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

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

Parent Directory Parent Directory | Revision Log Revision Log | View Patch Patch

revision 972 by ph10, Fri Jun 1 18:03:18 2012 UTC revision 1392 by ph10, Wed Nov 6 18:00:09 2013 UTC
# Line 1  Line 1 
1  .TH PCREPATTERN 3 "04 May 2012" "PCRE 8.31"  .TH PCREPATTERN 3 "05 November 2013" "PCRE 8.34"
2  .SH NAME  .SH NAME
3  PCRE - Perl-compatible regular expressions  PCRE - Perl-compatible regular expressions
4  .SH "PCRE REGULAR EXPRESSION DETAILS"  .SH "PCRE REGULAR EXPRESSION DETAILS"
# Line 20  have copious examples. Jeffrey Friedl's Line 20  have copious examples. Jeffrey Friedl's
20  published by O'Reilly, covers regular expressions in great detail. This  published by O'Reilly, covers regular expressions in great detail. This
21  description of PCRE's regular expressions is intended as reference material.  description of PCRE's regular expressions is intended as reference material.
22  .P  .P
23    This document discusses the patterns that are supported by PCRE when one its
24    main matching functions, \fBpcre_exec()\fP (8-bit) or \fBpcre[16|32]_exec()\fP
25    (16- or 32-bit), is used. PCRE also has alternative matching functions,
26    \fBpcre_dfa_exec()\fP and \fBpcre[16|32_dfa_exec()\fP, which match using a
27    different algorithm that is not Perl-compatible. Some of the features discussed
28    below are not available when DFA matching is used. The advantages and
29    disadvantages of the alternative functions, and how they differ from the normal
30    functions, are discussed in the
31    .\" HREF
32    \fBpcrematching\fP
33    .\"
34    page.
35    .
36    .
37    .SH "SPECIAL START-OF-PATTERN ITEMS"
38    .rs
39    .sp
40    A number of options that can be passed to \fBpcre_compile()\fP can also be set
41    by special items at the start of a pattern. These are not Perl-compatible, but
42    are provided to make these options accessible to pattern writers who are not
43    able to change the program that processes the pattern. Any number of these
44    items may appear, but they must all be together right at the start of the
45    pattern string, and the letters must be in upper case.
46    .
47    .
48    .SS "UTF support"
49    .rs
50    .sp
51  The original operation of PCRE was on strings of one-byte characters. However,  The original operation of PCRE was on strings of one-byte characters. However,
52  there is now also support for UTF-8 strings in the original library, and a  there is now also support for UTF-8 strings in the original library, an
53  second library that supports 16-bit and UTF-16 character strings. To use these  extra library that supports 16-bit and UTF-16 character strings, and a
54    third library that supports 32-bit and UTF-32 character strings. To use these
55  features, PCRE must be built to include appropriate support. When using UTF  features, PCRE must be built to include appropriate support. When using UTF
56  strings you must either call the compiling function with the PCRE_UTF8 or  strings you must either call the compiling function with the PCRE_UTF8,
57  PCRE_UTF16 option, or the pattern must start with one of these special  PCRE_UTF16, or PCRE_UTF32 option, or the pattern must start with one of
58  sequences:  these special sequences:
59  .sp  .sp
60    (*UTF8)    (*UTF8)
61    (*UTF16)    (*UTF16)
62      (*UTF32)
63      (*UTF)
64  .sp  .sp
65    (*UTF) is a generic sequence that can be used with any of the libraries.
66  Starting a pattern with such a sequence is equivalent to setting the relevant  Starting a pattern with such a sequence is equivalent to setting the relevant
67  option. This feature is not Perl-compatible. How setting a UTF mode affects  option. How setting a UTF mode affects pattern matching is mentioned in several
68  pattern matching is mentioned in several places below. There is also a summary  places below. There is also a summary of features in the
 of features in the  
69  .\" HREF  .\" HREF
70  \fBpcreunicode\fP  \fBpcreunicode\fP
71  .\"  .\"
72  page.  page.
73  .P  .P
74  Another special sequence that may appear at the start of a pattern or in  Some applications that allow their users to supply patterns may wish to
75  combination with (*UTF8) or (*UTF16) is:  restrict them to non-UTF data for security reasons. If the PCRE_NEVER_UTF
76    option is set at compile time, (*UTF) etc. are not allowed, and their
77    appearance causes an error.
78    .
79    .
80    .SS "Unicode property support"
81    .rs
82    .sp
83    Another special sequence that may appear at the start of a pattern is
84  .sp  .sp
85    (*UCP)    (*UCP)
86  .sp  .sp
# Line 49  This has the same effect as setting the Line 88  This has the same effect as setting the
88  such as \ed and \ew to use Unicode properties to determine character types,  such as \ed and \ew to use Unicode properties to determine character types,
89  instead of recognizing only characters with codes less than 128 via a lookup  instead of recognizing only characters with codes less than 128 via a lookup
90  table.  table.
91  .P  .
92    .
93    .SS "Disabling start-up optimizations"
94    .rs
95    .sp
96  If a pattern starts with (*NO_START_OPT), it has the same effect as setting the  If a pattern starts with (*NO_START_OPT), it has the same effect as setting the
97  PCRE_NO_START_OPTIMIZE option either at compile or matching time. There are  PCRE_NO_START_OPTIMIZE option either at compile or matching time.
 also some more of these special sequences that are concerned with the handling  
 of newlines; they are described below.  
 .P  
 The remainder of this document discusses the patterns that are supported by  
 PCRE when one its main matching functions, \fBpcre_exec()\fP (8-bit) or  
 \fBpcre16_exec()\fP (16-bit), is used. PCRE also has alternative matching  
 functions, \fBpcre_dfa_exec()\fP and \fBpcre16_dfa_exec()\fP, which match using  
 a different algorithm that is not Perl-compatible. Some of the features  
 discussed below are not available when DFA matching is used. The advantages and  
 disadvantages of the alternative functions, and how they differ from the normal  
 functions, are discussed in the  
 .\" HREF  
 \fBpcrematching\fP  
 .\"  
 page.  
98  .  .
99  .  .
100  .\" HTML <a name="newlines"></a>  .\" HTML <a name="newlines"></a>
101  .SH "NEWLINE CONVENTIONS"  .SS "Newline conventions"
102  .rs  .rs
103  .sp  .sp
104  PCRE supports five different conventions for indicating line breaks in  PCRE supports five different conventions for indicating line breaks in
# Line 103  example, on a Unix system where LF is th Line 131  example, on a Unix system where LF is th
131    (*CR)a.b    (*CR)a.b
132  .sp  .sp
133  changes the convention to CR. That pattern matches "a\enb" because LF is no  changes the convention to CR. That pattern matches "a\enb" because LF is no
134  longer a newline. Note that these special settings, which are not  longer a newline. If more than one of these settings is present, the last one
 Perl-compatible, are recognized only at the very start of a pattern, and that  
 they must be in upper case. If more than one of them is present, the last one  
135  is used.  is used.
136  .P  .P
137  The newline convention affects the interpretation of the dot metacharacter when  The newline convention affects where the circumflex and dollar assertions are
138  PCRE_DOTALL is not set, and also the behaviour of \eN. However, it does not  true. It also affects the interpretation of the dot metacharacter when
139  affect what the \eR escape sequence matches. By default, this is any Unicode  PCRE_DOTALL is not set, and the behaviour of \eN. However, it does not affect
140  newline sequence, for Perl compatibility. However, this can be changed; see the  what the \eR escape sequence matches. By default, this is any Unicode newline
141    sequence, for Perl compatibility. However, this can be changed; see the
142  description of \eR in the section entitled  description of \eR in the section entitled
143  .\" HTML <a href="#newlineseq">  .\" HTML <a href="#newlineseq">
144  .\" </a>  .\" </a>
# Line 121  below. A change of \eR setting can be co Line 148  below. A change of \eR setting can be co
148  convention.  convention.
149  .  .
150  .  .
151    .SS "Setting match and recursion limits"
152    .rs
153    .sp
154    The caller of \fBpcre_exec()\fP can set a limit on the number of times the
155    internal \fBmatch()\fP function is called and on the maximum depth of
156    recursive calls. These facilities are provided to catch runaway matches that
157    are provoked by patterns with huge matching trees (a typical example is a
158    pattern with nested unlimited repeats) and to avoid running out of system stack
159    by too much recursion. When one of these limits is reached, \fBpcre_exec()\fP
160    gives an error return. The limits can also be set by items at the start of the
161    pattern of the form
162    .sp
163      (*LIMIT_MATCH=d)
164      (*LIMIT_RECURSION=d)
165    .sp
166    where d is any number of decimal digits. However, the value of the setting must
167    be less than the value set (or defaulted) by the caller of \fBpcre_exec()\fP
168    for it to have any effect. In other words, the pattern writer can lower the
169    limits set by the programmer, but not raise them. If there is more than one
170    setting of one of these limits, the lower value is used.
171    .
172    .
173    .SH "EBCDIC CHARACTER CODES"
174    .rs
175    .sp
176    PCRE can be compiled to run in an environment that uses EBCDIC as its character
177    code rather than ASCII or Unicode (typically a mainframe system). In the
178    sections below, character code values are ASCII or Unicode; in an EBCDIC
179    environment these characters may have different code values, and there are no
180    code points greater than 255.
181    .
182    .
