ViewVC logotype

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

Parent Directory Parent Directory | Revision Log Revision Log

Revision 576 - (hide annotations) (download)
Sun Nov 21 18:45:10 2010 UTC (4 years, 5 months ago) by ph10
File size: 113806 byte(s)
Added support for (*NO_START_OPT)

1 nigel 79 .TH PCREPATTERN 3
2 nigel 63 .SH NAME
3     PCRE - Perl-compatible regular expressions
5 nigel 63 .rs
6     .sp
7 ph10 208 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 ph10 333 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 ph10 208 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 nigel 75 .P
23     The original operation of PCRE was on strings of one-byte characters. However,
24 ph10 461 there is now also support for UTF-8 character strings. To use this,
25 ph10 456 PCRE must be built to include UTF-8 support, and you must call
26     \fBpcre_compile()\fP or \fBpcre_compile2()\fP with the PCRE_UTF8 option. There
27     is also a special sequence that can be given at the start of a pattern:
28 ph10 412 .sp
29     (*UTF8)
30 ph10 416 .sp
31 ph10 412 Starting a pattern with this sequence is equivalent to setting the PCRE_UTF8
32     option. This feature is not Perl-compatible. How setting UTF-8 mode affects
33     pattern matching is mentioned in several places below. There is also a summary
34     of UTF-8 features in the
35 nigel 63 .\" HTML <a href="pcre.html#utf8support">
36     .\" </a>
37     section on UTF-8 support
38     .\"
39     in the main
40     .\" HREF
41 nigel 75 \fBpcre\fP
42 nigel 63 .\"
43     page.
44 nigel 75 .P
45 ph10 535 Another special sequence that may appear at the start of a pattern or in
46 ph10 518 combination with (*UTF8) is:
47     .sp
48     (*UCP)
49     .sp
50 ph10 535 This has the same effect as setting the PCRE_UCP option: it causes sequences
51     such as \ed and \ew to use Unicode properties to determine character types,
52     instead of recognizing only characters with codes less than 128 via a lookup
53 ph10 518 table.
54     .P
55 ph10 576 If a pattern starts with (*NO_START_OPT), it has the same effect as setting the
56     PCRE_NO_START_OPTIMIZE option either at compile or matching time. There are
57     also some more of these special sequences that are concerned with the handling
58     of newlines; they are described below.
59     .P
60 nigel 77 The remainder of this document discusses the patterns that are supported by
61     PCRE when its main matching function, \fBpcre_exec()\fP, is used.
62     From release 6.0, PCRE offers a second matching function,
63     \fBpcre_dfa_exec()\fP, which matches using a different algorithm that is not
64 ph10 172 Perl-compatible. Some of the features discussed below are not available when
65 ph10 168 \fBpcre_dfa_exec()\fP is used. The advantages and disadvantages of the
66     alternative function, and how it differs from the normal function, are
67     discussed in the
68 nigel 77 .\" HREF
69     \fBpcrematching\fP
70     .\"
71     page.
72 nigel 93 .
73     .
74 ph10 556 .\" HTML <a name="newlines"></a>
76     .rs
77     .sp
78     PCRE supports five different conventions for indicating line breaks in
79     strings: a single CR (carriage return) character, a single LF (linefeed)
80     character, the two-character sequence CRLF, any of the three preceding, or any
81     Unicode newline sequence. The
82     .\" HREF
83     \fBpcreapi\fP
84     .\"
85     page has
86     .\" HTML <a href="pcreapi.html#newlines">
87     .\" </a>
88     further discussion
89     .\"
90     about newlines, and shows how to set the newline convention in the
91     \fIoptions\fP arguments for the compiling and matching functions.
92     .P
93     It is also possible to specify a newline convention by starting a pattern
94     string with one of the following five sequences:
95     .sp
96     (*CR) carriage return
97     (*LF) linefeed
98     (*CRLF) carriage return, followed by linefeed
99     (*ANYCRLF) any of the three above
100     (*ANY) all Unicode newline sequences
101     .sp
102 ph10 461 These override the default and the options given to \fBpcre_compile()\fP or
103 ph10 456 \fBpcre_compile2()\fP. For example, on a Unix system where LF is the default
104     newline sequence, the pattern
105 ph10 227 .sp
106     (*CR)a.b
107     .sp
108     changes the convention to CR. That pattern matches "a\enb" because LF is no
109     longer a newline. Note that these special settings, which are not
110     Perl-compatible, are recognized only at the very start of a pattern, and that
111 ph10 231 they must be in upper case. If more than one of them is present, the last one
112     is used.
113     .P
114 ph10 514 The newline convention affects the interpretation of the dot metacharacter when
115     PCRE_DOTALL is not set, and also the behaviour of \eN. However, it does not
116     affect what the \eR escape sequence matches. By default, this is any Unicode
117     newline sequence, for Perl compatibility. However, this can be changed; see the
118     description of \eR in the section entitled
119 ph10 231 .\" HTML <a href="#newlineseq">
120     .\" </a>
121     "Newline sequences"
122     .\"
123 ph10 247 below. A change of \eR setting can be combined with a change of newline
124 ph10 246 convention.
125 ph10 227 .
126     .
128     .rs
129     .sp
130 nigel 63 A regular expression is a pattern that is matched against a subject string from
131     left to right. Most characters stand for themselves in a pattern, and match the
132     corresponding characters in the subject. As a trivial example, the pattern
133 nigel 75 .sp
134 nigel 63 The quick brown fox
135 nigel 75 .sp
136 nigel 77 matches a portion of a subject string that is identical to itself. When
137     caseless matching is specified (the PCRE_CASELESS option), letters are matched
138     independently of case. In UTF-8 mode, PCRE always understands the concept of
139     case for characters whose values are less than 128, so caseless matching is
140     always possible. For characters with higher values, the concept of case is
141     supported if PCRE is compiled with Unicode property support, but not otherwise.
142     If you want to use caseless matching for characters 128 and above, you must
143     ensure that PCRE is compiled with Unicode property support as well as with
144     UTF-8 support.
145     .P
146     The power of regular expressions comes from the ability to include alternatives
147     and repetitions in the pattern. These are encoded in the pattern by the use of
148 nigel 75 \fImetacharacters\fP, which do not stand for themselves but instead are
149 nigel 63 interpreted in some special way.
150 nigel 75 .P
151     There are two different sets of metacharacters: those that are recognized
152 nigel 63 anywhere in the pattern except within square brackets, and those that are
153 nigel 93 recognized within square brackets. Outside square brackets, the metacharacters
154     are as follows:
155 nigel 75 .sp
156     \e general escape character with several uses
157 nigel 63 ^ assert start of string (or line, in multiline mode)
158     $ assert end of string (or line, in multiline mode)
159     . match any character except newline (by default)
160     [ start character class definition
161     | start of alternative branch
162     ( start subpattern
163     ) end subpattern
164     ? extends the meaning of (
165     also 0 or 1 quantifier
166     also quantifier minimizer
167     * 0 or more quantifier
168     + 1 or more quantifier
169     also "possessive quantifier"
170     { start min/max quantifier
171 nigel 75 .sp
172 nigel 63 Part of a pattern that is in square brackets is called a "character class". In
173 nigel 75 a character class the only metacharacters are:
174     .sp
175     \e general escape character
176 nigel 63 ^ negate the class, but only if the first character
177     - indicates character range
178 nigel 75 .\" JOIN
179 nigel 63 [ POSIX character class (only if followed by POSIX
180     syntax)
181     ] terminates the character class
182 nigel 75 .sp
183     The following sections describe the use of each of the metacharacters.
184     .
185 nigel 93 .
186 nigel 63 .SH BACKSLASH
187     .rs
188     .sp
189     The backslash character has several uses. Firstly, if it is followed by a
190 ph10 574 character that is not a number or a letter, it takes away any special meaning
191     that character may have. This use of backslash as an escape character applies
192     both inside and outside character classes.
193 nigel 75 .P
194     For example, if you want to match a * character, you write \e* in the pattern.
195 nigel 63 This escaping action applies whether or not the following character would
196 nigel 75 otherwise be interpreted as a metacharacter, so it is always safe to precede a
197     non-alphanumeric with backslash to specify that it stands for itself. In
198     particular, if you want to match a backslash, you write \e\e.
199     .P
200 ph10 574 In UTF-8 mode, only ASCII numbers and letters have any special meaning after a
201     backslash. All other characters (in particular, those whose codepoints are
202     greater than 127) are treated as literals.
203     .P
204 nigel 63 If a pattern is compiled with the PCRE_EXTENDED option, whitespace in the
205     pattern (other than in a character class) and characters between a # outside
206 nigel 91 a character class and the next newline are ignored. An escaping backslash can
207     be used to include a whitespace or # character as part of the pattern.
208 nigel 75 .P
209 nigel 63 If you want to remove the special meaning from a sequence of characters, you
210 nigel 75 can do so by putting them between \eQ and \eE. This is different from Perl in
211     that $ and @ are handled as literals in \eQ...\eE sequences in PCRE, whereas in
212 nigel 63 Perl, $ and @ cause variable interpolation. Note the following examples:
213 nigel 75 .sp
214 nigel 63 Pattern PCRE matches Perl matches
215 nigel 75 .sp
216     .\" JOIN
217     \eQabc$xyz\eE abc$xyz abc followed by the
218 nigel 63 contents of $xyz
219 nigel 75 \eQabc\e$xyz\eE abc\e$xyz abc\e$xyz
220     \eQabc\eE\e$\eQxyz\eE abc$xyz abc$xyz
221     .sp
222     The \eQ...\eE sequence is recognized both inside and outside character classes.
223 ph10 555 An isolated \eE that is not preceded by \eQ is ignored.
224 nigel 75 .
225     .
226     .\" HTML <a name="digitsafterbackslash"></a>
227     .SS "Non-printing characters"
228     .rs
229     .sp
230 nigel 63 A second use of backslash provides a way of encoding non-printing characters
231     in patterns in a visible manner. There is no restriction on the appearance of
232     non-printing characters, apart from the binary zero that terminates a pattern,
233 ph10 456 but when a pattern is being prepared by text editing, it is often easier to use
234     one of the following escape sequences than the binary character it represents:
235 nigel 75 .sp
236     \ea alarm, that is, the BEL character (hex 07)
237 ph10 574 \ecx "control-x", where x is any ASCII character
238 nigel 75 \ee escape (hex 1B)
239     \ef formfeed (hex 0C)
240 ph10 227 \en linefeed (hex 0A)
241 nigel 75 \er carriage return (hex 0D)
242     \et tab (hex 09)
243 ph10 488 \eddd character with octal code ddd, or back reference
244 nigel 75 \exhh character with hex code hh
245 nigel 87 \ex{hhh..} character with hex code hhh..
246 nigel 75 .sp
247     The precise effect of \ecx is as follows: if x is a lower case letter, it
248 nigel 63 is converted to upper case. Then bit 6 of the character (hex 40) is inverted.
249 ph10 574 Thus \ecz becomes hex 1A (z is 7A), but \ec{ becomes hex 3B ({ is 7B), while
250     \ec; becomes hex 7B (; is 3B). If the byte following \ec has a value greater
251     than 127, a compile-time error occurs. This locks out non-ASCII characters in
252     both byte mode and UTF-8 mode. (When PCRE is compiled in EBCDIC mode, all byte
253     values are valid. A lower case letter is converted to upper case, and then the
254     0xc0 bits are flipped.)
255 nigel 75 .P
256     After \ex, from zero to two hexadecimal digits are read (letters can be in
257 nigel 87 upper or lower case). Any number of hexadecimal digits may appear between \ex{
258     and }, but the value of the character code must be less than 256 in non-UTF-8
259 ph10 211 mode, and less than 2**31 in UTF-8 mode. That is, the maximum value in
260     hexadecimal is 7FFFFFFF. Note that this is bigger than the largest Unicode code
261     point, which is 10FFFF.
262 nigel 75 .P
263 ph10 211 If characters other than hexadecimal digits appear between \ex{ and }, or if
264     there is no terminating }, this form of escape is not recognized. Instead, the
265     initial \ex will be interpreted as a basic hexadecimal escape, with no
266     following digits, giving a character whose value is zero.
267     .P
268 nigel 63 Characters whose value is less than 256 can be defined by either of the two
269 nigel 87 syntaxes for \ex. There is no difference in the way they are handled. For
270     example, \exdc is exactly the same as \ex{dc}.
271 nigel 75 .P
272 nigel 91 After \e0 up to two further octal digits are read. If there are fewer than two
273     digits, just those that are present are used. Thus the sequence \e0\ex\e07
274     specifies two binary zeros followed by a BEL character (code value 7). Make
275     sure you supply two digits after the initial zero if the pattern character that
276     follows is itself an octal digit.
277 nigel 75 .P
278 nigel 63 The handling of a backslash followed by a digit other than 0 is complicated.
279     Outside a character class, PCRE reads it and any following digits as a decimal
280     number. If the number is less than 10, or if there have been at least that many
281     previous capturing left parentheses in the expression, the entire sequence is
282 nigel 75 taken as a \fIback reference\fP. A description of how this works is given
283     .\" HTML <a href="#backreferences">
284     .\" </a>
285     later,
286     .\"
287     following the discussion of
288     .\" HTML <a href="#subpattern">
289     .\" </a>
290     parenthesized subpatterns.
291     .\"
292     .P
293 nigel 63 Inside a character class, or if the decimal number is greater than 9 and there
294     have not been that many capturing subpatterns, PCRE re-reads up to three octal
295 nigel 93 digits following the backslash, and uses them to generate a data character. Any
296 nigel 91 subsequent digits stand for themselves. In non-UTF-8 mode, the value of a
297     character specified in octal must be less than \e400. In UTF-8 mode, values up
298     to \e777 are permitted. For example:
299 nigel 75 .sp
300     \e040 is another way of writing a space
301     .\" JOIN
302     \e40 is the same, provided there are fewer than 40
303 nigel 63 previous capturing subpatterns
304 nigel 75 \e7 is always a back reference
305     .\" JOIN
306     \e11 might be a back reference, or another way of
307 nigel 63 writing a tab
308 nigel 75 \e011 is always a tab
309     \e0113 is a tab followed by the character "3"
310     .\" JOIN
311     \e113 might be a back reference, otherwise the
312 nigel 63 character with octal code 113
313 nigel 75 .\" JOIN
314     \e377 might be a back reference, otherwise
315 nigel 63 the byte consisting entirely of 1 bits
316 nigel 75 .\" JOIN
317     \e81 is either a back reference, or a binary zero
318 nigel 63 followed by the two characters "8" and "1"
319 nigel 75 .sp
320 nigel 63 Note that octal values of 100 or greater must not be introduced by a leading
321     zero, because no more than three octal digits are ever read.
322 nigel 75 .P
323 nigel 91 All the sequences that define a single character value can be used both inside
324     and outside character classes. In addition, inside a character class, the
325 ph10 513 sequence \eb is interpreted as the backspace character (hex 08). The sequences
326 ph10 514 \eB, \eN, \eR, and \eX are not special inside a character class. Like any other
327 ph10 513 unrecognized escape sequences, they are treated as the literal characters "B",
328 ph10 514 "N", "R", and "X" by default, but cause an error if the PCRE_EXTRA option is
329     set. Outside a character class, these sequences have different meanings.
330 nigel 75 .
331     .
332 nigel 93 .SS "Absolute and relative back references"
333     .rs
334     .sp
335 ph10 208 The sequence \eg followed by an unsigned or a negative number, optionally
336     enclosed in braces, is an absolute or relative back reference. A named back
337     reference can be coded as \eg{name}. Back references are discussed
338 nigel 93 .\" HTML <a href="#backreferences">
339     .\" </a>
340     later,
341     .\"
342     following the discussion of
343     .\" HTML <a href="#subpattern">
344     .\" </a>
345     parenthesized subpatterns.
346     .\"
347     .
348     .
349 ph10 333 .SS "Absolute and relative subroutine calls"
350     .rs
351     .sp
352 ph10 345 For compatibility with Oniguruma, the non-Perl syntax \eg followed by a name or
353     a number enclosed either in angle brackets or single quotes, is an alternative
354     syntax for referencing a subpattern as a "subroutine". Details are discussed
355 ph10 333 .\" HTML <a href="#onigurumasubroutines">
356     .\" </a>
357     later.
358     .\"
359 ph10 345 Note that \eg{...} (Perl syntax) and \eg<...> (Oniguruma syntax) are \fInot\fP
360 ph10 461 synonymous. The former is a back reference; the latter is a
361 ph10 454 .\" HTML <a href="#subpatternsassubroutines">
362     .\" </a>
363     subroutine
364     .\"
365     call.
366 ph10 333 .
367     .
