/[pcre]/code/trunk/doc/pcrepattern.3
ViewVC logotype

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

Parent Directory Parent Directory | Revision Log Revision Log


Revision 903 - (show annotations)
Sat Jan 21 16:37:17 2012 UTC (7 years, 6 months ago) by ph10
File size: 120090 byte(s)
Source file tidies for 8.30-RC1 release; fix Makefile.am bugs for building 
symbolic links to man pages.
1 .TH PCREPATTERN 3
2 .SH NAME
3 PCRE - Perl-compatible regular expressions
4 .SH "PCRE REGULAR EXPRESSION DETAILS"
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 The original operation of PCRE was on strings of one-byte characters. However,
24 there is now also support for UTF-8 strings in the original library, and a
25 second library that supports 16-bit and UTF-16 character strings. To use these
26 features, PCRE must be built to include appropriate support. When using UTF
27 strings you must either call the compiling function with the PCRE_UTF8 or
28 PCRE_UTF16 option, or the pattern must start with one of these special
29 sequences:
30 .sp
31 (*UTF8)
32 (*UTF16)
33 .sp
34 Starting a pattern with such a sequence is equivalent to setting the relevant
35 option. This feature is not Perl-compatible. How setting a UTF mode affects
36 pattern matching is mentioned in several places below. There is also a summary
37 of features in the
38 .\" HREF
39 \fBpcreunicode\fP
40 .\"
41 page.
42 .P
43 Another special sequence that may appear at the start of a pattern or in
44 combination with (*UTF8) or (*UTF16) is:
45 .sp
46 (*UCP)
47 .sp
48 This has the same effect as setting the PCRE_UCP option: it causes sequences
49 such as \ed and \ew to use Unicode properties to determine character types,
50 instead of recognizing only characters with codes less than 128 via a lookup
51 table.
52 .P
53 If a pattern starts with (*NO_START_OPT), it has the same effect as setting the
54 PCRE_NO_START_OPTIMIZE option either at compile or matching time. There are
55 also some more of these special sequences that are concerned with the handling
56 of newlines; they are described below.
57 .P
58 The remainder of this document discusses the patterns that are supported by
59 PCRE when one its main matching functions, \fBpcre_exec()\fP (8-bit) or
60 \fBpcre16_exec()\fP (16-bit), is used. PCRE also has alternative matching
61 functions, \fBpcre_dfa_exec()\fP and \fBpcre16_dfa_exec()\fP, which match using
62 a different algorithm that is not Perl-compatible. Some of the features
63 discussed below are not available when DFA matching is used. The advantages and
64 disadvantages of the alternative functions, and how they differ from the normal
65 functions, are discussed in the
66 .\" HREF
67 \fBpcrematching\fP
68 .\"
69 page.
70 .
71 .
72 .\" HTML <a name="newlines"></a>
73 .SH "NEWLINE CONVENTIONS"
74 .rs
75 .sp
76 PCRE supports five different conventions for indicating line breaks in
77 strings: a single CR (carriage return) character, a single LF (linefeed)
78 character, the two-character sequence CRLF, any of the three preceding, or any
79 Unicode newline sequence. The
80 .\" HREF
81 \fBpcreapi\fP
82 .\"
83 page has
84 .\" HTML <a href="pcreapi.html#newlines">
85 .\" </a>
86 further discussion
87 .\"
88 about newlines, and shows how to set the newline convention in the
89 \fIoptions\fP arguments for the compiling and matching functions.
90 .P
91 It is also possible to specify a newline convention by starting a pattern
92 string with one of the following five sequences:
93 .sp
94 (*CR) carriage return
95 (*LF) linefeed
96 (*CRLF) carriage return, followed by linefeed
97 (*ANYCRLF) any of the three above
98 (*ANY) all Unicode newline sequences
99 .sp
100 These override the default and the options given to the compiling function. For
101 example, on a Unix system where LF is the default newline sequence, the pattern
102 .sp
103 (*CR)a.b
104 .sp
105 changes the convention to CR. That pattern matches "a\enb" because LF is no
106 longer a newline. Note that these special settings, which are not
107 Perl-compatible, are recognized only at the very start of a pattern, and that
108 they must be in upper case. If more than one of them is present, the last one
109 is used.
110 .P
111 The newline convention affects the interpretation of the dot metacharacter when
112 PCRE_DOTALL is not set, and also the behaviour of \eN. However, it does not
113 affect what the \eR escape sequence matches. By default, this is any Unicode
114 newline sequence, for Perl compatibility. However, this can be changed; see the
115 description of \eR in the section entitled
116 .\" HTML <a href="#newlineseq">
117 .\" </a>
118 "Newline sequences"
119 .\"
120 below. A change of \eR setting can be combined with a change of newline
121 convention.
122 .
123 .
124 .SH "CHARACTERS AND METACHARACTERS"
125 .rs
126 .sp
127 A regular expression is a pattern that is matched against a subject string from
128 left to right. Most characters stand for themselves in a pattern, and match the
129 corresponding characters in the subject. As a trivial example, the pattern
130 .sp
131 The quick brown fox
132 .sp
133 matches a portion of a subject string that is identical to itself. When
134 caseless matching is specified (the PCRE_CASELESS option), letters are matched
135 independently of case. In a UTF mode, PCRE always understands the concept of
136 case for characters whose values are less than 128, so caseless matching is
137 always possible. For characters with higher values, the concept of case is
138 supported if PCRE is compiled with Unicode property support, but not otherwise.
139 If you want to use caseless matching for characters 128 and above, you must
140 ensure that PCRE is compiled with Unicode property support as well as with
141 UTF support.
142 .P
143 The power of regular expressions comes from the ability to include alternatives
144 and repetitions in the pattern. These are encoded in the pattern by the use of
145 \fImetacharacters\fP, which do not stand for themselves but instead are
146 interpreted in some special way.
147 .P
148 There are two different sets of metacharacters: those that are recognized
149 anywhere in the pattern except within square brackets, and those that are
150 recognized within square brackets. Outside square brackets, the metacharacters
151 are as follows:
152 .sp
153 \e general escape character with several uses
154 ^ assert start of string (or line, in multiline mode)
155 $ assert end of string (or line, in multiline mode)
156 . match any character except newline (by default)
157 [ start character class definition
158 | start of alternative branch
159 ( start subpattern
160 ) end subpattern
161 ? extends the meaning of (
162 also 0 or 1 quantifier
163 also quantifier minimizer
164 * 0 or more quantifier
165 + 1 or more quantifier
166 also "possessive quantifier"
167 { start min/max quantifier
168 .sp
169 Part of a pattern that is in square brackets is called a "character class". In
170 a character class the only metacharacters are:
171 .sp
172 \e general escape character
173 ^ negate the class, but only if the first character
174 - indicates character range
175 .\" JOIN
176 [ POSIX character class (only if followed by POSIX
177 syntax)
178 ] terminates the character class
179 .sp
180 The following sections describe the use of each of the metacharacters.
181 .
182 .
183 .SH BACKSLASH
184 .rs
185 .sp
186 The backslash character has several uses. Firstly, if it is followed by a
187 character that is not a number or a letter, it takes away any special meaning
188 that character may have. This use of backslash as an escape character applies
189 both inside and outside character classes.
190 .P
191 For example, if you want to match a * character, you write \e* in the pattern.
192 This escaping action applies whether or not the following character would
193 otherwise be interpreted as a metacharacter, so it is always safe to precede a
194 non-alphanumeric with backslash to specify that it stands for itself. In
195 particular, if you want to match a backslash, you write \e\e.
196 .P
197 In a UTF mode, only ASCII numbers and letters have any special meaning after a
198 backslash. All other characters (in particular, those whose codepoints are
199 greater than 127) are treated as literals.
200 .P
201 If a pattern is compiled with the PCRE_EXTENDED option, whitespace in the
202 pattern (other than in a character class) and characters between a # outside
203 a character class and the next newline are ignored. An escaping backslash can
204 be used to include a whitespace or # character as part of the pattern.
205 .P
206 If you want to remove the special meaning from a sequence of characters, you
207 can do so by putting them between \eQ and \eE. This is different from Perl in
208 that $ and @ are handled as literals in \eQ...\eE sequences in PCRE, whereas in
209 Perl, $ and @ cause variable interpolation. Note the following examples:
210 .sp
211 Pattern PCRE matches Perl matches
212 .sp
213 .\" JOIN
214 \eQabc$xyz\eE abc$xyz abc followed by the
215 contents of $xyz
216 \eQabc\e$xyz\eE abc\e$xyz abc\e$xyz
217 \eQabc\eE\e$\eQxyz\eE abc$xyz abc$xyz
218 .sp
219 The \eQ...\eE sequence is recognized both inside and outside character classes.
220 An isolated \eE that is not preceded by \eQ is ignored. If \eQ is not followed
221 by \eE later in the pattern, the literal interpretation continues to the end of
222 the pattern (that is, \eE is assumed at the end). If the isolated \eQ is inside
223 a character class, this causes an error, because the character class is not
224 terminated.
225 .
226 .
227 .\" HTML <a name="digitsafterbackslash"></a>
228 .SS "Non-printing characters"
229 .rs
230 .sp
231 A second use of backslash provides a way of encoding non-printing characters
232 in patterns in a visible manner. There is no restriction on the appearance of
233 non-printing characters, apart from the binary zero that terminates a pattern,
234 but when a pattern is being prepared by text editing, it is often easier to use
235 one of the following escape sequences than the binary character it represents:
236 .sp
237 \ea alarm, that is, the BEL character (hex 07)
238 \ecx "control-x", where x is any ASCII character
239 \ee escape (hex 1B)
240 \ef formfeed (hex 0C)
241 \en linefeed (hex 0A)
242 \er carriage return (hex 0D)
243 \et tab (hex 09)
244 \eddd character with octal code ddd, or back reference
245 \exhh character with hex code hh
246 \ex{hhh..} character with hex code hhh.. (non-JavaScript mode)
247 \euhhhh character with hex code hhhh (JavaScript mode only)
248 .sp
249 The precise effect of \ecx is as follows: if x is a lower case letter, it
250 is converted to upper case. Then bit 6 of the character (hex 40) is inverted.
251 Thus \ecz becomes hex 1A (z is 7A), but \ec{ becomes hex 3B ({ is 7B), while
252 \ec; becomes hex 7B (; is 3B). If the byte following \ec has a value greater
253 than 127, a compile-time error occurs. This locks out non-ASCII characters in
254 all modes. (When PCRE is compiled in EBCDIC mode, all byte values are valid. A
255 lower case letter is converted to upper case, and then the 0xc0 bits are
256 flipped.)
257 .P
258 By default, after \ex, from zero to two hexadecimal digits are read (letters
259 can be in upper or lower case). Any number of hexadecimal digits may appear
260 between \ex{ and }, but the character code is constrained as follows:
261 .sp
262 8-bit non-UTF mode less than 0x100
263 8-bit UTF-8 mode less than 0x10ffff and a valid codepoint
264 16-bit non-UTF mode less than 0x10000
265 16-bit UTF-16 mode less than 0x10ffff and a valid codepoint
266 .sp
267 Invalid Unicode codepoints are the range 0xd800 to 0xdfff (the so-called
268 "surrogate" codepoints).
269 .P
270 If characters other than hexadecimal digits appear between \ex{ and }, or if
271 there is no terminating }, this form of escape is not recognized. Instead, the
272 initial \ex will be interpreted as a basic hexadecimal escape, with no
273 following digits, giving a character whose value is zero.
274 .P
275 If the PCRE_JAVASCRIPT_COMPAT option is set, the interpretation of \ex is
276 as just described only when it is followed by two hexadecimal digits.
277 Otherwise, it matches a literal "x" character. In JavaScript mode, support for
278 code points greater than 256 is provided by \eu, which must be followed by
279 four hexadecimal digits; otherwise it matches a literal "u" character.
280 .P
281 Characters whose value is less than 256 can be defined by either of the two
282 syntaxes for \ex (or by \eu in JavaScript mode). There is no difference in the
283 way they are handled. For example, \exdc is exactly the same as \ex{dc} (or
284 \eu00dc in JavaScript mode).
285 .P
286 After \e0 up to two further octal digits are read. If there are fewer than two
287 digits, just those that are present are used. Thus the sequence \e0\ex\e07
288 specifies two binary zeros followed by a BEL character (code value 7). Make
289 sure you supply two digits after the initial zero if the pattern character that
290 follows is itself an octal digit.
291 .P
292 The handling of a backslash followed by a digit other than 0 is complicated.
293 Outside a character class, PCRE reads it and any following digits as a decimal
294 number. If the number is less than 10, or if there have been at least that many
295 previous capturing left parentheses in the expression, the entire sequence is
296 taken as a \fIback reference\fP. A description of how this works is given
297 .\" HTML <a href="#backreferences">
298 .\" </a>
299 later,
300 .\"
301 following the discussion of
302 .\" HTML <a href="#subpattern">
303 .\" </a>
304 parenthesized subpatterns.
305 .\"
306 .P
307 Inside a character class, or if the decimal number is greater than 9 and there
308 have not been that many capturing subpatterns, PCRE re-reads up to three octal
309 digits following the backslash, and uses them to generate a data character. Any
310 subsequent digits stand for themselves. The value of the character is
311 constrained in the same way as characters specified in hexadecimal.
312 For example:
313 .sp
314 \e040 is another way of writing a space
315 .\" JOIN
316 \e40 is the same, provided there are fewer than 40
317 previous capturing subpatterns
318 \e7 is always a back reference
319 .\" JOIN
320 \e11 might be a back reference, or another way of
321 writing a tab
322 \e011 is always a tab
323 \e0113 is a tab followed by the character "3"
324 .\" JOIN
325 \e113 might be a back reference, otherwise the
326 character with octal code 113
327 .\" JOIN
328 \e377 might be a back reference, otherwise
329 the value 255 (decimal)
330 .\" JOIN
331 \e81 is either a back reference, or a binary zero
332 followed by the two characters "8" and "1"
333 .sp
334 Note that octal values of 100 or greater must not be introduced by a leading
335 zero, because no more than three octal digits are ever read.
336 .P
337 All the sequences that define a single character value can be used both inside
338 and outside character classes. In addition, inside a character class, \eb is
339 interpreted as the backspace character (hex 08).
340 .P
341 \eN is not allowed in a character class. \eB, \eR, and \eX are not special
342 inside a character class. Like other unrecognized escape sequences, they are
343 treated as the literal characters "B", "R", and "X" by default, but cause an
344 error if the PCRE_EXTRA option is set. Outside a character class, these
345 sequences have different meanings.
346 .
347 .
348 .SS "Unsupported escape sequences"
349 .rs
350 .sp
351 In Perl, the sequences \el, \eL, \eu, and \eU are recognized by its string
352 handler and used to modify the case of following characters. By default, PCRE
353 does not support these escape sequences. However, if the PCRE_JAVASCRIPT_COMPAT
354 option is set, \eU matches a "U" character, and \eu can be used to define a
355 character by code point, as described in the previous section.
356 .
357 .
358 .SS "Absolute and relative back references"
359 .rs
360 .sp
361 The sequence \eg followed by an unsigned or a negative number, optionally
362 enclosed in braces, is an absolute or relative back reference. A named back
363 reference can be coded as \eg{name}. Back references are discussed
364 .\" HTML <a href="#backreferences">
365 .\" </a>
366 later,
367 .\"
368 following the discussion of
369 .\" HTML <a href="#subpattern">
370 .\" </a>
371 parenthesized subpatterns.
372 .\"
373 .
374 .
375 .SS "Absolute and relative subroutine calls"
376 .rs
377 .sp
378 For compatibility with Oniguruma, the non-Perl syntax \eg followed by a name or
379 a number enclosed either in angle brackets or single quotes, is an alternative
380 syntax for referencing a subpattern as a "subroutine". Details are discussed
381 .\" HTML <a href="#onigurumasubroutines">
382 .\" </a>
383 later.
