ViewVC logotype

Contents of /code/trunk/doc/pcrepattern.3

Parent Directory Parent Directory | Revision Log Revision Log

Revision 1490 - (show annotations)
Thu Jun 19 07:51:39 2014 UTC (11 months, 1 week ago) by chpe
File size: 135516 byte(s)
Error occurred while calculating annotation data.
Update to Unicode 7.0.0 release
1 .TH PCREPATTERN 3 "08 January 2014" "PCRE 8.35"
3 PCRE - Perl-compatible regular expressions
5 .rs
6 .sp
7 The syntax and semantics of the regular expressions that are supported by PCRE
8 are described in detail below. There is a quick-reference syntax summary in the
9 .\" HREF
10 \fBpcresyntax\fP
11 .\"
12 page. PCRE tries to match Perl syntax and semantics as closely as it can. PCRE
13 also supports some alternative regular expression syntax (which does not
14 conflict with the Perl syntax) in order to provide some compatibility with
15 regular expressions in Python, .NET, and Oniguruma.
16 .P
17 Perl's regular expressions are described in its own documentation, and
18 regular expressions in general are covered in a number of books, some of which
19 have copious examples. Jeffrey Friedl's "Mastering Regular Expressions",
20 published by O'Reilly, covers regular expressions in great detail. This
21 description of PCRE's regular expressions is intended as reference material.
22 .P
23 This document discusses the patterns that are supported by PCRE when one its
24 main matching functions, \fBpcre_exec()\fP (8-bit) or \fBpcre[16|32]_exec()\fP
25 (16- or 32-bit), is used. PCRE also has alternative matching functions,
26 \fBpcre_dfa_exec()\fP and \fBpcre[16|32_dfa_exec()\fP, which match using a
27 different algorithm that is not Perl-compatible. Some of the features discussed
28 below are not available when DFA matching is used. The advantages and
29 disadvantages of the alternative functions, and how they differ from the normal
30 functions, are discussed in the
31 .\" HREF
32 \fBpcrematching\fP
33 .\"
34 page.
35 .
36 .
38 .rs
39 .sp
40 A number of options that can be passed to \fBpcre_compile()\fP can also be set
41 by special items at the start of a pattern. These are not Perl-compatible, but
42 are provided to make these options accessible to pattern writers who are not
43 able to change the program that processes the pattern. Any number of these
44 items may appear, but they must all be together right at the start of the
45 pattern string, and the letters must be in upper case.
46 .
47 .
48 .SS "UTF support"
49 .rs
50 .sp
51 The original operation of PCRE was on strings of one-byte characters. However,
52 there is now also support for UTF-8 strings in the original library, an
53 extra library that supports 16-bit and UTF-16 character strings, and a
54 third library that supports 32-bit and UTF-32 character strings. To use these
55 features, PCRE must be built to include appropriate support. When using UTF
56 strings you must either call the compiling function with the PCRE_UTF8,
57 PCRE_UTF16, or PCRE_UTF32 option, or the pattern must start with one of
58 these special sequences:
59 .sp
60 (*UTF8)
61 (*UTF16)
62 (*UTF32)
63 (*UTF)
64 .sp
65 (*UTF) is a generic sequence that can be used with any of the libraries.
66 Starting a pattern with such a sequence is equivalent to setting the relevant
67 option. How setting a UTF mode affects pattern matching is mentioned in several
68 places below. There is also a summary of features in the
69 .\" HREF
70 \fBpcreunicode\fP
71 .\"
72 page.
73 .P
74 Some applications that allow their users to supply patterns may wish to
75 restrict them to non-UTF data for security reasons. If the PCRE_NEVER_UTF
76 option is set at compile time, (*UTF) etc. are not allowed, and their
77 appearance causes an error.
78 .
79 .
80 .SS "Unicode property support"
81 .rs
82 .sp
83 Another special sequence that may appear at the start of a pattern is (*UCP).
84 This has the same effect as setting the PCRE_UCP option: it causes sequences
85 such as \ed and \ew to use Unicode properties to determine character types,
86 instead of recognizing only characters with codes less than 128 via a lookup
87 table.
88 .
89 .
90 .SS "Disabling auto-possessification"
91 .rs
92 .sp
93 If a pattern starts with (*NO_AUTO_POSSESS), it has the same effect as setting
94 the PCRE_NO_AUTO_POSSESS option at compile time. This stops PCRE from making
95 quantifiers possessive when what follows cannot match the repeated item. For
96 example, by default a+b is treated as a++b. For more details, see the
97 .\" HREF
98 \fBpcreapi\fP
99 .\"
100 documentation.
101 .
102 .
103 .SS "Disabling start-up optimizations"
104 .rs
105 .sp
106 If a pattern starts with (*NO_START_OPT), it has the same effect as setting the
107 PCRE_NO_START_OPTIMIZE option either at compile or matching time. This disables
108 several optimizations for quickly reaching "no match" results. For more
109 details, see the
110 .\" HREF
111 \fBpcreapi\fP
112 .\"
113 documentation.
114 .
115 .
116 .\" HTML <a name="newlines"></a>
117 .SS "Newline conventions"
118 .rs
119 .sp
120 PCRE supports five different conventions for indicating line breaks in
121 strings: a single CR (carriage return) character, a single LF (linefeed)
122 character, the two-character sequence CRLF, any of the three preceding, or any
123 Unicode newline sequence. The
124 .\" HREF
125 \fBpcreapi\fP
126 .\"
127 page has
128 .\" HTML <a href="pcreapi.html#newlines">
129 .\" </a>
130 further discussion
131 .\"
132 about newlines, and shows how to set the newline convention in the
133 \fIoptions\fP arguments for the compiling and matching functions.
134 .P
135 It is also possible to specify a newline convention by starting a pattern
136 string with one of the following five sequences:
137 .sp
138 (*CR) carriage return
139 (*LF) linefeed
140 (*CRLF) carriage return, followed by linefeed
141 (*ANYCRLF) any of the three above
142 (*ANY) all Unicode newline sequences
143 .sp
144 These override the default and the options given to the compiling function. For
145 example, on a Unix system where LF is the default newline sequence, the pattern
146 .sp
147 (*CR)a.b
148 .sp
149 changes the convention to CR. That pattern matches "a\enb" because LF is no
150 longer a newline. If more than one of these settings is present, the last one
151 is used.
152 .P
153 The newline convention affects where the circumflex and dollar assertions are
154 true. It also affects the interpretation of the dot metacharacter when
155 PCRE_DOTALL is not set, and the behaviour of \eN. However, it does not affect
156 what the \eR escape sequence matches. By default, this is any Unicode newline
157 sequence, for Perl compatibility. However, this can be changed; see the
158 description of \eR in the section entitled
159 .\" HTML <a href="#newlineseq">
160 .\" </a>
161 "Newline sequences"
162 .\"
163 below. A change of \eR setting can be combined with a change of newline
164 convention.
165 .
166 .
167 .SS "Setting match and recursion limits"
168 .rs
169 .sp
170 The caller of \fBpcre_exec()\fP can set a limit on the number of times the
171 internal \fBmatch()\fP function is called and on the maximum depth of
172 recursive calls. These facilities are provided to catch runaway matches that
173 are provoked by patterns with huge matching trees (a typical example is a
174 pattern with nested unlimited repeats) and to avoid running out of system stack
175 by too much recursion. When one of these limits is reached, \fBpcre_exec()\fP
176 gives an error return. The limits can also be set by items at the start of the
177 pattern of the form
178 .sp
179 (*LIMIT_MATCH=d)
181 .sp
182 where d is any number of decimal digits. However, the value of the setting must
183 be less than the value set (or defaulted) by the caller of \fBpcre_exec()\fP
184 for it to have any effect. In other words, the pattern writer can lower the
185 limits set by the programmer, but not raise them. If there is more than one
186 setting of one of these limits, the lower value is used.
187 .
188 .
190 .rs
191 .sp
192 PCRE can be compiled to run in an environment that uses EBCDIC as its character
193 code rather than ASCII or Unicode (typically a mainframe system). In the
194 sections below, character code values are ASCII or Unicode; in an EBCDIC
195 environment these characters may have different code values, and there are no
196 code points greater than 255.
197 .
198 .
200 .rs
201 .sp
202 A regular expression is a pattern that is matched against a subject string from
203 left to right. Most characters stand for themselves in a pattern, and match the
204 corresponding characters in the subject. As a trivial example, the pattern
205 .sp
206 The quick brown fox
207 .sp
208 matches a portion of a subject string that is identical to itself. When
209 caseless matching is specified (the PCRE_CASELESS option), letters are matched
210 independently of case. In a UTF mode, PCRE always understands the concept of
211 case for characters whose values are less than 128, so caseless matching is
212 always possible. For characters with higher values, the concept of case is
213 supported if PCRE is compiled with Unicode property support, but not otherwise.
214 If you want to use caseless matching for characters 128 and above, you must
215 ensure that PCRE is compiled with Unicode property support as well as with
216 UTF support.
217 .P
218 The power of regular expressions comes from the ability to include alternatives
219 and repetitions in the pattern. These are encoded in the pattern by the use of
220 \fImetacharacters\fP, which do not stand for themselves but instead are
221 interpreted in some special way.
222 .P
223 There are two different sets of metacharacters: those that are recognized
224 anywhere in the pattern except within square brackets, and those that are
225 recognized within square brackets. Outside square brackets, the metacharacters
226 are as follows:
227 .sp
228 \e general escape character with several uses
229 ^ assert start of string (or line, in multiline mode)
230 $ assert end of string (or line, in multiline mode)
231 . match any character except newline (by default)
232 [ start character class definition
233 | start of alternative branch
234 ( start subpattern
235 ) end subpattern
236 ? extends the meaning of (
237 also 0 or 1 quantifier
238 also quantifier minimizer
239 * 0 or more quantifier
240 + 1 or more quantifier
241 also "possessive quantifier"
242 { start min/max quantifier
243 .sp
244 Part of a pattern that is in square brackets is called a "character class". In
245 a character class the only metacharacters are:
246 .sp
247 \e general escape character
248 ^ negate the class, but only if the first character
249 - indicates character range
250 .\" JOIN
251 [ POSIX character class (only if followed by POSIX
252 syntax)
253 ] terminates the character class
254 .sp
255 The following sections describe the use of each of the metacharacters.
256 .
257 .
259 .rs
260 .sp
261 The backslash character has several uses. Firstly, if it is followed by a
262 character that is not a number or a letter, it takes away any special meaning
263 that character may have. This use of backslash as an escape character applies
264 both inside and outside character classes.
265 .P
266 For example, if you want to match a * character, you write \e* in the pattern.
267 This escaping action applies whether or not the following character would
268 otherwise be interpreted as a metacharacter, so it is always safe to precede a
269 non-alphanumeric with backslash to specify that it stands for itself. In
270 particular, if you want to match a backslash, you write \e\e.
271 .P
272 In a UTF mode, only ASCII numbers and letters have any special meaning after a
273 backslash. All other characters (in particular, those whose codepoints are
274 greater than 127) are treated as literals.
275 .P
276 If a pattern is compiled with the PCRE_EXTENDED option, most white space in the
277 pattern (other than in a character class), and characters between a # outside a
278 character class and the next newline, inclusive, are ignored. An escaping
279 backslash can be used to include a white space or # character as part of the
280 pattern.
281 .P
282 If you want to remove the special meaning from a sequence of characters, you
283 can do so by putting them between \eQ and \eE. This is different from Perl in
284 that $ and @ are handled as literals in \eQ...\eE sequences in PCRE, whereas in
285 Perl, $ and @ cause variable interpolation. Note the following examples:
286 .sp
287 Pattern PCRE matches Perl matches
288 .sp
289 .\" JOIN
290 \eQabc$xyz\eE abc$xyz abc followed by the
291 contents of $xyz
292 \eQabc\e$xyz\eE abc\e$xyz abc\e$xyz
293 \eQabc\eE\e$\eQxyz\eE abc$xyz abc$xyz
294 .sp
295 The \eQ...\eE sequence is recognized both inside and outside character classes.
296 An isolated \eE that is not preceded by \eQ is ignored. If \eQ is not followed
297 by \eE later in the pattern, the literal interpretation continues to the end of
298 the pattern (that is, \eE is assumed at the end). If the isolated \eQ is inside
299 a character class, this causes an error, because the character class is not
300 terminated.
301 .
302 .
303 .\" HTML <a name="digitsafterbackslash"></a>
304 .SS "Non-printing characters"
305 .rs
306 .sp
307 A second use of backslash provides a way of encoding non-printing characters
308 in patterns in a visible manner. There is no restriction on the appearance of
309 non-printing characters, apart from the binary zero that terminates a pattern,
310 but when a pattern is being prepared by text editing, it is often easier to use
311 one of the following escape sequences than the binary character it represents:
312 .sp
313 \ea alarm, that is, the BEL character (hex 07)
314 \ecx "control-x", where x is any ASCII character
315 \ee escape (hex 1B)
316 \ef form feed (hex 0C)
317 \en linefeed (hex 0A)
318 \er carriage return (hex 0D)
319 \et tab (hex 09)
320 \e0dd character with octal code 0dd
321 \eddd character with octal code ddd, or back reference
322 \eo{ddd..} character with octal code ddd..
323 \exhh character with hex code hh
324 \ex{hhh..} character with hex code hhh.. (non-JavaScript mode)
325 \euhhhh character with hex code hhhh (JavaScript mode only)
326 .sp
327 The precise effect of \ecx on ASCII characters is as follows: if x is a lower
328 case letter, it is converted to upper case. Then bit 6 of the character (hex
329 40) is inverted. Thus \ecA to \ecZ become hex 01 to hex 1A (A is 41, Z is 5A),
330 but \ec{ becomes hex 3B ({ is 7B), and \ec; becomes hex 7B (; is 3B). If the
331 data item (byte or 16-bit value) following \ec has a value greater than 127, a
332 compile-time error occurs. This locks out non-ASCII characters in all modes.
333 .P
334 The \ec facility was designed for use with ASCII characters, but with the
335 extension to Unicode it is even less useful than it once was. It is, however,
336 recognized when PCRE is compiled in EBCDIC mode, where data items are always
337 bytes. In this mode, all values are valid after \ec. If the next character is a
338 lower case letter, it is converted to upper case. Then the 0xc0 bits of the
339 byte are inverted. Thus \ecA becomes hex 01, as in ASCII (A is C1), but because
340 the EBCDIC letters are disjoint, \ecZ becomes hex 29 (Z is E9), and other
341 characters also generate different values.
342 .P
343 After \e0 up to two further octal digits are read. If there are fewer than two
344 digits, just those that are present are used. Thus the sequence \e0\ex\e07
345 specifies two binary zeros followed by a BEL character (code value 7). Make
346 sure you supply two digits after the initial zero if the pattern character that
347 follows is itself an octal digit.
348 .P
349 The escape \eo must be followed by a sequence of octal digits, enclosed in
350 braces. An error occurs if this is not the case. This escape is a recent
351 addition to Perl; it provides way of specifying character code points as octal
352 numbers greater than 0777, and it also allows octal numbers and back references
353 to be unambiguously specified.
354 .P
355 For greater clarity and unambiguity, it is best to avoid following \e by a
356 digit greater than zero. Instead, use \eo{} or \ex{} to specify character
357 numbers, and \eg{} to specify back references. The following paragraphs
358 describe the old, ambiguous syntax.
359 .P
360 The handling of a backslash followed by a digit other than 0 is complicated,
361 and Perl has changed in recent releases, causing PCRE also to change. Outside a
362 character class, PCRE reads the digit and any following digits as a decimal
363 number. If the number is less than 8, or if there have been at least that many
364 previous capturing left parentheses in the expression, the entire sequence is
365 taken as a \fIback reference\fP. A description of how this works is given
366 .\" HTML <a href="#backreferences">
367 .\" </a>
368 later,
369 .\"
370 following the discussion of
371 .\" HTML <a href="#subpattern">
372 .\" </a>
373 parenthesized subpatterns.
374 .\"
375 .P
376 Inside a character class, or if the decimal number following \e is greater than
377 7 and there have not been that many capturing subpatterns, PCRE handles \e8 and
378 \e9 as the literal characters "8" and "9", and otherwise re-reads up to three
379 octal digits following the backslash, using them to generate a data character.
380 Any subsequent digits stand for themselves. For example:
381 .sp
382 \e040 is another way of writing an ASCII space
383 .\" JOIN
384 \e40 is the same, provided there are fewer than 40
385 previous capturing subpatterns
386 \e7 is always a back reference
387 .\" JOIN
388 \e11 might be a back reference, or another way of
389 writing a tab
390 \e011 is always a tab
391 \e0113 is a tab followed by the character "3"
392 .\" JOIN
393 \e113 might be a back reference, otherwise the
394 character with octal code 113
395 .\" JOIN
396 \e377 might be a back reference, otherwise
397 the value 255 (decimal)
398 .\" JOIN
399 \e81 is either a back reference, or the two
400 characters "8" and "1"
401 .sp
402 Note that octal values of 100 or greater that are specified using this syntax
403 must not be introduced by a leading zero, because no more than three octal
404 digits are ever read.
405 .P
406 By default, after \ex that is not followed by {, from zero to two hexadecimal
407 digits are read (letters can be in upper or lower case). Any number of
408 hexadecimal digits may appear between \ex{ and }. If a character other than
409 a hexadecimal digit appears between \ex{ and }, or if there is no terminating
410 }, an error occurs.
411 .P
412 If the PCRE_JAVASCRIPT_COMPAT option is set, the interpretation of \ex is
413 as just described only when it is followed by two hexadecimal digits.
414 Otherwise, it matches a literal "x" character. In JavaScript mode, support for
415 code points greater than 256 is provided by \eu, which must be followed by
416 four hexadecimal digits; otherwise it matches a literal "u" character.
