ViewVC logotype

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

Parent Directory Parent Directory | Revision Log Revision Log

Revision 99 - (hide annotations) (download)
Tue Mar 6 12:27:42 2007 UTC (8 years, 1 month ago) by ph10
File size: 74628 byte(s)
1. Move the comment about version numbers from pcre.h.in to configure.ac 
because that's where they are now set.
2. Update all the man pages to remove the use of .br and .in because this
causes trouble for some HTML converters. Also standardised the final sections 
giving author information and revision date.
3. Update the maintain/132html man page converter to handle .nf/.fi and to barf 
at .br/.in.

1 nigel 79 .TH PCREPATTERN 3
2 nigel 63 .SH NAME
3     PCRE - Perl-compatible regular expressions
5 nigel 63 .rs
6     .sp
7     The syntax and semantics of the regular expressions supported by PCRE are
8     described below. Regular expressions are also described in the Perl
9 nigel 75 documentation and in a number of books, some of which have copious examples.
10     Jeffrey Friedl's "Mastering Regular Expressions", published by O'Reilly, covers
11     regular expressions in great detail. This description of PCRE's regular
12     expressions is intended as reference material.
13     .P
14     The original operation of PCRE was on strings of one-byte characters. However,
15     there is now also support for UTF-8 character strings. To use this, you must
16     build PCRE to include UTF-8 support, and then call \fBpcre_compile()\fP with
17     the PCRE_UTF8 option. How this affects pattern matching is mentioned in several
18     places below. There is also a summary of UTF-8 features in the
19 nigel 63 .\" HTML <a href="pcre.html#utf8support">
20     .\" </a>
21     section on UTF-8 support
22     .\"
23     in the main
24     .\" HREF
25 nigel 75 \fBpcre\fP
26 nigel 63 .\"
27     page.
28 nigel 75 .P
29 nigel 77 The remainder of this document discusses the patterns that are supported by
30     PCRE when its main matching function, \fBpcre_exec()\fP, is used.
31     From release 6.0, PCRE offers a second matching function,
32     \fBpcre_dfa_exec()\fP, which matches using a different algorithm that is not
33     Perl-compatible. The advantages and disadvantages of the alternative function,
34     and how it differs from the normal function, are discussed in the
35     .\" HREF
36     \fBpcrematching\fP
37     .\"
38     page.
39 nigel 93 .
40     .
42     .rs
43     .sp
44 nigel 63 A regular expression is a pattern that is matched against a subject string from
45     left to right. Most characters stand for themselves in a pattern, and match the
46     corresponding characters in the subject. As a trivial example, the pattern
47 nigel 75 .sp
48 nigel 63 The quick brown fox
49 nigel 75 .sp
50 nigel 77 matches a portion of a subject string that is identical to itself. When
51     caseless matching is specified (the PCRE_CASELESS option), letters are matched
52     independently of case. In UTF-8 mode, PCRE always understands the concept of
53     case for characters whose values are less than 128, so caseless matching is
54     always possible. For characters with higher values, the concept of case is
55     supported if PCRE is compiled with Unicode property support, but not otherwise.
56     If you want to use caseless matching for characters 128 and above, you must
57     ensure that PCRE is compiled with Unicode property support as well as with
58     UTF-8 support.
59     .P
60     The power of regular expressions comes from the ability to include alternatives
61     and repetitions in the pattern. These are encoded in the pattern by the use of
62 nigel 75 \fImetacharacters\fP, which do not stand for themselves but instead are
63 nigel 63 interpreted in some special way.
64 nigel 75 .P
65     There are two different sets of metacharacters: those that are recognized
66 nigel 63 anywhere in the pattern except within square brackets, and those that are
67 nigel 93 recognized within square brackets. Outside square brackets, the metacharacters
68     are as follows:
69 nigel 75 .sp
70     \e general escape character with several uses
71 nigel 63 ^ assert start of string (or line, in multiline mode)
72     $ assert end of string (or line, in multiline mode)
73     . match any character except newline (by default)
74     [ start character class definition
75     | start of alternative branch
76     ( start subpattern
77     ) end subpattern
78     ? extends the meaning of (
79     also 0 or 1 quantifier
80     also quantifier minimizer
81     * 0 or more quantifier
82     + 1 or more quantifier
83     also "possessive quantifier"
84     { start min/max quantifier
85 nigel 75 .sp
86 nigel 63 Part of a pattern that is in square brackets is called a "character class". In
87 nigel 75 a character class the only metacharacters are:
88     .sp
89     \e general escape character
90 nigel 63 ^ negate the class, but only if the first character
91     - indicates character range
92 nigel 75 .\" JOIN
93 nigel 63 [ POSIX character class (only if followed by POSIX
94     syntax)
95     ] terminates the character class
96 nigel 75 .sp
97     The following sections describe the use of each of the metacharacters.
98     .
99 nigel 93 .
100 nigel 63 .SH BACKSLASH
101     .rs
102     .sp
103     The backslash character has several uses. Firstly, if it is followed by a
104 nigel 91 non-alphanumeric character, it takes away any special meaning that character
105     may have. This use of backslash as an escape character applies both inside and
106 nigel 63 outside character classes.
107 nigel 75 .P
108     For example, if you want to match a * character, you write \e* in the pattern.
109 nigel 63 This escaping action applies whether or not the following character would
110 nigel 75 otherwise be interpreted as a metacharacter, so it is always safe to precede a
111     non-alphanumeric with backslash to specify that it stands for itself. In
112     particular, if you want to match a backslash, you write \e\e.
113     .P
114 nigel 63 If a pattern is compiled with the PCRE_EXTENDED option, whitespace in the
115     pattern (other than in a character class) and characters between a # outside
116 nigel 91 a character class and the next newline are ignored. An escaping backslash can
117     be used to include a whitespace or # character as part of the pattern.
118 nigel 75 .P
119 nigel 63 If you want to remove the special meaning from a sequence of characters, you
120 nigel 75 can do so by putting them between \eQ and \eE. This is different from Perl in
121     that $ and @ are handled as literals in \eQ...\eE sequences in PCRE, whereas in
122 nigel 63 Perl, $ and @ cause variable interpolation. Note the following examples:
123 nigel 75 .sp
124 nigel 63 Pattern PCRE matches Perl matches
125 nigel 75 .sp
126     .\" JOIN
127     \eQabc$xyz\eE abc$xyz abc followed by the
128 nigel 63 contents of $xyz
129 nigel 75 \eQabc\e$xyz\eE abc\e$xyz abc\e$xyz
130     \eQabc\eE\e$\eQxyz\eE abc$xyz abc$xyz
131     .sp
132     The \eQ...\eE sequence is recognized both inside and outside character classes.
133     .
134     .
135     .\" HTML <a name="digitsafterbackslash"></a>
136     .SS "Non-printing characters"
137     .rs
138     .sp
139 nigel 63 A second use of backslash provides a way of encoding non-printing characters
140     in patterns in a visible manner. There is no restriction on the appearance of
141     non-printing characters, apart from the binary zero that terminates a pattern,
142     but when a pattern is being prepared by text editing, it is usually easier to
143     use one of the following escape sequences than the binary character it
144     represents:
145 nigel 75 .sp
146     \ea alarm, that is, the BEL character (hex 07)
147     \ecx "control-x", where x is any character
148     \ee escape (hex 1B)
149     \ef formfeed (hex 0C)
150     \en newline (hex 0A)
151     \er carriage return (hex 0D)
152     \et tab (hex 09)
153     \eddd character with octal code ddd, or backreference
154     \exhh character with hex code hh
155 nigel 87 \ex{hhh..} character with hex code hhh..
156 nigel 75 .sp
157     The precise effect of \ecx is as follows: if x is a lower case letter, it
158 nigel 63 is converted to upper case. Then bit 6 of the character (hex 40) is inverted.
159 nigel 75 Thus \ecz becomes hex 1A, but \ec{ becomes hex 3B, while \ec; becomes hex
160 nigel 63 7B.
161 nigel 75 .P
162     After \ex, from zero to two hexadecimal digits are read (letters can be in
163 nigel 87 upper or lower case). Any number of hexadecimal digits may appear between \ex{
164     and }, but the value of the character code must be less than 256 in non-UTF-8
165     mode, and less than 2**31 in UTF-8 mode (that is, the maximum hexadecimal value
166     is 7FFFFFFF). If characters other than hexadecimal digits appear between \ex{
167     and }, or if there is no terminating }, this form of escape is not recognized.
168     Instead, the initial \ex will be interpreted as a basic hexadecimal escape,
169     with no following digits, giving a character whose value is zero.
170 nigel 75 .P
171 nigel 63 Characters whose value is less than 256 can be defined by either of the two
172 nigel 87 syntaxes for \ex. There is no difference in the way they are handled. For
173     example, \exdc is exactly the same as \ex{dc}.
174 nigel 75 .P
175 nigel 91 After \e0 up to two further octal digits are read. If there are fewer than two
176     digits, just those that are present are used. Thus the sequence \e0\ex\e07
177     specifies two binary zeros followed by a BEL character (code value 7). Make
178     sure you supply two digits after the initial zero if the pattern character that
179     follows is itself an octal digit.
180 nigel 75 .P
181 nigel 63 The handling of a backslash followed by a digit other than 0 is complicated.
182     Outside a character class, PCRE reads it and any following digits as a decimal
183     number. If the number is less than 10, or if there have been at least that many
184     previous capturing left parentheses in the expression, the entire sequence is
185 nigel 75 taken as a \fIback reference\fP. A description of how this works is given
186     .\" HTML <a href="#backreferences">
187     .\" </a>
188     later,
189     .\"
190     following the discussion of
191     .\" HTML <a href="#subpattern">
192     .\" </a>
193     parenthesized subpatterns.
194     .\"
195     .P
196 nigel 63 Inside a character class, or if the decimal number is greater than 9 and there
197     have not been that many capturing subpatterns, PCRE re-reads up to three octal
198 nigel 93 digits following the backslash, and uses them to generate a data character. Any
199 nigel 91 subsequent digits stand for themselves. In non-UTF-8 mode, the value of a
200     character specified in octal must be less than \e400. In UTF-8 mode, values up
201     to \e777 are permitted. For example:
202 nigel 75 .sp
203     \e040 is another way of writing a space
204     .\" JOIN
205     \e40 is the same, provided there are fewer than 40
206 nigel 63 previous capturing subpatterns
207 nigel 75 \e7 is always a back reference
208     .\" JOIN
209     \e11 might be a back reference, or another way of
210 nigel 63 writing a tab
211 nigel 75 \e011 is always a tab
212     \e0113 is a tab followed by the character "3"
213     .\" JOIN
214     \e113 might be a back reference, otherwise the
215 nigel 63 character with octal code 113
216 nigel 75 .\" JOIN
217     \e377 might be a back reference, otherwise
218 nigel 63 the byte consisting entirely of 1 bits
219 nigel 75 .\" JOIN
220     \e81 is either a back reference, or a binary zero
221 nigel 63 followed by the two characters "8" and "1"
222 nigel 75 .sp
223 nigel 63 Note that octal values of 100 or greater must not be introduced by a leading
224     zero, because no more than three octal digits are ever read.