384 .\"
385 Note that \eg{...} (Perl syntax) and \eg<...> (Oniguruma syntax) are \fInot\fP
386 synonymous. The former is a back reference; the latter is a
387 .\" HTML <a href="#subpatternsassubroutines">
388 .\" </a>
389 subroutine
390 .\"
391 call.
392 .
393 .
394 .\" HTML <a name="genericchartypes"></a>
395 .SS "Generic character types"
396 .rs
397 .sp
398 Another use of backslash is for specifying generic character types:
399 .sp
400 \ed any decimal digit
401 \eD any character that is not a decimal digit
402 \eh any horizontal whitespace character
403 \eH any character that is not a horizontal whitespace character
404 \es any whitespace character
405 \eS any character that is not a whitespace character
406 \ev any vertical whitespace character
407 \eV any character that is not a vertical whitespace character
408 \ew any "word" character
409 \eW any "non-word" character
410 .sp
411 There is also the single sequence \eN, which matches a non-newline character.
412 This is the same as
413 .\" HTML <a href="#fullstopdot">
414 .\" </a>
415 the "." metacharacter
416 .\"
417 when PCRE_DOTALL is not set. Perl also uses \eN to match characters by name;
418 PCRE does not support this.
419 .P
420 Each pair of lower and upper case escape sequences partitions the complete set
421 of characters into two disjoint sets. Any given character matches one, and only
422 one, of each pair. The sequences can appear both inside and outside character
423 classes. They each match one character of the appropriate type. If the current
424 matching point is at the end of the subject string, all of them fail, because
425 there is no character to match.
426 .P
427 For compatibility with Perl, \es does not match the VT character (code 11).
428 This makes it different from the the POSIX "space" class. The \es characters
429 are HT (9), LF (10), FF (12), CR (13), and space (32). If "use locale;" is
430 included in a Perl script, \es may match the VT character. In PCRE, it never
431 does.
432 .P
433 A "word" character is an underscore or any character that is a letter or digit.
434 By default, the definition of letters and digits is controlled by PCRE's
435 low-valued character tables, and may vary if locale-specific matching is taking
436 place (see
437 .\" HTML <a href="pcreapi.html#localesupport">
438 .\" </a>
439 "Locale support"
440 .\"
441 in the
442 .\" HREF
443 \fBpcreapi\fP
444 .\"
445 page). For example, in a French locale such as "fr_FR" in Unix-like systems,
446 or "french" in Windows, some character codes greater than 128 are used for
447 accented letters, and these are then matched by \ew. The use of locales with
448 Unicode is discouraged.
449 .P
450 By default, in a UTF mode, characters with values greater than 128 never match
451 \ed, \es, or \ew, and always match \eD, \eS, and \eW. These sequences retain
452 their original meanings from before UTF support was available, mainly for
453 efficiency reasons. However, if PCRE is compiled with Unicode property support,
454 and the PCRE_UCP option is set, the behaviour is changed so that Unicode
455 properties are used to determine character types, as follows:
456 .sp
457 \ed any character that \ep{Nd} matches (decimal digit)
458 \es any character that \ep{Z} matches, plus HT, LF, FF, CR
459 \ew any character that \ep{L} or \ep{N} matches, plus underscore
460 .sp
461 The upper case escapes match the inverse sets of characters. Note that \ed
462 matches only decimal digits, whereas \ew matches any Unicode digit, as well as
463 any Unicode letter, and underscore. Note also that PCRE_UCP affects \eb, and
464 \eB because they are defined in terms of \ew and \eW. Matching these sequences
465 is noticeably slower when PCRE_UCP is set.
466 .P
467 The sequences \eh, \eH, \ev, and \eV are features that were added to Perl at
468 release 5.10. In contrast to the other sequences, which match only ASCII
469 characters by default, these always match certain high-valued codepoints,
470 whether or not PCRE_UCP is set. The horizontal space characters are:
471 .sp
472 U+0009 Horizontal tab
473 U+0020 Space
474 U+00A0 Non-break space
475 U+1680 Ogham space mark
476 U+180E Mongolian vowel separator
477 U+2000 En quad
478 U+2001 Em quad
479 U+2002 En space
480 U+2003 Em space
481 U+2004 Three-per-em space
482 U+2005 Four-per-em space
483 U+2006 Six-per-em space
484 U+2007 Figure space
485 U+2008 Punctuation space
486 U+2009 Thin space
487 U+200A Hair space
488 U+202F Narrow no-break space
489 U+205F Medium mathematical space
490 U+3000 Ideographic space
491 .sp
492 The vertical space characters are:
493 .sp
494 U+000A Linefeed
495 U+000B Vertical tab
496 U+000C Formfeed
497 U+000D Carriage return
498 U+0085 Next line
499 U+2028 Line separator
500 U+2029 Paragraph separator
501 .sp
502 In 8-bit, non-UTF-8 mode, only the characters with codepoints less than 256 are
503 relevant.
504 .
505 .
506 .\" HTML <a name="newlineseq"></a>
507 .SS "Newline sequences"
508 .rs
509 .sp
510 Outside a character class, by default, the escape sequence \eR matches any
511 Unicode newline sequence. In 8-bit non-UTF-8 mode \eR is equivalent to the
512 following:
513 .sp
514 (?>\er\en|\en|\ex0b|\ef|\er|\ex85)
515 .sp
516 This is an example of an "atomic group", details of which are given
517 .\" HTML <a href="#atomicgroup">
518 .\" </a>
519 below.
520 .\"
521 This particular group matches either the two-character sequence CR followed by
522 LF, or one of the single characters LF (linefeed, U+000A), VT (vertical tab,
523 U+000B), FF (formfeed, U+000C), CR (carriage return, U+000D), or NEL (next
524 line, U+0085). The two-character sequence is treated as a single unit that
525 cannot be split.
526 .P
527 In other modes, two additional characters whose codepoints are greater than 255
528 are added: LS (line separator, U+2028) and PS (paragraph separator, U+2029).
529 Unicode character property support is not needed for these characters to be
530 recognized.
531 .P
532 It is possible to restrict \eR to match only CR, LF, or CRLF (instead of the
533 complete set of Unicode line endings) by setting the option PCRE_BSR_ANYCRLF
534 either at compile time or when the pattern is matched. (BSR is an abbrevation
535 for "backslash R".) This can be made the default when PCRE is built; if this is
536 the case, the other behaviour can be requested via the PCRE_BSR_UNICODE option.
537 It is also possible to specify these settings by starting a pattern string with
538 one of the following sequences:
539 .sp
540 (*BSR_ANYCRLF) CR, LF, or CRLF only
541 (*BSR_UNICODE) any Unicode newline sequence
542 .sp
543 These override the default and the options given to the compiling function, but
544 they can themselves be overridden by options given to a matching function. Note
545 that these special settings, which are not Perl-compatible, are recognized only
546 at the very start of a pattern, and that they must be in upper case. If more
547 than one of them is present, the last one is used. They can be combined with a
548 change of newline convention; for example, a pattern can start with:
549 .sp
550 (*ANY)(*BSR_ANYCRLF)
551 .sp
552 They can also be combined with the (*UTF8), (*UTF16), or (*UCP) special
553 sequences. Inside a character class, \eR is treated as an unrecognized escape
554 sequence, and so matches the letter "R" by default, but causes an error if
555 PCRE_EXTRA is set.
556 .
557 .
558 .\" HTML <a name="uniextseq"></a>
559 .SS Unicode character properties
560 .rs
561 .sp
562 When PCRE is built with Unicode character property support, three additional
563 escape sequences that match characters with specific properties are available.
564 When in 8-bit non-UTF-8 mode, these sequences are of course limited to testing
565 characters whose codepoints are less than 256, but they do work in this mode.
566 The extra escape sequences are:
567 .sp
568 \ep{\fIxx\fP} a character with the \fIxx\fP property
569 \eP{\fIxx\fP} a character without the \fIxx\fP property
570 \eX an extended Unicode sequence
571 .sp
572 The property names represented by \fIxx\fP above are limited to the Unicode
573 script names, the general category properties, "Any", which matches any
574 character (including newline), and some special PCRE properties (described
575 in the
576 .\" HTML <a href="#extraprops">
577 .\" </a>
578 next section).
579 .\"
580 Other Perl properties such as "InMusicalSymbols" are not currently supported by
581 PCRE. Note that \eP{Any} does not match any characters, so always causes a
582 match failure.
583 .P
584 Sets of Unicode characters are defined as belonging to certain scripts. A
585 character from one of these sets can be matched using a script name. For
586 example:
587 .sp
588 \ep{Greek}
589 \eP{Han}
590 .sp
591 Those that are not part of an identified script are lumped together as
592 "Common". The current list of scripts is:
593 .P
594 Arabic,
595 Armenian,
596 Avestan,
597 Balinese,
598 Bamum,
599 Bengali,
600 Bopomofo,
601 Braille,
602 Buginese,
603 Buhid,
604 Canadian_Aboriginal,
605 Carian,
606 Cham,
607 Cherokee,
608 Common,
609 Coptic,
610 Cuneiform,
611 Cypriot,
612 Cyrillic,
613 Deseret,
614 Devanagari,
615 Egyptian_Hieroglyphs,
616 Ethiopic,
617 Georgian,
618 Glagolitic,
619 Gothic,
620 Greek,
621 Gujarati,
622 Gurmukhi,
623 Han,
624 Hangul,
625 Hanunoo,
626 Hebrew,
627 Hiragana,
628 Imperial_Aramaic,
629 Inherited,
630 Inscriptional_Pahlavi,
631 Inscriptional_Parthian,
632 Javanese,
633 Kaithi,
634 Kannada,
635 Katakana,
636 Kayah_Li,
637 Kharoshthi,
638 Khmer,
639 Lao,
640 Latin,
641 Lepcha,
642 Limbu,
643 Linear_B,
644 Lisu,
645 Lycian,
646 Lydian,
647 Malayalam,
648 Meetei_Mayek,
649 Mongolian,
650 Myanmar,
651 New_Tai_Lue,
652 Nko,
653 Ogham,
654 Old_Italic,
655 Old_Persian,
656 Old_South_Arabian,
657 Old_Turkic,
658 Ol_Chiki,
659 Oriya,
660 Osmanya,
661 Phags_Pa,
662 Phoenician,
663 Rejang,
664 Runic,
665 Samaritan,
666 Saurashtra,
667 Shavian,
668 Sinhala,
669 Sundanese,
670 Syloti_Nagri,
671 Syriac,
672 Tagalog,
673 Tagbanwa,
674 Tai_Le,
675 Tai_Tham,
676 Tai_Viet,
677 Tamil,
678 Telugu,
679 Thaana,
680 Thai,
681 Tibetan,
682 Tifinagh,
683 Ugaritic,
684 Vai,
685 Yi.
686 .P
687 Each character has exactly one Unicode general category property, specified by
688 a two-letter abbreviation. For compatibility with Perl, negation can be
689 specified by including a circumflex between the opening brace and the property
690 name. For example, \ep{^Lu} is the same as \eP{Lu}.
691 .P
692 If only one letter is specified with \ep or \eP, it includes all the general
693 category properties that start with that letter. In this case, in the absence
694 of negation, the curly brackets in the escape sequence are optional; these two
695 examples have the same effect:
696 .sp
697 \ep{L}
698 \epL
699 .sp
700 The following general category property codes are supported:
701 .sp
702 C Other
703 Cc Control
704 Cf Format
705 Cn Unassigned
706 Co Private use
707 Cs Surrogate
708 .sp
709 L Letter
710 Ll Lower case letter
711 Lm Modifier letter
712 Lo Other letter
713 Lt Title case letter
714 Lu Upper case letter
715 .sp
716 M Mark
717 Mc Spacing mark
718 Me Enclosing mark
719 Mn Non-spacing mark
720 .sp
721 N Number
722 Nd Decimal number
723 Nl Letter number
724 No Other number
725 .sp
726 P Punctuation
727 Pc Connector punctuation
728 Pd Dash punctuation
729 Pe Close punctuation
730 Pf Final punctuation
731 Pi Initial punctuation
732 Po Other punctuation
733 Ps Open punctuation
734 .sp
735 S Symbol
736 Sc Currency symbol
737 Sk Modifier symbol
738 Sm Mathematical symbol
739 So Other symbol
740 .sp
741 Z Separator
742 Zl Line separator
743 Zp Paragraph separator
744 Zs Space separator
745 .sp
746 The special property L& is also supported: it matches a character that has
747 the Lu, Ll, or Lt property, in other words, a letter that is not classified as
748 a modifier or "other".
749 .P
750 The Cs (Surrogate) property applies only to characters in the range U+D800 to
751 U+DFFF. Such characters are not valid in Unicode strings and so
752 cannot be tested by PCRE, unless UTF validity checking has been turned off
753 (see the discussion of PCRE_NO_UTF8_CHECK and PCRE_NO_UTF16_CHECK in the
754 .\" HREF
755 \fBpcreapi\fP
756 .\"
757 page). Perl does not support the Cs property.
758 .P
759 The long synonyms for property names that Perl supports (such as \ep{Letter})
760 are not supported by PCRE, nor is it permitted to prefix any of these
761 properties with "Is".
762 .P
763 No character that is in the Unicode table has the Cn (unassigned) property.
764 Instead, this property is assumed for any code point that is not in the
765 Unicode table.
766 .P
767 Specifying caseless matching does not affect these escape sequences. For
768 example, \ep{Lu} always matches only upper case letters.
769 .P
770 The \eX escape matches any number of Unicode characters that form an extended
771 Unicode sequence. \eX is equivalent to
772 .sp
773 (?>\ePM\epM*)
774 .sp
775 That is, it matches a character without the "mark" property, followed by zero
776 or more characters with the "mark" property, and treats the sequence as an
777 atomic group
778 .\" HTML <a href="#atomicgroup">
779 .\" </a>
780 (see below).
781 .\"
782 Characters with the "mark" property are typically accents that affect the
783 preceding character. None of them have codepoints less than 256, so in
784 8-bit non-UTF-8 mode \eX matches any one character.
785 .P
786 Note that recent versions of Perl have changed \eX to match what Unicode calls
787 an "extended grapheme cluster", which has a more complicated definition.
788 .P
789 Matching characters by Unicode property is not fast, because PCRE has to search
790 a structure that contains data for over fifteen thousand characters. That is
791 why the traditional escape sequences such as \ed and \ew do not use Unicode
792 properties in PCRE by default, though you can make them do so by setting the
793 PCRE_UCP option or by starting the pattern with (*UCP).
794 .
795 .
796 .\" HTML <a name="extraprops"></a>
797 .SS PCRE's additional properties
798 .rs
799 .sp
800 As well as the standard Unicode properties described in the previous
801 section, PCRE supports four more that make it possible to convert traditional
802 escape sequences such as \ew and \es and POSIX character classes to use Unicode
803 properties. PCRE uses these non-standard, non-Perl properties internally when
804 PCRE_UCP is set. They are:
805 .sp
806 Xan Any alphanumeric character
807 Xps Any POSIX space character
808 Xsp Any Perl space character
809 Xwd Any Perl "word" character
810 .sp
811 Xan matches characters that have either the L (letter) or the N (number)
812 property. Xps matches the characters tab, linefeed, vertical tab, formfeed, or
813 carriage return, and any other character that has the Z (separator) property.
814 Xsp is the same as Xps, except that vertical tab is excluded. Xwd matches the
815 same characters as Xan, plus underscore.
816 .
817 .
