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

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

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 51 by nigel, Sat Feb 24 21:39:37 2007 UTC revision 548 by ph10, Fri Jun 25 14:42:00 2010 UTC
# Line 1  Line 1 
1    -----------------------------------------------------------------------------
2    This file contains a concatenation of the PCRE man pages, converted to plain
3    text format for ease of searching with a text editor, or for use on systems
4    that do not have a man page processor. The small individual files that give
5    synopses of each function in the library have not been included. Neither has
6    the pcredemo program. There are separate text files for the pcregrep and
7    pcretest commands.
8    -----------------------------------------------------------------------------
9    
10    
11    PCRE(3)                                                                PCRE(3)
12    
13    
14    NAME
15           PCRE - Perl-compatible regular expressions
16    
17    
18    INTRODUCTION
19    
20           The  PCRE  library is a set of functions that implement regular expres-
21           sion pattern matching using the same syntax and semantics as Perl, with
22           just  a few differences. Some features that appeared in Python and PCRE
23           before they appeared in Perl are also available using the  Python  syn-
24           tax,  there  is  some  support for one or two .NET and Oniguruma syntax
25           items, and there is an option for requesting some  minor  changes  that
26           give better JavaScript compatibility.
27    
28           The  current implementation of PCRE corresponds approximately with Perl
29           5.10/5.11, including support for UTF-8 encoded strings and Unicode gen-
30           eral  category properties. However, UTF-8 and Unicode support has to be
31           explicitly enabled; it is not the default. The  Unicode  tables  corre-
32           spond to Unicode release 5.2.0.
33    
34           In  addition to the Perl-compatible matching function, PCRE contains an
35           alternative function that matches the same compiled patterns in a  dif-
36           ferent way. In certain circumstances, the alternative function has some
37           advantages.  For a discussion of the two matching algorithms,  see  the
38           pcrematching page.
39    
40           PCRE  is  written  in C and released as a C library. A number of people
41           have written wrappers and interfaces of various kinds.  In  particular,
42           Google  Inc.   have  provided  a comprehensive C++ wrapper. This is now
43           included as part of the PCRE distribution. The pcrecpp page has details
44           of  this  interface.  Other  people's contributions can be found in the
45           Contrib directory at the primary FTP site, which is:
46    
47           ftp://ftp.csx.cam.ac.uk/pub/software/programming/pcre
48    
49           Details of exactly which Perl regular expression features are  and  are
50           not supported by PCRE are given in separate documents. See the pcrepat-
51           tern and pcrecompat pages. There is a syntax summary in the  pcresyntax
52           page.
53    
54           Some  features  of  PCRE can be included, excluded, or changed when the
55           library is built. The pcre_config() function makes it  possible  for  a
56           client  to  discover  which  features are available. The features them-
57           selves are described in the pcrebuild page. Documentation about  build-
58           ing  PCRE  for various operating systems can be found in the README and
59           NON-UNIX-USE files in the source distribution.
60    
61           The library contains a number of undocumented  internal  functions  and
62           data  tables  that  are  used by more than one of the exported external
63           functions, but which are not intended  for  use  by  external  callers.
64           Their  names  all begin with "_pcre_", which hopefully will not provoke
65           any name clashes. In some environments, it is possible to control which
66           external  symbols  are  exported when a shared library is built, and in
67           these cases the undocumented symbols are not exported.
68    
69    
70    USER DOCUMENTATION
71    
72           The user documentation for PCRE comprises a number  of  different  sec-
73           tions.  In the "man" format, each of these is a separate "man page". In
74           the HTML format, each is a separate page, linked from the  index  page.
75           In  the  plain  text format, all the sections, except the pcredemo sec-
76           tion, are concatenated, for ease of searching. The sections are as fol-
77           lows:
78    
79             pcre              this document
80             pcre-config       show PCRE installation configuration information
81             pcreapi           details of PCRE's native C API
82             pcrebuild         options for building PCRE
83             pcrecallout       details of the callout feature
84             pcrecompat        discussion of Perl compatibility
85             pcrecpp           details of the C++ wrapper
86             pcredemo          a demonstration C program that uses PCRE
87             pcregrep          description of the pcregrep command
88             pcrematching      discussion of the two matching algorithms
89             pcrepartial       details of the partial matching facility
90             pcrepattern       syntax and semantics of supported
91                                 regular expressions
92             pcreperform       discussion of performance issues
93             pcreposix         the POSIX-compatible C API
94             pcreprecompile    details of saving and re-using precompiled patterns
95             pcresample        discussion of the pcredemo program
96             pcrestack         discussion of stack usage
97             pcresyntax        quick syntax reference
98             pcretest          description of the pcretest testing command
99    
100           In  addition,  in the "man" and HTML formats, there is a short page for
101           each C library function, listing its arguments and results.
102    
103    
104    LIMITATIONS
105    
106           There are some size limitations in PCRE but it is hoped that they  will
107           never in practice be relevant.
108    
109           The  maximum  length of a compiled pattern is 65539 (sic) bytes if PCRE
110           is compiled with the default internal linkage size of 2. If you want to
111           process  regular  expressions  that are truly enormous, you can compile
112           PCRE with an internal linkage size of 3 or 4 (see the  README  file  in
113           the  source  distribution and the pcrebuild documentation for details).
114           In these cases the limit is substantially larger.  However,  the  speed
115           of execution is slower.
116    
117           All values in repeating quantifiers must be less than 65536.
118    
119           There is no limit to the number of parenthesized subpatterns, but there
120           can be no more than 65535 capturing subpatterns.
121    
122           The maximum length of name for a named subpattern is 32 characters, and
123           the maximum number of named subpatterns is 10000.
124    
125           The  maximum  length of a subject string is the largest positive number
126           that an integer variable can hold. However, when using the  traditional
127           matching function, PCRE uses recursion to handle subpatterns and indef-
128           inite repetition.  This means that the available stack space may  limit
129           the size of a subject string that can be processed by certain patterns.
130           For a discussion of stack issues, see the pcrestack documentation.
131    
132    
133    UTF-8 AND UNICODE PROPERTY SUPPORT
134    
135           From release 3.3, PCRE has  had  some  support  for  character  strings
136           encoded  in the UTF-8 format. For release 4.0 this was greatly extended
137           to cover most common requirements, and in release 5.0  additional  sup-
138           port for Unicode general category properties was added.
139    
140           In  order  process  UTF-8 strings, you must build PCRE to include UTF-8
141           support in the code, and, in addition,  you  must  call  pcre_compile()
142           with  the  PCRE_UTF8  option  flag,  or the pattern must start with the
143           sequence (*UTF8). When either of these is the case,  both  the  pattern
144           and  any  subject  strings  that  are matched against it are treated as
145           UTF-8 strings instead of strings of 1-byte characters.
146    
147           If you compile PCRE with UTF-8 support, but do not use it at run  time,
148           the  library will be a bit bigger, but the additional run time overhead
149           is limited to testing the PCRE_UTF8 flag occasionally, so should not be
150           very big.
151    
152           If PCRE is built with Unicode character property support (which implies
153           UTF-8 support), the escape sequences \p{..}, \P{..}, and  \X  are  sup-
154           ported.  The available properties that can be tested are limited to the
155           general category properties such as Lu for an upper case letter  or  Nd
156           for  a  decimal number, the Unicode script names such as Arabic or Han,
157           and the derived properties Any and L&. A full  list  is  given  in  the
158           pcrepattern documentation. Only the short names for properties are sup-
159           ported. For example, \p{L} matches a letter. Its Perl synonym,  \p{Let-
160           ter},  is  not  supported.   Furthermore,  in Perl, many properties may
161           optionally be prefixed by "Is", for compatibility with Perl  5.6.  PCRE
162           does not support this.
163    
164       Validity of UTF-8 strings
165    
166           When  you  set  the  PCRE_UTF8 flag, the strings passed as patterns and
167           subjects are (by default) checked for validity on entry to the relevant
168           functions.  From  release 7.3 of PCRE, the check is according the rules
169           of RFC 3629, which are themselves derived from the  Unicode  specifica-
170           tion.  Earlier  releases  of PCRE followed the rules of RFC 2279, which
171           allows the full range of 31-bit values (0 to 0x7FFFFFFF).  The  current
172           check allows only values in the range U+0 to U+10FFFF, excluding U+D800
173           to U+DFFF.
174    
175           The excluded code points are the "Low Surrogate Area"  of  Unicode,  of
176           which  the Unicode Standard says this: "The Low Surrogate Area does not
177           contain any  character  assignments,  consequently  no  character  code
178           charts or namelists are provided for this area. Surrogates are reserved
179           for use with UTF-16 and then must be used in pairs."  The  code  points
180           that  are  encoded  by  UTF-16  pairs are available as independent code
181           points in the UTF-8 encoding. (In  other  words,  the  whole  surrogate
182           thing is a fudge for UTF-16 which unfortunately messes up UTF-8.)
183    
184           If  an  invalid  UTF-8  string  is  passed  to  PCRE,  an  error return
185           (PCRE_ERROR_BADUTF8) is given. In some situations, you may already know
186           that your strings are valid, and therefore want to skip these checks in
187           order to improve performance. If you set the PCRE_NO_UTF8_CHECK flag at
188           compile  time  or at run time, PCRE assumes that the pattern or subject
189           it is given (respectively) contains only valid  UTF-8  codes.  In  this
190           case, it does not diagnose an invalid UTF-8 string.
191    
192           If  you  pass  an  invalid UTF-8 string when PCRE_NO_UTF8_CHECK is set,
193           what happens depends on why the string is invalid. If the  string  con-
194           forms to the "old" definition of UTF-8 (RFC 2279), it is processed as a
195           string of characters in the range 0  to  0x7FFFFFFF.  In  other  words,
196           apart from the initial validity test, PCRE (when in UTF-8 mode) handles
197           strings according to the more liberal rules of RFC  2279.  However,  if
198           the  string does not even conform to RFC 2279, the result is undefined.
199           Your program may crash.
200    
201           If you want to process strings  of  values  in  the  full  range  0  to
202           0x7FFFFFFF,  encoded in a UTF-8-like manner as per the old RFC, you can
203           set PCRE_NO_UTF8_CHECK to bypass the more restrictive test. However, in
204           this situation, you will have to apply your own validity check.
205    
206       General comments about UTF-8 mode
207    
208           1.  An  unbraced  hexadecimal  escape sequence (such as \xb3) matches a
209           two-byte UTF-8 character if the value is greater than 127.
210    
211           2. Octal numbers up to \777 are recognized, and  match  two-byte  UTF-8
212           characters for values greater than \177.
213    
214           3.  Repeat quantifiers apply to complete UTF-8 characters, not to indi-
215           vidual bytes, for example: \x{100}{3}.
216    
217           4. The dot metacharacter matches one UTF-8 character instead of a  sin-
218           gle byte.
219    
220           5.  The  escape sequence \C can be used to match a single byte in UTF-8
221           mode, but its use can lead to some strange effects.  This  facility  is
222           not available in the alternative matching function, pcre_dfa_exec().
223    
224           6.  The  character escapes \b, \B, \d, \D, \s, \S, \w, and \W correctly
225           test characters of any code value, but, by default, the characters that
226           PCRE  recognizes  as digits, spaces, or word characters remain the same
227           set as before, all with values less than 256. This  remains  true  even
228           when  PCRE  is built to include Unicode property support, because to do
229           otherwise would slow down PCRE in many common  cases.  Note  that  this
230           also applies to \b, because it is defined in terms of \w and \W. If you
231           really want to test for a wider sense of, say,  "digit",  you  can  use
232           explicit  Unicode property tests such as \p{Nd}.  Alternatively, if you
233           set the PCRE_UCP option, the way that the  character  escapes  work  is
234           changed  so that Unicode properties are used to determine which charac-
235           ters match. There are more details in the section on generic  character
236           types in the pcrepattern documentation.
237    
238           7.  Similarly,  characters that match the POSIX named character classes
239           are all low-valued characters, unless the PCRE_UCP option is set.
240    
241           8. However, the Perl 5.10 horizontal and vertical  whitespace  matching
242           escapes (\h, \H, \v, and \V) do match all the appropriate Unicode char-
243           acters, whether or not PCRE_UCP is set.
244    
245           9. Case-insensitive matching applies only to  characters  whose  values
246           are  less than 128, unless PCRE is built with Unicode property support.
247           Even when Unicode property support is available, PCRE  still  uses  its
248           own  character  tables when checking the case of low-valued characters,
249           so as not to degrade performance.  The Unicode property information  is
250           used only for characters with higher values. Even when Unicode property
251           support is available, PCRE supports case-insensitive matching only when
252           there  is  a  one-to-one  mapping between a letter's cases. There are a
253           small number of many-to-one mappings in Unicode;  these  are  not  sup-
254           ported by PCRE.
255    
256    
257    AUTHOR
258    
259           Philip Hazel
260           University Computing Service
261           Cambridge CB2 3QH, England.
262    
263           Putting  an actual email address here seems to have been a spam magnet,
264           so I've taken it away. If you want to email me, use  my  two  initials,
265           followed by the two digits 10, at the domain cam.ac.uk.
266    
267    
268    REVISION
269    
270           Last updated: 12 May 2010
271           Copyright (c) 1997-2010 University of Cambridge.
272    ------------------------------------------------------------------------------
273    
274    
275    PCREBUILD(3)                                                      PCREBUILD(3)
276    
277    
278    NAME
279           PCRE - Perl-compatible regular expressions
280    
281    
282    PCRE BUILD-TIME OPTIONS
283    
284           This  document  describes  the  optional  features  of PCRE that can be
285           selected when the library is compiled. It assumes use of the  configure
286           script,  where the optional features are selected or deselected by pro-
287           viding options to configure before running the make  command.  However,
288           the  same  options  can be selected in both Unix-like and non-Unix-like
289           environments using the GUI facility of cmake-gui if you are using CMake
290           instead of configure to build PCRE.
291    
292           There  is  a  lot more information about building PCRE in non-Unix-like
293           environments in the file called NON_UNIX_USE, which is part of the PCRE
294           distribution.  You  should consult this file as well as the README file
295           if you are building in a non-Unix-like environment.
296    
297           The complete list of options for configure (which includes the standard
298           ones  such  as  the  selection  of  the  installation directory) can be
299           obtained by running
300    
301             ./configure --help
302    
303           The following sections include  descriptions  of  options  whose  names
304           begin with --enable or --disable. These settings specify changes to the
305           defaults for the configure command. Because of the way  that  configure
306           works,  --enable  and --disable always come in pairs, so the complemen-
307           tary option always exists as well, but as it specifies the default,  it
308           is not described.
309    
310    
311    C++ SUPPORT
312    
313           By default, the configure script will search for a C++ compiler and C++
314           header files. If it finds them, it automatically builds the C++ wrapper
315           library for PCRE. You can disable this by adding
316    
317             --disable-cpp
318    
319           to the configure command.
320    
321    
322    UTF-8 SUPPORT
323    
324           To build PCRE with support for UTF-8 Unicode character strings, add
325    
326             --enable-utf8
327    
328           to  the  configure  command.  Of  itself, this does not make PCRE treat
329           strings as UTF-8. As well as compiling PCRE with this option, you  also
330           have  have to set the PCRE_UTF8 option when you call the pcre_compile()
331           or pcre_compile2() functions.
332    
333           If you set --enable-utf8 when compiling in an EBCDIC environment,  PCRE
334           expects its input to be either ASCII or UTF-8 (depending on the runtime
335           option). It is not possible to support both EBCDIC and UTF-8  codes  in
336           the  same  version  of  the  library.  Consequently,  --enable-utf8 and
337           --enable-ebcdic are mutually exclusive.
338    
339    
340    UNICODE CHARACTER PROPERTY SUPPORT
341    
342           UTF-8 support allows PCRE to process character values greater than  255
343           in  the  strings that it handles. On its own, however, it does not pro-
344           vide any facilities for accessing the properties of such characters. If
345           you  want  to  be able to use the pattern escapes \P, \p, and \X, which
346           refer to Unicode character properties, you must add
347    
348             --enable-unicode-properties
349    
350           to the configure command. This implies UTF-8 support, even if you  have
351           not explicitly requested it.
352    
353           Including  Unicode  property  support  adds around 30K of tables to the
354           PCRE library. Only the general category properties such as  Lu  and  Nd
355           are supported. Details are given in the pcrepattern documentation.
356    
357    
358    CODE VALUE OF NEWLINE
359    
360           By  default,  PCRE interprets the linefeed (LF) character as indicating
361           the end of a line. This is the normal newline  character  on  Unix-like
362           systems.  You  can compile PCRE to use carriage return (CR) instead, by
363           adding
364    
365             --enable-newline-is-cr
366    
367           to the  configure  command.  There  is  also  a  --enable-newline-is-lf
368           option, which explicitly specifies linefeed as the newline character.
369    
370           Alternatively, you can specify that line endings are to be indicated by
371           the two character sequence CRLF. If you want this, add
372    
373             --enable-newline-is-crlf
374    
375           to the configure command. There is a fourth option, specified by
376    
377             --enable-newline-is-anycrlf
378    
379           which causes PCRE to recognize any of the three sequences  CR,  LF,  or
380           CRLF as indicating a line ending. Finally, a fifth option, specified by
381    
382             --enable-newline-is-any
383    
384           causes PCRE to recognize any Unicode newline sequence.
385    
386           Whatever  line  ending convention is selected when PCRE is built can be
387           overridden when the library functions are called. At build time  it  is
388           conventional to use the standard for your operating system.
389    
390    
391    WHAT \R MATCHES
392    
393           By  default,  the  sequence \R in a pattern matches any Unicode newline
394           sequence, whatever has been selected as the line  ending  sequence.  If
395           you specify
396    
397             --enable-bsr-anycrlf
398    
399           the  default  is changed so that \R matches only CR, LF, or CRLF. What-
400           ever is selected when PCRE is built can be overridden when the  library
401           functions are called.
402    
403    
404    BUILDING SHARED AND STATIC LIBRARIES
405    
406           The  PCRE building process uses libtool to build both shared and static
407           Unix libraries by default. You can suppress one of these by adding  one
408           of
409    
410             --disable-shared
411             --disable-static
412    
413           to the configure command, as required.
414    
415    
416    POSIX MALLOC USAGE
417    
418           When PCRE is called through the POSIX interface (see the pcreposix doc-
419           umentation), additional working storage is  required  for  holding  the
420           pointers  to capturing substrings, because PCRE requires three integers
421           per substring, whereas the POSIX interface provides only  two.  If  the
422           number of expected substrings is small, the wrapper function uses space
423           on the stack, because this is faster than using malloc() for each call.
424           The default threshold above which the stack is no longer used is 10; it
425           can be changed by adding a setting such as
426    
427             --with-posix-malloc-threshold=20
428    
429           to the configure command.
430    
431    
432    HANDLING VERY LARGE PATTERNS
433    
434           Within a compiled pattern, offset values are used  to  point  from  one
435           part  to another (for example, from an opening parenthesis to an alter-
436           nation metacharacter). By default, two-byte values are used  for  these
437           offsets,  leading  to  a  maximum size for a compiled pattern of around
438           64K. This is sufficient to handle all but the most  gigantic  patterns.
439           Nevertheless,  some  people do want to process truyl enormous patterns,
440           so it is possible to compile PCRE to use three-byte or  four-byte  off-
441           sets by adding a setting such as
442    
443             --with-link-size=3
444    
445           to  the  configure  command.  The value given must be 2, 3, or 4. Using
446           longer offsets slows down the operation of PCRE because it has to  load
447           additional bytes when handling them.