368 ph10 518 .\" HTML <a name="genericchartypes"></a>
369 nigel 75 .SS "Generic character types"
370     .rs
371     .sp
372 ph10 514 Another use of backslash is for specifying generic character types:
373 nigel 75 .sp
374 ph10 182 \ed any decimal digit
375 nigel 75 \eD any character that is not a decimal digit
376 ph10 178 \eh any horizontal whitespace character
377 ph10 182 \eH any character that is not a horizontal whitespace character
378 nigel 75 \es any whitespace character
379     \eS any character that is not a whitespace character
380 ph10 178 \ev any vertical whitespace character
381 ph10 182 \eV any character that is not a vertical whitespace character
382 nigel 75 \ew any "word" character
383     \eW any "non-word" character
384     .sp
385 ph10 535 There is also the single sequence \eN, which matches a non-newline character.
386     This is the same as
387 ph10 514 .\" HTML <a href="#fullstopdot">
388     .\" </a>
389 ph10 535 the "." metacharacter
390 ph10 514 .\"
391     when PCRE_DOTALL is not set.
392 nigel 75 .P
393 ph10 514 Each pair of lower and upper case escape sequences partitions the complete set
394     of characters into two disjoint sets. Any given character matches one, and only
395 ph10 518 one, of each pair. The sequences can appear both inside and outside character
396 nigel 75 classes. They each match one character of the appropriate type. If the current
397 ph10 518 matching point is at the end of the subject string, all of them fail, because
398 nigel 75 there is no character to match.
399     .P
400     For compatibility with Perl, \es does not match the VT character (code 11).
401     This makes it different from the the POSIX "space" class. The \es characters
402 ph10 178 are HT (9), LF (10), FF (12), CR (13), and space (32). If "use locale;" is
403 nigel 91 included in a Perl script, \es may match the VT character. In PCRE, it never
404 ph10 178 does.
405 nigel 75 .P
406 ph10 518 A "word" character is an underscore or any character that is a letter or digit.
407     By default, the definition of letters and digits is controlled by PCRE's
408     low-valued character tables, and may vary if locale-specific matching is taking
409     place (see
410     .\" HTML <a href="pcreapi.html#localesupport">
411     .\" </a>
412     "Locale support"
413     .\"
414     in the
415     .\" HREF
416     \fBpcreapi\fP
417     .\"
418     page). For example, in a French locale such as "fr_FR" in Unix-like systems,
419     or "french" in Windows, some character codes greater than 128 are used for
420     accented letters, and these are then matched by \ew. The use of locales with
421     Unicode is discouraged.
422 ph10 178 .P
423 ph10 518 By default, in UTF-8 mode, characters with values greater than 128 never match
424     \ed, \es, or \ew, and always match \eD, \eS, and \eW. These sequences retain
425     their original meanings from before UTF-8 support was available, mainly for
426 ph10 535 efficiency reasons. However, if PCRE is compiled with Unicode property support,
427 ph10 518 and the PCRE_UCP option is set, the behaviour is changed so that Unicode
428     properties are used to determine character types, as follows:
429     .sp
430     \ed any character that \ep{Nd} matches (decimal digit)
431     \es any character that \ep{Z} matches, plus HT, LF, FF, CR
432     \ew any character that \ep{L} or \ep{N} matches, plus underscore
433     .sp
434     The upper case escapes match the inverse sets of characters. Note that \ed
435 ph10 535 matches only decimal digits, whereas \ew matches any Unicode digit, as well as
436 ph10 518 any Unicode letter, and underscore. Note also that PCRE_UCP affects \eb, and
437     \eB because they are defined in terms of \ew and \eW. Matching these sequences
438     is noticeably slower when PCRE_UCP is set.
439     .P
440 ph10 572 The sequences \eh, \eH, \ev, and \eV are features that were added to Perl at
441     release 5.10. In contrast to the other sequences, which match only ASCII
442     characters by default, these always match certain high-valued codepoints in
443     UTF-8 mode, whether or not PCRE_UCP is set. The horizontal space characters
444     are:
445 ph10 178 .sp
446     U+0009 Horizontal tab
447     U+0020 Space
448     U+00A0 Non-break space
449     U+1680 Ogham space mark
450     U+180E Mongolian vowel separator
451     U+2000 En quad
452     U+2001 Em quad
453     U+2002 En space
454     U+2003 Em space
455     U+2004 Three-per-em space
456     U+2005 Four-per-em space
457     U+2006 Six-per-em space
458     U+2007 Figure space
459     U+2008 Punctuation space
460     U+2009 Thin space
461     U+200A Hair space
462     U+202F Narrow no-break space
463     U+205F Medium mathematical space
464     U+3000 Ideographic space
465     .sp
466     The vertical space characters are:
467     .sp
468     U+000A Linefeed
469     U+000B Vertical tab
470     U+000C Formfeed
471     U+000D Carriage return
472     U+0085 Next line
473     U+2028 Line separator
474     U+2029 Paragraph separator
475 nigel 75 .
476     .
477 ph10 231 .\" HTML <a name="newlineseq"></a>
478 nigel 93 .SS "Newline sequences"
479     .rs
480     .sp
481 ph10 231 Outside a character class, by default, the escape sequence \eR matches any
482 ph10 572 Unicode newline sequence. In non-UTF-8 mode \eR is equivalent to the following:
483 nigel 93 .sp
484     (?>\er\en|\en|\ex0b|\ef|\er|\ex85)
485     .sp
486     This is an example of an "atomic group", details of which are given
487     .\" HTML <a href="#atomicgroup">
488     .\" </a>
489     below.
490     .\"
491     This particular group matches either the two-character sequence CR followed by
492     LF, or one of the single characters LF (linefeed, U+000A), VT (vertical tab,
493     U+000B), FF (formfeed, U+000C), CR (carriage return, U+000D), or NEL (next
494     line, U+0085). The two-character sequence is treated as a single unit that
495     cannot be split.
496     .P
497     In UTF-8 mode, two additional characters whose codepoints are greater than 255
498     are added: LS (line separator, U+2028) and PS (paragraph separator, U+2029).
499     Unicode character property support is not needed for these characters to be
500     recognized.
501     .P
502 ph10 231 It is possible to restrict \eR to match only CR, LF, or CRLF (instead of the
503     complete set of Unicode line endings) by setting the option PCRE_BSR_ANYCRLF
504 ph10 247 either at compile time or when the pattern is matched. (BSR is an abbrevation
505 ph10 246 for "backslash R".) This can be made the default when PCRE is built; if this is
506     the case, the other behaviour can be requested via the PCRE_BSR_UNICODE option.
507     It is also possible to specify these settings by starting a pattern string with
508     one of the following sequences:
509 ph10 231 .sp
510     (*BSR_ANYCRLF) CR, LF, or CRLF only
511     (*BSR_UNICODE) any Unicode newline sequence
512     .sp
513 ph10 461 These override the default and the options given to \fBpcre_compile()\fP or
514 ph10 456 \fBpcre_compile2()\fP, but they can be overridden by options given to
515     \fBpcre_exec()\fP or \fBpcre_dfa_exec()\fP. Note that these special settings,
516     which are not Perl-compatible, are recognized only at the very start of a
517     pattern, and that they must be in upper case. If more than one of them is
518     present, the last one is used. They can be combined with a change of newline
519 ph10 518 convention; for example, a pattern can start with:
520 ph10 246 .sp
521     (*ANY)(*BSR_ANYCRLF)
522     .sp
523 ph10 518 They can also be combined with the (*UTF8) or (*UCP) special sequences. Inside
524     a character class, \eR is treated as an unrecognized escape sequence, and so
525     matches the letter "R" by default, but causes an error if PCRE_EXTRA is set.
526 nigel 93 .
527     .
528 nigel 75 .\" HTML <a name="uniextseq"></a>
529     .SS Unicode character properties
530     .rs
531     .sp
532     When PCRE is built with Unicode character property support, three additional
533 ph10 184 escape sequences that match characters with specific properties are available.
534     When not in UTF-8 mode, these sequences are of course limited to testing
535     characters whose codepoints are less than 256, but they do work in this mode.
536     The extra escape sequences are:
537 nigel 75 .sp
538 nigel 87 \ep{\fIxx\fP} a character with the \fIxx\fP property
539     \eP{\fIxx\fP} a character without the \fIxx\fP property
540     \eX an extended Unicode sequence
541 nigel 75 .sp
542 nigel 87 The property names represented by \fIxx\fP above are limited to the Unicode
543 ph10 517 script names, the general category properties, "Any", which matches any
544     character (including newline), and some special PCRE properties (described
545 ph10 535 in the
546 ph10 517 .\" HTML <a href="#extraprops">
547     .\" </a>
548 ph10 535 next section).
549 ph10 517 .\"
550     Other Perl properties such as "InMusicalSymbols" are not currently supported by
551     PCRE. Note that \eP{Any} does not match any characters, so always causes a
552     match failure.
553 nigel 75 .P
554 nigel 87 Sets of Unicode characters are defined as belonging to certain scripts. A
555     character from one of these sets can be matched using a script name. For
556     example:
557 nigel 75 .sp
558 nigel 87 \ep{Greek}
559     \eP{Han}
560     .sp
561     Those that are not part of an identified script are lumped together as
562     "Common". The current list of scripts is:
563     .P
564     Arabic,
565     Armenian,
566 ph10 491 Avestan,
567 nigel 93 Balinese,
568 ph10 491 Bamum,
569 nigel 87 Bengali,
570     Bopomofo,
571     Braille,
572     Buginese,
573     Buhid,
574     Canadian_Aboriginal,
575 ph10 491 Carian,
576     Cham,
577 nigel 87 Cherokee,
578     Common,
579     Coptic,
580 nigel 93 Cuneiform,
581 nigel 87 Cypriot,
582     Cyrillic,
583     Deseret,
584     Devanagari,
585 ph10 491 Egyptian_Hieroglyphs,
586 nigel 87 Ethiopic,
587     Georgian,
588     Glagolitic,
589     Gothic,
590     Greek,
591     Gujarati,
592     Gurmukhi,
593     Han,
594     Hangul,
595     Hanunoo,
596     Hebrew,
597     Hiragana,
598 ph10 491 Imperial_Aramaic,
599 nigel 87 Inherited,
600 ph10 491 Inscriptional_Pahlavi,
601     Inscriptional_Parthian,
602     Javanese,
603     Kaithi,
604 nigel 87 Kannada,
605     Katakana,
606 ph10 491 Kayah_Li,
607 nigel 87 Kharoshthi,
608     Khmer,
609     Lao,
610     Latin,
611 ph10 491 Lepcha,
612 nigel 87 Limbu,
613     Linear_B,
614 ph10 491 Lisu,
615     Lycian,
616     Lydian,
617 nigel 87 Malayalam,
618 ph10 491 Meetei_Mayek,
619 nigel 87 Mongolian,
620     Myanmar,
621     New_Tai_Lue,
622 nigel 93 Nko,
623 nigel 87 Ogham,
624     Old_Italic,
625     Old_Persian,
626 ph10 491 Old_South_Arabian,
627     Old_Turkic,
628     Ol_Chiki,
629 nigel 87 Oriya,
630     Osmanya,
631 nigel 93 Phags_Pa,
632     Phoenician,
633 ph10 491 Rejang,
634 nigel 87 Runic,
635 ph10 491 Samaritan,
636     Saurashtra,
637 nigel 87 Shavian,
638     Sinhala,
639 ph10 491 Sundanese,
640 nigel 87 Syloti_Nagri,
641     Syriac,
642     Tagalog,
643     Tagbanwa,
644     Tai_Le,
645 ph10 491 Tai_Tham,
646     Tai_Viet,
647 nigel 87 Tamil,
648     Telugu,
649     Thaana,
650     Thai,
651     Tibetan,
652     Tifinagh,
653     Ugaritic,
654 ph10 491 Vai,
655 nigel 87 Yi.
656     .P
657 ph10 517 Each character has exactly one Unicode general category property, specified by
658     a two-letter abbreviation. For compatibility with Perl, negation can be
659     specified by including a circumflex between the opening brace and the property
660     name. For example, \ep{^Lu} is the same as \eP{Lu}.
661 nigel 87 .P
662     If only one letter is specified with \ep or \eP, it includes all the general
663     category properties that start with that letter. In this case, in the absence
664     of negation, the curly brackets in the escape sequence are optional; these two
665     examples have the same effect:
666     .sp
667 nigel 75 \ep{L}
668     \epL
669     .sp
670 nigel 87 The following general category property codes are supported:
671 nigel 75 .sp
672     C Other
673     Cc Control
674     Cf Format
675     Cn Unassigned
676     Co Private use
677     Cs Surrogate
678     .sp
679     L Letter
680     Ll Lower case letter
681     Lm Modifier letter
682     Lo Other letter
683     Lt Title case letter
684     Lu Upper case letter
685     .sp
686     M Mark
687     Mc Spacing mark
688     Me Enclosing mark
689     Mn Non-spacing mark
690     .sp
691     N Number
692     Nd Decimal number
693     Nl Letter number
694     No Other number
695     .sp
696     P Punctuation
697     Pc Connector punctuation
698     Pd Dash punctuation
699     Pe Close punctuation
700     Pf Final punctuation
701     Pi Initial punctuation
702     Po Other punctuation
703     Ps Open punctuation
704     .sp
705     S Symbol
706     Sc Currency symbol
707     Sk Modifier symbol
708     Sm Mathematical symbol
709     So Other symbol
710     .sp
711     Z Separator
712     Zl Line separator
713     Zp Paragraph separator
714     Zs Space separator
715     .sp
716 nigel 87 The special property L& is also supported: it matches a character that has
717     the Lu, Ll, or Lt property, in other words, a letter that is not classified as
718     a modifier or "other".
719 nigel 75 .P
720 ph10 211 The Cs (Surrogate) property applies only to characters in the range U+D800 to
721     U+DFFF. Such characters are not valid in UTF-8 strings (see RFC 3629) and so
722     cannot be tested by PCRE, unless UTF-8 validity checking has been turned off
723     (see the discussion of PCRE_NO_UTF8_CHECK in the
724     .\" HREF
725     \fBpcreapi\fP
726     .\"
727 ph10 451 page). Perl does not support the Cs property.
728 ph10 211 .P
729 ph10 451 The long synonyms for property names that Perl supports (such as \ep{Letter})
730 nigel 91 are not supported by PCRE, nor is it permitted to prefix any of these
731 nigel 87 properties with "Is".
732     .P
733     No character that is in the Unicode table has the Cn (unassigned) property.
734     Instead, this property is assumed for any code point that is not in the
735     Unicode table.
736     .P
737 nigel 75 Specifying caseless matching does not affect these escape sequences. For
738     example, \ep{Lu} always matches only upper case letters.
739     .P
740     The \eX escape matches any number of Unicode characters that form an extended
741     Unicode sequence. \eX is equivalent to
742     .sp
743     (?>\ePM\epM*)
744     .sp
745     That is, it matches a character without the "mark" property, followed by zero
746     or more characters with the "mark" property, and treats the sequence as an
747     atomic group
748     .\" HTML <a href="#atomicgroup">
749     .\" </a>
750     (see below).
751     .\"
752     Characters with the "mark" property are typically accents that affect the
753 ph10 185 preceding character. None of them have codepoints less than 256, so in
754 ph10 184 non-UTF-8 mode \eX matches any one character.
755 nigel 75 .P
756     Matching characters by Unicode property is not fast, because PCRE has to search
757     a structure that contains data for over fifteen thousand characters. That is
758     why the traditional escape sequences such as \ed and \ew do not use Unicode
759 ph10 535 properties in PCRE by default, though you can make them do so by setting the
760 ph10 518 PCRE_UCP option for \fBpcre_compile()\fP or by starting the pattern with
761     (*UCP).
762 nigel 75 .
763     .
764 ph10 517 .\" HTML <a name="extraprops"></a>
765     .SS PCRE's additional properties
766     .rs
767     .sp
768 ph10 535 As well as the standard Unicode properties described in the previous
769     section, PCRE supports four more that make it possible to convert traditional
770 ph10 517 escape sequences such as \ew and \es and POSIX character classes to use Unicode
771 ph10 518 properties. PCRE uses these non-standard, non-Perl properties internally when
772     PCRE_UCP is set. They are:
773 ph10 517 .sp
774     Xan Any alphanumeric character
775     Xps Any POSIX space character
776     Xsp Any Perl space character
777     Xwd Any Perl "word" character
778     .sp
779 ph10 535 Xan matches characters that have either the L (letter) or the N (number)
780     property. Xps matches the characters tab, linefeed, vertical tab, formfeed, or
781 ph10 517 carriage return, and any other character that has the Z (separator) property.
782 ph10 535 Xsp is the same as Xps, except that vertical tab is excluded. Xwd matches the
783 ph10 517 same characters as Xan, plus underscore.
784     .
785     .