818 .\" HTML <a name="resetmatchstart"></a>
819 .SS "Resetting the match start"
820 .rs
821 .sp
822 The escape sequence \eK causes any previously matched characters not to be
823 included in the final matched sequence. For example, the pattern:
824 .sp
825 foo\eKbar
826 .sp
827 matches "foobar", but reports that it has matched "bar". This feature is
828 similar to a lookbehind assertion
829 .\" HTML <a href="#lookbehind">
830 .\" </a>
831 (described below).
832 .\"
833 However, in this case, the part of the subject before the real match does not
834 have to be of fixed length, as lookbehind assertions do. The use of \eK does
835 not interfere with the setting of
836 .\" HTML <a href="#subpattern">
837 .\" </a>
838 captured substrings.
839 .\"
840 For example, when the pattern
841 .sp
842 (foo)\eKbar
843 .sp
844 matches "foobar", the first substring is still set to "foo".
845 .P
846 Perl documents that the use of \eK within assertions is "not well defined". In
847 PCRE, \eK is acted upon when it occurs inside positive assertions, but is
848 ignored in negative assertions.
849 .
850 .
851 .\" HTML <a name="smallassertions"></a>
852 .SS "Simple assertions"
853 .rs
854 .sp
855 The final use of backslash is for certain simple assertions. An assertion
856 specifies a condition that has to be met at a particular point in a match,
857 without consuming any characters from the subject string. The use of
858 subpatterns for more complicated assertions is described
859 .\" HTML <a href="#bigassertions">
860 .\" </a>
861 below.
862 .\"
863 The backslashed assertions are:
864 .sp
865 \eb matches at a word boundary
866 \eB matches when not at a word boundary
867 \eA matches at the start of the subject
868 \eZ matches at the end of the subject
869 also matches before a newline at the end of the subject
870 \ez matches only at the end of the subject
871 \eG matches at the first matching position in the subject
872 .sp
873 Inside a character class, \eb has a different meaning; it matches the backspace
874 character. If any other of these assertions appears in a character class, by
875 default it matches the corresponding literal character (for example, \eB
876 matches the letter B). However, if the PCRE_EXTRA option is set, an "invalid
877 escape sequence" error is generated instead.
878 .P
879 A word boundary is a position in the subject string where the current character
880 and the previous character do not both match \ew or \eW (i.e. one matches
881 \ew and the other matches \eW), or the start or end of the string if the
882 first or last character matches \ew, respectively. In a UTF mode, the meanings
883 of \ew and \eW can be changed by setting the PCRE_UCP option. When this is
884 done, it also affects \eb and \eB. Neither PCRE nor Perl has a separate "start
885 of word" or "end of word" metasequence. However, whatever follows \eb normally
886 determines which it is. For example, the fragment \eba matches "a" at the start
887 of a word.
888 .P
889 The \eA, \eZ, and \ez assertions differ from the traditional circumflex and
890 dollar (described in the next section) in that they only ever match at the very
891 start and end of the subject string, whatever options are set. Thus, they are
892 independent of multiline mode. These three assertions are not affected by the
893 PCRE_NOTBOL or PCRE_NOTEOL options, which affect only the behaviour of the
894 circumflex and dollar metacharacters. However, if the \fIstartoffset\fP
895 argument of \fBpcre_exec()\fP is non-zero, indicating that matching is to start
896 at a point other than the beginning of the subject, \eA can never match. The
897 difference between \eZ and \ez is that \eZ matches before a newline at the end
898 of the string as well as at the very end, whereas \ez matches only at the end.
899 .P
900 The \eG assertion is true only when the current matching position is at the
901 start point of the match, as specified by the \fIstartoffset\fP argument of
902 \fBpcre_exec()\fP. It differs from \eA when the value of \fIstartoffset\fP is
903 non-zero. By calling \fBpcre_exec()\fP multiple times with appropriate
904 arguments, you can mimic Perl's /g option, and it is in this kind of
905 implementation where \eG can be useful.
906 .P
907 Note, however, that PCRE's interpretation of \eG, as the start of the current
908 match, is subtly different from Perl's, which defines it as the end of the
909 previous match. In Perl, these can be different when the previously matched
910 string was empty. Because PCRE does just one match at a time, it cannot
911 reproduce this behaviour.
912 .P
913 If all the alternatives of a pattern begin with \eG, the expression is anchored
914 to the starting match position, and the "anchored" flag is set in the compiled
915 regular expression.
916 .
917 .
918 .SH "CIRCUMFLEX AND DOLLAR"
919 .rs
920 .sp
921 Outside a character class, in the default matching mode, the circumflex
922 character is an assertion that is true only if the current matching point is
923 at the start of the subject string. If the \fIstartoffset\fP argument of
924 \fBpcre_exec()\fP is non-zero, circumflex can never match if the PCRE_MULTILINE
925 option is unset. Inside a character class, circumflex has an entirely different
926 meaning
927 .\" HTML <a href="#characterclass">
928 .\" </a>
929 (see below).
930 .\"
931 .P
932 Circumflex need not be the first character of the pattern if a number of
933 alternatives are involved, but it should be the first thing in each alternative
934 in which it appears if the pattern is ever to match that branch. If all
935 possible alternatives start with a circumflex, that is, if the pattern is
936 constrained to match only at the start of the subject, it is said to be an
937 "anchored" pattern. (There are also other constructs that can cause a pattern
938 to be anchored.)
939 .P
940 A dollar character is an assertion that is true only if the current matching
941 point is at the end of the subject string, or immediately before a newline
942 at the end of the string (by default). Dollar need not be the last character of
943 the pattern if a number of alternatives are involved, but it should be the last
944 item in any branch in which it appears. Dollar has no special meaning in a
945 character class.
946 .P
947 The meaning of dollar can be changed so that it matches only at the very end of
948 the string, by setting the PCRE_DOLLAR_ENDONLY option at compile time. This
949 does not affect the \eZ assertion.
950 .P
951 The meanings of the circumflex and dollar characters are changed if the
952 PCRE_MULTILINE option is set. When this is the case, a circumflex matches
953 immediately after internal newlines as well as at the start of the subject
954 string. It does not match after a newline that ends the string. A dollar
955 matches before any newlines in the string, as well as at the very end, when
956 PCRE_MULTILINE is set. When newline is specified as the two-character
957 sequence CRLF, isolated CR and LF characters do not indicate newlines.
958 .P
959 For example, the pattern /^abc$/ matches the subject string "def\enabc" (where
960 \en represents a newline) in multiline mode, but not otherwise. Consequently,
961 patterns that are anchored in single line mode because all branches start with
962 ^ are not anchored in multiline mode, and a match for circumflex is possible
963 when the \fIstartoffset\fP argument of \fBpcre_exec()\fP is non-zero. The
964 PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
965 .P
966 Note that the sequences \eA, \eZ, and \ez can be used to match the start and
967 end of the subject in both modes, and if all branches of a pattern start with
968 \eA it is always anchored, whether or not PCRE_MULTILINE is set.
969 .
970 .
971 .\" HTML <a name="fullstopdot"></a>
972 .SH "FULL STOP (PERIOD, DOT) AND \eN"
973 .rs
974 .sp
975 Outside a character class, a dot in the pattern matches any one character in
976 the subject string except (by default) a character that signifies the end of a
977 line.
978 .P
979 When a line ending is defined as a single character, dot never matches that
980 character; when the two-character sequence CRLF is used, dot does not match CR
981 if it is immediately followed by LF, but otherwise it matches all characters
982 (including isolated CRs and LFs). When any Unicode line endings are being
983 recognized, dot does not match CR or LF or any of the other line ending
984 characters.
985 .P
986 The behaviour of dot with regard to newlines can be changed. If the PCRE_DOTALL
987 option is set, a dot matches any one character, without exception. If the
988 two-character sequence CRLF is present in the subject string, it takes two dots
989 to match it.
990 .P
991 The handling of dot is entirely independent of the handling of circumflex and
992 dollar, the only relationship being that they both involve newlines. Dot has no
993 special meaning in a character class.
994 .P
995 The escape sequence \eN behaves like a dot, except that it is not affected by
996 the PCRE_DOTALL option. In other words, it matches any character except one
997 that signifies the end of a line. Perl also uses \eN to match characters by
998 name; PCRE does not support this.
999 .
1000 .
1001 .SH "MATCHING A SINGLE DATA UNIT"
1002 .rs
1003 .sp
1004 Outside a character class, the escape sequence \eC matches any one data unit,
1005 whether or not a UTF mode is set. In the 8-bit library, one data unit is one
1006 byte; in the 16-bit library it is a 16-bit unit. Unlike a dot, \eC always
1007 matches line-ending characters. The feature is provided in Perl in order to
1008 match individual bytes in UTF-8 mode, but it is unclear how it can usefully be
1009 used. Because \eC breaks up characters into individual data units, matching one
1010 unit with \eC in a UTF mode means that the rest of the string may start with a
1011 malformed UTF character. This has undefined results, because PCRE assumes that
1012 it is dealing with valid UTF strings (and by default it checks this at the
1013 start of processing unless the PCRE_NO_UTF8_CHECK option is used).
1014 .P
1015 PCRE does not allow \eC to appear in lookbehind assertions
1016 .\" HTML <a href="#lookbehind">
1017 .\" </a>
1018 (described below)
1019 .\"
1020 in a UTF mode, because this would make it impossible to calculate the length of
1021 the lookbehind.
1022 .P
1023 In general, the \eC escape sequence is best avoided. However, one
1024 way of using it that avoids the problem of malformed UTF characters is to use a
1025 lookahead to check the length of the next character, as in this pattern, which
1026 could be used with a UTF-8 string (ignore white space and line breaks):
1027 .sp
1028 (?| (?=[\ex00-\ex7f])(\eC) |
1029 (?=[\ex80-\ex{7ff}])(\eC)(\eC) |
1030 (?=[\ex{800}-\ex{ffff}])(\eC)(\eC)(\eC) |
1031 (?=[\ex{10000}-\ex{1fffff}])(\eC)(\eC)(\eC)(\eC))
1032 .sp
1033 A group that starts with (?| resets the capturing parentheses numbers in each
1034 alternative (see
1035 .\" HTML <a href="#dupsubpatternnumber">
1036 .\" </a>
1037 "Duplicate Subpattern Numbers"
1038 .\"
1039 below). The assertions at the start of each branch check the next UTF-8
1040 character for values whose encoding uses 1, 2, 3, or 4 bytes, respectively. The
1041 character's individual bytes are then captured by the appropriate number of
1042 groups.
1043 .
1044 .
1045 .\" HTML <a name="characterclass"></a>
1046 .SH "SQUARE BRACKETS AND CHARACTER CLASSES"
1047 .rs
1048 .sp
1049 An opening square bracket introduces a character class, terminated by a closing
1050 square bracket. A closing square bracket on its own is not special by default.
1051 However, if the PCRE_JAVASCRIPT_COMPAT option is set, a lone closing square
1052 bracket causes a compile-time error. If a closing square bracket is required as
1053 a member of the class, it should be the first data character in the class
1054 (after an initial circumflex, if present) or escaped with a backslash.
1055 .P
1056 A character class matches a single character in the subject. In a UTF mode, the
1057 character may be more than one data unit long. A matched character must be in
1058 the set of characters defined by the class, unless the first character in the
1059 class definition is a circumflex, in which case the subject character must not
1060 be in the set defined by the class. If a circumflex is actually required as a
1061 member of the class, ensure it is not the first character, or escape it with a
1062 backslash.
1063 .P
1064 For example, the character class [aeiou] matches any lower case vowel, while
1065 [^aeiou] matches any character that is not a lower case vowel. Note that a
1066 circumflex is just a convenient notation for specifying the characters that
1067 are in the class by enumerating those that are not. A class that starts with a
1068 circumflex is not an assertion; it still consumes a character from the subject
1069 string, and therefore it fails if the current pointer is at the end of the
1070 string.
1071 .P
1072 In UTF-8 (UTF-16) mode, characters with values greater than 255 (0xffff) can be
1073 included in a class as a literal string of data units, or by using the \ex{
1074 escaping mechanism.
1075 .P
1076 When caseless matching is set, any letters in a class represent both their
1077 upper case and lower case versions, so for example, a caseless [aeiou] matches
1078 "A" as well as "a", and a caseless [^aeiou] does not match "A", whereas a
1079 caseful version would. In a UTF mode, PCRE always understands the concept of
1080 case for characters whose values are less than 128, so caseless matching is
1081 always possible. For characters with higher values, the concept of case is
1082 supported if PCRE is compiled with Unicode property support, but not otherwise.
1083 If you want to use caseless matching in a UTF mode for characters 128 and
1084 above, you must ensure that PCRE is compiled with Unicode property support as
1085 well as with UTF support.
1086 .P
1087 Characters that might indicate line breaks are never treated in any special way
1088 when matching character classes, whatever line-ending sequence is in use, and
1089 whatever setting of the PCRE_DOTALL and PCRE_MULTILINE options is used. A class
1090 such as [^a] always matches one of these characters.
1091 .P
1092 The minus (hyphen) character can be used to specify a range of characters in a
1093 character class. For example, [d-m] matches any letter between d and m,
1094 inclusive. If a minus character is required in a class, it must be escaped with
1095 a backslash or appear in a position where it cannot be interpreted as
1096 indicating a range, typically as the first or last character in the class.
1097 .P
1098 It is not possible to have the literal character "]" as the end character of a
1099 range. A pattern such as [W-]46] is interpreted as a class of two characters
1100 ("W" and "-") followed by a literal string "46]", so it would match "W46]" or
1101 "-46]". However, if the "]" is escaped with a backslash it is interpreted as
1102 the end of range, so [W-\e]46] is interpreted as a class containing a range
1103 followed by two other characters. The octal or hexadecimal representation of
1104 "]" can also be used to end a range.
1105 .P
1106 Ranges operate in the collating sequence of character values. They can also be
1107 used for characters specified numerically, for example [\e000-\e037]. Ranges
1108 can include any characters that are valid for the current mode.
1109 .P
1110 If a range that includes letters is used when caseless matching is set, it
1111 matches the letters in either case. For example, [W-c] is equivalent to
1112 [][\e\e^_`wxyzabc], matched caselessly, and in a non-UTF mode, if character
1113 tables for a French locale are in use, [\exc8-\excb] matches accented E
1114 characters in both cases. In UTF modes, PCRE supports the concept of case for
1115 characters with values greater than 128 only when it is compiled with Unicode
1116 property support.
1117 .P
1118 The character escape sequences \ed, \eD, \eh, \eH, \ep, \eP, \es, \eS, \ev,
1119 \eV, \ew, and \eW may appear in a character class, and add the characters that
1120 they match to the class. For example, [\edABCDEF] matches any hexadecimal
1121 digit. In UTF modes, the PCRE_UCP option affects the meanings of \ed, \es, \ew
1122 and their upper case partners, just as it does when they appear outside a
1123 character class, as described in the section entitled
1124 .\" HTML <a href="#genericchartypes">
1125 .\" </a>
1126 "Generic character types"
1127 .\"
1128 above. The escape sequence \eb has a different meaning inside a character
1129 class; it matches the backspace character. The sequences \eB, \eN, \eR, and \eX
1130 are not special inside a character class. Like any other unrecognized escape
1131 sequences, they are treated as the literal characters "B", "N", "R", and "X" by
1132 default, but cause an error if the PCRE_EXTRA option is set.
1133 .P
1134 A circumflex can conveniently be used with the upper case character types to
1135 specify a more restricted set of characters than the matching lower case type.
1136 For example, the class [^\eW_] matches any letter or digit, but not underscore,
1137 whereas [\ew] includes underscore. A positive character class should be read as
1138 "something OR something OR ..." and a negative class as "NOT something AND NOT
1139 something AND NOT ...".
1140 .P
1141 The only metacharacters that are recognized in character classes are backslash,
1142 hyphen (only where it can be interpreted as specifying a range), circumflex
1143 (only at the start), opening square bracket (only when it can be interpreted as
1144 introducing a POSIX class name - see the next section), and the terminating
1145 closing square bracket. However, escaping other non-alphanumeric characters
1146 does no harm.