448    
449    
450    AVOIDING EXCESSIVE STACK USAGE
451    
452           When matching with the pcre_exec() function, PCRE implements backtrack-
453           ing by making recursive calls to an internal function  called  match().
454           In  environments  where  the size of the stack is limited, this can se-
455           verely limit PCRE's operation. (The Unix environment does  not  usually
456           suffer from this problem, but it may sometimes be necessary to increase
457           the maximum stack size.  There is a discussion in the  pcrestack  docu-
458           mentation.)  An alternative approach to recursion that uses memory from
459           the heap to remember data, instead of using recursive  function  calls,
460           has  been  implemented to work round the problem of limited stack size.
461           If you want to build a version of PCRE that works this way, add
462    
463             --disable-stack-for-recursion
464    
465           to the configure command. With this configuration, PCRE  will  use  the
466           pcre_stack_malloc  and pcre_stack_free variables to call memory manage-
467           ment functions. By default these point to malloc() and free(), but  you
468           can replace the pointers so that your own functions are used instead.
469    
470           Separate  functions  are  provided  rather  than  using pcre_malloc and
471           pcre_free because the  usage  is  very  predictable:  the  block  sizes
472           requested  are  always  the  same,  and  the blocks are always freed in
473           reverse order. A calling program might be able to  implement  optimized
474           functions  that  perform  better  than  malloc()  and free(). PCRE runs
475           noticeably more slowly when built in this way. This option affects only
476           the pcre_exec() function; it is not relevant for pcre_dfa_exec().
477    
478    
479    LIMITING PCRE RESOURCE USAGE
480    
481           Internally,  PCRE has a function called match(), which it calls repeat-
482           edly  (sometimes  recursively)  when  matching  a  pattern   with   the
483           pcre_exec()  function.  By controlling the maximum number of times this
484           function may be called during a single matching operation, a limit  can
485           be  placed  on  the resources used by a single call to pcre_exec(). The
486           limit can be changed at run time, as described in the pcreapi  documen-
487           tation.  The default is 10 million, but this can be changed by adding a
488           setting such as
489    
490             --with-match-limit=500000
491    
492           to  the  configure  command.  This  setting  has  no  effect   on   the
493           pcre_dfa_exec() matching function.
494    
495           In  some  environments  it is desirable to limit the depth of recursive
496           calls of match() more strictly than the total number of calls, in order
497           to  restrict  the maximum amount of stack (or heap, if --disable-stack-
498           for-recursion is specified) that is used. A second limit controls this;
499           it  defaults  to  the  value  that is set for --with-match-limit, which
500           imposes no additional constraints. However, you can set a  lower  limit
501           by adding, for example,
502    
503             --with-match-limit-recursion=10000
504    
505           to  the  configure  command.  This  value can also be overridden at run
506           time.
507    
508    
509    CREATING CHARACTER TABLES AT BUILD TIME
510    
511           PCRE uses fixed tables for processing characters whose code values  are
512           less  than 256. By default, PCRE is built with a set of tables that are
513           distributed in the file pcre_chartables.c.dist. These  tables  are  for
514           ASCII codes only. If you add
515    
516             --enable-rebuild-chartables
517    
518           to  the  configure  command, the distributed tables are no longer used.
519           Instead, a program called dftables is compiled and  run.  This  outputs
520           the source for new set of tables, created in the default locale of your
521           C runtime system. (This method of replacing the tables does not work if
522           you  are cross compiling, because dftables is run on the local host. If
523           you need to create alternative tables when cross  compiling,  you  will
524           have to do so "by hand".)
525    
526    
527    USING EBCDIC CODE
528    
529           PCRE  assumes  by  default that it will run in an environment where the
530           character code is ASCII (or Unicode, which is  a  superset  of  ASCII).
531           This  is  the  case for most computer operating systems. PCRE can, how-
532           ever, be compiled to run in an EBCDIC environment by adding
533    
534             --enable-ebcdic
535    
536           to the configure command. This setting implies --enable-rebuild-charta-
537           bles.  You  should  only  use  it if you know that you are in an EBCDIC
538           environment (for example,  an  IBM  mainframe  operating  system).  The
539           --enable-ebcdic option is incompatible with --enable-utf8.
540    
541    
542    PCREGREP OPTIONS FOR COMPRESSED FILE SUPPORT
543    
544           By default, pcregrep reads all files as plain text. You can build it so
545           that it recognizes files whose names end in .gz or .bz2, and reads them
546           with libz or libbz2, respectively, by adding one or both of
547    
548             --enable-pcregrep-libz
549             --enable-pcregrep-libbz2
550    
551           to the configure command. These options naturally require that the rel-
552           evant libraries are installed on your system. Configuration  will  fail
553           if they are not.
554    
555    
556    PCRETEST OPTION FOR LIBREADLINE SUPPORT
557    
558           If you add
559    
560             --enable-pcretest-libreadline
561    
562           to  the  configure  command,  pcretest  is  linked with the libreadline
563           library, and when its input is from a terminal, it reads it  using  the
564           readline() function. This provides line-editing and history facilities.
565           Note that libreadline is GPL-licensed, so if you distribute a binary of
566           pcretest linked in this way, there may be licensing issues.
567    
568           Setting  this  option  causes  the -lreadline option to be added to the
569           pcretest build. In many operating environments with  a  sytem-installed
570           libreadline this is sufficient. However, in some environments (e.g.  if
571           an unmodified distribution version of readline is in use),  some  extra
572           configuration  may  be necessary. The INSTALL file for libreadline says
573           this:
574    
575             "Readline uses the termcap functions, but does not link with the
576             termcap or curses library itself, allowing applications which link
577             with readline the to choose an appropriate library."
578    
579           If your environment has not been set up so that an appropriate  library
580           is automatically included, you may need to add something like
581    
582             LIBS="-ncurses"
583    
584           immediately before the configure command.
585    
586    
587    SEE ALSO
588    
589           pcreapi(3), pcre_config(3).
590    
591    
592    AUTHOR
593    
594           Philip Hazel
595           University Computing Service
596           Cambridge CB2 3QH, England.
597    
598    
599    REVISION
600    
601           Last updated: 29 September 2009
602           Copyright (c) 1997-2009 University of Cambridge.
603    ------------------------------------------------------------------------------
604    
605    
606    PCREMATCHING(3)                                                PCREMATCHING(3)
607    
608    
609    NAME
610           PCRE - Perl-compatible regular expressions
611    
612    
613    PCRE MATCHING ALGORITHMS
614    
615           This document describes the two different algorithms that are available
616           in PCRE for matching a compiled regular expression against a given sub-
617           ject  string.  The  "standard"  algorithm  is  the  one provided by the
618           pcre_exec() function.  This works in the same was  as  Perl's  matching
619           function, and provides a Perl-compatible matching operation.
620    
621           An  alternative  algorithm is provided by the pcre_dfa_exec() function;
622           this operates in a different way, and is not  Perl-compatible.  It  has
623           advantages  and disadvantages compared with the standard algorithm, and
624           these are described below.
625    
626           When there is only one possible way in which a given subject string can
627           match  a pattern, the two algorithms give the same answer. A difference
628           arises, however, when there are multiple possibilities. For example, if
629           the pattern
630    
631             ^<.*>
632    
633           is matched against the string
634    
635             <something> <something else> <something further>
636    
637           there are three possible answers. The standard algorithm finds only one
638           of them, whereas the alternative algorithm finds all three.
639    
640    
641    REGULAR EXPRESSIONS AS TREES
642    
643           The set of strings that are matched by a regular expression can be rep-
644           resented  as  a  tree structure. An unlimited repetition in the pattern
645           makes the tree of infinite size, but it is still a tree.  Matching  the
646           pattern  to a given subject string (from a given starting point) can be
647           thought of as a search of the tree.  There are two  ways  to  search  a
648           tree:  depth-first  and  breadth-first, and these correspond to the two
649           matching algorithms provided by PCRE.
650    
651    
652    THE STANDARD MATCHING ALGORITHM
653    
654           In the terminology of Jeffrey Friedl's book "Mastering Regular  Expres-
655           sions",  the  standard  algorithm  is an "NFA algorithm". It conducts a
656           depth-first search of the pattern tree. That is, it  proceeds  along  a
657           single path through the tree, checking that the subject matches what is
658           required. When there is a mismatch, the algorithm  tries  any  alterna-
659           tives  at  the  current point, and if they all fail, it backs up to the
660           previous branch point in the  tree,  and  tries  the  next  alternative
661           branch  at  that  level.  This often involves backing up (moving to the
662           left) in the subject string as well.  The  order  in  which  repetition
663           branches  are  tried  is controlled by the greedy or ungreedy nature of
664           the quantifier.
665    
666           If a leaf node is reached, a matching string has  been  found,  and  at
667           that  point the algorithm stops. Thus, if there is more than one possi-
668           ble match, this algorithm returns the first one that it finds.  Whether
669           this  is the shortest, the longest, or some intermediate length depends
670           on the way the greedy and ungreedy repetition quantifiers are specified
671           in the pattern.
672    
673           Because  it  ends  up  with a single path through the tree, it is rela-
674           tively straightforward for this algorithm to keep  track  of  the  sub-
675           strings  that  are  matched  by portions of the pattern in parentheses.
676           This provides support for capturing parentheses and back references.
677    
678    
679    THE ALTERNATIVE MATCHING ALGORITHM
680    
681           This algorithm conducts a breadth-first search of  the  tree.  Starting
682           from  the  first  matching  point  in the subject, it scans the subject
683           string from left to right, once, character by character, and as it does
684           this,  it remembers all the paths through the tree that represent valid
685           matches. In Friedl's terminology, this is a kind  of  "DFA  algorithm",
686           though  it is not implemented as a traditional finite state machine (it
687           keeps multiple states active simultaneously).
688    
689           Although the general principle of this matching algorithm  is  that  it
690           scans  the subject string only once, without backtracking, there is one
691           exception: when a lookaround assertion is encountered,  the  characters
692           following  or  preceding  the  current  point  have to be independently
693           inspected.
694    
695           The scan continues until either the end of the subject is  reached,  or
696           there  are  no more unterminated paths. At this point, terminated paths
697           represent the different matching possibilities (if there are none,  the
698           match  has  failed).   Thus,  if there is more than one possible match,
699           this algorithm finds all of them, and in particular, it finds the long-
700           est.  There  is  an  option to stop the algorithm after the first match
701           (which is necessarily the shortest) is found.
702    
703           Note that all the matches that are found start at the same point in the
704           subject. If the pattern
705    
706             cat(er(pillar)?)
707    
708           is  matched  against the string "the caterpillar catchment", the result
709           will be the three strings "cat", "cater", and "caterpillar" that  start
710           at the fourth character of the subject. The algorithm does not automat-
711           ically move on to find matches that start at later positions.
712    
713           There are a number of features of PCRE regular expressions that are not
714           supported by the alternative matching algorithm. They are as follows:
715    
716           1.  Because  the  algorithm  finds  all possible matches, the greedy or
717           ungreedy nature of repetition quantifiers is not relevant.  Greedy  and
718           ungreedy quantifiers are treated in exactly the same way. However, pos-
719           sessive quantifiers can make a difference when what follows could  also
720           match what is quantified, for example in a pattern like this:
721    
722             ^a++\w!
723    
724           This  pattern matches "aaab!" but not "aaa!", which would be matched by
725           a non-possessive quantifier. Similarly, if an atomic group is  present,
726           it  is matched as if it were a standalone pattern at the current point,
727           and the longest match is then "locked in" for the rest of  the  overall
728           pattern.
729    
730           2. When dealing with multiple paths through the tree simultaneously, it
731           is not straightforward to keep track of  captured  substrings  for  the
732           different  matching  possibilities,  and  PCRE's implementation of this
733           algorithm does not attempt to do this. This means that no captured sub-
734           strings are available.
735    
736           3.  Because no substrings are captured, back references within the pat-
737           tern are not supported, and cause errors if encountered.
738    
739           4. For the same reason, conditional expressions that use  a  backrefer-
740           ence  as  the  condition or test for a specific group recursion are not
741           supported.
742    
743           5. Because many paths through the tree may be  active,  the  \K  escape
744           sequence, which resets the start of the match when encountered (but may
745           be on some paths and not on others), is not  supported.  It  causes  an
746           error if encountered.
747    
748           6.  Callouts  are  supported, but the value of the capture_top field is
749           always 1, and the value of the capture_last field is always -1.
750    
751           7. The \C escape sequence, which (in the standard algorithm) matches  a
752           single  byte, even in UTF-8 mode, is not supported because the alterna-
753           tive algorithm moves through the subject  string  one  character  at  a
754           time, for all active paths through the tree.
755    
756           8.  Except for (*FAIL), the backtracking control verbs such as (*PRUNE)
757           are not supported. (*FAIL) is supported, and  behaves  like  a  failing
758           negative assertion.
759    
760    
761    ADVANTAGES OF THE ALTERNATIVE ALGORITHM
762    
763           Using  the alternative matching algorithm provides the following advan-
764           tages:
765    
766           1. All possible matches (at a single point in the subject) are automat-
767           ically  found,  and  in particular, the longest match is found. To find
768           more than one match using the standard algorithm, you have to do kludgy
769           things with callouts.
770    
771           2.  Because  the  alternative  algorithm  scans the subject string just
772           once, and never needs to backtrack, it is possible to  pass  very  long
773           subject  strings  to  the matching function in several pieces, checking
774           for partial matching each time.  The  pcrepartial  documentation  gives
775           details of partial matching.
776    
777    
778    DISADVANTAGES OF THE ALTERNATIVE ALGORITHM
779    
780           The alternative algorithm suffers from a number of disadvantages:
781    
782           1.  It  is  substantially  slower  than the standard algorithm. This is
783           partly because it has to search for all possible matches, but  is  also
784           because it is less susceptible to optimization.
785    
786           2. Capturing parentheses and back references are not supported.
787    
788           3. Although atomic groups are supported, their use does not provide the
789           performance advantage that it does for the standard algorithm.
790    
791    
792    AUTHOR
793    
794           Philip Hazel
795           University Computing Service
796           Cambridge CB2 3QH, England.
797    
798    
799    REVISION
800    
801           Last updated: 29 September 2009
802           Copyright (c) 1997-2009 University of Cambridge.
803    ------------------------------------------------------------------------------
804    
805    
806    PCREAPI(3)                                                          PCREAPI(3)
807    
808    
809    NAME
810           PCRE - Perl-compatible regular expressions
811    
812    
813    PCRE NATIVE API
814    
815           #include <pcre.h>
816    
817           pcre *pcre_compile(const char *pattern, int options,
818                const char **errptr, int *erroffset,
819                const unsigned char *tableptr);
820    
821           pcre *pcre_compile2(const char *pattern, int options,
822                int *errorcodeptr,
823                const char **errptr, int *erroffset,
824                const unsigned char *tableptr);
825    
826           pcre_extra *pcre_study(const pcre *code, int options,
827                const char **errptr);
828    
829           int pcre_exec(const pcre *code, const pcre_extra *extra,
830                const char *subject, int length, int startoffset,
831                int options, int *ovector, int ovecsize);
832    
833           int pcre_dfa_exec(const pcre *code, const pcre_extra *extra,
834                const char *subject, int length, int startoffset,
835                int options, int *ovector, int ovecsize,
836                int *workspace, int wscount);
837    
838           int pcre_copy_named_substring(const pcre *code,
839                const char *subject, int *ovector,
840                int stringcount, const char *stringname,
841                char *buffer, int buffersize);
842    
843           int pcre_copy_substring(const char *subject, int *ovector,
844                int stringcount, int stringnumber, char *buffer,
845                int buffersize);
846    
847           int pcre_get_named_substring(const pcre *code,
848                const char *subject, int *ovector,
849                int stringcount, const char *stringname,
850                const char **stringptr);
851    
852           int pcre_get_stringnumber(const pcre *code,
853                const char *name);
854    
855           int pcre_get_stringtable_entries(const pcre *code,
856                const char *name, char **first, char **last);
857    
858           int pcre_get_substring(const char *subject, int *ovector,
859                int stringcount, int stringnumber,
860                const char **stringptr);
861    
862           int pcre_get_substring_list(const char *subject,
863                int *ovector, int stringcount, const char ***listptr);
864    
865           void pcre_free_substring(const char *stringptr);
866    
867           void pcre_free_substring_list(const char **stringptr);
868    
869           const unsigned char *pcre_maketables(void);
870    
871           int pcre_fullinfo(const pcre *code, const pcre_extra *extra,
872                int what, void *where);
873    
874           int pcre_info(const pcre *code, int *optptr, int *firstcharptr);
875    
876           int pcre_refcount(pcre *code, int adjust);
877    
878           int pcre_config(int what, void *where);
879    
880           char *pcre_version(void);
881    
882           void *(*pcre_malloc)(size_t);
883    
884           void (*pcre_free)(void *);
885    
886           void *(*pcre_stack_malloc)(size_t);
887    
888           void (*pcre_stack_free)(void *);
889    
890           int (*pcre_callout)(pcre_callout_block *);
891    
892    
893    PCRE API OVERVIEW
894    
895           PCRE has its own native API, which is described in this document. There
896           are also some wrapper functions that correspond to  the  POSIX  regular
897           expression  API.  These  are  described in the pcreposix documentation.
898           Both of these APIs define a set of C function calls. A C++  wrapper  is
899           distributed with PCRE. It is documented in the pcrecpp page.
900    
901           The  native  API  C  function prototypes are defined in the header file
902           pcre.h, and on Unix systems the library itself is called  libpcre.   It
903           can normally be accessed by adding -lpcre to the command for linking an
904           application  that  uses  PCRE.  The  header  file  defines  the  macros
905           PCRE_MAJOR  and  PCRE_MINOR to contain the major and minor release num-
906           bers for the library.  Applications can use these  to  include  support
907           for different releases of PCRE.
908    
909           In a Windows environment, if you want to statically link an application
910           program against a non-dll pcre.a  file,  you  must  define  PCRE_STATIC
911           before  including  pcre.h or pcrecpp.h, because otherwise the pcre_mal-
912           loc()   and   pcre_free()   exported   functions   will   be   declared
913           __declspec(dllimport), with unwanted results.
914    
915           The   functions   pcre_compile(),  pcre_compile2(),  pcre_study(),  and
916           pcre_exec() are used for compiling and matching regular expressions  in
917           a  Perl-compatible  manner. A sample program that demonstrates the sim-
918           plest way of using them is provided in the file  called  pcredemo.c  in
919           the PCRE source distribution. A listing of this program is given in the
920           pcredemo documentation, and the pcresample documentation describes  how
921           to compile and run it.
922    
923           A second matching function, pcre_dfa_exec(), which is not Perl-compati-
924           ble, is also provided. This uses a different algorithm for  the  match-
925           ing.  The  alternative algorithm finds all possible matches (at a given
926           point in the subject), and scans the subject just  once  (unless  there
927           are  lookbehind  assertions).  However,  this algorithm does not return
928           captured substrings. A description of the two matching  algorithms  and
929           their  advantages  and disadvantages is given in the pcrematching docu-
930           mentation.
931    
932           In addition to the main compiling and  matching  functions,  there  are
933           convenience functions for extracting captured substrings from a subject
934           string that is matched by pcre_exec(). They are:
935    
936             pcre_copy_substring()
937             pcre_copy_named_substring()
938             pcre_get_substring()
939             pcre_get_named_substring()
940             pcre_get_substring_list()
941             pcre_get_stringnumber()
942             pcre_get_stringtable_entries()
943    
944           pcre_free_substring() and pcre_free_substring_list() are also provided,
945           to free the memory used for extracted strings.