417 .P
418 Characters whose value is less than 256 can be defined by either of the two
419 syntaxes for \ex (or by \eu in JavaScript mode). There is no difference in the
420 way they are handled. For example, \exdc is exactly the same as \ex{dc} (or
421 \eu00dc in JavaScript mode).
422 .
423 .
424 .SS "Constraints on character values"
425 .rs
426 .sp
427 Characters that are specified using octal or hexadecimal numbers are
428 limited to certain values, as follows:
429 .sp
430 8-bit non-UTF mode less than 0x100
431 8-bit UTF-8 mode less than 0x10ffff and a valid codepoint
432 16-bit non-UTF mode less than 0x10000
433 16-bit UTF-16 mode less than 0x10ffff and a valid codepoint
434 32-bit non-UTF mode less than 0x100000000
435 32-bit UTF-32 mode less than 0x10ffff and a valid codepoint
436 .sp
437 Invalid Unicode codepoints are the range 0xd800 to 0xdfff (the so-called
438 "surrogate" codepoints), and 0xffef.
439 .
440 .
441 .SS "Escape sequences in character classes"
442 .rs
443 .sp
444 All the sequences that define a single character value can be used both inside
445 and outside character classes. In addition, inside a character class, \eb is
446 interpreted as the backspace character (hex 08).
447 .P
448 \eN is not allowed in a character class. \eB, \eR, and \eX are not special
449 inside a character class. Like other unrecognized escape sequences, they are
450 treated as the literal characters "B", "R", and "X" by default, but cause an
451 error if the PCRE_EXTRA option is set. Outside a character class, these
452 sequences have different meanings.
453 .
454 .
455 .SS "Unsupported escape sequences"
456 .rs
457 .sp
458 In Perl, the sequences \el, \eL, \eu, and \eU are recognized by its string
459 handler and used to modify the case of following characters. By default, PCRE
460 does not support these escape sequences. However, if the PCRE_JAVASCRIPT_COMPAT
461 option is set, \eU matches a "U" character, and \eu can be used to define a
462 character by code point, as described in the previous section.
463 .
464 .
465 .SS "Absolute and relative back references"
466 .rs
467 .sp
468 The sequence \eg followed by an unsigned or a negative number, optionally
469 enclosed in braces, is an absolute or relative back reference. A named back
470 reference can be coded as \eg{name}. Back references are discussed
471 .\" HTML <a href="#backreferences">
472 .\" </a>
473 later,
474 .\"
475 following the discussion of
476 .\" HTML <a href="#subpattern">
477 .\" </a>
478 parenthesized subpatterns.
479 .\"
480 .
481 .
482 .SS "Absolute and relative subroutine calls"
483 .rs
484 .sp
485 For compatibility with Oniguruma, the non-Perl syntax \eg followed by a name or
486 a number enclosed either in angle brackets or single quotes, is an alternative
487 syntax for referencing a subpattern as a "subroutine". Details are discussed
488 .\" HTML <a href="#onigurumasubroutines">
489 .\" </a>
490 later.
491 .\"
492 Note that \eg{...} (Perl syntax) and \eg<...> (Oniguruma syntax) are \fInot\fP
493 synonymous. The former is a back reference; the latter is a
494 .\" HTML <a href="#subpatternsassubroutines">
495 .\" </a>
496 subroutine
497 .\"
498 call.
499 .
500 .
501 .\" HTML <a name="genericchartypes"></a>
502 .SS "Generic character types"
503 .rs
504 .sp
505 Another use of backslash is for specifying generic character types:
506 .sp
507 \ed any decimal digit
508 \eD any character that is not a decimal digit
509 \eh any horizontal white space character
510 \eH any character that is not a horizontal white space character
511 \es any white space character
512 \eS any character that is not a white space character
513 \ev any vertical white space character
514 \eV any character that is not a vertical white space character
515 \ew any "word" character
516 \eW any "non-word" character
517 .sp
518 There is also the single sequence \eN, which matches a non-newline character.
519 This is the same as
520 .\" HTML <a href="#fullstopdot">
521 .\" </a>
522 the "." metacharacter
523 .\"
524 when PCRE_DOTALL is not set. Perl also uses \eN to match characters by name;
525 PCRE does not support this.
526 .P
527 Each pair of lower and upper case escape sequences partitions the complete set
528 of characters into two disjoint sets. Any given character matches one, and only
529 one, of each pair. The sequences can appear both inside and outside character
530 classes. They each match one character of the appropriate type. If the current
531 matching point is at the end of the subject string, all of them fail, because
532 there is no character to match.
533 .P
534 For compatibility with Perl, \es did not used to match the VT character (code
535 11), which made it different from the the POSIX "space" class. However, Perl
536 added VT at release 5.18, and PCRE followed suit at release 8.34. The default
537 \es characters are now HT (9), LF (10), VT (11), FF (12), CR (13), and space
538 (32), which are defined as white space in the "C" locale. This list may vary if
539 locale-specific matching is taking place. For example, in some locales the
540 "non-breaking space" character (\exA0) is recognized as white space, and in
541 others the VT character is not.
542 .P
543 A "word" character is an underscore or any character that is a letter or digit.
544 By default, the definition of letters and digits is controlled by PCRE's
545 low-valued character tables, and may vary if locale-specific matching is taking
546 place (see
547 .\" HTML <a href="pcreapi.html#localesupport">
548 .\" </a>
549 "Locale support"
550 .\"
551 in the
552 .\" HREF
553 \fBpcreapi\fP
554 .\"
555 page). For example, in a French locale such as "fr_FR" in Unix-like systems,
556 or "french" in Windows, some character codes greater than 127 are used for
557 accented letters, and these are then matched by \ew. The use of locales with
558 Unicode is discouraged.
559 .P
560 By default, characters whose code points are greater than 127 never match \ed,
561 \es, or \ew, and always match \eD, \eS, and \eW, although this may vary for
562 characters in the range 128-255 when locale-specific matching is happening.
563 These escape sequences retain their original meanings from before Unicode
564 support was available, mainly for efficiency reasons. If PCRE is compiled with
565 Unicode property support, and the PCRE_UCP option is set, the behaviour is
566 changed so that Unicode properties are used to determine character types, as
567 follows:
568 .sp
569 \ed any character that matches \ep{Nd} (decimal digit)
570 \es any character that matches \ep{Z} or \eh or \ev
571 \ew any character that matches \ep{L} or \ep{N}, plus underscore
572 .sp
573 The upper case escapes match the inverse sets of characters. Note that \ed
574 matches only decimal digits, whereas \ew matches any Unicode digit, as well as
575 any Unicode letter, and underscore. Note also that PCRE_UCP affects \eb, and
576 \eB because they are defined in terms of \ew and \eW. Matching these sequences
577 is noticeably slower when PCRE_UCP is set.
578 .P
579 The sequences \eh, \eH, \ev, and \eV are features that were added to Perl at
580 release 5.10. In contrast to the other sequences, which match only ASCII
581 characters by default, these always match certain high-valued code points,
582 whether or not PCRE_UCP is set. The horizontal space characters are:
583 .sp
584 U+0009 Horizontal tab (HT)
585 U+0020 Space
586 U+00A0 Non-break space
587 U+1680 Ogham space mark
588 U+180E Mongolian vowel separator
589 U+2000 En quad
590 U+2001 Em quad
591 U+2002 En space
592 U+2003 Em space
593 U+2004 Three-per-em space
594 U+2005 Four-per-em space
595 U+2006 Six-per-em space
596 U+2007 Figure space
597 U+2008 Punctuation space
598 U+2009 Thin space
599 U+200A Hair space
600 U+202F Narrow no-break space
601 U+205F Medium mathematical space
602 U+3000 Ideographic space
603 .sp
604 The vertical space characters are:
605 .sp
606 U+000A Linefeed (LF)
607 U+000B Vertical tab (VT)
608 U+000C Form feed (FF)
609 U+000D Carriage return (CR)
610 U+0085 Next line (NEL)
611 U+2028 Line separator
612 U+2029 Paragraph separator
613 .sp
614 In 8-bit, non-UTF-8 mode, only the characters with codepoints less than 256 are
615 relevant.
616 .
617 .
618 .\" HTML <a name="newlineseq"></a>
619 .SS "Newline sequences"
620 .rs
621 .sp
622 Outside a character class, by default, the escape sequence \eR matches any
623 Unicode newline sequence. In 8-bit non-UTF-8 mode \eR is equivalent to the
624 following:
625 .sp
626 (?>\er\en|\en|\ex0b|\ef|\er|\ex85)
627 .sp
628 This is an example of an "atomic group", details of which are given
629 .\" HTML <a href="#atomicgroup">
630 .\" </a>
631 below.
632 .\"
633 This particular group matches either the two-character sequence CR followed by
634 LF, or one of the single characters LF (linefeed, U+000A), VT (vertical tab,
635 U+000B), FF (form feed, U+000C), CR (carriage return, U+000D), or NEL (next
636 line, U+0085). The two-character sequence is treated as a single unit that
637 cannot be split.
638 .P
639 In other modes, two additional characters whose codepoints are greater than 255
640 are added: LS (line separator, U+2028) and PS (paragraph separator, U+2029).
641 Unicode character property support is not needed for these characters to be
642 recognized.
643 .P
644 It is possible to restrict \eR to match only CR, LF, or CRLF (instead of the
645 complete set of Unicode line endings) by setting the option PCRE_BSR_ANYCRLF
646 either at compile time or when the pattern is matched. (BSR is an abbrevation
647 for "backslash R".) This can be made the default when PCRE is built; if this is
648 the case, the other behaviour can be requested via the PCRE_BSR_UNICODE option.
649 It is also possible to specify these settings by starting a pattern string with
650 one of the following sequences:
651 .sp
652 (*BSR_ANYCRLF) CR, LF, or CRLF only
653 (*BSR_UNICODE) any Unicode newline sequence
654 .sp
655 These override the default and the options given to the compiling function, but
656 they can themselves be overridden by options given to a matching function. Note
657 that these special settings, which are not Perl-compatible, are recognized only
658 at the very start of a pattern, and that they must be in upper case. If more
659 than one of them is present, the last one is used. They can be combined with a
660 change of newline convention; for example, a pattern can start with:
661 .sp
663 .sp
664 They can also be combined with the (*UTF8), (*UTF16), (*UTF32), (*UTF) or
665 (*UCP) special sequences. Inside a character class, \eR is treated as an
666 unrecognized escape sequence, and so matches the letter "R" by default, but
667 causes an error if PCRE_EXTRA is set.
668 .
669 .
670 .\" HTML <a name="uniextseq"></a>
671 .SS Unicode character properties
672 .rs
673 .sp
674 When PCRE is built with Unicode character property support, three additional
675 escape sequences that match characters with specific properties are available.
676 When in 8-bit non-UTF-8 mode, these sequences are of course limited to testing
677 characters whose codepoints are less than 256, but they do work in this mode.
678 The extra escape sequences are:
679 .sp
680 \ep{\fIxx\fP} a character with the \fIxx\fP property
681 \eP{\fIxx\fP} a character without the \fIxx\fP property
682 \eX a Unicode extended grapheme cluster
683 .sp
684 The property names represented by \fIxx\fP above are limited to the Unicode
685 script names, the general category properties, "Any", which matches any
686 character (including newline), and some special PCRE properties (described
687 in the
688 .\" HTML <a href="#extraprops">
689 .\" </a>
690 next section).
691 .\"
692 Other Perl properties such as "InMusicalSymbols" are not currently supported by
693 PCRE. Note that \eP{Any} does not match any characters, so always causes a
694 match failure.
695 .P
696 Sets of Unicode characters are defined as belonging to certain scripts. A
697 character from one of these sets can be matched using a script name. For
698 example:
699 .sp
700 \ep{Greek}
701 \eP{Han}
702 .sp
703 Those that are not part of an identified script are lumped together as
704 "Common". The current list of scripts is:
705 .P
706 Arabic,
707 Armenian,
708 Avestan,
709 Balinese,
710 Bamum,
711 Bassa_Vah,
712 Batak,
713 Bengali,
714 Bopomofo,
715 Brahmi,
716 Braille,
717 Buginese,
718 Buhid,
719 Canadian_Aboriginal,
720 Carian,
721 Caucasian_Albanian,
722 Chakma,
723 Cham,
724 Cherokee,
725 Common,
726 Coptic,
727 Cuneiform,
728 Cypriot,
729 Cyrillic,
730 Deseret,
731 Devanagari,
732 Duployan,
733 Egyptian_Hieroglyphs,
734 Elbasan,
735 Ethiopic,
736 Georgian,
737 Glagolitic,
738 Gothic,
739 Grantha,
740 Greek,
741 Gujarati,
742 Gurmukhi,
743 Han,
744 Hangul,
745 Hanunoo,
746 Hebrew,
747 Hiragana,
748 Imperial_Aramaic,
749 Inherited,
750 Inscriptional_Pahlavi,
751 Inscriptional_Parthian,
752 Javanese,
753 Kaithi,
754 Kannada,
755 Katakana,
756 Kayah_Li,
757 Kharoshthi,
758 Khmer,
759 Khojki,
760 Khudawadi,
761 Lao,
762 Latin,
763 Lepcha,
764 Limbu,
765 Linear_A,
766 Linear_B,
767 Lisu,
768 Lycian,
769 Lydian,
770 Mahajani,
771 Malayalam,
772 Mandaic,
773 Manichaean,
774 Meetei_Mayek,
775 Mende_Kikakui,
776 Meroitic_Cursive,
777 Meroitic_Hieroglyphs,
778 Miao,
779 Modi,
780 Mongolian,
781 Mro,
782 Myanmar,
783 Nabataean,
784 New_Tai_Lue,
785 Nko,
786 Ogham,
787 Ol_Chiki,
788 Old_Italic,
789 Old_North_Arabian,
790 Old_Permic,
791 Old_Persian,
792 Old_South_Arabian,
793 Old_Turkic,
794 Oriya,
795 Osmanya,
796 Pahawh_Hmong,
797 Palmyrene,
798 Pau_Cin_Hau,
799 Phags_Pa,
800 Phoenician,
801 Psalter_Pahlavi,
802 Rejang,
803 Runic,
804 Samaritan,
805 Saurashtra,
806 Sharada,
807 Shavian,
808 Siddham,
809 Sinhala,
810 Sora_Sompeng,
811 Sundanese,
812 Syloti_Nagri,
813 Syriac,
814 Tagalog,
815 Tagbanwa,
816 Tai_Le,
817 Tai_Tham,
818 Tai_Viet,
819 Takri,
820 Tamil,
821 Telugu,
822 Thaana,
823 Thai,
824 Tibetan,
825 Tifinagh,
826 Tirhuta,
827 Ugaritic,
828 Vai,
829 Warang_Citi,
830 Yi.
831 .P
832 Each character has exactly one Unicode general category property, specified by
833 a two-letter abbreviation. For compatibility with Perl, negation can be
834 specified by including a circumflex between the opening brace and the property
835 name. For example, \ep{^Lu} is the same as \eP{Lu}.
836 .P
837 If only one letter is specified with \ep or \eP, it includes all the general
838 category properties that start with that letter. In this case, in the absence
839 of negation, the curly brackets in the escape sequence are optional; these two
840 examples have the same effect:
841 .sp
842 \ep{L}
843 \epL
844 .sp
845 The following general category property codes are supported:
846 .sp
847 C Other
848 Cc Control
849 Cf Format
850 Cn Unassigned
851 Co Private use
852 Cs Surrogate
853 .sp
854 L Letter
855 Ll Lower case letter
856 Lm Modifier letter
857 Lo Other letter
858 Lt Title case letter
859 Lu Upper case letter
860 .sp
861 M Mark
862 Mc Spacing mark
863 Me Enclosing mark
864 Mn Non-spacing mark
865 .sp
866 N Number
867 Nd Decimal number
868 Nl Letter number
869 No Other number
870 .sp
871 P Punctuation
872 Pc Connector punctuation
873 Pd Dash punctuation
874 Pe Close punctuation
875 Pf Final punctuation
876 Pi Initial punctuation
877 Po Other punctuation
878 Ps Open punctuation
879 .sp
880 S Symbol
881 Sc Currency symbol
882 Sk Modifier symbol
883 Sm Mathematical symbol
884 So Other symbol
885 .sp
886 Z Separator
887 Zl Line separator
888 Zp Paragraph separator
889 Zs Space separator
890 .sp
891 The special property L& is also supported: it matches a character that has
892 the Lu, Ll, or Lt property, in other words, a letter that is not classified as
893 a modifier or "other".
894 .P
895 The Cs (Surrogate) property applies only to characters in the range U+D800 to
896 U+DFFF. Such characters are not valid in Unicode strings and so
897 cannot be tested by PCRE, unless UTF validity checking has been turned off
898 (see the discussion of PCRE_NO_UTF8_CHECK, PCRE_NO_UTF16_CHECK and
899 PCRE_NO_UTF32_CHECK in the
900 .\" HREF
901 \fBpcreapi\fP
902 .\"
903 page). Perl does not support the Cs property.
904 .P
905 The long synonyms for property names that Perl supports (such as \ep{Letter})
906 are not supported by PCRE, nor is it permitted to prefix any of these
907 properties with "Is".
908 .P
909 No character that is in the Unicode table has the Cn (unassigned) property.
910 Instead, this property is assumed for any code point that is not in the
911 Unicode table.
912 .P
913 Specifying caseless matching does not affect these escape sequences. For
914 example, \ep{Lu} always matches only upper case letters. This is different from
915 the behaviour of current versions of Perl.
916 .P
917 Matching characters by Unicode property is not fast, because PCRE has to do a
918 multistage table lookup in order to find a character's property. That is why
919 the traditional escape sequences such as \ed and \ew do not use Unicode
920 properties in PCRE by default, though you can make them do so by setting the
921 PCRE_UCP option or by starting the pattern with (*UCP).
922 .
923 .