225 nigel 75 .P
226 nigel 91 All the sequences that define a single character value can be used both inside
227     and outside character classes. In addition, inside a character class, the
228     sequence \eb is interpreted as the backspace character (hex 08), and the
229 nigel 93 sequences \eR and \eX are interpreted as the characters "R" and "X",
230     respectively. Outside a character class, these sequences have different
231     meanings
232 nigel 75 .\" HTML <a href="#uniextseq">
233     .\" </a>
234     (see below).
235     .\"
236     .
237     .
238 nigel 93 .SS "Absolute and relative back references"
239     .rs
240     .sp
241     The sequence \eg followed by a positive or negative number, optionally enclosed
242     in braces, is an absolute or relative back reference. Back references are
243     discussed
244     .\" HTML <a href="#backreferences">
245     .\" </a>
246     later,
247     .\"
248     following the discussion of
249     .\" HTML <a href="#subpattern">
250     .\" </a>
251     parenthesized subpatterns.
252     .\"
253     .
254     .
255 nigel 75 .SS "Generic character types"
256     .rs
257     .sp
258 nigel 93 Another use of backslash is for specifying generic character types. The
259 nigel 75 following are always recognized:
260     .sp
261     \ed any decimal digit
262     \eD any character that is not a decimal digit
263     \es any whitespace character
264     \eS any character that is not a whitespace character
265     \ew any "word" character
266     \eW any "non-word" character
267     .sp
268 nigel 63 Each pair of escape sequences partitions the complete set of characters into
269     two disjoint sets. Any given character matches one, and only one, of each pair.
270 nigel 75 .P
271     These character type sequences can appear both inside and outside character
272     classes. They each match one character of the appropriate type. If the current
273     matching point is at the end of the subject string, all of them fail, since
274     there is no character to match.
275     .P
276     For compatibility with Perl, \es does not match the VT character (code 11).
277     This makes it different from the the POSIX "space" class. The \es characters
278 nigel 91 are HT (9), LF (10), FF (12), CR (13), and space (32). (If "use locale;" is
279     included in a Perl script, \es may match the VT character. In PCRE, it never
280     does.)
281 nigel 75 .P
282     A "word" character is an underscore or any character less than 256 that is a
283     letter or digit. The definition of letters and digits is controlled by PCRE's
284     low-valued character tables, and may vary if locale-specific matching is taking
285     place (see
286 nigel 63 .\" HTML <a href="pcreapi.html#localesupport">
287     .\" </a>
288     "Locale support"
289     .\"
290     in the
291     .\" HREF
292 nigel 75 \fBpcreapi\fP
293 nigel 63 .\"
294 nigel 75 page). For example, in the "fr_FR" (French) locale, some character codes
295     greater than 128 are used for accented letters, and these are matched by \ew.
296     .P
297     In UTF-8 mode, characters with values greater than 128 never match \ed, \es, or
298     \ew, and always match \eD, \eS, and \eW. This is true even when Unicode
299 nigel 87 character property support is available. The use of locales with Unicode is
300     discouraged.
301 nigel 75 .
302     .
303 nigel 93 .SS "Newline sequences"
304     .rs
305     .sp
306     Outside a character class, the escape sequence \eR matches any Unicode newline
307     sequence. This is an extension to Perl. In non-UTF-8 mode \eR is equivalent to
308     the following:
309     .sp
310     (?>\er\en|\en|\ex0b|\ef|\er|\ex85)
311     .sp
312     This is an example of an "atomic group", details of which are given
313     .\" HTML <a href="#atomicgroup">
314     .\" </a>
315     below.
316     .\"
317     This particular group matches either the two-character sequence CR followed by
318     LF, or one of the single characters LF (linefeed, U+000A), VT (vertical tab,
319     U+000B), FF (formfeed, U+000C), CR (carriage return, U+000D), or NEL (next
320     line, U+0085). The two-character sequence is treated as a single unit that
321     cannot be split.
322     .P
323     In UTF-8 mode, two additional characters whose codepoints are greater than 255
324     are added: LS (line separator, U+2028) and PS (paragraph separator, U+2029).
325     Unicode character property support is not needed for these characters to be
326     recognized.
327     .P
328     Inside a character class, \eR matches the letter "R".
329     .
330     .
331 nigel 75 .\" HTML <a name="uniextseq"></a>
332     .SS Unicode character properties
333     .rs
334     .sp
335     When PCRE is built with Unicode character property support, three additional
336 nigel 87 escape sequences to match character properties are available when UTF-8 mode
337 nigel 75 is selected. They are:
338     .sp
339 nigel 87 \ep{\fIxx\fP} a character with the \fIxx\fP property
340     \eP{\fIxx\fP} a character without the \fIxx\fP property
341     \eX an extended Unicode sequence
342 nigel 75 .sp
343 nigel 87 The property names represented by \fIxx\fP above are limited to the Unicode
344     script names, the general category properties, and "Any", which matches any
345     character (including newline). Other properties such as "InMusicalSymbols" are
346     not currently supported by PCRE. Note that \eP{Any} does not match any
347     characters, so always causes a match failure.
348 nigel 75 .P
349 nigel 87 Sets of Unicode characters are defined as belonging to certain scripts. A
350     character from one of these sets can be matched using a script name. For
351     example:
352 nigel 75 .sp
353 nigel 87 \ep{Greek}
354     \eP{Han}
355     .sp
356     Those that are not part of an identified script are lumped together as
357     "Common". The current list of scripts is:
358     .P
359     Arabic,
360     Armenian,
361 nigel 93 Balinese,
362 nigel 87 Bengali,
363     Bopomofo,
364     Braille,
365     Buginese,
366     Buhid,
367     Canadian_Aboriginal,
368     Cherokee,
369     Common,
370     Coptic,
371 nigel 93 Cuneiform,
372 nigel 87 Cypriot,
373     Cyrillic,
374     Deseret,
375     Devanagari,
376     Ethiopic,
377     Georgian,
378     Glagolitic,
379     Gothic,
380     Greek,
381     Gujarati,
382     Gurmukhi,
383     Han,
384     Hangul,
385     Hanunoo,
386     Hebrew,
387     Hiragana,
388     Inherited,
389     Kannada,
390     Katakana,
391     Kharoshthi,
392     Khmer,
393     Lao,
394     Latin,
395     Limbu,
396     Linear_B,
397     Malayalam,
398     Mongolian,
399     Myanmar,
400     New_Tai_Lue,
401 nigel 93 Nko,
402 nigel 87 Ogham,
403     Old_Italic,
404     Old_Persian,
405     Oriya,
406     Osmanya,
407 nigel 93 Phags_Pa,
408     Phoenician,
409 nigel 87 Runic,
410     Shavian,
411     Sinhala,
412     Syloti_Nagri,
413     Syriac,
414     Tagalog,
415     Tagbanwa,
416     Tai_Le,
417     Tamil,
418     Telugu,
419     Thaana,
420     Thai,
421     Tibetan,
422     Tifinagh,
423     Ugaritic,
424     Yi.
425     .P
426     Each character has exactly one general category property, specified by a
427     two-letter abbreviation. For compatibility with Perl, negation can be specified
428     by including a circumflex between the opening brace and the property name. For
429     example, \ep{^Lu} is the same as \eP{Lu}.
430     .P
431     If only one letter is specified with \ep or \eP, it includes all the general
432     category properties that start with that letter. In this case, in the absence
433     of negation, the curly brackets in the escape sequence are optional; these two
434     examples have the same effect:
435     .sp
436 nigel 75 \ep{L}
437     \epL
438     .sp
439 nigel 87 The following general category property codes are supported:
440 nigel 75 .sp
441     C Other
442     Cc Control
443     Cf Format
444     Cn Unassigned
445     Co Private use
446     Cs Surrogate
447     .sp
448     L Letter
449     Ll Lower case letter
450     Lm Modifier letter
451     Lo Other letter
452     Lt Title case letter
453     Lu Upper case letter
454     .sp
455     M Mark
456     Mc Spacing mark
457     Me Enclosing mark
458     Mn Non-spacing mark
459     .sp
460     N Number
461     Nd Decimal number
462     Nl Letter number
463     No Other number
464     .sp
465     P Punctuation
466     Pc Connector punctuation
467     Pd Dash punctuation
468     Pe Close punctuation
469     Pf Final punctuation
470     Pi Initial punctuation
471     Po Other punctuation
472     Ps Open punctuation
473     .sp
474     S Symbol
475     Sc Currency symbol
476     Sk Modifier symbol
477     Sm Mathematical symbol
478     So Other symbol
479     .sp
480     Z Separator
481     Zl Line separator
482     Zp Paragraph separator
483     Zs Space separator
484     .sp
485 nigel 87 The special property L& is also supported: it matches a character that has
486     the Lu, Ll, or Lt property, in other words, a letter that is not classified as
487     a modifier or "other".
488 nigel 75 .P
489 nigel 87 The long synonyms for these properties that Perl supports (such as \ep{Letter})
490 nigel 91 are not supported by PCRE, nor is it permitted to prefix any of these
491 nigel 87 properties with "Is".
492     .P
493     No character that is in the Unicode table has the Cn (unassigned) property.
494     Instead, this property is assumed for any code point that is not in the
495     Unicode table.
496     .P
497 nigel 75 Specifying caseless matching does not affect these escape sequences. For
498     example, \ep{Lu} always matches only upper case letters.
499     .P
500     The \eX escape matches any number of Unicode characters that form an extended
501     Unicode sequence. \eX is equivalent to
502     .sp
503     (?>\ePM\epM*)
504     .sp
505     That is, it matches a character without the "mark" property, followed by zero
506     or more characters with the "mark" property, and treats the sequence as an
507     atomic group
508     .\" HTML <a href="#atomicgroup">
509     .\" </a>
510     (see below).
511     .\"
512     Characters with the "mark" property are typically accents that affect the
513     preceding character.
514     .P
515     Matching characters by Unicode property is not fast, because PCRE has to search
516     a structure that contains data for over fifteen thousand characters. That is
517     why the traditional escape sequences such as \ed and \ew do not use Unicode
518     properties in PCRE.
519     .
520     .
521     .\" HTML <a name="smallassertions"></a>
522     .SS "Simple assertions"
523     .rs
524     .sp
525 nigel 93 The final use of backslash is for certain simple assertions. An assertion
526 nigel 63 specifies a condition that has to be met at a particular point in a match,
527     without consuming any characters from the subject string. The use of
528 nigel 75 subpatterns for more complicated assertions is described
529     .\" HTML <a href="#bigassertions">
530     .\" </a>
531     below.