1147 .
1148 .
1149 .SH "POSIX CHARACTER CLASSES"
1150 .rs
1151 .sp
1152 Perl supports the POSIX notation for character classes. This uses names
1153 enclosed by [: and :] within the enclosing square brackets. PCRE also supports
1154 this notation. For example,
1155 .sp
1156 [01[:alpha:]%]
1157 .sp
1158 matches "0", "1", any alphabetic character, or "%". The supported class names
1159 are:
1160 .sp
1161 alnum letters and digits
1162 alpha letters
1163 ascii character codes 0 - 127
1164 blank space or tab only
1165 cntrl control characters
1166 digit decimal digits (same as \ed)
1167 graph printing characters, excluding space
1168 lower lower case letters
1169 print printing characters, including space
1170 punct printing characters, excluding letters and digits and space
1171 space white space (not quite the same as \es)
1172 upper upper case letters
1173 word "word" characters (same as \ew)
1174 xdigit hexadecimal digits
1175 .sp
1176 The "space" characters are HT (9), LF (10), VT (11), FF (12), CR (13), and
1177 space (32). Notice that this list includes the VT character (code 11). This
1178 makes "space" different to \es, which does not include VT (for Perl
1179 compatibility).
1180 .P
1181 The name "word" is a Perl extension, and "blank" is a GNU extension from Perl
1182 5.8. Another Perl extension is negation, which is indicated by a ^ character
1183 after the colon. For example,
1184 .sp
1185 [12[:^digit:]]
1186 .sp
1187 matches "1", "2", or any non-digit. PCRE (and Perl) also recognize the POSIX
1188 syntax [.ch.] and [=ch=] where "ch" is a "collating element", but these are not
1189 supported, and an error is given if they are encountered.
1190 .P
1191 By default, in UTF modes, characters with values greater than 128 do not match
1192 any of the POSIX character classes. However, if the PCRE_UCP option is passed
1193 to \fBpcre_compile()\fP, some of the classes are changed so that Unicode
1194 character properties are used. This is achieved by replacing the POSIX classes
1195 by other sequences, as follows:
1196 .sp
1197 [:alnum:] becomes \ep{Xan}
1198 [:alpha:] becomes \ep{L}
1199 [:blank:] becomes \eh
1200 [:digit:] becomes \ep{Nd}
1201 [:lower:] becomes \ep{Ll}
1202 [:space:] becomes \ep{Xps}
1203 [:upper:] becomes \ep{Lu}
1204 [:word:] becomes \ep{Xwd}
1205 .sp
1206 Negated versions, such as [:^alpha:] use \eP instead of \ep. The other POSIX
1207 classes are unchanged, and match only characters with code points less than
1208 128.
1209 .
1210 .
1211 .SH "VERTICAL BAR"
1212 .rs
1213 .sp
1214 Vertical bar characters are used to separate alternative patterns. For example,
1215 the pattern
1216 .sp
1217 gilbert|sullivan
1218 .sp
1219 matches either "gilbert" or "sullivan". Any number of alternatives may appear,
1220 and an empty alternative is permitted (matching the empty string). The matching
1221 process tries each alternative in turn, from left to right, and the first one
1222 that succeeds is used. If the alternatives are within a subpattern
1223 .\" HTML <a href="#subpattern">
1224 .\" </a>
1225 (defined below),
1226 .\"
1227 "succeeds" means matching the rest of the main pattern as well as the
1228 alternative in the subpattern.
1229 .
1230 .
1231 .SH "INTERNAL OPTION SETTING"
1232 .rs
1233 .sp
1234 The settings of the PCRE_CASELESS, PCRE_MULTILINE, PCRE_DOTALL, and
1235 PCRE_EXTENDED options (which are Perl-compatible) can be changed from within
1236 the pattern by a sequence of Perl option letters enclosed between "(?" and ")".
1237 The option letters are
1238 .sp
1239 i for PCRE_CASELESS
1240 m for PCRE_MULTILINE
1241 s for PCRE_DOTALL
1242 x for PCRE_EXTENDED
1243 .sp
1244 For example, (?im) sets caseless, multiline matching. It is also possible to
1245 unset these options by preceding the letter with a hyphen, and a combined
1246 setting and unsetting such as (?im-sx), which sets PCRE_CASELESS and
1247 PCRE_MULTILINE while unsetting PCRE_DOTALL and PCRE_EXTENDED, is also
1248 permitted. If a letter appears both before and after the hyphen, the option is
1249 unset.
1250 .P
1251 The PCRE-specific options PCRE_DUPNAMES, PCRE_UNGREEDY, and PCRE_EXTRA can be
1252 changed in the same way as the Perl-compatible options by using the characters
1253 J, U and X respectively.
1254 .P
1255 When one of these option changes occurs at top level (that is, not inside
1256 subpattern parentheses), the change applies to the remainder of the pattern
1257 that follows. If the change is placed right at the start of a pattern, PCRE
1258 extracts it into the global options (and it will therefore show up in data
1259 extracted by the \fBpcre_fullinfo()\fP function).
1260 .P
1261 An option change within a subpattern (see below for a description of
1262 subpatterns) affects only that part of the subpattern that follows it, so
1263 .sp
1264 (a(?i)b)c
1265 .sp
1266 matches abc and aBc and no other strings (assuming PCRE_CASELESS is not used).
1267 By this means, options can be made to have different settings in different
1268 parts of the pattern. Any changes made in one alternative do carry on
1269 into subsequent branches within the same subpattern. For example,
1270 .sp
1271 (a(?i)b|c)
1272 .sp
1273 matches "ab", "aB", "c", and "C", even though when matching "C" the first
1274 branch is abandoned before the option setting. This is because the effects of
1275 option settings happen at compile time. There would be some very weird
1276 behaviour otherwise.
1277 .P
1278 \fBNote:\fP There are other PCRE-specific options that can be set by the
1279 application when the compiling or matching functions are called. In some cases
1280 the pattern can contain special leading sequences such as (*CRLF) to override
1281 what the application has set or what has been defaulted. Details are given in
1282 the section entitled
1283 .\" HTML <a href="#newlineseq">
1284 .\" </a>
1285 "Newline sequences"
1286 .\"
1287 above. There are also the (*UTF8), (*UTF16), and (*UCP) leading sequences that
1288 can be used to set UTF and Unicode property modes; they are equivalent to
1289 setting the PCRE_UTF8, PCRE_UTF16, and the PCRE_UCP options, respectively.
1290 .
1291 .
1292 .\" HTML <a name="subpattern"></a>
1293 .SH SUBPATTERNS
1294 .rs
1295 .sp
1296 Subpatterns are delimited by parentheses (round brackets), which can be nested.
1297 Turning part of a pattern into a subpattern does two things:
1298 .sp
1299 1. It localizes a set of alternatives. For example, the pattern
1300 .sp
1301 cat(aract|erpillar|)
1302 .sp
1303 matches "cataract", "caterpillar", or "cat". Without the parentheses, it would
1304 match "cataract", "erpillar" or an empty string.
1305 .sp
1306 2. It sets up the subpattern as a capturing subpattern. This means that, when
1307 the whole pattern matches, that portion of the subject string that matched the
1308 subpattern is passed back to the caller via the \fIovector\fP argument of the
1309 matching function. (This applies only to the traditional matching functions;
1310 the DFA matching functions do not support capturing.)
1311 .P
1312 Opening parentheses are counted from left to right (starting from 1) to obtain
1313 numbers for the capturing subpatterns. For example, if the string "the red
1314 king" is matched against the pattern
1315 .sp
1316 the ((red|white) (king|queen))
1317 .sp
1318 the captured substrings are "red king", "red", and "king", and are numbered 1,
1319 2, and 3, respectively.
1320 .P
1321 The fact that plain parentheses fulfil two functions is not always helpful.
1322 There are often times when a grouping subpattern is required without a
1323 capturing requirement. If an opening parenthesis is followed by a question mark
1324 and a colon, the subpattern does not do any capturing, and is not counted when
1325 computing the number of any subsequent capturing subpatterns. For example, if
1326 the string "the white queen" is matched against the pattern
1327 .sp
1328 the ((?:red|white) (king|queen))
1329 .sp
1330 the captured substrings are "white queen" and "queen", and are numbered 1 and
1331 2. The maximum number of capturing subpatterns is 65535.
1332 .P
1333 As a convenient shorthand, if any option settings are required at the start of
1334 a non-capturing subpattern, the option letters may appear between the "?" and
1335 the ":". Thus the two patterns
1336 .sp
1337 (?i:saturday|sunday)
1338 (?:(?i)saturday|sunday)
1339 .sp
1340 match exactly the same set of strings. Because alternative branches are tried
1341 from left to right, and options are not reset until the end of the subpattern
1342 is reached, an option setting in one branch does affect subsequent branches, so
1343 the above patterns match "SUNDAY" as well as "Saturday".
1344 .
1345 .
1346 .\" HTML <a name="dupsubpatternnumber"></a>
1347 .SH "DUPLICATE SUBPATTERN NUMBERS"
1348 .rs
1349 .sp
1350 Perl 5.10 introduced a feature whereby each alternative in a subpattern uses
1351 the same numbers for its capturing parentheses. Such a subpattern starts with
1352 (?| and is itself a non-capturing subpattern. For example, consider this
1353 pattern:
1354 .sp
1355 (?|(Sat)ur|(Sun))day
1356 .sp
1357 Because the two alternatives are inside a (?| group, both sets of capturing
1358 parentheses are numbered one. Thus, when the pattern matches, you can look
1359 at captured substring number one, whichever alternative matched. This construct
1360 is useful when you want to capture part, but not all, of one of a number of
1361 alternatives. Inside a (?| group, parentheses are numbered as usual, but the
1362 number is reset at the start of each branch. The numbers of any capturing
1363 parentheses that follow the subpattern start after the highest number used in
1364 any branch. The following example is taken from the Perl documentation. The
1365 numbers underneath show in which buffer the captured content will be stored.
1366 .sp
1367 # before ---------------branch-reset----------- after
1368 / ( a ) (?| x ( y ) z | (p (q) r) | (t) u (v) ) ( z ) /x
1369 # 1 2 2 3 2 3 4
1370 .sp
1371 A back reference to a numbered subpattern uses the most recent value that is
1372 set for that number by any subpattern. The following pattern matches "abcabc"
1373 or "defdef":
1374 .sp
1375 /(?|(abc)|(def))\e1/
1376 .sp
1377 In contrast, a subroutine call to a numbered subpattern always refers to the
1378 first one in the pattern with the given number. The following pattern matches
1379 "abcabc" or "defabc":
1380 .sp
1381 /(?|(abc)|(def))(?1)/
1382 .sp
1383 If a
1384 .\" HTML <a href="#conditions">
1385 .\" </a>
1386 condition test
1387 .\"
1388 for a subpattern's having matched refers to a non-unique number, the test is
1389 true if any of the subpatterns of that number have matched.
1390 .P
1391 An alternative approach to using this "branch reset" feature is to use
1392 duplicate named subpatterns, as described in the next section.
1393 .
1394 .
1395 .SH "NAMED SUBPATTERNS"
1396 .rs
1397 .sp
1398 Identifying capturing parentheses by number is simple, but it can be very hard
1399 to keep track of the numbers in complicated regular expressions. Furthermore,
1400 if an expression is modified, the numbers may change. To help with this
1401 difficulty, PCRE supports the naming of subpatterns. This feature was not
1402 added to Perl until release 5.10. Python had the feature earlier, and PCRE
1403 introduced it at release 4.0, using the Python syntax. PCRE now supports both
1404 the Perl and the Python syntax. Perl allows identically numbered subpatterns to
1405 have different names, but PCRE does not.
1406 .P
1407 In PCRE, a subpattern can be named in one of three ways: (?<name>...) or
1408 (?'name'...) as in Perl, or (?P<name>...) as in Python. References to capturing
1409 parentheses from other parts of the pattern, such as
1410 .\" HTML <a href="#backreferences">
1411 .\" </a>
1412 back references,
1413 .\"
1414 .\" HTML <a href="#recursion">
1415 .\" </a>
1416 recursion,
1417 .\"
1418 and
1419 .\" HTML <a href="#conditions">
1420 .\" </a>
1421 conditions,
1422 .\"
1423 can be made by name as well as by number.
1424 .P
1425 Names consist of up to 32 alphanumeric characters and underscores. Named
1426 capturing parentheses are still allocated numbers as well as names, exactly as
1427 if the names were not present. The PCRE API provides function calls for
1428 extracting the name-to-number translation table from a compiled pattern. There
1429 is also a convenience function for extracting a captured substring by name.
1430 .P
1431 By default, a name must be unique within a pattern, but it is possible to relax
1432 this constraint by setting the PCRE_DUPNAMES option at compile time. (Duplicate
1433 names are also always permitted for subpatterns with the same number, set up as
1434 described in the previous section.) Duplicate names can be useful for patterns
1435 where only one instance of the named parentheses can match. Suppose you want to
1436 match the name of a weekday, either as a 3-letter abbreviation or as the full
1437 name, and in both cases you want to extract the abbreviation. This pattern
1438 (ignoring the line breaks) does the job:
1439 .sp
1440 (?<DN>Mon|Fri|Sun)(?:day)?|
1441 (?<DN>Tue)(?:sday)?|
1442 (?<DN>Wed)(?:nesday)?|
1443 (?<DN>Thu)(?:rsday)?|
1444 (?<DN>Sat)(?:urday)?
1445 .sp
1446 There are five capturing substrings, but only one is ever set after a match.
1447 (An alternative way of solving this problem is to use a "branch reset"
1448 subpattern, as described in the previous section.)
1449 .P
1450 The convenience function for extracting the data by name returns the substring
1451 for the first (and in this example, the only) subpattern of that name that
1452 matched. This saves searching to find which numbered subpattern it was.
1453 .P
1454 If you make a back reference to a non-unique named subpattern from elsewhere in
1455 the pattern, the one that corresponds to the first occurrence of the name is
1456 used. In the absence of duplicate numbers (see the previous section) this is
1457 the one with the lowest number. If you use a named reference in a condition
1458 test (see the
1459 .\"
1460 .\" HTML <a href="#conditions">
1461 .\" </a>
1462 section about conditions
1463 .\"
1464 below), either to check whether a subpattern has matched, or to check for
1465 recursion, all subpatterns with the same name are tested. If the condition is
1466 true for any one of them, the overall condition is true. This is the same
1467 behaviour as testing by number. For further details of the interfaces for
1468 handling named subpatterns, see the
1469 .\" HREF
1470 \fBpcreapi\fP
1471 .\"
1472 documentation.
1473 .P
1474 \fBWarning:\fP You cannot use different names to distinguish between two
1475 subpatterns with the same number because PCRE uses only the numbers when
1476 matching. For this reason, an error is given at compile time if different names
1477 are given to subpatterns with the same number. However, you can give the same
1478 name to subpatterns with the same number, even when PCRE_DUPNAMES is not set.
1479 .
1480 .