924 .SS Extended grapheme clusters
925 .rs
926 .sp
927 The \eX escape matches any number of Unicode characters that form an "extended
928 grapheme cluster", and treats the sequence as an atomic group
929 .\" HTML <a href="#atomicgroup">
930 .\" </a>
931 (see below).
932 .\"
933 Up to and including release 8.31, PCRE matched an earlier, simpler definition
934 that was equivalent to
935 .sp
936 (?>\ePM\epM*)
937 .sp
938 That is, it matched a character without the "mark" property, followed by zero
939 or more characters with the "mark" property. Characters with the "mark"
940 property are typically non-spacing accents that affect the preceding character.
941 .P
942 This simple definition was extended in Unicode to include more complicated
943 kinds of composite character by giving each character a grapheme breaking
944 property, and creating rules that use these properties to define the boundaries
945 of extended grapheme clusters. In releases of PCRE later than 8.31, \eX matches
946 one of these clusters.
947 .P
948 \eX always matches at least one character. Then it decides whether to add
949 additional characters according to the following rules for ending a cluster:
950 .P
951 1. End at the end of the subject string.
952 .P
953 2. Do not end between CR and LF; otherwise end after any control character.
954 .P
955 3. Do not break Hangul (a Korean script) syllable sequences. Hangul characters
956 are of five types: L, V, T, LV, and LVT. An L character may be followed by an
957 L, V, LV, or LVT character; an LV or V character may be followed by a V or T
958 character; an LVT or T character may be follwed only by a T character.
959 .P
960 4. Do not end before extending characters or spacing marks. Characters with
961 the "mark" property always have the "extend" grapheme breaking property.
962 .P
963 5. Do not end after prepend characters.
964 .P
965 6. Otherwise, end the cluster.
966 .
967 .
968 .\" HTML <a name="extraprops"></a>
969 .SS PCRE's additional properties
970 .rs
971 .sp
972 As well as the standard Unicode properties described above, PCRE supports four
973 more that make it possible to convert traditional escape sequences such as \ew
974 and \es to use Unicode properties. PCRE uses these non-standard, non-Perl
975 properties internally when PCRE_UCP is set. However, they may also be used
976 explicitly. These properties are:
977 .sp
978 Xan Any alphanumeric character
979 Xps Any POSIX space character
980 Xsp Any Perl space character
981 Xwd Any Perl "word" character
982 .sp
983 Xan matches characters that have either the L (letter) or the N (number)
984 property. Xps matches the characters tab, linefeed, vertical tab, form feed, or
985 carriage return, and any other character that has the Z (separator) property.
986 Xsp is the same as Xps; it used to exclude vertical tab, for Perl
987 compatibility, but Perl changed, and so PCRE followed at release 8.34. Xwd
988 matches the same characters as Xan, plus underscore.
989 .P
990 There is another non-standard property, Xuc, which matches any character that
991 can be represented by a Universal Character Name in C++ and other programming
992 languages. These are the characters $, @, ` (grave accent), and all characters
993 with Unicode code points greater than or equal to U+00A0, except for the
994 surrogates U+D800 to U+DFFF. Note that most base (ASCII) characters are
995 excluded. (Universal Character Names are of the form \euHHHH or \eUHHHHHHHH
996 where H is a hexadecimal digit. Note that the Xuc property does not match these
997 sequences but the characters that they represent.)
998 .
999 .
1000 .\" HTML <a name="resetmatchstart"></a>
1001 .SS "Resetting the match start"
1002 .rs
1003 .sp
1004 The escape sequence \eK causes any previously matched characters not to be
1005 included in the final matched sequence. For example, the pattern:
1006 .sp
1007 foo\eKbar
1008 .sp
1009 matches "foobar", but reports that it has matched "bar". This feature is
1010 similar to a lookbehind assertion
1011 .\" HTML <a href="#lookbehind">
1012 .\" </a>
1013 (described below).
1014 .\"
1015 However, in this case, the part of the subject before the real match does not
1016 have to be of fixed length, as lookbehind assertions do. The use of \eK does
1017 not interfere with the setting of
1018 .\" HTML <a href="#subpattern">
1019 .\" </a>
1020 captured substrings.
1021 .\"
1022 For example, when the pattern
1023 .sp
1024 (foo)\eKbar
1025 .sp
1026 matches "foobar", the first substring is still set to "foo".
1027 .P
1028 Perl documents that the use of \eK within assertions is "not well defined". In
1029 PCRE, \eK is acted upon when it occurs inside positive assertions, but is
1030 ignored in negative assertions. Note that when a pattern such as (?=ab\eK)
1031 matches, the reported start of the match can be greater than the end of the
1032 match.
1033 .
1034 .
1035 .\" HTML <a name="smallassertions"></a>
1036 .SS "Simple assertions"
1037 .rs
1038 .sp
1039 The final use of backslash is for certain simple assertions. An assertion
1040 specifies a condition that has to be met at a particular point in a match,
1041 without consuming any characters from the subject string. The use of
1042 subpatterns for more complicated assertions is described
1043 .\" HTML <a href="#bigassertions">
1044 .\" </a>
1045 below.
1046 .\"
1047 The backslashed assertions are:
1048 .sp
1049 \eb matches at a word boundary
1050 \eB matches when not at a word boundary
1051 \eA matches at the start of the subject
1052 \eZ matches at the end of the subject
1053 also matches before a newline at the end of the subject
1054 \ez matches only at the end of the subject
1055 \eG matches at the first matching position in the subject
1056 .sp
1057 Inside a character class, \eb has a different meaning; it matches the backspace
1058 character. If any other of these assertions appears in a character class, by
1059 default it matches the corresponding literal character (for example, \eB
1060 matches the letter B). However, if the PCRE_EXTRA option is set, an "invalid
1061 escape sequence" error is generated instead.
1062 .P
1063 A word boundary is a position in the subject string where the current character
1064 and the previous character do not both match \ew or \eW (i.e. one matches
1065 \ew and the other matches \eW), or the start or end of the string if the
1066 first or last character matches \ew, respectively. In a UTF mode, the meanings
1067 of \ew and \eW can be changed by setting the PCRE_UCP option. When this is
1068 done, it also affects \eb and \eB. Neither PCRE nor Perl has a separate "start
1069 of word" or "end of word" metasequence. However, whatever follows \eb normally
1070 determines which it is. For example, the fragment \eba matches "a" at the start
1071 of a word.
1072 .P
1073 The \eA, \eZ, and \ez assertions differ from the traditional circumflex and
1074 dollar (described in the next section) in that they only ever match at the very
1075 start and end of the subject string, whatever options are set. Thus, they are
1076 independent of multiline mode. These three assertions are not affected by the
1077 PCRE_NOTBOL or PCRE_NOTEOL options, which affect only the behaviour of the
1078 circumflex and dollar metacharacters. However, if the \fIstartoffset\fP
1079 argument of \fBpcre_exec()\fP is non-zero, indicating that matching is to start
1080 at a point other than the beginning of the subject, \eA can never match. The
1081 difference between \eZ and \ez is that \eZ matches before a newline at the end
1082 of the string as well as at the very end, whereas \ez matches only at the end.
1083 .P
1084 The \eG assertion is true only when the current matching position is at the
1085 start point of the match, as specified by the \fIstartoffset\fP argument of
1086 \fBpcre_exec()\fP. It differs from \eA when the value of \fIstartoffset\fP is
1087 non-zero. By calling \fBpcre_exec()\fP multiple times with appropriate
1088 arguments, you can mimic Perl's /g option, and it is in this kind of
1089 implementation where \eG can be useful.
1090 .P
1091 Note, however, that PCRE's interpretation of \eG, as the start of the current
1092 match, is subtly different from Perl's, which defines it as the end of the
1093 previous match. In Perl, these can be different when the previously matched
1094 string was empty. Because PCRE does just one match at a time, it cannot
1095 reproduce this behaviour.
1096 .P
1097 If all the alternatives of a pattern begin with \eG, the expression is anchored
1098 to the starting match position, and the "anchored" flag is set in the compiled
1099 regular expression.
1100 .
1101 .
1103 .rs
1104 .sp
1105 The circumflex and dollar metacharacters are zero-width assertions. That is,
1106 they test for a particular condition being true without consuming any
1107 characters from the subject string.
1108 .P
1109 Outside a character class, in the default matching mode, the circumflex
1110 character is an assertion that is true only if the current matching point is at
1111 the start of the subject string. If the \fIstartoffset\fP argument of
1112 \fBpcre_exec()\fP is non-zero, circumflex can never match if the PCRE_MULTILINE
1113 option is unset. Inside a character class, circumflex has an entirely different
1114 meaning
1115 .\" HTML <a href="#characterclass">
1116 .\" </a>
1117 (see below).
1118 .\"
1119 .P
1120 Circumflex need not be the first character of the pattern if a number of
1121 alternatives are involved, but it should be the first thing in each alternative
1122 in which it appears if the pattern is ever to match that branch. If all
1123 possible alternatives start with a circumflex, that is, if the pattern is
1124 constrained to match only at the start of the subject, it is said to be an
1125 "anchored" pattern. (There are also other constructs that can cause a pattern
1126 to be anchored.)
1127 .P
1128 The dollar character is an assertion that is true only if the current matching
1129 point is at the end of the subject string, or immediately before a newline at
1130 the end of the string (by default). Note, however, that it does not actually
1131 match the newline. Dollar need not be the last character of the pattern if a
1132 number of alternatives are involved, but it should be the last item in any
1133 branch in which it appears. Dollar has no special meaning in a character class.
1134 .P
1135 The meaning of dollar can be changed so that it matches only at the very end of
1136 the string, by setting the PCRE_DOLLAR_ENDONLY option at compile time. This
1137 does not affect the \eZ assertion.
1138 .P
1139 The meanings of the circumflex and dollar characters are changed if the
1140 PCRE_MULTILINE option is set. When this is the case, a circumflex matches
1141 immediately after internal newlines as well as at the start of the subject
1142 string. It does not match after a newline that ends the string. A dollar
1143 matches before any newlines in the string, as well as at the very end, when
1144 PCRE_MULTILINE is set. When newline is specified as the two-character
1145 sequence CRLF, isolated CR and LF characters do not indicate newlines.
1146 .P
1147 For example, the pattern /^abc$/ matches the subject string "def\enabc" (where
1148 \en represents a newline) in multiline mode, but not otherwise. Consequently,
1149 patterns that are anchored in single line mode because all branches start with
1150 ^ are not anchored in multiline mode, and a match for circumflex is possible
1151 when the \fIstartoffset\fP argument of \fBpcre_exec()\fP is non-zero. The
1152 PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
1153 .P
1154 Note that the sequences \eA, \eZ, and \ez can be used to match the start and
1155 end of the subject in both modes, and if all branches of a pattern start with
1156 \eA it is always anchored, whether or not PCRE_MULTILINE is set.
1157 .
1158 .
1159 .\" HTML <a name="fullstopdot"></a>
1161 .rs
1162 .sp
1163 Outside a character class, a dot in the pattern matches any one character in
1164 the subject string except (by default) a character that signifies the end of a
1165 line.
1166 .P
1167 When a line ending is defined as a single character, dot never matches that
1168 character; when the two-character sequence CRLF is used, dot does not match CR
1169 if it is immediately followed by LF, but otherwise it matches all characters
1170 (including isolated CRs and LFs). When any Unicode line endings are being
1171 recognized, dot does not match CR or LF or any of the other line ending
1172 characters.
1173 .P
1174 The behaviour of dot with regard to newlines can be changed. If the PCRE_DOTALL
1175 option is set, a dot matches any one character, without exception. If the
1176 two-character sequence CRLF is present in the subject string, it takes two dots
1177 to match it.
1178 .P
1179 The handling of dot is entirely independent of the handling of circumflex and
1180 dollar, the only relationship being that they both involve newlines. Dot has no
1181 special meaning in a character class.
1182 .P
1183 The escape sequence \eN behaves like a dot, except that it is not affected by
1184 the PCRE_DOTALL option. In other words, it matches any character except one
1185 that signifies the end of a line. Perl also uses \eN to match characters by
1186 name; PCRE does not support this.
1187 .
1188 .
1190 .rs
1191 .sp
1192 Outside a character class, the escape sequence \eC matches any one data unit,
1193 whether or not a UTF mode is set. In the 8-bit library, one data unit is one
1194 byte; in the 16-bit library it is a 16-bit unit; in the 32-bit library it is
1195 a 32-bit unit. Unlike a dot, \eC always
1196 matches line-ending characters. The feature is provided in Perl in order to
1197 match individual bytes in UTF-8 mode, but it is unclear how it can usefully be
1198 used. Because \eC breaks up characters into individual data units, matching one
1199 unit with \eC in a UTF mode means that the rest of the string may start with a
1200 malformed UTF character. This has undefined results, because PCRE assumes that
1201 it is dealing with valid UTF strings (and by default it checks this at the
1202 start of processing unless the PCRE_NO_UTF8_CHECK, PCRE_NO_UTF16_CHECK or
1203 PCRE_NO_UTF32_CHECK option is used).
1204 .P
1205 PCRE does not allow \eC to appear in lookbehind assertions
1206 .\" HTML <a href="#lookbehind">
1207 .\" </a>
1208 (described below)
1209 .\"
1210 in a UTF mode, because this would make it impossible to calculate the length of
1211 the lookbehind.
1212 .P
1213 In general, the \eC escape sequence is best avoided. However, one
1214 way of using it that avoids the problem of malformed UTF characters is to use a
1215 lookahead to check the length of the next character, as in this pattern, which
1216 could be used with a UTF-8 string (ignore white space and line breaks):
1217 .sp
1218 (?| (?=[\ex00-\ex7f])(\eC) |
1219 (?=[\ex80-\ex{7ff}])(\eC)(\eC) |
1220 (?=[\ex{800}-\ex{ffff}])(\eC)(\eC)(\eC) |
1221 (?=[\ex{10000}-\ex{1fffff}])(\eC)(\eC)(\eC)(\eC))
1222 .sp
1223 A group that starts with (?| resets the capturing parentheses numbers in each
1224 alternative (see
1225 .\" HTML <a href="#dupsubpatternnumber">
1226 .\" </a>
1227 "Duplicate Subpattern Numbers"
1228 .\"
1229 below). The assertions at the start of each branch check the next UTF-8
1230 character for values whose encoding uses 1, 2, 3, or 4 bytes, respectively. The
1231 character's individual bytes are then captured by the appropriate number of
1232 groups.
1233 .
1234 .
1235 .\" HTML <a name="characterclass"></a>
1237 .rs
1238 .sp
1239 An opening square bracket introduces a character class, terminated by a closing
1240 square bracket. A closing square bracket on its own is not special by default.
1241 However, if the PCRE_JAVASCRIPT_COMPAT option is set, a lone closing square
1242 bracket causes a compile-time error. If a closing square bracket is required as
1243 a member of the class, it should be the first data character in the class
1244 (after an initial circumflex, if present) or escaped with a backslash.
1245 .P
1246 A character class matches a single character in the subject. In a UTF mode, the
1247 character may be more than one data unit long. A matched character must be in
1248 the set of characters defined by the class, unless the first character in the
1249 class definition is a circumflex, in which case the subject character must not
1250 be in the set defined by the class. If a circumflex is actually required as a
1251 member of the class, ensure it is not the first character, or escape it with a
1252 backslash.
1253 .P
1254 For example, the character class [aeiou] matches any lower case vowel, while
1255 [^aeiou] matches any character that is not a lower case vowel. Note that a
1256 circumflex is just a convenient notation for specifying the characters that
1257 are in the class by enumerating those that are not. A class that starts with a
1258 circumflex is not an assertion; it still consumes a character from the subject
1259 string, and therefore it fails if the current pointer is at the end of the
1260 string.
1261 .P
1262 In UTF-8 (UTF-16, UTF-32) mode, characters with values greater than 255 (0xffff)
1263 can be included in a class as a literal string of data units, or by using the
1264 \ex{ escaping mechanism.
1265 .P
1266 When caseless matching is set, any letters in a class represent both their
1267 upper case and lower case versions, so for example, a caseless [aeiou] matches
1268 "A" as well as "a", and a caseless [^aeiou] does not match "A", whereas a
1269 caseful version would. In a UTF mode, PCRE always understands the concept of
1270 case for characters whose values are less than 128, so caseless matching is
1271 always possible. For characters with higher values, the concept of case is
1272 supported if PCRE is compiled with Unicode property support, but not otherwise.
1273 If you want to use caseless matching in a UTF mode for characters 128 and
1274 above, you must ensure that PCRE is compiled with Unicode property support as
1275 well as with UTF support.
1276 .P
1277 Characters that might indicate line breaks are never treated in any special way
1278 when matching character classes, whatever line-ending sequence is in use, and
1279 whatever setting of the PCRE_DOTALL and PCRE_MULTILINE options is used. A class
1280 such as [^a] always matches one of these characters.
1281 .P
1282 The minus (hyphen) character can be used to specify a range of characters in a
1283 character class. For example, [d-m] matches any letter between d and m,
1284 inclusive. If a minus character is required in a class, it must be escaped with
1285 a backslash or appear in a position where it cannot be interpreted as
1286 indicating a range, typically as the first or last character in the class, or
1287 immediately after a range. For example, [b-d-z] matches letters in the range b
1288 to d, a hyphen character, or z.
1289 .P
1290 It is not possible to have the literal character "]" as the end character of a
1291 range. A pattern such as [W-]46] is interpreted as a class of two characters
1292 ("W" and "-") followed by a literal string "46]", so it would match "W46]" or
1293 "-46]". However, if the "]" is escaped with a backslash it is interpreted as
1294 the end of range, so [W-\e]46] is interpreted as a class containing a range
1295 followed by two other characters. The octal or hexadecimal representation of
1296 "]" can also be used to end a range.
1297 .P
1298 An error is generated if a POSIX character class (see below) or an escape
1299 sequence other than one that defines a single character appears at a point
1300 where a range ending character is expected. For example, [z-\exff] is valid,
1301 but [A-\ed] and [A-[:digit:]] are not.
