Annotation of win32/pcre/pcre.txt, revision 1.7
1.1 misha 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
1.4 misha 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.
1.1 misha 8: -----------------------------------------------------------------------------
9:
10:
1.6 misha 11: PCRE(3) Library Functions Manual PCRE(3)
12:
1.1 misha 13:
14:
15: NAME
1.7 ! moko 16: PCRE - Perl-compatible regular expressions (original API)
! 17:
! 18: PLEASE TAKE NOTE
! 19:
! 20: This document relates to PCRE releases that use the original API, with
! 21: library names libpcre, libpcre16, and libpcre32. January 2015 saw the
! 22: first release of a new API, known as PCRE2, with release numbers start-
! 23: ing at 10.00 and library names libpcre2-8, libpcre2-16, and
! 24: libpcre2-32. The old libraries (now called PCRE1) are still being main-
! 25: tained for bug fixes, but there will be no new development. New
! 26: projects are advised to use the new PCRE2 libraries.
! 27:
1.1 misha 28:
29: INTRODUCTION
30:
31: The PCRE library is a set of functions that implement regular expres-
32: sion pattern matching using the same syntax and semantics as Perl, with
1.4 misha 33: just a few differences. Some features that appeared in Python and PCRE
34: before they appeared in Perl are also available using the Python syn-
35: tax, there is some support for one or two .NET and Oniguruma syntax
36: items, and there is an option for requesting some minor changes that
37: give better JavaScript compatibility.
38:
1.5 misha 39: Starting with release 8.30, it is possible to compile two separate PCRE
40: libraries: the original, which supports 8-bit character strings
41: (including UTF-8 strings), and a second library that supports 16-bit
42: character strings (including UTF-16 strings). The build process allows
43: either one or both to be built. The majority of the work to make this
44: possible was done by Zoltan Herczeg.
45:
1.6 misha 46: Starting with release 8.32 it is possible to compile a third separate
47: PCRE library that supports 32-bit character strings (including UTF-32
48: strings). The build process allows any combination of the 8-, 16- and
49: 32-bit libraries. The work to make this possible was done by Christian
50: Persch.
51:
52: The three libraries contain identical sets of functions, except that
53: the names in the 16-bit library start with pcre16_ instead of pcre_,
54: and the names in the 32-bit library start with pcre32_ instead of
55: pcre_. To avoid over-complication and reduce the documentation mainte-
56: nance load, most of the documentation describes the 8-bit library, with
57: the differences for the 16-bit and 32-bit libraries described sepa-
58: rately in the pcre16 and pcre32 pages. References to functions or
59: structures of the form pcre[16|32]_xxx should be read as meaning
60: "pcre_xxx when using the 8-bit library, pcre16_xxx when using the
61: 16-bit library, or pcre32_xxx when using the 32-bit library".
1.5 misha 62:
1.4 misha 63: The current implementation of PCRE corresponds approximately with Perl
1.6 misha 64: 5.12, including support for UTF-8/16/32 encoded strings and Unicode
65: general category properties. However, UTF-8/16/32 and Unicode support
66: has to be explicitly enabled; it is not the default. The Unicode tables
1.7 ! moko 67: correspond to Unicode release 6.3.0.
1.1 misha 68:
69: In addition to the Perl-compatible matching function, PCRE contains an
1.4 misha 70: alternative function that matches the same compiled patterns in a dif-
71: ferent way. In certain circumstances, the alternative function has some
72: advantages. For a discussion of the two matching algorithms, see the
73: pcrematching page.
1.1 misha 74:
75: PCRE is written in C and released as a C library. A number of people
76: have written wrappers and interfaces of various kinds. In particular,
1.5 misha 77: Google Inc. have provided a comprehensive C++ wrapper for the 8-bit
78: library. This is now included as part of the PCRE distribution. The
79: pcrecpp page has details of this interface. Other people's contribu-
80: tions can be found in the Contrib directory at the primary FTP site,
81: which is:
1.1 misha 82:
83: ftp://ftp.csx.cam.ac.uk/pub/software/programming/pcre
84:
1.5 misha 85: Details of exactly which Perl regular expression features are and are
1.1 misha 86: not supported by PCRE are given in separate documents. See the pcrepat-
1.5 misha 87: tern and pcrecompat pages. There is a syntax summary in the pcresyntax
1.1 misha 88: page.
89:
1.5 misha 90: Some features of PCRE can be included, excluded, or changed when the
91: library is built. The pcre_config() function makes it possible for a
92: client to discover which features are available. The features them-
93: selves are described in the pcrebuild page. Documentation about build-
94: ing PCRE for various operating systems can be found in the README and
1.6 misha 95: NON-AUTOTOOLS_BUILD files in the source distribution.
1.1 misha 96:
1.5 misha 97: The libraries contains a number of undocumented internal functions and
98: data tables that are used by more than one of the exported external
99: functions, but which are not intended for use by external callers.
1.6 misha 100: Their names all begin with "_pcre_" or "_pcre16_" or "_pcre32_", which
101: hopefully will not provoke any name clashes. In some environments, it
102: is possible to control which external symbols are exported when a
103: shared library is built, and in these cases the undocumented symbols
104: are not exported.
105:
106:
107: SECURITY CONSIDERATIONS
108:
109: If you are using PCRE in a non-UTF application that permits users to
110: supply arbitrary patterns for compilation, you should be aware of a
111: feature that allows users to turn on UTF support from within a pattern,
112: provided that PCRE was built with UTF support. For example, an 8-bit
113: pattern that begins with "(*UTF8)" or "(*UTF)" turns on UTF-8 mode,
114: which interprets patterns and subjects as strings of UTF-8 characters
115: instead of individual 8-bit characters. This causes both the pattern
116: and any data against which it is matched to be checked for UTF-8 valid-
117: ity. If the data string is very long, such a check might use suffi-
118: ciently many resources as to cause your application to lose perfor-
119: mance.
120:
121: One way of guarding against this possibility is to use the
122: pcre_fullinfo() function to check the compiled pattern's options for
123: UTF. Alternatively, from release 8.33, you can set the PCRE_NEVER_UTF
124: option at compile time. This causes an compile time error if a pattern
125: contains a UTF-setting sequence.
126:
127: If your application is one that supports UTF, be aware that validity
128: checking can take time. If the same data string is to be matched many
129: times, you can use the PCRE_NO_UTF[8|16|32]_CHECK option for the second
130: and subsequent matches to save redundant checks.
131:
132: Another way that performance can be hit is by running a pattern that
133: has a very large search tree against a string that will never match.
134: Nested unlimited repeats in a pattern are a common example. PCRE pro-
135: vides some protection against this: see the PCRE_EXTRA_MATCH_LIMIT fea-
136: ture in the pcreapi page.
1.1 misha 137:
138:
139: USER DOCUMENTATION
140:
1.5 misha 141: The user documentation for PCRE comprises a number of different sec-
142: tions. In the "man" format, each of these is a separate "man page". In
143: the HTML format, each is a separate page, linked from the index page.
1.7 ! moko 144: In the plain text format, the descriptions of the pcregrep and pcretest
! 145: programs are in files called pcregrep.txt and pcretest.txt, respec-
! 146: tively. The remaining sections, except for the pcredemo section (which
! 147: is a program listing), are concatenated in pcre.txt, for ease of
! 148: searching. The sections are as follows:
1.1 misha 149:
150: pcre this document
1.6 misha 151: pcre-config show PCRE installation configuration information
1.5 misha 152: pcre16 details of the 16-bit library
1.6 misha 153: pcre32 details of the 32-bit library
1.1 misha 154: pcreapi details of PCRE's native C API
1.6 misha 155: pcrebuild building PCRE
1.1 misha 156: pcrecallout details of the callout feature
157: pcrecompat discussion of Perl compatibility
1.5 misha 158: pcrecpp details of the C++ wrapper for the 8-bit library
1.4 misha 159: pcredemo a demonstration C program that uses PCRE
1.5 misha 160: pcregrep description of the pcregrep command (8-bit only)
161: pcrejit discussion of the just-in-time optimization support
162: pcrelimits details of size and other limits
1.1 misha 163: pcrematching discussion of the two matching algorithms
164: pcrepartial details of the partial matching facility
165: pcrepattern syntax and semantics of supported
166: regular expressions
167: pcreperform discussion of performance issues
1.5 misha 168: pcreposix the POSIX-compatible C API for the 8-bit library
1.1 misha 169: pcreprecompile details of saving and re-using precompiled patterns
1.4 misha 170: pcresample discussion of the pcredemo program
1.1 misha 171: pcrestack discussion of stack usage
1.4 misha 172: pcresyntax quick syntax reference
1.1 misha 173: pcretest description of the pcretest testing command
1.6 misha 174: pcreunicode discussion of Unicode and UTF-8/16/32 support
1.5 misha 175:
1.7 ! moko 176: In the "man" and HTML formats, there is also a short page for each C
! 177: library function, listing its arguments and results.
1.5 misha 178:
179:
180: AUTHOR
181:
182: Philip Hazel
183: University Computing Service
184: Cambridge CB2 3QH, England.
185:
186: Putting an actual email address here seems to have been a spam magnet,
187: so I've taken it away. If you want to email me, use my two initials,
188: followed by the two digits 10, at the domain cam.ac.uk.
189:
190:
191: REVISION
192:
1.7 ! moko 193: Last updated: 10 February 2015
! 194: Copyright (c) 1997-2015 University of Cambridge.
1.5 misha 195: ------------------------------------------------------------------------------
196:
197:
1.6 misha 198: PCRE(3) Library Functions Manual PCRE(3)
199:
1.5 misha 200:
201:
202: NAME
203: PCRE - Perl-compatible regular expressions
204:
205: #include <pcre.h>
206:
207:
208: PCRE 16-BIT API BASIC FUNCTIONS
209:
210: pcre16 *pcre16_compile(PCRE_SPTR16 pattern, int options,
211: const char **errptr, int *erroffset,
212: const unsigned char *tableptr);
213:
214: pcre16 *pcre16_compile2(PCRE_SPTR16 pattern, int options,
215: int *errorcodeptr,
216: const char **errptr, int *erroffset,
217: const unsigned char *tableptr);
218:
219: pcre16_extra *pcre16_study(const pcre16 *code, int options,
220: const char **errptr);
221:
222: void pcre16_free_study(pcre16_extra *extra);
223:
224: int pcre16_exec(const pcre16 *code, const pcre16_extra *extra,
225: PCRE_SPTR16 subject, int length, int startoffset,
226: int options, int *ovector, int ovecsize);
227:
228: int pcre16_dfa_exec(const pcre16 *code, const pcre16_extra *extra,
229: PCRE_SPTR16 subject, int length, int startoffset,
230: int options, int *ovector, int ovecsize,
231: int *workspace, int wscount);
232:
233:
234: PCRE 16-BIT API STRING EXTRACTION FUNCTIONS
235:
236: int pcre16_copy_named_substring(const pcre16 *code,
237: PCRE_SPTR16 subject, int *ovector,
238: int stringcount, PCRE_SPTR16 stringname,
239: PCRE_UCHAR16 *buffer, int buffersize);
240:
241: int pcre16_copy_substring(PCRE_SPTR16 subject, int *ovector,
242: int stringcount, int stringnumber, PCRE_UCHAR16 *buffer,
243: int buffersize);
244:
245: int pcre16_get_named_substring(const pcre16 *code,
246: PCRE_SPTR16 subject, int *ovector,
247: int stringcount, PCRE_SPTR16 stringname,
248: PCRE_SPTR16 *stringptr);
249:
250: int pcre16_get_stringnumber(const pcre16 *code,
251: PCRE_SPTR16 name);
252:
253: int pcre16_get_stringtable_entries(const pcre16 *code,
254: PCRE_SPTR16 name, PCRE_UCHAR16 **first, PCRE_UCHAR16 **last);
255:
256: int pcre16_get_substring(PCRE_SPTR16 subject, int *ovector,
257: int stringcount, int stringnumber,
258: PCRE_SPTR16 *stringptr);
259:
260: int pcre16_get_substring_list(PCRE_SPTR16 subject,
261: int *ovector, int stringcount, PCRE_SPTR16 **listptr);
1.1 misha 262:
1.5 misha 263: void pcre16_free_substring(PCRE_SPTR16 stringptr);
1.1 misha 264:
1.5 misha 265: void pcre16_free_substring_list(PCRE_SPTR16 *stringptr);
1.1 misha 266:
267:
1.5 misha 268: PCRE 16-BIT API AUXILIARY FUNCTIONS
269:
270: pcre16_jit_stack *pcre16_jit_stack_alloc(int startsize, int maxsize);
271:
272: void pcre16_jit_stack_free(pcre16_jit_stack *stack);
273:
274: void pcre16_assign_jit_stack(pcre16_extra *extra,
275: pcre16_jit_callback callback, void *data);
276:
277: const unsigned char *pcre16_maketables(void);
278:
279: int pcre16_fullinfo(const pcre16 *code, const pcre16_extra *extra,
280: int what, void *where);
281:
282: int pcre16_refcount(pcre16 *code, int adjust);
283:
284: int pcre16_config(int what, void *where);
285:
286: const char *pcre16_version(void);
287:
288: int pcre16_pattern_to_host_byte_order(pcre16 *code,
289: pcre16_extra *extra, const unsigned char *tables);
290:
291:
292: PCRE 16-BIT API INDIRECTED FUNCTIONS
293:
294: void *(*pcre16_malloc)(size_t);
295:
296: void (*pcre16_free)(void *);
297:
298: void *(*pcre16_stack_malloc)(size_t);
299:
300: void (*pcre16_stack_free)(void *);
301:
302: int (*pcre16_callout)(pcre16_callout_block *);
303:
304:
305: PCRE 16-BIT API 16-BIT-ONLY FUNCTION
306:
307: int pcre16_utf16_to_host_byte_order(PCRE_UCHAR16 *output,
308: PCRE_SPTR16 input, int length, int *byte_order,
309: int keep_boms);
310:
311:
312: THE PCRE 16-BIT LIBRARY
313:
314: Starting with release 8.30, it is possible to compile a PCRE library
315: that supports 16-bit character strings, including UTF-16 strings, as
316: well as or instead of the original 8-bit library. The majority of the
317: work to make this possible was done by Zoltan Herczeg. The two
318: libraries contain identical sets of functions, used in exactly the same
319: way. Only the names of the functions and the data types of their argu-
320: ments and results are different. To avoid over-complication and reduce
321: the documentation maintenance load, most of the PCRE documentation
322: describes the 8-bit library, with only occasional references to the
323: 16-bit library. This page describes what is different when you use the
324: 16-bit library.
325:
326: WARNING: A single application can be linked with both libraries, but
327: you must take care when processing any particular pattern to use func-
328: tions from just one library. For example, if you want to study a pat-
329: tern that was compiled with pcre16_compile(), you must do so with
330: pcre16_study(), not pcre_study(), and you must free the study data with
331: pcre16_free_study().
332:
333:
334: THE HEADER FILE
335:
336: There is only one header file, pcre.h. It contains prototypes for all
1.6 misha 337: the functions in all libraries, as well as definitions of flags, struc-
338: tures, error codes, etc.
1.5 misha 339:
340:
341: THE LIBRARY NAME
342:
343: In Unix-like systems, the 16-bit library is called libpcre16, and can
344: normally be accesss by adding -lpcre16 to the command for linking an
345: application that uses PCRE.
346:
347:
348: STRING TYPES
349:
350: In the 8-bit library, strings are passed to PCRE library functions as
351: vectors of bytes with the C type "char *". In the 16-bit library,
352: strings are passed as vectors of unsigned 16-bit quantities. The macro
353: PCRE_UCHAR16 specifies an appropriate data type, and PCRE_SPTR16 is
354: defined as "const PCRE_UCHAR16 *". In very many environments, "short
355: int" is a 16-bit data type. When PCRE is built, it defines PCRE_UCHAR16
1.6 misha 356: as "unsigned short int", but checks that it really is a 16-bit data
357: type. If it is not, the build fails with an error message telling the
358: maintainer to modify the definition appropriately.
1.5 misha 359:
360:
361: STRUCTURE TYPES
362:
363: The types of the opaque structures that are used for compiled 16-bit
364: patterns and JIT stacks are pcre16 and pcre16_jit_stack respectively.
365: The type of the user-accessible structure that is returned by
366: pcre16_study() is pcre16_extra, and the type of the structure that is
367: used for passing data to a callout function is pcre16_callout_block.
368: These structures contain the same fields, with the same names, as their
369: 8-bit counterparts. The only difference is that pointers to character
370: strings are 16-bit instead of 8-bit types.
371:
372:
373: 16-BIT FUNCTIONS
374:
375: For every function in the 8-bit library there is a corresponding func-
376: tion in the 16-bit library with a name that starts with pcre16_ instead
377: of pcre_. The prototypes are listed above. In addition, there is one
378: extra function, pcre16_utf16_to_host_byte_order(). This is a utility
379: function that converts a UTF-16 character string to host byte order if
380: necessary. The other 16-bit functions expect the strings they are
381: passed to be in host byte order.
382:
383: The input and output arguments of pcre16_utf16_to_host_byte_order() may
384: point to the same address, that is, conversion in place is supported.
385: The output buffer must be at least as long as the input.
386:
387: The length argument specifies the number of 16-bit data units in the
388: input string; a negative value specifies a zero-terminated string.
389:
390: If byte_order is NULL, it is assumed that the string starts off in host
391: byte order. This may be changed by byte-order marks (BOMs) anywhere in
392: the string (commonly as the first character).
393:
394: If byte_order is not NULL, a non-zero value of the integer to which it
395: points means that the input starts off in host byte order, otherwise
396: the opposite order is assumed. Again, BOMs in the string can change
397: this. The final byte order is passed back at the end of processing.
398:
399: If keep_boms is not zero, byte-order mark characters (0xfeff) are
400: copied into the output string. Otherwise they are discarded.
401:
402: The result of the function is the number of 16-bit units placed into
403: the output buffer, including the zero terminator if the string was
404: zero-terminated.
405:
406:
407: SUBJECT STRING OFFSETS
408:
1.6 misha 409: The lengths and starting offsets of subject strings must be specified
410: in 16-bit data units, and the offsets within subject strings that are
411: returned by the matching functions are in also 16-bit units rather than
412: bytes.
1.5 misha 413:
414:
415: NAMED SUBPATTERNS
416:
417: The name-to-number translation table that is maintained for named sub-
418: patterns uses 16-bit characters. The pcre16_get_stringtable_entries()
419: function returns the length of each entry in the table as the number of
420: 16-bit data units.
421:
422:
423: OPTION NAMES
424:
425: There are two new general option names, PCRE_UTF16 and
426: PCRE_NO_UTF16_CHECK, which correspond to PCRE_UTF8 and
427: PCRE_NO_UTF8_CHECK in the 8-bit library. In fact, these new options
1.6 misha 428: define the same bits in the options word. There is a discussion about
429: the validity of UTF-16 strings in the pcreunicode page.
1.5 misha 430:
1.6 misha 431: For the pcre16_config() function there is an option PCRE_CONFIG_UTF16
432: that returns 1 if UTF-16 support is configured, otherwise 0. If this
433: option is given to pcre_config() or pcre32_config(), or if the
434: PCRE_CONFIG_UTF8 or PCRE_CONFIG_UTF32 option is given to pcre16_con-
435: fig(), the result is the PCRE_ERROR_BADOPTION error.
1.1 misha 436:
437:
1.5 misha 438: CHARACTER CODES
1.1 misha 439:
1.5 misha 440: In 16-bit mode, when PCRE_UTF16 is not set, character values are
441: treated in the same way as in 8-bit, non UTF-8 mode, except, of course,
442: that they can range from 0 to 0xffff instead of 0 to 0xff. Character
443: types for characters less than 0xff can therefore be influenced by the
444: locale in the same way as before. Characters greater than 0xff have
445: only one case, and no "type" (such as letter or digit).
1.1 misha 446:
1.5 misha 447: In UTF-16 mode, the character code is Unicode, in the range 0 to
448: 0x10ffff, with the exception of values in the range 0xd800 to 0xdfff
449: because those are "surrogate" values that are used in pairs to encode
450: values greater than 0xffff.
1.1 misha 451:
1.5 misha 452: A UTF-16 string can indicate its endianness by special code knows as a
453: byte-order mark (BOM). The PCRE functions do not handle this, expecting
454: strings to be in host byte order. A utility function called
455: pcre16_utf16_to_host_byte_order() is provided to help with this (see
456: above).
1.1 misha 457:
458:
1.5 misha 459: ERROR NAMES
1.1 misha 460:
1.5 misha 461: The errors PCRE_ERROR_BADUTF16_OFFSET and PCRE_ERROR_SHORTUTF16 corre-
462: spond to their 8-bit counterparts. The error PCRE_ERROR_BADMODE is
463: given when a compiled pattern is passed to a function that processes
464: patterns in the other mode, for example, if a pattern compiled with
465: pcre_compile() is passed to pcre16_exec().
1.1 misha 466:
1.5 misha 467: There are new error codes whose names begin with PCRE_UTF16_ERR for
468: invalid UTF-16 strings, corresponding to the PCRE_UTF8_ERR codes for
469: UTF-8 strings that are described in the section entitled "Reason codes
470: for invalid UTF-8 strings" in the main pcreapi page. The UTF-16 errors
471: are:
1.1 misha 472:
1.5 misha 473: PCRE_UTF16_ERR1 Missing low surrogate at end of string
474: PCRE_UTF16_ERR2 Invalid low surrogate follows high surrogate
475: PCRE_UTF16_ERR3 Isolated low surrogate
1.6 misha 476: PCRE_UTF16_ERR4 Non-character
1.1 misha 477:
478:
1.5 misha 479: ERROR TEXTS
1.1 misha 480:
1.5 misha 481: If there is an error while compiling a pattern, the error text that is
482: passed back by pcre16_compile() or pcre16_compile2() is still an 8-bit
483: character string, zero-terminated.
1.1 misha 484:
485:
1.5 misha 486: CALLOUTS
1.1 misha 487:
1.5 misha 488: The subject and mark fields in the callout block that is passed to a
489: callout function point to 16-bit vectors.
1.1 misha 490:
491:
1.5 misha 492: TESTING
1.1 misha 493:
1.5 misha 494: The pcretest program continues to operate with 8-bit input and output
495: files, but it can be used for testing the 16-bit library. If it is run
496: with the command line option -16, patterns and subject strings are con-
497: verted from 8-bit to 16-bit before being passed to PCRE, and the 16-bit
498: library functions are used instead of the 8-bit ones. Returned 16-bit
1.6 misha 499: strings are converted to 8-bit for output. If both the 8-bit and the
500: 32-bit libraries were not compiled, pcretest defaults to 16-bit and the
501: -16 option is ignored.
502:
503: When PCRE is being built, the RunTest script that is called by "make
504: check" uses the pcretest -C option to discover which of the 8-bit,
505: 16-bit and 32-bit libraries has been built, and runs the tests appro-
506: priately.
1.4 misha 507:
508:
1.5 misha 509: NOT SUPPORTED IN 16-BIT MODE
1.4 misha 510:
1.5 misha 511: Not all the features of the 8-bit library are available with the 16-bit
512: library. The C++ and POSIX wrapper functions support only the 8-bit
513: library, and the pcregrep program is at present 8-bit only.
1.1 misha 514:
515:
516: AUTHOR
517:
518: Philip Hazel
519: University Computing Service
520: Cambridge CB2 3QH, England.
521:
522:
523: REVISION
524:
1.6 misha 525: Last updated: 12 May 2013
526: Copyright (c) 1997-2013 University of Cambridge.
527: ------------------------------------------------------------------------------
528:
529:
530: PCRE(3) Library Functions Manual PCRE(3)
531:
532:
533:
534: NAME
535: PCRE - Perl-compatible regular expressions
536:
537: #include <pcre.h>
538:
539:
540: PCRE 32-BIT API BASIC FUNCTIONS
541:
542: pcre32 *pcre32_compile(PCRE_SPTR32 pattern, int options,
543: const char **errptr, int *erroffset,
544: const unsigned char *tableptr);
545:
546: pcre32 *pcre32_compile2(PCRE_SPTR32 pattern, int options,
547: int *errorcodeptr,
548: const unsigned char *tableptr);
549:
550: pcre32_extra *pcre32_study(const pcre32 *code, int options,
551: const char **errptr);
552:
553: void pcre32_free_study(pcre32_extra *extra);
554:
555: int pcre32_exec(const pcre32 *code, const pcre32_extra *extra,
556: PCRE_SPTR32 subject, int length, int startoffset,
557: int options, int *ovector, int ovecsize);
558:
559: int pcre32_dfa_exec(const pcre32 *code, const pcre32_extra *extra,
560: PCRE_SPTR32 subject, int length, int startoffset,
561: int options, int *ovector, int ovecsize,
562: int *workspace, int wscount);
563:
564:
565: PCRE 32-BIT API STRING EXTRACTION FUNCTIONS
566:
567: int pcre32_copy_named_substring(const pcre32 *code,
568: PCRE_SPTR32 subject, int *ovector,
569: int stringcount, PCRE_SPTR32 stringname,
570: PCRE_UCHAR32 *buffer, int buffersize);
571:
572: int pcre32_copy_substring(PCRE_SPTR32 subject, int *ovector,
573: int stringcount, int stringnumber, PCRE_UCHAR32 *buffer,
574: int buffersize);
575:
576: int pcre32_get_named_substring(const pcre32 *code,
577: PCRE_SPTR32 subject, int *ovector,
578: int stringcount, PCRE_SPTR32 stringname,
579: PCRE_SPTR32 *stringptr);
580:
581: int pcre32_get_stringnumber(const pcre32 *code,
582: PCRE_SPTR32 name);
583:
584: int pcre32_get_stringtable_entries(const pcre32 *code,
585: PCRE_SPTR32 name, PCRE_UCHAR32 **first, PCRE_UCHAR32 **last);
586:
587: int pcre32_get_substring(PCRE_SPTR32 subject, int *ovector,
588: int stringcount, int stringnumber,
589: PCRE_SPTR32 *stringptr);
590:
591: int pcre32_get_substring_list(PCRE_SPTR32 subject,
592: int *ovector, int stringcount, PCRE_SPTR32 **listptr);
593:
594: void pcre32_free_substring(PCRE_SPTR32 stringptr);
595:
596: void pcre32_free_substring_list(PCRE_SPTR32 *stringptr);
597:
598:
599: PCRE 32-BIT API AUXILIARY FUNCTIONS
600:
601: pcre32_jit_stack *pcre32_jit_stack_alloc(int startsize, int maxsize);
602:
603: void pcre32_jit_stack_free(pcre32_jit_stack *stack);
604:
605: void pcre32_assign_jit_stack(pcre32_extra *extra,
606: pcre32_jit_callback callback, void *data);
607:
608: const unsigned char *pcre32_maketables(void);
609:
610: int pcre32_fullinfo(const pcre32 *code, const pcre32_extra *extra,
611: int what, void *where);
612:
613: int pcre32_refcount(pcre32 *code, int adjust);
614:
615: int pcre32_config(int what, void *where);
616:
617: const char *pcre32_version(void);
618:
619: int pcre32_pattern_to_host_byte_order(pcre32 *code,
620: pcre32_extra *extra, const unsigned char *tables);
621:
622:
623: PCRE 32-BIT API INDIRECTED FUNCTIONS
624:
625: void *(*pcre32_malloc)(size_t);
626:
627: void (*pcre32_free)(void *);
628:
629: void *(*pcre32_stack_malloc)(size_t);
630:
631: void (*pcre32_stack_free)(void *);
632:
633: int (*pcre32_callout)(pcre32_callout_block *);
634:
635:
636: PCRE 32-BIT API 32-BIT-ONLY FUNCTION
637:
638: int pcre32_utf32_to_host_byte_order(PCRE_UCHAR32 *output,
639: PCRE_SPTR32 input, int length, int *byte_order,
640: int keep_boms);
641:
642:
643: THE PCRE 32-BIT LIBRARY
644:
645: Starting with release 8.32, it is possible to compile a PCRE library
646: that supports 32-bit character strings, including UTF-32 strings, as
647: well as or instead of the original 8-bit library. This work was done by
648: Christian Persch, based on the work done by Zoltan Herczeg for the
649: 16-bit library. All three libraries contain identical sets of func-
650: tions, used in exactly the same way. Only the names of the functions
651: and the data types of their arguments and results are different. To
652: avoid over-complication and reduce the documentation maintenance load,
653: most of the PCRE documentation describes the 8-bit library, with only
654: occasional references to the 16-bit and 32-bit libraries. This page
655: describes what is different when you use the 32-bit library.
656:
657: WARNING: A single application can be linked with all or any of the
658: three libraries, but you must take care when processing any particular
659: pattern to use functions from just one library. For example, if you
660: want to study a pattern that was compiled with pcre32_compile(), you
661: must do so with pcre32_study(), not pcre_study(), and you must free the
662: study data with pcre32_free_study().
663:
664:
665: THE HEADER FILE
666:
667: There is only one header file, pcre.h. It contains prototypes for all
668: the functions in all libraries, as well as definitions of flags, struc-
669: tures, error codes, etc.
670:
671:
672: THE LIBRARY NAME
673:
674: In Unix-like systems, the 32-bit library is called libpcre32, and can
675: normally be accesss by adding -lpcre32 to the command for linking an
676: application that uses PCRE.
677:
678:
679: STRING TYPES
680:
681: In the 8-bit library, strings are passed to PCRE library functions as
682: vectors of bytes with the C type "char *". In the 32-bit library,
683: strings are passed as vectors of unsigned 32-bit quantities. The macro
684: PCRE_UCHAR32 specifies an appropriate data type, and PCRE_SPTR32 is
685: defined as "const PCRE_UCHAR32 *". In very many environments, "unsigned
686: int" is a 32-bit data type. When PCRE is built, it defines PCRE_UCHAR32
687: as "unsigned int", but checks that it really is a 32-bit data type. If
688: it is not, the build fails with an error message telling the maintainer
689: to modify the definition appropriately.
690:
691:
692: STRUCTURE TYPES
693:
694: The types of the opaque structures that are used for compiled 32-bit
695: patterns and JIT stacks are pcre32 and pcre32_jit_stack respectively.
696: The type of the user-accessible structure that is returned by
697: pcre32_study() is pcre32_extra, and the type of the structure that is
698: used for passing data to a callout function is pcre32_callout_block.
699: These structures contain the same fields, with the same names, as their
700: 8-bit counterparts. The only difference is that pointers to character
701: strings are 32-bit instead of 8-bit types.
702:
703:
704: 32-BIT FUNCTIONS
705:
706: For every function in the 8-bit library there is a corresponding func-
707: tion in the 32-bit library with a name that starts with pcre32_ instead
708: of pcre_. The prototypes are listed above. In addition, there is one
709: extra function, pcre32_utf32_to_host_byte_order(). This is a utility
710: function that converts a UTF-32 character string to host byte order if
711: necessary. The other 32-bit functions expect the strings they are
712: passed to be in host byte order.
713:
714: The input and output arguments of pcre32_utf32_to_host_byte_order() may
715: point to the same address, that is, conversion in place is supported.
716: The output buffer must be at least as long as the input.
717:
718: The length argument specifies the number of 32-bit data units in the
719: input string; a negative value specifies a zero-terminated string.
720:
721: If byte_order is NULL, it is assumed that the string starts off in host
722: byte order. This may be changed by byte-order marks (BOMs) anywhere in
723: the string (commonly as the first character).
724:
725: If byte_order is not NULL, a non-zero value of the integer to which it
726: points means that the input starts off in host byte order, otherwise
727: the opposite order is assumed. Again, BOMs in the string can change
728: this. The final byte order is passed back at the end of processing.
729:
730: If keep_boms is not zero, byte-order mark characters (0xfeff) are
731: copied into the output string. Otherwise they are discarded.
732:
733: The result of the function is the number of 32-bit units placed into
734: the output buffer, including the zero terminator if the string was
735: zero-terminated.
736:
737:
738: SUBJECT STRING OFFSETS
739:
740: The lengths and starting offsets of subject strings must be specified
741: in 32-bit data units, and the offsets within subject strings that are
742: returned by the matching functions are in also 32-bit units rather than
743: bytes.
744:
745:
746: NAMED SUBPATTERNS
747:
748: The name-to-number translation table that is maintained for named sub-
749: patterns uses 32-bit characters. The pcre32_get_stringtable_entries()
750: function returns the length of each entry in the table as the number of
751: 32-bit data units.
752:
753:
754: OPTION NAMES
755:
756: There are two new general option names, PCRE_UTF32 and
757: PCRE_NO_UTF32_CHECK, which correspond to PCRE_UTF8 and
758: PCRE_NO_UTF8_CHECK in the 8-bit library. In fact, these new options
759: define the same bits in the options word. There is a discussion about
760: the validity of UTF-32 strings in the pcreunicode page.
761:
762: For the pcre32_config() function there is an option PCRE_CONFIG_UTF32
763: that returns 1 if UTF-32 support is configured, otherwise 0. If this
764: option is given to pcre_config() or pcre16_config(), or if the
765: PCRE_CONFIG_UTF8 or PCRE_CONFIG_UTF16 option is given to pcre32_con-
766: fig(), the result is the PCRE_ERROR_BADOPTION error.
767:
768:
769: CHARACTER CODES
770:
771: In 32-bit mode, when PCRE_UTF32 is not set, character values are
772: treated in the same way as in 8-bit, non UTF-8 mode, except, of course,
773: that they can range from 0 to 0x7fffffff instead of 0 to 0xff. Charac-
774: ter types for characters less than 0xff can therefore be influenced by
775: the locale in the same way as before. Characters greater than 0xff
776: have only one case, and no "type" (such as letter or digit).
777:
778: In UTF-32 mode, the character code is Unicode, in the range 0 to
779: 0x10ffff, with the exception of values in the range 0xd800 to 0xdfff
780: because those are "surrogate" values that are ill-formed in UTF-32.
781:
782: A UTF-32 string can indicate its endianness by special code knows as a
783: byte-order mark (BOM). The PCRE functions do not handle this, expecting
784: strings to be in host byte order. A utility function called
785: pcre32_utf32_to_host_byte_order() is provided to help with this (see
786: above).
787:
788:
789: ERROR NAMES
790:
791: The error PCRE_ERROR_BADUTF32 corresponds to its 8-bit counterpart.
792: The error PCRE_ERROR_BADMODE is given when a compiled pattern is passed
793: to a function that processes patterns in the other mode, for example,
794: if a pattern compiled with pcre_compile() is passed to pcre32_exec().
795:
796: There are new error codes whose names begin with PCRE_UTF32_ERR for
797: invalid UTF-32 strings, corresponding to the PCRE_UTF8_ERR codes for
798: UTF-8 strings that are described in the section entitled "Reason codes
799: for invalid UTF-8 strings" in the main pcreapi page. The UTF-32 errors
800: are:
801:
802: PCRE_UTF32_ERR1 Surrogate character (range from 0xd800 to 0xdfff)
803: PCRE_UTF32_ERR2 Non-character
804: PCRE_UTF32_ERR3 Character > 0x10ffff
805:
806:
807: ERROR TEXTS
808:
809: If there is an error while compiling a pattern, the error text that is
810: passed back by pcre32_compile() or pcre32_compile2() is still an 8-bit
811: character string, zero-terminated.
812:
813:
814: CALLOUTS
815:
816: The subject and mark fields in the callout block that is passed to a
817: callout function point to 32-bit vectors.
818:
819:
820: TESTING
821:
822: The pcretest program continues to operate with 8-bit input and output
823: files, but it can be used for testing the 32-bit library. If it is run
824: with the command line option -32, patterns and subject strings are con-
825: verted from 8-bit to 32-bit before being passed to PCRE, and the 32-bit
826: library functions are used instead of the 8-bit ones. Returned 32-bit
827: strings are converted to 8-bit for output. If both the 8-bit and the
828: 16-bit libraries were not compiled, pcretest defaults to 32-bit and the
829: -32 option is ignored.
830:
831: When PCRE is being built, the RunTest script that is called by "make
832: check" uses the pcretest -C option to discover which of the 8-bit,
833: 16-bit and 32-bit libraries has been built, and runs the tests appro-
834: priately.
835:
836:
837: NOT SUPPORTED IN 32-BIT MODE
838:
839: Not all the features of the 8-bit library are available with the 32-bit
840: library. The C++ and POSIX wrapper functions support only the 8-bit
841: library, and the pcregrep program is at present 8-bit only.
842:
843:
844: AUTHOR
845:
846: Philip Hazel
847: University Computing Service
848: Cambridge CB2 3QH, England.
849:
850:
851: REVISION
852:
853: Last updated: 12 May 2013
854: Copyright (c) 1997-2013 University of Cambridge.
1.1 misha 855: ------------------------------------------------------------------------------
856:
857:
1.6 misha 858: PCREBUILD(3) Library Functions Manual PCREBUILD(3)
859:
1.1 misha 860:
861:
862: NAME
863: PCRE - Perl-compatible regular expressions
864:
1.6 misha 865: BUILDING PCRE
866:
867: PCRE is distributed with a configure script that can be used to build
868: the library in Unix-like environments using the applications known as
869: Autotools. Also in the distribution are files to support building
870: using CMake instead of configure. The text file README contains general
871: information about building with Autotools (some of which is repeated
872: below), and also has some comments about building on various operating
873: systems. There is a lot more information about building PCRE without
874: using Autotools (including information about using CMake and building
875: "by hand") in the text file called NON-AUTOTOOLS-BUILD. You should
876: consult this file as well as the README file if you are building in a
877: non-Unix-like environment.
878:
1.1 misha 879:
880: PCRE BUILD-TIME OPTIONS
881:
1.6 misha 882: The rest of this document describes the optional features of PCRE that
883: can be selected when the library is compiled. It assumes use of the
884: configure script, where the optional features are selected or dese-
885: lected by providing options to configure before running the make com-
886: mand. However, the same options can be selected in both Unix-like and
887: non-Unix-like environments using the GUI facility of cmake-gui if you
888: are using CMake instead of configure to build PCRE.
889:
890: If you are not using Autotools or CMake, option selection can be done
891: by editing the config.h file, or by passing parameter settings to the
892: compiler, as described in NON-AUTOTOOLS-BUILD.
1.1 misha 893:
894: The complete list of options for configure (which includes the standard
1.6 misha 895: ones such as the selection of the installation directory) can be
1.1 misha 896: obtained by running
897:
898: ./configure --help
899:
1.6 misha 900: The following sections include descriptions of options whose names
1.1 misha 901: begin with --enable or --disable. These settings specify changes to the
1.6 misha 902: defaults for the configure command. Because of the way that configure
903: works, --enable and --disable always come in pairs, so the complemen-
904: tary option always exists as well, but as it specifies the default, it
1.1 misha 905: is not described.
906:
907:
1.6 misha 908: BUILDING 8-BIT, 16-BIT AND 32-BIT LIBRARIES
1.5 misha 909:
1.6 misha 910: By default, a library called libpcre is built, containing functions
911: that take string arguments contained in vectors of bytes, either as
912: single-byte characters, or interpreted as UTF-8 strings. You can also
913: build a separate library, called libpcre16, in which strings are con-
914: tained in vectors of 16-bit data units and interpreted either as sin-
1.5 misha 915: gle-unit characters or UTF-16 strings, by adding
916:
917: --enable-pcre16
918:
1.6 misha 919: to the configure command. You can also build yet another separate
920: library, called libpcre32, in which strings are contained in vectors of
921: 32-bit data units and interpreted either as single-unit characters or
922: UTF-32 strings, by adding
923:
924: --enable-pcre32
925:
1.5 misha 926: to the configure command. If you do not want the 8-bit library, add
927:
928: --disable-pcre8
929:
1.6 misha 930: as well. At least one of the three libraries must be built. Note that
931: the C++ and POSIX wrappers are for the 8-bit library only, and that
932: pcregrep is an 8-bit program. None of these are built if you select
933: only the 16-bit or 32-bit libraries.
1.5 misha 934:
935:
936: BUILDING SHARED AND STATIC LIBRARIES
937:
1.6 misha 938: The Autotools PCRE building process uses libtool to build both shared
939: and static libraries by default. You can suppress one of these by
940: adding one of
1.5 misha 941:
942: --disable-shared
943: --disable-static
944:
945: to the configure command, as required.
946:
947:
1.1 misha 948: C++ SUPPORT
949:
1.5 misha 950: By default, if the 8-bit library is being built, the configure script
951: will search for a C++ compiler and C++ header files. If it finds them,
952: it automatically builds the C++ wrapper library (which supports only
953: 8-bit strings). You can disable this by adding
1.1 misha 954:
955: --disable-cpp
956:
957: to the configure command.
958:
959:
1.6 misha 960: UTF-8, UTF-16 AND UTF-32 SUPPORT
1.1 misha 961:
1.5 misha 962: To build PCRE with support for UTF Unicode character strings, add
1.1 misha 963:
1.5 misha 964: --enable-utf
1.1 misha 965:
1.6 misha 966: to the configure command. This setting applies to all three libraries,
967: adding support for UTF-8 to the 8-bit library, support for UTF-16 to
968: the 16-bit library, and support for UTF-32 to the to the 32-bit
969: library. There are no separate options for enabling UTF-8, UTF-16 and
970: UTF-32 independently because that would allow ridiculous settings such
971: as requesting UTF-16 support while building only the 8-bit library. It
972: is not possible to build one library with UTF support and another with-
973: out in the same configuration. (For backwards compatibility, --enable-
974: utf8 is a synonym of --enable-utf.)
975:
976: Of itself, this setting does not make PCRE treat strings as UTF-8,
977: UTF-16 or UTF-32. As well as compiling PCRE with this option, you also
978: have have to set the PCRE_UTF8, PCRE_UTF16 or PCRE_UTF32 option (as
979: appropriate) when you call one of the pattern compiling functions.
980:
981: If you set --enable-utf when compiling in an EBCDIC environment, PCRE
982: expects its input to be either ASCII or UTF-8 (depending on the run-
983: time option). It is not possible to support both EBCDIC and UTF-8 codes
984: in the same version of the library. Consequently, --enable-utf and
1.3 misha 985: --enable-ebcdic are mutually exclusive.
986:
1.1 misha 987:
988: UNICODE CHARACTER PROPERTY SUPPORT
989:
1.6 misha 990: UTF support allows the libraries to process character codepoints up to
991: 0x10ffff in the strings that they handle. On its own, however, it does
1.5 misha 992: not provide any facilities for accessing the properties of such charac-
993: ters. If you want to be able to use the pattern escapes \P, \p, and \X,
994: which refer to Unicode character properties, you must add
1.1 misha 995:
996: --enable-unicode-properties
997:
1.6 misha 998: to the configure command. This implies UTF support, even if you have
1.1 misha 999: not explicitly requested it.
1000:
1.6 misha 1001: Including Unicode property support adds around 30K of tables to the
1002: PCRE library. Only the general category properties such as Lu and Nd
1.1 misha 1003: are supported. Details are given in the pcrepattern documentation.
1004:
1005:
1.5 misha 1006: JUST-IN-TIME COMPILER SUPPORT
1007:
1008: Just-in-time compiler support is included in the build by specifying
1009:
1010: --enable-jit
1011:
1.6 misha 1012: This support is available only for certain hardware architectures. If
1013: this option is set for an unsupported architecture, a compile time
1014: error occurs. See the pcrejit documentation for a discussion of JIT
1.5 misha 1015: usage. When JIT support is enabled, pcregrep automatically makes use of
1016: it, unless you add
1017:
1018: --disable-pcregrep-jit
1019:
1020: to the "configure" command.
1021:
1022:
1.1 misha 1023: CODE VALUE OF NEWLINE
1024:
1.6 misha 1025: By default, PCRE interprets the linefeed (LF) character as indicating
1026: the end of a line. This is the normal newline character on Unix-like
1027: systems. You can compile PCRE to use carriage return (CR) instead, by
1.3 misha 1028: adding
1.1 misha 1029:
1030: --enable-newline-is-cr
1031:
1.6 misha 1032: to the configure command. There is also a --enable-newline-is-lf
1.1 misha 1033: option, which explicitly specifies linefeed as the newline character.
1034:
1035: Alternatively, you can specify that line endings are to be indicated by
1036: the two character sequence CRLF. If you want this, add
1037:
1038: --enable-newline-is-crlf
1039:
1040: to the configure command. There is a fourth option, specified by
1041:
1042: --enable-newline-is-anycrlf
1043:
1.6 misha 1044: which causes PCRE to recognize any of the three sequences CR, LF, or
1.1 misha 1045: CRLF as indicating a line ending. Finally, a fifth option, specified by
1046:
1047: --enable-newline-is-any
1048:
1049: causes PCRE to recognize any Unicode newline sequence.
1050:
1.6 misha 1051: Whatever line ending convention is selected when PCRE is built can be
1052: overridden when the library functions are called. At build time it is
1.1 misha 1053: conventional to use the standard for your operating system.
1054:
1055:
1056: WHAT \R MATCHES
1057:
1.6 misha 1058: By default, the sequence \R in a pattern matches any Unicode newline
1059: sequence, whatever has been selected as the line ending sequence. If
1.1 misha 1060: you specify
1061:
1062: --enable-bsr-anycrlf
1063:
1.6 misha 1064: the default is changed so that \R matches only CR, LF, or CRLF. What-
1065: ever is selected when PCRE is built can be overridden when the library
1.1 misha 1066: functions are called.
1067:
1068:
1069: POSIX MALLOC USAGE
1070:
1.6 misha 1071: When the 8-bit library is called through the POSIX interface (see the
1072: pcreposix documentation), additional working storage is required for
1073: holding the pointers to capturing substrings, because PCRE requires
1.5 misha 1074: three integers per substring, whereas the POSIX interface provides only
1.6 misha 1075: two. If the number of expected substrings is small, the wrapper func-
1076: tion uses space on the stack, because this is faster than using mal-
1077: loc() for each call. The default threshold above which the stack is no
1.5 misha 1078: longer used is 10; it can be changed by adding a setting such as
1.1 misha 1079:
1080: --with-posix-malloc-threshold=20
1081:
1082: to the configure command.
1083:
1084:
1085: HANDLING VERY LARGE PATTERNS
1086:
1.6 misha 1087: Within a compiled pattern, offset values are used to point from one
1088: part to another (for example, from an opening parenthesis to an alter-
1089: nation metacharacter). By default, in the 8-bit and 16-bit libraries,
1090: two-byte values are used for these offsets, leading to a maximum size
1091: for a compiled pattern of around 64K. This is sufficient to handle all
1092: but the most gigantic patterns. Nevertheless, some people do want to
1093: process truly enormous patterns, so it is possible to compile PCRE to
1094: use three-byte or four-byte offsets by adding a setting such as
1.1 misha 1095:
1096: --with-link-size=3
1097:
1.6 misha 1098: to the configure command. The value given must be 2, 3, or 4. For the
1099: 16-bit library, a value of 3 is rounded up to 4. In these libraries,
1100: using longer offsets slows down the operation of PCRE because it has to
1101: load additional data when handling them. For the 32-bit library the
1102: value is always 4 and cannot be overridden; the value of --with-link-
1103: size is ignored.
1.1 misha 1104:
1105:
1106: AVOIDING EXCESSIVE STACK USAGE
1107:
1108: When matching with the pcre_exec() function, PCRE implements backtrack-
1.6 misha 1109: ing by making recursive calls to an internal function called match().
1110: In environments where the size of the stack is limited, this can se-
1111: verely limit PCRE's operation. (The Unix environment does not usually
1.1 misha 1112: suffer from this problem, but it may sometimes be necessary to increase
1.6 misha 1113: the maximum stack size. There is a discussion in the pcrestack docu-
1114: mentation.) An alternative approach to recursion that uses memory from
1115: the heap to remember data, instead of using recursive function calls,
1116: has been implemented to work round the problem of limited stack size.
1.1 misha 1117: If you want to build a version of PCRE that works this way, add
1118:
1119: --disable-stack-for-recursion
1120:
1.6 misha 1121: to the configure command. With this configuration, PCRE will use the
1122: pcre_stack_malloc and pcre_stack_free variables to call memory manage-
1123: ment functions. By default these point to malloc() and free(), but you
1.4 misha 1124: can replace the pointers so that your own functions are used instead.
1125:
1.6 misha 1126: Separate functions are provided rather than using pcre_malloc and
1127: pcre_free because the usage is very predictable: the block sizes
1128: requested are always the same, and the blocks are always freed in
1129: reverse order. A calling program might be able to implement optimized
1130: functions that perform better than malloc() and free(). PCRE runs
1.1 misha 1131: noticeably more slowly when built in this way. This option affects only
1.4 misha 1132: the pcre_exec() function; it is not relevant for pcre_dfa_exec().
1.1 misha 1133:
1134:
1135: LIMITING PCRE RESOURCE USAGE
1136:
1.6 misha 1137: Internally, PCRE has a function called match(), which it calls repeat-
1138: edly (sometimes recursively) when matching a pattern with the
1139: pcre_exec() function. By controlling the maximum number of times this
1140: function may be called during a single matching operation, a limit can
1141: be placed on the resources used by a single call to pcre_exec(). The
1142: limit can be changed at run time, as described in the pcreapi documen-
1143: tation. The default is 10 million, but this can be changed by adding a
1.1 misha 1144: setting such as
1145:
1146: --with-match-limit=500000
1147:
1.6 misha 1148: to the configure command. This setting has no effect on the
1.1 misha 1149: pcre_dfa_exec() matching function.
1150:
1.6 misha 1151: In some environments it is desirable to limit the depth of recursive
1.1 misha 1152: calls of match() more strictly than the total number of calls, in order
1.6 misha 1153: to restrict the maximum amount of stack (or heap, if --disable-stack-
1.1 misha 1154: for-recursion is specified) that is used. A second limit controls this;
1.6 misha 1155: it defaults to the value that is set for --with-match-limit, which
1156: imposes no additional constraints. However, you can set a lower limit
1.1 misha 1157: by adding, for example,
1158:
1159: --with-match-limit-recursion=10000
1160:
1.6 misha 1161: to the configure command. This value can also be overridden at run
1.1 misha 1162: time.
1163:
1164:
1165: CREATING CHARACTER TABLES AT BUILD TIME
1166:
1.6 misha 1167: PCRE uses fixed tables for processing characters whose code values are
1168: less than 256. By default, PCRE is built with a set of tables that are
1169: distributed in the file pcre_chartables.c.dist. These tables are for
1.1 misha 1170: ASCII codes only. If you add
1171:
1172: --enable-rebuild-chartables
1173:
1.6 misha 1174: to the configure command, the distributed tables are no longer used.
1175: Instead, a program called dftables is compiled and run. This outputs
1.1 misha 1176: the source for new set of tables, created in the default locale of your
1.6 misha 1177: C run-time system. (This method of replacing the tables does not work
1178: if you are cross compiling, because dftables is run on the local host.
1179: If you need to create alternative tables when cross compiling, you will
1.1 misha 1180: have to do so "by hand".)
1181:
1182:
1183: USING EBCDIC CODE
1184:
1.6 misha 1185: PCRE assumes by default that it will run in an environment where the
1186: character code is ASCII (or Unicode, which is a superset of ASCII).
1187: This is the case for most computer operating systems. PCRE can, how-
1.1 misha 1188: ever, be compiled to run in an EBCDIC environment by adding
1189:
1190: --enable-ebcdic
1191:
1192: to the configure command. This setting implies --enable-rebuild-charta-
1.6 misha 1193: bles. You should only use it if you know that you are in an EBCDIC
1194: environment (for example, an IBM mainframe operating system). The
1.5 misha 1195: --enable-ebcdic option is incompatible with --enable-utf.
1.1 misha 1196:
1.6 misha 1197: The EBCDIC character that corresponds to an ASCII LF is assumed to have
1198: the value 0x15 by default. However, in some EBCDIC environments, 0x25
1199: is used. In such an environment you should use
1200:
1201: --enable-ebcdic-nl25
1202:
1203: as well as, or instead of, --enable-ebcdic. The EBCDIC character for CR
1204: has the same value as in ASCII, namely, 0x0d. Whichever of 0x15 and
1205: 0x25 is not chosen as LF is made to correspond to the Unicode NEL char-
1206: acter (which, in Unicode, is 0x85).
1207:
1208: The options that select newline behaviour, such as --enable-newline-is-
1209: cr, and equivalent run-time options, refer to these character values in
1210: an EBCDIC environment.
1211:
1.1 misha 1212:
1213: PCREGREP OPTIONS FOR COMPRESSED FILE SUPPORT
1214:
1215: By default, pcregrep reads all files as plain text. You can build it so
1216: that it recognizes files whose names end in .gz or .bz2, and reads them
1217: with libz or libbz2, respectively, by adding one or both of
1218:
1219: --enable-pcregrep-libz
1220: --enable-pcregrep-libbz2
1221:
1222: to the configure command. These options naturally require that the rel-
1.5 misha 1223: evant libraries are installed on your system. Configuration will fail
1.1 misha 1224: if they are not.
1225:
1226:
1.5 misha 1227: PCREGREP BUFFER SIZE
1228:
1229: pcregrep uses an internal buffer to hold a "window" on the file it is
1230: scanning, in order to be able to output "before" and "after" lines when
1231: it finds a match. The size of the buffer is controlled by a parameter
1232: whose default value is 20K. The buffer itself is three times this size,
1233: but because of the way it is used for holding "before" lines, the long-
1234: est line that is guaranteed to be processable is the parameter size.
1235: You can change the default parameter value by adding, for example,
1236:
1237: --with-pcregrep-bufsize=50K
1238:
1239: to the configure command. The caller of pcregrep can, however, override
1240: this value by specifying a run-time option.
1241:
1242:
1.1 misha 1243: PCRETEST OPTION FOR LIBREADLINE SUPPORT
1244:
1245: If you add
1246:
1247: --enable-pcretest-libreadline
1248:
1249: to the configure command, pcretest is linked with the libreadline
1250: library, and when its input is from a terminal, it reads it using the
1251: readline() function. This provides line-editing and history facilities.
1.4 misha 1252: Note that libreadline is GPL-licensed, so if you distribute a binary of
1.1 misha 1253: pcretest linked in this way, there may be licensing issues.
1254:
1255: Setting this option causes the -lreadline option to be added to the
1256: pcretest build. In many operating environments with a sytem-installed
1257: libreadline this is sufficient. However, in some environments (e.g. if
1258: an unmodified distribution version of readline is in use), some extra
1259: configuration may be necessary. The INSTALL file for libreadline says
1260: this:
1261:
1262: "Readline uses the termcap functions, but does not link with the
1263: termcap or curses library itself, allowing applications which link
1264: with readline the to choose an appropriate library."
1265:
1266: If your environment has not been set up so that an appropriate library
1267: is automatically included, you may need to add something like
1268:
1269: LIBS="-ncurses"
1270:
1271: immediately before the configure command.
1272:
1273:
1.6 misha 1274: DEBUGGING WITH VALGRIND SUPPORT
1275:
1276: By adding the
1277:
1278: --enable-valgrind
1279:
1280: option to to the configure command, PCRE will use valgrind annotations
1281: to mark certain memory regions as unaddressable. This allows it to
1282: detect invalid memory accesses, and is mostly useful for debugging PCRE
1283: itself.
1284:
1285:
1286: CODE COVERAGE REPORTING
1287:
1288: If your C compiler is gcc, you can build a version of PCRE that can
1289: generate a code coverage report for its test suite. To enable this, you
1290: must install lcov version 1.6 or above. Then specify
1291:
1292: --enable-coverage
1293:
1294: to the configure command and build PCRE in the usual way.
1295:
1296: Note that using ccache (a caching C compiler) is incompatible with code
1297: coverage reporting. If you have configured ccache to run automatically
1298: on your system, you must set the environment variable
1299:
1300: CCACHE_DISABLE=1
1301:
1302: before running make to build PCRE, so that ccache is not used.
1303:
1304: When --enable-coverage is used, the following addition targets are
1305: added to the Makefile:
1306:
1307: make coverage
1308:
1309: This creates a fresh coverage report for the PCRE test suite. It is
1310: equivalent to running "make coverage-reset", "make coverage-baseline",
1311: "make check", and then "make coverage-report".
1312:
1313: make coverage-reset
1314:
1315: This zeroes the coverage counters, but does nothing else.
1316:
1317: make coverage-baseline
1318:
1319: This captures baseline coverage information.
1320:
1321: make coverage-report
1322:
1323: This creates the coverage report.
1324:
1325: make coverage-clean-report
1326:
1327: This removes the generated coverage report without cleaning the cover-
1328: age data itself.
1329:
1330: make coverage-clean-data
1331:
1332: This removes the captured coverage data without removing the coverage
1333: files created at compile time (*.gcno).
1334:
1335: make coverage-clean
1336:
1337: This cleans all coverage data including the generated coverage report.
1338: For more information about code coverage, see the gcov and lcov docu-
1339: mentation.
1340:
1341:
1.1 misha 1342: SEE ALSO
1343:
1.6 misha 1344: pcreapi(3), pcre16, pcre32, pcre_config(3).
1.1 misha 1345:
1346:
1347: AUTHOR
1348:
1349: Philip Hazel
1350: University Computing Service
1351: Cambridge CB2 3QH, England.
1352:
1353:
1354: REVISION
1355:
1.6 misha 1356: Last updated: 12 May 2013
1357: Copyright (c) 1997-2013 University of Cambridge.
1.1 misha 1358: ------------------------------------------------------------------------------
1359:
1360:
1.6 misha 1361: PCREMATCHING(3) Library Functions Manual PCREMATCHING(3)
1362:
1.1 misha 1363:
1364:
1365: NAME
1366: PCRE - Perl-compatible regular expressions
1367:
1368: PCRE MATCHING ALGORITHMS
1369:
1370: This document describes the two different algorithms that are available
1371: in PCRE for matching a compiled regular expression against a given sub-
1372: ject string. The "standard" algorithm is the one provided by the
1.6 misha 1373: pcre_exec(), pcre16_exec() and pcre32_exec() functions. These work in
1374: the same as as Perl's matching function, and provide a Perl-compatible
1375: matching operation. The just-in-time (JIT) optimization that is
1376: described in the pcrejit documentation is compatible with these func-
1377: tions.
1378:
1379: An alternative algorithm is provided by the pcre_dfa_exec(),
1380: pcre16_dfa_exec() and pcre32_dfa_exec() functions; they operate in a
1381: different way, and are not Perl-compatible. This alternative has advan-
1382: tages and disadvantages compared with the standard algorithm, and these
1383: are described below.
1.1 misha 1384:
1385: When there is only one possible way in which a given subject string can
1386: match a pattern, the two algorithms give the same answer. A difference
1387: arises, however, when there are multiple possibilities. For example, if
1388: the pattern
1389:
1390: ^<.*>
1391:
1392: is matched against the string
1393:
1394: <something> <something else> <something further>
1395:
1396: there are three possible answers. The standard algorithm finds only one
1397: of them, whereas the alternative algorithm finds all three.
1398:
1399:
1400: REGULAR EXPRESSIONS AS TREES
1401:
1402: The set of strings that are matched by a regular expression can be rep-
1403: resented as a tree structure. An unlimited repetition in the pattern
1404: makes the tree of infinite size, but it is still a tree. Matching the
1405: pattern to a given subject string (from a given starting point) can be
1406: thought of as a search of the tree. There are two ways to search a
1407: tree: depth-first and breadth-first, and these correspond to the two
1408: matching algorithms provided by PCRE.
1409:
1410:
1411: THE STANDARD MATCHING ALGORITHM
1412:
1413: In the terminology of Jeffrey Friedl's book "Mastering Regular Expres-
1414: sions", the standard algorithm is an "NFA algorithm". It conducts a
1415: depth-first search of the pattern tree. That is, it proceeds along a
1416: single path through the tree, checking that the subject matches what is
1417: required. When there is a mismatch, the algorithm tries any alterna-
1418: tives at the current point, and if they all fail, it backs up to the
1419: previous branch point in the tree, and tries the next alternative
1420: branch at that level. This often involves backing up (moving to the
1421: left) in the subject string as well. The order in which repetition
1422: branches are tried is controlled by the greedy or ungreedy nature of
1423: the quantifier.
1424:
1425: If a leaf node is reached, a matching string has been found, and at
1426: that point the algorithm stops. Thus, if there is more than one possi-
1427: ble match, this algorithm returns the first one that it finds. Whether
1428: this is the shortest, the longest, or some intermediate length depends
1429: on the way the greedy and ungreedy repetition quantifiers are specified
1430: in the pattern.
1431:
1432: Because it ends up with a single path through the tree, it is rela-
1433: tively straightforward for this algorithm to keep track of the sub-
1434: strings that are matched by portions of the pattern in parentheses.
1435: This provides support for capturing parentheses and back references.
1436:
1437:
1438: THE ALTERNATIVE MATCHING ALGORITHM
1439:
1440: This algorithm conducts a breadth-first search of the tree. Starting
1441: from the first matching point in the subject, it scans the subject
1442: string from left to right, once, character by character, and as it does
1443: this, it remembers all the paths through the tree that represent valid
1444: matches. In Friedl's terminology, this is a kind of "DFA algorithm",
1445: though it is not implemented as a traditional finite state machine (it
1446: keeps multiple states active simultaneously).
1447:
1.4 misha 1448: Although the general principle of this matching algorithm is that it
1449: scans the subject string only once, without backtracking, there is one
1450: exception: when a lookaround assertion is encountered, the characters
1451: following or preceding the current point have to be independently
1452: inspected.
1453:
1.1 misha 1454: The scan continues until either the end of the subject is reached, or
1455: there are no more unterminated paths. At this point, terminated paths
1456: represent the different matching possibilities (if there are none, the
1457: match has failed). Thus, if there is more than one possible match,
1458: this algorithm finds all of them, and in particular, it finds the long-
1.4 misha 1459: est. The matches are returned in decreasing order of length. There is
1460: an option to stop the algorithm after the first match (which is neces-
1461: sarily the shortest) is found.
1.1 misha 1462:
1463: Note that all the matches that are found start at the same point in the
1464: subject. If the pattern
1465:
1.4 misha 1466: cat(er(pillar)?)?
1.1 misha 1467:
1.4 misha 1468: is matched against the string "the caterpillar catchment", the result
1469: will be the three strings "caterpillar", "cater", and "cat" that start
1470: at the fifth character of the subject. The algorithm does not automati-
1471: cally move on to find matches that start at later positions.
1.1 misha 1472:
1.7 ! moko 1473: PCRE's "auto-possessification" optimization usually applies to charac-
! 1474: ter repeats at the end of a pattern (as well as internally). For exam-
! 1475: ple, the pattern "a\d+" is compiled as if it were "a\d++" because there
! 1476: is no point even considering the possibility of backtracking into the
! 1477: repeated digits. For DFA matching, this means that only one possible
! 1478: match is found. If you really do want multiple matches in such cases,
! 1479: either use an ungreedy repeat ("a\d+?") or set the PCRE_NO_AUTO_POSSESS
! 1480: option when compiling.
! 1481:
1.1 misha 1482: There are a number of features of PCRE regular expressions that are not
1483: supported by the alternative matching algorithm. They are as follows:
1484:
1.7 ! moko 1485: 1. Because the algorithm finds all possible matches, the greedy or
! 1486: ungreedy nature of repetition quantifiers is not relevant. Greedy and
1.1 misha 1487: ungreedy quantifiers are treated in exactly the same way. However, pos-
1.7 ! moko 1488: sessive quantifiers can make a difference when what follows could also
1.1 misha 1489: match what is quantified, for example in a pattern like this:
1490:
1491: ^a++\w!
1492:
1.7 ! moko 1493: This pattern matches "aaab!" but not "aaa!", which would be matched by
! 1494: a non-possessive quantifier. Similarly, if an atomic group is present,
! 1495: it is matched as if it were a standalone pattern at the current point,
! 1496: and the longest match is then "locked in" for the rest of the overall
1.1 misha 1497: pattern.
1498:
1499: 2. When dealing with multiple paths through the tree simultaneously, it
1.7 ! moko 1500: is not straightforward to keep track of captured substrings for the
! 1501: different matching possibilities, and PCRE's implementation of this
1.1 misha 1502: algorithm does not attempt to do this. This means that no captured sub-
1503: strings are available.
1504:
1.7 ! moko 1505: 3. Because no substrings are captured, back references within the pat-
1.1 misha 1506: tern are not supported, and cause errors if encountered.
1507:
1.7 ! moko 1508: 4. For the same reason, conditional expressions that use a backrefer-
! 1509: ence as the condition or test for a specific group recursion are not
1.1 misha 1510: supported.
1511:
1.7 ! moko 1512: 5. Because many paths through the tree may be active, the \K escape
1.1 misha 1513: sequence, which resets the start of the match when encountered (but may
1.7 ! moko 1514: be on some paths and not on others), is not supported. It causes an
1.1 misha 1515: error if encountered.