532     .\"
533 nigel 91 The backslashed assertions are:
534 nigel 75 .sp
535     \eb matches at a word boundary
536     \eB matches when not at a word boundary
537 nigel 93 \eA matches at the start of the subject
538     \eZ matches at the end of the subject
539     also matches before a newline at the end of the subject
540     \ez matches only at the end of the subject
541     \eG matches at the first matching position in the subject
542 nigel 75 .sp
543     These assertions may not appear in character classes (but note that \eb has a
544 nigel 63 different meaning, namely the backspace character, inside a character class).
545 nigel 75 .P
546 nigel 63 A word boundary is a position in the subject string where the current character
547 nigel 75 and the previous character do not both match \ew or \eW (i.e. one matches
548     \ew and the other matches \eW), or the start or end of the string if the
549     first or last character matches \ew, respectively.
550     .P
551     The \eA, \eZ, and \ez assertions differ from the traditional circumflex and
552     dollar (described in the next section) in that they only ever match at the very
553     start and end of the subject string, whatever options are set. Thus, they are
554     independent of multiline mode. These three assertions are not affected by the
555     PCRE_NOTBOL or PCRE_NOTEOL options, which affect only the behaviour of the
556     circumflex and dollar metacharacters. However, if the \fIstartoffset\fP
557     argument of \fBpcre_exec()\fP is non-zero, indicating that matching is to start
558     at a point other than the beginning of the subject, \eA can never match. The
559 nigel 91 difference between \eZ and \ez is that \eZ matches before a newline at the end
560     of the string as well as at the very end, whereas \ez matches only at the end.
561 nigel 75 .P
562     The \eG assertion is true only when the current matching position is at the
563     start point of the match, as specified by the \fIstartoffset\fP argument of
564     \fBpcre_exec()\fP. It differs from \eA when the value of \fIstartoffset\fP is
565     non-zero. By calling \fBpcre_exec()\fP multiple times with appropriate
566 nigel 63 arguments, you can mimic Perl's /g option, and it is in this kind of
567 nigel 75 implementation where \eG can be useful.
568     .P
569     Note, however, that PCRE's interpretation of \eG, as the start of the current
570 nigel 63 match, is subtly different from Perl's, which defines it as the end of the
571     previous match. In Perl, these can be different when the previously matched
572     string was empty. Because PCRE does just one match at a time, it cannot
573     reproduce this behaviour.
574 nigel 75 .P
575     If all the alternatives of a pattern begin with \eG, the expression is anchored
576 nigel 63 to the starting match position, and the "anchored" flag is set in the compiled
577     regular expression.
578 nigel 75 .
579     .
581 nigel 63 .rs
582     .sp
583     Outside a character class, in the default matching mode, the circumflex
584 nigel 75 character is an assertion that is true only if the current matching point is
585     at the start of the subject string. If the \fIstartoffset\fP argument of
586     \fBpcre_exec()\fP is non-zero, circumflex can never match if the PCRE_MULTILINE
587 nigel 63 option is unset. Inside a character class, circumflex has an entirely different
588 nigel 75 meaning
589     .\" HTML <a href="#characterclass">
590     .\" </a>
591     (see below).
592     .\"
593     .P
594 nigel 63 Circumflex need not be the first character of the pattern if a number of
595     alternatives are involved, but it should be the first thing in each alternative
596     in which it appears if the pattern is ever to match that branch. If all
597     possible alternatives start with a circumflex, that is, if the pattern is
598     constrained to match only at the start of the subject, it is said to be an
599     "anchored" pattern. (There are also other constructs that can cause a pattern
600     to be anchored.)
601 nigel 75 .P
602     A dollar character is an assertion that is true only if the current matching
603 nigel 63 point is at the end of the subject string, or immediately before a newline
604 nigel 91 at the end of the string (by default). Dollar need not be the last character of
605     the pattern if a number of alternatives are involved, but it should be the last
606     item in any branch in which it appears. Dollar has no special meaning in a
607     character class.
608 nigel 75 .P
609 nigel 63 The meaning of dollar can be changed so that it matches only at the very end of
610     the string, by setting the PCRE_DOLLAR_ENDONLY option at compile time. This
611 nigel 75 does not affect the \eZ assertion.
612     .P
613 nigel 63 The meanings of the circumflex and dollar characters are changed if the
614 nigel 91 PCRE_MULTILINE option is set. When this is the case, a circumflex matches
615     immediately after internal newlines as well as at the start of the subject
616     string. It does not match after a newline that ends the string. A dollar
617     matches before any newlines in the string, as well as at the very end, when
618     PCRE_MULTILINE is set. When newline is specified as the two-character
619     sequence CRLF, isolated CR and LF characters do not indicate newlines.
620 nigel 75 .P
621 nigel 91 For example, the pattern /^abc$/ matches the subject string "def\enabc" (where
622     \en represents a newline) in multiline mode, but not otherwise. Consequently,
623     patterns that are anchored in single line mode because all branches start with
624     ^ are not anchored in multiline mode, and a match for circumflex is possible
625     when the \fIstartoffset\fP argument of \fBpcre_exec()\fP is non-zero. The
626     PCRE_DOLLAR_ENDONLY option is ignored if PCRE_MULTILINE is set.
627     .P
628 nigel 75 Note that the sequences \eA, \eZ, and \ez can be used to match the start and
629 nigel 63 end of the subject in both modes, and if all branches of a pattern start with
630 nigel 91 \eA it is always anchored, whether or not PCRE_MULTILINE is set.
631 nigel 75 .
632     .
634 nigel 63 .rs
635     .sp
636     Outside a character class, a dot in the pattern matches any one character in
637 nigel 91 the subject string except (by default) a character that signifies the end of a
638 nigel 93 line. In UTF-8 mode, the matched character may be more than one byte long.
639 nigel 91 .P
640 nigel 93 When a line ending is defined as a single character, dot never matches that
641     character; when the two-character sequence CRLF is used, dot does not match CR
642     if it is immediately followed by LF, but otherwise it matches all characters
643     (including isolated CRs and LFs). When any Unicode line endings are being
644     recognized, dot does not match CR or LF or any of the other line ending
645     characters.
646     .P
647 nigel 91 The behaviour of dot with regard to newlines can be changed. If the PCRE_DOTALL
648 nigel 93 option is set, a dot matches any one character, without exception. If the
649     two-character sequence CRLF is present in the subject string, it takes two dots
650     to match it.
651 nigel 91 .P
652     The handling of dot is entirely independent of the handling of circumflex and
653     dollar, the only relationship being that they both involve newlines. Dot has no
654     special meaning in a character class.
655 nigel 75 .
656     .
658 nigel 63 .rs
659     .sp
660 nigel 75 Outside a character class, the escape sequence \eC matches any one byte, both
661 nigel 93 in and out of UTF-8 mode. Unlike a dot, it always matches any line-ending
662     characters. The feature is provided in Perl in order to match individual bytes
663     in UTF-8 mode. Because it breaks up UTF-8 characters into individual bytes,
664     what remains in the string may be a malformed UTF-8 string. For this reason,
665     the \eC escape sequence is best avoided.
666 nigel 75 .P
667     PCRE does not allow \eC to appear in lookbehind assertions
668     .\" HTML <a href="#lookbehind">
669     .\" </a>
670     (described below),
671     .\"
672     because in UTF-8 mode this would make it impossible to calculate the length of
673     the lookbehind.
674     .
675     .
676     .\" HTML <a name="characterclass"></a>
678 nigel 63 .rs
679     .sp
680     An opening square bracket introduces a character class, terminated by a closing
681     square bracket. A closing square bracket on its own is not special. If a
682     closing square bracket is required as a member of the class, it should be the
683     first data character in the class (after an initial circumflex, if present) or
684     escaped with a backslash.
685 nigel 75 .P
686 nigel 63 A character class matches a single character in the subject. In UTF-8 mode, the
687     character may occupy more than one byte. A matched character must be in the set
688     of characters defined by the class, unless the first character in the class
689     definition is a circumflex, in which case the subject character must not be in
690     the set defined by the class. If a circumflex is actually required as a member
691     of the class, ensure it is not the first character, or escape it with a
692     backslash.
693 nigel 75 .P
694 nigel 63 For example, the character class [aeiou] matches any lower case vowel, while
695     [^aeiou] matches any character that is not a lower case vowel. Note that a
696 nigel 75 circumflex is just a convenient notation for specifying the characters that
697     are in the class by enumerating those that are not. A class that starts with a
698     circumflex is not an assertion: it still consumes a character from the subject
699     string, and therefore it fails if the current pointer is at the end of the
700     string.
701     .P
702 nigel 63 In UTF-8 mode, characters with values greater than 255 can be included in a
703 nigel 75 class as a literal string of bytes, or by using the \ex{ escaping mechanism.
704     .P
705 nigel 63 When caseless matching is set, any letters in a class represent both their
706     upper case and lower case versions, so for example, a caseless [aeiou] matches
707     "A" as well as "a", and a caseless [^aeiou] does not match "A", whereas a
708 nigel 77 caseful version would. In UTF-8 mode, PCRE always understands the concept of
709     case for characters whose values are less than 128, so caseless matching is
710     always possible. For characters with higher values, the concept of case is
711     supported if PCRE is compiled with Unicode property support, but not otherwise.
712     If you want to use caseless matching for characters 128 and above, you must
713     ensure that PCRE is compiled with Unicode property support as well as with
714     UTF-8 support.
715 nigel 75 .P
716 nigel 93 Characters that might indicate line breaks are never treated in any special way
717     when matching character classes, whatever line-ending sequence is in use, and
718     whatever setting of the PCRE_DOTALL and PCRE_MULTILINE options is used. A class
719     such as [^a] always matches one of these characters.
720 nigel 75 .P
721 nigel 63 The minus (hyphen) character can be used to specify a range of characters in a
722     character class. For example, [d-m] matches any letter between d and m,
723     inclusive. If a minus character is required in a class, it must be escaped with
724     a backslash or appear in a position where it cannot be interpreted as
725     indicating a range, typically as the first or last character in the class.
726 nigel 75 .P
727 nigel 63 It is not possible to have the literal character "]" as the end character of a
728     range. A pattern such as [W-]46] is interpreted as a class of two characters
729     ("W" and "-") followed by a literal string "46]", so it would match "W46]" or
730     "-46]". However, if the "]" is escaped with a backslash it is interpreted as
731 nigel 75 the end of range, so [W-\e]46] is interpreted as a class containing a range
732     followed by two other characters. The octal or hexadecimal representation of
733     "]" can also be used to end a range.
734     .P
735 nigel 63 Ranges operate in the collating sequence of character values. They can also be
736 nigel 75 used for characters specified numerically, for example [\e000-\e037]. In UTF-8
737 nigel 63 mode, ranges can include characters whose values are greater than 255, for
738 nigel 75 example [\ex{100}-\ex{2ff}].