183  .SH "CHARACTERS AND METACHARACTERS"  .SH "CHARACTERS AND METACHARACTERS"
184  .rs  .rs
185  .sp  .sp
# Line 241  one of the following escape sequences th Line 300  one of the following escape sequences th
300    \en        linefeed (hex 0A)    \en        linefeed (hex 0A)
301    \er        carriage return (hex 0D)    \er        carriage return (hex 0D)
302    \et        tab (hex 09)    \et        tab (hex 09)
303      \e0dd      character with octal code 0dd
304    \eddd      character with octal code ddd, or back reference    \eddd      character with octal code ddd, or back reference
305      \eo{ddd..} character with octal code ddd..
306    \exhh      character with hex code hh    \exhh      character with hex code hh
307    \ex{hhh..} character with hex code hhh.. (non-JavaScript mode)    \ex{hhh..} character with hex code hhh.. (non-JavaScript mode)
308    \euhhhh    character with hex code hhhh (JavaScript mode only)    \euhhhh    character with hex code hhhh (JavaScript mode only)
309  .sp  .sp
310  The precise effect of \ecx is as follows: if x is a lower case letter, it  The precise effect of \ecx on ASCII characters is as follows: if x is a lower
311  is converted to upper case. Then bit 6 of the character (hex 40) is inverted.  case letter, it is converted to upper case. Then bit 6 of the character (hex
312  Thus \ecz becomes hex 1A (z is 7A), but \ec{ becomes hex 3B ({ is 7B), while  40) is inverted. Thus \ecA to \ecZ become hex 01 to hex 1A (A is 41, Z is 5A),
313  \ec; becomes hex 7B (; is 3B). If the byte following \ec has a value greater  but \ec{ becomes hex 3B ({ is 7B), and \ec; becomes hex 7B (; is 3B). If the
314  than 127, a compile-time error occurs. This locks out non-ASCII characters in  data item (byte or 16-bit value) following \ec has a value greater than 127, a
315  all modes. (When PCRE is compiled in EBCDIC mode, all byte values are valid. A  compile-time error occurs. This locks out non-ASCII characters in all modes.
316  lower case letter is converted to upper case, and then the 0xc0 bits are  .P
317  flipped.)  The \ec facility was designed for use with ASCII characters, but with the
318  .P  extension to Unicode it is even less useful than it once was. It is, however,
319  By default, after \ex, from zero to two hexadecimal digits are read (letters  recognized when PCRE is compiled in EBCDIC mode, where data items are always
320  can be in upper or lower case). Any number of hexadecimal digits may appear  bytes. In this mode, all values are valid after \ec. If the next character is a
321  between \ex{ and }, but the character code is constrained as follows:  lower case letter, it is converted to upper case. Then the 0xc0 bits of the
322  .sp  byte are inverted. Thus \ecA becomes hex 01, as in ASCII (A is C1), but because
323    8-bit non-UTF mode    less than 0x100  the EBCDIC letters are disjoint, \ecZ becomes hex 29 (Z is E9), and other
324    8-bit UTF-8 mode      less than 0x10ffff and a valid codepoint  characters also generate different values.
   16-bit non-UTF mode   less than 0x10000  
   16-bit UTF-16 mode    less than 0x10ffff and a valid codepoint  
 .sp  
 Invalid Unicode codepoints are the range 0xd800 to 0xdfff (the so-called  
 "surrogate" codepoints).  
 .P  
 If characters other than hexadecimal digits appear between \ex{ and }, or if  
 there is no terminating }, this form of escape is not recognized. Instead, the  
 initial \ex will be interpreted as a basic hexadecimal escape, with no  
 following digits, giving a character whose value is zero.  
 .P  
 If the PCRE_JAVASCRIPT_COMPAT option is set, the interpretation of \ex is  
 as just described only when it is followed by two hexadecimal digits.  
 Otherwise, it matches a literal "x" character. In JavaScript mode, support for  
 code points greater than 256 is provided by \eu, which must be followed by  
 four hexadecimal digits; otherwise it matches a literal "u" character.  
 .P  
 Characters whose value is less than 256 can be defined by either of the two  
 syntaxes for \ex (or by \eu in JavaScript mode). There is no difference in the  
 way they are handled. For example, \exdc is exactly the same as \ex{dc} (or  
 \eu00dc in JavaScript mode).  
325  .P  .P
326  After \e0 up to two further octal digits are read. If there are fewer than two  After \e0 up to two further octal digits are read. If there are fewer than two
327  digits, just those that are present are used. Thus the sequence \e0\ex\e07  digits, just those that are present are used. Thus the sequence \e0\ex\e07
# Line 289  specifies two binary zeros followed by a Line 329  specifies two binary zeros followed by a
329  sure you supply two digits after the initial zero if the pattern character that  sure you supply two digits after the initial zero if the pattern character that
330  follows is itself an octal digit.  follows is itself an octal digit.
331  .P  .P
332  The handling of a backslash followed by a digit other than 0 is complicated.  The escape \eo must be followed by a sequence of octal digits, enclosed in
333  Outside a character class, PCRE reads it and any following digits as a decimal  braces. An error occurs if this is not the case. This escape is a recent
334  number. If the number is less than 10, or if there have been at least that many  addition to Perl; it provides way of specifying character code points as octal
335    numbers greater than 0777, and it also allows octal numbers and back references
336    to be unambiguously specified.
337    .P
338    For greater clarity and unambiguity, it is best to avoid following \e by a
339    digit greater than zero. Instead, use \eo{} or \ex{} to specify character
340    numbers, and \eg{} to specify back references. The following paragraphs
341    describe the old, ambiguous syntax.
342    .P
343    The handling of a backslash followed by a digit other than 0 is complicated,
344    and Perl has changed in recent releases, causing PCRE also to change. Outside a
345    character class, PCRE reads the digit and any following digits as a decimal
346    number. If the number is less than 8, or if there have been at least that many
347  previous capturing left parentheses in the expression, the entire sequence is  previous capturing left parentheses in the expression, the entire sequence is
348  taken as a \fIback reference\fP. A description of how this works is given  taken as a \fIback reference\fP. A description of how this works is given
349  .\" HTML <a href="#backreferences">  .\" HTML <a href="#backreferences">
# Line 304  following the discussion of Line 356  following the discussion of
356  parenthesized subpatterns.  parenthesized subpatterns.
357  .\"  .\"
358  .P  .P
359  Inside a character class, or if the decimal number is greater than 9 and there  Inside a character class, or if the decimal number following \e is greater than
360  have not been that many capturing subpatterns, PCRE re-reads up to three octal  7 and there have not been that many capturing subpatterns, PCRE handles \e8 and
361  digits following the backslash, and uses them to generate a data character. Any  \e9 as the literal characters "8" and "9", and otherwise re-reads up to three
362  subsequent digits stand for themselves. The value of the character is  octal digits following the backslash, using them to generate a data character.
363  constrained in the same way as characters specified in hexadecimal.  Any subsequent digits stand for themselves. For example:
 For example:  
364  .sp  .sp
365    \e040   is another way of writing a space    \e040   is another way of writing an ASCII space
366  .\" JOIN  .\" JOIN
367    \e40    is the same, provided there are fewer than 40    \e40    is the same, provided there are fewer than 40
368              previous capturing subpatterns              previous capturing subpatterns
# Line 328  For example: Line 379  For example:
379    \e377   might be a back reference, otherwise    \e377   might be a back reference, otherwise
380              the value 255 (decimal)              the value 255 (decimal)
381  .\" JOIN  .\" JOIN
382    \e81    is either a back reference, or a binary zero    \e81    is either a back reference, or the two
383              followed by the two characters "8" and "1"              characters "8" and "1"
384  .sp  .sp
385  Note that octal values of 100 or greater must not be introduced by a leading  Note that octal values of 100 or greater that are specified using this syntax
386  zero, because no more than three octal digits are ever read.  must not be introduced by a leading zero, because no more than three octal
387    digits are ever read.
388    .P
389    By default, after \ex that is not followed by {, from zero to two hexadecimal
390    digits are read (letters can be in upper or lower case). Any number of
391    hexadecimal digits may appear between \ex{ and }. If a character other than
392    a hexadecimal digit appears between \ex{ and }, or if there is no terminating
393    }, an error occurs.
394    .P
395    If the PCRE_JAVASCRIPT_COMPAT option is set, the interpretation of \ex is
396    as just described only when it is followed by two hexadecimal digits.
397    Otherwise, it matches a literal "x" character. In JavaScript mode, support for
398    code points greater than 256 is provided by \eu, which must be followed by
399    four hexadecimal digits; otherwise it matches a literal "u" character.
400  .P  .P
401    Characters whose value is less than 256 can be defined by either of the two
402    syntaxes for \ex (or by \eu in JavaScript mode). There is no difference in the
403    way they are handled. For example, \exdc is exactly the same as \ex{dc} (or
404    \eu00dc in JavaScript mode).
405    .
406    .
407    .SS "Constraints on character values"
408    .rs
409    .sp
410    Characters that are specified using octal or hexadecimal numbers are
411    limited to certain values, as follows:
412    .sp
413      8-bit non-UTF mode    less than 0x100
414      8-bit UTF-8 mode      less than 0x10ffff and a valid codepoint
415      16-bit non-UTF mode   less than 0x10000
416      16-bit UTF-16 mode    less than 0x10ffff and a valid codepoint
417      32-bit non-UTF mode   less than 0x100000000
418      32-bit UTF-32 mode    less than 0x10ffff and a valid codepoint
419    .sp
420    Invalid Unicode codepoints are the range 0xd800 to 0xdfff (the so-called
421    "surrogate" codepoints), and 0xffef.
