The tables field is used to pass a character tables pointer to
1701     pcre_exec(); this overrides the value that is stored with the compiled
1702     pattern. A non-NULL value is stored with the compiled pattern only if
1703     custom tables were supplied to pcre_compile() via its tableptr argu-
1704 nigel 75 ment. If NULL is passed to pcre_exec() using this mechanism, it forces
1705 nigel 87 PCRE's internal tables to be used. This facility is helpful when re-
1706     using patterns that have been saved after compiling with an external
1707     set of tables, because the external tables might be at a different
1708     address when pcre_exec() is called. See the pcreprecompile documenta-
1709 nigel 75 tion for a discussion of saving compiled patterns for later use.
1710 nigel 41
1711 nigel 75 Option bits for pcre_exec()
1712 nigel 71
1713 nigel 87 The unused bits of the options argument for pcre_exec() must be zero.
1714 nigel 91 The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_xxx,
1715     PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK and
1716     PCRE_PARTIAL.
1717 nigel 41
1718 nigel 75 PCRE_ANCHORED
1719 nigel 41
1720 nigel 91 The PCRE_ANCHORED option limits pcre_exec() to matching at the first
1721     matching position. If a pattern was compiled with PCRE_ANCHORED, or
1722     turned out to be anchored by virtue of its contents, it cannot be made
1723 nigel 75 unachored at matching time.
1724    
1725 nigel 91 PCRE_NEWLINE_CR
1726     PCRE_NEWLINE_LF
1727     PCRE_NEWLINE_CRLF
1728 nigel 93 PCRE_NEWLINE_ANY
1729 nigel 91
1730     These options override the newline definition that was chosen or
1731     defaulted when the pattern was compiled. For details, see the descrip-
1732 nigel 93 tion of pcre_compile() above. During matching, the newline choice
1733     affects the behaviour of the dot, circumflex, and dollar metacharac-
1734     ters. It may also alter the way the match position is advanced after a
1735     match failure for an unanchored pattern. When PCRE_NEWLINE_CRLF or
1736     PCRE_NEWLINE_ANY is set, and a match attempt fails when the current
1737     position is at a CRLF sequence, the match position is advanced by two
1738     characters instead of one, in other words, to after the CRLF.
1739 nigel 91
1740 nigel 73 PCRE_NOTBOL
1741 nigel 41
1742 nigel 75 This option specifies that first character of the subject string is not
1743 nigel 93 the beginning of a line, so the circumflex metacharacter should not
1744     match before it. Setting this without PCRE_MULTILINE (at compile time)
1745     causes circumflex never to match. This option affects only the behav-
1746 nigel 77 iour of the circumflex metacharacter. It does not affect \A.
1747 nigel 41
1748 nigel 73 PCRE_NOTEOL
1749 nigel 41
1750 nigel 75 This option specifies that the end of the subject string is not the end
1751 nigel 93 of a line, so the dollar metacharacter should not match it nor (except
1752     in multiline mode) a newline immediately before it. Setting this with-
1753 nigel 75 out PCRE_MULTILINE (at compile time) causes dollar never to match. This
1754 nigel 93 option affects only the behaviour of the dollar metacharacter. It does
1755 nigel 75 not affect \Z or \z.
1756 nigel 41
1757 nigel 73 PCRE_NOTEMPTY
1758 nigel 41
1759 nigel 73 An empty string is not considered to be a valid match if this option is
1760 nigel 93 set. If there are alternatives in the pattern, they are tried. If all
1761     the alternatives match the empty string, the entire match fails. For
1762 nigel 73 example, if the pattern
1763 nigel 41
1764 nigel 73 a?b?
1765 nigel 41
1766 nigel 93 is applied to a string not beginning with "a" or "b", it matches the
1767     empty string at the start of the subject. With PCRE_NOTEMPTY set, this
1768 nigel 73 match is not valid, so PCRE searches further into the string for occur-
1769     rences of "a" or "b".
1770 nigel 41
1771 nigel 73 Perl has no direct equivalent of PCRE_NOTEMPTY, but it does make a spe-
1772 nigel 93 cial case of a pattern match of the empty string within its split()
1773     function, and when using the /g modifier. It is possible to emulate
1774 nigel 73 Perl's behaviour after matching a null string by first trying the match
1775 nigel 75 again at the same offset with PCRE_NOTEMPTY and PCRE_ANCHORED, and then
1776 nigel 93 if that fails by advancing the starting offset (see below) and trying
1777 nigel 75 an ordinary match again. There is some code that demonstrates how to do
1778     this in the pcredemo.c sample program.
1779 nigel 41
1780 nigel 75 PCRE_NO_UTF8_CHECK
1781    
1782     When PCRE_UTF8 is set at compile time, the validity of the subject as a
1783 nigel 93 UTF-8 string is automatically checked when pcre_exec() is subsequently
1784     called. The value of startoffset is also checked to ensure that it
1785     points to the start of a UTF-8 character. If an invalid UTF-8 sequence
1786 nigel 75 of bytes is found, pcre_exec() returns the error PCRE_ERROR_BADUTF8. If
1787 nigel 93 startoffset contains an invalid value, PCRE_ERROR_BADUTF8_OFFSET is
1788 nigel 75 returned.
1789    
1790 nigel 93 If you already know that your subject is valid, and you want to skip
1791     these checks for performance reasons, you can set the
1792     PCRE_NO_UTF8_CHECK option when calling pcre_exec(). You might want to
1793     do this for the second and subsequent calls to pcre_exec() if you are
1794     making repeated calls to find all the matches in a single subject
1795     string. However, you should be sure that the value of startoffset
1796     points to the start of a UTF-8 character. When PCRE_NO_UTF8_CHECK is
1797     set, the effect of passing an invalid UTF-8 string as a subject, or a
1798     value of startoffset that does not point to the start of a UTF-8 char-
1799 nigel 75 acter, is undefined. Your program may crash.
1800    
1801     PCRE_PARTIAL
1802    
1803 nigel 93 This option turns on the partial matching feature. If the subject
1804     string fails to match the pattern, but at some point during the match-
1805     ing process the end of the subject was reached (that is, the subject
1806     partially matches the pattern and the failure to match occurred only
1807     because there were not enough subject characters), pcre_exec() returns
1808     PCRE_ERROR_PARTIAL instead of PCRE_ERROR_NOMATCH. When PCRE_PARTIAL is
1809     used, there are restrictions on what may appear in the pattern. These
1810 nigel 75 are discussed in the pcrepartial documentation.
1811    
1812     The string to be matched by pcre_exec()
1813    
1814 nigel 93 The subject string is passed to pcre_exec() as a pointer in subject, a
1815     length in length, and a starting byte offset in startoffset. In UTF-8
1816     mode, the byte offset must point to the start of a UTF-8 character.
1817     Unlike the pattern string, the subject may contain binary zero bytes.
1818     When the starting offset is zero, the search for a match starts at the
1819 nigel 75 beginning of the subject, and this is by far the most common case.
1820 nigel 63
1821 nigel 93 A non-zero starting offset is useful when searching for another match
1822     in the same subject by calling pcre_exec() again after a previous suc-
1823     cess. Setting startoffset differs from just passing over a shortened
1824     string and setting PCRE_NOTBOL in the case of a pattern that begins
1825 nigel 73 with any kind of lookbehind. For example, consider the pattern
1826 nigel 41
1827 nigel 73 \Biss\B
1828 nigel 41
1829 nigel 93 which finds occurrences of "iss" in the middle of words. (\B matches
1830     only if the current position in the subject is not a word boundary.)
1831     When applied to the string "Mississipi" the first call to pcre_exec()
1832     finds the first occurrence. If pcre_exec() is called again with just
1833     the remainder of the subject, namely "issipi", it does not match,
1834 nigel 73 because \B is always false at the start of the subject, which is deemed
1835 nigel 93 to be a word boundary. However, if pcre_exec() is passed the entire
1836 nigel 75 string again, but with startoffset set to 4, it finds the second occur-
1837 nigel 93 rence of "iss" because it is able to look behind the starting point to
1838 nigel 75 discover that it is preceded by a letter.
1839 nigel 41
1840 nigel 93 If a non-zero starting offset is passed when the pattern is anchored,
1841 nigel 75 one attempt to match at the given offset is made. This can only succeed
1842 nigel 93 if the pattern does not require the match to be at the start of the
1843 nigel 75 subject.
1844 nigel 41
1845 nigel 75 How pcre_exec() returns captured substrings
1846    
1847 nigel 93 In general, a pattern matches a certain portion of the subject, and in
1848     addition, further substrings from the subject may be picked out by
1849     parts of the pattern. Following the usage in Jeffrey Friedl's book,
1850     this is called "capturing" in what follows, and the phrase "capturing
1851     subpattern" is used for a fragment of a pattern that picks out a sub-
1852     string. PCRE supports several other kinds of parenthesized subpattern
1853 nigel 73 that do not cause substrings to be captured.
1854 nigel 65
1855 nigel 93 Captured substrings are returned to the caller via a vector of integer
1856     offsets whose address is passed in ovector. The number of elements in
1857     the vector is passed in ovecsize, which must be a non-negative number.
1858 nigel 75 Note: this argument is NOT the size of ovector in bytes.
1859 nigel 41
1860 nigel 93 The first two-thirds of the vector is used to pass back captured sub-
1861     strings, each substring using a pair of integers. The remaining third
1862     of the vector is used as workspace by pcre_exec() while matching cap-
1863     turing subpatterns, and is not available for passing back information.
1864     The length passed in ovecsize should always be a multiple of three. If
1865 nigel 75 it is not, it is rounded down.
1866    
1867 nigel 93 When a match is successful, information about captured substrings is
1868     returned in pairs of integers, starting at the beginning of ovector,
1869     and continuing up to two-thirds of its length at the most. The first
1870 nigel 73 element of a pair is set to the offset of the first character in a sub-
1871 nigel 93 string, and the second is set to the offset of the first character
1872     after the end of a substring. The first pair, ovector[0] and ovec-
1873     tor[1], identify the portion of the subject string matched by the
1874     entire pattern. The next pair is used for the first capturing subpat-
1875 nigel 91 tern, and so on. The value returned by pcre_exec() is one more than the
1876     highest numbered pair that has been set. For example, if two substrings
1877 nigel 93 have been captured, the returned value is 3. If there are no capturing
1878     subpatterns, the return value from a successful match is 1, indicating
1879 nigel 91 that just the first pair of offsets has been set.
1880 nigel 41
1881 nigel 73 If a capturing subpattern is matched repeatedly, it is the last portion
1882 nigel 75 of the string that it matched that is returned.
1883 nigel 41
1884 nigel 93 If the vector is too small to hold all the captured substring offsets,
1885 nigel 75 it is used as far as possible (up to two-thirds of its length), and the
1886 nigel 93 function returns a value of zero. In particular, if the substring off-
1887 nigel 75 sets are not of interest, pcre_exec() may be called with ovector passed
1888 nigel 93 as NULL and ovecsize as zero. However, if the pattern contains back
1889     references and the ovector is not big enough to remember the related
1890     substrings, PCRE has to get additional memory for use during matching.