739     .P
740 nigel 63 If a range that includes letters is used when caseless matching is set, it
741     matches the letters in either case. For example, [W-c] is equivalent to
742 nigel 75 [][\e\e^_`wxyzabc], matched caselessly, and in non-UTF-8 mode, if character
743     tables for the "fr_FR" locale are in use, [\exc8-\excb] matches accented E
744     characters in both cases. In UTF-8 mode, PCRE supports the concept of case for
745     characters with values greater than 128 only when it is compiled with Unicode
746     property support.
747     .P
748     The character types \ed, \eD, \ep, \eP, \es, \eS, \ew, and \eW may also appear
749     in a character class, and add the characters that they match to the class. For
750     example, [\edABCDEF] matches any hexadecimal digit. A circumflex can
751 nigel 63 conveniently be used with the upper case character types to specify a more
752     restricted set of characters than the matching lower case type. For example,
753 nigel 75 the class [^\eW_] matches any letter or digit, but not underscore.
754     .P
755     The only metacharacters that are recognized in character classes are backslash,
756     hyphen (only where it can be interpreted as specifying a range), circumflex
757     (only at the start), opening square bracket (only when it can be interpreted as
758     introducing a POSIX class name - see the next section), and the terminating
759     closing square bracket. However, escaping other non-alphanumeric characters
760     does no harm.
761     .
762     .
764 nigel 63 .rs
765     .sp
766 nigel 75 Perl supports the POSIX notation for character classes. This uses names
767 nigel 63 enclosed by [: and :] within the enclosing square brackets. PCRE also supports
768     this notation. For example,
769 nigel 75 .sp
770 nigel 63 [01[:alpha:]%]
771 nigel 75 .sp
772 nigel 63 matches "0", "1", any alphabetic character, or "%". The supported class names
773     are
774 nigel 75 .sp
775 nigel 63 alnum letters and digits
776     alpha letters
777     ascii character codes 0 - 127
778     blank space or tab only
779     cntrl control characters
780 nigel 75 digit decimal digits (same as \ed)
781 nigel 63 graph printing characters, excluding space
782     lower lower case letters
783     print printing characters, including space
784     punct printing characters, excluding letters and digits
785 nigel 75 space white space (not quite the same as \es)
786 nigel 63 upper upper case letters
787 nigel 75 word "word" characters (same as \ew)
788 nigel 63 xdigit hexadecimal digits
789 nigel 75 .sp
790 nigel 63 The "space" characters are HT (9), LF (10), VT (11), FF (12), CR (13), and
791     space (32). Notice that this list includes the VT character (code 11). This
792 nigel 75 makes "space" different to \es, which does not include VT (for Perl
793 nigel 63 compatibility).
794 nigel 75 .P
795 nigel 63 The name "word" is a Perl extension, and "blank" is a GNU extension from Perl
796     5.8. Another Perl extension is negation, which is indicated by a ^ character
797     after the colon. For example,
798 nigel 75 .sp
799 nigel 63 [12[:^digit:]]
800 nigel 75 .sp
801 nigel 63 matches "1", "2", or any non-digit. PCRE (and Perl) also recognize the POSIX
802     syntax [.ch.] and [=ch=] where "ch" is a "collating element", but these are not
803     supported, and an error is given if they are encountered.
804 nigel 75 .P
805     In UTF-8 mode, characters with values greater than 128 do not match any of
806 nigel 63 the POSIX character classes.
807 nigel 75 .
808     .
809     .SH "VERTICAL BAR"
810 nigel 63 .rs
811     .sp
812     Vertical bar characters are used to separate alternative patterns. For example,
813     the pattern
814 nigel 75 .sp
815 nigel 63 gilbert|sullivan
816 nigel 75 .sp
817 nigel 63 matches either "gilbert" or "sullivan". Any number of alternatives may appear,
818 nigel 91 and an empty alternative is permitted (matching the empty string). The matching
819     process tries each alternative in turn, from left to right, and the first one
820     that succeeds is used. If the alternatives are within a subpattern
821 nigel 75 .\" HTML <a href="#subpattern">
822     .\" </a>
823     (defined below),
824     .\"
825     "succeeds" means matching the rest of the main pattern as well as the
826     alternative in the subpattern.
827     .
828     .
830 nigel 63 .rs
831     .sp
832     The settings of the PCRE_CASELESS, PCRE_MULTILINE, PCRE_DOTALL, and
833     PCRE_EXTENDED options can be changed from within the pattern by a sequence of
834     Perl option letters enclosed between "(?" and ")". The option letters are
835 nigel 75 .sp
836 nigel 63 i for PCRE_CASELESS
837     m for PCRE_MULTILINE
838     s for PCRE_DOTALL
839     x for PCRE_EXTENDED
840 nigel 75 .sp
841 nigel 63 For example, (?im) sets caseless, multiline matching. It is also possible to
842     unset these options by preceding the letter with a hyphen, and a combined
843     setting and unsetting such as (?im-sx), which sets PCRE_CASELESS and
844     PCRE_MULTILINE while unsetting PCRE_DOTALL and PCRE_EXTENDED, is also
845     permitted. If a letter appears both before and after the hyphen, the option is
846     unset.
847 nigel 75 .P
848 nigel 63 When an option change occurs at top level (that is, not inside subpattern
849     parentheses), the change applies to the remainder of the pattern that follows.
850     If the change is placed right at the start of a pattern, PCRE extracts it into
851     the global options (and it will therefore show up in data extracted by the
852 nigel 75 \fBpcre_fullinfo()\fP function).
853     .P
854 nigel 93 An option change within a subpattern (see below for a description of
855     subpatterns) affects only that part of the current pattern that follows it, so
856 nigel 75 .sp
857 nigel 63 (a(?i)b)c
858 nigel 75 .sp
859 nigel 63 matches abc and aBc and no other strings (assuming PCRE_CASELESS is not used).
860     By this means, options can be made to have different settings in different
861     parts of the pattern. Any changes made in one alternative do carry on
862     into subsequent branches within the same subpattern. For example,
863 nigel 75 .sp
864 nigel 63 (a(?i)b|c)
865 nigel 75 .sp
866 nigel 63 matches "ab", "aB", "c", and "C", even though when matching "C" the first
867     branch is abandoned before the option setting. This is because the effects of
868     option settings happen at compile time. There would be some very weird
869     behaviour otherwise.
870 nigel 75 .P
871 nigel 91 The PCRE-specific options PCRE_DUPNAMES, PCRE_UNGREEDY, and PCRE_EXTRA can be
872     changed in the same way as the Perl-compatible options by using the characters
873     J, U and X respectively.
874 nigel 75 .
875     .
876     .\" HTML <a name="subpattern"></a>
877 nigel 63 .SH SUBPATTERNS
878     .rs
879     .sp
880     Subpatterns are delimited by parentheses (round brackets), which can be nested.
881 nigel 75 Turning part of a pattern into a subpattern does two things:
882     .sp
883 nigel 63 1. It localizes a set of alternatives. For example, the pattern
884 nigel 75 .sp
885 nigel 63 cat(aract|erpillar|)
886 nigel 75 .sp
887 nigel 63 matches one of the words "cat", "cataract", or "caterpillar". Without the
888 nigel 93 parentheses, it would match "cataract", "erpillar" or an empty string.
889 nigel 75 .sp
890     2. It sets up the subpattern as a capturing subpattern. This means that, when
891     the whole pattern matches, that portion of the subject string that matched the
892     subpattern is passed back to the caller via the \fIovector\fP argument of
893     \fBpcre_exec()\fP. Opening parentheses are counted from left to right (starting
894     from 1) to obtain numbers for the capturing subpatterns.
895     .P
896 nigel 63 For example, if the string "the red king" is matched against the pattern
897 nigel 75 .sp
898 nigel 63 the ((red|white) (king|queen))
899 nigel 75 .sp
900 nigel 63 the captured substrings are "red king", "red", and "king", and are numbered 1,
901     2, and 3, respectively.
902 nigel 75 .P
903 nigel 63 The fact that plain parentheses fulfil two functions is not always helpful.
904     There are often times when a grouping subpattern is required without a
905     capturing requirement. If an opening parenthesis is followed by a question mark
906     and a colon, the subpattern does not do any capturing, and is not counted when
907     computing the number of any subsequent capturing subpatterns. For example, if
908     the string "the white queen" is matched against the pattern
909 nigel 75 .sp
910 nigel 63 the ((?:red|white) (king|queen))
911 nigel 75 .sp
912 nigel 63 the captured substrings are "white queen" and "queen", and are numbered 1 and
913 nigel 93 2. The maximum number of capturing subpatterns is 65535.
914 nigel 75 .P
915 nigel 63 As a convenient shorthand, if any option settings are required at the start of
916     a non-capturing subpattern, the option letters may appear between the "?" and
917     the ":". Thus the two patterns
918 nigel 75 .sp
919 nigel 63 (?i:saturday|sunday)
920     (?:(?i)saturday|sunday)
921 nigel 75 .sp
922 nigel 63 match exactly the same set of strings. Because alternative branches are tried
923     from left to right, and options are not reset until the end of the subpattern
924     is reached, an option setting in one branch does affect subsequent branches, so
925     the above patterns match "SUNDAY" as well as "Saturday".
926 nigel 75 .
927     .
929 nigel 63 .rs
930     .sp
931     Identifying capturing parentheses by number is simple, but it can be very hard
932     to keep track of the numbers in complicated regular expressions. Furthermore,
933 nigel 75 if an expression is modified, the numbers may change. To help with this
934 nigel 93 difficulty, PCRE supports the naming of subpatterns. This feature was not
935     added to Perl until release 5.10. Python had the feature earlier, and PCRE
936     introduced it at release 4.0, using the Python syntax. PCRE now supports both
937     the Perl and the Python syntax.
938     .P
939     In PCRE, a subpattern can be named in one of three ways: (?<name>...) or
940     (?'name'...) as in Perl, or (?P<name>...) as in Python. References to capturing
941 nigel 91 parentheses from other parts of the pattern, such as
942     .\" HTML <a href="#backreferences">
943     .\" </a>
944     backreferences,
945     .\"
946     .\" HTML <a href="#recursion">
947     .\" </a>
948     recursion,
949     .\"
950     and
951     .\" HTML <a href="#conditions">
952     .\" </a>
953     conditions,
954     .\"
955     can be made by name as well as by number.
956 nigel 75 .P
957 nigel 91 Names consist of up to 32 alphanumeric characters and underscores. Named
958 nigel 93 capturing parentheses are still allocated numbers as well as names, exactly as
959     if the names were not present. The PCRE API provides function calls for
960     extracting the name-to-number translation table from a compiled pattern. There
961     is also a convenience function for extracting a captured substring by name.