946    
947           The  function  pcre_maketables()  is  used  to build a set of character
948           tables  in  the  current  locale   for   passing   to   pcre_compile(),
949           pcre_exec(),  or  pcre_dfa_exec(). This is an optional facility that is
950           provided for specialist use.  Most  commonly,  no  special  tables  are
951           passed,  in  which case internal tables that are generated when PCRE is
952           built are used.
953    
954           The function pcre_fullinfo() is used to find out  information  about  a
955           compiled  pattern; pcre_info() is an obsolete version that returns only
956           some of the available information, but is retained for  backwards  com-
957           patibility.   The function pcre_version() returns a pointer to a string
958           containing the version of PCRE and its date of release.
959    
960           The function pcre_refcount() maintains a  reference  count  in  a  data
961           block  containing  a compiled pattern. This is provided for the benefit
962           of object-oriented applications.
963    
964           The global variables pcre_malloc and pcre_free  initially  contain  the
965           entry  points  of  the  standard malloc() and free() functions, respec-
966           tively. PCRE calls the memory management functions via these variables,
967           so  a  calling  program  can replace them if it wishes to intercept the
968           calls. This should be done before calling any PCRE functions.
969    
970           The global variables pcre_stack_malloc  and  pcre_stack_free  are  also
971           indirections  to  memory  management functions. These special functions
972           are used only when PCRE is compiled to use  the  heap  for  remembering
973           data, instead of recursive function calls, when running the pcre_exec()
974           function. See the pcrebuild documentation for  details  of  how  to  do
975           this.  It  is  a non-standard way of building PCRE, for use in environ-
976           ments that have limited stacks. Because of the greater  use  of  memory
977           management,  it  runs  more  slowly. Separate functions are provided so
978           that special-purpose external code can be  used  for  this  case.  When
979           used,  these  functions  are always called in a stack-like manner (last
980           obtained, first freed), and always for memory blocks of the same  size.
981           There  is  a discussion about PCRE's stack usage in the pcrestack docu-
982           mentation.
983    
984           The global variable pcre_callout initially contains NULL. It can be set
985           by  the  caller  to  a "callout" function, which PCRE will then call at
986           specified points during a matching operation. Details are given in  the
987           pcrecallout documentation.
988    
989    
990    NEWLINES
991    
992           PCRE  supports five different conventions for indicating line breaks in
993           strings: a single CR (carriage return) character, a  single  LF  (line-
994           feed) character, the two-character sequence CRLF, any of the three pre-
995           ceding, or any Unicode newline sequence. The Unicode newline  sequences
996           are  the  three just mentioned, plus the single characters VT (vertical
997           tab, U+000B), FF (formfeed, U+000C), NEL (next line, U+0085), LS  (line
998           separator, U+2028), and PS (paragraph separator, U+2029).
999    
1000           Each  of  the first three conventions is used by at least one operating
1001           system as its standard newline sequence. When PCRE is built, a  default
1002           can  be  specified.  The default default is LF, which is the Unix stan-
1003           dard. When PCRE is run, the default can be overridden,  either  when  a
1004           pattern is compiled, or when it is matched.
1005    
1006           At compile time, the newline convention can be specified by the options
1007           argument of pcre_compile(), or it can be specified by special  text  at
1008           the start of the pattern itself; this overrides any other settings. See
1009           the pcrepattern page for details of the special character sequences.
1010    
1011           In the PCRE documentation the word "newline" is used to mean "the char-
1012           acter  or pair of characters that indicate a line break". The choice of
1013           newline convention affects the handling of  the  dot,  circumflex,  and
1014           dollar metacharacters, the handling of #-comments in /x mode, and, when
1015           CRLF is a recognized line ending sequence, the match position  advance-
1016           ment for a non-anchored pattern. There is more detail about this in the
1017           section on pcre_exec() options below.
1018    
1019           The choice of newline convention does not affect the interpretation  of
1020           the  \n  or  \r  escape  sequences, nor does it affect what \R matches,
1021           which is controlled in a similar way, but by separate options.
1022    
1023    
1024    MULTITHREADING
1025    
1026           The PCRE functions can be used in  multi-threading  applications,  with
1027           the  proviso  that  the  memory  management  functions  pointed  to  by
1028           pcre_malloc, pcre_free, pcre_stack_malloc, and pcre_stack_free, and the
1029           callout function pointed to by pcre_callout, are shared by all threads.
1030    
1031           The  compiled form of a regular expression is not altered during match-
1032           ing, so the same compiled pattern can safely be used by several threads
1033           at once.
1034    
1035    
1036    SAVING PRECOMPILED PATTERNS FOR LATER USE
1037    
1038           The compiled form of a regular expression can be saved and re-used at a
1039           later time, possibly by a different program, and even on a  host  other
1040           than  the  one  on  which  it  was  compiled.  Details are given in the
1041           pcreprecompile documentation. However, compiling a  regular  expression
1042           with  one version of PCRE for use with a different version is not guar-
1043           anteed to work and may cause crashes.
1044    
1045    
1046    CHECKING BUILD-TIME OPTIONS
1047    
1048           int pcre_config(int what, void *where);
1049    
1050           The function pcre_config() makes it possible for a PCRE client to  dis-
1051           cover which optional features have been compiled into the PCRE library.
1052           The pcrebuild documentation has more details about these optional  fea-
1053           tures.
1054    
1055           The  first  argument  for pcre_config() is an integer, specifying which
1056           information is required; the second argument is a pointer to a variable
1057           into  which  the  information  is  placed. The following information is
1058           available:
1059    
1060             PCRE_CONFIG_UTF8
1061    
1062           The output is an integer that is set to one if UTF-8 support is  avail-
1063           able; otherwise it is set to zero.
1064    
1065             PCRE_CONFIG_UNICODE_PROPERTIES
1066    
1067           The  output  is  an  integer  that is set to one if support for Unicode
1068           character properties is available; otherwise it is set to zero.
1069    
1070             PCRE_CONFIG_NEWLINE
1071    
1072           The output is an integer whose value specifies  the  default  character
1073           sequence  that is recognized as meaning "newline". The four values that
1074           are supported are: 10 for LF, 13 for CR, 3338 for CRLF, -2 for ANYCRLF,
1075           and  -1  for  ANY.  Though they are derived from ASCII, the same values
1076           are returned in EBCDIC environments. The default should normally corre-
1077           spond to the standard sequence for your operating system.
1078    
1079             PCRE_CONFIG_BSR
1080    
1081           The output is an integer whose value indicates what character sequences
1082           the \R escape sequence matches by default. A value of 0 means  that  \R
1083           matches  any  Unicode  line ending sequence; a value of 1 means that \R
1084           matches only CR, LF, or CRLF. The default can be overridden when a pat-
1085           tern is compiled or matched.
1086    
1087             PCRE_CONFIG_LINK_SIZE
1088    
1089           The  output  is  an  integer that contains the number of bytes used for
1090           internal linkage in compiled regular expressions. The value is 2, 3, or
1091           4.  Larger  values  allow larger regular expressions to be compiled, at
1092           the expense of slower matching. The default value of  2  is  sufficient
1093           for  all  but  the  most massive patterns, since it allows the compiled
1094           pattern to be up to 64K in size.
1095    
1096             PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
1097    
1098           The output is an integer that contains the threshold  above  which  the
1099           POSIX  interface  uses malloc() for output vectors. Further details are
1100           given in the pcreposix documentation.
1101    
1102             PCRE_CONFIG_MATCH_LIMIT
1103    
1104           The output is a long integer that gives the default limit for the  num-
1105           ber  of  internal  matching  function calls in a pcre_exec() execution.
1106           Further details are given with pcre_exec() below.
1107    
1108             PCRE_CONFIG_MATCH_LIMIT_RECURSION
1109    
1110           The output is a long integer that gives the default limit for the depth
1111           of   recursion  when  calling  the  internal  matching  function  in  a
1112           pcre_exec() execution.  Further  details  are  given  with  pcre_exec()
1113           below.
1114    
1115             PCRE_CONFIG_STACKRECURSE
1116    
1117           The  output is an integer that is set to one if internal recursion when
1118           running pcre_exec() is implemented by recursive function calls that use
1119           the  stack  to remember their state. This is the usual way that PCRE is
1120           compiled. The output is zero if PCRE was compiled to use blocks of data
1121           on  the  heap  instead  of  recursive  function  calls.  In  this case,
1122           pcre_stack_malloc and  pcre_stack_free  are  called  to  manage  memory
1123           blocks on the heap, thus avoiding the use of the stack.
1124    
1125    
1126    COMPILING A PATTERN
1127    
1128           pcre *pcre_compile(const char *pattern, int options,
1129                const char **errptr, int *erroffset,
1130                const unsigned char *tableptr);
1131    
1132           pcre *pcre_compile2(const char *pattern, int options,
1133                int *errorcodeptr,
1134                const char **errptr, int *erroffset,
1135                const unsigned char *tableptr);
1136    
1137           Either of the functions pcre_compile() or pcre_compile2() can be called
1138           to compile a pattern into an internal form. The only difference between
1139           the  two interfaces is that pcre_compile2() has an additional argument,
1140           errorcodeptr, via which a numerical error  code  can  be  returned.  To
1141           avoid  too  much repetition, we refer just to pcre_compile() below, but
1142           the information applies equally to pcre_compile2().
1143    
1144           The pattern is a C string terminated by a binary zero, and is passed in
1145           the  pattern  argument.  A  pointer to a single block of memory that is
1146           obtained via pcre_malloc is returned. This contains the  compiled  code
1147           and related data. The pcre type is defined for the returned block; this
1148           is a typedef for a structure whose contents are not externally defined.
1149           It is up to the caller to free the memory (via pcre_free) when it is no
1150           longer required.
1151    
1152           Although the compiled code of a PCRE regex is relocatable, that is,  it
1153           does not depend on memory location, the complete pcre data block is not
1154           fully relocatable, because it may contain a copy of the tableptr  argu-
1155           ment, which is an address (see below).
1156    
1157           The options argument contains various bit settings that affect the com-
1158           pilation. It should be zero if no options are required.  The  available
1159           options  are  described  below. Some of them (in particular, those that
1160           are compatible with Perl, but some others as well) can also be set  and
1161           unset  from  within  the  pattern  (see the detailed description in the
1162           pcrepattern documentation). For those options that can be different  in
1163           different  parts  of  the pattern, the contents of the options argument
1164           specifies their settings at the start of compilation and execution. The
1165           PCRE_ANCHORED, PCRE_BSR_xxx, and PCRE_NEWLINE_xxx options can be set at
1166           the time of matching as well as at compile time.
1167    
1168           If errptr is NULL, pcre_compile() returns NULL immediately.  Otherwise,
1169           if  compilation  of  a  pattern fails, pcre_compile() returns NULL, and
1170           sets the variable pointed to by errptr to point to a textual error mes-
1171           sage. This is a static string that is part of the library. You must not
1172           try to free it. The byte offset from the start of the  pattern  to  the
1173           character  that  was  being  processed when the error was discovered is
1174           placed in the variable pointed to by erroffset, which must not be NULL.
1175           If  it  is,  an  immediate error is given. Some errors are not detected
1176           until checks are carried out when the whole pattern has  been  scanned;
1177           in this case the offset is set to the end of the pattern.
1178    
1179           If  pcre_compile2()  is  used instead of pcre_compile(), and the error-
1180           codeptr argument is not NULL, a non-zero error code number is  returned
1181           via  this argument in the event of an error. This is in addition to the
1182           textual error message. Error codes and messages are listed below.
1183    
1184           If the final argument, tableptr, is NULL, PCRE uses a  default  set  of
1185           character  tables  that  are  built  when  PCRE  is compiled, using the
1186           default C locale. Otherwise, tableptr must be an address  that  is  the
1187           result  of  a  call to pcre_maketables(). This value is stored with the
1188           compiled pattern, and used again by pcre_exec(), unless  another  table
1189           pointer is passed to it. For more discussion, see the section on locale
1190           support below.
1191    
1192           This code fragment shows a typical straightforward  call  to  pcre_com-
1193           pile():
1194    
1195             pcre *re;
1196             const char *error;
1197             int erroffset;
1198             re = pcre_compile(
1199               "^A.*Z",          /* the pattern */
1200               0,                /* default options */
1201               &error,           /* for error message */
1202               &erroffset,       /* for error offset */
1203               NULL);            /* use default character tables */
1204    
1205           The  following  names  for option bits are defined in the pcre.h header
1206           file:
1207    
1208             PCRE_ANCHORED
1209    
1210           If this bit is set, the pattern is forced to be "anchored", that is, it
1211           is  constrained to match only at the first matching point in the string
1212           that is being searched (the "subject string"). This effect can also  be
1213           achieved  by appropriate constructs in the pattern itself, which is the
1214           only way to do it in Perl.
1215    
1216             PCRE_AUTO_CALLOUT
1217    
1218           If this bit is set, pcre_compile() automatically inserts callout items,
1219           all  with  number  255, before each pattern item. For discussion of the
1220           callout facility, see the pcrecallout documentation.
1221    
1222             PCRE_BSR_ANYCRLF
1223             PCRE_BSR_UNICODE
1224    
1225           These options (which are mutually exclusive) control what the \R escape
1226           sequence  matches.  The choice is either to match only CR, LF, or CRLF,
1227           or to match any Unicode newline sequence. The default is specified when
1228           PCRE is built. It can be overridden from within the pattern, or by set-
1229           ting an option when a compiled pattern is matched.
1230    
1231             PCRE_CASELESS
1232    
1233           If this bit is set, letters in the pattern match both upper  and  lower
1234           case  letters.  It  is  equivalent  to  Perl's /i option, and it can be
1235           changed within a pattern by a (?i) option setting. In UTF-8 mode,  PCRE
1236           always  understands the concept of case for characters whose values are
1237           less than 128, so caseless matching is always possible. For  characters
1238           with  higher  values,  the concept of case is supported if PCRE is com-
1239           piled with Unicode property support, but not otherwise. If you want  to
1240           use  caseless  matching  for  characters 128 and above, you must ensure
1241           that PCRE is compiled with Unicode property support  as  well  as  with
1242           UTF-8 support.
1243    
1244             PCRE_DOLLAR_ENDONLY
1245    
1246           If  this bit is set, a dollar metacharacter in the pattern matches only
1247           at the end of the subject string. Without this option,  a  dollar  also
1248           matches  immediately before a newline at the end of the string (but not
1249           before any other newlines). The PCRE_DOLLAR_ENDONLY option  is  ignored
1250           if  PCRE_MULTILINE  is  set.   There is no equivalent to this option in
1251           Perl, and no way to set it within a pattern.
1252    
1253             PCRE_DOTALL
1254    
1255           If this bit is set, a dot metacharater in the pattern matches all char-
1256           acters,  including  those that indicate newline. Without it, a dot does
1257           not match when the current position is at a  newline.  This  option  is
1258           equivalent  to Perl's /s option, and it can be changed within a pattern
1259           by a (?s) option setting. A negative class such as [^a] always  matches
1260           newline characters, independent of the setting of this option.
1261    
1262             PCRE_DUPNAMES
1263    
1264           If  this  bit is set, names used to identify capturing subpatterns need
1265           not be unique. This can be helpful for certain types of pattern when it
1266           is  known  that  only  one instance of the named subpattern can ever be
1267           matched. There are more details of named subpatterns  below;  see  also
1268           the pcrepattern documentation.
1269    
1270             PCRE_EXTENDED
1271    
1272           If  this  bit  is  set,  whitespace  data characters in the pattern are
1273           totally ignored except when escaped or inside a character class. White-
1274           space does not include the VT character (code 11). In addition, charac-
1275           ters between an unescaped # outside a character class and the next new-
1276           line,  inclusive,  are  also  ignored.  This is equivalent to Perl's /x
1277           option, and it can be changed within a pattern by a  (?x)  option  set-
1278           ting.
1279    
1280           This  option  makes  it possible to include comments inside complicated
1281           patterns.  Note, however, that this applies only  to  data  characters.
1282           Whitespace   characters  may  never  appear  within  special  character
1283           sequences in a pattern, for  example  within  the  sequence  (?(  which
1284           introduces a conditional subpattern.
1285    
1286             PCRE_EXTRA
1287    
1288           This  option  was invented in order to turn on additional functionality
1289           of PCRE that is incompatible with Perl, but it  is  currently  of  very
1290           little  use. When set, any backslash in a pattern that is followed by a
1291           letter that has no special meaning  causes  an  error,  thus  reserving
1292           these  combinations  for  future  expansion.  By default, as in Perl, a
1293           backslash followed by a letter with no special meaning is treated as  a
1294           literal. (Perl can, however, be persuaded to give an error for this, by
1295           running it with the -w option.) There are at present no other  features
1296           controlled  by this option. It can also be set by a (?X) option setting
1297           within a pattern.
1298    
1299             PCRE_FIRSTLINE
1300    
1301           If this option is set, an  unanchored  pattern  is  required  to  match
1302           before  or  at  the  first  newline  in  the subject string, though the
1303           matched text may continue over the newline.
1304    
1305             PCRE_JAVASCRIPT_COMPAT
1306    
1307           If this option is set, PCRE's behaviour is changed in some ways so that
1308           it  is  compatible with JavaScript rather than Perl. The changes are as
1309           follows:
1310    
1311           (1) A lone closing square bracket in a pattern  causes  a  compile-time
1312           error,  because this is illegal in JavaScript (by default it is treated
1313           as a data character). Thus, the pattern AB]CD becomes illegal when this
1314           option is set.
1315    
1316           (2)  At run time, a back reference to an unset subpattern group matches
1317           an empty string (by default this causes the current  matching  alterna-
1318           tive  to  fail). A pattern such as (\1)(a) succeeds when this option is
1319           set (assuming it can find an "a" in the subject), whereas it  fails  by
1320           default, for Perl compatibility.
1321    
1322             PCRE_MULTILINE
1323    
1324           By  default,  PCRE  treats the subject string as consisting of a single
1325           line of characters (even if it actually contains newlines). The  "start
1326           of  line"  metacharacter  (^)  matches only at the start of the string,
1327           while the "end of line" metacharacter ($) matches only at  the  end  of
1328           the string, or before a terminating newline (unless PCRE_DOLLAR_ENDONLY
1329           is set). This is the same as Perl.
1330    
1331           When PCRE_MULTILINE it is set, the "start of line" and  "end  of  line"
1332           constructs  match  immediately following or immediately before internal
1333           newlines in the subject string, respectively, as well as  at  the  very
1334           start  and  end.  This is equivalent to Perl's /m option, and it can be
1335           changed within a pattern by a (?m) option setting. If there are no new-
1336           lines  in  a  subject string, or no occurrences of ^ or $ in a pattern,
1337           setting PCRE_MULTILINE has no effect.
1338    
1339             PCRE_NEWLINE_CR
1340             PCRE_NEWLINE_LF
1341             PCRE_NEWLINE_CRLF
1342             PCRE_NEWLINE_ANYCRLF
1343             PCRE_NEWLINE_ANY
1344    
1345           These options override the default newline definition that  was  chosen
1346           when  PCRE  was built. Setting the first or the second specifies that a
1347           newline is indicated by a single character (CR  or  LF,  respectively).