1516:
1.7 ! moko 1517: 6. Callouts are supported, but the value of the capture_top field is
1.1 misha 1518: always 1, and the value of the capture_last field is always -1.
1519:
1.7 ! moko 1520: 7. The \C escape sequence, which (in the standard algorithm) always
! 1521: matches a single data unit, even in UTF-8, UTF-16 or UTF-32 modes, is
! 1522: not supported in these modes, because the alternative algorithm moves
1.6 misha 1523: through the subject string one character (not data unit) at a time, for
1524: all active paths through the tree.
1.1 misha 1525:
1.7 ! moko 1526: 8. Except for (*FAIL), the backtracking control verbs such as (*PRUNE)
! 1527: are not supported. (*FAIL) is supported, and behaves like a failing
1.1 misha 1528: negative assertion.
1529:
1530:
1531: ADVANTAGES OF THE ALTERNATIVE ALGORITHM
1532:
1.7 ! moko 1533: Using the alternative matching algorithm provides the following advan-
1.1 misha 1534: tages:
1535:
1536: 1. All possible matches (at a single point in the subject) are automat-
1.7 ! moko 1537: ically found, and in particular, the longest match is found. To find
1.1 misha 1538: more than one match using the standard algorithm, you have to do kludgy
1539: things with callouts.
1540:
1.7 ! moko 1541: 2. Because the alternative algorithm scans the subject string just
1.5 misha 1542: once, and never needs to backtrack (except for lookbehinds), it is pos-
1.7 ! moko 1543: sible to pass very long subject strings to the matching function in
1.5 misha 1544: several pieces, checking for partial matching each time. Although it is
1.7 ! moko 1545: possible to do multi-segment matching using the standard algorithm by
! 1546: retaining partially matched substrings, it is more complicated. The
! 1547: pcrepartial documentation gives details of partial matching and dis-
1.5 misha 1548: cusses multi-segment matching.
1.1 misha 1549:
1550:
1551: DISADVANTAGES OF THE ALTERNATIVE ALGORITHM
1552:
1553: The alternative algorithm suffers from a number of disadvantages:
1554:
1.7 ! moko 1555: 1. It is substantially slower than the standard algorithm. This is
! 1556: partly because it has to search for all possible matches, but is also
1.1 misha 1557: because it is less susceptible to optimization.
1558:
1559: 2. Capturing parentheses and back references are not supported.
1560:
1561: 3. Although atomic groups are supported, their use does not provide the
1562: performance advantage that it does for the standard algorithm.
1563:
1564:
1565: AUTHOR
1566:
1567: Philip Hazel
1568: University Computing Service
1569: Cambridge CB2 3QH, England.
1570:
1571:
1572: REVISION
1573:
1.7 ! moko 1574: Last updated: 12 November 2013
1.5 misha 1575: Copyright (c) 1997-2012 University of Cambridge.
1.1 misha 1576: ------------------------------------------------------------------------------
1577:
1578:
1.6 misha 1579: PCREAPI(3) Library Functions Manual PCREAPI(3)
1580:
1.1 misha 1581:
1582:
1583: NAME
1584: PCRE - Perl-compatible regular expressions
1585:
1.5 misha 1586: #include <pcre.h>
1.1 misha 1587:
1588:
1.5 misha 1589: PCRE NATIVE API BASIC FUNCTIONS
1.1 misha 1590:
1591: pcre *pcre_compile(const char *pattern, int options,
1592: const char **errptr, int *erroffset,
1593: const unsigned char *tableptr);
1594:
1595: pcre *pcre_compile2(const char *pattern, int options,
1596: int *errorcodeptr,
1597: const char **errptr, int *erroffset,
1598: const unsigned char *tableptr);
1599:
1600: pcre_extra *pcre_study(const pcre *code, int options,
1601: const char **errptr);
1602:
1.5 misha 1603: void pcre_free_study(pcre_extra *extra);
1604:
1.1 misha 1605: int pcre_exec(const pcre *code, const pcre_extra *extra,
1606: const char *subject, int length, int startoffset,
1607: int options, int *ovector, int ovecsize);
1608:
1609: int pcre_dfa_exec(const pcre *code, const pcre_extra *extra,
1610: const char *subject, int length, int startoffset,
1611: int options, int *ovector, int ovecsize,
1612: int *workspace, int wscount);
1613:
1.5 misha 1614:
1615: PCRE NATIVE API STRING EXTRACTION FUNCTIONS
1616:
1.1 misha 1617: int pcre_copy_named_substring(const pcre *code,
1618: const char *subject, int *ovector,
1619: int stringcount, const char *stringname,
1620: char *buffer, int buffersize);
1621:
1622: int pcre_copy_substring(const char *subject, int *ovector,
1623: int stringcount, int stringnumber, char *buffer,
1624: int buffersize);
1625:
1626: int pcre_get_named_substring(const pcre *code,
1627: const char *subject, int *ovector,
1628: int stringcount, const char *stringname,
1629: const char **stringptr);
1630:
1631: int pcre_get_stringnumber(const pcre *code,
1632: const char *name);
1633:
1634: int pcre_get_stringtable_entries(const pcre *code,
1635: const char *name, char **first, char **last);
1636:
1637: int pcre_get_substring(const char *subject, int *ovector,
1638: int stringcount, int stringnumber,
1639: const char **stringptr);
1640:
1641: int pcre_get_substring_list(const char *subject,
1642: int *ovector, int stringcount, const char ***listptr);
1643:
1644: void pcre_free_substring(const char *stringptr);
1645:
1646: void pcre_free_substring_list(const char **stringptr);
1647:
1.5 misha 1648:
1649: PCRE NATIVE API AUXILIARY FUNCTIONS
1650:
1.6 misha 1651: int pcre_jit_exec(const pcre *code, const pcre_extra *extra,
1652: const char *subject, int length, int startoffset,
1653: int options, int *ovector, int ovecsize,
1654: pcre_jit_stack *jstack);
1655:
1.5 misha 1656: pcre_jit_stack *pcre_jit_stack_alloc(int startsize, int maxsize);
1657:
1658: void pcre_jit_stack_free(pcre_jit_stack *stack);
1659:
1660: void pcre_assign_jit_stack(pcre_extra *extra,
1661: pcre_jit_callback callback, void *data);
1662:
1.1 misha 1663: const unsigned char *pcre_maketables(void);
1664:
1665: int pcre_fullinfo(const pcre *code, const pcre_extra *extra,
1666: int what, void *where);
1667:
1668: int pcre_refcount(pcre *code, int adjust);
1669:
1670: int pcre_config(int what, void *where);
1671:
1.5 misha 1672: const char *pcre_version(void);
1673:
1674: int pcre_pattern_to_host_byte_order(pcre *code,
1675: pcre_extra *extra, const unsigned char *tables);
1676:
1677:
1678: PCRE NATIVE API INDIRECTED FUNCTIONS
1.1 misha 1679:
1680: void *(*pcre_malloc)(size_t);
1681:
1682: void (*pcre_free)(void *);
1683:
1684: void *(*pcre_stack_malloc)(size_t);
1685:
1686: void (*pcre_stack_free)(void *);
1687:
1688: int (*pcre_callout)(pcre_callout_block *);
1689:
1.7 ! moko 1690: int (*pcre_stack_guard)(void);
! 1691:
1.1 misha 1692:
1.6 misha 1693: PCRE 8-BIT, 16-BIT, AND 32-BIT LIBRARIES
1.5 misha 1694:
1.6 misha 1695: As well as support for 8-bit character strings, PCRE also supports
1696: 16-bit strings (from release 8.30) and 32-bit strings (from release
1697: 8.32), by means of two additional libraries. They can be built as well
1698: as, or instead of, the 8-bit library. To avoid too much complication,
1699: this document describes the 8-bit versions of the functions, with only
1700: occasional references to the 16-bit and 32-bit libraries.
1701:
1702: The 16-bit and 32-bit functions operate in the same way as their 8-bit
1703: counterparts; they just use different data types for their arguments
1704: and results, and their names start with pcre16_ or pcre32_ instead of
1705: pcre_. For every option that has UTF8 in its name (for example,
1706: PCRE_UTF8), there are corresponding 16-bit and 32-bit names with UTF8
1707: replaced by UTF16 or UTF32, respectively. This facility is in fact just
1708: cosmetic; the 16-bit and 32-bit option names define the same bit val-
1.5 misha 1709: ues.
1710:
1711: References to bytes and UTF-8 in this document should be read as refer-
1.6 misha 1712: ences to 16-bit data units and UTF-16 when using the 16-bit library, or
1713: 32-bit data units and UTF-32 when using the 32-bit library, unless
1714: specified otherwise. More details of the specific differences for the
1715: 16-bit and 32-bit libraries are given in the pcre16 and pcre32 pages.
1.5 misha 1716:
1717:
1.1 misha 1718: PCRE API OVERVIEW
1719:
1720: PCRE has its own native API, which is described in this document. There
1.6 misha 1721: are also some wrapper functions (for the 8-bit library only) that cor-
1722: respond to the POSIX regular expression API, but they do not give
1723: access to all the functionality. They are described in the pcreposix
1724: documentation. Both of these APIs define a set of C function calls. A
1.5 misha 1725: C++ wrapper (again for the 8-bit library only) is also distributed with
1726: PCRE. It is documented in the pcrecpp page.
1727:
1.6 misha 1728: The native API C function prototypes are defined in the header file
1729: pcre.h, and on Unix-like systems the (8-bit) library itself is called
1730: libpcre. It can normally be accessed by adding -lpcre to the command
1731: for linking an application that uses PCRE. The header file defines the
1.5 misha 1732: macros PCRE_MAJOR and PCRE_MINOR to contain the major and minor release
1.6 misha 1733: numbers for the library. Applications can use these to include support
1.1 misha 1734: for different releases of PCRE.
1735:
1.4 misha 1736: In a Windows environment, if you want to statically link an application
1.6 misha 1737: program against a non-dll pcre.a file, you must define PCRE_STATIC
1738: before including pcre.h or pcrecpp.h, because otherwise the pcre_mal-
1.4 misha 1739: loc() and pcre_free() exported functions will be declared
1740: __declspec(dllimport), with unwanted results.
1741:
1.6 misha 1742: The functions pcre_compile(), pcre_compile2(), pcre_study(), and
1743: pcre_exec() are used for compiling and matching regular expressions in
1744: a Perl-compatible manner. A sample program that demonstrates the sim-
1745: plest way of using them is provided in the file called pcredemo.c in
1.4 misha 1746: the PCRE source distribution. A listing of this program is given in the
1.6 misha 1747: pcredemo documentation, and the pcresample documentation describes how
1.4 misha 1748: to compile and run it.
1.1 misha 1749:
1.6 misha 1750: Just-in-time compiler support is an optional feature of PCRE that can
1.5 misha 1751: be built in appropriate hardware environments. It greatly speeds up the
1.6 misha 1752: matching performance of many patterns. Simple programs can easily
1753: request that it be used if available, by setting an option that is
1754: ignored when it is not relevant. More complicated programs might need
1755: to make use of the functions pcre_jit_stack_alloc(),
1756: pcre_jit_stack_free(), and pcre_assign_jit_stack() in order to control
1757: the JIT code's memory usage.
1758:
1759: From release 8.32 there is also a direct interface for JIT execution,
1760: which gives improved performance. The JIT-specific functions are dis-
1761: cussed in the pcrejit documentation.
1.5 misha 1762:
1.1 misha 1763: A second matching function, pcre_dfa_exec(), which is not Perl-compati-
1.5 misha 1764: ble, is also provided. This uses a different algorithm for the match-
1765: ing. The alternative algorithm finds all possible matches (at a given
1766: point in the subject), and scans the subject just once (unless there
1767: are lookbehind assertions). However, this algorithm does not return
1768: captured substrings. A description of the two matching algorithms and
1769: their advantages and disadvantages is given in the pcrematching docu-
1.4 misha 1770: mentation.
1.1 misha 1771:
1.5 misha 1772: In addition to the main compiling and matching functions, there are
1.1 misha 1773: convenience functions for extracting captured substrings from a subject
1774: string that is matched by pcre_exec(). They are:
1775:
1776: pcre_copy_substring()
1777: pcre_copy_named_substring()
1778: pcre_get_substring()
1779: pcre_get_named_substring()
1780: pcre_get_substring_list()
1781: pcre_get_stringnumber()
1782: pcre_get_stringtable_entries()
1783:
1784: pcre_free_substring() and pcre_free_substring_list() are also provided,
1785: to free the memory used for extracted strings.
1786:
1.5 misha 1787: The function pcre_maketables() is used to build a set of character
1788: tables in the current locale for passing to pcre_compile(),
1789: pcre_exec(), or pcre_dfa_exec(). This is an optional facility that is
1790: provided for specialist use. Most commonly, no special tables are
1791: passed, in which case internal tables that are generated when PCRE is
1.1 misha 1792: built are used.
1793:
1.5 misha 1794: The function pcre_fullinfo() is used to find out information about a
1795: compiled pattern. The function pcre_version() returns a pointer to a
1796: string containing the version of PCRE and its date of release.
1.1 misha 1797:
1.5 misha 1798: The function pcre_refcount() maintains a reference count in a data
1799: block containing a compiled pattern. This is provided for the benefit
1.1 misha 1800: of object-oriented applications.
1801:
1.5 misha 1802: The global variables pcre_malloc and pcre_free initially contain the
1803: entry points of the standard malloc() and free() functions, respec-
1.1 misha 1804: tively. PCRE calls the memory management functions via these variables,
1.5 misha 1805: so a calling program can replace them if it wishes to intercept the
1.1 misha 1806: calls. This should be done before calling any PCRE functions.
1807:
1.5 misha 1808: The global variables pcre_stack_malloc and pcre_stack_free are also
1809: indirections to memory management functions. These special functions
1810: are used only when PCRE is compiled to use the heap for remembering
1.1 misha 1811: data, instead of recursive function calls, when running the pcre_exec()
1.5 misha 1812: function. See the pcrebuild documentation for details of how to do
1813: this. It is a non-standard way of building PCRE, for use in environ-
1814: ments that have limited stacks. Because of the greater use of memory
1815: management, it runs more slowly. Separate functions are provided so
1816: that special-purpose external code can be used for this case. When
1817: used, these functions are always called in a stack-like manner (last
1818: obtained, first freed), and always for memory blocks of the same size.
1819: There is a discussion about PCRE's stack usage in the pcrestack docu-
1.1 misha 1820: mentation.
1821:
1822: The global variable pcre_callout initially contains NULL. It can be set
1.5 misha 1823: by the caller to a "callout" function, which PCRE will then call at
1824: specified points during a matching operation. Details are given in the
1.1 misha 1825: pcrecallout documentation.
1826:
1.7 ! moko 1827: The global variable pcre_stack_guard initially contains NULL. It can be
! 1828: set by the caller to a function that is called by PCRE whenever it
! 1829: starts to compile a parenthesized part of a pattern. When parentheses
! 1830: are nested, PCRE uses recursive function calls, which use up the system
! 1831: stack. This function is provided so that applications with restricted
! 1832: stacks can force a compilation error if the stack runs out. The func-
! 1833: tion should return zero if all is well, or non-zero to force an error.
! 1834:
1.1 misha 1835:
1836: NEWLINES
1837:
1.5 misha 1838: PCRE supports five different conventions for indicating line breaks in
1839: strings: a single CR (carriage return) character, a single LF (line-
1.1 misha 1840: feed) character, the two-character sequence CRLF, any of the three pre-
1.5 misha 1841: ceding, or any Unicode newline sequence. The Unicode newline sequences
1842: are the three just mentioned, plus the single characters VT (vertical
1.6 misha 1843: tab, U+000B), FF (form feed, U+000C), NEL (next line, U+0085), LS (line
1.1 misha 1844: separator, U+2028), and PS (paragraph separator, U+2029).
1845:
1.5 misha 1846: Each of the first three conventions is used by at least one operating
1847: system as its standard newline sequence. When PCRE is built, a default
1848: can be specified. The default default is LF, which is the Unix stan-
1849: dard. When PCRE is run, the default can be overridden, either when a
1.1 misha 1850: pattern is compiled, or when it is matched.
1851:
1852: At compile time, the newline convention can be specified by the options
1.5 misha 1853: argument of pcre_compile(), or it can be specified by special text at
1.1 misha 1854: the start of the pattern itself; this overrides any other settings. See
1855: the pcrepattern page for details of the special character sequences.
1856:
1857: In the PCRE documentation the word "newline" is used to mean "the char-
1.5 misha 1858: acter or pair of characters that indicate a line break". The choice of
1859: newline convention affects the handling of the dot, circumflex, and
1.1 misha 1860: dollar metacharacters, the handling of #-comments in /x mode, and, when
1.5 misha 1861: CRLF is a recognized line ending sequence, the match position advance-
1.1 misha 1862: ment for a non-anchored pattern. There is more detail about this in the
1863: section on pcre_exec() options below.
1864:
1.5 misha 1865: The choice of newline convention does not affect the interpretation of
1866: the \n or \r escape sequences, nor does it affect what \R matches,
1.1 misha 1867: which is controlled in a similar way, but by separate options.
1868:
1869:
1870: MULTITHREADING
1871:
1.5 misha 1872: The PCRE functions can be used in multi-threading applications, with
1.1 misha 1873: the proviso that the memory management functions pointed to by
1874: pcre_malloc, pcre_free, pcre_stack_malloc, and pcre_stack_free, and the
1.7 ! moko 1875: callout and stack-checking functions pointed to by pcre_callout and
! 1876: pcre_stack_guard, are shared by all threads.
1.1 misha 1877:
1.7 ! moko 1878: The compiled form of a regular expression is not altered during match-
1.1 misha 1879: ing, so the same compiled pattern can safely be used by several threads
1880: at once.
1881:
1.7 ! moko 1882: If the just-in-time optimization feature is being used, it needs sepa-
! 1883: rate memory stack areas for each thread. See the pcrejit documentation
1.5 misha 1884: for more details.
1885:
1.1 misha 1886:
1887: SAVING PRECOMPILED PATTERNS FOR LATER USE
1888:
1889: The compiled form of a regular expression can be saved and re-used at a
1.7 ! moko 1890: later time, possibly by a different program, and even on a host other
! 1891: than the one on which it was compiled. Details are given in the
! 1892: pcreprecompile documentation, which includes a description of the
! 1893: pcre_pattern_to_host_byte_order() function. However, compiling a regu-
! 1894: lar expression with one version of PCRE for use with a different ver-
1.5 misha 1895: sion is not guaranteed to work and may cause crashes.
1.1 misha 1896:
1897:
1898: CHECKING BUILD-TIME OPTIONS
1899:
1900: int pcre_config(int what, void *where);
1901:
1.7 ! moko 1902: The function pcre_config() makes it possible for a PCRE client to dis-
1.1 misha 1903: cover which optional features have been compiled into the PCRE library.
1.7 ! moko 1904: The pcrebuild documentation has more details about these optional fea-
1.1 misha 1905: tures.
1906:
1.7 ! moko 1907: The first argument for pcre_config() is an integer, specifying which
1.1 misha 1908: information is required; the second argument is a pointer to a variable
1.7 ! moko 1909: into which the information is placed. The returned value is zero on
! 1910: success, or the negative error code PCRE_ERROR_BADOPTION if the value
! 1911: in the first argument is not recognized. The following information is
1.1 misha 1912: available:
1913:
1914: PCRE_CONFIG_UTF8
1915:
1.7 ! moko 1916: The output is an integer that is set to one if UTF-8 support is avail-
! 1917: able; otherwise it is set to zero. This value should normally be given
1.6 misha 1918: to the 8-bit version of this function, pcre_config(). If it is given to
1.7 ! moko 1919: the 16-bit or 32-bit version of this function, the result is
1.5 misha 1920: PCRE_ERROR_BADOPTION.
1921:
1922: PCRE_CONFIG_UTF16
1923:
1924: The output is an integer that is set to one if UTF-16 support is avail-
1.7 ! moko 1925: able; otherwise it is set to zero. This value should normally be given
1.5 misha 1926: to the 16-bit version of this function, pcre16_config(). If it is given
1.7 ! moko 1927: to the 8-bit or 32-bit version of this function, the result is
1.6 misha 1928: PCRE_ERROR_BADOPTION.
1929:
1930: PCRE_CONFIG_UTF32
1931:
1932: The output is an integer that is set to one if UTF-32 support is avail-
1.7 ! moko 1933: able; otherwise it is set to zero. This value should normally be given
1.6 misha 1934: to the 32-bit version of this function, pcre32_config(). If it is given
1.7 ! moko 1935: to the 8-bit or 16-bit version of this function, the result is
1.6 misha 1936: PCRE_ERROR_BADOPTION.
1.1 misha 1937:
1938: PCRE_CONFIG_UNICODE_PROPERTIES
1939:
1.7 ! moko 1940: The output is an integer that is set to one if support for Unicode
1.1 misha 1941: character properties is available; otherwise it is set to zero.
1942:
1.5 misha 1943: PCRE_CONFIG_JIT
1944:
1945: The output is an integer that is set to one if support for just-in-time
1946: compiling is available; otherwise it is set to zero.
1947:
1948: PCRE_CONFIG_JITTARGET
1949:
1.7 ! moko 1950: The output is a pointer to a zero-terminated "const char *" string. If
1.5 misha 1951: JIT support is available, the string contains the name of the architec-
1.7 ! moko 1952: ture for which the JIT compiler is configured, for example "x86 32bit
! 1953: (little endian + unaligned)". If JIT support is not available, the
1.5 misha 1954: result is NULL.
1955:
1.1 misha 1956: PCRE_CONFIG_NEWLINE
1957:
1.7 ! moko 1958: The output is an integer whose value specifies the default character
! 1959: sequence that is recognized as meaning "newline". The values that are
1.6 misha 1960: supported in ASCII/Unicode environments are: 10 for LF, 13 for CR, 3338
1.7 ! moko 1961: for CRLF, -2 for ANYCRLF, and -1 for ANY. In EBCDIC environments, CR,
! 1962: ANYCRLF, and ANY yield the same values. However, the value for LF is
! 1963: normally 21, though some EBCDIC environments use 37. The corresponding
! 1964: values for CRLF are 3349 and 3365. The default should normally corre-
1.3 misha 1965: spond to the standard sequence for your operating system.
1.1 misha 1966:
1967: PCRE_CONFIG_BSR
1968:
1969: The output is an integer whose value indicates what character sequences
1.7 ! moko 1970: the \R escape sequence matches by default. A value of 0 means that \R
! 1971: matches any Unicode line ending sequence; a value of 1 means that \R
1.1 misha 1972: matches only CR, LF, or CRLF. The default can be overridden when a pat-
1973: tern is compiled or matched.
1974:
1975: PCRE_CONFIG_LINK_SIZE
1976:
1.7 ! moko 1977: The output is an integer that contains the number of bytes used for
1.5 misha 1978: internal linkage in compiled regular expressions. For the 8-bit
1979: library, the value can be 2, 3, or 4. For the 16-bit library, the value
1.7 ! moko 1980: is either 2 or 4 and is still a number of bytes. For the 32-bit
1.6 misha 1981: library, the value is either 2 or 4 and is still a number of bytes. The
1982: default value of 2 is sufficient for all but the most massive patterns,
1.7 ! moko 1983: since it allows the compiled pattern to be up to 64K in size. Larger
! 1984: values allow larger regular expressions to be compiled, at the expense
1.6 misha 1985: of slower matching.
1.1 misha 1986:
1987: PCRE_CONFIG_POSIX_MALLOC_THRESHOLD
1988:
1.7 ! moko 1989: The output is an integer that contains the threshold above which the
! 1990: POSIX interface uses malloc() for output vectors. Further details are
1.1 misha 1991: given in the pcreposix documentation.
1992:
1.7 ! moko 1993: PCRE_CONFIG_PARENS_LIMIT
! 1994:
! 1995: The output is a long integer that gives the maximum depth of nesting of
! 1996: parentheses (of any kind) in a pattern. This limit is imposed to cap
! 1997: the amount of system stack used when a pattern is compiled. It is spec-
! 1998: ified when PCRE is built; the default is 250. This limit does not take
! 1999: into account the stack that may already be used by the calling applica-
! 2000: tion. For finer control over compilation stack usage, you can set a
! 2001: pointer to an external checking function in pcre_stack_guard.
! 2002:
1.1 misha 2003: PCRE_CONFIG_MATCH_LIMIT
2004:
1.7 ! moko 2005: The output is a long integer that gives the default limit for the num-
! 2006: ber of internal matching function calls in a pcre_exec() execution.
1.3 misha 2007: Further details are given with pcre_exec() below.
1.1 misha 2008:
2009: PCRE_CONFIG_MATCH_LIMIT_RECURSION
2010:
1.3 misha 2011: The output is a long integer that gives the default limit for the depth
1.7 ! moko 2012: of recursion when calling the internal matching function in a
! 2013: pcre_exec() execution. Further details are given with pcre_exec()
1.3 misha 2014: below.
1.1 misha 2015:
2016: PCRE_CONFIG_STACKRECURSE
2017:
1.7 ! moko 2018: The output is an integer that is set to one if internal recursion when
1.1 misha 2019: running pcre_exec() is implemented by recursive function calls that use
1.7 ! moko 2020: the stack to remember their state. This is the usual way that PCRE is
1.1 misha 2021: compiled. The output is zero if PCRE was compiled to use blocks of data
1.7 ! moko 2022: on the heap instead of recursive function calls. In this case,
! 2023: pcre_stack_malloc and pcre_stack_free are called to manage memory
1.1 misha 2024: blocks on the heap, thus avoiding the use of the stack.
2025:
2026:
2027: COMPILING A PATTERN
2028:
2029: pcre *pcre_compile(const char *pattern, int options,
2030: const char **errptr, int *erroffset,
2031: const unsigned char *tableptr);
2032:
2033: pcre *pcre_compile2(const char *pattern, int options,
2034: int *errorcodeptr,
2035: const char **errptr, int *erroffset,
2036: const unsigned char *tableptr);
2037:
2038: Either of the functions pcre_compile() or pcre_compile2() can be called
2039: to compile a pattern into an internal form. The only difference between
1.7 ! moko 2040: the two interfaces is that pcre_compile2() has an additional argument,
! 2041: errorcodeptr, via which a numerical error code can be returned. To
! 2042: avoid too much repetition, we refer just to pcre_compile() below, but
1.4 misha 2043: the information applies equally to pcre_compile2().
1.1 misha 2044:
2045: The pattern is a C string terminated by a binary zero, and is passed in
1.7 ! moko 2046: the pattern argument. A pointer to a single block of memory that is
! 2047: obtained via pcre_malloc is returned. This contains the compiled code
1.1 misha 2048: and related data. The pcre type is defined for the returned block; this
2049: is a typedef for a structure whose contents are not externally defined.
2050: It is up to the caller to free the memory (via pcre_free) when it is no
2051: longer required.
2052:
1.7 ! moko 2053: Although the compiled code of a PCRE regex is relocatable, that is, it
1.1 misha 2054: does not depend on memory location, the complete pcre data block is not
1.7 ! moko 2055: fully relocatable, because it may contain a copy of the tableptr argu-
1.1 misha 2056: ment, which is an address (see below).
2057:
2058: The options argument contains various bit settings that affect the com-
1.7 ! moko 2059: pilation. It should be zero if no options are required. The available
! 2060: options are described below. Some of them (in particular, those that
! 2061: are compatible with Perl, but some others as well) can also be set and
! 2062: unset from within the pattern (see the detailed description in the
! 2063: pcrepattern documentation). For those options that can be different in
! 2064: different parts of the pattern, the contents of the options argument
1.4 misha 2065: specifies their settings at the start of compilation and execution. The
1.7 ! moko 2066: PCRE_ANCHORED, PCRE_BSR_xxx, PCRE_NEWLINE_xxx, PCRE_NO_UTF8_CHECK, and
! 2067: PCRE_NO_START_OPTIMIZE options can be set at the time of matching as
1.6 misha 2068: well as at compile time.
1.1 misha 2069:
2070: If errptr is NULL, pcre_compile() returns NULL immediately. Otherwise,
1.7 ! moko 2071: if compilation of a pattern fails, pcre_compile() returns NULL, and
1.1 misha 2072: sets the variable pointed to by errptr to point to a textual error mes-
2073: sage. This is a static string that is part of the library. You must not
1.7 ! moko 2074: try to free it. Normally, the offset from the start of the pattern to
1.6 misha 2075: the data unit that was being processed when the error was discovered is
1.7 ! moko 2076: placed in the variable pointed to by erroffset, which must not be NULL
! 2077: (if it is, an immediate error is given). However, for an invalid UTF-8
! 2078: or UTF-16 string, the offset is that of the first data unit of the
1.6 misha 2079: failing character.
1.4 misha 2080:
1.7 ! moko 2081: Some errors are not detected until the whole pattern has been scanned;
! 2082: in these cases, the offset passed back is the length of the pattern.
! 2083: Note that the offset is in data units, not characters, even in a UTF
1.6 misha 2084: mode. It may sometimes point into the middle of a UTF-8 or UTF-16 char-
2085: acter.
1.4 misha 2086:
1.7 ! moko 2087: If pcre_compile2() is used instead of pcre_compile(), and the error-
! 2088: codeptr argument is not NULL, a non-zero error code number is returned
! 2089: via this argument in the event of an error. This is in addition to the
1.1 misha 2090: textual error message. Error codes and messages are listed below.
2091:
1.7 ! moko 2092: If the final argument, tableptr, is NULL, PCRE uses a default set of
! 2093: character tables that are built when PCRE is compiled, using the
! 2094: default C locale. Otherwise, tableptr must be an address that is the
! 2095: result of a call to pcre_maketables(). This value is stored with the
! 2096: compiled pattern, and used again by pcre_exec() and pcre_dfa_exec()
! 2097: when the pattern is matched. For more discussion, see the section on
! 2098: locale support below.
1.1 misha 2099:
1.7 ! moko 2100: This code fragment shows a typical straightforward call to pcre_com-
1.1 misha 2101: pile():
2102:
2103: pcre *re;
2104: const char *error;
2105: int erroffset;
2106: re = pcre_compile(
2107: "^A.*Z", /* the pattern */
2108: 0, /* default options */
2109: &error, /* for error message */
2110: &erroffset, /* for error offset */
2111: NULL); /* use default character tables */
2112:
1.7 ! moko 2113: The following names for option bits are defined in the pcre.h header
1.1 misha 2114: file:
2115:
2116: PCRE_ANCHORED
2117:
2118: If this bit is set, the pattern is forced to be "anchored", that is, it
1.7 ! moko 2119: is constrained to match only at the first matching point in the string
! 2120: that is being searched (the "subject string"). This effect can also be
! 2121: achieved by appropriate constructs in the pattern itself, which is the
1.1 misha 2122: only way to do it in Perl.
2123:
2124: PCRE_AUTO_CALLOUT
2125:
2126: If this bit is set, pcre_compile() automatically inserts callout items,
1.7 ! moko 2127: all with number 255, before each pattern item. For discussion of the
1.1 misha 2128: callout facility, see the pcrecallout documentation.
2129:
2130: PCRE_BSR_ANYCRLF
2131: PCRE_BSR_UNICODE
2132:
2133: These options (which are mutually exclusive) control what the \R escape
1.7 ! moko 2134: sequence matches. The choice is either to match only CR, LF, or CRLF,
1.1 misha 2135: or to match any Unicode newline sequence. The default is specified when
2136: PCRE is built. It can be overridden from within the pattern, or by set-
2137: ting an option when a compiled pattern is matched.
2138:
2139: PCRE_CASELESS
2140:
1.7 ! moko 2141: If this bit is set, letters in the pattern match both upper and lower
! 2142: case letters. It is equivalent to Perl's /i option, and it can be
! 2143: changed within a pattern by a (?i) option setting. In UTF-8 mode, PCRE
! 2144: always understands the concept of case for characters whose values are
! 2145: less than 128, so caseless matching is always possible. For characters
! 2146: with higher values, the concept of case is supported if PCRE is com-
! 2147: piled with Unicode property support, but not otherwise. If you want to
! 2148: use caseless matching for characters 128 and above, you must ensure
! 2149: that PCRE is compiled with Unicode property support as well as with
1.1 misha 2150: UTF-8 support.
2151:
2152: PCRE_DOLLAR_ENDONLY
2153:
1.7 ! moko 2154: If this bit is set, a dollar metacharacter in the pattern matches only
! 2155: at the end of the subject string. Without this option, a dollar also
! 2156: matches immediately before a newline at the end of the string (but not
! 2157: before any other newlines). The PCRE_DOLLAR_ENDONLY option is ignored
! 2158: if PCRE_MULTILINE is set. There is no equivalent to this option in
1.1 misha 2159: Perl, and no way to set it within a pattern.
2160:
2161: PCRE_DOTALL
2162:
1.7 ! moko 2163: If this bit is set, a dot metacharacter in the pattern matches a char-
1.4 misha 2164: acter of any value, including one that indicates a newline. However, it
1.7 ! moko 2165: only ever matches one character, even if newlines are coded as CRLF.
! 2166: Without this option, a dot does not match when the current position is
1.4 misha 2167: at a newline. This option is equivalent to Perl's /s option, and it can
1.7 ! moko 2168: be changed within a pattern by a (?s) option setting. A negative class
1.4 misha 2169: such as [^a] always matches newline characters, independent of the set-
2170: ting of this option.
1.1 misha 2171:
2172: PCRE_DUPNAMES
2173:
1.7 ! moko 2174: If this bit is set, names used to identify capturing subpatterns need
1.1 misha 2175: not be unique. This can be helpful for certain types of pattern when it
1.7 ! moko 2176: is known that only one instance of the named subpattern can ever be
! 2177: matched. There are more details of named subpatterns below; see also
1.1 misha 2178: the pcrepattern documentation.
2179:
2180: PCRE_EXTENDED
2181:
1.7 ! moko 2182: If this bit is set, most white space characters in the pattern are
! 2183: totally ignored except when escaped or inside a character class. How-
! 2184: ever, white space is not allowed within sequences such as (?> that
! 2185: introduce various parenthesized subpatterns, nor within a numerical
! 2186: quantifier such as {1,3}. However, ignorable white space is permitted
! 2187: between an item and a following quantifier and between a quantifier and
! 2188: a following + that indicates possessiveness.
! 2189:
! 2190: White space did not used to include the VT character (code 11), because
! 2191: Perl did not treat this character as white space. However, Perl changed
! 2192: at release 5.18, so PCRE followed at release 8.34, and VT is now
! 2193: treated as white space.
! 2194:
! 2195: PCRE_EXTENDED also causes characters between an unescaped # outside a
! 2196: character class and the next newline, inclusive, to be ignored.
! 2197: PCRE_EXTENDED is equivalent to Perl's /x option, and it can be changed
! 2198: within a pattern by a (?x) option setting.
! 2199:
! 2200: Which characters are interpreted as newlines is controlled by the
! 2201: options passed to pcre_compile() or by a special sequence at the start
! 2202: of the pattern, as described in the section entitled "Newline conven-
1.4 misha 2203: tions" in the pcrepattern documentation. Note that the end of this type
1.7 ! moko 2204: of comment is a literal newline sequence in the pattern; escape
1.4 misha 2205: sequences that happen to represent a newline do not count.
2206:
1.7 ! moko 2207: This option makes it possible to include comments inside complicated
! 2208: patterns. Note, however, that this applies only to data characters.
! 2209: White space characters may never appear within special character
1.4 misha 2210: sequences in a pattern, for example within the sequence (?( that intro-
2211: duces a conditional subpattern.
1.1 misha 2212:
2213: PCRE_EXTRA
2214:
1.7 ! moko 2215: This option was invented in order to turn on additional functionality
! 2216: of PCRE that is incompatible with Perl, but it is currently of very
! 2217: little use. When set, any backslash in a pattern that is followed by a
! 2218: letter that has no special meaning causes an error, thus reserving
! 2219: these combinations for future expansion. By default, as in Perl, a
! 2220: backslash followed by a letter with no special meaning is treated as a
1.4 misha 2221: literal. (Perl can, however, be persuaded to give an error for this, by
1.7 ! moko 2222: running it with the -w option.) There are at present no other features
! 2223: controlled by this option. It can also be set by a (?X) option setting
1.4 misha 2224: within a pattern.
1.1 misha 2225:
2226: PCRE_FIRSTLINE
2227:
1.7 ! moko 2228: If this option is set, an unanchored pattern is required to match
! 2229: before or at the first newline in the subject string, though the
1.1 misha 2230: matched text may continue over the newline.
2231:
2232: PCRE_JAVASCRIPT_COMPAT
2233:
2234: If this option is set, PCRE's behaviour is changed in some ways so that
1.7 ! moko 2235: it is compatible with JavaScript rather than Perl. The changes are as
1.1 misha 2236: follows:
2237:
1.7 ! moko 2238: (1) A lone closing square bracket in a pattern causes a compile-time
! 2239: error, because this is illegal in JavaScript (by default it is treated
1.1 misha 2240: as a data character). Thus, the pattern AB]CD becomes illegal when this
2241: option is set.
2242:
1.7 ! moko 2243: (2) At run time, a back reference to an unset subpattern group matches
! 2244: an empty string (by default this causes the current matching alterna-
! 2245: tive to fail). A pattern such as (\1)(a) succeeds when this option is
! 2246: set (assuming it can find an "a" in the subject), whereas it fails by
1.1 misha 2247: default, for Perl compatibility.
2248:
1.5 misha 2249: (3) \U matches an upper case "U" character; by default \U causes a com-
2250: pile time error (Perl uses \U to upper case subsequent characters).
2251:
2252: (4) \u matches a lower case "u" character unless it is followed by four
1.7 ! moko 2253: hexadecimal digits, in which case the hexadecimal number defines the
! 2254: code point to match. By default, \u causes a compile time error (Perl
1.5 misha 2255: uses it to upper case the following character).
2256:
1.7 ! moko 2257: (5) \x matches a lower case "x" character unless it is followed by two
! 2258: hexadecimal digits, in which case the hexadecimal number defines the
! 2259: code point to match. By default, as in Perl, a hexadecimal number is
1.5 misha 2260: always expected after \x, but it may have zero, one, or two digits (so,
2261: for example, \xz matches a binary zero character followed by z).
2262:
1.1 misha 2263: PCRE_MULTILINE
2264:
1.7 ! moko 2265: By default, for the purposes of matching "start of line" and "end of
1.6 misha 2266: line", PCRE treats the subject string as consisting of a single line of
1.7 ! moko 2267: characters, even if it actually contains newlines. The "start of line"
1.6 misha 2268: metacharacter (^) matches only at the start of the string, and the "end
1.7 ! moko 2269: of line" metacharacter ($) matches only at the end of the string, or
! 2270: before a terminating newline (except when PCRE_DOLLAR_ENDONLY is set).
! 2271: Note, however, that unless PCRE_DOTALL is set, the "any character"
! 2272: metacharacter (.) does not match at a newline. This behaviour (for ^,
1.6 misha 2273: $, and dot) is the same as Perl.
2274:
1.7 ! moko 2275: When PCRE_MULTILINE it is set, the "start of line" and "end of line"
! 2276: constructs match immediately following or immediately before internal
! 2277: newlines in the subject string, respectively, as well as at the very
! 2278: start and end. This is equivalent to Perl's /m option, and it can be
1.1 misha 2279: changed within a pattern by a (?m) option setting. If there are no new-
1.7 ! moko 2280: lines in a subject string, or no occurrences of ^ or $ in a pattern,
1.1 misha 2281: setting PCRE_MULTILINE has no effect.
2282:
1.6 misha 2283: PCRE_NEVER_UTF
2284:
2285: This option locks out interpretation of the pattern as UTF-8 (or UTF-16
1.7 ! moko 2286: or UTF-32 in the 16-bit and 32-bit libraries). In particular, it pre-
! 2287: vents the creator of the pattern from switching to UTF interpretation
1.6 misha 2288: by starting the pattern with (*UTF). This may be useful in applications
2289: that process patterns from external sources. The combination of
2290: PCRE_UTF8 and PCRE_NEVER_UTF also causes an error.
2291:
1.1 misha 2292: PCRE_NEWLINE_CR
2293: PCRE_NEWLINE_LF
2294: PCRE_NEWLINE_CRLF
2295: PCRE_NEWLINE_ANYCRLF
2296: PCRE_NEWLINE_ANY
2297:
1.7 ! moko 2298: These options override the default newline definition that was chosen
! 2299: when PCRE was built. Setting the first or the second specifies that a
! 2300: newline is indicated by a single character (CR or LF, respectively).
! 2301: Setting PCRE_NEWLINE_CRLF specifies that a newline is indicated by the
! 2302: two-character CRLF sequence. Setting PCRE_NEWLINE_ANYCRLF specifies
1.1 misha 2303: that any of the three preceding sequences should be recognized. Setting
1.7 ! moko 2304: PCRE_NEWLINE_ANY specifies that any Unicode newline sequence should be
1.6 misha 2305: recognized.
2306:
1.7 ! moko 2307: In an ASCII/Unicode environment, the Unicode newline sequences are the
! 2308: three just mentioned, plus the single characters VT (vertical tab,
1.6 misha 2309: U+000B), FF (form feed, U+000C), NEL (next line, U+0085), LS (line sep-
1.7 ! moko 2310: arator, U+2028), and PS (paragraph separator, U+2029). For the 8-bit
1.6 misha 2311: library, the last two are recognized only in UTF-8 mode.
2312:
1.7 ! moko 2313: When PCRE is compiled to run in an EBCDIC (mainframe) environment, the
1.6 misha 2314: code for CR is 0x0d, the same as ASCII. However, the character code for
1.7 ! moko 2315: LF is normally 0x15, though in some EBCDIC environments 0x25 is used.
! 2316: Whichever of these is not LF is made to correspond to Unicode's NEL
! 2317: character. EBCDIC codes are all less than 256. For more details, see
1.6 misha 2318: the pcrebuild documentation.
1.1 misha 2319:
1.7 ! moko 2320: The newline setting in the options word uses three bits that are
1.1 misha 2321: treated as a number, giving eight possibilities. Currently only six are
1.7 ! moko 2322: used (default plus the five values above). This means that if you set
! 2323: more than one newline option, the combination may or may not be sensi-
1.1 misha 2324: ble. For example, PCRE_NEWLINE_CR with PCRE_NEWLINE_LF is equivalent to
1.7 ! moko 2325: PCRE_NEWLINE_CRLF, but other combinations may yield unused numbers and
1.1 misha 2326: cause an error.
2327:
1.7 ! moko 2328: The only time that a line break in a pattern is specially recognized
! 2329: when compiling is when PCRE_EXTENDED is set. CR and LF are white space
! 2330: characters, and so are ignored in this mode. Also, an unescaped # out-
! 2331: side a character class indicates a comment that lasts until after the
! 2332: next line break sequence. In other circumstances, line break sequences
1.4 misha 2333: in patterns are treated as literal data.
1.1 misha 2334:
2335: The newline option that is set at compile time becomes the default that
1.3 misha 2336: is used for pcre_exec() and pcre_dfa_exec(), but it can be overridden.
1.1 misha 2337:
2338: PCRE_NO_AUTO_CAPTURE
2339:
2340: If this option is set, it disables the use of numbered capturing paren-
1.7 ! moko 2341: theses in the pattern. Any opening parenthesis that is not followed by
! 2342: ? behaves as if it were followed by ?: but named parentheses can still
! 2343: be used for capturing (and they acquire numbers in the usual way).
1.1 misha 2344: There is no equivalent of this option in Perl.
2345:
1.7 ! moko 2346: PCRE_NO_AUTO_POSSESS
! 2347:
! 2348: If this option is set, it disables "auto-possessification". This is an
! 2349: optimization that, for example, turns a+b into a++b in order to avoid
! 2350: backtracks into a+ that can never be successful. However, if callouts
! 2351: are in use, auto-possessification means that some of them are never
! 2352: taken. You can set this option if you want the matching functions to do
! 2353: a full unoptimized search and run all the callouts, but it is mainly
! 2354: provided for testing purposes.
! 2355:
1.6 misha 2356: PCRE_NO_START_OPTIMIZE
1.4 misha 2357:
1.7 ! moko 2358: This is an option that acts at matching time; that is, it is really an
! 2359: option for pcre_exec() or pcre_dfa_exec(). If it is set at compile
! 2360: time, it is remembered with the compiled pattern and assumed at match-
! 2361: ing time. This is necessary if you want to use JIT execution, because
! 2362: the JIT compiler needs to know whether or not this option is set. For
1.6 misha 2363: details see the discussion of PCRE_NO_START_OPTIMIZE below.
1.4 misha 2364:
2365: PCRE_UCP
2366:
1.7 ! moko 2367: This option changes the way PCRE processes \B, \b, \D, \d, \S, \s, \W,
! 2368: \w, and some of the POSIX character classes. By default, only ASCII
! 2369: characters are recognized, but if PCRE_UCP is set, Unicode properties
! 2370: are used instead to classify characters. More details are given in the
! 2371: section on generic character types in the pcrepattern page. If you set
! 2372: PCRE_UCP, matching one of the items it affects takes much longer. The
! 2373: option is available only if PCRE has been compiled with Unicode prop-
1.4 misha 2374: erty support.
2375:
1.1 misha 2376: PCRE_UNGREEDY
2377:
1.7 ! moko 2378: This option inverts the "greediness" of the quantifiers so that they
! 2379: are not greedy by default, but become greedy if followed by "?". It is
! 2380: not compatible with Perl. It can also be set by a (?U) option setting
1.1 misha 2381: within the pattern.
2382:
2383: PCRE_UTF8
2384:
1.7 ! moko 2385: This option causes PCRE to regard both the pattern and the subject as
1.5 misha 2386: strings of UTF-8 characters instead of single-byte strings. However, it
1.7 ! moko 2387: is available only when PCRE is built to include UTF support. If not,
! 2388: the use of this option provokes an error. Details of how this option
1.5 misha 2389: changes the behaviour of PCRE are given in the pcreunicode page.
1.1 misha 2390:
2391: PCRE_NO_UTF8_CHECK
2392:
2393: When PCRE_UTF8 is set, the validity of the pattern as a UTF-8 string is
1.7 ! moko 2394: automatically checked. There is a discussion about the validity of
! 2395: UTF-8 strings in the pcreunicode page. If an invalid UTF-8 sequence is
! 2396: found, pcre_compile() returns an error. If you already know that your
! 2397: pattern is valid, and you want to skip this check for performance rea-
! 2398: sons, you can set the PCRE_NO_UTF8_CHECK option. When it is set, the
1.5 misha 2399: effect of passing an invalid UTF-8 string as a pattern is undefined. It
1.7 ! moko 2400: may cause your program to crash or loop. Note that this option can also
! 2401: be passed to pcre_exec() and pcre_dfa_exec(), to suppress the validity
! 2402: checking of subject strings only. If the same string is being matched
! 2403: many times, the option can be safely set for the second and subsequent
1.6 misha 2404: matchings to improve performance.
1.1 misha 2405:
2406:
2407: COMPILATION ERROR CODES
2408:
1.7 ! moko 2409: The following table lists the error codes than may be returned by
! 2410: pcre_compile2(), along with the error messages that may be returned by
! 2411: both compiling functions. Note that error messages are always 8-bit
! 2412: ASCII strings, even in 16-bit or 32-bit mode. As PCRE has developed,
! 2413: some error codes have fallen out of use. To avoid confusion, they have
1.6 misha 2414: not been re-used.
1.1 misha 2415:
2416: 0 no error
2417: 1 \ at end of pattern
2418: 2 \c at end of pattern
2419: 3 unrecognized character follows \
2420: 4 numbers out of order in {} quantifier
2421: 5 number too big in {} quantifier
2422: 6 missing terminating ] for character class
2423: 7 invalid escape sequence in character class
2424: 8 range out of order in character class
2425: 9 nothing to repeat
2426: 10 [this code is not in use]
2427: 11 internal error: unexpected repeat
2428: 12 unrecognized character after (? or (?-
2429: 13 POSIX named classes are supported only within a class
2430: 14 missing )
2431: 15 reference to non-existent subpattern
2432: 16 erroffset passed as NULL
2433: 17 unknown option bit(s) set
2434: 18 missing ) after comment
2435: 19 [this code is not in use]
2436: 20 regular expression is too large
2437: 21 failed to get memory
2438: 22 unmatched parentheses
2439: 23 internal error: code overflow
2440: 24 unrecognized character after (?<
2441: 25 lookbehind assertion is not fixed length
2442: 26 malformed number or name after (?(
2443: 27 conditional group contains more than two branches
2444: 28 assertion expected after (?(
2445: 29 (?R or (?[+-]digits must be followed by )
2446: 30 unknown POSIX class name
2447: 31 POSIX collating elements are not supported
1.5 misha 2448: 32 this version of PCRE is compiled without UTF support
1.1 misha 2449: 33 [this code is not in use]
1.7 ! moko 2450: 34 character value in \x{} or \o{} is too large
1.1 misha 2451: 35 invalid condition (?(0)
2452: 36 \C not allowed in lookbehind assertion
1.5 misha 2453: 37 PCRE does not support \L, \l, \N{name}, \U, or \u
1.1 misha 2454: 38 number after (?C is > 255
2455: 39 closing ) for (?C expected
2456: 40 recursive call could loop indefinitely
2457: 41 unrecognized character after (?P
2458: 42 syntax error in subpattern name (missing terminator)
2459: 43 two named subpatterns have the same name
1.5 misha 2460: 44 invalid UTF-8 string (specifically UTF-8)
1.1 misha 2461: 45 support for \P, \p, and \X has not been compiled
2462: 46 malformed \P or \p sequence
2463: 47 unknown property name after \P or \p
2464: 48 subpattern name is too long (maximum 32 characters)
2465: 49 too many named subpatterns (maximum 10000)
2466: 50 [this code is not in use]
1.5 misha 2467: 51 octal value is greater than \377 in 8-bit non-UTF-8 mode
1.1 misha 2468: 52 internal error: overran compiling workspace
1.4 misha 2469: 53 internal error: previously-checked referenced subpattern
2470: not found
1.1 misha 2471: 54 DEFINE group contains more than one branch
2472: 55 repeating a DEFINE group is not allowed
2473: 56 inconsistent NEWLINE options
2474: 57 \g is not followed by a braced, angle-bracketed, or quoted
2475: name/number or by a plain number
2476: 58 a numbered reference must not be zero
1.4 misha 2477: 59 an argument is not allowed for (*ACCEPT), (*FAIL), or (*COMMIT)
1.6 misha 2478: 60 (*VERB) not recognized or malformed
1.1 misha 2479: 61 number is too big
2480: 62 subpattern name expected
2481: 63 digit expected after (?+
2482: 64 ] is an invalid data character in JavaScript compatibility mode
1.4 misha 2483: 65 different names for subpatterns of the same number are
2484: not allowed
2485: 66 (*MARK) must have an argument
1.5 misha 2486: 67 this version of PCRE is not compiled with Unicode property
2487: support
2488: 68 \c must be followed by an ASCII character
2489: 69 \k is not followed by a braced, angle-bracketed, or quoted name
2490: 70 internal error: unknown opcode in find_fixedlength()
2491: 71 \N is not supported in a class
2492: 72 too many forward references
2493: 73 disallowed Unicode code point (>= 0xd800 && <= 0xdfff)
2494: 74 invalid UTF-16 string (specifically UTF-16)
1.6 misha 2495: 75 name is too long in (*MARK), (*PRUNE), (*SKIP), or (*THEN)
2496: 76 character value in \u.... sequence is too large
2497: 77 invalid UTF-32 string (specifically UTF-32)
1.7 ! moko 2498: 78 setting UTF is disabled by the application
! 2499: 79 non-hex character in \x{} (closing brace missing?)
! 2500: 80 non-octal character in \o{} (closing brace missing?)
! 2501: 81 missing opening brace after \o
! 2502: 82 parentheses are too deeply nested
! 2503: 83 invalid range in character class
! 2504: 84 group name must start with a non-digit
! 2505: 85 parentheses are too deeply nested (stack check)
1.1 misha 2506:
1.7 ! moko 2507: The numbers 32 and 10000 in errors 48 and 49 are defaults; different
1.1 misha 2508: values may be used if the limits were changed when PCRE was built.
2509:
2510:
2511: STUDYING A PATTERN
2512:
1.7 ! moko 2513: pcre_extra *pcre_study(const pcre *code, int options,
1.1 misha 2514: const char **errptr);
2515:
1.7 ! moko 2516: If a compiled pattern is going to be used several times, it is worth
1.1 misha 2517: spending more time analyzing it in order to speed up the time taken for
1.7 ! moko 2518: matching. The function pcre_study() takes a pointer to a compiled pat-
1.1 misha 2519: tern as its first argument. If studying the pattern produces additional
1.7 ! moko 2520: information that will help speed up matching, pcre_study() returns a
! 2521: pointer to a pcre_extra block, in which the study_data field points to
1.1 misha 2522: the results of the study.
2523:
2524: The returned value from pcre_study() can be passed directly to
1.7 ! moko 2525: pcre_exec() or pcre_dfa_exec(). However, a pcre_extra block also con-
! 2526: tains other fields that can be set by the caller before the block is
1.4 misha 2527: passed; these are described below in the section on matching a pattern.
1.1 misha 2528:
1.7 ! moko 2529: If studying the pattern does not produce any useful information,
! 2530: pcre_study() returns NULL by default. In that circumstance, if the
1.6 misha 2531: calling program wants to pass any of the other fields to pcre_exec() or
1.7 ! moko 2532: pcre_dfa_exec(), it must set up its own pcre_extra block. However, if
! 2533: pcre_study() is called with the PCRE_STUDY_EXTRA_NEEDED option, it
1.6 misha 2534: returns a pcre_extra block even if studying did not find any additional
1.7 ! moko 2535: information. It may still return NULL, however, if an error occurs in
1.6 misha 2536: pcre_study().
2537:
1.7 ! moko 2538: The second argument of pcre_study() contains option bits. There are
1.6 misha 2539: three further options in addition to PCRE_STUDY_EXTRA_NEEDED:
2540:
2541: PCRE_STUDY_JIT_COMPILE
2542: PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE
2543: PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE
2544:
1.7 ! moko 2545: If any of these are set, and the just-in-time compiler is available,
! 2546: the pattern is further compiled into machine code that executes much
! 2547: faster than the pcre_exec() interpretive matching function. If the
! 2548: just-in-time compiler is not available, these options are ignored. All
1.6 misha 2549: undefined bits in the options argument must be zero.
1.5 misha 2550:
1.7 ! moko 2551: JIT compilation is a heavyweight optimization. It can take some time
! 2552: for patterns to be analyzed, and for one-off matches and simple pat-
! 2553: terns the benefit of faster execution might be offset by a much slower
1.5 misha 2554: study time. Not all patterns can be optimized by the JIT compiler. For
1.7 ! moko 2555: those that cannot be handled, matching automatically falls back to the
! 2556: pcre_exec() interpreter. For more details, see the pcrejit documenta-
1.5 misha 2557: tion.
1.1 misha 2558:
1.7 ! moko 2559: The third argument for pcre_study() is a pointer for an error message.
! 2560: If studying succeeds (even if no data is returned), the variable it
! 2561: points to is set to NULL. Otherwise it is set to point to a textual
1.1 misha 2562: error message. This is a static string that is part of the library. You
1.7 ! moko 2563: must not try to free it. You should test the error pointer for NULL
1.1 misha 2564: after calling pcre_study(), to be sure that it has run successfully.
2565:
1.7 ! moko 2566: When you are finished with a pattern, you can free the memory used for
1.5 misha 2567: the study data by calling pcre_free_study(). This function was added to
1.7 ! moko 2568: the API for release 8.20. For earlier versions, the memory could be
! 2569: freed with pcre_free(), just like the pattern itself. This will still
! 2570: work in cases where JIT optimization is not used, but it is advisable
1.6 misha 2571: to change to the new function when convenient.
1.5 misha 2572:
1.7 ! moko 2573: This is a typical way in which pcre_study() is used (except that in a
1.5 misha 2574: real application there should be tests for errors):
1.1 misha 2575:
1.5 misha 2576: int rc;
2577: pcre *re;
2578: pcre_extra *sd;
2579: re = pcre_compile("pattern", 0, &error, &erroroffset, NULL);
2580: sd = pcre_study(
1.1 misha 2581: re, /* result of pcre_compile() */
1.5 misha 2582: 0, /* no options */
1.1 misha 2583: &error); /* set to NULL or points to a message */
1.5 misha 2584: rc = pcre_exec( /* see below for details of pcre_exec() options */
2585: re, sd, "subject", 7, 0, 0, ovector, 30);
2586: ...
2587: pcre_free_study(sd);
2588: pcre_free(re);
1.1 misha 2589:
1.4 misha 2590: Studying a pattern does two things: first, a lower bound for the length
2591: of subject string that is needed to match the pattern is computed. This
2592: does not mean that there are any strings of that length that match, but
1.7 ! moko 2593: it does guarantee that no shorter strings match. The value is used to
1.6 misha 2594: avoid wasting time by trying to match strings that are shorter than the
1.7 ! moko 2595: lower bound. You can find out the value in a calling program via the
1.6 misha 2596: pcre_fullinfo() function.
1.4 misha 2597:
2598: Studying a pattern is also useful for non-anchored patterns that do not
1.7 ! moko 2599: have a single fixed starting character. A bitmap of possible starting
! 2600: bytes is created. This speeds up finding a position in the subject at
1.5 misha 2601: which to start matching. (In 16-bit mode, the bitmap is used for 16-bit
1.7 ! moko 2602: values less than 256. In 32-bit mode, the bitmap is used for 32-bit
1.5 misha 2603: values less than 256.)
1.4 misha 2604:
1.7 ! moko 2605: These two optimizations apply to both pcre_exec() and pcre_dfa_exec(),
! 2606: and the information is also used by the JIT compiler. The optimiza-
! 2607: tions can be disabled by setting the PCRE_NO_START_OPTIMIZE option.
! 2608: You might want to do this if your pattern contains callouts or (*MARK)
! 2609: and you want to make use of these facilities in cases where matching
1.6 misha 2610: fails.
2611:
1.7 ! moko 2612: PCRE_NO_START_OPTIMIZE can be specified at either compile time or exe-
! 2613: cution time. However, if PCRE_NO_START_OPTIMIZE is passed to
1.6 misha 2614: pcre_exec(), (that is, after any JIT compilation has happened) JIT exe-
1.7 ! moko 2615: cution is disabled. For JIT execution to work with PCRE_NO_START_OPTI-
1.6 misha 2616: MIZE, the option must be set at compile time.
2617:
2618: There is a longer discussion of PCRE_NO_START_OPTIMIZE below.
1.1 misha 2619:
2620:
2621: LOCALE SUPPORT
2622:
1.7 ! moko 2623: PCRE handles caseless matching, and determines whether characters are
! 2624: letters, digits, or whatever, by reference to a set of tables, indexed
! 2625: by character code point. When running in UTF-8 mode, or in the 16- or
! 2626: 32-bit libraries, this applies only to characters with code points less
! 2627: than 256. By default, higher-valued code points never match escapes
! 2628: such as \w or \d. However, if PCRE is built with Unicode property sup-
! 2629: port, all characters can be tested with \p and \P, or, alternatively,
! 2630: the PCRE_UCP option can be set when a pattern is compiled; this causes
! 2631: \w and friends to use Unicode property support instead of the built-in
! 2632: tables.
! 2633:
! 2634: The use of locales with Unicode is discouraged. If you are handling
! 2635: characters with code points greater than 128, you should either use
! 2636: Unicode support, or use locales, but not try to mix the two.
1.1 misha 2637:
1.3 misha 2638: PCRE contains an internal set of tables that are used when the final
2639: argument of pcre_compile() is NULL. These are sufficient for many
1.1 misha 2640: applications. Normally, the internal tables recognize only ASCII char-
2641: acters. However, when PCRE is built, it is possible to cause the inter-
2642: nal tables to be rebuilt in the default "C" locale of the local system,
2643: which may cause them to be different.
2644:
1.3 misha 2645: The internal tables can always be overridden by tables supplied by the
1.1 misha 2646: application that calls PCRE. These may be created in a different locale
1.3 misha 2647: from the default. As more and more applications change to using Uni-
1.1 misha 2648: code, the need for this locale support is expected to die away.
2649:
1.3 misha 2650: External tables are built by calling the pcre_maketables() function,
2651: which has no arguments, in the relevant locale. The result can then be
1.7 ! moko 2652: passed to pcre_compile() as often as necessary. For example, to build
! 2653: and use tables that are appropriate for the French locale (where
! 2654: accented characters with values greater than 128 are treated as let-
! 2655: ters), the following code could be used:
1.1 misha 2656:
2657: setlocale(LC_CTYPE, "fr_FR");
2658: tables = pcre_maketables();
2659: re = pcre_compile(..., tables);
2660:
1.3 misha 2661: The locale name "fr_FR" is used on Linux and other Unix-like systems;
1.1 misha 2662: if you are using Windows, the name for the French locale is "french".
2663:
1.3 misha 2664: When pcre_maketables() runs, the tables are built in memory that is
2665: obtained via pcre_malloc. It is the caller's responsibility to ensure
2666: that the memory containing the tables remains available for as long as
1.1 misha 2667: it is needed.
2668:
2669: The pointer that is passed to pcre_compile() is saved with the compiled
1.3 misha 2670: pattern, and the same tables are used via this pointer by pcre_study()
1.7 ! moko 2671: and also by pcre_exec() and pcre_dfa_exec(). Thus, for any single pat-
1.1 misha 2672: tern, compilation, studying and matching all happen in the same locale,
1.7 ! moko 2673: but different patterns can be processed in different locales.
1.1 misha 2674:
1.3 misha 2675: It is possible to pass a table pointer or NULL (indicating the use of
1.7 ! moko 2676: the internal tables) to pcre_exec() or pcre_dfa_exec() (see the discus-
! 2677: sion below in the section on matching a pattern). This facility is pro-
! 2678: vided for use with pre-compiled patterns that have been saved and
! 2679: reloaded. Character tables are not saved with patterns, so if a non-
! 2680: standard table was used at compile time, it must be provided again when
! 2681: the reloaded pattern is matched. Attempting to use this facility to
! 2682: match a pattern in a different locale from the one in which it was com-
! 2683: piled is likely to lead to anomalous (usually incorrect) results.
1.1 misha 2684:
2685:
2686: INFORMATION ABOUT A PATTERN
2687:
2688: int pcre_fullinfo(const pcre *code, const pcre_extra *extra,
2689: int what, void *where);
2690:
1.3 misha 2691: The pcre_fullinfo() function returns information about a compiled pat-
1.5 misha 2692: tern. It replaces the pcre_info() function, which was removed from the
2693: library at version 8.30, after more than 10 years of obsolescence.
1.1 misha 2694:
1.3 misha 2695: The first argument for pcre_fullinfo() is a pointer to the compiled
2696: pattern. The second argument is the result of pcre_study(), or NULL if
2697: the pattern was not studied. The third argument specifies which piece
2698: of information is required, and the fourth argument is a pointer to a
2699: variable to receive the data. The yield of the function is zero for
1.1 misha 2700: success, or one of the following negative numbers:
2701:
1.5 misha 2702: PCRE_ERROR_NULL the argument code was NULL
2703: the argument where was NULL
2704: PCRE_ERROR_BADMAGIC the "magic number" was not found
2705: PCRE_ERROR_BADENDIANNESS the pattern was compiled with different
2706: endianness
2707: PCRE_ERROR_BADOPTION the value of what was invalid
1.6 misha 2708: PCRE_ERROR_UNSET the requested field is not set
1.1 misha 2709:
1.3 misha 2710: The "magic number" is placed at the start of each compiled pattern as
1.5 misha 2711: an simple check against passing an arbitrary memory pointer. The endi-
2712: anness error can occur if a compiled pattern is saved and reloaded on a
2713: different host. Here is a typical call of pcre_fullinfo(), to obtain
2714: the length of the compiled pattern:
1.1 misha 2715:
2716: int rc;
2717: size_t length;
2718: rc = pcre_fullinfo(
2719: re, /* result of pcre_compile() */
1.5 misha 2720: sd, /* result of pcre_study(), or NULL */
1.1 misha 2721: PCRE_INFO_SIZE, /* what is required */
2722: &length); /* where to put the data */
2723:
1.5 misha 2724: The possible values for the third argument are defined in pcre.h, and
1.1 misha 2725: are as follows:
2726:
2727: PCRE_INFO_BACKREFMAX
2728:
1.5 misha 2729: Return the number of the highest back reference in the pattern. The
2730: fourth argument should point to an int variable. Zero is returned if
1.1 misha 2731: there are no back references.