962 nigel 91 .P
963     By default, a name must be unique within a pattern, but it is possible to relax
964     this constraint by setting the PCRE_DUPNAMES option at compile time. This can
965     be useful for patterns where only one instance of the named parentheses can
966     match. Suppose you want to match the name of a weekday, either as a 3-letter
967     abbreviation or as the full name, and in both cases you want to extract the
968     abbreviation. This pattern (ignoring the line breaks) does the job:
969     .sp
970 nigel 93 (?<DN>Mon|Fri|Sun)(?:day)?|
971     (?<DN>Tue)(?:sday)?|
972     (?<DN>Wed)(?:nesday)?|
973     (?<DN>Thu)(?:rsday)?|
974     (?<DN>Sat)(?:urday)?
975 nigel 91 .sp
976     There are five capturing substrings, but only one is ever set after a match.
977     The convenience function for extracting the data by name returns the substring
978 nigel 93 for the first (and in this example, the only) subpattern of that name that
979 nigel 91 matched. This saves searching to find which numbered subpattern it was. If you
980     make a reference to a non-unique named subpattern from elsewhere in the
981     pattern, the one that corresponds to the lowest number is used. For further
982     details of the interfaces for handling named subpatterns, see the
983 nigel 63 .\" HREF
984 nigel 75 \fBpcreapi\fP
985 nigel 63 .\"
986     documentation.
987 nigel 75 .
988     .
989 nigel 63 .SH REPETITION
990     .rs
991     .sp
992     Repetition is specified by quantifiers, which can follow any of the following
993     items:
994 nigel 75 .sp
995 nigel 63 a literal data character
996 nigel 93 the dot metacharacter
997 nigel 75 the \eC escape sequence
998     the \eX escape sequence (in UTF-8 mode with Unicode properties)
999 nigel 93 the \eR escape sequence
1000 nigel 75 an escape such as \ed that matches a single character
1001 nigel 63 a character class
1002     a back reference (see next section)
1003     a parenthesized subpattern (unless it is an assertion)
1004 nigel 75 .sp
1005 nigel 63 The general repetition quantifier specifies a minimum and maximum number of
1006     permitted matches, by giving the two numbers in curly brackets (braces),
1007     separated by a comma. The numbers must be less than 65536, and the first must
1008     be less than or equal to the second. For example:
1009 nigel 75 .sp
1010 nigel 63 z{2,4}
1011 nigel 75 .sp
1012 nigel 63 matches "zz", "zzz", or "zzzz". A closing brace on its own is not a special
1013     character. If the second number is omitted, but the comma is present, there is
1014     no upper limit; if the second number and the comma are both omitted, the
1015     quantifier specifies an exact number of required matches. Thus
1016 nigel 75 .sp
1017 nigel 63 [aeiou]{3,}
1018 nigel 75 .sp
1019 nigel 63 matches at least 3 successive vowels, but may match many more, while
1020 nigel 75 .sp
1021     \ed{8}
1022     .sp
1023 nigel 63 matches exactly 8 digits. An opening curly bracket that appears in a position
1024     where a quantifier is not allowed, or one that does not match the syntax of a
1025     quantifier, is taken as a literal character. For example, {,6} is not a
1026     quantifier, but a literal string of four characters.
1027 nigel 75 .P
1028 nigel 63 In UTF-8 mode, quantifiers apply to UTF-8 characters rather than to individual
1029 nigel 75 bytes. Thus, for example, \ex{100}{2} matches two UTF-8 characters, each of
1030     which is represented by a two-byte sequence. Similarly, when Unicode property
1031     support is available, \eX{3} matches three Unicode extended sequences, each of
1032     which may be several bytes long (and they may be of different lengths).
1033     .P
1034 nigel 63 The quantifier {0} is permitted, causing the expression to behave as if the
1035     previous item and the quantifier were not present.
1036 nigel 75 .P
1037 nigel 93 For convenience, the three most common quantifiers have single-character
1038     abbreviations:
1039 nigel 75 .sp
1040 nigel 63 * is equivalent to {0,}
1041     + is equivalent to {1,}
1042     ? is equivalent to {0,1}
1043 nigel 75 .sp
1044 nigel 63 It is possible to construct infinite loops by following a subpattern that can
1045     match no characters with a quantifier that has no upper limit, for example:
1046 nigel 75 .sp
1047 nigel 63 (a?)*
1048 nigel 75 .sp
1049 nigel 63 Earlier versions of Perl and PCRE used to give an error at compile time for
1050     such patterns. However, because there are cases where this can be useful, such
1051     patterns are now accepted, but if any repetition of the subpattern does in fact
1052     match no characters, the loop is forcibly broken.
1053 nigel 75 .P
1054 nigel 63 By default, the quantifiers are "greedy", that is, they match as much as
1055     possible (up to the maximum number of permitted times), without causing the
1056     rest of the pattern to fail. The classic example of where this gives problems
1057 nigel 75 is in trying to match comments in C programs. These appear between /* and */
1058     and within the comment, individual * and / characters may appear. An attempt to
1059     match C comments by applying the pattern
1060     .sp
1061     /\e*.*\e*/
1062     .sp
1063 nigel 63 to the string
1064 nigel 75 .sp
1065     /* first comment */ not comment /* second comment */
1066     .sp
1067 nigel 63 fails, because it matches the entire string owing to the greediness of the .*
1068     item.
1069 nigel 75 .P
1070 nigel 63 However, if a quantifier is followed by a question mark, it ceases to be
1071     greedy, and instead matches the minimum number of times possible, so the
1072     pattern
1073 nigel 75 .sp
1074     /\e*.*?\e*/
1075     .sp
1076 nigel 63 does the right thing with the C comments. The meaning of the various
1077     quantifiers is not otherwise changed, just the preferred number of matches.
1078     Do not confuse this use of question mark with its use as a quantifier in its
1079     own right. Because it has two uses, it can sometimes appear doubled, as in
1080 nigel 75 .sp
1081     \ed??\ed
1082     .sp
1083 nigel 63 which matches one digit by preference, but can match two if that is the only
1084     way the rest of the pattern matches.
1085 nigel 75 .P
1086 nigel 93 If the PCRE_UNGREEDY option is set (an option that is not available in Perl),
1087 nigel 63 the quantifiers are not greedy by default, but individual ones can be made
1088     greedy by following them with a question mark. In other words, it inverts the
1089     default behaviour.
1090 nigel 75 .P
1091 nigel 63 When a parenthesized subpattern is quantified with a minimum repeat count that
1092 nigel 75 is greater than 1 or with a limited maximum, more memory is required for the
1093 nigel 63 compiled pattern, in proportion to the size of the minimum or maximum.
1094 nigel 75 .P
1095 nigel 63 If a pattern starts with .* or .{0,} and the PCRE_DOTALL option (equivalent
1096 nigel 93 to Perl's /s) is set, thus allowing the dot to match newlines, the pattern is
1097 nigel 63 implicitly anchored, because whatever follows will be tried against every
1098     character position in the subject string, so there is no point in retrying the
1099     overall match at any position after the first. PCRE normally treats such a
1100 nigel 75 pattern as though it were preceded by \eA.
1101     .P
1102 nigel 63 In cases where it is known that the subject string contains no newlines, it is
1103     worth setting PCRE_DOTALL in order to obtain this optimization, or
1104     alternatively using ^ to indicate anchoring explicitly.
1105 nigel 75 .P
1106 nigel 63 However, there is one situation where the optimization cannot be used. When .*
1107     is inside capturing parentheses that are the subject of a backreference
1108 nigel 93 elsewhere in the pattern, a match at the start may fail where a later one
1109     succeeds. Consider, for example:
1110 nigel 75 .sp
1111     (.*)abc\e1
1112     .sp
1113 nigel 63 If the subject is "xyz123abc123" the match point is the fourth character. For
1114     this reason, such a pattern is not implicitly anchored.
1115 nigel 75 .P
1116 nigel 63 When a capturing subpattern is repeated, the value captured is the substring
1117     that matched the final iteration. For example, after
1118 nigel 75 .sp
1119     (tweedle[dume]{3}\es*)+
1120     .sp
1121 nigel 63 has matched "tweedledum tweedledee" the value of the captured substring is
1122     "tweedledee". However, if there are nested capturing subpatterns, the
1123     corresponding captured values may have been set in previous iterations. For
1124     example, after
1125 nigel 75 .sp
1126 nigel 63 /(a|(b))+/
1127 nigel 75 .sp
1128 nigel 63 matches "aba" the value of the second captured substring is "b".
1129 nigel 75 .
1130     .
1131     .\" HTML <a name="atomicgroup"></a>
1133 nigel 63 .rs
1134     .sp
1135 nigel 93 With both maximizing ("greedy") and minimizing ("ungreedy" or "lazy")
1136     repetition, failure of what follows normally causes the repeated item to be
1137     re-evaluated to see if a different number of repeats allows the rest of the
1138     pattern to match. Sometimes it is useful to prevent this, either to change the
1139     nature of the match, or to cause it fail earlier than it otherwise might, when
1140     the author of the pattern knows there is no point in carrying on.
1141 nigel 75 .P
1142     Consider, for example, the pattern \ed+foo when applied to the subject line
1143     .sp
1144 nigel 63 123456bar
1145 nigel 75 .sp
1146 nigel 63 After matching all 6 digits and then failing to match "foo", the normal
1147 nigel 75 action of the matcher is to try again with only 5 digits matching the \ed+
1148 nigel 63 item, and then with 4, and so on, before ultimately failing. "Atomic grouping"
1149     (a term taken from Jeffrey Friedl's book) provides the means for specifying
1150     that once a subpattern has matched, it is not to be re-evaluated in this way.
1151 nigel 75 .P
1152 nigel 93 If we use atomic grouping for the previous example, the matcher gives up
1153 nigel 63 immediately on failing to match "foo" the first time. The notation is a kind of
1154     special parenthesis, starting with (?> as in this example:
1155 nigel 75 .sp
1156     (?>\ed+)foo
1157     .sp
1158 nigel 63 This kind of parenthesis "locks up" the part of the pattern it contains once
1159     it has matched, and a failure further into the pattern is prevented from
1160     backtracking into it. Backtracking past it to previous items, however, works as
1161     normal.
1162 nigel 75 .P
1163 nigel 63 An alternative description is that a subpattern of this type matches the string
1164     of characters that an identical standalone pattern would match, if anchored at
1165     the current point in the subject string.
1166 nigel 75 .P
1167 nigel 63 Atomic grouping subpatterns are not capturing subpatterns. Simple cases such as
1168     the above example can be thought of as a maximizing repeat that must swallow
1169 nigel 75 everything it can. So, while both \ed+ and \ed+? are prepared to adjust the
1170 nigel 63 number of digits they match in order to make the rest of the pattern match,
1171 nigel 75 (?>\ed+) can only match an entire sequence of digits.