1348           Setting  PCRE_NEWLINE_CRLF specifies that a newline is indicated by the
1349           two-character CRLF  sequence.  Setting  PCRE_NEWLINE_ANYCRLF  specifies
1350           that any of the three preceding sequences should be recognized. Setting
1351           PCRE_NEWLINE_ANY specifies that any Unicode newline sequence should  be
1352           recognized. The Unicode newline sequences are the three just mentioned,
1353           plus the single characters VT (vertical  tab,  U+000B),  FF  (formfeed,
1354           U+000C),  NEL  (next line, U+0085), LS (line separator, U+2028), and PS
1355           (paragraph separator, U+2029). The last  two  are  recognized  only  in
1356           UTF-8 mode.
1357    
1358           The  newline  setting  in  the  options  word  uses three bits that are
1359           treated as a number, giving eight possibilities. Currently only six are
1360           used  (default  plus the five values above). This means that if you set
1361           more than one newline option, the combination may or may not be  sensi-
1362           ble. For example, PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to
1363           PCRE_NEWLINE_CRLF, but other combinations may yield unused numbers  and
1364           cause an error.
1365    
1366           The  only time that a line break is specially recognized when compiling
1367           a pattern is if PCRE_EXTENDED is set, and  an  unescaped  #  outside  a
1368           character  class  is  encountered.  This indicates a comment that lasts
1369           until after the next line break sequence. In other circumstances,  line
1370           break   sequences   are   treated  as  literal  data,  except  that  in
1371           PCRE_EXTENDED mode, both CR and LF are treated as whitespace characters
1372           and are therefore ignored.
1373    
1374           The newline option that is set at compile time becomes the default that
1375           is used for pcre_exec() and pcre_dfa_exec(), but it can be overridden.
1376    
1377             PCRE_NO_AUTO_CAPTURE
1378    
1379           If this option is set, it disables the use of numbered capturing paren-
1380           theses  in the pattern. Any opening parenthesis that is not followed by
1381           ? behaves as if it were followed by ?: but named parentheses can  still
1382           be  used  for  capturing  (and  they acquire numbers in the usual way).
1383           There is no equivalent of this option in Perl.
1384    
1385             PCRE_UCP
1386    
1387           This option changes the way PCRE processes \b, \d, \s, \w, and some  of
1388           the POSIX character classes. By default, only ASCII characters are rec-
1389           ognized, but if PCRE_UCP is set, Unicode properties are used instead to
1390           classify  characters.  More details are given in the section on generic
1391           character types in the pcrepattern page. If you set PCRE_UCP,  matching
1392           one  of the items it affects takes much longer. The option is available
1393           only if PCRE has been compiled with Unicode property support.
1394    
1395             PCRE_UNGREEDY
1396    
1397           This option inverts the "greediness" of the quantifiers  so  that  they
1398           are  not greedy by default, but become greedy if followed by "?". It is
1399           not compatible with Perl. It can also be set by a (?U)  option  setting
1400           within the pattern.
1401    
1402             PCRE_UTF8
1403    
1404           This  option  causes PCRE to regard both the pattern and the subject as
1405           strings of UTF-8 characters instead of single-byte  character  strings.
1406           However,  it is available only when PCRE is built to include UTF-8 sup-
1407           port. If not, the use of this option provokes an error. Details of  how
1408           this  option  changes the behaviour of PCRE are given in the section on
1409           UTF-8 support in the main pcre page.
1410    
1411             PCRE_NO_UTF8_CHECK
1412    
1413           When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
1414           automatically  checked.  There  is  a  discussion about the validity of
1415           UTF-8 strings in the main pcre page. If an invalid  UTF-8  sequence  of
1416           bytes  is  found,  pcre_compile() returns an error. If you already know
1417           that your pattern is valid, and you want to skip this check for perfor-
1418           mance  reasons,  you  can set the PCRE_NO_UTF8_CHECK option. When it is
1419           set, the effect of passing an invalid UTF-8  string  as  a  pattern  is
1420           undefined.  It  may  cause your program to crash. Note that this option
1421           can also be passed to pcre_exec() and pcre_dfa_exec(), to suppress  the
1422           UTF-8 validity checking of subject strings.
1423    
1424    
1425    COMPILATION ERROR CODES
1426    
1427           The  following  table  lists  the  error  codes than may be returned by
1428           pcre_compile2(), along with the error messages that may be returned  by
1429           both  compiling functions. As PCRE has developed, some error codes have
1430           fallen out of use. To avoid confusion, they have not been re-used.
1431    
1432              0  no error
1433              1  \ at end of pattern
1434              2  \c at end of pattern
1435              3  unrecognized character follows \
1436              4  numbers out of order in {} quantifier
1437              5  number too big in {} quantifier
1438              6  missing terminating ] for character class
1439              7  invalid escape sequence in character class
1440              8  range out of order in character class
1441              9  nothing to repeat
1442             10  [this code is not in use]
1443             11  internal error: unexpected repeat
1444             12  unrecognized character after (? or (?-
1445             13  POSIX named classes are supported only within a class
1446             14  missing )
1447             15  reference to non-existent subpattern
1448             16  erroffset passed as NULL
1449             17  unknown option bit(s) set
1450             18  missing ) after comment
1451             19  [this code is not in use]
1452             20  regular expression is too large
1453             21  failed to get memory
1454             22  unmatched parentheses
1455             23  internal error: code overflow
1456             24  unrecognized character after (?<
1457             25  lookbehind assertion is not fixed length
1458             26  malformed number or name after (?(
1459             27  conditional group contains more than two branches
1460             28  assertion expected after (?(
1461             29  (?R or (?[+-]digits must be followed by )
1462             30  unknown POSIX class name
1463             31  POSIX collating elements are not supported
1464             32  this version of PCRE is not compiled with PCRE_UTF8 support
1465             33  [this code is not in use]
1466             34  character value in \x{...} sequence is too large
1467             35  invalid condition (?(0)
1468             36  \C not allowed in lookbehind assertion
1469             37  PCRE does not support \L, \l, \N, \U, or \u
1470             38  number after (?C is > 255
1471             39  closing ) for (?C expected
1472             40  recursive call could loop indefinitely
1473             41  unrecognized character after (?P
1474             42  syntax error in subpattern name (missing terminator)
1475             43  two named subpatterns have the same name
1476             44  invalid UTF-8 string
1477             45  support for \P, \p, and \X has not been compiled
1478             46  malformed \P or \p sequence
1479             47  unknown property name after \P or \p
1480             48  subpattern name is too long (maximum 32 characters)
1481             49  too many named subpatterns (maximum 10000)
1482             50  [this code is not in use]
1483             51  octal value is greater than \377 (not in UTF-8 mode)
1484             52  internal error: overran compiling workspace
1485             53  internal error: previously-checked referenced subpattern
1486                   not found
1487             54  DEFINE group contains more than one branch
1488             55  repeating a DEFINE group is not allowed
1489             56  inconsistent NEWLINE options
1490             57  \g is not followed by a braced, angle-bracketed, or quoted
1491                   name/number or by a plain number
1492             58  a numbered reference must not be zero
1493             59  an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
1494             60  (*VERB) not recognized
1495             61  number is too big
1496             62  subpattern name expected
1497             63  digit expected after (?+
1498             64  ] is an invalid data character in JavaScript compatibility mode
1499             65  different names for subpatterns of the same number are
1500                   not allowed
1501             66  (*MARK) must have an argument
1502             67  this version of PCRE is not compiled with PCRE_UCP support
1503    
1504           The numbers 32 and 10000 in errors 48 and 49  are  defaults;  different
1505           values may be used if the limits were changed when PCRE was built.
1506    
1507    
1508    STUDYING A PATTERN
1509    
1510           pcre_extra *pcre_study(const pcre *code, int options
1511                const char **errptr);
1512    
1513           If  a  compiled  pattern is going to be used several times, it is worth
1514           spending more time analyzing it in order to speed up the time taken for
1515           matching.  The function pcre_study() takes a pointer to a compiled pat-
1516           tern as its first argument. If studying the pattern produces additional
1517           information  that  will  help speed up matching, pcre_study() returns a
1518           pointer to a pcre_extra block, in which the study_data field points  to
1519           the results of the study.
1520    
1521           The  returned  value  from  pcre_study()  can  be  passed  directly  to
1522           pcre_exec() or pcre_dfa_exec(). However, a pcre_extra block  also  con-
1523           tains  other  fields  that can be set by the caller before the block is
1524           passed; these are described below in the section on matching a pattern.
1525    
1526           If studying the  pattern  does  not  produce  any  useful  information,
1527           pcre_study() returns NULL. In that circumstance, if the calling program
1528           wants  to  pass  any  of   the   other   fields   to   pcre_exec()   or
1529           pcre_dfa_exec(), it must set up its own pcre_extra block.
1530    
1531           The  second  argument of pcre_study() contains option bits. At present,
1532           no options are defined, and this argument should always be zero.
1533    
1534           The third argument for pcre_study() is a pointer for an error  message.
1535           If  studying  succeeds  (even  if no data is returned), the variable it
1536           points to is set to NULL. Otherwise it is set to  point  to  a  textual
1537           error message. This is a static string that is part of the library. You
1538           must not try to free it. You should test the  error  pointer  for  NULL
1539           after calling pcre_study(), to be sure that it has run successfully.
1540    
1541           This is a typical call to pcre_study():
1542    
1543             pcre_extra *pe;
1544             pe = pcre_study(
1545               re,             /* result of pcre_compile() */
1546               0,              /* no options exist */
1547               &error);        /* set to NULL or points to a message */
1548    
1549           Studying a pattern does two things: first, a lower bound for the length
1550           of subject string that is needed to match the pattern is computed. This
1551           does not mean that there are any strings of that length that match, but
1552           it does guarantee that no shorter strings match. The value is  used  by
1553           pcre_exec()  and  pcre_dfa_exec()  to  avoid  wasting time by trying to
1554           match strings that are shorter than the lower bound. You can  find  out
1555           the value in a calling program via the pcre_fullinfo() function.
1556    
1557           Studying a pattern is also useful for non-anchored patterns that do not
1558           have a single fixed starting character. A bitmap of  possible  starting
1559           bytes  is  created. This speeds up finding a position in the subject at
1560           which to start matching.
1561    
1562           The two optimizations just described can be  disabled  by  setting  the
1563           PCRE_NO_START_OPTIMIZE    option    when    calling    pcre_exec()   or
1564           pcre_dfa_exec(). You might want to do this  if  your  pattern  contains
1565           callouts,  or  make  use of (*MARK), and you make use of these in cases
1566           where matching fails.  See  the  discussion  of  PCRE_NO_START_OPTIMIZE
1567           below.
1568    
1569    
1570    LOCALE SUPPORT
1571    
1572           PCRE  handles  caseless matching, and determines whether characters are
1573           letters, digits, or whatever, by reference to a set of tables,  indexed
1574           by  character  value.  When running in UTF-8 mode, this applies only to
1575           characters with codes less than 128. By  default,  higher-valued  codes
1576           never match escapes such as \w or \d, but they can be tested with \p if
1577           PCRE is built with Unicode character property  support.  Alternatively,
1578           the  PCRE_UCP  option  can  be  set at compile time; this causes \w and
1579           friends to use Unicode property support instead of built-in tables. The
1580           use of locales with Unicode is discouraged. If you are handling charac-
1581           ters with codes greater than 128, you should either use UTF-8 and  Uni-
1582           code, or use locales, but not try to mix the two.
1583    
1584           PCRE  contains  an  internal set of tables that are used when the final
1585           argument of pcre_compile() is  NULL.  These  are  sufficient  for  many
1586           applications.  Normally, the internal tables recognize only ASCII char-
1587           acters. However, when PCRE is built, it is possible to cause the inter-
1588           nal tables to be rebuilt in the default "C" locale of the local system,
1589           which may cause them to be different.
1590    
1591           The internal tables can always be overridden by tables supplied by  the
1592           application that calls PCRE. These may be created in a different locale
1593           from the default. As more and more applications change  to  using  Uni-
1594           code, the need for this locale support is expected to die away.
1595    
1596           External  tables  are  built by calling the pcre_maketables() function,
1597           which has no arguments, in the relevant locale. The result can then  be
1598           passed  to  pcre_compile()  or  pcre_exec()  as often as necessary. For
1599           example, to build and use tables that are appropriate  for  the  French
1600           locale  (where  accented  characters  with  values greater than 128 are
1601           treated as letters), the following code could be used:
1602    
1603             setlocale(LC_CTYPE, "fr_FR");
1604             tables = pcre_maketables();
1605             re = pcre_compile(..., tables);
1606    
1607           The locale name "fr_FR" is used on Linux and other  Unix-like  systems;
1608           if you are using Windows, the name for the French locale is "french".
1609    
1610           When  pcre_maketables()  runs,  the  tables are built in memory that is
1611           obtained via pcre_malloc. It is the caller's responsibility  to  ensure
1612           that  the memory containing the tables remains available for as long as
1613           it is needed.
1614    
1615           The pointer that is passed to pcre_compile() is saved with the compiled
1616           pattern,  and the same tables are used via this pointer by pcre_study()
1617           and normally also by pcre_exec(). Thus, by default, for any single pat-
1618           tern, compilation, studying and matching all happen in the same locale,
1619           but different patterns can be compiled in different locales.
1620    
1621           It is possible to pass a table pointer or NULL (indicating the  use  of
1622           the  internal  tables)  to  pcre_exec(). Although not intended for this
1623           purpose, this facility could be used to match a pattern in a  different
1624           locale from the one in which it was compiled. Passing table pointers at
1625           run time is discussed below in the section on matching a pattern.
1626    
1627    
1628    INFORMATION ABOUT A PATTERN
1629    
1630           int pcre_fullinfo(const pcre *code, const pcre_extra *extra,
1631                int what, void *where);
1632    
1633           The pcre_fullinfo() function returns information about a compiled  pat-
1634           tern. It replaces the obsolete pcre_info() function, which is neverthe-
1635           less retained for backwards compability (and is documented below).
1636    
1637           The first argument for pcre_fullinfo() is a  pointer  to  the  compiled
1638           pattern.  The second argument is the result of pcre_study(), or NULL if
1639           the pattern was not studied. The third argument specifies  which  piece
1640           of  information  is required, and the fourth argument is a pointer to a
1641           variable to receive the data. The yield of the  function  is  zero  for
1642           success, or one of the following negative numbers:
1643    
1644             PCRE_ERROR_NULL       the argument code was NULL
1645                                   the argument where was NULL
1646             PCRE_ERROR_BADMAGIC   the "magic number" was not found
1647             PCRE_ERROR_BADOPTION  the value of what was invalid
1648    
1649           The  "magic  number" is placed at the start of each compiled pattern as
1650           an simple check against passing an arbitrary memory pointer. Here is  a
1651           typical  call  of pcre_fullinfo(), to obtain the length of the compiled
1652           pattern:
1653    
1654             int rc;
1655             size_t length;
1656             rc = pcre_fullinfo(
1657               re,               /* result of pcre_compile() */
1658               pe,               /* result of pcre_study(), or NULL */
1659               PCRE_INFO_SIZE,   /* what is required */
1660               &length);         /* where to put the data */
1661    
1662           The possible values for the third argument are defined in  pcre.h,  and
1663           are as follows:
1664    
1665             PCRE_INFO_BACKREFMAX
1666    
1667           Return  the  number  of  the highest back reference in the pattern. The
1668           fourth argument should point to an int variable. Zero  is  returned  if
1669           there are no back references.
1670    
1671             PCRE_INFO_CAPTURECOUNT
1672    
1673           Return  the  number of capturing subpatterns in the pattern. The fourth
1674           argument should point to an int variable.
1675    
1676             PCRE_INFO_DEFAULT_TABLES
1677    
1678           Return a pointer to the internal default character tables within  PCRE.
1679           The  fourth  argument should point to an unsigned char * variable. This
1680           information call is provided for internal use by the pcre_study() func-
1681           tion.  External  callers  can  cause PCRE to use its internal tables by
1682           passing a NULL table pointer.
1683    
1684             PCRE_INFO_FIRSTBYTE
1685    
1686           Return information about the first byte of any matched  string,  for  a
1687           non-anchored  pattern. The fourth argument should point to an int vari-
1688           able. (This option used to be called PCRE_INFO_FIRSTCHAR; the old  name
1689           is still recognized for backwards compatibility.)
1690    
1691           If  there  is  a  fixed first byte, for example, from a pattern such as
1692           (cat|cow|coyote), its value is returned. Otherwise, if either
1693    
1694           (a) the pattern was compiled with the PCRE_MULTILINE option, and  every
1695           branch starts with "^", or
1696    
1697           (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not
1698           set (if it were set, the pattern would be anchored),
1699    
1700           -1 is returned, indicating that the pattern matches only at  the  start
1701           of  a  subject string or after any newline within the string. Otherwise
1702           -2 is returned. For anchored patterns, -2 is returned.
1703    
1704             PCRE_INFO_FIRSTTABLE
1705    
1706           If the pattern was studied, and this resulted in the construction of  a
1707           256-bit table indicating a fixed set of bytes for the first byte in any
1708           matching string, a pointer to the table is returned. Otherwise NULL  is
1709           returned.  The fourth argument should point to an unsigned char * vari-
1710           able.
1711    
1712             PCRE_INFO_HASCRORLF
1713    
1714           Return 1 if the pattern contains any explicit  matches  for  CR  or  LF
1715           characters,  otherwise  0.  The  fourth argument should point to an int
1716           variable. An explicit match is either a literal CR or LF character,  or
1717           \r or \n.
1718    
1719             PCRE_INFO_JCHANGED
1720    
1721           Return  1  if  the (?J) or (?-J) option setting is used in the pattern,
1722           otherwise 0. The fourth argument should point to an int variable.  (?J)
1723           and (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
1724    
1725             PCRE_INFO_LASTLITERAL
1726    
1727           Return  the  value of the rightmost literal byte that must exist in any
1728           matched string, other than at its  start,  if  such  a  byte  has  been
1729           recorded. The fourth argument should point to an int variable. If there
1730           is no such byte, -1 is returned. For anchored patterns, a last  literal
1731           byte  is  recorded only if it follows something of variable length. For
1732           example, for the pattern /^a\d+z\d+/ the returned value is "z", but for
1733           /^a\dz\d/ the returned value is -1.
1734    
1735             PCRE_INFO_MINLENGTH
1736    
1737           If  the  pattern  was studied and a minimum length for matching subject
1738           strings was computed, its value is  returned.  Otherwise  the  returned
1739           value  is  -1. The value is a number of characters, not bytes (this may
1740           be relevant in UTF-8 mode). The fourth argument should point to an  int
1741           variable.  A  non-negative  value is a lower bound to the length of any
1742           matching string. There may not be any strings of that  length  that  do
1743           actually match, but every string that does match is at least that long.
1744    
1745             PCRE_INFO_NAMECOUNT
1746             PCRE_INFO_NAMEENTRYSIZE
1747             PCRE_INFO_NAMETABLE
1748    
1749           PCRE  supports the use of named as well as numbered capturing parenthe-
1750           ses. The names are just an additional way of identifying the  parenthe-
1751           ses, which still acquire numbers. Several convenience functions such as
1752           pcre_get_named_substring() are provided for  extracting  captured  sub-
1753           strings  by  name. It is also possible to extract the data directly, by
1754           first converting the name to a number in order to  access  the  correct
1755           pointers in the output vector (described with pcre_exec() below). To do
1756           the conversion, you need  to  use  the  name-to-number  map,  which  is
1757           described by these three values.
1758    
1759           The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT
1760           gives the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size
1761           of  each  entry;  both  of  these  return  an int value. The entry size
1762           depends on the length of the longest name. PCRE_INFO_NAMETABLE  returns
1763           a  pointer  to  the  first  entry of the table (a pointer to char). The
1764           first two bytes of each entry are the number of the capturing parenthe-
1765           sis,  most  significant byte first. The rest of the entry is the corre-
1766           sponding name, zero terminated.