2732:
2733: PCRE_INFO_CAPTURECOUNT
2734:
1.5 misha 2735: Return the number of capturing subpatterns in the pattern. The fourth
1.1 misha 2736: argument should point to an int variable.
2737:
2738: PCRE_INFO_DEFAULT_TABLES
2739:
1.5 misha 2740: Return a pointer to the internal default character tables within PCRE.
2741: The fourth argument should point to an unsigned char * variable. This
1.1 misha 2742: information call is provided for internal use by the pcre_study() func-
1.5 misha 2743: tion. External callers can cause PCRE to use its internal tables by
1.1 misha 2744: passing a NULL table pointer.
2745:
1.7 ! moko 2746: PCRE_INFO_FIRSTBYTE (deprecated)
1.1 misha 2747:
1.5 misha 2748: Return information about the first data unit of any matched string, for
1.7 ! moko 2749: a non-anchored pattern. The name of this option refers to the 8-bit
! 2750: library, where data units are bytes. The fourth argument should point
! 2751: to an int variable. Negative values are used for special cases. How-
! 2752: ever, this means that when the 32-bit library is in non-UTF-32 mode,
! 2753: the full 32-bit range of characters cannot be returned. For this rea-
! 2754: son, this value is deprecated; use PCRE_INFO_FIRSTCHARACTERFLAGS and
! 2755: PCRE_INFO_FIRSTCHARACTER instead.
1.5 misha 2756:
2757: If there is a fixed first value, for example, the letter "c" from a
2758: pattern such as (cat|cow|coyote), its value is returned. In the 8-bit
1.6 misha 2759: library, the value is always less than 256. In the 16-bit library the
2760: value can be up to 0xffff. In the 32-bit library the value can be up to
2761: 0x10ffff.
1.1 misha 2762:
1.5 misha 2763: If there is no fixed first value, and if either
1.1 misha 2764:
1.6 misha 2765: (a) the pattern was compiled with the PCRE_MULTILINE option, and every
1.1 misha 2766: branch starts with "^", or
2767:
2768: (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not
2769: set (if it were set, the pattern would be anchored),
2770:
1.6 misha 2771: -1 is returned, indicating that the pattern matches only at the start
2772: of a subject string or after any newline within the string. Otherwise
1.1 misha 2773: -2 is returned. For anchored patterns, -2 is returned.
2774:
1.7 ! moko 2775: PCRE_INFO_FIRSTCHARACTER
! 2776:
! 2777: Return the value of the first data unit (non-UTF character) of any
! 2778: matched string in the situation where PCRE_INFO_FIRSTCHARACTERFLAGS
! 2779: returns 1; otherwise return 0. The fourth argument should point to an
! 2780: uint_t variable.
! 2781:
! 2782: In the 8-bit library, the value is always less than 256. In the 16-bit
! 2783: library the value can be up to 0xffff. In the 32-bit library in UTF-32
! 2784: mode the value can be up to 0x10ffff, and up to 0xffffffff when not
! 2785: using UTF-32 mode.
! 2786:
! 2787: PCRE_INFO_FIRSTCHARACTERFLAGS
! 2788:
! 2789: Return information about the first data unit of any matched string, for
! 2790: a non-anchored pattern. The fourth argument should point to an int
! 2791: variable.
! 2792:
! 2793: If there is a fixed first value, for example, the letter "c" from a
! 2794: pattern such as (cat|cow|coyote), 1 is returned, and the character
! 2795: value can be retrieved using PCRE_INFO_FIRSTCHARACTER. If there is no
! 2796: fixed first value, and if either
! 2797:
! 2798: (a) the pattern was compiled with the PCRE_MULTILINE option, and every
! 2799: branch starts with "^", or
! 2800:
! 2801: (b) every branch of the pattern starts with ".*" and PCRE_DOTALL is not
! 2802: set (if it were set, the pattern would be anchored),
! 2803:
! 2804: 2 is returned, indicating that the pattern matches only at the start of
! 2805: a subject string or after any newline within the string. Otherwise 0 is
! 2806: returned. For anchored patterns, 0 is returned.
1.6 misha 2807:
1.1 misha 2808: PCRE_INFO_FIRSTTABLE
2809:
1.5 misha 2810: If the pattern was studied, and this resulted in the construction of a
2811: 256-bit table indicating a fixed set of values for the first data unit
2812: in any matching string, a pointer to the table is returned. Otherwise
2813: NULL is returned. The fourth argument should point to an unsigned char
2814: * variable.
1.1 misha 2815:
2816: PCRE_INFO_HASCRORLF
2817:
1.5 misha 2818: Return 1 if the pattern contains any explicit matches for CR or LF
2819: characters, otherwise 0. The fourth argument should point to an int
2820: variable. An explicit match is either a literal CR or LF character, or
1.1 misha 2821: \r or \n.
2822:
2823: PCRE_INFO_JCHANGED
2824:
1.5 misha 2825: Return 1 if the (?J) or (?-J) option setting is used in the pattern,
2826: otherwise 0. The fourth argument should point to an int variable. (?J)
1.1 misha 2827: and (?-J) set and unset the local PCRE_DUPNAMES option, respectively.
2828:
1.5 misha 2829: PCRE_INFO_JIT
2830:
1.6 misha 2831: Return 1 if the pattern was studied with one of the JIT options, and
2832: just-in-time compiling was successful. The fourth argument should point
2833: to an int variable. A return value of 0 means that JIT support is not
2834: available in this version of PCRE, or that the pattern was not studied
2835: with a JIT option, or that the JIT compiler could not handle this par-
2836: ticular pattern. See the pcrejit documentation for details of what can
2837: and cannot be handled.
1.5 misha 2838:
2839: PCRE_INFO_JITSIZE
2840:
1.6 misha 2841: If the pattern was successfully studied with a JIT option, return the
2842: size of the JIT compiled code, otherwise return zero. The fourth argu-
2843: ment should point to a size_t variable.
1.5 misha 2844:
1.1 misha 2845: PCRE_INFO_LASTLITERAL
2846:
1.5 misha 2847: Return the value of the rightmost literal data unit that must exist in
2848: any matched string, other than at its start, if such a value has been
1.1 misha 2849: recorded. The fourth argument should point to an int variable. If there
1.5 misha 2850: is no such value, -1 is returned. For anchored patterns, a last literal
2851: value is recorded only if it follows something of variable length. For
1.1 misha 2852: example, for the pattern /^a\d+z\d+/ the returned value is "z", but for
2853: /^a\dz\d/ the returned value is -1.
2854:
1.6 misha 2855: Since for the 32-bit library using the non-UTF-32 mode, this function
1.7 ! moko 2856: is unable to return the full 32-bit range of characters, this value is
! 2857: deprecated; instead the PCRE_INFO_REQUIREDCHARFLAGS and
1.6 misha 2858: PCRE_INFO_REQUIREDCHAR values should be used.
2859:
1.7 ! moko 2860: PCRE_INFO_MATCH_EMPTY
! 2861:
! 2862: Return 1 if the pattern can match an empty string, otherwise 0. The
! 2863: fourth argument should point to an int variable.
! 2864:
1.6 misha 2865: PCRE_INFO_MATCHLIMIT
2866:
1.7 ! moko 2867: If the pattern set a match limit by including an item of the form
! 2868: (*LIMIT_MATCH=nnnn) at the start, the value is returned. The fourth
! 2869: argument should point to an unsigned 32-bit integer. If no such value
! 2870: has been set, the call to pcre_fullinfo() returns the error
1.6 misha 2871: PCRE_ERROR_UNSET.
2872:
2873: PCRE_INFO_MAXLOOKBEHIND
2874:
1.7 ! moko 2875: Return the number of characters (NB not data units) in the longest
! 2876: lookbehind assertion in the pattern. This information is useful when
! 2877: doing multi-segment matching using the partial matching facilities.
1.6 misha 2878: Note that the simple assertions \b and \B require a one-character look-
1.7 ! moko 2879: behind. \A also registers a one-character lookbehind, though it does
! 2880: not actually inspect the previous character. This is to ensure that at
1.6 misha 2881: least one character from the old segment is retained when a new segment
2882: is processed. Otherwise, if there are no lookbehinds in the pattern, \A
2883: might match incorrectly at the start of a new segment.
2884:
1.4 misha 2885: PCRE_INFO_MINLENGTH
2886:
1.7 ! moko 2887: If the pattern was studied and a minimum length for matching subject
! 2888: strings was computed, its value is returned. Otherwise the returned
1.6 misha 2889: value is -1. The value is a number of characters, which in UTF mode may
1.7 ! moko 2890: be different from the number of data units. The fourth argument should
! 2891: point to an int variable. A non-negative value is a lower bound to the
! 2892: length of any matching string. There may not be any strings of that
! 2893: length that do actually match, but every string that does match is at
1.5 misha 2894: least that long.
1.4 misha 2895:
1.1 misha 2896: PCRE_INFO_NAMECOUNT
2897: PCRE_INFO_NAMEENTRYSIZE
2898: PCRE_INFO_NAMETABLE
2899:
1.7 ! moko 2900: PCRE supports the use of named as well as numbered capturing parenthe-
! 2901: ses. The names are just an additional way of identifying the parenthe-
1.1 misha 2902: ses, which still acquire numbers. Several convenience functions such as
1.7 ! moko 2903: pcre_get_named_substring() are provided for extracting captured sub-
! 2904: strings by name. It is also possible to extract the data directly, by
! 2905: first converting the name to a number in order to access the correct
1.1 misha 2906: pointers in the output vector (described with pcre_exec() below). To do
1.7 ! moko 2907: the conversion, you need to use the name-to-number map, which is
1.1 misha 2908: described by these three values.
2909:
2910: The map consists of a number of fixed-size entries. PCRE_INFO_NAMECOUNT
2911: gives the number of entries, and PCRE_INFO_NAMEENTRYSIZE gives the size
1.7 ! moko 2912: of each entry; both of these return an int value. The entry size
! 2913: depends on the length of the longest name. PCRE_INFO_NAMETABLE returns
1.5 misha 2914: a pointer to the first entry of the table. This is a pointer to char in
2915: the 8-bit library, where the first two bytes of each entry are the num-
1.7 ! moko 2916: ber of the capturing parenthesis, most significant byte first. In the
! 2917: 16-bit library, the pointer points to 16-bit data units, the first of
! 2918: which contains the parenthesis number. In the 32-bit library, the
! 2919: pointer points to 32-bit data units, the first of which contains the
! 2920: parenthesis number. The rest of the entry is the corresponding name,
1.6 misha 2921: zero terminated.
1.4 misha 2922:
1.7 ! moko 2923: The names are in alphabetical order. If (?| is used to create multiple
! 2924: groups with the same number, as described in the section on duplicate
! 2925: subpattern numbers in the pcrepattern page, the groups may be given the
! 2926: same name, but there is only one entry in the table. Different names
! 2927: for groups of the same number are not permitted. Duplicate names for
! 2928: subpatterns with different numbers are permitted, but only if PCRE_DUP-
! 2929: NAMES is set. They appear in the table in the order in which they were
! 2930: found in the pattern. In the absence of (?| this is the order of
! 2931: increasing number; when (?| is used this is not necessarily the case
! 2932: because later subpatterns may have lower numbers.
1.4 misha 2933:
1.6 misha 2934: As a simple example of the name/number table, consider the following
1.5 misha 2935: pattern after compilation by the 8-bit library (assume PCRE_EXTENDED is
2936: set, so white space - including newlines - is ignored):
1.1 misha 2937:
2938: (?<date> (?<year>(\d\d)?\d\d) -
2939: (?<month>\d\d) - (?<day>\d\d) )
2940:
1.6 misha 2941: There are four named subpatterns, so the table has four entries, and
2942: each entry in the table is eight bytes long. The table is as follows,
1.1 misha 2943: with non-printing bytes shows in hexadecimal, and undefined bytes shown
2944: as ??:
2945:
2946: 00 01 d a t e 00 ??
2947: 00 05 d a y 00 ?? ??
2948: 00 04 m o n t h 00
2949: 00 02 y e a r 00 ??
2950:
1.6 misha 2951: When writing code to extract data from named subpatterns using the
2952: name-to-number map, remember that the length of the entries is likely
1.1 misha 2953: to be different for each compiled pattern.
2954:
2955: PCRE_INFO_OKPARTIAL
2956:
1.6 misha 2957: Return 1 if the pattern can be used for partial matching with
2958: pcre_exec(), otherwise 0. The fourth argument should point to an int
2959: variable. From release 8.00, this always returns 1, because the
2960: restrictions that previously applied to partial matching have been
2961: lifted. The pcrepartial documentation gives details of partial match-
1.4 misha 2962: ing.
1.1 misha 2963:
2964: PCRE_INFO_OPTIONS
2965:
1.6 misha 2966: Return a copy of the options with which the pattern was compiled. The
2967: fourth argument should point to an unsigned long int variable. These
1.1 misha 2968: option bits are those specified in the call to pcre_compile(), modified
2969: by any top-level option settings at the start of the pattern itself. In
1.6 misha 2970: other words, they are the options that will be in force when matching
2971: starts. For example, if the pattern /(?im)abc(?-i)d/ is compiled with
2972: the PCRE_EXTENDED option, the result is PCRE_CASELESS, PCRE_MULTILINE,
1.1 misha 2973: and PCRE_EXTENDED.
2974:
1.6 misha 2975: A pattern is automatically anchored by PCRE if all of its top-level
1.1 misha 2976: alternatives begin with one of the following:
2977:
2978: ^ unless PCRE_MULTILINE is set
2979: \A always
2980: \G always
2981: .* if PCRE_DOTALL is set and there are no back
2982: references to the subpattern in which .* appears
2983:
2984: For such patterns, the PCRE_ANCHORED bit is set in the options returned
2985: by pcre_fullinfo().
2986:
1.6 misha 2987: PCRE_INFO_RECURSIONLIMIT
2988:
2989: If the pattern set a recursion limit by including an item of the form
2990: (*LIMIT_RECURSION=nnnn) at the start, the value is returned. The fourth
2991: argument should point to an unsigned 32-bit integer. If no such value
2992: has been set, the call to pcre_fullinfo() returns the error
2993: PCRE_ERROR_UNSET.
2994:
1.1 misha 2995: PCRE_INFO_SIZE
2996:
1.6 misha 2997: Return the size of the compiled pattern in bytes (for all three
2998: libraries). The fourth argument should point to a size_t variable. This
2999: value does not include the size of the pcre structure that is returned
3000: by pcre_compile(). The value that is passed as the argument to
3001: pcre_malloc() when pcre_compile() is getting memory in which to place
3002: the compiled data is the value returned by this option plus the size of
3003: the pcre structure. Studying a compiled pattern, with or without JIT,
3004: does not alter the value returned by this option.
1.1 misha 3005:
3006: PCRE_INFO_STUDYSIZE
3007:
1.6 misha 3008: Return the size in bytes (for all three libraries) of the data block
3009: pointed to by the study_data field in a pcre_extra block. If pcre_extra
3010: is NULL, or there is no study data, zero is returned. The fourth argu-
3011: ment should point to a size_t variable. The study_data field is set by
3012: pcre_study() to record information that will speed up matching (see the
3013: section entitled "Studying a pattern" above). The format of the
3014: study_data block is private, but its length is made available via this
3015: option so that it can be saved and restored (see the pcreprecompile
3016: documentation for details).
3017:
3018: PCRE_INFO_REQUIREDCHARFLAGS
3019:
1.7 ! moko 3020: Returns 1 if there is a rightmost literal data unit that must exist in
1.6 misha 3021: any matched string, other than at its start. The fourth argument should
1.7 ! moko 3022: point to an int variable. If there is no such value, 0 is returned. If
1.6 misha 3023: returning 1, the character value itself can be retrieved using
3024: PCRE_INFO_REQUIREDCHAR.
3025:
3026: For anchored patterns, a last literal value is recorded only if it fol-
1.7 ! moko 3027: lows something of variable length. For example, for the pattern
! 3028: /^a\d+z\d+/ the returned value 1 (with "z" returned from
1.6 misha 3029: PCRE_INFO_REQUIREDCHAR), but for /^a\dz\d/ the returned value is 0.
3030:
3031: PCRE_INFO_REQUIREDCHAR
3032:
1.7 ! moko 3033: Return the value of the rightmost literal data unit that must exist in
! 3034: any matched string, other than at its start, if such a value has been
! 3035: recorded. The fourth argument should point to an uint32_t variable. If
1.6 misha 3036: there is no such value, 0 is returned.
1.1 misha 3037:
3038:
3039: REFERENCE COUNTS
3040:
3041: int pcre_refcount(pcre *code, int adjust);
3042:
1.7 ! moko 3043: The pcre_refcount() function is used to maintain a reference count in
1.1 misha 3044: the data block that contains a compiled pattern. It is provided for the
1.7 ! moko 3045: benefit of applications that operate in an object-oriented manner,
1.1 misha 3046: where different parts of the application may be using the same compiled
3047: pattern, but you want to free the block when they are all done.
3048:
3049: When a pattern is compiled, the reference count field is initialized to
1.7 ! moko 3050: zero. It is changed only by calling this function, whose action is to
! 3051: add the adjust value (which may be positive or negative) to it. The
1.1 misha 3052: yield of the function is the new value. However, the value of the count
1.7 ! moko 3053: is constrained to lie between 0 and 65535, inclusive. If the new value
1.1 misha 3054: is outside these limits, it is forced to the appropriate limit value.
3055:
1.7 ! moko 3056: Except when it is zero, the reference count is not correctly preserved
! 3057: if a pattern is compiled on one host and then transferred to a host
1.1 misha 3058: whose byte-order is different. (This seems a highly unlikely scenario.)
3059:
3060:
3061: MATCHING A PATTERN: THE TRADITIONAL FUNCTION
3062:
3063: int pcre_exec(const pcre *code, const pcre_extra *extra,
3064: const char *subject, int length, int startoffset,
3065: int options, int *ovector, int ovecsize);
3066:
1.7 ! moko 3067: The function pcre_exec() is called to match a subject string against a
! 3068: compiled pattern, which is passed in the code argument. If the pattern
! 3069: was studied, the result of the study should be passed in the extra
! 3070: argument. You can call pcre_exec() with the same code and extra argu-
! 3071: ments as many times as you like, in order to match different subject
1.5 misha 3072: strings with the same pattern.
3073:
1.7 ! moko 3074: This function is the main matching facility of the library, and it
! 3075: operates in a Perl-like manner. For specialist use there is also an
! 3076: alternative matching function, which is described below in the section
1.5 misha 3077: about the pcre_dfa_exec() function.
1.1 misha 3078:
1.7 ! moko 3079: In most applications, the pattern will have been compiled (and option-
! 3080: ally studied) in the same process that calls pcre_exec(). However, it
1.1 misha 3081: is possible to save compiled patterns and study data, and then use them
1.7 ! moko 3082: later in different processes, possibly even on different hosts. For a
1.1 misha 3083: discussion about this, see the pcreprecompile documentation.
3084:
3085: Here is an example of a simple call to pcre_exec():
3086:
3087: int rc;
3088: int ovector[30];
3089: rc = pcre_exec(
3090: re, /* result of pcre_compile() */
3091: NULL, /* we didn't study the pattern */
3092: "some string", /* the subject string */
3093: 11, /* the length of the subject string */
3094: 0, /* start at offset 0 in the subject */
3095: 0, /* default options */
3096: ovector, /* vector of integers for substring information */
3097: 30); /* number of elements (NOT size in bytes) */
3098:
3099: Extra data for pcre_exec()
3100:
1.7 ! moko 3101: If the extra argument is not NULL, it must point to a pcre_extra data
! 3102: block. The pcre_study() function returns such a block (when it doesn't
! 3103: return NULL), but you can also create one for yourself, and pass addi-
! 3104: tional information in it. The pcre_extra block contains the following
1.1 misha 3105: fields (not necessarily in this order):
3106:
3107: unsigned long int flags;
3108: void *study_data;
1.5 misha 3109: void *executable_jit;
1.1 misha 3110: unsigned long int match_limit;
3111: unsigned long int match_limit_recursion;
3112: void *callout_data;
3113: const unsigned char *tables;
1.4 misha 3114: unsigned char **mark;
1.1 misha 3115:
1.7 ! moko 3116: In the 16-bit version of this structure, the mark field has type
1.5 misha 3117: "PCRE_UCHAR16 **".
3118:
1.7 ! moko 3119: In the 32-bit version of this structure, the mark field has type
1.6 misha 3120: "PCRE_UCHAR32 **".
3121:
1.7 ! moko 3122: The flags field is used to specify which of the other fields are set.
1.6 misha 3123: The flag bits are:
1.1 misha 3124:
1.6 misha 3125: PCRE_EXTRA_CALLOUT_DATA
1.5 misha 3126: PCRE_EXTRA_EXECUTABLE_JIT
1.6 misha 3127: PCRE_EXTRA_MARK
1.1 misha 3128: PCRE_EXTRA_MATCH_LIMIT
3129: PCRE_EXTRA_MATCH_LIMIT_RECURSION
1.6 misha 3130: PCRE_EXTRA_STUDY_DATA
1.1 misha 3131: PCRE_EXTRA_TABLES
3132:
1.7 ! moko 3133: Other flag bits should be set to zero. The study_data field and some-
! 3134: times the executable_jit field are set in the pcre_extra block that is
! 3135: returned by pcre_study(), together with the appropriate flag bits. You
! 3136: should not set these yourself, but you may add to the block by setting
1.6 misha 3137: other fields and their corresponding flag bits.
1.1 misha 3138:
3139: The match_limit field provides a means of preventing PCRE from using up
1.7 ! moko 3140: a vast amount of resources when running patterns that are not going to
! 3141: match, but which have a very large number of possibilities in their
! 3142: search trees. The classic example is a pattern that uses nested unlim-
1.4 misha 3143: ited repeats.
3144:
1.7 ! moko 3145: Internally, pcre_exec() uses a function called match(), which it calls
! 3146: repeatedly (sometimes recursively). The limit set by match_limit is
! 3147: imposed on the number of times this function is called during a match,
! 3148: which has the effect of limiting the amount of backtracking that can
1.5 misha 3149: take place. For patterns that are not anchored, the count restarts from
3150: zero for each position in the subject string.
3151:
3152: When pcre_exec() is called with a pattern that was successfully studied
1.7 ! moko 3153: with a JIT option, the way that the matching is executed is entirely
1.6 misha 3154: different. However, there is still the possibility of runaway matching
3155: that goes on for a very long time, and so the match_limit value is also
3156: used in this case (but in a different way) to limit how long the match-
3157: ing can continue.
3158:
1.7 ! moko 3159: The default value for the limit can be set when PCRE is built; the
! 3160: default default is 10 million, which handles all but the most extreme
! 3161: cases. You can override the default by suppling pcre_exec() with a
! 3162: pcre_extra block in which match_limit is set, and
! 3163: PCRE_EXTRA_MATCH_LIMIT is set in the flags field. If the limit is
1.1 misha 3164: exceeded, pcre_exec() returns PCRE_ERROR_MATCHLIMIT.
3165:
1.7 ! moko 3166: A value for the match limit may also be supplied by an item at the
1.6 misha 3167: start of a pattern of the form
3168:
3169: (*LIMIT_MATCH=d)
3170:
1.7 ! moko 3171: where d is a decimal number. However, such a setting is ignored unless
! 3172: d is less than the limit set by the caller of pcre_exec() or, if no
1.6 misha 3173: such limit is set, less than the default.
3174:
1.7 ! moko 3175: The match_limit_recursion field is similar to match_limit, but instead
1.1 misha 3176: of limiting the total number of times that match() is called, it limits
1.7 ! moko 3177: the depth of recursion. The recursion depth is a smaller number than
! 3178: the total number of calls, because not all calls to match() are recur-
1.1 misha 3179: sive. This limit is of use only if it is set smaller than match_limit.
3180:
1.7 ! moko 3181: Limiting the recursion depth limits the amount of machine stack that
! 3182: can be used, or, when PCRE has been compiled to use memory on the heap
! 3183: instead of the stack, the amount of heap memory that can be used. This
! 3184: limit is not relevant, and is ignored, when matching is done using JIT
1.6 misha 3185: compiled code.
1.5 misha 3186:
1.7 ! moko 3187: The default value for match_limit_recursion can be set when PCRE is
! 3188: built; the default default is the same value as the default for
! 3189: match_limit. You can override the default by suppling pcre_exec() with
! 3190: a pcre_extra block in which match_limit_recursion is set, and
! 3191: PCRE_EXTRA_MATCH_LIMIT_RECURSION is set in the flags field. If the
1.1 misha 3192: limit is exceeded, pcre_exec() returns PCRE_ERROR_RECURSIONLIMIT.
3193:
1.7 ! moko 3194: A value for the recursion limit may also be supplied by an item at the
1.6 misha 3195: start of a pattern of the form
3196:
3197: (*LIMIT_RECURSION=d)
3198:
1.7 ! moko 3199: where d is a decimal number. However, such a setting is ignored unless
! 3200: d is less than the limit set by the caller of pcre_exec() or, if no
1.6 misha 3201: such limit is set, less than the default.
3202:
1.7 ! moko 3203: The callout_data field is used in conjunction with the "callout" fea-
1.4 misha 3204: ture, and is described in the pcrecallout documentation.
1.1 misha 3205:
1.7 ! moko 3206: The tables field is provided for use with patterns that have been pre-
! 3207: compiled using custom character tables, saved to disc or elsewhere, and
! 3208: then reloaded, because the tables that were used to compile a pattern
! 3209: are not saved with it. See the pcreprecompile documentation for a dis-
! 3210: cussion of saving compiled patterns for later use. If NULL is passed
! 3211: using this mechanism, it forces PCRE's internal tables to be used.
! 3212:
! 3213: Warning: The tables that pcre_exec() uses must be the same as those
! 3214: that were used when the pattern was compiled. If this is not the case,
! 3215: the behaviour of pcre_exec() is undefined. Therefore, when a pattern is
! 3216: compiled and matched in the same process, this field should never be
! 3217: set. In this (the most common) case, the correct table pointer is auto-
! 3218: matically passed with the compiled pattern from pcre_compile() to
! 3219: pcre_exec().
! 3220:
! 3221: If PCRE_EXTRA_MARK is set in the flags field, the mark field must be
! 3222: set to point to a suitable variable. If the pattern contains any back-
! 3223: tracking control verbs such as (*MARK:NAME), and the execution ends up
! 3224: with a name to pass back, a pointer to the name string (zero termi-
! 3225: nated) is placed in the variable pointed to by the mark field. The
! 3226: names are within the compiled pattern; if you wish to retain such a
! 3227: name you must copy it before freeing the memory of a compiled pattern.
! 3228: If there is no name to pass back, the variable pointed to by the mark
! 3229: field is set to NULL. For details of the backtracking control verbs,
1.5 misha 3230: see the section entitled "Backtracking control" in the pcrepattern doc-
3231: umentation.
1.4 misha 3232:
1.1 misha 3233: Option bits for pcre_exec()
3234:
1.7 ! moko 3235: The unused bits of the options argument for pcre_exec() must be zero.
! 3236: The only bits that may be set are PCRE_ANCHORED, PCRE_NEWLINE_xxx,
! 3237: PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
! 3238: PCRE_NO_START_OPTIMIZE, PCRE_NO_UTF8_CHECK, PCRE_PARTIAL_HARD, and
1.6 misha 3239: PCRE_PARTIAL_SOFT.
3240:
1.7 ! moko 3241: If the pattern was successfully studied with one of the just-in-time
1.6 misha 3242: (JIT) compile options, the only supported options for JIT execution are
1.7 ! moko 3243: PCRE_NO_UTF8_CHECK, PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY,
! 3244: PCRE_NOTEMPTY_ATSTART, PCRE_PARTIAL_HARD, and PCRE_PARTIAL_SOFT. If an
! 3245: unsupported option is used, JIT execution is disabled and the normal
1.6 misha 3246: interpretive code in pcre_exec() is run.
1.5 misha 3247:
1.1 misha 3248: PCRE_ANCHORED
3249:
1.7 ! moko 3250: The PCRE_ANCHORED option limits pcre_exec() to matching at the first
! 3251: matching position. If a pattern was compiled with PCRE_ANCHORED, or
! 3252: turned out to be anchored by virtue of its contents, it cannot be made
1.1 misha 3253: unachored at matching time.
3254:
3255: PCRE_BSR_ANYCRLF
3256: PCRE_BSR_UNICODE
3257:
3258: These options (which are mutually exclusive) control what the \R escape
1.7 ! moko 3259: sequence matches. The choice is either to match only CR, LF, or CRLF,
! 3260: or to match any Unicode newline sequence. These options override the
1.1 misha 3261: choice that was made or defaulted when the pattern was compiled.
3262:
3263: PCRE_NEWLINE_CR
3264: PCRE_NEWLINE_LF
3265: PCRE_NEWLINE_CRLF
3266: PCRE_NEWLINE_ANYCRLF
3267: PCRE_NEWLINE_ANY
3268:
1.7 ! moko 3269: These options override the newline definition that was chosen or
! 3270: defaulted when the pattern was compiled. For details, see the descrip-
! 3271: tion of pcre_compile() above. During matching, the newline choice
! 3272: affects the behaviour of the dot, circumflex, and dollar metacharac-
! 3273: ters. It may also alter the way the match position is advanced after a
1.1 misha 3274: match failure for an unanchored pattern.
3275:
1.7 ! moko 3276: When PCRE_NEWLINE_CRLF, PCRE_NEWLINE_ANYCRLF, or PCRE_NEWLINE_ANY is
! 3277: set, and a match attempt for an unanchored pattern fails when the cur-
! 3278: rent position is at a CRLF sequence, and the pattern contains no
! 3279: explicit matches for CR or LF characters, the match position is
1.1 misha 3280: advanced by two characters instead of one, in other words, to after the
3281: CRLF.
3282:
3283: The above rule is a compromise that makes the most common cases work as
1.7 ! moko 3284: expected. For example, if the pattern is .+A (and the PCRE_DOTALL
1.1 misha 3285: option is not set), it does not match the string "\r\nA" because, after
1.7 ! moko 3286: failing at the start, it skips both the CR and the LF before retrying.
! 3287: However, the pattern [\r\n]A does match that string, because it con-
1.1 misha 3288: tains an explicit CR or LF reference, and so advances only by one char-
3289: acter after the first failure.
3290:
3291: An explicit match for CR of LF is either a literal appearance of one of
1.7 ! moko 3292: those characters, or one of the \r or \n escape sequences. Implicit
! 3293: matches such as [^X] do not count, nor does \s (which includes CR and
1.1 misha 3294: LF in the characters that it matches).
3295:
1.7 ! moko 3296: Notwithstanding the above, anomalous effects may still occur when CRLF
1.1 misha 3297: is a valid newline sequence and explicit \r or \n escapes appear in the
3298: pattern.
3299:
3300: PCRE_NOTBOL
3301:
3302: This option specifies that first character of the subject string is not
1.7 ! moko 3303: the beginning of a line, so the circumflex metacharacter should not
! 3304: match before it. Setting this without PCRE_MULTILINE (at compile time)
! 3305: causes circumflex never to match. This option affects only the behav-
1.1 misha 3306: iour of the circumflex metacharacter. It does not affect \A.
3307:
3308: PCRE_NOTEOL
3309:
3310: This option specifies that the end of the subject string is not the end
1.7 ! moko 3311: of a line, so the dollar metacharacter should not match it nor (except
! 3312: in multiline mode) a newline immediately before it. Setting this with-
1.1 misha 3313: out PCRE_MULTILINE (at compile time) causes dollar never to match. This
1.7 ! moko 3314: option affects only the behaviour of the dollar metacharacter. It does
1.1 misha 3315: not affect \Z or \z.
3316:
3317: PCRE_NOTEMPTY
3318:
3319: An empty string is not considered to be a valid match if this option is
1.7 ! moko 3320: set. If there are alternatives in the pattern, they are tried. If all
! 3321: the alternatives match the empty string, the entire match fails. For
1.1 misha 3322: example, if the pattern
3323:
3324: a?b?
3325:
1.7 ! moko 3326: is applied to a string not beginning with "a" or "b", it matches an
! 3327: empty string at the start of the subject. With PCRE_NOTEMPTY set, this
1.1 misha 3328: match is not valid, so PCRE searches further into the string for occur-
3329: rences of "a" or "b".
3330:
1.4 misha 3331: PCRE_NOTEMPTY_ATSTART
3332:
1.7 ! moko 3333: This is like PCRE_NOTEMPTY, except that an empty string match that is
! 3334: not at the start of the subject is permitted. If the pattern is
1.4 misha 3335: anchored, such a match can occur only if the pattern contains \K.
3336:
1.7 ! moko 3337: Perl has no direct equivalent of PCRE_NOTEMPTY or
! 3338: PCRE_NOTEMPTY_ATSTART, but it does make a special case of a pattern
! 3339: match of the empty string within its split() function, and when using
! 3340: the /g modifier. It is possible to emulate Perl's behaviour after
1.4 misha 3341: matching a null string by first trying the match again at the same off-
1.7 ! moko 3342: set with PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED, and then if that
1.4 misha 3343: fails, by advancing the starting offset (see below) and trying an ordi-
1.7 ! moko 3344: nary match again. There is some code that demonstrates how to do this
! 3345: in the pcredemo sample program. In the most general case, you have to
! 3346: check to see if the newline convention recognizes CRLF as a newline,
! 3347: and if so, and the current character is CR followed by LF, advance the
1.4 misha 3348: starting offset by two characters instead of one.
1.1 misha 3349:
1.3 misha 3350: PCRE_NO_START_OPTIMIZE
3351:
1.7 ! moko 3352: There are a number of optimizations that pcre_exec() uses at the start
! 3353: of a match, in order to speed up the process. For example, if it is
1.4 misha 3354: known that an unanchored match must start with a specific character, it
1.7 ! moko 3355: searches the subject for that character, and fails immediately if it
! 3356: cannot find it, without actually running the main matching function.
1.4 misha 3357: This means that a special item such as (*COMMIT) at the start of a pat-
1.7 ! moko 3358: tern is not considered until after a suitable starting point for the
! 3359: match has been found. Also, when callouts or (*MARK) items are in use,
1.6 misha 3360: these "start-up" optimizations can cause them to be skipped if the pat-
3361: tern is never actually used. The start-up optimizations are in effect a
3362: pre-scan of the subject that takes place before the pattern is run.
3363:
1.7 ! moko 3364: The PCRE_NO_START_OPTIMIZE option disables the start-up optimizations,
! 3365: possibly causing performance to suffer, but ensuring that in cases
! 3366: where the result is "no match", the callouts do occur, and that items
1.4 misha 3367: such as (*COMMIT) and (*MARK) are considered at every possible starting
1.7 ! moko 3368: position in the subject string. If PCRE_NO_START_OPTIMIZE is set at
! 3369: compile time, it cannot be unset at matching time. The use of
1.6 misha 3370: PCRE_NO_START_OPTIMIZE at matching time (that is, passing it to
1.7 ! moko 3371: pcre_exec()) disables JIT execution; in this situation, matching is
1.6 misha 3372: always done using interpretively.
1.4 misha 3373:
1.7 ! moko 3374: Setting PCRE_NO_START_OPTIMIZE can change the outcome of a matching
1.4 misha 3375: operation. Consider the pattern
3376:
3377: (*COMMIT)ABC
3378:
1.7 ! moko 3379: When this is compiled, PCRE records the fact that a match must start
! 3380: with the character "A". Suppose the subject string is "DEFABC". The
! 3381: start-up optimization scans along the subject, finds "A" and runs the
! 3382: first match attempt from there. The (*COMMIT) item means that the pat-
! 3383: tern must match the current starting position, which in this case, it
! 3384: does. However, if the same match is run with PCRE_NO_START_OPTIMIZE
! 3385: set, the initial scan along the subject string does not happen. The
! 3386: first match attempt is run starting from "D" and when this fails,
! 3387: (*COMMIT) prevents any further matches being tried, so the overall
! 3388: result is "no match". If the pattern is studied, more start-up opti-
! 3389: mizations may be used. For example, a minimum length for the subject
1.4 misha 3390: may be recorded. Consider the pattern
3391:
3392: (*MARK:A)(X|Y)
3393:
1.7 ! moko 3394: The minimum length for a match is one character. If the subject is
! 3395: "ABC", there will be attempts to match "ABC", "BC", "C", and then
! 3396: finally an empty string. If the pattern is studied, the final attempt
! 3397: does not take place, because PCRE knows that the subject is too short,
! 3398: and so the (*MARK) is never encountered. In this case, studying the
! 3399: pattern does not affect the overall match result, which is still "no
1.4 misha 3400: match", but it does affect the auxiliary information that is returned.
1.3 misha 3401:
1.1 misha 3402: PCRE_NO_UTF8_CHECK
3403:
3404: When PCRE_UTF8 is set at compile time, the validity of the subject as a
1.7 ! moko 3405: UTF-8 string is automatically checked when pcre_exec() is subsequently
1.6 misha 3406: called. The entire string is checked before any other processing takes
1.7 ! moko 3407: place. The value of startoffset is also checked to ensure that it
! 3408: points to the start of a UTF-8 character. There is a discussion about
! 3409: the validity of UTF-8 strings in the pcreunicode page. If an invalid
! 3410: sequence of bytes is found, pcre_exec() returns the error
1.5 misha 3411: PCRE_ERROR_BADUTF8 or, if PCRE_PARTIAL_HARD is set and the problem is a
3412: truncated character at the end of the subject, PCRE_ERROR_SHORTUTF8. In
1.7 ! moko 3413: both cases, information about the precise nature of the error may also
! 3414: be returned (see the descriptions of these errors in the section enti-
! 3415: tled Error return values from pcre_exec() below). If startoffset con-
1.5 misha 3416: tains a value that does not point to the start of a UTF-8 character (or
3417: to the end of the subject), PCRE_ERROR_BADUTF8_OFFSET is returned.
1.4 misha 3418:
1.7 ! moko 3419: If you already know that your subject is valid, and you want to skip
! 3420: these checks for performance reasons, you can set the
! 3421: PCRE_NO_UTF8_CHECK option when calling pcre_exec(). You might want to
! 3422: do this for the second and subsequent calls to pcre_exec() if you are
! 3423: making repeated calls to find all the matches in a single subject
! 3424: string. However, you should be sure that the value of startoffset
! 3425: points to the start of a character (or the end of the subject). When
1.5 misha 3426: PCRE_NO_UTF8_CHECK is set, the effect of passing an invalid string as a
1.7 ! moko 3427: subject or an invalid value of startoffset is undefined. Your program
! 3428: may crash or loop.
1.1 misha 3429:
1.4 misha 3430: PCRE_PARTIAL_HARD
3431: PCRE_PARTIAL_SOFT
3432:
1.7 ! moko 3433: These options turn on the partial matching feature. For backwards com-
! 3434: patibility, PCRE_PARTIAL is a synonym for PCRE_PARTIAL_SOFT. A partial
! 3435: match occurs if the end of the subject string is reached successfully,
! 3436: but there are not enough subject characters to complete the match. If
1.4 misha 3437: this happens when PCRE_PARTIAL_SOFT (but not PCRE_PARTIAL_HARD) is set,
1.7 ! moko 3438: matching continues by testing any remaining alternatives. Only if no
! 3439: complete match can be found is PCRE_ERROR_PARTIAL returned instead of
! 3440: PCRE_ERROR_NOMATCH. In other words, PCRE_PARTIAL_SOFT says that the
! 3441: caller is prepared to handle a partial match, but only if no complete
1.4 misha 3442: match can be found.
3443:
1.7 ! moko 3444: If PCRE_PARTIAL_HARD is set, it overrides PCRE_PARTIAL_SOFT. In this
! 3445: case, if a partial match is found, pcre_exec() immediately returns
! 3446: PCRE_ERROR_PARTIAL, without considering any other alternatives. In
! 3447: other words, when PCRE_PARTIAL_HARD is set, a partial match is consid-
1.4 misha 3448: ered to be more important that an alternative complete match.
3449:
1.7 ! moko 3450: In both cases, the portion of the string that was inspected when the
1.4 misha 3451: partial match was found is set as the first matching string. There is a
1.7 ! moko 3452: more detailed discussion of partial and multi-segment matching, with
1.4 misha 3453: examples, in the pcrepartial documentation.
1.1 misha 3454:
3455: The string to be matched by pcre_exec()
3456:
1.7 ! moko 3457: The subject string is passed to pcre_exec() as a pointer in subject, a
! 3458: length in length, and a starting offset in startoffset. The units for
! 3459: length and startoffset are bytes for the 8-bit library, 16-bit data
! 3460: items for the 16-bit library, and 32-bit data items for the 32-bit
1.6 misha 3461: library.
3462:
1.7 ! moko 3463: If startoffset is negative or greater than the length of the subject,
! 3464: pcre_exec() returns PCRE_ERROR_BADOFFSET. When the starting offset is
! 3465: zero, the search for a match starts at the beginning of the subject,
! 3466: and this is by far the most common case. In UTF-8 or UTF-16 mode, the
! 3467: offset must point to the start of a character, or the end of the sub-
! 3468: ject (in UTF-32 mode, one data unit equals one character, so all off-
! 3469: sets are valid). Unlike the pattern string, the subject may contain
1.6 misha 3470: binary zeroes.
1.2 misha 3471:
1.7 ! moko 3472: A non-zero starting offset is useful when searching for another match
! 3473: in the same subject by calling pcre_exec() again after a previous suc-
! 3474: cess. Setting startoffset differs from just passing over a shortened
! 3475: string and setting PCRE_NOTBOL in the case of a pattern that begins
1.1 misha 3476: with any kind of lookbehind. For example, consider the pattern
3477:
3478: \Biss\B
3479:
1.7 ! moko 3480: which finds occurrences of "iss" in the middle of words. (\B matches
! 3481: only if the current position in the subject is not a word boundary.)
! 3482: When applied to the string "Mississipi" the first call to pcre_exec()
! 3483: finds the first occurrence. If pcre_exec() is called again with just
! 3484: the remainder of the subject, namely "issipi", it does not match,
1.1 misha 3485: because \B is always false at the start of the subject, which is deemed
1.7 ! moko 3486: to be a word boundary. However, if pcre_exec() is passed the entire
1.1 misha 3487: string again, but with startoffset set to 4, it finds the second occur-
1.7 ! moko 3488: rence of "iss" because it is able to look behind the starting point to
1.1 misha 3489: discover that it is preceded by a letter.
3490:
1.7 ! moko 3491: Finding all the matches in a subject is tricky when the pattern can
1.4 misha 3492: match an empty string. It is possible to emulate Perl's /g behaviour by
1.7 ! moko 3493: first trying the match again at the same offset, with the
! 3494: PCRE_NOTEMPTY_ATSTART and PCRE_ANCHORED options, and then if that
! 3495: fails, advancing the starting offset and trying an ordinary match
1.4 misha 3496: again. There is some code that demonstrates how to do this in the pcre-
3497: demo sample program. In the most general case, you have to check to see
1.7 ! moko 3498: if the newline convention recognizes CRLF as a newline, and if so, and
1.4 misha 3499: the current character is CR followed by LF, advance the starting offset
3500: by two characters instead of one.
3501:
1.7 ! moko 3502: If a non-zero starting offset is passed when the pattern is anchored,
1.1 misha 3503: one attempt to match at the given offset is made. This can only succeed
1.7 ! moko 3504: if the pattern does not require the match to be at the start of the
1.1 misha 3505: subject.
3506:
3507: How pcre_exec() returns captured substrings
3508:
1.7 ! moko 3509: In general, a pattern matches a certain portion of the subject, and in
! 3510: addition, further substrings from the subject may be picked out by
! 3511: parts of the pattern. Following the usage in Jeffrey Friedl's book,
! 3512: this is called "capturing" in what follows, and the phrase "capturing
! 3513: subpattern" is used for a fragment of a pattern that picks out a sub-
! 3514: string. PCRE supports several other kinds of parenthesized subpattern
1.1 misha 3515: that do not cause substrings to be captured.
3516:
1.2 misha 3517: Captured substrings are returned to the caller via a vector of integers
1.7 ! moko 3518: whose address is passed in ovector. The number of elements in the vec-
! 3519: tor is passed in ovecsize, which must be a non-negative number. Note:
1.2 misha 3520: this argument is NOT the size of ovector in bytes.
3521:
1.7 ! moko 3522: The first two-thirds of the vector is used to pass back captured sub-
! 3523: strings, each substring using a pair of integers. The remaining third
! 3524: of the vector is used as workspace by pcre_exec() while matching cap-
! 3525: turing subpatterns, and is not available for passing back information.
! 3526: The number passed in ovecsize should always be a multiple of three. If
1.1 misha 3527: it is not, it is rounded down.
3528:
1.7 ! moko 3529: When a match is successful, information about captured substrings is
! 3530: returned in pairs of integers, starting at the beginning of ovector,
! 3531: and continuing up to two-thirds of its length at the most. The first
! 3532: element of each pair is set to the offset of the first character in a
! 3533: substring, and the second is set to the offset of the first character
! 3534: after the end of a substring. These values are always data unit off-
! 3535: sets, even in UTF mode. They are byte offsets in the 8-bit library,
! 3536: 16-bit data item offsets in the 16-bit library, and 32-bit data item
1.6 misha 3537: offsets in the 32-bit library. Note: they are not character counts.
1.2 misha 3538:
1.7 ! moko 3539: The first pair of integers, ovector[0] and ovector[1], identify the
! 3540: portion of the subject string matched by the entire pattern. The next
! 3541: pair is used for the first capturing subpattern, and so on. The value
1.2 misha 3542: returned by pcre_exec() is one more than the highest numbered pair that
1.7 ! moko 3543: has been set. For example, if two substrings have been captured, the
! 3544: returned value is 3. If there are no capturing subpatterns, the return
1.2 misha 3545: value from a successful match is 1, indicating that just the first pair
3546: of offsets has been set.
1.1 misha 3547:
3548: If a capturing subpattern is matched repeatedly, it is the last portion
3549: of the string that it matched that is returned.
3550:
1.7 ! moko 3551: If the vector is too small to hold all the captured substring offsets,
1.1 misha 3552: it is used as far as possible (up to two-thirds of its length), and the
1.7 ! moko 3553: function returns a value of zero. If neither the actual string matched
! 3554: nor any captured substrings are of interest, pcre_exec() may be called
! 3555: with ovector passed as NULL and ovecsize as zero. However, if the pat-
! 3556: tern contains back references and the ovector is not big enough to
! 3557: remember the related substrings, PCRE has to get additional memory for
! 3558: use during matching. Thus it is usually advisable to supply an ovector
1.5 misha 3559: of reasonable size.
3560:
1.7 ! moko 3561: There are some cases where zero is returned (indicating vector over-
! 3562: flow) when in fact the vector is exactly the right size for the final
1.5 misha 3563: match. For example, consider the pattern
3564:
3565: (a)(?:(b)c|bd)
3566:
1.7 ! moko 3567: If a vector of 6 elements (allowing for only 1 captured substring) is
1.5 misha 3568: given with subject string "abd", pcre_exec() will try to set the second
3569: captured string, thereby recording a vector overflow, before failing to
1.7 ! moko 3570: match "c" and backing up to try the second alternative. The zero
! 3571: return, however, does correctly indicate that the maximum number of
1.5 misha 3572: slots (namely 2) have been filled. In similar cases where there is tem-
1.7 ! moko 3573: porary overflow, but the final number of used slots is actually less
1.5 misha 3574: than the maximum, a non-zero value is returned.
1.1 misha 3575:
1.4 misha 3576: The pcre_fullinfo() function can be used to find out how many capturing
1.7 ! moko 3577: subpatterns there are in a compiled pattern. The smallest size for
! 3578: ovector that will allow for n captured substrings, in addition to the
1.1 misha 3579: offsets of the substring matched by the whole pattern, is (n+1)*3.
3580:
1.7 ! moko 3581: It is possible for capturing subpattern number n+1 to match some part
1.1 misha 3582: of the subject when subpattern n has not been used at all. For example,
1.7 ! moko 3583: if the string "abc" is matched against the pattern (a|(z))(bc) the
1.1 misha 3584: return from the function is 4, and subpatterns 1 and 3 are matched, but
1.7 ! moko 3585: 2 is not. When this happens, both values in the offset pairs corre-
1.1 misha 3586: sponding to unused subpatterns are set to -1.
3587:
1.7 ! moko 3588: Offset values that correspond to unused subpatterns at the end of the
! 3589: expression are also set to -1. For example, if the string "abc" is
! 3590: matched against the pattern (abc)(x(yz)?)? subpatterns 2 and 3 are not
! 3591: matched. The return from the function is 2, because the highest used
! 3592: capturing subpattern number is 1, and the offsets for for the second
! 3593: and third capturing subpatterns (assuming the vector is large enough,
1.4 misha 3594: of course) are set to -1.
3595:
1.7 ! moko 3596: Note: Elements in the first two-thirds of ovector that do not corre-
! 3597: spond to capturing parentheses in the pattern are never changed. That
! 3598: is, if a pattern contains n capturing parentheses, no more than ovec-
! 3599: tor[0] to ovector[2n+1] are set by pcre_exec(). The other elements (in
1.5 misha 3600: the first two-thirds) retain whatever values they previously had.
1.1 misha 3601:
1.7 ! moko 3602: Some convenience functions are provided for extracting the captured
1.1 misha 3603: substrings as separate strings. These are described below.
3604:
3605: Error return values from pcre_exec()
3606:
1.7 ! moko 3607: If pcre_exec() fails, it returns a negative number. The following are
1.1 misha 3608: defined in the header file:
3609:
3610: PCRE_ERROR_NOMATCH (-1)
3611:
3612: The subject string did not match the pattern.
3613:
3614: PCRE_ERROR_NULL (-2)
3615:
1.7 ! moko 3616: Either code or subject was passed as NULL, or ovector was NULL and
1.1 misha 3617: ovecsize was not zero.
3618:
3619: PCRE_ERROR_BADOPTION (-3)
3620:
3621: An unrecognized bit was set in the options argument.
3622:
3623: PCRE_ERROR_BADMAGIC (-4)
3624:
1.7 ! moko 3625: PCRE stores a 4-byte "magic number" at the start of the compiled code,
1.1 misha 3626: to catch the case when it is passed a junk pointer and to detect when a
3627: pattern that was compiled in an environment of one endianness is run in
1.7 ! moko 3628: an environment with the other endianness. This is the error that PCRE
1.1 misha 3629: gives when the magic number is not present.
3630:
3631: PCRE_ERROR_UNKNOWN_OPCODE (-5)
3632:
3633: While running the pattern match, an unknown item was encountered in the
1.7 ! moko 3634: compiled pattern. This error could be caused by a bug in PCRE or by
1.1 misha 3635: overwriting of the compiled pattern.
3636:
3637: PCRE_ERROR_NOMEMORY (-6)
3638:
1.7 ! moko 3639: If a pattern contains back references, but the ovector that is passed
1.1 misha 3640: to pcre_exec() is not big enough to remember the referenced substrings,
1.7 ! moko 3641: PCRE gets a block of memory at the start of matching to use for this
! 3642: purpose. If the call via pcre_malloc() fails, this error is given. The
1.1 misha 3643: memory is automatically freed at the end of matching.
3644:
1.7 ! moko 3645: This error is also given if pcre_stack_malloc() fails in pcre_exec().
! 3646: This can happen only when PCRE has been compiled with --disable-stack-
1.4 misha 3647: for-recursion.
3648:
1.1 misha 3649: PCRE_ERROR_NOSUBSTRING (-7)
3650:
1.7 ! moko 3651: This error is used by the pcre_copy_substring(), pcre_get_substring(),
1.1 misha 3652: and pcre_get_substring_list() functions (see below). It is never
3653: returned by pcre_exec().
3654:
3655: PCRE_ERROR_MATCHLIMIT (-8)
3656:
1.7 ! moko 3657: The backtracking limit, as specified by the match_limit field in a
! 3658: pcre_extra structure (or defaulted) was reached. See the description
1.1 misha 3659: above.
3660:
3661: PCRE_ERROR_CALLOUT (-9)
3662:
3663: This error is never generated by pcre_exec() itself. It is provided for
1.7 ! moko 3664: use by callout functions that want to yield a distinctive error code.
1.1 misha 3665: See the pcrecallout documentation for details.
3666:
3667: PCRE_ERROR_BADUTF8 (-10)
3668:
1.7 ! moko 3669: A string that contains an invalid UTF-8 byte sequence was passed as a
! 3670: subject, and the PCRE_NO_UTF8_CHECK option was not set. If the size of
! 3671: the output vector (ovecsize) is at least 2, the byte offset to the
! 3672: start of the the invalid UTF-8 character is placed in the first ele-
! 3673: ment, and a reason code is placed in the second element. The reason
1.5 misha 3674: codes are listed in the following section. For backward compatibility,
1.7 ! moko 3675: if PCRE_PARTIAL_HARD is set and the problem is a truncated UTF-8 char-
! 3676: acter at the end of the subject (reason codes 1 to 5),
1.5 misha 3677: PCRE_ERROR_SHORTUTF8 is returned instead of PCRE_ERROR_BADUTF8.
1.1 misha 3678:
3679: PCRE_ERROR_BADUTF8_OFFSET (-11)
3680:
1.7 ! moko 3681: The UTF-8 byte sequence that was passed as a subject was checked and
! 3682: found to be valid (the PCRE_NO_UTF8_CHECK option was not set), but the
! 3683: value of startoffset did not point to the beginning of a UTF-8 charac-
1.4 misha 3684: ter or the end of the subject.
1.1 misha 3685:
3686: PCRE_ERROR_PARTIAL (-12)
3687:
1.7 ! moko 3688: The subject string did not match, but it did match partially. See the
1.1 misha 3689: pcrepartial documentation for details of partial matching.
3690:
3691: PCRE_ERROR_BADPARTIAL (-13)
3692:
1.7 ! moko 3693: This code is no longer in use. It was formerly returned when the
! 3694: PCRE_PARTIAL option was used with a compiled pattern containing items
! 3695: that were not supported for partial matching. From release 8.00
1.4 misha 3696: onwards, there are no restrictions on partial matching.
1.1 misha 3697:
3698: PCRE_ERROR_INTERNAL (-14)
3699:
1.7 ! moko 3700: An unexpected internal error has occurred. This error could be caused
1.1 misha 3701: by a bug in PCRE or by overwriting of the compiled pattern.
3702:
3703: PCRE_ERROR_BADCOUNT (-15)
3704:
1.3 misha 3705: This error is given if the value of the ovecsize argument is negative.
1.1 misha 3706:
3707: PCRE_ERROR_RECURSIONLIMIT (-21)
3708:
3709: The internal recursion limit, as specified by the match_limit_recursion
1.7 ! moko 3710: field in a pcre_extra structure (or defaulted) was reached. See the
1.1 misha 3711: description above.
3712:
3713: PCRE_ERROR_BADNEWLINE (-23)
3714:
3715: An invalid combination of PCRE_NEWLINE_xxx options was given.
3716:
1.4 misha 3717: PCRE_ERROR_BADOFFSET (-24)
3718:
3719: The value of startoffset was negative or greater than the length of the
3720: subject, that is, the value in length.
3721:
3722: PCRE_ERROR_SHORTUTF8 (-25)
3723:
1.7 ! moko 3724: This error is returned instead of PCRE_ERROR_BADUTF8 when the subject
! 3725: string ends with a truncated UTF-8 character and the PCRE_PARTIAL_HARD
! 3726: option is set. Information about the failure is returned as for
! 3727: PCRE_ERROR_BADUTF8. It is in fact sufficient to detect this case, but
! 3728: this special error code for PCRE_PARTIAL_HARD precedes the implementa-
! 3729: tion of returned information; it is retained for backwards compatibil-
1.5 misha 3730: ity.
3731:
3732: PCRE_ERROR_RECURSELOOP (-26)
3733:
3734: This error is returned when pcre_exec() detects a recursion loop within
1.7 ! moko 3735: the pattern. Specifically, it means that either the whole pattern or a
! 3736: subpattern has been called recursively for the second time at the same
1.5 misha 3737: position in the subject string. Some simple patterns that might do this
1.7 ! moko 3738: are detected and faulted at compile time, but more complicated cases,
1.5 misha 3739: in particular mutual recursions between two different subpatterns, can-
3740: not be detected until run time.
3741:
3742: PCRE_ERROR_JIT_STACKLIMIT (-27)
3743:
1.7 ! moko 3744: This error is returned when a pattern that was successfully studied
! 3745: using a JIT compile option is being matched, but the memory available
! 3746: for the just-in-time processing stack is not large enough. See the
1.6 misha 3747: pcrejit documentation for more details.
1.5 misha 3748:
1.6 misha 3749: PCRE_ERROR_BADMODE (-28)
1.5 misha 3750:
3751: This error is given if a pattern that was compiled by the 8-bit library
1.6 misha 3752: is passed to a 16-bit or 32-bit library function, or vice versa.
1.5 misha 3753:
1.6 misha 3754: PCRE_ERROR_BADENDIANNESS (-29)
1.5 misha 3755:
1.7 ! moko 3756: This error is given if a pattern that was compiled and saved is
! 3757: reloaded on a host with different endianness. The utility function
1.5 misha 3758: pcre_pattern_to_host_byte_order() can be used to convert such a pattern
3759: so that it runs on the new host.
1.4 misha 3760:
1.6 misha 3761: PCRE_ERROR_JIT_BADOPTION
3762:
1.7 ! moko 3763: This error is returned when a pattern that was successfully studied
! 3764: using a JIT compile option is being matched, but the matching mode
! 3765: (partial or complete match) does not correspond to any JIT compilation
! 3766: mode. When the JIT fast path function is used, this error may be also
! 3767: given for invalid options. See the pcrejit documentation for more
1.6 misha 3768: details.
3769:
3770: PCRE_ERROR_BADLENGTH (-32)
3771:
1.7 ! moko 3772: This error is given if pcre_exec() is called with a negative value for
1.6 misha 3773: the length argument.
3774:
3775: Error numbers -16 to -20, -22, and 30 are not used by pcre_exec().
1.1 misha 3776:
1.5 misha 3777: Reason codes for invalid UTF-8 strings
3778:
1.7 ! moko 3779: This section applies only to the 8-bit library. The corresponding
! 3780: information for the 16-bit and 32-bit libraries is given in the pcre16
1.6 misha 3781: and pcre32 pages.
1.5 misha 3782:
3783: When pcre_exec() returns either PCRE_ERROR_BADUTF8 or PCRE_ERROR_SHORT-
1.7 ! moko 3784: UTF8, and the size of the output vector (ovecsize) is at least 2, the
! 3785: offset of the start of the invalid UTF-8 character is placed in the
1.5 misha 3786: first output vector element (ovector[0]) and a reason code is placed in
1.7 ! moko 3787: the second element (ovector[1]). The reason codes are given names in
1.5 misha 3788: the pcre.h header file:
3789:
3790: PCRE_UTF8_ERR1
3791: PCRE_UTF8_ERR2
3792: PCRE_UTF8_ERR3
3793: PCRE_UTF8_ERR4
3794: PCRE_UTF8_ERR5
3795:
1.7 ! moko 3796: The string ends with a truncated UTF-8 character; the code specifies
! 3797: how many bytes are missing (1 to 5). Although RFC 3629 restricts UTF-8
! 3798: characters to be no longer than 4 bytes, the encoding scheme (origi-
! 3799: nally defined by RFC 2279) allows for up to 6 bytes, and this is
1.5 misha 3800: checked first; hence the possibility of 4 or 5 missing bytes.
3801:
3802: PCRE_UTF8_ERR6
3803: PCRE_UTF8_ERR7
3804: PCRE_UTF8_ERR8
3805: PCRE_UTF8_ERR9
3806: PCRE_UTF8_ERR10
3807:
3808: The two most significant bits of the 2nd, 3rd, 4th, 5th, or 6th byte of
1.7 ! moko 3809: the character do not have the binary value 0b10 (that is, either the
1.5 misha 3810: most significant bit is 0, or the next bit is 1).
3811:
3812: PCRE_UTF8_ERR11
3813: PCRE_UTF8_ERR12
3814:
1.7 ! moko 3815: A character that is valid by the RFC 2279 rules is either 5 or 6 bytes
1.5 misha 3816: long; these code points are excluded by RFC 3629.
3817:
3818: PCRE_UTF8_ERR13
3819:
1.7 ! moko 3820: A 4-byte character has a value greater than 0x10fff; these code points
1.5 misha 3821: are excluded by RFC 3629.
3822:
3823: PCRE_UTF8_ERR14
3824:
1.7 ! moko 3825: A 3-byte character has a value in the range 0xd800 to 0xdfff; this
! 3826: range of code points are reserved by RFC 3629 for use with UTF-16, and
1.5 misha 3827: so are excluded from UTF-8.
3828:
3829: PCRE_UTF8_ERR15
3830: PCRE_UTF8_ERR16
3831: PCRE_UTF8_ERR17
3832: PCRE_UTF8_ERR18
3833: PCRE_UTF8_ERR19
3834:
1.7 ! moko 3835: A 2-, 3-, 4-, 5-, or 6-byte character is "overlong", that is, it codes
! 3836: for a value that can be represented by fewer bytes, which is invalid.
! 3837: For example, the two bytes 0xc0, 0xae give the value 0x2e, whose cor-
1.5 misha 3838: rect coding uses just one byte.
3839:
3840: PCRE_UTF8_ERR20
3841:
3842: The two most significant bits of the first byte of a character have the
1.7 ! moko 3843: binary value 0b10 (that is, the most significant bit is 1 and the sec-
! 3844: ond is 0). Such a byte can only validly occur as the second or subse-
1.5 misha 3845: quent byte of a multi-byte character.
3846:
3847: PCRE_UTF8_ERR21
3848:
1.7 ! moko 3849: The first byte of a character has the value 0xfe or 0xff. These values
1.5 misha 3850: can never occur in a valid UTF-8 string.
3851:
1.6 misha 3852: PCRE_UTF8_ERR22
3853:
1.7 ! moko 3854: This error code was formerly used when the presence of a so-called
! 3855: "non-character" caused an error. Unicode corrigendum #9 makes it clear
! 3856: that such characters should not cause a string to be rejected, and so
1.6 misha 3857: this code is no longer in use and is never returned.
3858:
1.1 misha 3859:
3860: EXTRACTING CAPTURED SUBSTRINGS BY NUMBER
3861:
3862: int pcre_copy_substring(const char *subject, int *ovector,
3863: int stringcount, int stringnumber, char *buffer,
3864: int buffersize);
3865:
3866: int pcre_get_substring(const char *subject, int *ovector,
3867: int stringcount, int stringnumber,
3868: const char **stringptr);
3869:
3870: int pcre_get_substring_list(const char *subject,
3871: int *ovector, int stringcount, const char ***listptr);
3872:
1.7 ! moko 3873: Captured substrings can be accessed directly by using the offsets
! 3874: returned by pcre_exec() in ovector. For convenience, the functions
1.1 misha 3875: pcre_copy_substring(), pcre_get_substring(), and pcre_get_sub-
1.7 ! moko 3876: string_list() are provided for extracting captured substrings as new,
! 3877: separate, zero-terminated strings. These functions identify substrings
! 3878: by number. The next section describes functions for extracting named
1.1 misha 3879: substrings.
3880:
1.7 ! moko 3881: A substring that contains a binary zero is correctly extracted and has
! 3882: a further zero added on the end, but the result is not, of course, a C
! 3883: string. However, you can process such a string by referring to the
! 3884: length that is returned by pcre_copy_substring() and pcre_get_sub-
1.1 misha 3885: string(). Unfortunately, the interface to pcre_get_substring_list() is
1.7 ! moko 3886: not adequate for handling strings containing binary zeros, because the
1.1 misha 3887: end of the final string is not independently indicated.
3888:
1.7 ! moko 3889: The first three arguments are the same for all three of these func-
! 3890: tions: subject is the subject string that has just been successfully
1.1 misha 3891: matched, ovector is a pointer to the vector of integer offsets that was
3892: passed to pcre_exec(), and stringcount is the number of substrings that
1.7 ! moko 3893: were captured by the match, including the substring that matched the
1.1 misha 3894: entire regular expression. This is the value returned by pcre_exec() if
1.7 ! moko 3895: it is greater than zero. If pcre_exec() returned zero, indicating that
! 3896: it ran out of space in ovector, the value passed as stringcount should
1.1 misha 3897: be the number of elements in the vector divided by three.