1891 nigel 73 Thus it is usually advisable to supply an ovector.
1892 nigel 41
1893 nigel 93 The pcre_info() function can be used to find out how many capturing
1894     subpatterns there are in a compiled pattern. The smallest size for
1895     ovector that will allow for n captured substrings, in addition to the
1896 nigel 91 offsets of the substring matched by the whole pattern, is (n+1)*3.
1897 nigel 41
1898 nigel 93 It is possible for capturing subpattern number n+1 to match some part
1899 nigel 91 of the subject when subpattern n has not been used at all. For example,
1900 nigel 93 if the string "abc" is matched against the pattern (a|(z))(bc) the
1901 nigel 91 return from the function is 4, and subpatterns 1 and 3 are matched, but
1902 nigel 93 2 is not. When this happens, both values in the offset pairs corre-
1903 nigel 91 sponding to unused subpatterns are set to -1.
1904 nigel 75
1905 nigel 93 Offset values that correspond to unused subpatterns at the end of the
1906     expression are also set to -1. For example, if the string "abc" is
1907     matched against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not
1908     matched. The return from the function is 2, because the highest used
1909 nigel 91 capturing subpattern number is 1. However, you can refer to the offsets
1910 nigel 93 for the second and third capturing subpatterns if you wish (assuming
1911 nigel 91 the vector is large enough, of course).
1912    
1913 nigel 93 Some convenience functions are provided for extracting the captured
1914 nigel 91 substrings as separate strings. These are described below.
1915    
1916     Error return values from pcre_exec()
1917    
1918 nigel 93 If pcre_exec() fails, it returns a negative number. The following are
1919 nigel 73 defined in the header file:
1920 nigel 41
1921 nigel 73 PCRE_ERROR_NOMATCH (-1)
1922 nigel 41
1923 nigel 73 The subject string did not match the pattern.
1924 nigel 41
1925 nigel 73 PCRE_ERROR_NULL (-2)
1926 nigel 41
1927 nigel 93 Either code or subject was passed as NULL, or ovector was NULL and
1928 nigel 73 ovecsize was not zero.
1929 nigel 41
1930 nigel 73 PCRE_ERROR_BADOPTION (-3)
1931 nigel 41
1932 nigel 73 An unrecognized bit was set in the options argument.
1933 nigel 41
1934 nigel 73 PCRE_ERROR_BADMAGIC (-4)
1935 nigel 41
1936 nigel 93 PCRE stores a 4-byte "magic number" at the start of the compiled code,
1937 nigel 75 to catch the case when it is passed a junk pointer and to detect when a
1938     pattern that was compiled in an environment of one endianness is run in
1939 nigel 93 an environment with the other endianness. This is the error that PCRE
1940 nigel 75 gives when the magic number is not present.
1941 nigel 41
1942 nigel 93 PCRE_ERROR_UNKNOWN_OPCODE (-5)
1943 nigel 41
1944 nigel 73 While running the pattern match, an unknown item was encountered in the
1945 nigel 93 compiled pattern. This error could be caused by a bug in PCRE or by
1946 nigel 73 overwriting of the compiled pattern.
1947 nigel 41
1948 nigel 73 PCRE_ERROR_NOMEMORY (-6)
1949 nigel 41
1950 nigel 93 If a pattern contains back references, but the ovector that is passed
1951 nigel 73 to pcre_exec() is not big enough to remember the referenced substrings,
1952 nigel 93 PCRE gets a block of memory at the start of matching to use for this
1953     purpose. If the call via pcre_malloc() fails, this error is given. The
1954 nigel 75 memory is automatically freed at the end of matching.
1955 nigel 41
1956 nigel 73 PCRE_ERROR_NOSUBSTRING (-7)
1957 nigel 53
1958 nigel 93 This error is used by the pcre_copy_substring(), pcre_get_substring(),
1959 nigel 73 and pcre_get_substring_list() functions (see below). It is never
1960     returned by pcre_exec().
1961 nigel 63
1962 nigel 73 PCRE_ERROR_MATCHLIMIT (-8)
1963 nigel 63
1964 nigel 93 The backtracking limit, as specified by the match_limit field in a
1965     pcre_extra structure (or defaulted) was reached. See the description
1966 nigel 87 above.
1967    
1968 nigel 73 PCRE_ERROR_CALLOUT (-9)
1969 nigel 63
1970 nigel 73 This error is never generated by pcre_exec() itself. It is provided for
1971 nigel 93 use by callout functions that want to yield a distinctive error code.
1972 nigel 73 See the pcrecallout documentation for details.
1973 nigel 71
1974 nigel 73 PCRE_ERROR_BADUTF8 (-10)
1975 nigel 71
1976 nigel 93 A string that contains an invalid UTF-8 byte sequence was passed as a
1977 nigel 73 subject.
1978    
1979     PCRE_ERROR_BADUTF8_OFFSET (-11)
1980    
1981     The UTF-8 byte sequence that was passed as a subject was valid, but the
1982 nigel 93 value of startoffset did not point to the beginning of a UTF-8 charac-
1983 nigel 73 ter.
1984    
1985 nigel 77 PCRE_ERROR_PARTIAL (-12)
1986 nigel 73
1987 nigel 93 The subject string did not match, but it did match partially. See the
1988 nigel 75 pcrepartial documentation for details of partial matching.
1989    
1990 nigel 77 PCRE_ERROR_BADPARTIAL (-13)
1991 nigel 75
1992 nigel 93 The PCRE_PARTIAL option was used with a compiled pattern containing
1993     items that are not supported for partial matching. See the pcrepartial
1994 nigel 75 documentation for details of partial matching.
1995    
1996 nigel 77 PCRE_ERROR_INTERNAL (-14)
1997 nigel 75
1998 nigel 93 An unexpected internal error has occurred. This error could be caused
1999 nigel 75 by a bug in PCRE or by overwriting of the compiled pattern.
2000    
2001 nigel 77 PCRE_ERROR_BADCOUNT (-15)
2002 nigel 75
2003 nigel 93 This error is given if the value of the ovecsize argument is negative.
2004 nigel 75
2005 nigel 93 PCRE_ERROR_RECURSIONLIMIT (-21)
2006 nigel 75
2007 nigel 93 The internal recursion limit, as specified by the match_limit_recursion
2008     field in a pcre_extra structure (or defaulted) was reached. See the
2009     description above.
2010    
2011     PCRE_ERROR_NULLWSLIMIT (-22)
2012    
2013     When a group that can match an empty substring is repeated with an
2014     unbounded upper limit, the subject position at the start of the group
2015     must be remembered, so that a test for an empty string can be made when
2016     the end of the group is reached. Some workspace is required for this;
2017     if it runs out, this error is given.
2018    
2019     PCRE_ERROR_BADNEWLINE (-23)
2020    
2021     An invalid combination of PCRE_NEWLINE_xxx options was given.
2022    
2023     Error numbers -16 to -20 are not used by pcre_exec().
2024    
2025    
2026 nigel 63 EXTRACTING CAPTURED SUBSTRINGS BY NUMBER
2027    
2028 nigel 73 int pcre_copy_substring(const char *subject, int *ovector,
2029     int stringcount, int stringnumber, char *buffer,
2030     int buffersize);
2031 nigel 63
2032 nigel 73 int pcre_get_substring(const char *subject, int *ovector,
2033     int stringcount, int stringnumber,
2034     const char **stringptr);
2035 nigel 63
2036 nigel 73 int pcre_get_substring_list(const char *subject,
2037     int *ovector, int stringcount, const char ***listptr);
2038 nigel 63
2039 nigel 93 Captured substrings can be accessed directly by using the offsets
2040     returned by pcre_exec() in ovector. For convenience, the functions
2041 nigel 73 pcre_copy_substring(), pcre_get_substring(), and pcre_get_sub-
2042 nigel 93 string_list() are provided for extracting captured substrings as new,
2043     separate, zero-terminated strings. These functions identify substrings
2044     by number. The next section describes functions for extracting named
2045 nigel 91 substrings.
2046 nigel 41
2047 nigel 93 A substring that contains a binary zero is correctly extracted and has
2048     a further zero added on the end, but the result is not, of course, a C
2049     string. However, you can process such a string by referring to the
2050     length that is returned by pcre_copy_substring() and pcre_get_sub-
2051 nigel 91 string(). Unfortunately, the interface to pcre_get_substring_list() is
2052 nigel 93 not adequate for handling strings containing binary zeros, because the
2053 nigel 91 end of the final string is not independently indicated.
2054    
2055 nigel 93 The first three arguments are the same for all three of these func-
2056     tions: subject is the subject string that has just been successfully
2057 nigel 73 matched, ovector is a pointer to the vector of integer offsets that was
2058     passed to pcre_exec(), and stringcount is the number of substrings that
2059 nigel 93 were captured by the match, including the substring that matched the
2060 nigel 75 entire regular expression. This is the value returned by pcre_exec() if
2061 nigel 93 it is greater than zero. If pcre_exec() returned zero, indicating that
2062     it ran out of space in ovector, the value passed as stringcount should
2063 nigel 75 be the number of elements in the vector divided by three.
2064 nigel 41
2065 nigel 93 The functions pcre_copy_substring() and pcre_get_substring() extract a
2066     single substring, whose number is given as stringnumber. A value of
2067     zero extracts the substring that matched the entire pattern, whereas
2068     higher values extract the captured substrings. For pcre_copy_sub-
2069     string(), the string is placed in buffer, whose length is given by
2070     buffersize, while for pcre_get_substring() a new block of memory is
2071     obtained via pcre_malloc, and its address is returned via stringptr.
2072     The yield of the function is the length of the string, not including
2073     the terminating zero, or one of these error codes:
2074 nigel 41
2075 nigel 73 PCRE_ERROR_NOMEMORY (-6)
2076 nigel 41
2077 nigel 93 The buffer was too small for pcre_copy_substring(), or the attempt to
2078 nigel 73 get memory failed for pcre_get_substring().
2079 nigel 41
2080 nigel 73 PCRE_ERROR_NOSUBSTRING (-7)
2081 nigel 41
2082 nigel 73 There is no substring whose number is stringnumber.
2083 nigel 41
2084 nigel 93 The pcre_get_substring_list() function extracts all available sub-
2085     strings and builds a list of pointers to them. All this is done in a
2086 nigel 75 single block of memory that is obtained via pcre_malloc. The address of
2087 nigel 93 the memory block is returned via listptr, which is also the start of
2088     the list of string pointers. The end of the list is marked by a NULL
2089     pointer. The yield of the function is zero if all went well, or the
2090     error code
2091 nigel 41
2092 nigel 73 PCRE_ERROR_NOMEMORY (-6)
2093 nigel 41
2094 nigel 73 if the attempt to get the memory block failed.
2095 nigel 41
2096 nigel 91 When any of these functions encounter a substring that is unset, which
2097     can happen when capturing subpattern number n+1 matches some part of
2098     the subject, but subpattern n has not been used at all, they return an
2099 nigel 73 empty string. This can be distinguished from a genuine zero-length sub-
2100 nigel 91 string by inspecting the appropriate offset in ovector, which is nega-
2101 nigel 73 tive for unset substrings.