1767    
1768           The names are in alphabetical order. Duplicate names may appear if  (?|
1769           is used to create multiple groups with the same number, as described in
1770           the section on duplicate subpattern numbers in  the  pcrepattern  page.
1771           Duplicate  names  for  subpatterns with different numbers are permitted
1772           only if PCRE_DUPNAMES is set. In all cases  of  duplicate  names,  they
1773           appear  in  the table in the order in which they were found in the pat-
1774           tern. In the absence of (?| this is the  order  of  increasing  number;
1775           when (?| is used this is not necessarily the case because later subpat-
1776           terns may have lower numbers.
1777    
1778           As a simple example of the name/number table,  consider  the  following
1779           pattern  (assume  PCRE_EXTENDED is set, so white space - including new-
1780           lines - is ignored):
1781    
1782             (?<date> (?<year>(\d\d)?\d\d) -
1783             (?<month>\d\d) - (?<day>\d\d) )
1784    
1785           There are four named subpatterns, so the table has  four  entries,  and
1786           each  entry  in the table is eight bytes long. The table is as follows,
1787           with non-printing bytes shows in hexadecimal, and undefined bytes shown
1788           as ??:
1789    
1790             00 01 d  a  t  e  00 ??
1791             00 05 d  a  y  00 ?? ??
1792             00 04 m  o  n  t  h  00
1793             00 02 y  e  a  r  00 ??
1794    
1795           When  writing  code  to  extract  data from named subpatterns using the
1796           name-to-number map, remember that the length of the entries  is  likely
1797           to be different for each compiled pattern.
1798    
1799             PCRE_INFO_OKPARTIAL
1800    
1801           Return  1  if  the  pattern  can  be  used  for  partial  matching with
1802           pcre_exec(), otherwise 0. The fourth argument should point  to  an  int
1803           variable.  From  release  8.00,  this  always  returns  1,  because the
1804           restrictions that previously applied  to  partial  matching  have  been
1805           lifted.  The  pcrepartial documentation gives details of partial match-
1806           ing.
1807    
1808             PCRE_INFO_OPTIONS
1809    
1810           Return a copy of the options with which the pattern was  compiled.  The
1811           fourth  argument  should  point to an unsigned long int variable. These
1812           option bits are those specified in the call to pcre_compile(), modified
1813           by any top-level option settings at the start of the pattern itself. In
1814           other words, they are the options that will be in force  when  matching
1815           starts.  For  example, if the pattern /(?im)abc(?-i)d/ is compiled with
1816           the PCRE_EXTENDED option, the result is PCRE_CASELESS,  PCRE_MULTILINE,
1817           and PCRE_EXTENDED.
1818    
1819           A  pattern  is  automatically  anchored by PCRE if all of its top-level
1820           alternatives begin with one of the following:
1821    
1822             ^     unless PCRE_MULTILINE is set
1823             \A    always
1824             \G    always
1825             .*    if PCRE_DOTALL is set and there are no back
1826                     references to the subpattern in which .* appears
1827    
1828           For such patterns, the PCRE_ANCHORED bit is set in the options returned
1829           by pcre_fullinfo().
1830    
1831             PCRE_INFO_SIZE
1832    
1833           Return  the  size  of the compiled pattern, that is, the value that was
1834           passed as the argument to pcre_malloc() when PCRE was getting memory in
1835           which to place the compiled data. The fourth argument should point to a
1836           size_t variable.
1837    
1838             PCRE_INFO_STUDYSIZE
1839    
1840           Return the size of the data block pointed to by the study_data field in
1841           a  pcre_extra  block.  That  is,  it  is  the  value that was passed to
1842           pcre_malloc() when PCRE was getting memory into which to place the data
1843           created  by  pcre_study().  If pcre_extra is NULL, or there is no study
1844           data, zero is returned. The fourth argument should point  to  a  size_t
1845           variable.
1846    
1847    
1848    OBSOLETE INFO FUNCTION
1849    
1850           int pcre_info(const pcre *code, int *optptr, int *firstcharptr);
1851    
1852           The  pcre_info()  function is now obsolete because its interface is too
1853           restrictive to return all the available data about a compiled  pattern.
1854           New   programs   should  use  pcre_fullinfo()  instead.  The  yield  of
1855           pcre_info() is the number of capturing subpatterns, or one of the  fol-
1856           lowing negative numbers:
1857    
1858             PCRE_ERROR_NULL       the argument code was NULL
1859             PCRE_ERROR_BADMAGIC   the "magic number" was not found
1860    
1861           If  the  optptr  argument is not NULL, a copy of the options with which
1862           the pattern was compiled is placed in the integer  it  points  to  (see
1863           PCRE_INFO_OPTIONS above).
1864    
1865           If  the  pattern  is  not anchored and the firstcharptr argument is not
1866           NULL, it is used to pass back information about the first character  of
1867           any matched string (see PCRE_INFO_FIRSTBYTE above).
1868    
1869    
1870    REFERENCE COUNTS
1871    
1872           int pcre_refcount(pcre *code, int adjust);
1873    
1874           The  pcre_refcount()  function is used to maintain a reference count in
1875           the data block that contains a compiled pattern. It is provided for the
1876           benefit  of  applications  that  operate  in an object-oriented manner,
1877           where different parts of the application may be using the same compiled
1878           pattern, but you want to free the block when they are all done.
1879    
1880           When a pattern is compiled, the reference count field is initialized to
1881           zero.  It is changed only by calling this function, whose action is  to
1882           add  the  adjust  value  (which may be positive or negative) to it. The
1883           yield of the function is the new value. However, the value of the count
1884           is  constrained to lie between 0 and 65535, inclusive. If the new value
1885           is outside these limits, it is forced to the appropriate limit value.
1886    
1887           Except when it is zero, the reference count is not correctly  preserved
1888           if  a  pattern  is  compiled on one host and then transferred to a host
1889           whose byte-order is different. (This seems a highly unlikely scenario.)
1890    
1891    
1892    MATCHING A PATTERN: THE TRADITIONAL FUNCTION
1893    
1894           int pcre_exec(const pcre *code, const pcre_extra *extra,
1895                const char *subject, int length, int startoffset,
1896                int options, int *ovector, int ovecsize);
1897    
1898           The function pcre_exec() is called to match a subject string against  a
1899           compiled  pattern, which is passed in the code argument. If the pattern
1900           was studied, the result of the study should  be  passed  in  the  extra
1901           argument.  This  function is the main matching facility of the library,
1902           and it operates in a Perl-like manner. For specialist use there is also
1903           an  alternative matching function, which is described below in the sec-
1904           tion about the pcre_dfa_exec() function.
1905    
1906           In most applications, the pattern will have been compiled (and  option-
1907           ally  studied)  in the same process that calls pcre_exec(). However, it
1908           is possible to save compiled patterns and study data, and then use them
1909           later  in  different processes, possibly even on different hosts. For a
1910           discussion about this, see the pcreprecompile documentation.
1911    
1912           Here is an example of a simple call to pcre_exec():
1913    
1914             int rc;
1915             int ovector[30];
1916             rc = pcre_exec(
1917               re,             /* result of pcre_compile() */
1918               NULL,           /* we didn't study the pattern */
1919               "some string",  /* the subject string */
1920               11,             /* the length of the subject string */
1921               0,              /* start at offset 0 in the subject */
1922               0,              /* default options */
1923               ovector,        /* vector of integers for substring information */
1924               30);            /* number of elements (NOT size in bytes) */
1925    
1926       Extra data for pcre_exec()
1927    
1928           If the extra argument is not NULL, it must point to a  pcre_extra  data
1929           block.  The pcre_study() function returns such a block (when it doesn't
1930           return NULL), but you can also create one for yourself, and pass  addi-
1931           tional  information  in it. The pcre_extra block contains the following
1932           fields (not necessarily in this order):
1933    
1934             unsigned long int flags;
1935             void *study_data;
1936             unsigned long int match_limit;
1937             unsigned long int match_limit_recursion;
1938             void *callout_data;
1939             const unsigned char *tables;
1940             unsigned char **mark;
1941    
1942           The flags field is a bitmap that specifies which of  the  other  fields
1943           are set. The flag bits are:
1944    
1945             PCRE_EXTRA_STUDY_DATA
1946             PCRE_EXTRA_MATCH_LIMIT
1947             PCRE_EXTRA_MATCH_LIMIT_RECURSION
1948             PCRE_EXTRA_CALLOUT_DATA
1949             PCRE_EXTRA_TABLES
1950             PCRE_EXTRA_MARK
1951    
1952           Other  flag  bits should be set to zero. The study_data field is set in
1953           the pcre_extra block that is returned by  pcre_study(),  together  with
1954           the appropriate flag bit. You should not set this yourself, but you may
1955           add to the block by setting the other fields  and  their  corresponding
1956           flag bits.
1957    
1958           The match_limit field provides a means of preventing PCRE from using up
1959           a vast amount of resources when running patterns that are not going  to
1960           match,  but  which  have  a very large number of possibilities in their
1961           search trees. The classic example is a pattern that uses nested  unlim-
1962           ited repeats.
1963    
1964           Internally,  PCRE uses a function called match() which it calls repeat-
1965           edly (sometimes recursively). The limit set by match_limit  is  imposed
1966           on  the  number  of times this function is called during a match, which
1967           has the effect of limiting the amount of  backtracking  that  can  take
1968           place. For patterns that are not anchored, the count restarts from zero
1969           for each position in the subject string.
1970    
1971           The default value for the limit can be set  when  PCRE  is  built;  the
1972           default  default  is 10 million, which handles all but the most extreme
1973           cases. You can override the default  by  suppling  pcre_exec()  with  a
1974           pcre_extra     block    in    which    match_limit    is    set,    and
1975           PCRE_EXTRA_MATCH_LIMIT is set in the  flags  field.  If  the  limit  is
1976           exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
1977    
1978           The  match_limit_recursion field is similar to match_limit, but instead
1979           of limiting the total number of times that match() is called, it limits
1980           the  depth  of  recursion. The recursion depth is a smaller number than
1981           the total number of calls, because not all calls to match() are  recur-
1982           sive.  This limit is of use only if it is set smaller than match_limit.
1983    
1984           Limiting  the  recursion  depth  limits the amount of stack that can be
1985           used, or, when PCRE has been compiled to use memory on the heap instead
1986           of the stack, the amount of heap memory that can be used.
1987    
1988           The  default  value  for  match_limit_recursion can be set when PCRE is
1989           built; the default default  is  the  same  value  as  the  default  for
1990           match_limit.  You can override the default by suppling pcre_exec() with
1991           a  pcre_extra  block  in  which  match_limit_recursion  is   set,   and
1992           PCRE_EXTRA_MATCH_LIMIT_RECURSION  is  set  in  the  flags field. If the
1993           limit is exceeded, pcre_exec() returns PCRE_ERROR_RECURSIONLIMIT.
1994    
1995           The callout_data field is used in conjunction with the  "callout"  fea-
1996           ture, and is described in the pcrecallout documentation.
1997    
1998           The  tables  field  is  used  to  pass  a  character  tables pointer to
1999           pcre_exec(); this overrides the value that is stored with the  compiled
2000           pattern.  A  non-NULL value is stored with the compiled pattern only if
2001           custom tables were supplied to pcre_compile() via  its  tableptr  argu-
2002           ment.  If NULL is passed to pcre_exec() using this mechanism, it forces
2003           PCRE's internal tables to be used. This facility is  helpful  when  re-
2004           using  patterns  that  have been saved after compiling with an external
2005           set of tables, because the external tables  might  be  at  a  different
2006           address  when  pcre_exec() is called. See the pcreprecompile documenta-
2007           tion for a discussion of saving compiled patterns for later use.
2008    
2009           If PCRE_EXTRA_MARK is set in the flags field, the mark  field  must  be
2010           set  to  point  to a char * variable. If the pattern contains any back-
2011           tracking control verbs such as (*MARK:NAME), and the execution ends  up
2012           with  a  name  to  pass back, a pointer to the name string (zero termi-
2013           nated) is placed in the variable pointed to  by  the  mark  field.  The
2014           names  are  within  the  compiled pattern; if you wish to retain such a
2015           name you must copy it before freeing the memory of a compiled  pattern.
2016           If  there  is no name to pass back, the variable pointed to by the mark
2017           field set to NULL. For details of the backtracking control  verbs,  see
2018           the section entitled "Backtracking control" in the pcrepattern documen-
2019           tation.
2020    
2021       Option bits for pcre_exec()
2022    
2023           The unused bits of the options argument for pcre_exec() must  be  zero.
2024           The  only  bits  that  may  be set are PCRE_ANCHORED, PCRE_NEWLINE_xxx,
2025           PCRE_NOTBOL,   PCRE_NOTEOL,    PCRE_NOTEMPTY,    PCRE_NOTEMPTY_ATSTART,
2026           PCRE_NO_START_OPTIMIZE,   PCRE_NO_UTF8_CHECK,   PCRE_PARTIAL_SOFT,  and
2027           PCRE_PARTIAL_HARD.
2028    
2029             PCRE_ANCHORED
2030    
2031           The PCRE_ANCHORED option limits pcre_exec() to matching  at  the  first
2032           matching  position.  If  a  pattern was compiled with PCRE_ANCHORED, or
2033           turned out to be anchored by virtue of its contents, it cannot be  made
2034           unachored at matching time.
2035    
2036             PCRE_BSR_ANYCRLF
2037             PCRE_BSR_UNICODE
2038    
2039           These options (which are mutually exclusive) control what the \R escape
2040           sequence matches. The choice is either to match only CR, LF,  or  CRLF,
2041           or  to  match  any Unicode newline sequence. These options override the
2042           choice that was made or defaulted when the pattern was compiled.
2043    
2044             PCRE_NEWLINE_CR
2045             PCRE_NEWLINE_LF
2046             PCRE_NEWLINE_CRLF
2047             PCRE_NEWLINE_ANYCRLF
2048             PCRE_NEWLINE_ANY
2049    
2050           These options override  the  newline  definition  that  was  chosen  or
2051           defaulted  when the pattern was compiled. For details, see the descrip-
2052           tion of pcre_compile()  above.  During  matching,  the  newline  choice
2053           affects  the  behaviour  of the dot, circumflex, and dollar metacharac-
2054           ters. It may also alter the way the match position is advanced after  a
2055           match failure for an unanchored pattern.
2056    
2057           When  PCRE_NEWLINE_CRLF,  PCRE_NEWLINE_ANYCRLF,  or PCRE_NEWLINE_ANY is
2058           set, and a match attempt for an unanchored pattern fails when the  cur-
2059           rent  position  is  at  a  CRLF  sequence,  and the pattern contains no
2060           explicit matches for  CR  or  LF  characters,  the  match  position  is
2061           advanced by two characters instead of one, in other words, to after the
2062           CRLF.
2063    
2064           The above rule is a compromise that makes the most common cases work as
2065           expected.  For  example,  if  the  pattern  is .+A (and the PCRE_DOTALL
2066           option is not set), it does not match the string "\r\nA" because, after
2067           failing  at the start, it skips both the CR and the LF before retrying.
2068           However, the pattern [\r\n]A does match that string,  because  it  con-
2069           tains an explicit CR or LF reference, and so advances only by one char-
2070           acter after the first failure.
2071    
2072           An explicit match for CR of LF is either a literal appearance of one of
2073           those  characters,  or  one  of the \r or \n escape sequences. Implicit
2074           matches such as [^X] do not count, nor does \s (which includes  CR  and
2075           LF in the characters that it matches).
2076    
2077           Notwithstanding  the above, anomalous effects may still occur when CRLF
2078           is a valid newline sequence and explicit \r or \n escapes appear in the
2079           pattern.
2080    
2081             PCRE_NOTBOL
2082    
2083           This option specifies that first character of the subject string is not
2084           the beginning of a line, so the  circumflex  metacharacter  should  not
2085           match  before it. Setting this without PCRE_MULTILINE (at compile time)
2086           causes circumflex never to match. This option affects only  the  behav-
2087           iour of the circumflex metacharacter. It does not affect \A.
2088    
2089             PCRE_NOTEOL
2090    
2091           This option specifies that the end of the subject string is not the end
2092           of a line, so the dollar metacharacter should not match it nor  (except
2093           in  multiline mode) a newline immediately before it. Setting this with-
2094           out PCRE_MULTILINE (at compile time) causes dollar never to match. This
2095           option  affects only the behaviour of the dollar metacharacter. It does
2096           not affect \Z or \z.
2097    
2098             PCRE_NOTEMPTY
2099    
2100           An empty string is not considered to be a valid match if this option is
2101           set.  If  there are alternatives in the pattern, they are tried. If all
2102           the alternatives match the empty string, the entire  match  fails.  For
2103           example, if the pattern
2104    
2105             a?b?
2106    
2107           is  applied  to  a  string not beginning with "a" or "b", it matches an
2108           empty string at the start of the subject. With PCRE_NOTEMPTY set,  this
2109           match is not valid, so PCRE searches further into the string for occur-
2110           rences of "a" or "b".
2111    
2112             PCRE_NOTEMPTY_ATSTART
2113    
2114           This is like PCRE_NOTEMPTY, except that an empty string match  that  is
2115           not  at  the  start  of  the  subject  is  permitted. If the pattern is
2116           anchored, such a match can occur only if the pattern contains \K.
2117    
2118           Perl    has    no    direct    equivalent    of    PCRE_NOTEMPTY     or
2119           PCRE_NOTEMPTY_ATSTART,  but  it  does  make a special case of a pattern
2120           match of the empty string within its split() function, and  when  using
2121           the  /g  modifier.  It  is  possible  to emulate Perl's behaviour after
2122           matching a null string by first trying the match again at the same off-
2123           set  with  PCRE_NOTEMPTY_ATSTART  and  PCRE_ANCHORED,  and then if that
2124           fails, by advancing the starting offset (see below) and trying an ordi-
2125           nary  match  again. There is some code that demonstrates how to do this
2126           in the pcredemo sample program.
2127    
2128             PCRE_NO_START_OPTIMIZE
2129    
2130           There are a number of optimizations that pcre_exec() uses at the  start
2131           of  a  match,  in  order to speed up the process. For example, if it is
2132           known that an unanchored match must start with a specific character, it
2133           searches  the  subject  for that character, and fails immediately if it
2134           cannot find it, without actually running the  main  matching  function.
2135           This means that a special item such as (*COMMIT) at the start of a pat-
2136           tern is not considered until after a suitable starting  point  for  the
2137           match  has been found. When callouts or (*MARK) items are in use, these
2138           "start-up" optimizations can cause them to be skipped if the pattern is
2139           never  actually  used.  The start-up optimizations are in effect a pre-
2140           scan of the subject that takes place before the pattern is run.
2141    
2142           The PCRE_NO_START_OPTIMIZE option disables the start-up  optimizations,
2143           possibly  causing  performance  to  suffer,  but ensuring that in cases
2144           where the result is "no match", the callouts do occur, and  that  items
2145           such as (*COMMIT) and (*MARK) are considered at every possible starting
2146           position in the subject  string.   Setting  PCRE_NO_START_OPTIMIZE  can
2147           change the outcome of a matching operation.  Consider the pattern
2148    
2149             (*COMMIT)ABC
2150    
2151           When  this  is  compiled, PCRE records the fact that a match must start
2152           with the character "A". Suppose the subject  string  is  "DEFABC".  The
2153           start-up  optimization  scans along the subject, finds "A" and runs the
2154           first match attempt from there. The (*COMMIT) item means that the  pat-
2155           tern  must  match the current starting position, which in this case, it
2156           does. However, if the same match  is  run  with  PCRE_NO_START_OPTIMIZE
2157           set,  the  initial  scan  along the subject string does not happen. The
2158           first match attempt is run starting  from  "D"  and  when  this  fails,
2159           (*COMMIT)  prevents  any  further  matches  being tried, so the overall
2160           result is "no match". If the pattern is studied,  more  start-up  opti-
2161           mizations  may  be  used. For example, a minimum length for the subject
2162           may be recorded. Consider the pattern
2163    
2164             (*MARK:A)(X|Y)
2165    
2166           The minimum length for a match is one  character.  If  the  subject  is
2167           "ABC",  there  will  be  attempts  to  match "ABC", "BC", "C", and then
2168           finally an empty string.  If the pattern is studied, the final  attempt
2169           does  not take place, because PCRE knows that the subject is too short,
2170           and so the (*MARK) is never encountered.  In this  case,  studying  the
2171           pattern  does  not  affect the overall match result, which is still "no
2172           match", but it does affect the auxiliary information that is returned.