786 ph10 168 .\" HTML <a name="resetmatchstart"></a>
787     .SS "Resetting the match start"
788     .rs
789     .sp
790 ph10 572 The escape sequence \eK causes any previously matched characters not to be
791     included in the final matched sequence. For example, the pattern:
792 ph10 168 .sp
793     foo\eKbar
794     .sp
795 ph10 172 matches "foobar", but reports that it has matched "bar". This feature is
796 ph10 168 similar to a lookbehind assertion
797     .\" HTML <a href="#lookbehind">
798     .\" </a>
799     (described below).
800     .\"
801 ph10 172 However, in this case, the part of the subject before the real match does not
802     have to be of fixed length, as lookbehind assertions do. The use of \eK does
803 ph10 168 not interfere with the setting of
804     .\" HTML <a href="#subpattern">
805     .\" </a>
806     captured substrings.
807 ph10 172 .\"
808 ph10 168 For example, when the pattern
809     .sp
810     (foo)\eKbar
811     .sp
812 ph10 172 matches "foobar", the first substring is still set to "foo".
813 ph10 500 .P
814 ph10 507 Perl documents that the use of \eK within assertions is "not well defined". In
815     PCRE, \eK is acted upon when it occurs inside positive assertions, but is
816 ph10 500 ignored in negative assertions.
817 ph10 168 .
818     .
819 nigel 75 .\" HTML <a name="smallassertions"></a>
820     .SS "Simple assertions"
821     .rs
822     .sp
823 nigel 93 The final use of backslash is for certain simple assertions. An assertion
824 nigel 63 specifies a condition that has to be met at a particular point in a match,
825     without consuming any characters from the subject string. The use of
826 nigel 75 subpatterns for more complicated assertions is described
827     .\" HTML <a href="#bigassertions">
828     .\" </a>
829     below.
830     .\"
831 nigel 91 The backslashed assertions are:
832 nigel 75 .sp
833     \eb matches at a word boundary
834     \eB matches when not at a word boundary
835 nigel 93 \eA matches at the start of the subject
836     \eZ matches at the end of the subject
837     also matches before a newline at the end of the subject
838     \ez matches only at the end of the subject
839     \eG matches at the first matching position in the subject
840 nigel 75 .sp
841 ph10 513 Inside a character class, \eb has a different meaning; it matches the backspace
842 ph10 535 character. If any other of these assertions appears in a character class, by
843 ph10 513 default it matches the corresponding literal character (for example, \eB
844     matches the letter B). However, if the PCRE_EXTRA option is set, an "invalid
845     escape sequence" error is generated instead.
846 nigel 75 .P
847 nigel 63 A word boundary is a position in the subject string where the current character
848 nigel 75 and the previous character do not both match \ew or \eW (i.e. one matches
849     \ew and the other matches \eW), or the start or end of the string if the
850 ph10 518 first or last character matches \ew, respectively. In UTF-8 mode, the meanings
851     of \ew and \eW can be changed by setting the PCRE_UCP option. When this is
852     done, it also affects \eb and \eB. Neither PCRE nor Perl has a separate "start
853     of word" or "end of word" metasequence. However, whatever follows \eb normally
854     determines which it is. For example, the fragment \eba matches "a" at the start
855     of a word.
856 nigel 75 .P
857     The \eA, \eZ, and \ez assertions differ from the traditional circumflex and
858     dollar (described in the next section) in that they only ever match at the very
859     start and end of the subject string, whatever options are set. Thus, they are
860     independent of multiline mode. These three assertions are not affected by the
861     PCRE_NOTBOL or PCRE_NOTEOL options, which affect only the behaviour of the
862     circumflex and dollar metacharacters. However, if the \fIstartoffset\fP
863     argument of \fBpcre_exec()\fP is non-zero, indicating that matching is to start
864     at a point other than the beginning of the subject, \eA can never match. The
865 nigel 91 difference between \eZ and \ez is that \eZ matches before a newline at the end
866     of the string as well as at the very end, whereas \ez matches only at the end.
867 nigel 75 .P
868     The \eG assertion is true only when the current matching position is at the
869     start point of the match, as specified by the \fIstartoffset\fP argument of
870     \fBpcre_exec()\fP. It differs from \eA when the value of \fIstartoffset\fP is
871     non-zero. By calling \fBpcre_exec()\fP multiple times with appropriate
872 nigel 63 arguments, you can mimic Perl's /g option, and it is in this kind of
873 nigel 75 implementation where \eG can be useful.
874     .P
875     Note, however, that PCRE's interpretation of \eG, as the start of the current
876 nigel 63 match, is subtly different from Perl's, which defines it as the end of the
877     previous match. In Perl, these can be different when the previously matched
878     string was empty. Because PCRE does just one match at a time, it cannot
879     reproduce this behaviour.
880 nigel 75 .P
881     If all the alternatives of a pattern begin with \eG, the expression is anchored
882 nigel 63 to the starting match position, and the "anchored" flag is set in the compiled
883     regular expression.
884 nigel 75 .
885     .
887 nigel 63 .rs
888     .sp
889     Outside a character class, in the default matching mode, the circumflex
890 nigel 75 character is an assertion that is true only if the current matching point is
891     at the start of the subject string. If the \fIstartoffset\fP argument of
892     \fBpcre_exec()\fP is non-zero, circumflex can never match if the PCRE_MULTILINE
893 nigel 63 option is unset. Inside a character class, circumflex has an entirely different
894 nigel 75 meaning
895     .\" HTML <a href="#characterclass">
896     .\" </a>
897     (see below).
898     .\"
899     .P
900 nigel 63 Circumflex need not be the first character of the pattern if a number of
901     alternatives are involved, but it should be the first thing in each alternative
902     in which it appears if the pattern is ever to match that branch. If all
903     possible alternatives start with a circumflex, that is, if the pattern is
904     constrained to match only at the start of the subject, it is said to be an
905     "anchored" pattern. (There are also other constructs that can cause a pattern
906     to be anchored.)
907 nigel 75 .P
908     A dollar character is an assertion that is true only if the current matching
909 nigel 63 point is at the end of the subject string, or immediately before a newline
910 nigel 91 at the end of the string (by default). Dollar need not be the last character of
911     the pattern if a number of alternatives are involved, but it should be the last
912     item in any branch in which it appears. Dollar has no special meaning in a
913     character class.
914 nigel 75 .P
915 nigel 63 The meaning of dollar can be changed so that it matches only at the very end of
916     the string, by setting the PCRE_DOLLAR_ENDONLY option at compile time. This
917 nigel 75 does not affect the \eZ assertion.
918     .P
919 nigel 63 The meanings of the circumflex and dollar characters are changed if the
920 nigel 91 PCRE_MULTILINE option is set. When this is the case, a circumflex matches
921     immediately after internal newlines as well as at the start of the subject
922     string. It does not match after a newline that ends the string. A dollar
923     matches before any newlines in the string, as well as at the very end, when
924     PCRE_MULTILINE is set. When newline is specified as the two-character
925     sequence CRLF, isolated CR and LF characters do not indicate newlines.
926 nigel 75 .P
927 nigel 91 For example, the pattern /^abc$/ matches the subject string "def\enabc" (where
928     \en represents a newline) in multiline mode, but not otherwise. Consequently,
929     patterns that are anchored in single line mode because all branches start with
930     ^ are not anchored in multiline mode, and a match for circumflex is possible
931     when the \fIstartoffset\fP argument of \fBpcre_exec()\fP is non-zero. The
932     PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
933     .P
934 nigel 75 Note that the sequences \eA, \eZ, and \ez can be used to match the start and
935 nigel 63 end of the subject in both modes, and if all branches of a pattern start with
936 nigel 91 \eA it is always anchored, whether or not PCRE_MULTILINE is set.
937 nigel 75 .
938     .
939 ph10 514 .\" HTML <a name="fullstopdot"></a>
941 nigel 63 .rs
942     .sp
943     Outside a character class, a dot in the pattern matches any one character in
944 nigel 91 the subject string except (by default) a character that signifies the end of a
945 nigel 93 line. In UTF-8 mode, the matched character may be more than one byte long.
946 nigel 91 .P
947 nigel 93 When a line ending is defined as a single character, dot never matches that
948     character; when the two-character sequence CRLF is used, dot does not match CR
949     if it is immediately followed by LF, but otherwise it matches all characters
950     (including isolated CRs and LFs). When any Unicode line endings are being
951     recognized, dot does not match CR or LF or any of the other line ending
952     characters.
953     .P
954 nigel 91 The behaviour of dot with regard to newlines can be changed. If the PCRE_DOTALL
955 nigel 93 option is set, a dot matches any one character, without exception. If the
956     two-character sequence CRLF is present in the subject string, it takes two dots
957     to match it.
958 nigel 91 .P
959     The handling of dot is entirely independent of the handling of circumflex and
960     dollar, the only relationship being that they both involve newlines. Dot has no
961     special meaning in a character class.
962 ph10 514 .P
963 ph10 572 The escape sequence \eN behaves like a dot, except that it is not affected by
964     the PCRE_DOTALL option. In other words, it matches any character except one
965     that signifies the end of a line.
966 nigel 75 .
967     .
969 nigel 63 .rs
970     .sp
971 nigel 75 Outside a character class, the escape sequence \eC matches any one byte, both
972 nigel 93 in and out of UTF-8 mode. Unlike a dot, it always matches any line-ending
973     characters. The feature is provided in Perl in order to match individual bytes
974 ph10 572 in UTF-8 mode. Because it breaks up UTF-8 characters into individual bytes, the
975     rest of the string may start with a malformed UTF-8 character. For this reason,
976 nigel 93 the \eC escape sequence is best avoided.
977 nigel 75 .P
978     PCRE does not allow \eC to appear in lookbehind assertions
979     .\" HTML <a href="#lookbehind">
980     .\" </a>
981     (described below),
982     .\"
983     because in UTF-8 mode this would make it impossible to calculate the length of
984     the lookbehind.
985     .
986     .
987     .\" HTML <a name="characterclass"></a>
989 nigel 63 .rs
990     .sp
991     An opening square bracket introduces a character class, terminated by a closing
992 ph10 461 square bracket. A closing square bracket on its own is not special by default.
993     However, if the PCRE_JAVASCRIPT_COMPAT option is set, a lone closing square
994 ph10 456 bracket causes a compile-time error. If a closing square bracket is required as
995     a member of the class, it should be the first data character in the class
996     (after an initial circumflex, if present) or escaped with a backslash.
997 nigel 75 .P
998 nigel 63 A character class matches a single character in the subject. In UTF-8 mode, the
999 ph10 456 character may be more than one byte long. A matched character must be in the
1000     set of characters defined by the class, unless the first character in the class
1001 nigel 63 definition is a circumflex, in which case the subject character must not be in
1002     the set defined by the class. If a circumflex is actually required as a member
1003     of the class, ensure it is not the first character, or escape it with a
1004     backslash.
1005 nigel 75 .P
1006 nigel 63 For example, the character class [aeiou] matches any lower case vowel, while
1007     [^aeiou] matches any character that is not a lower case vowel. Note that a
1008 nigel 75 circumflex is just a convenient notation for specifying the characters that
1009     are in the class by enumerating those that are not. A class that starts with a
1010 ph10 456 circumflex is not an assertion; it still consumes a character from the subject
1011 nigel 75 string, and therefore it fails if the current pointer is at the end of the
1012     string.
1013     .P
1014 nigel 63 In UTF-8 mode, characters with values greater than 255 can be included in a
1015 nigel 75 class as a literal string of bytes, or by using the \ex{ escaping mechanism.
1016     .P
1017 nigel 63 When caseless matching is set, any letters in a class represent both their
1018     upper case and lower case versions, so for example, a caseless [aeiou] matches
1019     "A" as well as "a", and a caseless [^aeiou] does not match "A", whereas a
1020 nigel 77 caseful version would. In UTF-8 mode, PCRE always understands the concept of
1021     case for characters whose values are less than 128, so caseless matching is
1022     always possible. For characters with higher values, the concept of case is
1023     supported if PCRE is compiled with Unicode property support, but not otherwise.
1024 ph10 456 If you want to use caseless matching in UTF8-mode for characters 128 and above,
1025     you must ensure that PCRE is compiled with Unicode property support as well as
1026     with UTF-8 support.
1027 nigel 75 .P
1028 nigel 93 Characters that might indicate line breaks are never treated in any special way
1029     when matching character classes, whatever line-ending sequence is in use, and
1030     whatever setting of the PCRE_DOTALL and PCRE_MULTILINE options is used. A class
1031     such as [^a] always matches one of these characters.
1032 nigel 75 .P
1033 nigel 63 The minus (hyphen) character can be used to specify a range of characters in a
1034     character class. For example, [d-m] matches any letter between d and m,
1035     inclusive. If a minus character is required in a class, it must be escaped with
1036     a backslash or appear in a position where it cannot be interpreted as
1037     indicating a range, typically as the first or last character in the class.
1038 nigel 75 .P
1039 nigel 63 It is not possible to have the literal character "]" as the end character of a
1040     range. A pattern such as [W-]46] is interpreted as a class of two characters
1041     ("W" and "-") followed by a literal string "46]", so it would match "W46]" or
1042     "-46]". However, if the "]" is escaped with a backslash it is interpreted as
1043 nigel 75 the end of range, so [W-\e]46] is interpreted as a class containing a range
1044     followed by two other characters. The octal or hexadecimal representation of
1045     "]" can also be used to end a range.
1046     .P
1047 nigel 63 Ranges operate in the collating sequence of character values. They can also be
1048 nigel 75 used for characters specified numerically, for example [\e000-\e037]. In UTF-8
1049 nigel 63 mode, ranges can include characters whose values are greater than 255, for
1050 nigel 75 example [\ex{100}-\ex{2ff}].
1051     .P
1052 nigel 63 If a range that includes letters is used when caseless matching is set, it
1053     matches the letters in either case. For example, [W-c] is equivalent to
1054 nigel 75 [][\e\e^_`wxyzabc], matched caselessly, and in non-UTF-8 mode, if character
1055 ph10 139 tables for a French locale are in use, [\exc8-\excb] matches accented E
1056 nigel 75 characters in both cases. In UTF-8 mode, PCRE supports the concept of case for
1057     characters with values greater than 128 only when it is compiled with Unicode
1058     property support.
1059     .P
1060 ph10 575 The character escape sequences \ed, \eD, \eh, \eH, \ep, \eP, \es, \eS, \ev,
1061     \eV, \ew, and \eW may appear in a character class, and add the characters that
1062     they match to the class. For example, [\edABCDEF] matches any hexadecimal
1063     digit. In UTF-8 mode, the PCRE_UCP option affects the meanings of \ed, \es, \ew
1064     and their upper case partners, just as it does when they appear outside a
1065     character class, as described in the section entitled
1066     .\" HTML <a href="#genericchartypes">
1067     .\" </a>
1068     "Generic character types"
1069     .\"
1070     above. The escape sequence \eb has a different meaning inside a character
1071     class; it matches the backspace character. The sequences \eB, \eN, \eR, and \eX
1072     are not special inside a character class. Like any other unrecognized escape
1073     sequences, they are treated as the literal characters "B", "N", "R", and "X" by
1074     default, but cause an error if the PCRE_EXTRA option is set.
1075     .P
1076     A circumflex can conveniently be used with the upper case character types to
1077 ph10 518 specify a more restricted set of characters than the matching lower case type.
1078 ph10 575 For example, the class [^\eW_] matches any letter or digit, but not underscore,
1079     whereas [\ew] includes underscore. A positive character class should be read as
1080     "something OR something OR ..." and a negative class as "NOT something AND NOT
1081     something AND NOT ...".
1082 nigel 75 .P
1083     The only metacharacters that are recognized in character classes are backslash,
1084     hyphen (only where it can be interpreted as specifying a range), circumflex
1085     (only at the start), opening square bracket (only when it can be interpreted as
1086     introducing a POSIX class name - see the next section), and the terminating
1087     closing square bracket. However, escaping other non-alphanumeric characters
1088     does no harm.
1089     .
1090     .
1092 nigel 63 .rs
1093     .sp
1094 nigel 75 Perl supports the POSIX notation for character classes. This uses names
1095 nigel 63 enclosed by [: and :] within the enclosing square brackets. PCRE also supports
1096     this notation. For example,
1097 nigel 75 .sp
1098 nigel 63 [01[:alpha:]%]
1099 nigel 75 .sp
1100 nigel 63 matches "0", "1", any alphabetic character, or "%". The supported class names
1101 ph10 518 are:
1102 nigel 75 .sp
1103 nigel 63 alnum letters and digits
1104     alpha letters
1105     ascii character codes 0 - 127
1106     blank space or tab only
1107     cntrl control characters
1108 nigel 75 digit decimal digits (same as \ed)
1109 nigel 63 graph printing characters, excluding space
1110     lower lower case letters
1111     print printing characters, including space
1112 ph10 518 punct printing characters, excluding letters and digits and space
1113 nigel 75 space white space (not quite the same as \es)
1114 nigel 63 upper upper case letters
1115 nigel 75 word "word" characters (same as \ew)
1116 nigel 63 xdigit hexadecimal digits
1117 nigel 75 .sp
1118 nigel 63 The "space" characters are HT (9), LF (10), VT (11), FF (12), CR (13), and
1119     space (32). Notice that this list includes the VT character (code 11). This
1120 nigel 75 makes "space" different to \es, which does not include VT (for Perl
1121 nigel 63 compatibility).