2102 nigel 41
2103 nigel 91 The two convenience functions pcre_free_substring() and pcre_free_sub-
2104     string_list() can be used to free the memory returned by a previous
2105 nigel 75 call of pcre_get_substring() or pcre_get_substring_list(), respec-
2106 nigel 91 tively. They do nothing more than call the function pointed to by
2107     pcre_free, which of course could be called directly from a C program.
2108     However, PCRE is used in some situations where it is linked via a spe-
2109     cial interface to another programming language that cannot use
2110     pcre_free directly; it is for these cases that the functions are pro-
2111 nigel 77 vided.
2112 nigel 41
2113 nigel 73
2114 nigel 63 EXTRACTING CAPTURED SUBSTRINGS BY NAME
2115 nigel 41
2116 nigel 75 int pcre_get_stringnumber(const pcre *code,
2117     const char *name);
2118    
2119 nigel 73 int pcre_copy_named_substring(const pcre *code,
2120     const char *subject, int *ovector,
2121     int stringcount, const char *stringname,
2122     char *buffer, int buffersize);
2123 nigel 41
2124 nigel 73 int pcre_get_named_substring(const pcre *code,
2125     const char *subject, int *ovector,
2126     int stringcount, const char *stringname,
2127     const char **stringptr);
2128 nigel 41
2129 nigel 91 To extract a substring by name, you first have to find associated num-
2130 nigel 75 ber. For example, for this pattern
2131 nigel 41
2132 nigel 93 (a+)b(?<xxx>\d+)...
2133 nigel 63
2134 nigel 91 the number of the subpattern called "xxx" is 2. If the name is known to
2135     be unique (PCRE_DUPNAMES was not set), you can find the number from the
2136     name by calling pcre_get_stringnumber(). The first argument is the com-
2137     piled pattern, and the second is the name. The yield of the function is
2138     the subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no
2139     subpattern of that name.
2140 nigel 63
2141 nigel 75 Given the number, you can extract the substring directly, or use one of
2142     the functions described in the previous section. For convenience, there
2143     are also two functions that do the whole job.
2144    
2145 nigel 87 Most of the arguments of pcre_copy_named_substring() and
2146     pcre_get_named_substring() are the same as those for the similarly
2147     named functions that extract by number. As these are described in the
2148     previous section, they are not re-described here. There are just two
2149 nigel 75 differences:
2150 nigel 63
2151 nigel 87 First, instead of a substring number, a substring name is given. Sec-
2152 nigel 73 ond, there is an extra argument, given at the start, which is a pointer
2153 nigel 87 to the compiled pattern. This is needed in order to gain access to the
2154 nigel 73 name-to-number translation table.
2155 nigel 63
2156 nigel 87 These functions call pcre_get_stringnumber(), and if it succeeds, they
2157     then call pcre_copy_substring() or pcre_get_substring(), as appropri-
2158 ph10 128 ate. NOTE: If PCRE_DUPNAMES is set and there are duplicate names, the
2159     behaviour may not be what you want (see the next section).
2160 nigel 63
2161 nigel 77
2162 nigel 91 DUPLICATE SUBPATTERN NAMES
2163    
2164     int pcre_get_stringtable_entries(const pcre *code,
2165     const char *name, char **first, char **last);
2166    
2167 ph10 128 When a pattern is compiled with the PCRE_DUPNAMES option, names for
2168     subpatterns are not required to be unique. Normally, patterns with
2169     duplicate names are such that in any one match, only one of the named
2170     subpatterns participates. An example is shown in the pcrepattern docu-
2171 nigel 91 mentation. When duplicates are present, pcre_copy_named_substring() and
2172 ph10 128 pcre_get_named_substring() return the first substring corresponding to
2173     the given name that is set. If none are set, an empty string is
2174 nigel 91 returned. The pcre_get_stringnumber() function returns one of the num-
2175 ph10 128 bers that are associated with the name, but it is not defined which it
2176 nigel 91 is.
2177    
2178 ph10 128 If you want to get full details of all captured substrings for a given
2179     name, you must use the pcre_get_stringtable_entries() function. The
2180 nigel 91 first argument is the compiled pattern, and the second is the name. The
2181 ph10 128 third and fourth are pointers to variables which are updated by the
2182 nigel 91 function. After it has run, they point to the first and last entries in
2183 ph10 128 the name-to-number table for the given name. The function itself
2184     returns the length of each entry, or PCRE_ERROR_NOSUBSTRING (-7) if
2185     there are none. The format of the table is described above in the sec-
2186     tion entitled Information about a pattern. Given all the relevant
2187     entries for the name, you can extract each of their numbers, and hence
2188 nigel 93 the captured data, if any.
2189 nigel 91
2190    
2191 nigel 77 FINDING ALL POSSIBLE MATCHES
2192    
2193 ph10 128 The traditional matching function uses a similar algorithm to Perl,
2194 nigel 77 which stops when it finds the first match, starting at a given point in
2195 ph10 128 the subject. If you want to find all possible matches, or the longest
2196     possible match, consider using the alternative matching function (see
2197     below) instead. If you cannot use the alternative function, but still
2198     need to find all possible matches, you can kludge it up by making use
2199 nigel 77 of the callout facility, which is described in the pcrecallout documen-
2200     tation.
2201    
2202     What you have to do is to insert a callout right at the end of the pat-
2203 ph10 128 tern. When your callout function is called, extract and save the cur-
2204     rent matched substring. Then return 1, which forces pcre_exec() to
2205     backtrack and try other alternatives. Ultimately, when it runs out of
2206 nigel 77 matches, pcre_exec() will yield PCRE_ERROR_NOMATCH.
2207    
2208    
2209     MATCHING A PATTERN: THE ALTERNATIVE FUNCTION
2210    
2211     int pcre_dfa_exec(const pcre *code, const pcre_extra *extra,
2212     const char *subject, int length, int startoffset,
2213     int options, int *ovector, int ovecsize,
2214     int *workspace, int wscount);
2215    
2216 ph10 128 The function pcre_dfa_exec() is called to match a subject string
2217     against a compiled pattern, using a matching algorithm that scans the
2218     subject string just once, and does not backtrack. This has different
2219     characteristics to the normal algorithm, and is not compatible with
2220     Perl. Some of the features of PCRE patterns are not supported. Never-
2221     theless, there are times when this kind of matching can be useful. For
2222 nigel 93 a discussion of the two matching algorithms, see the pcrematching docu-
2223     mentation.
2224 nigel 77
2225 ph10 128 The arguments for the pcre_dfa_exec() function are the same as for
2226 nigel 77 pcre_exec(), plus two extras. The ovector argument is used in a differ-
2227 ph10 128 ent way, and this is described below. The other common arguments are
2228     used in the same way as for pcre_exec(), so their description is not
2229 nigel 77 repeated here.
2230    
2231 ph10 128 The two additional arguments provide workspace for the function. The
2232     workspace vector should contain at least 20 elements. It is used for
2233 nigel 77 keeping track of multiple paths through the pattern tree. More
2234 ph10 128 workspace will be needed for patterns and subjects where there are a
2235 nigel 91 lot of potential matches.
2236 nigel 77
2237 nigel 87 Here is an example of a simple call to pcre_dfa_exec():
2238 nigel 77
2239     int rc;
2240     int ovector[10];
2241     int wspace[20];
2242 nigel 87 rc = pcre_dfa_exec(
2243 nigel 77 re, /* result of pcre_compile() */
2244     NULL, /* we didn't study the pattern */
2245     "some string", /* the subject string */
2246     11, /* the length of the subject string */
2247     0, /* start at offset 0 in the subject */
2248     0, /* default options */
2249     ovector, /* vector of integers for substring information */
2250     10, /* number of elements (NOT size in bytes) */
2251     wspace, /* working space vector */
2252     20); /* number of elements (NOT size in bytes) */
2253    
2254     Option bits for pcre_dfa_exec()
2255    
2256 ph10 128 The unused bits of the options argument for pcre_dfa_exec() must be
2257     zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEW-
2258     LINE_xxx, PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NO_UTF8_CHECK,
2259 nigel 91 PCRE_PARTIAL, PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last
2260     three of these are the same as for pcre_exec(), so their description is
2261     not repeated here.
2262 nigel 77
2263     PCRE_PARTIAL
2264    
2265 ph10 128 This has the same general effect as it does for pcre_exec(), but the
2266     details are slightly different. When PCRE_PARTIAL is set for
2267     pcre_dfa_exec(), the return code PCRE_ERROR_NOMATCH is converted into
2268     PCRE_ERROR_PARTIAL if the end of the subject is reached, there have
2269 nigel 77 been no complete matches, but there is still at least one matching pos-
2270 ph10 128 sibility. The portion of the string that provided the partial match is
2271 nigel 77 set as the first matching string.
2272    
2273     PCRE_DFA_SHORTEST
2274    
2275 ph10 128 Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to
2276 nigel 93 stop as soon as it has found one match. Because of the way the alterna-
2277 ph10 128 tive algorithm works, this is necessarily the shortest possible match
2278 nigel 93 at the first possible matching point in the subject string.
2279 nigel 77
2280     PCRE_DFA_RESTART
2281    
2282 ph10 128 When pcre_dfa_exec() is called with the PCRE_PARTIAL option, and
2283     returns a partial match, it is possible to call it again, with addi-
2284     tional subject characters, and have it continue with the same match.
2285     The PCRE_DFA_RESTART option requests this action; when it is set, the
2286     workspace and wscount options must reference the same vector as before
2287     because data about the match so far is left in them after a partial
2288     match. There is more discussion of this facility in the pcrepartial
2289 nigel 77 documentation.
2290    
2291     Successful returns from pcre_dfa_exec()
2292    
2293 ph10 128 When pcre_dfa_exec() succeeds, it may have matched more than one sub-
2294 nigel 77 string in the subject. Note, however, that all the matches from one run
2295 ph10 128 of the function start at the same point in the subject. The shorter
2296     matches are all initial substrings of the longer matches. For example,
2297 nigel 77 if the pattern
2298    
2299     <.*>
2300    
2301     is matched against the string
2302    
2303     This is <something> <something else> <something further> no more
2304    
2305     the three matched strings are
2306    
2307     <something>
2308     <something> <something else>
2309     <something> <something else> <something further>
2310    
2311 ph10 128 On success, the yield of the function is a number greater than zero,
2312     which is the number of matched substrings. The substrings themselves
2313     are returned in ovector. Each string uses two elements; the first is
2314     the offset to the start, and the second is the offset to the end. In
2315     fact, all the strings have the same start offset. (Space could have
2316     been saved by giving this only once, but it was decided to retain some
2317     compatibility with the way pcre_exec() returns data, even though the
2318 nigel 93 meaning of the strings is different.)
2319 nigel 77
2320     The strings are returned in reverse order of length; that is, the long-
2321 ph10 128 est matching string is given first. If there were too many matches to
2322     fit into ovector, the yield of the function is zero, and the vector is
2323 nigel 77 filled with the longest matches.