2173    
2174             PCRE_NO_UTF8_CHECK
2175    
2176           When PCRE_UTF8 is set at compile time, the validity of the subject as a
2177           UTF-8  string is automatically checked when pcre_exec() is subsequently
2178           called.  The value of startoffset is also checked  to  ensure  that  it
2179           points  to  the start of a UTF-8 character. There is a discussion about
2180           the validity of UTF-8 strings in the section on UTF-8  support  in  the
2181           main  pcre  page.  If  an  invalid  UTF-8  sequence  of bytes is found,
2182           pcre_exec() returns the error PCRE_ERROR_BADUTF8. If  startoffset  con-
2183           tains an invalid value, PCRE_ERROR_BADUTF8_OFFSET is returned.
2184    
2185           If  you  already  know that your subject is valid, and you want to skip
2186           these   checks   for   performance   reasons,   you   can    set    the
2187           PCRE_NO_UTF8_CHECK  option  when calling pcre_exec(). You might want to
2188           do this for the second and subsequent calls to pcre_exec() if  you  are
2189           making  repeated  calls  to  find  all  the matches in a single subject
2190           string. However, you should be  sure  that  the  value  of  startoffset
2191           points  to  the  start of a UTF-8 character. When PCRE_NO_UTF8_CHECK is
2192           set, the effect of passing an invalid UTF-8 string as a subject,  or  a
2193           value  of startoffset that does not point to the start of a UTF-8 char-
2194           acter, is undefined. Your program may crash.
2195    
2196             PCRE_PARTIAL_HARD
2197             PCRE_PARTIAL_SOFT
2198    
2199           These options turn on the partial matching feature. For backwards  com-
2200           patibility,  PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial
2201           match occurs if the end of the subject string is reached  successfully,
2202           but  there  are not enough subject characters to complete the match. If
2203           this happens when PCRE_PARTIAL_HARD  is  set,  pcre_exec()  immediately
2204           returns  PCRE_ERROR_PARTIAL.  Otherwise,  if  PCRE_PARTIAL_SOFT is set,
2205           matching continues by testing any other alternatives. Only if they  all
2206           fail  is  PCRE_ERROR_PARTIAL  returned (instead of PCRE_ERROR_NOMATCH).
2207           The portion of the string that was inspected when the partial match was
2208           found  is  set  as  the first matching string. There is a more detailed
2209           discussion in the pcrepartial documentation.
2210    
2211       The string to be matched by pcre_exec()
2212    
2213           The subject string is passed to pcre_exec() as a pointer in subject,  a
2214           length (in bytes) in length, and a starting byte offset in startoffset.
2215           In UTF-8 mode, the byte offset must point to the start of a UTF-8 char-
2216           acter.  Unlike  the pattern string, the subject may contain binary zero
2217           bytes. When the starting offset is zero, the search for a match  starts
2218           at  the  beginning  of  the subject, and this is by far the most common
2219           case.
2220    
2221           A non-zero starting offset is useful when searching for  another  match
2222           in  the same subject by calling pcre_exec() again after a previous suc-
2223           cess.  Setting startoffset differs from just passing over  a  shortened
2224           string  and  setting  PCRE_NOTBOL  in the case of a pattern that begins
2225           with any kind of lookbehind. For example, consider the pattern
2226    
2227             \Biss\B
2228    
2229           which finds occurrences of "iss" in the middle of  words.  (\B  matches
2230           only  if  the  current position in the subject is not a word boundary.)
2231           When applied to the string "Mississipi" the first call  to  pcre_exec()
2232           finds  the  first  occurrence. If pcre_exec() is called again with just
2233           the remainder of the subject,  namely  "issipi",  it  does  not  match,
2234           because \B is always false at the start of the subject, which is deemed
2235           to be a word boundary. However, if pcre_exec()  is  passed  the  entire
2236           string again, but with startoffset set to 4, it finds the second occur-
2237           rence of "iss" because it is able to look behind the starting point  to
2238           discover that it is preceded by a letter.
2239    
2240           If  a  non-zero starting offset is passed when the pattern is anchored,
2241           one attempt to match at the given offset is made. This can only succeed
2242           if  the  pattern  does  not require the match to be at the start of the
2243           subject.
2244    
2245       How pcre_exec() returns captured substrings
2246    
2247           In general, a pattern matches a certain portion of the subject, and  in
2248           addition,  further  substrings  from  the  subject may be picked out by
2249           parts of the pattern. Following the usage  in  Jeffrey  Friedl's  book,
2250           this  is  called "capturing" in what follows, and the phrase "capturing
2251           subpattern" is used for a fragment of a pattern that picks out  a  sub-
2252           string.  PCRE  supports several other kinds of parenthesized subpattern
2253           that do not cause substrings to be captured.
2254    
2255           Captured substrings are returned to the caller via a vector of integers
2256           whose  address is passed in ovector. The number of elements in the vec-
2257           tor is passed in ovecsize, which must be a non-negative  number.  Note:
2258           this argument is NOT the size of ovector in bytes.
2259    
2260           The  first  two-thirds of the vector is used to pass back captured sub-
2261           strings, each substring using a pair of integers. The  remaining  third
2262           of  the  vector is used as workspace by pcre_exec() while matching cap-
2263           turing subpatterns, and is not available for passing back  information.
2264           The  number passed in ovecsize should always be a multiple of three. If
2265           it is not, it is rounded down.
2266    
2267           When a match is successful, information about  captured  substrings  is
2268           returned  in  pairs  of integers, starting at the beginning of ovector,
2269           and continuing up to two-thirds of its length at the  most.  The  first
2270           element  of  each pair is set to the byte offset of the first character
2271           in a substring, and the second is set to the byte offset of  the  first
2272           character  after  the end of a substring. Note: these values are always
2273           byte offsets, even in UTF-8 mode. They are not character counts.
2274    
2275           The first pair of integers, ovector[0]  and  ovector[1],  identify  the
2276           portion  of  the subject string matched by the entire pattern. The next
2277           pair is used for the first capturing subpattern, and so on.  The  value
2278           returned by pcre_exec() is one more than the highest numbered pair that
2279           has been set.  For example, if two substrings have been  captured,  the
2280           returned  value is 3. If there are no capturing subpatterns, the return
2281           value from a successful match is 1, indicating that just the first pair
2282           of offsets has been set.
2283    
2284           If a capturing subpattern is matched repeatedly, it is the last portion
2285           of the string that it matched that is returned.
2286    
2287           If the vector is too small to hold all the captured substring  offsets,
2288           it is used as far as possible (up to two-thirds of its length), and the
2289           function returns a value of zero. If the substring offsets are  not  of
2290           interest,  pcre_exec()  may  be  called with ovector passed as NULL and
2291           ovecsize as zero. However, if the pattern contains back references  and
2292           the  ovector is not big enough to remember the related substrings, PCRE
2293           has to get additional memory for use during matching. Thus it  is  usu-
2294           ally advisable to supply an ovector.
2295    
2296           The pcre_fullinfo() function can be used to find out how many capturing
2297           subpatterns there are in a compiled  pattern.  The  smallest  size  for
2298           ovector  that  will allow for n captured substrings, in addition to the
2299           offsets of the substring matched by the whole pattern, is (n+1)*3.
2300    
2301           It is possible for capturing subpattern number n+1 to match  some  part
2302           of the subject when subpattern n has not been used at all. For example,
2303           if the string "abc" is matched  against  the  pattern  (a|(z))(bc)  the
2304           return from the function is 4, and subpatterns 1 and 3 are matched, but
2305           2 is not. When this happens, both values in  the  offset  pairs  corre-
2306           sponding to unused subpatterns are set to -1.
2307    
2308           Offset  values  that correspond to unused subpatterns at the end of the
2309           expression are also set to -1. For example,  if  the  string  "abc"  is
2310           matched  against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not
2311           matched. The return from the function is 2, because  the  highest  used
2312           capturing subpattern number is 1. However, you can refer to the offsets
2313           for the second and third capturing subpatterns if  you  wish  (assuming
2314           the vector is large enough, of course).
2315    
2316           Some  convenience  functions  are  provided for extracting the captured
2317           substrings as separate strings. These are described below.
2318    
2319       Error return values from pcre_exec()
2320    
2321           If pcre_exec() fails, it returns a negative number. The  following  are
2322           defined in the header file:
2323    
2324             PCRE_ERROR_NOMATCH        (-1)
2325    
2326           The subject string did not match the pattern.
2327    
2328             PCRE_ERROR_NULL           (-2)
2329    
2330           Either  code  or  subject  was  passed as NULL, or ovector was NULL and
2331           ovecsize was not zero.
2332    
2333             PCRE_ERROR_BADOPTION      (-3)
2334    
2335           An unrecognized bit was set in the options argument.
2336    
2337             PCRE_ERROR_BADMAGIC       (-4)
2338    
2339           PCRE stores a 4-byte "magic number" at the start of the compiled  code,
2340           to catch the case when it is passed a junk pointer and to detect when a
2341           pattern that was compiled in an environment of one endianness is run in
2342           an  environment  with the other endianness. This is the error that PCRE
2343           gives when the magic number is not present.
2344    
2345             PCRE_ERROR_UNKNOWN_OPCODE (-5)
2346    
2347           While running the pattern match, an unknown item was encountered in the
2348           compiled  pattern.  This  error  could be caused by a bug in PCRE or by
2349           overwriting of the compiled pattern.
2350    
2351             PCRE_ERROR_NOMEMORY       (-6)
2352    
2353           If a pattern contains back references, but the ovector that  is  passed
2354           to pcre_exec() is not big enough to remember the referenced substrings,
2355           PCRE gets a block of memory at the start of matching to  use  for  this
2356           purpose.  If the call via pcre_malloc() fails, this error is given. The
2357           memory is automatically freed at the end of matching.
2358    
2359           This error is also given if pcre_stack_malloc() fails  in  pcre_exec().
2360           This  can happen only when PCRE has been compiled with --disable-stack-
2361           for-recursion.
2362    
2363             PCRE_ERROR_NOSUBSTRING    (-7)
2364    
2365           This error is used by the pcre_copy_substring(),  pcre_get_substring(),
2366           and  pcre_get_substring_list()  functions  (see  below).  It  is  never
2367           returned by pcre_exec().
2368    
2369             PCRE_ERROR_MATCHLIMIT     (-8)
2370    
2371           The backtracking limit, as specified by  the  match_limit  field  in  a
2372           pcre_extra  structure  (or  defaulted) was reached. See the description
2373           above.
2374    
2375             PCRE_ERROR_CALLOUT        (-9)
2376    
2377           This error is never generated by pcre_exec() itself. It is provided for
2378           use  by  callout functions that want to yield a distinctive error code.
2379           See the pcrecallout documentation for details.
2380    
2381             PCRE_ERROR_BADUTF8        (-10)
2382    
2383           A string that contains an invalid UTF-8 byte sequence was passed  as  a
2384           subject.
2385    
2386             PCRE_ERROR_BADUTF8_OFFSET (-11)
2387    
2388           The UTF-8 byte sequence that was passed as a subject was valid, but the
2389           value of startoffset did not point to the beginning of a UTF-8  charac-
2390           ter.
2391    
2392             PCRE_ERROR_PARTIAL        (-12)
2393    
2394           The  subject  string did not match, but it did match partially. See the
2395           pcrepartial documentation for details of partial matching.
2396    
2397             PCRE_ERROR_BADPARTIAL     (-13)
2398    
2399           This code is no longer in  use.  It  was  formerly  returned  when  the
2400           PCRE_PARTIAL  option  was used with a compiled pattern containing items
2401           that were  not  supported  for  partial  matching.  From  release  8.00
2402           onwards, there are no restrictions on partial matching.
2403    
2404             PCRE_ERROR_INTERNAL       (-14)
2405    
2406           An  unexpected  internal error has occurred. This error could be caused
2407           by a bug in PCRE or by overwriting of the compiled pattern.
2408    
2409             PCRE_ERROR_BADCOUNT       (-15)
2410    
2411           This error is given if the value of the ovecsize argument is negative.
2412    
2413             PCRE_ERROR_RECURSIONLIMIT (-21)
2414    
2415           The internal recursion limit, as specified by the match_limit_recursion
2416           field  in  a  pcre_extra  structure (or defaulted) was reached. See the
2417           description above.
2418    
2419             PCRE_ERROR_BADNEWLINE     (-23)
2420    
2421           An invalid combination of PCRE_NEWLINE_xxx options was given.
2422    
2423           Error numbers -16 to -20 and -22 are not used by pcre_exec().
2424    
2425    
2426    EXTRACTING CAPTURED SUBSTRINGS BY NUMBER
2427    
2428           int pcre_copy_substring(const char *subject, int *ovector,
2429                int stringcount, int stringnumber, char *buffer,
2430                int buffersize);
2431    
2432           int pcre_get_substring(const char *subject, int *ovector,
2433                int stringcount, int stringnumber,
2434                const char **stringptr);
2435    
2436           int pcre_get_substring_list(const char *subject,
2437                int *ovector, int stringcount, const char ***listptr);
2438    
2439           Captured substrings can be  accessed  directly  by  using  the  offsets
2440           returned  by  pcre_exec()  in  ovector.  For convenience, the functions
2441           pcre_copy_substring(),    pcre_get_substring(),    and    pcre_get_sub-
2442           string_list()  are  provided for extracting captured substrings as new,
2443           separate, zero-terminated strings. These functions identify  substrings
2444           by  number.  The  next section describes functions for extracting named
2445           substrings.
2446    
2447           A substring that contains a binary zero is correctly extracted and  has
2448           a  further zero added on the end, but the result is not, of course, a C
2449           string.  However, you can process such a string  by  referring  to  the
2450           length  that  is  returned  by  pcre_copy_substring() and pcre_get_sub-
2451           string().  Unfortunately, the interface to pcre_get_substring_list() is
2452           not  adequate for handling strings containing binary zeros, because the
2453           end of the final string is not independently indicated.
2454    
2455           The first three arguments are the same for all  three  of  these  func-
2456           tions:  subject  is  the subject string that has just been successfully
2457           matched, ovector is a pointer to the vector of integer offsets that was
2458           passed to pcre_exec(), and stringcount is the number of substrings that
2459           were captured by the match, including the substring  that  matched  the
2460           entire regular expression. This is the value returned by pcre_exec() if
2461           it is greater than zero. If pcre_exec() returned zero, indicating  that
2462           it  ran out of space in ovector, the value passed as stringcount should
2463           be the number of elements in the vector divided by three.
2464    
2465           The functions pcre_copy_substring() and pcre_get_substring() extract  a
2466           single  substring,  whose  number  is given as stringnumber. A value of
2467           zero extracts the substring that matched the  entire  pattern,  whereas
2468           higher  values  extract  the  captured  substrings.  For pcre_copy_sub-
2469           string(), the string is placed in buffer,  whose  length  is  given  by
2470           buffersize,  while  for  pcre_get_substring()  a new block of memory is
2471           obtained via pcre_malloc, and its address is  returned  via  stringptr.
2472           The  yield  of  the function is the length of the string, not including
2473           the terminating zero, or one of these error codes:
2474    
2475             PCRE_ERROR_NOMEMORY       (-6)
2476    
2477           The buffer was too small for pcre_copy_substring(), or the  attempt  to
2478           get memory failed for pcre_get_substring().
2479    
2480             PCRE_ERROR_NOSUBSTRING    (-7)
2481    
2482           There is no substring whose number is stringnumber.
2483    
2484           The  pcre_get_substring_list()  function  extracts  all  available sub-
2485           strings and builds a list of pointers to them. All this is  done  in  a
2486           single block of memory that is obtained via pcre_malloc. The address of
2487           the memory block is returned via listptr, which is also  the  start  of
2488           the  list  of  string pointers. The end of the list is marked by a NULL
2489           pointer. The yield of the function is zero if all  went  well,  or  the
2490           error code
2491    
2492             PCRE_ERROR_NOMEMORY       (-6)
2493    
2494           if the attempt to get the memory block failed.
2495    
2496           When  any of these functions encounter a substring that is unset, which
2497           can happen when capturing subpattern number n+1 matches  some  part  of
2498           the  subject, but subpattern n has not been used at all, they return an
2499           empty string. This can be distinguished from a genuine zero-length sub-
2500           string  by inspecting the appropriate offset in ovector, which is nega-
2501           tive for unset substrings.
2502    
2503           The two convenience functions pcre_free_substring() and  pcre_free_sub-
2504           string_list()  can  be  used  to free the memory returned by a previous
2505           call  of  pcre_get_substring()  or  pcre_get_substring_list(),  respec-
2506           tively.  They  do  nothing  more  than  call the function pointed to by
2507           pcre_free, which of course could be called directly from a  C  program.
2508           However,  PCRE is used in some situations where it is linked via a spe-
2509           cial  interface  to  another  programming  language  that  cannot   use
2510           pcre_free  directly;  it is for these cases that the functions are pro-
2511           vided.
2512    
2513    
2514    EXTRACTING CAPTURED SUBSTRINGS BY NAME
2515    
2516           int pcre_get_stringnumber(const pcre *code,
2517                const char *name);
2518    
2519           int pcre_copy_named_substring(const pcre *code,
2520                const char *subject, int *ovector,
2521                int stringcount, const char *stringname,
2522                char *buffer, int buffersize);
2523    
2524           int pcre_get_named_substring(const pcre *code,
2525                const char *subject, int *ovector,
2526                int stringcount, const char *stringname,
2527                const char **stringptr);
2528    
2529           To extract a substring by name, you first have to find associated  num-
2530           ber.  For example, for this pattern
2531    
2532             (a+)b(?<xxx>\d+)...
2533    
2534           the number of the subpattern called "xxx" is 2. If the name is known to
2535           be unique (PCRE_DUPNAMES was not set), you can find the number from the
2536           name by calling pcre_get_stringnumber(). The first argument is the com-
2537           piled pattern, and the second is the name. The yield of the function is
2538           the  subpattern  number,  or PCRE_ERROR_NOSUBSTRING (-7) if there is no
2539           subpattern of that name.
2540    
2541           Given the number, you can extract the substring directly, or use one of
2542           the functions described in the previous section. For convenience, there
2543           are also two functions that do the whole job.
2544    
2545           Most   of   the   arguments    of    pcre_copy_named_substring()    and
2546           pcre_get_named_substring()  are  the  same  as  those for the similarly
2547           named functions that extract by number. As these are described  in  the
2548           previous  section,  they  are not re-described here. There are just two
2549           differences:
2550    
2551           First, instead of a substring number, a substring name is  given.  Sec-
2552           ond, there is an extra argument, given at the start, which is a pointer
2553           to the compiled pattern. This is needed in order to gain access to  the
2554           name-to-number translation table.