3898:
1.7 ! moko 3899: The functions pcre_copy_substring() and pcre_get_substring() extract a
! 3900: single substring, whose number is given as stringnumber. A value of
! 3901: zero extracts the substring that matched the entire pattern, whereas
! 3902: higher values extract the captured substrings. For pcre_copy_sub-
! 3903: string(), the string is placed in buffer, whose length is given by
! 3904: buffersize, while for pcre_get_substring() a new block of memory is
! 3905: obtained via pcre_malloc, and its address is returned via stringptr.
! 3906: The yield of the function is the length of the string, not including
1.1 misha 3907: the terminating zero, or one of these error codes:
3908:
3909: PCRE_ERROR_NOMEMORY (-6)
3910:
1.7 ! moko 3911: The buffer was too small for pcre_copy_substring(), or the attempt to
1.1 misha 3912: get memory failed for pcre_get_substring().
3913:
3914: PCRE_ERROR_NOSUBSTRING (-7)
3915:
3916: There is no substring whose number is stringnumber.
3917:
1.7 ! moko 3918: The pcre_get_substring_list() function extracts all available sub-
! 3919: strings and builds a list of pointers to them. All this is done in a
1.1 misha 3920: single block of memory that is obtained via pcre_malloc. The address of
1.7 ! moko 3921: the memory block is returned via listptr, which is also the start of
! 3922: the list of string pointers. The end of the list is marked by a NULL
! 3923: pointer. The yield of the function is zero if all went well, or the
1.1 misha 3924: error code
3925:
3926: PCRE_ERROR_NOMEMORY (-6)
3927:
3928: if the attempt to get the memory block failed.
3929:
1.7 ! moko 3930: When any of these functions encounter a substring that is unset, which
! 3931: can happen when capturing subpattern number n+1 matches some part of
! 3932: the subject, but subpattern n has not been used at all, they return an
1.1 misha 3933: empty string. This can be distinguished from a genuine zero-length sub-
1.7 ! moko 3934: string by inspecting the appropriate offset in ovector, which is nega-
1.1 misha 3935: tive for unset substrings.
3936:
1.7 ! moko 3937: The two convenience functions pcre_free_substring() and pcre_free_sub-
! 3938: string_list() can be used to free the memory returned by a previous
1.1 misha 3939: call of pcre_get_substring() or pcre_get_substring_list(), respec-
1.7 ! moko 3940: tively. They do nothing more than call the function pointed to by
! 3941: pcre_free, which of course could be called directly from a C program.
! 3942: However, PCRE is used in some situations where it is linked via a spe-
! 3943: cial interface to another programming language that cannot use
! 3944: pcre_free directly; it is for these cases that the functions are pro-
1.1 misha 3945: vided.
3946:
3947:
3948: EXTRACTING CAPTURED SUBSTRINGS BY NAME
3949:
3950: int pcre_get_stringnumber(const pcre *code,
3951: const char *name);
3952:
3953: int pcre_copy_named_substring(const pcre *code,
3954: const char *subject, int *ovector,
3955: int stringcount, const char *stringname,
3956: char *buffer, int buffersize);
3957:
3958: int pcre_get_named_substring(const pcre *code,
3959: const char *subject, int *ovector,
3960: int stringcount, const char *stringname,
3961: const char **stringptr);
3962:
1.7 ! moko 3963: To extract a substring by name, you first have to find associated num-
1.1 misha 3964: ber. For example, for this pattern
3965:
3966: (a+)b(?<xxx>\d+)...
3967:
3968: the number of the subpattern called "xxx" is 2. If the name is known to
3969: be unique (PCRE_DUPNAMES was not set), you can find the number from the
3970: name by calling pcre_get_stringnumber(). The first argument is the com-
3971: piled pattern, and the second is the name. The yield of the function is
1.7 ! moko 3972: the subpattern number, or PCRE_ERROR_NOSUBSTRING (-7) if there is no
1.1 misha 3973: subpattern of that name.
3974:
3975: Given the number, you can extract the substring directly, or use one of
3976: the functions described in the previous section. For convenience, there
3977: are also two functions that do the whole job.
3978:
1.7 ! moko 3979: Most of the arguments of pcre_copy_named_substring() and
! 3980: pcre_get_named_substring() are the same as those for the similarly
! 3981: named functions that extract by number. As these are described in the
! 3982: previous section, they are not re-described here. There are just two
1.1 misha 3983: differences:
3984:
1.7 ! moko 3985: First, instead of a substring number, a substring name is given. Sec-
1.1 misha 3986: ond, there is an extra argument, given at the start, which is a pointer
1.7 ! moko 3987: to the compiled pattern. This is needed in order to gain access to the
1.1 misha 3988: name-to-number translation table.
3989:
1.7 ! moko 3990: These functions call pcre_get_stringnumber(), and if it succeeds, they
! 3991: then call pcre_copy_substring() or pcre_get_substring(), as appropri-
! 3992: ate. NOTE: If PCRE_DUPNAMES is set and there are duplicate names, the
1.1 misha 3993: behaviour may not be what you want (see the next section).
3994:
1.4 misha 3995: Warning: If the pattern uses the (?| feature to set up multiple subpat-
1.7 ! moko 3996: terns with the same number, as described in the section on duplicate
! 3997: subpattern numbers in the pcrepattern page, you cannot use names to
! 3998: distinguish the different subpatterns, because names are not included
! 3999: in the compiled code. The matching process uses only numbers. For this
! 4000: reason, the use of different names for subpatterns of the same number
1.4 misha 4001: causes an error at compile time.
1.3 misha 4002:
1.1 misha 4003:
4004: DUPLICATE SUBPATTERN NAMES
4005:
4006: int pcre_get_stringtable_entries(const pcre *code,
4007: const char *name, char **first, char **last);
4008:
1.7 ! moko 4009: When a pattern is compiled with the PCRE_DUPNAMES option, names for
! 4010: subpatterns are not required to be unique. (Duplicate names are always
! 4011: allowed for subpatterns with the same number, created by using the (?|
! 4012: feature. Indeed, if such subpatterns are named, they are required to
1.4 misha 4013: use the same names.)
4014:
4015: Normally, patterns with duplicate names are such that in any one match,
1.7 ! moko 4016: only one of the named subpatterns participates. An example is shown in
1.4 misha 4017: the pcrepattern documentation.
1.1 misha 4018:
1.7 ! moko 4019: When duplicates are present, pcre_copy_named_substring() and
! 4020: pcre_get_named_substring() return the first substring corresponding to
! 4021: the given name that is set. If none are set, PCRE_ERROR_NOSUBSTRING
! 4022: (-7) is returned; no data is returned. The pcre_get_stringnumber()
! 4023: function returns one of the numbers that are associated with the name,
1.1 misha 4024: but it is not defined which it is.
4025:
1.7 ! moko 4026: If you want to get full details of all captured substrings for a given
! 4027: name, you must use the pcre_get_stringtable_entries() function. The
1.1 misha 4028: first argument is the compiled pattern, and the second is the name. The
1.7 ! moko 4029: third and fourth are pointers to variables which are updated by the
1.1 misha 4030: function. After it has run, they point to the first and last entries in
1.7 ! moko 4031: the name-to-number table for the given name. The function itself
! 4032: returns the length of each entry, or PCRE_ERROR_NOSUBSTRING (-7) if
! 4033: there are none. The format of the table is described above in the sec-
! 4034: tion entitled Information about a pattern above. Given all the rele-
! 4035: vant entries for the name, you can extract each of their numbers, and
1.5 misha 4036: hence the captured data, if any.
1.1 misha 4037:
4038:
4039: FINDING ALL POSSIBLE MATCHES
4040:
1.7 ! moko 4041: The traditional matching function uses a similar algorithm to Perl,
1.1 misha 4042: which stops when it finds the first match, starting at a given point in
1.7 ! moko 4043: the subject. If you want to find all possible matches, or the longest
! 4044: possible match, consider using the alternative matching function (see
! 4045: below) instead. If you cannot use the alternative function, but still
! 4046: need to find all possible matches, you can kludge it up by making use
1.1 misha 4047: of the callout facility, which is described in the pcrecallout documen-
4048: tation.
4049:
4050: What you have to do is to insert a callout right at the end of the pat-
1.7 ! moko 4051: tern. When your callout function is called, extract and save the cur-
! 4052: rent matched substring. Then return 1, which forces pcre_exec() to
! 4053: backtrack and try other alternatives. Ultimately, when it runs out of
1.1 misha 4054: matches, pcre_exec() will yield PCRE_ERROR_NOMATCH.
4055:
4056:
1.5 misha 4057: OBTAINING AN ESTIMATE OF STACK USAGE
4058:
1.7 ! moko 4059: Matching certain patterns using pcre_exec() can use a lot of process
! 4060: stack, which in certain environments can be rather limited in size.
! 4061: Some users find it helpful to have an estimate of the amount of stack
! 4062: that is used by pcre_exec(), to help them set recursion limits, as
! 4063: described in the pcrestack documentation. The estimate that is output
1.5 misha 4064: by pcretest when called with the -m and -C options is obtained by call-
1.7 ! moko 4065: ing pcre_exec with the values NULL, NULL, NULL, -999, and -999 for its
1.5 misha 4066: first five arguments.
4067:
1.7 ! moko 4068: Normally, if its first argument is NULL, pcre_exec() immediately
! 4069: returns the negative error code PCRE_ERROR_NULL, but with this special
! 4070: combination of arguments, it returns instead a negative number whose
! 4071: absolute value is the approximate stack frame size in bytes. (A nega-
! 4072: tive number is used so that it is clear that no match has happened.)
! 4073: The value is approximate because in some cases, recursive calls to
1.5 misha 4074: pcre_exec() occur when there are one or two additional variables on the
4075: stack.
4076:
1.7 ! moko 4077: If PCRE has been compiled to use the heap instead of the stack for
! 4078: recursion, the value returned is the size of each block that is
1.5 misha 4079: obtained from the heap.
4080:
4081:
1.1 misha 4082: MATCHING A PATTERN: THE ALTERNATIVE FUNCTION
4083:
4084: int pcre_dfa_exec(const pcre *code, const pcre_extra *extra,
4085: const char *subject, int length, int startoffset,
4086: int options, int *ovector, int ovecsize,
4087: int *workspace, int wscount);
4088:
1.7 ! moko 4089: The function pcre_dfa_exec() is called to match a subject string
! 4090: against a compiled pattern, using a matching algorithm that scans the
! 4091: subject string just once, and does not backtrack. This has different
! 4092: characteristics to the normal algorithm, and is not compatible with
! 4093: Perl. Some of the features of PCRE patterns are not supported. Never-
! 4094: theless, there are times when this kind of matching can be useful. For
! 4095: a discussion of the two matching algorithms, and a list of features
! 4096: that pcre_dfa_exec() does not support, see the pcrematching documenta-
1.4 misha 4097: tion.
1.1 misha 4098:
1.7 ! moko 4099: The arguments for the pcre_dfa_exec() function are the same as for
1.1 misha 4100: pcre_exec(), plus two extras. The ovector argument is used in a differ-
1.7 ! moko 4101: ent way, and this is described below. The other common arguments are
! 4102: used in the same way as for pcre_exec(), so their description is not
1.1 misha 4103: repeated here.
4104:
1.7 ! moko 4105: The two additional arguments provide workspace for the function. The
! 4106: workspace vector should contain at least 20 elements. It is used for
1.1 misha 4107: keeping track of multiple paths through the pattern tree. More
1.7 ! moko 4108: workspace will be needed for patterns and subjects where there are a
1.1 misha 4109: lot of potential matches.
4110:
4111: Here is an example of a simple call to pcre_dfa_exec():
4112:
4113: int rc;
4114: int ovector[10];
4115: int wspace[20];
4116: rc = pcre_dfa_exec(
4117: re, /* result of pcre_compile() */
4118: NULL, /* we didn't study the pattern */
4119: "some string", /* the subject string */
4120: 11, /* the length of the subject string */
4121: 0, /* start at offset 0 in the subject */
4122: 0, /* default options */
4123: ovector, /* vector of integers for substring information */
4124: 10, /* number of elements (NOT size in bytes) */
4125: wspace, /* working space vector */
4126: 20); /* number of elements (NOT size in bytes) */
4127:
4128: Option bits for pcre_dfa_exec()
4129:
1.7 ! moko 4130: The unused bits of the options argument for pcre_dfa_exec() must be
! 4131: zero. The only bits that may be set are PCRE_ANCHORED, PCRE_NEW-
1.4 misha 4132: LINE_xxx, PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY,
1.7 ! moko 4133: PCRE_NOTEMPTY_ATSTART, PCRE_NO_UTF8_CHECK, PCRE_BSR_ANYCRLF,
! 4134: PCRE_BSR_UNICODE, PCRE_NO_START_OPTIMIZE, PCRE_PARTIAL_HARD, PCRE_PAR-
! 4135: TIAL_SOFT, PCRE_DFA_SHORTEST, and PCRE_DFA_RESTART. All but the last
! 4136: four of these are exactly the same as for pcre_exec(), so their
1.4 misha 4137: description is not repeated here.
4138:
4139: PCRE_PARTIAL_HARD
4140: PCRE_PARTIAL_SOFT
4141:
1.7 ! moko 4142: These have the same general effect as they do for pcre_exec(), but the
! 4143: details are slightly different. When PCRE_PARTIAL_HARD is set for
! 4144: pcre_dfa_exec(), it returns PCRE_ERROR_PARTIAL if the end of the sub-
! 4145: ject is reached and there is still at least one matching possibility
1.4 misha 4146: that requires additional characters. This happens even if some complete
4147: matches have also been found. When PCRE_PARTIAL_SOFT is set, the return
4148: code PCRE_ERROR_NOMATCH is converted into PCRE_ERROR_PARTIAL if the end
1.7 ! moko 4149: of the subject is reached, there have been no complete matches, but
! 4150: there is still at least one matching possibility. The portion of the
! 4151: string that was inspected when the longest partial match was found is
! 4152: set as the first matching string in both cases. There is a more
! 4153: detailed discussion of partial and multi-segment matching, with exam-
1.4 misha 4154: ples, in the pcrepartial documentation.
1.1 misha 4155:
4156: PCRE_DFA_SHORTEST
4157:
1.7 ! moko 4158: Setting the PCRE_DFA_SHORTEST option causes the matching algorithm to
1.1 misha 4159: stop as soon as it has found one match. Because of the way the alterna-
1.7 ! moko 4160: tive algorithm works, this is necessarily the shortest possible match
1.1 misha 4161: at the first possible matching point in the subject string.
4162:
4163: PCRE_DFA_RESTART
4164:
1.4 misha 4165: When pcre_dfa_exec() returns a partial match, it is possible to call it
1.7 ! moko 4166: again, with additional subject characters, and have it continue with
! 4167: the same match. The PCRE_DFA_RESTART option requests this action; when
! 4168: it is set, the workspace and wscount options must reference the same
! 4169: vector as before because data about the match so far is left in them
1.4 misha 4170: after a partial match. There is more discussion of this facility in the
4171: pcrepartial documentation.
1.1 misha 4172:
4173: Successful returns from pcre_dfa_exec()
4174:
1.7 ! moko 4175: When pcre_dfa_exec() succeeds, it may have matched more than one sub-
1.1 misha 4176: string in the subject. Note, however, that all the matches from one run
1.7 ! moko 4177: of the function start at the same point in the subject. The shorter
! 4178: matches are all initial substrings of the longer matches. For example,
1.1 misha 4179: if the pattern
4180:
4181: <.*>
4182:
4183: is matched against the string
4184:
4185: This is <something> <something else> <something further> no more
4186:
4187: the three matched strings are
4188:
4189: <something>
4190: <something> <something else>
4191: <something> <something else> <something further>
4192:
1.7 ! moko 4193: On success, the yield of the function is a number greater than zero,
! 4194: which is the number of matched substrings. The substrings themselves
! 4195: are returned in ovector. Each string uses two elements; the first is
! 4196: the offset to the start, and the second is the offset to the end. In
! 4197: fact, all the strings have the same start offset. (Space could have
! 4198: been saved by giving this only once, but it was decided to retain some
! 4199: compatibility with the way pcre_exec() returns data, even though the
1.1 misha 4200: meaning of the strings is different.)
4201:
4202: The strings are returned in reverse order of length; that is, the long-
1.7 ! moko 4203: est matching string is given first. If there were too many matches to
! 4204: fit into ovector, the yield of the function is zero, and the vector is
! 4205: filled with the longest matches. Unlike pcre_exec(), pcre_dfa_exec()
1.5 misha 4206: can use the entire ovector for returning matched strings.
1.1 misha 4207:
1.7 ! moko 4208: NOTE: PCRE's "auto-possessification" optimization usually applies to
! 4209: character repeats at the end of a pattern (as well as internally). For
! 4210: example, the pattern "a\d+" is compiled as if it were "a\d++" because
! 4211: there is no point even considering the possibility of backtracking into
! 4212: the repeated digits. For DFA matching, this means that only one possi-
! 4213: ble match is found. If you really do want multiple matches in such
! 4214: cases, either use an ungreedy repeat ("a\d+?") or set the
! 4215: PCRE_NO_AUTO_POSSESS option when compiling.
! 4216:
1.1 misha 4217: Error returns from pcre_dfa_exec()
4218:
1.5 misha 4219: The pcre_dfa_exec() function returns a negative number when it fails.
4220: Many of the errors are the same as for pcre_exec(), and these are
4221: described above. There are in addition the following errors that are
1.1 misha 4222: specific to pcre_dfa_exec():
4223:
4224: PCRE_ERROR_DFA_UITEM (-16)
4225:
1.5 misha 4226: This return is given if pcre_dfa_exec() encounters an item in the pat-
4227: tern that it does not support, for instance, the use of \C or a back
1.1 misha 4228: reference.
4229:
4230: PCRE_ERROR_DFA_UCOND (-17)
4231:
1.5 misha 4232: This return is given if pcre_dfa_exec() encounters a condition item
4233: that uses a back reference for the condition, or a test for recursion
1.1 misha 4234: in a specific group. These are not supported.
4235:
4236: PCRE_ERROR_DFA_UMLIMIT (-18)
4237:
1.5 misha 4238: This return is given if pcre_dfa_exec() is called with an extra block
4239: that contains a setting of the match_limit or match_limit_recursion
4240: fields. This is not supported (these fields are meaningless for DFA
4241: matching).
1.1 misha 4242:
4243: PCRE_ERROR_DFA_WSSIZE (-19)
4244:
4245: This return is given if pcre_dfa_exec() runs out of space in the
4246: workspace vector.
4247:
4248: PCRE_ERROR_DFA_RECURSE (-20)
4249:
4250: When a recursive subpattern is processed, the matching function calls
4251: itself recursively, using private vectors for ovector and workspace.
4252: This error is given if the output vector is not large enough. This
4253: should be extremely rare, as a vector of size 1000 is used.
4254:
1.6 misha 4255: PCRE_ERROR_DFA_BADRESTART (-30)
4256:
4257: When pcre_dfa_exec() is called with the PCRE_DFA_RESTART option, some
4258: plausibility checks are made on the contents of the workspace, which
4259: should contain data about the previous partial match. If any of these
4260: checks fail, this error is given.
4261:
1.1 misha 4262:
4263: SEE ALSO
4264:
1.6 misha 4265: pcre16(3), pcre32(3), pcrebuild(3), pcrecallout(3), pcrecpp(3)(3),
4266: pcrematching(3), pcrepartial(3), pcreposix(3), pcreprecompile(3), pcre-
4267: sample(3), pcrestack(3).
1.1 misha 4268:
4269:
4270: AUTHOR
4271:
4272: Philip Hazel
4273: University Computing Service
4274: Cambridge CB2 3QH, England.
4275:
4276:
4277: REVISION
4278:
1.7 ! moko 4279: Last updated: 09 February 2014
! 4280: Copyright (c) 1997-2014 University of Cambridge.
1.1 misha 4281: ------------------------------------------------------------------------------
4282:
4283:
1.6 misha 4284: PCRECALLOUT(3) Library Functions Manual PCRECALLOUT(3)
4285:
1.1 misha 4286:
4287:
4288: NAME
4289: PCRE - Perl-compatible regular expressions
4290:
1.6 misha 4291: SYNOPSIS
1.1 misha 4292:
1.6 misha 4293: #include <pcre.h>
1.1 misha 4294:
4295: int (*pcre_callout)(pcre_callout_block *);
4296:
1.5 misha 4297: int (*pcre16_callout)(pcre16_callout_block *);
4298:
1.6 misha 4299: int (*pcre32_callout)(pcre32_callout_block *);
4300:
4301:
4302: DESCRIPTION
4303:
1.1 misha 4304: PCRE provides a feature called "callout", which is a means of temporar-
4305: ily passing control to the caller of PCRE in the middle of pattern
4306: matching. The caller of PCRE provides an external function by putting
1.5 misha 4307: its entry point in the global variable pcre_callout (pcre16_callout for
1.6 misha 4308: the 16-bit library, pcre32_callout for the 32-bit library). By default,
4309: this variable contains NULL, which disables all calling out.
1.5 misha 4310:
4311: Within a regular expression, (?C) indicates the points at which the
4312: external function is to be called. Different callout points can be
4313: identified by putting a number less than 256 after the letter C. The
4314: default value is zero. For example, this pattern has two callout
1.1 misha 4315: points:
4316:
4317: (?C1)abc(?C2)def
4318:
1.5 misha 4319: If the PCRE_AUTO_CALLOUT option bit is set when a pattern is compiled,
4320: PCRE automatically inserts callouts, all with number 255, before each
4321: item in the pattern. For example, if PCRE_AUTO_CALLOUT is used with the
4322: pattern
1.1 misha 4323:
4324: A(\d{2}|--)
4325:
4326: it is processed as if it were
4327:
4328: (?C255)A(?C255)((?C255)\d{2}(?C255)|(?C255)-(?C255)-(?C255))(?C255)
4329:
1.5 misha 4330: Notice that there is a callout before and after each parenthesis and
1.6 misha 4331: alternation bar. If the pattern contains a conditional group whose con-
4332: dition is an assertion, an automatic callout is inserted immediately
4333: before the condition. Such a callout may also be inserted explicitly,
4334: for example:
4335:
4336: (?(?C9)(?=a)ab|de)
4337:
4338: This applies only to assertion conditions (because they are themselves
4339: independent groups).
4340:
4341: Automatic callouts can be used for tracking the progress of pattern
1.7 ! moko 4342: matching. The pcretest program has a pattern qualifier (/C) that sets
! 4343: automatic callouts; when it is used, the output indicates how the pat-
! 4344: tern is being matched. This is useful information when you are trying
! 4345: to optimize the performance of a particular pattern.
1.5 misha 4346:
1.1 misha 4347:
4348: MISSING CALLOUTS
4349:
1.7 ! moko 4350: You should be aware that, because of optimizations in the way PCRE com-
! 4351: piles and matches patterns, callouts sometimes do not happen exactly as
! 4352: you might expect.
! 4353:
! 4354: At compile time, PCRE "auto-possessifies" repeated items when it knows
! 4355: that what follows cannot be part of the repeat. For example, a+[bc] is
! 4356: compiled as if it were a++[bc]. The pcretest output when this pattern
! 4357: is anchored and then applied with automatic callouts to the string
! 4358: "aaaa" is:
! 4359:
! 4360: --->aaaa
! 4361: +0 ^ ^
! 4362: +1 ^ a+
! 4363: +3 ^ ^ [bc]
! 4364: No match
! 4365:
! 4366: This indicates that when matching [bc] fails, there is no backtracking
! 4367: into a+ and therefore the callouts that would be taken for the back-
! 4368: tracks do not occur. You can disable the auto-possessify feature by
! 4369: passing PCRE_NO_AUTO_POSSESS to pcre_compile(), or starting the pattern
! 4370: with (*NO_AUTO_POSSESS). If this is done in pcretest (using the /O
! 4371: qualifier), the output changes to this:
! 4372:
! 4373: --->aaaa
! 4374: +0 ^ ^
! 4375: +1 ^ a+
! 4376: +3 ^ ^ [bc]
! 4377: +3 ^ ^ [bc]
! 4378: +3 ^ ^ [bc]
! 4379: +3 ^^ [bc]
! 4380: No match
! 4381:
! 4382: This time, when matching [bc] fails, the matcher backtracks into a+ and
! 4383: tries again, repeatedly, until a+ itself fails.
! 4384:
! 4385: Other optimizations that provide fast "no match" results also affect
! 4386: callouts. For example, if the pattern is
1.1 misha 4387:
4388: ab(?C4)cd
4389:
4390: PCRE knows that any matching string must contain the letter "d". If the
1.7 ! moko 4391: subject string is "abyz", the lack of "d" means that matching doesn't
! 4392: ever start, and the callout is never reached. However, with "abyd",
1.1 misha 4393: though the result is still no match, the callout is obeyed.
4394:
1.7 ! moko 4395: If the pattern is studied, PCRE knows the minimum length of a matching
! 4396: string, and will immediately give a "no match" return without actually
! 4397: running a match if the subject is not long enough, or, for unanchored
1.4 misha 4398: patterns, if it has been scanned far enough.
4399:
1.7 ! moko 4400: You can disable these optimizations by passing the PCRE_NO_START_OPTI-
! 4401: MIZE option to the matching function, or by starting the pattern with
! 4402: (*NO_START_OPT). This slows down the matching process, but does ensure
1.5 misha 4403: that callouts such as the example above are obeyed.
1.3 misha 4404:
1.1 misha 4405:
4406: THE CALLOUT INTERFACE
4407:
1.7 ! moko 4408: During matching, when PCRE reaches a callout point, the external func-
1.6 misha 4409: tion defined by pcre_callout or pcre[16|32]_callout is called (if it is
1.7 ! moko 4410: set). This applies to both normal and DFA matching. The only argument
! 4411: to the callout function is a pointer to a pcre_callout or
! 4412: pcre[16|32]_callout block. These structures contains the following
1.6 misha 4413: fields:
1.5 misha 4414:
4415: int version;
4416: int callout_number;
4417: int *offset_vector;
4418: const char *subject; (8-bit version)
4419: PCRE_SPTR16 subject; (16-bit version)
1.6 misha 4420: PCRE_SPTR32 subject; (32-bit version)
1.5 misha 4421: int subject_length;
4422: int start_match;
4423: int current_position;
4424: int capture_top;
4425: int capture_last;
4426: void *callout_data;
4427: int pattern_position;
4428: int next_item_length;
4429: const unsigned char *mark; (8-bit version)
4430: const PCRE_UCHAR16 *mark; (16-bit version)
1.6 misha 4431: const PCRE_UCHAR32 *mark; (32-bit version)
1.1 misha 4432:
1.7 ! moko 4433: The version field is an integer containing the version number of the
! 4434: block format. The initial version was 0; the current version is 2. The
! 4435: version number will change again in future if additional fields are
1.1 misha 4436: added, but the intention is never to remove any of the existing fields.
4437:
1.7 ! moko 4438: The callout_number field contains the number of the callout, as com-
! 4439: piled into the pattern (that is, the number after ?C for manual call-
1.1 misha 4440: outs, and 255 for automatically generated callouts).
4441:
1.7 ! moko 4442: The offset_vector field is a pointer to the vector of offsets that was
! 4443: passed by the caller to the matching function. When pcre_exec() or
! 4444: pcre[16|32]_exec() is used, the contents can be inspected, in order to
! 4445: extract substrings that have been matched so far, in the same way as
! 4446: for extracting substrings after a match has completed. For the DFA
1.5 misha 4447: matching functions, this field is not useful.
1.1 misha 4448:
4449: The subject and subject_length fields contain copies of the values that
1.5 misha 4450: were passed to the matching function.
1.1 misha 4451:
1.7 ! moko 4452: The start_match field normally contains the offset within the subject
! 4453: at which the current match attempt started. However, if the escape
! 4454: sequence \K has been encountered, this value is changed to reflect the
! 4455: modified starting point. If the pattern is not anchored, the callout
1.1 misha 4456: function may be called several times from the same point in the pattern
4457: for different starting points in the subject.
4458:
1.7 ! moko 4459: The current_position field contains the offset within the subject of
1.1 misha 4460: the current match pointer.
4461:
1.7 ! moko 4462: When the pcre_exec() or pcre[16|32]_exec() is used, the capture_top
! 4463: field contains one more than the number of the highest numbered cap-
! 4464: tured substring so far. If no substrings have been captured, the value
! 4465: of capture_top is one. This is always the case when the DFA functions
1.6 misha 4466: are used, because they do not support captured substrings.
4467:
1.7 ! moko 4468: The capture_last field contains the number of the most recently cap-
! 4469: tured substring. However, when a recursion exits, the value reverts to
! 4470: what it was outside the recursion, as do the values of all captured
! 4471: substrings. If no substrings have been captured, the value of cap-
! 4472: ture_last is -1. This is always the case for the DFA matching func-
1.6 misha 4473: tions.
1.1 misha 4474:
1.7 ! moko 4475: The callout_data field contains a value that is passed to a matching
! 4476: function specifically so that it can be passed back in callouts. It is
! 4477: passed in the callout_data field of a pcre_extra or pcre[16|32]_extra
! 4478: data structure. If no such data was passed, the value of callout_data
! 4479: in a callout block is NULL. There is a description of the pcre_extra
1.6 misha 4480: structure in the pcreapi documentation.
1.1 misha 4481:
1.7 ! moko 4482: The pattern_position field is present from version 1 of the callout
1.5 misha 4483: structure. It contains the offset to the next item to be matched in the
4484: pattern string.
4485:
1.7 ! moko 4486: The next_item_length field is present from version 1 of the callout
1.5 misha 4487: structure. It contains the length of the next item to be matched in the
1.7 ! moko 4488: pattern string. When the callout immediately precedes an alternation
! 4489: bar, a closing parenthesis, or the end of the pattern, the length is
! 4490: zero. When the callout precedes an opening parenthesis, the length is
1.5 misha 4491: that of the entire subpattern.
1.1 misha 4492:
1.7 ! moko 4493: The pattern_position and next_item_length fields are intended to help
! 4494: in distinguishing between different automatic callouts, which all have
1.1 misha 4495: the same callout number. However, they are set for all callouts.
4496:
1.7 ! moko 4497: The mark field is present from version 2 of the callout structure. In
! 4498: callouts from pcre_exec() or pcre[16|32]_exec() it contains a pointer
! 4499: to the zero-terminated name of the most recently passed (*MARK),
! 4500: (*PRUNE), or (*THEN) item in the match, or NULL if no such items have
! 4501: been passed. Instances of (*PRUNE) or (*THEN) without a name do not
! 4502: obliterate a previous (*MARK). In callouts from the DFA matching func-
1.6 misha 4503: tions this field always contains NULL.
1.5 misha 4504:
1.1 misha 4505:
4506: RETURN VALUES
4507:
1.7 ! moko 4508: The external callout function returns an integer to PCRE. If the value
! 4509: is zero, matching proceeds as normal. If the value is greater than
! 4510: zero, matching fails at the current point, but the testing of other
1.1 misha 4511: matching possibilities goes ahead, just as if a lookahead assertion had
1.7 ! moko 4512: failed. If the value is less than zero, the match is abandoned, the
1.5 misha 4513: matching function returns the negative value.
1.1 misha 4514:
1.7 ! moko 4515: Negative values should normally be chosen from the set of
1.1 misha 4516: PCRE_ERROR_xxx values. In particular, PCRE_ERROR_NOMATCH forces a stan-
1.7 ! moko 4517: dard "no match" failure. The error number PCRE_ERROR_CALLOUT is
! 4518: reserved for use by callout functions; it will never be used by PCRE
1.1 misha 4519: itself.
4520:
4521:
4522: AUTHOR
4523:
4524: Philip Hazel
4525: University Computing Service
4526: Cambridge CB2 3QH, England.
4527:
4528:
4529: REVISION
4530:
1.7 ! moko 4531: Last updated: 12 November 2013
1.6 misha 4532: Copyright (c) 1997-2013 University of Cambridge.
1.1 misha 4533: ------------------------------------------------------------------------------
4534:
4535:
1.6 misha 4536: PCRECOMPAT(3) Library Functions Manual PCRECOMPAT(3)
4537:
1.1 misha 4538:
4539:
4540: NAME
4541: PCRE - Perl-compatible regular expressions
4542:
4543: DIFFERENCES BETWEEN PCRE AND PERL
4544:
4545: This document describes the differences in the ways that PCRE and Perl
1.4 misha 4546: handle regular expressions. The differences described here are with
4547: respect to Perl versions 5.10 and above.
1.1 misha 4548:
1.5 misha 4549: 1. PCRE has only a subset of Perl's Unicode support. Details of what it
4550: does have are given in the pcreunicode page.
4551:
4552: 2. PCRE allows repeat quantifiers only on parenthesized assertions, but
4553: they do not mean what you might think. For example, (?!a){3} does not
4554: assert that the next three characters are not "a". It just asserts that
4555: the next character is not "a" three times (in principle: PCRE optimizes
4556: this to run the assertion just once). Perl allows repeat quantifiers on
4557: other assertions such as \b, but these do not seem to have any use.
4558:
4559: 3. Capturing subpatterns that occur inside negative lookahead asser-
4560: tions are counted, but their entries in the offsets vector are never
1.6 misha 4561: set. Perl sometimes (but not always) sets its numerical variables from
4562: inside negative assertions.
1.1 misha 4563:
1.5 misha 4564: 4. Though binary zero characters are supported in the subject string,
1.1 misha 4565: they are not allowed in a pattern string because it is passed as a nor-
4566: mal C string, terminated by zero. The escape sequence \0 can be used in
4567: the pattern to represent a binary zero.
4568:
1.5 misha 4569: 5. The following Perl escape sequences are not supported: \l, \u, \L,
4570: \U, and \N when followed by a character name or Unicode value. (\N on
4571: its own, matching a non-newline character, is supported.) In fact these
4572: are implemented by Perl's general string-handling and are not part of
4573: its pattern matching engine. If any of these are encountered by PCRE,
4574: an error is generated by default. However, if the PCRE_JAVASCRIPT_COM-
4575: PAT option is set, \U and \u are interpreted as JavaScript interprets
4576: them.
4577:
4578: 6. The Perl escape sequences \p, \P, and \X are supported only if PCRE
4579: is built with Unicode character property support. The properties that
4580: can be tested with \p and \P are limited to the general category prop-
4581: erties such as Lu and Nd, script names such as Greek or Han, and the
4582: derived properties Any and L&. PCRE does support the Cs (surrogate)
4583: property, which Perl does not; the Perl documentation says "Because
1.4 misha 4584: Perl hides the need for the user to understand the internal representa-
1.5 misha 4585: tion of Unicode characters, there is no need to implement the somewhat
1.4 misha 4586: messy concept of surrogates."
1.1 misha 4587:
1.6 misha 4588: 7. PCRE does support the \Q...\E escape for quoting substrings. Charac-
4589: ters in between are treated as literals. This is slightly different
4590: from Perl in that $ and @ are also handled as literals inside the
4591: quotes. In Perl, they cause variable interpolation (but of course PCRE
1.1 misha 4592: does not have variables). Note the following examples:
4593:
4594: Pattern PCRE matches Perl matches
4595:
4596: \Qabc$xyz\E abc$xyz abc followed by the
4597: contents of $xyz
4598: \Qabc\$xyz\E abc\$xyz abc\$xyz
4599: \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
4600:
1.6 misha 4601: The \Q...\E sequence is recognized both inside and outside character
1.1 misha 4602: classes.
4603:
1.6 misha 4604: 8. Fairly obviously, PCRE does not support the (?{code}) and (??{code})
4605: constructions. However, there is support for recursive patterns. This
4606: is not available in Perl 5.8, but it is in Perl 5.10. Also, the PCRE
4607: "callout" feature allows an external function to be called during pat-
1.1 misha 4608: tern matching. See the pcrecallout documentation for details.
4609:
1.6 misha 4610: 9. Subpatterns that are called as subroutines (whether or not recur-
4611: sively) are always treated as atomic groups in PCRE. This is like
4612: Python, but unlike Perl. Captured values that are set outside a sub-
4613: routine call can be reference from inside in PCRE, but not in Perl.
1.5 misha 4614: There is a discussion that explains these differences in more detail in
4615: the section on recursion differences from Perl in the pcrepattern page.
4616:
1.6 misha 4617: 10. If any of the backtracking control verbs are used in a subpattern
4618: that is called as a subroutine (whether or not recursively), their
4619: effect is confined to that subpattern; it does not extend to the sur-
4620: rounding pattern. This is not always the case in Perl. In particular,
4621: if (*THEN) is present in a group that is called as a subroutine, its
4622: action is limited to that group, even if the group does not contain any
4623: | characters. Note that such subpatterns are processed as anchored at
4624: the point where they are tested.
4625:
4626: 11. If a pattern contains more than one backtracking control verb, the
4627: first one that is backtracked onto acts. For example, in the pattern
4628: A(*COMMIT)B(*PRUNE)C a failure in B triggers (*COMMIT), but a failure
4629: in C triggers (*PRUNE). Perl's behaviour is more complex; in many cases
4630: it is the same as PCRE, but there are examples where it differs.
4631:
4632: 12. Most backtracking verbs in assertions have their normal actions.
4633: They are not confined to the assertion.
4634:
4635: 13. There are some differences that are concerned with the settings of
4636: captured strings when part of a pattern is repeated. For example,
4637: matching "aba" against the pattern /^(a(b)?)+$/ in Perl leaves $2
1.1 misha 4638: unset, but in PCRE it is set to "b".
4639:
1.6 misha 4640: 14. PCRE's handling of duplicate subpattern numbers and duplicate sub-
1.4 misha 4641: pattern names is not as general as Perl's. This is a consequence of the
4642: fact the PCRE works internally just with numbers, using an external ta-
1.6 misha 4643: ble to translate between numbers and names. In particular, a pattern
4644: such as (?|(?<a>A)|(?<b)B), where the two capturing parentheses have
4645: the same number but different names, is not supported, and causes an
4646: error at compile time. If it were allowed, it would not be possible to
4647: distinguish which parentheses matched, because both names map to cap-
1.4 misha 4648: turing subpattern number 1. To avoid this confusing situation, an error
4649: is given at compile time.
4650:
1.6 misha 4651: 15. Perl recognizes comments in some places that PCRE does not, for
4652: example, between the ( and ? at the start of a subpattern. If the /x
1.7 ! moko 4653: modifier is set, Perl allows white space between ( and ? (though cur-
! 4654: rent Perls warn that this is deprecated) but PCRE never does, even if
! 4655: the PCRE_EXTENDED option is set.
! 4656:
! 4657: 16. Perl, when in warning mode, gives warnings for character classes
! 4658: such as [A-\d] or [a-[:digit:]]. It then treats the hyphens as liter-
! 4659: als. PCRE has no warning features, so it gives an error in these cases
! 4660: because they are almost certainly user mistakes.
1.5 misha 4661:
1.7 ! moko 4662: 17. In PCRE, the upper/lower case character properties Lu and Ll are
1.6 misha 4663: not affected when case-independent matching is specified. For example,
4664: \p{Lu} always matches an upper case letter. I think Perl has changed in
4665: this respect; in the release at the time of writing (5.16), \p{Lu} and
4666: \p{Ll} match all letters, regardless of case, when case independence is
4667: specified.
4668:
1.7 ! moko 4669: 18. PCRE provides some extensions to the Perl regular expression facil-
1.5 misha 4670: ities. Perl 5.10 includes new features that are not in earlier ver-
4671: sions of Perl, some of which (such as named parentheses) have been in
1.4 misha 4672: PCRE for some time. This list is with respect to Perl 5.10:
4673:
1.5 misha 4674: (a) Although lookbehind assertions in PCRE must match fixed length
4675: strings, each alternative branch of a lookbehind assertion can match a
4676: different length of string. Perl requires them all to have the same
1.4 misha 4677: length.
1.1 misha 4678:
1.5 misha 4679: (b) If PCRE_DOLLAR_ENDONLY is set and PCRE_MULTILINE is not set, the $
1.1 misha 4680: meta-character matches only at the very end of the string.
4681:
4682: (c) If PCRE_EXTRA is set, a backslash followed by a letter with no spe-
4683: cial meaning is faulted. Otherwise, like Perl, the backslash is quietly
4684: ignored. (Perl can be made to issue a warning.)
4685:
1.5 misha 4686: (d) If PCRE_UNGREEDY is set, the greediness of the repetition quanti-
1.1 misha 4687: fiers is inverted, that is, by default they are not greedy, but if fol-
4688: lowed by a question mark they are.
4689:
4690: (e) PCRE_ANCHORED can be used at matching time to force a pattern to be
4691: tried only at the first matching position in the subject string.
4692:
1.4 misha 4693: (f) The PCRE_NOTBOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART,
1.5 misha 4694: and PCRE_NO_AUTO_CAPTURE options for pcre_exec() have no Perl equiva-
1.4 misha 4695: lents.
1.1 misha 4696:
1.5 misha 4697: (g) The \R escape sequence can be restricted to match only CR, LF, or
1.1 misha 4698: CRLF by the PCRE_BSR_ANYCRLF option.
4699:
4700: (h) The callout facility is PCRE-specific.
4701:
4702: (i) The partial matching facility is PCRE-specific.
4703:
4704: (j) Patterns compiled by PCRE can be saved and re-used at a later time,
1.5 misha 4705: even on different hosts that have the other endianness. However, this
4706: does not apply to optimized data created by the just-in-time compiler.
1.1 misha 4707:
1.6 misha 4708: (k) The alternative matching functions (pcre_dfa_exec(),
4709: pcre16_dfa_exec() and pcre32_dfa_exec(),) match in a different way and
4710: are not Perl-compatible.
1.1 misha 4711:
1.5 misha 4712: (l) PCRE recognizes some special sequences such as (*CR) at the start
1.1 misha 4713: of a pattern that set overall options that cannot be changed within the
4714: pattern.
4715:
4716:
4717: AUTHOR
4718:
4719: Philip Hazel
4720: University Computing Service
4721: Cambridge CB2 3QH, England.
4722:
4723:
4724: REVISION
4725:
1.7 ! moko 4726: Last updated: 10 November 2013
1.6 misha 4727: Copyright (c) 1997-2013 University of Cambridge.
1.1 misha 4728: ------------------------------------------------------------------------------
4729:
4730:
1.6 misha 4731: PCREPATTERN(3) Library Functions Manual PCREPATTERN(3)
4732:
1.1 misha 4733:
4734:
4735: NAME
4736: PCRE - Perl-compatible regular expressions
4737:
4738: PCRE REGULAR EXPRESSION DETAILS
4739:
4740: The syntax and semantics of the regular expressions that are supported
4741: by PCRE are described in detail below. There is a quick-reference syn-
4742: tax summary in the pcresyntax page. PCRE tries to match Perl syntax and
4743: semantics as closely as it can. PCRE also supports some alternative
4744: regular expression syntax (which does not conflict with the Perl syn-
4745: tax) in order to provide some compatibility with regular expressions in
4746: Python, .NET, and Oniguruma.
4747:
4748: Perl's regular expressions are described in its own documentation, and
4749: regular expressions in general are covered in a number of books, some
4750: of which have copious examples. Jeffrey Friedl's "Mastering Regular
4751: Expressions", published by O'Reilly, covers regular expressions in
4752: great detail. This description of PCRE's regular expressions is
4753: intended as reference material.
4754:
1.6 misha 4755: This document discusses the patterns that are supported by PCRE when
4756: one its main matching functions, pcre_exec() (8-bit) or
4757: pcre[16|32]_exec() (16- or 32-bit), is used. PCRE also has alternative
4758: matching functions, pcre_dfa_exec() and pcre[16|32_dfa_exec(), which
4759: match using a different algorithm that is not Perl-compatible. Some of
4760: the features discussed below are not available when DFA matching is
4761: used. The advantages and disadvantages of the alternative functions,
4762: and how they differ from the normal functions, are discussed in the
4763: pcrematching page.
4764:
4765:
4766: SPECIAL START-OF-PATTERN ITEMS
4767:
4768: A number of options that can be passed to pcre_compile() can also be
4769: set by special items at the start of a pattern. These are not Perl-com-
4770: patible, but are provided to make these options accessible to pattern
4771: writers who are not able to change the program that processes the pat-
4772: tern. Any number of these items may appear, but they must all be
4773: together right at the start of the pattern string, and the letters must
4774: be in upper case.
4775:
4776: UTF support
4777:
1.1 misha 4778: The original operation of PCRE was on strings of one-byte characters.
1.5 misha 4779: However, there is now also support for UTF-8 strings in the original
1.6 misha 4780: library, an extra library that supports 16-bit and UTF-16 character
4781: strings, and a third library that supports 32-bit and UTF-32 character
1.5 misha 4782: strings. To use these features, PCRE must be built to include appropri-
1.6 misha 4783: ate support. When using UTF strings you must either call the compiling
4784: function with the PCRE_UTF8, PCRE_UTF16, or PCRE_UTF32 option, or the
4785: pattern must start with one of these special sequences:
1.3 misha 4786:
4787: (*UTF8)
1.5 misha 4788: (*UTF16)
1.6 misha 4789: (*UTF32)
4790: (*UTF)
4791:
4792: (*UTF) is a generic sequence that can be used with any of the
4793: libraries. Starting a pattern with such a sequence is equivalent to
4794: setting the relevant option. How setting a UTF mode affects pattern
4795: matching is mentioned in several places below. There is also a summary
4796: of features in the pcreunicode page.
4797:
4798: Some applications that allow their users to supply patterns may wish to
4799: restrict them to non-UTF data for security reasons. If the
4800: PCRE_NEVER_UTF option is set at compile time, (*UTF) etc. are not
4801: allowed, and their appearance causes an error.
1.3 misha 4802:
1.6 misha 4803: Unicode property support
1.1 misha 4804:
1.7 ! moko 4805: Another special sequence that may appear at the start of a pattern is
! 4806: (*UCP). This has the same effect as setting the PCRE_UCP option: it
! 4807: causes sequences such as \d and \w to use Unicode properties to deter-
! 4808: mine character types, instead of recognizing only characters with codes
! 4809: less than 128 via a lookup table.
! 4810:
! 4811: Disabling auto-possessification
! 4812:
! 4813: If a pattern starts with (*NO_AUTO_POSSESS), it has the same effect as
! 4814: setting the PCRE_NO_AUTO_POSSESS option at compile time. This stops
! 4815: PCRE from making quantifiers possessive when what follows cannot match
! 4816: the repeated item. For example, by default a+b is treated as a++b. For
! 4817: more details, see the pcreapi documentation.
1.4 misha 4818:
1.6 misha 4819: Disabling start-up optimizations
4820:
1.7 ! moko 4821: If a pattern starts with (*NO_START_OPT), it has the same effect as
1.4 misha 4822: setting the PCRE_NO_START_OPTIMIZE option either at compile or matching
1.7 ! moko 4823: time. This disables several optimizations for quickly reaching "no
! 4824: match" results. For more details, see the pcreapi documentation.
1.1 misha 4825:
1.6 misha 4826: Newline conventions
1.1 misha 4827:
1.6 misha 4828: PCRE supports five different conventions for indicating line breaks in
4829: strings: a single CR (carriage return) character, a single LF (line-
1.1 misha 4830: feed) character, the two-character sequence CRLF, any of the three pre-
1.6 misha 4831: ceding, or any Unicode newline sequence. The pcreapi page has further
4832: discussion about newlines, and shows how to set the newline convention
1.1 misha 4833: in the options arguments for the compiling and matching functions.
4834:
1.6 misha 4835: It is also possible to specify a newline convention by starting a pat-
1.1 misha 4836: tern string with one of the following five sequences:
4837:
4838: (*CR) carriage return
4839: (*LF) linefeed
4840: (*CRLF) carriage return, followed by linefeed
4841: (*ANYCRLF) any of the three above
4842: (*ANY) all Unicode newline sequences
4843:
1.5 misha 4844: These override the default and the options given to the compiling func-
1.6 misha 4845: tion. For example, on a Unix system where LF is the default newline
1.5 misha 4846: sequence, the pattern
1.1 misha 4847:
4848: (*CR)a.b
4849:
4850: changes the convention to CR. That pattern matches "a\nb" because LF is
1.6 misha 4851: no longer a newline. If more than one of these settings is present, the
4852: last one is used.
4853:
4854: The newline convention affects where the circumflex and dollar asser-
4855: tions are true. It also affects the interpretation of the dot metachar-
4856: acter when PCRE_DOTALL is not set, and the behaviour of \N. However, it
4857: does not affect what the \R escape sequence matches. By default, this
4858: is any Unicode newline sequence, for Perl compatibility. However, this
4859: can be changed; see the description of \R in the section entitled "New-
4860: line sequences" below. A change of \R setting can be combined with a
4861: change of newline convention.
4862:
4863: Setting match and recursion limits
4864:
4865: The caller of pcre_exec() can set a limit on the number of times the
4866: internal match() function is called and on the maximum depth of recur-
4867: sive calls. These facilities are provided to catch runaway matches that
4868: are provoked by patterns with huge matching trees (a typical example is
4869: a pattern with nested unlimited repeats) and to avoid running out of
4870: system stack by too much recursion. When one of these limits is
4871: reached, pcre_exec() gives an error return. The limits can also be set
4872: by items at the start of the pattern of the form
4873:
4874: (*LIMIT_MATCH=d)
4875: (*LIMIT_RECURSION=d)
4876:
4877: where d is any number of decimal digits. However, the value of the set-
1.7 ! moko 4878: ting must be less than the value set (or defaulted) by the caller of
! 4879: pcre_exec() for it to have any effect. In other words, the pattern
! 4880: writer can lower the limits set by the programmer, but not raise them.
! 4881: If there is more than one setting of one of these limits, the lower
! 4882: value is used.
1.6 misha 4883:
4884:
4885: EBCDIC CHARACTER CODES
4886:
1.7 ! moko 4887: PCRE can be compiled to run in an environment that uses EBCDIC as its
1.6 misha 4888: character code rather than ASCII or Unicode (typically a mainframe sys-
1.7 ! moko 4889: tem). In the sections below, character code values are ASCII or Uni-
1.6 misha 4890: code; in an EBCDIC environment these characters may have different code
4891: values, and there are no code points greater than 255.
1.1 misha 4892:
4893:
4894: CHARACTERS AND METACHARACTERS
4895:
1.7 ! moko 4896: A regular expression is a pattern that is matched against a subject
! 4897: string from left to right. Most characters stand for themselves in a
! 4898: pattern, and match the corresponding characters in the subject. As a
1.1 misha 4899: trivial example, the pattern
4900:
4901: The quick brown fox
4902:
4903: matches a portion of a subject string that is identical to itself. When
1.7 ! moko 4904: caseless matching is specified (the PCRE_CASELESS option), letters are
! 4905: matched independently of case. In a UTF mode, PCRE always understands
! 4906: the concept of case for characters whose values are less than 128, so
! 4907: caseless matching is always possible. For characters with higher val-
! 4908: ues, the concept of case is supported if PCRE is compiled with Unicode
! 4909: property support, but not otherwise. If you want to use caseless
! 4910: matching for characters 128 and above, you must ensure that PCRE is
1.5 misha 4911: compiled with Unicode property support as well as with UTF support.
1.1 misha 4912:
1.7 ! moko 4913: The power of regular expressions comes from the ability to include
! 4914: alternatives and repetitions in the pattern. These are encoded in the
1.1 misha 4915: pattern by the use of metacharacters, which do not stand for themselves
4916: but instead are interpreted in some special way.
4917:
1.7 ! moko 4918: There are two different sets of metacharacters: those that are recog-
! 4919: nized anywhere in the pattern except within square brackets, and those
! 4920: that are recognized within square brackets. Outside square brackets,
1.1 misha 4921: the metacharacters are as follows:
4922:
4923: \ general escape character with several uses
4924: ^ assert start of string (or line, in multiline mode)
4925: $ assert end of string (or line, in multiline mode)
4926: . match any character except newline (by default)
4927: [ start character class definition
4928: | start of alternative branch
4929: ( start subpattern
4930: ) end subpattern
4931: ? extends the meaning of (
4932: also 0 or 1 quantifier
4933: also quantifier minimizer
4934: * 0 or more quantifier
4935: + 1 or more quantifier
4936: also "possessive quantifier"
4937: { start min/max quantifier
4938:
1.7 ! moko 4939: Part of a pattern that is in square brackets is called a "character
1.1 misha 4940: class". In a character class the only metacharacters are:
4941:
4942: \ general escape character
4943: ^ negate the class, but only if the first character
4944: - indicates character range
4945: [ POSIX character class (only if followed by POSIX
4946: syntax)
4947: ] terminates the character class
4948:
1.3 misha 4949: The following sections describe the use of each of the metacharacters.
1.1 misha 4950:
4951:
4952: BACKSLASH
4953:
4954: The backslash character has several uses. Firstly, if it is followed by
1.4 misha 4955: a character that is not a number or a letter, it takes away any special
1.7 ! moko 4956: meaning that character may have. This use of backslash as an escape
1.4 misha 4957: character applies both inside and outside character classes.
4958:
1.7 ! moko 4959: For example, if you want to match a * character, you write \* in the
! 4960: pattern. This escaping action applies whether or not the following
! 4961: character would otherwise be interpreted as a metacharacter, so it is
! 4962: always safe to precede a non-alphanumeric with backslash to specify
! 4963: that it stands for itself. In particular, if you want to match a back-
1.1 misha 4964: slash, you write \\.
4965:
1.7 ! moko 4966: In a UTF mode, only ASCII numbers and letters have any special meaning
! 4967: after a backslash. All other characters (in particular, those whose
1.4 misha 4968: codepoints are greater than 127) are treated as literals.
4969:
1.7 ! moko 4970: If a pattern is compiled with the PCRE_EXTENDED option, most white
! 4971: space in the pattern (other than in a character class), and characters
! 4972: between a # outside a character class and the next newline, inclusive,
! 4973: are ignored. An escaping backslash can be used to include a white space
! 4974: or # character as part of the pattern.
! 4975:
! 4976: If you want to remove the special meaning from a sequence of charac-
! 4977: ters, you can do so by putting them between \Q and \E. This is differ-
! 4978: ent from Perl in that $ and @ are handled as literals in \Q...\E
! 4979: sequences in PCRE, whereas in Perl, $ and @ cause variable interpola-
1.1 misha 4980: tion. Note the following examples:
4981:
4982: Pattern PCRE matches Perl matches
4983:
4984: \Qabc$xyz\E abc$xyz abc followed by the
4985: contents of $xyz
4986: \Qabc\$xyz\E abc\$xyz abc\$xyz
4987: \Qabc\E\$\Qxyz\E abc$xyz abc$xyz
4988:
1.7 ! moko 4989: The \Q...\E sequence is recognized both inside and outside character
! 4990: classes. An isolated \E that is not preceded by \Q is ignored. If \Q
! 4991: is not followed by \E later in the pattern, the literal interpretation
! 4992: continues to the end of the pattern (that is, \E is assumed at the
! 4993: end). If the isolated \Q is inside a character class, this causes an
1.5 misha 4994: error, because the character class is not terminated.
1.1 misha 4995:
4996: Non-printing characters
4997:
4998: A second use of backslash provides a way of encoding non-printing char-
1.7 ! moko 4999: acters in patterns in a visible manner. There is no restriction on the
! 5000: appearance of non-printing characters, apart from the binary zero that
! 5001: terminates a pattern, but when a pattern is being prepared by text
! 5002: editing, it is often easier to use one of the following escape
1.1 misha 5003: sequences than the binary character it represents:
5004:
5005: \a alarm, that is, the BEL character (hex 07)
1.4 misha 5006: \cx "control-x", where x is any ASCII character
1.1 misha 5007: \e escape (hex 1B)
1.6 misha 5008: \f form feed (hex 0C)
1.1 misha 5009: \n linefeed (hex 0A)
5010: \r carriage return (hex 0D)
5011: \t tab (hex 09)
1.7 ! moko 5012: \0dd character with octal code 0dd
1.4 misha 5013: \ddd character with octal code ddd, or back reference
1.7 ! moko 5014: \o{ddd..} character with octal code ddd..
1.1 misha 5015: \xhh character with hex code hh
1.5 misha 5016: \x{hhh..} character with hex code hhh.. (non-JavaScript mode)
5017: \uhhhh character with hex code hhhh (JavaScript mode only)
1.1 misha 5018:
1.7 ! moko 5019: The precise effect of \cx on ASCII characters is as follows: if x is a
! 5020: lower case letter, it is converted to upper case. Then bit 6 of the
1.6 misha 5021: character (hex 40) is inverted. Thus \cA to \cZ become hex 01 to hex 1A
1.7 ! moko 5022: (A is 41, Z is 5A), but \c{ becomes hex 3B ({ is 7B), and \c; becomes
! 5023: hex 7B (; is 3B). If the data item (byte or 16-bit value) following \c
! 5024: has a value greater than 127, a compile-time error occurs. This locks
1.6 misha 5025: out non-ASCII characters in all modes.
5026:
1.7 ! moko 5027: The \c facility was designed for use with ASCII characters, but with
! 5028: the extension to Unicode it is even less useful than it once was. It
! 5029: is, however, recognized when PCRE is compiled in EBCDIC mode, where
! 5030: data items are always bytes. In this mode, all values are valid after
! 5031: \c. If the next character is a lower case letter, it is converted to
! 5032: upper case. Then the 0xc0 bits of the byte are inverted. Thus \cA
! 5033: becomes hex 01, as in ASCII (A is C1), but because the EBCDIC letters
! 5034: are disjoint, \cZ becomes hex 29 (Z is E9), and other characters also
1.6 misha 5035: generate different values.
1.5 misha 5036:
1.7 ! moko 5037: After \0 up to two further octal digits are read. If there are fewer
! 5038: than two digits, just those that are present are used. Thus the
1.1 misha 5039: sequence \0\x\07 specifies two binary zeros followed by a BEL character
1.7 ! moko 5040: (code value 7). Make sure you supply two digits after the initial zero
1.1 misha 5041: if the pattern character that follows is itself an octal digit.
5042:
1.7 ! moko 5043: The escape \o must be followed by a sequence of octal digits, enclosed
! 5044: in braces. An error occurs if this is not the case. This escape is a
! 5045: recent addition to Perl; it provides way of specifying character code
! 5046: points as octal numbers greater than 0777, and it also allows octal
! 5047: numbers and back references to be unambiguously specified.
! 5048:
! 5049: For greater clarity and unambiguity, it is best to avoid following \ by
! 5050: a digit greater than zero. Instead, use \o{} or \x{} to specify charac-
! 5051: ter numbers, and \g{} to specify back references. The following para-
! 5052: graphs describe the old, ambiguous syntax.
! 5053:
1.1 misha 5054: The handling of a backslash followed by a digit other than 0 is compli-
1.7 ! moko 5055: cated, and Perl has changed in recent releases, causing PCRE also to
! 5056: change. Outside a character class, PCRE reads the digit and any follow-
! 5057: ing digits as a decimal number. If the number is less than 8, or if
! 5058: there have been at least that many previous capturing left parentheses
! 5059: in the expression, the entire sequence is taken as a back reference. A
! 5060: description of how this works is given later, following the discussion
1.1 misha 5061: of parenthesized subpatterns.
5062:
1.7 ! moko 5063: Inside a character class, or if the decimal number following \ is
! 5064: greater than 7 and there have not been that many capturing subpatterns,
! 5065: PCRE handles \8 and \9 as the literal characters "8" and "9", and oth-
! 5066: erwise re-reads up to three octal digits following the backslash, using
! 5067: them to generate a data character. Any subsequent digits stand for
! 5068: themselves. For example:
1.1 misha 5069:
1.6 misha 5070: \040 is another way of writing an ASCII space
1.1 misha 5071: \40 is the same, provided there are fewer than 40
5072: previous capturing subpatterns
5073: \7 is always a back reference
5074: \11 might be a back reference, or another way of
5075: writing a tab
5076: \011 is always a tab
5077: \0113 is a tab followed by the character "3"
5078: \113 might be a back reference, otherwise the
5079: character with octal code 113
5080: \377 might be a back reference, otherwise
1.5 misha 5081: the value 255 (decimal)
1.7 ! moko 5082: \81 is either a back reference, or the two
! 5083: characters "8" and "1"
1.1 misha 5084:
1.7 ! moko 5085: Note that octal values of 100 or greater that are specified using this
! 5086: syntax must not be introduced by a leading zero, because no more than
! 5087: three octal digits are ever read.
! 5088:
! 5089: By default, after \x that is not followed by {, from zero to two hexa-
! 5090: decimal digits are read (letters can be in upper or lower case). Any
! 5091: number of hexadecimal digits may appear between \x{ and }. If a charac-
! 5092: ter other than a hexadecimal digit appears between \x{ and }, or if
! 5093: there is no terminating }, an error occurs.
! 5094:
! 5095: If the PCRE_JAVASCRIPT_COMPAT option is set, the interpretation of \x
! 5096: is as just described only when it is followed by two hexadecimal dig-
! 5097: its. Otherwise, it matches a literal "x" character. In JavaScript
! 5098: mode, support for code points greater than 256 is provided by \u, which
! 5099: must be followed by four hexadecimal digits; otherwise it matches a
! 5100: literal "u" character.
! 5101:
! 5102: Characters whose value is less than 256 can be defined by either of the
! 5103: two syntaxes for \x (or by \u in JavaScript mode). There is no differ-
! 5104: ence in the way they are handled. For example, \xdc is exactly the same
! 5105: as \x{dc} (or \u00dc in JavaScript mode).
! 5106:
! 5107: Constraints on character values
! 5108:
! 5109: Characters that are specified using octal or hexadecimal numbers are
! 5110: limited to certain values, as follows:
! 5111:
! 5112: 8-bit non-UTF mode less than 0x100
! 5113: 8-bit UTF-8 mode less than 0x10ffff and a valid codepoint
! 5114: 16-bit non-UTF mode less than 0x10000
! 5115: 16-bit UTF-16 mode less than 0x10ffff and a valid codepoint
! 5116: 32-bit non-UTF mode less than 0x100000000
! 5117: 32-bit UTF-32 mode less than 0x10ffff and a valid codepoint
! 5118:
! 5119: Invalid Unicode codepoints are the range 0xd800 to 0xdfff (the so-
! 5120: called "surrogate" codepoints), and 0xffef.
! 5121:
! 5122: Escape sequences in character classes
1.1 misha 5123:
5124: All the sequences that define a single character value can be used both
1.4 misha 5125: inside and outside character classes. In addition, inside a character
1.5 misha 5126: class, \b is interpreted as the backspace character (hex 08).
5127:
5128: \N is not allowed in a character class. \B, \R, and \X are not special
5129: inside a character class. Like other unrecognized escape sequences,
5130: they are treated as the literal characters "B", "R", and "X" by
5131: default, but cause an error if the PCRE_EXTRA option is set. Outside a
5132: character class, these sequences have different meanings.
5133:
5134: Unsupported escape sequences
5135:
5136: In Perl, the sequences \l, \L, \u, and \U are recognized by its string
5137: handler and used to modify the case of following characters. By
5138: default, PCRE does not support these escape sequences. However, if the
5139: PCRE_JAVASCRIPT_COMPAT option is set, \U matches a "U" character, and
5140: \u can be used to define a character by code point, as described in the
5141: previous section.
1.1 misha 5142:
5143: Absolute and relative back references
5144:
1.4 misha 5145: The sequence \g followed by an unsigned or a negative number, option-
5146: ally enclosed in braces, is an absolute or relative back reference. A
1.1 misha 5147: named back reference can be coded as \g{name}. Back references are dis-
5148: cussed later, following the discussion of parenthesized subpatterns.
5149:
5150: Absolute and relative subroutine calls
5151:
1.4 misha 5152: For compatibility with Oniguruma, the non-Perl syntax \g followed by a
1.1 misha 5153: name or a number enclosed either in angle brackets or single quotes, is
1.4 misha 5154: an alternative syntax for referencing a subpattern as a "subroutine".
5155: Details are discussed later. Note that \g{...} (Perl syntax) and
5156: \g<...> (Oniguruma syntax) are not synonymous. The former is a back
1.1 misha 5157: reference; the latter is a subroutine call.
5158:
5159: Generic character types
5160:
1.4 misha 5161: Another use of backslash is for specifying generic character types:
1.1 misha 5162:
5163: \d any decimal digit
5164: \D any character that is not a decimal digit
1.6 misha 5165: \h any horizontal white space character
5166: \H any character that is not a horizontal white space character
5167: \s any white space character
5168: \S any character that is not a white space character
5169: \v any vertical white space character
5170: \V any character that is not a vertical white space character
1.1 misha 5171: \w any "word" character
5172: \W any "non-word" character
5173:
1.4 misha 5174: There is also the single sequence \N, which matches a non-newline char-
5175: acter. This is the same as the "." metacharacter when PCRE_DOTALL is
1.5 misha 5176: not set. Perl also uses \N to match characters by name; PCRE does not
5177: support this.