2324    
2325     Error returns from pcre_dfa_exec()
2326    
2327 ph10 128 The pcre_dfa_exec() function returns a negative number when it fails.
2328     Many of the errors are the same as for pcre_exec(), and these are
2329     described above. There are in addition the following errors that are
2330 nigel 77 specific to pcre_dfa_exec():
2331    
2332     PCRE_ERROR_DFA_UITEM (-16)
2333    
2334 ph10 128 This return is given if pcre_dfa_exec() encounters an item in the pat-
2335     tern that it does not support, for instance, the use of \C or a back
2336 nigel 77 reference.
2337    
2338     PCRE_ERROR_DFA_UCOND (-17)
2339    
2340 ph10 128 This return is given if pcre_dfa_exec() encounters a condition item
2341     that uses a back reference for the condition, or a test for recursion
2342 nigel 93 in a specific group. These are not supported.
2343 nigel 77
2344     PCRE_ERROR_DFA_UMLIMIT (-18)
2345    
2346 ph10 128 This return is given if pcre_dfa_exec() is called with an extra block
2347 nigel 77 that contains a setting of the match_limit field. This is not supported
2348     (it is meaningless).
2349    
2350     PCRE_ERROR_DFA_WSSIZE (-19)
2351    
2352 ph10 128 This return is given if pcre_dfa_exec() runs out of space in the
2353 nigel 77 workspace vector.
2354    
2355     PCRE_ERROR_DFA_RECURSE (-20)
2356    
2357 ph10 128 When a recursive subpattern is processed, the matching function calls
2358     itself recursively, using private vectors for ovector and workspace.
2359     This error is given if the output vector is not large enough. This
2360 nigel 77 should be extremely rare, as a vector of size 1000 is used.
2361    
2362 nigel 93
2363     SEE ALSO
2364    
2365 ph10 128 pcrebuild(3), pcrecallout(3), pcrecpp(3)(3), pcrematching(3), pcrepar-
2366     tial(3), pcreposix(3), pcreprecompile(3), pcresample(3), pcrestack(3).
2367 nigel 93
2368 nigel 63
2369 ph10 99 AUTHOR
2370 nigel 63
2371 ph10 99 Philip Hazel
2372     University Computing Service
2373     Cambridge CB2 3QH, England.
2374    
2375    
2376     REVISION
2377    
2378     Last updated: 06 March 2007
2379     Copyright (c) 1997-2007 University of Cambridge.
2380     ------------------------------------------------------------------------------
2381 ph10 111
2382    
2383 nigel 79 PCRECALLOUT(3) PCRECALLOUT(3)
2384 nigel 63
2385 nigel 79
2386 nigel 73 NAME
2387     PCRE - Perl-compatible regular expressions
2388    
2389 nigel 77
2390 nigel 63 PCRE CALLOUTS
2391    
2392 nigel 73 int (*pcre_callout)(pcre_callout_block *);
2393 nigel 63
2394 nigel 73 PCRE provides a feature called "callout", which is a means of temporar-
2395     ily passing control to the caller of PCRE in the middle of pattern
2396     matching. The caller of PCRE provides an external function by putting
2397     its entry point in the global variable pcre_callout. By default, this
2398     variable contains NULL, which disables all calling out.
2399 nigel 63
2400 nigel 73 Within a regular expression, (?C) indicates the points at which the
2401     external function is to be called. Different callout points can be
2402     identified by putting a number less than 256 after the letter C. The
2403     default value is zero. For example, this pattern has two callout
2404     points:
2405 nigel 63
2406 nigel 75 (?C1)eabc(?C2)def
2407 nigel 63
2408 nigel 75 If the PCRE_AUTO_CALLOUT option bit is set when pcre_compile() is
2409     called, PCRE automatically inserts callouts, all with number 255,
2410     before each item in the pattern. For example, if PCRE_AUTO_CALLOUT is
2411     used with the pattern
2412 nigel 63
2413 nigel 75 A(\d{2}|--)
2414    
2415     it is processed as if it were
2416    
2417     (?C255)A(?C255)((?C255)\d{2}(?C255)|(?C255)-(?C255)-(?C255))(?C255)
2418    
2419     Notice that there is a callout before and after each parenthesis and
2420     alternation bar. Automatic callouts can be used for tracking the
2421     progress of pattern matching. The pcretest command has an option that
2422     sets automatic callouts; when it is used, the output indicates how the
2423     pattern is matched. This is useful information when you are trying to
2424     optimize the performance of a particular pattern.
2425    
2426    
2427     MISSING CALLOUTS
2428    
2429     You should be aware that, because of optimizations in the way PCRE
2430     matches patterns, callouts sometimes do not happen. For example, if the
2431     pattern is
2432    
2433     ab(?C4)cd
2434    
2435     PCRE knows that any matching string must contain the letter "d". If the
2436     subject string is "abyz", the lack of "d" means that matching doesn't
2437     ever start, and the callout is never reached. However, with "abyd",
2438     though the result is still no match, the callout is obeyed.
2439    
2440    
2441     THE CALLOUT INTERFACE
2442    
2443     During matching, when PCRE reaches a callout point, the external func-
2444 nigel 77 tion defined by pcre_callout is called (if it is set). This applies to
2445     both the pcre_exec() and the pcre_dfa_exec() matching functions. The
2446     only argument to the callout function is a pointer to a pcre_callout
2447     block. This structure contains the following fields:
2448 nigel 75
2449 nigel 73 int version;
2450     int callout_number;
2451     int *offset_vector;
2452     const char *subject;
2453     int subject_length;
2454     int start_match;
2455     int current_position;
2456     int capture_top;
2457     int capture_last;
2458     void *callout_data;
2459 nigel 75 int pattern_position;
2460     int next_item_length;
2461 nigel 63
2462 nigel 77 The version field is an integer containing the version number of the
2463     block format. The initial version was 0; the current version is 1. The
2464     version number will change again in future if additional fields are
2465 nigel 75 added, but the intention is never to remove any of the existing fields.
2466 nigel 63
2467 nigel 77 The callout_number field contains the number of the callout, as com-
2468     piled into the pattern (that is, the number after ?C for manual call-
2469 nigel 75 outs, and 255 for automatically generated callouts).
2470 nigel 63
2471 nigel 77 The offset_vector field is a pointer to the vector of offsets that was
2472     passed by the caller to pcre_exec() or pcre_dfa_exec(). When
2473     pcre_exec() is used, the contents can be inspected in order to extract
2474     substrings that have been matched so far, in the same way as for
2475     extracting substrings after a match has completed. For pcre_dfa_exec()
2476     this field is not useful.
2477 nigel 63
2478 nigel 75 The subject and subject_length fields contain copies of the values that
2479 nigel 73 were passed to pcre_exec().
2480 nigel 63
2481 nigel 77 The start_match field contains the offset within the subject at which
2482     the current match attempt started. If the pattern is not anchored, the
2483 nigel 75 callout function may be called several times from the same point in the
2484     pattern for different starting points in the subject.
2485 nigel 63
2486 nigel 77 The current_position field contains the offset within the subject of
2487 nigel 73 the current match pointer.
2488 nigel 63
2489 nigel 77 When the pcre_exec() function is used, the capture_top field contains
2490     one more than the number of the highest numbered captured substring so
2491     far. If no substrings have been captured, the value of capture_top is
2492     one. This is always the case when pcre_dfa_exec() is used, because it
2493     does not support captured substrings.
2494 nigel 63
2495 nigel 77 The capture_last field contains the number of the most recently cap-
2496     tured substring. If no substrings have been captured, its value is -1.
2497     This is always the case when pcre_dfa_exec() is used.
2498 nigel 63
2499 nigel 77 The callout_data field contains a value that is passed to pcre_exec()
2500     or pcre_dfa_exec() specifically so that it can be passed back in call-
2501     outs. It is passed in the pcre_callout field of the pcre_extra data
2502     structure. If no such data was passed, the value of callout_data in a
2503     pcre_callout block is NULL. There is a description of the pcre_extra
2504 nigel 73 structure in the pcreapi documentation.
2505 nigel 63
2506 nigel 77 The pattern_position field is present from version 1 of the pcre_call-
2507 nigel 75 out structure. It contains the offset to the next item to be matched in
2508     the pattern string.
2509 nigel 63
2510 nigel 77 The next_item_length field is present from version 1 of the pcre_call-
2511 nigel 75 out structure. It contains the length of the next item to be matched in
2512 nigel 77 the pattern string. When the callout immediately precedes an alterna-
2513     tion bar, a closing parenthesis, or the end of the pattern, the length
2514     is zero. When the callout precedes an opening parenthesis, the length
2515 nigel 75 is that of the entire subpattern.
2516 nigel 73
2517 nigel 77 The pattern_position and next_item_length fields are intended to help
2518     in distinguishing between different automatic callouts, which all have
2519 nigel 75 the same callout number. However, they are set for all callouts.
2520    
2521    
2522 nigel 63 RETURN VALUES
2523    
2524 nigel 77 The external callout function returns an integer to PCRE. If the value
2525     is zero, matching proceeds as normal. If the value is greater than
2526     zero, matching fails at the current point, but the testing of other
2527     matching possibilities goes ahead, just as if a lookahead assertion had
2528     failed. If the value is less than zero, the match is abandoned, and
2529     pcre_exec() (or pcre_dfa_exec()) returns the negative value.
2530 nigel 63
2531 nigel 77 Negative values should normally be chosen from the set of
2532 nigel 73 PCRE_ERROR_xxx values. In particular, PCRE_ERROR_NOMATCH forces a stan-
2533 nigel 77 dard "no match" failure. The error number PCRE_ERROR_CALLOUT is
2534     reserved for use by callout functions; it will never be used by PCRE
2535 nigel 73 itself.
2536 nigel 63
2537    
2538 ph10 99 AUTHOR
2539 nigel 63
2540 ph10 99 Philip Hazel
2541     University Computing Service
2542     Cambridge CB2 3QH, England.
2543    
2544    
2545     REVISION
2546    
2547     Last updated: 06 March 2007
2548     Copyright (c) 1997-2007 University of Cambridge.
2549     ------------------------------------------------------------------------------
2550 ph10 111
2551    
2552 nigel 79 PCRECOMPAT(3) PCRECOMPAT(3)
2553 nigel 63
2554 nigel 79
2555 nigel 73 NAME
2556     PCRE - Perl-compatible regular expressions
2557    
2558 nigel 77
2559 nigel 75 DIFFERENCES BETWEEN PCRE AND PERL
2560 nigel 41
2561 nigel 73 This document describes the differences in the ways that PCRE and Perl
2562 nigel 93 handle regular expressions. The differences described here are mainly
2563     with respect to Perl 5.8, though PCRE version 7.0 contains some fea-
2564     tures that are expected to be in the forthcoming Perl 5.10.
2565 nigel 41
2566 nigel 93 1. PCRE has only a subset of Perl's UTF-8 and Unicode support. Details
2567     of what it does have are given in the section on UTF-8 support in the
2568 nigel 87 main pcre page.