2555    
2556           These  functions call pcre_get_stringnumber(), and if it succeeds, they
2557           then call pcre_copy_substring() or pcre_get_substring(),  as  appropri-
2558           ate.  NOTE:  If PCRE_DUPNAMES is set and there are duplicate names, the
2559           behaviour may not be what you want (see the next section).
2560    
2561           Warning: If the pattern uses the (?| feature to set up multiple subpat-
2562           terns  with  the  same number, as described in the section on duplicate
2563           subpattern numbers in the pcrepattern page, you  cannot  use  names  to
2564           distinguish  the  different subpatterns, because names are not included
2565           in the compiled code. The matching process uses only numbers. For  this
2566           reason,  the  use of different names for subpatterns of the same number
2567           causes an error at compile time.
2568    
2569    
2570    DUPLICATE SUBPATTERN NAMES
2571    
2572           int pcre_get_stringtable_entries(const pcre *code,
2573                const char *name, char **first, char **last);
2574    
2575           When a pattern is compiled with the  PCRE_DUPNAMES  option,  names  for
2576           subpatterns  are not required to be unique. (Duplicate names are always
2577           allowed for subpatterns with the same number, created by using the  (?|
2578           feature.  Indeed,  if  such subpatterns are named, they are required to
2579           use the same names.)
2580    
2581           Normally, patterns with duplicate names are such that in any one match,
2582           only  one of the named subpatterns participates. An example is shown in
2583           the pcrepattern documentation.
2584    
2585           When   duplicates   are   present,   pcre_copy_named_substring()    and
2586           pcre_get_named_substring()  return the first substring corresponding to
2587           the given name that is set. If  none  are  set,  PCRE_ERROR_NOSUBSTRING
2588           (-7)  is  returned;  no  data  is returned. The pcre_get_stringnumber()
2589           function returns one of the numbers that are associated with the  name,
2590           but it is not defined which it is.
2591    
2592           If  you want to get full details of all captured substrings for a given
2593           name, you must use  the  pcre_get_stringtable_entries()  function.  The
2594           first argument is the compiled pattern, and the second is the name. The
2595           third and fourth are pointers to variables which  are  updated  by  the
2596           function. After it has run, they point to the first and last entries in
2597           the name-to-number table  for  the  given  name.  The  function  itself
2598           returns  the  length  of  each entry, or PCRE_ERROR_NOSUBSTRING (-7) if
2599           there are none. The format of the table is described above in the  sec-
2600           tion  entitled  Information  about  a  pattern.  Given all the relevant
2601           entries for the name, you can extract each of their numbers, and  hence
2602           the captured data, if any.
2603    
2604    
2605    FINDING ALL POSSIBLE MATCHES
2606    
2607           The  traditional  matching  function  uses a similar algorithm to Perl,
2608           which stops when it finds the first match, starting at a given point in
2609           the  subject.  If you want to find all possible matches, or the longest
2610           possible match, consider using the alternative matching  function  (see
2611           below)  instead.  If you cannot use the alternative function, but still
2612           need to find all possible matches, you can kludge it up by  making  use
2613           of the callout facility, which is described in the pcrecallout documen-
2614           tation.
2615    
2616           What you have to do is to insert a callout right at the end of the pat-
2617           tern.   When your callout function is called, extract and save the cur-
2618           rent matched substring. Then return  1,  which  forces  pcre_exec()  to
2619           backtrack  and  try other alternatives. Ultimately, when it runs out of
2620           matches, pcre_exec() will yield PCRE_ERROR_NOMATCH.
2621    
2622    
2623    MATCHING A PATTERN: THE ALTERNATIVE FUNCTION
2624    
2625           int pcre_dfa_exec(const pcre *code, const pcre_extra *extra,
2626                const char *subject, int length, int startoffset,
2627                int options, int *ovector, int ovecsize,
2628                int *workspace, int wscount);
2629    
2630           The function pcre_dfa_exec()  is  called  to  match  a  subject  string
2631           against  a  compiled pattern, using a matching algorithm that scans the
2632           subject string just once, and does not backtrack.  This  has  different
2633           characteristics  to  the  normal  algorithm, and is not compatible with
2634           Perl. Some of the features of PCRE patterns are not  supported.  Never-
2635           theless,  there are times when this kind of matching can be useful. For
2636           a discussion of the two matching algorithms, and  a  list  of  features
2637           that  pcre_dfa_exec() does not support, see the pcrematching documenta-
2638           tion.
2639    
2640           The arguments for the pcre_dfa_exec() function  are  the  same  as  for
2641           pcre_exec(), plus two extras. The ovector argument is used in a differ-
2642           ent way, and this is described below. The other  common  arguments  are
2643           used  in  the  same way as for pcre_exec(), so their description is not
2644           repeated here.
2645    
2646           The two additional arguments provide workspace for  the  function.  The
2647           workspace  vector  should  contain at least 20 elements. It is used for
2648           keeping  track  of  multiple  paths  through  the  pattern  tree.  More
2649           workspace  will  be  needed for patterns and subjects where there are a
2650           lot of potential matches.
2651    
2652           Here is an example of a simple call to pcre_dfa_exec():
2653    
2654             int rc;
2655             int ovector[10];
2656             int wspace[20];
2657             rc = pcre_dfa_exec(
2658               re,             /* result of pcre_compile() */
2659               NULL,           /* we didn't study the pattern */
2660               "some string",  /* the subject string */
2661               11,             /* the length of the subject string */
2662               0,              /* start at offset 0 in the subject */
2663               0,              /* default options */
2664               ovector,        /* vector of integers for substring information */
2665               10,             /* number of elements (NOT size in bytes) */
2666               wspace,         /* working space vector */
2667               20);            /* number of elements (NOT size in bytes) */
2668    
2669       Option bits for pcre_dfa_exec()
2670    
2671           The unused bits of the options argument  for  pcre_dfa_exec()  must  be
2672           zero.  The  only  bits  that  may  be  set are PCRE_ANCHORED, PCRE_NEW-
2673           LINE_xxx,        PCRE_NOTBOL,        PCRE_NOTEOL,        PCRE_NOTEMPTY,
2674           PCRE_NOTEMPTY_ATSTART,       PCRE_NO_UTF8_CHECK,      PCRE_BSR_ANYCRLF,
2675           PCRE_BSR_UNICODE, PCRE_NO_START_OPTIMIZE, PCRE_PARTIAL_HARD,  PCRE_PAR-
2676           TIAL_SOFT,  PCRE_DFA_SHORTEST,  and PCRE_DFA_RESTART.  All but the last
2677           four of these are  exactly  the  same  as  for  pcre_exec(),  so  their
2678           description is not repeated here.
2679    
2680             PCRE_PARTIAL_HARD
2681             PCRE_PARTIAL_SOFT
2682    
2683           These  have the same general effect as they do for pcre_exec(), but the
2684           details are slightly  different.  When  PCRE_PARTIAL_HARD  is  set  for
2685           pcre_dfa_exec(),  it  returns PCRE_ERROR_PARTIAL if the end of the sub-
2686           ject is reached and there is still at least  one  matching  possibility
2687           that requires additional characters. This happens even if some complete
2688           matches have also been found. When PCRE_PARTIAL_SOFT is set, the return
2689           code PCRE_ERROR_NOMATCH is converted into PCRE_ERROR_PARTIAL if the end
2690           of the subject is reached, there have been  no  complete  matches,  but
2691           there  is  still  at least one matching possibility. The portion of the
2692           string that was inspected when the longest partial match was  found  is
2693           set as the first matching string in both cases.
2694    
2695             PCRE_DFA_SHORTEST
2696    
2697           Setting  the  PCRE_DFA_SHORTEST option causes the matching algorithm to
2698           stop as soon as it has found one match. Because of the way the alterna-
2699           tive  algorithm  works, this is necessarily the shortest possible match
2700           at the first possible matching point in the subject string.
2701    
2702             PCRE_DFA_RESTART
2703    
2704           When pcre_dfa_exec() returns a partial match, it is possible to call it
2705           again,  with  additional  subject characters, and have it continue with
2706           the same match. The PCRE_DFA_RESTART option requests this action;  when
2707           it  is  set,  the workspace and wscount options must reference the same
2708           vector as before because data about the match so far is  left  in  them
2709           after a partial match. There is more discussion of this facility in the
2710           pcrepartial documentation.
2711    
2712       Successful returns from pcre_dfa_exec()
2713    
2714           When pcre_dfa_exec() succeeds, it may have matched more than  one  sub-
2715           string in the subject. Note, however, that all the matches from one run
2716           of the function start at the same point in  the  subject.  The  shorter
2717           matches  are all initial substrings of the longer matches. For example,
2718           if the pattern
2719    
2720             <.*>
2721    
2722           is matched against the string
2723    
2724             This is <something> <something else> <something further> no more
2725    
2726           the three matched strings are
2727    
2728             <something>
2729             <something> <something else>
2730             <something> <something else> <something further>
2731    
2732           On success, the yield of the function is a number  greater  than  zero,
2733           which  is  the  number of matched substrings. The substrings themselves
2734           are returned in ovector. Each string uses two elements;  the  first  is
2735           the  offset  to  the start, and the second is the offset to the end. In
2736           fact, all the strings have the same start  offset.  (Space  could  have
2737           been  saved by giving this only once, but it was decided to retain some
2738           compatibility with the way pcre_exec() returns data,  even  though  the
2739           meaning of the strings is different.)
2740    
2741           The strings are returned in reverse order of length; that is, the long-
2742           est matching string is given first. If there were too many  matches  to
2743           fit  into ovector, the yield of the function is zero, and the vector is
2744           filled with the longest matches.
2745    
2746       Error returns from pcre_dfa_exec()
2747    
2748           The pcre_dfa_exec() function returns a negative number when  it  fails.
2749           Many  of  the  errors  are  the  same as for pcre_exec(), and these are
2750           described above.  There are in addition the following errors  that  are
2751           specific to pcre_dfa_exec():
2752    
2753             PCRE_ERROR_DFA_UITEM      (-16)
2754    
2755           This  return is given if pcre_dfa_exec() encounters an item in the pat-
2756           tern that it does not support, for instance, the use of \C  or  a  back
2757           reference.
2758    
2759             PCRE_ERROR_DFA_UCOND      (-17)
2760    
2761           This  return  is  given  if pcre_dfa_exec() encounters a condition item
2762           that uses a back reference for the condition, or a test  for  recursion
2763           in a specific group. These are not supported.
2764    
2765             PCRE_ERROR_DFA_UMLIMIT    (-18)
2766    
2767           This  return  is given if pcre_dfa_exec() is called with an extra block
2768           that contains a setting of the match_limit field. This is not supported
2769           (it is meaningless).
2770    
2771             PCRE_ERROR_DFA_WSSIZE     (-19)
2772    
2773           This  return  is  given  if  pcre_dfa_exec()  runs  out of space in the
2774           workspace vector.
2775    
2776             PCRE_ERROR_DFA_RECURSE    (-20)
2777    
2778           When a recursive subpattern is processed, the matching  function  calls
2779           itself  recursively,  using  private vectors for ovector and workspace.
2780           This error is given if the output vector  is  not  large  enough.  This
2781           should be extremely rare, as a vector of size 1000 is used.
2782    
2783    
2784    SEE ALSO
2785    
2786           pcrebuild(3),  pcrecallout(3), pcrecpp(3)(3), pcrematching(3), pcrepar-
2787           tial(3), pcreposix(3), pcreprecompile(3), pcresample(3), pcrestack(3).
2788    
2789    
2790    AUTHOR
2791    
2792           Philip Hazel
2793           University Computing Service
2794           Cambridge CB2 3QH, England.
2795    
2796    
2797    REVISION
2798    
2799           Last updated: 21 June 2010
2800           Copyright (c) 1997-2010 University of Cambridge.
2801    ------------------------------------------------------------------------------
2802    
2803    
2804    PCRECALLOUT(3)                                                  PCRECALLOUT(3)
2805    
2806    
2807    NAME
2808           PCRE - Perl-compatible regular expressions
2809    
2810    
2811    PCRE CALLOUTS
2812    
2813           int (*pcre_callout)(pcre_callout_block *);
2814    
2815           PCRE provides a feature called "callout", which is a means of temporar-
2816           ily passing control to the caller of PCRE  in  the  middle  of  pattern
2817           matching.  The  caller of PCRE provides an external function by putting
2818           its entry point in the global variable pcre_callout. By  default,  this
2819           variable contains NULL, which disables all calling out.
2820    
2821           Within  a  regular  expression,  (?C) indicates the points at which the
2822           external function is to be called.  Different  callout  points  can  be
2823           identified  by  putting  a number less than 256 after the letter C. The
2824           default value is zero.  For  example,  this  pattern  has  two  callout
2825           points:
2826    
2827             (?C1)abc(?C2)def
2828    
2829           If  the  PCRE_AUTO_CALLOUT  option  bit  is  set when pcre_compile() or
2830           pcre_compile2() is called, PCRE  automatically  inserts  callouts,  all
2831           with  number  255,  before  each  item  in the pattern. For example, if
2832           PCRE_AUTO_CALLOUT is used with the pattern
2833    
2834             A(\d{2}|--)
2835    
2836           it is processed as if it were
2837    
2838           (?C255)A(?C255)((?C255)\d{2}(?C255)|(?C255)-(?C255)-(?C255))(?C255)
2839    
2840           Notice that there is a callout before and after  each  parenthesis  and
2841           alternation  bar.  Automatic  callouts  can  be  used  for tracking the
2842           progress of pattern matching. The pcretest command has an  option  that
2843           sets  automatic callouts; when it is used, the output indicates how the
2844           pattern is matched. This is useful information when you are  trying  to
2845           optimize the performance of a particular pattern.
2846    
2847    
2848    MISSING CALLOUTS
2849    
2850           You  should  be  aware  that,  because of optimizations in the way PCRE
2851           matches patterns by default, callouts  sometimes  do  not  happen.  For
2852           example, if the pattern is
2853    
2854             ab(?C4)cd
2855    
2856           PCRE knows that any matching string must contain the letter "d". If the
2857           subject string is "abyz", the lack of "d" means that  matching  doesn't
2858           ever  start,  and  the  callout is never reached. However, with "abyd",
2859           though the result is still no match, the callout is obeyed.
2860    
2861           If the pattern is studied, PCRE knows the minimum length of a  matching
2862           string,  and will immediately give a "no match" return without actually
2863           running a match if the subject is not long enough, or,  for  unanchored
2864           patterns, if it has been scanned far enough.
2865    
2866           You  can disable these optimizations by passing the PCRE_NO_START_OPTI-
2867           MIZE option to pcre_exec() or  pcre_dfa_exec().  This  slows  down  the
2868           matching  process,  but  does  ensure that callouts such as the example
2869           above are obeyed.
2870    
2871    
2872    THE CALLOUT INTERFACE
2873    
2874           During matching, when PCRE reaches a callout point, the external  func-
2875           tion  defined by pcre_callout is called (if it is set). This applies to
2876           both the pcre_exec() and the pcre_dfa_exec()  matching  functions.  The
2877           only  argument  to  the callout function is a pointer to a pcre_callout
2878           block. This structure contains the following fields:
2879    
2880             int          version;
2881             int          callout_number;
2882             int         *offset_vector;
2883             const char  *subject;
2884             int          subject_length;
2885             int          start_match;
2886             int          current_position;
2887             int          capture_top;
2888             int          capture_last;
2889             void        *callout_data;
2890             int          pattern_position;
2891             int          next_item_length;
2892    
2893           The version field is an integer containing the version  number  of  the
2894           block  format. The initial version was 0; the current version is 1. The
2895           version number will change again in future  if  additional  fields  are
2896           added, but the intention is never to remove any of the existing fields.
2897    
2898           The  callout_number  field  contains the number of the callout, as com-
2899           piled into the pattern (that is, the number after ?C for  manual  call-
2900           outs, and 255 for automatically generated callouts).
2901    
2902           The  offset_vector field is a pointer to the vector of offsets that was
2903           passed  by  the  caller  to  pcre_exec()   or   pcre_dfa_exec().   When
2904           pcre_exec()  is used, the contents can be inspected in order to extract
2905           substrings that have been matched so  far,  in  the  same  way  as  for
2906           extracting  substrings after a match has completed. For pcre_dfa_exec()
2907           this field is not useful.
2908    
2909           The subject and subject_length fields contain copies of the values that
2910           were passed to pcre_exec().
2911    
2912           The  start_match  field normally contains the offset within the subject
2913           at which the current match attempt  started.  However,  if  the  escape
2914           sequence  \K has been encountered, this value is changed to reflect the
2915           modified starting point. If the pattern is not  anchored,  the  callout
2916           function may be called several times from the same point in the pattern
2917           for different starting points in the subject.
2918    
2919           The current_position field contains the offset within  the  subject  of
2920           the current match pointer.
2921    
2922           When  the  pcre_exec() function is used, the capture_top field contains
2923           one more than the number of the highest numbered captured substring  so
2924           far.  If  no substrings have been captured, the value of capture_top is
2925           one. This is always the case when pcre_dfa_exec() is used,  because  it
2926           does not support captured substrings.
2927    
2928           The  capture_last  field  contains the number of the most recently cap-
2929           tured substring. If no substrings have been captured, its value is  -1.
2930           This is always the case when pcre_dfa_exec() is used.
2931    
2932           The  callout_data  field contains a value that is passed to pcre_exec()
2933           or pcre_dfa_exec() specifically so that it can be passed back in  call-
2934           outs.  It  is  passed  in the pcre_callout field of the pcre_extra data
2935           structure. If no such data was passed, the value of callout_data  in  a
2936           pcre_callout  block  is  NULL. There is a description of the pcre_extra
2937           structure in the pcreapi documentation.
2938    
2939           The pattern_position field is present from version 1 of the  pcre_call-
2940           out structure. It contains the offset to the next item to be matched in
2941           the pattern string.
2942    
2943           The next_item_length field is present from version 1 of the  pcre_call-
2944           out structure. It contains the length of the next item to be matched in
2945           the pattern string. When the callout immediately precedes  an  alterna-
2946           tion  bar, a closing parenthesis, or the end of the pattern, the length
2947           is zero. When the callout precedes an opening parenthesis,  the  length
2948           is that of the entire subpattern.
2949    
2950           The  pattern_position  and next_item_length fields are intended to help
2951           in distinguishing between different automatic callouts, which all  have
2952           the same callout number. However, they are set for all callouts.
2953    
2954    
2955    RETURN VALUES
2956    
2957           The  external callout function returns an integer to PCRE. If the value
2958           is zero, matching proceeds as normal. If  the  value  is  greater  than
2959           zero,  matching  fails  at  the current point, but the testing of other
2960           matching possibilities goes ahead, just as if a lookahead assertion had
2961           failed.  If  the  value  is less than zero, the match is abandoned, and
2962           pcre_exec() or pcre_dfa_exec() returns the negative value.
2963    
2964           Negative  values  should  normally  be   chosen   from   the   set   of
2965           PCRE_ERROR_xxx values. In particular, PCRE_ERROR_NOMATCH forces a stan-
2966           dard "no  match"  failure.   The  error  number  PCRE_ERROR_CALLOUT  is
2967           reserved  for  use  by callout functions; it will never be used by PCRE
2968           itself.