1172     .P
1173 nigel 63 Atomic groups in general can of course contain arbitrarily complicated
1174     subpatterns, and can be nested. However, when the subpattern for an atomic
1175     group is just a single repeated item, as in the example above, a simpler
1176     notation, called a "possessive quantifier" can be used. This consists of an
1177     additional + character following a quantifier. Using this notation, the
1178     previous example can be rewritten as
1179 nigel 75 .sp
1180     \ed++foo
1181     .sp
1182 nigel 63 Possessive quantifiers are always greedy; the setting of the PCRE_UNGREEDY
1183     option is ignored. They are a convenient notation for the simpler forms of
1184 nigel 93 atomic group. However, there is no difference in the meaning of a possessive
1185     quantifier and the equivalent atomic group, though there may be a performance
1186     difference; possessive quantifiers should be slightly faster.
1187 nigel 75 .P
1188 nigel 93 The possessive quantifier syntax is an extension to the Perl 5.8 syntax.
1189     Jeffrey Friedl originated the idea (and the name) in the first edition of his
1190     book. Mike McCloskey liked it, so implemented it when he built Sun's Java
1191     package, and PCRE copied it from there. It ultimately found its way into Perl
1192     at release 5.10.
1193 nigel 75 .P
1194 nigel 93 PCRE has an optimization that automatically "possessifies" certain simple
1195     pattern constructs. For example, the sequence A+B is treated as A++B because
1196     there is no point in backtracking into a sequence of A's when B must follow.
1197     .P
1198 nigel 63 When a pattern contains an unlimited repeat inside a subpattern that can itself
1199     be repeated an unlimited number of times, the use of an atomic group is the
1200     only way to avoid some failing matches taking a very long time indeed. The
1201     pattern
1202 nigel 75 .sp
1203     (\eD+|<\ed+>)*[!?]
1204     .sp
1205 nigel 63 matches an unlimited number of substrings that either consist of non-digits, or
1206     digits enclosed in <>, followed by either ! or ?. When it matches, it runs
1207     quickly. However, if it is applied to
1208 nigel 75 .sp
1209 nigel 63 aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
1210 nigel 75 .sp
1211 nigel 63 it takes a long time before reporting failure. This is because the string can
1212 nigel 75 be divided between the internal \eD+ repeat and the external * repeat in a
1213     large number of ways, and all have to be tried. (The example uses [!?] rather
1214     than a single character at the end, because both PCRE and Perl have an
1215     optimization that allows for fast failure when a single character is used. They
1216     remember the last single character that is required for a match, and fail early
1217     if it is not present in the string.) If the pattern is changed so that it uses
1218     an atomic group, like this:
1219     .sp
1220     ((?>\eD+)|<\ed+>)*[!?]
1221     .sp
1222 nigel 63 sequences of non-digits cannot be broken, and failure happens quickly.
1223 nigel 75 .
1224     .
1225     .\" HTML <a name="backreferences"></a>
1227 nigel 63 .rs
1228     .sp
1229     Outside a character class, a backslash followed by a digit greater than 0 (and
1230     possibly further digits) is a back reference to a capturing subpattern earlier
1231     (that is, to its left) in the pattern, provided there have been that many
1232     previous capturing left parentheses.
1233 nigel 75 .P
1234 nigel 63 However, if the decimal number following the backslash is less than 10, it is
1235     always taken as a back reference, and causes an error only if there are not
1236     that many capturing left parentheses in the entire pattern. In other words, the
1237     parentheses that are referenced need not be to the left of the reference for
1238 nigel 91 numbers less than 10. A "forward back reference" of this type can make sense
1239     when a repetition is involved and the subpattern to the right has participated
1240     in an earlier iteration.
1241     .P
1242 nigel 93 It is not possible to have a numerical "forward back reference" to a subpattern
1243     whose number is 10 or more using this syntax because a sequence such as \e50 is
1244     interpreted as a character defined in octal. See the subsection entitled
1245 nigel 91 "Non-printing characters"
1246 nigel 75 .\" HTML <a href="#digitsafterbackslash">
1247     .\" </a>
1248     above
1249     .\"
1250 nigel 93 for further details of the handling of digits following a backslash. There is
1251     no such problem when named parentheses are used. A back reference to any
1252     subpattern is possible using named parentheses (see below).
1253 nigel 75 .P
1254 nigel 93 Another way of avoiding the ambiguity inherent in the use of digits following a
1255     backslash is to use the \eg escape sequence, which is a feature introduced in
1256     Perl 5.10. This escape must be followed by a positive or a negative number,
1257     optionally enclosed in braces. These examples are all identical:
1258     .sp
1259     (ring), \e1
1260     (ring), \eg1
1261     (ring), \eg{1}
1262     .sp
1263     A positive number specifies an absolute reference without the ambiguity that is
1264     present in the older syntax. It is also useful when literal digits follow the
1265     reference. A negative number is a relative reference. Consider this example:
1266     .sp
1267     (abc(def)ghi)\eg{-1}
1268     .sp
1269     The sequence \eg{-1} is a reference to the most recently started capturing
1270     subpattern before \eg, that is, is it equivalent to \e2. Similarly, \eg{-2}
1271     would be equivalent to \e1. The use of relative references can be helpful in
1272     long patterns, and also in patterns that are created by joining together
1273     fragments that contain references within themselves.
1274     .P
1275 nigel 63 A back reference matches whatever actually matched the capturing subpattern in
1276     the current subject string, rather than anything matching the subpattern
1277     itself (see
1278     .\" HTML <a href="#subpatternsassubroutines">
1279     .\" </a>
1280     "Subpatterns as subroutines"
1281     .\"
1282     below for a way of doing that). So the pattern
1283 nigel 75 .sp
1284     (sens|respons)e and \e1ibility
1285     .sp
1286 nigel 63 matches "sense and sensibility" and "response and responsibility", but not
1287     "sense and responsibility". If caseful matching is in force at the time of the
1288     back reference, the case of letters is relevant. For example,
1289 nigel 75 .sp
1290     ((?i)rah)\es+\e1
1291     .sp
1292 nigel 63 matches "rah rah" and "RAH RAH", but not "RAH rah", even though the original
1293     capturing subpattern is matched caselessly.
1294 nigel 75 .P
1295 nigel 93 Back references to named subpatterns use the Perl syntax \ek<name> or \ek'name'
1296     or the Python syntax (?P=name). We could rewrite the above example in either of
1297     the following ways:
1298 nigel 75 .sp
1299 nigel 93 (?<p1>(?i)rah)\es+\ek<p1>
1300 nigel 91 (?P<p1>(?i)rah)\es+(?P=p1)
1301 nigel 75 .sp
1302 nigel 91 A subpattern that is referenced by name may appear in the pattern before or
1303     after the reference.
1304     .P
1305 nigel 63 There may be more than one back reference to the same subpattern. If a
1306     subpattern has not actually been used in a particular match, any back
1307     references to it always fail. For example, the pattern
1308 nigel 75 .sp
1309     (a|(bc))\e2
1310     .sp
1311 nigel 63 always fails if it starts to match "a" rather than "bc". Because there may be
1312     many capturing parentheses in a pattern, all digits following the backslash are
1313     taken as part of a potential back reference number. If the pattern continues
1314     with a digit character, some delimiter must be used to terminate the back
1315     reference. If the PCRE_EXTENDED option is set, this can be whitespace.
1316 nigel 75 Otherwise an empty comment (see
1317     .\" HTML <a href="#comments">
1318     .\" </a>
1319     "Comments"
1320     .\"
1321     below) can be used.
1322     .P
1323 nigel 63 A back reference that occurs inside the parentheses to which it refers fails
1324 nigel 75 when the subpattern is first used, so, for example, (a\e1) never matches.
1325 nigel 63 However, such references can be useful inside repeated subpatterns. For
1326     example, the pattern
1327 nigel 75 .sp
1328     (a|b\e1)+
1329     .sp
1330 nigel 63 matches any number of "a"s and also "aba", "ababbaa" etc. At each iteration of
1331     the subpattern, the back reference matches the character string corresponding
1332     to the previous iteration. In order for this to work, the pattern must be such
1333     that the first iteration does not need to match the back reference. This can be
1334     done using alternation, as in the example above, or by a quantifier with a
1335     minimum of zero.
1336 nigel 75 .
1337     .
1338     .\" HTML <a name="bigassertions"></a>
1339 nigel 63 .SH ASSERTIONS
1340     .rs
1341     .sp
1342     An assertion is a test on the characters following or preceding the current
1343     matching point that does not actually consume any characters. The simple
1344 nigel 75 assertions coded as \eb, \eB, \eA, \eG, \eZ, \ez, ^ and $ are described
1345     .\" HTML <a href="#smallassertions">
1346     .\" </a>
1347     above.
1348     .\"
1349     .P
1350 nigel 63 More complicated assertions are coded as subpatterns. There are two kinds:
1351     those that look ahead of the current position in the subject string, and those
1352 nigel 75 that look behind it. An assertion subpattern is matched in the normal way,
1353     except that it does not cause the current matching position to be changed.
1354     .P
1355     Assertion subpatterns are not capturing subpatterns, and may not be repeated,
1356     because it makes no sense to assert the same thing several times. If any kind
1357     of assertion contains capturing subpatterns within it, these are counted for
1358     the purposes of numbering the capturing subpatterns in the whole pattern.
1359     However, substring capturing is carried out only for positive assertions,
1360     because it does not make sense for negative assertions.
1361     .
1362     .
1363     .SS "Lookahead assertions"
1364     .rs
1365     .sp
1366 nigel 91 Lookahead assertions start with (?= for positive assertions and (?! for
1367     negative assertions. For example,
1368 nigel 75 .sp
1369     \ew+(?=;)
1370     .sp
1371 nigel 63 matches a word followed by a semicolon, but does not include the semicolon in
1372     the match, and
1373 nigel 75 .sp
1374 nigel 63 foo(?!bar)
1375 nigel 75 .sp
1376 nigel 63 matches any occurrence of "foo" that is not followed by "bar". Note that the
1377     apparently similar pattern
1378 nigel 75 .sp
1379 nigel 63 (?!foo)bar
1380 nigel 75 .sp
1381 nigel 63 does not find an occurrence of "bar" that is preceded by something other than
1382     "foo"; it finds any occurrence of "bar" whatsoever, because the assertion
1383     (?!foo) is always true when the next three characters are "bar". A
1384 nigel 75 lookbehind assertion is needed to achieve the other effect.
1385     .P
1386 nigel 63 If you want to force a matching failure at some point in a pattern, the most
1387     convenient way to do it is with (?!) because an empty string always matches, so
1388     an assertion that requires there not to be an empty string must always fail.
1389 nigel 75 .
1390     .