422    .
423    .
424    .SS "Escape sequences in character classes"
425    .rs
426    .sp
427  All the sequences that define a single character value can be used both inside  All the sequences that define a single character value can be used both inside
428  and outside character classes. In addition, inside a character class, \eb is  and outside character classes. In addition, inside a character class, \eb is
429  interpreted as the backspace character (hex 08).  interpreted as the backspace character (hex 08).
# Line 424  classes. They each match one character o Line 514  classes. They each match one character o
514  matching point is at the end of the subject string, all of them fail, because  matching point is at the end of the subject string, all of them fail, because
515  there is no character to match.  there is no character to match.
516  .P  .P
517  For compatibility with Perl, \es does not match the VT character (code 11).  For compatibility with Perl, \es did not used to match the VT character (code
518  This makes it different from the the POSIX "space" class. The \es characters  11), which made it different from the the POSIX "space" class. However, Perl
519  are HT (9), LF (10), FF (12), CR (13), and space (32). If "use locale;" is  added VT at release 5.18, and PCRE followed suit at release 8.34. The \es
520  included in a Perl script, \es may match the VT character. In PCRE, it never  characters are now HT (9), LF (10), VT (11), FF (12), CR (13), and space (32).
 does.  
521  .P  .P
522  A "word" character is an underscore or any character that is a letter or digit.  A "word" character is an underscore or any character that is a letter or digit.
523  By default, the definition of letters and digits is controlled by PCRE's  By default, the definition of letters and digits is controlled by PCRE's
# Line 454  efficiency reasons. However, if PCRE is Line 543  efficiency reasons. However, if PCRE is
543  and the PCRE_UCP option is set, the behaviour is changed so that Unicode  and the PCRE_UCP option is set, the behaviour is changed so that Unicode
544  properties are used to determine character types, as follows:  properties are used to determine character types, as follows:
545  .sp  .sp
546    \ed  any character that \ep{Nd} matches (decimal digit)    \ed  any character that matches \ep{Nd} (decimal digit)
547    \es  any character that \ep{Z} matches, plus HT, LF, FF, CR    \es  any character that matches \ep{Z} or \eh or \ev
548    \ew  any character that \ep{L} or \ep{N} matches, plus underscore    \ew  any character that matches \ep{L} or \ep{N}, plus underscore
549  .sp  .sp
550  The upper case escapes match the inverse sets of characters. Note that \ed  The upper case escapes match the inverse sets of characters. Note that \ed
551  matches only decimal digits, whereas \ew matches any Unicode digit, as well as  matches only decimal digits, whereas \ew matches any Unicode digit, as well as
# Line 469  release 5.10. In contrast to the other s Line 558  release 5.10. In contrast to the other s
558  characters by default, these always match certain high-valued codepoints,  characters by default, these always match certain high-valued codepoints,
559  whether or not PCRE_UCP is set. The horizontal space characters are:  whether or not PCRE_UCP is set. The horizontal space characters are:
560  .sp  .sp
561    U+0009     Horizontal tab    U+0009     Horizontal tab (HT)
562    U+0020     Space    U+0020     Space
563    U+00A0     Non-break space    U+00A0     Non-break space
564    U+1680     Ogham space mark    U+1680     Ogham space mark
# Line 491  whether or not PCRE_UCP is set. The hori Line 580  whether or not PCRE_UCP is set. The hori
580  .sp  .sp
581  The vertical space characters are:  The vertical space characters are:
582  .sp  .sp
583    U+000A     Linefeed    U+000A     Linefeed (LF)
584    U+000B     Vertical tab    U+000B     Vertical tab (VT)
585    U+000C     Form feed    U+000C     Form feed (FF)
586    U+000D     Carriage return    U+000D     Carriage return (CR)
587    U+0085     Next line    U+0085     Next line (NEL)
588    U+2028     Line separator    U+2028     Line separator
589    U+2029     Paragraph separator    U+2029     Paragraph separator
590  .sp  .sp
# Line 549  change of newline convention; for exampl Line 638  change of newline convention; for exampl
638  .sp  .sp
639    (*ANY)(*BSR_ANYCRLF)    (*ANY)(*BSR_ANYCRLF)
640  .sp  .sp
641  They can also be combined with the (*UTF8), (*UTF16), or (*UCP) special  They can also be combined with the (*UTF8), (*UTF16), (*UTF32), (*UTF) or
642  sequences. Inside a character class, \eR is treated as an unrecognized escape  (*UCP) special sequences. Inside a character class, \eR is treated as an
643  sequence, and so matches the letter "R" by default, but causes an error if  unrecognized escape sequence, and so matches the letter "R" by default, but
644  PCRE_EXTRA is set.  causes an error if PCRE_EXTRA is set.
645  .  .
646  .  .
647  .\" HTML <a name="uniextseq"></a>  .\" HTML <a name="uniextseq"></a>
# Line 567  The extra escape sequences are: Line 656  The extra escape sequences are:
656  .sp  .sp
657    \ep{\fIxx\fP}   a character with the \fIxx\fP property    \ep{\fIxx\fP}   a character with the \fIxx\fP property
658    \eP{\fIxx\fP}   a character without the \fIxx\fP property    \eP{\fIxx\fP}   a character without the \fIxx\fP property
659    \eX       an extended Unicode sequence    \eX       a Unicode extended grapheme cluster
660  .sp  .sp
661  The property names represented by \fIxx\fP above are limited to the Unicode  The property names represented by \fIxx\fP above are limited to the Unicode
662  script names, the general category properties, "Any", which matches any  script names, the general category properties, "Any", which matches any
# Line 760  a modifier or "other". Line 849  a modifier or "other".
849  The Cs (Surrogate) property applies only to characters in the range U+D800 to  The Cs (Surrogate) property applies only to characters in the range U+D800 to
850  U+DFFF. Such characters are not valid in Unicode strings and so  U+DFFF. Such characters are not valid in Unicode strings and so
851  cannot be tested by PCRE, unless UTF validity checking has been turned off  cannot be tested by PCRE, unless UTF validity checking has been turned off
852  (see the discussion of PCRE_NO_UTF8_CHECK and PCRE_NO_UTF16_CHECK in the  (see the discussion of PCRE_NO_UTF8_CHECK, PCRE_NO_UTF16_CHECK and
853    PCRE_NO_UTF32_CHECK in the
854  .\" HREF  .\" HREF
855  \fBpcreapi\fP  \fBpcreapi\fP
856  .\"  .\"
# Line 775  Instead, this property is assumed for an Line 865  Instead, this property is assumed for an
865  Unicode table.  Unicode table.
866  .P  .P
867  Specifying caseless matching does not affect these escape sequences. For  Specifying caseless matching does not affect these escape sequences. For
868  example, \ep{Lu} always matches only upper case letters.  example, \ep{Lu} always matches only upper case letters. This is different from
869    the behaviour of current versions of Perl.
870  .P  .P
871  The \eX escape matches any number of Unicode characters that form an extended  Matching characters by Unicode property is not fast, because PCRE has to do a
872  Unicode sequence. \eX is equivalent to  multistage table lookup in order to find a character's property. That is why
873  .sp  the traditional escape sequences such as \ed and \ew do not use Unicode
874    (?>\ePM\epM*)  properties in PCRE by default, though you can make them do so by setting the
875    PCRE_UCP option or by starting the pattern with (*UCP).
876    .
877    .
878    .SS Extended grapheme clusters
879    .rs
880  .sp  .sp
881  That is, it matches a character without the "mark" property, followed by zero  The \eX escape matches any number of Unicode characters that form an "extended
882  or more characters with the "mark" property, and treats the sequence as an  grapheme cluster", and treats the sequence as an atomic group
 atomic group  
883  .\" HTML <a href="#atomicgroup">  .\" HTML <a href="#atomicgroup">
884  .\" </a>  .\" </a>
885  (see below).  (see below).
886  .\"  .\"
887  Characters with the "mark" property are typically accents that affect the  Up to and including release 8.31, PCRE matched an earlier, simpler definition
888  preceding character. None of them have codepoints less than 256, so in  that was equivalent to
889  8-bit non-UTF-8 mode \eX matches any one character.  .sp
890  .P    (?>\ePM\epM*)
891  Note that recent versions of Perl have changed \eX to match what Unicode calls  .sp
892  an "extended grapheme cluster", which has a more complicated definition.  That is, it matched a character without the "mark" property, followed by zero
893  .P  or more characters with the "mark" property. Characters with the "mark"
894  Matching characters by Unicode property is not fast, because PCRE has to search  property are typically non-spacing accents that affect the preceding character.
895  a structure that contains data for over fifteen thousand characters. That is  .P
896  why the traditional escape sequences such as \ed and \ew do not use Unicode  This simple definition was extended in Unicode to include more complicated
897  properties in PCRE by default, though you can make them do so by setting the  kinds of composite character by giving each character a grapheme breaking
898  PCRE_UCP option or by starting the pattern with (*UCP).  property, and creating rules that use these properties to define the boundaries
899    of extended grapheme clusters. In releases of PCRE later than 8.31, \eX matches
900    one of these clusters.
901    .P
902    \eX always matches at least one character. Then it decides whether to add
903    additional characters according to the following rules for ending a cluster:
904    .P
905    1. End at the end of the subject string.
906    .P
907    2. Do not end between CR and LF; otherwise end after any control character.