1302 .P
1303 Ranges operate in the collating sequence of character values. They can also be
1304 used for characters specified numerically, for example [\e000-\e037]. Ranges
1305 can include any characters that are valid for the current mode.
1306 .P
1307 If a range that includes letters is used when caseless matching is set, it
1308 matches the letters in either case. For example, [W-c] is equivalent to
1309 [][\e\e^_`wxyzabc], matched caselessly, and in a non-UTF mode, if character
1310 tables for a French locale are in use, [\exc8-\excb] matches accented E
1311 characters in both cases. In UTF modes, PCRE supports the concept of case for
1312 characters with values greater than 128 only when it is compiled with Unicode
1313 property support.
1314 .P
1315 The character escape sequences \ed, \eD, \eh, \eH, \ep, \eP, \es, \eS, \ev,
1316 \eV, \ew, and \eW may appear in a character class, and add the characters that
1317 they match to the class. For example, [\edABCDEF] matches any hexadecimal
1318 digit. In UTF modes, the PCRE_UCP option affects the meanings of \ed, \es, \ew
1319 and their upper case partners, just as it does when they appear outside a
1320 character class, as described in the section entitled
1321 .\" HTML <a href="#genericchartypes">
1322 .\" </a>
1323 "Generic character types"
1324 .\"
1325 above. The escape sequence \eb has a different meaning inside a character
1326 class; it matches the backspace character. The sequences \eB, \eN, \eR, and \eX
1327 are not special inside a character class. Like any other unrecognized escape
1328 sequences, they are treated as the literal characters "B", "N", "R", and "X" by
1329 default, but cause an error if the PCRE_EXTRA option is set.
1330 .P
1331 A circumflex can conveniently be used with the upper case character types to
1332 specify a more restricted set of characters than the matching lower case type.
1333 For example, the class [^\eW_] matches any letter or digit, but not underscore,
1334 whereas [\ew] includes underscore. A positive character class should be read as
1335 "something OR something OR ..." and a negative class as "NOT something AND NOT
1336 something AND NOT ...".
1337 .P
1338 The only metacharacters that are recognized in character classes are backslash,
1339 hyphen (only where it can be interpreted as specifying a range), circumflex
1340 (only at the start), opening square bracket (only when it can be interpreted as
1341 introducing a POSIX class name, or for a special compatibility feature - see
1342 the next two sections), and the terminating closing square bracket. However,
1343 escaping other non-alphanumeric characters does no harm.
1344 .
1345 .
1347 .rs
1348 .sp
1349 Perl supports the POSIX notation for character classes. This uses names
1350 enclosed by [: and :] within the enclosing square brackets. PCRE also supports
1351 this notation. For example,
1352 .sp
1353 [01[:alpha:]%]
1354 .sp
1355 matches "0", "1", any alphabetic character, or "%". The supported class names
1356 are:
1357 .sp
1358 alnum letters and digits
1359 alpha letters
1360 ascii character codes 0 - 127
1361 blank space or tab only
1362 cntrl control characters
1363 digit decimal digits (same as \ed)
1364 graph printing characters, excluding space
1365 lower lower case letters
1366 print printing characters, including space
1367 punct printing characters, excluding letters and digits and space
1368 space white space (the same as \es from PCRE 8.34)
1369 upper upper case letters
1370 word "word" characters (same as \ew)
1371 xdigit hexadecimal digits
1372 .sp
1373 The default "space" characters are HT (9), LF (10), VT (11), FF (12), CR (13),
1374 and space (32). If locale-specific matching is taking place, the list of space
1375 characters may be different; there may be fewer or more of them. "Space" used
1376 to be different to \es, which did not include VT, for Perl compatibility.
1377 However, Perl changed at release 5.18, and PCRE followed at release 8.34.
1378 "Space" and \es now match the same set of characters.
1379 .P
1380 The name "word" is a Perl extension, and "blank" is a GNU extension from Perl
1381 5.8. Another Perl extension is negation, which is indicated by a ^ character
1382 after the colon. For example,
1383 .sp
1384 [12[:^digit:]]
1385 .sp
1386 matches "1", "2", or any non-digit. PCRE (and Perl) also recognize the POSIX
1387 syntax [.ch.] and [=ch=] where "ch" is a "collating element", but these are not
1388 supported, and an error is given if they are encountered.
1389 .P
1390 By default, characters with values greater than 128 do not match any of the
1391 POSIX character classes. However, if the PCRE_UCP option is passed to
1392 \fBpcre_compile()\fP, some of the classes are changed so that Unicode character
1393 properties are used. This is achieved by replacing certain POSIX classes by
1394 other sequences, as follows:
1395 .sp
1396 [:alnum:] becomes \ep{Xan}
1397 [:alpha:] becomes \ep{L}
1398 [:blank:] becomes \eh
1399 [:digit:] becomes \ep{Nd}
1400 [:lower:] becomes \ep{Ll}
1401 [:space:] becomes \ep{Xps}
1402 [:upper:] becomes \ep{Lu}
1403 [:word:] becomes \ep{Xwd}
1404 .sp
1405 Negated versions, such as [:^alpha:] use \eP instead of \ep. Three other POSIX
1406 classes are handled specially in UCP mode:
1407 .TP 10
1408 [:graph:]
1409 This matches characters that have glyphs that mark the page when printed. In
1410 Unicode property terms, it matches all characters with the L, M, N, P, S, or Cf
1411 properties, except for:
1412 .sp
1413 U+061C Arabic Letter Mark
1414 U+180E Mongolian Vowel Separator
1415 U+2066 - U+2069 Various "isolate"s
1416 .sp
1417 .TP 10
1418 [:print:]
1419 This matches the same characters as [:graph:] plus space characters that are
1420 not controls, that is, characters with the Zs property.
1421 .TP 10
1422 [:punct:]
1423 This matches all characters that have the Unicode P (punctuation) property,
1424 plus those characters whose code points are less than 128 that have the S
1425 (Symbol) property.
1426 .P
1427 The other POSIX classes are unchanged, and match only characters with code
1428 points less than 128.
1429 .
1430 .
1432 .rs
1433 .sp
1434 In the POSIX.2 compliant library that was included in 4.4BSD Unix, the ugly
1435 syntax [[:<:]] and [[:>:]] is used for matching "start of word" and "end of
1436 word". PCRE treats these items as follows:
1437 .sp
1438 [[:<:]] is converted to \eb(?=\ew)
1439 [[:>:]] is converted to \eb(?<=\ew)
1440 .sp
1441 Only these exact character sequences are recognized. A sequence such as
1442 [a[:<:]b] provokes error for an unrecognized POSIX class name. This support is
1443 not compatible with Perl. It is provided to help migrations from other
1444 environments, and is best not used in any new patterns. Note that \eb matches
1445 at the start and the end of a word (see
1446 .\" HTML <a href="#smallassertions">
1447 .\" </a>
1448 "Simple assertions"
1449 .\"
1450 above), and in a Perl-style pattern the preceding or following character
1451 normally shows which is wanted, without the need for the assertions that are
1452 used above in order to give exactly the POSIX behaviour.
1453 .
1454 .
1456 .rs
1457 .sp
1458 Vertical bar characters are used to separate alternative patterns. For example,
1459 the pattern
1460 .sp
1461 gilbert|sullivan
1462 .sp
1463 matches either "gilbert" or "sullivan". Any number of alternatives may appear,
1464 and an empty alternative is permitted (matching the empty string). The matching
1465 process tries each alternative in turn, from left to right, and the first one
1466 that succeeds is used. If the alternatives are within a subpattern
1467 .\" HTML <a href="#subpattern">
1468 .\" </a>
1469 (defined below),
1470 .\"
1471 "succeeds" means matching the rest of the main pattern as well as the
1472 alternative in the subpattern.
1473 .
1474 .
1476 .rs
1477 .sp
1478 The settings of the PCRE_CASELESS, PCRE_MULTILINE, PCRE_DOTALL, and
1479 PCRE_EXTENDED options (which are Perl-compatible) can be changed from within
1480 the pattern by a sequence of Perl option letters enclosed between "(?" and ")".
1481 The option letters are
1482 .sp
1483 i for PCRE_CASELESS
1485 s for PCRE_DOTALL
1486 x for PCRE_EXTENDED
1487 .sp
1488 For example, (?im) sets caseless, multiline matching. It is also possible to
1489 unset these options by preceding the letter with a hyphen, and a combined
1490 setting and unsetting such as (?im-sx), which sets PCRE_CASELESS and
1491 PCRE_MULTILINE while unsetting PCRE_DOTALL and PCRE_EXTENDED, is also
1492 permitted. If a letter appears both before and after the hyphen, the option is
1493 unset.
1494 .P
1495 The PCRE-specific options PCRE_DUPNAMES, PCRE_UNGREEDY, and PCRE_EXTRA can be
1496 changed in the same way as the Perl-compatible options by using the characters
1497 J, U and X respectively.
1498 .P
1499 When one of these option changes occurs at top level (that is, not inside
1500 subpattern parentheses), the change applies to the remainder of the pattern
1501 that follows. If the change is placed right at the start of a pattern, PCRE
1502 extracts it into the global options (and it will therefore show up in data
1503 extracted by the \fBpcre_fullinfo()\fP function).
1504 .P
1505 An option change within a subpattern (see below for a description of
1506 subpatterns) affects only that part of the subpattern that follows it, so
1507 .sp
1508 (a(?i)b)c
1509 .sp
1510 matches abc and aBc and no other strings (assuming PCRE_CASELESS is not used).
1511 By this means, options can be made to have different settings in different
1512 parts of the pattern. Any changes made in one alternative do carry on
1513 into subsequent branches within the same subpattern. For example,
1514 .sp
1515 (a(?i)b|c)
1516 .sp
1517 matches "ab", "aB", "c", and "C", even though when matching "C" the first
1518 branch is abandoned before the option setting. This is because the effects of
1519 option settings happen at compile time. There would be some very weird
1520 behaviour otherwise.
1521 .P
1522 \fBNote:\fP There are other PCRE-specific options that can be set by the
1523 application when the compiling or matching functions are called. In some cases
1524 the pattern can contain special leading sequences such as (*CRLF) to override
1525 what the application has set or what has been defaulted. Details are given in
1526 the section entitled
1527 .\" HTML <a href="#newlineseq">
1528 .\" </a>
1529 "Newline sequences"
1530 .\"
1531 above. There are also the (*UTF8), (*UTF16),(*UTF32), and (*UCP) leading
1532 sequences that can be used to set UTF and Unicode property modes; they are
1533 equivalent to setting the PCRE_UTF8, PCRE_UTF16, PCRE_UTF32 and the PCRE_UCP
1534 options, respectively. The (*UTF) sequence is a generic version that can be
1535 used with any of the libraries. However, the application can set the
1536 PCRE_NEVER_UTF option, which locks out the use of the (*UTF) sequences.
1537 .
1538 .
1539 .\" HTML <a name="subpattern"></a>
1541 .rs
1542 .sp
1543 Subpatterns are delimited by parentheses (round brackets), which can be nested.
1544 Turning part of a pattern into a subpattern does two things:
1545 .sp
1546 1. It localizes a set of alternatives. For example, the pattern
1547 .sp
1548 cat(aract|erpillar|)
1549 .sp
1550 matches "cataract", "caterpillar", or "cat". Without the parentheses, it would
1551 match "cataract", "erpillar" or an empty string.
1552 .sp
1553 2. It sets up the subpattern as a capturing subpattern. This means that, when
1554 the whole pattern matches, that portion of the subject string that matched the
1555 subpattern is passed back to the caller via the \fIovector\fP argument of the
1556 matching function. (This applies only to the traditional matching functions;
1557 the DFA matching functions do not support capturing.)
1558 .P
1559 Opening parentheses are counted from left to right (starting from 1) to obtain
1560 numbers for the capturing subpatterns. For example, if the string "the red
1561 king" is matched against the pattern
1562 .sp
1563 the ((red|white) (king|queen))
1564 .sp
1565 the captured substrings are "red king", "red", and "king", and are numbered 1,
1566 2, and 3, respectively.
1567 .P
1568 The fact that plain parentheses fulfil two functions is not always helpful.
1569 There are often times when a grouping subpattern is required without a
1570 capturing requirement. If an opening parenthesis is followed by a question mark
1571 and a colon, the subpattern does not do any capturing, and is not counted when
1572 computing the number of any subsequent capturing subpatterns. For example, if
1573 the string "the white queen" is matched against the pattern
1574 .sp
1575 the ((?:red|white) (king|queen))
1576 .sp
1577 the captured substrings are "white queen" and "queen", and are numbered 1 and
1578 2. The maximum number of capturing subpatterns is 65535.
1579 .P
1580 As a convenient shorthand, if any option settings are required at the start of
1581 a non-capturing subpattern, the option letters may appear between the "?" and
1582 the ":". Thus the two patterns
1583 .sp
1584 (?i:saturday|sunday)
1585 (?:(?i)saturday|sunday)
1586 .sp
1587 match exactly the same set of strings. Because alternative branches are tried
1588 from left to right, and options are not reset until the end of the subpattern
1589 is reached, an option setting in one branch does affect subsequent branches, so
1590 the above patterns match "SUNDAY" as well as "Saturday".
1591 .
1592 .
1593 .\" HTML <a name="dupsubpatternnumber"></a>
1595 .rs
1596 .sp
1597 Perl 5.10 introduced a feature whereby each alternative in a subpattern uses
1598 the same numbers for its capturing parentheses. Such a subpattern starts with
1599 (?| and is itself a non-capturing subpattern. For example, consider this
1600 pattern:
1601 .sp
1602 (?|(Sat)ur|(Sun))day
1603 .sp
1604 Because the two alternatives are inside a (?| group, both sets of capturing
1605 parentheses are numbered one. Thus, when the pattern matches, you can look
1606 at captured substring number one, whichever alternative matched. This construct
1607 is useful when you want to capture part, but not all, of one of a number of
1608 alternatives. Inside a (?| group, parentheses are numbered as usual, but the
1609 number is reset at the start of each branch. The numbers of any capturing
1610 parentheses that follow the subpattern start after the highest number used in
1611 any branch. The following example is taken from the Perl documentation. The
1612 numbers underneath show in which buffer the captured content will be stored.
1613 .sp
1614 # before ---------------branch-reset----------- after
1615 / ( a ) (?| x ( y ) z | (p (q) r) | (t) u (v) ) ( z ) /x
1616 # 1 2 2 3 2 3 4
1617 .sp
1618 A back reference to a numbered subpattern uses the most recent value that is
1619 set for that number by any subpattern. The following pattern matches "abcabc"
1620 or "defdef":
1621 .sp
1622 /(?|(abc)|(def))\e1/
1623 .sp
1624 In contrast, a subroutine call to a numbered subpattern always refers to the
1625 first one in the pattern with the given number. The following pattern matches
1626 "abcabc" or "defabc":
1627 .sp
1628 /(?|(abc)|(def))(?1)/
1629 .sp
1630 If a
1631 .\" HTML <a href="#conditions">
1632 .\" </a>
1633 condition test
1634 .\"
1635 for a subpattern's having matched refers to a non-unique number, the test is
1636 true if any of the subpatterns of that number have matched.
1637 .P
1638 An alternative approach to using this "branch reset" feature is to use
1639 duplicate named subpatterns, as described in the next section.
1640 .
1641 .
1643 .rs
1644 .sp
1645 Identifying capturing parentheses by number is simple, but it can be very hard
1646 to keep track of the numbers in complicated regular expressions. Furthermore,
1647 if an expression is modified, the numbers may change. To help with this
1648 difficulty, PCRE supports the naming of subpatterns. This feature was not
1649 added to Perl until release 5.10. Python had the feature earlier, and PCRE
1650 introduced it at release 4.0, using the Python syntax. PCRE now supports both
1651 the Perl and the Python syntax. Perl allows identically numbered subpatterns to
1652 have different names, but PCRE does not.
1653 .P
1654 In PCRE, a subpattern can be named in one of three ways: (?<name>...) or
1655 (?'name'...) as in Perl, or (?P<name>...) as in Python. References to capturing
1656 parentheses from other parts of the pattern, such as
1657 .\" HTML <a href="#backreferences">
1658 .\" </a>
1659 back references,
1660 .\"
1661 .\" HTML <a href="#recursion">
1662 .\" </a>
1663 recursion,
1664 .\"
1665 and
1666 .\" HTML <a href="#conditions">
1667 .\" </a>
1668 conditions,
1669 .\"
1670 can be made by name as well as by number.
1671 .P
1672 Names consist of up to 32 alphanumeric characters and underscores, but must
1673 start with a non-digit. Named capturing parentheses are still allocated numbers
1674 as well as names, exactly as if the names were not present. The PCRE API
1675 provides function calls for extracting the name-to-number translation table
1676 from a compiled pattern. There is also a convenience function for extracting a
1677 captured substring by name.
1678 .P
1679 By default, a name must be unique within a pattern, but it is possible to relax
1680 this constraint by setting the PCRE_DUPNAMES option at compile time. (Duplicate
1681 names are also always permitted for subpatterns with the same number, set up as
1682 described in the previous section.) Duplicate names can be useful for patterns
1683 where only one instance of the named parentheses can match. Suppose you want to
1684 match the name of a weekday, either as a 3-letter abbreviation or as the full
1685 name, and in both cases you want to extract the abbreviation. This pattern
1686 (ignoring the line breaks) does the job:
1687 .sp
1688 (?<DN>Mon|Fri|Sun)(?:day)?|
1689 (?<DN>Tue)(?:sday)?|
1690 (?<DN>Wed)(?:nesday)?|
1691 (?<DN>Thu)(?:rsday)?|
1692 (?<DN>Sat)(?:urday)?