1481 .SH REPETITION
1482 .rs
1483 .sp
1484 Repetition is specified by quantifiers, which can follow any of the following
1485 items:
1486 .sp
1487 a literal data character
1488 the dot metacharacter
1489 the \eC escape sequence
1490 the \eX escape sequence
1491 the \eR escape sequence
1492 an escape such as \ed or \epL that matches a single character
1493 a character class
1494 a back reference (see next section)
1495 a parenthesized subpattern (including assertions)
1496 a subroutine call to a subpattern (recursive or otherwise)
1497 .sp
1498 The general repetition quantifier specifies a minimum and maximum number of
1499 permitted matches, by giving the two numbers in curly brackets (braces),
1500 separated by a comma. The numbers must be less than 65536, and the first must
1501 be less than or equal to the second. For example:
1502 .sp
1503 z{2,4}
1504 .sp
1505 matches "zz", "zzz", or "zzzz". A closing brace on its own is not a special
1506 character. If the second number is omitted, but the comma is present, there is
1507 no upper limit; if the second number and the comma are both omitted, the
1508 quantifier specifies an exact number of required matches. Thus
1509 .sp
1510 [aeiou]{3,}
1511 .sp
1512 matches at least 3 successive vowels, but may match many more, while
1513 .sp
1514 \ed{8}
1515 .sp
1516 matches exactly 8 digits. An opening curly bracket that appears in a position
1517 where a quantifier is not allowed, or one that does not match the syntax of a
1518 quantifier, is taken as a literal character. For example, {,6} is not a
1519 quantifier, but a literal string of four characters.
1520 .P
1521 In UTF modes, quantifiers apply to characters rather than to individual data
1522 units. Thus, for example, \ex{100}{2} matches two characters, each of
1523 which is represented by a two-byte sequence in a UTF-8 string. Similarly,
1524 \eX{3} matches three Unicode extended sequences, each of which may be several
1525 data units long (and they may be of different lengths).
1526 .P
1527 The quantifier {0} is permitted, causing the expression to behave as if the
1528 previous item and the quantifier were not present. This may be useful for
1529 subpatterns that are referenced as
1530 .\" HTML <a href="#subpatternsassubroutines">
1531 .\" </a>
1532 subroutines
1533 .\"
1534 from elsewhere in the pattern (but see also the section entitled
1535 .\" HTML <a href="#subdefine">
1536 .\" </a>
1537 "Defining subpatterns for use by reference only"
1538 .\"
1539 below). Items other than subpatterns that have a {0} quantifier are omitted
1540 from the compiled pattern.
1541 .P
1542 For convenience, the three most common quantifiers have single-character
1543 abbreviations:
1544 .sp
1545 * is equivalent to {0,}
1546 + is equivalent to {1,}
1547 ? is equivalent to {0,1}
1548 .sp
1549 It is possible to construct infinite loops by following a subpattern that can
1550 match no characters with a quantifier that has no upper limit, for example:
1551 .sp
1552 (a?)*
1553 .sp
1554 Earlier versions of Perl and PCRE used to give an error at compile time for
1555 such patterns. However, because there are cases where this can be useful, such
1556 patterns are now accepted, but if any repetition of the subpattern does in fact
1557 match no characters, the loop is forcibly broken.
1558 .P
1559 By default, the quantifiers are "greedy", that is, they match as much as
1560 possible (up to the maximum number of permitted times), without causing the
1561 rest of the pattern to fail. The classic example of where this gives problems
1562 is in trying to match comments in C programs. These appear between /* and */
1563 and within the comment, individual * and / characters may appear. An attempt to
1564 match C comments by applying the pattern
1565 .sp
1566 /\e*.*\e*/
1567 .sp
1568 to the string
1569 .sp
1570 /* first comment */ not comment /* second comment */
1571 .sp
1572 fails, because it matches the entire string owing to the greediness of the .*
1573 item.
1574 .P
1575 However, if a quantifier is followed by a question mark, it ceases to be
1576 greedy, and instead matches the minimum number of times possible, so the
1577 pattern
1578 .sp
1579 /\e*.*?\e*/
1580 .sp
1581 does the right thing with the C comments. The meaning of the various
1582 quantifiers is not otherwise changed, just the preferred number of matches.
1583 Do not confuse this use of question mark with its use as a quantifier in its
1584 own right. Because it has two uses, it can sometimes appear doubled, as in
1585 .sp
1586 \ed??\ed
1587 .sp
1588 which matches one digit by preference, but can match two if that is the only
1589 way the rest of the pattern matches.
1590 .P
1591 If the PCRE_UNGREEDY option is set (an option that is not available in Perl),
1592 the quantifiers are not greedy by default, but individual ones can be made
1593 greedy by following them with a question mark. In other words, it inverts the
1594 default behaviour.
1595 .P
1596 When a parenthesized subpattern is quantified with a minimum repeat count that
1597 is greater than 1 or with a limited maximum, more memory is required for the
1598 compiled pattern, in proportion to the size of the minimum or maximum.
1599 .P
1600 If a pattern starts with .* or .{0,} and the PCRE_DOTALL option (equivalent
1601 to Perl's /s) is set, thus allowing the dot to match newlines, the pattern is
1602 implicitly anchored, because whatever follows will be tried against every
1603 character position in the subject string, so there is no point in retrying the
1604 overall match at any position after the first. PCRE normally treats such a
1605 pattern as though it were preceded by \eA.
1606 .P
1607 In cases where it is known that the subject string contains no newlines, it is
1608 worth setting PCRE_DOTALL in order to obtain this optimization, or
1609 alternatively using ^ to indicate anchoring explicitly.
1610 .P
1611 However, there is one situation where the optimization cannot be used. When .*
1612 is inside capturing parentheses that are the subject of a back reference
1613 elsewhere in the pattern, a match at the start may fail where a later one
1614 succeeds. Consider, for example:
1615 .sp
1616 (.*)abc\e1
1617 .sp
1618 If the subject is "xyz123abc123" the match point is the fourth character. For
1619 this reason, such a pattern is not implicitly anchored.
1620 .P
1621 When a capturing subpattern is repeated, the value captured is the substring
1622 that matched the final iteration. For example, after
1623 .sp
1624 (tweedle[dume]{3}\es*)+
1625 .sp
1626 has matched "tweedledum tweedledee" the value of the captured substring is
1627 "tweedledee". However, if there are nested capturing subpatterns, the
1628 corresponding captured values may have been set in previous iterations. For
1629 example, after
1630 .sp
1631 /(a|(b))+/
1632 .sp
1633 matches "aba" the value of the second captured substring is "b".
1634 .
1635 .
1636 .\" HTML <a name="atomicgroup"></a>
1637 .SH "ATOMIC GROUPING AND POSSESSIVE QUANTIFIERS"
1638 .rs
1639 .sp
1640 With both maximizing ("greedy") and minimizing ("ungreedy" or "lazy")
1641 repetition, failure of what follows normally causes the repeated item to be
1642 re-evaluated to see if a different number of repeats allows the rest of the
1643 pattern to match. Sometimes it is useful to prevent this, either to change the
1644 nature of the match, or to cause it fail earlier than it otherwise might, when
1645 the author of the pattern knows there is no point in carrying on.
1646 .P
1647 Consider, for example, the pattern \ed+foo when applied to the subject line
1648 .sp
1649 123456bar
1650 .sp
1651 After matching all 6 digits and then failing to match "foo", the normal
1652 action of the matcher is to try again with only 5 digits matching the \ed+
1653 item, and then with 4, and so on, before ultimately failing. "Atomic grouping"
1654 (a term taken from Jeffrey Friedl's book) provides the means for specifying
1655 that once a subpattern has matched, it is not to be re-evaluated in this way.
1656 .P
1657 If we use atomic grouping for the previous example, the matcher gives up
1658 immediately on failing to match "foo" the first time. The notation is a kind of
1659 special parenthesis, starting with (?> as in this example:
1660 .sp
1661 (?>\ed+)foo
1662 .sp
1663 This kind of parenthesis "locks up" the part of the pattern it contains once
1664 it has matched, and a failure further into the pattern is prevented from
1665 backtracking into it. Backtracking past it to previous items, however, works as
1666 normal.
1667 .P
1668 An alternative description is that a subpattern of this type matches the string
1669 of characters that an identical standalone pattern would match, if anchored at
1670 the current point in the subject string.
1671 .P
1672 Atomic grouping subpatterns are not capturing subpatterns. Simple cases such as
1673 the above example can be thought of as a maximizing repeat that must swallow
1674 everything it can. So, while both \ed+ and \ed+? are prepared to adjust the
1675 number of digits they match in order to make the rest of the pattern match,
1676 (?>\ed+) can only match an entire sequence of digits.
1677 .P
1678 Atomic groups in general can of course contain arbitrarily complicated
1679 subpatterns, and can be nested. However, when the subpattern for an atomic
1680 group is just a single repeated item, as in the example above, a simpler
1681 notation, called a "possessive quantifier" can be used. This consists of an
1682 additional + character following a quantifier. Using this notation, the
1683 previous example can be rewritten as
1684 .sp
1685 \ed++foo
1686 .sp
1687 Note that a possessive quantifier can be used with an entire group, for
1688 example:
1689 .sp
1690 (abc|xyz){2,3}+
1691 .sp
1692 Possessive quantifiers are always greedy; the setting of the PCRE_UNGREEDY
1693 option is ignored. They are a convenient notation for the simpler forms of
1694 atomic group. However, there is no difference in the meaning of a possessive
1695 quantifier and the equivalent atomic group, though there may be a performance
1696 difference; possessive quantifiers should be slightly faster.
1697 .P
1698 The possessive quantifier syntax is an extension to the Perl 5.8 syntax.
1699 Jeffrey Friedl originated the idea (and the name) in the first edition of his
1700 book. Mike McCloskey liked it, so implemented it when he built Sun's Java
1701 package, and PCRE copied it from there. It ultimately found its way into Perl
1702 at release 5.10.
1703 .P
1704 PCRE has an optimization that automatically "possessifies" certain simple
1705 pattern constructs. For example, the sequence A+B is treated as A++B because
1706 there is no point in backtracking into a sequence of A's when B must follow.
1707 .P
1708 When a pattern contains an unlimited repeat inside a subpattern that can itself
1709 be repeated an unlimited number of times, the use of an atomic group is the
1710 only way to avoid some failing matches taking a very long time indeed. The
1711 pattern
1712 .sp
1713 (\eD+|<\ed+>)*[!?]
1714 .sp
1715 matches an unlimited number of substrings that either consist of non-digits, or
1716 digits enclosed in <>, followed by either ! or ?. When it matches, it runs
1717 quickly. However, if it is applied to
1718 .sp
1719 aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
1720 .sp
1721 it takes a long time before reporting failure. This is because the string can
1722 be divided between the internal \eD+ repeat and the external * repeat in a
1723 large number of ways, and all have to be tried. (The example uses [!?] rather
1724 than a single character at the end, because both PCRE and Perl have an
1725 optimization that allows for fast failure when a single character is used. They
1726 remember the last single character that is required for a match, and fail early
1727 if it is not present in the string.) If the pattern is changed so that it uses
1728 an atomic group, like this:
1729 .sp
1730 ((?>\eD+)|<\ed+>)*[!?]
1731 .sp
1732 sequences of non-digits cannot be broken, and failure happens quickly.
1733 .
1734 .
1735 .\" HTML <a name="backreferences"></a>
1736 .SH "BACK REFERENCES"
1737 .rs
1738 .sp
1739 Outside a character class, a backslash followed by a digit greater than 0 (and
1740 possibly further digits) is a back reference to a capturing subpattern earlier
1741 (that is, to its left) in the pattern, provided there have been that many
1742 previous capturing left parentheses.
1743 .P
1744 However, if the decimal number following the backslash is less than 10, it is
1745 always taken as a back reference, and causes an error only if there are not
1746 that many capturing left parentheses in the entire pattern. In other words, the
1747 parentheses that are referenced need not be to the left of the reference for
1748 numbers less than 10. A "forward back reference" of this type can make sense
1749 when a repetition is involved and the subpattern to the right has participated
1750 in an earlier iteration.
1751 .P
1752 It is not possible to have a numerical "forward back reference" to a subpattern
1753 whose number is 10 or more using this syntax because a sequence such as \e50 is
1754 interpreted as a character defined in octal. See the subsection entitled
1755 "Non-printing characters"
1756 .\" HTML <a href="#digitsafterbackslash">
1757 .\" </a>
1758 above
1759 .\"
1760 for further details of the handling of digits following a backslash. There is
1761 no such problem when named parentheses are used. A back reference to any
1762 subpattern is possible using named parentheses (see below).
1763 .P
1764 Another way of avoiding the ambiguity inherent in the use of digits following a
1765 backslash is to use the \eg escape sequence. This escape must be followed by an
1766 unsigned number or a negative number, optionally enclosed in braces. These
1767 examples are all identical:
1768 .sp
1769 (ring), \e1
1770 (ring), \eg1
1771 (ring), \eg{1}
1772 .sp
1773 An unsigned number specifies an absolute reference without the ambiguity that
1774 is present in the older syntax. It is also useful when literal digits follow
1775 the reference. A negative number is a relative reference. Consider this
1776 example:
1777 .sp
1778 (abc(def)ghi)\eg{-1}
1779 .sp
1780 The sequence \eg{-1} is a reference to the most recently started capturing
1781 subpattern before \eg, that is, is it equivalent to \e2 in this example.
1782 Similarly, \eg{-2} would be equivalent to \e1. The use of relative references
1783 can be helpful in long patterns, and also in patterns that are created by
1784 joining together fragments that contain references within themselves.
1785 .P
1786 A back reference matches whatever actually matched the capturing subpattern in
1787 the current subject string, rather than anything matching the subpattern
1788 itself (see
1789 .\" HTML <a href="#subpatternsassubroutines">
1790 .\" </a>
1791 "Subpatterns as subroutines"
1792 .\"
1793 below for a way of doing that). So the pattern
1794 .sp
1795 (sens|respons)e and \e1ibility
1796 .sp
1797 matches "sense and sensibility" and "response and responsibility", but not
1798 "sense and responsibility". If caseful matching is in force at the time of the
1799 back reference, the case of letters is relevant. For example,
1800 .sp
1801 ((?i)rah)\es+\e1
1802 .sp
1803 matches "rah rah" and "RAH RAH", but not "RAH rah", even though the original
1804 capturing subpattern is matched caselessly.
1805 .P
1806 There are several different ways of writing back references to named
1807 subpatterns. The .NET syntax \ek{name} and the Perl syntax \ek<name> or
1808 \ek'name' are supported, as is the Python syntax (?P=name). Perl 5.10's unified
1809 back reference syntax, in which \eg can be used for both numeric and named
1810 references, is also supported. We could rewrite the above example in any of
1811 the following ways:
1812 .sp
1813 (?<p1>(?i)rah)\es+\ek<p1>
1814 (?'p1'(?i)rah)\es+\ek{p1}
1815 (?P<p1>(?i)rah)\es+(?P=p1)
1816 (?<p1>(?i)rah)\es+\eg{p1}
1817 .sp
1818 A subpattern that is referenced by name may appear in the pattern before or
1819 after the reference.
1820 .P
1821 There may be more than one back reference to the same subpattern. If a
1822 subpattern has not actually been used in a particular match, any back
1823 references to it always fail by default. For example, the pattern
1824 .sp
1825 (a|(bc))\e2
1826 .sp
1827 always fails if it starts to match "a" rather than "bc". However, if the
1828 PCRE_JAVASCRIPT_COMPAT option is set at compile time, a back reference to an
1829 unset value matches an empty string.
1830 .P
1831 Because there may be many capturing parentheses in a pattern, all digits
1832 following a backslash are taken as part of a potential back reference number.
1833 If the pattern continues with a digit character, some delimiter must be used to
1834 terminate the back reference. If the PCRE_EXTENDED option is set, this can be
1835 whitespace. Otherwise, the \eg{ syntax or an empty comment (see
1836 .\" HTML <a href="#comments">
1837 .\" </a>
1838 "Comments"
1839 .\"
1840 below) can be used.
1841 .
1842 .SS "Recursive back references"
1843 .rs
1844 .sp
1845 A back reference that occurs inside the parentheses to which it refers fails
1846 when the subpattern is first used, so, for example, (a\e1) never matches.