2569 nigel 41
2570 nigel 73 2. PCRE does not allow repeat quantifiers on lookahead assertions. Perl
2571 nigel 93 permits them, but they do not mean what you might think. For example,
2572 nigel 73 (?!a){3} does not assert that the next three characters are not "a". It
2573     just asserts that the next character is not "a" three times.
2574 nigel 41
2575 nigel 93 3. Capturing subpatterns that occur inside negative lookahead asser-
2576     tions are counted, but their entries in the offsets vector are never
2577     set. Perl sets its numerical variables from any such patterns that are
2578 nigel 73 matched before the assertion fails to match something (thereby succeed-
2579 nigel 93 ing), but only if the negative lookahead assertion contains just one
2580 nigel 73 branch.
2581 nigel 41
2582 nigel 93 4. Though binary zero characters are supported in the subject string,
2583 nigel 73 they are not allowed in a pattern string because it is passed as a nor-
2584 nigel 75 mal C string, terminated by zero. The escape sequence \0 can be used in
2585     the pattern to represent a binary zero.
2586 nigel 41
2587 nigel 93 5. The following Perl escape sequences are not supported: \l, \u, \L,
2588 nigel 75 \U, and \N. In fact these are implemented by Perl's general string-han-
2589 nigel 93 dling and are not part of its pattern matching engine. If any of these
2590 nigel 75 are encountered by PCRE, an error is generated.
2591 nigel 41
2592 nigel 93 6. The Perl escape sequences \p, \P, and \X are supported only if PCRE
2593     is built with Unicode character property support. The properties that
2594     can be tested with \p and \P are limited to the general category prop-
2595     erties such as Lu and Nd, script names such as Greek or Han, and the
2596 nigel 87 derived properties Any and L&.
2597 nigel 75
2598     7. PCRE does support the \Q...\E escape for quoting substrings. Charac-
2599 nigel 93 ters in between are treated as literals. This is slightly different
2600     from Perl in that $ and @ are also handled as literals inside the
2601     quotes. In Perl, they cause variable interpolation (but of course PCRE
2602 nigel 73 does not have variables). Note the following examples:
2603 nigel 49
2604 nigel 73 Pattern PCRE matches Perl matches
2605 nigel 41
2606 nigel 73 \Qabc$xyz\E abc$xyz abc followed by the
2607     contents of $xyz
2608     \Qabc\$xyz\E abc\$xyz abc\$xyz
2609     \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2610 nigel 41
2611 nigel 93 The \Q...\E sequence is recognized both inside and outside character
2612 nigel 73 classes.
2613 nigel 41
2614 nigel 93 8. Fairly obviously, PCRE does not support the (?{code}) and (??{code})
2615     constructions. However, there is support for recursive patterns. This
2616     is not available in Perl 5.8, but will be in Perl 5.10. Also, the PCRE
2617     "callout" feature allows an external function to be called during pat-
2618 nigel 75 tern matching. See the pcrecallout documentation for details.
2619 nigel 63
2620 nigel 93 9. Subpatterns that are called recursively or as "subroutines" are
2621     always treated as atomic groups in PCRE. This is like Python, but
2622     unlike Perl.
2623    
2624     10. There are some differences that are concerned with the settings of
2625     captured strings when part of a pattern is repeated. For example,
2626     matching "aba" against the pattern /^(a(b)?)+$/ in Perl leaves $2
2627 nigel 73 unset, but in PCRE it is set to "b".
2628 nigel 41
2629 nigel 93 11. PCRE provides some extensions to the Perl regular expression facil-
2630     ities. Perl 5.10 will include new features that are not in earlier
2631     versions, some of which (such as named parentheses) have been in PCRE
2632     for some time. This list is with respect to Perl 5.10:
2633 nigel 41
2634 nigel 93 (a) Although lookbehind assertions must match fixed length strings,
2635 nigel 73 each alternative branch of a lookbehind assertion can match a different
2636     length of string. Perl requires them all to have the same length.
2637 nigel 41
2638 nigel 93 (b) If PCRE_DOLLAR_ENDONLY is set and PCRE_MULTILINE is not set, the $
2639 nigel 73 meta-character matches only at the very end of the string.
2640 nigel 41
2641 nigel 73 (c) If PCRE_EXTRA is set, a backslash followed by a letter with no spe-
2642 nigel 93 cial meaning is faulted. Otherwise, like Perl, the backslash is
2643 nigel 91 ignored. (Perl can be made to issue a warning.)
2644 nigel 41
2645 nigel 93 (d) If PCRE_UNGREEDY is set, the greediness of the repetition quanti-
2646 nigel 73 fiers is inverted, that is, by default they are not greedy, but if fol-
2647     lowed by a question mark they are.
2648 nigel 41
2649 nigel 75 (e) PCRE_ANCHORED can be used at matching time to force a pattern to be
2650     tried only at the first matching position in the subject string.
2651 nigel 41
2652 nigel 93 (f) The PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, and PCRE_NO_AUTO_CAP-
2653 nigel 73 TURE options for pcre_exec() have no Perl equivalents.
2654 nigel 41
2655 nigel 93 (g) The callout facility is PCRE-specific.
2656 nigel 41
2657 nigel 93 (h) The partial matching facility is PCRE-specific.
2658 nigel 43
2659 nigel 93 (i) Patterns compiled by PCRE can be saved and re-used at a later time,
2660 nigel 75 even on different hosts that have the other endianness.
2661    
2662 nigel 93 (j) The alternative matching function (pcre_dfa_exec()) matches in a
2663 nigel 77 different way and is not Perl-compatible.
2664    
2665 nigel 63
2666 ph10 99 AUTHOR
2667 nigel 63
2668 ph10 99 Philip Hazel
2669     University Computing Service
2670     Cambridge CB2 3QH, England.
2671    
2672    
2673     REVISION
2674    
2675     Last updated: 06 March 2007
2676     Copyright (c) 1997-2007 University of Cambridge.
2677     ------------------------------------------------------------------------------
2678 ph10 111
2679    
2680 nigel 79 PCREPATTERN(3) PCREPATTERN(3)
2681 nigel 63
2682 nigel 79
2683 nigel 73 NAME
2684     PCRE - Perl-compatible regular expressions
2685    
2686 nigel 77
2687 nigel 63 PCRE REGULAR EXPRESSION DETAILS
2688    
2689 nigel 73 The syntax and semantics of the regular expressions supported by PCRE
2690     are described below. Regular expressions are also described in the Perl
2691 nigel 75 documentation and in a number of books, some of which have copious
2692     examples. Jeffrey Friedl's "Mastering Regular Expressions", published
2693     by O'Reilly, covers regular expressions in great detail. This descrip-
2694     tion of PCRE's regular expressions is intended as reference material.
2695 nigel 49
2696 nigel 75 The original operation of PCRE was on strings of one-byte characters.
2697     However, there is now also support for UTF-8 character strings. To use
2698     this, you must build PCRE to include UTF-8 support, and then call
2699     pcre_compile() with the PCRE_UTF8 option. How this affects pattern
2700     matching is mentioned in several places below. There is also a summary
2701     of UTF-8 features in the section on UTF-8 support in the main pcre
2702     page.
2703 nigel 41
2704 nigel 77 The remainder of this document discusses the patterns that are sup-
2705     ported by PCRE when its main matching function, pcre_exec(), is used.
2706     From release 6.0, PCRE offers a second matching function,
2707     pcre_dfa_exec(), which matches using a different algorithm that is not
2708     Perl-compatible. The advantages and disadvantages of the alternative
2709     function, and how it differs from the normal function, are discussed in
2710     the pcrematching page.
2711    
2712 nigel 93
2713     CHARACTERS AND METACHARACTERS
2714    
2715 nigel 75 A regular expression is a pattern that is matched against a subject
2716     string from left to right. Most characters stand for themselves in a
2717     pattern, and match the corresponding characters in the subject. As a
2718 nigel 73 trivial example, the pattern
2719 nigel 41
2720 nigel 73 The quick brown fox
2721 nigel 41
2722 nigel 77 matches a portion of a subject string that is identical to itself. When
2723     caseless matching is specified (the PCRE_CASELESS option), letters are
2724     matched independently of case. In UTF-8 mode, PCRE always understands
2725     the concept of case for characters whose values are less than 128, so
2726     caseless matching is always possible. For characters with higher val-
2727     ues, the concept of case is supported if PCRE is compiled with Unicode
2728     property support, but not otherwise. If you want to use caseless
2729     matching for characters 128 and above, you must ensure that PCRE is
2730     compiled with Unicode property support as well as with UTF-8 support.
2731 nigel 41
2732 nigel 77 The power of regular expressions comes from the ability to include
2733     alternatives and repetitions in the pattern. These are encoded in the
2734     pattern by the use of metacharacters, which do not stand for themselves
2735     but instead are interpreted in some special way.
2736    
2737     There are two different sets of metacharacters: those that are recog-
2738     nized anywhere in the pattern except within square brackets, and those
2739 nigel 93 that are recognized within square brackets. Outside square brackets,
2740     the metacharacters are as follows:
2741 nigel 41
2742 nigel 73 \ general escape character with several uses
2743     ^ assert start of string (or line, in multiline mode)
2744     $ assert end of string (or line, in multiline mode)
2745     . match any character except newline (by default)
2746     [ start character class definition
2747     | start of alternative branch
2748     ( start subpattern
2749     ) end subpattern
2750     ? extends the meaning of (
2751     also 0 or 1 quantifier
2752     also quantifier minimizer
2753     * 0 or more quantifier
2754     + 1 or more quantifier
2755     also "possessive quantifier"
2756     { start min/max quantifier
2757 nigel 41
2758 nigel 77 Part of a pattern that is in square brackets is called a "character
2759 nigel 75 class". In a character class the only metacharacters are:
2760 nigel 41
2761 nigel 73 \ general escape character
2762     ^ negate the class, but only if the first character
2763     - indicates character range
2764     [ POSIX character class (only if followed by POSIX
2765     syntax)
2766     ] terminates the character class
2767 nigel 41
2768 nigel 77 The following sections describe the use of each of the metacharacters.
2769 nigel 41
2770    
2771 nigel 63 BACKSLASH
2772 nigel 41
2773 nigel 73 The backslash character has several uses. Firstly, if it is followed by
2774 nigel 77 a non-alphanumeric character, it takes away any special meaning that
2775     character may have. This use of backslash as an escape character
2776 nigel 73 applies both inside and outside character classes.
2777 nigel 41
2778 nigel 77 For example, if you want to match a * character, you write \* in the
2779     pattern. This escaping action applies whether or not the following
2780     character would otherwise be interpreted as a metacharacter, so it is
2781     always safe to precede a non-alphanumeric with backslash to specify
2782     that it stands for itself. In particular, if you want to match a back-
2783 nigel 75 slash, you write \\.
2784 nigel 41
2785 nigel 77 If a pattern is compiled with the PCRE_EXTENDED option, whitespace in
2786     the pattern (other than in a character class) and characters between a
2787 nigel 91 # outside a character class and the next newline are ignored. An escap-
2788     ing backslash can be used to include a whitespace or # character as
2789     part of the pattern.