1391     .\" HTML <a name="lookbehind"></a>
1392     .SS "Lookbehind assertions"
1393     .rs
1394     .sp
1395 nigel 63 Lookbehind assertions start with (?<= for positive assertions and (?<! for
1396     negative assertions. For example,
1397 nigel 75 .sp
1398 nigel 63 (?<!foo)bar
1399 nigel 75 .sp
1400 nigel 63 does find an occurrence of "bar" that is not preceded by "foo". The contents of
1401     a lookbehind assertion are restricted such that all the strings it matches must
1402 nigel 91 have a fixed length. However, if there are several top-level alternatives, they
1403     do not all have to have the same fixed length. Thus
1404 nigel 75 .sp
1405 nigel 63 (?<=bullock|donkey)
1406 nigel 75 .sp
1407 nigel 63 is permitted, but
1408 nigel 75 .sp
1409 nigel 63 (?<!dogs?|cats?)
1410 nigel 75 .sp
1411 nigel 63 causes an error at compile time. Branches that match different length strings
1412     are permitted only at the top level of a lookbehind assertion. This is an
1413     extension compared with Perl (at least for 5.8), which requires all branches to
1414     match the same length of string. An assertion such as
1415 nigel 75 .sp
1416 nigel 63 (?<=ab(c|de))
1417 nigel 75 .sp
1418 nigel 63 is not permitted, because its single top-level branch can match two different
1419     lengths, but it is acceptable if rewritten to use two top-level branches:
1420 nigel 75 .sp
1421 nigel 63 (?<=abc|abde)
1422 nigel 75 .sp
1423 nigel 63 The implementation of lookbehind assertions is, for each alternative, to
1424 nigel 93 temporarily move the current position back by the fixed length and then try to
1425 nigel 63 match. If there are insufficient characters before the current position, the
1426 nigel 93 assertion fails.
1427 nigel 75 .P
1428     PCRE does not allow the \eC escape (which matches a single byte in UTF-8 mode)
1429 nigel 63 to appear in lookbehind assertions, because it makes it impossible to calculate
1430 nigel 93 the length of the lookbehind. The \eX and \eR escapes, which can match
1431     different numbers of bytes, are also not permitted.
1432 nigel 75 .P
1433 nigel 93 Possessive quantifiers can be used in conjunction with lookbehind assertions to
1434     specify efficient matching at the end of the subject string. Consider a simple
1435     pattern such as
1436 nigel 75 .sp
1437 nigel 63 abcd$
1438 nigel 75 .sp
1439 nigel 63 when applied to a long string that does not match. Because matching proceeds
1440     from left to right, PCRE will look for each "a" in the subject and then see if
1441     what follows matches the rest of the pattern. If the pattern is specified as
1442 nigel 75 .sp
1443 nigel 63 ^.*abcd$
1444 nigel 75 .sp
1445 nigel 63 the initial .* matches the entire string at first, but when this fails (because
1446     there is no following "a"), it backtracks to match all but the last character,
1447     then all but the last two characters, and so on. Once again the search for "a"
1448     covers the entire string, from right to left, so we are no better off. However,
1449     if the pattern is written as
1450 nigel 75 .sp
1451 nigel 63 ^.*+(?<=abcd)
1452 nigel 75 .sp
1453 nigel 93 there can be no backtracking for the .*+ item; it can match only the entire
1454 nigel 63 string. The subsequent lookbehind assertion does a single test on the last four
1455     characters. If it fails, the match fails immediately. For long strings, this
1456     approach makes a significant difference to the processing time.
1457 nigel 75 .
1458     .
1459     .SS "Using multiple assertions"
1460     .rs
1461     .sp
1462 nigel 63 Several assertions (of any sort) may occur in succession. For example,
1463 nigel 75 .sp
1464     (?<=\ed{3})(?<!999)foo
1465     .sp
1466 nigel 63 matches "foo" preceded by three digits that are not "999". Notice that each of
1467     the assertions is applied independently at the same point in the subject
1468     string. First there is a check that the previous three characters are all
1469     digits, and then there is a check that the same three characters are not "999".
1470 nigel 75 This pattern does \fInot\fP match "foo" preceded by six characters, the first
1471 nigel 63 of which are digits and the last three of which are not "999". For example, it
1472     doesn't match "123abcfoo". A pattern to do that is
1473 nigel 75 .sp
1474     (?<=\ed{3}...)(?<!999)foo
1475     .sp
1476 nigel 63 This time the first assertion looks at the preceding six characters, checking
1477     that the first three are digits, and then the second assertion checks that the
1478     preceding three characters are not "999".
1479 nigel 75 .P
1480 nigel 63 Assertions can be nested in any combination. For example,
1481 nigel 75 .sp
1482 nigel 63 (?<=(?<!foo)bar)baz
1483 nigel 75 .sp
1484 nigel 63 matches an occurrence of "baz" that is preceded by "bar" which in turn is not
1485     preceded by "foo", while
1486 nigel 75 .sp
1487     (?<=\ed{3}(?!999)...)foo
1488     .sp
1489     is another pattern that matches "foo" preceded by three digits and any three
1490 nigel 63 characters that are not "999".
1491 nigel 75 .
1492     .
1493 nigel 91 .\" HTML <a name="conditions"></a>
1495 nigel 63 .rs
1496     .sp
1497     It is possible to cause the matching process to obey a subpattern
1498     conditionally or to choose between two alternative subpatterns, depending on
1499     the result of an assertion, or whether a previous capturing subpattern matched
1500     or not. The two possible forms of conditional subpattern are
1501 nigel 75 .sp
1502 nigel 63 (?(condition)yes-pattern)
1503     (?(condition)yes-pattern|no-pattern)
1504 nigel 75 .sp
1505 nigel 63 If the condition is satisfied, the yes-pattern is used; otherwise the
1506     no-pattern (if present) is used. If there are more than two alternatives in the
1507     subpattern, a compile-time error occurs.
1508 nigel 75 .P
1509 nigel 93 There are four kinds of condition: references to subpatterns, references to
1510     recursion, a pseudo-condition called DEFINE, and assertions.
1511     .
1512     .SS "Checking for a used subpattern by number"
1513     .rs
1514     .sp
1515     If the text between the parentheses consists of a sequence of digits, the
1516     condition is true if the capturing subpattern of that number has previously
1517     matched.
1518 nigel 91 .P
1519     Consider the following pattern, which contains non-significant white space to
1520     make it more readable (assume the PCRE_EXTENDED option) and to divide it into
1521     three parts for ease of discussion:
1522 nigel 75 .sp
1523     ( \e( )? [^()]+ (?(1) \e) )
1524     .sp
1525 nigel 63 The first part matches an optional opening parenthesis, and if that
1526     character is present, sets it as the first captured substring. The second part
1527     matches one or more characters that are not parentheses. The third part is a
1528     conditional subpattern that tests whether the first set of parentheses matched
1529     or not. If they did, that is, if subject started with an opening parenthesis,
1530     the condition is true, and so the yes-pattern is executed and a closing
1531     parenthesis is required. Otherwise, since no-pattern is not present, the
1532     subpattern matches nothing. In other words, this pattern matches a sequence of
1533 nigel 93 non-parentheses, optionally enclosed in parentheses.
1534     .
1535     .SS "Checking for a used subpattern by name"
1536     .rs
1537 nigel 91 .sp
1538 nigel 93 Perl uses the syntax (?(<name>)...) or (?('name')...) to test for a used
1539     subpattern by name. For compatibility with earlier versions of PCRE, which had
1540     this facility before Perl, the syntax (?(name)...) is also recognized. However,
1541     there is a possible ambiguity with this syntax, because subpattern names may
1542     consist entirely of digits. PCRE looks first for a named subpattern; if it
1543     cannot find one and the name consists entirely of digits, PCRE looks for a
1544     subpattern of that number, which must be greater than zero. Using subpattern
1545     names that consist entirely of digits is not recommended.
1546     .P
1547     Rewriting the above example to use a named subpattern gives this:
1548 nigel 91 .sp
1549 nigel 93 (?<OPEN> \e( )? [^()]+ (?(<OPEN>) \e) )
1550     .sp
1551     .
1552     .SS "Checking for pattern recursion"
1553     .rs
1554     .sp
1555 nigel 91 If the condition is the string (R), and there is no subpattern with the name R,
1556 nigel 93 the condition is true if a recursive call to the whole pattern or any
1557     subpattern has been made. If digits or a name preceded by ampersand follow the
1558     letter R, for example:
1559     .sp
1560     (?(R3)...) or (?(R&name)...)
1561     .sp
1562     the condition is true if the most recent recursion is into the subpattern whose
1563     number or name is given. This condition does not check the entire recursion
1564     stack.
1565 nigel 75 .P
1566 nigel 93 At "top level", all these recursion test conditions are false. Recursive
1567     patterns are described below.
1568     .
1569     .SS "Defining subpatterns for use by reference only"
1570     .rs
1571     .sp
1572     If the condition is the string (DEFINE), and there is no subpattern with the
1573     name DEFINE, the condition is always false. In this case, there may be only one
1574     alternative in the subpattern. It is always skipped if control reaches this
1575     point in the pattern; the idea of DEFINE is that it can be used to define
1576     "subroutines" that can be referenced from elsewhere. (The use of "subroutines"
1577     is described below.) For example, a pattern to match an IPv4 address could be
1578     written like this (ignore whitespace and line breaks):
1579     .sp
1580     (?(DEFINE) (?<byte> 2[0-4]\ed | 25[0-5] | 1\ed\ed | [1-9]?\ed) )
1581     \eb (?&byte) (\e.(?&byte)){3} \eb
1582     .sp
1583     The first part of the pattern is a DEFINE group inside which a another group
1584     named "byte" is defined. This matches an individual component of an IPv4
1585     address (a number less than 256). When matching takes place, this part of the
1586     pattern is skipped because DEFINE acts like a false condition.
1587     .P
1588     The rest of the pattern uses references to the named group to match the four
1589     dot-separated components of an IPv4 address, insisting on a word boundary at
1590     each end.
1591     .
1592     .SS "Assertion conditions"
1593     .rs
1594     .sp
1595     If the condition is not in any of the above formats, it must be an assertion.
1596 nigel 63 This may be a positive or negative lookahead or lookbehind assertion. Consider
1597     this pattern, again containing non-significant white space, and with the two
1598     alternatives on the second line:
1599 nigel 75 .sp
1600 nigel 63 (?(?=[^a-z]*[a-z])
1601 nigel 75 \ed{2}-[a-z]{3}-\ed{2} | \ed{2}-\ed{2}-\ed{2} )
1602     .sp
1603 nigel 63 The condition is a positive lookahead assertion that matches an optional
1604     sequence of non-letters followed by a letter. In other words, it tests for the
1605     presence of at least one letter in the subject. If a letter is found, the
1606     subject is matched against the first alternative; otherwise it is matched
1607     against the second. This pattern matches strings in one of the two forms
1608     dd-aaa-dd or dd-dd-dd, where aaa are letters and dd are digits.
1609 nigel 75 .
1610     .