1.4 misha 5178:
1.5 misha 5179: Each pair of lower and upper case escape sequences partitions the com-
5180: plete set of characters into two disjoint sets. Any given character
5181: matches one, and only one, of each pair. The sequences can appear both
5182: inside and outside character classes. They each match one character of
5183: the appropriate type. If the current matching point is at the end of
5184: the subject string, all of them fail, because there is no character to
1.4 misha 5185: match.
5186:
1.7 ! moko 5187: For compatibility with Perl, \s did not used to match the VT character
! 5188: (code 11), which made it different from the the POSIX "space" class.
! 5189: However, Perl added VT at release 5.18, and PCRE followed suit at
! 5190: release 8.34. The default \s characters are now HT (9), LF (10), VT
! 5191: (11), FF (12), CR (13), and space (32), which are defined as white
! 5192: space in the "C" locale. This list may vary if locale-specific matching
! 5193: is taking place. For example, in some locales the "non-breaking space"
! 5194: character (\xA0) is recognized as white space, and in others the VT
! 5195: character is not.
1.1 misha 5196:
1.5 misha 5197: A "word" character is an underscore or any character that is a letter
5198: or digit. By default, the definition of letters and digits is con-
5199: trolled by PCRE's low-valued character tables, and may vary if locale-
5200: specific matching is taking place (see "Locale support" in the pcreapi
5201: page). For example, in a French locale such as "fr_FR" in Unix-like
1.7 ! moko 5202: systems, or "french" in Windows, some character codes greater than 127
1.5 misha 5203: are used for accented letters, and these are then matched by \w. The
1.4 misha 5204: use of locales with Unicode is discouraged.
5205:
1.7 ! moko 5206: By default, characters whose code points are greater than 127 never
! 5207: match \d, \s, or \w, and always match \D, \S, and \W, although this may
! 5208: vary for characters in the range 128-255 when locale-specific matching
! 5209: is happening. These escape sequences retain their original meanings
! 5210: from before Unicode support was available, mainly for efficiency rea-
! 5211: sons. If PCRE is compiled with Unicode property support, and the
! 5212: PCRE_UCP option is set, the behaviour is changed so that Unicode prop-
! 5213: erties are used to determine character types, as follows:
! 5214:
! 5215: \d any character that matches \p{Nd} (decimal digit)
! 5216: \s any character that matches \p{Z} or \h or \v
! 5217: \w any character that matches \p{L} or \p{N}, plus underscore
! 5218:
! 5219: The upper case escapes match the inverse sets of characters. Note that
! 5220: \d matches only decimal digits, whereas \w matches any Unicode digit,
! 5221: as well as any Unicode letter, and underscore. Note also that PCRE_UCP
! 5222: affects \b, and \B because they are defined in terms of \w and \W.
1.4 misha 5223: Matching these sequences is noticeably slower when PCRE_UCP is set.
5224:
1.7 ! moko 5225: The sequences \h, \H, \v, and \V are features that were added to Perl
! 5226: at release 5.10. In contrast to the other sequences, which match only
! 5227: ASCII characters by default, these always match certain high-valued
! 5228: code points, whether or not PCRE_UCP is set. The horizontal space char-
1.5 misha 5229: acters are:
1.1 misha 5230:
1.6 misha 5231: U+0009 Horizontal tab (HT)
1.1 misha 5232: U+0020 Space
5233: U+00A0 Non-break space
5234: U+1680 Ogham space mark
5235: U+180E Mongolian vowel separator
5236: U+2000 En quad
5237: U+2001 Em quad
5238: U+2002 En space
5239: U+2003 Em space
5240: U+2004 Three-per-em space
5241: U+2005 Four-per-em space
5242: U+2006 Six-per-em space
5243: U+2007 Figure space
5244: U+2008 Punctuation space
5245: U+2009 Thin space
5246: U+200A Hair space
5247: U+202F Narrow no-break space
5248: U+205F Medium mathematical space
5249: U+3000 Ideographic space
5250:
5251: The vertical space characters are:
5252:
1.6 misha 5253: U+000A Linefeed (LF)
5254: U+000B Vertical tab (VT)
5255: U+000C Form feed (FF)
5256: U+000D Carriage return (CR)
5257: U+0085 Next line (NEL)
1.1 misha 5258: U+2028 Line separator
5259: U+2029 Paragraph separator
5260:
1.5 misha 5261: In 8-bit, non-UTF-8 mode, only the characters with codepoints less than
5262: 256 are relevant.
5263:
1.1 misha 5264: Newline sequences
5265:
1.7 ! moko 5266: Outside a character class, by default, the escape sequence \R matches
! 5267: any Unicode newline sequence. In 8-bit non-UTF-8 mode \R is equivalent
1.5 misha 5268: to the following:
1.1 misha 5269:
5270: (?>\r\n|\n|\x0b|\f|\r|\x85)
5271:
1.7 ! moko 5272: This is an example of an "atomic group", details of which are given
1.1 misha 5273: below. This particular group matches either the two-character sequence
1.7 ! moko 5274: CR followed by LF, or one of the single characters LF (linefeed,
! 5275: U+000A), VT (vertical tab, U+000B), FF (form feed, U+000C), CR (car-
! 5276: riage return, U+000D), or NEL (next line, U+0085). The two-character
1.6 misha 5277: sequence is treated as a single unit that cannot be split.
1.1 misha 5278:
1.7 ! moko 5279: In other modes, two additional characters whose codepoints are greater
1.1 misha 5280: than 255 are added: LS (line separator, U+2028) and PS (paragraph sepa-
1.7 ! moko 5281: rator, U+2029). Unicode character property support is not needed for
1.1 misha 5282: these characters to be recognized.
5283:
5284: It is possible to restrict \R to match only CR, LF, or CRLF (instead of
1.7 ! moko 5285: the complete set of Unicode line endings) by setting the option
1.1 misha 5286: PCRE_BSR_ANYCRLF either at compile time or when the pattern is matched.
5287: (BSR is an abbrevation for "backslash R".) This can be made the default
1.7 ! moko 5288: when PCRE is built; if this is the case, the other behaviour can be
! 5289: requested via the PCRE_BSR_UNICODE option. It is also possible to
! 5290: specify these settings by starting a pattern string with one of the
1.1 misha 5291: following sequences:
5292:
5293: (*BSR_ANYCRLF) CR, LF, or CRLF only
5294: (*BSR_UNICODE) any Unicode newline sequence
5295:
1.5 misha 5296: These override the default and the options given to the compiling func-
1.7 ! moko 5297: tion, but they can themselves be overridden by options given to a
! 5298: matching function. Note that these special settings, which are not
! 5299: Perl-compatible, are recognized only at the very start of a pattern,
! 5300: and that they must be in upper case. If more than one of them is
! 5301: present, the last one is used. They can be combined with a change of
1.4 misha 5302: newline convention; for example, a pattern can start with:
1.1 misha 5303:
5304: (*ANY)(*BSR_ANYCRLF)
5305:
1.7 ! moko 5306: They can also be combined with the (*UTF8), (*UTF16), (*UTF32), (*UTF)
1.6 misha 5307: or (*UCP) special sequences. Inside a character class, \R is treated as
1.7 ! moko 5308: an unrecognized escape sequence, and so matches the letter "R" by
1.6 misha 5309: default, but causes an error if PCRE_EXTRA is set.
1.1 misha 5310:
5311: Unicode character properties
5312:
5313: When PCRE is built with Unicode character property support, three addi-
1.7 ! moko 5314: tional escape sequences that match characters with specific properties
! 5315: are available. When in 8-bit non-UTF-8 mode, these sequences are of
! 5316: course limited to testing characters whose codepoints are less than
1.5 misha 5317: 256, but they do work in this mode. The extra escape sequences are:
1.1 misha 5318:
5319: \p{xx} a character with the xx property
5320: \P{xx} a character without the xx property
1.6 misha 5321: \X a Unicode extended grapheme cluster
1.1 misha 5322:
1.7 ! moko 5323: The property names represented by xx above are limited to the Unicode
1.4 misha 5324: script names, the general category properties, "Any", which matches any
1.7 ! moko 5325: character (including newline), and some special PCRE properties
! 5326: (described in the next section). Other Perl properties such as "InMu-
! 5327: sicalSymbols" are not currently supported by PCRE. Note that \P{Any}
1.4 misha 5328: does not match any characters, so always causes a match failure.
1.1 misha 5329:
5330: Sets of Unicode characters are defined as belonging to certain scripts.
1.7 ! moko 5331: A character from one of these sets can be matched using a script name.
1.1 misha 5332: For example:
5333:
5334: \p{Greek}
5335: \P{Han}
5336:
1.7 ! moko 5337: Those that are not part of an identified script are lumped together as
1.1 misha 5338: "Common". The current list of scripts is:
5339:
1.7 ! moko 5340: Arabic, Armenian, Avestan, Balinese, Bamum, Bassa_Vah, Batak, Bengali,
! 5341: Bopomofo, Brahmi, Braille, Buginese, Buhid, Canadian_Aboriginal, Car-
! 5342: ian, Caucasian_Albanian, Chakma, Cham, Cherokee, Common, Coptic, Cunei-
! 5343: form, Cypriot, Cyrillic, Deseret, Devanagari, Duployan, Egyptian_Hiero-
! 5344: glyphs, Elbasan, Ethiopic, Georgian, Glagolitic, Gothic, Grantha,
! 5345: Greek, Gujarati, Gurmukhi, Han, Hangul, Hanunoo, Hebrew, Hiragana,
! 5346: Imperial_Aramaic, Inherited, Inscriptional_Pahlavi, Inscrip-
1.6 misha 5347: tional_Parthian, Javanese, Kaithi, Kannada, Katakana, Kayah_Li,
1.7 ! moko 5348: Kharoshthi, Khmer, Khojki, Khudawadi, Lao, Latin, Lepcha, Limbu, Lin-
! 5349: ear_A, Linear_B, Lisu, Lycian, Lydian, Mahajani, Malayalam, Mandaic,
! 5350: Manichaean, Meetei_Mayek, Mende_Kikakui, Meroitic_Cursive,
! 5351: Meroitic_Hieroglyphs, Miao, Modi, Mongolian, Mro, Myanmar, Nabataean,
! 5352: New_Tai_Lue, Nko, Ogham, Ol_Chiki, Old_Italic, Old_North_Arabian,
! 5353: Old_Permic, Old_Persian, Old_South_Arabian, Old_Turkic, Oriya, Osmanya,
! 5354: Pahawh_Hmong, Palmyrene, Pau_Cin_Hau, Phags_Pa, Phoenician,
! 5355: Psalter_Pahlavi, Rejang, Runic, Samaritan, Saurashtra, Sharada, Sha-
! 5356: vian, Siddham, Sinhala, Sora_Sompeng, Sundanese, Syloti_Nagri, Syriac,
! 5357: Tagalog, Tagbanwa, Tai_Le, Tai_Tham, Tai_Viet, Takri, Tamil, Telugu,
! 5358: Thaana, Thai, Tibetan, Tifinagh, Tirhuta, Ugaritic, Vai, Warang_Citi,
1.6 misha 5359: Yi.
1.4 misha 5360:
5361: Each character has exactly one Unicode general category property, spec-
1.7 ! moko 5362: ified by a two-letter abbreviation. For compatibility with Perl, nega-
! 5363: tion can be specified by including a circumflex between the opening
! 5364: brace and the property name. For example, \p{^Lu} is the same as
1.4 misha 5365: \P{Lu}.
1.1 misha 5366:
5367: If only one letter is specified with \p or \P, it includes all the gen-
1.7 ! moko 5368: eral category properties that start with that letter. In this case, in
! 5369: the absence of negation, the curly brackets in the escape sequence are
1.1 misha 5370: optional; these two examples have the same effect:
5371:
5372: \p{L}
5373: \pL
5374:
5375: The following general category property codes are supported:
5376:
5377: C Other
5378: Cc Control
5379: Cf Format
5380: Cn Unassigned
5381: Co Private use
5382: Cs Surrogate
5383:
5384: L Letter
5385: Ll Lower case letter
5386: Lm Modifier letter
5387: Lo Other letter
5388: Lt Title case letter
5389: Lu Upper case letter
5390:
5391: M Mark
5392: Mc Spacing mark
5393: Me Enclosing mark
5394: Mn Non-spacing mark
5395:
5396: N Number
5397: Nd Decimal number
5398: Nl Letter number
5399: No Other number
5400:
5401: P Punctuation
5402: Pc Connector punctuation
5403: Pd Dash punctuation
5404: Pe Close punctuation
5405: Pf Final punctuation
5406: Pi Initial punctuation
5407: Po Other punctuation
5408: Ps Open punctuation
5409:
5410: S Symbol
5411: Sc Currency symbol
5412: Sk Modifier symbol
5413: Sm Mathematical symbol
5414: So Other symbol
5415:
5416: Z Separator
5417: Zl Line separator
5418: Zp Paragraph separator
5419: Zs Space separator
5420:
1.7 ! moko 5421: The special property L& is also supported: it matches a character that
! 5422: has the Lu, Ll, or Lt property, in other words, a letter that is not
1.1 misha 5423: classified as a modifier or "other".
5424:
1.7 ! moko 5425: The Cs (Surrogate) property applies only to characters in the range
! 5426: U+D800 to U+DFFF. Such characters are not valid in Unicode strings and
! 5427: so cannot be tested by PCRE, unless UTF validity checking has been
1.6 misha 5428: turned off (see the discussion of PCRE_NO_UTF8_CHECK,
1.7 ! moko 5429: PCRE_NO_UTF16_CHECK and PCRE_NO_UTF32_CHECK in the pcreapi page). Perl
1.6 misha 5430: does not support the Cs property.
1.1 misha 5431:
1.7 ! moko 5432: The long synonyms for property names that Perl supports (such as
! 5433: \p{Letter}) are not supported by PCRE, nor is it permitted to prefix
1.1 misha 5434: any of these properties with "Is".
5435:
5436: No character that is in the Unicode table has the Cn (unassigned) prop-
5437: erty. Instead, this property is assumed for any code point that is not
5438: in the Unicode table.
5439:
1.7 ! moko 5440: Specifying caseless matching does not affect these escape sequences.
! 5441: For example, \p{Lu} always matches only upper case letters. This is
1.6 misha 5442: different from the behaviour of current versions of Perl.
5443:
1.7 ! moko 5444: Matching characters by Unicode property is not fast, because PCRE has
! 5445: to do a multistage table lookup in order to find a character's prop-
1.6 misha 5446: erty. That is why the traditional escape sequences such as \d and \w do
5447: not use Unicode properties in PCRE by default, though you can make them
1.7 ! moko 5448: do so by setting the PCRE_UCP option or by starting the pattern with
1.6 misha 5449: (*UCP).
5450:
5451: Extended grapheme clusters
1.1 misha 5452:
1.7 ! moko 5453: The \X escape matches any number of Unicode characters that form an
1.6 misha 5454: "extended grapheme cluster", and treats the sequence as an atomic group
1.7 ! moko 5455: (see below). Up to and including release 8.31, PCRE matched an ear-
1.6 misha 5456: lier, simpler definition that was equivalent to
1.1 misha 5457:
5458: (?>\PM\pM*)
5459:
1.7 ! moko 5460: That is, it matched a character without the "mark" property, followed
! 5461: by zero or more characters with the "mark" property. Characters with
! 5462: the "mark" property are typically non-spacing accents that affect the
1.6 misha 5463: preceding character.
5464:
1.7 ! moko 5465: This simple definition was extended in Unicode to include more compli-
! 5466: cated kinds of composite character by giving each character a grapheme
! 5467: breaking property, and creating rules that use these properties to
! 5468: define the boundaries of extended grapheme clusters. In releases of
1.6 misha 5469: PCRE later than 8.31, \X matches one of these clusters.
5470:
1.7 ! moko 5471: \X always matches at least one character. Then it decides whether to
1.6 misha 5472: add additional characters according to the following rules for ending a
5473: cluster:
5474:
5475: 1. End at the end of the subject string.
5476:
1.7 ! moko 5477: 2. Do not end between CR and LF; otherwise end after any control char-
1.6 misha 5478: acter.
5479:
1.7 ! moko 5480: 3. Do not break Hangul (a Korean script) syllable sequences. Hangul
! 5481: characters are of five types: L, V, T, LV, and LVT. An L character may
! 5482: be followed by an L, V, LV, or LVT character; an LV or V character may
1.6 misha 5483: be followed by a V or T character; an LVT or T character may be follwed
5484: only by a T character.
5485:
1.7 ! moko 5486: 4. Do not end before extending characters or spacing marks. Characters
! 5487: with the "mark" property always have the "extend" grapheme breaking
1.6 misha 5488: property.
5489:
5490: 5. Do not end after prepend characters.
5491:
5492: 6. Otherwise, end the cluster.
1.4 misha 5493:
5494: PCRE's additional properties
5495:
1.7 ! moko 5496: As well as the standard Unicode properties described above, PCRE sup-
! 5497: ports four more that make it possible to convert traditional escape
! 5498: sequences such as \w and \s to use Unicode properties. PCRE uses these
! 5499: non-standard, non-Perl properties internally when PCRE_UCP is set. How-
! 5500: ever, they may also be used explicitly. These properties are:
1.4 misha 5501:
5502: Xan Any alphanumeric character
5503: Xps Any POSIX space character
5504: Xsp Any Perl space character
5505: Xwd Any Perl "word" character
5506:
1.6 misha 5507: Xan matches characters that have either the L (letter) or the N (num-
5508: ber) property. Xps matches the characters tab, linefeed, vertical tab,
5509: form feed, or carriage return, and any other character that has the Z
1.7 ! moko 5510: (separator) property. Xsp is the same as Xps; it used to exclude ver-
! 5511: tical tab, for Perl compatibility, but Perl changed, and so PCRE fol-
! 5512: lowed at release 8.34. Xwd matches the same characters as Xan, plus
! 5513: underscore.
1.1 misha 5514:
1.6 misha 5515: There is another non-standard property, Xuc, which matches any charac-
5516: ter that can be represented by a Universal Character Name in C++ and
5517: other programming languages. These are the characters $, @, ` (grave
5518: accent), and all characters with Unicode code points greater than or
5519: equal to U+00A0, except for the surrogates U+D800 to U+DFFF. Note that
5520: most base (ASCII) characters are excluded. (Universal Character Names
5521: are of the form \uHHHH or \UHHHHHHHH where H is a hexadecimal digit.
5522: Note that the Xuc property does not match these sequences but the char-
5523: acters that they represent.)
5524:
1.1 misha 5525: Resetting the match start
5526:
1.6 misha 5527: The escape sequence \K causes any previously matched characters not to
1.4 misha 5528: be included in the final matched sequence. For example, the pattern:
1.1 misha 5529:
5530: foo\Kbar
5531:
1.6 misha 5532: matches "foobar", but reports that it has matched "bar". This feature
5533: is similar to a lookbehind assertion (described below). However, in
5534: this case, the part of the subject before the real match does not have
5535: to be of fixed length, as lookbehind assertions do. The use of \K does
5536: not interfere with the setting of captured substrings. For example,
1.1 misha 5537: when the pattern
5538:
5539: (foo)\Kbar
5540:
5541: matches "foobar", the first substring is still set to "foo".
5542:
1.6 misha 5543: Perl documents that the use of \K within assertions is "not well
5544: defined". In PCRE, \K is acted upon when it occurs inside positive
1.7 ! moko 5545: assertions, but is ignored in negative assertions. Note that when a
! 5546: pattern such as (?=ab\K) matches, the reported start of the match can
! 5547: be greater than the end of the match.
1.4 misha 5548:
1.1 misha 5549: Simple assertions
5550:
1.6 misha 5551: The final use of backslash is for certain simple assertions. An asser-
5552: tion specifies a condition that has to be met at a particular point in
5553: a match, without consuming any characters from the subject string. The
5554: use of subpatterns for more complicated assertions is described below.
1.1 misha 5555: The backslashed assertions are:
5556:
5557: \b matches at a word boundary
5558: \B matches when not at a word boundary
5559: \A matches at the start of the subject
5560: \Z matches at the end of the subject
5561: also matches before a newline at the end of the subject
5562: \z matches only at the end of the subject
5563: \G matches at the first matching position in the subject
5564:
1.6 misha 5565: Inside a character class, \b has a different meaning; it matches the
5566: backspace character. If any other of these assertions appears in a
5567: character class, by default it matches the corresponding literal char-
1.4 misha 5568: acter (for example, \B matches the letter B). However, if the
1.6 misha 5569: PCRE_EXTRA option is set, an "invalid escape sequence" error is gener-
1.4 misha 5570: ated instead.
1.1 misha 5571:
1.6 misha 5572: A word boundary is a position in the subject string where the current
5573: character and the previous character do not both match \w or \W (i.e.
5574: one matches \w and the other matches \W), or the start or end of the
5575: string if the first or last character matches \w, respectively. In a
5576: UTF mode, the meanings of \w and \W can be changed by setting the
5577: PCRE_UCP option. When this is done, it also affects \b and \B. Neither
5578: PCRE nor Perl has a separate "start of word" or "end of word" metase-
5579: quence. However, whatever follows \b normally determines which it is.
1.4 misha 5580: For example, the fragment \ba matches "a" at the start of a word.
1.1 misha 5581:
1.6 misha 5582: The \A, \Z, and \z assertions differ from the traditional circumflex
1.1 misha 5583: and dollar (described in the next section) in that they only ever match
1.6 misha 5584: at the very start and end of the subject string, whatever options are
5585: set. Thus, they are independent of multiline mode. These three asser-
1.1 misha 5586: tions are not affected by the PCRE_NOTBOL or PCRE_NOTEOL options, which
1.6 misha 5587: affect only the behaviour of the circumflex and dollar metacharacters.
5588: However, if the startoffset argument of pcre_exec() is non-zero, indi-
1.1 misha 5589: cating that matching is to start at a point other than the beginning of
1.6 misha 5590: the subject, \A can never match. The difference between \Z and \z is
1.1 misha 5591: that \Z matches before a newline at the end of the string as well as at
5592: the very end, whereas \z matches only at the end.
5593:
1.6 misha 5594: The \G assertion is true only when the current matching position is at
5595: the start point of the match, as specified by the startoffset argument
5596: of pcre_exec(). It differs from \A when the value of startoffset is
5597: non-zero. By calling pcre_exec() multiple times with appropriate argu-
1.1 misha 5598: ments, you can mimic Perl's /g option, and it is in this kind of imple-
5599: mentation where \G can be useful.
5600:
1.6 misha 5601: Note, however, that PCRE's interpretation of \G, as the start of the
1.1 misha 5602: current match, is subtly different from Perl's, which defines it as the
1.6 misha 5603: end of the previous match. In Perl, these can be different when the
5604: previously matched string was empty. Because PCRE does just one match
1.1 misha 5605: at a time, it cannot reproduce this behaviour.
5606:
1.6 misha 5607: If all the alternatives of a pattern begin with \G, the expression is
1.1 misha 5608: anchored to the starting match position, and the "anchored" flag is set
5609: in the compiled regular expression.
5610:
5611:
5612: CIRCUMFLEX AND DOLLAR
5613:
1.6 misha 5614: The circumflex and dollar metacharacters are zero-width assertions.
5615: That is, they test for a particular condition being true without con-
5616: suming any characters from the subject string.
5617:
1.1 misha 5618: Outside a character class, in the default matching mode, the circumflex
1.6 misha 5619: character is an assertion that is true only if the current matching
5620: point is at the start of the subject string. If the startoffset argu-
5621: ment of pcre_exec() is non-zero, circumflex can never match if the
5622: PCRE_MULTILINE option is unset. Inside a character class, circumflex
1.1 misha 5623: has an entirely different meaning (see below).
5624:
1.6 misha 5625: Circumflex need not be the first character of the pattern if a number
5626: of alternatives are involved, but it should be the first thing in each
5627: alternative in which it appears if the pattern is ever to match that
5628: branch. If all possible alternatives start with a circumflex, that is,
5629: if the pattern is constrained to match only at the start of the sub-
5630: ject, it is said to be an "anchored" pattern. (There are also other
1.1 misha 5631: constructs that can cause a pattern to be anchored.)
5632:
1.6 misha 5633: The dollar character is an assertion that is true only if the current
5634: matching point is at the end of the subject string, or immediately
5635: before a newline at the end of the string (by default). Note, however,
5636: that it does not actually match the newline. Dollar need not be the
5637: last character of the pattern if a number of alternatives are involved,
5638: but it should be the last item in any branch in which it appears. Dol-
5639: lar has no special meaning in a character class.
1.1 misha 5640:
1.5 misha 5641: The meaning of dollar can be changed so that it matches only at the
5642: very end of the string, by setting the PCRE_DOLLAR_ENDONLY option at
1.1 misha 5643: compile time. This does not affect the \Z assertion.
5644:
5645: The meanings of the circumflex and dollar characters are changed if the
1.5 misha 5646: PCRE_MULTILINE option is set. When this is the case, a circumflex
5647: matches immediately after internal newlines as well as at the start of
5648: the subject string. It does not match after a newline that ends the
5649: string. A dollar matches before any newlines in the string, as well as
5650: at the very end, when PCRE_MULTILINE is set. When newline is specified
5651: as the two-character sequence CRLF, isolated CR and LF characters do
1.1 misha 5652: not indicate newlines.
5653:
1.5 misha 5654: For example, the pattern /^abc$/ matches the subject string "def\nabc"
5655: (where \n represents a newline) in multiline mode, but not otherwise.
5656: Consequently, patterns that are anchored in single line mode because
5657: all branches start with ^ are not anchored in multiline mode, and a
5658: match for circumflex is possible when the startoffset argument of
5659: pcre_exec() is non-zero. The PCRE_DOLLAR_ENDONLY option is ignored if
1.1 misha 5660: PCRE_MULTILINE is set.
5661:
1.5 misha 5662: Note that the sequences \A, \Z, and \z can be used to match the start
5663: and end of the subject in both modes, and if all branches of a pattern
5664: start with \A it is always anchored, whether or not PCRE_MULTILINE is
1.1 misha 5665: set.
5666:
5667:
1.4 misha 5668: FULL STOP (PERIOD, DOT) AND \N
1.1 misha 5669:
5670: Outside a character class, a dot in the pattern matches any one charac-
1.5 misha 5671: ter in the subject string except (by default) a character that signi-
5672: fies the end of a line.
1.1 misha 5673:
1.4 misha 5674: When a line ending is defined as a single character, dot never matches
5675: that character; when the two-character sequence CRLF is used, dot does
5676: not match CR if it is immediately followed by LF, but otherwise it
5677: matches all characters (including isolated CRs and LFs). When any Uni-
5678: code line endings are being recognized, dot does not match CR or LF or
1.1 misha 5679: any of the other line ending characters.
5680:
1.4 misha 5681: The behaviour of dot with regard to newlines can be changed. If the
5682: PCRE_DOTALL option is set, a dot matches any one character, without
1.1 misha 5683: exception. If the two-character sequence CRLF is present in the subject
5684: string, it takes two dots to match it.
5685:
1.4 misha 5686: The handling of dot is entirely independent of the handling of circum-
5687: flex and dollar, the only relationship being that they both involve
1.1 misha 5688: newlines. Dot has no special meaning in a character class.
5689:
1.4 misha 5690: The escape sequence \N behaves like a dot, except that it is not
5691: affected by the PCRE_DOTALL option. In other words, it matches any
1.5 misha 5692: character except one that signifies the end of a line. Perl also uses
5693: \N to match characters by name; PCRE does not support this.
1.4 misha 5694:
1.1 misha 5695:
1.5 misha 5696: MATCHING A SINGLE DATA UNIT
1.1 misha 5697:
1.5 misha 5698: Outside a character class, the escape sequence \C matches any one data
5699: unit, whether or not a UTF mode is set. In the 8-bit library, one data
1.6 misha 5700: unit is one byte; in the 16-bit library it is a 16-bit unit; in the
5701: 32-bit library it is a 32-bit unit. Unlike a dot, \C always matches
5702: line-ending characters. The feature is provided in Perl in order to
5703: match individual bytes in UTF-8 mode, but it is unclear how it can use-
5704: fully be used. Because \C breaks up characters into individual data
5705: units, matching one unit with \C in a UTF mode means that the rest of
5706: the string may start with a malformed UTF character. This has undefined
5707: results, because PCRE assumes that it is dealing with valid UTF strings
5708: (and by default it checks this at the start of processing unless the
5709: PCRE_NO_UTF8_CHECK, PCRE_NO_UTF16_CHECK or PCRE_NO_UTF32_CHECK option
5710: is used).
1.1 misha 5711:
1.5 misha 5712: PCRE does not allow \C to appear in lookbehind assertions (described
5713: below) in a UTF mode, because this would make it impossible to calcu-
1.1 misha 5714: late the length of the lookbehind.
5715:
1.5 misha 5716: In general, the \C escape sequence is best avoided. However, one way of
5717: using it that avoids the problem of malformed UTF characters is to use
5718: a lookahead to check the length of the next character, as in this pat-
5719: tern, which could be used with a UTF-8 string (ignore white space and
5720: line breaks):
5721:
5722: (?| (?=[\x00-\x7f])(\C) |
5723: (?=[\x80-\x{7ff}])(\C)(\C) |
5724: (?=[\x{800}-\x{ffff}])(\C)(\C)(\C) |
5725: (?=[\x{10000}-\x{1fffff}])(\C)(\C)(\C)(\C))
5726:
5727: A group that starts with (?| resets the capturing parentheses numbers
5728: in each alternative (see "Duplicate Subpattern Numbers" below). The
5729: assertions at the start of each branch check the next UTF-8 character
5730: for values whose encoding uses 1, 2, 3, or 4 bytes, respectively. The
5731: character's individual bytes are then captured by the appropriate num-
5732: ber of groups.
5733:
1.1 misha 5734:
5735: SQUARE BRACKETS AND CHARACTER CLASSES
5736:
5737: An opening square bracket introduces a character class, terminated by a
5738: closing square bracket. A closing square bracket on its own is not spe-
1.4 misha 5739: cial by default. However, if the PCRE_JAVASCRIPT_COMPAT option is set,
5740: a lone closing square bracket causes a compile-time error. If a closing
5741: square bracket is required as a member of the class, it should be the
5742: first data character in the class (after an initial circumflex, if
5743: present) or escaped with a backslash.
1.1 misha 5744:
1.5 misha 5745: A character class matches a single character in the subject. In a UTF
5746: mode, the character may be more than one data unit long. A matched
5747: character must be in the set of characters defined by the class, unless
5748: the first character in the class definition is a circumflex, in which
5749: case the subject character must not be in the set defined by the class.
5750: If a circumflex is actually required as a member of the class, ensure
5751: it is not the first character, or escape it with a backslash.
1.1 misha 5752:
1.4 misha 5753: For example, the character class [aeiou] matches any lower case vowel,
5754: while [^aeiou] matches any character that is not a lower case vowel.
1.1 misha 5755: Note that a circumflex is just a convenient notation for specifying the
1.4 misha 5756: characters that are in the class by enumerating those that are not. A
5757: class that starts with a circumflex is not an assertion; it still con-
5758: sumes a character from the subject string, and therefore it fails if
1.1 misha 5759: the current pointer is at the end of the string.
5760:
1.6 misha 5761: In UTF-8 (UTF-16, UTF-32) mode, characters with values greater than 255
1.5 misha 5762: (0xffff) can be included in a class as a literal string of data units,
5763: or by using the \x{ escaping mechanism.
1.1 misha 5764:
1.4 misha 5765: When caseless matching is set, any letters in a class represent both
5766: their upper case and lower case versions, so for example, a caseless
5767: [aeiou] matches "A" as well as "a", and a caseless [^aeiou] does not
1.5 misha 5768: match "A", whereas a caseful version would. In a UTF mode, PCRE always
1.4 misha 5769: understands the concept of case for characters whose values are less
5770: than 128, so caseless matching is always possible. For characters with
5771: higher values, the concept of case is supported if PCRE is compiled
5772: with Unicode property support, but not otherwise. If you want to use
1.5 misha 5773: caseless matching in a UTF mode for characters 128 and above, you must
1.4 misha 5774: ensure that PCRE is compiled with Unicode property support as well as
1.5 misha 5775: with UTF support.
1.4 misha 5776:
5777: Characters that might indicate line breaks are never treated in any
5778: special way when matching character classes, whatever line-ending
5779: sequence is in use, and whatever setting of the PCRE_DOTALL and
1.1 misha 5780: PCRE_MULTILINE options is used. A class such as [^a] always matches one
5781: of these characters.
5782:
1.4 misha 5783: The minus (hyphen) character can be used to specify a range of charac-
5784: ters in a character class. For example, [d-m] matches any letter
5785: between d and m, inclusive. If a minus character is required in a
5786: class, it must be escaped with a backslash or appear in a position
5787: where it cannot be interpreted as indicating a range, typically as the
1.7 ! moko 5788: first or last character in the class, or immediately after a range. For
! 5789: example, [b-d-z] matches letters in the range b to d, a hyphen charac-
! 5790: ter, or z.
1.1 misha 5791:
5792: It is not possible to have the literal character "]" as the end charac-
1.4 misha 5793: ter of a range. A pattern such as [W-]46] is interpreted as a class of
5794: two characters ("W" and "-") followed by a literal string "46]", so it
5795: would match "W46]" or "-46]". However, if the "]" is escaped with a
5796: backslash it is interpreted as the end of range, so [W-\]46] is inter-
5797: preted as a class containing a range followed by two other characters.
5798: The octal or hexadecimal representation of "]" can also be used to end
1.1 misha 5799: a range.
5800:
1.7 ! moko 5801: An error is generated if a POSIX character class (see below) or an
! 5802: escape sequence other than one that defines a single character appears
! 5803: at a point where a range ending character is expected. For example,
! 5804: [z-\xff] is valid, but [A-\d] and [A-[:digit:]] are not.
! 5805:
! 5806: Ranges operate in the collating sequence of character values. They can
! 5807: also be used for characters specified numerically, for example
! 5808: [\000-\037]. Ranges can include any characters that are valid for the
1.5 misha 5809: current mode.
1.1 misha 5810:
5811: If a range that includes letters is used when caseless matching is set,
5812: it matches the letters in either case. For example, [W-c] is equivalent
1.7 ! moko 5813: to [][\\^_`wxyzabc], matched caselessly, and in a non-UTF mode, if
! 5814: character tables for a French locale are in use, [\xc8-\xcb] matches
! 5815: accented E characters in both cases. In UTF modes, PCRE supports the
! 5816: concept of case for characters with values greater than 128 only when
1.1 misha 5817: it is compiled with Unicode property support.
5818:
1.7 ! moko 5819: The character escape sequences \d, \D, \h, \H, \p, \P, \s, \S, \v, \V,
1.4 misha 5820: \w, and \W may appear in a character class, and add the characters that
1.7 ! moko 5821: they match to the class. For example, [\dABCDEF] matches any hexadeci-
! 5822: mal digit. In UTF modes, the PCRE_UCP option affects the meanings of
! 5823: \d, \s, \w and their upper case partners, just as it does when they
! 5824: appear outside a character class, as described in the section entitled
1.4 misha 5825: "Generic character types" above. The escape sequence \b has a different
1.7 ! moko 5826: meaning inside a character class; it matches the backspace character.
! 5827: The sequences \B, \N, \R, and \X are not special inside a character
! 5828: class. Like any other unrecognized escape sequences, they are treated
! 5829: as the literal characters "B", "N", "R", and "X" by default, but cause
1.4 misha 5830: an error if the PCRE_EXTRA option is set.
5831:
1.7 ! moko 5832: A circumflex can conveniently be used with the upper case character
! 5833: types to specify a more restricted set of characters than the matching
! 5834: lower case type. For example, the class [^\W_] matches any letter or
1.4 misha 5835: digit, but not underscore, whereas [\w] includes underscore. A positive
5836: character class should be read as "something OR something OR ..." and a
5837: negative class as "NOT something AND NOT something AND NOT ...".
5838:
1.7 ! moko 5839: The only metacharacters that are recognized in character classes are
! 5840: backslash, hyphen (only where it can be interpreted as specifying a
! 5841: range), circumflex (only at the start), opening square bracket (only
! 5842: when it can be interpreted as introducing a POSIX class name, or for a
! 5843: special compatibility feature - see the next two sections), and the
! 5844: terminating closing square bracket. However, escaping other non-
! 5845: alphanumeric characters does no harm.
1.1 misha 5846:
5847:
5848: POSIX CHARACTER CLASSES
5849:
5850: Perl supports the POSIX notation for character classes. This uses names
1.4 misha 5851: enclosed by [: and :] within the enclosing square brackets. PCRE also
1.1 misha 5852: supports this notation. For example,
5853:
5854: [01[:alpha:]%]
5855:
5856: matches "0", "1", any alphabetic character, or "%". The supported class
1.4 misha 5857: names are:
1.1 misha 5858:
5859: alnum letters and digits
5860: alpha letters
5861: ascii character codes 0 - 127
5862: blank space or tab only
5863: cntrl control characters
5864: digit decimal digits (same as \d)
5865: graph printing characters, excluding space
5866: lower lower case letters
5867: print printing characters, including space
1.4 misha 5868: punct printing characters, excluding letters and digits and space
1.7 ! moko 5869: space white space (the same as \s from PCRE 8.34)
1.1 misha 5870: upper upper case letters
5871: word "word" characters (same as \w)
5872: xdigit hexadecimal digits
5873:
1.7 ! moko 5874: The default "space" characters are HT (9), LF (10), VT (11), FF (12),
! 5875: CR (13), and space (32). If locale-specific matching is taking place,
! 5876: the list of space characters may be different; there may be fewer or
! 5877: more of them. "Space" used to be different to \s, which did not include
! 5878: VT, for Perl compatibility. However, Perl changed at release 5.18, and
! 5879: PCRE followed at release 8.34. "Space" and \s now match the same set
! 5880: of characters.
1.1 misha 5881:
1.7 ! moko 5882: The name "word" is a Perl extension, and "blank" is a GNU extension
! 5883: from Perl 5.8. Another Perl extension is negation, which is indicated
1.1 misha 5884: by a ^ character after the colon. For example,
5885:
5886: [12[:^digit:]]
5887:
1.7 ! moko 5888: matches "1", "2", or any non-digit. PCRE (and Perl) also recognize the
1.1 misha 5889: POSIX syntax [.ch.] and [=ch=] where "ch" is a "collating element", but
5890: these are not supported, and an error is given if they are encountered.
5891:
1.7 ! moko 5892: By default, characters with values greater than 128 do not match any of
! 5893: the POSIX character classes. However, if the PCRE_UCP option is passed
! 5894: to pcre_compile(), some of the classes are changed so that Unicode
! 5895: character properties are used. This is achieved by replacing certain
! 5896: POSIX classes by other sequences, as follows:
1.4 misha 5897:
5898: [:alnum:] becomes \p{Xan}
5899: [:alpha:] becomes \p{L}
5900: [:blank:] becomes \h
5901: [:digit:] becomes \p{Nd}
5902: [:lower:] becomes \p{Ll}
5903: [:space:] becomes \p{Xps}
5904: [:upper:] becomes \p{Lu}
5905: [:word:] becomes \p{Xwd}
5906:
1.7 ! moko 5907: Negated versions, such as [:^alpha:] use \P instead of \p. Three other
! 5908: POSIX classes are handled specially in UCP mode:
! 5909:
! 5910: [:graph:] This matches characters that have glyphs that mark the page
! 5911: when printed. In Unicode property terms, it matches all char-
! 5912: acters with the L, M, N, P, S, or Cf properties, except for:
! 5913:
! 5914: U+061C Arabic Letter Mark
! 5915: U+180E Mongolian Vowel Separator
! 5916: U+2066 - U+2069 Various "isolate"s
! 5917:
! 5918:
! 5919: [:print:] This matches the same characters as [:graph:] plus space
! 5920: characters that are not controls, that is, characters with
! 5921: the Zs property.
! 5922:
! 5923: [:punct:] This matches all characters that have the Unicode P (punctua-
! 5924: tion) property, plus those characters whose code points are
! 5925: less than 128 that have the S (Symbol) property.
! 5926:
! 5927: The other POSIX classes are unchanged, and match only characters with
! 5928: code points less than 128.
! 5929:
! 5930:
! 5931: COMPATIBILITY FEATURE FOR WORD BOUNDARIES
! 5932:
! 5933: In the POSIX.2 compliant library that was included in 4.4BSD Unix, the
! 5934: ugly syntax [[:<:]] and [[:>:]] is used for matching "start of word"
! 5935: and "end of word". PCRE treats these items as follows:
! 5936:
! 5937: [[:<:]] is converted to \b(?=\w)
! 5938: [[:>:]] is converted to \b(?<=\w)
! 5939:
! 5940: Only these exact character sequences are recognized. A sequence such as
! 5941: [a[:<:]b] provokes error for an unrecognized POSIX class name. This
! 5942: support is not compatible with Perl. It is provided to help migrations
! 5943: from other environments, and is best not used in any new patterns. Note
! 5944: that \b matches at the start and the end of a word (see "Simple asser-
! 5945: tions" above), and in a Perl-style pattern the preceding or following
! 5946: character normally shows which is wanted, without the need for the
! 5947: assertions that are used above in order to give exactly the POSIX be-
! 5948: haviour.
1.1 misha 5949:
5950:
5951: VERTICAL BAR
5952:
1.7 ! moko 5953: Vertical bar characters are used to separate alternative patterns. For
1.1 misha 5954: example, the pattern
5955:
5956: gilbert|sullivan
5957:
1.7 ! moko 5958: matches either "gilbert" or "sullivan". Any number of alternatives may
! 5959: appear, and an empty alternative is permitted (matching the empty
1.1 misha 5960: string). The matching process tries each alternative in turn, from left
1.7 ! moko 5961: to right, and the first one that succeeds is used. If the alternatives
! 5962: are within a subpattern (defined below), "succeeds" means matching the
1.3 misha 5963: rest of the main pattern as well as the alternative in the subpattern.
1.1 misha 5964:
5965:
5966: INTERNAL OPTION SETTING
5967:
1.7 ! moko 5968: The settings of the PCRE_CASELESS, PCRE_MULTILINE, PCRE_DOTALL, and
! 5969: PCRE_EXTENDED options (which are Perl-compatible) can be changed from
! 5970: within the pattern by a sequence of Perl option letters enclosed
1.1 misha 5971: between "(?" and ")". The option letters are
5972:
5973: i for PCRE_CASELESS
5974: m for PCRE_MULTILINE
5975: s for PCRE_DOTALL
5976: x for PCRE_EXTENDED
5977:
5978: For example, (?im) sets caseless, multiline matching. It is also possi-
5979: ble to unset these options by preceding the letter with a hyphen, and a
1.7 ! moko 5980: combined setting and unsetting such as (?im-sx), which sets PCRE_CASE-
! 5981: LESS and PCRE_MULTILINE while unsetting PCRE_DOTALL and PCRE_EXTENDED,
! 5982: is also permitted. If a letter appears both before and after the
1.1 misha 5983: hyphen, the option is unset.
5984:
1.7 ! moko 5985: The PCRE-specific options PCRE_DUPNAMES, PCRE_UNGREEDY, and PCRE_EXTRA
! 5986: can be changed in the same way as the Perl-compatible options by using
1.1 misha 5987: the characters J, U and X respectively.
5988:
1.7 ! moko 5989: When one of these option changes occurs at top level (that is, not
! 5990: inside subpattern parentheses), the change applies to the remainder of
1.3 misha 5991: the pattern that follows. If the change is placed right at the start of
5992: a pattern, PCRE extracts it into the global options (and it will there-
5993: fore show up in data extracted by the pcre_fullinfo() function).
1.1 misha 5994:
1.7 ! moko 5995: An option change within a subpattern (see below for a description of
! 5996: subpatterns) affects only that part of the subpattern that follows it,
1.4 misha 5997: so
1.1 misha 5998:
5999: (a(?i)b)c
6000:
6001: matches abc and aBc and no other strings (assuming PCRE_CASELESS is not
1.7 ! moko 6002: used). By this means, options can be made to have different settings
! 6003: in different parts of the pattern. Any changes made in one alternative
! 6004: do carry on into subsequent branches within the same subpattern. For
1.1 misha 6005: example,
6006:
6007: (a(?i)b|c)
6008:
1.7 ! moko 6009: matches "ab", "aB", "c", and "C", even though when matching "C" the
! 6010: first branch is abandoned before the option setting. This is because
! 6011: the effects of option settings happen at compile time. There would be
1.1 misha 6012: some very weird behaviour otherwise.
6013:
1.7 ! moko 6014: Note: There are other PCRE-specific options that can be set by the
! 6015: application when the compiling or matching functions are called. In
! 6016: some cases the pattern can contain special leading sequences such as
! 6017: (*CRLF) to override what the application has set or what has been
! 6018: defaulted. Details are given in the section entitled "Newline
! 6019: sequences" above. There are also the (*UTF8), (*UTF16),(*UTF32), and
! 6020: (*UCP) leading sequences that can be used to set UTF and Unicode prop-
! 6021: erty modes; they are equivalent to setting the PCRE_UTF8, PCRE_UTF16,
! 6022: PCRE_UTF32 and the PCRE_UCP options, respectively. The (*UTF) sequence
! 6023: is a generic version that can be used with any of the libraries. How-
! 6024: ever, the application can set the PCRE_NEVER_UTF option, which locks
1.6 misha 6025: out the use of the (*UTF) sequences.
1.1 misha 6026:
6027:
6028: SUBPATTERNS
6029:
6030: Subpatterns are delimited by parentheses (round brackets), which can be
6031: nested. Turning part of a pattern into a subpattern does two things:
6032:
6033: 1. It localizes a set of alternatives. For example, the pattern
6034:
6035: cat(aract|erpillar|)
6036:
1.7 ! moko 6037: matches "cataract", "caterpillar", or "cat". Without the parentheses,
1.4 misha 6038: it would match "cataract", "erpillar" or an empty string.
1.1 misha 6039:
1.7 ! moko 6040: 2. It sets up the subpattern as a capturing subpattern. This means
! 6041: that, when the whole pattern matches, that portion of the subject
1.1 misha 6042: string that matched the subpattern is passed back to the caller via the
1.7 ! moko 6043: ovector argument of the matching function. (This applies only to the
! 6044: traditional matching functions; the DFA matching functions do not sup-
1.5 misha 6045: port capturing.)
6046:
6047: Opening parentheses are counted from left to right (starting from 1) to
1.7 ! moko 6048: obtain numbers for the capturing subpatterns. For example, if the
1.5 misha 6049: string "the red king" is matched against the pattern
1.1 misha 6050:
6051: the ((red|white) (king|queen))
6052:
6053: the captured substrings are "red king", "red", and "king", and are num-
6054: bered 1, 2, and 3, respectively.
6055:
1.7 ! moko 6056: The fact that plain parentheses fulfil two functions is not always
! 6057: helpful. There are often times when a grouping subpattern is required
! 6058: without a capturing requirement. If an opening parenthesis is followed
! 6059: by a question mark and a colon, the subpattern does not do any captur-
! 6060: ing, and is not counted when computing the number of any subsequent
! 6061: capturing subpatterns. For example, if the string "the white queen" is
1.1 misha 6062: matched against the pattern
6063:
6064: the ((?:red|white) (king|queen))
6065:
6066: the captured substrings are "white queen" and "queen", and are numbered
6067: 1 and 2. The maximum number of capturing subpatterns is 65535.
6068:
1.7 ! moko 6069: As a convenient shorthand, if any option settings are required at the
! 6070: start of a non-capturing subpattern, the option letters may appear
1.1 misha 6071: between the "?" and the ":". Thus the two patterns
6072:
6073: (?i:saturday|sunday)
6074: (?:(?i)saturday|sunday)
6075:
6076: match exactly the same set of strings. Because alternative branches are
1.7 ! moko 6077: tried from left to right, and options are not reset until the end of
! 6078: the subpattern is reached, an option setting in one branch does affect
! 6079: subsequent branches, so the above patterns match "SUNDAY" as well as
1.1 misha 6080: "Saturday".
6081:
6082:
6083: DUPLICATE SUBPATTERN NUMBERS
6084:
6085: Perl 5.10 introduced a feature whereby each alternative in a subpattern
1.7 ! moko 6086: uses the same numbers for its capturing parentheses. Such a subpattern
! 6087: starts with (?| and is itself a non-capturing subpattern. For example,
1.1 misha 6088: consider this pattern:
6089:
6090: (?|(Sat)ur|(Sun))day
6091:
1.7 ! moko 6092: Because the two alternatives are inside a (?| group, both sets of cap-
! 6093: turing parentheses are numbered one. Thus, when the pattern matches,
! 6094: you can look at captured substring number one, whichever alternative
! 6095: matched. This construct is useful when you want to capture part, but
1.1 misha 6096: not all, of one of a number of alternatives. Inside a (?| group, paren-
1.7 ! moko 6097: theses are numbered as usual, but the number is reset at the start of
! 6098: each branch. The numbers of any capturing parentheses that follow the
! 6099: subpattern start after the highest number used in any branch. The fol-
1.4 misha 6100: lowing example is taken from the Perl documentation. The numbers under-
1.1 misha 6101: neath show in which buffer the captured content will be stored.
6102:
6103: # before ---------------branch-reset----------- after
6104: / ( a ) (?| x ( y ) z | (p (q) r) | (t) u (v) ) ( z ) /x
6105: # 1 2 2 3 2 3 4
6106:
1.7 ! moko 6107: A back reference to a numbered subpattern uses the most recent value
! 6108: that is set for that number by any subpattern. The following pattern
1.4 misha 6109: matches "abcabc" or "defdef":
6110:
6111: /(?|(abc)|(def))\1/
1.1 misha 6112:
1.7 ! moko 6113: In contrast, a subroutine call to a numbered subpattern always refers
! 6114: to the first one in the pattern with the given number. The following
1.5 misha 6115: pattern matches "abcabc" or "defabc":
1.4 misha 6116:
6117: /(?|(abc)|(def))(?1)/
6118:
1.7 ! moko 6119: If a condition test for a subpattern's having matched refers to a non-
! 6120: unique number, the test is true if any of the subpatterns of that num-
1.4 misha 6121: ber have matched.
6122:
1.7 ! moko 6123: An alternative approach to using this "branch reset" feature is to use
1.1 misha 6124: duplicate named subpatterns, as described in the next section.
6125:
6126:
6127: NAMED SUBPATTERNS
6128:
1.7 ! moko 6129: Identifying capturing parentheses by number is simple, but it can be
! 6130: very hard to keep track of the numbers in complicated regular expres-
! 6131: sions. Furthermore, if an expression is modified, the numbers may
! 6132: change. To help with this difficulty, PCRE supports the naming of sub-
1.1 misha 6133: patterns. This feature was not added to Perl until release 5.10. Python
1.7 ! moko 6134: had the feature earlier, and PCRE introduced it at release 4.0, using
! 6135: the Python syntax. PCRE now supports both the Perl and the Python syn-
! 6136: tax. Perl allows identically numbered subpatterns to have different
1.4 misha 6137: names, but PCRE does not.
1.1 misha 6138:
1.7 ! moko 6139: In PCRE, a subpattern can be named in one of three ways: (?<name>...)
! 6140: or (?'name'...) as in Perl, or (?P<name>...) as in Python. References
! 6141: to capturing parentheses from other parts of the pattern, such as back
! 6142: references, recursion, and conditions, can be made by name as well as
1.1 misha 6143: by number.
6144:
1.7 ! moko 6145: Names consist of up to 32 alphanumeric characters and underscores, but
! 6146: must start with a non-digit. Named capturing parentheses are still
! 6147: allocated numbers as well as names, exactly as if the names were not
! 6148: present. The PCRE API provides function calls for extracting the name-
! 6149: to-number translation table from a compiled pattern. There is also a
! 6150: convenience function for extracting a captured substring by name.
1.1 misha 6151:
1.7 ! moko 6152: By default, a name must be unique within a pattern, but it is possible
1.1 misha 6153: to relax this constraint by setting the PCRE_DUPNAMES option at compile
1.7 ! moko 6154: time. (Duplicate names are also always permitted for subpatterns with
! 6155: the same number, set up as described in the previous section.) Dupli-
! 6156: cate names can be useful for patterns where only one instance of the
! 6157: named parentheses can match. Suppose you want to match the name of a
! 6158: weekday, either as a 3-letter abbreviation or as the full name, and in
1.1 misha 6159: both cases you want to extract the abbreviation. This pattern (ignoring
6160: the line breaks) does the job:
6161:
6162: (?<DN>Mon|Fri|Sun)(?:day)?|
6163: (?<DN>Tue)(?:sday)?|
6164: (?<DN>Wed)(?:nesday)?|
6165: (?<DN>Thu)(?:rsday)?|
6166: (?<DN>Sat)(?:urday)?
6167:
1.7 ! moko 6168: There are five capturing substrings, but only one is ever set after a
1.1 misha 6169: match. (An alternative way of solving this problem is to use a "branch
6170: reset" subpattern, as described in the previous section.)
6171:
1.7 ! moko 6172: The convenience function for extracting the data by name returns the
! 6173: substring for the first (and in this example, the only) subpattern of
! 6174: that name that matched. This saves searching to find which numbered
1.4 misha 6175: subpattern it was.
6176:
1.7 ! moko 6177: If you make a back reference to a non-unique named subpattern from
! 6178: elsewhere in the pattern, the subpatterns to which the name refers are
! 6179: checked in the order in which they appear in the overall pattern. The
! 6180: first one that is set is used for the reference. For example, this pat-
! 6181: tern matches both "foofoo" and "barbar" but not "foobar" or "barfoo":
! 6182:
! 6183: (?:(?<n>foo)|(?<n>bar))\k<n>
! 6184:
! 6185:
! 6186: If you make a subroutine call to a non-unique named subpattern, the one
! 6187: that corresponds to the first occurrence of the name is used. In the
! 6188: absence of duplicate numbers (see the previous section) this is the one
! 6189: with the lowest number.
! 6190:
! 6191: If you use a named reference in a condition test (see the section about
! 6192: conditions below), either to check whether a subpattern has matched, or
! 6193: to check for recursion, all subpatterns with the same name are tested.
! 6194: If the condition is true for any one of them, the overall condition is
! 6195: true. This is the same behaviour as testing by number. For further
! 6196: details of the interfaces for handling named subpatterns, see the
! 6197: pcreapi documentation.
1.1 misha 6198:
1.3 misha 6199: Warning: You cannot use different names to distinguish between two sub-
1.6 misha 6200: patterns with the same number because PCRE uses only the numbers when
1.4 misha 6201: matching. For this reason, an error is given at compile time if differ-
1.6 misha 6202: ent names are given to subpatterns with the same number. However, you
1.7 ! moko 6203: can always give the same name to subpatterns with the same number, even
! 6204: when PCRE_DUPNAMES is not set.
1.3 misha 6205:
1.1 misha 6206:
6207: REPETITION
6208:
1.6 misha 6209: Repetition is specified by quantifiers, which can follow any of the
1.1 misha 6210: following items:
6211:
6212: a literal data character
6213: the dot metacharacter
6214: the \C escape sequence
1.5 misha 6215: the \X escape sequence
1.1 misha 6216: the \R escape sequence
1.4 misha 6217: an escape such as \d or \pL that matches a single character
1.1 misha 6218: a character class
6219: a back reference (see next section)
1.5 misha 6220: a parenthesized subpattern (including assertions)
6221: a subroutine call to a subpattern (recursive or otherwise)
1.1 misha 6222:
1.6 misha 6223: The general repetition quantifier specifies a minimum and maximum num-
6224: ber of permitted matches, by giving the two numbers in curly brackets
6225: (braces), separated by a comma. The numbers must be less than 65536,
1.1 misha 6226: and the first must be less than or equal to the second. For example:
6227:
6228: z{2,4}
6229:
1.6 misha 6230: matches "zz", "zzz", or "zzzz". A closing brace on its own is not a
6231: special character. If the second number is omitted, but the comma is
6232: present, there is no upper limit; if the second number and the comma
6233: are both omitted, the quantifier specifies an exact number of required
1.1 misha 6234: matches. Thus
6235:
6236: [aeiou]{3,}
6237:
6238: matches at least 3 successive vowels, but may match many more, while
6239:
6240: \d{8}
6241:
1.6 misha 6242: matches exactly 8 digits. An opening curly bracket that appears in a
6243: position where a quantifier is not allowed, or one that does not match
6244: the syntax of a quantifier, is taken as a literal character. For exam-
1.1 misha 6245: ple, {,6} is not a quantifier, but a literal string of four characters.
6246:
1.5 misha 6247: In UTF modes, quantifiers apply to characters rather than to individual
1.6 misha 6248: data units. Thus, for example, \x{100}{2} matches two characters, each
1.5 misha 6249: of which is represented by a two-byte sequence in a UTF-8 string. Simi-
1.6 misha 6250: larly, \X{3} matches three Unicode extended grapheme clusters, each of
6251: which may be several data units long (and they may be of different
6252: lengths).
1.1 misha 6253:
6254: The quantifier {0} is permitted, causing the expression to behave as if
6255: the previous item and the quantifier were not present. This may be use-
1.5 misha 6256: ful for subpatterns that are referenced as subroutines from elsewhere
1.4 misha 6257: in the pattern (but see also the section entitled "Defining subpatterns
1.5 misha 6258: for use by reference only" below). Items other than subpatterns that
1.4 misha 6259: have a {0} quantifier are omitted from the compiled pattern.
1.1 misha 6260:
1.5 misha 6261: For convenience, the three most common quantifiers have single-charac-
1.1 misha 6262: ter abbreviations:
6263:
6264: * is equivalent to {0,}
6265: + is equivalent to {1,}
6266: ? is equivalent to {0,1}
6267:
1.5 misha 6268: It is possible to construct infinite loops by following a subpattern
1.1 misha 6269: that can match no characters with a quantifier that has no upper limit,
6270: for example:
6271:
6272: (a?)*
6273:
6274: Earlier versions of Perl and PCRE used to give an error at compile time
1.5 misha 6275: for such patterns. However, because there are cases where this can be
6276: useful, such patterns are now accepted, but if any repetition of the
6277: subpattern does in fact match no characters, the loop is forcibly bro-
1.1 misha 6278: ken.
6279:
1.5 misha 6280: By default, the quantifiers are "greedy", that is, they match as much
6281: as possible (up to the maximum number of permitted times), without
6282: causing the rest of the pattern to fail. The classic example of where
1.1 misha 6283: this gives problems is in trying to match comments in C programs. These
1.5 misha 6284: appear between /* and */ and within the comment, individual * and /
6285: characters may appear. An attempt to match C comments by applying the
1.1 misha 6286: pattern
6287:
6288: /\*.*\*/
6289:
6290: to the string
6291:
6292: /* first comment */ not comment /* second comment */
6293:
1.5 misha 6294: fails, because it matches the entire string owing to the greediness of
1.1 misha 6295: the .* item.
6296:
1.5 misha 6297: However, if a quantifier is followed by a question mark, it ceases to
1.1 misha 6298: be greedy, and instead matches the minimum number of times possible, so
6299: the pattern
6300:
6301: /\*.*?\*/
6302:
1.5 misha 6303: does the right thing with the C comments. The meaning of the various
6304: quantifiers is not otherwise changed, just the preferred number of
6305: matches. Do not confuse this use of question mark with its use as a
6306: quantifier in its own right. Because it has two uses, it can sometimes
1.1 misha 6307: appear doubled, as in
6308:
6309: \d??\d
6310:
6311: which matches one digit by preference, but can match two if that is the
6312: only way the rest of the pattern matches.
6313:
1.5 misha 6314: If the PCRE_UNGREEDY option is set (an option that is not available in
6315: Perl), the quantifiers are not greedy by default, but individual ones
6316: can be made greedy by following them with a question mark. In other
1.1 misha 6317: words, it inverts the default behaviour.
6318:
1.5 misha 6319: When a parenthesized subpattern is quantified with a minimum repeat
6320: count that is greater than 1 or with a limited maximum, more memory is
6321: required for the compiled pattern, in proportion to the size of the
1.1 misha 6322: minimum or maximum.
6323:
6324: If a pattern starts with .* or .{0,} and the PCRE_DOTALL option (equiv-
1.5 misha 6325: alent to Perl's /s) is set, thus allowing the dot to match newlines,
6326: the pattern is implicitly anchored, because whatever follows will be
6327: tried against every character position in the subject string, so there
6328: is no point in retrying the overall match at any position after the
6329: first. PCRE normally treats such a pattern as though it were preceded
1.1 misha 6330: by \A.
6331:
1.5 misha 6332: In cases where it is known that the subject string contains no new-
6333: lines, it is worth setting PCRE_DOTALL in order to obtain this opti-
1.1 misha 6334: mization, or alternatively using ^ to indicate anchoring explicitly.
6335:
1.6 misha 6336: However, there are some cases where the optimization cannot be used.
1.4 misha 6337: When .* is inside capturing parentheses that are the subject of a back
6338: reference elsewhere in the pattern, a match at the start may fail where
6339: a later one succeeds. Consider, for example:
1.1 misha 6340:
6341: (.*)abc\1
6342:
1.5 misha 6343: If the subject is "xyz123abc123" the match point is the fourth charac-
1.1 misha 6344: ter. For this reason, such a pattern is not implicitly anchored.
6345:
1.6 misha 6346: Another case where implicit anchoring is not applied is when the lead-
6347: ing .* is inside an atomic group. Once again, a match at the start may
6348: fail where a later one succeeds. Consider this pattern:
6349:
6350: (?>.*?a)b
6351:
6352: It matches "ab" in the subject "aab". The use of the backtracking con-
6353: trol verbs (*PRUNE) and (*SKIP) also disable this optimization.
6354:
1.1 misha 6355: When a capturing subpattern is repeated, the value captured is the sub-
6356: string that matched the final iteration. For example, after
6357:
6358: (tweedle[dume]{3}\s*)+
6359:
6360: has matched "tweedledum tweedledee" the value of the captured substring
1.6 misha 6361: is "tweedledee". However, if there are nested capturing subpatterns,
6362: the corresponding captured values may have been set in previous itera-
1.1 misha 6363: tions. For example, after
6364:
6365: /(a|(b))+/
6366:
6367: matches "aba" the value of the second captured substring is "b".
6368:
6369:
6370: ATOMIC GROUPING AND POSSESSIVE QUANTIFIERS
6371:
1.6 misha 6372: With both maximizing ("greedy") and minimizing ("ungreedy" or "lazy")
6373: repetition, failure of what follows normally causes the repeated item
6374: to be re-evaluated to see if a different number of repeats allows the
6375: rest of the pattern to match. Sometimes it is useful to prevent this,
6376: either to change the nature of the match, or to cause it fail earlier
6377: than it otherwise might, when the author of the pattern knows there is
1.1 misha 6378: no point in carrying on.
6379:
1.6 misha 6380: Consider, for example, the pattern \d+foo when applied to the subject
1.1 misha 6381: line
6382:
6383: 123456bar
6384:
6385: After matching all 6 digits and then failing to match "foo", the normal
1.6 misha 6386: action of the matcher is to try again with only 5 digits matching the
6387: \d+ item, and then with 4, and so on, before ultimately failing.
6388: "Atomic grouping" (a term taken from Jeffrey Friedl's book) provides
6389: the means for specifying that once a subpattern has matched, it is not
1.1 misha 6390: to be re-evaluated in this way.
6391:
1.6 misha 6392: If we use atomic grouping for the previous example, the matcher gives
6393: up immediately on failing to match "foo" the first time. The notation
1.1 misha 6394: is a kind of special parenthesis, starting with (?> as in this example:
6395:
6396: (?>\d+)foo
6397:
1.6 misha 6398: This kind of parenthesis "locks up" the part of the pattern it con-
6399: tains once it has matched, and a failure further into the pattern is
6400: prevented from backtracking into it. Backtracking past it to previous
1.1 misha 6401: items, however, works as normal.
6402:
1.6 misha 6403: An alternative description is that a subpattern of this type matches
6404: the string of characters that an identical standalone pattern would
1.1 misha 6405: match, if anchored at the current point in the subject string.
6406:
6407: Atomic grouping subpatterns are not capturing subpatterns. Simple cases
6408: such as the above example can be thought of as a maximizing repeat that
1.6 misha 6409: must swallow everything it can. So, while both \d+ and \d+? are pre-
6410: pared to adjust the number of digits they match in order to make the
1.1 misha 6411: rest of the pattern match, (?>\d+) can only match an entire sequence of
6412: digits.