2790 nigel 41
2791 nigel 77 If you want to remove the special meaning from a sequence of charac-
2792     ters, you can do so by putting them between \Q and \E. This is differ-
2793     ent from Perl in that $ and @ are handled as literals in \Q...\E
2794     sequences in PCRE, whereas in Perl, $ and @ cause variable interpola-
2795 nigel 73 tion. Note the following examples:
2796 nigel 63
2797 nigel 73 Pattern PCRE matches Perl matches
2798 nigel 63
2799 nigel 73 \Qabc$xyz\E abc$xyz abc followed by the
2800     contents of $xyz
2801     \Qabc\$xyz\E abc\$xyz abc\$xyz
2802     \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
2803 nigel 63
2804 nigel 77 The \Q...\E sequence is recognized both inside and outside character
2805 nigel 73 classes.
2806 nigel 63
2807 nigel 75 Non-printing characters
2808    
2809 nigel 73 A second use of backslash provides a way of encoding non-printing char-
2810 nigel 77 acters in patterns in a visible manner. There is no restriction on the
2811     appearance of non-printing characters, apart from the binary zero that
2812     terminates a pattern, but when a pattern is being prepared by text
2813     editing, it is usually easier to use one of the following escape
2814 nigel 73 sequences than the binary character it represents:
2815 nigel 63
2816 nigel 73 \a alarm, that is, the BEL character (hex 07)
2817     \cx "control-x", where x is any character
2818     \e escape (hex 1B)
2819     \f formfeed (hex 0C)
2820     \n newline (hex 0A)
2821     \r carriage return (hex 0D)
2822     \t tab (hex 09)
2823     \ddd character with octal code ddd, or backreference
2824     \xhh character with hex code hh
2825 nigel 87 \x{hhh..} character with hex code hhh..
2826 nigel 41
2827 nigel 77 The precise effect of \cx is as follows: if x is a lower case letter,
2828     it is converted to upper case. Then bit 6 of the character (hex 40) is
2829     inverted. Thus \cz becomes hex 1A, but \c{ becomes hex 3B, while \c;
2830 nigel 73 becomes hex 7B.
2831 nigel 41
2832 nigel 77 After \x, from zero to two hexadecimal digits are read (letters can be
2833 nigel 87 in upper or lower case). Any number of hexadecimal digits may appear
2834     between \x{ and }, but the value of the character code must be less
2835     than 256 in non-UTF-8 mode, and less than 2**31 in UTF-8 mode (that is,
2836     the maximum hexadecimal value is 7FFFFFFF). If characters other than
2837     hexadecimal digits appear between \x{ and }, or if there is no termi-
2838     nating }, this form of escape is not recognized. Instead, the initial
2839     \x will be interpreted as a basic hexadecimal escape, with no following
2840     digits, giving a character whose value is zero.
2841 nigel 41
2842 nigel 73 Characters whose value is less than 256 can be defined by either of the
2843 nigel 87 two syntaxes for \x. There is no difference in the way they are han-
2844     dled. For example, \xdc is exactly the same as \x{dc}.
2845 nigel 41
2846 nigel 91 After \0 up to two further octal digits are read. If there are fewer
2847     than two digits, just those that are present are used. Thus the
2848     sequence \0\x\07 specifies two binary zeros followed by a BEL character
2849     (code value 7). Make sure you supply two digits after the initial zero
2850     if the pattern character that follows is itself an octal digit.
2851 nigel 63
2852 nigel 73 The handling of a backslash followed by a digit other than 0 is compli-
2853     cated. Outside a character class, PCRE reads it and any following dig-
2854 nigel 91 its as a decimal number. If the number is less than 10, or if there
2855 nigel 73 have been at least that many previous capturing left parentheses in the
2856 nigel 91 expression, the entire sequence is taken as a back reference. A
2857     description of how this works is given later, following the discussion
2858 nigel 73 of parenthesized subpatterns.
2859 nigel 41
2860 nigel 91 Inside a character class, or if the decimal number is greater than 9
2861     and there have not been that many capturing subpatterns, PCRE re-reads
2862 nigel 93 up to three octal digits following the backslash, and uses them to gen-
2863 nigel 91 erate a data character. Any subsequent digits stand for themselves. In
2864     non-UTF-8 mode, the value of a character specified in octal must be
2865     less than \400. In UTF-8 mode, values up to \777 are permitted. For
2866     example:
2867 nigel 41
2868 nigel 73 \040 is another way of writing a space
2869     \40 is the same, provided there are fewer than 40
2870     previous capturing subpatterns
2871     \7 is always a back reference
2872     \11 might be a back reference, or another way of
2873     writing a tab
2874     \011 is always a tab
2875     \0113 is a tab followed by the character "3"
2876     \113 might be a back reference, otherwise the
2877     character with octal code 113
2878     \377 might be a back reference, otherwise
2879     the byte consisting entirely of 1 bits
2880     \81 is either a back reference, or a binary zero
2881     followed by the two characters "8" and "1"
2882 nigel 41
2883 nigel 91 Note that octal values of 100 or greater must not be introduced by a
2884 nigel 73 leading zero, because no more than three octal digits are ever read.
2885 nigel 41
2886 nigel 91 All the sequences that define a single character value can be used both
2887     inside and outside character classes. In addition, inside a character
2888     class, the sequence \b is interpreted as the backspace character (hex
2889 nigel 93 08), and the sequences \R and \X are interpreted as the characters "R"
2890     and "X", respectively. Outside a character class, these sequences have
2891     different meanings (see below).
2892 nigel 43
2893 nigel 93 Absolute and relative back references
2894    
2895     The sequence \g followed by a positive or negative number, optionally
2896     enclosed in braces, is an absolute or relative back reference. Back
2897     references are discussed later, following the discussion of parenthe-
2898     sized subpatterns.
2899    
2900 nigel 75 Generic character types
2901 nigel 41
2902 nigel 93 Another use of backslash is for specifying generic character types. The
2903     following are always recognized:
2904 nigel 75
2905 nigel 73 \d any decimal digit
2906     \D any character that is not a decimal digit
2907     \s any whitespace character
2908     \S any character that is not a whitespace character
2909     \w any "word" character
2910     \W any "non-word" character
2911 nigel 41
2912 nigel 73 Each pair of escape sequences partitions the complete set of characters
2913 nigel 87 into two disjoint sets. Any given character matches one, and only one,
2914 nigel 73 of each pair.
2915 nigel 41
2916 nigel 75 These character type sequences can appear both inside and outside char-
2917 nigel 87 acter classes. They each match one character of the appropriate type.
2918     If the current matching point is at the end of the subject string, all
2919 nigel 75 of them fail, since there is no character to match.
2920 nigel 41
2921 nigel 87 For compatibility with Perl, \s does not match the VT character (code
2922     11). This makes it different from the the POSIX "space" class. The \s
2923 nigel 91 characters are HT (9), LF (10), FF (12), CR (13), and space (32). (If
2924     "use locale;" is included in a Perl script, \s may match the VT charac-
2925     ter. In PCRE, it never does.)
2926 nigel 63
2927 nigel 75 A "word" character is an underscore or any character less than 256 that
2928 nigel 87 is a letter or digit. The definition of letters and digits is con-
2929     trolled by PCRE's low-valued character tables, and may vary if locale-
2930     specific matching is taking place (see "Locale support" in the pcreapi
2931 ph10 142 page). For example, in a French locale such as "fr_FR" in Unix-like
2932     systems, or "french" in Windows, some character codes greater than 128
2933     are used for accented letters, and these are matched by \w.
2934 nigel 63
2935 nigel 87 In UTF-8 mode, characters with values greater than 128 never match \d,
2936 nigel 75 \s, or \w, and always match \D, \S, and \W. This is true even when Uni-
2937 nigel 87 code character property support is available. The use of locales with
2938     Unicode is discouraged.
2939 nigel 41
2940 nigel 93 Newline sequences
2941    
2942     Outside a character class, the escape sequence \R matches any Unicode
2943     newline sequence. This is an extension to Perl. In non-UTF-8 mode \R is
2944     equivalent to the following:
2945    
2946     (?>\r\n|\n|\x0b|\f|\r|\x85)
2947    
2948     This is an example of an "atomic group", details of which are given
2949     below. This particular group matches either the two-character sequence
2950     CR followed by LF, or one of the single characters LF (linefeed,
2951     U+000A), VT (vertical tab, U+000B), FF (formfeed, U+000C), CR (carriage
2952     return, U+000D), or NEL (next line, U+0085). The two-character sequence
2953     is treated as a single unit that cannot be split.
2954    
2955     In UTF-8 mode, two additional characters whose codepoints are greater
2956     than 255 are added: LS (line separator, U+2028) and PS (paragraph sepa-
2957     rator, U+2029). Unicode character property support is not needed for
2958     these characters to be recognized.
2959    
2960     Inside a character class, \R matches the letter "R".
2961    
2962 nigel 75 Unicode character properties
2963    
2964     When PCRE is built with Unicode character property support, three addi-
2965 nigel 87 tional escape sequences to match character properties are available
2966 nigel 75 when UTF-8 mode is selected. They are:
2967    
2968 nigel 87 \p{xx} a character with the xx property
2969     \P{xx} a character without the xx property
2970     \X an extended Unicode sequence
2971 nigel 75
2972 nigel 77 The property names represented by xx above are limited to the Unicode
2973 nigel 87 script names, the general category properties, and "Any", which matches
2974     any character (including newline). Other properties such as "InMusical-
2975     Symbols" are not currently supported by PCRE. Note that \P{Any} does
2976     not match any characters, so always causes a match failure.
2977 nigel 75
2978 nigel 87 Sets of Unicode characters are defined as belonging to certain scripts.
2979     A character from one of these sets can be matched using a script name.
2980     For example:
2981 nigel 75
2982 nigel 87 \p{Greek}
2983     \P{Han}
2984    
2985     Those that are not part of an identified script are lumped together as
2986     "Common". The current list of scripts is:
2987    
2988 nigel 93 Arabic, Armenian, Balinese, Bengali, Bopomofo, Braille, Buginese,
2989     Buhid, Canadian_Aboriginal, Cherokee, Common, Coptic, Cuneiform,
2990     Cypriot, Cyrillic, Deseret, Devanagari, Ethiopic, Georgian, Glagolitic,
2991     Gothic, Greek, Gujarati, Gurmukhi, Han, Hangul, Hanunoo, Hebrew, Hira-
2992     gana, Inherited, Kannada, Katakana, Kharoshthi, Khmer, Lao, Latin,
2993     Limbu, Linear_B, Malayalam, Mongolian, Myanmar, New_Tai_Lue, Nko,
2994     Ogham, Old_Italic, Old_Persian, Oriya, Osmanya, Phags_Pa, Phoenician,
2995     Runic, Shavian, Sinhala, Syloti_Nagri, Syriac, Tagalog, Tagbanwa,
2996     Tai_Le, Tamil, Telugu, Thaana, Thai, Tibetan, Tifinagh, Ugaritic, Yi.
2997 nigel 87
2998     Each character has exactly one general category property, specified by
2999     a two-letter abbreviation. For compatibility with Perl, negation can be
3000     specified by including a circumflex between the opening brace and the
3001     property name. For example, \p{^Lu} is the same as \P{Lu}.