1693 .sp
1694 There are five capturing substrings, but only one is ever set after a match.
1695 (An alternative way of solving this problem is to use a "branch reset"
1696 subpattern, as described in the previous section.)
1697 .P
1698 The convenience function for extracting the data by name returns the substring
1699 for the first (and in this example, the only) subpattern of that name that
1700 matched. This saves searching to find which numbered subpattern it was.
1701 .P
1702 If you make a back reference to a non-unique named subpattern from elsewhere in
1703 the pattern, the subpatterns to which the name refers are checked in the order
1704 in which they appear in the overall pattern. The first one that is set is used
1705 for the reference. For example, this pattern matches both "foofoo" and
1706 "barbar" but not "foobar" or "barfoo":
1707 .sp
1708 (?:(?<n>foo)|(?<n>bar))\ek<n>
1709 .sp
1710 .P
1711 If you make a subroutine call to a non-unique named subpattern, the one that
1712 corresponds to the first occurrence of the name is used. In the absence of
1713 duplicate numbers (see the previous section) this is the one with the lowest
1714 number.
1715 .P
1716 If you use a named reference in a condition
1717 test (see the
1718 .\"
1719 .\" HTML <a href="#conditions">
1720 .\" </a>
1721 section about conditions
1722 .\"
1723 below), either to check whether a subpattern has matched, or to check for
1724 recursion, all subpatterns with the same name are tested. If the condition is
1725 true for any one of them, the overall condition is true. This is the same
1726 behaviour as testing by number. For further details of the interfaces for
1727 handling named subpatterns, see the
1728 .\" HREF
1729 \fBpcreapi\fP
1730 .\"
1731 documentation.
1732 .P
1733 \fBWarning:\fP You cannot use different names to distinguish between two
1734 subpatterns with the same number because PCRE uses only the numbers when
1735 matching. For this reason, an error is given at compile time if different names
1736 are given to subpatterns with the same number. However, you can always give the
1737 same name to subpatterns with the same number, even when PCRE_DUPNAMES is not
1738 set.
1739 .
1740 .
1742 .rs
1743 .sp
1744 Repetition is specified by quantifiers, which can follow any of the following
1745 items:
1746 .sp
1747 a literal data character
1748 the dot metacharacter
1749 the \eC escape sequence
1750 the \eX escape sequence
1751 the \eR escape sequence
1752 an escape such as \ed or \epL that matches a single character
1753 a character class
1754 a back reference (see next section)
1755 a parenthesized subpattern (including assertions)
1756 a subroutine call to a subpattern (recursive or otherwise)
1757 .sp
1758 The general repetition quantifier specifies a minimum and maximum number of
1759 permitted matches, by giving the two numbers in curly brackets (braces),
1760 separated by a comma. The numbers must be less than 65536, and the first must
1761 be less than or equal to the second. For example:
1762 .sp
1763 z{2,4}
1764 .sp
1765 matches "zz", "zzz", or "zzzz". A closing brace on its own is not a special
1766 character. If the second number is omitted, but the comma is present, there is
1767 no upper limit; if the second number and the comma are both omitted, the
1768 quantifier specifies an exact number of required matches. Thus
1769 .sp
1770 [aeiou]{3,}
1771 .sp
1772 matches at least 3 successive vowels, but may match many more, while
1773 .sp
1774 \ed{8}
1775 .sp
1776 matches exactly 8 digits. An opening curly bracket that appears in a position
1777 where a quantifier is not allowed, or one that does not match the syntax of a
1778 quantifier, is taken as a literal character. For example, {,6} is not a
1779 quantifier, but a literal string of four characters.
1780 .P
1781 In UTF modes, quantifiers apply to characters rather than to individual data
1782 units. Thus, for example, \ex{100}{2} matches two characters, each of
1783 which is represented by a two-byte sequence in a UTF-8 string. Similarly,
1784 \eX{3} matches three Unicode extended grapheme clusters, each of which may be
1785 several data units long (and they may be of different lengths).
1786 .P
1787 The quantifier {0} is permitted, causing the expression to behave as if the
1788 previous item and the quantifier were not present. This may be useful for
1789 subpatterns that are referenced as
1790 .\" HTML <a href="#subpatternsassubroutines">
1791 .\" </a>
1792 subroutines
1793 .\"
1794 from elsewhere in the pattern (but see also the section entitled
1795 .\" HTML <a href="#subdefine">
1796 .\" </a>
1797 "Defining subpatterns for use by reference only"
1798 .\"
1799 below). Items other than subpatterns that have a {0} quantifier are omitted
1800 from the compiled pattern.
1801 .P
1802 For convenience, the three most common quantifiers have single-character
1803 abbreviations:
1804 .sp
1805 * is equivalent to {0,}
1806 + is equivalent to {1,}
1807 ? is equivalent to {0,1}
1808 .sp
1809 It is possible to construct infinite loops by following a subpattern that can
1810 match no characters with a quantifier that has no upper limit, for example:
1811 .sp
1812 (a?)*
1813 .sp
1814 Earlier versions of Perl and PCRE used to give an error at compile time for
1815 such patterns. However, because there are cases where this can be useful, such
1816 patterns are now accepted, but if any repetition of the subpattern does in fact
1817 match no characters, the loop is forcibly broken.
1818 .P
1819 By default, the quantifiers are "greedy", that is, they match as much as
1820 possible (up to the maximum number of permitted times), without causing the
1821 rest of the pattern to fail. The classic example of where this gives problems
1822 is in trying to match comments in C programs. These appear between /* and */
1823 and within the comment, individual * and / characters may appear. An attempt to
1824 match C comments by applying the pattern
1825 .sp
1826 /\e*.*\e*/
1827 .sp
1828 to the string
1829 .sp
1830 /* first comment */ not comment /* second comment */
1831 .sp
1832 fails, because it matches the entire string owing to the greediness of the .*
1833 item.
1834 .P
1835 However, if a quantifier is followed by a question mark, it ceases to be
1836 greedy, and instead matches the minimum number of times possible, so the
1837 pattern
1838 .sp
1839 /\e*.*?\e*/
1840 .sp
1841 does the right thing with the C comments. The meaning of the various
1842 quantifiers is not otherwise changed, just the preferred number of matches.
1843 Do not confuse this use of question mark with its use as a quantifier in its
1844 own right. Because it has two uses, it can sometimes appear doubled, as in
1845 .sp
1846 \ed??\ed
1847 .sp
1848 which matches one digit by preference, but can match two if that is the only
1849 way the rest of the pattern matches.
1850 .P
1851 If the PCRE_UNGREEDY option is set (an option that is not available in Perl),
1852 the quantifiers are not greedy by default, but individual ones can be made
1853 greedy by following them with a question mark. In other words, it inverts the
1854 default behaviour.
1855 .P
1856 When a parenthesized subpattern is quantified with a minimum repeat count that
1857 is greater than 1 or with a limited maximum, more memory is required for the
1858 compiled pattern, in proportion to the size of the minimum or maximum.
1859 .P
1860 If a pattern starts with .* or .{0,} and the PCRE_DOTALL option (equivalent
1861 to Perl's /s) is set, thus allowing the dot to match newlines, the pattern is
1862 implicitly anchored, because whatever follows will be tried against every
1863 character position in the subject string, so there is no point in retrying the
1864 overall match at any position after the first. PCRE normally treats such a
1865 pattern as though it were preceded by \eA.
1866 .P
1867 In cases where it is known that the subject string contains no newlines, it is
1868 worth setting PCRE_DOTALL in order to obtain this optimization, or
1869 alternatively using ^ to indicate anchoring explicitly.
1870 .P
1871 However, there are some cases where the optimization cannot be used. When .*
1872 is inside capturing parentheses that are the subject of a back reference
1873 elsewhere in the pattern, a match at the start may fail where a later one
1874 succeeds. Consider, for example:
1875 .sp
1876 (.*)abc\e1
1877 .sp
1878 If the subject is "xyz123abc123" the match point is the fourth character. For
1879 this reason, such a pattern is not implicitly anchored.
1880 .P
1881 Another case where implicit anchoring is not applied is when the leading .* is
1882 inside an atomic group. Once again, a match at the start may fail where a later
1883 one succeeds. Consider this pattern:
1884 .sp
1885 (?>.*?a)b
1886 .sp
1887 It matches "ab" in the subject "aab". The use of the backtracking control verbs
1888 (*PRUNE) and (*SKIP) also disable this optimization.
1889 .P
1890 When a capturing subpattern is repeated, the value captured is the substring
1891 that matched the final iteration. For example, after
1892 .sp
1893 (tweedle[dume]{3}\es*)+
1894 .sp
1895 has matched "tweedledum tweedledee" the value of the captured substring is
1896 "tweedledee". However, if there are nested capturing subpatterns, the
1897 corresponding captured values may have been set in previous iterations. For
1898 example, after
1899 .sp
1900 /(a|(b))+/
1901 .sp
1902 matches "aba" the value of the second captured substring is "b".
1903 .
1904 .
1905 .\" HTML <a name="atomicgroup"></a>
1907 .rs
1908 .sp
1909 With both maximizing ("greedy") and minimizing ("ungreedy" or "lazy")
1910 repetition, failure of what follows normally causes the repeated item to be
1911 re-evaluated to see if a different number of repeats allows the rest of the
1912 pattern to match. Sometimes it is useful to prevent this, either to change the
1913 nature of the match, or to cause it fail earlier than it otherwise might, when
1914 the author of the pattern knows there is no point in carrying on.
1915 .P
1916 Consider, for example, the pattern \ed+foo when applied to the subject line
1917 .sp
1918 123456bar
1919 .sp
1920 After matching all 6 digits and then failing to match "foo", the normal
1921 action of the matcher is to try again with only 5 digits matching the \ed+
1922 item, and then with 4, and so on, before ultimately failing. "Atomic grouping"
1923 (a term taken from Jeffrey Friedl's book) provides the means for specifying
1924 that once a subpattern has matched, it is not to be re-evaluated in this way.
1925 .P
1926 If we use atomic grouping for the previous example, the matcher gives up
1927 immediately on failing to match "foo" the first time. The notation is a kind of
1928 special parenthesis, starting with (?> as in this example:
1929 .sp
1930 (?>\ed+)foo
1931 .sp
1932 This kind of parenthesis "locks up" the part of the pattern it contains once
1933 it has matched, and a failure further into the pattern is prevented from
1934 backtracking into it. Backtracking past it to previous items, however, works as
1935 normal.
1936 .P
1937 An alternative description is that a subpattern of this type matches the string
1938 of characters that an identical standalone pattern would match, if anchored at
1939 the current point in the subject string.
1940 .P
1941 Atomic grouping subpatterns are not capturing subpatterns. Simple cases such as
1942 the above example can be thought of as a maximizing repeat that must swallow
1943 everything it can. So, while both \ed+ and \ed+? are prepared to adjust the
1944 number of digits they match in order to make the rest of the pattern match,
1945 (?>\ed+) can only match an entire sequence of digits.
1946 .P
1947 Atomic groups in general can of course contain arbitrarily complicated
1948 subpatterns, and can be nested. However, when the subpattern for an atomic
1949 group is just a single repeated item, as in the example above, a simpler
1950 notation, called a "possessive quantifier" can be used. This consists of an
1951 additional + character following a quantifier. Using this notation, the
1952 previous example can be rewritten as
1953 .sp
1954 \ed++foo
1955 .sp
1956 Note that a possessive quantifier can be used with an entire group, for
1957 example:
1958 .sp
1959 (abc|xyz){2,3}+
1960 .sp
1961 Possessive quantifiers are always greedy; the setting of the PCRE_UNGREEDY
1962 option is ignored. They are a convenient notation for the simpler forms of
1963 atomic group. However, there is no difference in the meaning of a possessive
1964 quantifier and the equivalent atomic group, though there may be a performance
1965 difference; possessive quantifiers should be slightly faster.
1966 .P
1967 The possessive quantifier syntax is an extension to the Perl 5.8 syntax.
1968 Jeffrey Friedl originated the idea (and the name) in the first edition of his
1969 book. Mike McCloskey liked it, so implemented it when he built Sun's Java
1970 package, and PCRE copied it from there. It ultimately found its way into Perl
1971 at release 5.10.
1972 .P
1973 PCRE has an optimization that automatically "possessifies" certain simple
1974 pattern constructs. For example, the sequence A+B is treated as A++B because
1975 there is no point in backtracking into a sequence of A's when B must follow.
1976 .P
1977 When a pattern contains an unlimited repeat inside a subpattern that can itself
1978 be repeated an unlimited number of times, the use of an atomic group is the
1979 only way to avoid some failing matches taking a very long time indeed. The
1980 pattern
1981 .sp
1982 (\eD+|<\ed+>)*[!?]
1983 .sp
1984 matches an unlimited number of substrings that either consist of non-digits, or
1985 digits enclosed in <>, followed by either ! or ?. When it matches, it runs
1986 quickly. However, if it is applied to
1987 .sp
1988 aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
1989 .sp
1990 it takes a long time before reporting failure. This is because the string can
1991 be divided between the internal \eD+ repeat and the external * repeat in a
1992 large number of ways, and all have to be tried. (The example uses [!?] rather
1993 than a single character at the end, because both PCRE and Perl have an
1994 optimization that allows for fast failure when a single character is used. They
1995 remember the last single character that is required for a match, and fail early
1996 if it is not present in the string.) If the pattern is changed so that it uses
1997 an atomic group, like this:
1998 .sp
1999 ((?>\eD+)|<\ed+>)*[!?]
2000 .sp
2001 sequences of non-digits cannot be broken, and failure happens quickly.
2002 .
2003 .
2004 .\" HTML <a name="backreferences"></a>
2006 .rs
2007 .sp
2008 Outside a character class, a backslash followed by a digit greater than 0 (and
2009 possibly further digits) is a back reference to a capturing subpattern earlier
2010 (that is, to its left) in the pattern, provided there have been that many
2011 previous capturing left parentheses.
2012 .P
2013 However, if the decimal number following the backslash is less than 10, it is
2014 always taken as a back reference, and causes an error only if there are not
2015 that many capturing left parentheses in the entire pattern. In other words, the
2016 parentheses that are referenced need not be to the left of the reference for
2017 numbers less than 10. A "forward back reference" of this type can make sense
2018 when a repetition is involved and the subpattern to the right has participated
2019 in an earlier iteration.
2020 .P
2021 It is not possible to have a numerical "forward back reference" to a subpattern
2022 whose number is 10 or more using this syntax because a sequence such as \e50 is
2023 interpreted as a character defined in octal. See the subsection entitled
2024 "Non-printing characters"
2025 .\" HTML <a href="#digitsafterbackslash">
2026 .\" </a>
2027 above
2028 .\"
2029 for further details of the handling of digits following a backslash. There is
2030 no such problem when named parentheses are used. A back reference to any
2031 subpattern is possible using named parentheses (see below).
2032 .P
2033 Another way of avoiding the ambiguity inherent in the use of digits following a
2034 backslash is to use the \eg escape sequence. This escape must be followed by an
2035 unsigned number or a negative number, optionally enclosed in braces. These
2036 examples are all identical:
2037 .sp
2038 (ring), \e1
2039 (ring), \eg1
2040 (ring), \eg{1}
2041 .sp
2042 An unsigned number specifies an absolute reference without the ambiguity that
2043 is present in the older syntax. It is also useful when literal digits follow
2044 the reference. A negative number is a relative reference. Consider this
2045 example:
2046 .sp
2047 (abc(def)ghi)\eg{-1}
2048 .sp
2049 The sequence \eg{-1} is a reference to the most recently started capturing
2050 subpattern before \eg, that is, is it equivalent to \e2 in this example.
2051 Similarly, \eg{-2} would be equivalent to \e1. The use of relative references
2052 can be helpful in long patterns, and also in patterns that are created by
2053 joining together fragments that contain references within themselves.
2054 .P
2055 A back reference matches whatever actually matched the capturing subpattern in
2056 the current subject string, rather than anything matching the subpattern
2057 itself (see
2058 .\" HTML <a href="#subpatternsassubroutines">
2059 .\" </a>
2060 "Subpatterns as subroutines"
2061 .\"
2062 below for a way of doing that). So the pattern
2063 .sp
2064 (sens|respons)e and \e1ibility
2065 .sp
2066 matches "sense and sensibility" and "response and responsibility", but not
2067 "sense and responsibility". If caseful matching is in force at the time of the
2068 back reference, the case of letters is relevant. For example,
2069 .sp
2070 ((?i)rah)\es+\e1
2071 .sp
2072 matches "rah rah" and "RAH RAH", but not "RAH rah", even though the original
2073 capturing subpattern is matched caselessly.
2074 .P
2075 There are several different ways of writing back references to named
2076 subpatterns. The .NET syntax \ek{name} and the Perl syntax \ek<name> or
2077 \ek'name' are supported, as is the Python syntax (?P=name). Perl 5.10's unified
2078 back reference syntax, in which \eg can be used for both numeric and named
2079 references, is also supported. We could rewrite the above example in any of
2080 the following ways:
2081 .sp
2082 (?<p1>(?i)rah)\es+\ek<p1>
2083 (?'p1'(?i)rah)\es+\ek{p1}
2084 (?P<p1>(?i)rah)\es+(?P=p1)
2085 (?<p1>(?i)rah)\es+\eg{p1}
2086 .sp
2087 A subpattern that is referenced by name may appear in the pattern before or
2088 after the reference.
2089 .P
2090 There may be more than one back reference to the same subpattern. If a
2091 subpattern has not actually been used in a particular match, any back
2092 references to it always fail by default. For example, the pattern
2093 .sp
2094 (a|(bc))\e2
2095 .sp
2096 always fails if it starts to match "a" rather than "bc". However, if the
2097 PCRE_JAVASCRIPT_COMPAT option is set at compile time, a back reference to an
2098 unset value matches an empty string.
2099 .P
2100 Because there may be many capturing parentheses in a pattern, all digits
2101 following a backslash are taken as part of a potential back reference number.