908    .P
909    3. Do not break Hangul (a Korean script) syllable sequences. Hangul characters
910    are of five types: L, V, T, LV, and LVT. An L character may be followed by an
911    L, V, LV, or LVT character; an LV or V character may be followed by a V or T
912    character; an LVT or T character may be follwed only by a T character.
913    .P
914    4. Do not end before extending characters or spacing marks. Characters with
915    the "mark" property always have the "extend" grapheme breaking property.
916    .P
917    5. Do not end after prepend characters.
918    .P
919    6. Otherwise, end the cluster.
920  .  .
921  .  .
922  .\" HTML <a name="extraprops"></a>  .\" HTML <a name="extraprops"></a>
923  .SS PCRE's additional properties  .SS PCRE's additional properties
924  .rs  .rs
925  .sp  .sp
926  As well as the standard Unicode properties described in the previous  As well as the standard Unicode properties described above, PCRE supports four
927  section, PCRE supports four more that make it possible to convert traditional  more that make it possible to convert traditional escape sequences such as \ew
928  escape sequences such as \ew and \es and POSIX character classes to use Unicode  and \es to use Unicode properties. PCRE uses these non-standard, non-Perl
929  properties. PCRE uses these non-standard, non-Perl properties internally when  properties internally when PCRE_UCP is set. However, they may also be used
930  PCRE_UCP is set. They are:  explicitly. These properties are:
931  .sp  .sp
932    Xan   Any alphanumeric character    Xan   Any alphanumeric character
933    Xps   Any POSIX space character    Xps   Any POSIX space character
# Line 821  PCRE_UCP is set. They are: Line 937  PCRE_UCP is set. They are:
937  Xan matches characters that have either the L (letter) or the N (number)  Xan matches characters that have either the L (letter) or the N (number)
938  property. Xps matches the characters tab, linefeed, vertical tab, form feed, or  property. Xps matches the characters tab, linefeed, vertical tab, form feed, or
939  carriage return, and any other character that has the Z (separator) property.  carriage return, and any other character that has the Z (separator) property.
940  Xsp is the same as Xps, except that vertical tab is excluded. Xwd matches the  Xsp is the same as Xps; it used to exclude vertical tab, for Perl
941  same characters as Xan, plus underscore.  compatibility, but Perl changed, and so PCRE followed at release 8.34. Xwd
942    matches the same characters as Xan, plus underscore.
943    .P
944    There is another non-standard property, Xuc, which matches any character that
945    can be represented by a Universal Character Name in C++ and other programming
946    languages. These are the characters $, @, ` (grave accent), and all characters
947    with Unicode code points greater than or equal to U+00A0, except for the
948    surrogates U+D800 to U+DFFF. Note that most base (ASCII) characters are
949    excluded. (Universal Character Names are of the form \euHHHH or \eUHHHHHHHH
950    where H is a hexadecimal digit. Note that the Xuc property does not match these
951    sequences but the characters that they represent.)
952  .  .
953  .  .
954  .\" HTML <a name="resetmatchstart"></a>  .\" HTML <a name="resetmatchstart"></a>
# Line 928  regular expression. Line 1054  regular expression.
1054  .SH "CIRCUMFLEX AND DOLLAR"  .SH "CIRCUMFLEX AND DOLLAR"
1055  .rs  .rs
1056  .sp  .sp
1057    The circumflex and dollar metacharacters are zero-width assertions. That is,
1058    they test for a particular condition being true without consuming any
1059    characters from the subject string.
1060    .P
1061  Outside a character class, in the default matching mode, the circumflex  Outside a character class, in the default matching mode, the circumflex
1062  character is an assertion that is true only if the current matching point is  character is an assertion that is true only if the current matching point is at
1063  at the start of the subject string. If the \fIstartoffset\fP argument of  the start of the subject string. If the \fIstartoffset\fP argument of
1064  \fBpcre_exec()\fP is non-zero, circumflex can never match if the PCRE_MULTILINE  \fBpcre_exec()\fP is non-zero, circumflex can never match if the PCRE_MULTILINE
1065  option is unset. Inside a character class, circumflex has an entirely different  option is unset. Inside a character class, circumflex has an entirely different
1066  meaning  meaning
# Line 947  constrained to match only at the start o Line 1077  constrained to match only at the start o
1077  "anchored" pattern. (There are also other constructs that can cause a pattern  "anchored" pattern. (There are also other constructs that can cause a pattern
1078  to be anchored.)  to be anchored.)
1079  .P  .P
1080  A dollar character is an assertion that is true only if the current matching  The dollar character is an assertion that is true only if the current matching
1081  point is at the end of the subject string, or immediately before a newline  point is at the end of the subject string, or immediately before a newline at
1082  at the end of the string (by default). Dollar need not be the last character of  the end of the string (by default). Note, however, that it does not actually
1083  the pattern if a number of alternatives are involved, but it should be the last  match the newline. Dollar need not be the last character of the pattern if a
1084  item in any branch in which it appears. Dollar has no special meaning in a  number of alternatives are involved, but it should be the last item in any
1085  character class.  branch in which it appears. Dollar has no special meaning in a character class.
1086  .P  .P
1087  The meaning of dollar can be changed so that it matches only at the very end of  The meaning of dollar can be changed so that it matches only at the very end of
1088  the string, by setting the PCRE_DOLLAR_ENDONLY option at compile time. This  the string, by setting the PCRE_DOLLAR_ENDONLY option at compile time. This
# Line 1013  name; PCRE does not support this. Line 1143  name; PCRE does not support this.
1143  .sp  .sp
1144  Outside a character class, the escape sequence \eC matches any one data unit,  Outside a character class, the escape sequence \eC matches any one data unit,
1145  whether or not a UTF mode is set. In the 8-bit library, one data unit is one  whether or not a UTF mode is set. In the 8-bit library, one data unit is one
1146  byte; in the 16-bit library it is a 16-bit unit. Unlike a dot, \eC always  byte; in the 16-bit library it is a 16-bit unit; in the 32-bit library it is
1147    a 32-bit unit. Unlike a dot, \eC always
1148  matches line-ending characters. The feature is provided in Perl in order to  matches line-ending characters. The feature is provided in Perl in order to
1149  match individual bytes in UTF-8 mode, but it is unclear how it can usefully be  match individual bytes in UTF-8 mode, but it is unclear how it can usefully be
1150  used. Because \eC breaks up characters into individual data units, matching one  used. Because \eC breaks up characters into individual data units, matching one
1151  unit with \eC in a UTF mode means that the rest of the string may start with a  unit with \eC in a UTF mode means that the rest of the string may start with a
1152  malformed UTF character. This has undefined results, because PCRE assumes that  malformed UTF character. This has undefined results, because PCRE assumes that
1153  it is dealing with valid UTF strings (and by default it checks this at the  it is dealing with valid UTF strings (and by default it checks this at the
1154  start of processing unless the PCRE_NO_UTF8_CHECK or PCRE_NO_UTF16_CHECK option  start of processing unless the PCRE_NO_UTF8_CHECK, PCRE_NO_UTF16_CHECK or
1155  is used).  PCRE_NO_UTF32_CHECK option is used).
1156  .P  .P
1157  PCRE does not allow \eC to appear in lookbehind assertions  PCRE does not allow \eC to appear in lookbehind assertions
1158  .\" HTML <a href="#lookbehind">  .\" HTML <a href="#lookbehind">
# Line 1080  circumflex is not an assertion; it still Line 1211  circumflex is not an assertion; it still
1211  string, and therefore it fails if the current pointer is at the end of the  string, and therefore it fails if the current pointer is at the end of the
1212  string.  string.
1213  .P  .P
1214  In UTF-8 (UTF-16) mode, characters with values greater than 255 (0xffff) can be  In UTF-8 (UTF-16, UTF-32) mode, characters with values greater than 255 (0xffff)
1215  included in a class as a literal string of data units, or by using the \ex{  can be included in a class as a literal string of data units, or by using the
1216  escaping mechanism.  \ex{ escaping mechanism.
1217  .P  .P
1218  When caseless matching is set, any letters in a class represent both their  When caseless matching is set, any letters in a class represent both their
1219  upper case and lower case versions, so for example, a caseless [aeiou] matches  upper case and lower case versions, so for example, a caseless [aeiou] matches
# Line 1104  The minus (hyphen) character can be used Line 1235  The minus (hyphen) character can be used
1235  character class. For example, [d-m] matches any letter between d and m,  character class. For example, [d-m] matches any letter between d and m,
1236  inclusive. If a minus character is required in a class, it must be escaped with  inclusive. If a minus character is required in a class, it must be escaped with
1237  a backslash or appear in a position where it cannot be interpreted as  a backslash or appear in a position where it cannot be interpreted as
1238  indicating a range, typically as the first or last character in the class.  indicating a range, typically as the first or last character in the class, or
1239    immediately after a range. For example, [b-d-z] matches letters in the range b
1240    to d, a hyphen character, or z.
1241  .P  .P
1242  It is not possible to have the literal character "]" as the end character of a  It is not possible to have the literal character "]" as the end character of a
1243  range. A pattern such as [W-]46] is interpreted as a class of two characters  range. A pattern such as [W-]46] is interpreted as a class of two characters
# Line 1114  the end of range, so [W-\e]46] is interp Line 1247  the end of range, so [W-\e]46] is interp
1247  followed by two other characters. The octal or hexadecimal representation of  followed by two other characters. The octal or hexadecimal representation of
1248  "]" can also be used to end a range.  "]" can also be used to end a range.