2969    
2970    
2971    AUTHOR
2972    
2973           Philip Hazel
2974           University Computing Service
2975           Cambridge CB2 3QH, England.
2976    
2977    
2978    REVISION
2979    
2980           Last updated: 29 September 2009
2981           Copyright (c) 1997-2009 University of Cambridge.
2982    ------------------------------------------------------------------------------
2983    
2984    
2985    PCRECOMPAT(3)                                                    PCRECOMPAT(3)
2986    
2987    
2988    NAME
2989           PCRE - Perl-compatible regular expressions
2990    
2991    
2992    DIFFERENCES BETWEEN PCRE AND PERL
2993    
2994           This  document describes the differences in the ways that PCRE and Perl
2995           handle regular expressions. The differences  described  here  are  with
2996           respect to Perl 5.10/5.11.
2997    
2998           1.  PCRE has only a subset of Perl's UTF-8 and Unicode support. Details
2999           of what it does have are given in the section on UTF-8 support  in  the
3000           main pcre page.
3001    
3002           2. PCRE does not allow repeat quantifiers on lookahead assertions. Perl
3003           permits them, but they do not mean what you might think.  For  example,
3004           (?!a){3} does not assert that the next three characters are not "a". It
3005           just asserts that the next character is not "a" three times.
3006    
3007           3. Capturing subpatterns that occur inside  negative  lookahead  asser-
3008           tions  are  counted,  but their entries in the offsets vector are never
3009           set. Perl sets its numerical variables from any such patterns that  are
3010           matched before the assertion fails to match something (thereby succeed-
3011           ing), but only if the negative lookahead assertion  contains  just  one
3012           branch.
3013    
3014           4.  Though  binary zero characters are supported in the subject string,
3015           they are not allowed in a pattern string because it is passed as a nor-
3016           mal C string, terminated by zero. The escape sequence \0 can be used in
3017           the pattern to represent a binary zero.
3018    
3019           5. The following Perl escape sequences are not supported: \l,  \u,  \L,
3020           \U, and \N. In fact these are implemented by Perl's general string-han-
3021           dling and are not part of its pattern matching engine. If any of  these
3022           are encountered by PCRE, an error is generated.
3023    
3024           6.  The Perl escape sequences \p, \P, and \X are supported only if PCRE
3025           is built with Unicode character property support. The  properties  that
3026           can  be tested with \p and \P are limited to the general category prop-
3027           erties such as Lu and Nd, script names such as Greek or  Han,  and  the
3028           derived  properties  Any  and  L&. PCRE does support the Cs (surrogate)
3029           property, which Perl does not; the  Perl  documentation  says  "Because
3030           Perl hides the need for the user to understand the internal representa-
3031           tion of Unicode characters, there is no need to implement the  somewhat
3032           messy concept of surrogates."
3033    
3034           7. PCRE does support the \Q...\E escape for quoting substrings. Charac-
3035           ters in between are treated as literals.  This  is  slightly  different
3036           from  Perl  in  that  $  and  @ are also handled as literals inside the
3037           quotes. In Perl, they cause variable interpolation (but of course  PCRE
3038           does not have variables). Note the following examples:
3039    
3040               Pattern            PCRE matches      Perl matches
3041    
3042               \Qabc$xyz\E        abc$xyz           abc followed by the
3043                                                      contents of $xyz
3044               \Qabc\$xyz\E       abc\$xyz          abc\$xyz
3045               \Qabc\E\$\Qxyz\E   abc$xyz           abc$xyz
3046    
3047           The  \Q...\E  sequence  is recognized both inside and outside character
3048           classes.
3049    
3050           8. Fairly obviously, PCRE does not support the (?{code}) and (??{code})
3051           constructions.  However,  there is support for recursive patterns. This
3052           is not available in Perl 5.8, but it is in Perl 5.10.  Also,  the  PCRE
3053           "callout"  feature allows an external function to be called during pat-
3054           tern matching. See the pcrecallout documentation for details.
3055    
3056           9. Subpatterns that are called  recursively  or  as  "subroutines"  are
3057           always  treated  as  atomic  groups  in  PCRE. This is like Python, but
3058           unlike Perl. There is a discussion of an example that explains this  in
3059           more  detail  in  the section on recursion differences from Perl in the
3060           pcrepattern page.
3061    
3062           10. There are some differences that are concerned with the settings  of
3063           captured  strings  when  part  of  a  pattern is repeated. For example,
3064           matching "aba" against the  pattern  /^(a(b)?)+$/  in  Perl  leaves  $2
3065           unset, but in PCRE it is set to "b".
3066    
3067           11.  PCRE's handling of duplicate subpattern numbers and duplicate sub-
3068           pattern names is not as general as Perl's. This is a consequence of the
3069           fact the PCRE works internally just with numbers, using an external ta-
3070           ble to translate between numbers and names. In  particular,  a  pattern
3071           such  as  (?|(?<a>A)|(?<b)B),  where the two capturing parentheses have
3072           the same number but different names, is not supported,  and  causes  an
3073           error  at compile time. If it were allowed, it would not be possible to
3074           distinguish which parentheses matched, because both names map  to  cap-
3075           turing subpattern number 1. To avoid this confusing situation, an error
3076           is given at compile time.
3077    
3078           12. PCRE provides some extensions to the Perl regular expression facil-
3079           ities.   Perl  5.10  includes new features that are not in earlier ver-
3080           sions of Perl, some of which (such as named parentheses) have  been  in
3081           PCRE for some time. This list is with respect to Perl 5.10:
3082    
3083           (a)  Although  lookbehind  assertions  in  PCRE must match fixed length
3084           strings, each alternative branch of a lookbehind assertion can match  a
3085           different  length  of  string.  Perl requires them all to have the same
3086           length.
3087    
3088           (b) If PCRE_DOLLAR_ENDONLY is set and PCRE_MULTILINE is not set, the  $
3089           meta-character matches only at the very end of the string.
3090    
3091           (c) If PCRE_EXTRA is set, a backslash followed by a letter with no spe-
3092           cial meaning is faulted. Otherwise, like Perl, the backslash is quietly
3093           ignored.  (Perl can be made to issue a warning.)
3094    
3095           (d)  If  PCRE_UNGREEDY is set, the greediness of the repetition quanti-
3096           fiers is inverted, that is, by default they are not greedy, but if fol-
3097           lowed by a question mark they are.
3098    
3099           (e) PCRE_ANCHORED can be used at matching time to force a pattern to be
3100           tried only at the first matching position in the subject string.
3101    
3102           (f) The PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
3103           and  PCRE_NO_AUTO_CAPTURE  options for pcre_exec() have no Perl equiva-
3104           lents.
3105    
3106           (g) The \R escape sequence can be restricted to match only CR,  LF,  or
3107           CRLF by the PCRE_BSR_ANYCRLF option.
3108    
3109           (h) The callout facility is PCRE-specific.
3110    
3111           (i) The partial matching facility is PCRE-specific.
3112    
3113           (j) Patterns compiled by PCRE can be saved and re-used at a later time,
3114           even on different hosts that have the other endianness.
3115    
3116           (k) The alternative matching function (pcre_dfa_exec())  matches  in  a
3117           different way and is not Perl-compatible.
3118    
3119           (l)  PCRE  recognizes some special sequences such as (*CR) at the start
3120           of a pattern that set overall options that cannot be changed within the
3121           pattern.
3122    
3123    
3124    AUTHOR
3125    
3126           Philip Hazel
3127           University Computing Service
3128           Cambridge CB2 3QH, England.
3129    
3130    
3131    REVISION
3132    
3133           Last updated: 12 May 2010
3134           Copyright (c) 1997-2010 University of Cambridge.
3135    ------------------------------------------------------------------------------
3136    
3137    
3138    PCREPATTERN(3)                                                  PCREPATTERN(3)
3139    
3140    
3141    NAME
3142           PCRE - Perl-compatible regular expressions
3143    
3144    
3145    PCRE REGULAR EXPRESSION DETAILS
3146    
3147           The  syntax and semantics of the regular expressions that are supported
3148           by PCRE are described in detail below. There is a quick-reference  syn-
3149           tax summary in the pcresyntax page. PCRE tries to match Perl syntax and
3150           semantics as closely as it can. PCRE  also  supports  some  alternative
3151           regular  expression  syntax (which does not conflict with the Perl syn-
3152           tax) in order to provide some compatibility with regular expressions in
3153           Python, .NET, and Oniguruma.
3154    
3155           Perl's  regular expressions are described in its own documentation, and
3156           regular expressions in general are covered in a number of  books,  some
3157           of  which  have  copious  examples. Jeffrey Friedl's "Mastering Regular
3158           Expressions", published by  O'Reilly,  covers  regular  expressions  in
3159           great  detail.  This  description  of  PCRE's  regular  expressions  is
3160           intended as reference material.
3161    
3162           The original operation of PCRE was on strings of  one-byte  characters.
3163           However,  there is now also support for UTF-8 character strings. To use
3164           this, PCRE must be built to include UTF-8 support, and  you  must  call
3165           pcre_compile()  or  pcre_compile2() with the PCRE_UTF8 option. There is
3166           also a special sequence that can be given at the start of a pattern:
3167    
3168             (*UTF8)
3169    
3170           Starting a pattern with this sequence  is  equivalent  to  setting  the
3171           PCRE_UTF8  option.  This  feature  is  not Perl-compatible. How setting
3172           UTF-8 mode affects pattern matching  is  mentioned  in  several  places
3173           below.  There  is  also  a  summary of UTF-8 features in the section on
3174           UTF-8 support in the main pcre page.
3175    
3176           Another special sequence that may appear at the start of a  pattern  or
3177           in combination with (*UTF8) is:
3178    
3179             (*UCP)
3180    
3181           This  has  the  same  effect  as setting the PCRE_UCP option: it causes
3182           sequences such as \d and \w to  use  Unicode  properties  to  determine
3183           character types, instead of recognizing only characters with codes less
3184           than 128 via a lookup table.
3185    
3186           The remainder of this document discusses the  patterns  that  are  sup-
3187           ported  by  PCRE when its main matching function, pcre_exec(), is used.
3188           From  release  6.0,   PCRE   offers   a   second   matching   function,
3189           pcre_dfa_exec(),  which matches using a different algorithm that is not
3190           Perl-compatible. Some of the features discussed below are not available
3191           when  pcre_dfa_exec()  is used. The advantages and disadvantages of the
3192           alternative function, and how it differs from the normal function,  are
3193           discussed in the pcrematching page.
3194    
3195    
3196    NEWLINE CONVENTIONS
3197    
3198           PCRE  supports five different conventions for indicating line breaks in
3199           strings: a single CR (carriage return) character, a  single  LF  (line-
3200           feed) character, the two-character sequence CRLF, any of the three pre-
3201           ceding, or any Unicode newline sequence. The pcreapi page  has  further
3202           discussion  about newlines, and shows how to set the newline convention
3203           in the options arguments for the compiling and matching functions.
3204    
3205           It is also possible to specify a newline convention by starting a  pat-
3206           tern string with one of the following five sequences:
3207    
3208             (*CR)        carriage return
3209             (*LF)        linefeed
3210             (*CRLF)      carriage return, followed by linefeed
3211             (*ANYCRLF)   any of the three above
3212             (*ANY)       all Unicode newline sequences
3213    
3214           These  override  the default and the options given to pcre_compile() or
3215           pcre_compile2(). For example, on a Unix system where LF is the  default
3216           newline sequence, the pattern
3217    
3218             (*CR)a.b
3219    
3220           changes the convention to CR. That pattern matches "a\nb" because LF is
3221           no longer a newline. Note that these special settings,  which  are  not
3222           Perl-compatible,  are  recognized  only at the very start of a pattern,
3223           and that they must be in upper case.  If  more  than  one  of  them  is
3224           present, the last one is used.
3225    
3226           The  newline convention affects the interpretation of the dot metachar-
3227           acter when PCRE_DOTALL is not set, and also the behaviour of  \N.  How-
3228           ever,  it  does  not  affect  what  the  \R escape sequence matches. By
3229           default, this is any Unicode newline sequence, for Perl  compatibility.
3230           However,  this can be changed; see the description of \R in the section
3231           entitled "Newline sequences" below. A change of \R setting can be  com-
3232           bined with a change of newline convention.
3233    
3234    
3235    CHARACTERS AND METACHARACTERS
3236    
3237           A  regular  expression  is  a pattern that is matched against a subject
3238           string from left to right. Most characters stand for  themselves  in  a
3239           pattern,  and  match  the corresponding characters in the subject. As a
3240           trivial example, the pattern
3241    
3242             The quick brown fox
3243    
3244           matches a portion of a subject string that is identical to itself. When
3245           caseless  matching is specified (the PCRE_CASELESS option), letters are
3246           matched independently of case. In UTF-8 mode, PCRE  always  understands
3247           the  concept  of case for characters whose values are less than 128, so
3248           caseless matching is always possible. For characters with  higher  val-
3249           ues,  the concept of case is supported if PCRE is compiled with Unicode
3250           property support, but not otherwise.   If  you  want  to  use  caseless
3251           matching  for  characters  128  and above, you must ensure that PCRE is
3252           compiled with Unicode property support as well as with UTF-8 support.
3253    
3254           The power of regular expressions comes  from  the  ability  to  include
3255           alternatives  and  repetitions in the pattern. These are encoded in the
3256           pattern by the use of metacharacters, which do not stand for themselves
3257           but instead are interpreted in some special way.
3258    
3259           There  are  two different sets of metacharacters: those that are recog-
3260           nized anywhere in the pattern except within square brackets, and  those
3261           that  are  recognized  within square brackets. Outside square brackets,
3262           the metacharacters are as follows:
3263    
3264             \      general escape character with several uses
3265             ^      assert start of string (or line, in multiline mode)
3266             $      assert end of string (or line, in multiline mode)
3267             .      match any character except newline (by default)
3268             [      start character class definition
3269             |      start of alternative branch
3270             (      start subpattern
3271             )      end subpattern
3272             ?      extends the meaning of (
3273                    also 0 or 1 quantifier
3274                    also quantifier minimizer
3275             *      0 or more quantifier
3276             +      1 or more quantifier
3277                    also "possessive quantifier"
3278             {      start min/max quantifier
3279    
3280           Part of a pattern that is in square brackets  is  called  a  "character
3281           class". In a character class the only metacharacters are:
3282    
3283             \      general escape character
3284             ^      negate the class, but only if the first character
3285             -      indicates character range
3286             [      POSIX character class (only if followed by POSIX
3287                      syntax)
3288             ]      terminates the character class
3289    
3290           The following sections describe the use of each of the metacharacters.
3291    
3292    
3293    BACKSLASH
3294    
3295           The backslash character has several uses. Firstly, if it is followed by
3296           a non-alphanumeric character, it takes away any  special  meaning  that
3297           character  may  have.  This  use  of  backslash  as an escape character
3298           applies both inside and outside character classes.
3299    
3300           For example, if you want to match a * character, you write  \*  in  the
3301           pattern.   This  escaping  action  applies whether or not the following
3302           character would otherwise be interpreted as a metacharacter, so  it  is
3303           always  safe  to  precede  a non-alphanumeric with backslash to specify
3304           that it stands for itself. In particular, if you want to match a  back-
3305           slash, you write \\.
3306    
3307           If  a  pattern is compiled with the PCRE_EXTENDED option, whitespace in
3308           the pattern (other than in a character class) and characters between  a
3309           # outside a character class and the next newline are ignored. An escap-
3310           ing backslash can be used to include a whitespace  or  #  character  as
3311           part of the pattern.
3312    
3313           If  you  want  to remove the special meaning from a sequence of charac-
3314           ters, you can do so by putting them between \Q and \E. This is  differ-
3315           ent  from  Perl  in  that  $  and  @ are handled as literals in \Q...\E
3316           sequences in PCRE, whereas in Perl, $ and @ cause  variable  interpola-
3317           tion. Note the following examples:
3318    
3319             Pattern            PCRE matches   Perl matches
3320    
3321             \Qabc$xyz\E        abc$xyz        abc followed by the
3322                                                 contents of $xyz
3323             \Qabc\$xyz\E       abc\$xyz       abc\$xyz
3324             \Qabc\E\$\Qxyz\E   abc$xyz        abc$xyz
3325    
3326           The  \Q...\E  sequence  is recognized both inside and outside character
3327           classes.
3328    
3329       Non-printing characters
3330    
3331           A second use of backslash provides a way of encoding non-printing char-
3332           acters  in patterns in a visible manner. There is no restriction on the
3333           appearance of non-printing characters, apart from the binary zero  that
3334           terminates  a  pattern,  but  when  a pattern is being prepared by text
3335           editing, it is  often  easier  to  use  one  of  the  following  escape
3336           sequences than the binary character it represents:
3337    
3338             \a        alarm, that is, the BEL character (hex 07)
3339             \cx       "control-x", where x is any character
3340             \e        escape (hex 1B)
3341             \f        formfeed (hex 0C)
3342             \n        linefeed (hex 0A)
3343             \r        carriage return (hex 0D)
3344             \t        tab (hex 09)
3345             \ddd      character with octal code ddd, or back reference
3346             \xhh      character with hex code hh
3347             \x{hhh..} character with hex code hhh..
3348    
3349           The  precise  effect of \cx is as follows: if x is a lower case letter,
3350           it is converted to upper case. Then bit 6 of the character (hex 40)  is
3351           inverted.   Thus  \cz becomes hex 1A, but \c{ becomes hex 3B, while \c;
3352           becomes hex 7B.
3353    
3354           After \x, from zero to two hexadecimal digits are read (letters can  be
3355           in  upper  or  lower case). Any number of hexadecimal digits may appear
3356           between \x{ and }, but the value of the character  code  must  be  less
3357           than 256 in non-UTF-8 mode, and less than 2**31 in UTF-8 mode. That is,
3358           the maximum value in hexadecimal is 7FFFFFFF. Note that this is  bigger
3359           than the largest Unicode code point, which is 10FFFF.
3360    
3361           If  characters  other than hexadecimal digits appear between \x{ and },
3362           or if there is no terminating }, this form of escape is not recognized.
3363           Instead,  the  initial  \x  will  be interpreted as a basic hexadecimal
3364           escape, with no following digits, giving a  character  whose  value  is
3365           zero.
3366    
3367           Characters whose value is less than 256 can be defined by either of the
3368           two syntaxes for \x. There is no difference in the way  they  are  han-
3369           dled. For example, \xdc is exactly the same as \x{dc}.
3370    
3371           After  \0  up  to two further octal digits are read. If there are fewer
3372           than two digits, just  those  that  are  present  are  used.  Thus  the
3373           sequence \0\x\07 specifies two binary zeros followed by a BEL character
3374           (code value 7). Make sure you supply two digits after the initial  zero
3375           if the pattern character that follows is itself an octal digit.
3376    
3377           The handling of a backslash followed by a digit other than 0 is compli-
3378           cated.  Outside a character class, PCRE reads it and any following dig-
3379           its  as  a  decimal  number. If the number is less than 10, or if there
3380           have been at least that many previous capturing left parentheses in the
3381           expression,  the  entire  sequence  is  taken  as  a  back reference. A
3382           description of how this works is given later, following the  discussion
3383           of parenthesized subpatterns.
3384    
3385           Inside  a  character  class, or if the decimal number is greater than 9
3386           and there have not been that many capturing subpatterns, PCRE  re-reads
3387           up to three octal digits following the backslash, and uses them to gen-
3388           erate a data character. Any subsequent digits stand for themselves.  In
3389           non-UTF-8  mode,  the  value  of a character specified in octal must be
3390           less than \400. In UTF-8 mode, values up to  \777  are  permitted.  For
3391           example:
3392    
3393             \040   is another way of writing a space
3394             \40    is the same, provided there are fewer than 40
3395                       previous capturing subpatterns
3396             \7     is always a back reference
3397             \11    might be a back reference, or another way of
3398                       writing a tab
3399             \011   is always a tab
3400             \0113  is