6413:
1.6 misha 6414: Atomic groups in general can of course contain arbitrarily complicated
6415: subpatterns, and can be nested. However, when the subpattern for an
1.1 misha 6416: atomic group is just a single repeated item, as in the example above, a
1.6 misha 6417: simpler notation, called a "possessive quantifier" can be used. This
6418: consists of an additional + character following a quantifier. Using
1.1 misha 6419: this notation, the previous example can be rewritten as
6420:
6421: \d++foo
6422:
6423: Note that a possessive quantifier can be used with an entire group, for
6424: example:
6425:
6426: (abc|xyz){2,3}+
6427:
1.6 misha 6428: Possessive quantifiers are always greedy; the setting of the
1.1 misha 6429: PCRE_UNGREEDY option is ignored. They are a convenient notation for the
1.6 misha 6430: simpler forms of atomic group. However, there is no difference in the
6431: meaning of a possessive quantifier and the equivalent atomic group,
6432: though there may be a performance difference; possessive quantifiers
1.1 misha 6433: should be slightly faster.
6434:
1.6 misha 6435: The possessive quantifier syntax is an extension to the Perl 5.8 syn-
6436: tax. Jeffrey Friedl originated the idea (and the name) in the first
1.1 misha 6437: edition of his book. Mike McCloskey liked it, so implemented it when he
1.6 misha 6438: built Sun's Java package, and PCRE copied it from there. It ultimately
1.1 misha 6439: found its way into Perl at release 5.10.
6440:
6441: PCRE has an optimization that automatically "possessifies" certain sim-
1.6 misha 6442: ple pattern constructs. For example, the sequence A+B is treated as
6443: A++B because there is no point in backtracking into a sequence of A's
1.1 misha 6444: when B must follow.
6445:
1.6 misha 6446: When a pattern contains an unlimited repeat inside a subpattern that
6447: can itself be repeated an unlimited number of times, the use of an
6448: atomic group is the only way to avoid some failing matches taking a
1.1 misha 6449: very long time indeed. The pattern
6450:
6451: (\D+|<\d+>)*[!?]
6452:
1.6 misha 6453: matches an unlimited number of substrings that either consist of non-
6454: digits, or digits enclosed in <>, followed by either ! or ?. When it
1.1 misha 6455: matches, it runs quickly. However, if it is applied to
6456:
6457: aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
6458:
1.6 misha 6459: it takes a long time before reporting failure. This is because the
6460: string can be divided between the internal \D+ repeat and the external
6461: * repeat in a large number of ways, and all have to be tried. (The
6462: example uses [!?] rather than a single character at the end, because
6463: both PCRE and Perl have an optimization that allows for fast failure
6464: when a single character is used. They remember the last single charac-
6465: ter that is required for a match, and fail early if it is not present
6466: in the string.) If the pattern is changed so that it uses an atomic
1.1 misha 6467: group, like this:
6468:
6469: ((?>\D+)|<\d+>)*[!?]
6470:
1.3 misha 6471: sequences of non-digits cannot be broken, and failure happens quickly.
1.1 misha 6472:
6473:
6474: BACK REFERENCES
6475:
6476: Outside a character class, a backslash followed by a digit greater than
6477: 0 (and possibly further digits) is a back reference to a capturing sub-
1.6 misha 6478: pattern earlier (that is, to its left) in the pattern, provided there
1.1 misha 6479: have been that many previous capturing left parentheses.
6480:
6481: However, if the decimal number following the backslash is less than 10,
1.6 misha 6482: it is always taken as a back reference, and causes an error only if
6483: there are not that many capturing left parentheses in the entire pat-
6484: tern. In other words, the parentheses that are referenced need not be
6485: to the left of the reference for numbers less than 10. A "forward back
6486: reference" of this type can make sense when a repetition is involved
6487: and the subpattern to the right has participated in an earlier itera-
1.1 misha 6488: tion.
6489:
1.6 misha 6490: It is not possible to have a numerical "forward back reference" to a
6491: subpattern whose number is 10 or more using this syntax because a
6492: sequence such as \50 is interpreted as a character defined in octal.
1.1 misha 6493: See the subsection entitled "Non-printing characters" above for further
1.6 misha 6494: details of the handling of digits following a backslash. There is no
6495: such problem when named parentheses are used. A back reference to any
1.1 misha 6496: subpattern is possible using named parentheses (see below).
6497:
1.6 misha 6498: Another way of avoiding the ambiguity inherent in the use of digits
6499: following a backslash is to use the \g escape sequence. This escape
1.4 misha 6500: must be followed by an unsigned number or a negative number, optionally
6501: enclosed in braces. These examples are all identical:
1.1 misha 6502:
6503: (ring), \1
6504: (ring), \g1
6505: (ring), \g{1}
6506:
1.6 misha 6507: An unsigned number specifies an absolute reference without the ambigu-
1.1 misha 6508: ity that is present in the older syntax. It is also useful when literal
6509: digits follow the reference. A negative number is a relative reference.
6510: Consider this example:
6511:
6512: (abc(def)ghi)\g{-1}
6513:
6514: The sequence \g{-1} is a reference to the most recently started captur-
1.4 misha 6515: ing subpattern before \g, that is, is it equivalent to \2 in this exam-
1.6 misha 6516: ple. Similarly, \g{-2} would be equivalent to \1. The use of relative
6517: references can be helpful in long patterns, and also in patterns that
6518: are created by joining together fragments that contain references
1.4 misha 6519: within themselves.
1.1 misha 6520:
1.6 misha 6521: A back reference matches whatever actually matched the capturing sub-
6522: pattern in the current subject string, rather than anything matching
1.1 misha 6523: the subpattern itself (see "Subpatterns as subroutines" below for a way
6524: of doing that). So the pattern
6525:
6526: (sens|respons)e and \1ibility
6527:
1.6 misha 6528: matches "sense and sensibility" and "response and responsibility", but
6529: not "sense and responsibility". If caseful matching is in force at the
6530: time of the back reference, the case of letters is relevant. For exam-
1.1 misha 6531: ple,
6532:
6533: ((?i)rah)\s+\1
6534:
1.6 misha 6535: matches "rah rah" and "RAH RAH", but not "RAH rah", even though the
1.1 misha 6536: original capturing subpattern is matched caselessly.
6537:
1.6 misha 6538: There are several different ways of writing back references to named
6539: subpatterns. The .NET syntax \k{name} and the Perl syntax \k<name> or
6540: \k'name' are supported, as is the Python syntax (?P=name). Perl 5.10's
1.1 misha 6541: unified back reference syntax, in which \g can be used for both numeric
1.6 misha 6542: and named references, is also supported. We could rewrite the above
1.1 misha 6543: example in any of the following ways:
6544:
6545: (?<p1>(?i)rah)\s+\k<p1>
6546: (?'p1'(?i)rah)\s+\k{p1}
6547: (?P<p1>(?i)rah)\s+(?P=p1)
6548: (?<p1>(?i)rah)\s+\g{p1}
6549:
1.6 misha 6550: A subpattern that is referenced by name may appear in the pattern
1.1 misha 6551: before or after the reference.
6552:
1.6 misha 6553: There may be more than one back reference to the same subpattern. If a
6554: subpattern has not actually been used in a particular match, any back
1.4 misha 6555: references to it always fail by default. For example, the pattern
1.1 misha 6556:
6557: (a|(bc))\2
6558:
1.6 misha 6559: always fails if it starts to match "a" rather than "bc". However, if
1.4 misha 6560: the PCRE_JAVASCRIPT_COMPAT option is set at compile time, a back refer-
6561: ence to an unset value matches an empty string.
6562:
1.6 misha 6563: Because there may be many capturing parentheses in a pattern, all dig-
6564: its following a backslash are taken as part of a potential back refer-
6565: ence number. If the pattern continues with a digit character, some
6566: delimiter must be used to terminate the back reference. If the
6567: PCRE_EXTENDED option is set, this can be white space. Otherwise, the
6568: \g{ syntax or an empty comment (see "Comments" below) can be used.
1.4 misha 6569:
6570: Recursive back references
1.1 misha 6571:
1.6 misha 6572: A back reference that occurs inside the parentheses to which it refers
6573: fails when the subpattern is first used, so, for example, (a\1) never
6574: matches. However, such references can be useful inside repeated sub-
1.1 misha 6575: patterns. For example, the pattern
6576:
6577: (a|b\1)+
6578:
6579: matches any number of "a"s and also "aba", "ababbaa" etc. At each iter-
1.6 misha 6580: ation of the subpattern, the back reference matches the character
6581: string corresponding to the previous iteration. In order for this to
6582: work, the pattern must be such that the first iteration does not need
6583: to match the back reference. This can be done using alternation, as in
1.1 misha 6584: the example above, or by a quantifier with a minimum of zero.
6585:
1.6 misha 6586: Back references of this type cause the group that they reference to be
6587: treated as an atomic group. Once the whole group has been matched, a
6588: subsequent matching failure cannot cause backtracking into the middle
1.4 misha 6589: of the group.
6590:
1.1 misha 6591:
6592: ASSERTIONS
6593:
1.6 misha 6594: An assertion is a test on the characters following or preceding the
6595: current matching point that does not actually consume any characters.
6596: The simple assertions coded as \b, \B, \A, \G, \Z, \z, ^ and $ are
1.1 misha 6597: described above.
6598:
1.6 misha 6599: More complicated assertions are coded as subpatterns. There are two
6600: kinds: those that look ahead of the current position in the subject
6601: string, and those that look behind it. An assertion subpattern is
6602: matched in the normal way, except that it does not cause the current
1.1 misha 6603: matching position to be changed.
6604:
1.6 misha 6605: Assertion subpatterns are not capturing subpatterns. If such an asser-
6606: tion contains capturing subpatterns within it, these are counted for
6607: the purposes of numbering the capturing subpatterns in the whole pat-
6608: tern. However, substring capturing is carried out only for positive
6609: assertions. (Perl sometimes, but not always, does do capturing in nega-
6610: tive assertions.)
1.5 misha 6611:
6612: For compatibility with Perl, assertion subpatterns may be repeated;
6613: though it makes no sense to assert the same thing several times, the
6614: side effect of capturing parentheses may occasionally be useful. In
6615: practice, there only three cases:
6616:
6617: (1) If the quantifier is {0}, the assertion is never obeyed during
6618: matching. However, it may contain internal capturing parenthesized
6619: groups that are called from elsewhere via the subroutine mechanism.
6620:
6621: (2) If quantifier is {0,n} where n is greater than zero, it is treated
6622: as if it were {0,1}. At run time, the rest of the pattern match is
6623: tried with and without the assertion, the order depending on the greed-
6624: iness of the quantifier.
6625:
6626: (3) If the minimum repetition is greater than zero, the quantifier is
6627: ignored. The assertion is obeyed just once when encountered during
6628: matching.
1.1 misha 6629:
6630: Lookahead assertions
6631:
6632: Lookahead assertions start with (?= for positive assertions and (?! for
6633: negative assertions. For example,
6634:
6635: \w+(?=;)
6636:
1.5 misha 6637: matches a word followed by a semicolon, but does not include the semi-
1.1 misha 6638: colon in the match, and
6639:
6640: foo(?!bar)
6641:
1.5 misha 6642: matches any occurrence of "foo" that is not followed by "bar". Note
1.1 misha 6643: that the apparently similar pattern
6644:
6645: (?!foo)bar
6646:
1.5 misha 6647: does not find an occurrence of "bar" that is preceded by something
6648: other than "foo"; it finds any occurrence of "bar" whatsoever, because
1.1 misha 6649: the assertion (?!foo) is always true when the next three characters are
6650: "bar". A lookbehind assertion is needed to achieve the other effect.
6651:
6652: If you want to force a matching failure at some point in a pattern, the
1.5 misha 6653: most convenient way to do it is with (?!) because an empty string
6654: always matches, so an assertion that requires there not to be an empty
1.4 misha 6655: string must always fail. The backtracking control verb (*FAIL) or (*F)
6656: is a synonym for (?!).
1.1 misha 6657:
6658: Lookbehind assertions
6659:
1.5 misha 6660: Lookbehind assertions start with (?<= for positive assertions and (?<!
1.1 misha 6661: for negative assertions. For example,
6662:
6663: (?<!foo)bar
6664:
1.5 misha 6665: does find an occurrence of "bar" that is not preceded by "foo". The
6666: contents of a lookbehind assertion are restricted such that all the
1.1 misha 6667: strings it matches must have a fixed length. However, if there are sev-
1.5 misha 6668: eral top-level alternatives, they do not all have to have the same
1.1 misha 6669: fixed length. Thus
6670:
6671: (?<=bullock|donkey)
6672:
6673: is permitted, but
6674:
6675: (?<!dogs?|cats?)
6676:
1.5 misha 6677: causes an error at compile time. Branches that match different length
6678: strings are permitted only at the top level of a lookbehind assertion.
1.4 misha 6679: This is an extension compared with Perl, which requires all branches to
6680: match the same length of string. An assertion such as
1.1 misha 6681:
6682: (?<=ab(c|de))
6683:
1.5 misha 6684: is not permitted, because its single top-level branch can match two
1.4 misha 6685: different lengths, but it is acceptable to PCRE if rewritten to use two
6686: top-level branches:
1.1 misha 6687:
6688: (?<=abc|abde)
6689:
1.5 misha 6690: In some cases, the escape sequence \K (see above) can be used instead
1.4 misha 6691: of a lookbehind assertion to get round the fixed-length restriction.
1.1 misha 6692:
1.5 misha 6693: The implementation of lookbehind assertions is, for each alternative,
6694: to temporarily move the current position back by the fixed length and
1.1 misha 6695: then try to match. If there are insufficient characters before the cur-
6696: rent position, the assertion fails.
6697:
1.5 misha 6698: In a UTF mode, PCRE does not allow the \C escape (which matches a sin-
6699: gle data unit even in a UTF mode) to appear in lookbehind assertions,
6700: because it makes it impossible to calculate the length of the lookbe-
6701: hind. The \X and \R escapes, which can match different numbers of data
6702: units, are also not permitted.
1.1 misha 6703:
1.4 misha 6704: "Subroutine" calls (see below) such as (?2) or (?&X) are permitted in
6705: lookbehinds, as long as the subpattern matches a fixed-length string.
6706: Recursion, however, is not supported.
6707:
1.3 misha 6708: Possessive quantifiers can be used in conjunction with lookbehind
1.4 misha 6709: assertions to specify efficient matching of fixed-length strings at the
6710: end of subject strings. Consider a simple pattern such as
1.1 misha 6711:
6712: abcd$
6713:
1.3 misha 6714: when applied to a long string that does not match. Because matching
1.1 misha 6715: proceeds from left to right, PCRE will look for each "a" in the subject
1.3 misha 6716: and then see if what follows matches the rest of the pattern. If the
1.1 misha 6717: pattern is specified as
6718:
6719: ^.*abcd$
6720:
1.3 misha 6721: the initial .* matches the entire string at first, but when this fails
1.1 misha 6722: (because there is no following "a"), it backtracks to match all but the
1.3 misha 6723: last character, then all but the last two characters, and so on. Once
6724: again the search for "a" covers the entire string, from right to left,
1.1 misha 6725: so we are no better off. However, if the pattern is written as
6726:
6727: ^.*+(?<=abcd)
6728:
1.3 misha 6729: there can be no backtracking for the .*+ item; it can match only the
6730: entire string. The subsequent lookbehind assertion does a single test
6731: on the last four characters. If it fails, the match fails immediately.
6732: For long strings, this approach makes a significant difference to the
1.1 misha 6733: processing time.
6734:
6735: Using multiple assertions
6736:
6737: Several assertions (of any sort) may occur in succession. For example,
6738:
6739: (?<=\d{3})(?<!999)foo
6740:
1.3 misha 6741: matches "foo" preceded by three digits that are not "999". Notice that
6742: each of the assertions is applied independently at the same point in
6743: the subject string. First there is a check that the previous three
6744: characters are all digits, and then there is a check that the same
1.1 misha 6745: three characters are not "999". This pattern does not match "foo" pre-
1.3 misha 6746: ceded by six characters, the first of which are digits and the last
6747: three of which are not "999". For example, it doesn't match "123abc-
1.1 misha 6748: foo". A pattern to do that is
6749:
6750: (?<=\d{3}...)(?<!999)foo
6751:
1.3 misha 6752: This time the first assertion looks at the preceding six characters,
1.1 misha 6753: checking that the first three are digits, and then the second assertion
6754: checks that the preceding three characters are not "999".
6755:
6756: Assertions can be nested in any combination. For example,
6757:
6758: (?<=(?<!foo)bar)baz
6759:
1.3 misha 6760: matches an occurrence of "baz" that is preceded by "bar" which in turn
1.1 misha 6761: is not preceded by "foo", while
6762:
6763: (?<=\d{3}(?!999)...)foo
6764:
1.3 misha 6765: is another pattern that matches "foo" preceded by three digits and any
1.1 misha 6766: three characters that are not "999".
6767:
6768:
6769: CONDITIONAL SUBPATTERNS
6770:
1.3 misha 6771: It is possible to cause the matching process to obey a subpattern con-
6772: ditionally or to choose between two alternative subpatterns, depending
1.4 misha 6773: on the result of an assertion, or whether a specific capturing subpat-
6774: tern has already been matched. The two possible forms of conditional
6775: subpattern are:
1.1 misha 6776:
6777: (?(condition)yes-pattern)
6778: (?(condition)yes-pattern|no-pattern)
6779:
1.3 misha 6780: If the condition is satisfied, the yes-pattern is used; otherwise the
6781: no-pattern (if present) is used. If there are more than two alterna-
1.4 misha 6782: tives in the subpattern, a compile-time error occurs. Each of the two
6783: alternatives may itself contain nested subpatterns of any form, includ-
6784: ing conditional subpatterns; the restriction to two alternatives
6785: applies only at the level of the condition. This pattern fragment is an
6786: example where the alternatives are complex:
6787:
6788: (?(1) (A|B|C) | (D | (?(2)E|F) | E) )
6789:
1.1 misha 6790:
1.3 misha 6791: There are four kinds of condition: references to subpatterns, refer-
1.1 misha 6792: ences to recursion, a pseudo-condition called DEFINE, and assertions.
6793:
6794: Checking for a used subpattern by number
6795:
1.3 misha 6796: If the text between the parentheses consists of a sequence of digits,
1.4 misha 6797: the condition is true if a capturing subpattern of that number has pre-
6798: viously matched. If there is more than one capturing subpattern with
6799: the same number (see the earlier section about duplicate subpattern
6800: numbers), the condition is true if any of them have matched. An alter-
6801: native notation is to precede the digits with a plus or minus sign. In
6802: this case, the subpattern number is relative rather than absolute. The
6803: most recently opened parentheses can be referenced by (?(-1), the next
6804: most recent by (?(-2), and so on. Inside loops it can also make sense
6805: to refer to subsequent groups. The next parentheses to be opened can be
6806: referenced as (?(+1), and so on. (The value zero in any of these forms
6807: is not used; it provokes a compile-time error.)
1.1 misha 6808:
1.3 misha 6809: Consider the following pattern, which contains non-significant white
1.1 misha 6810: space to make it more readable (assume the PCRE_EXTENDED option) and to
6811: divide it into three parts for ease of discussion:
6812:
6813: ( \( )? [^()]+ (?(1) \) )
6814:
1.3 misha 6815: The first part matches an optional opening parenthesis, and if that
1.1 misha 6816: character is present, sets it as the first captured substring. The sec-
1.3 misha 6817: ond part matches one or more characters that are not parentheses. The
1.4 misha 6818: third part is a conditional subpattern that tests whether or not the
6819: first set of parentheses matched. If they did, that is, if subject
6820: started with an opening parenthesis, the condition is true, and so the
6821: yes-pattern is executed and a closing parenthesis is required. Other-
6822: wise, since no-pattern is not present, the subpattern matches nothing.
6823: In other words, this pattern matches a sequence of non-parentheses,
1.1 misha 6824: optionally enclosed in parentheses.
6825:
1.3 misha 6826: If you were embedding this pattern in a larger one, you could use a
1.1 misha 6827: relative reference:
6828:
6829: ...other stuff... ( \( )? [^()]+ (?(-1) \) ) ...
6830:
1.3 misha 6831: This makes the fragment independent of the parentheses in the larger
1.1 misha 6832: pattern.
6833:
6834: Checking for a used subpattern by name
6835:
1.3 misha 6836: Perl uses the syntax (?(<name>)...) or (?('name')...) to test for a
6837: used subpattern by name. For compatibility with earlier versions of
6838: PCRE, which had this facility before Perl, the syntax (?(name)...) is
1.7 ! moko 6839: also recognized.
1.1 misha 6840:
6841: Rewriting the above example to use a named subpattern gives this:
6842:
6843: (?<OPEN> \( )? [^()]+ (?(<OPEN>) \) )
6844:
1.7 ! moko 6845: If the name used in a condition of this kind is a duplicate, the test
! 6846: is applied to all subpatterns of the same name, and is true if any one
1.4 misha 6847: of them has matched.
1.1 misha 6848:
6849: Checking for pattern recursion
6850:
6851: If the condition is the string (R), and there is no subpattern with the
1.7 ! moko 6852: name R, the condition is true if a recursive call to the whole pattern
1.1 misha 6853: or any subpattern has been made. If digits or a name preceded by amper-
6854: sand follow the letter R, for example:
6855:
6856: (?(R3)...) or (?(R&name)...)
6857:
1.4 misha 6858: the condition is true if the most recent recursion is into a subpattern
6859: whose number or name is given. This condition does not check the entire
1.7 ! moko 6860: recursion stack. If the name used in a condition of this kind is a
1.4 misha 6861: duplicate, the test is applied to all subpatterns of the same name, and
6862: is true if any one of them is the most recent recursion.
1.1 misha 6863:
1.7 ! moko 6864: At "top level", all these recursion test conditions are false. The
1.4 misha 6865: syntax for recursive patterns is described below.
1.1 misha 6866:
6867: Defining subpatterns for use by reference only
6868:
1.7 ! moko 6869: If the condition is the string (DEFINE), and there is no subpattern
! 6870: with the name DEFINE, the condition is always false. In this case,
! 6871: there may be only one alternative in the subpattern. It is always
! 6872: skipped if control reaches this point in the pattern; the idea of
! 6873: DEFINE is that it can be used to define subroutines that can be refer-
! 6874: enced from elsewhere. (The use of subroutines is described below.) For
! 6875: example, a pattern to match an IPv4 address such as "192.168.23.245"
1.6 misha 6876: could be written like this (ignore white space and line breaks):
1.1 misha 6877:
6878: (?(DEFINE) (?<byte> 2[0-4]\d | 25[0-5] | 1\d\d | [1-9]?\d) )
6879: \b (?&byte) (\.(?&byte)){3} \b
6880:
1.7 ! moko 6881: The first part of the pattern is a DEFINE group inside which a another
! 6882: group named "byte" is defined. This matches an individual component of
! 6883: an IPv4 address (a number less than 256). When matching takes place,
! 6884: this part of the pattern is skipped because DEFINE acts like a false
! 6885: condition. The rest of the pattern uses references to the named group
! 6886: to match the four dot-separated components of an IPv4 address, insist-
1.4 misha 6887: ing on a word boundary at each end.
1.1 misha 6888:
6889: Assertion conditions
6890:
1.7 ! moko 6891: If the condition is not in any of the above formats, it must be an
! 6892: assertion. This may be a positive or negative lookahead or lookbehind
! 6893: assertion. Consider this pattern, again containing non-significant
1.1 misha 6894: white space, and with the two alternatives on the second line:
6895:
6896: (?(?=[^a-z]*[a-z])
6897: \d{2}-[a-z]{3}-\d{2} | \d{2}-\d{2}-\d{2} )
6898:
1.7 ! moko 6899: The condition is a positive lookahead assertion that matches an
! 6900: optional sequence of non-letters followed by a letter. In other words,
! 6901: it tests for the presence of at least one letter in the subject. If a
! 6902: letter is found, the subject is matched against the first alternative;
! 6903: otherwise it is matched against the second. This pattern matches
! 6904: strings in one of the two forms dd-aaa-dd or dd-dd-dd, where aaa are
1.1 misha 6905: letters and dd are digits.
6906:
6907:
6908: COMMENTS
6909:
1.4 misha 6910: There are two ways of including comments in patterns that are processed
6911: by PCRE. In both cases, the start of the comment must not be in a char-
6912: acter class, nor in the middle of any other sequence of related charac-
1.7 ! moko 6913: ters such as (?: or a subpattern name or number. The characters that
1.4 misha 6914: make up a comment play no part in the pattern matching.
6915:
1.7 ! moko 6916: The sequence (?# marks the start of a comment that continues up to the
! 6917: next closing parenthesis. Nested parentheses are not permitted. If the
1.4 misha 6918: PCRE_EXTENDED option is set, an unescaped # character also introduces a
1.7 ! moko 6919: comment, which in this case continues to immediately after the next
! 6920: newline character or character sequence in the pattern. Which charac-
1.4 misha 6921: ters are interpreted as newlines is controlled by the options passed to
1.7 ! moko 6922: a compiling function or by a special sequence at the start of the pat-
1.5 misha 6923: tern, as described in the section entitled "Newline conventions" above.
6924: Note that the end of this type of comment is a literal newline sequence
1.7 ! moko 6925: in the pattern; escape sequences that happen to represent a newline do
! 6926: not count. For example, consider this pattern when PCRE_EXTENDED is
1.5 misha 6927: set, and the default newline convention is in force:
1.4 misha 6928:
6929: abc #comment \n still comment
6930:
1.7 ! moko 6931: On encountering the # character, pcre_compile() skips along, looking
! 6932: for a newline in the pattern. The sequence \n is still literal at this
! 6933: stage, so it does not terminate the comment. Only an actual character
1.4 misha 6934: with the code value 0x0a (the default newline) does so.
1.1 misha 6935:
6936:
6937: RECURSIVE PATTERNS
6938:
1.7 ! moko 6939: Consider the problem of matching a string in parentheses, allowing for
! 6940: unlimited nested parentheses. Without the use of recursion, the best
! 6941: that can be done is to use a pattern that matches up to some fixed
! 6942: depth of nesting. It is not possible to handle an arbitrary nesting
1.1 misha 6943: depth.
6944:
6945: For some time, Perl has provided a facility that allows regular expres-
1.7 ! moko 6946: sions to recurse (amongst other things). It does this by interpolating
! 6947: Perl code in the expression at run time, and the code can refer to the
1.1 misha 6948: expression itself. A Perl pattern using code interpolation to solve the
6949: parentheses problem can be created like this:
6950:
6951: $re = qr{\( (?: (?>[^()]+) | (?p{$re}) )* \)}x;
6952:
6953: The (?p{...}) item interpolates Perl code at run time, and in this case
6954: refers recursively to the pattern in which it appears.
6955:
6956: Obviously, PCRE cannot support the interpolation of Perl code. Instead,
1.7 ! moko 6957: it supports special syntax for recursion of the entire pattern, and
! 6958: also for individual subpattern recursion. After its introduction in
! 6959: PCRE and Python, this kind of recursion was subsequently introduced
1.4 misha 6960: into Perl at release 5.10.
1.1 misha 6961:
1.7 ! moko 6962: A special item that consists of (? followed by a number greater than
! 6963: zero and a closing parenthesis is a recursive subroutine call of the
! 6964: subpattern of the given number, provided that it occurs inside that
! 6965: subpattern. (If not, it is a non-recursive subroutine call, which is
! 6966: described in the next section.) The special item (?R) or (?0) is a
1.5 misha 6967: recursive call of the entire regular expression.
1.1 misha 6968:
1.7 ! moko 6969: This PCRE pattern solves the nested parentheses problem (assume the
1.1 misha 6970: PCRE_EXTENDED option is set so that white space is ignored):
6971:
1.4 misha 6972: \( ( [^()]++ | (?R) )* \)
1.1 misha 6973:
1.7 ! moko 6974: First it matches an opening parenthesis. Then it matches any number of
! 6975: substrings which can either be a sequence of non-parentheses, or a
! 6976: recursive match of the pattern itself (that is, a correctly parenthe-
1.4 misha 6977: sized substring). Finally there is a closing parenthesis. Note the use
6978: of a possessive quantifier to avoid backtracking into sequences of non-
6979: parentheses.
1.1 misha 6980:
1.7 ! moko 6981: If this were part of a larger pattern, you would not want to recurse
1.1 misha 6982: the entire pattern, so instead you could use this:
6983:
1.4 misha 6984: ( \( ( [^()]++ | (?1) )* \) )
1.1 misha 6985:
1.7 ! moko 6986: We have put the pattern into parentheses, and caused the recursion to
1.1 misha 6987: refer to them instead of the whole pattern.
6988:
1.7 ! moko 6989: In a larger pattern, keeping track of parenthesis numbers can be
! 6990: tricky. This is made easier by the use of relative references. Instead
1.4 misha 6991: of (?1) in the pattern above you can write (?-2) to refer to the second
1.7 ! moko 6992: most recently opened parentheses preceding the recursion. In other
! 6993: words, a negative number counts capturing parentheses leftwards from
1.4 misha 6994: the point at which it is encountered.
6995:
1.7 ! moko 6996: It is also possible to refer to subsequently opened parentheses, by
! 6997: writing references such as (?+2). However, these cannot be recursive
! 6998: because the reference is not inside the parentheses that are refer-
! 6999: enced. They are always non-recursive subroutine calls, as described in
1.5 misha 7000: the next section.
1.1 misha 7001:
1.7 ! moko 7002: An alternative approach is to use named parentheses instead. The Perl
! 7003: syntax for this is (?&name); PCRE's earlier syntax (?P>name) is also
1.1 misha 7004: supported. We could rewrite the above example as follows:
7005:
1.4 misha 7006: (?<pn> \( ( [^()]++ | (?&pn) )* \) )
1.1 misha 7007:
1.7 ! moko 7008: If there is more than one subpattern with the same name, the earliest
1.1 misha 7009: one is used.
7010:
1.7 ! moko 7011: This particular example pattern that we have been looking at contains
1.4 misha 7012: nested unlimited repeats, and so the use of a possessive quantifier for
7013: matching strings of non-parentheses is important when applying the pat-
1.7 ! moko 7014: tern to strings that do not match. For example, when this pattern is
1.4 misha 7015: applied to
1.1 misha 7016:
7017: (aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa()
7018:
1.7 ! moko 7019: it yields "no match" quickly. However, if a possessive quantifier is
! 7020: not used, the match runs for a very long time indeed because there are
! 7021: so many different ways the + and * repeats can carve up the subject,
1.4 misha 7022: and all have to be tested before failure can be reported.
7023:
1.7 ! moko 7024: At the end of a match, the values of capturing parentheses are those
! 7025: from the outermost level. If you want to obtain intermediate values, a
! 7026: callout function can be used (see below and the pcrecallout documenta-
1.4 misha 7027: tion). If the pattern above is matched against
1.1 misha 7028:
7029: (ab(cd)ef)
7030:
1.7 ! moko 7031: the value for the inner capturing parentheses (numbered 2) is "ef",
! 7032: which is the last value taken on at the top level. If a capturing sub-
! 7033: pattern is not matched at the top level, its final captured value is
! 7034: unset, even if it was (temporarily) set at a deeper level during the
1.5 misha 7035: matching process.
1.4 misha 7036:
1.7 ! moko 7037: If there are more than 15 capturing parentheses in a pattern, PCRE has
! 7038: to obtain extra memory to store data during a recursion, which it does
1.4 misha 7039: by using pcre_malloc, freeing it via pcre_free afterwards. If no memory
7040: can be obtained, the match fails with the PCRE_ERROR_NOMEMORY error.
7041:
1.7 ! moko 7042: Do not confuse the (?R) item with the condition (R), which tests for
! 7043: recursion. Consider this pattern, which matches text in angle brack-
! 7044: ets, allowing for arbitrary nesting. Only digits are allowed in nested
! 7045: brackets (that is, when recursing), whereas any characters are permit-
1.1 misha 7046: ted at the outer level.
7047:
7048: < (?: (?(R) \d++ | [^<>]*+) | (?R)) * >
7049:
1.7 ! moko 7050: In this pattern, (?(R) is the start of a conditional subpattern, with
! 7051: two different alternatives for the recursive and non-recursive cases.
1.1 misha 7052: The (?R) item is the actual recursive call.
7053:
1.5 misha 7054: Differences in recursion processing between PCRE and Perl
1.4 misha 7055:
1.7 ! moko 7056: Recursion processing in PCRE differs from Perl in two important ways.
! 7057: In PCRE (like Python, but unlike Perl), a recursive subpattern call is
1.4 misha 7058: always treated as an atomic group. That is, once it has matched some of
7059: the subject string, it is never re-entered, even if it contains untried
1.7 ! moko 7060: alternatives and there is a subsequent matching failure. This can be
! 7061: illustrated by the following pattern, which purports to match a palin-
! 7062: dromic string that contains an odd number of characters (for example,
1.4 misha 7063: "a", "aba", "abcba", "abcdcba"):
7064:
7065: ^(.|(.)(?1)\2)$
7066:
7067: The idea is that it either matches a single character, or two identical
1.7 ! moko 7068: characters surrounding a sub-palindrome. In Perl, this pattern works;
! 7069: in PCRE it does not if the pattern is longer than three characters.
1.4 misha 7070: Consider the subject string "abcba":
7071:
1.7 ! moko 7072: At the top level, the first character is matched, but as it is not at
1.4 misha 7073: the end of the string, the first alternative fails; the second alterna-
7074: tive is taken and the recursion kicks in. The recursive call to subpat-
1.7 ! moko 7075: tern 1 successfully matches the next character ("b"). (Note that the
1.4 misha 7076: beginning and end of line tests are not part of the recursion).
7077:
1.7 ! moko 7078: Back at the top level, the next character ("c") is compared with what
! 7079: subpattern 2 matched, which was "a". This fails. Because the recursion
! 7080: is treated as an atomic group, there are now no backtracking points,
! 7081: and so the entire match fails. (Perl is able, at this point, to re-
! 7082: enter the recursion and try the second alternative.) However, if the
1.4 misha 7083: pattern is written with the alternatives in the other order, things are
7084: different:
7085:
7086: ^((.)(?1)\2|.)$
7087:
1.7 ! moko 7088: This time, the recursing alternative is tried first, and continues to
! 7089: recurse until it runs out of characters, at which point the recursion
! 7090: fails. But this time we do have another alternative to try at the
! 7091: higher level. That is the big difference: in the previous case the
1.4 misha 7092: remaining alternative is at a deeper recursion level, which PCRE cannot
7093: use.
7094:
1.7 ! moko 7095: To change the pattern so that it matches all palindromic strings, not
! 7096: just those with an odd number of characters, it is tempting to change
1.4 misha 7097: the pattern to this:
7098:
7099: ^((.)(?1)\2|.?)$
7100:
1.7 ! moko 7101: Again, this works in Perl, but not in PCRE, and for the same reason.
! 7102: When a deeper recursion has matched a single character, it cannot be
! 7103: entered again in order to match an empty string. The solution is to
! 7104: separate the two cases, and write out the odd and even cases as alter-
1.4 misha 7105: natives at the higher level:
7106:
7107: ^(?:((.)(?1)\2|)|((.)(?3)\4|.))
7108:
1.7 ! moko 7109: If you want to match typical palindromic phrases, the pattern has to
1.4 misha 7110: ignore all non-word characters, which can be done like this:
7111:
7112: ^\W*+(?:((.)\W*+(?1)\W*+\2|)|((.)\W*+(?3)\W*+\4|\W*+.\W*+))\W*+$
7113:
7114: If run with the PCRE_CASELESS option, this pattern matches phrases such
7115: as "A man, a plan, a canal: Panama!" and it works well in both PCRE and
1.7 ! moko 7116: Perl. Note the use of the possessive quantifier *+ to avoid backtrack-
! 7117: ing into sequences of non-word characters. Without this, PCRE takes a
! 7118: great deal longer (ten times or more) to match typical phrases, and
1.4 misha 7119: Perl takes so long that you think it has gone into a loop.
7120:
1.7 ! moko 7121: WARNING: The palindrome-matching patterns above work only if the sub-
! 7122: ject string does not start with a palindrome that is shorter than the
! 7123: entire string. For example, although "abcba" is correctly matched, if
! 7124: the subject is "ababa", PCRE finds the palindrome "aba" at the start,
! 7125: then fails at top level because the end of the string does not follow.
! 7126: Once again, it cannot jump back into the recursion to try other alter-
1.4 misha 7127: natives, so the entire match fails.
7128:
1.7 ! moko 7129: The second way in which PCRE and Perl differ in their recursion pro-
! 7130: cessing is in the handling of captured values. In Perl, when a subpat-
! 7131: tern is called recursively or as a subpattern (see the next section),
! 7132: it has no access to any values that were captured outside the recur-
! 7133: sion, whereas in PCRE these values can be referenced. Consider this
1.5 misha 7134: pattern:
7135:
7136: ^(.)(\1|a(?2))
7137:
1.7 ! moko 7138: In PCRE, this pattern matches "bab". The first capturing parentheses
! 7139: match "b", then in the second group, when the back reference \1 fails
! 7140: to match "b", the second alternative matches "a" and then recurses. In
! 7141: the recursion, \1 does now match "b" and so the whole match succeeds.
! 7142: In Perl, the pattern fails to match because inside the recursive call
1.5 misha 7143: \1 cannot access the externally set value.
7144:
1.1 misha 7145:
7146: SUBPATTERNS AS SUBROUTINES
7147:
1.7 ! moko 7148: If the syntax for a recursive subpattern call (either by number or by
! 7149: name) is used outside the parentheses to which it refers, it operates
! 7150: like a subroutine in a programming language. The called subpattern may
! 7151: be defined before or after the reference. A numbered reference can be
1.5 misha 7152: absolute or relative, as in these examples:
1.1 misha 7153:
7154: (...(absolute)...)...(?2)...
7155: (...(relative)...)...(?-1)...
7156: (...(?+1)...(relative)...
7157:
7158: An earlier example pointed out that the pattern
7159:
7160: (sens|respons)e and \1ibility
7161:
1.7 ! moko 7162: matches "sense and sensibility" and "response and responsibility", but
1.1 misha 7163: not "sense and responsibility". If instead the pattern
7164:
7165: (sens|respons)e and (?1)ibility
7166:
1.7 ! moko 7167: is used, it does match "sense and responsibility" as well as the other
! 7168: two strings. Another example is given in the discussion of DEFINE
1.1 misha 7169: above.
7170:
1.7 ! moko 7171: All subroutine calls, whether recursive or not, are always treated as
! 7172: atomic groups. That is, once a subroutine has matched some of the sub-
1.5 misha 7173: ject string, it is never re-entered, even if it contains untried alter-
1.7 ! moko 7174: natives and there is a subsequent matching failure. Any capturing
! 7175: parentheses that are set during the subroutine call revert to their
1.5 misha 7176: previous values afterwards.
1.1 misha 7177:
1.7 ! moko 7178: Processing options such as case-independence are fixed when a subpat-
! 7179: tern is defined, so if it is used as a subroutine, such options cannot
1.1 misha 7180: be changed for different calls. For example, consider this pattern:
7181:
7182: (abc)(?i:(?-1))
7183:
1.7 ! moko 7184: It matches "abcabc". It does not match "abcABC" because the change of
1.1 misha 7185: processing option does not affect the called subpattern.
7186:
7187:
7188: ONIGURUMA SUBROUTINE SYNTAX
7189:
1.7 ! moko 7190: For compatibility with Oniguruma, the non-Perl syntax \g followed by a
1.1 misha 7191: name or a number enclosed either in angle brackets or single quotes, is
1.7 ! moko 7192: an alternative syntax for referencing a subpattern as a subroutine,
! 7193: possibly recursively. Here are two of the examples used above, rewrit-
1.1 misha 7194: ten using this syntax:
7195:
7196: (?<pn> \( ( (?>[^()]+) | \g<pn> )* \) )
7197: (sens|respons)e and \g'1'ibility
7198:
1.7 ! moko 7199: PCRE supports an extension to Oniguruma: if a number is preceded by a
1.1 misha 7200: plus or a minus sign it is taken as a relative reference. For example:
7201:
7202: (abc)(?i:\g<-1>)
7203:
1.7 ! moko 7204: Note that \g{...} (Perl syntax) and \g<...> (Oniguruma syntax) are not
! 7205: synonymous. The former is a back reference; the latter is a subroutine
1.1 misha 7206: call.
7207:
7208:
7209: CALLOUTS
7210:
7211: Perl has a feature whereby using the sequence (?{...}) causes arbitrary
1.7 ! moko 7212: Perl code to be obeyed in the middle of matching a regular expression.
1.1 misha 7213: This makes it possible, amongst other things, to extract different sub-
7214: strings that match the same pair of parentheses when there is a repeti-
7215: tion.
7216:
7217: PCRE provides a similar feature, but of course it cannot obey arbitrary
7218: Perl code. The feature is called "callout". The caller of PCRE provides
1.7 ! moko 7219: an external function by putting its entry point in the global variable
! 7220: pcre_callout (8-bit library) or pcre[16|32]_callout (16-bit or 32-bit
! 7221: library). By default, this variable contains NULL, which disables all
1.6 misha 7222: calling out.
7223:
1.7 ! moko 7224: Within a regular expression, (?C) indicates the points at which the
! 7225: external function is to be called. If you want to identify different
! 7226: callout points, you can put a number less than 256 after the letter C.
! 7227: The default value is zero. For example, this pattern has two callout
1.1 misha 7228: points:
7229:
7230: (?C1)abc(?C2)def
7231:
1.7 ! moko 7232: If the PCRE_AUTO_CALLOUT flag is passed to a compiling function, call-
! 7233: outs are automatically installed before each item in the pattern. They
! 7234: are all numbered 255. If there is a conditional group in the pattern
1.6 misha 7235: whose condition is an assertion, an additional callout is inserted just
7236: before the condition. An explicit callout may also be set at this posi-
7237: tion, as in this example:
7238:
7239: (?(?C9)(?=a)abc|def)
7240:
7241: Note that this applies only to assertion conditions, not to other types
7242: of condition.
7243:
1.7 ! moko 7244: During matching, when PCRE reaches a callout point, the external func-
! 7245: tion is called. It is provided with the number of the callout, the
! 7246: position in the pattern, and, optionally, one item of data originally
! 7247: supplied by the caller of the matching function. The callout function
! 7248: may cause matching to proceed, to backtrack, or to fail altogether.
! 7249:
! 7250: By default, PCRE implements a number of optimizations at compile time
! 7251: and matching time, and one side-effect is that sometimes callouts are
! 7252: skipped. If you need all possible callouts to happen, you need to set
! 7253: options that disable the relevant optimizations. More details, and a
! 7254: complete description of the interface to the callout function, are
! 7255: given in the pcrecallout documentation.
1.1 misha 7256:
7257:
7258: BACKTRACKING CONTROL
7259:
1.6 misha 7260: Perl 5.10 introduced a number of "Special Backtracking Control Verbs",
7261: which are still described in the Perl documentation as "experimental
7262: and subject to change or removal in a future version of Perl". It goes
7263: on to say: "Their usage in production code should be noted to avoid
7264: problems during upgrades." The same remarks apply to the PCRE features
7265: described in this section.
7266:
7267: The new verbs make use of what was previously invalid syntax: an open-
7268: ing parenthesis followed by an asterisk. They are generally of the form
7269: (*VERB) or (*VERB:NAME). Some may take either form, possibly behaving
7270: differently depending on whether or not a name is present. A name is
7271: any sequence of characters that does not include a closing parenthesis.
7272: The maximum length of name is 255 in the 8-bit library and 65535 in the
7273: 16-bit and 32-bit libraries. If the name is empty, that is, if the
7274: closing parenthesis immediately follows the colon, the effect is as if
7275: the colon were not there. Any number of these verbs may occur in a
7276: pattern.
1.1 misha 7277:
1.5 misha 7278: Since these verbs are specifically related to backtracking, most of
7279: them can be used only when the pattern is to be matched using one of
1.6 misha 7280: the traditional matching functions, because these use a backtracking
7281: algorithm. With the exception of (*FAIL), which behaves like a failing
7282: negative assertion, the backtracking control verbs cause an error if
7283: encountered by a DFA matching function.
7284:
7285: The behaviour of these verbs in repeated groups, assertions, and in
7286: subpatterns called as subroutines (whether or not recursively) is docu-
7287: mented below.
1.4 misha 7288:
1.6 misha 7289: Optimizations that affect backtracking verbs
1.4 misha 7290:
1.6 misha 7291: PCRE contains some optimizations that are used to speed up matching by
1.4 misha 7292: running some checks at the start of each match attempt. For example, it
1.6 misha 7293: may know the minimum length of matching subject, or that a particular
7294: character must be present. When one of these optimizations bypasses the
7295: running of a match, any included backtracking verbs will not, of
1.4 misha 7296: course, be processed. You can suppress the start-of-match optimizations
1.6 misha 7297: by setting the PCRE_NO_START_OPTIMIZE option when calling pcre_com-
1.4 misha 7298: pile() or pcre_exec(), or by starting the pattern with (*NO_START_OPT).
1.6 misha 7299: There is more discussion of this option in the section entitled "Option
7300: bits for pcre_exec()" in the pcreapi documentation.
1.1 misha 7301:
1.6 misha 7302: Experiments with Perl suggest that it too has similar optimizations,
1.5 misha 7303: sometimes leading to anomalous results.
7304:
1.1 misha 7305: Verbs that act immediately
7306:
1.6 misha 7307: The following verbs act as soon as they are encountered. They may not
1.4 misha 7308: be followed by a name.
1.1 misha 7309:
7310: (*ACCEPT)
7311:
1.6 misha 7312: This verb causes the match to end successfully, skipping the remainder
7313: of the pattern. However, when it is inside a subpattern that is called
7314: as a subroutine, only that subpattern is ended successfully. Matching
7315: then continues at the outer level. If (*ACCEPT) in triggered in a posi-
7316: tive assertion, the assertion succeeds; in a negative assertion, the
7317: assertion fails.
7318:
7319: If (*ACCEPT) is inside capturing parentheses, the data so far is cap-
7320: tured. For example:
1.1 misha 7321:
1.4 misha 7322: A((?:A|B(*ACCEPT)|C)D)
1.1 misha 7323:
1.6 misha 7324: This matches "AB", "AAD", or "ACD"; when it matches "AB", "B" is cap-
1.4 misha 7325: tured by the outer parentheses.
1.1 misha 7326:
7327: (*FAIL) or (*F)
7328:
1.6 misha 7329: This verb causes a matching failure, forcing backtracking to occur. It
7330: is equivalent to (?!) but easier to read. The Perl documentation notes
7331: that it is probably useful only when combined with (?{}) or (??{}).
7332: Those are, of course, Perl features that are not present in PCRE. The
7333: nearest equivalent is the callout feature, as for example in this pat-
1.1 misha 7334: tern:
7335:
7336: a+(?C)(*FAIL)
7337:
1.6 misha 7338: A match with the string "aaaa" always fails, but the callout is taken
1.1 misha 7339: before each backtrack happens (in this example, 10 times).
7340:
1.4 misha 7341: Recording which path was taken
7342:
1.6 misha 7343: There is one verb whose main purpose is to track how a match was
7344: arrived at, though it also has a secondary use in conjunction with
1.4 misha 7345: advancing the match starting point (see (*SKIP) below).
7346:
7347: (*MARK:NAME) or (*:NAME)
7348:
1.6 misha 7349: A name is always required with this verb. There may be as many
7350: instances of (*MARK) as you like in a pattern, and their names do not
1.4 misha 7351: have to be unique.
7352:
1.6 misha 7353: When a match succeeds, the name of the last-encountered (*MARK:NAME),
7354: (*PRUNE:NAME), or (*THEN:NAME) on the matching path is passed back to
7355: the caller as described in the section entitled "Extra data for
7356: pcre_exec()" in the pcreapi documentation. Here is an example of
7357: pcretest output, where the /K modifier requests the retrieval and out-
7358: putting of (*MARK) data:
1.4 misha 7359:
1.5 misha 7360: re> /X(*MARK:A)Y|X(*MARK:B)Z/K
7361: data> XY
1.4 misha 7362: 0: XY
7363: MK: A
7364: XZ
7365: 0: XZ
7366: MK: B
7367:
7368: The (*MARK) name is tagged with "MK:" in this output, and in this exam-
7369: ple it indicates which of the two alternatives matched. This is a more
7370: efficient way of obtaining this information than putting each alterna-
7371: tive in its own capturing parentheses.
7372:
1.6 misha 7373: If a verb with a name is encountered in a positive assertion that is
7374: true, the name is recorded and passed back if it is the last-encoun-
7375: tered. This does not happen for negative assertions or failing positive
7376: assertions.
1.4 misha 7377:
1.6 misha 7378: After a partial match or a failed match, the last encountered name in
7379: the entire match process is returned. For example:
1.4 misha 7380:
1.5 misha 7381: re> /X(*MARK:A)Y|X(*MARK:B)Z/K
7382: data> XP
1.4 misha 7383: No match, mark = B
7384:
1.6 misha 7385: Note that in this unanchored example the mark is retained from the
7386: match attempt that started at the letter "X" in the subject. Subsequent
7387: match attempts starting at "P" and then with an empty string do not get
7388: as far as the (*MARK) item, but nevertheless do not reset it.
7389:
7390: If you are interested in (*MARK) values after failed matches, you
7391: should probably set the PCRE_NO_START_OPTIMIZE option (see above) to
7392: ensure that the match is always attempted.
1.4 misha 7393:
1.1 misha 7394: Verbs that act after backtracking
7395:
7396: The following verbs do nothing when they are encountered. Matching con-
1.6 misha 7397: tinues with what follows, but if there is no subsequent match, causing
7398: a backtrack to the verb, a failure is forced. That is, backtracking
7399: cannot pass to the left of the verb. However, when one of these verbs
7400: appears inside an atomic group or an assertion that is true, its effect
7401: is confined to that group, because once the group has been matched,
7402: there is never any backtracking into it. In this situation, backtrack-
7403: ing can "jump back" to the left of the entire atomic group or asser-
7404: tion. (Remember also, as stated above, that this localization also
7405: applies in subroutine calls.)
1.4 misha 7406:
1.5 misha 7407: These verbs differ in exactly what kind of failure occurs when back-
1.6 misha 7408: tracking reaches them. The behaviour described below is what happens
7409: when the verb is not in a subroutine or an assertion. Subsequent sec-
7410: tions cover these special cases.
1.1 misha 7411:
7412: (*COMMIT)
7413:
1.5 misha 7414: This verb, which may not be followed by a name, causes the whole match
1.6 misha 7415: to fail outright if there is a later matching failure that causes back-
7416: tracking to reach it. Even if the pattern is unanchored, no further
7417: attempts to find a match by advancing the starting point take place. If
7418: (*COMMIT) is the only backtracking verb that is encountered, once it
7419: has been passed pcre_exec() is committed to finding a match at the cur-
7420: rent starting point, or not at all. For example:
1.1 misha 7421:
7422: a+(*COMMIT)b
7423:
1.6 misha 7424: This matches "xxaab" but not "aacaab". It can be thought of as a kind
1.4 misha 7425: of dynamic anchor, or "I've started, so I must finish." The name of the
1.6 misha 7426: most recently passed (*MARK) in the path is passed back when (*COMMIT)
1.4 misha 7427: forces a match failure.
7428:
1.6 misha 7429: If there is more than one backtracking verb in a pattern, a different
7430: one that follows (*COMMIT) may be triggered first, so merely passing
7431: (*COMMIT) during a match does not always guarantee that a match must be
7432: at this starting point.
7433:
1.5 misha 7434: Note that (*COMMIT) at the start of a pattern is not the same as an
7435: anchor, unless PCRE's start-of-match optimizations are turned off, as
1.7 ! moko 7436: shown in this output from pcretest:
1.4 misha 7437:
1.5 misha 7438: re> /(*COMMIT)abc/
7439: data> xyzabc
1.4 misha 7440: 0: abc
1.7 ! moko 7441: data> xyzabc\Y
1.4 misha 7442: No match
1.1 misha 7443:
1.7 ! moko 7444: For this pattern, PCRE knows that any match must start with "a", so the
! 7445: optimization skips along the subject to "a" before applying the pattern
! 7446: to the first set of data. The match attempt then succeeds. In the sec-
! 7447: ond set of data, the escape sequence \Y is interpreted by the pcretest
! 7448: program. It causes the PCRE_NO_START_OPTIMIZE option to be set when
! 7449: pcre_exec() is called. This disables the optimization that skips along
! 7450: to the first character. The pattern is now applied starting at "x", and
! 7451: so the (*COMMIT) causes the match to fail without trying any other
! 7452: starting points.
1.4 misha 7453:
7454: (*PRUNE) or (*PRUNE:NAME)
7455:
1.5 misha 7456: This verb causes the match to fail at the current starting position in
1.6 misha 7457: the subject if there is a later matching failure that causes backtrack-
7458: ing to reach it. If the pattern is unanchored, the normal "bumpalong"
7459: advance to the next starting character then happens. Backtracking can
7460: occur as usual to the left of (*PRUNE), before it is reached, or when
7461: matching to the right of (*PRUNE), but if there is no match to the
7462: right, backtracking cannot cross (*PRUNE). In simple cases, the use of
7463: (*PRUNE) is just an alternative to an atomic group or possessive quan-
7464: tifier, but there are some uses of (*PRUNE) that cannot be expressed in
7465: any other way. In an anchored pattern (*PRUNE) has the same effect as
7466: (*COMMIT).
7467:
7468: The behaviour of (*PRUNE:NAME) is the not the same as
7469: (*MARK:NAME)(*PRUNE). It is like (*MARK:NAME) in that the name is
7470: remembered for passing back to the caller. However, (*SKIP:NAME)
7471: searches only for names set with (*MARK).
1.1 misha 7472:
7473: (*SKIP)
7474:
1.6 misha 7475: This verb, when given without a name, is like (*PRUNE), except that if
7476: the pattern is unanchored, the "bumpalong" advance is not to the next
1.4 misha 7477: character, but to the position in the subject where (*SKIP) was encoun-
1.6 misha 7478: tered. (*SKIP) signifies that whatever text was matched leading up to
1.4 misha 7479: it cannot be part of a successful match. Consider:
1.1 misha 7480:
7481: a+(*SKIP)b
7482:
1.6 misha 7483: If the subject is "aaaac...", after the first match attempt fails
7484: (starting at the first character in the string), the starting point
1.1 misha 7485: skips on to start the next attempt at "c". Note that a possessive quan-
1.6 misha 7486: tifer does not have the same effect as this example; although it would
7487: suppress backtracking during the first match attempt, the second
7488: attempt would start at the second character instead of skipping on to
1.1 misha 7489: "c".
7490:
1.4 misha 7491: (*SKIP:NAME)
1.1 misha 7492:
1.6 misha 7493: When (*SKIP) has an associated name, its behaviour is modified. When it
7494: is triggered, the previous path through the pattern is searched for the
7495: most recent (*MARK) that has the same name. If one is found, the
7496: "bumpalong" advance is to the subject position that corresponds to that
7497: (*MARK) instead of to where (*SKIP) was encountered. If no (*MARK) with
7498: a matching name is found, the (*SKIP) is ignored.
7499:
7500: Note that (*SKIP:NAME) searches only for names set by (*MARK:NAME). It
7501: ignores names that are set by (*PRUNE:NAME) or (*THEN:NAME).
1.4 misha 7502:
7503: (*THEN) or (*THEN:NAME)
7504:
1.6 misha 7505: This verb causes a skip to the next innermost alternative when back-
7506: tracking reaches it. That is, it cancels any further backtracking
7507: within the current alternative. Its name comes from the observation
7508: that it can be used for a pattern-based if-then-else block:
1.1 misha 7509:
7510: ( COND1 (*THEN) FOO | COND2 (*THEN) BAR | COND3 (*THEN) BAZ ) ...
7511:
1.3 misha 7512: If the COND1 pattern matches, FOO is tried (and possibly further items
1.5 misha 7513: after the end of the group if FOO succeeds); on failure, the matcher
1.3 misha 7514: skips to the second alternative and tries COND2, without backtracking
1.6 misha 7515: into COND1. If that succeeds and BAR fails, COND3 is tried. If subse-
7516: quently BAZ fails, there are no more alternatives, so there is a back-
7517: track to whatever came before the entire group. If (*THEN) is not
7518: inside an alternation, it acts like (*PRUNE).
7519:
7520: The behaviour of (*THEN:NAME) is the not the same as
7521: (*MARK:NAME)(*THEN). It is like (*MARK:NAME) in that the name is
7522: remembered for passing back to the caller. However, (*SKIP:NAME)
7523: searches only for names set with (*MARK).
7524:
7525: A subpattern that does not contain a | character is just a part of the
7526: enclosing alternative; it is not a nested alternation with only one
7527: alternative. The effect of (*THEN) extends beyond such a subpattern to
7528: the enclosing alternative. Consider this pattern, where A, B, etc. are
7529: complex pattern fragments that do not contain any | characters at this
7530: level:
1.5 misha 7531:
7532: A (B(*THEN)C) | D
7533:
7534: If A and B are matched, but there is a failure in C, matching does not
7535: backtrack into A; instead it moves to the next alternative, that is, D.
7536: However, if the subpattern containing (*THEN) is given an alternative,
7537: it behaves differently:
7538:
7539: A (B(*THEN)C | (*FAIL)) | D
7540:
7541: The effect of (*THEN) is now confined to the inner subpattern. After a
7542: failure in C, matching moves to (*FAIL), which causes the whole subpat-
7543: tern to fail because there are no more alternatives to try. In this
7544: case, matching does now backtrack into A.
7545:
1.6 misha 7546: Note that a conditional subpattern is not considered as having two
1.5 misha 7547: alternatives, because only one is ever used. In other words, the |
7548: character in a conditional subpattern has a different meaning. Ignoring
7549: white space, consider:
7550:
7551: ^.*? (?(?=a) a | b(*THEN)c )
7552:
7553: If the subject is "ba", this pattern does not match. Because .*? is
7554: ungreedy, it initially matches zero characters. The condition (?=a)
7555: then fails, the character "b" is matched, but "c" is not. At this
7556: point, matching does not backtrack to .*? as might perhaps be expected
7557: from the presence of the | character. The conditional subpattern is
7558: part of the single alternative that comprises the whole pattern, and so
7559: the match fails. (If there was a backtrack into .*?, allowing it to
7560: match "b", the match would succeed.)
7561:
7562: The verbs just described provide four different "strengths" of control
7563: when subsequent matching fails. (*THEN) is the weakest, carrying on the
7564: match at the next alternative. (*PRUNE) comes next, failing the match
7565: at the current starting position, but allowing an advance to the next
7566: character (for an unanchored pattern). (*SKIP) is similar, except that
7567: the advance may be more than one character. (*COMMIT) is the strongest,
1.4 misha 7568: causing the entire match to fail.
7569:
1.6 misha 7570: More than one backtracking verb
7571:
7572: If more than one backtracking verb is present in a pattern, the one
7573: that is backtracked onto first acts. For example, consider this pat-
7574: tern, where A, B, etc. are complex pattern fragments:
7575:
7576: (A(*COMMIT)B(*THEN)C|ABD)
7577:
7578: If A matches but B fails, the backtrack to (*COMMIT) causes the entire
7579: match to fail. However, if A and B match, but C fails, the backtrack to
7580: (*THEN) causes the next alternative (ABD) to be tried. This behaviour
7581: is consistent, but is not always the same as Perl's. It means that if
7582: two or more backtracking verbs appear in succession, all the the last
7583: of them has no effect. Consider this example:
7584:
7585: ...(*COMMIT)(*PRUNE)...
7586:
7587: If there is a matching failure to the right, backtracking onto (*PRUNE)
1.7 ! moko 7588: causes it to be triggered, and its action is taken. There can never be
! 7589: a backtrack onto (*COMMIT).
1.6 misha 7590:
7591: Backtracking verbs in repeated groups
7592:
7593: PCRE differs from Perl in its handling of backtracking verbs in
7594: repeated groups. For example, consider:
7595:
7596: /(a(*COMMIT)b)+ac/
7597:
7598: If the subject is "abac", Perl matches, but PCRE fails because the
7599: (*COMMIT) in the second repeat of the group acts.
7600:
7601: Backtracking verbs in assertions
7602:
7603: (*FAIL) in an assertion has its normal effect: it forces an immediate
7604: backtrack.
7605:
7606: (*ACCEPT) in a positive assertion causes the assertion to succeed with-
7607: out any further processing. In a negative assertion, (*ACCEPT) causes
7608: the assertion to fail without any further processing.
7609:
7610: The other backtracking verbs are not treated specially if they appear
7611: in a positive assertion. In particular, (*THEN) skips to the next
7612: alternative in the innermost enclosing group that has alternations,
7613: whether or not this is within the assertion.
7614:
7615: Negative assertions are, however, different, in order to ensure that
7616: changing a positive assertion into a negative assertion changes its
7617: result. Backtracking into (*COMMIT), (*SKIP), or (*PRUNE) causes a neg-
7618: ative assertion to be true, without considering any further alternative
7619: branches in the assertion. Backtracking into (*THEN) causes it to skip
7620: to the next enclosing alternative within the assertion (the normal be-
7621: haviour), but if the assertion does not have such an alternative,
7622: (*THEN) behaves like (*PRUNE).
7623:
7624: Backtracking verbs in subroutines
7625:
7626: These behaviours occur whether or not the subpattern is called recur-
7627: sively. Perl's treatment of subroutines is different in some cases.
7628:
7629: (*FAIL) in a subpattern called as a subroutine has its normal effect:
7630: it forces an immediate backtrack.
7631:
7632: (*ACCEPT) in a subpattern called as a subroutine causes the subroutine
7633: match to succeed without any further processing. Matching then contin-
7634: ues after the subroutine call.
7635:
7636: (*COMMIT), (*SKIP), and (*PRUNE) in a subpattern called as a subroutine
7637: cause the subroutine match to fail.
7638:
7639: (*THEN) skips to the next alternative in the innermost enclosing group
7640: within the subpattern that has alternatives. If there is no such group
7641: within the subpattern, (*THEN) causes the subroutine match to fail.
1.1 misha 7642:
7643:
7644: SEE ALSO
7645:
1.5 misha 7646: pcreapi(3), pcrecallout(3), pcrematching(3), pcresyntax(3), pcre(3),
1.6 misha 7647: pcre16(3), pcre32(3).
1.1 misha 7648:
7649:
7650: AUTHOR
7651:
7652: Philip Hazel
7653: University Computing Service
7654: Cambridge CB2 3QH, England.
7655:
7656:
7657: REVISION
7658:
1.7 ! moko 7659: Last updated: 08 January 2014
! 7660: Copyright (c) 1997-2014 University of Cambridge.
1.1 misha 7661: ------------------------------------------------------------------------------
7662:
7663:
1.6 misha 7664: PCRESYNTAX(3) Library Functions Manual PCRESYNTAX(3)
7665:
1.1 misha 7666:
7667:
7668: NAME
7669: PCRE - Perl-compatible regular expressions
7670:
7671: PCRE REGULAR EXPRESSION SYNTAX SUMMARY
7672:
7673: The full syntax and semantics of the regular expressions that are sup-
7674: ported by PCRE are described in the pcrepattern documentation. This
1.5 misha 7675: document contains a quick-reference summary of the syntax.
1.1 misha 7676:
7677:
7678: QUOTING
7679:
7680: \x where x is non-alphanumeric is a literal x
7681: \Q...\E treat enclosed characters as literal
7682:
7683:
7684: CHARACTERS
7685:
7686: \a alarm, that is, the BEL character (hex 07)
1.4 misha 7687: \cx "control-x", where x is any ASCII character
1.1 misha 7688: \e escape (hex 1B)
1.6 misha 7689: \f form feed (hex 0C)
1.1 misha 7690: \n newline (hex 0A)
7691: \r carriage return (hex 0D)
7692: \t tab (hex 09)
1.7 ! moko 7693: \0dd character with octal code 0dd
1.1 misha 7694: \ddd character with octal code ddd, or backreference
1.7 ! moko 7695: \o{ddd..} character with octal code ddd..
1.1 misha 7696: \xhh character with hex code hh
7697: \x{hhh..} character with hex code hhh..
7698:
1.7 ! moko 7699: Note that \0dd is always an octal code, and that \8 and \9 are the lit-
! 7700: eral characters "8" and "9".