3002    
3003     If only one letter is specified with \p or \P, it includes all the gen-
3004     eral category properties that start with that letter. In this case, in
3005     the absence of negation, the curly brackets in the escape sequence are
3006     optional; these two examples have the same effect:
3007    
3008 nigel 75 \p{L}
3009     \pL
3010    
3011 nigel 87 The following general category property codes are supported:
3012 nigel 75
3013     C Other
3014     Cc Control
3015     Cf Format
3016     Cn Unassigned
3017     Co Private use
3018     Cs Surrogate
3019    
3020     L Letter
3021     Ll Lower case letter
3022     Lm Modifier letter
3023     Lo Other letter
3024     Lt Title case letter
3025     Lu Upper case letter
3026    
3027     M Mark
3028     Mc Spacing mark
3029     Me Enclosing mark
3030     Mn Non-spacing mark
3031    
3032     N Number
3033     Nd Decimal number
3034     Nl Letter number
3035     No Other number
3036    
3037     P Punctuation
3038     Pc Connector punctuation
3039     Pd Dash punctuation
3040     Pe Close punctuation
3041     Pf Final punctuation
3042     Pi Initial punctuation
3043     Po Other punctuation
3044     Ps Open punctuation
3045    
3046     S Symbol
3047     Sc Currency symbol
3048     Sk Modifier symbol
3049     Sm Mathematical symbol
3050     So Other symbol
3051    
3052     Z Separator
3053     Zl Line separator
3054     Zp Paragraph separator
3055     Zs Space separator
3056    
3057 nigel 87 The special property L& is also supported: it matches a character that
3058     has the Lu, Ll, or Lt property, in other words, a letter that is not
3059     classified as a modifier or "other".
3060 nigel 75
3061 nigel 87 The long synonyms for these properties that Perl supports (such as
3062 nigel 91 \p{Letter}) are not supported by PCRE, nor is it permitted to prefix
3063 nigel 87 any of these properties with "Is".
3064    
3065     No character that is in the Unicode table has the Cn (unassigned) prop-
3066     erty. Instead, this property is assumed for any code point that is not
3067     in the Unicode table.
3068    
3069 nigel 77 Specifying caseless matching does not affect these escape sequences.
3070 nigel 75 For example, \p{Lu} always matches only upper case letters.
3071    
3072 nigel 77 The \X escape matches any number of Unicode characters that form an
3073 nigel 75 extended Unicode sequence. \X is equivalent to
3074    
3075     (?>\PM\pM*)
3076    
3077 nigel 77 That is, it matches a character without the "mark" property, followed
3078     by zero or more characters with the "mark" property, and treats the
3079     sequence as an atomic group (see below). Characters with the "mark"
3080 nigel 75 property are typically accents that affect the preceding character.
3081    
3082 nigel 77 Matching characters by Unicode property is not fast, because PCRE has
3083     to search a structure that contains data for over fifteen thousand
3084 nigel 75 characters. That is why the traditional escape sequences such as \d and
3085     \w do not use Unicode properties in PCRE.
3086    
3087     Simple assertions
3088    
3089 nigel 93 The final use of backslash is for certain simple assertions. An asser-
3090 nigel 77 tion specifies a condition that has to be met at a particular point in
3091     a match, without consuming any characters from the subject string. The
3092     use of subpatterns for more complicated assertions is described below.
3093 nigel 75 The backslashed assertions are:
3094 nigel 41
3095 nigel 73 \b matches at a word boundary
3096     \B matches when not at a word boundary
3097 nigel 93 \A matches at the start of the subject
3098     \Z matches at the end of the subject
3099     also matches before a newline at the end of the subject
3100     \z matches only at the end of the subject
3101     \G matches at the first matching position in the subject
3102 nigel 41
3103 nigel 77 These assertions may not appear in character classes (but note that \b
3104 nigel 73 has a different meaning, namely the backspace character, inside a char-
3105     acter class).
3106 nigel 41
3107 nigel 77 A word boundary is a position in the subject string where the current
3108     character and the previous character do not both match \w or \W (i.e.
3109     one matches \w and the other matches \W), or the start or end of the
3110 nigel 73 string if the first or last character matches \w, respectively.
3111 nigel 43
3112 nigel 77 The \A, \Z, and \z assertions differ from the traditional circumflex
3113 nigel 75 and dollar (described in the next section) in that they only ever match
3114 nigel 77 at the very start and end of the subject string, whatever options are
3115     set. Thus, they are independent of multiline mode. These three asser-
3116 nigel 75 tions are not affected by the PCRE_NOTBOL or PCRE_NOTEOL options, which
3117 nigel 77 affect only the behaviour of the circumflex and dollar metacharacters.
3118     However, if the startoffset argument of pcre_exec() is non-zero, indi-
3119 nigel 75 cating that matching is to start at a point other than the beginning of
3120 nigel 77 the subject, \A can never match. The difference between \Z and \z is
3121 nigel 91 that \Z matches before a newline at the end of the string as well as at
3122     the very end, whereas \z matches only at the end.
3123 nigel 63
3124 nigel 91 The \G assertion is true only when the current matching position is at
3125     the start point of the match, as specified by the startoffset argument
3126     of pcre_exec(). It differs from \A when the value of startoffset is
3127     non-zero. By calling pcre_exec() multiple times with appropriate argu-
3128 nigel 73 ments, you can mimic Perl's /g option, and it is in this kind of imple-
3129     mentation where \G can be useful.
3130 nigel 41
3131 nigel 91 Note, however, that PCRE's interpretation of \G, as the start of the
3132 nigel 73 current match, is subtly different from Perl's, which defines it as the
3133 nigel 91 end of the previous match. In Perl, these can be different when the
3134     previously matched string was empty. Because PCRE does just one match
3135 nigel 73 at a time, it cannot reproduce this behaviour.
3136 nigel 41
3137 nigel 91 If all the alternatives of a pattern begin with \G, the expression is
3138 nigel 73 anchored to the starting match position, and the "anchored" flag is set
3139     in the compiled regular expression.
3140 nigel 63
3141    
3142 nigel 41 CIRCUMFLEX AND DOLLAR
3143 nigel 63
3144 nigel 73 Outside a character class, in the default matching mode, the circumflex
3145 nigel 91 character is an assertion that is true only if the current matching
3146     point is at the start of the subject string. If the startoffset argu-
3147     ment of pcre_exec() is non-zero, circumflex can never match if the
3148     PCRE_MULTILINE option is unset. Inside a character class, circumflex
3149 nigel 73 has an entirely different meaning (see below).
3150 nigel 41
3151 nigel 91 Circumflex need not be the first character of the pattern if a number
3152     of alternatives are involved, but it should be the first thing in each
3153     alternative in which it appears if the pattern is ever to match that
3154     branch. If all possible alternatives start with a circumflex, that is,
3155     if the pattern is constrained to match only at the start of the sub-
3156     ject, it is said to be an "anchored" pattern. (There are also other
3157 nigel 73 constructs that can cause a pattern to be anchored.)
3158 nigel 41
3159 nigel 91 A dollar character is an assertion that is true only if the current
3160     matching point is at the end of the subject string, or immediately
3161     before a newline at the end of the string (by default). Dollar need not
3162     be the last character of the pattern if a number of alternatives are
3163     involved, but it should be the last item in any branch in which it
3164     appears. Dollar has no special meaning in a character class.
3165 nigel 41
3166 nigel 77 The meaning of dollar can be changed so that it matches only at the
3167     very end of the string, by setting the PCRE_DOLLAR_ENDONLY option at
3168 nigel 73 compile time. This does not affect the \Z assertion.
3169 nigel 41
3170 nigel 73 The meanings of the circumflex and dollar characters are changed if the
3171 nigel 91 PCRE_MULTILINE option is set. When this is the case, a circumflex
3172     matches immediately after internal newlines as well as at the start of
3173     the subject string. It does not match after a newline that ends the
3174     string. A dollar matches before any newlines in the string, as well as
3175     at the very end, when PCRE_MULTILINE is set. When newline is specified
3176     as the two-character sequence CRLF, isolated CR and LF characters do
3177     not indicate newlines.
3178 nigel 41
3179 nigel 91 For example, the pattern /^abc$/ matches the subject string "def\nabc"
3180     (where \n represents a newline) in multiline mode, but not otherwise.
3181     Consequently, patterns that are anchored in single line mode because
3182     all branches start with ^ are not anchored in multiline mode, and a
3183     match for circumflex is possible when the startoffset argument of
3184     pcre_exec() is non-zero. The PCRE_DOLLAR_ENDONLY option is ignored if
3185     PCRE_MULTILINE is set.
3186 nigel 41
3187 nigel 91 Note that the sequences \A, \Z, and \z can be used to match the start
3188     and end of the subject in both modes, and if all branches of a pattern
3189     start with \A it is always anchored, whether or not PCRE_MULTILINE is
3190     set.
3191 nigel 41
3192 nigel 91
3193 nigel 63 FULL STOP (PERIOD, DOT)
3194 nigel 41
3195 nigel 73 Outside a character class, a dot in the pattern matches any one charac-
3196 nigel 91 ter in the subject string except (by default) a character that signi-
3197     fies the end of a line. In UTF-8 mode, the matched character may be
3198 nigel 93 more than one byte long.
3199 nigel 41
3200 nigel 93 When a line ending is defined as a single character, dot never matches
3201     that character; when the two-character sequence CRLF is used, dot does
3202     not match CR if it is immediately followed by LF, but otherwise it
3203     matches all characters (including isolated CRs and LFs). When any Uni-
3204     code line endings are being recognized, dot does not match CR or LF or
3205     any of the other line ending characters.
3206 nigel 41
3207 nigel 93 The behaviour of dot with regard to newlines can be changed. If the
3208     PCRE_DOTALL option is set, a dot matches any one character, without
3209     exception. If the two-character sequence CRLF is present in the subject
3210     string, it takes two dots to match it.
3211    
3212     The handling of dot is entirely independent of the handling of circum-
3213     flex and dollar, the only relationship being that they both involve
3214 nigel 91 newlines. Dot has no special meaning in a character class.
3215    
3216    
3217 nigel 63 MATCHING A SINGLE BYTE
3218    
3219 nigel 73 Outside a character class, the escape sequence \C matches any one byte,
3220 nigel 93 both in and out of UTF-8 mode. Unlike a dot, it always matches any
3221     line-ending characters. The feature is provided in Perl in order to
3222     match individual bytes in UTF-8 mode. Because it breaks up UTF-8 char-
3223     acters into individual bytes, what remains in the string may be a mal-
3224     formed UTF-8 string. For this reason, the \C escape sequence is best
3225     avoided.