2102 If the pattern continues with a digit character, some delimiter must be used to
2103 terminate the back reference. If the PCRE_EXTENDED option is set, this can be
2104 white space. Otherwise, the \eg{ syntax or an empty comment (see
2105 .\" HTML <a href="#comments">
2106 .\" </a>
2107 "Comments"
2108 .\"
2109 below) can be used.
2110 .
2111 .SS "Recursive back references"
2112 .rs
2113 .sp
2114 A back reference that occurs inside the parentheses to which it refers fails
2115 when the subpattern is first used, so, for example, (a\e1) never matches.
2116 However, such references can be useful inside repeated subpatterns. For
2117 example, the pattern
2118 .sp
2119 (a|b\e1)+
2120 .sp
2121 matches any number of "a"s and also "aba", "ababbaa" etc. At each iteration of
2122 the subpattern, the back reference matches the character string corresponding
2123 to the previous iteration. In order for this to work, the pattern must be such
2124 that the first iteration does not need to match the back reference. This can be
2125 done using alternation, as in the example above, or by a quantifier with a
2126 minimum of zero.
2127 .P
2128 Back references of this type cause the group that they reference to be treated
2129 as an
2130 .\" HTML <a href="#atomicgroup">
2131 .\" </a>
2132 atomic group.
2133 .\"
2134 Once the whole group has been matched, a subsequent matching failure cannot
2135 cause backtracking into the middle of the group.
2136 .
2137 .
2138 .\" HTML <a name="bigassertions"></a>
2140 .rs
2141 .sp
2142 An assertion is a test on the characters following or preceding the current
2143 matching point that does not actually consume any characters. The simple
2144 assertions coded as \eb, \eB, \eA, \eG, \eZ, \ez, ^ and $ are described
2145 .\" HTML <a href="#smallassertions">
2146 .\" </a>
2147 above.
2148 .\"
2149 .P
2150 More complicated assertions are coded as subpatterns. There are two kinds:
2151 those that look ahead of the current position in the subject string, and those
2152 that look behind it. An assertion subpattern is matched in the normal way,
2153 except that it does not cause the current matching position to be changed.
2154 .P
2155 Assertion subpatterns are not capturing subpatterns. If such an assertion
2156 contains capturing subpatterns within it, these are counted for the purposes of
2157 numbering the capturing subpatterns in the whole pattern. However, substring
2158 capturing is carried out only for positive assertions. (Perl sometimes, but not
2159 always, does do capturing in negative assertions.)
2160 .P
2161 For compatibility with Perl, assertion subpatterns may be repeated; though
2162 it makes no sense to assert the same thing several times, the side effect of
2163 capturing parentheses may occasionally be useful. In practice, there only three
2164 cases:
2165 .sp
2166 (1) If the quantifier is {0}, the assertion is never obeyed during matching.
2167 However, it may contain internal capturing parenthesized groups that are called
2168 from elsewhere via the
2169 .\" HTML <a href="#subpatternsassubroutines">
2170 .\" </a>
2171 subroutine mechanism.
2172 .\"
2173 .sp
2174 (2) If quantifier is {0,n} where n is greater than zero, it is treated as if it
2175 were {0,1}. At run time, the rest of the pattern match is tried with and
2176 without the assertion, the order depending on the greediness of the quantifier.
2177 .sp
2178 (3) If the minimum repetition is greater than zero, the quantifier is ignored.
2179 The assertion is obeyed just once when encountered during matching.
2180 .
2181 .
2182 .SS "Lookahead assertions"
2183 .rs
2184 .sp
2185 Lookahead assertions start with (?= for positive assertions and (?! for
2186 negative assertions. For example,
2187 .sp
2188 \ew+(?=;)
2189 .sp
2190 matches a word followed by a semicolon, but does not include the semicolon in
2191 the match, and
2192 .sp
2193 foo(?!bar)
2194 .sp
2195 matches any occurrence of "foo" that is not followed by "bar". Note that the
2196 apparently similar pattern
2197 .sp
2198 (?!foo)bar
2199 .sp
2200 does not find an occurrence of "bar" that is preceded by something other than
2201 "foo"; it finds any occurrence of "bar" whatsoever, because the assertion
2202 (?!foo) is always true when the next three characters are "bar". A
2203 lookbehind assertion is needed to achieve the other effect.
2204 .P
2205 If you want to force a matching failure at some point in a pattern, the most
2206 convenient way to do it is with (?!) because an empty string always matches, so
2207 an assertion that requires there not to be an empty string must always fail.
2208 The backtracking control verb (*FAIL) or (*F) is a synonym for (?!).
2209 .
2210 .
2211 .\" HTML <a name="lookbehind"></a>
2212 .SS "Lookbehind assertions"
2213 .rs
2214 .sp
2215 Lookbehind assertions start with (?<= for positive assertions and (?<! for
2216 negative assertions. For example,
2217 .sp
2218 (?<!foo)bar
2219 .sp
2220 does find an occurrence of "bar" that is not preceded by "foo". The contents of
2221 a lookbehind assertion are restricted such that all the strings it matches must
2222 have a fixed length. However, if there are several top-level alternatives, they
2223 do not all have to have the same fixed length. Thus
2224 .sp
2225 (?<=bullock|donkey)
2226 .sp
2227 is permitted, but
2228 .sp
2229 (?<!dogs?|cats?)
2230 .sp
2231 causes an error at compile time. Branches that match different length strings
2232 are permitted only at the top level of a lookbehind assertion. This is an
2233 extension compared with Perl, which requires all branches to match the same
2234 length of string. An assertion such as
2235 .sp
2236 (?<=ab(c|de))
2237 .sp
2238 is not permitted, because its single top-level branch can match two different
2239 lengths, but it is acceptable to PCRE if rewritten to use two top-level
2240 branches:
2241 .sp
2242 (?<=abc|abde)
2243 .sp
2244 In some cases, the escape sequence \eK
2245 .\" HTML <a href="#resetmatchstart">
2246 .\" </a>
2247 (see above)
2248 .\"
2249 can be used instead of a lookbehind assertion to get round the fixed-length
2250 restriction.
2251 .P
2252 The implementation of lookbehind assertions is, for each alternative, to
2253 temporarily move the current position back by the fixed length and then try to
2254 match. If there are insufficient characters before the current position, the
2255 assertion fails.
2256 .P
2257 In a UTF mode, PCRE does not allow the \eC escape (which matches a single data
2258 unit even in a UTF mode) to appear in lookbehind assertions, because it makes
2259 it impossible to calculate the length of the lookbehind. The \eX and \eR
2260 escapes, which can match different numbers of data units, are also not
2261 permitted.
2262 .P
2263 .\" HTML <a href="#subpatternsassubroutines">
2264 .\" </a>
2265 "Subroutine"
2266 .\"
2267 calls (see below) such as (?2) or (?&X) are permitted in lookbehinds, as long
2268 as the subpattern matches a fixed-length string.
2269 .\" HTML <a href="#recursion">
2270 .\" </a>
2271 Recursion,
2272 .\"
2273 however, is not supported.
2274 .P
2275 Possessive quantifiers can be used in conjunction with lookbehind assertions to
2276 specify efficient matching of fixed-length strings at the end of subject
2277 strings. Consider a simple pattern such as
2278 .sp
2279 abcd$
2280 .sp
2281 when applied to a long string that does not match. Because matching proceeds
2282 from left to right, PCRE will look for each "a" in the subject and then see if
2283 what follows matches the rest of the pattern. If the pattern is specified as
2284 .sp
2285 ^.*abcd$
2286 .sp
2287 the initial .* matches the entire string at first, but when this fails (because
2288 there is no following "a"), it backtracks to match all but the last character,
2289 then all but the last two characters, and so on. Once again the search for "a"
2290 covers the entire string, from right to left, so we are no better off. However,
2291 if the pattern is written as
2292 .sp
2293 ^.*+(?<=abcd)
2294 .sp
2295 there can be no backtracking for the .*+ item; it can match only the entire
2296 string. The subsequent lookbehind assertion does a single test on the last four
2297 characters. If it fails, the match fails immediately. For long strings, this
2298 approach makes a significant difference to the processing time.
2299 .
2300 .
2301 .SS "Using multiple assertions"
2302 .rs
2303 .sp
2304 Several assertions (of any sort) may occur in succession. For example,
2305 .sp
2306 (?<=\ed{3})(?<!999)foo
2307 .sp
2308 matches "foo" preceded by three digits that are not "999". Notice that each of
2309 the assertions is applied independently at the same point in the subject
2310 string. First there is a check that the previous three characters are all
2311 digits, and then there is a check that the same three characters are not "999".
2312 This pattern does \fInot\fP match "foo" preceded by six characters, the first
2313 of which are digits and the last three of which are not "999". For example, it
2314 doesn't match "123abcfoo". A pattern to do that is
2315 .sp
2316 (?<=\ed{3}...)(?<!999)foo
2317 .sp
2318 This time the first assertion looks at the preceding six characters, checking
2319 that the first three are digits, and then the second assertion checks that the
2320 preceding three characters are not "999".
2321 .P
2322 Assertions can be nested in any combination. For example,
2323 .sp
2324 (?<=(?<!foo)bar)baz
2325 .sp
2326 matches an occurrence of "baz" that is preceded by "bar" which in turn is not
2327 preceded by "foo", while
2328 .sp
2329 (?<=\ed{3}(?!999)...)foo
2330 .sp
2331 is another pattern that matches "foo" preceded by three digits and any three
2332 characters that are not "999".
2333 .
2334 .
2335 .\" HTML <a name="conditions"></a>
2337 .rs
2338 .sp
2339 It is possible to cause the matching process to obey a subpattern
2340 conditionally or to choose between two alternative subpatterns, depending on
2341 the result of an assertion, or whether a specific capturing subpattern has
2342 already been matched. The two possible forms of conditional subpattern are:
2343 .sp
2344 (?(condition)yes-pattern)
2345 (?(condition)yes-pattern|no-pattern)
2346 .sp
2347 If the condition is satisfied, the yes-pattern is used; otherwise the
2348 no-pattern (if present) is used. If there are more than two alternatives in the
2349 subpattern, a compile-time error occurs. Each of the two alternatives may
2350 itself contain nested subpatterns of any form, including conditional
2351 subpatterns; the restriction to two alternatives applies only at the level of
2352 the condition. This pattern fragment is an example where the alternatives are
2353 complex:
2354 .sp
2355 (?(1) (A|B|C) | (D | (?(2)E|F) | E) )
2356 .sp
2357 .P
2358 There are four kinds of condition: references to subpatterns, references to
2359 recursion, a pseudo-condition called DEFINE, and assertions.
2360 .
2361 .SS "Checking for a used subpattern by number"
2362 .rs
2363 .sp
2364 If the text between the parentheses consists of a sequence of digits, the
2365 condition is true if a capturing subpattern of that number has previously
2366 matched. If there is more than one capturing subpattern with the same number
2367 (see the earlier
2368 .\"
2369 .\" HTML <a href="#recursion">
2370 .\" </a>
2371 section about duplicate subpattern numbers),
2372 .\"
2373 the condition is true if any of them have matched. An alternative notation is
2374 to precede the digits with a plus or minus sign. In this case, the subpattern
2375 number is relative rather than absolute. The most recently opened parentheses
2376 can be referenced by (?(-1), the next most recent by (?(-2), and so on. Inside
2377 loops it can also make sense to refer to subsequent groups. The next
2378 parentheses to be opened can be referenced as (?(+1), and so on. (The value
2379 zero in any of these forms is not used; it provokes a compile-time error.)
2380 .P
2381 Consider the following pattern, which contains non-significant white space to
2382 make it more readable (assume the PCRE_EXTENDED option) and to divide it into
2383 three parts for ease of discussion:
2384 .sp
2385 ( \e( )? [^()]+ (?(1) \e) )
2386 .sp
2387 The first part matches an optional opening parenthesis, and if that
2388 character is present, sets it as the first captured substring. The second part
2389 matches one or more characters that are not parentheses. The third part is a
2390 conditional subpattern that tests whether or not the first set of parentheses
2391 matched. If they did, that is, if subject started with an opening parenthesis,
2392 the condition is true, and so the yes-pattern is executed and a closing
2393 parenthesis is required. Otherwise, since no-pattern is not present, the
2394 subpattern matches nothing. In other words, this pattern matches a sequence of
2395 non-parentheses, optionally enclosed in parentheses.
2396 .P
2397 If you were embedding this pattern in a larger one, you could use a relative
2398 reference:
2399 .sp
2400 ...other stuff... ( \e( )? [^()]+ (?(-1) \e) ) ...
2401 .sp
2402 This makes the fragment independent of the parentheses in the larger pattern.
2403 .
2404 .SS "Checking for a used subpattern by name"
2405 .rs
2406 .sp
2407 Perl uses the syntax (?(<name>)...) or (?('name')...) to test for a used
2408 subpattern by name. For compatibility with earlier versions of PCRE, which had
2409 this facility before Perl, the syntax (?(name)...) is also recognized.
2410 .P
2411 Rewriting the above example to use a named subpattern gives this:
2412 .sp
2413 (?<OPEN> \e( )? [^()]+ (?(<OPEN>) \e) )
2414 .sp
2415 If the name used in a condition of this kind is a duplicate, the test is
2416 applied to all subpatterns of the same name, and is true if any one of them has
2417 matched.
2418 .
2419 .SS "Checking for pattern recursion"
2420 .rs
2421 .sp
2422 If the condition is the string (R), and there is no subpattern with the name R,
2423 the condition is true if a recursive call to the whole pattern or any
2424 subpattern has been made. If digits or a name preceded by ampersand follow the
2425 letter R, for example:
2426 .sp
2427 (?(R3)...) or (?(R&name)...)
2428 .sp
2429 the condition is true if the most recent recursion is into a subpattern whose
2430 number or name is given. This condition does not check the entire recursion
2431 stack. If the name used in a condition of this kind is a duplicate, the test is
2432 applied to all subpatterns of the same name, and is true if any one of them is
2433 the most recent recursion.
2434 .P
2435 At "top level", all these recursion test conditions are false.
2436 .\" HTML <a href="#recursion">
2437 .\" </a>
2438 The syntax for recursive patterns
2439 .\"
2440 is described below.
2441 .
2442 .\" HTML <a name="subdefine"></a>
2443 .SS "Defining subpatterns for use by reference only"
2444 .rs
2445 .sp
2446 If the condition is the string (DEFINE), and there is no subpattern with the
2447 name DEFINE, the condition is always false. In this case, there may be only one
2448 alternative in the subpattern. It is always skipped if control reaches this
2449 point in the pattern; the idea of DEFINE is that it can be used to define
2450 subroutines that can be referenced from elsewhere. (The use of
2451 .\" HTML <a href="#subpatternsassubroutines">
2452 .\" </a>
2453 subroutines
2454 .\"
2455 is described below.) For example, a pattern to match an IPv4 address such as
2456 "" could be written like this (ignore white space and line
2457 breaks):
2458 .sp
2459 (?(DEFINE) (?<byte> 2[0-4]\ed | 25[0-5] | 1\ed\ed | [1-9]?\ed) )
2460 \eb (?&byte) (\e.(?&byte)){3} \eb
2461 .sp
2462 The first part of the pattern is a DEFINE group inside which a another group
2463 named "byte" is defined. This matches an individual component of an IPv4
2464 address (a number less than 256). When matching takes place, this part of the
2465 pattern is skipped because DEFINE acts like a false condition. The rest of the
2466 pattern uses references to the named group to match the four dot-separated
2467 components of an IPv4 address, insisting on a word boundary at each end.
2468 .
2469 .SS "Assertion conditions"
2470 .rs
2471 .sp
2472 If the condition is not in any of the above formats, it must be an assertion.
2473 This may be a positive or negative lookahead or lookbehind assertion. Consider
2474 this pattern, again containing non-significant white space, and with the two
2475 alternatives on the second line:
2476 .sp
2477 (?(?=[^a-z]*[a-z])
2478 \ed{2}-[a-z]{3}-\ed{2} | \ed{2}-\ed{2}-\ed{2} )
2479 .sp
2480 The condition is a positive lookahead assertion that matches an optional
2481 sequence of non-letters followed by a letter. In other words, it tests for the
2482 presence of at least one letter in the subject. If a letter is found, the
2483 subject is matched against the first alternative; otherwise it is matched
2484 against the second. This pattern matches strings in one of the two forms
2485 dd-aaa-dd or dd-dd-dd, where aaa are letters and dd are digits.
2486 .
2487 .
2488 .\" HTML <a name="comments"></a>
2490 .rs
2491 .sp
2492 There are two ways of including comments in patterns that are processed by
2493 PCRE. In both cases, the start of the comment must not be in a character class,
2494 nor in the middle of any other sequence of related characters such as (?: or a
2495 subpattern name or number. The characters that make up a comment play no part
2496 in the pattern matching.
2497 .P
2498 The sequence (?# marks the start of a comment that continues up to the next
2499 closing parenthesis. Nested parentheses are not permitted. If the PCRE_EXTENDED
2500 option is set, an unescaped # character also introduces a comment, which in
2501 this case continues to immediately after the next newline character or
2502 character sequence in the pattern. Which characters are interpreted as newlines
2503 is controlled by the options passed to a compiling function or by a special
2504 sequence at the start of the pattern, as described in the section entitled
2505 .\" HTML <a href="#newlines">
2506 .\" </a>
2507 "Newline conventions"
2508 .\"
2509 above. Note that the end of this type of comment is a literal newline sequence
2510 in the pattern; escape sequences that happen to represent a newline do not
2511 count. For example, consider this pattern when PCRE_EXTENDED is set, and the
2512 default newline convention is in force:
2513 .sp
2514 abc #comment \en still comment
2515 .sp
2516 On encountering the # character, \fBpcre_compile()\fP skips along, looking for
2517 a newline in the pattern. The sequence \en is still literal at this stage, so
2518 it does not terminate the comment. Only an actual character with the code value
2519 0x0a (the default newline) does so.