! 7701:
1.1 misha 7702:
7703: CHARACTER TYPES
7704:
7705: . any character except newline;
7706: in dotall mode, any character whatsoever
1.5 misha 7707: \C one data unit, even in UTF mode (best avoided)
1.1 misha 7708: \d a decimal digit
7709: \D a character that is not a decimal digit
1.6 misha 7710: \h a horizontal white space character
7711: \H a character that is not a horizontal white space character
1.4 misha 7712: \N a character that is not a newline
1.1 misha 7713: \p{xx} a character with the xx property
7714: \P{xx} a character without the xx property
7715: \R a newline sequence
1.6 misha 7716: \s a white space character
7717: \S a character that is not a white space character
7718: \v a vertical white space character
7719: \V a character that is not a vertical white space character
1.1 misha 7720: \w a "word" character
7721: \W a "non-word" character
1.6 misha 7722: \X a Unicode extended grapheme cluster
1.1 misha 7723:
1.7 ! moko 7724: By default, \d, \s, and \w match only ASCII characters, even in UTF-8
! 7725: mode or in the 16- bit and 32-bit libraries. However, if locale-spe-
! 7726: cific matching is happening, \s and \w may also match characters with
! 7727: code points in the range 128-255. If the PCRE_UCP option is set, the
! 7728: behaviour of these escape sequences is changed to use Unicode proper-
! 7729: ties and they match many more characters.
1.1 misha 7730:
7731:
1.4 misha 7732: GENERAL CATEGORY PROPERTIES FOR \p and \P
1.1 misha 7733:
7734: C Other
7735: Cc Control
7736: Cf Format
7737: Cn Unassigned
7738: Co Private use
7739: Cs Surrogate
7740:
7741: L Letter
7742: Ll Lower case letter
7743: Lm Modifier letter
7744: Lo Other letter
7745: Lt Title case letter
7746: Lu Upper case letter
7747: L& Ll, Lu, or Lt
7748:
7749: M Mark
7750: Mc Spacing mark
7751: Me Enclosing mark
7752: Mn Non-spacing mark
7753:
7754: N Number
7755: Nd Decimal number
7756: Nl Letter number
7757: No Other number
7758:
7759: P Punctuation
7760: Pc Connector punctuation
7761: Pd Dash punctuation
7762: Pe Close punctuation
7763: Pf Final punctuation
7764: Pi Initial punctuation
7765: Po Other punctuation
7766: Ps Open punctuation
7767:
7768: S Symbol
7769: Sc Currency symbol
7770: Sk Modifier symbol
7771: Sm Mathematical symbol
7772: So Other symbol
7773:
7774: Z Separator
7775: Zl Line separator
7776: Zp Paragraph separator
7777: Zs Space separator
7778:
7779:
1.4 misha 7780: PCRE SPECIAL CATEGORY PROPERTIES FOR \p and \P
7781:
7782: Xan Alphanumeric: union of properties L and N
7783: Xps POSIX space: property Z or tab, NL, VT, FF, CR
1.7 ! moko 7784: Xsp Perl space: property Z or tab, NL, VT, FF, CR
1.6 misha 7785: Xuc Univerally-named character: one that can be
7786: represented by a Universal Character Name
1.4 misha 7787: Xwd Perl word: property Xan or underscore
7788:
1.7 ! moko 7789: Perl and POSIX space are now the same. Perl added VT to its space char-
! 7790: acter set at release 5.18 and PCRE changed at release 8.34.
! 7791:
1.4 misha 7792:
1.1 misha 7793: SCRIPT NAMES FOR \p AND \P
7794:
1.7 ! moko 7795: Arabic, Armenian, Avestan, Balinese, Bamum, Bassa_Vah, Batak, Bengali,
! 7796: Bopomofo, Brahmi, Braille, Buginese, Buhid, Canadian_Aboriginal, Car-
! 7797: ian, Caucasian_Albanian, Chakma, Cham, Cherokee, Common, Coptic, Cunei-
! 7798: form, Cypriot, Cyrillic, Deseret, Devanagari, Duployan, Egyptian_Hiero-
! 7799: glyphs, Elbasan, Ethiopic, Georgian, Glagolitic, Gothic, Grantha,
! 7800: Greek, Gujarati, Gurmukhi, Han, Hangul, Hanunoo, Hebrew, Hiragana,
! 7801: Imperial_Aramaic, Inherited, Inscriptional_Pahlavi, Inscrip-
1.6 misha 7802: tional_Parthian, Javanese, Kaithi, Kannada, Katakana, Kayah_Li,
1.7 ! moko 7803: Kharoshthi, Khmer, Khojki, Khudawadi, Lao, Latin, Lepcha, Limbu, Lin-
! 7804: ear_A, Linear_B, Lisu, Lycian, Lydian, Mahajani, Malayalam, Mandaic,
! 7805: Manichaean, Meetei_Mayek, Mende_Kikakui, Meroitic_Cursive,
! 7806: Meroitic_Hieroglyphs, Miao, Modi, Mongolian, Mro, Myanmar, Nabataean,
! 7807: New_Tai_Lue, Nko, Ogham, Ol_Chiki, Old_Italic, Old_North_Arabian,
! 7808: Old_Permic, Old_Persian, Old_South_Arabian, Old_Turkic, Oriya, Osmanya,
! 7809: Pahawh_Hmong, Palmyrene, Pau_Cin_Hau, Phags_Pa, Phoenician,
! 7810: Psalter_Pahlavi, Rejang, Runic, Samaritan, Saurashtra, Sharada, Sha-
! 7811: vian, Siddham, Sinhala, Sora_Sompeng, Sundanese, Syloti_Nagri, Syriac,
! 7812: Tagalog, Tagbanwa, Tai_Le, Tai_Tham, Tai_Viet, Takri, Tamil, Telugu,
! 7813: Thaana, Thai, Tibetan, Tifinagh, Tirhuta, Ugaritic, Vai, Warang_Citi,
1.6 misha 7814: Yi.
1.1 misha 7815:
7816:
7817: CHARACTER CLASSES
7818:
7819: [...] positive character class
7820: [^...] negative character class
7821: [x-y] range (can be used for hex characters)
7822: [[:xxx:]] positive POSIX named set
7823: [[:^xxx:]] negative POSIX named set
7824:
7825: alnum alphanumeric
7826: alpha alphabetic
7827: ascii 0-127
7828: blank space or tab
7829: cntrl control character
7830: digit decimal digit
7831: graph printing, excluding space
7832: lower lower case letter
7833: print printing, including space
7834: punct printing, excluding alphanumeric
1.6 misha 7835: space white space
1.1 misha 7836: upper upper case letter
7837: word same as \w
7838: xdigit hexadecimal digit
7839:
1.7 ! moko 7840: In PCRE, POSIX character set names recognize only ASCII characters by
! 7841: default, but some of them use Unicode properties if PCRE_UCP is set.
1.4 misha 7842: You can use \Q...\E inside a character class.
1.1 misha 7843:
7844:
7845: QUANTIFIERS
7846:
7847: ? 0 or 1, greedy
7848: ?+ 0 or 1, possessive
7849: ?? 0 or 1, lazy
7850: * 0 or more, greedy
7851: *+ 0 or more, possessive
7852: *? 0 or more, lazy
7853: + 1 or more, greedy
7854: ++ 1 or more, possessive
7855: +? 1 or more, lazy
7856: {n} exactly n
7857: {n,m} at least n, no more than m, greedy
7858: {n,m}+ at least n, no more than m, possessive
7859: {n,m}? at least n, no more than m, lazy
7860: {n,} n or more, greedy
7861: {n,}+ n or more, possessive
7862: {n,}? n or more, lazy
7863:
7864:
7865: ANCHORS AND SIMPLE ASSERTIONS
7866:
1.4 misha 7867: \b word boundary
1.1 misha 7868: \B not a word boundary
7869: ^ start of subject
7870: also after internal newline in multiline mode
7871: \A start of subject
7872: $ end of subject
7873: also before newline at end of subject
7874: also before internal newline in multiline mode
7875: \Z end of subject
7876: also before newline at end of subject
7877: \z end of subject
7878: \G first matching position in subject
7879:
7880:
7881: MATCH POINT RESET
7882:
7883: \K reset start of match
7884:
1.7 ! moko 7885: \K is honoured in positive assertions, but ignored in negative ones.
! 7886:
1.1 misha 7887:
7888: ALTERNATION
7889:
7890: expr|expr|expr...
7891:
7892:
7893: CAPTURING
7894:
1.3 misha 7895: (...) capturing group
7896: (?<name>...) named capturing group (Perl)
7897: (?'name'...) named capturing group (Perl)
7898: (?P<name>...) named capturing group (Python)
7899: (?:...) non-capturing group
7900: (?|...) non-capturing group; reset group numbers for
7901: capturing groups in each alternative
1.1 misha 7902:
7903:
7904: ATOMIC GROUPS
7905:
1.3 misha 7906: (?>...) atomic, non-capturing group
1.1 misha 7907:
7908:
7909: COMMENT
7910:
1.3 misha 7911: (?#....) comment (not nestable)
1.1 misha 7912:
7913:
7914: OPTION SETTING
7915:
1.3 misha 7916: (?i) caseless
7917: (?J) allow duplicate names
7918: (?m) multiline
7919: (?s) single line (dotall)
7920: (?U) default ungreedy (lazy)
7921: (?x) extended (ignore white space)
7922: (?-...) unset option(s)
7923:
1.7 ! moko 7924: The following are recognized only at the very start of a pattern or
! 7925: after one of the newline or \R options with similar syntax. More than
! 7926: one of them may appear.
1.3 misha 7927:
1.6 misha 7928: (*LIMIT_MATCH=d) set the match limit to d (decimal number)
7929: (*LIMIT_RECURSION=d) set the recursion limit to d (decimal number)
1.7 ! moko 7930: (*NO_AUTO_POSSESS) no auto-possessification (PCRE_NO_AUTO_POSSESS)
1.4 misha 7931: (*NO_START_OPT) no start-match optimization (PCRE_NO_START_OPTIMIZE)
1.5 misha 7932: (*UTF8) set UTF-8 mode: 8-bit library (PCRE_UTF8)
7933: (*UTF16) set UTF-16 mode: 16-bit library (PCRE_UTF16)
1.6 misha 7934: (*UTF32) set UTF-32 mode: 32-bit library (PCRE_UTF32)
7935: (*UTF) set appropriate UTF mode for the library in use
1.4 misha 7936: (*UCP) set PCRE_UCP (use Unicode properties for \d etc)
1.1 misha 7937:
1.7 ! moko 7938: Note that LIMIT_MATCH and LIMIT_RECURSION can only reduce the value of
! 7939: the limits set by the caller of pcre_exec(), not increase them.
! 7940:
! 7941:
! 7942: NEWLINE CONVENTION
! 7943:
! 7944: These are recognized only at the very start of the pattern or after
! 7945: option settings with a similar syntax.
! 7946:
! 7947: (*CR) carriage return only
! 7948: (*LF) linefeed only
! 7949: (*CRLF) carriage return followed by linefeed
! 7950: (*ANYCRLF) all three of the above
! 7951: (*ANY) any Unicode newline sequence
! 7952:
! 7953:
! 7954: WHAT \R MATCHES
! 7955:
! 7956: These are recognized only at the very start of the pattern or after
! 7957: option setting with a similar syntax.
! 7958:
! 7959: (*BSR_ANYCRLF) CR, LF, or CRLF
! 7960: (*BSR_UNICODE) any Unicode newline sequence
! 7961:
1.1 misha 7962:
7963: LOOKAHEAD AND LOOKBEHIND ASSERTIONS
7964:
1.3 misha 7965: (?=...) positive look ahead
7966: (?!...) negative look ahead
7967: (?<=...) positive look behind
7968: (?<!...) negative look behind
1.1 misha 7969:
7970: Each top-level branch of a look behind must be of a fixed length.
7971:
7972:
7973: BACKREFERENCES
7974:
1.3 misha 7975: \n reference by number (can be ambiguous)
7976: \gn reference by number
7977: \g{n} reference by number
7978: \g{-n} relative reference by number
7979: \k<name> reference by name (Perl)
7980: \k'name' reference by name (Perl)
7981: \g{name} reference by name (Perl)
7982: \k{name} reference by name (.NET)
7983: (?P=name) reference by name (Python)
1.1 misha 7984:
7985:
7986: SUBROUTINE REFERENCES (POSSIBLY RECURSIVE)
7987:
1.3 misha 7988: (?R) recurse whole pattern
7989: (?n) call subpattern by absolute number
7990: (?+n) call subpattern by relative number
7991: (?-n) call subpattern by relative number
7992: (?&name) call subpattern by name (Perl)
7993: (?P>name) call subpattern by name (Python)
7994: \g<name> call subpattern by name (Oniguruma)
7995: \g'name' call subpattern by name (Oniguruma)
7996: \g<n> call subpattern by absolute number (Oniguruma)
7997: \g'n' call subpattern by absolute number (Oniguruma)
7998: \g<+n> call subpattern by relative number (PCRE extension)
7999: \g'+n' call subpattern by relative number (PCRE extension)
8000: \g<-n> call subpattern by relative number (PCRE extension)
8001: \g'-n' call subpattern by relative number (PCRE extension)
1.1 misha 8002:
8003:
8004: CONDITIONAL PATTERNS
8005:
8006: (?(condition)yes-pattern)
8007: (?(condition)yes-pattern|no-pattern)
8008:
1.3 misha 8009: (?(n)... absolute reference condition
8010: (?(+n)... relative reference condition
8011: (?(-n)... relative reference condition
8012: (?(<name>)... named reference condition (Perl)
8013: (?('name')... named reference condition (Perl)
8014: (?(name)... named reference condition (PCRE)
8015: (?(R)... overall recursion condition
8016: (?(Rn)... specific group recursion condition
8017: (?(R&name)... specific recursion condition
8018: (?(DEFINE)... define subpattern for reference
8019: (?(assert)... assertion condition
1.1 misha 8020:
8021:
8022: BACKTRACKING CONTROL
8023:
8024: The following act immediately they are reached:
8025:
1.3 misha 8026: (*ACCEPT) force successful match
8027: (*FAIL) force backtrack; synonym (*F)
1.5 misha 8028: (*MARK:NAME) set name to be passed back; synonym (*:NAME)
1.1 misha 8029:
1.7 ! moko 8030: The following act only when a subsequent match failure causes a back-
1.1 misha 8031: track to reach them. They all force a match failure, but they differ in
8032: what happens afterwards. Those that advance the start-of-match point do
8033: so only if the pattern is not anchored.
8034:
1.3 misha 8035: (*COMMIT) overall failure, no advance of starting point
8036: (*PRUNE) advance to next starting character
1.5 misha 8037: (*PRUNE:NAME) equivalent to (*MARK:NAME)(*PRUNE)
8038: (*SKIP) advance to current matching position
8039: (*SKIP:NAME) advance to position corresponding to an earlier
8040: (*MARK:NAME); if not found, the (*SKIP) is ignored
1.3 misha 8041: (*THEN) local failure, backtrack to next alternation
1.5 misha 8042: (*THEN:NAME) equivalent to (*MARK:NAME)(*THEN)
1.1 misha 8043:
8044:
8045: CALLOUTS
8046:
8047: (?C) callout
8048: (?Cn) callout with data n
8049:
8050:
8051: SEE ALSO
8052:
8053: pcrepattern(3), pcreapi(3), pcrecallout(3), pcrematching(3), pcre(3).
8054:
8055:
8056: AUTHOR
8057:
8058: Philip Hazel
8059: University Computing Service
8060: Cambridge CB2 3QH, England.
8061:
8062:
8063: REVISION
8064:
1.7 ! moko 8065: Last updated: 08 January 2014
! 8066: Copyright (c) 1997-2014 University of Cambridge.
1.5 misha 8067: ------------------------------------------------------------------------------
8068:
8069:
1.6 misha 8070: PCREUNICODE(3) Library Functions Manual PCREUNICODE(3)
8071:
1.5 misha 8072:
8073:
8074: NAME
8075: PCRE - Perl-compatible regular expressions
8076:
1.6 misha 8077: UTF-8, UTF-16, UTF-32, AND UNICODE PROPERTY SUPPORT
1.5 misha 8078:
1.6 misha 8079: As well as UTF-8 support, PCRE also supports UTF-16 (from release 8.30)
8080: and UTF-32 (from release 8.32), by means of two additional libraries.
8081: They can be built as well as, or instead of, the 8-bit library.
1.5 misha 8082:
8083:
8084: UTF-8 SUPPORT
8085:
8086: In order process UTF-8 strings, you must build PCRE's 8-bit library
8087: with UTF support, and, in addition, you must call pcre_compile() with
8088: the PCRE_UTF8 option flag, or the pattern must start with the sequence
1.6 misha 8089: (*UTF8) or (*UTF). When either of these is the case, both the pattern
8090: and any subject strings that are matched against it are treated as
8091: UTF-8 strings instead of strings of individual 1-byte characters.
8092:
8093:
8094: UTF-16 AND UTF-32 SUPPORT
8095:
8096: In order process UTF-16 or UTF-32 strings, you must build PCRE's 16-bit
8097: or 32-bit library with UTF support, and, in addition, you must call
8098: pcre16_compile() or pcre32_compile() with the PCRE_UTF16 or PCRE_UTF32
8099: option flag, as appropriate. Alternatively, the pattern must start with
8100: the sequence (*UTF16), (*UTF32), as appropriate, or (*UTF), which can
8101: be used with either library. When UTF mode is set, both the pattern and
8102: any subject strings that are matched against it are treated as UTF-16
8103: or UTF-32 strings instead of strings of individual 16-bit or 32-bit
8104: characters.
1.5 misha 8105:
8106:
8107: UTF SUPPORT OVERHEAD
8108:
1.6 misha 8109: If you compile PCRE with UTF support, but do not use it at run time,
8110: the library will be a bit bigger, but the additional run time overhead
8111: is limited to testing the PCRE_UTF[8|16|32] flag occasionally, so
8112: should not be very big.
1.5 misha 8113:
8114:
8115: UNICODE PROPERTY SUPPORT
8116:
8117: If PCRE is built with Unicode character property support (which implies
1.6 misha 8118: UTF support), the escape sequences \p{..}, \P{..}, and \X can be used.
8119: The available properties that can be tested are limited to the general
8120: category properties such as Lu for an upper case letter or Nd for a
1.5 misha 8121: decimal number, the Unicode script names such as Arabic or Han, and the
1.6 misha 8122: derived properties Any and L&. Full lists is given in the pcrepattern
8123: and pcresyntax documentation. Only the short names for properties are
8124: supported. For example, \p{L} matches a letter. Its Perl synonym,
8125: \p{Letter}, is not supported. Furthermore, in Perl, many properties
8126: may optionally be prefixed by "Is", for compatibility with Perl 5.6.
8127: PCRE does not support this.
1.5 misha 8128:
8129: Validity of UTF-8 strings
8130:
1.6 misha 8131: When you set the PCRE_UTF8 flag, the byte strings passed as patterns
1.5 misha 8132: and subjects are (by default) checked for validity on entry to the rel-
1.6 misha 8133: evant functions. The entire string is checked before any other process-
8134: ing takes place. From release 7.3 of PCRE, the check is according the
1.5 misha 8135: rules of RFC 3629, which are themselves derived from the Unicode speci-
8136: fication. Earlier releases of PCRE followed the rules of RFC 2279,
8137: which allows the full range of 31-bit values (0 to 0x7FFFFFFF). The
8138: current check allows only values in the range U+0 to U+10FFFF, exclud-
1.6 misha 8139: ing the surrogate area. (From release 8.33 the so-called "non-charac-
8140: ter" code points are no longer excluded because Unicode corrigendum #9
8141: makes it clear that they should not be.)
8142:
8143: Characters in the "Surrogate Area" of Unicode are reserved for use by
8144: UTF-16, where they are used in pairs to encode codepoints with values
8145: greater than 0xFFFF. The code points that are encoded by UTF-16 pairs
8146: are available independently in the UTF-8 and UTF-32 encodings. (In
8147: other words, the whole surrogate thing is a fudge for UTF-16 which
8148: unfortunately messes up UTF-8 and UTF-32.)
1.5 misha 8149:
8150: If an invalid UTF-8 string is passed to PCRE, an error return is given.
8151: At compile time, the only additional information is the offset to the
1.6 misha 8152: first byte of the failing character. The run-time functions pcre_exec()
1.5 misha 8153: and pcre_dfa_exec() also pass back this information, as well as a more
8154: detailed reason code if the caller has provided memory in which to do
8155: this.
8156:
8157: In some situations, you may already know that your strings are valid,
8158: and therefore want to skip these checks in order to improve perfor-
1.6 misha 8159: mance, for example in the case of a long subject string that is being
8160: scanned repeatedly. If you set the PCRE_NO_UTF8_CHECK flag at compile
8161: time or at run time, PCRE assumes that the pattern or subject it is
8162: given (respectively) contains only valid UTF-8 codes. In this case, it
8163: does not diagnose an invalid UTF-8 string.
8164:
8165: Note that passing PCRE_NO_UTF8_CHECK to pcre_compile() just disables
8166: the check for the pattern; it does not also apply to subject strings.
8167: If you want to disable the check for a subject string you must pass
8168: this option to pcre_exec() or pcre_dfa_exec().
8169:
8170: If you pass an invalid UTF-8 string when PCRE_NO_UTF8_CHECK is set, the
8171: result is undefined and your program may crash.
1.5 misha 8172:
8173: Validity of UTF-16 strings
8174:
8175: When you set the PCRE_UTF16 flag, the strings of 16-bit data units that
8176: are passed as patterns and subjects are (by default) checked for valid-
8177: ity on entry to the relevant functions. Values other than those in the
8178: surrogate range U+D800 to U+DFFF are independent code points. Values in
8179: the surrogate range must be used in pairs in the correct manner.
8180:
8181: If an invalid UTF-16 string is passed to PCRE, an error return is
8182: given. At compile time, the only additional information is the offset
1.6 misha 8183: to the first data unit of the failing character. The run-time functions
1.5 misha 8184: pcre16_exec() and pcre16_dfa_exec() also pass back this information, as
8185: well as a more detailed reason code if the caller has provided memory
8186: in which to do this.
8187:
8188: In some situations, you may already know that your strings are valid,
8189: and therefore want to skip these checks in order to improve perfor-
8190: mance. If you set the PCRE_NO_UTF16_CHECK flag at compile time or at
8191: run time, PCRE assumes that the pattern or subject it is given (respec-
8192: tively) contains only valid UTF-16 sequences. In this case, it does not
1.6 misha 8193: diagnose an invalid UTF-16 string. However, if an invalid string is
8194: passed, the result is undefined.
8195:
8196: Validity of UTF-32 strings
8197:
8198: When you set the PCRE_UTF32 flag, the strings of 32-bit data units that
8199: are passed as patterns and subjects are (by default) checked for valid-
8200: ity on entry to the relevant functions. This check allows only values
8201: in the range U+0 to U+10FFFF, excluding the surrogate area U+D800 to
8202: U+DFFF.
8203:
8204: If an invalid UTF-32 string is passed to PCRE, an error return is
8205: given. At compile time, the only additional information is the offset
8206: to the first data unit of the failing character. The run-time functions
8207: pcre32_exec() and pcre32_dfa_exec() also pass back this information, as
8208: well as a more detailed reason code if the caller has provided memory
8209: in which to do this.
8210:
8211: In some situations, you may already know that your strings are valid,
8212: and therefore want to skip these checks in order to improve perfor-
8213: mance. If you set the PCRE_NO_UTF32_CHECK flag at compile time or at
8214: run time, PCRE assumes that the pattern or subject it is given (respec-
8215: tively) contains only valid UTF-32 sequences. In this case, it does not
8216: diagnose an invalid UTF-32 string. However, if an invalid string is
8217: passed, the result is undefined.
1.5 misha 8218:
8219: General comments about UTF modes
8220:
1.6 misha 8221: 1. Codepoints less than 256 can be specified in patterns by either
8222: braced or unbraced hexadecimal escape sequences (for example, \x{b3} or
8223: \xb3). Larger values have to use braced sequences.
1.5 misha 8224:
1.6 misha 8225: 2. Octal numbers up to \777 are recognized, and in UTF-8 mode they
1.5 misha 8226: match two-byte characters for values greater than \177.
8227:
8228: 3. Repeat quantifiers apply to complete UTF characters, not to individ-
8229: ual data units, for example: \x{100}{3}.
8230:
8231: 4. The dot metacharacter matches one UTF character instead of a single
8232: data unit.
8233:
8234: 5. The escape sequence \C can be used to match a single byte in UTF-8
1.6 misha 8235: mode, or a single 16-bit data unit in UTF-16 mode, or a single 32-bit
8236: data unit in UTF-32 mode, but its use can lead to some strange effects
8237: because it breaks up multi-unit characters (see the description of \C
8238: in the pcrepattern documentation). The use of \C is not supported in
8239: the alternative matching function pcre[16|32]_dfa_exec(), nor is it
8240: supported in UTF mode by the JIT optimization of pcre[16|32]_exec(). If
8241: JIT optimization is requested for a UTF pattern that contains \C, it
8242: will not succeed, and so the matching will be carried out by the normal
8243: interpretive function.
1.5 misha 8244:
1.6 misha 8245: 6. The character escapes \b, \B, \d, \D, \s, \S, \w, and \W correctly
1.5 misha 8246: test characters of any code value, but, by default, the characters that
1.6 misha 8247: PCRE recognizes as digits, spaces, or word characters remain the same
8248: set as in non-UTF mode, all with values less than 256. This remains
8249: true even when PCRE is built to include Unicode property support,
1.5 misha 8250: because to do otherwise would slow down PCRE in many common cases. Note
1.6 misha 8251: in particular that this applies to \b and \B, because they are defined
1.5 misha 8252: in terms of \w and \W. If you really want to test for a wider sense of,
1.6 misha 8253: say, "digit", you can use explicit Unicode property tests such as
1.5 misha 8254: \p{Nd}. Alternatively, if you set the PCRE_UCP option, the way that the
1.6 misha 8255: character escapes work is changed so that Unicode properties are used
1.5 misha 8256: to determine which characters match. There are more details in the sec-
8257: tion on generic character types in the pcrepattern documentation.
8258:
1.6 misha 8259: 7. Similarly, characters that match the POSIX named character classes
1.5 misha 8260: are all low-valued characters, unless the PCRE_UCP option is set.
8261:
1.6 misha 8262: 8. However, the horizontal and vertical white space matching escapes
8263: (\h, \H, \v, and \V) do match all the appropriate Unicode characters,
1.5 misha 8264: whether or not PCRE_UCP is set.
8265:
1.6 misha 8266: 9. Case-insensitive matching applies only to characters whose values
8267: are less than 128, unless PCRE is built with Unicode property support.
8268: A few Unicode characters such as Greek sigma have more than two code-
8269: points that are case-equivalent. Up to and including PCRE release 8.31,
8270: only one-to-one case mappings were supported, but later releases (with
8271: Unicode property support) do treat as case-equivalent all versions of
8272: characters such as Greek sigma.
1.5 misha 8273:
8274:
8275: AUTHOR
8276:
8277: Philip Hazel
8278: University Computing Service
8279: Cambridge CB2 3QH, England.
8280:
8281:
8282: REVISION
8283:
1.6 misha 8284: Last updated: 27 February 2013
8285: Copyright (c) 1997-2013 University of Cambridge.
1.5 misha 8286: ------------------------------------------------------------------------------
8287:
8288:
1.6 misha 8289: PCREJIT(3) Library Functions Manual PCREJIT(3)
8290:
1.5 misha 8291:
8292:
8293: NAME
8294: PCRE - Perl-compatible regular expressions
8295:
8296: PCRE JUST-IN-TIME COMPILER SUPPORT
8297:
8298: Just-in-time compiling is a heavyweight optimization that can greatly
8299: speed up pattern matching. However, it comes at the cost of extra pro-
8300: cessing before the match is performed. Therefore, it is of most benefit
8301: when the same pattern is going to be matched many times. This does not
8302: necessarily mean many calls of a matching function; if the pattern is
8303: not anchored, matching attempts may take place many times at various
8304: positions in the subject, even for a single call. Therefore, if the
8305: subject string is very long, it may still pay to use JIT for one-off
8306: matches.
8307:
8308: JIT support applies only to the traditional Perl-compatible matching
8309: function. It does not apply when the DFA matching function is being
8310: used. The code for this support was written by Zoltan Herczeg.
8311:
8312:
1.6 misha 8313: 8-BIT, 16-BIT AND 32-BIT SUPPORT
1.5 misha 8314:
1.6 misha 8315: JIT support is available for all of the 8-bit, 16-bit and 32-bit PCRE
8316: libraries. To keep this documentation simple, only the 8-bit interface
8317: is described in what follows. If you are using the 16-bit library, sub-
8318: stitute the 16-bit functions and 16-bit structures (for example,
8319: pcre16_jit_stack instead of pcre_jit_stack). If you are using the
8320: 32-bit library, substitute the 32-bit functions and 32-bit structures
8321: (for example, pcre32_jit_stack instead of pcre_jit_stack).
1.5 misha 8322:
8323:
8324: AVAILABILITY OF JIT SUPPORT
8325:
8326: JIT support is an optional feature of PCRE. The "configure" option
8327: --enable-jit (or equivalent CMake option) must be set when PCRE is
8328: built if you want to use JIT. The support is limited to the following
8329: hardware platforms:
8330:
8331: ARM v5, v7, and Thumb2
8332: Intel x86 32-bit and 64-bit
8333: MIPS 32-bit
8334: Power PC 32-bit and 64-bit
1.6 misha 8335: SPARC 32-bit (experimental)
1.5 misha 8336:
1.6 misha 8337: If --enable-jit is set on an unsupported platform, compilation fails.
1.5 misha 8338:
8339: A program that is linked with PCRE 8.20 or later can tell if JIT sup-
8340: port is available by calling pcre_config() with the PCRE_CONFIG_JIT
8341: option. The result is 1 when JIT is available, and 0 otherwise. How-
8342: ever, a simple program does not need to check this in order to use JIT.
1.6 misha 8343: The normal API is implemented in a way that falls back to the interpre-
8344: tive code if JIT is not available. For programs that need the best pos-
8345: sible performance, there is also a "fast path" API that is JIT-spe-
8346: cific.
1.5 misha 8347:
8348: If your program may sometimes be linked with versions of PCRE that are
8349: older than 8.20, but you want to use JIT when it is available, you can
8350: test the values of PCRE_MAJOR and PCRE_MINOR, or the existence of a JIT
8351: macro such as PCRE_CONFIG_JIT, for compile-time control of your code.
8352:
8353:
8354: SIMPLE USE OF JIT
8355:
8356: You have to do two things to make use of the JIT support in the sim-
8357: plest way:
8358:
8359: (1) Call pcre_study() with the PCRE_STUDY_JIT_COMPILE option for
8360: each compiled pattern, and pass the resulting pcre_extra block to
8361: pcre_exec().
8362:
8363: (2) Use pcre_free_study() to free the pcre_extra block when it is
1.6 misha 8364: no longer needed, instead of just freeing it yourself. This
8365: ensures that
8366: any JIT data is also freed.
1.5 misha 8367:
1.6 misha 8368: For a program that may be linked with pre-8.20 versions of PCRE, you
1.5 misha 8369: can insert
8370:
8371: #ifndef PCRE_STUDY_JIT_COMPILE
8372: #define PCRE_STUDY_JIT_COMPILE 0
8373: #endif
8374:
1.6 misha 8375: so that no option is passed to pcre_study(), and then use something
1.5 misha 8376: like this to free the study data:
8377:
8378: #ifdef PCRE_CONFIG_JIT
8379: pcre_free_study(study_ptr);
8380: #else
8381: pcre_free(study_ptr);
8382: #endif
8383:
1.6 misha 8384: PCRE_STUDY_JIT_COMPILE requests the JIT compiler to generate code for
8385: complete matches. If you want to run partial matches using the
8386: PCRE_PARTIAL_HARD or PCRE_PARTIAL_SOFT options of pcre_exec(), you
8387: should set one or both of the following options in addition to, or
8388: instead of, PCRE_STUDY_JIT_COMPILE when you call pcre_study():
8389:
8390: PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE
8391: PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE
8392:
8393: The JIT compiler generates different optimized code for each of the
8394: three modes (normal, soft partial, hard partial). When pcre_exec() is
8395: called, the appropriate code is run if it is available. Otherwise, the
8396: pattern is matched using interpretive code.
8397:
1.5 misha 8398: In some circumstances you may need to call additional functions. These
8399: are described in the section entitled "Controlling the JIT stack"
8400: below.
8401:
1.6 misha 8402: If JIT support is not available, PCRE_STUDY_JIT_COMPILE etc. are
8403: ignored, and no JIT data is created. Otherwise, the compiled pattern is
8404: passed to the JIT compiler, which turns it into machine code that exe-
8405: cutes much faster than the normal interpretive code. When pcre_exec()
8406: is passed a pcre_extra block containing a pointer to JIT code of the
8407: appropriate mode (normal or hard/soft partial), it obeys that code
8408: instead of running the interpreter. The result is identical, but the
8409: compiled JIT code runs much faster.
8410:
8411: There are some pcre_exec() options that are not supported for JIT exe-
8412: cution. There are also some pattern items that JIT cannot handle.
8413: Details are given below. In both cases, execution automatically falls
8414: back to the interpretive code. If you want to know whether JIT was
8415: actually used for a particular match, you should arrange for a JIT
8416: callback function to be set up as described in the section entitled
8417: "Controlling the JIT stack" below, even if you do not need to supply a
8418: non-default JIT stack. Such a callback function is called whenever JIT
8419: code is about to be obeyed. If the execution options are not right for
8420: JIT execution, the callback function is not obeyed.
8421:
8422: If the JIT compiler finds an unsupported item, no JIT data is gener-
8423: ated. You can find out if JIT execution is available after studying a
8424: pattern by calling pcre_fullinfo() with the PCRE_INFO_JIT option. A
8425: result of 1 means that JIT compilation was successful. A result of 0
1.5 misha 8426: means that JIT support is not available, or the pattern was not studied
1.6 misha 8427: with PCRE_STUDY_JIT_COMPILE etc., or the JIT compiler was not able to
8428: handle the pattern.
1.5 misha 8429:
8430: Once a pattern has been studied, with or without JIT, it can be used as
8431: many times as you like for matching different subject strings.
8432:
8433:
8434: UNSUPPORTED OPTIONS AND PATTERN ITEMS
8435:
1.6 misha 8436: The only pcre_exec() options that are supported for JIT execution are
8437: PCRE_NO_UTF8_CHECK, PCRE_NO_UTF16_CHECK, PCRE_NO_UTF32_CHECK, PCRE_NOT-
8438: BOL, PCRE_NOTEOL, PCRE_NOTEMPTY, PCRE_NOTEMPTY_ATSTART, PCRE_PAR-
8439: TIAL_HARD, and PCRE_PARTIAL_SOFT.
8440:
8441: The only unsupported pattern items are \C (match a single data unit)
8442: when running in a UTF mode, and a callout immediately before an asser-
8443: tion condition in a conditional group.
1.5 misha 8444:
8445:
8446: RETURN VALUES FROM JIT EXECUTION
8447:
1.6 misha 8448: When a pattern is matched using JIT execution, the return values are
8449: the same as those given by the interpretive pcre_exec() code, with the
8450: addition of one new error code: PCRE_ERROR_JIT_STACKLIMIT. This means
8451: that the memory used for the JIT stack was insufficient. See "Control-
1.5 misha 8452: ling the JIT stack" below for a discussion of JIT stack usage. For com-
1.6 misha 8453: patibility with the interpretive pcre_exec() code, no more than two-
8454: thirds of the ovector argument is used for passing back captured sub-
1.5 misha 8455: strings.
8456:
1.6 misha 8457: The error code PCRE_ERROR_MATCHLIMIT is returned by the JIT code if
8458: searching a very large pattern tree goes on for too long, as it is in
8459: the same circumstance when JIT is not used, but the details of exactly
8460: what is counted are not the same. The PCRE_ERROR_RECURSIONLIMIT error
1.5 misha 8461: code is never returned by JIT execution.
8462:
8463:
8464: SAVING AND RESTORING COMPILED PATTERNS
8465:
1.6 misha 8466: The code that is generated by the JIT compiler is architecture-spe-
8467: cific, and is also position dependent. For those reasons it cannot be
8468: saved (in a file or database) and restored later like the bytecode and
8469: other data of a compiled pattern. Saving and restoring compiled pat-
8470: terns is not something many people do. More detail about this facility
8471: is given in the pcreprecompile documentation. It should be possible to
8472: run pcre_study() on a saved and restored pattern, and thereby recreate
8473: the JIT data, but because JIT compilation uses significant resources,
8474: it is probably not worth doing this; you might as well recompile the
1.5 misha 8475: original pattern.
8476:
8477:
8478: CONTROLLING THE JIT STACK
8479:
8480: When the compiled JIT code runs, it needs a block of memory to use as a
1.6 misha 8481: stack. By default, it uses 32K on the machine stack. However, some
8482: large or complicated patterns need more than this. The error
8483: PCRE_ERROR_JIT_STACKLIMIT is given when there is not enough stack.
8484: Three functions are provided for managing blocks of memory for use as
8485: JIT stacks. There is further discussion about the use of JIT stacks in
1.5 misha 8486: the section entitled "JIT stack FAQ" below.
8487:
1.6 misha 8488: The pcre_jit_stack_alloc() function creates a JIT stack. Its arguments
8489: are a starting size and a maximum size, and it returns a pointer to an
8490: opaque structure of type pcre_jit_stack, or NULL if there is an error.
8491: The pcre_jit_stack_free() function can be used to free a stack that is
8492: no longer needed. (For the technically minded: the address space is
1.5 misha 8493: allocated by mmap or VirtualAlloc.)
8494:
1.6 misha 8495: JIT uses far less memory for recursion than the interpretive code, and
8496: a maximum stack size of 512K to 1M should be more than enough for any
1.5 misha 8497: pattern.
8498:
1.6 misha 8499: The pcre_assign_jit_stack() function specifies which stack JIT code
1.5 misha 8500: should use. Its arguments are as follows:
8501:
8502: pcre_extra *extra
8503: pcre_jit_callback callback
8504: void *data
8505:
1.6 misha 8506: The extra argument must be the result of studying a pattern with
8507: PCRE_STUDY_JIT_COMPILE etc. There are three cases for the values of the
1.5 misha 8508: other two options:
8509:
8510: (1) If callback is NULL and data is NULL, an internal 32K block
8511: on the machine stack is used.
8512:
8513: (2) If callback is NULL and data is not NULL, data must be
8514: a valid JIT stack, the result of calling pcre_jit_stack_alloc().
8515:
1.6 misha 8516: (3) If callback is not NULL, it must point to a function that is
8517: called with data as an argument at the start of matching, in
8518: order to set up a JIT stack. If the return from the callback
8519: function is NULL, the internal 32K stack is used; otherwise the
8520: return value must be a valid JIT stack, the result of calling
8521: pcre_jit_stack_alloc().
8522:
8523: A callback function is obeyed whenever JIT code is about to be run; it
8524: is not obeyed when pcre_exec() is called with options that are incom-
8525: patible for JIT execution. A callback function can therefore be used to
8526: determine whether a match operation was executed by JIT or by the
8527: interpreter.
8528:
8529: You may safely use the same JIT stack for more than one pattern (either
8530: by assigning directly or by callback), as long as the patterns are all
8531: matched sequentially in the same thread. In a multithread application,
8532: if you do not specify a JIT stack, or if you assign or pass back NULL
8533: from a callback, that is thread-safe, because each thread has its own
8534: machine stack. However, if you assign or pass back a non-NULL JIT
8535: stack, this must be a different stack for each thread so that the
8536: application is thread-safe.
8537:
8538: Strictly speaking, even more is allowed. You can assign the same non-
8539: NULL stack to any number of patterns as long as they are not used for
8540: matching by multiple threads at the same time. For example, you can
8541: assign the same stack to all compiled patterns, and use a global mutex
8542: in the callback to wait until the stack is available for use. However,
8543: this is an inefficient solution, and not recommended.
1.5 misha 8544:
1.6 misha 8545: This is a suggestion for how a multithreaded program that needs to set
8546: up non-default JIT stacks might operate:
1.5 misha 8547:
8548: During thread initalization
8549: thread_local_var = pcre_jit_stack_alloc(...)
8550:
8551: During thread exit
8552: pcre_jit_stack_free(thread_local_var)
8553:
8554: Use a one-line callback function
8555: return thread_local_var
8556:
8557: All the functions described in this section do nothing if JIT is not
8558: available, and pcre_assign_jit_stack() does nothing unless the extra
8559: argument is non-NULL and points to a pcre_extra block that is the
1.6 misha 8560: result of a successful study with PCRE_STUDY_JIT_COMPILE etc.
1.5 misha 8561:
8562:
8563: JIT STACK FAQ
8564:
8565: (1) Why do we need JIT stacks?
8566:
8567: PCRE (and JIT) is a recursive, depth-first engine, so it needs a stack
8568: where the local data of the current node is pushed before checking its
8569: child nodes. Allocating real machine stack on some platforms is diffi-
8570: cult. For example, the stack chain needs to be updated every time if we
8571: extend the stack on PowerPC. Although it is possible, its updating
8572: time overhead decreases performance. So we do the recursion in memory.
8573:
8574: (2) Why don't we simply allocate blocks of memory with malloc()?
8575:
8576: Modern operating systems have a nice feature: they can reserve an
8577: address space instead of allocating memory. We can safely allocate mem-
8578: ory pages inside this address space, so the stack could grow without
8579: moving memory data (this is important because of pointers). Thus we can
8580: allocate 1M address space, and use only a single memory page (usually
8581: 4K) if that is enough. However, we can still grow up to 1M anytime if
8582: needed.
8583:
8584: (3) Who "owns" a JIT stack?
8585:
8586: The owner of the stack is the user program, not the JIT studied pattern
8587: or anything else. The user program must ensure that if a stack is used
8588: by pcre_exec(), (that is, it is assigned to the pattern currently run-
8589: ning), that stack must not be used by any other threads (to avoid over-
8590: writing the same memory area). The best practice for multithreaded pro-
8591: grams is to allocate a stack for each thread, and return this stack
8592: through the JIT callback function.
8593:
8594: (4) When should a JIT stack be freed?
8595:
8596: You can free a JIT stack at any time, as long as it will not be used by
8597: pcre_exec() again. When you assign the stack to a pattern, only a
8598: pointer is set. There is no reference counting or any other magic. You
8599: can free the patterns and stacks in any order, anytime. Just do not
8600: call pcre_exec() with a pattern pointing to an already freed stack, as
8601: that will cause SEGFAULT. (Also, do not free a stack currently used by
8602: pcre_exec() in another thread). You can also replace the stack for a
8603: pattern at any time. You can even free the previous stack before
8604: assigning a replacement.
8605:
8606: (5) Should I allocate/free a stack every time before/after calling
8607: pcre_exec()?
8608:
8609: No, because this is too costly in terms of resources. However, you
8610: could implement some clever idea which release the stack if it is not
1.6 misha 8611: used in let's say two minutes. The JIT callback can help to achieve
8612: this without keeping a list of the currently JIT studied patterns.
1.5 misha 8613:
8614: (6) OK, the stack is for long term memory allocation. But what happens
8615: if a pattern causes stack overflow with a stack of 1M? Is that 1M kept
8616: until the stack is freed?
8617:
8618: Especially on embedded sytems, it might be a good idea to release mem-
8619: ory sometimes without freeing the stack. There is no API for this at
1.6 misha 8620: the moment. Probably a function call which returns with the currently
1.5 misha 8621: allocated memory for any stack and another which allows releasing mem-
8622: ory (shrinking the stack) would be a good idea if someone needs this.
8623:
8624: (7) This is too much of a headache. Isn't there any better solution for
8625: JIT stack handling?
8626:
8627: No, thanks to Windows. If POSIX threads were used everywhere, we could
8628: throw out this complicated API.
8629:
8630:
8631: EXAMPLE CODE
8632:
8633: This is a single-threaded example that specifies a JIT stack without
8634: using a callback.
8635:
8636: int rc;
8637: int ovector[30];
8638: pcre *re;
8639: pcre_extra *extra;
8640: pcre_jit_stack *jit_stack;
8641:
8642: re = pcre_compile(pattern, 0, &error, &erroffset, NULL);
8643: /* Check for errors */
8644: extra = pcre_study(re, PCRE_STUDY_JIT_COMPILE, &error);
8645: jit_stack = pcre_jit_stack_alloc(32*1024, 512*1024);
8646: /* Check for error (NULL) */
8647: pcre_assign_jit_stack(extra, NULL, jit_stack);
8648: rc = pcre_exec(re, extra, subject, length, 0, 0, ovector, 30);
8649: /* Check results */
8650: pcre_free(re);
8651: pcre_free_study(extra);
8652: pcre_jit_stack_free(jit_stack);
8653:
8654:
1.6 misha 8655: JIT FAST PATH API
8656:
8657: Because the API described above falls back to interpreted execution
8658: when JIT is not available, it is convenient for programs that are writ-
8659: ten for general use in many environments. However, calling JIT via
8660: pcre_exec() does have a performance impact. Programs that are written
8661: for use where JIT is known to be available, and which need the best
8662: possible performance, can instead use a "fast path" API to call JIT
8663: execution directly instead of calling pcre_exec() (obviously only for
8664: patterns that have been successfully studied by JIT).
8665:
8666: The fast path function is called pcre_jit_exec(), and it takes exactly
8667: the same arguments as pcre_exec(), plus one additional argument that
8668: must point to a JIT stack. The JIT stack arrangements described above
8669: do not apply. The return values are the same as for pcre_exec().
8670:
8671: When you call pcre_exec(), as well as testing for invalid options, a
8672: number of other sanity checks are performed on the arguments. For exam-
8673: ple, if the subject pointer is NULL, or its length is negative, an
8674: immediate error is given. Also, unless PCRE_NO_UTF[8|16|32] is set, a
8675: UTF subject string is tested for validity. In the interests of speed,
8676: these checks do not happen on the JIT fast path, and if invalid data is
8677: passed, the result is undefined.
8678:
8679: Bypassing the sanity checks and the pcre_exec() wrapping can give
8680: speedups of more than 10%.
8681:
8682:
1.5 misha 8683: SEE ALSO
8684:
8685: pcreapi(3)
8686:
8687:
8688: AUTHOR
8689:
8690: Philip Hazel (FAQ by Zoltan Herczeg)
8691: University Computing Service
8692: Cambridge CB2 3QH, England.
8693:
8694:
8695: REVISION
8696:
1.6 misha 8697: Last updated: 17 March 2013
8698: Copyright (c) 1997-2013 University of Cambridge.
1.1 misha 8699: ------------------------------------------------------------------------------
8700:
8701:
1.6 misha 8702: PCREPARTIAL(3) Library Functions Manual PCREPARTIAL(3)
8703:
1.1 misha 8704:
8705:
8706: NAME
8707: PCRE - Perl-compatible regular expressions
8708:
8709: PARTIAL MATCHING IN PCRE
8710:
1.5 misha 8711: In normal use of PCRE, if the subject string that is passed to a match-
8712: ing function matches as far as it goes, but is too short to match the
8713: entire pattern, PCRE_ERROR_NOMATCH is returned. There are circumstances
8714: where it might be helpful to distinguish this case from other cases in
8715: which there is no match.
1.1 misha 8716:
8717: Consider, for example, an application where a human is required to type
8718: in data for a field with specific formatting requirements. An example
8719: might be a date in the form ddmmmyy, defined by this pattern:
8720:
8721: ^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$
8722:
8723: If the application sees the user's keystrokes one by one, and can check
8724: that what has been typed so far is potentially valid, it is able to
1.4 misha 8725: raise an error as soon as a mistake is made, by beeping and not
8726: reflecting the character that has been typed, for example. This immedi-
8727: ate feedback is likely to be a better user interface than a check that
8728: is delayed until the entire string has been entered. Partial matching
8729: can also be useful when the subject string is very long and is not all
8730: available at once.
8731:
8732: PCRE supports partial matching by means of the PCRE_PARTIAL_SOFT and
1.5 misha 8733: PCRE_PARTIAL_HARD options, which can be set when calling any of the
8734: matching functions. For backwards compatibility, PCRE_PARTIAL is a syn-
8735: onym for PCRE_PARTIAL_SOFT. The essential difference between the two
8736: options is whether or not a partial match is preferred to an alterna-
8737: tive complete match, though the details differ between the two types of
8738: matching function. If both options are set, PCRE_PARTIAL_HARD takes
8739: precedence.
8740:
1.6 misha 8741: If you want to use partial matching with just-in-time optimized code,
8742: you must call pcre_study(), pcre16_study() or pcre32_study() with one
8743: or both of these options:
8744:
8745: PCRE_STUDY_JIT_PARTIAL_SOFT_COMPILE
8746: PCRE_STUDY_JIT_PARTIAL_HARD_COMPILE
8747:
8748: PCRE_STUDY_JIT_COMPILE should also be set if you are going to run non-
8749: partial matches on the same pattern. If the appropriate JIT study mode
8750: has not been set for a match, the interpretive matching code is used.
8751:
8752: Setting a partial matching option disables two of PCRE's standard opti-
8753: mizations. PCRE remembers the last literal data unit in a pattern, and
8754: abandons matching immediately if it is not present in the subject
1.5 misha 8755: string. This optimization cannot be used for a subject string that
8756: might match only partially. If the pattern was studied, PCRE knows the
8757: minimum length of a matching string, and does not bother to run the
8758: matching function on shorter strings. This optimization is also dis-
8759: abled for partial matching.
8760:
8761:
1.6 misha 8762: PARTIAL MATCHING USING pcre_exec() OR pcre[16|32]_exec()
1.5 misha 8763:
1.6 misha 8764: A partial match occurs during a call to pcre_exec() or
8765: pcre[16|32]_exec() when the end of the subject string is reached suc-
8766: cessfully, but matching cannot continue because more characters are
8767: needed. However, at least one character in the subject must have been
8768: inspected. This character need not form part of the final matched
8769: string; lookbehind assertions and the \K escape sequence provide ways
8770: of inspecting characters before the start of a matched substring. The
8771: requirement for inspecting at least one character exists because an
8772: empty string can always be matched; without such a restriction there
8773: would always be a partial match of an empty string at the end of the
8774: subject.
1.5 misha 8775:
1.6 misha 8776: If there are at least two slots in the offsets vector when a partial
8777: match is returned, the first slot is set to the offset of the earliest
1.5 misha 8778: character that was inspected. For convenience, the second offset points
8779: to the end of the subject so that a substring can easily be identified.
1.6 misha 8780: If there are at least three slots in the offsets vector, the third slot
8781: is set to the offset of the character where matching started.
1.4 misha 8782:
1.6 misha 8783: For the majority of patterns, the contents of the first and third slots
8784: will be the same. However, for patterns that contain lookbehind asser-
8785: tions, or begin with \b or \B, characters before the one where matching
8786: started may have been inspected while carrying out the match. For exam-
8787: ple, consider this pattern:
1.4 misha 8788:
8789: /(?<=abc)123/
8790:
8791: This pattern matches "123", but only if it is preceded by "abc". If the
1.6 misha 8792: subject string is "xyzabc12", the first two offsets after a partial
8793: match are for the substring "abc12", because all these characters were
8794: inspected. However, the third offset is set to 6, because that is the
8795: offset where matching began.
1.4 misha 8796:
8797: What happens when a partial match is identified depends on which of the
8798: two partial matching options are set.
8799:
1.6 misha 8800: PCRE_PARTIAL_SOFT WITH pcre_exec() OR pcre[16|32]_exec()
1.4 misha 8801:
1.6 misha 8802: If PCRE_PARTIAL_SOFT is set when pcre_exec() or pcre[16|32]_exec()
8803: identifies a partial match, the partial match is remembered, but match-
8804: ing continues as normal, and other alternatives in the pattern are
8805: tried. If no complete match can be found, PCRE_ERROR_PARTIAL is
8806: returned instead of PCRE_ERROR_NOMATCH.
8807:
8808: This option is "soft" because it prefers a complete match over a par-
8809: tial match. All the various matching items in a pattern behave as if
8810: the subject string is potentially complete. For example, \z, \Z, and $
8811: match at the end of the subject, as normal, and for \b and \B the end
1.4 misha 8812: of the subject is treated as a non-alphanumeric.
8813:
1.6 misha 8814: If there is more than one partial match, the first one that was found
1.4 misha 8815: provides the data that is returned. Consider this pattern:
8816:
8817: /123\w+X|dogY/
8818:
1.6 misha 8819: If this is matched against the subject string "abc123dog", both alter-
8820: natives fail to match, but the end of the subject is reached during
8821: matching, so PCRE_ERROR_PARTIAL is returned. The offsets are set to 3
8822: and 9, identifying "123dog" as the first partial match that was found.
8823: (In this example, there are two partial matches, because "dog" on its
1.4 misha 8824: own partially matches the second alternative.)
8825:
1.6 misha 8826: PCRE_PARTIAL_HARD WITH pcre_exec() OR pcre[16|32]_exec()
1.4 misha 8827:
1.6 misha 8828: If PCRE_PARTIAL_HARD is set for pcre_exec() or pcre[16|32]_exec(),
8829: PCRE_ERROR_PARTIAL is returned as soon as a partial match is found,
1.5 misha 8830: without continuing to search for possible complete matches. This option
8831: is "hard" because it prefers an earlier partial match over a later com-
1.6 misha 8832: plete match. For this reason, the assumption is made that the end of
8833: the supplied subject string may not be the true end of the available
1.5 misha 8834: data, and so, if \z, \Z, \b, \B, or $ are encountered at the end of the
1.6 misha 8835: subject, the result is PCRE_ERROR_PARTIAL, provided that at least one
1.5 misha 8836: character in the subject has been inspected.
8837:
8838: Setting PCRE_PARTIAL_HARD also affects the way UTF-8 and UTF-16 subject
1.6 misha 8839: strings are checked for validity. Normally, an invalid sequence causes
8840: the error PCRE_ERROR_BADUTF8 or PCRE_ERROR_BADUTF16. However, in the
8841: special case of a truncated character at the end of the subject,
8842: PCRE_ERROR_SHORTUTF8 or PCRE_ERROR_SHORTUTF16 is returned when
1.5 misha 8843: PCRE_PARTIAL_HARD is set.
1.4 misha 8844:
8845: Comparing hard and soft partial matching
8846:
1.6 misha 8847: The difference between the two partial matching options can be illus-
1.4 misha 8848: trated by a pattern such as:
8849:
8850: /dog(sbody)?/
8851:
1.6 misha 8852: This matches either "dog" or "dogsbody", greedily (that is, it prefers
8853: the longer string if possible). If it is matched against the string
8854: "dog" with PCRE_PARTIAL_SOFT, it yields a complete match for "dog".
1.4 misha 8855: However, if PCRE_PARTIAL_HARD is set, the result is PCRE_ERROR_PARTIAL.
1.6 misha 8856: On the other hand, if the pattern is made ungreedy the result is dif-
1.4 misha 8857: ferent:
8858:
8859: /dog(sbody)??/
8860:
1.6 misha 8861: In this case the result is always a complete match because that is
8862: found first, and matching never continues after finding a complete
1.5 misha 8863: match. It might be easier to follow this explanation by thinking of the
8864: two patterns like this:
1.4 misha 8865:
8866: /dog(sbody)?/ is the same as /dogsbody|dog/
8867: /dog(sbody)??/ is the same as /dog|dogsbody/
8868:
1.6 misha 8869: The second pattern will never match "dogsbody", because it will always
1.5 misha 8870: find the shorter match first.
1.4 misha 8871:
8872:
1.6 misha 8873: PARTIAL MATCHING USING pcre_dfa_exec() OR pcre[16|32]_dfa_exec()
1.4 misha 8874:
1.5 misha 8875: The DFA functions move along the subject string character by character,
1.6 misha 8876: without backtracking, searching for all possible matches simultane-
8877: ously. If the end of the subject is reached before the end of the pat-
8878: tern, there is the possibility of a partial match, again provided that
1.5 misha 8879: at least one character has been inspected.
1.4 misha 8880:
1.6 misha 8881: When PCRE_PARTIAL_SOFT is set, PCRE_ERROR_PARTIAL is returned only if
8882: there have been no complete matches. Otherwise, the complete matches
8883: are returned. However, if PCRE_PARTIAL_HARD is set, a partial match
8884: takes precedence over any complete matches. The portion of the string
8885: that was inspected when the longest partial match was found is set as
1.4 misha 8886: the first matching string, provided there are at least two slots in the
8887: offsets vector.
8888:
1.6 misha 8889: Because the DFA functions always search for all possible matches, and
8890: there is no difference between greedy and ungreedy repetition, their
8891: behaviour is different from the standard functions when PCRE_PAR-
8892: TIAL_HARD is set. Consider the string "dog" matched against the
1.5 misha 8893: ungreedy pattern shown above:
1.4 misha 8894:
8895: /dog(sbody)??/
8896:
1.6 misha 8897: Whereas the standard functions stop as soon as they find the complete
8898: match for "dog", the DFA functions also find the partial match for
1.5 misha 8899: "dogsbody", and so return that when PCRE_PARTIAL_HARD is set.
1.4 misha 8900:
8901:
8902: PARTIAL MATCHING AND WORD BOUNDARIES
8903:
1.6 misha 8904: If a pattern ends with one of sequences \b or \B, which test for word
8905: boundaries, partial matching with PCRE_PARTIAL_SOFT can give counter-
1.4 misha 8906: intuitive results. Consider this pattern:
8907:
8908: /\bcat\b/
8909:
8910: This matches "cat", provided there is a word boundary at either end. If
8911: the subject string is "the cat", the comparison of the final "t" with a
1.6 misha 8912: following character cannot take place, so a partial match is found.
8913: However, normal matching carries on, and \b matches at the end of the
8914: subject when the last character is a letter, so a complete match is
8915: found. The result, therefore, is not PCRE_ERROR_PARTIAL. Using
8916: PCRE_PARTIAL_HARD in this case does yield PCRE_ERROR_PARTIAL, because
1.5 misha 8917: then the partial match takes precedence.
1.4 misha 8918:
8919:
8920: FORMERLY RESTRICTED PATTERNS
8921:
8922: For releases of PCRE prior to 8.00, because of the way certain internal
1.6 misha 8923: optimizations were implemented in the pcre_exec() function, the
8924: PCRE_PARTIAL option (predecessor of PCRE_PARTIAL_SOFT) could not be
8925: used with all patterns. From release 8.00 onwards, the restrictions no
8926: longer apply, and partial matching with can be requested for any pat-
1.5 misha 8927: tern.
1.4 misha 8928:
8929: Items that were formerly restricted were repeated single characters and
1.6 misha 8930: repeated metasequences. If PCRE_PARTIAL was set for a pattern that did
8931: not conform to the restrictions, pcre_exec() returned the error code
8932: PCRE_ERROR_BADPARTIAL (-13). This error code is no longer in use. The
8933: PCRE_INFO_OKPARTIAL call to pcre_fullinfo() to find out if a compiled
1.4 misha 8934: pattern can be used for partial matching now always returns 1.
1.1 misha 8935:
8936:
8937: EXAMPLE OF PARTIAL MATCHING USING PCRETEST
8938:
1.6 misha 8939: If the escape sequence \P is present in a pcretest data line, the
8940: PCRE_PARTIAL_SOFT option is used for the match. Here is a run of
1.4 misha 8941: pcretest that uses the date example quoted above:
1.1 misha 8942:
8943: re> /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
8944: data> 25jun04\P
8945: 0: 25jun04
8946: 1: jun
8947: data> 25dec3\P
1.4 misha 8948: Partial match: 23dec3
1.1 misha 8949: data> 3ju\P
1.4 misha 8950: Partial match: 3ju
1.1 misha 8951: data> 3juj\P
8952: No match
8953: data> j\P
8954: No match
8955:
1.6 misha 8956: The first data string is matched completely, so pcretest shows the
8957: matched substrings. The remaining four strings do not match the com-
1.4 misha 8958: plete pattern, but the first two are partial matches. Similar output is
1.5 misha 8959: obtained if DFA matching is used.
1.1 misha 8960:
1.6 misha 8961: If the escape sequence \P is present more than once in a pcretest data
1.4 misha 8962: line, the PCRE_PARTIAL_HARD option is set for the match.
1.1 misha 8963:
8964:
1.6 misha 8965: MULTI-SEGMENT MATCHING WITH pcre_dfa_exec() OR pcre[16|32]_dfa_exec()
1.1 misha 8966:
1.6 misha 8967: When a partial match has been found using a DFA matching function, it
8968: is possible to continue the match by providing additional subject data
8969: and calling the function again with the same compiled regular expres-
8970: sion, this time setting the PCRE_DFA_RESTART option. You must pass the
1.4 misha 8971: same working space as before, because this is where details of the pre-
1.6 misha 8972: vious partial match are stored. Here is an example using pcretest,
8973: using the \R escape sequence to set the PCRE_DFA_RESTART option (\D
1.5 misha 8974: specifies the use of the DFA matching function):
1.1 misha 8975:
8976: re> /^\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d$/
8977: data> 23ja\P\D
8978: Partial match: 23ja
8979: data> n05\R\D
8980: 0: n05
8981:
1.6 misha 8982: The first call has "23ja" as the subject, and requests partial match-
8983: ing; the second call has "n05" as the subject for the continued
8984: (restarted) match. Notice that when the match is complete, only the
8985: last part is shown; PCRE does not retain the previously partially-
8986: matched string. It is up to the calling program to do that if it needs
1.1 misha 8987: to.
8988:
1.7 ! moko 8989: That means that, for an unanchored pattern, if a continued match fails,
! 8990: it is not possible to try again at a new starting point. All this
! 8991: facility is capable of doing is continuing with the previous match
! 8992: attempt. In the previous example, if the second set of data is "ug23"
! 8993: the result is no match, even though there would be a match for "aug23"
! 8994: if the entire string were given at once. Depending on the application,
! 8995: this may or may not be what you want. The only way to allow for start-
! 8996: ing again at the next character is to retain the matched part of the
! 8997: subject and try a new complete match.
! 8998:
1.6 misha 8999: You can set the PCRE_PARTIAL_SOFT or PCRE_PARTIAL_HARD options with
9000: PCRE_DFA_RESTART to continue partial matching over multiple segments.
9001: This facility can be used to pass very long subject strings to the DFA
1.5 misha 9002: matching functions.
9003:
9004:
1.6 misha 9005: MULTI-SEGMENT MATCHING WITH pcre_exec() OR pcre[16|32]_exec()
1.5 misha 9006:
1.6 misha 9007: From release 8.00, the standard matching functions can also be used to
1.5 misha 9008: do multi-segment matching. Unlike the DFA functions, it is not possible
1.6 misha 9009: to restart the previous match with a new segment of data. Instead, new
1.5 misha 9010: data must be added to the previous subject string, and the entire match
1.6 misha 9011: re-run, starting from the point where the partial match occurred. Ear-
1.5 misha 9012: lier data can be discarded.
9013:
1.6 misha 9014: It is best to use PCRE_PARTIAL_HARD in this situation, because it does
9015: not treat the end of a segment as the end of the subject when matching
9016: \z, \Z, \b, \B, and $. Consider an unanchored pattern that matches
1.5 misha 9017: dates:
1.1 misha 9018:
1.4 misha 9019: re> /\d?\d(jan|feb|mar|apr|may|jun|jul|aug|sep|oct|nov|dec)\d\d/
9020: data> The date is 23ja\P\P
9021: Partial match: 23ja
9022:
1.6 misha 9023: At this stage, an application could discard the text preceding "23ja",
9024: add on text from the next segment, and call the matching function
9025: again. Unlike the DFA matching functions, the entire matching string
9026: must always be available, and the complete matching process occurs for
1.5 misha 9027: each call, so more memory and more processing time is needed.
9028:
1.6 misha 9029: Note: If the pattern contains lookbehind assertions, or \K, or starts
1.5 misha 9030: with \b or \B, the string that is returned for a partial match includes
1.6 misha 9031: characters that precede the start of what would be returned for a com-
9032: plete match, because it contains all the characters that were inspected
9033: during the partial match.
1.4 misha 9034:
9035:
9036: ISSUES WITH MULTI-SEGMENT MATCHING
9037:
9038: Certain types of pattern may give problems with multi-segment matching,
9039: whichever matching function is used.
9040:
9041: 1. If the pattern contains a test for the beginning of a line, you need
1.6 misha 9042: to pass the PCRE_NOTBOL option when the subject string for any call
9043: does start at the beginning of a line. There is also a PCRE_NOTEOL
1.4 misha 9044: option, but in practice when doing multi-segment matching you should be
9045: using PCRE_PARTIAL_HARD, which includes the effect of PCRE_NOTEOL.
9046:
1.6 misha 9047: 2. Lookbehind assertions that have already been obeyed are catered for
9048: in the offsets that are returned for a partial match. However a lookbe-
9049: hind assertion later in the pattern could require even earlier charac-
9050: ters to be inspected. You can handle this case by using the
9051: PCRE_INFO_MAXLOOKBEHIND option of the pcre_fullinfo() or
9052: pcre[16|32]_fullinfo() functions to obtain the length of the longest
9053: lookbehind in the pattern. This length is given in characters, not
9054: bytes. If you always retain at least that many characters before the
9055: partially matched string, all should be well. (Of course, near the
9056: start of the subject, fewer characters may be present; in that case all
9057: characters should be retained.)