1122 nigel 75 .P
1123 nigel 63 The name "word" is a Perl extension, and "blank" is a GNU extension from Perl
1124     5.8. Another Perl extension is negation, which is indicated by a ^ character
1125     after the colon. For example,
1126 nigel 75 .sp
1127 nigel 63 [12[:^digit:]]
1128 nigel 75 .sp
1129 nigel 63 matches "1", "2", or any non-digit. PCRE (and Perl) also recognize the POSIX
1130     syntax [.ch.] and [=ch=] where "ch" is a "collating element", but these are not
1131     supported, and an error is given if they are encountered.
1132 nigel 75 .P
1133 ph10 518 By default, in UTF-8 mode, characters with values greater than 128 do not match
1134     any of the POSIX character classes. However, if the PCRE_UCP option is passed
1135 ph10 535 to \fBpcre_compile()\fP, some of the classes are changed so that Unicode
1136     character properties are used. This is achieved by replacing the POSIX classes
1137 ph10 518 by other sequences, as follows:
1138     .sp
1139     [:alnum:] becomes \ep{Xan}
1140     [:alpha:] becomes \ep{L}
1141 ph10 535 [:blank:] becomes \eh
1142 ph10 518 [:digit:] becomes \ep{Nd}
1143     [:lower:] becomes \ep{Ll}
1144 ph10 535 [:space:] becomes \ep{Xps}
1145 ph10 518 [:upper:] becomes \ep{Lu}
1146     [:word:] becomes \ep{Xwd}
1147     .sp
1148     Negated versions, such as [:^alpha:] use \eP instead of \ep. The other POSIX
1149     classes are unchanged, and match only characters with code points less than
1150     128.
1151 nigel 75 .
1152     .
1153     .SH "VERTICAL BAR"
1154 nigel 63 .rs
1155     .sp
1156     Vertical bar characters are used to separate alternative patterns. For example,
1157     the pattern
1158 nigel 75 .sp
1159 nigel 63 gilbert|sullivan
1160 nigel 75 .sp
1161 nigel 63 matches either "gilbert" or "sullivan". Any number of alternatives may appear,
1162 nigel 91 and an empty alternative is permitted (matching the empty string). The matching
1163     process tries each alternative in turn, from left to right, and the first one
1164     that succeeds is used. If the alternatives are within a subpattern
1165 nigel 75 .\" HTML <a href="#subpattern">
1166     .\" </a>
1167     (defined below),
1168     .\"
1169     "succeeds" means matching the rest of the main pattern as well as the
1170     alternative in the subpattern.
1171     .
1172     .
1174 nigel 63 .rs
1175     .sp
1176     The settings of the PCRE_CASELESS, PCRE_MULTILINE, PCRE_DOTALL, and
1177 ph10 231 PCRE_EXTENDED options (which are Perl-compatible) can be changed from within
1178     the pattern by a sequence of Perl option letters enclosed between "(?" and ")".
1179     The option letters are
1180 nigel 75 .sp
1181 nigel 63 i for PCRE_CASELESS
1182     m for PCRE_MULTILINE
1183     s for PCRE_DOTALL
1184     x for PCRE_EXTENDED
1185 nigel 75 .sp
1186 nigel 63 For example, (?im) sets caseless, multiline matching. It is also possible to
1187     unset these options by preceding the letter with a hyphen, and a combined
1188     setting and unsetting such as (?im-sx), which sets PCRE_CASELESS and
1189     PCRE_MULTILINE while unsetting PCRE_DOTALL and PCRE_EXTENDED, is also
1190     permitted. If a letter appears both before and after the hyphen, the option is
1191     unset.
1192 nigel 75 .P
1193 ph10 231 The PCRE-specific options PCRE_DUPNAMES, PCRE_UNGREEDY, and PCRE_EXTRA can be
1194     changed in the same way as the Perl-compatible options by using the characters
1195     J, U and X respectively.
1196     .P
1197 ph10 412 When one of these option changes occurs at top level (that is, not inside
1198     subpattern parentheses), the change applies to the remainder of the pattern
1199     that follows. If the change is placed right at the start of a pattern, PCRE
1200     extracts it into the global options (and it will therefore show up in data
1201     extracted by the \fBpcre_fullinfo()\fP function).
1202 nigel 75 .P
1203 nigel 93 An option change within a subpattern (see below for a description of
1204 ph10 572 subpatterns) affects only that part of the subpattern that follows it, so
1205 nigel 75 .sp
1206 nigel 63 (a(?i)b)c
1207 nigel 75 .sp
1208 nigel 63 matches abc and aBc and no other strings (assuming PCRE_CASELESS is not used).
1209     By this means, options can be made to have different settings in different
1210     parts of the pattern. Any changes made in one alternative do carry on
1211     into subsequent branches within the same subpattern. For example,
1212 nigel 75 .sp
1213 nigel 63 (a(?i)b|c)
1214 nigel 75 .sp
1215 nigel 63 matches "ab", "aB", "c", and "C", even though when matching "C" the first
1216     branch is abandoned before the option setting. This is because the effects of
1217     option settings happen at compile time. There would be some very weird
1218     behaviour otherwise.
1219 ph10 251 .P
1220     \fBNote:\fP There are other PCRE-specific options that can be set by the
1221     application when the compile or match functions are called. In some cases the
1222 ph10 412 pattern can contain special leading sequences such as (*CRLF) to override what
1223     the application has set or what has been defaulted. Details are given in the
1224     section entitled
1225 ph10 251 .\" HTML <a href="#newlineseq">
1226     .\" </a>
1227     "Newline sequences"
1228     .\"
1229 ph10 518 above. There are also the (*UTF8) and (*UCP) leading sequences that can be used
1230     to set UTF-8 and Unicode property modes; they are equivalent to setting the
1231     PCRE_UTF8 and the PCRE_UCP options, respectively.
1232 nigel 75 .
1233     .
1234     .\" HTML <a name="subpattern"></a>
1235 nigel 63 .SH SUBPATTERNS
1236     .rs
1237     .sp
1238     Subpatterns are delimited by parentheses (round brackets), which can be nested.
1239 nigel 75 Turning part of a pattern into a subpattern does two things:
1240     .sp
1241 nigel 63 1. It localizes a set of alternatives. For example, the pattern
1242 nigel 75 .sp
1243 nigel 63 cat(aract|erpillar|)
1244 nigel 75 .sp
1245 ph10 572 matches "cataract", "caterpillar", or "cat". Without the parentheses, it would
1246     match "cataract", "erpillar" or an empty string.
1247 nigel 75 .sp
1248     2. It sets up the subpattern as a capturing subpattern. This means that, when
1249     the whole pattern matches, that portion of the subject string that matched the
1250     subpattern is passed back to the caller via the \fIovector\fP argument of
1251     \fBpcre_exec()\fP. Opening parentheses are counted from left to right (starting
1252 ph10 572 from 1) to obtain numbers for the capturing subpatterns. For example, if the
1253     string "the red king" is matched against the pattern
1254 nigel 75 .sp
1255 nigel 63 the ((red|white) (king|queen))
1256 nigel 75 .sp
1257 nigel 63 the captured substrings are "red king", "red", and "king", and are numbered 1,
1258     2, and 3, respectively.
1259 nigel 75 .P
1260 nigel 63 The fact that plain parentheses fulfil two functions is not always helpful.
1261     There are often times when a grouping subpattern is required without a
1262     capturing requirement. If an opening parenthesis is followed by a question mark
1263     and a colon, the subpattern does not do any capturing, and is not counted when
1264     computing the number of any subsequent capturing subpatterns. For example, if
1265     the string "the white queen" is matched against the pattern
1266 nigel 75 .sp
1267 nigel 63 the ((?:red|white) (king|queen))
1268 nigel 75 .sp
1269 nigel 63 the captured substrings are "white queen" and "queen", and are numbered 1 and
1270 nigel 93 2. The maximum number of capturing subpatterns is 65535.
1271 nigel 75 .P
1272 nigel 63 As a convenient shorthand, if any option settings are required at the start of
1273     a non-capturing subpattern, the option letters may appear between the "?" and
1274     the ":". Thus the two patterns
1275 nigel 75 .sp
1276 nigel 63 (?i:saturday|sunday)
1277     (?:(?i)saturday|sunday)
1278 nigel 75 .sp
1279 nigel 63 match exactly the same set of strings. Because alternative branches are tried
1280     from left to right, and options are not reset until the end of the subpattern
1281     is reached, an option setting in one branch does affect subsequent branches, so
1282     the above patterns match "SUNDAY" as well as "Saturday".
1283 nigel 75 .
1284     .
1285 ph10 456 .\" HTML <a name="dupsubpatternnumber"></a>
1287     .rs
1288     .sp
1289 ph10 182 Perl 5.10 introduced a feature whereby each alternative in a subpattern uses
1290     the same numbers for its capturing parentheses. Such a subpattern starts with
1291     (?| and is itself a non-capturing subpattern. For example, consider this
1292 ph10 175 pattern:
1293     .sp
1294     (?|(Sat)ur|(Sun))day
1295 ph10 182 .sp
1296     Because the two alternatives are inside a (?| group, both sets of capturing
1297     parentheses are numbered one. Thus, when the pattern matches, you can look
1298     at captured substring number one, whichever alternative matched. This construct
1299     is useful when you want to capture part, but not all, of one of a number of
1300     alternatives. Inside a (?| group, parentheses are numbered as usual, but the
1301 ph10 175 number is reset at the start of each branch. The numbers of any capturing
1302 ph10 572 parentheses that follow the subpattern start after the highest number used in
1303     any branch. The following example is taken from the Perl documentation. The
1304     numbers underneath show in which buffer the captured content will be stored.
1305 ph10 175 .sp
1306     # before ---------------branch-reset----------- after
1307     / ( a ) (?| x ( y ) z | (p (q) r) | (t) u (v) ) ( z ) /x
1308     # 1 2 2 3 2 3 4
1309 ph10 182 .sp
1310 ph10 488 A back reference to a numbered subpattern uses the most recent value that is
1311     set for that number by any subpattern. The following pattern matches "abcabc"
1312     or "defdef":
1313 ph10 456 .sp
1314 ph10 461 /(?|(abc)|(def))\e1/
1315 ph10 456 .sp
1316     In contrast, a recursive or "subroutine" call to a numbered subpattern always
1317 ph10 461 refers to the first one in the pattern with the given number. The following
1318 ph10 456 pattern matches "abcabc" or "defabc":
1319     .sp
1320     /(?|(abc)|(def))(?1)/
1321     .sp
1322 ph10 459 If a
1323     .\" HTML <a href="#conditions">
1324     .\" </a>
1325     condition test
1326     .\"
1327     for a subpattern's having matched refers to a non-unique number, the test is
1328     true if any of the subpatterns of that number have matched.
1329     .P
1330     An alternative approach to using this "branch reset" feature is to use
1331 ph10 175 duplicate named subpatterns, as described in the next section.
1332     .
1333     .
1334 nigel 75 .SH "NAMED SUBPATTERNS"
1335 nigel 63 .rs
1336     .sp
1337     Identifying capturing parentheses by number is simple, but it can be very hard
1338     to keep track of the numbers in complicated regular expressions. Furthermore,
1339 nigel 75 if an expression is modified, the numbers may change. To help with this
1340 nigel 93 difficulty, PCRE supports the naming of subpatterns. This feature was not
1341     added to Perl until release 5.10. Python had the feature earlier, and PCRE
1342     introduced it at release 4.0, using the Python syntax. PCRE now supports both
1343 ph10 459 the Perl and the Python syntax. Perl allows identically numbered subpatterns to
1344     have different names, but PCRE does not.
1345 nigel 93 .P
1346     In PCRE, a subpattern can be named in one of three ways: (?<name>...) or
1347     (?'name'...) as in Perl, or (?P<name>...) as in Python. References to capturing
1348 nigel 91 parentheses from other parts of the pattern, such as
1349     .\" HTML <a href="#backreferences">
1350     .\" </a>
1351 ph10 488 back references,
1352 nigel 91 .\"
1353     .\" HTML <a href="#recursion">
1354     .\" </a>
1355     recursion,
1356     .\"
1357     and
1358     .\" HTML <a href="#conditions">
1359     .\" </a>
1360     conditions,
1361     .\"
1362     can be made by name as well as by number.
1363 nigel 75 .P
1364 nigel 91 Names consist of up to 32 alphanumeric characters and underscores. Named
1365 nigel 93 capturing parentheses are still allocated numbers as well as names, exactly as
1366     if the names were not present. The PCRE API provides function calls for
1367     extracting the name-to-number translation table from a compiled pattern. There
1368     is also a convenience function for extracting a captured substring by name.
1369 nigel 91 .P
1370     By default, a name must be unique within a pattern, but it is possible to relax
1371 ph10 457 this constraint by setting the PCRE_DUPNAMES option at compile time. (Duplicate
1372 ph10 461 names are also always permitted for subpatterns with the same number, set up as
1373 ph10 457 described in the previous section.) Duplicate names can be useful for patterns
1374     where only one instance of the named parentheses can match. Suppose you want to
1375     match the name of a weekday, either as a 3-letter abbreviation or as the full
1376     name, and in both cases you want to extract the abbreviation. This pattern
1377     (ignoring the line breaks) does the job:
1378 nigel 91 .sp
1379 nigel 93 (?<DN>Mon|Fri|Sun)(?:day)?|
1380     (?<DN>Tue)(?:sday)?|
1381     (?<DN>Wed)(?:nesday)?|
1382     (?<DN>Thu)(?:rsday)?|
1383     (?<DN>Sat)(?:urday)?
1384 nigel 91 .sp
1385     There are five capturing substrings, but only one is ever set after a match.
1386 ph10 182 (An alternative way of solving this problem is to use a "branch reset"
1387 ph10 175 subpattern, as described in the previous section.)
1388     .P
1389 nigel 91 The convenience function for extracting the data by name returns the substring
1390 nigel 93 for the first (and in this example, the only) subpattern of that name that
1391 ph10 461 matched. This saves searching to find which numbered subpattern it was.
1392 ph10 459 .P
1393 ph10 488 If you make a back reference to a non-unique named subpattern from elsewhere in
1394 ph10 459 the pattern, the one that corresponds to the first occurrence of the name is
1395     used. In the absence of duplicate numbers (see the previous section) this is
1396     the one with the lowest number. If you use a named reference in a condition
1397     test (see the
1398     .\"
1399     .\" HTML <a href="#conditions">
1400     .\" </a>
1401     section about conditions
1402     .\"
1403 ph10 461 below), either to check whether a subpattern has matched, or to check for
1404 ph10 459 recursion, all subpatterns with the same name are tested. If the condition is
1405     true for any one of them, the overall condition is true. This is the same
1406     behaviour as testing by number. For further details of the interfaces for
1407     handling named subpatterns, see the
1408 nigel 63 .\" HREF
1409 nigel 75 \fBpcreapi\fP
1410 nigel 63 .\"
1411     documentation.
1412 ph10 385 .P
1413     \fBWarning:\fP You cannot use different names to distinguish between two
1414 ph10 457 subpatterns with the same number because PCRE uses only the numbers when
1415     matching. For this reason, an error is given at compile time if different names
1416     are given to subpatterns with the same number. However, you can give the same
1417     name to subpatterns with the same number, even when PCRE_DUPNAMES is not set.
1418 nigel 75 .
1419     .
1420 nigel 63 .SH REPETITION
1421     .rs
1422     .sp
1423     Repetition is specified by quantifiers, which can follow any of the following
1424     items:
1425 nigel 75 .sp
1426 nigel 63 a literal data character
1427 nigel 93 the dot metacharacter
1428 nigel 75 the \eC escape sequence
1429     the \eX escape sequence (in UTF-8 mode with Unicode properties)
1430 nigel 93 the \eR escape sequence
1431 ph10 572 an escape such as \ed or \epL that matches a single character
1432 nigel 63 a character class
1433     a back reference (see next section)
1434     a parenthesized subpattern (unless it is an assertion)
1435 ph10 461 a recursive or "subroutine" call to a subpattern
1436 nigel 75 .sp
1437 nigel 63 The general repetition quantifier specifies a minimum and maximum number of
1438     permitted matches, by giving the two numbers in curly brackets (braces),
1439     separated by a comma. The numbers must be less than 65536, and the first must
1440     be less than or equal to the second. For example:
1441 nigel 75 .sp
1442 nigel 63 z{2,4}
1443 nigel 75 .sp
1444 nigel 63 matches "zz", "zzz", or "zzzz". A closing brace on its own is not a special
1445     character. If the second number is omitted, but the comma is present, there is
1446     no upper limit; if the second number and the comma are both omitted, the
1447     quantifier specifies an exact number of required matches. Thus
1448 nigel 75 .sp
1449 nigel 63 [aeiou]{3,}
1450 nigel 75 .sp
1451 nigel 63 matches at least 3 successive vowels, but may match many more, while
1452 nigel 75 .sp
1453     \ed{8}
1454     .sp
1455 nigel 63 matches exactly 8 digits. An opening curly bracket that appears in a position
1456     where a quantifier is not allowed, or one that does not match the syntax of a
1457     quantifier, is taken as a literal character. For example, {,6} is not a
1458     quantifier, but a literal string of four characters.