1847 However, such references can be useful inside repeated subpatterns. For
1848 example, the pattern
1849 .sp
1850 (a|b\e1)+
1851 .sp
1852 matches any number of "a"s and also "aba", "ababbaa" etc. At each iteration of
1853 the subpattern, the back reference matches the character string corresponding
1854 to the previous iteration. In order for this to work, the pattern must be such
1855 that the first iteration does not need to match the back reference. This can be
1856 done using alternation, as in the example above, or by a quantifier with a
1857 minimum of zero.
1858 .P
1859 Back references of this type cause the group that they reference to be treated
1860 as an
1861 .\" HTML <a href="#atomicgroup">
1862 .\" </a>
1863 atomic group.
1864 .\"
1865 Once the whole group has been matched, a subsequent matching failure cannot
1866 cause backtracking into the middle of the group.
1867 .
1868 .
1869 .\" HTML <a name="bigassertions"></a>
1870 .SH ASSERTIONS
1871 .rs
1872 .sp
1873 An assertion is a test on the characters following or preceding the current
1874 matching point that does not actually consume any characters. The simple
1875 assertions coded as \eb, \eB, \eA, \eG, \eZ, \ez, ^ and $ are described
1876 .\" HTML <a href="#smallassertions">
1877 .\" </a>
1878 above.
1879 .\"
1880 .P
1881 More complicated assertions are coded as subpatterns. There are two kinds:
1882 those that look ahead of the current position in the subject string, and those
1883 that look behind it. An assertion subpattern is matched in the normal way,
1884 except that it does not cause the current matching position to be changed.
1885 .P
1886 Assertion subpatterns are not capturing subpatterns. If such an assertion
1887 contains capturing subpatterns within it, these are counted for the purposes of
1888 numbering the capturing subpatterns in the whole pattern. However, substring
1889 capturing is carried out only for positive assertions, because it does not make
1890 sense for negative assertions.
1891 .P
1892 For compatibility with Perl, assertion subpatterns may be repeated; though
1893 it makes no sense to assert the same thing several times, the side effect of
1894 capturing parentheses may occasionally be useful. In practice, there only three
1895 cases:
1896 .sp
1897 (1) If the quantifier is {0}, the assertion is never obeyed during matching.
1898 However, it may contain internal capturing parenthesized groups that are called
1899 from elsewhere via the
1900 .\" HTML <a href="#subpatternsassubroutines">
1901 .\" </a>
1902 subroutine mechanism.
1903 .\"
1904 .sp
1905 (2) If quantifier is {0,n} where n is greater than zero, it is treated as if it
1906 were {0,1}. At run time, the rest of the pattern match is tried with and
1907 without the assertion, the order depending on the greediness of the quantifier.
1908 .sp
1909 (3) If the minimum repetition is greater than zero, the quantifier is ignored.
1910 The assertion is obeyed just once when encountered during matching.
1911 .
1912 .
1913 .SS "Lookahead assertions"
1914 .rs
1915 .sp
1916 Lookahead assertions start with (?= for positive assertions and (?! for
1917 negative assertions. For example,
1918 .sp
1919 \ew+(?=;)
1920 .sp
1921 matches a word followed by a semicolon, but does not include the semicolon in
1922 the match, and
1923 .sp
1924 foo(?!bar)
1925 .sp
1926 matches any occurrence of "foo" that is not followed by "bar". Note that the
1927 apparently similar pattern
1928 .sp
1929 (?!foo)bar
1930 .sp
1931 does not find an occurrence of "bar" that is preceded by something other than
1932 "foo"; it finds any occurrence of "bar" whatsoever, because the assertion
1933 (?!foo) is always true when the next three characters are "bar". A
1934 lookbehind assertion is needed to achieve the other effect.
1935 .P
1936 If you want to force a matching failure at some point in a pattern, the most
1937 convenient way to do it is with (?!) because an empty string always matches, so
1938 an assertion that requires there not to be an empty string must always fail.
1939 The backtracking control verb (*FAIL) or (*F) is a synonym for (?!).
1940 .
1941 .
1942 .\" HTML <a name="lookbehind"></a>
1943 .SS "Lookbehind assertions"
1944 .rs
1945 .sp
1946 Lookbehind assertions start with (?<= for positive assertions and (?<! for
1947 negative assertions. For example,
1948 .sp
1949 (?<!foo)bar
1950 .sp
1951 does find an occurrence of "bar" that is not preceded by "foo". The contents of
1952 a lookbehind assertion are restricted such that all the strings it matches must
1953 have a fixed length. However, if there are several top-level alternatives, they
1954 do not all have to have the same fixed length. Thus
1955 .sp
1956 (?<=bullock|donkey)
1957 .sp
1958 is permitted, but
1959 .sp
1960 (?<!dogs?|cats?)
1961 .sp
1962 causes an error at compile time. Branches that match different length strings
1963 are permitted only at the top level of a lookbehind assertion. This is an
1964 extension compared with Perl, which requires all branches to match the same
1965 length of string. An assertion such as
1966 .sp
1967 (?<=ab(c|de))
1968 .sp
1969 is not permitted, because its single top-level branch can match two different
1970 lengths, but it is acceptable to PCRE if rewritten to use two top-level
1971 branches:
1972 .sp
1973 (?<=abc|abde)
1974 .sp
1975 In some cases, the escape sequence \eK
1976 .\" HTML <a href="#resetmatchstart">
1977 .\" </a>
1978 (see above)
1979 .\"
1980 can be used instead of a lookbehind assertion to get round the fixed-length
1981 restriction.
1982 .P
1983 The implementation of lookbehind assertions is, for each alternative, to
1984 temporarily move the current position back by the fixed length and then try to
1985 match. If there are insufficient characters before the current position, the
1986 assertion fails.
1987 .P
1988 In a UTF mode, PCRE does not allow the \eC escape (which matches a single data
1989 unit even in a UTF mode) to appear in lookbehind assertions, because it makes
1990 it impossible to calculate the length of the lookbehind. The \eX and \eR
1991 escapes, which can match different numbers of data units, are also not
1992 permitted.
1993 .P
1994 .\" HTML <a href="#subpatternsassubroutines">
1995 .\" </a>
1996 "Subroutine"
1997 .\"
1998 calls (see below) such as (?2) or (?&X) are permitted in lookbehinds, as long
1999 as the subpattern matches a fixed-length string.
2000 .\" HTML <a href="#recursion">
2001 .\" </a>
2002 Recursion,
2003 .\"
2004 however, is not supported.
2005 .P
2006 Possessive quantifiers can be used in conjunction with lookbehind assertions to
2007 specify efficient matching of fixed-length strings at the end of subject
2008 strings. Consider a simple pattern such as
2009 .sp
2010 abcd$
2011 .sp
2012 when applied to a long string that does not match. Because matching proceeds
2013 from left to right, PCRE will look for each "a" in the subject and then see if
2014 what follows matches the rest of the pattern. If the pattern is specified as
2015 .sp
2016 ^.*abcd$
2017 .sp
2018 the initial .* matches the entire string at first, but when this fails (because
2019 there is no following "a"), it backtracks to match all but the last character,
2020 then all but the last two characters, and so on. Once again the search for "a"
2021 covers the entire string, from right to left, so we are no better off. However,
2022 if the pattern is written as
2023 .sp
2024 ^.*+(?<=abcd)
2025 .sp
2026 there can be no backtracking for the .*+ item; it can match only the entire
2027 string. The subsequent lookbehind assertion does a single test on the last four
2028 characters. If it fails, the match fails immediately. For long strings, this
2029 approach makes a significant difference to the processing time.
2030 .
2031 .
2032 .SS "Using multiple assertions"
2033 .rs
2034 .sp
2035 Several assertions (of any sort) may occur in succession. For example,
2036 .sp
2037 (?<=\ed{3})(?<!999)foo
2038 .sp
2039 matches "foo" preceded by three digits that are not "999". Notice that each of
2040 the assertions is applied independently at the same point in the subject
2041 string. First there is a check that the previous three characters are all
2042 digits, and then there is a check that the same three characters are not "999".
2043 This pattern does \fInot\fP match "foo" preceded by six characters, the first
2044 of which are digits and the last three of which are not "999". For example, it
2045 doesn't match "123abcfoo". A pattern to do that is
2046 .sp
2047 (?<=\ed{3}...)(?<!999)foo
2048 .sp
2049 This time the first assertion looks at the preceding six characters, checking
2050 that the first three are digits, and then the second assertion checks that the
2051 preceding three characters are not "999".
2052 .P
2053 Assertions can be nested in any combination. For example,
2054 .sp
2055 (?<=(?<!foo)bar)baz
2056 .sp
2057 matches an occurrence of "baz" that is preceded by "bar" which in turn is not
2058 preceded by "foo", while
2059 .sp
2060 (?<=\ed{3}(?!999)...)foo
2061 .sp
2062 is another pattern that matches "foo" preceded by three digits and any three
2063 characters that are not "999".
2064 .
2065 .
2066 .\" HTML <a name="conditions"></a>
2067 .SH "CONDITIONAL SUBPATTERNS"
2068 .rs
2069 .sp
2070 It is possible to cause the matching process to obey a subpattern
2071 conditionally or to choose between two alternative subpatterns, depending on
2072 the result of an assertion, or whether a specific capturing subpattern has
2073 already been matched. The two possible forms of conditional subpattern are:
2074 .sp
2075 (?(condition)yes-pattern)
2076 (?(condition)yes-pattern|no-pattern)
2077 .sp
2078 If the condition is satisfied, the yes-pattern is used; otherwise the
2079 no-pattern (if present) is used. If there are more than two alternatives in the
2080 subpattern, a compile-time error occurs. Each of the two alternatives may
2081 itself contain nested subpatterns of any form, including conditional
2082 subpatterns; the restriction to two alternatives applies only at the level of
2083 the condition. This pattern fragment is an example where the alternatives are
2084 complex:
2085 .sp
2086 (?(1) (A|B|C) | (D | (?(2)E|F) | E) )
2087 .sp
2088 .P
2089 There are four kinds of condition: references to subpatterns, references to
2090 recursion, a pseudo-condition called DEFINE, and assertions.
2091 .
2092 .SS "Checking for a used subpattern by number"
2093 .rs
2094 .sp
2095 If the text between the parentheses consists of a sequence of digits, the
2096 condition is true if a capturing subpattern of that number has previously
2097 matched. If there is more than one capturing subpattern with the same number
2098 (see the earlier
2099 .\"
2100 .\" HTML <a href="#recursion">
2101 .\" </a>
2102 section about duplicate subpattern numbers),
2103 .\"
2104 the condition is true if any of them have matched. An alternative notation is
2105 to precede the digits with a plus or minus sign. In this case, the subpattern
2106 number is relative rather than absolute. The most recently opened parentheses
2107 can be referenced by (?(-1), the next most recent by (?(-2), and so on. Inside
2108 loops it can also make sense to refer to subsequent groups. The next
2109 parentheses to be opened can be referenced as (?(+1), and so on. (The value
2110 zero in any of these forms is not used; it provokes a compile-time error.)
2111 .P
2112 Consider the following pattern, which contains non-significant white space to
2113 make it more readable (assume the PCRE_EXTENDED option) and to divide it into
2114 three parts for ease of discussion:
2115 .sp
2116 ( \e( )? [^()]+ (?(1) \e) )
2117 .sp
2118 The first part matches an optional opening parenthesis, and if that
2119 character is present, sets it as the first captured substring. The second part
2120 matches one or more characters that are not parentheses. The third part is a
2121 conditional subpattern that tests whether or not the first set of parentheses
2122 matched. If they did, that is, if subject started with an opening parenthesis,
2123 the condition is true, and so the yes-pattern is executed and a closing
2124 parenthesis is required. Otherwise, since no-pattern is not present, the
2125 subpattern matches nothing. In other words, this pattern matches a sequence of
2126 non-parentheses, optionally enclosed in parentheses.
2127 .P
2128 If you were embedding this pattern in a larger one, you could use a relative
2129 reference:
2130 .sp
2131 ...other stuff... ( \e( )? [^()]+ (?(-1) \e) ) ...
2132 .sp
2133 This makes the fragment independent of the parentheses in the larger pattern.
2134 .
2135 .SS "Checking for a used subpattern by name"
2136 .rs
2137 .sp
2138 Perl uses the syntax (?(<name>)...) or (?('name')...) to test for a used
2139 subpattern by name. For compatibility with earlier versions of PCRE, which had
2140 this facility before Perl, the syntax (?(name)...) is also recognized. However,
2141 there is a possible ambiguity with this syntax, because subpattern names may
2142 consist entirely of digits. PCRE looks first for a named subpattern; if it
2143 cannot find one and the name consists entirely of digits, PCRE looks for a
2144 subpattern of that number, which must be greater than zero. Using subpattern
2145 names that consist entirely of digits is not recommended.
2146 .P
2147 Rewriting the above example to use a named subpattern gives this:
2148 .sp
2149 (?<OPEN> \e( )? [^()]+ (?(<OPEN>) \e) )
2150 .sp
2151 If the name used in a condition of this kind is a duplicate, the test is
2152 applied to all subpatterns of the same name, and is true if any one of them has
2153 matched.
2154 .
2155 .SS "Checking for pattern recursion"
2156 .rs
2157 .sp
2158 If the condition is the string (R), and there is no subpattern with the name R,
2159 the condition is true if a recursive call to the whole pattern or any
2160 subpattern has been made. If digits or a name preceded by ampersand follow the
2161 letter R, for example:
2162 .sp
2163 (?(R3)...) or (?(R&name)...)
2164 .sp
2165 the condition is true if the most recent recursion is into a subpattern whose
2166 number or name is given. This condition does not check the entire recursion
2167 stack. If the name used in a condition of this kind is a duplicate, the test is
2168 applied to all subpatterns of the same name, and is true if any one of them is
2169 the most recent recursion.
2170 .P
2171 At "top level", all these recursion test conditions are false.
2172 .\" HTML <a href="#recursion">
2173 .\" </a>
2174 The syntax for recursive patterns
2175 .\"
2176 is described below.
2177 .
2178 .\" HTML <a name="subdefine"></a>
2179 .SS "Defining subpatterns for use by reference only"
2180 .rs
2181 .sp
2182 If the condition is the string (DEFINE), and there is no subpattern with the
2183 name DEFINE, the condition is always false. In this case, there may be only one
2184 alternative in the subpattern. It is always skipped if control reaches this
2185 point in the pattern; the idea of DEFINE is that it can be used to define
2186 subroutines that can be referenced from elsewhere. (The use of
2187 .\" HTML <a href="#subpatternsassubroutines">
2188 .\" </a>
2189 subroutines
2190 .\"
2191 is described below.) For example, a pattern to match an IPv4 address such as
2192 "192.168.23.245" could be written like this (ignore whitespace and line
2193 breaks):
2194 .sp
2195 (?(DEFINE) (?<byte> 2[0-4]\ed | 25[0-5] | 1\ed\ed | [1-9]?\ed) )
2196 \eb (?&byte) (\e.(?&byte)){3} \eb
2197 .sp
2198 The first part of the pattern is a DEFINE group inside which a another group
2199 named "byte" is defined. This matches an individual component of an IPv4
2200 address (a number less than 256). When matching takes place, this part of the
2201 pattern is skipped because DEFINE acts like a false condition. The rest of the
2202 pattern uses references to the named group to match the four dot-separated
2203 components of an IPv4 address, insisting on a word boundary at each end.
2204 .
2205 .SS "Assertion conditions"
2206 .rs
2207 .sp
2208 If the condition is not in any of the above formats, it must be an assertion.
2209 This may be a positive or negative lookahead or lookbehind assertion. Consider
2210 this pattern, again containing non-significant white space, and with the two
2211 alternatives on the second line:
2212 .sp
2213 (?(?=[^a-z]*[a-z])
2214 \ed{2}-[a-z]{3}-\ed{2} | \ed{2}-\ed{2}-\ed{2} )
2215 .sp
2216 The condition is a positive lookahead assertion that matches an optional
2217 sequence of non-letters followed by a letter. In other words, it tests for the
2218 presence of at least one letter in the subject. If a letter is found, the
2219 subject is matched against the first alternative; otherwise it is matched
2220 against the second. This pattern matches strings in one of the two forms
2221 dd-aaa-dd or dd-dd-dd, where aaa are letters and dd are digits.