2520 .
2521 .
2522 .\" HTML <a name="recursion"></a>
2524 .rs
2525 .sp
2526 Consider the problem of matching a string in parentheses, allowing for
2527 unlimited nested parentheses. Without the use of recursion, the best that can
2528 be done is to use a pattern that matches up to some fixed depth of nesting. It
2529 is not possible to handle an arbitrary nesting depth.
2530 .P
2531 For some time, Perl has provided a facility that allows regular expressions to
2532 recurse (amongst other things). It does this by interpolating Perl code in the
2533 expression at run time, and the code can refer to the expression itself. A Perl
2534 pattern using code interpolation to solve the parentheses problem can be
2535 created like this:
2536 .sp
2537 $re = qr{\e( (?: (?>[^()]+) | (?p{$re}) )* \e)}x;
2538 .sp
2539 The (?p{...}) item interpolates Perl code at run time, and in this case refers
2540 recursively to the pattern in which it appears.
2541 .P
2542 Obviously, PCRE cannot support the interpolation of Perl code. Instead, it
2543 supports special syntax for recursion of the entire pattern, and also for
2544 individual subpattern recursion. After its introduction in PCRE and Python,
2545 this kind of recursion was subsequently introduced into Perl at release 5.10.
2546 .P
2547 A special item that consists of (? followed by a number greater than zero and a
2548 closing parenthesis is a recursive subroutine call of the subpattern of the
2549 given number, provided that it occurs inside that subpattern. (If not, it is a
2550 .\" HTML <a href="#subpatternsassubroutines">
2551 .\" </a>
2552 non-recursive subroutine
2553 .\"
2554 call, which is described in the next section.) The special item (?R) or (?0) is
2555 a recursive call of the entire regular expression.
2556 .P
2557 This PCRE pattern solves the nested parentheses problem (assume the
2558 PCRE_EXTENDED option is set so that white space is ignored):
2559 .sp
2560 \e( ( [^()]++ | (?R) )* \e)
2561 .sp
2562 First it matches an opening parenthesis. Then it matches any number of
2563 substrings which can either be a sequence of non-parentheses, or a recursive
2564 match of the pattern itself (that is, a correctly parenthesized substring).
2565 Finally there is a closing parenthesis. Note the use of a possessive quantifier
2566 to avoid backtracking into sequences of non-parentheses.
2567 .P
2568 If this were part of a larger pattern, you would not want to recurse the entire
2569 pattern, so instead you could use this:
2570 .sp
2571 ( \e( ( [^()]++ | (?1) )* \e) )
2572 .sp
2573 We have put the pattern into parentheses, and caused the recursion to refer to
2574 them instead of the whole pattern.
2575 .P
2576 In a larger pattern, keeping track of parenthesis numbers can be tricky. This
2577 is made easier by the use of relative references. Instead of (?1) in the
2578 pattern above you can write (?-2) to refer to the second most recently opened
2579 parentheses preceding the recursion. In other words, a negative number counts
2580 capturing parentheses leftwards from the point at which it is encountered.
2581 .P
2582 It is also possible to refer to subsequently opened parentheses, by writing
2583 references such as (?+2). However, these cannot be recursive because the
2584 reference is not inside the parentheses that are referenced. They are always
2585 .\" HTML <a href="#subpatternsassubroutines">
2586 .\" </a>
2587 non-recursive subroutine
2588 .\"
2589 calls, as described in the next section.
2590 .P
2591 An alternative approach is to use named parentheses instead. The Perl syntax
2592 for this is (?&name); PCRE's earlier syntax (?P>name) is also supported. We
2593 could rewrite the above example as follows:
2594 .sp
2595 (?<pn> \e( ( [^()]++ | (?&pn) )* \e) )
2596 .sp
2597 If there is more than one subpattern with the same name, the earliest one is
2598 used.
2599 .P
2600 This particular example pattern that we have been looking at contains nested
2601 unlimited repeats, and so the use of a possessive quantifier for matching
2602 strings of non-parentheses is important when applying the pattern to strings
2603 that do not match. For example, when this pattern is applied to
2604 .sp
2605 (aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa()
2606 .sp
2607 it yields "no match" quickly. However, if a possessive quantifier is not used,
2608 the match runs for a very long time indeed because there are so many different
2609 ways the + and * repeats can carve up the subject, and all have to be tested
2610 before failure can be reported.
2611 .P
2612 At the end of a match, the values of capturing parentheses are those from
2613 the outermost level. If you want to obtain intermediate values, a callout
2614 function can be used (see below and the
2615 .\" HREF
2616 \fBpcrecallout\fP
2617 .\"
2618 documentation). If the pattern above is matched against
2619 .sp
2620 (ab(cd)ef)
2621 .sp
2622 the value for the inner capturing parentheses (numbered 2) is "ef", which is
2623 the last value taken on at the top level. If a capturing subpattern is not
2624 matched at the top level, its final captured value is unset, even if it was
2625 (temporarily) set at a deeper level during the matching process.
2626 .P
2627 If there are more than 15 capturing parentheses in a pattern, PCRE has to
2628 obtain extra memory to store data during a recursion, which it does by using
2629 \fBpcre_malloc\fP, freeing it via \fBpcre_free\fP afterwards. If no memory can
2630 be obtained, the match fails with the PCRE_ERROR_NOMEMORY error.
2631 .P
2632 Do not confuse the (?R) item with the condition (R), which tests for recursion.
2633 Consider this pattern, which matches text in angle brackets, allowing for
2634 arbitrary nesting. Only digits are allowed in nested brackets (that is, when
2635 recursing), whereas any characters are permitted at the outer level.
2636 .sp
2637 < (?: (?(R) \ed++ | [^<>]*+) | (?R)) * >
2638 .sp
2639 In this pattern, (?(R) is the start of a conditional subpattern, with two
2640 different alternatives for the recursive and non-recursive cases. The (?R) item
2641 is the actual recursive call.
2642 .
2643 .
2644 .\" HTML <a name="recursiondifference"></a>
2645 .SS "Differences in recursion processing between PCRE and Perl"
2646 .rs
2647 .sp
2648 Recursion processing in PCRE differs from Perl in two important ways. In PCRE
2649 (like Python, but unlike Perl), a recursive subpattern call is always treated
2650 as an atomic group. That is, once it has matched some of the subject string, it
2651 is never re-entered, even if it contains untried alternatives and there is a
2652 subsequent matching failure. This can be illustrated by the following pattern,
2653 which purports to match a palindromic string that contains an odd number of
2654 characters (for example, "a", "aba", "abcba", "abcdcba"):
2655 .sp
2656 ^(.|(.)(?1)\e2)$
2657 .sp
2658 The idea is that it either matches a single character, or two identical
2659 characters surrounding a sub-palindrome. In Perl, this pattern works; in PCRE
2660 it does not if the pattern is longer than three characters. Consider the
2661 subject string "abcba":
2662 .P
2663 At the top level, the first character is matched, but as it is not at the end
2664 of the string, the first alternative fails; the second alternative is taken
2665 and the recursion kicks in. The recursive call to subpattern 1 successfully
2666 matches the next character ("b"). (Note that the beginning and end of line
2667 tests are not part of the recursion).
2668 .P
2669 Back at the top level, the next character ("c") is compared with what
2670 subpattern 2 matched, which was "a". This fails. Because the recursion is
2671 treated as an atomic group, there are now no backtracking points, and so the
2672 entire match fails. (Perl is able, at this point, to re-enter the recursion and
2673 try the second alternative.) However, if the pattern is written with the
2674 alternatives in the other order, things are different:
2675 .sp
2676 ^((.)(?1)\e2|.)$
2677 .sp
2678 This time, the recursing alternative is tried first, and continues to recurse
2679 until it runs out of characters, at which point the recursion fails. But this
2680 time we do have another alternative to try at the higher level. That is the big
2681 difference: in the previous case the remaining alternative is at a deeper
2682 recursion level, which PCRE cannot use.
2683 .P
2684 To change the pattern so that it matches all palindromic strings, not just
2685 those with an odd number of characters, it is tempting to change the pattern to
2686 this:
2687 .sp
2688 ^((.)(?1)\e2|.?)$
2689 .sp
2690 Again, this works in Perl, but not in PCRE, and for the same reason. When a
2691 deeper recursion has matched a single character, it cannot be entered again in
2692 order to match an empty string. The solution is to separate the two cases, and
2693 write out the odd and even cases as alternatives at the higher level:
2694 .sp
2695 ^(?:((.)(?1)\e2|)|((.)(?3)\e4|.))
2696 .sp
2697 If you want to match typical palindromic phrases, the pattern has to ignore all
2698 non-word characters, which can be done like this:
2699 .sp
2700 ^\eW*+(?:((.)\eW*+(?1)\eW*+\e2|)|((.)\eW*+(?3)\eW*+\e4|\eW*+.\eW*+))\eW*+$
2701 .sp
2702 If run with the PCRE_CASELESS option, this pattern matches phrases such as "A
2703 man, a plan, a canal: Panama!" and it works well in both PCRE and Perl. Note
2704 the use of the possessive quantifier *+ to avoid backtracking into sequences of
2705 non-word characters. Without this, PCRE takes a great deal longer (ten times or
2706 more) to match typical phrases, and Perl takes so long that you think it has
2707 gone into a loop.
2708 .P
2709 \fBWARNING\fP: The palindrome-matching patterns above work only if the subject
2710 string does not start with a palindrome that is shorter than the entire string.
2711 For example, although "abcba" is correctly matched, if the subject is "ababa",
2712 PCRE finds the palindrome "aba" at the start, then fails at top level because
2713 the end of the string does not follow. Once again, it cannot jump back into the
2714 recursion to try other alternatives, so the entire match fails.
2715 .P
2716 The second way in which PCRE and Perl differ in their recursion processing is
2717 in the handling of captured values. In Perl, when a subpattern is called
2718 recursively or as a subpattern (see the next section), it has no access to any
2719 values that were captured outside the recursion, whereas in PCRE these values
2720 can be referenced. Consider this pattern:
2721 .sp
2722 ^(.)(\e1|a(?2))
2723 .sp
2724 In PCRE, this pattern matches "bab". The first capturing parentheses match "b",
2725 then in the second group, when the back reference \e1 fails to match "b", the
2726 second alternative matches "a" and then recurses. In the recursion, \e1 does
2727 now match "b" and so the whole match succeeds. In Perl, the pattern fails to
2728 match because inside the recursive call \e1 cannot access the externally set
2729 value.
2730 .
2731 .
2732 .\" HTML <a name="subpatternsassubroutines"></a>
2734 .rs
2735 .sp
2736 If the syntax for a recursive subpattern call (either by number or by
2737 name) is used outside the parentheses to which it refers, it operates like a
2738 subroutine in a programming language. The called subpattern may be defined
2739 before or after the reference. A numbered reference can be absolute or
2740 relative, as in these examples:
2741 .sp
2742 (...(absolute)...)...(?2)...
2743 (...(relative)...)...(?-1)...
2744 (...(?+1)...(relative)...
2745 .sp
2746 An earlier example pointed out that the pattern
2747 .sp
2748 (sens|respons)e and \e1ibility
2749 .sp
2750 matches "sense and sensibility" and "response and responsibility", but not
2751 "sense and responsibility". If instead the pattern
2752 .sp
2753 (sens|respons)e and (?1)ibility
2754 .sp
2755 is used, it does match "sense and responsibility" as well as the other two
2756 strings. Another example is given in the discussion of DEFINE above.
2757 .P
2758 All subroutine calls, whether recursive or not, are always treated as atomic
2759 groups. That is, once a subroutine has matched some of the subject string, it
2760 is never re-entered, even if it contains untried alternatives and there is a
2761 subsequent matching failure. Any capturing parentheses that are set during the
2762 subroutine call revert to their previous values afterwards.
2763 .P
2764 Processing options such as case-independence are fixed when a subpattern is
2765 defined, so if it is used as a subroutine, such options cannot be changed for
2766 different calls. For example, consider this pattern:
2767 .sp
2768 (abc)(?i:(?-1))
2769 .sp
2770 It matches "abcabc". It does not match "abcABC" because the change of
2771 processing option does not affect the called subpattern.
2772 .
2773 .
2774 .\" HTML <a name="onigurumasubroutines"></a>
2776 .rs
2777 .sp
2778 For compatibility with Oniguruma, the non-Perl syntax \eg followed by a name or
2779 a number enclosed either in angle brackets or single quotes, is an alternative
2780 syntax for referencing a subpattern as a subroutine, possibly recursively. Here
2781 are two of the examples used above, rewritten using this syntax:
2782 .sp
2783 (?<pn> \e( ( (?>[^()]+) | \eg<pn> )* \e) )
2784 (sens|respons)e and \eg'1'ibility
2785 .sp
2786 PCRE supports an extension to Oniguruma: if a number is preceded by a
2787 plus or a minus sign it is taken as a relative reference. For example:
2788 .sp
2789 (abc)(?i:\eg<-1>)
2790 .sp
2791 Note that \eg{...} (Perl syntax) and \eg<...> (Oniguruma syntax) are \fInot\fP
2792 synonymous. The former is a back reference; the latter is a subroutine call.
2793 .
2794 .
2796 .rs
2797 .sp
2798 Perl has a feature whereby using the sequence (?{...}) causes arbitrary Perl
2799 code to be obeyed in the middle of matching a regular expression. This makes it
2800 possible, amongst other things, to extract different substrings that match the
2801 same pair of parentheses when there is a repetition.
2802 .P
2803 PCRE provides a similar feature, but of course it cannot obey arbitrary Perl
2804 code. The feature is called "callout". The caller of PCRE provides an external
2805 function by putting its entry point in the global variable \fIpcre_callout\fP
2806 (8-bit library) or \fIpcre[16|32]_callout\fP (16-bit or 32-bit library).
2807 By default, this variable contains NULL, which disables all calling out.
2808 .P
2809 Within a regular expression, (?C) indicates the points at which the external
2810 function is to be called. If you want to identify different callout points, you
2811 can put a number less than 256 after the letter C. The default value is zero.
2812 For example, this pattern has two callout points:
2813 .sp
2814 (?C1)abc(?C2)def
2815 .sp
2816 If the PCRE_AUTO_CALLOUT flag is passed to a compiling function, callouts are
2817 automatically installed before each item in the pattern. They are all numbered
2818 255. If there is a conditional group in the pattern whose condition is an
2819 assertion, an additional callout is inserted just before the condition. An
2820 explicit callout may also be set at this position, as in this example:
2821 .sp
2822 (?(?C9)(?=a)abc|def)
2823 .sp
2824 Note that this applies only to assertion conditions, not to other types of
2825 condition.
2826 .P
2827 During matching, when PCRE reaches a callout point, the external function is
2828 called. It is provided with the number of the callout, the position in the
2829 pattern, and, optionally, one item of data originally supplied by the caller of
2830 the matching function. The callout function may cause matching to proceed, to
2831 backtrack, or to fail altogether.
2832 .P
2833 By default, PCRE implements a number of optimizations at compile time and
2834 matching time, and one side-effect is that sometimes callouts are skipped. If
2835 you need all possible callouts to happen, you need to set options that disable
2836 the relevant optimizations. More details, and a complete description of the
2837 interface to the callout function, are given in the
2838 .\" HREF
2839 \fBpcrecallout\fP
2840 .\"
2841 documentation.
2842 .
2843 .
2844 .\" HTML <a name="backtrackcontrol"></a>
2846 .rs
2847 .sp
2848 Perl 5.10 introduced a number of "Special Backtracking Control Verbs", which
2849 are still described in the Perl documentation as "experimental and subject to
2850 change or removal in a future version of Perl". It goes on to say: "Their usage
2851 in production code should be noted to avoid problems during upgrades." The same
2852 remarks apply to the PCRE features described in this section.
2853 .P
2854 The new verbs make use of what was previously invalid syntax: an opening
2855 parenthesis followed by an asterisk. They are generally of the form
2856 (*VERB) or (*VERB:NAME). Some may take either form, possibly behaving
2857 differently depending on whether or not a name is present. A name is any
2858 sequence of characters that does not include a closing parenthesis. The maximum
2859 length of name is 255 in the 8-bit library and 65535 in the 16-bit and 32-bit
2860 libraries. If the name is empty, that is, if the closing parenthesis
2861 immediately follows the colon, the effect is as if the colon were not there.
2862 Any number of these verbs may occur in a pattern.
2863 .P
2864 Since these verbs are specifically related to backtracking, most of them can be
2865 used only when the pattern is to be matched using one of the traditional
2866 matching functions, because these use a backtracking algorithm. With the
2867 exception of (*FAIL), which behaves like a failing negative assertion, the
2868 backtracking control verbs cause an error if encountered by a DFA matching
2869 function.
2870 .P
2871 The behaviour of these verbs in
2872 .\" HTML <a href="#btrepeat">
2873 .\" </a>
2874 repeated groups,
2875 .\"
2876 .\" HTML <a href="#btassert">
2877 .\" </a>
2878 assertions,
2879 .\"
2880 and in
2881 .\" HTML <a href="#btsub">
2882 .\" </a>
2883 subpatterns called as subroutines
2884 .\"
2885 (whether or not recursively) is documented below.
2886 .
2887 .
2888 .\" HTML <a name="nooptimize"></a>
2889 .SS "Optimizations that affect backtracking verbs"
2890 .rs
2891 .sp
2892 PCRE contains some optimizations that are used to speed up matching by running
2893 some checks at the start of each match attempt. For example, it may know the
2894 minimum length of matching subject, or that a particular character must be
2895 present. When one of these optimizations bypasses the running of a match, any
2896 included backtracking verbs will not, of course, be processed. You can suppress
2897 the start-of-match optimizations by setting the PCRE_NO_START_OPTIMIZE option
2898 when calling \fBpcre_compile()\fP or \fBpcre_exec()\fP, or by starting the
2899 pattern with (*NO_START_OPT). There is more discussion of this option in the
2900 section entitled
2901 .\" HTML <a href="pcreapi.html#execoptions">
2902 .\" </a>
2903 "Option bits for \fBpcre_exec()\fP"
2904 .\"
2905 in the
2906 .\" HREF
2907 \fBpcreapi\fP
2908 .\"
2909 documentation.