1459 nigel 75 .P
1460 nigel 63 In UTF-8 mode, quantifiers apply to UTF-8 characters rather than to individual
1461 nigel 75 bytes. Thus, for example, \ex{100}{2} matches two UTF-8 characters, each of
1462     which is represented by a two-byte sequence. Similarly, when Unicode property
1463     support is available, \eX{3} matches three Unicode extended sequences, each of
1464     which may be several bytes long (and they may be of different lengths).
1465     .P
1466 nigel 63 The quantifier {0} is permitted, causing the expression to behave as if the
1467 ph10 345 previous item and the quantifier were not present. This may be useful for
1468     subpatterns that are referenced as
1469 ph10 335 .\" HTML <a href="#subpatternsassubroutines">
1470     .\" </a>
1471     subroutines
1472     .\"
1473 ph10 572 from elsewhere in the pattern (but see also the section entitled
1474     .\" HTML <a href="#subdefine">
1475     .\" </a>
1476     "Defining subpatterns for use by reference only"
1477     .\"
1478     below). Items other than subpatterns that have a {0} quantifier are omitted
1479     from the compiled pattern.
1480 nigel 75 .P
1481 nigel 93 For convenience, the three most common quantifiers have single-character
1482     abbreviations:
1483 nigel 75 .sp
1484 nigel 63 * is equivalent to {0,}
1485     + is equivalent to {1,}
1486     ? is equivalent to {0,1}
1487 nigel 75 .sp
1488 nigel 63 It is possible to construct infinite loops by following a subpattern that can
1489     match no characters with a quantifier that has no upper limit, for example:
1490 nigel 75 .sp
1491 nigel 63 (a?)*
1492 nigel 75 .sp
1493 nigel 63 Earlier versions of Perl and PCRE used to give an error at compile time for
1494     such patterns. However, because there are cases where this can be useful, such
1495     patterns are now accepted, but if any repetition of the subpattern does in fact
1496     match no characters, the loop is forcibly broken.
1497 nigel 75 .P
1498 nigel 63 By default, the quantifiers are "greedy", that is, they match as much as
1499     possible (up to the maximum number of permitted times), without causing the
1500     rest of the pattern to fail. The classic example of where this gives problems
1501 nigel 75 is in trying to match comments in C programs. These appear between /* and */
1502     and within the comment, individual * and / characters may appear. An attempt to
1503     match C comments by applying the pattern
1504     .sp
1505     /\e*.*\e*/
1506     .sp
1507 nigel 63 to the string
1508 nigel 75 .sp
1509     /* first comment */ not comment /* second comment */
1510     .sp
1511 nigel 63 fails, because it matches the entire string owing to the greediness of the .*
1512     item.
1513 nigel 75 .P
1514 nigel 63 However, if a quantifier is followed by a question mark, it ceases to be
1515     greedy, and instead matches the minimum number of times possible, so the
1516     pattern
1517 nigel 75 .sp
1518     /\e*.*?\e*/
1519     .sp
1520 nigel 63 does the right thing with the C comments. The meaning of the various
1521     quantifiers is not otherwise changed, just the preferred number of matches.
1522     Do not confuse this use of question mark with its use as a quantifier in its
1523     own right. Because it has two uses, it can sometimes appear doubled, as in
1524 nigel 75 .sp
1525     \ed??\ed
1526     .sp
1527 nigel 63 which matches one digit by preference, but can match two if that is the only
1528     way the rest of the pattern matches.
1529 nigel 75 .P
1530 nigel 93 If the PCRE_UNGREEDY option is set (an option that is not available in Perl),
1531 nigel 63 the quantifiers are not greedy by default, but individual ones can be made
1532     greedy by following them with a question mark. In other words, it inverts the
1533     default behaviour.
1534 nigel 75 .P
1535 nigel 63 When a parenthesized subpattern is quantified with a minimum repeat count that
1536 nigel 75 is greater than 1 or with a limited maximum, more memory is required for the
1537 nigel 63 compiled pattern, in proportion to the size of the minimum or maximum.
1538 nigel 75 .P
1539 nigel 63 If a pattern starts with .* or .{0,} and the PCRE_DOTALL option (equivalent
1540 nigel 93 to Perl's /s) is set, thus allowing the dot to match newlines, the pattern is
1541 nigel 63 implicitly anchored, because whatever follows will be tried against every
1542     character position in the subject string, so there is no point in retrying the
1543     overall match at any position after the first. PCRE normally treats such a
1544 nigel 75 pattern as though it were preceded by \eA.
1545     .P
1546 nigel 63 In cases where it is known that the subject string contains no newlines, it is
1547     worth setting PCRE_DOTALL in order to obtain this optimization, or
1548     alternatively using ^ to indicate anchoring explicitly.
1549 nigel 75 .P
1550 nigel 63 However, there is one situation where the optimization cannot be used. When .*
1551 ph10 488 is inside capturing parentheses that are the subject of a back reference
1552 nigel 93 elsewhere in the pattern, a match at the start may fail where a later one
1553     succeeds. Consider, for example:
1554 nigel 75 .sp
1555     (.*)abc\e1
1556     .sp
1557 nigel 63 If the subject is "xyz123abc123" the match point is the fourth character. For
1558     this reason, such a pattern is not implicitly anchored.
1559 nigel 75 .P
1560 nigel 63 When a capturing subpattern is repeated, the value captured is the substring
1561     that matched the final iteration. For example, after
1562 nigel 75 .sp
1563     (tweedle[dume]{3}\es*)+
1564     .sp
1565 nigel 63 has matched "tweedledum tweedledee" the value of the captured substring is
1566     "tweedledee". However, if there are nested capturing subpatterns, the
1567     corresponding captured values may have been set in previous iterations. For
1568     example, after
1569 nigel 75 .sp
1570 nigel 63 /(a|(b))+/
1571 nigel 75 .sp
1572 nigel 63 matches "aba" the value of the second captured substring is "b".
1573 nigel 75 .
1574     .
1575     .\" HTML <a name="atomicgroup"></a>
1577 nigel 63 .rs
1578     .sp
1579 nigel 93 With both maximizing ("greedy") and minimizing ("ungreedy" or "lazy")
1580     repetition, failure of what follows normally causes the repeated item to be
1581     re-evaluated to see if a different number of repeats allows the rest of the
1582     pattern to match. Sometimes it is useful to prevent this, either to change the
1583     nature of the match, or to cause it fail earlier than it otherwise might, when
1584     the author of the pattern knows there is no point in carrying on.
1585 nigel 75 .P
1586     Consider, for example, the pattern \ed+foo when applied to the subject line
1587     .sp
1588 nigel 63 123456bar
1589 nigel 75 .sp
1590 nigel 63 After matching all 6 digits and then failing to match "foo", the normal
1591 nigel 75 action of the matcher is to try again with only 5 digits matching the \ed+
1592 nigel 63 item, and then with 4, and so on, before ultimately failing. "Atomic grouping"
1593     (a term taken from Jeffrey Friedl's book) provides the means for specifying
1594     that once a subpattern has matched, it is not to be re-evaluated in this way.
1595 nigel 75 .P
1596 nigel 93 If we use atomic grouping for the previous example, the matcher gives up
1597 nigel 63 immediately on failing to match "foo" the first time. The notation is a kind of
1598     special parenthesis, starting with (?> as in this example:
1599 nigel 75 .sp
1600     (?>\ed+)foo
1601     .sp
1602 nigel 63 This kind of parenthesis "locks up" the part of the pattern it contains once
1603     it has matched, and a failure further into the pattern is prevented from
1604     backtracking into it. Backtracking past it to previous items, however, works as
1605     normal.
1606 nigel 75 .P
1607 nigel 63 An alternative description is that a subpattern of this type matches the string
1608     of characters that an identical standalone pattern would match, if anchored at
1609     the current point in the subject string.
1610 nigel 75 .P
1611 nigel 63 Atomic grouping subpatterns are not capturing subpatterns. Simple cases such as
1612     the above example can be thought of as a maximizing repeat that must swallow
1613 nigel 75 everything it can. So, while both \ed+ and \ed+? are prepared to adjust the
1614 nigel 63 number of digits they match in order to make the rest of the pattern match,
1615 nigel 75 (?>\ed+) can only match an entire sequence of digits.
1616     .P
1617 nigel 63 Atomic groups in general can of course contain arbitrarily complicated
1618     subpatterns, and can be nested. However, when the subpattern for an atomic
1619     group is just a single repeated item, as in the example above, a simpler
1620     notation, called a "possessive quantifier" can be used. This consists of an
1621     additional + character following a quantifier. Using this notation, the
1622     previous example can be rewritten as
1623 nigel 75 .sp
1624     \ed++foo
1625     .sp
1626 ph10 208 Note that a possessive quantifier can be used with an entire group, for
1627     example:
1628     .sp
1629     (abc|xyz){2,3}+
1630     .sp
1631 nigel 63 Possessive quantifiers are always greedy; the setting of the PCRE_UNGREEDY
1632     option is ignored. They are a convenient notation for the simpler forms of
1633 nigel 93 atomic group. However, there is no difference in the meaning of a possessive
1634     quantifier and the equivalent atomic group, though there may be a performance
1635     difference; possessive quantifiers should be slightly faster.
1636 nigel 75 .P
1637 nigel 93 The possessive quantifier syntax is an extension to the Perl 5.8 syntax.
1638     Jeffrey Friedl originated the idea (and the name) in the first edition of his
1639     book. Mike McCloskey liked it, so implemented it when he built Sun's Java
1640     package, and PCRE copied it from there. It ultimately found its way into Perl
1641     at release 5.10.
1642 nigel 75 .P
1643 nigel 93 PCRE has an optimization that automatically "possessifies" certain simple
1644     pattern constructs. For example, the sequence A+B is treated as A++B because
1645     there is no point in backtracking into a sequence of A's when B must follow.
1646     .P
1647 nigel 63 When a pattern contains an unlimited repeat inside a subpattern that can itself
1648     be repeated an unlimited number of times, the use of an atomic group is the
1649     only way to avoid some failing matches taking a very long time indeed. The
1650     pattern
1651 nigel 75 .sp
1652     (\eD+|<\ed+>)*[!?]
1653     .sp
1654 nigel 63 matches an unlimited number of substrings that either consist of non-digits, or
1655     digits enclosed in <>, followed by either ! or ?. When it matches, it runs
1656     quickly. However, if it is applied to
1657 nigel 75 .sp
1658 nigel 63 aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
1659 nigel 75 .sp
1660 nigel 63 it takes a long time before reporting failure. This is because the string can
1661 nigel 75 be divided between the internal \eD+ repeat and the external * repeat in a
1662     large number of ways, and all have to be tried. (The example uses [!?] rather
1663     than a single character at the end, because both PCRE and Perl have an
1664     optimization that allows for fast failure when a single character is used. They
1665     remember the last single character that is required for a match, and fail early
1666     if it is not present in the string.) If the pattern is changed so that it uses
1667     an atomic group, like this:
1668     .sp
1669     ((?>\eD+)|<\ed+>)*[!?]
1670     .sp
1671 nigel 63 sequences of non-digits cannot be broken, and failure happens quickly.
1672 nigel 75 .
1673     .
1674     .\" HTML <a name="backreferences"></a>
1676 nigel 63 .rs
1677     .sp
1678     Outside a character class, a backslash followed by a digit greater than 0 (and
1679     possibly further digits) is a back reference to a capturing subpattern earlier
1680     (that is, to its left) in the pattern, provided there have been that many
1681     previous capturing left parentheses.
1682 nigel 75 .P
1683 nigel 63 However, if the decimal number following the backslash is less than 10, it is
1684     always taken as a back reference, and causes an error only if there are not
1685     that many capturing left parentheses in the entire pattern. In other words, the
1686     parentheses that are referenced need not be to the left of the reference for
1687 nigel 91 numbers less than 10. A "forward back reference" of this type can make sense
1688     when a repetition is involved and the subpattern to the right has participated
1689     in an earlier iteration.
1690     .P
1691 nigel 93 It is not possible to have a numerical "forward back reference" to a subpattern
1692     whose number is 10 or more using this syntax because a sequence such as \e50 is
1693     interpreted as a character defined in octal. See the subsection entitled
1694 nigel 91 "Non-printing characters"
1695 nigel 75 .\" HTML <a href="#digitsafterbackslash">
1696     .\" </a>
1697     above
1698     .\"
1699 nigel 93 for further details of the handling of digits following a backslash. There is
1700     no such problem when named parentheses are used. A back reference to any
1701     subpattern is possible using named parentheses (see below).
1702 nigel 75 .P
1703 nigel 93 Another way of avoiding the ambiguity inherent in the use of digits following a
1704 ph10 572 backslash is to use the \eg escape sequence. This escape must be followed by an
1705     unsigned number or a negative number, optionally enclosed in braces. These
1706     examples are all identical:
1707 nigel 93 .sp
1708     (ring), \e1
1709     (ring), \eg1
1710     (ring), \eg{1}
1711     .sp
1712 ph10 208 An unsigned number specifies an absolute reference without the ambiguity that
1713     is present in the older syntax. It is also useful when literal digits follow
1714     the reference. A negative number is a relative reference. Consider this
1715     example:
1716 nigel 93 .sp
1717     (abc(def)ghi)\eg{-1}
1718     .sp
1719     The sequence \eg{-1} is a reference to the most recently started capturing
1720 ph10 572 subpattern before \eg, that is, is it equivalent to \e2 in this example.
1721     Similarly, \eg{-2} would be equivalent to \e1. The use of relative references
1722     can be helpful in long patterns, and also in patterns that are created by
1723     joining together fragments that contain references within themselves.
1724 nigel 93 .P
1725 nigel 63 A back reference matches whatever actually matched the capturing subpattern in
1726     the current subject string, rather than anything matching the subpattern
1727     itself (see
1728     .\" HTML <a href="#subpatternsassubroutines">
1729     .\" </a>
1730     "Subpatterns as subroutines"
1731     .\"
1732     below for a way of doing that). So the pattern
1733 nigel 75 .sp
1734     (sens|respons)e and \e1ibility
1735     .sp
1736 nigel 63 matches "sense and sensibility" and "response and responsibility", but not
1737     "sense and responsibility". If caseful matching is in force at the time of the
1738     back reference, the case of letters is relevant. For example,
1739 nigel 75 .sp
1740     ((?i)rah)\es+\e1
1741     .sp
1742 nigel 63 matches "rah rah" and "RAH RAH", but not "RAH rah", even though the original
1743     capturing subpattern is matched caselessly.
1744 nigel 75 .P
1745 ph10 171 There are several different ways of writing back references to named
1746     subpatterns. The .NET syntax \ek{name} and the Perl syntax \ek<name> or
1747     \ek'name' are supported, as is the Python syntax (?P=name). Perl 5.10's unified
1748     back reference syntax, in which \eg can be used for both numeric and named
1749     references, is also supported. We could rewrite the above example in any of
1750 nigel 93 the following ways:
1751 nigel 75 .sp
1752 nigel 93 (?<p1>(?i)rah)\es+\ek<p1>
1753 ph10 171 (?'p1'(?i)rah)\es+\ek{p1}
1754 nigel 91 (?P<p1>(?i)rah)\es+(?P=p1)
1755 ph10 171 (?<p1>(?i)rah)\es+\eg{p1}
1756 nigel 75 .sp
1757 nigel 91 A subpattern that is referenced by name may appear in the pattern before or
1758     after the reference.
1759     .P
1760 nigel 63 There may be more than one back reference to the same subpattern. If a
1761     subpattern has not actually been used in a particular match, any back
1762 ph10 456 references to it always fail by default. For example, the pattern
1763 nigel 75 .sp
1764     (a|(bc))\e2
1765     .sp
1766 ph10 461 always fails if it starts to match "a" rather than "bc". However, if the
1767     PCRE_JAVASCRIPT_COMPAT option is set at compile time, a back reference to an
1768 ph10 456 unset value matches an empty string.
1769     .P
1770     Because there may be many capturing parentheses in a pattern, all digits
1771     following a backslash are taken as part of a potential back reference number.