1249  .P  .P
1250    An error is generated if a POSIX character class (see below) or an escape
1251    sequence other than one that defines a single character appears at a point
1252    where a range ending character is expected. For example, [z-\exff] is valid,
1253    but [A-\ed] and [A-[:digit:]] are not.
1254    .P
1255  Ranges operate in the collating sequence of character values. They can also be  Ranges operate in the collating sequence of character values. They can also be
1256  used for characters specified numerically, for example [\e000-\e037]. Ranges  used for characters specified numerically, for example [\e000-\e037]. Ranges
1257  can include any characters that are valid for the current mode.  can include any characters that are valid for the current mode.
# Line 1179  are: Line 1317  are:
1317    lower    lower case letters    lower    lower case letters
1318    print    printing characters, including space    print    printing characters, including space
1319    punct    printing characters, excluding letters and digits and space    punct    printing characters, excluding letters and digits and space
1320    space    white space (not quite the same as \es)    space    white space (the same as \es from PCRE 8.34)
1321    upper    upper case letters    upper    upper case letters
1322    word     "word" characters (same as \ew)    word     "word" characters (same as \ew)
1323    xdigit   hexadecimal digits    xdigit   hexadecimal digits
1324  .sp  .sp
1325  The "space" characters are HT (9), LF (10), VT (11), FF (12), CR (13), and  The "space" characters are HT (9), LF (10), VT (11), FF (12), CR (13), and
1326  space (32). Notice that this list includes the VT character (code 11). This  space (32). "Space" used to be different to \es, which did not include VT, for
1327  makes "space" different to \es, which does not include VT (for Perl  Perl compatibility. However, Perl changed at release 5.18, and PCRE followed at
1328  compatibility).  release 8.34. "Space" and \es now match the same set of characters.
1329  .P  .P
1330  The name "word" is a Perl extension, and "blank" is a GNU extension from Perl  The name "word" is a Perl extension, and "blank" is a GNU extension from Perl
1331  5.8. Another Perl extension is negation, which is indicated by a ^ character  5.8. Another Perl extension is negation, which is indicated by a ^ character
# Line 1202  supported, and an error is given if they Line 1340  supported, and an error is given if they
1340  By default, in UTF modes, characters with values greater than 128 do not match  By default, in UTF modes, characters with values greater than 128 do not match
1341  any of the POSIX character classes. However, if the PCRE_UCP option is passed  any of the POSIX character classes. However, if the PCRE_UCP option is passed
1342  to \fBpcre_compile()\fP, some of the classes are changed so that Unicode  to \fBpcre_compile()\fP, some of the classes are changed so that Unicode
1343  character properties are used. This is achieved by replacing the POSIX classes  character properties are used. This is achieved by replacing certain POSIX
1344  by other sequences, as follows:  classes by other sequences, as follows:
1345  .sp  .sp
1346    [:alnum:]  becomes  \ep{Xan}    [:alnum:]  becomes  \ep{Xan}
1347    [:alpha:]  becomes  \ep{L}    [:alpha:]  becomes  \ep{L}
# Line 1214  by other sequences, as follows: Line 1352  by other sequences, as follows:
1352    [:upper:]  becomes  \ep{Lu}    [:upper:]  becomes  \ep{Lu}
1353    [:word:]   becomes  \ep{Xwd}    [:word:]   becomes  \ep{Xwd}
1354  .sp  .sp
1355  Negated versions, such as [:^alpha:] use \eP instead of \ep. The other POSIX  Negated versions, such as [:^alpha:] use \eP instead of \ep. Three other POSIX
1356  classes are unchanged, and match only characters with code points less than  classes are handled specially in UCP mode:
1357  128.  .TP 10
1358    [:graph:]
1359    This matches characters that have glyphs that mark the page when printed. In
1360    Unicode property terms, it matches all characters with the L, M, N, P, S, or Cf
1361    properties, except for:
1362    .sp
1363      U+061C           Arabic Letter Mark
1364      U+180E           Mongolian Vowel Separator
1365      U+2066 - U+2069  Various "isolate"s
1366    .sp
1367    .TP 10
1368    [:print:]
1369    This matches the same characters as [:graph:] plus space characters that are
1370    not controls, that is, characters with the Zs property.
1371    .TP 10
1372    [:punct:]
1373    This matches all characters that have the Unicode P (punctuation) property,
1374    plus those characters whose code points are less than 128 that have the S
1375    (Symbol) property.
1376    .P
1377    The other POSIX classes are unchanged, and match only characters with code
1378    points less than 128.
1379  .  .
1380  .  .
1381  .SH "VERTICAL BAR"  .SH "VERTICAL BAR"
# Line 1295  the section entitled Line 1454  the section entitled
1454  .\" </a>  .\" </a>
1455  "Newline sequences"  "Newline sequences"
1456  .\"  .\"
1457  above. There are also the (*UTF8), (*UTF16), and (*UCP) leading sequences that  above. There are also the (*UTF8), (*UTF16),(*UTF32), and (*UCP) leading
1458  can be used to set UTF and Unicode property modes; they are equivalent to  sequences that can be used to set UTF and Unicode property modes; they are
1459  setting the PCRE_UTF8, PCRE_UTF16, and the PCRE_UCP options, respectively.  equivalent to setting the PCRE_UTF8, PCRE_UTF16, PCRE_UTF32 and the PCRE_UCP
1460    options, respectively. The (*UTF) sequence is a generic version that can be
1461    used with any of the libraries. However, the application can set the
1462    PCRE_NEVER_UTF option, which locks out the use of the (*UTF) sequences.
1463  .  .
1464  .  .
1465  .\" HTML <a name="subpattern"></a>  .\" HTML <a name="subpattern"></a>
# Line 1463  for the first (and in this example, the Line 1625  for the first (and in this example, the
1625  matched. This saves searching to find which numbered subpattern it was.  matched. This saves searching to find which numbered subpattern it was.
1626  .P  .P
1627  If you make a back reference to a non-unique named subpattern from elsewhere in  If you make a back reference to a non-unique named subpattern from elsewhere in
1628  the pattern, the one that corresponds to the first occurrence of the name is  the pattern, the subpatterns to which the name refers are checked in the order
1629  used. In the absence of duplicate numbers (see the previous section) this is  in which they appear in the overall pattern. The first one that is set is used
1630  the one with the lowest number. If you use a named reference in a condition  for the reference. For example, this pattern matches both "foofoo" and
1631    "barbar" but not "foobar" or "barfoo":
1632    .sp
1633      (?:(?<n>foo)|(?<n>bar))\k<n>
1634    .sp
1635    .P
1636    If you make a subroutine call to a non-unique named subpattern, the one that
1637    corresponds to the first occurrence of the name is used. In the absence of
1638    duplicate numbers (see the previous section) this is the one with the lowest
1639    number.
1640    .P
1641    If you use a named reference in a condition
1642  test (see the  test (see the
1643  .\"  .\"
1644  .\" HTML <a href="#conditions">  .\" HTML <a href="#conditions">
# Line 1485  documentation. Line 1658  documentation.
1658  \fBWarning:\fP You cannot use different names to distinguish between two  \fBWarning:\fP You cannot use different names to distinguish between two
1659  subpatterns with the same number because PCRE uses only the numbers when  subpatterns with the same number because PCRE uses only the numbers when
1660  matching. For this reason, an error is given at compile time if different names  matching. For this reason, an error is given at compile time if different names
1661  are given to subpatterns with the same number. However, you can give the same  are given to subpatterns with the same number. However, you can always give the
1662  name to subpatterns with the same number, even when PCRE_DUPNAMES is not set.  same name to subpatterns with the same number, even when PCRE_DUPNAMES is not
1663    set.
1664  .  .
1665  .  .
1666  .SH REPETITION  .SH REPETITION
# Line 1532  quantifier, but a literal string of four Line 1706  quantifier, but a literal string of four
1706  In UTF modes, quantifiers apply to characters rather than to individual data  In UTF modes, quantifiers apply to characters rather than to individual data
1707  units. Thus, for example, \ex{100}{2} matches two characters, each of  units. Thus, for example, \ex{100}{2} matches two characters, each of
1708  which is represented by a two-byte sequence in a UTF-8 string. Similarly,  which is represented by a two-byte sequence in a UTF-8 string. Similarly,
1709  \eX{3} matches three Unicode extended sequences, each of which may be several  \eX{3} matches three Unicode extended grapheme clusters, each of which may be
1710  data units long (and they may be of different lengths).  several data units long (and they may be of different lengths).
1711  .P  .P
1712  The quantifier {0} is permitted, causing the expression to behave as if the  The quantifier {0} is permitted, causing the expression to behave as if the
1713  previous item and the quantifier were not present. This may be useful for  previous item and the quantifier were not present. This may be useful for
# Line 1619  In cases where it is known that the subj Line 1793  In cases where it is known that the subj
1793  worth setting PCRE_DOTALL in order to obtain this optimization, or  worth setting PCRE_DOTALL in order to obtain this optimization, or
1794  alternatively using ^ to indicate anchoring explicitly.  alternatively using ^ to indicate anchoring explicitly.
1795  .P  .P
1796  However, there is one situation where the optimization cannot be used. When .*  However, there are some cases where the optimization cannot be used. When .*
1797  is inside capturing parentheses that are the subject of a back reference  is inside capturing parentheses that are the subject of a back reference
1798  elsewhere in the pattern, a match at the start may fail where a later one  elsewhere in the pattern, a match at the start may fail where a later one
1799  succeeds. Consider, for example:  succeeds. Consider, for example:
# Line 1629  succeeds. Consider, for example: Line 1803  succeeds. Consider, for example:
1803  If the subject is "xyz123abc123" the match point is the fourth character. For  If the subject is "xyz123abc123" the match point is the fourth character. For
1804  this reason, such a pattern is not implicitly anchored.  this reason, such a pattern is not implicitly anchored.