2222 .
2223 .
2224 .\" HTML <a name="comments"></a>
2225 .SH COMMENTS
2226 .rs
2227 .sp
2228 There are two ways of including comments in patterns that are processed by
2229 PCRE. In both cases, the start of the comment must not be in a character class,
2230 nor in the middle of any other sequence of related characters such as (?: or a
2231 subpattern name or number. The characters that make up a comment play no part
2232 in the pattern matching.
2233 .P
2234 The sequence (?# marks the start of a comment that continues up to the next
2235 closing parenthesis. Nested parentheses are not permitted. If the PCRE_EXTENDED
2236 option is set, an unescaped # character also introduces a comment, which in
2237 this case continues to immediately after the next newline character or
2238 character sequence in the pattern. Which characters are interpreted as newlines
2239 is controlled by the options passed to a compiling function or by a special
2240 sequence at the start of the pattern, as described in the section entitled
2241 .\" HTML <a href="#newlines">
2242 .\" </a>
2243 "Newline conventions"
2244 .\"
2245 above. Note that the end of this type of comment is a literal newline sequence
2246 in the pattern; escape sequences that happen to represent a newline do not
2247 count. For example, consider this pattern when PCRE_EXTENDED is set, and the
2248 default newline convention is in force:
2249 .sp
2250 abc #comment \en still comment
2251 .sp
2252 On encountering the # character, \fBpcre_compile()\fP skips along, looking for
2253 a newline in the pattern. The sequence \en is still literal at this stage, so
2254 it does not terminate the comment. Only an actual character with the code value
2255 0x0a (the default newline) does so.
2256 .
2257 .
2258 .\" HTML <a name="recursion"></a>
2259 .SH "RECURSIVE PATTERNS"
2260 .rs
2261 .sp
2262 Consider the problem of matching a string in parentheses, allowing for
2263 unlimited nested parentheses. Without the use of recursion, the best that can
2264 be done is to use a pattern that matches up to some fixed depth of nesting. It
2265 is not possible to handle an arbitrary nesting depth.
2266 .P
2267 For some time, Perl has provided a facility that allows regular expressions to
2268 recurse (amongst other things). It does this by interpolating Perl code in the
2269 expression at run time, and the code can refer to the expression itself. A Perl
2270 pattern using code interpolation to solve the parentheses problem can be
2271 created like this:
2272 .sp
2273 $re = qr{\e( (?: (?>[^()]+) | (?p{$re}) )* \e)}x;
2274 .sp
2275 The (?p{...}) item interpolates Perl code at run time, and in this case refers
2276 recursively to the pattern in which it appears.
2277 .P
2278 Obviously, PCRE cannot support the interpolation of Perl code. Instead, it
2279 supports special syntax for recursion of the entire pattern, and also for
2280 individual subpattern recursion. After its introduction in PCRE and Python,
2281 this kind of recursion was subsequently introduced into Perl at release 5.10.
2282 .P
2283 A special item that consists of (? followed by a number greater than zero and a
2284 closing parenthesis is a recursive subroutine call of the subpattern of the
2285 given number, provided that it occurs inside that subpattern. (If not, it is a
2286 .\" HTML <a href="#subpatternsassubroutines">
2287 .\" </a>
2288 non-recursive subroutine
2289 .\"
2290 call, which is described in the next section.) The special item (?R) or (?0) is
2291 a recursive call of the entire regular expression.
2292 .P
2293 This PCRE pattern solves the nested parentheses problem (assume the
2294 PCRE_EXTENDED option is set so that white space is ignored):
2295 .sp
2296 \e( ( [^()]++ | (?R) )* \e)
2297 .sp
2298 First it matches an opening parenthesis. Then it matches any number of
2299 substrings which can either be a sequence of non-parentheses, or a recursive
2300 match of the pattern itself (that is, a correctly parenthesized substring).
2301 Finally there is a closing parenthesis. Note the use of a possessive quantifier
2302 to avoid backtracking into sequences of non-parentheses.
2303 .P
2304 If this were part of a larger pattern, you would not want to recurse the entire
2305 pattern, so instead you could use this:
2306 .sp
2307 ( \e( ( [^()]++ | (?1) )* \e) )
2308 .sp
2309 We have put the pattern into parentheses, and caused the recursion to refer to
2310 them instead of the whole pattern.
2311 .P
2312 In a larger pattern, keeping track of parenthesis numbers can be tricky. This
2313 is made easier by the use of relative references. Instead of (?1) in the
2314 pattern above you can write (?-2) to refer to the second most recently opened
2315 parentheses preceding the recursion. In other words, a negative number counts
2316 capturing parentheses leftwards from the point at which it is encountered.
2317 .P
2318 It is also possible to refer to subsequently opened parentheses, by writing
2319 references such as (?+2). However, these cannot be recursive because the
2320 reference is not inside the parentheses that are referenced. They are always
2321 .\" HTML <a href="#subpatternsassubroutines">
2322 .\" </a>
2323 non-recursive subroutine
2324 .\"
2325 calls, as described in the next section.
2326 .P
2327 An alternative approach is to use named parentheses instead. The Perl syntax
2328 for this is (?&name); PCRE's earlier syntax (?P>name) is also supported. We
2329 could rewrite the above example as follows:
2330 .sp
2331 (?<pn> \e( ( [^()]++ | (?&pn) )* \e) )
2332 .sp
2333 If there is more than one subpattern with the same name, the earliest one is
2334 used.
2335 .P
2336 This particular example pattern that we have been looking at contains nested
2337 unlimited repeats, and so the use of a possessive quantifier for matching
2338 strings of non-parentheses is important when applying the pattern to strings
2339 that do not match. For example, when this pattern is applied to
2340 .sp
2341 (aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa()
2342 .sp
2343 it yields "no match" quickly. However, if a possessive quantifier is not used,
2344 the match runs for a very long time indeed because there are so many different
2345 ways the + and * repeats can carve up the subject, and all have to be tested
2346 before failure can be reported.
2347 .P
2348 At the end of a match, the values of capturing parentheses are those from
2349 the outermost level. If you want to obtain intermediate values, a callout
2350 function can be used (see below and the
2351 .\" HREF
2352 \fBpcrecallout\fP
2353 .\"
2354 documentation). If the pattern above is matched against
2355 .sp
2356 (ab(cd)ef)
2357 .sp
2358 the value for the inner capturing parentheses (numbered 2) is "ef", which is
2359 the last value taken on at the top level. If a capturing subpattern is not
2360 matched at the top level, its final captured value is unset, even if it was
2361 (temporarily) set at a deeper level during the matching process.
2362 .P
2363 If there are more than 15 capturing parentheses in a pattern, PCRE has to
2364 obtain extra memory to store data during a recursion, which it does by using
2365 \fBpcre_malloc\fP, freeing it via \fBpcre_free\fP afterwards. If no memory can
2366 be obtained, the match fails with the PCRE_ERROR_NOMEMORY error.
2367 .P
2368 Do not confuse the (?R) item with the condition (R), which tests for recursion.
2369 Consider this pattern, which matches text in angle brackets, allowing for
2370 arbitrary nesting. Only digits are allowed in nested brackets (that is, when
2371 recursing), whereas any characters are permitted at the outer level.
2372 .sp
2373 < (?: (?(R) \ed++ | [^<>]*+) | (?R)) * >
2374 .sp
2375 In this pattern, (?(R) is the start of a conditional subpattern, with two
2376 different alternatives for the recursive and non-recursive cases. The (?R) item
2377 is the actual recursive call.
2378 .
2379 .
2380 .\" HTML <a name="recursiondifference"></a>
2381 .SS "Differences in recursion processing between PCRE and Perl"
2382 .rs
2383 .sp
2384 Recursion processing in PCRE differs from Perl in two important ways. In PCRE
2385 (like Python, but unlike Perl), a recursive subpattern call is always treated
2386 as an atomic group. That is, once it has matched some of the subject string, it
2387 is never re-entered, even if it contains untried alternatives and there is a
2388 subsequent matching failure. This can be illustrated by the following pattern,
2389 which purports to match a palindromic string that contains an odd number of
2390 characters (for example, "a", "aba", "abcba", "abcdcba"):
2391 .sp
2392 ^(.|(.)(?1)\e2)$
2393 .sp
2394 The idea is that it either matches a single character, or two identical
2395 characters surrounding a sub-palindrome. In Perl, this pattern works; in PCRE
2396 it does not if the pattern is longer than three characters. Consider the
2397 subject string "abcba":
2398 .P
2399 At the top level, the first character is matched, but as it is not at the end
2400 of the string, the first alternative fails; the second alternative is taken
2401 and the recursion kicks in. The recursive call to subpattern 1 successfully
2402 matches the next character ("b"). (Note that the beginning and end of line
2403 tests are not part of the recursion).
2404 .P
2405 Back at the top level, the next character ("c") is compared with what
2406 subpattern 2 matched, which was "a". This fails. Because the recursion is
2407 treated as an atomic group, there are now no backtracking points, and so the
2408 entire match fails. (Perl is able, at this point, to re-enter the recursion and
2409 try the second alternative.) However, if the pattern is written with the
2410 alternatives in the other order, things are different:
2411 .sp
2412 ^((.)(?1)\e2|.)$
2413 .sp
2414 This time, the recursing alternative is tried first, and continues to recurse
2415 until it runs out of characters, at which point the recursion fails. But this
2416 time we do have another alternative to try at the higher level. That is the big
2417 difference: in the previous case the remaining alternative is at a deeper
2418 recursion level, which PCRE cannot use.
2419 .P
2420 To change the pattern so that it matches all palindromic strings, not just
2421 those with an odd number of characters, it is tempting to change the pattern to
2422 this:
2423 .sp
2424 ^((.)(?1)\e2|.?)$
2425 .sp
2426 Again, this works in Perl, but not in PCRE, and for the same reason. When a
2427 deeper recursion has matched a single character, it cannot be entered again in
2428 order to match an empty string. The solution is to separate the two cases, and
2429 write out the odd and even cases as alternatives at the higher level:
2430 .sp
2431 ^(?:((.)(?1)\e2|)|((.)(?3)\e4|.))
2432 .sp
2433 If you want to match typical palindromic phrases, the pattern has to ignore all
2434 non-word characters, which can be done like this:
2435 .sp
2436 ^\eW*+(?:((.)\eW*+(?1)\eW*+\e2|)|((.)\eW*+(?3)\eW*+\e4|\eW*+.\eW*+))\eW*+$
2437 .sp
2438 If run with the PCRE_CASELESS option, this pattern matches phrases such as "A
2439 man, a plan, a canal: Panama!" and it works well in both PCRE and Perl. Note
2440 the use of the possessive quantifier *+ to avoid backtracking into sequences of
2441 non-word characters. Without this, PCRE takes a great deal longer (ten times or
2442 more) to match typical phrases, and Perl takes so long that you think it has
2443 gone into a loop.
2444 .P
2445 \fBWARNING\fP: The palindrome-matching patterns above work only if the subject
2446 string does not start with a palindrome that is shorter than the entire string.
2447 For example, although "abcba" is correctly matched, if the subject is "ababa",
2448 PCRE finds the palindrome "aba" at the start, then fails at top level because
2449 the end of the string does not follow. Once again, it cannot jump back into the
2450 recursion to try other alternatives, so the entire match fails.
2451 .P
2452 The second way in which PCRE and Perl differ in their recursion processing is
2453 in the handling of captured values. In Perl, when a subpattern is called
2454 recursively or as a subpattern (see the next section), it has no access to any
2455 values that were captured outside the recursion, whereas in PCRE these values
2456 can be referenced. Consider this pattern:
2457 .sp
2458 ^(.)(\e1|a(?2))
2459 .sp
2460 In PCRE, this pattern matches "bab". The first capturing parentheses match "b",
2461 then in the second group, when the back reference \e1 fails to match "b", the
2462 second alternative matches "a" and then recurses. In the recursion, \e1 does
2463 now match "b" and so the whole match succeeds. In Perl, the pattern fails to
2464 match because inside the recursive call \e1 cannot access the externally set
2465 value.
2466 .
2467 .
2468 .\" HTML <a name="subpatternsassubroutines"></a>
2469 .SH "SUBPATTERNS AS SUBROUTINES"
2470 .rs
2471 .sp
2472 If the syntax for a recursive subpattern call (either by number or by
2473 name) is used outside the parentheses to which it refers, it operates like a
2474 subroutine in a programming language. The called subpattern may be defined
2475 before or after the reference. A numbered reference can be absolute or
2476 relative, as in these examples:
2477 .sp
2478 (...(absolute)...)...(?2)...
2479 (...(relative)...)...(?-1)...
2480 (...(?+1)...(relative)...
2481 .sp
2482 An earlier example pointed out that the pattern
2483 .sp
2484 (sens|respons)e and \e1ibility
2485 .sp
2486 matches "sense and sensibility" and "response and responsibility", but not
2487 "sense and responsibility". If instead the pattern
2488 .sp
2489 (sens|respons)e and (?1)ibility
2490 .sp
2491 is used, it does match "sense and responsibility" as well as the other two
2492 strings. Another example is given in the discussion of DEFINE above.
2493 .P
2494 All subroutine calls, whether recursive or not, are always treated as atomic
2495 groups. That is, once a subroutine has matched some of the subject string, it
2496 is never re-entered, even if it contains untried alternatives and there is a
2497 subsequent matching failure. Any capturing parentheses that are set during the
2498 subroutine call revert to their previous values afterwards.
2499 .P
2500 Processing options such as case-independence are fixed when a subpattern is
2501 defined, so if it is used as a subroutine, such options cannot be changed for
2502 different calls. For example, consider this pattern:
2503 .sp
2504 (abc)(?i:(?-1))
2505 .sp
2506 It matches "abcabc". It does not match "abcABC" because the change of
2507 processing option does not affect the called subpattern.
2508 .
2509 .
2510 .\" HTML <a name="onigurumasubroutines"></a>
2511 .SH "ONIGURUMA SUBROUTINE SYNTAX"
2512 .rs
2513 .sp
2514 For compatibility with Oniguruma, the non-Perl syntax \eg followed by a name or
2515 a number enclosed either in angle brackets or single quotes, is an alternative
2516 syntax for referencing a subpattern as a subroutine, possibly recursively. Here
2517 are two of the examples used above, rewritten using this syntax:
2518 .sp
2519 (?<pn> \e( ( (?>[^()]+) | \eg<pn> )* \e) )
2520 (sens|respons)e and \eg'1'ibility
2521 .sp
2522 PCRE supports an extension to Oniguruma: if a number is preceded by a
2523 plus or a minus sign it is taken as a relative reference. For example:
2524 .sp
2525 (abc)(?i:\eg<-1>)
2526 .sp
2527 Note that \eg{...} (Perl syntax) and \eg<...> (Oniguruma syntax) are \fInot\fP
2528 synonymous. The former is a back reference; the latter is a subroutine call.
2529 .
2530 .
2531 .SH CALLOUTS
2532 .rs
2533 .sp
2534 Perl has a feature whereby using the sequence (?{...}) causes arbitrary Perl
2535 code to be obeyed in the middle of matching a regular expression. This makes it
2536 possible, amongst other things, to extract different substrings that match the
2537 same pair of parentheses when there is a repetition.
2538 .P
2539 PCRE provides a similar feature, but of course it cannot obey arbitrary Perl
2540 code. The feature is called "callout". The caller of PCRE provides an external
2541 function by putting its entry point in the global variable \fIpcre_callout\fP
2542 (8-bit library) or \fIpcre16_callout\fP (16-bit library). By default, this
2543 variable contains NULL, which disables all calling out.