1772     If the pattern continues with a digit character, some delimiter must be used to
1773     terminate the back reference. If the PCRE_EXTENDED option is set, this can be
1774     whitespace. Otherwise, the \eg{ syntax or an empty comment (see
1775 nigel 75 .\" HTML <a href="#comments">
1776     .\" </a>
1777     "Comments"
1778     .\"
1779     below) can be used.
1780 ph10 488 .
1781     .SS "Recursive back references"
1782     .rs
1783     .sp
1784 nigel 63 A back reference that occurs inside the parentheses to which it refers fails
1785 nigel 75 when the subpattern is first used, so, for example, (a\e1) never matches.
1786 nigel 63 However, such references can be useful inside repeated subpatterns. For
1787     example, the pattern
1788 nigel 75 .sp
1789     (a|b\e1)+
1790     .sp
1791 nigel 63 matches any number of "a"s and also "aba", "ababbaa" etc. At each iteration of
1792     the subpattern, the back reference matches the character string corresponding
1793     to the previous iteration. In order for this to work, the pattern must be such
1794     that the first iteration does not need to match the back reference. This can be
1795     done using alternation, as in the example above, or by a quantifier with a
1796     minimum of zero.
1797 ph10 488 .P
1798     Back references of this type cause the group that they reference to be treated
1799     as an
1800     .\" HTML <a href="#atomicgroup">
1801     .\" </a>
1802     atomic group.
1803     .\"
1804     Once the whole group has been matched, a subsequent matching failure cannot
1805     cause backtracking into the middle of the group.
1806 nigel 75 .
1807     .
1808     .\" HTML <a name="bigassertions"></a>
1809 nigel 63 .SH ASSERTIONS
1810     .rs
1811     .sp
1812     An assertion is a test on the characters following or preceding the current
1813     matching point that does not actually consume any characters. The simple
1814 nigel 75 assertions coded as \eb, \eB, \eA, \eG, \eZ, \ez, ^ and $ are described
1815     .\" HTML <a href="#smallassertions">
1816     .\" </a>
1817     above.
1818     .\"
1819     .P
1820 nigel 63 More complicated assertions are coded as subpatterns. There are two kinds:
1821     those that look ahead of the current position in the subject string, and those
1822 nigel 75 that look behind it. An assertion subpattern is matched in the normal way,
1823     except that it does not cause the current matching position to be changed.
1824     .P
1825     Assertion subpatterns are not capturing subpatterns, and may not be repeated,
1826     because it makes no sense to assert the same thing several times. If any kind
1827     of assertion contains capturing subpatterns within it, these are counted for
1828     the purposes of numbering the capturing subpatterns in the whole pattern.
1829     However, substring capturing is carried out only for positive assertions,
1830     because it does not make sense for negative assertions.
1831     .
1832     .
1833     .SS "Lookahead assertions"
1834     .rs
1835     .sp
1836 nigel 91 Lookahead assertions start with (?= for positive assertions and (?! for
1837     negative assertions. For example,
1838 nigel 75 .sp
1839     \ew+(?=;)
1840     .sp
1841 nigel 63 matches a word followed by a semicolon, but does not include the semicolon in
1842     the match, and
1843 nigel 75 .sp
1844 nigel 63 foo(?!bar)
1845 nigel 75 .sp
1846 nigel 63 matches any occurrence of "foo" that is not followed by "bar". Note that the
1847     apparently similar pattern
1848 nigel 75 .sp
1849 nigel 63 (?!foo)bar
1850 nigel 75 .sp
1851 nigel 63 does not find an occurrence of "bar" that is preceded by something other than
1852     "foo"; it finds any occurrence of "bar" whatsoever, because the assertion
1853     (?!foo) is always true when the next three characters are "bar". A
1854 nigel 75 lookbehind assertion is needed to achieve the other effect.
1855     .P
1856 nigel 63 If you want to force a matching failure at some point in a pattern, the most
1857     convenient way to do it is with (?!) because an empty string always matches, so
1858     an assertion that requires there not to be an empty string must always fail.
1859 ph10 572 The backtracking control verb (*FAIL) or (*F) is a synonym for (?!).
1860 nigel 75 .
1861     .
1862     .\" HTML <a name="lookbehind"></a>
1863     .SS "Lookbehind assertions"
1864     .rs
1865     .sp
1866 nigel 63 Lookbehind assertions start with (?<= for positive assertions and (?<! for
1867     negative assertions. For example,
1868 nigel 75 .sp
1869 nigel 63 (?<!foo)bar
1870 nigel 75 .sp
1871 nigel 63 does find an occurrence of "bar" that is not preceded by "foo". The contents of
1872     a lookbehind assertion are restricted such that all the strings it matches must
1873 nigel 91 have a fixed length. However, if there are several top-level alternatives, they
1874     do not all have to have the same fixed length. Thus
1875 nigel 75 .sp
1876 nigel 63 (?<=bullock|donkey)
1877 nigel 75 .sp
1878 nigel 63 is permitted, but
1879 nigel 75 .sp
1880 nigel 63 (?<!dogs?|cats?)
1881 nigel 75 .sp
1882 nigel 63 causes an error at compile time. Branches that match different length strings
1883     are permitted only at the top level of a lookbehind assertion. This is an
1884 ph10 572 extension compared with Perl, which requires all branches to match the same
1885     length of string. An assertion such as
1886 nigel 75 .sp
1887 nigel 63 (?<=ab(c|de))
1888 nigel 75 .sp
1889 nigel 63 is not permitted, because its single top-level branch can match two different
1890 ph10 454 lengths, but it is acceptable to PCRE if rewritten to use two top-level
1891     branches:
1892 nigel 75 .sp
1893 nigel 63 (?<=abc|abde)
1894 nigel 75 .sp
1895 ph10 572 In some cases, the escape sequence \eK
1896 ph10 168 .\" HTML <a href="#resetmatchstart">
1897     .\" </a>
1898     (see above)
1899     .\"
1900 ph10 461 can be used instead of a lookbehind assertion to get round the fixed-length
1901 ph10 454 restriction.
1902 ph10 168 .P
1903 nigel 63 The implementation of lookbehind assertions is, for each alternative, to
1904 nigel 93 temporarily move the current position back by the fixed length and then try to
1905 nigel 63 match. If there are insufficient characters before the current position, the
1906 nigel 93 assertion fails.
1907 nigel 75 .P
1908     PCRE does not allow the \eC escape (which matches a single byte in UTF-8 mode)
1909 nigel 63 to appear in lookbehind assertions, because it makes it impossible to calculate
1910 nigel 93 the length of the lookbehind. The \eX and \eR escapes, which can match
1911     different numbers of bytes, are also not permitted.
1912 nigel 75 .P
1913 ph10 454 .\" HTML <a href="#subpatternsassubroutines">
1914     .\" </a>
1915     "Subroutine"
1916     .\"
1917     calls (see below) such as (?2) or (?&X) are permitted in lookbehinds, as long
1918 ph10 461 as the subpattern matches a fixed-length string.
1919 ph10 454 .\" HTML <a href="#recursion">
1920     .\" </a>
1921     Recursion,
1922     .\"
1923     however, is not supported.
1924     .P
1925 nigel 93 Possessive quantifiers can be used in conjunction with lookbehind assertions to
1926 ph10 456 specify efficient matching of fixed-length strings at the end of subject
1927     strings. Consider a simple pattern such as
1928 nigel 75 .sp
1929 nigel 63 abcd$
1930 nigel 75 .sp
1931 nigel 63 when applied to a long string that does not match. Because matching proceeds
1932     from left to right, PCRE will look for each "a" in the subject and then see if
1933     what follows matches the rest of the pattern. If the pattern is specified as
1934 nigel 75 .sp
1935 nigel 63 ^.*abcd$
1936 nigel 75 .sp
1937 nigel 63 the initial .* matches the entire string at first, but when this fails (because
1938     there is no following "a"), it backtracks to match all but the last character,
1939     then all but the last two characters, and so on. Once again the search for "a"
1940     covers the entire string, from right to left, so we are no better off. However,
1941     if the pattern is written as
1942 nigel 75 .sp
1943 nigel 63 ^.*+(?<=abcd)
1944 nigel 75 .sp
1945 nigel 93 there can be no backtracking for the .*+ item; it can match only the entire
1946 nigel 63 string. The subsequent lookbehind assertion does a single test on the last four
1947     characters. If it fails, the match fails immediately. For long strings, this
1948     approach makes a significant difference to the processing time.
1949 nigel 75 .
1950     .
1951     .SS "Using multiple assertions"
1952     .rs
1953     .sp
1954 nigel 63 Several assertions (of any sort) may occur in succession. For example,
1955 nigel 75 .sp
1956     (?<=\ed{3})(?<!999)foo
1957     .sp
1958 nigel 63 matches "foo" preceded by three digits that are not "999". Notice that each of
1959     the assertions is applied independently at the same point in the subject
1960     string. First there is a check that the previous three characters are all
1961     digits, and then there is a check that the same three characters are not "999".
1962 nigel 75 This pattern does \fInot\fP match "foo" preceded by six characters, the first
1963 nigel 63 of which are digits and the last three of which are not "999". For example, it
1964     doesn't match "123abcfoo". A pattern to do that is
1965 nigel 75 .sp
1966     (?<=\ed{3}...)(?<!999)foo
1967     .sp
1968 nigel 63 This time the first assertion looks at the preceding six characters, checking
1969     that the first three are digits, and then the second assertion checks that the
1970     preceding three characters are not "999".
1971 nigel 75 .P
1972 nigel 63 Assertions can be nested in any combination. For example,
1973 nigel 75 .sp
1974 nigel 63 (?<=(?<!foo)bar)baz
1975 nigel 75 .sp
1976 nigel 63 matches an occurrence of "baz" that is preceded by "bar" which in turn is not
1977     preceded by "foo", while
1978 nigel 75 .sp
1979     (?<=\ed{3}(?!999)...)foo
1980     .sp
1981     is another pattern that matches "foo" preceded by three digits and any three
1982 nigel 63 characters that are not "999".
1983 nigel 75 .
1984     .
1985 nigel 91 .\" HTML <a name="conditions"></a>
1987 nigel 63 .rs
1988     .sp
1989     It is possible to cause the matching process to obey a subpattern
1990     conditionally or to choose between two alternative subpatterns, depending on
1991 ph10 461 the result of an assertion, or whether a specific capturing subpattern has
1992 ph10 456 already been matched. The two possible forms of conditional subpattern are:
1993 nigel 75 .sp
1994 nigel 63 (?(condition)yes-pattern)
1995     (?(condition)yes-pattern|no-pattern)
1996 nigel 75 .sp
1997 nigel 63 If the condition is satisfied, the yes-pattern is used; otherwise the
1998     no-pattern (if present) is used. If there are more than two alternatives in the
1999 ph10 557 subpattern, a compile-time error occurs. Each of the two alternatives may
2000     itself contain nested subpatterns of any form, including conditional
2001     subpatterns; the restriction to two alternatives applies only at the level of
2002     the condition. This pattern fragment is an example where the alternatives are
2003     complex:
2004     .sp
2005     (?(1) (A|B|C) | (D | (?(2)E|F) | E) )
2006     .sp
2007 nigel 75 .P
2008 nigel 93 There are four kinds of condition: references to subpatterns, references to
2009     recursion, a pseudo-condition called DEFINE, and assertions.
2010     .
2011     .SS "Checking for a used subpattern by number"
2012     .rs
2013     .sp
2014     If the text between the parentheses consists of a sequence of digits, the
2015 ph10 456 condition is true if a capturing subpattern of that number has previously
2016 ph10 461 matched. If there is more than one capturing subpattern with the same number
2017     (see the earlier
2018 ph10 456 .\"
2019     .\" HTML <a href="#recursion">
2020     .\" </a>
2021     section about duplicate subpattern numbers),
2022     .\"
2023 ph10 572 the condition is true if any of them have matched. An alternative notation is
2024 ph10 456 to precede the digits with a plus or minus sign. In this case, the subpattern
2025     number is relative rather than absolute. The most recently opened parentheses
2026 ph10 572 can be referenced by (?(-1), the next most recent by (?(-2), and so on. Inside
2027     loops it can also make sense to refer to subsequent groups. The next
2028     parentheses to be opened can be referenced as (?(+1), and so on. (The value
2029     zero in any of these forms is not used; it provokes a compile-time error.)
2030 nigel 91 .P
2031     Consider the following pattern, which contains non-significant white space to
2032     make it more readable (assume the PCRE_EXTENDED option) and to divide it into
2033     three parts for ease of discussion:
2034 nigel 75 .sp
2035     ( \e( )? [^()]+ (?(1) \e) )
2036     .sp
2037 nigel 63 The first part matches an optional opening parenthesis, and if that
2038     character is present, sets it as the first captured substring. The second part
2039     matches one or more characters that are not parentheses. The third part is a
2040 ph10 572 conditional subpattern that tests whether or not the first set of parentheses
2041     matched. If they did, that is, if subject started with an opening parenthesis,
2042 nigel 63 the condition is true, and so the yes-pattern is executed and a closing
2043     parenthesis is required. Otherwise, since no-pattern is not present, the
2044     subpattern matches nothing. In other words, this pattern matches a sequence of
2045 nigel 93 non-parentheses, optionally enclosed in parentheses.
2046 ph10 167 .P
2047 ph10 172 If you were embedding this pattern in a larger one, you could use a relative
2048 ph10 167 reference:
2049     .sp
2050     ...other stuff... ( \e( )? [^()]+ (?(-1) \e) ) ...
2051     .sp
2052     This makes the fragment independent of the parentheses in the larger pattern.
2053 nigel 93 .
2054     .SS "Checking for a used subpattern by name"
2055     .rs
2056 nigel 91 .sp
2057 nigel 93 Perl uses the syntax (?(<name>)...) or (?('name')...) to test for a used
2058     subpattern by name. For compatibility with earlier versions of PCRE, which had
2059     this facility before Perl, the syntax (?(name)...) is also recognized. However,
2060     there is a possible ambiguity with this syntax, because subpattern names may
2061     consist entirely of digits. PCRE looks first for a named subpattern; if it
2062     cannot find one and the name consists entirely of digits, PCRE looks for a
2063     subpattern of that number, which must be greater than zero. Using subpattern
2064     names that consist entirely of digits is not recommended.
2065     .P
2066     Rewriting the above example to use a named subpattern gives this:
2067 nigel 91 .sp
2068 nigel 93 (?<OPEN> \e( )? [^()]+ (?(<OPEN>) \e) )
2069     .sp
2070 ph10 461 If the name used in a condition of this kind is a duplicate, the test is
2071     applied to all subpatterns of the same name, and is true if any one of them has
2072 ph10 459 matched.
2073 nigel 93 .
2074     .SS "Checking for pattern recursion"
2075     .rs
2076     .sp
2077 nigel 91 If the condition is the string (R), and there is no subpattern with the name R,
2078 nigel 93 the condition is true if a recursive call to the whole pattern or any
2079     subpattern has been made. If digits or a name preceded by ampersand follow the
2080     letter R, for example:
2081     .sp
2082     (?(R3)...) or (?(R&name)...)
2083     .sp
2084 ph10 456 the condition is true if the most recent recursion is into a subpattern whose
2085 nigel 93 number or name is given. This condition does not check the entire recursion
2086 ph10 461 stack. If the name used in a condition of this kind is a duplicate, the test is
2087     applied to all subpatterns of the same name, and is true if any one of them is
2088     the most recent recursion.
2089 nigel 75 .P
2090 ph10 461 At "top level", all these recursion test conditions are false.
2091 ph10 454 .\" HTML <a href="#recursion">
2092     .\" </a>
2093 ph10 459 The syntax for recursive patterns
2094 ph10 454 .\"
2095 ph10 459 is described below.
2096 nigel 93 .
2097 ph10 572 .\" HTML <a name="subdefine"></a>
2098 nigel 93 .SS "Defining subpatterns for use by reference only"
2099     .rs
2100     .sp
2101     If the condition is the string (DEFINE), and there is no subpattern with the
2102     name DEFINE, the condition is always false. In this case, there may be only one
2103     alternative in the subpattern. It is always skipped if control reaches this
2104     point in the pattern; the idea of DEFINE is that it can be used to define
2105 ph10 461 "subroutines" that can be referenced from elsewhere. (The use of
2106 ph10 454 .\" HTML <a href="#subpatternsassubroutines">
2107     .\" </a>
2108     "subroutines"
2109     .\"
2110 ph10 572 is described below.) For example, a pattern to match an IPv4 address such as
2111     "" could be written like this (ignore whitespace and line
2112     breaks):
2113 nigel 93 .sp
2114     (?(DEFINE) (?<byte> 2[0-4]\ed | 25[0-5] | 1\ed\ed | [1-9]?\ed) )
2115     \eb (?&byte) (\e.(?&byte)){3} \eb
2116     .sp
2117     The first part of the pattern is a DEFINE group inside which a another group
2118     named "byte" is defined. This matches an individual component of an IPv4
2119     address (a number less than 256). When matching takes place, this part of the
2120 ph10 456 pattern is skipped because DEFINE acts like a false condition. The rest of the
2121     pattern uses references to the named group to match the four dot-separated
2122     components of an IPv4 address, insisting on a word boundary at each end.