9058:
9059: From release 8.33, there is a more accurate way of deciding which char-
9060: acters to retain. Instead of subtracting the length of the longest
9061: lookbehind from the earliest inspected character (offsets[0]), the
9062: match start position (offsets[2]) should be used, and the next match
9063: attempt started at the offsets[2] character by setting the startoffset
9064: argument of pcre_exec() or pcre_dfa_exec().
9065:
9066: For example, if the pattern "(?<=123)abc" is partially matched against
9067: the string "xx123a", the three offset values returned are 2, 6, and 5.
9068: This indicates that the matching process that gave a partial match
9069: started at offset 5, but the characters "123a" were all inspected. The
9070: maximum lookbehind for that pattern is 3, so taking that away from 5
9071: shows that we need only keep "123a", and the next match attempt can be
9072: started at offset 3 (that is, at "a") when further characters have been
9073: added. When the match start is not the earliest inspected character,
9074: pcretest shows it explicitly:
9075:
9076: re> "(?<=123)abc"
9077: data> xx123a\P\P
9078: Partial match at offset 5: 123a
9079:
9080: 3. Because a partial match must always contain at least one character,
9081: what might be considered a partial match of an empty string actually
9082: gives a "no match" result. For example:
9083:
9084: re> /c(?<=abc)x/
9085: data> ab\P
9086: No match
9087:
9088: If the next segment begins "cx", a match should be found, but this will
9089: only happen if characters from the previous segment are retained. For
9090: this reason, a "no match" result should be interpreted as "partial
9091: match of an empty string" when the pattern contains lookbehinds.
9092:
9093: 4. Matching a subject string that is split into multiple segments may
9094: not always produce exactly the same result as matching over one single
9095: long string, especially when PCRE_PARTIAL_SOFT is used. The section
9096: "Partial Matching and Word Boundaries" above describes an issue that
9097: arises if the pattern ends with \b or \B. Another kind of difference
9098: may occur when there are multiple matching possibilities, because (for
9099: PCRE_PARTIAL_SOFT) a partial match result is given only when there are
1.4 misha 9100: no completed matches. This means that as soon as the shortest match has
1.6 misha 9101: been found, continuation to a new subject segment is no longer possi-
1.4 misha 9102: ble. Consider again this pcretest example:
1.1 misha 9103:
9104: re> /dog(sbody)?/
1.4 misha 9105: data> dogsb\P
9106: 0: dog
1.1 misha 9107: data> do\P\D
9108: Partial match: do
9109: data> gsb\R\P\D
9110: 0: g
9111: data> dogsbody\D
9112: 0: dogsbody
9113: 1: dog
9114:
1.6 misha 9115: The first data line passes the string "dogsb" to a standard matching
9116: function, setting the PCRE_PARTIAL_SOFT option. Although the string is
9117: a partial match for "dogsbody", the result is not PCRE_ERROR_PARTIAL,
9118: because the shorter string "dog" is a complete match. Similarly, when
9119: the subject is presented to a DFA matching function in several parts
9120: ("do" and "gsb" being the first two) the match stops when "dog" has
9121: been found, and it is not possible to continue. On the other hand, if
9122: "dogsbody" is presented as a single string, a DFA matching function
1.5 misha 9123: finds both matches.
1.4 misha 9124:
1.6 misha 9125: Because of these problems, it is best to use PCRE_PARTIAL_HARD when
9126: matching multi-segment data. The example above then behaves differ-
1.4 misha 9127: ently:
1.1 misha 9128:
1.4 misha 9129: re> /dog(sbody)?/
9130: data> dogsb\P\P
9131: Partial match: dogsb
9132: data> do\P\D
9133: Partial match: do
9134: data> gsb\R\P\P\D
9135: Partial match: gsb
1.1 misha 9136:
1.6 misha 9137: 5. Patterns that contain alternatives at the top level which do not all
9138: start with the same pattern item may not work as expected when
1.5 misha 9139: PCRE_DFA_RESTART is used. For example, consider this pattern:
1.1 misha 9140:
9141: 1234|3789
9142:
1.6 misha 9143: If the first part of the subject is "ABC123", a partial match of the
9144: first alternative is found at offset 3. There is no partial match for
1.1 misha 9145: the second alternative, because such a match does not start at the same
1.6 misha 9146: point in the subject string. Attempting to continue with the string
9147: "7890" does not yield a match because only those alternatives that
9148: match at one point in the subject are remembered. The problem arises
9149: because the start of the second alternative matches within the first
9150: alternative. There is no problem with anchored patterns or patterns
1.4 misha 9151: such as:
1.1 misha 9152:
9153: 1234|ABCD
9154:
1.6 misha 9155: where no string can be a partial match for both alternatives. This is
9156: not a problem if a standard matching function is used, because the
1.5 misha 9157: entire match has to be rerun each time:
1.4 misha 9158:
9159: re> /1234|3789/
9160: data> ABC123\P\P
9161: Partial match: 123
9162: data> 1237890
9163: 0: 3789
9164:
9165: Of course, instead of using PCRE_DFA_RESTART, the same technique of re-
1.6 misha 9166: running the entire match can also be used with the DFA matching func-
9167: tions. Another possibility is to work with two buffers. If a partial
9168: match at offset n in the first buffer is followed by "no match" when
9169: PCRE_DFA_RESTART is used on the second buffer, you can then try a new
1.5 misha 9170: match starting at offset n+1 in the first buffer.
1.1 misha 9171:
9172:
9173: AUTHOR
9174:
9175: Philip Hazel
9176: University Computing Service
9177: Cambridge CB2 3QH, England.
9178:
9179:
9180: REVISION
9181:
1.7 ! moko 9182: Last updated: 02 July 2013
1.6 misha 9183: Copyright (c) 1997-2013 University of Cambridge.
1.1 misha 9184: ------------------------------------------------------------------------------
9185:
9186:
1.6 misha 9187: PCREPRECOMPILE(3) Library Functions Manual PCREPRECOMPILE(3)
9188:
1.1 misha 9189:
9190:
9191: NAME
9192: PCRE - Perl-compatible regular expressions
9193:
9194: SAVING AND RE-USING PRECOMPILED PCRE PATTERNS
9195:
9196: If you are running an application that uses a large number of regular
9197: expression patterns, it may be useful to store them in a precompiled
9198: form instead of having to compile them every time the application is
9199: run. If you are not using any private character tables (see the
9200: pcre_maketables() documentation), this is relatively straightforward.
1.5 misha 9201: If you are using private tables, it is a little bit more complicated.
9202: However, if you are using the just-in-time optimization feature, it is
9203: not possible to save and reload the JIT data.
1.1 misha 9204:
9205: If you save compiled patterns to a file, you can copy them to a differ-
1.5 misha 9206: ent host and run them there. If the two hosts have different endianness
1.6 misha 9207: (byte order), you should run the pcre[16|32]_pat-
9208: tern_to_host_byte_order() function on the new host before trying to
9209: match the pattern. The matching functions return PCRE_ERROR_BADENDIAN-
9210: NESS if they detect a pattern with the wrong endianness.
1.5 misha 9211:
9212: Compiling regular expressions with one version of PCRE for use with a
9213: different version is not guaranteed to work and may cause crashes, and
9214: saving and restoring a compiled pattern loses any JIT optimization
9215: data.
1.1 misha 9216:
9217:
9218: SAVING A COMPILED PATTERN
1.4 misha 9219:
1.6 misha 9220: The value returned by pcre[16|32]_compile() points to a single block of
1.5 misha 9221: memory that holds the compiled pattern and associated data. You can
1.6 misha 9222: find the length of this block in bytes by calling
9223: pcre[16|32]_fullinfo() with an argument of PCRE_INFO_SIZE. You can then
9224: save the data in any appropriate manner. Here is sample code for the
9225: 8-bit library that compiles a pattern and writes it to a file. It
9226: assumes that the variable fd refers to a file that is open for output:
1.1 misha 9227:
9228: int erroroffset, rc, size;
9229: char *error;
9230: pcre *re;
9231:
9232: re = pcre_compile("my pattern", 0, &error, &erroroffset, NULL);
9233: if (re == NULL) { ... handle errors ... }
9234: rc = pcre_fullinfo(re, NULL, PCRE_INFO_SIZE, &size);
9235: if (rc < 0) { ... handle errors ... }
9236: rc = fwrite(re, 1, size, fd);
9237: if (rc != size) { ... handle errors ... }
9238:
9239: In this example, the bytes that comprise the compiled pattern are
9240: copied exactly. Note that this is binary data that may contain any of
9241: the 256 possible byte values. On systems that make a distinction
9242: between binary and non-binary data, be sure that the file is opened for
9243: binary output.
9244:
9245: If you want to write more than one pattern to a file, you will have to
9246: devise a way of separating them. For binary data, preceding each pat-
9247: tern with its length is probably the most straightforward approach.
9248: Another possibility is to write out the data in hexadecimal instead of
9249: binary, one pattern to a line.
9250:
9251: Saving compiled patterns in a file is only one possible way of storing
9252: them for later use. They could equally well be saved in a database, or
9253: in the memory of some daemon process that passes them via sockets to
9254: the processes that want them.
9255:
1.5 misha 9256: If the pattern has been studied, it is also possible to save the normal
9257: study data in a similar way to the compiled pattern itself. However, if
9258: the PCRE_STUDY_JIT_COMPILE was used, the just-in-time data that is cre-
9259: ated cannot be saved because it is too dependent on the current envi-
9260: ronment. When studying generates additional information,
1.6 misha 9261: pcre[16|32]_study() returns a pointer to a pcre[16|32]_extra data
9262: block. Its format is defined in the section on matching a pattern in
9263: the pcreapi documentation. The study_data field points to the binary
9264: study data, and this is what you must save (not the pcre[16|32]_extra
9265: block itself). The length of the study data can be obtained by calling
9266: pcre[16|32]_fullinfo() with an argument of PCRE_INFO_STUDYSIZE. Remem-
9267: ber to check that pcre[16|32]_study() did return a non-NULL value
9268: before trying to save the study data.
1.1 misha 9269:
9270:
9271: RE-USING A PRECOMPILED PATTERN
9272:
1.5 misha 9273: Re-using a precompiled pattern is straightforward. Having reloaded it
1.6 misha 9274: into main memory, called pcre[16|32]_pattern_to_host_byte_order() if
9275: necessary, you pass its pointer to pcre[16|32]_exec() or
9276: pcre[16|32]_dfa_exec() in the usual way.
1.1 misha 9277:
9278: However, if you passed a pointer to custom character tables when the
1.6 misha 9279: pattern was compiled (the tableptr argument of pcre[16|32]_compile()),
9280: you must now pass a similar pointer to pcre[16|32]_exec() or
9281: pcre[16|32]_dfa_exec(), because the value saved with the compiled pat-
9282: tern will obviously be nonsense. A field in a pcre[16|32]_extra() block
9283: is used to pass this data, as described in the section on matching a
9284: pattern in the pcreapi documentation.
1.1 misha 9285:
1.7 ! moko 9286: Warning: The tables that pcre_exec() and pcre_dfa_exec() use must be
! 9287: the same as those that were used when the pattern was compiled. If this
! 9288: is not the case, the behaviour is undefined.
! 9289:
1.1 misha 9290: If you did not provide custom character tables when the pattern was
1.5 misha 9291: compiled, the pointer in the compiled pattern is NULL, which causes the
9292: matching functions to use PCRE's internal tables. Thus, you do not need
9293: to take any special action at run time in this case.
1.1 misha 9294:
9295: If you saved study data with the compiled pattern, you need to create
1.6 misha 9296: your own pcre[16|32]_extra data block and set the study_data field to
1.5 misha 9297: point to the reloaded study data. You must also set the
9298: PCRE_EXTRA_STUDY_DATA bit in the flags field to indicate that study
1.6 misha 9299: data is present. Then pass the pcre[16|32]_extra block to the matching
1.5 misha 9300: function in the usual way. If the pattern was studied for just-in-time
9301: optimization, that data cannot be saved, and so is lost by a
9302: save/restore cycle.
1.1 misha 9303:
9304:
9305: COMPATIBILITY WITH DIFFERENT PCRE RELEASES
9306:
9307: In general, it is safest to recompile all saved patterns when you
9308: update to a new PCRE release, though not all updates actually require
1.4 misha 9309: this.
1.1 misha 9310:
9311:
9312: AUTHOR
9313:
9314: Philip Hazel
9315: University Computing Service
9316: Cambridge CB2 3QH, England.
9317:
9318:
9319: REVISION
9320:
1.7 ! moko 9321: Last updated: 12 November 2013
! 9322: Copyright (c) 1997-2013 University of Cambridge.
1.1 misha 9323: ------------------------------------------------------------------------------
9324:
9325:
1.6 misha 9326: PCREPERFORM(3) Library Functions Manual PCREPERFORM(3)
9327:
1.1 misha 9328:
9329:
9330: NAME
9331: PCRE - Perl-compatible regular expressions
9332:
9333: PCRE PERFORMANCE
9334:
9335: Two aspects of performance are discussed below: memory usage and pro-
9336: cessing time. The way you express your pattern as a regular expression
9337: can affect both of them.
9338:
9339:
1.4 misha 9340: COMPILED PATTERN MEMORY USAGE
1.1 misha 9341:
1.5 misha 9342: Patterns are compiled by PCRE into a reasonably efficient interpretive
9343: code, so that most simple patterns do not use much memory. However,
9344: there is one case where the memory usage of a compiled pattern can be
9345: unexpectedly large. If a parenthesized subpattern has a quantifier with
9346: a minimum greater than 1 and/or a limited maximum, the whole subpattern
9347: is repeated in the compiled code. For example, the pattern
1.1 misha 9348:
9349: (abc|def){2,4}
9350:
9351: is compiled as if it were
9352:
9353: (abc|def)(abc|def)((abc|def)(abc|def)?)?
9354:
9355: (Technical aside: It is done this way so that backtrack points within
9356: each of the repetitions can be independently maintained.)
9357:
9358: For regular expressions whose quantifiers use only small numbers, this
9359: is not usually a problem. However, if the numbers are large, and par-
9360: ticularly if such repetitions are nested, the memory usage can become
9361: an embarrassment. For example, the very simple pattern
9362:
9363: ((ab){1,1000}c){1,3}
9364:
1.5 misha 9365: uses 51K bytes when compiled using the 8-bit library. When PCRE is com-
9366: piled with its default internal pointer size of two bytes, the size
9367: limit on a compiled pattern is 64K data units, and this is reached with
9368: the above pattern if the outer repetition is increased from 3 to 4.
9369: PCRE can be compiled to use larger internal pointers and thus handle
9370: larger compiled patterns, but it is better to try to rewrite your pat-
9371: tern to use less memory if you can.
1.1 misha 9372:
1.5 misha 9373: One way of reducing the memory usage for such patterns is to make use
1.1 misha 9374: of PCRE's "subroutine" facility. Re-writing the above pattern as
9375:
9376: ((ab)(?2){0,999}c)(?1){0,2}
9377:
9378: reduces the memory requirements to 18K, and indeed it remains under 20K
1.5 misha 9379: even with the outer repetition increased to 100. However, this pattern
9380: is not exactly equivalent, because the "subroutine" calls are treated
9381: as atomic groups into which there can be no backtracking if there is a
9382: subsequent matching failure. Therefore, PCRE cannot do this kind of
9383: rewriting automatically. Furthermore, there is a noticeable loss of
9384: speed when executing the modified pattern. Nevertheless, if the atomic
9385: grouping is not a problem and the loss of speed is acceptable, this
9386: kind of rewriting will allow you to process patterns that PCRE cannot
1.1 misha 9387: otherwise handle.
9388:
9389:
1.4 misha 9390: STACK USAGE AT RUN TIME
9391:
1.6 misha 9392: When pcre_exec() or pcre[16|32]_exec() is used for matching, certain
9393: kinds of pattern can cause it to use large amounts of the process
9394: stack. In some environments the default process stack is quite small,
9395: and if it runs out the result is often SIGSEGV. This issue is probably
9396: the most frequently raised problem with PCRE. Rewriting your pattern
9397: can often help. The pcrestack documentation discusses this issue in
9398: detail.
1.4 misha 9399:
9400:
1.1 misha 9401: PROCESSING TIME
9402:
1.6 misha 9403: Certain items in regular expression patterns are processed more effi-
1.1 misha 9404: ciently than others. It is more efficient to use a character class like
1.6 misha 9405: [aeiou] than a set of single-character alternatives such as
9406: (a|e|i|o|u). In general, the simplest construction that provides the
1.1 misha 9407: required behaviour is usually the most efficient. Jeffrey Friedl's book
1.6 misha 9408: contains a lot of useful general discussion about optimizing regular
9409: expressions for efficient performance. This document contains a few
1.1 misha 9410: observations about PCRE.
9411:
1.6 misha 9412: Using Unicode character properties (the \p, \P, and \X escapes) is
9413: slow, because PCRE has to use a multi-stage table lookup whenever it
9414: needs a character's property. If you can find an alternative pattern
9415: that does not use character properties, it will probably be faster.
1.1 misha 9416:
1.5 misha 9417: By default, the escape sequences \b, \d, \s, and \w, and the POSIX
9418: character classes such as [:alpha:] do not use Unicode properties,
1.4 misha 9419: partly for backwards compatibility, and partly for performance reasons.
1.5 misha 9420: However, you can set PCRE_UCP if you want Unicode character properties
9421: to be used. This can double the matching time for items such as \d,
9422: when matched with a traditional matching function; the performance loss
9423: is less with a DFA matching function, and in both cases there is not
9424: much difference for \b.
1.4 misha 9425:
9426: When a pattern begins with .* not in parentheses, or in parentheses
1.1 misha 9427: that are not the subject of a backreference, and the PCRE_DOTALL option
1.4 misha 9428: is set, the pattern is implicitly anchored by PCRE, since it can match
9429: only at the start of a subject string. However, if PCRE_DOTALL is not
9430: set, PCRE cannot make this optimization, because the . metacharacter
9431: does not then match a newline, and if the subject string contains new-
9432: lines, the pattern may match from the character immediately following
1.1 misha 9433: one of them instead of from the very start. For example, the pattern
9434:
9435: .*second
9436:
1.4 misha 9437: matches the subject "first\nand second" (where \n stands for a newline
9438: character), with the match starting at the seventh character. In order
1.1 misha 9439: to do this, PCRE has to retry the match starting after every newline in
9440: the subject.
9441:
1.4 misha 9442: If you are using such a pattern with subject strings that do not con-
1.1 misha 9443: tain newlines, the best performance is obtained by setting PCRE_DOTALL,
1.4 misha 9444: or starting the pattern with ^.* or ^.*? to indicate explicit anchor-
9445: ing. That saves PCRE from having to scan along the subject looking for
1.1 misha 9446: a newline to restart at.
9447:
1.4 misha 9448: Beware of patterns that contain nested indefinite repeats. These can
9449: take a long time to run when applied to a string that does not match.
1.1 misha 9450: Consider the pattern fragment
9451:
9452: ^(a+)*
9453:
1.4 misha 9454: This can match "aaaa" in 16 different ways, and this number increases
9455: very rapidly as the string gets longer. (The * repeat can match 0, 1,
9456: 2, 3, or 4 times, and for each of those cases other than 0 or 4, the +
9457: repeats can match different numbers of times.) When the remainder of
1.1 misha 9458: the pattern is such that the entire match is going to fail, PCRE has in
1.4 misha 9459: principle to try every possible variation, and this can take an
1.1 misha 9460: extremely long time, even for relatively short strings.
9461:
9462: An optimization catches some of the more simple cases such as
9463:
9464: (a+)*b
9465:
1.4 misha 9466: where a literal character follows. Before embarking on the standard
9467: matching procedure, PCRE checks that there is a "b" later in the sub-
9468: ject string, and if there is not, it fails the match immediately. How-
9469: ever, when there is no following literal this optimization cannot be
1.1 misha 9470: used. You can see the difference by comparing the behaviour of
9471:
9472: (a+)*\d
9473:
1.4 misha 9474: with the pattern above. The former gives a failure almost instantly
9475: when applied to a whole line of "a" characters, whereas the latter
1.1 misha 9476: takes an appreciable time with strings longer than about 20 characters.
9477:
9478: In many cases, the solution to this kind of performance issue is to use
9479: an atomic group or a possessive quantifier.
9480:
9481:
9482: AUTHOR
9483:
9484: Philip Hazel
9485: University Computing Service
9486: Cambridge CB2 3QH, England.
9487:
9488:
9489: REVISION
9490:
1.6 misha 9491: Last updated: 25 August 2012
1.5 misha 9492: Copyright (c) 1997-2012 University of Cambridge.
1.1 misha 9493: ------------------------------------------------------------------------------
9494:
9495:
1.6 misha 9496: PCREPOSIX(3) Library Functions Manual PCREPOSIX(3)
9497:
1.1 misha 9498:
9499:
9500: NAME
9501: PCRE - Perl-compatible regular expressions.
9502:
1.7 ! moko 9503: SYNOPSIS
1.1 misha 9504:
9505: #include <pcreposix.h>
9506:
9507: int regcomp(regex_t *preg, const char *pattern,
9508: int cflags);
9509:
9510: int regexec(regex_t *preg, const char *string,
9511: size_t nmatch, regmatch_t pmatch[], int eflags);
1.7 ! moko 9512: size_t regerror(int errcode, const regex_t *preg,
1.1 misha 9513: char *errbuf, size_t errbuf_size);
9514:
9515: void regfree(regex_t *preg);
9516:
9517:
9518: DESCRIPTION
9519:
1.5 misha 9520: This set of functions provides a POSIX-style API for the PCRE regular
9521: expression 8-bit library. See the pcreapi documentation for a descrip-
9522: tion of PCRE's native API, which contains much additional functional-
1.6 misha 9523: ity. There is no POSIX-style wrapper for PCRE's 16-bit and 32-bit
9524: library.
1.1 misha 9525:
9526: The functions described here are just wrapper functions that ultimately
9527: call the PCRE native API. Their prototypes are defined in the
1.6 misha 9528: pcreposix.h header file, and on Unix systems the library itself is
9529: called pcreposix.a, so can be accessed by adding -lpcreposix to the
9530: command for linking an application that uses them. Because the POSIX
1.1 misha 9531: functions call the native ones, it is also necessary to add -lpcre.
9532:
1.6 misha 9533: I have implemented only those POSIX option bits that can be reasonably
9534: mapped to PCRE native options. In addition, the option REG_EXTENDED is
9535: defined with the value zero. This has no effect, but since programs
9536: that are written to the POSIX interface often use it, this makes it
9537: easier to slot in PCRE as a replacement library. Other POSIX options
1.3 misha 9538: are not even defined.
1.1 misha 9539:
1.6 misha 9540: There are also some other options that are not defined by POSIX. These
1.4 misha 9541: have been added at the request of users who want to make use of certain
9542: PCRE-specific features via the POSIX calling interface.
9543:
1.6 misha 9544: When PCRE is called via these functions, it is only the API that is
9545: POSIX-like in style. The syntax and semantics of the regular expres-
9546: sions themselves are still those of Perl, subject to the setting of
9547: various PCRE options, as described below. "POSIX-like in style" means
9548: that the API approximates to the POSIX definition; it is not fully
9549: POSIX-compatible, and in multi-byte encoding domains it is probably
1.1 misha 9550: even less compatible.
9551:
1.6 misha 9552: The header for these functions is supplied as pcreposix.h to avoid any
9553: potential clash with other POSIX libraries. It can, of course, be
1.1 misha 9554: renamed or aliased as regex.h, which is the "correct" name. It provides
1.6 misha 9555: two structure types, regex_t for compiled internal forms, and reg-
9556: match_t for returning captured substrings. It also defines some con-
9557: stants whose names start with "REG_"; these are used for setting
1.1 misha 9558: options and identifying error codes.
9559:
9560:
9561: COMPILING A PATTERN
9562:
1.6 misha 9563: The function regcomp() is called to compile a pattern into an internal
9564: form. The pattern is a C string terminated by a binary zero, and is
9565: passed in the argument pattern. The preg argument is a pointer to a
9566: regex_t structure that is used as a base for storing information about
1.1 misha 9567: the compiled regular expression.
9568:
9569: The argument cflags is either zero, or contains one or more of the bits
9570: defined by the following macros:
9571:
9572: REG_DOTALL
9573:
9574: The PCRE_DOTALL option is set when the regular expression is passed for
9575: compilation to the native function. Note that REG_DOTALL is not part of
9576: the POSIX standard.
9577:
9578: REG_ICASE
9579:
1.6 misha 9580: The PCRE_CASELESS option is set when the regular expression is passed
1.1 misha 9581: for compilation to the native function.
9582:
9583: REG_NEWLINE
9584:
1.6 misha 9585: The PCRE_MULTILINE option is set when the regular expression is passed
9586: for compilation to the native function. Note that this does not mimic
9587: the defined POSIX behaviour for REG_NEWLINE (see the following sec-
1.1 misha 9588: tion).
9589:
9590: REG_NOSUB
9591:
1.6 misha 9592: The PCRE_NO_AUTO_CAPTURE option is set when the regular expression is
1.1 misha 9593: passed for compilation to the native function. In addition, when a pat-
1.6 misha 9594: tern that is compiled with this flag is passed to regexec() for match-
9595: ing, the nmatch and pmatch arguments are ignored, and no captured
1.1 misha 9596: strings are returned.
9597:
1.4 misha 9598: REG_UCP
9599:
1.6 misha 9600: The PCRE_UCP option is set when the regular expression is passed for
9601: compilation to the native function. This causes PCRE to use Unicode
9602: properties when matchine \d, \w, etc., instead of just recognizing
1.4 misha 9603: ASCII values. Note that REG_UTF8 is not part of the POSIX standard.
9604:
9605: REG_UNGREEDY
9606:
1.6 misha 9607: The PCRE_UNGREEDY option is set when the regular expression is passed
9608: for compilation to the native function. Note that REG_UNGREEDY is not
1.4 misha 9609: part of the POSIX standard.
9610:
1.1 misha 9611: REG_UTF8
9612:
1.6 misha 9613: The PCRE_UTF8 option is set when the regular expression is passed for
9614: compilation to the native function. This causes the pattern itself and
9615: all data strings used for matching it to be treated as UTF-8 strings.
1.1 misha 9616: Note that REG_UTF8 is not part of the POSIX standard.
9617:
1.6 misha 9618: In the absence of these flags, no options are passed to the native
9619: function. This means the the regex is compiled with PCRE default
9620: semantics. In particular, the way it handles newline characters in the
9621: subject string is the Perl way, not the POSIX way. Note that setting
9622: PCRE_MULTILINE has only some of the effects specified for REG_NEWLINE.
9623: It does not affect the way newlines are matched by . (they are not) or
1.1 misha 9624: by a negative class such as [^a] (they are).
9625:
1.6 misha 9626: The yield of regcomp() is zero on success, and non-zero otherwise. The
1.1 misha 9627: preg structure is filled in on success, and one member of the structure
1.6 misha 9628: is public: re_nsub contains the number of capturing subpatterns in the
1.1 misha 9629: regular expression. Various error codes are defined in the header file.
9630:
1.6 misha 9631: NOTE: If the yield of regcomp() is non-zero, you must not attempt to
1.4 misha 9632: use the contents of the preg structure. If, for example, you pass it to
9633: regexec(), the result is undefined and your program is likely to crash.
9634:
1.1 misha 9635:
9636: MATCHING NEWLINE CHARACTERS
9637:
9638: This area is not simple, because POSIX and Perl take different views of
1.6 misha 9639: things. It is not possible to get PCRE to obey POSIX semantics, but
9640: then PCRE was never intended to be a POSIX engine. The following table
9641: lists the different possibilities for matching newline characters in
1.1 misha 9642: PCRE:
9643:
9644: Default Change with
9645:
9646: . matches newline no PCRE_DOTALL
9647: newline matches [^a] yes not changeable
9648: $ matches \n at end yes PCRE_DOLLARENDONLY
9649: $ matches \n in middle no PCRE_MULTILINE
9650: ^ matches \n in middle no PCRE_MULTILINE
9651:
9652: This is the equivalent table for POSIX:
9653:
9654: Default Change with
9655:
9656: . matches newline yes REG_NEWLINE
9657: newline matches [^a] yes REG_NEWLINE
9658: $ matches \n at end no REG_NEWLINE
9659: $ matches \n in middle no REG_NEWLINE
9660: ^ matches \n in middle no REG_NEWLINE
9661:
9662: PCRE's behaviour is the same as Perl's, except that there is no equiva-
1.6 misha 9663: lent for PCRE_DOLLAR_ENDONLY in Perl. In both PCRE and Perl, there is
1.1 misha 9664: no way to stop newline from matching [^a].
9665:
1.6 misha 9666: The default POSIX newline handling can be obtained by setting
9667: PCRE_DOTALL and PCRE_DOLLAR_ENDONLY, but there is no way to make PCRE
1.1 misha 9668: behave exactly as for the REG_NEWLINE action.
9669:
9670:
9671: MATCHING A PATTERN
9672:
1.6 misha 9673: The function regexec() is called to match a compiled pattern preg
9674: against a given string, which is by default terminated by a zero byte
9675: (but see REG_STARTEND below), subject to the options in eflags. These
1.1 misha 9676: can be:
9677:
9678: REG_NOTBOL
9679:
9680: The PCRE_NOTBOL option is set when calling the underlying PCRE matching
9681: function.
9682:
1.3 misha 9683: REG_NOTEMPTY
9684:
9685: The PCRE_NOTEMPTY option is set when calling the underlying PCRE match-
9686: ing function. Note that REG_NOTEMPTY is not part of the POSIX standard.
9687: However, setting this option can give more POSIX-like behaviour in some
9688: situations.
9689:
1.1 misha 9690: REG_NOTEOL
9691:
9692: The PCRE_NOTEOL option is set when calling the underlying PCRE matching
9693: function.
9694:
9695: REG_STARTEND
9696:
1.6 misha 9697: The string is considered to start at string + pmatch[0].rm_so and to
9698: have a terminating NUL located at string + pmatch[0].rm_eo (there need
9699: not actually be a NUL at that location), regardless of the value of
9700: nmatch. This is a BSD extension, compatible with but not specified by
9701: IEEE Standard 1003.2 (POSIX.2), and should be used with caution in
1.1 misha 9702: software intended to be portable to other systems. Note that a non-zero
9703: rm_so does not imply REG_NOTBOL; REG_STARTEND affects only the location
9704: of the string, not how it is matched.
9705:
1.6 misha 9706: If the pattern was compiled with the REG_NOSUB flag, no data about any
9707: matched strings is returned. The nmatch and pmatch arguments of
1.1 misha 9708: regexec() are ignored.
9709:
1.4 misha 9710: If the value of nmatch is zero, or if the value pmatch is NULL, no data
9711: about any matched strings is returned.
9712:
1.1 misha 9713: Otherwise,the portion of the string that was matched, and also any cap-
9714: tured substrings, are returned via the pmatch argument, which points to
1.6 misha 9715: an array of nmatch structures of type regmatch_t, containing the mem-
9716: bers rm_so and rm_eo. These contain the offset to the first character
9717: of each substring and the offset to the first character after the end
9718: of each substring, respectively. The 0th element of the vector relates
9719: to the entire portion of string that was matched; subsequent elements
9720: relate to the capturing subpatterns of the regular expression. Unused
1.1 misha 9721: entries in the array have both structure members set to -1.
9722:
1.6 misha 9723: A successful match yields a zero return; various error codes are
9724: defined in the header file, of which REG_NOMATCH is the "expected"
1.1 misha 9725: failure code.
9726:
9727:
9728: ERROR MESSAGES
9729:
9730: The regerror() function maps a non-zero errorcode from either regcomp()
1.6 misha 9731: or regexec() to a printable message. If preg is not NULL, the error
1.1 misha 9732: should have arisen from the use of that structure. A message terminated
1.6 misha 9733: by a binary zero is placed in errbuf. The length of the message,
9734: including the zero, is limited to errbuf_size. The yield of the func-
1.1 misha 9735: tion is the size of buffer needed to hold the whole message.
9736:
9737:
9738: MEMORY USAGE
9739:
1.6 misha 9740: Compiling a regular expression causes memory to be allocated and asso-
9741: ciated with the preg structure. The function regfree() frees all such
9742: memory, after which preg may no longer be used as a compiled expres-
1.1 misha 9743: sion.
9744:
9745:
9746: AUTHOR
9747:
9748: Philip Hazel
9749: University Computing Service
9750: Cambridge CB2 3QH, England.
9751:
9752:
9753: REVISION
9754:
1.5 misha 9755: Last updated: 09 January 2012
9756: Copyright (c) 1997-2012 University of Cambridge.
1.1 misha 9757: ------------------------------------------------------------------------------
9758:
9759:
1.6 misha 9760: PCRECPP(3) Library Functions Manual PCRECPP(3)
9761:
1.1 misha 9762:
9763:
9764: NAME
9765: PCRE - Perl-compatible regular expressions.
9766:
9767: SYNOPSIS OF C++ WRAPPER
9768:
9769: #include <pcrecpp.h>
9770:
9771:
9772: DESCRIPTION
9773:
9774: The C++ wrapper for PCRE was provided by Google Inc. Some additional
9775: functionality was added by Giuseppe Maxia. This brief man page was con-
9776: structed from the notes in the pcrecpp.h file, which should be con-
1.5 misha 9777: sulted for further details. Note that the C++ wrapper supports only the
1.6 misha 9778: original 8-bit PCRE library. There is no 16-bit or 32-bit support at
9779: present.
1.1 misha 9780:
9781:
9782: MATCHING INTERFACE
9783:
1.6 misha 9784: The "FullMatch" operation checks that supplied text matches a supplied
9785: pattern exactly. If pointer arguments are supplied, it copies matched
1.1 misha 9786: sub-strings that match sub-patterns into them.
9787:
9788: Example: successful match
9789: pcrecpp::RE re("h.*o");
9790: re.FullMatch("hello");
9791:
9792: Example: unsuccessful match (requires full match):
9793: pcrecpp::RE re("e");
9794: !re.FullMatch("hello");
9795:
9796: Example: creating a temporary RE object:
9797: pcrecpp::RE("h.*o").FullMatch("hello");
9798:
1.6 misha 9799: You can pass in a "const char*" or a "string" for "text". The examples
9800: below tend to use a const char*. You can, as in the different examples
9801: above, store the RE object explicitly in a variable or use a temporary
9802: RE object. The examples below use one mode or the other arbitrarily.
1.1 misha 9803: Either could correctly be used for any of these examples.
9804:
9805: You must supply extra pointer arguments to extract matched subpieces.
9806:
9807: Example: extracts "ruby" into "s" and 1234 into "i"
9808: int i;
9809: string s;
9810: pcrecpp::RE re("(\\w+):(\\d+)");
9811: re.FullMatch("ruby:1234", &s, &i);
9812:
9813: Example: does not try to extract any extra sub-patterns
9814: re.FullMatch("ruby:1234", &s);
9815:
9816: Example: does not try to extract into NULL
9817: re.FullMatch("ruby:1234", NULL, &i);
9818:
9819: Example: integer overflow causes failure
9820: !re.FullMatch("ruby:1234567891234", NULL, &i);
9821:
9822: Example: fails because there aren't enough sub-patterns:
9823: !pcrecpp::RE("\\w+:\\d+").FullMatch("ruby:1234", &s);
9824:
9825: Example: fails because string cannot be stored in integer
9826: !pcrecpp::RE("(.*)").FullMatch("ruby", &i);
9827:
1.6 misha 9828: The provided pointer arguments can be pointers to any scalar numeric
1.1 misha 9829: type, or one of:
9830:
9831: string (matched piece is copied to string)
9832: StringPiece (StringPiece is mutated to point to matched piece)
9833: T (where "bool T::ParseFrom(const char*, int)" exists)
9834: NULL (the corresponding matched sub-pattern is not copied)
9835:
1.6 misha 9836: The function returns true iff all of the following conditions are sat-
1.1 misha 9837: isfied:
9838:
9839: a. "text" matches "pattern" exactly;
9840:
9841: b. The number of matched sub-patterns is >= number of supplied
9842: pointers;
9843:
9844: c. The "i"th argument has a suitable type for holding the
9845: string captured as the "i"th sub-pattern. If you pass in
9846: void * NULL for the "i"th argument, or a non-void * NULL
9847: of the correct type, or pass fewer arguments than the
9848: number of sub-patterns, "i"th captured sub-pattern is
9849: ignored.
9850:
1.6 misha 9851: CAVEAT: An optional sub-pattern that does not exist in the matched
9852: string is assigned the empty string. Therefore, the following will
1.1 misha 9853: return false (because the empty string is not a valid number):
9854:
9855: int number;
9856: pcrecpp::RE::FullMatch("abc", "[a-z]+(\\d+)?", &number);
9857:
1.6 misha 9858: The matching interface supports at most 16 arguments per call. If you
9859: need more, consider using the more general interface
1.1 misha 9860: pcrecpp::RE::DoMatch. See pcrecpp.h for the signature for DoMatch.
9861:
1.6 misha 9862: NOTE: Do not use no_arg, which is used internally to mark the end of a
9863: list of optional arguments, as a placeholder for missing arguments, as
1.3 misha 9864: this can lead to segfaults.
9865:
1.1 misha 9866:
9867: QUOTING METACHARACTERS
9868:
1.6 misha 9869: You can use the "QuoteMeta" operation to insert backslashes before all
9870: potentially meaningful characters in a string. The returned string,
1.1 misha 9871: used as a regular expression, will exactly match the original string.
9872:
9873: Example:
9874: string quoted = RE::QuoteMeta(unquoted);
9875:
1.6 misha 9876: Note that it's legal to escape a character even if it has no special
9877: meaning in a regular expression -- so this function does that. (This
9878: also makes it identical to the perl function of the same name; see
9879: "perldoc -f quotemeta".) For example, "1.5-2.0?" becomes
1.1 misha 9880: "1\.5\-2\.0\?".
9881:
9882:
9883: PARTIAL MATCHES
9884:
1.6 misha 9885: You can use the "PartialMatch" operation when you want the pattern to
1.1 misha 9886: match any substring of the text.
9887:
9888: Example: simple search for a string:
9889: pcrecpp::RE("ell").PartialMatch("hello");
9890:
9891: Example: find first number in a string:
9892: int number;
9893: pcrecpp::RE re("(\\d+)");
9894: re.PartialMatch("x*100 + 20", &number);
9895: assert(number == 100);
9896:
9897:
9898: UTF-8 AND THE MATCHING INTERFACE
9899:
1.6 misha 9900: By default, pattern and text are plain text, one byte per character.
9901: The UTF8 flag, passed to the constructor, causes both pattern and
1.1 misha 9902: string to be treated as UTF-8 text, still a byte stream but potentially
1.6 misha 9903: multiple bytes per character. In practice, the text is likelier to be
9904: UTF-8 than the pattern, but the match returned may depend on the UTF8
9905: flag, so always use it when matching UTF8 text. For example, "." will
9906: match one byte normally but with UTF8 set may match up to three bytes
1.1 misha 9907: of a multi-byte character.
9908:
9909: Example:
9910: pcrecpp::RE_Options options;
9911: options.set_utf8();
9912: pcrecpp::RE re(utf8_pattern, options);
9913: re.FullMatch(utf8_string);
9914:
9915: Example: using the convenience function UTF8():
9916: pcrecpp::RE re(utf8_pattern, pcrecpp::UTF8());
9917: re.FullMatch(utf8_string);
9918:
9919: NOTE: The UTF8 flag is ignored if pcre was not configured with the
9920: --enable-utf8 flag.
9921:
9922:
9923: PASSING MODIFIERS TO THE REGULAR EXPRESSION ENGINE
9924:
1.6 misha 9925: PCRE defines some modifiers to change the behavior of the regular
9926: expression engine. The C++ wrapper defines an auxiliary class,
9927: RE_Options, as a vehicle to pass such modifiers to a RE class. Cur-
1.1 misha 9928: rently, the following modifiers are supported:
9929:
9930: modifier description Perl corresponding
9931:
9932: PCRE_CASELESS case insensitive match /i
9933: PCRE_MULTILINE multiple lines match /m
9934: PCRE_DOTALL dot matches newlines /s
9935: PCRE_DOLLAR_ENDONLY $ matches only at end N/A
9936: PCRE_EXTRA strict escape parsing N/A
1.6 misha 9937: PCRE_EXTENDED ignore white spaces /x
1.1 misha 9938: PCRE_UTF8 handles UTF8 chars built-in
9939: PCRE_UNGREEDY reverses * and *? N/A
9940: PCRE_NO_AUTO_CAPTURE disables capturing parens N/A (*)
9941:
1.6 misha 9942: (*) Both Perl and PCRE allow non capturing parentheses by means of the
9943: "?:" modifier within the pattern itself. e.g. (?:ab|cd) does not cap-
1.1 misha 9944: ture, while (ab|cd) does.
9945:
1.6 misha 9946: For a full account on how each modifier works, please check the PCRE
1.1 misha 9947: API reference page.
9948:
1.6 misha 9949: For each modifier, there are two member functions whose name is made
9950: out of the modifier in lowercase, without the "PCRE_" prefix. For
1.1 misha 9951: instance, PCRE_CASELESS is handled by
9952:
9953: bool caseless()
9954:
9955: which returns true if the modifier is set, and
9956:
9957: RE_Options & set_caseless(bool)
9958:
9959: which sets or unsets the modifier. Moreover, PCRE_EXTRA_MATCH_LIMIT can
1.6 misha 9960: be accessed through the set_match_limit() and match_limit() member
9961: functions. Setting match_limit to a non-zero value will limit the exe-
9962: cution of pcre to keep it from doing bad things like blowing the stack
9963: or taking an eternity to return a result. A value of 5000 is good
9964: enough to stop stack blowup in a 2MB thread stack. Setting match_limit
9965: to zero disables match limiting. Alternatively, you can call
9966: match_limit_recursion() which uses PCRE_EXTRA_MATCH_LIMIT_RECURSION to
9967: limit how much PCRE recurses. match_limit() limits the number of
1.1 misha 9968: matches PCRE does; match_limit_recursion() limits the depth of internal
9969: recursion, and therefore the amount of stack that is used.
9970:
1.6 misha 9971: Normally, to pass one or more modifiers to a RE class, you declare a
1.1 misha 9972: RE_Options object, set the appropriate options, and pass this object to
9973: a RE constructor. Example:
9974:
1.5 misha 9975: RE_Options opt;
1.1 misha 9976: opt.set_caseless(true);
9977: if (RE("HELLO", opt).PartialMatch("hello world")) ...
9978:
9979: RE_options has two constructors. The default constructor takes no argu-
1.6 misha 9980: ments and creates a set of flags that are off by default. The optional
9981: parameter option_flags is to facilitate transfer of legacy code from C
1.1 misha 9982: programs. This lets you do
9983:
9984: RE(pattern,
9985: RE_Options(PCRE_CASELESS|PCRE_MULTILINE)).PartialMatch(str);
9986:
9987: However, new code is better off doing
9988:
9989: RE(pattern,
9990: RE_Options().set_caseless(true).set_multiline(true))
9991: .PartialMatch(str);
9992:
9993: If you are going to pass one of the most used modifiers, there are some
9994: convenience functions that return a RE_Options class with the appropri-
1.6 misha 9995: ate modifier already set: CASELESS(), UTF8(), MULTILINE(), DOTALL(),
1.1 misha 9996: and EXTENDED().
9997:
1.6 misha 9998: If you need to set several options at once, and you don't want to go
9999: through the pains of declaring a RE_Options object and setting several
10000: options, there is a parallel method that give you such ability on the
10001: fly. You can concatenate several set_xxxxx() member functions, since
10002: each of them returns a reference to its class object. For example, to
10003: pass PCRE_CASELESS, PCRE_EXTENDED, and PCRE_MULTILINE to a RE with one
1.1 misha 10004: statement, you may write:
10005:
10006: RE(" ^ xyz \\s+ .* blah$",
10007: RE_Options()
10008: .set_caseless(true)
10009: .set_extended(true)
10010: .set_multiline(true)).PartialMatch(sometext);
10011:
10012:
10013: SCANNING TEXT INCREMENTALLY
10014:
1.6 misha 10015: The "Consume" operation may be useful if you want to repeatedly match
1.1 misha 10016: regular expressions at the front of a string and skip over them as they
1.6 misha 10017: match. This requires use of the "StringPiece" type, which represents a
10018: sub-range of a real string. Like RE, StringPiece is defined in the
1.1 misha 10019: pcrecpp namespace.
10020:
10021: Example: read lines of the form "var = value" from a string.
10022: string contents = ...; // Fill string somehow
10023: pcrecpp::StringPiece input(contents); // Wrap in a StringPiece
10024:
10025: string var;
10026: int value;
10027: pcrecpp::RE re("(\\w+) = (\\d+)\n");
10028: while (re.Consume(&input, &var, &value)) {
10029: ...;
10030: }
10031:
1.6 misha 10032: Each successful call to "Consume" will set "var/value", and also
1.1 misha 10033: advance "input" so it points past the matched text.
10034:
1.6 misha 10035: The "FindAndConsume" operation is similar to "Consume" but does not
10036: anchor your match at the beginning of the string. For example, you
1.1 misha 10037: could extract all words from a string by repeatedly calling
10038:
10039: pcrecpp::RE("(\\w+)").FindAndConsume(&input, &word)
10040:
10041:
10042: PARSING HEX/OCTAL/C-RADIX NUMBERS
10043:
10044: By default, if you pass a pointer to a numeric value, the corresponding
1.6 misha 10045: text is interpreted as a base-10 number. You can instead wrap the
1.1 misha 10046: pointer with a call to one of the operators Hex(), Octal(), or CRadix()
1.6 misha 10047: to interpret the text in another base. The CRadix operator interprets
10048: C-style "0" (base-8) and "0x" (base-16) prefixes, but defaults to
1.1 misha 10049: base-10.
10050:
10051: Example:
10052: int a, b, c, d;
10053: pcrecpp::RE re("(.*) (.*) (.*) (.*)");
10054: re.FullMatch("100 40 0100 0x40",
10055: pcrecpp::Octal(&a), pcrecpp::Hex(&b),
10056: pcrecpp::CRadix(&c), pcrecpp::CRadix(&d));
10057:
10058: will leave 64 in a, b, c, and d.
10059:
10060:
10061: REPLACING PARTS OF STRINGS
10062:
1.6 misha 10063: You can replace the first match of "pattern" in "str" with "rewrite".
10064: Within "rewrite", backslash-escaped digits (\1 to \9) can be used to
10065: insert text matching corresponding parenthesized group from the pat-
1.1 misha 10066: tern. \0 in "rewrite" refers to the entire matching text. For example:
10067:
10068: string s = "yabba dabba doo";
10069: pcrecpp::RE("b+").Replace("d", &s);
10070:
1.6 misha 10071: will leave "s" containing "yada dabba doo". The result is true if the
1.1 misha 10072: pattern matches and a replacement occurs, false otherwise.
10073:
1.6 misha 10074: GlobalReplace is like Replace except that it replaces all occurrences
10075: of the pattern in the string with the rewrite. Replacements are not
1.1 misha 10076: subject to re-matching. For example:
10077:
10078: string s = "yabba dabba doo";
10079: pcrecpp::RE("b+").GlobalReplace("d", &s);
10080:
1.6 misha 10081: will leave "s" containing "yada dada doo". It returns the number of
1.1 misha 10082: replacements made.
10083:
1.6 misha 10084: Extract is like Replace, except that if the pattern matches, "rewrite"
10085: is copied into "out" (an additional argument) with substitutions. The
10086: non-matching portions of "text" are ignored. Returns true iff a match
1.1 misha 10087: occurred and the extraction happened successfully; if no match occurs,
10088: the string is left unaffected.
10089:
10090:
10091: AUTHOR
10092:
10093: The C++ wrapper was contributed by Google Inc.
10094: Copyright (c) 2007 Google Inc.
10095:
10096:
10097: REVISION
10098:
1.5 misha 10099: Last updated: 08 January 2012
1.1 misha 10100: ------------------------------------------------------------------------------
10101:
10102:
1.6 misha 10103: PCRESAMPLE(3) Library Functions Manual PCRESAMPLE(3)
10104:
1.1 misha 10105:
10106:
10107: NAME
10108: PCRE - Perl-compatible regular expressions
10109:
10110: PCRE SAMPLE PROGRAM
10111:
10112: A simple, complete demonstration program, to get you started with using
1.4 misha 10113: PCRE, is supplied in the file pcredemo.c in the PCRE distribution. A
10114: listing of this program is given in the pcredemo documentation. If you
10115: do not have a copy of the PCRE distribution, you can save this listing
10116: to re-create pcredemo.c.
1.1 misha 10117:
1.5 misha 10118: The demonstration program, which uses the original PCRE 8-bit library,
10119: compiles the regular expression that is its first argument, and matches
10120: it against the subject string in its second argument. No PCRE options
10121: are set, and default character tables are used. If matching succeeds,
10122: the program outputs the portion of the subject that matched, together
10123: with the contents of any captured substrings.
1.1 misha 10124:
10125: If the -g option is given on the command line, the program then goes on
10126: to check for further matches of the same regular expression in the same
1.5 misha 10127: subject string. The logic is a little bit tricky because of the possi-
10128: bility of matching an empty string. Comments in the code explain what
1.1 misha 10129: is going on.
10130:
1.5 misha 10131: If PCRE is installed in the standard include and library directories
1.4 misha 10132: for your operating system, you should be able to compile the demonstra-
10133: tion program using this command:
1.1 misha 10134:
10135: gcc -o pcredemo pcredemo.c -lpcre
10136:
1.5 misha 10137: If PCRE is installed elsewhere, you may need to add additional options
10138: to the command line. For example, on a Unix-like system that has PCRE
10139: installed in /usr/local, you can compile the demonstration program
1.1 misha 10140: using a command like this:
10141:
10142: gcc -o pcredemo -I/usr/local/include pcredemo.c \
10143: -L/usr/local/lib -lpcre
10144:
1.5 misha 10145: In a Windows environment, if you want to statically link the program
1.4 misha 10146: against a non-dll pcre.a file, you must uncomment the line that defines
1.5 misha 10147: PCRE_STATIC before including pcre.h, because otherwise the pcre_mal-
1.4 misha 10148: loc() and pcre_free() exported functions will be declared
10149: __declspec(dllimport), with unwanted results.
10150:
1.5 misha 10151: Once you have compiled and linked the demonstration program, you can
1.4 misha 10152: run simple tests like this:
1.1 misha 10153:
10154: ./pcredemo 'cat|dog' 'the cat sat on the mat'
10155: ./pcredemo -g 'cat|dog' 'the dog sat on the cat'
10156:
1.5 misha 10157: Note that there is a much more comprehensive test program, called
10158: pcretest, which supports many more facilities for testing regular
10159: expressions and both PCRE libraries. The pcredemo program is provided
10160: as a simple coding example.
1.1 misha 10161:
1.5 misha 10162: If you try to run pcredemo when PCRE is not installed in the standard
10163: library directory, you may get an error like this on some operating
1.4 misha 10164: systems (e.g. Solaris):
1.1 misha 10165:
1.5 misha 10166: ld.so.1: a.out: fatal: libpcre.so.0: open failed: No such file or
1.1 misha 10167: directory
10168:
1.5 misha 10169: This is caused by the way shared library support works on those sys-
1.1 misha 10170: tems. You need to add
10171:
10172: -R/usr/local/lib
10173:
10174: (for example) to the compile command to get round this problem.
10175:
10176:
10177: AUTHOR
10178:
10179: Philip Hazel
10180: University Computing Service
10181: Cambridge CB2 3QH, England.
10182:
10183:
10184: REVISION
10185:
1.5 misha 10186: Last updated: 10 January 2012
10187: Copyright (c) 1997-2012 University of Cambridge.
1.1 misha 10188: ------------------------------------------------------------------------------
1.6 misha 10189: PCRELIMITS(3) Library Functions Manual PCRELIMITS(3)
10190:
1.5 misha 10191:
10192:
10193: NAME
10194: PCRE - Perl-compatible regular expressions
10195:
10196: SIZE AND OTHER LIMITATIONS
10197:
10198: There are some size limitations in PCRE but it is hoped that they will
10199: never in practice be relevant.
10200:
10201: The maximum length of a compiled pattern is approximately 64K data
1.7 ! moko 10202: units (bytes for the 8-bit library, 16-bit units for the 16-bit
1.6 misha 10203: library, and 32-bit units for the 32-bit library) if PCRE is compiled
1.7 ! moko 10204: with the default internal linkage size, which is 2 bytes for the 8-bit
! 10205: and 16-bit libraries, and 4 bytes for the 32-bit library. If you want
! 10206: to process regular expressions that are truly enormous, you can compile
! 10207: PCRE with an internal linkage size of 3 or 4 (when building the 16-bit
! 10208: or 32-bit library, 3 is rounded up to 4). See the README file in the
! 10209: source distribution and the pcrebuild documentation for details. In
! 10210: these cases the limit is substantially larger. However, the speed of
1.6 misha 10211: execution is slower.
1.5 misha 10212:
10213: All values in repeating quantifiers must be less than 65536.
10214:
10215: There is no limit to the number of parenthesized subpatterns, but there
1.7 ! moko 10216: can be no more than 65535 capturing subpatterns. There is, however, a
! 10217: limit to the depth of nesting of parenthesized subpatterns of all
! 10218: kinds. This is imposed in order to limit the amount of system stack
! 10219: used at compile time. The limit can be specified when PCRE is built;
! 10220: the default is 250.
1.5 misha 10221:
10222: There is a limit to the number of forward references to subsequent sub-
1.7 ! moko 10223: patterns of around 200,000. Repeated forward references with fixed
! 10224: upper limits, for example, (?2){0,100} when subpattern number 2 is to
! 10225: the right, are included in the count. There is no limit to the number
1.5 misha 10226: of backward references.
10227:
10228: The maximum length of name for a named subpattern is 32 characters, and
10229: the maximum number of named subpatterns is 10000.
10230:
1.7 ! moko 10231: The maximum length of a name in a (*MARK), (*PRUNE), (*SKIP), or
! 10232: (*THEN) verb is 255 for the 8-bit library and 65535 for the 16-bit and
! 10233: 32-bit libraries.
1.6 misha 10234:
1.7 ! moko 10235: The maximum length of a subject string is the largest positive number
! 10236: that an integer variable can hold. However, when using the traditional
1.5 misha 10237: matching function, PCRE uses recursion to handle subpatterns and indef-
1.7 ! moko 10238: inite repetition. This means that the available stack space may limit
1.5 misha 10239: the size of a subject string that can be processed by certain patterns.
10240: For a discussion of stack issues, see the pcrestack documentation.
10241:
10242:
10243: AUTHOR
10244:
10245: Philip Hazel
10246: University Computing Service
10247: Cambridge CB2 3QH, England.
10248:
10249:
10250: REVISION
10251:
1.7 ! moko 10252: Last updated: 05 November 2013
! 10253: Copyright (c) 1997-2013 University of Cambridge.
1.5 misha 10254: ------------------------------------------------------------------------------
10255:
10256:
1.6 misha 10257: PCRESTACK(3) Library Functions Manual PCRESTACK(3)
10258:
1.1 misha 10259:
10260:
10261: NAME
10262: PCRE - Perl-compatible regular expressions
10263:
10264: PCRE DISCUSSION OF STACK USAGE
10265:
1.6 misha 10266: When you call pcre[16|32]_exec(), it makes use of an internal function
1.5 misha 10267: called match(). This calls itself recursively at branch points in the
10268: pattern, in order to remember the state of the match so that it can
10269: back up and try a different alternative if the first one fails. As
10270: matching proceeds deeper and deeper into the tree of possibilities, the
10271: recursion depth increases. The match() function is also called in other
10272: circumstances, for example, whenever a parenthesized sub-pattern is
10273: entered, and in certain cases of repetition.
1.1 misha 10274:
10275: Not all calls of match() increase the recursion depth; for an item such
10276: as a* it may be called several times at the same level, after matching
10277: different numbers of a's. Furthermore, in a number of cases where the
10278: result of the recursive call would immediately be passed back as the
10279: result of the current call (a "tail recursion"), the function is just
10280: restarted instead.
10281:
1.6 misha 10282: The above comments apply when pcre[16|32]_exec() is run in its normal
1.5 misha 10283: interpretive manner. If the pattern was studied with the
10284: PCRE_STUDY_JIT_COMPILE option, and just-in-time compiling was success-
1.6 misha 10285: ful, and the options passed to pcre[16|32]_exec() were not incompati-
10286: ble, the matching process uses the JIT-compiled code instead of the
10287: match() function. In this case, the memory requirements are handled
10288: entirely differently. See the pcrejit documentation for details.
10289:
10290: The pcre[16|32]_dfa_exec() function operates in an entirely different
10291: way, and uses recursion only when there is a regular expression recur-
10292: sion or subroutine call in the pattern. This includes the processing of
10293: assertion and "once-only" subpatterns, which are handled like subrou-
10294: tine calls. Normally, these are never very deep, and the limit on the
10295: complexity of pcre[16|32]_dfa_exec() is controlled by the amount of
10296: workspace it is given. However, it is possible to write patterns with
10297: runaway infinite recursions; such patterns will cause
10298: pcre[16|32]_dfa_exec() to run out of stack. At present, there is no
10299: protection against this.
10300:
10301: The comments that follow do NOT apply to pcre[16|32]_dfa_exec(); they
10302: are relevant only for pcre[16|32]_exec() without the JIT optimization.
10303:
10304: Reducing pcre[16|32]_exec()'s stack usage
10305:
10306: Each time that match() is actually called recursively, it uses memory
10307: from the process stack. For certain kinds of pattern and data, very
10308: large amounts of stack may be needed, despite the recognition of "tail
10309: recursion". You can often reduce the amount of recursion, and there-
10310: fore the amount of stack used, by modifying the pattern that is being
1.1 misha 10311: matched. Consider, for example, this pattern:
10312:
10313: ([^<]|<(?!inet))+
10314:
1.6 misha 10315: It matches from wherever it starts until it encounters "<inet" or the
10316: end of the data, and is the kind of pattern that might be used when
1.1 misha 10317: processing an XML file. Each iteration of the outer parentheses matches
1.6 misha 10318: either one character that is not "<" or a "<" that is not followed by
10319: "inet". However, each time a parenthesis is processed, a recursion
1.1 misha 10320: occurs, so this formulation uses a stack frame for each matched charac-
1.6 misha 10321: ter. For a long string, a lot of stack is required. Consider now this
1.1 misha 10322: rewritten pattern, which matches exactly the same strings:
10323:
10324: ([^<]++|<(?!inet))+
10325:
1.6 misha 10326: This uses very much less stack, because runs of characters that do not
10327: contain "<" are "swallowed" in one item inside the parentheses. Recur-
10328: sion happens only when a "<" character that is not followed by "inet"
10329: is encountered (and we assume this is relatively rare). A possessive
10330: quantifier is used to stop any backtracking into the runs of non-"<"
1.1 misha 10331: characters, but that is not related to stack usage.
10332:
1.6 misha 10333: This example shows that one way of avoiding stack problems when match-
1.1 misha 10334: ing long subject strings is to write repeated parenthesized subpatterns
10335: to match more than one character whenever possible.
10336:
1.6 misha 10337: Compiling PCRE to use heap instead of stack for pcre[16|32]_exec()
1.2 misha 10338:
1.6 misha 10339: In environments where stack memory is constrained, you might want to
10340: compile PCRE to use heap memory instead of stack for remembering back-
10341: up points when pcre[16|32]_exec() is running. This makes it run a lot
10342: more slowly, however. Details of how to do this are given in the pcre-
10343: build documentation. When built in this way, instead of using the
10344: stack, PCRE obtains and frees memory by calling the functions that are
10345: pointed to by the pcre[16|32]_stack_malloc and pcre[16|32]_stack_free
10346: variables. By default, these point to malloc() and free(), but you can
10347: replace the pointers to cause PCRE to use your own functions. Since the
10348: block sizes are always the same, and are always freed in reverse order,
10349: it may be possible to implement customized memory handlers that are
10350: more efficient than the standard functions.
10351:
10352: Limiting pcre[16|32]_exec()'s stack usage
10353:
10354: You can set limits on the number of times that match() is called, both
10355: in total and recursively. If a limit is exceeded, pcre[16|32]_exec()
10356: returns an error code. Setting suitable limits should prevent it from
10357: running out of stack. The default values of the limits are very large,
10358: and unlikely ever to operate. They can be changed when PCRE is built,
10359: and they can also be set when pcre[16|32]_exec() is called. For details
10360: of these interfaces, see the pcrebuild documentation and the section on
10361: extra data for pcre[16|32]_exec() in the pcreapi documentation.
1.2 misha 10362:
10363: As a very rough rule of thumb, you should reckon on about 500 bytes per
1.6 misha 10364: recursion. Thus, if you want to limit your stack usage to 8Mb, you
10365: should set the limit at 16000 recursions. A 64Mb stack, on the other
1.4 misha 10366: hand, can support around 128000 recursions.
10367:
10368: In Unix-like environments, the pcretest test program has a command line
10369: option (-S) that can be used to increase the size of its stack. As long
1.6 misha 10370: as the stack is large enough, another option (-M) can be used to find
10371: the smallest limits that allow a particular pattern to match a given
10372: subject string. This is done by calling pcre[16|32]_exec() repeatedly
10373: with different limits.
1.2 misha 10374:
1.5 misha 10375: Obtaining an estimate of stack usage
10376:
1.6 misha 10377: The actual amount of stack used per recursion can vary quite a lot,
1.5 misha 10378: depending on the compiler that was used to build PCRE and the optimiza-
10379: tion or debugging options that were set for it. The rule of thumb value
1.6 misha 10380: of 500 bytes mentioned above may be larger or smaller than what is
1.5 misha 10381: actually needed. A better approximation can be obtained by running this
10382: command:
10383:
10384: pcretest -m -C
10385:
1.6 misha 10386: The -C option causes pcretest to output information about the options
1.5 misha 10387: with which PCRE was compiled. When -m is also given (before -C), infor-
10388: mation about stack use is given in a line like this:
10389:
10390: Match recursion uses stack: approximate frame size = 640 bytes
10391:
10392: The value is approximate because some recursions need a bit more (up to
10393: perhaps 16 more bytes).
10394:
1.6 misha 10395: If the above command is given when PCRE is compiled to use the heap
10396: instead of the stack for recursion, the value that is output is the
1.5 misha 10397: size of each block that is obtained from the heap.
10398:
1.2 misha 10399: Changing stack size in Unix-like systems
10400:
1.6 misha 10401: In Unix-like environments, there is not often a problem with the stack
10402: unless very long strings are involved, though the default limit on
10403: stack size varies from system to system. Values from 8Mb to 64Mb are
1.1 misha 10404: common. You can find your default limit by running the command:
10405:
10406: ulimit -s
10407:
1.6 misha 10408: Unfortunately, the effect of running out of stack is often SIGSEGV,
10409: though sometimes a more explicit error message is given. You can nor-
1.1 misha 10410: mally increase the limit on stack size by code such as this:
10411:
10412: struct rlimit rlim;
10413: getrlimit(RLIMIT_STACK, &rlim);
10414: rlim.rlim_cur = 100*1024*1024;
10415: setrlimit(RLIMIT_STACK, &rlim);
10416:
1.6 misha 10417: This reads the current limits (soft and hard) using getrlimit(), then
10418: attempts to increase the soft limit to 100Mb using setrlimit(). You
10419: must do this before calling pcre[16|32]_exec().
1.1 misha 10420:
1.2 misha 10421: Changing stack size in Mac OS X
1.1 misha 10422:
1.2 misha 10423: Using setrlimit(), as described above, should also work on Mac OS X. It
10424: is also possible to set a stack size when linking a program. There is a
1.6 misha 10425: discussion about stack sizes in Mac OS X at this web site:
1.2 misha 10426: http://developer.apple.com/qa/qa2005/qa1419.html.
1.1 misha 10427:
10428:
10429: AUTHOR
10430:
10431: Philip Hazel
10432: University Computing Service
10433: Cambridge CB2 3QH, England.
10434:
10435:
10436: REVISION
10437:
1.6 misha 10438: Last updated: 24 June 2012
1.5 misha 10439: Copyright (c) 1997-2012 University of Cambridge.
1.1 misha 10440: ------------------------------------------------------------------------------
10441:
10442:
E-mail:
![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to pcre.txt CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [parser3project] / win32 / pcre