3226 nigel 63
3227 nigel 77 PCRE does not allow \C to appear in lookbehind assertions (described
3228     below), because in UTF-8 mode this would make it impossible to calcu-
3229 nigel 75 late the length of the lookbehind.
3230 nigel 63
3231    
3232 nigel 75 SQUARE BRACKETS AND CHARACTER CLASSES
3233 nigel 63
3234 nigel 73 An opening square bracket introduces a character class, terminated by a
3235     closing square bracket. A closing square bracket on its own is not spe-
3236     cial. If a closing square bracket is required as a member of the class,
3237 nigel 77 it should be the first data character in the class (after an initial
3238 nigel 73 circumflex, if present) or escaped with a backslash.
3239 nigel 41
3240 nigel 77 A character class matches a single character in the subject. In UTF-8
3241     mode, the character may occupy more than one byte. A matched character
3242 nigel 73 must be in the set of characters defined by the class, unless the first
3243 nigel 77 character in the class definition is a circumflex, in which case the
3244     subject character must not be in the set defined by the class. If a
3245     circumflex is actually required as a member of the class, ensure it is
3246 nigel 73 not the first character, or escape it with a backslash.
3247 nigel 41
3248 nigel 77 For example, the character class [aeiou] matches any lower case vowel,
3249     while [^aeiou] matches any character that is not a lower case vowel.
3250 nigel 73 Note that a circumflex is just a convenient notation for specifying the
3251 nigel 77 characters that are in the class by enumerating those that are not. A
3252     class that starts with a circumflex is not an assertion: it still con-
3253     sumes a character from the subject string, and therefore it fails if
3254 nigel 75 the current pointer is at the end of the string.
3255 nigel 41
3256 nigel 77 In UTF-8 mode, characters with values greater than 255 can be included
3257     in a class as a literal string of bytes, or by using the \x{ escaping
3258 nigel 73 mechanism.
3259 nigel 63
3260 nigel 77 When caseless matching is set, any letters in a class represent both
3261     their upper case and lower case versions, so for example, a caseless
3262     [aeiou] matches "A" as well as "a", and a caseless [^aeiou] does not
3263     match "A", whereas a caseful version would. In UTF-8 mode, PCRE always
3264     understands the concept of case for characters whose values are less
3265     than 128, so caseless matching is always possible. For characters with
3266     higher values, the concept of case is supported if PCRE is compiled
3267     with Unicode property support, but not otherwise. If you want to use
3268     caseless matching for characters 128 and above, you must ensure that
3269     PCRE is compiled with Unicode property support as well as with UTF-8
3270     support.
3271 nigel 41
3272 nigel 93 Characters that might indicate line breaks are never treated in any
3273     special way when matching character classes, whatever line-ending
3274     sequence is in use, and whatever setting of the PCRE_DOTALL and
3275     PCRE_MULTILINE options is used. A class such as [^a] always matches one
3276     of these characters.
3277 nigel 41
3278 nigel 75 The minus (hyphen) character can be used to specify a range of charac-
3279     ters in a character class. For example, [d-m] matches any letter
3280     between d and m, inclusive. If a minus character is required in a
3281     class, it must be escaped with a backslash or appear in a position
3282     where it cannot be interpreted as indicating a range, typically as the
3283 nigel 73 first or last character in the class.
3284 nigel 41
3285 nigel 73 It is not possible to have the literal character "]" as the end charac-
3286 nigel 75 ter of a range. A pattern such as [W-]46] is interpreted as a class of
3287     two characters ("W" and "-") followed by a literal string "46]", so it
3288     would match "W46]" or "-46]". However, if the "]" is escaped with a
3289     backslash it is interpreted as the end of range, so [W-\]46] is inter-
3290     preted as a class containing a range followed by two other characters.
3291     The octal or hexadecimal representation of "]" can also be used to end
3292     a range.
3293 nigel 41
3294 nigel 75 Ranges operate in the collating sequence of character values. They can
3295     also be used for characters specified numerically, for example
3296     [\000-\037]. In UTF-8 mode, ranges can include characters whose values
3297 nigel 73 are greater than 255, for example [\x{100}-\x{2ff}].
3298 nigel 63
3299 nigel 73 If a range that includes letters is used when caseless matching is set,
3300     it matches the letters in either case. For example, [W-c] is equivalent
3301 nigel 75 to [][\\^_`wxyzabc], matched caselessly, and in non-UTF-8 mode, if
3302 ph10 142 character tables for a French locale are in use, [\xc8-\xcb] matches
3303 nigel 75 accented E characters in both cases. In UTF-8 mode, PCRE supports the
3304     concept of case for characters with values greater than 128 only when
3305     it is compiled with Unicode property support.
3306 nigel 41
3307 nigel 75 The character types \d, \D, \p, \P, \s, \S, \w, and \W may also appear
3308     in a character class, and add the characters that they match to the
3309     class. For example, [\dABCDEF] matches any hexadecimal digit. A circum-
3310     flex can conveniently be used with the upper case character types to
3311     specify a more restricted set of characters than the matching lower
3312     case type. For example, the class [^\W_] matches any letter or digit,
3313     but not underscore.
3314 nigel 41
3315 nigel 75 The only metacharacters that are recognized in character classes are
3316     backslash, hyphen (only where it can be interpreted as specifying a
3317     range), circumflex (only at the start), opening square bracket (only
3318     when it can be interpreted as introducing a POSIX class name - see the
3319     next section), and the terminating closing square bracket. However,
3320     escaping other non-alphanumeric characters does no harm.
3321 nigel 41
3322 nigel 73
3323 nigel 43 POSIX CHARACTER CLASSES
3324    
3325 nigel 75 Perl supports the POSIX notation for character classes. This uses names
3326     enclosed by [: and :] within the enclosing square brackets. PCRE also
3327     supports this notation. For example,
3328 nigel 63
3329 nigel 73 [01[:alpha:]%]
3330 nigel 43
3331 nigel 73 matches "0", "1", any alphabetic character, or "%". The supported class
3332     names are
3333 nigel 43
3334 nigel 73 alnum letters and digits
3335     alpha letters
3336     ascii character codes 0 - 127
3337     blank space or tab only
3338     cntrl control characters
3339     digit decimal digits (same as \d)
3340     graph printing characters, excluding space
3341     lower lower case letters
3342     print printing characters, including space
3343     punct printing characters, excluding letters and digits
3344     space white space (not quite the same as \s)
3345     upper upper case letters
3346     word "word" characters (same as \w)
3347     xdigit hexadecimal digits
3348 nigel 43
3349 nigel 73 The "space" characters are HT (9), LF (10), VT (11), FF (12), CR (13),
3350     and space (32). Notice that this list includes the VT character (code
3351     11). This makes "space" different to \s, which does not include VT (for
3352     Perl compatibility).
3353 nigel 43
3354 nigel 73 The name "word" is a Perl extension, and "blank" is a GNU extension
3355     from Perl 5.8. Another Perl extension is negation, which is indicated
3356     by a ^ character after the colon. For example,
3357 nigel 63
3358 nigel 73 [12[:^digit:]]
3359 nigel 43
3360 nigel 73 matches "1", "2", or any non-digit. PCRE (and Perl) also recognize the
3361     POSIX syntax [.ch.] and [=ch=] where "ch" is a "collating element", but
3362     these are not supported, and an error is given if they are encountered.
3363 nigel 43
3364 nigel 75 In UTF-8 mode, characters with values greater than 128 do not match any
3365 nigel 73 of the POSIX character classes.
3366 nigel 43
3367    
3368 nigel 41 VERTICAL BAR
3369 nigel 63
3370 nigel 73 Vertical bar characters are used to separate alternative patterns. For
3371     example, the pattern
3372 nigel 41
3373 nigel 73 gilbert|sullivan
3374 nigel 41
3375 nigel 73 matches either "gilbert" or "sullivan". Any number of alternatives may
3376     appear, and an empty alternative is permitted (matching the empty
3377 nigel 91 string). The matching process tries each alternative in turn, from left
3378     to right, and the first one that succeeds is used. If the alternatives
3379     are within a subpattern (defined below), "succeeds" means matching the
3380     rest of the main pattern as well as the alternative in the subpattern.
3381 nigel 41
3382    
3383     INTERNAL OPTION SETTING
3384    
3385 nigel 73 The settings of the PCRE_CASELESS, PCRE_MULTILINE, PCRE_DOTALL, and
3386     PCRE_EXTENDED options can be changed from within the pattern by a
3387     sequence of Perl option letters enclosed between "(?" and ")". The
3388     option letters are
3389 nigel 63
3390 nigel 73 i for PCRE_CASELESS
3391     m for PCRE_MULTILINE
3392     s for PCRE_DOTALL
3393     x for PCRE_EXTENDED
3394 nigel 41
3395 nigel 73 For example, (?im) sets caseless, multiline matching. It is also possi-
3396     ble to unset these options by preceding the letter with a hyphen, and a
3397     combined setting and unsetting such as (?im-sx), which sets PCRE_CASE-
3398     LESS and PCRE_MULTILINE while unsetting PCRE_DOTALL and PCRE_EXTENDED,
3399     is also permitted. If a letter appears both before and after the
3400     hyphen, the option is unset.
3401 nigel 41
3402 nigel 73 When an option change occurs at top level (that is, not inside subpat-
3403     tern parentheses), the change applies to the remainder of the pattern
3404     that follows. If the change is placed right at the start of a pattern,
3405     PCRE extracts it into the global options (and it will therefore show up
3406     in data extracted by the pcre_fullinfo() function).
3407 nigel 41
3408 nigel 93 An option change within a subpattern (see below for a description of
3409     subpatterns) affects only that part of the current pattern that follows
3410     it, so
3411 nigel 41
3412 nigel 73 (a(?i)b)c
3413 nigel 41
3414 nigel 73 matches abc and aBc and no other strings (assuming PCRE_CASELESS is not
3415 nigel 93 used). By this means, options can be made to have different settings
3416     in different parts of the pattern. Any changes made in one alternative
3417     do carry on into subsequent branches within the same subpattern. For
3418 nigel 73 example,
3419 nigel 41
3420 nigel 73 (a(?i)b|c)
3421 nigel 41
3422 nigel 93 matches "ab", "aB", "c", and "C", even though when matching "C" the
3423     first branch is abandoned before the option setting. This is because
3424     the effects of option settings happen at compile time. There would be
3425 nigel 73 some very weird behaviour otherwise.
3426 nigel 41
3427 nigel 93 The PCRE-specific options PCRE_DUPNAMES, PCRE_UNGREEDY, and PCRE_EXTRA
3428     can be changed in the same way as the Perl-compatible options by using
3429 nigel 91 the characters J, U and X respectively.
3430 nigel 41
3431    
3432 nigel 63 SUBPATTERNS
3433 nigel 41
3434 nigel 73 Subpatterns are delimited by parentheses (round brackets), which can be
3435 nigel 75 nested. Turning part of a pattern into a subpattern does two things:
3436 nigel 41
3437 nigel 73 1. It localizes a set of alternatives. For example, the pattern
3438 nigel 41
3439 nigel 73 cat(aract|erpillar|)
3440 nigel 41