2123 nigel 93 .
2124     .SS "Assertion conditions"
2125     .rs
2126     .sp
2127     If the condition is not in any of the above formats, it must be an assertion.
2128 nigel 63 This may be a positive or negative lookahead or lookbehind assertion. Consider
2129     this pattern, again containing non-significant white space, and with the two
2130     alternatives on the second line:
2131 nigel 75 .sp
2132 nigel 63 (?(?=[^a-z]*[a-z])
2133 nigel 75 \ed{2}-[a-z]{3}-\ed{2} | \ed{2}-\ed{2}-\ed{2} )
2134     .sp
2135 nigel 63 The condition is a positive lookahead assertion that matches an optional
2136     sequence of non-letters followed by a letter. In other words, it tests for the
2137     presence of at least one letter in the subject. If a letter is found, the
2138     subject is matched against the first alternative; otherwise it is matched
2139     against the second. This pattern matches strings in one of the two forms
2140     dd-aaa-dd or dd-dd-dd, where aaa are letters and dd are digits.
2141 nigel 75 .
2142     .
2143     .\" HTML <a name="comments"></a>
2144 nigel 63 .SH COMMENTS
2145     .rs
2146     .sp
2147 ph10 562 There are two ways of including comments in patterns that are processed by
2148     PCRE. In both cases, the start of the comment must not be in a character class,
2149     nor in the middle of any other sequence of related characters such as (?: or a
2150     subpattern name or number. The characters that make up a comment play no part
2151     in the pattern matching.
2152     .P
2153 nigel 75 The sequence (?# marks the start of a comment that continues up to the next
2154 ph10 562 closing parenthesis. Nested parentheses are not permitted. If the PCRE_EXTENDED
2155     option is set, an unescaped # character also introduces a comment, which in
2156     this case continues to immediately after the next newline character or
2157     character sequence in the pattern. Which characters are interpreted as newlines
2158     is controlled by the options passed to \fBpcre_compile()\fP or by a special
2159     sequence at the start of the pattern, as described in the section entitled
2160 ph10 572 .\" HTML <a href="#newlines">
2161 ph10 556 .\" </a>
2162     "Newline conventions"
2163     .\"
2164 ph10 572 above. Note that the end of this type of comment is a literal newline sequence
2165     in the pattern; escape sequences that happen to represent a newline do not
2166     count. For example, consider this pattern when PCRE_EXTENDED is set, and the
2167     default newline convention is in force:
2168 ph10 556 .sp
2169     abc #comment \en still comment
2170     .sp
2171     On encountering the # character, \fBpcre_compile()\fP skips along, looking for
2172     a newline in the pattern. The sequence \en is still literal at this stage, so
2173     it does not terminate the comment. Only an actual character with the code value
2174 ph10 562 0x0a (the default newline) does so.
2175 nigel 75 .
2176     .
2177 nigel 91 .\" HTML <a name="recursion"></a>
2178 nigel 75 .SH "RECURSIVE PATTERNS"
2179 nigel 63 .rs
2180     .sp
2181     Consider the problem of matching a string in parentheses, allowing for
2182     unlimited nested parentheses. Without the use of recursion, the best that can
2183     be done is to use a pattern that matches up to some fixed depth of nesting. It
2184 nigel 93 is not possible to handle an arbitrary nesting depth.
2185     .P
2186     For some time, Perl has provided a facility that allows regular expressions to
2187     recurse (amongst other things). It does this by interpolating Perl code in the
2188     expression at run time, and the code can refer to the expression itself. A Perl
2189     pattern using code interpolation to solve the parentheses problem can be
2190     created like this:
2191 nigel 75 .sp
2192     $re = qr{\e( (?: (?>[^()]+) | (?p{$re}) )* \e)}x;
2193     .sp
2194 nigel 63 The (?p{...}) item interpolates Perl code at run time, and in this case refers
2195 nigel 93 recursively to the pattern in which it appears.
2196 nigel 75 .P
2197 nigel 93 Obviously, PCRE cannot support the interpolation of Perl code. Instead, it
2198     supports special syntax for recursion of the entire pattern, and also for
2199     individual subpattern recursion. After its introduction in PCRE and Python,
2200 ph10 453 this kind of recursion was subsequently introduced into Perl at release 5.10.
2201 nigel 75 .P
2202 nigel 93 A special item that consists of (? followed by a number greater than zero and a
2203     closing parenthesis is a recursive call of the subpattern of the given number,
2204 ph10 461 provided that it occurs inside that subpattern. (If not, it is a
2205 ph10 454 .\" HTML <a href="#subpatternsassubroutines">
2206     .\" </a>
2207     "subroutine"
2208     .\"
2209 nigel 93 call, which is described in the next section.) The special item (?R) or (?0) is
2210     a recursive call of the entire regular expression.
2211 nigel 87 .P
2212     This PCRE pattern solves the nested parentheses problem (assume the
2213     PCRE_EXTENDED option is set so that white space is ignored):
2214 nigel 75 .sp
2215 ph10 456 \e( ( [^()]++ | (?R) )* \e)
2216 nigel 75 .sp
2217 nigel 63 First it matches an opening parenthesis. Then it matches any number of
2218     substrings which can either be a sequence of non-parentheses, or a recursive
2219 nigel 87 match of the pattern itself (that is, a correctly parenthesized substring).
2220 ph10 461 Finally there is a closing parenthesis. Note the use of a possessive quantifier
2221 ph10 456 to avoid backtracking into sequences of non-parentheses.
2222 nigel 75 .P
2223 nigel 63 If this were part of a larger pattern, you would not want to recurse the entire
2224     pattern, so instead you could use this:
2225 nigel 75 .sp
2226 ph10 456 ( \e( ( [^()]++ | (?1) )* \e) )
2227 nigel 75 .sp
2228 nigel 63 We have put the pattern into parentheses, and caused the recursion to refer to
2229 ph10 172 them instead of the whole pattern.
2230 ph10 166 .P
2231     In a larger pattern, keeping track of parenthesis numbers can be tricky. This
2232 ph10 572 is made easier by the use of relative references. Instead of (?1) in the
2233     pattern above you can write (?-2) to refer to the second most recently opened
2234     parentheses preceding the recursion. In other words, a negative number counts
2235     capturing parentheses leftwards from the point at which it is encountered.
2236 ph10 166 .P
2237     It is also possible to refer to subsequently opened parentheses, by writing
2238     references such as (?+2). However, these cannot be recursive because the
2239     reference is not inside the parentheses that are referenced. They are always
2240 ph10 454 .\" HTML <a href="#subpatternsassubroutines">
2241     .\" </a>
2242     "subroutine"
2243     .\"
2244     calls, as described in the next section.
2245 ph10 166 .P
2246     An alternative approach is to use named parentheses instead. The Perl syntax
2247     for this is (?&name); PCRE's earlier syntax (?P>name) is also supported. We
2248     could rewrite the above example as follows:
2249 nigel 75 .sp
2250 ph10 456 (?<pn> \e( ( [^()]++ | (?&pn) )* \e) )
2251 nigel 75 .sp
2252 nigel 93 If there is more than one subpattern with the same name, the earliest one is
2253 ph10 172 used.
2254 ph10 166 .P
2255     This particular example pattern that we have been looking at contains nested
2256 ph10 456 unlimited repeats, and so the use of a possessive quantifier for matching
2257     strings of non-parentheses is important when applying the pattern to strings
2258     that do not match. For example, when this pattern is applied to
2259 nigel 75 .sp
2260 nigel 63 (aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa()
2261 nigel 75 .sp
2262 ph10 456 it yields "no match" quickly. However, if a possessive quantifier is not used,
2263 nigel 63 the match runs for a very long time indeed because there are so many different
2264     ways the + and * repeats can carve up the subject, and all have to be tested
2265     before failure can be reported.
2266 nigel 75 .P
2267 ph10 464 At the end of a match, the values of capturing parentheses are those from
2268     the outermost level. If you want to obtain intermediate values, a callout
2269     function can be used (see below and the
2270 nigel 63 .\" HREF
2271 nigel 75 \fBpcrecallout\fP
2272 nigel 63 .\"
2273     documentation). If the pattern above is matched against
2274 nigel 75 .sp
2275 nigel 63 (ab(cd)ef)
2276 nigel 75 .sp
2277 ph10 464 the value for the inner capturing parentheses (numbered 2) is "ef", which is
2278     the last value taken on at the top level. If a capturing subpattern is not
2279     matched at the top level, its final value is unset, even if it is (temporarily)
2280     set at a deeper level.
2281 nigel 75 .P
2282 ph10 464 If there are more than 15 capturing parentheses in a pattern, PCRE has to
2283     obtain extra memory to store data during a recursion, which it does by using
2284     \fBpcre_malloc\fP, freeing it via \fBpcre_free\fP afterwards. If no memory can
2285     be obtained, the match fails with the PCRE_ERROR_NOMEMORY error.
2286     .P
2287 nigel 63 Do not confuse the (?R) item with the condition (R), which tests for recursion.
2288     Consider this pattern, which matches text in angle brackets, allowing for
2289     arbitrary nesting. Only digits are allowed in nested brackets (that is, when
2290     recursing), whereas any characters are permitted at the outer level.
2291 nigel 75 .sp
2292     < (?: (?(R) \ed++ | [^<>]*+) | (?R)) * >
2293     .sp
2294 nigel 63 In this pattern, (?(R) is the start of a conditional subpattern, with two
2295     different alternatives for the recursive and non-recursive cases. The (?R) item
2296     is the actual recursive call.
2297 nigel 75 .
2298     .
2299 ph10 453 .\" HTML <a name="recursiondifference"></a>
2300     .SS "Recursion difference from Perl"
2301     .rs
2302     .sp
2303     In PCRE (like Python, but unlike Perl), a recursive subpattern call is always
2304     treated as an atomic group. That is, once it has matched some of the subject
2305     string, it is never re-entered, even if it contains untried alternatives and
2306 ph10 461 there is a subsequent matching failure. This can be illustrated by the
2307     following pattern, which purports to match a palindromic string that contains
2308 ph10 453 an odd number of characters (for example, "a", "aba", "abcba", "abcdcba"):
2309     .sp
2310     ^(.|(.)(?1)\e2)$
2311     .sp
2312 ph10 461 The idea is that it either matches a single character, or two identical
2313     characters surrounding a sub-palindrome. In Perl, this pattern works; in PCRE
2314 ph10 453 it does not if the pattern is longer than three characters. Consider the
2315     subject string "abcba":
2316     .P
2317 ph10 461 At the top level, the first character is matched, but as it is not at the end
2318 ph10 453 of the string, the first alternative fails; the second alternative is taken
2319     and the recursion kicks in. The recursive call to subpattern 1 successfully
2320     matches the next character ("b"). (Note that the beginning and end of line
2321     tests are not part of the recursion).
2322     .P
2323     Back at the top level, the next character ("c") is compared with what
2324 ph10 461 subpattern 2 matched, which was "a". This fails. Because the recursion is
2325 ph10 453 treated as an atomic group, there are now no backtracking points, and so the
2326     entire match fails. (Perl is able, at this point, to re-enter the recursion and
2327     try the second alternative.) However, if the pattern is written with the
2328     alternatives in the other order, things are different:
2329     .sp
2330     ^((.)(?1)\e2|.)$
2331     .sp
2332 ph10 461 This time, the recursing alternative is tried first, and continues to recurse
2333     until it runs out of characters, at which point the recursion fails. But this
2334     time we do have another alternative to try at the higher level. That is the big
2335 ph10 453 difference: in the previous case the remaining alternative is at a deeper
2336     recursion level, which PCRE cannot use.
2337     .P
2338 ph10 572 To change the pattern so that it matches all palindromic strings, not just
2339     those with an odd number of characters, it is tempting to change the pattern to
2340     this:
2341 ph10 453 .sp
2342     ^((.)(?1)\e2|.?)$
2343     .sp
2344 ph10 461 Again, this works in Perl, but not in PCRE, and for the same reason. When a
2345     deeper recursion has matched a single character, it cannot be entered again in
2346     order to match an empty string. The solution is to separate the two cases, and
2347 ph10 453 write out the odd and even cases as alternatives at the higher level:
2348     .sp
2349     ^(?:((.)(?1)\e2|)|((.)(?3)\e4|.))
2350 ph10 461 .sp
2351     If you want to match typical palindromic phrases, the pattern has to ignore all
2352 ph10 453 non-word characters, which can be done like this:
2353     .sp
2354 ph10 461 ^\eW*+(?:((.)\eW*+(?1)\eW*+\e2|)|((.)\eW*+(?3)\eW*+\e4|\eW*+.\eW*+))\eW*+$
2355 ph10 453 .sp
2356 ph10 461 If run with the PCRE_CASELESS option, this pattern matches phrases such as "A
2357     man, a plan, a canal: Panama!" and it works well in both PCRE and Perl. Note
2358     the use of the possessive quantifier *+ to avoid backtracking into sequences of
2359 ph10 453 non-word characters. Without this, PCRE takes a great deal longer (ten times or
2360     more) to match typical phrases, and Perl takes so long that you think it has
2361     gone into a loop.
2362 ph10 456 .P
2363     \fBWARNING\fP: The palindrome-matching patterns above work only if the subject
2364     string does not start with a palindrome that is shorter than the entire string.
2365     For example, although "abcba" is correctly matched, if the subject is "ababa",
2366     PCRE finds the palindrome "aba" at the start, then fails at top level because
2367     the end of the string does not follow. Once again, it cannot jump back into the
2368     recursion to try other alternatives, so the entire match fails.
2369 ph10 453 .
2370     .
2371 nigel 63 .\" HTML <a name="subpatternsassubroutines"></a>
2373 nigel 63 .rs
2374     .sp
2375     If the syntax for a recursive subpattern reference (either by number or by
2376     name) is used outside the parentheses to which it refers, it operates like a
2377 nigel 93 subroutine in a programming language. The "called" subpattern may be defined
2378 ph10 166 before or after the reference. A numbered reference can be absolute or
2379     relative, as in these examples:
2380 nigel 75 .sp
2381 ph10 166 (...(absolute)...)...(?2)...
2382     (...(relative)...)...(?-1)...
2383 ph10 172 (...(?+1)...(relative)...
2384 ph10 166 .sp
2385     An earlier example pointed out that the pattern
2386     .sp
2387 nigel 75 (sens|respons)e and \e1ibility
2388     .sp
2389 nigel 63 matches "sense and sensibility" and "response and responsibility", but not
2390     "sense and responsibility". If instead the pattern
2391 nigel 75 .sp
2392 nigel 63 (sens|respons)e and (?1)ibility
2393 nigel 75 .sp
2394 nigel 63 is used, it does match "sense and responsibility" as well as the other two
2395 nigel 93 strings. Another example is given in the discussion of DEFINE above.
2396 nigel 87 .P
2397 ph10 464 Like recursive subpatterns, a subroutine call is always treated as an atomic
2398 nigel 87 group. That is, once it has matched some of the subject string, it is never
2399     re-entered, even if it contains untried alternatives and there is a subsequent
2400 ph10 469 matching failure. Any capturing parentheses that are set during the subroutine
2401 ph10 464 call revert to their previous values afterwards.
2402 nigel 93 .P
2403     When a subpattern is used as a subroutine, processing options such as
2404     case-independence are fixed when the subpattern is defined. They cannot be
2405     changed for different calls. For example, consider this pattern:
2406     .sp
2407 ph10 166 (abc)(?i:(?-1))
2408 nigel 93 .sp
2409     It matches "abcabc". It does not match "abcABC" because the change of
2410     processing option does not affect the called subpattern.
2411 nigel 75 .
2412     .
2413 ph10 333 .\" HTML <a name="onigurumasubroutines"></a>
2415     .rs
2416     .sp
2417 ph10 345 For compatibility with Oniguruma, the non-Perl syntax \eg followed by a name or
2418     a number enclosed either in angle brackets or single quotes, is an alternative
2419     syntax for referencing a subpattern as a subroutine, possibly recursively. Here
2420 ph10 333 are two of the examples used above, rewritten using this syntax:
2421     .sp
2422     (?<pn> \e( ( (?>[^()]+) | \eg<pn> )* \e) )
2423     (sens|respons)e and \eg'1'ibility
2424     .sp
2425 ph10 345 PCRE supports an extension to Oniguruma: if a number is preceded by a
2426 ph10 333 plus or a minus sign it is taken as a relative reference. For example:
2427     .sp
2428     (abc)(?i:\eg<-1>)
2429     .sp
2430 ph10 345 Note that \eg{...} (Perl syntax) and \eg<...> (Oniguruma syntax) are \fInot\fP
2431 ph10 333 synonymous. The former is a back reference; the latter is a subroutine call.