2544 .P
2545 Within a regular expression, (?C) indicates the points at which the external
2546 function is to be called. If you want to identify different callout points, you
2547 can put a number less than 256 after the letter C. The default value is zero.
2548 For example, this pattern has two callout points:
2549 .sp
2550 (?C1)abc(?C2)def
2551 .sp
2552 If the PCRE_AUTO_CALLOUT flag is passed to a compiling function, callouts are
2553 automatically installed before each item in the pattern. They are all numbered
2554 255.
2555 .P
2556 During matching, when PCRE reaches a callout point, the external function is
2557 called. It is provided with the number of the callout, the position in the
2558 pattern, and, optionally, one item of data originally supplied by the caller of
2559 the matching function. The callout function may cause matching to proceed, to
2560 backtrack, or to fail altogether. A complete description of the interface to
2561 the callout function is given in the
2562 .\" HREF
2563 \fBpcrecallout\fP
2564 .\"
2565 documentation.
2566 .
2567 .
2568 .\" HTML <a name="backtrackcontrol"></a>
2569 .SH "BACKTRACKING CONTROL"
2570 .rs
2571 .sp
2572 Perl 5.10 introduced a number of "Special Backtracking Control Verbs", which
2573 are described in the Perl documentation as "experimental and subject to change
2574 or removal in a future version of Perl". It goes on to say: "Their usage in
2575 production code should be noted to avoid problems during upgrades." The same
2576 remarks apply to the PCRE features described in this section.
2577 .P
2578 Since these verbs are specifically related to backtracking, most of them can be
2579 used only when the pattern is to be matched using one of the traditional
2580 matching functions, which use a backtracking algorithm. With the exception of
2581 (*FAIL), which behaves like a failing negative assertion, they cause an error
2582 if encountered by a DFA matching function.
2583 .P
2584 If any of these verbs are used in an assertion or in a subpattern that is
2585 called as a subroutine (whether or not recursively), their effect is confined
2586 to that subpattern; it does not extend to the surrounding pattern, with one
2587 exception: the name from a *(MARK), (*PRUNE), or (*THEN) that is encountered in
2588 a successful positive assertion \fIis\fP passed back when a match succeeds
2589 (compare capturing parentheses in assertions). Note that such subpatterns are
2590 processed as anchored at the point where they are tested. Note also that Perl's
2591 treatment of subroutines is different in some cases.
2592 .P
2593 The new verbs make use of what was previously invalid syntax: an opening
2594 parenthesis followed by an asterisk. They are generally of the form
2595 (*VERB) or (*VERB:NAME). Some may take either form, with differing behaviour,
2596 depending on whether or not an argument is present. A name is any sequence of
2597 characters that does not include a closing parenthesis. If the name is empty,
2598 that is, if the closing parenthesis immediately follows the colon, the effect
2599 is as if the colon were not there. Any number of these verbs may occur in a
2600 pattern.
2601 .P
2602 PCRE contains some optimizations that are used to speed up matching by running
2603 some checks at the start of each match attempt. For example, it may know the
2604 minimum length of matching subject, or that a particular character must be
2605 present. When one of these optimizations suppresses the running of a match, any
2606 included backtracking verbs will not, of course, be processed. You can suppress
2607 the start-of-match optimizations by setting the PCRE_NO_START_OPTIMIZE option
2608 when calling \fBpcre_compile()\fP or \fBpcre_exec()\fP, or by starting the
2609 pattern with (*NO_START_OPT).
2610 .P
2611 Experiments with Perl suggest that it too has similar optimizations, sometimes
2612 leading to anomalous results.
2613 .
2614 .
2615 .SS "Verbs that act immediately"
2616 .rs
2617 .sp
2618 The following verbs act as soon as they are encountered. They may not be
2619 followed by a name.
2620 .sp
2621 (*ACCEPT)
2622 .sp
2623 This verb causes the match to end successfully, skipping the remainder of the
2624 pattern. However, when it is inside a subpattern that is called as a
2625 subroutine, only that subpattern is ended successfully. Matching then continues
2626 at the outer level. If (*ACCEPT) is inside capturing parentheses, the data so
2627 far is captured. For example:
2628 .sp
2629 A((?:A|B(*ACCEPT)|C)D)
2630 .sp
2631 This matches "AB", "AAD", or "ACD"; when it matches "AB", "B" is captured by
2632 the outer parentheses.
2633 .sp
2634 (*FAIL) or (*F)
2635 .sp
2636 This verb causes a matching failure, forcing backtracking to occur. It is
2637 equivalent to (?!) but easier to read. The Perl documentation notes that it is
2638 probably useful only when combined with (?{}) or (??{}). Those are, of course,
2639 Perl features that are not present in PCRE. The nearest equivalent is the
2640 callout feature, as for example in this pattern:
2641 .sp
2642 a+(?C)(*FAIL)
2643 .sp
2644 A match with the string "aaaa" always fails, but the callout is taken before
2645 each backtrack happens (in this example, 10 times).
2646 .
2647 .
2648 .SS "Recording which path was taken"
2649 .rs
2650 .sp
2651 There is one verb whose main purpose is to track how a match was arrived at,
2652 though it also has a secondary use in conjunction with advancing the match
2653 starting point (see (*SKIP) below).
2654 .sp
2655 (*MARK:NAME) or (*:NAME)
2656 .sp
2657 A name is always required with this verb. There may be as many instances of
2658 (*MARK) as you like in a pattern, and their names do not have to be unique.
2659 .P
2660 When a match succeeds, the name of the last-encountered (*MARK) on the matching
2661 path is passed back to the caller as described in the section entitled
2662 .\" HTML <a href="pcreapi.html#extradata">
2663 .\" </a>
2664 "Extra data for \fBpcre_exec()\fP"
2665 .\"
2666 in the
2667 .\" HREF
2668 \fBpcreapi\fP
2669 .\"
2670 documentation. Here is an example of \fBpcretest\fP output, where the /K
2671 modifier requests the retrieval and outputting of (*MARK) data:
2672 .sp
2673 re> /X(*MARK:A)Y|X(*MARK:B)Z/K
2674 data> XY
2675 0: XY
2676 MK: A
2677 XZ
2678 0: XZ
2679 MK: B
2680 .sp
2681 The (*MARK) name is tagged with "MK:" in this output, and in this example it
2682 indicates which of the two alternatives matched. This is a more efficient way
2683 of obtaining this information than putting each alternative in its own
2684 capturing parentheses.
2685 .P
2686 If (*MARK) is encountered in a positive assertion, its name is recorded and
2687 passed back if it is the last-encountered. This does not happen for negative
2688 assertions.
2689 .P
2690 After a partial match or a failed match, the name of the last encountered
2691 (*MARK) in the entire match process is returned. For example:
2692 .sp
2693 re> /X(*MARK:A)Y|X(*MARK:B)Z/K
2694 data> XP
2695 No match, mark = B
2696 .sp
2697 Note that in this unanchored example the mark is retained from the match
2698 attempt that started at the letter "X". Subsequent match attempts starting at
2699 "P" and then with an empty string do not get as far as the (*MARK) item, but
2700 nevertheless do not reset it.
2701 .
2702 .
2703 .SS "Verbs that act after backtracking"
2704 .rs
2705 .sp
2706 The following verbs do nothing when they are encountered. Matching continues
2707 with what follows, but if there is no subsequent match, causing a backtrack to
2708 the verb, a failure is forced. That is, backtracking cannot pass to the left of
2709 the verb. However, when one of these verbs appears inside an atomic group, its
2710 effect is confined to that group, because once the group has been matched,
2711 there is never any backtracking into it. In this situation, backtracking can
2712 "jump back" to the left of the entire atomic group. (Remember also, as stated
2713 above, that this localization also applies in subroutine calls and assertions.)
2714 .P
2715 These verbs differ in exactly what kind of failure occurs when backtracking
2716 reaches them.
2717 .sp
2718 (*COMMIT)
2719 .sp
2720 This verb, which may not be followed by a name, causes the whole match to fail
2721 outright if the rest of the pattern does not match. Even if the pattern is
2722 unanchored, no further attempts to find a match by advancing the starting point
2723 take place. Once (*COMMIT) has been passed, \fBpcre_exec()\fP is committed to
2724 finding a match at the current starting point, or not at all. For example:
2725 .sp
2726 a+(*COMMIT)b
2727 .sp
2728 This matches "xxaab" but not "aacaab". It can be thought of as a kind of
2729 dynamic anchor, or "I've started, so I must finish." The name of the most
2730 recently passed (*MARK) in the path is passed back when (*COMMIT) forces a
2731 match failure.
2732 .P
2733 Note that (*COMMIT) at the start of a pattern is not the same as an anchor,
2734 unless PCRE's start-of-match optimizations are turned off, as shown in this
2735 \fBpcretest\fP example:
2736 .sp
2737 re> /(*COMMIT)abc/
2738 data> xyzabc
2739 0: abc
2740 xyzabc\eY
2741 No match
2742 .sp
2743 PCRE knows that any match must start with "a", so the optimization skips along
2744 the subject to "a" before running the first match attempt, which succeeds. When
2745 the optimization is disabled by the \eY escape in the second subject, the match
2746 starts at "x" and so the (*COMMIT) causes it to fail without trying any other
2747 starting points.
2748 .sp
2749 (*PRUNE) or (*PRUNE:NAME)
2750 .sp
2751 This verb causes the match to fail at the current starting position in the
2752 subject if the rest of the pattern does not match. If the pattern is
2753 unanchored, the normal "bumpalong" advance to the next starting character then
2754 happens. Backtracking can occur as usual to the left of (*PRUNE), before it is
2755 reached, or when matching to the right of (*PRUNE), but if there is no match to
2756 the right, backtracking cannot cross (*PRUNE). In simple cases, the use of
2757 (*PRUNE) is just an alternative to an atomic group or possessive quantifier,
2758 but there are some uses of (*PRUNE) that cannot be expressed in any other way.
2759 The behaviour of (*PRUNE:NAME) is the same as (*MARK:NAME)(*PRUNE). In an
2760 anchored pattern (*PRUNE) has the same effect as (*COMMIT).
2761 .sp
2762 (*SKIP)
2763 .sp
2764 This verb, when given without a name, is like (*PRUNE), except that if the
2765 pattern is unanchored, the "bumpalong" advance is not to the next character,
2766 but to the position in the subject where (*SKIP) was encountered. (*SKIP)
2767 signifies that whatever text was matched leading up to it cannot be part of a
2768 successful match. Consider:
2769 .sp
2770 a+(*SKIP)b
2771 .sp
2772 If the subject is "aaaac...", after the first match attempt fails (starting at
2773 the first character in the string), the starting point skips on to start the
2774 next attempt at "c". Note that a possessive quantifer does not have the same
2775 effect as this example; although it would suppress backtracking during the
2776 first match attempt, the second attempt would start at the second character
2777 instead of skipping on to "c".
2778 .sp
2779 (*SKIP:NAME)
2780 .sp
2781 When (*SKIP) has an associated name, its behaviour is modified. If the
2782 following pattern fails to match, the previous path through the pattern is
2783 searched for the most recent (*MARK) that has the same name. If one is found,
2784 the "bumpalong" advance is to the subject position that corresponds to that
2785 (*MARK) instead of to where (*SKIP) was encountered. If no (*MARK) with a
2786 matching name is found, the (*SKIP) is ignored.
2787 .sp
2788 (*THEN) or (*THEN:NAME)
2789 .sp
2790 This verb causes a skip to the next innermost alternative if the rest of the
2791 pattern does not match. That is, it cancels pending backtracking, but only
2792 within the current alternative. Its name comes from the observation that it can
2793 be used for a pattern-based if-then-else block:
2794 .sp
2795 ( COND1 (*THEN) FOO | COND2 (*THEN) BAR | COND3 (*THEN) BAZ ) ...
2796 .sp
2797 If the COND1 pattern matches, FOO is tried (and possibly further items after
2798 the end of the group if FOO succeeds); on failure, the matcher skips to the
2799 second alternative and tries COND2, without backtracking into COND1. The
2800 behaviour of (*THEN:NAME) is exactly the same as (*MARK:NAME)(*THEN).
2801 If (*THEN) is not inside an alternation, it acts like (*PRUNE).
2802 .P
2803 Note that a subpattern that does not contain a | character is just a part of
2804 the enclosing alternative; it is not a nested alternation with only one
2805 alternative. The effect of (*THEN) extends beyond such a subpattern to the
2806 enclosing alternative. Consider this pattern, where A, B, etc. are complex
2807 pattern fragments that do not contain any | characters at this level:
2808 .sp
2809 A (B(*THEN)C) | D
2810 .sp
2811 If A and B are matched, but there is a failure in C, matching does not
2812 backtrack into A; instead it moves to the next alternative, that is, D.
2813 However, if the subpattern containing (*THEN) is given an alternative, it
2814 behaves differently:
2815 .sp
2816 A (B(*THEN)C | (*FAIL)) | D
2817 .sp
2818 The effect of (*THEN) is now confined to the inner subpattern. After a failure
2819 in C, matching moves to (*FAIL), which causes the whole subpattern to fail
2820 because there are no more alternatives to try. In this case, matching does now
2821 backtrack into A.
2822 .P
2823 Note also that a conditional subpattern is not considered as having two
2824 alternatives, because only one is ever used. In other words, the | character in
2825 a conditional subpattern has a different meaning. Ignoring white space,
2826 consider:
2827 .sp
2828 ^.*? (?(?=a) a | b(*THEN)c )
2829 .sp
2830 If the subject is "ba", this pattern does not match. Because .*? is ungreedy,
2831 it initially matches zero characters. The condition (?=a) then fails, the
2832 character "b" is matched, but "c" is not. At this point, matching does not
2833 backtrack to .*? as might perhaps be expected from the presence of the |
2834 character. The conditional subpattern is part of the single alternative that
2835 comprises the whole pattern, and so the match fails. (If there was a backtrack
2836 into .*?, allowing it to match "b", the match would succeed.)
2837 .P
2838 The verbs just described provide four different "strengths" of control when
2839 subsequent matching fails. (*THEN) is the weakest, carrying on the match at the
2840 next alternative. (*PRUNE) comes next, failing the match at the current
2841 starting position, but allowing an advance to the next character (for an
2842 unanchored pattern). (*SKIP) is similar, except that the advance may be more
2843 than one character. (*COMMIT) is the strongest, causing the entire match to
2844 fail.
2845 .P
2846 If more than one such verb is present in a pattern, the "strongest" one wins.
2847 For example, consider this pattern, where A, B, etc. are complex pattern
2848 fragments:
2849 .sp
2850 (A(*COMMIT)B(*THEN)C|D)
2851 .sp
2852 Once A has matched, PCRE is committed to this match, at the current starting
2853 position. If subsequently B matches, but C does not, the normal (*THEN) action
2854 of trying the next alternative (that is, D) does not happen because (*COMMIT)
2855 overrides.
2856 .
2857 .
2858 .SH "SEE ALSO"
2859 .rs
2860 .sp
2861 \fBpcreapi\fP(3), \fBpcrecallout\fP(3), \fBpcrematching\fP(3),
2862 \fBpcresyntax\fP(3), \fBpcre\fP(3), \fBpcre16(3)\fP.
2863 .
2864 .
2865 .SH AUTHOR
2866 .rs
2867 .sp
2868 .nf
2869 Philip Hazel
2870 University Computing Service
2871 Cambridge CB2 3QH, England.
2872 .fi
2873 .
2874 .
2875 .SH REVISION
2876 .rs
2877 .sp
2878 .nf
2879 Last updated: 09 January 2012
2880 Copyright (c) 1997-2012 University of Cambridge.
2881 .fi

Properties

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

  ViewVC Help
Powered by ViewVC 1.1.5