1805  .P  .P
1806    Another case where implicit anchoring is not applied is when the leading .* is
1807    inside an atomic group. Once again, a match at the start may fail where a later
1808    one succeeds. Consider this pattern:
1809    .sp
1810      (?>.*?a)b
1811    .sp
1812    It matches "ab" in the subject "aab". The use of the backtracking control verbs
1813    (*PRUNE) and (*SKIP) also disable this optimization.
1814    .P
1815  When a capturing subpattern is repeated, the value captured is the substring  When a capturing subpattern is repeated, the value captured is the substring
1816  that matched the final iteration. For example, after  that matched the final iteration. For example, after
1817  .sp  .sp
# Line 1897  except that it does not cause the curren Line 2080  except that it does not cause the curren
2080  Assertion subpatterns are not capturing subpatterns. If such an assertion  Assertion subpatterns are not capturing subpatterns. If such an assertion
2081  contains capturing subpatterns within it, these are counted for the purposes of  contains capturing subpatterns within it, these are counted for the purposes of
2082  numbering the capturing subpatterns in the whole pattern. However, substring  numbering the capturing subpatterns in the whole pattern. However, substring
2083  capturing is carried out only for positive assertions, because it does not make  capturing is carried out only for positive assertions. (Perl sometimes, but not
2084  sense for negative assertions.  always, does do capturing in negative assertions.)
2085  .P  .P
2086  For compatibility with Perl, assertion subpatterns may be repeated; though  For compatibility with Perl, assertion subpatterns may be repeated; though
2087  it makes no sense to assert the same thing several times, the side effect of  it makes no sense to assert the same thing several times, the side effect of
# Line 2550  same pair of parentheses when there is a Line 2733  same pair of parentheses when there is a
2733  PCRE provides a similar feature, but of course it cannot obey arbitrary Perl  PCRE provides a similar feature, but of course it cannot obey arbitrary Perl
2734  code. The feature is called "callout". The caller of PCRE provides an external  code. The feature is called "callout". The caller of PCRE provides an external
2735  function by putting its entry point in the global variable \fIpcre_callout\fP  function by putting its entry point in the global variable \fIpcre_callout\fP
2736  (8-bit library) or \fIpcre16_callout\fP (16-bit library). By default, this  (8-bit library) or \fIpcre[16|32]_callout\fP (16-bit or 32-bit library).
2737  variable contains NULL, which disables all calling out.  By default, this variable contains NULL, which disables all calling out.
2738  .P  .P
2739  Within a regular expression, (?C) indicates the points at which the external  Within a regular expression, (?C) indicates the points at which the external
2740  function is to be called. If you want to identify different callout points, you  function is to be called. If you want to identify different callout points, you
# Line 2562  For example, this pattern has two callou Line 2745  For example, this pattern has two callou
2745  .sp  .sp
2746  If the PCRE_AUTO_CALLOUT flag is passed to a compiling function, callouts are  If the PCRE_AUTO_CALLOUT flag is passed to a compiling function, callouts are
2747  automatically installed before each item in the pattern. They are all numbered  automatically installed before each item in the pattern. They are all numbered
2748  255.  255. If there is a conditional group in the pattern whose condition is an
2749    assertion, an additional callout is inserted just before the condition. An
2750    explicit callout may also be set at this position, as in this example:
2751    .sp
2752      (?(?C9)(?=a)abc|def)
2753    .sp
2754    Note that this applies only to assertion conditions, not to other types of
2755    condition.
2756  .P  .P
2757  During matching, when PCRE reaches a callout point, the external function is  During matching, when PCRE reaches a callout point, the external function is
2758  called. It is provided with the number of the callout, the position in the  called. It is provided with the number of the callout, the position in the
# Line 2581  documentation. Line 2771  documentation.
2771  .rs  .rs
2772  .sp  .sp
2773  Perl 5.10 introduced a number of "Special Backtracking Control Verbs", which  Perl 5.10 introduced a number of "Special Backtracking Control Verbs", which
2774  are described in the Perl documentation as "experimental and subject to change  are still described in the Perl documentation as "experimental and subject to
2775  or removal in a future version of Perl". It goes on to say: "Their usage in  change or removal in a future version of Perl". It goes on to say: "Their usage
2776  production code should be noted to avoid problems during upgrades." The same  in production code should be noted to avoid problems during upgrades." The same
2777  remarks apply to the PCRE features described in this section.  remarks apply to the PCRE features described in this section.
2778  .P  .P
2779    The new verbs make use of what was previously invalid syntax: an opening
2780    parenthesis followed by an asterisk. They are generally of the form
2781    (*VERB) or (*VERB:NAME). Some may take either form, possibly behaving
2782    differently depending on whether or not a name is present. A name is any
2783    sequence of characters that does not include a closing parenthesis. The maximum
2784    length of name is 255 in the 8-bit library and 65535 in the 16-bit and 32-bit
2785    libraries. If the name is empty, that is, if the closing parenthesis
2786    immediately follows the colon, the effect is as if the colon were not there.
2787    Any number of these verbs may occur in a pattern.
2788    .P
2789  Since these verbs are specifically related to backtracking, most of them can be  Since these verbs are specifically related to backtracking, most of them can be
2790  used only when the pattern is to be matched using one of the traditional  used only when the pattern is to be matched using one of the traditional
2791  matching functions, which use a backtracking algorithm. With the exception of  matching functions, because these use a backtracking algorithm. With the
2792  (*FAIL), which behaves like a failing negative assertion, they cause an error  exception of (*FAIL), which behaves like a failing negative assertion, the
2793  if encountered by a DFA matching function.  backtracking control verbs cause an error if encountered by a DFA matching
2794  .P  function.
 If any of these verbs are used in an assertion or in a subpattern that is  
 called as a subroutine (whether or not recursively), their effect is confined  
 to that subpattern; it does not extend to the surrounding pattern, with one  
 exception: the name from a *(MARK), (*PRUNE), or (*THEN) that is encountered in  
 a successful positive assertion \fIis\fP passed back when a match succeeds  
 (compare capturing parentheses in assertions). Note that such subpatterns are  
 processed as anchored at the point where they are tested. Note also that Perl's  
 treatment of subroutines and assertions is different in some cases.  
2795  .P  .P
2796  The new verbs make use of what was previously invalid syntax: an opening  The behaviour of these verbs in
2797  parenthesis followed by an asterisk. They are generally of the form  .\" HTML <a href="#btrepeat">
2798  (*VERB) or (*VERB:NAME). Some may take either form, with differing behaviour,  .\" </a>
2799  depending on whether or not an argument is present. A name is any sequence of  repeated groups,
2800  characters that does not include a closing parenthesis. The maximum length of  .\"
2801  name is 255 in the 8-bit library and 65535 in the 16-bit library. If the name  .\" HTML <a href="#btassert">
2802  is empty, that is, if the closing parenthesis immediately follows the colon,  .\" </a>
2803  the effect is as if the colon were not there. Any number of these verbs may  assertions,
2804  occur in a pattern.  .\"
2805    and in
2806    .\" HTML <a href="#btsub">
2807    .\" </a>
2808    subpatterns called as subroutines
2809    .\"
2810    (whether or not recursively) is documented below.
2811  .  .
2812  .  .
2813  .\" HTML <a name="nooptimize"></a>  .\" HTML <a name="nooptimize"></a>
# Line 2619  occur in a pattern. Line 2817  occur in a pattern.
2817  PCRE contains some optimizations that are used to speed up matching by running  PCRE contains some optimizations that are used to speed up matching by running
2818  some checks at the start of each match attempt. For example, it may know the  some checks at the start of each match attempt. For example, it may know the
2819  minimum length of matching subject, or that a particular character must be  minimum length of matching subject, or that a particular character must be
2820  present. When one of these optimizations suppresses the running of a match, any  present. When one of these optimizations bypasses the running of a match, any
2821  included backtracking verbs will not, of course, be processed. You can suppress  included backtracking verbs will not, of course, be processed. You can suppress
2822  the start-of-match optimizations by setting the PCRE_NO_START_OPTIMIZE option  the start-of-match optimizations by setting the PCRE_NO_START_OPTIMIZE option
2823  when calling \fBpcre_compile()\fP or \fBpcre_exec()\fP, or by starting the  when calling \fBpcre_compile()\fP or \fBpcre_exec()\fP, or by starting the
# Line 2633  in the Line 2831  in the
2831  .\" HREF  .\" HREF
2832  \fBpcreapi\fP  \fBpcreapi\fP
2833  .\"  .\"
2834  documentation.  documentation.
2835  .P  .P
2836  Experiments with Perl suggest that it too has similar optimizations, sometimes  Experiments with Perl suggest that it too has similar optimizations, sometimes
2837  leading to anomalous results.  leading to anomalous results.
# Line 2650  followed by a name. Line 2848  followed by a name.
2848  This verb causes the match to end successfully, skipping the remainder of the  This verb causes the match to end successfully, skipping the remainder of the
2849  pattern. However, when it is inside a subpattern that is called as a  pattern. However, when it is inside a subpattern that is called as a
2850  subroutine, only that subpattern is ended successfully. Matching then continues  subroutine, only that subpattern is ended successfully. Matching then continues
2851  at the outer level. If (*ACCEPT) is inside capturing parentheses, the data so  at the outer level. If (*ACCEPT) in triggered in a positive assertion, the
2852  far is captured. For example:  assertion succeeds; in a negative assertion, the assertion fails.