2432     .
2433     .
2434 nigel 63 .SH CALLOUTS
2435     .rs
2436     .sp
2437     Perl has a feature whereby using the sequence (?{...}) causes arbitrary Perl
2438     code to be obeyed in the middle of matching a regular expression. This makes it
2439     possible, amongst other things, to extract different substrings that match the
2440     same pair of parentheses when there is a repetition.
2441 nigel 75 .P
2442 nigel 63 PCRE provides a similar feature, but of course it cannot obey arbitrary Perl
2443     code. The feature is called "callout". The caller of PCRE provides an external
2444 nigel 75 function by putting its entry point in the global variable \fIpcre_callout\fP.
2445 nigel 63 By default, this variable contains NULL, which disables all calling out.
2446 nigel 75 .P
2447 nigel 63 Within a regular expression, (?C) indicates the points at which the external
2448     function is to be called. If you want to identify different callout points, you
2449     can put a number less than 256 after the letter C. The default value is zero.
2450     For example, this pattern has two callout points:
2451 nigel 75 .sp
2452 ph10 155 (?C1)abc(?C2)def
2453 nigel 75 .sp
2454     If the PCRE_AUTO_CALLOUT flag is passed to \fBpcre_compile()\fP, callouts are
2455     automatically installed before each item in the pattern. They are all numbered
2456     255.
2457     .P
2458     During matching, when PCRE reaches a callout point (and \fIpcre_callout\fP is
2459 nigel 63 set), the external function is called. It is provided with the number of the
2460 nigel 75 callout, the position in the pattern, and, optionally, one item of data
2461     originally supplied by the caller of \fBpcre_exec()\fP. The callout function
2462     may cause matching to proceed, to backtrack, or to fail altogether. A complete
2463     description of the interface to the callout function is given in the
2464 nigel 63 .\" HREF
2465 nigel 75 \fBpcrecallout\fP
2466 nigel 63 .\"
2467     documentation.
2468 nigel 93 .
2469     .
2470 ph10 510 .\" HTML <a name="backtrackcontrol"></a>
2472 ph10 210 .rs
2473     .sp
2474 ph10 211 Perl 5.10 introduced a number of "Special Backtracking Control Verbs", which
2475 ph10 210 are described in the Perl documentation as "experimental and subject to change
2476 ph10 211 or removal in a future version of Perl". It goes on to say: "Their usage in
2477     production code should be noted to avoid problems during upgrades." The same
2478 ph10 210 remarks apply to the PCRE features described in this section.
2479     .P
2480 ph10 341 Since these verbs are specifically related to backtracking, most of them can be
2481     used only when the pattern is to be matched using \fBpcre_exec()\fP, which uses
2482 ph10 345 a backtracking algorithm. With the exception of (*FAIL), which behaves like a
2483 ph10 341 failing negative assertion, they cause an error if encountered by
2484 ph10 210 \fBpcre_dfa_exec()\fP.
2485     .P
2486 ph10 469 If any of these verbs are used in an assertion or subroutine subpattern
2487 ph10 464 (including recursive subpatterns), their effect is confined to that subpattern;
2488     it does not extend to the surrounding pattern. Note that such subpatterns are
2489     processed as anchored at the point where they are tested.
2490 ph10 445 .P
2491 ph10 211 The new verbs make use of what was previously invalid syntax: an opening
2492 ph10 510 parenthesis followed by an asterisk. They are generally of the form
2493 ph10 512 (*VERB) or (*VERB:NAME). Some may take either form, with differing behaviour,
2494 ph10 510 depending on whether or not an argument is present. An name is a sequence of
2495     letters, digits, and underscores. If the name is empty, that is, if the closing
2496     parenthesis immediately follows the colon, the effect is as if the colon were
2497     not there. Any number of these verbs may occur in a pattern.
2498     .P
2499 ph10 512 PCRE contains some optimizations that are used to speed up matching by running
2500     some checks at the start of each match attempt. For example, it may know the
2501     minimum length of matching subject, or that a particular character must be
2502     present. When one of these optimizations suppresses the running of a match, any
2503     included backtracking verbs will not, of course, be processed. You can suppress
2504     the start-of-match optimizations by setting the PCRE_NO_START_OPTIMIZE option
2505 ph10 510 when calling \fBpcre_exec()\fP.
2506 ph10 210 .
2507 ph10 510 .
2508 ph10 210 .SS "Verbs that act immediately"
2509     .rs
2510     .sp
2511 ph10 512 The following verbs act as soon as they are encountered. They may not be
2512 ph10 510 followed by a name.
2513 ph10 210 .sp
2514     (*ACCEPT)
2515     .sp
2516     This verb causes the match to end successfully, skipping the remainder of the
2517 ph10 211 pattern. When inside a recursion, only the innermost pattern is ended
2518 ph10 456 immediately. If (*ACCEPT) is inside capturing parentheses, the data so far is
2519     captured. (This feature was added to PCRE at release 8.00.) For example:
2520 ph10 210 .sp
2521 ph10 447 A((?:A|B(*ACCEPT)|C)D)
2522 ph10 210 .sp
2523 ph10 461 This matches "AB", "AAD", or "ACD"; when it matches "AB", "B" is captured by
2524 ph10 447 the outer parentheses.
2525 ph10 210 .sp
2526     (*FAIL) or (*F)
2527     .sp
2528 ph10 211 This verb causes the match to fail, forcing backtracking to occur. It is
2529 ph10 210 equivalent to (?!) but easier to read. The Perl documentation notes that it is
2530     probably useful only when combined with (?{}) or (??{}). Those are, of course,
2531     Perl features that are not present in PCRE. The nearest equivalent is the
2532     callout feature, as for example in this pattern:
2533     .sp
2534     a+(?C)(*FAIL)
2535     .sp
2536 ph10 211 A match with the string "aaaa" always fails, but the callout is taken before
2537     each backtrack happens (in this example, 10 times).
2538 ph10 210 .
2539 ph10 510 .
2540     .SS "Recording which path was taken"
2541     .rs
2542     .sp
2543 ph10 512 There is one verb whose main purpose is to track how a match was arrived at,
2544     though it also has a secondary use in conjunction with advancing the match
2545 ph10 510 starting point (see (*SKIP) below).
2546     .sp
2547     (*MARK:NAME) or (*:NAME)
2548     .sp
2549     A name is always required with this verb. There may be as many instances of
2550     (*MARK) as you like in a pattern, and their names do not have to be unique.
2551     .P
2552     When a match succeeds, the name of the last-encountered (*MARK) is passed back
2553     to the caller via the \fIpcre_extra\fP data structure, as described in the
2554     .\" HTML <a href="pcreapi.html#extradata">
2555     .\" </a>
2556     section on \fIpcre_extra\fP
2557     .\"
2558 ph10 512 in the
2559 ph10 510 .\" HREF
2560     \fBpcreapi\fP
2561     .\"
2562     documentation. No data is returned for a partial match. Here is an example of
2563     \fBpcretest\fP output, where the /K modifier requests the retrieval and
2564     outputting of (*MARK) data:
2565     .sp
2566     /X(*MARK:A)Y|X(*MARK:B)Z/K
2567     XY
2568     0: XY
2569     MK: A
2570     XZ
2571     0: XZ
2572     MK: B
2573     .sp
2574 ph10 512 The (*MARK) name is tagged with "MK:" in this output, and in this example it
2575     indicates which of the two alternatives matched. This is a more efficient way
2576 ph10 510 of obtaining this information than putting each alternative in its own
2577     capturing parentheses.
2578     .P
2579     A name may also be returned after a failed match if the final path through the
2580     pattern involves (*MARK). However, unless (*MARK) used in conjunction with
2581     (*COMMIT), this is unlikely to happen for an unanchored pattern because, as the
2582     starting point for matching is advanced, the final check is often with an empty
2583     string, causing a failure before (*MARK) is reached. For example:
2584     .sp
2585     /X(*MARK:A)Y|X(*MARK:B)Z/K
2586     XP
2587     No match
2588     .sp
2589     There are three potential starting points for this match (starting with X,
2590 ph10 512 starting with P, and with an empty string). If the pattern is anchored, the
2591 ph10 510 result is different:
2592     .sp
2593     /^X(*MARK:A)Y|^X(*MARK:B)Z/K
2594     XP
2595     No match, mark = B
2596     .sp
2597 ph10 512 PCRE's start-of-match optimizations can also interfere with this. For example,
2598     if, as a result of a call to \fBpcre_study()\fP, it knows the minimum
2599 ph10 510 subject length for a match, a shorter subject will not be scanned at all.
2600     .P
2601 ph10 512 Note that similar anomalies (though different in detail) exist in Perl, no
2602     doubt for the same reasons. The use of (*MARK) data after a failed match of an
2603 ph10 510 unanchored pattern is not recommended, unless (*COMMIT) is involved.
2604     .
2605     .
2606 ph10 210 .SS "Verbs that act after backtracking"
2607     .rs
2608     .sp
2609 ph10 211 The following verbs do nothing when they are encountered. Matching continues
2610 ph10 510 with what follows, but if there is no subsequent match, causing a backtrack to
2611     the verb, a failure is forced. That is, backtracking cannot pass to the left of
2612     the verb. However, when one of these verbs appears inside an atomic group, its
2613     effect is confined to that group, because once the group has been matched,
2614     there is never any backtracking into it. In this situation, backtracking can
2615 ph10 512 "jump back" to the left of the entire atomic group. (Remember also, as stated
2616 ph10 510 above, that this localization also applies in subroutine calls and assertions.)
2617     .P
2618     These verbs differ in exactly what kind of failure occurs when backtracking
2619     reaches them.
2620 ph10 210 .sp
2621     (*COMMIT)
2622     .sp
2623 ph10 510 This verb, which may not be followed by a name, causes the whole match to fail
2624     outright if the rest of the pattern does not match. Even if the pattern is
2625     unanchored, no further attempts to find a match by advancing the starting point
2626     take place. Once (*COMMIT) has been passed, \fBpcre_exec()\fP is committed to
2627     finding a match at the current starting point, or not at all. For example:
2628 ph10 210 .sp
2629     a+(*COMMIT)b
2630     .sp
2631 ph10 211 This matches "xxaab" but not "aacaab". It can be thought of as a kind of
2632 ph10 512 dynamic anchor, or "I've started, so I must finish." The name of the most
2633     recently passed (*MARK) in the path is passed back when (*COMMIT) forces a
2634 ph10 510 match failure.
2635     .P
2636 ph10 512 Note that (*COMMIT) at the start of a pattern is not the same as an anchor,
2637     unless PCRE's start-of-match optimizations are turned off, as shown in this
2638 ph10 510 \fBpcretest\fP example:
2639 ph10 210 .sp
2640 ph10 510 /(*COMMIT)abc/
2641     xyzabc
2642     0: abc
2643     xyzabc\eY
2644     No match
2645 ph10 210 .sp
2646 ph10 512 PCRE knows that any match must start with "a", so the optimization skips along
2647 ph10 510 the subject to "a" before running the first match attempt, which succeeds. When
2648     the optimization is disabled by the \eY escape in the second subject, the match
2649 ph10 512 starts at "x" and so the (*COMMIT) causes it to fail without trying any other
2650 ph10 510 starting points.
2651 ph10 210 .sp
2652 ph10 510 (*PRUNE) or (*PRUNE:NAME)
2653     .sp
2654 ph10 512 This verb causes the match to fail at the current starting position in the
2655 ph10 510 subject if the rest of the pattern does not match. If the pattern is
2656     unanchored, the normal "bumpalong" advance to the next starting character then
2657     happens. Backtracking can occur as usual to the left of (*PRUNE), before it is
2658     reached, or when matching to the right of (*PRUNE), but if there is no match to
2659     the right, backtracking cannot cross (*PRUNE). In simple cases, the use of
2660     (*PRUNE) is just an alternative to an atomic group or possessive quantifier,
2661     but there are some uses of (*PRUNE) that cannot be expressed in any other way.
2662     The behaviour of (*PRUNE:NAME) is the same as (*MARK:NAME)(*PRUNE) when the
2663     match fails completely; the name is passed back if this is the final attempt.
2664     (*PRUNE:NAME) does not pass back a name if the match succeeds. In an anchored
2665     pattern (*PRUNE) has the same effect as (*COMMIT).
2666     .sp
2667 ph10 210 (*SKIP)
2668     .sp
2669 ph10 510 This verb, when given without a name, is like (*PRUNE), except that if the
2670     pattern is unanchored, the "bumpalong" advance is not to the next character,
2671     but to the position in the subject where (*SKIP) was encountered. (*SKIP)
2672     signifies that whatever text was matched leading up to it cannot be part of a
2673     successful match. Consider:
2674 ph10 210 .sp
2675     a+(*SKIP)b
2676     .sp
2677 ph10 211 If the subject is "aaaac...", after the first match attempt fails (starting at
2678 ph10 210 the first character in the string), the starting point skips on to start the
2679 ph10 211 next attempt at "c". Note that a possessive quantifer does not have the same
2680 ph10 456 effect as this example; although it would suppress backtracking during the
2681 ph10 210 first match attempt, the second attempt would start at the second character
2682     instead of skipping on to "c".
2683     .sp
2684 ph10 510 (*SKIP:NAME)
2685 ph10 211 .sp
2686 ph10 512 When (*SKIP) has an associated name, its behaviour is modified. If the
2687     following pattern fails to match, the previous path through the pattern is
2688     searched for the most recent (*MARK) that has the same name. If one is found,
2689     the "bumpalong" advance is to the subject position that corresponds to that
2690     (*MARK) instead of to where (*SKIP) was encountered. If no (*MARK) with a
2691     matching name is found, normal "bumpalong" of one character happens (the
2692 ph10 510 (*SKIP) is ignored).
2693     .sp
2694     (*THEN) or (*THEN:NAME)
2695     .sp
2696 ph10 550 This verb causes a skip to the next alternation in the innermost enclosing
2697     group if the rest of the pattern does not match. That is, it cancels pending
2698     backtracking, but only within the current alternation. Its name comes from the
2699     observation that it can be used for a pattern-based if-then-else block:
2700 ph10 210 .sp
2701     ( COND1 (*THEN) FOO | COND2 (*THEN) BAR | COND3 (*THEN) BAZ ) ...
2702     .sp
2703 ph10 211 If the COND1 pattern matches, FOO is tried (and possibly further items after
2704 ph10 210 the end of the group if FOO succeeds); on failure the matcher skips to the
2705 ph10 510 second alternative and tries COND2, without backtracking into COND1. The
2706     behaviour of (*THEN:NAME) is exactly the same as (*MARK:NAME)(*THEN) if the
2707     overall match fails. If (*THEN) is not directly inside an alternation, it acts
2708     like (*PRUNE).
2709 ph10 210 .
2710 ph10 551 .P
2711     The above verbs provide four different "strengths" of control when subsequent
2712     matching fails. (*THEN) is the weakest, carrying on the match at the next
2713     alternation. (*PRUNE) comes next, failing the match at the current starting
2714     position, but allowing an advance to the next character (for an unanchored
2715     pattern). (*SKIP) is similar, except that the advance may be more than one
2716     character. (*COMMIT) is the strongest, causing the entire match to fail.
2717     .P
2718     If more than one is present in a pattern, the "stongest" one wins. For example,
2719     consider this pattern, where A, B, etc. are complex pattern fragments:
2720     .sp
2721     (A(*COMMIT)B(*THEN)C|D)
2722     .sp
2723     Once A has matched, PCRE is committed to this match, at the current starting
2724     position. If subsequently B matches, but C does not, the normal (*THEN) action
2725     of trying the next alternation (that is, D) does not happen because (*COMMIT)
2726     overrides.
2727 ph10 210 .
2728 ph10 551 .
2729 nigel 93 .SH "SEE ALSO"
2730     .rs
2731     .sp
2732 ph10 461 \fBpcreapi\fP(3), \fBpcrecallout\fP(3), \fBpcrematching\fP(3),
2733 ph10 456 \fBpcresyntax\fP(3), \fBpcre\fP(3).
2734 ph10 99 .
2735     .
2736     .SH AUTHOR
2737     .rs
2738     .sp
2739     .nf
2740     Philip Hazel
2741     University Computing Service
2742     Cambridge CB2 3QH, England.
2743     .fi
2744     .
2745     .
2746     .SH REVISION
2747     .rs
2748     .sp
2749     .nf
2750 ph10 575 Last updated: 21 November 2010
2751 ph10 488 Copyright (c) 1997-2010 University of Cambridge.
2752 ph10 99 .fi


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

ViewVC Help
Powered by ViewVC 1.1.12