2910 .P
2911 Experiments with Perl suggest that it too has similar optimizations, sometimes
2912 leading to anomalous results.
2913 .
2914 .
2915 .SS "Verbs that act immediately"
2916 .rs
2917 .sp
2918 The following verbs act as soon as they are encountered. They may not be
2919 followed by a name.
2920 .sp
2921 (*ACCEPT)
2922 .sp
2923 This verb causes the match to end successfully, skipping the remainder of the
2924 pattern. However, when it is inside a subpattern that is called as a
2925 subroutine, only that subpattern is ended successfully. Matching then continues
2926 at the outer level. If (*ACCEPT) in triggered in a positive assertion, the
2927 assertion succeeds; in a negative assertion, the assertion fails.
2928 .P
2929 If (*ACCEPT) is inside capturing parentheses, the data so far is captured. For
2930 example:
2931 .sp
2932 A((?:A|B(*ACCEPT)|C)D)
2933 .sp
2934 This matches "AB", "AAD", or "ACD"; when it matches "AB", "B" is captured by
2935 the outer parentheses.
2936 .sp
2937 (*FAIL) or (*F)
2938 .sp
2939 This verb causes a matching failure, forcing backtracking to occur. It is
2940 equivalent to (?!) but easier to read. The Perl documentation notes that it is
2941 probably useful only when combined with (?{}) or (??{}). Those are, of course,
2942 Perl features that are not present in PCRE. The nearest equivalent is the
2943 callout feature, as for example in this pattern:
2944 .sp
2945 a+(?C)(*FAIL)
2946 .sp
2947 A match with the string "aaaa" always fails, but the callout is taken before
2948 each backtrack happens (in this example, 10 times).
2949 .
2950 .
2951 .SS "Recording which path was taken"
2952 .rs
2953 .sp
2954 There is one verb whose main purpose is to track how a match was arrived at,
2955 though it also has a secondary use in conjunction with advancing the match
2956 starting point (see (*SKIP) below).
2957 .sp
2958 (*MARK:NAME) or (*:NAME)
2959 .sp
2960 A name is always required with this verb. There may be as many instances of
2961 (*MARK) as you like in a pattern, and their names do not have to be unique.
2962 .P
2963 When a match succeeds, the name of the last-encountered (*MARK:NAME),
2964 (*PRUNE:NAME), or (*THEN:NAME) on the matching path is passed back to the
2965 caller as described in the section entitled
2966 .\" HTML <a href="pcreapi.html#extradata">
2967 .\" </a>
2968 "Extra data for \fBpcre_exec()\fP"
2969 .\"
2970 in the
2971 .\" HREF
2972 \fBpcreapi\fP
2973 .\"
2974 documentation. Here is an example of \fBpcretest\fP output, where the /K
2975 modifier requests the retrieval and outputting of (*MARK) data:
2976 .sp
2977 re> /X(*MARK:A)Y|X(*MARK:B)Z/K
2978 data> XY
2979 0: XY
2980 MK: A
2981 XZ
2982 0: XZ
2983 MK: B
2984 .sp
2985 The (*MARK) name is tagged with "MK:" in this output, and in this example it
2986 indicates which of the two alternatives matched. This is a more efficient way
2987 of obtaining this information than putting each alternative in its own
2988 capturing parentheses.
2989 .P
2990 If a verb with a name is encountered in a positive assertion that is true, the
2991 name is recorded and passed back if it is the last-encountered. This does not
2992 happen for negative assertions or failing positive assertions.
2993 .P
2994 After a partial match or a failed match, the last encountered name in the
2995 entire match process is returned. For example:
2996 .sp
2997 re> /X(*MARK:A)Y|X(*MARK:B)Z/K
2998 data> XP
2999 No match, mark = B
3000 .sp
3001 Note that in this unanchored example the mark is retained from the match
3002 attempt that started at the letter "X" in the subject. Subsequent match
3003 attempts starting at "P" and then with an empty string do not get as far as the
3004 (*MARK) item, but nevertheless do not reset it.
3005 .P
3006 If you are interested in (*MARK) values after failed matches, you should
3007 probably set the PCRE_NO_START_OPTIMIZE option
3008 .\" HTML <a href="#nooptimize">
3009 .\" </a>
3010 (see above)
3011 .\"
3012 to ensure that the match is always attempted.
3013 .
3014 .
3015 .SS "Verbs that act after backtracking"
3016 .rs
3017 .sp
3018 The following verbs do nothing when they are encountered. Matching continues
3019 with what follows, but if there is no subsequent match, causing a backtrack to
3020 the verb, a failure is forced. That is, backtracking cannot pass to the left of
3021 the verb. However, when one of these verbs appears inside an atomic group or an
3022 assertion that is true, its effect is confined to that group, because once the
3023 group has been matched, there is never any backtracking into it. In this
3024 situation, backtracking can "jump back" to the left of the entire atomic group
3025 or assertion. (Remember also, as stated above, that this localization also
3026 applies in subroutine calls.)
3027 .P
3028 These verbs differ in exactly what kind of failure occurs when backtracking
3029 reaches them. The behaviour described below is what happens when the verb is
3030 not in a subroutine or an assertion. Subsequent sections cover these special
3031 cases.
3032 .sp
3033 (*COMMIT)
3034 .sp
3035 This verb, which may not be followed by a name, causes the whole match to fail
3036 outright if there is a later matching failure that causes backtracking to reach
3037 it. Even if the pattern is unanchored, no further attempts to find a match by
3038 advancing the starting point take place. If (*COMMIT) is the only backtracking
3039 verb that is encountered, once it has been passed \fBpcre_exec()\fP is
3040 committed to finding a match at the current starting point, or not at all. For
3041 example:
3042 .sp
3043 a+(*COMMIT)b
3044 .sp
3045 This matches "xxaab" but not "aacaab". It can be thought of as a kind of
3046 dynamic anchor, or "I've started, so I must finish." The name of the most
3047 recently passed (*MARK) in the path is passed back when (*COMMIT) forces a
3048 match failure.
3049 .P
3050 If there is more than one backtracking verb in a pattern, a different one that
3051 follows (*COMMIT) may be triggered first, so merely passing (*COMMIT) during a
3052 match does not always guarantee that a match must be at this starting point.
3053 .P
3054 Note that (*COMMIT) at the start of a pattern is not the same as an anchor,
3055 unless PCRE's start-of-match optimizations are turned off, as shown in this
3056 output from \fBpcretest\fP:
3057 .sp
3058 re> /(*COMMIT)abc/
3059 data> xyzabc
3060 0: abc
3061 data> xyzabc\eY
3062 No match
3063 .sp
3064 For this pattern, PCRE knows that any match must start with "a", so the
3065 optimization skips along the subject to "a" before applying the pattern to the
3066 first set of data. The match attempt then succeeds. In the second set of data,
3067 the escape sequence \eY is interpreted by the \fBpcretest\fP program. It causes
3068 the PCRE_NO_START_OPTIMIZE option to be set when \fBpcre_exec()\fP is called.
3069 This disables the optimization that skips along to the first character. The
3070 pattern is now applied starting at "x", and so the (*COMMIT) causes the match
3071 to fail without trying any other starting points.
3072 .sp
3073 (*PRUNE) or (*PRUNE:NAME)
3074 .sp
3075 This verb causes the match to fail at the current starting position in the
3076 subject if there is a later matching failure that causes backtracking to reach
3077 it. If the pattern is unanchored, the normal "bumpalong" advance to the next
3078 starting character then happens. Backtracking can occur as usual to the left of
3079 (*PRUNE), before it is reached, or when matching to the right of (*PRUNE), but
3080 if there is no match to the right, backtracking cannot cross (*PRUNE). In
3081 simple cases, the use of (*PRUNE) is just an alternative to an atomic group or
3082 possessive quantifier, but there are some uses of (*PRUNE) that cannot be
3083 expressed in any other way. In an anchored pattern (*PRUNE) has the same effect
3084 as (*COMMIT).
3085 .P
3086 The behaviour of (*PRUNE:NAME) is the not the same as (*MARK:NAME)(*PRUNE).
3087 It is like (*MARK:NAME) in that the name is remembered for passing back to the
3088 caller. However, (*SKIP:NAME) searches only for names set with (*MARK).
3089 .sp
3090 (*SKIP)
3091 .sp
3092 This verb, when given without a name, is like (*PRUNE), except that if the
3093 pattern is unanchored, the "bumpalong" advance is not to the next character,
3094 but to the position in the subject where (*SKIP) was encountered. (*SKIP)
3095 signifies that whatever text was matched leading up to it cannot be part of a
3096 successful match. Consider:
3097 .sp
3098 a+(*SKIP)b
3099 .sp
3100 If the subject is "aaaac...", after the first match attempt fails (starting at
3101 the first character in the string), the starting point skips on to start the
3102 next attempt at "c". Note that a possessive quantifer does not have the same
3103 effect as this example; although it would suppress backtracking during the
3104 first match attempt, the second attempt would start at the second character
3105 instead of skipping on to "c".
3106 .sp
3107 (*SKIP:NAME)
3108 .sp
3109 When (*SKIP) has an associated name, its behaviour is modified. When it is
3110 triggered, the previous path through the pattern is searched for the most
3111 recent (*MARK) that has the same name. If one is found, the "bumpalong" advance
3112 is to the subject position that corresponds to that (*MARK) instead of to where
3113 (*SKIP) was encountered. If no (*MARK) with a matching name is found, the
3114 (*SKIP) is ignored.
3115 .P
3116 Note that (*SKIP:NAME) searches only for names set by (*MARK:NAME). It ignores
3117 names that are set by (*PRUNE:NAME) or (*THEN:NAME).
3118 .sp
3119 (*THEN) or (*THEN:NAME)
3120 .sp
3121 This verb causes a skip to the next innermost alternative when backtracking
3122 reaches it. That is, it cancels any further backtracking within the current
3123 alternative. Its name comes from the observation that it can be used for a
3124 pattern-based if-then-else block:
3125 .sp
3126 ( COND1 (*THEN) FOO | COND2 (*THEN) BAR | COND3 (*THEN) BAZ ) ...
3127 .sp
3128 If the COND1 pattern matches, FOO is tried (and possibly further items after
3129 the end of the group if FOO succeeds); on failure, the matcher skips to the
3130 second alternative and tries COND2, without backtracking into COND1. If that
3131 succeeds and BAR fails, COND3 is tried. If subsequently BAZ fails, there are no
3132 more alternatives, so there is a backtrack to whatever came before the entire
3133 group. If (*THEN) is not inside an alternation, it acts like (*PRUNE).
3134 .P
3135 The behaviour of (*THEN:NAME) is the not the same as (*MARK:NAME)(*THEN).
3136 It is like (*MARK:NAME) in that the name is remembered for passing back to the
3137 caller. However, (*SKIP:NAME) searches only for names set with (*MARK).
3138 .P
3139 A subpattern that does not contain a | character is just a part of the
3140 enclosing alternative; it is not a nested alternation with only one
3141 alternative. The effect of (*THEN) extends beyond such a subpattern to the
3142 enclosing alternative. Consider this pattern, where A, B, etc. are complex
3143 pattern fragments that do not contain any | characters at this level:
3144 .sp
3145 A (B(*THEN)C) | D
3146 .sp
3147 If A and B are matched, but there is a failure in C, matching does not
3148 backtrack into A; instead it moves to the next alternative, that is, D.
3149 However, if the subpattern containing (*THEN) is given an alternative, it
3150 behaves differently:
3151 .sp
3152 A (B(*THEN)C | (*FAIL)) | D
3153 .sp
3154 The effect of (*THEN) is now confined to the inner subpattern. After a failure
3155 in C, matching moves to (*FAIL), which causes the whole subpattern to fail
3156 because there are no more alternatives to try. In this case, matching does now
3157 backtrack into A.
3158 .P
3159 Note that a conditional subpattern is not considered as having two
3160 alternatives, because only one is ever used. In other words, the | character in
3161 a conditional subpattern has a different meaning. Ignoring white space,
3162 consider:
3163 .sp
3164 ^.*? (?(?=a) a | b(*THEN)c )
3165 .sp
3166 If the subject is "ba", this pattern does not match. Because .*? is ungreedy,
3167 it initially matches zero characters. The condition (?=a) then fails, the
3168 character "b" is matched, but "c" is not. At this point, matching does not
3169 backtrack to .*? as might perhaps be expected from the presence of the |
3170 character. The conditional subpattern is part of the single alternative that
3171 comprises the whole pattern, and so the match fails. (If there was a backtrack
3172 into .*?, allowing it to match "b", the match would succeed.)
3173 .P
3174 The verbs just described provide four different "strengths" of control when
3175 subsequent matching fails. (*THEN) is the weakest, carrying on the match at the
3176 next alternative. (*PRUNE) comes next, failing the match at the current
3177 starting position, but allowing an advance to the next character (for an
3178 unanchored pattern). (*SKIP) is similar, except that the advance may be more
3179 than one character. (*COMMIT) is the strongest, causing the entire match to
3180 fail.
3181 .
3182 .
3183 .SS "More than one backtracking verb"
3184 .rs
3185 .sp
3186 If more than one backtracking verb is present in a pattern, the one that is
3187 backtracked onto first acts. For example, consider this pattern, where A, B,
3188 etc. are complex pattern fragments:
3189 .sp
3191 .sp
3192 If A matches but B fails, the backtrack to (*COMMIT) causes the entire match to
3193 fail. However, if A and B match, but C fails, the backtrack to (*THEN) causes
3194 the next alternative (ABD) to be tried. This behaviour is consistent, but is
3195 not always the same as Perl's. It means that if two or more backtracking verbs
3196 appear in succession, all the the last of them has no effect. Consider this
3197 example:
3198 .sp
3199 ...(*COMMIT)(*PRUNE)...
3200 .sp
3201 If there is a matching failure to the right, backtracking onto (*PRUNE) causes
3202 it to be triggered, and its action is taken. There can never be a backtrack
3203 onto (*COMMIT).
3204 .
3205 .
3206 .\" HTML <a name="btrepeat"></a>
3207 .SS "Backtracking verbs in repeated groups"
3208 .rs
3209 .sp
3210 PCRE differs from Perl in its handling of backtracking verbs in repeated
3211 groups. For example, consider:
3212 .sp
3213 /(a(*COMMIT)b)+ac/
3214 .sp
3215 If the subject is "abac", Perl matches, but PCRE fails because the (*COMMIT) in
3216 the second repeat of the group acts.
3217 .
3218 .
3219 .\" HTML <a name="btassert"></a>
3220 .SS "Backtracking verbs in assertions"
3221 .rs
3222 .sp
3223 (*FAIL) in an assertion has its normal effect: it forces an immediate backtrack.
3224 .P
3225 (*ACCEPT) in a positive assertion causes the assertion to succeed without any
3226 further processing. In a negative assertion, (*ACCEPT) causes the assertion to
3227 fail without any further processing.
3228 .P
3229 The other backtracking verbs are not treated specially if they appear in a
3230 positive assertion. In particular, (*THEN) skips to the next alternative in the
3231 innermost enclosing group that has alternations, whether or not this is within
3232 the assertion.
3233 .P
3234 Negative assertions are, however, different, in order to ensure that changing a
3235 positive assertion into a negative assertion changes its result. Backtracking
3236 into (*COMMIT), (*SKIP), or (*PRUNE) causes a negative assertion to be true,
3237 without considering any further alternative branches in the assertion.
3238 Backtracking into (*THEN) causes it to skip to the next enclosing alternative
3239 within the assertion (the normal behaviour), but if the assertion does not have
3240 such an alternative, (*THEN) behaves like (*PRUNE).
3241 .
3242 .
3243 .\" HTML <a name="btsub"></a>
3244 .SS "Backtracking verbs in subroutines"
3245 .rs
3246 .sp
3247 These behaviours occur whether or not the subpattern is called recursively.
3248 Perl's treatment of subroutines is different in some cases.
3249 .P
3250 (*FAIL) in a subpattern called as a subroutine has its normal effect: it forces
3251 an immediate backtrack.
3252 .P
3253 (*ACCEPT) in a subpattern called as a subroutine causes the subroutine match to
3254 succeed without any further processing. Matching then continues after the
3255 subroutine call.
3256 .P
3257 (*COMMIT), (*SKIP), and (*PRUNE) in a subpattern called as a subroutine cause
3258 the subroutine match to fail.
3259 .P
3260 (*THEN) skips to the next alternative in the innermost enclosing group within
3261 the subpattern that has alternatives. If there is no such group within the
3262 subpattern, (*THEN) causes the subroutine match to fail.
3263 .
3264 .
3265 .SH "SEE ALSO"
3266 .rs
3267 .sp
3268 \fBpcreapi\fP(3), \fBpcrecallout\fP(3), \fBpcrematching\fP(3),
3269 \fBpcresyntax\fP(3), \fBpcre\fP(3), \fBpcre16(3)\fP, \fBpcre32(3)\fP.
3270 .
3271 .
3273 .rs
3274 .sp
3275 .nf
3276 Philip Hazel
3277 University Computing Service
3278 Cambridge CB2 3QH, England.
3279 .fi
3280 .
3281 .
3283 .rs
3284 .sp
3285 .nf
3286 Last updated: 08 January 2014
3287 Copyright (c) 1997-2014 University of Cambridge.
3288 .fi


Name Value
svn:eol-style native
svn:keywords "Author Date Id Revision Url"

  ViewVC Help
Powered by ViewVC 1.1.5