1611     .\" HTML <a name="comments"></a>
1612 nigel 63 .SH COMMENTS
1613     .rs
1614     .sp
1615 nigel 75 The sequence (?# marks the start of a comment that continues up to the next
1616 nigel 63 closing parenthesis. Nested parentheses are not permitted. The characters
1617     that make up a comment play no part in the pattern matching at all.
1618 nigel 75 .P
1619 nigel 63 If the PCRE_EXTENDED option is set, an unescaped # character outside a
1620 nigel 91 character class introduces a comment that continues to immediately after the
1621     next newline in the pattern.
1622 nigel 75 .
1623     .
1624 nigel 91 .\" HTML <a name="recursion"></a>
1625 nigel 75 .SH "RECURSIVE PATTERNS"
1626 nigel 63 .rs
1627     .sp
1628     Consider the problem of matching a string in parentheses, allowing for
1629     unlimited nested parentheses. Without the use of recursion, the best that can
1630     be done is to use a pattern that matches up to some fixed depth of nesting. It
1631 nigel 93 is not possible to handle an arbitrary nesting depth.
1632     .P
1633     For some time, Perl has provided a facility that allows regular expressions to
1634     recurse (amongst other things). It does this by interpolating Perl code in the
1635     expression at run time, and the code can refer to the expression itself. A Perl
1636     pattern using code interpolation to solve the parentheses problem can be
1637     created like this:
1638 nigel 75 .sp
1639     $re = qr{\e( (?: (?>[^()]+) | (?p{$re}) )* \e)}x;
1640     .sp
1641 nigel 63 The (?p{...}) item interpolates Perl code at run time, and in this case refers
1642 nigel 93 recursively to the pattern in which it appears.
1643 nigel 75 .P
1644 nigel 93 Obviously, PCRE cannot support the interpolation of Perl code. Instead, it
1645     supports special syntax for recursion of the entire pattern, and also for
1646     individual subpattern recursion. After its introduction in PCRE and Python,
1647     this kind of recursion was introduced into Perl at release 5.10.
1648 nigel 75 .P
1649 nigel 93 A special item that consists of (? followed by a number greater than zero and a
1650     closing parenthesis is a recursive call of the subpattern of the given number,
1651     provided that it occurs inside that subpattern. (If not, it is a "subroutine"
1652     call, which is described in the next section.) The special item (?R) or (?0) is
1653     a recursive call of the entire regular expression.
1654 nigel 87 .P
1655 nigel 93 In PCRE (like Python, but unlike Perl), a recursive subpattern call is always
1656     treated as an atomic group. That is, once it has matched some of the subject
1657     string, it is never re-entered, even if it contains untried alternatives and
1658     there is a subsequent matching failure.
1659     .P
1660 nigel 87 This PCRE pattern solves the nested parentheses problem (assume the
1661     PCRE_EXTENDED option is set so that white space is ignored):
1662 nigel 75 .sp
1663     \e( ( (?>[^()]+) | (?R) )* \e)
1664     .sp
1665 nigel 63 First it matches an opening parenthesis. Then it matches any number of
1666     substrings which can either be a sequence of non-parentheses, or a recursive
1667 nigel 87 match of the pattern itself (that is, a correctly parenthesized substring).
1668 nigel 63 Finally there is a closing parenthesis.
1669 nigel 75 .P
1670 nigel 63 If this were part of a larger pattern, you would not want to recurse the entire
1671     pattern, so instead you could use this:
1672 nigel 75 .sp
1673     ( \e( ( (?>[^()]+) | (?1) )* \e) )
1674     .sp
1675 nigel 63 We have put the pattern into parentheses, and caused the recursion to refer to
1676     them instead of the whole pattern. In a larger pattern, keeping track of
1677     parenthesis numbers can be tricky. It may be more convenient to use named
1678 nigel 93 parentheses instead. The Perl syntax for this is (?&name); PCRE's earlier
1679     syntax (?P>name) is also supported. We could rewrite the above example as
1680     follows:
1681 nigel 75 .sp
1682 nigel 93 (?<pn> \e( ( (?>[^()]+) | (?&pn) )* \e) )
1683 nigel 75 .sp
1684 nigel 93 If there is more than one subpattern with the same name, the earliest one is
1685     used. This particular example pattern contains nested unlimited repeats, and so
1686     the use of atomic grouping for matching strings of non-parentheses is important
1687 nigel 63 when applying the pattern to strings that do not match. For example, when this
1688     pattern is applied to
1689 nigel 75 .sp
1690 nigel 63 (aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa()
1691 nigel 75 .sp
1692 nigel 63 it yields "no match" quickly. However, if atomic grouping is not used,
1693     the match runs for a very long time indeed because there are so many different
1694     ways the + and * repeats can carve up the subject, and all have to be tested
1695     before failure can be reported.
1696 nigel 75 .P
1697 nigel 63 At the end of a match, the values set for any capturing subpatterns are those
1698     from the outermost level of the recursion at which the subpattern value is set.
1699     If you want to obtain intermediate values, a callout function can be used (see
1700 nigel 93 below and the
1701 nigel 63 .\" HREF
1702 nigel 75 \fBpcrecallout\fP
1703 nigel 63 .\"
1704     documentation). If the pattern above is matched against
1705 nigel 75 .sp
1706 nigel 63 (ab(cd)ef)
1707 nigel 75 .sp
1708 nigel 63 the value for the capturing parentheses is "ef", which is the last value taken
1709     on at the top level. If additional parentheses are added, giving
1710 nigel 75 .sp
1711     \e( ( ( (?>[^()]+) | (?R) )* ) \e)
1712 nigel 63 ^ ^
1713     ^ ^
1714 nigel 75 .sp
1715 nigel 63 the string they capture is "ab(cd)ef", the contents of the top level
1716     parentheses. If there are more than 15 capturing parentheses in a pattern, PCRE
1717     has to obtain extra memory to store data during a recursion, which it does by
1718 nigel 75 using \fBpcre_malloc\fP, freeing it via \fBpcre_free\fP afterwards. If no
1719 nigel 63 memory can be obtained, the match fails with the PCRE_ERROR_NOMEMORY error.
1720 nigel 75 .P
1721 nigel 63 Do not confuse the (?R) item with the condition (R), which tests for recursion.
1722     Consider this pattern, which matches text in angle brackets, allowing for
1723     arbitrary nesting. Only digits are allowed in nested brackets (that is, when
1724     recursing), whereas any characters are permitted at the outer level.
1725 nigel 75 .sp
1726     < (?: (?(R) \ed++ | [^<>]*+) | (?R)) * >
1727     .sp
1728 nigel 63 In this pattern, (?(R) is the start of a conditional subpattern, with two
1729     different alternatives for the recursive and non-recursive cases. The (?R) item
1730     is the actual recursive call.
1731 nigel 75 .
1732     .
1733 nigel 63 .\" HTML <a name="subpatternsassubroutines"></a>
1735 nigel 63 .rs
1736     .sp
1737     If the syntax for a recursive subpattern reference (either by number or by
1738     name) is used outside the parentheses to which it refers, it operates like a
1739 nigel 93 subroutine in a programming language. The "called" subpattern may be defined
1740     before or after the reference. An earlier example pointed out that the pattern
1741 nigel 75 .sp
1742     (sens|respons)e and \e1ibility
1743     .sp
1744 nigel 63 matches "sense and sensibility" and "response and responsibility", but not
1745     "sense and responsibility". If instead the pattern
1746 nigel 75 .sp
1747 nigel 63 (sens|respons)e and (?1)ibility
1748 nigel 75 .sp
1749 nigel 63 is used, it does match "sense and responsibility" as well as the other two
1750 nigel 93 strings. Another example is given in the discussion of DEFINE above.
1751 nigel 87 .P
1752     Like recursive subpatterns, a "subroutine" call is always treated as an atomic
1753     group. That is, once it has matched some of the subject string, it is never
1754     re-entered, even if it contains untried alternatives and there is a subsequent
1755     matching failure.
1756 nigel 93 .P
1757     When a subpattern is used as a subroutine, processing options such as
1758     case-independence are fixed when the subpattern is defined. They cannot be
1759     changed for different calls. For example, consider this pattern:
1760     .sp
1761     (abc)(?i:(?1))
1762     .sp
1763     It matches "abcabc". It does not match "abcABC" because the change of
1764     processing option does not affect the called subpattern.
1765 nigel 75 .
1766     .
1767 nigel 63 .SH CALLOUTS
1768     .rs
1769     .sp
1770     Perl has a feature whereby using the sequence (?{...}) causes arbitrary Perl
1771     code to be obeyed in the middle of matching a regular expression. This makes it
1772     possible, amongst other things, to extract different substrings that match the
1773     same pair of parentheses when there is a repetition.
1774 nigel 75 .P
1775 nigel 63 PCRE provides a similar feature, but of course it cannot obey arbitrary Perl
1776     code. The feature is called "callout". The caller of PCRE provides an external
1777 nigel 75 function by putting its entry point in the global variable \fIpcre_callout\fP.
1778 nigel 63 By default, this variable contains NULL, which disables all calling out.
1779 nigel 75 .P
1780 nigel 63 Within a regular expression, (?C) indicates the points at which the external
1781     function is to be called. If you want to identify different callout points, you
1782     can put a number less than 256 after the letter C. The default value is zero.
1783     For example, this pattern has two callout points:
1784 nigel 75 .sp
1785 nigel 63 (?C1)\dabc(?C2)def
1786 nigel 75 .sp
1787     If the PCRE_AUTO_CALLOUT flag is passed to \fBpcre_compile()\fP, callouts are
1788     automatically installed before each item in the pattern. They are all numbered
1789     255.
1790     .P
1791     During matching, when PCRE reaches a callout point (and \fIpcre_callout\fP is
1792 nigel 63 set), the external function is called. It is provided with the number of the
1793 nigel 75 callout, the position in the pattern, and, optionally, one item of data
1794     originally supplied by the caller of \fBpcre_exec()\fP. The callout function
1795     may cause matching to proceed, to backtrack, or to fail altogether. A complete
1796     description of the interface to the callout function is given in the
1797 nigel 63 .\" HREF
1798 nigel 75 \fBpcrecallout\fP
1799 nigel 63 .\"
1800     documentation.
1801 nigel 93 .
1802     .
1803     .SH "SEE ALSO"
1804     .rs
1805     .sp
1806     \fBpcreapi\fP(3), \fBpcrecallout\fP(3), \fBpcrematching\fP(3), \fBpcre\fP(3).
1807 ph10 99 .
1808     .
1809     .SH AUTHOR
1810     .rs
1811     .sp
1812     .nf
1813     Philip Hazel
1814     University Computing Service
1815     Cambridge CB2 3QH, England.
1816     .fi
1817     .
1818     .
1819     .SH REVISION
1820     .rs
1821     .sp
1822     .nf
1823     Last updated: 06 March 2007
1824     Copyright (c) 1997-2007 University of Cambridge.
1825     .fi


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

ViewVC Help
Powered by ViewVC 1.1.12