2853    .P
2854    If (*ACCEPT) is inside capturing parentheses, the data so far is captured. For
2855    example:
2856  .sp  .sp
2857    A((?:A|B(*ACCEPT)|C)D)    A((?:A|B(*ACCEPT)|C)D)
2858  .sp  .sp
# Line 2684  starting point (see (*SKIP) below). Line 2885  starting point (see (*SKIP) below).
2885  A name is always required with this verb. There may be as many instances of  A name is always required with this verb. There may be as many instances of
2886  (*MARK) as you like in a pattern, and their names do not have to be unique.  (*MARK) as you like in a pattern, and their names do not have to be unique.
2887  .P  .P
2888  When a match succeeds, the name of the last-encountered (*MARK) on the matching  When a match succeeds, the name of the last-encountered (*MARK:NAME),
2889  path is passed back to the caller as described in the section entitled  (*PRUNE:NAME), or (*THEN:NAME) on the matching path is passed back to the
2890    caller as described in the section entitled
2891  .\" HTML <a href="pcreapi.html#extradata">  .\" HTML <a href="pcreapi.html#extradata">
2892  .\" </a>  .\" </a>
2893  "Extra data for \fBpcre_exec()\fP"  "Extra data for \fBpcre_exec()\fP"
# Line 2710  indicates which of the two alternatives Line 2912  indicates which of the two alternatives
2912  of obtaining this information than putting each alternative in its own  of obtaining this information than putting each alternative in its own
2913  capturing parentheses.  capturing parentheses.
2914  .P  .P
2915  If (*MARK) is encountered in a positive assertion, its name is recorded and  If a verb with a name is encountered in a positive assertion that is true, the
2916  passed back if it is the last-encountered. This does not happen for negative  name is recorded and passed back if it is the last-encountered. This does not
2917  assertions.  happen for negative assertions or failing positive assertions.
2918  .P  .P
2919  After a partial match or a failed match, the name of the last encountered  After a partial match or a failed match, the last encountered name in the
2920  (*MARK) in the entire match process is returned. For example:  entire match process is returned. For example:
2921  .sp  .sp
2922      re> /X(*MARK:A)Y|X(*MARK:B)Z/K      re> /X(*MARK:A)Y|X(*MARK:B)Z/K
2923    data> XP    data> XP
# Line 2727  attempts starting at "P" and then with a Line 2929  attempts starting at "P" and then with a
2929  (*MARK) item, but nevertheless do not reset it.  (*MARK) item, but nevertheless do not reset it.
2930  .P  .P
2931  If you are interested in (*MARK) values after failed matches, you should  If you are interested in (*MARK) values after failed matches, you should
2932  probably set the PCRE_NO_START_OPTIMIZE option  probably set the PCRE_NO_START_OPTIMIZE option
2933  .\" HTML <a href="#nooptimize">  .\" HTML <a href="#nooptimize">
2934  .\" </a>  .\" </a>
2935  (see above)  (see above)
2936  .\"  .\"
2937  to ensure that the match is always attempted.  to ensure that the match is always attempted.
2938  .  .
# Line 2741  to ensure that the match is always attem Line 2943  to ensure that the match is always attem
2943  The following verbs do nothing when they are encountered. Matching continues  The following verbs do nothing when they are encountered. Matching continues
2944  with what follows, but if there is no subsequent match, causing a backtrack to  with what follows, but if there is no subsequent match, causing a backtrack to
2945  the verb, a failure is forced. That is, backtracking cannot pass to the left of  the verb, a failure is forced. That is, backtracking cannot pass to the left of
2946  the verb. However, when one of these verbs appears inside an atomic group, its  the verb. However, when one of these verbs appears inside an atomic group or an
2947  effect is confined to that group, because once the group has been matched,  assertion that is true, its effect is confined to that group, because once the
2948  there is never any backtracking into it. In this situation, backtracking can  group has been matched, there is never any backtracking into it. In this
2949  "jump back" to the left of the entire atomic group. (Remember also, as stated  situation, backtracking can "jump back" to the left of the entire atomic group
2950  above, that this localization also applies in subroutine calls and assertions.)  or assertion. (Remember also, as stated above, that this localization also
2951    applies in subroutine calls.)
2952  .P  .P
2953  These verbs differ in exactly what kind of failure occurs when backtracking  These verbs differ in exactly what kind of failure occurs when backtracking
2954  reaches them.  reaches them. The behaviour described below is what happens when the verb is
2955    not in a subroutine or an assertion. Subsequent sections cover these special
2956    cases.
2957  .sp  .sp
2958    (*COMMIT)    (*COMMIT)
2959  .sp  .sp
2960  This verb, which may not be followed by a name, causes the whole match to fail  This verb, which may not be followed by a name, causes the whole match to fail
2961  outright if the rest of the pattern does not match. Even if the pattern is  outright if there is a later matching failure that causes backtracking to reach
2962  unanchored, no further attempts to find a match by advancing the starting point  it. Even if the pattern is unanchored, no further attempts to find a match by
2963  take place. Once (*COMMIT) has been passed, \fBpcre_exec()\fP is committed to  advancing the starting point take place. If (*COMMIT) is the only backtracking
2964  finding a match at the current starting point, or not at all. For example:  verb that is encountered, once it has been passed \fBpcre_exec()\fP is
2965    committed to finding a match at the current starting point, or not at all. For
2966    example:
2967  .sp  .sp
2968    a+(*COMMIT)b    a+(*COMMIT)b
2969  .sp  .sp
# Line 2765  dynamic anchor, or "I've started, so I m Line 2972  dynamic anchor, or "I've started, so I m
2972  recently passed (*MARK) in the path is passed back when (*COMMIT) forces a  recently passed (*MARK) in the path is passed back when (*COMMIT) forces a
2973  match failure.  match failure.
2974  .P  .P
2975    If there is more than one backtracking verb in a pattern, a different one that
2976    follows (*COMMIT) may be triggered first, so merely passing (*COMMIT) during a
2977    match does not always guarantee that a match must be at this starting point.
2978    .P
2979  Note that (*COMMIT) at the start of a pattern is not the same as an anchor,  Note that (*COMMIT) at the start of a pattern is not the same as an anchor,
2980  unless PCRE's start-of-match optimizations are turned off, as shown in this  unless PCRE's start-of-match optimizations are turned off, as shown in this
2981  \fBpcretest\fP example:  \fBpcretest\fP example:
# Line 2784  starting points. Line 2995  starting points.
2995    (*PRUNE) or (*PRUNE:NAME)    (*PRUNE) or (*PRUNE:NAME)
2996  .sp  .sp
2997  This verb causes the match to fail at the current starting position in the  This verb causes the match to fail at the current starting position in the
2998  subject if the rest of the pattern does not match. If the pattern is  subject if there is a later matching failure that causes backtracking to reach
2999  unanchored, the normal "bumpalong" advance to the next starting character then  it. If the pattern is unanchored, the normal "bumpalong" advance to the next
3000  happens. Backtracking can occur as usual to the left of (*PRUNE), before it is  starting character then happens. Backtracking can occur as usual to the left of
3001  reached, or when matching to the right of (*PRUNE), but if there is no match to  (*PRUNE), before it is reached, or when matching to the right of (*PRUNE), but
3002  the right, backtracking cannot cross (*PRUNE). In simple cases, the use of  if there is no match to the right, backtracking cannot cross (*PRUNE). In
3003  (*PRUNE) is just an alternative to an atomic group or possessive quantifier,  simple cases, the use of (*PRUNE) is just an alternative to an atomic group or
3004  but there are some uses of (*PRUNE) that cannot be expressed in any other way.  possessive quantifier, but there are some uses of (*PRUNE) that cannot be
3005  The behaviour of (*PRUNE:NAME) is the same as (*MARK:NAME)(*PRUNE). In an  expressed in any other way. In an anchored pattern (*PRUNE) has the same effect
3006  anchored pattern (*PRUNE) has the same effect as (*COMMIT).  as (*COMMIT).
3007    .P
3008    The behaviour of (*PRUNE:NAME) is the not the same as (*MARK:NAME)(*PRUNE).
3009    It is like (*MARK:NAME) in that the name is remembered for passing back to the
3010    caller. However, (*SKIP:NAME) searches only for names set with (*MARK).
3011  .sp  .sp
3012    (*SKIP)    (*SKIP)
3013  .sp  .sp
# Line 2813  instead of skipping on to "c". Line 3028  instead of skipping on to "c".
3028  .sp  .sp
3029    (*SKIP:NAME)    (*SKIP:NAME)
3030  .sp  .sp
3031  When (*SKIP) has an associated name, its behaviour is modified. If the  When (*SKIP) has an associated name, its behaviour is modified. When it is
3032  following pattern fails to match, the previous path through the pattern is  triggered, the previous path through the pattern is searched for the most
3033  searched for the most recent (*MARK) that has the same name. If one is found,  recent (*MARK) that has the same name. If one is found, the "bumpalong" advance
3034  the "bumpalong" advance is to the subject position that corresponds to that  is to the subject position that corresponds to that (*MARK) instead of to where
3035  (*MARK) instead of to where (*SKIP) was encountered. If no (*MARK) with a  (*SKIP) was encountered. If no (*MARK) with a matching name is found, the
3036  matching name is found, the (*SKIP) is ignored.  (*SKIP) is ignored.
3037    .P
3038    Note that (*SKIP:NAME) searches only for names set by (*MARK:NAME). It ignores
3039    names that are set by (*PRUNE:NAME) or (*THEN:NAME).
3040  .sp  .sp
3041    (*THEN) or (*THEN:NAME)    (*THEN) or (*THEN:NAME)
3042  .sp  .sp
3043  This verb causes a skip to the next innermost alternative if the rest of the  This verb causes a skip to the next innermost alternative when backtracking
3044  pattern does not match. That is, it cancels pending backtracking, but only  reaches it. That is, it cancels any further backtracking within the current
3045  within the current alternative. Its name comes from the observation that it can  alternative. Its name comes from the observation that it can be used for a
3046  be used for a pattern-based if-then-else block:  pattern-based if-then-else block:
3047  .sp  .sp
3048    ( COND1 (*THEN) FOO | COND2 (*THEN) BAR | COND3 (*THEN) BAZ ) ...    ( COND1 (*THEN) FOO | COND2 (*THEN) BAR | COND3 (*THEN) BAZ ) ...
3049  .sp  .sp
3050  If the COND1 pattern matches, FOO is tried (and possibly further items after  If the COND1 pattern matches, FOO is tried (and possibly further items after
3051  the end of the group if FOO succeeds); on failure, the matcher skips to the  the end of the group if FOO succeeds); on failure, the matcher skips to the
3052  second alternative and tries COND2, without backtracking into COND1. The  second alternative and tries COND2, without backtracking into COND1. If that
3053  behaviour of (*THEN:NAME) is exactly the same as (*MARK:NAME)(*THEN).  succeeds and BAR fails, COND3 is tried. If subsequently BAZ fails, there are no
3054  If (*THEN) is not inside an alternation, it acts like (*PRUNE).  more alternatives, so there is a backtrack to whatever came before the entire
3055    group. If (*THEN) is not inside an alternation, it acts like (*PRUNE).
3056    .P
3057    The behaviour of (*THEN:NAME) is the not the same as (*MARK:NAME)(*THEN).
3058    It is like (*MARK:NAME) in that the name is remembered for passing back to the
3059    caller. However, (*SKIP:NAME) searches only for names set with (*MARK).
3060  .P  .P
3061  Note that a subpattern that does not contain a | character is just a part of  A subpattern that does not contain a | character is just a part of the
3062  the enclosing alternative; it is not a nested alternation with only one  enclosing alternative; it is not a nested alternation with only one
3063  alternative. The effect of (*THEN) extends beyond such a subpattern to the  alternative. The effect of (*THEN) extends beyond such a subpattern to the
3064  enclosing alternative. Consider this pattern, where A, B, etc. are complex  enclosing alternative. Consider this pattern, where A, B, etc. are complex
3065  pattern fragments that do not contain any | characters at this level:  pattern fragments that do not contain any | characters at this level:
# Line 2855  in C, matching moves to (*FAIL), which c Line 3078  in C, matching moves to (*FAIL), which c
3078  because there are no more alternatives to try. In this case, matching does now  because there are no more alternatives to try. In this case, matching does now
3079  backtrack into A.  backtrack into A.
3080  .P  .P
3081  Note also that a conditional subpattern is not considered as having two  Note that a conditional subpattern is not considered as having two
3082  alternatives, because only one is ever used. In other words, the | character in  alternatives, because only one is ever used. In other words, the | character in
3083  a conditional subpattern has a different meaning. Ignoring white space,  a conditional subpattern has a different meaning. Ignoring white space,
3084  consider:  consider:
# Line 2877  starting position, but allowing an advan Line 3100  starting position, but allowing an advan
3100  unanchored pattern). (*SKIP) is similar, except that the advance may be more  unanchored pattern). (*SKIP) is similar, except that the advance may be more
3101  than one character. (*COMMIT) is the strongest, causing the entire match to  than one character. (*COMMIT) is the strongest, causing the entire match to
3102  fail.  fail.
3103    .
3104    .
3105    .SS "More than one backtracking verb"
3106    .rs
3107    .sp
3108    If more than one backtracking verb is present in a pattern, the one that is
3109    backtracked onto first acts. For example, consider this pattern, where A, B,
3110    etc. are complex pattern fragments:
3111    .sp
3112      (A(*COMMIT)B(*THEN)C|ABD)
3113    .sp
3114    If A matches but B fails, the backtrack to (*COMMIT) causes the entire match to
3115    fail. However, if A and B match, but C fails, the backtrack to (*THEN) causes
3116    the next alternative (ABD) to be tried. This behaviour is consistent, but is
3117    not always the same as Perl's. It means that if two or more backtracking verbs
3118    appear in succession, all the the last of them has no effect. Consider this
3119    example:
3120    .sp
3121      ...(*COMMIT)(*PRUNE)...
3122    .sp
3123    If there is a matching failure to the right, backtracking onto (*PRUNE) causes
3124    it to be triggered, and its action is taken. There can never be a backtrack
3125    onto (*COMMIT).
3126    .
3127    .
3128    .\" HTML <a name="btrepeat"></a>
3129    .SS "Backtracking verbs in repeated groups"
3130    .rs
3131    .sp
3132    PCRE differs from Perl in its handling of backtracking verbs in repeated
3133    groups. For example, consider:
3134    .sp
3135      /(a(*COMMIT)b)+ac/
3136    .sp
3137    If the subject is "abac", Perl matches, but PCRE fails because the (*COMMIT) in
3138    the second repeat of the group acts.
3139    .
3140    .
3141    .\" HTML <a name="btassert"></a>
3142    .SS "Backtracking verbs in assertions"
3143    .rs
3144    .sp
3145    (*FAIL) in an assertion has its normal effect: it forces an immediate backtrack.
3146    .P
3147    (*ACCEPT) in a positive assertion causes the assertion to succeed without any
3148    further processing. In a negative assertion, (*ACCEPT) causes the assertion to
3149    fail without any further processing.
3150    .P
3151    The other backtracking verbs are not treated specially if they appear in a
3152    positive assertion. In particular, (*THEN) skips to the next alternative in the
3153    innermost enclosing group that has alternations, whether or not this is within
3154    the assertion.
3155    .P
3156    Negative assertions are, however, different, in order to ensure that changing a
3157    positive assertion into a negative assertion changes its result. Backtracking
3158    into (*COMMIT), (*SKIP), or (*PRUNE) causes a negative assertion to be true,
3159    without considering any further alternative branches in the assertion.
3160    Backtracking into (*THEN) causes it to skip to the next enclosing alternative
3161    within the assertion (the normal behaviour), but if the assertion does not have
3162    such an alternative, (*THEN) behaves like (*PRUNE).
3163    .
3164    .
3165    .\" HTML <a name="btsub"></a>
3166    .SS "Backtracking verbs in subroutines"
3167    .rs
3168    .sp
3169    These behaviours occur whether or not the subpattern is called recursively.
3170    Perl's treatment of subroutines is different in some cases.
3171    .P
3172    (*FAIL) in a subpattern called as a subroutine has its normal effect: it forces
3173    an immediate backtrack.
3174    .P
3175    (*ACCEPT) in a subpattern called as a subroutine causes the subroutine match to
3176    succeed without any further processing. Matching then continues after the
3177    subroutine call.
3178    .P
3179    (*COMMIT), (*SKIP), and (*PRUNE) in a subpattern called as a subroutine cause
3180    the subroutine match to fail.
3181  .P  .P
3182  If more than one such verb is present in a pattern, the "strongest" one wins.  (*THEN) skips to the next alternative in the innermost enclosing group within
3183  For example, consider this pattern, where A, B, etc. are complex pattern  the subpattern that has alternatives. If there is no such group within the
3184  fragments:  subpattern, (*THEN) causes the subroutine match to fail.
 .sp  
   (A(*COMMIT)B(*THEN)C|D)  
 .sp  
 Once A has matched, PCRE is committed to this match, at the current starting  
 position. If subsequently B matches, but C does not, the normal (*THEN) action  
 of trying the next alternative (that is, D) does not happen because (*COMMIT)  
 overrides.  
3185  .  .
3186  .  .
3187  .SH "SEE ALSO"  .SH "SEE ALSO"
3188  .rs  .rs
3189  .sp  .sp
3190  \fBpcreapi\fP(3), \fBpcrecallout\fP(3), \fBpcrematching\fP(3),  \fBpcreapi\fP(3), \fBpcrecallout\fP(3), \fBpcrematching\fP(3),
3191  \fBpcresyntax\fP(3), \fBpcre\fP(3), \fBpcre16(3)\fP.  \fBpcresyntax\fP(3), \fBpcre\fP(3), \fBpcre16(3)\fP, \fBpcre32(3)\fP.
3192  .  .
3193  .  .
3194  .SH AUTHOR  .SH AUTHOR
# Line 2911  Cambridge CB2 3QH, England. Line 3205  Cambridge CB2 3QH, England.
3205  .rs  .rs
3206  .sp  .sp
3207  .nf  .nf
3208  Last updated: 01 June 2012  Last updated: 05 November 2013
3209  Copyright (c) 1997-2012 University of Cambridge.  Copyright (c) 1997-2013 University of Cambridge.
3210  .fi  .fi

Legend:
Removed from v.972  
changed lines
  Added in v.1392

  ViewVC Help
Powered by ViewVC 1.1.5