PCRE Syntax Quick Reference

The full syntax and semantics of the regular expressions that are supported by PCRE are described in the pcrepattern documentation. This document contains just a quick-reference summary of the syntax.

PCRE REGULAR EXPRESSION SYNTAX SUMMARY

QUOTING

\x         where x is non-alphanumeric is a literal x
\Q…\E    treat enclosed characters as literal

CHARACTERS

\a         alarm, that is, the BEL character (hex 07)
\cx        “control-x”, where x is any character
\e         escape (hex 1B)
\f         formfeed (hex 0C)
\n         newline (hex 0A)
\r         carriage return (hex 0D)
\t         tab (hex 09)
\ddd       character with octal code ddd, or backreference
\xhh       character with hex code hh
\x{hhh..}  character with hex code hhh..

CHARACTER TYPES

.          any character except newline;
in dotall mode, any character whatsoever
\C         one byte, even in UTF-8 mode (best avoided)
\d         a decimal digit
\D         a character that is not a decimal digit
\h         a horizontal whitespace character
\H         a character that is not a horizontal whitespace character
\p{xx}     a character with the xx property
\P{xx}     a character without the xx property
\R         a newline sequence
\s         a whitespace character
\S         a character that is not a whitespace character
\v         a vertical whitespace character
\V         a character that is not a vertical whitespace character
\w         a “word” character
\W         a “non-word” character
\X         an extended Unicode sequence

In PCRE, \d, \D, \s, \S, \w, and \W recognize only ASCII characters.

GENERAL CATEGORY PROPERTY CODES FOR \p and \P

C          Other
Cc         Control
Cf         Format
Cn         Unassigned
Co         Private use
Cs         Surrogate

L          Letter
Ll         Lower case letter
Lm         Modifier letter
Lo         Other letter
Lt         Title case letter
Lu         Upper case letter
L&         Ll, Lu, or Lt

M          Mark
Mc         Spacing mark
Me         Enclosing mark
Mn         Non-spacing mark

N          Number
Nd         Decimal number
Nl         Letter number
No         Other number

P          Punctuation
Pc         Connector punctuation
Pd         Dash punctuation
Pe         Close punctuation
Pf         Final punctuation
Pi         Initial punctuation
Po         Other punctuation
Ps         Open punctuation

S          Symbol
Sc         Currency symbol
Sk         Modifier symbol
Sm         Mathematical symbol
So         Other symbol

Z          Separator
Zl         Line separator
Zp         Paragraph separator
Zs         Space separator

SCRIPT NAMES FOR \p AND \P

Arabic, Armenian, Balinese, Bengali, Bopomofo, Braille, Buginese, Buhid, Canadian_Aboriginal, Cherokee, Common, Coptic, Cuneiform, Cypriot, Cyrillic, Deseret, Devanagari, Ethiopic, Georgian, Glagolitic, Gothic, Greek, Gujarati, Gurmukhi, Han, Hangul, Hanunoo, Hebrew, Hiragana, Inherited, Kannada, Katakana, Kharoshthi, Khmer, Lao, Latin, Limbu, Linear_B, Malayalam, Mongolian, Myanmar, New_Tai_Lue, Nko, Ogham, Old_Italic, Old_Persian, Oriya, Osmanya, Phags_Pa, Phoenician, Runic, Shavian, Sinhala, Syloti_Nagri, Syriac, Tagalog, Tagbanwa, Tai_Le, Tamil, Telugu, Thaana, Thai, Tibetan, Tifinagh, Ugaritic, Yi.

CHARACTER CLASSES

[…]       positive character class
[^…]      negative character class
[x-y]       range (can be used for hex characters)
[[:xxx:]]   positive POSIX named set
[[^:xxx:]]  negative POSIX named set

alnum       alphanumeric
alpha       alphabetic
ascii       0-127
blank       space or tab
cntrl       control character
digit       decimal digit
graph       printing, excluding space
lower       lower case letter
print       printing, including space
punct       printing, excluding alphanumeric
space       whitespace
upper       upper case letter
word        same as \w
xdigit      hexadecimal digit

In PCRE, POSIX character set names recognize only ASCII characters. You can use \Q…\E inside a character class.

QUANTIFIERS

?           0 or 1, greedy
?+          0 or 1, possessive
??          0 or 1, lazy
*           0 or more, greedy
*+          0 or more, possessive
*?          0 or more, lazy
+           1 or more, greedy
++          1 or more, possessive
+?          1 or more, lazy
{n}         exactly n
{n,m}       at least n, no more than m, greedy
{n,m}+      at least n, no more than m, possessive
{n,m}?      at least n, no more than m, lazy
{n,}        n or more, greedy
{n,}+       n or more, possessive
{n,}?       n or more, lazy

ANCHORS AND SIMPLE ASSERTIONS

\b          word boundary
\B          not a word boundary
^           start of subject
also after internal newline in multiline mode
\A          start of subject
$           end of subject
also before newline at end of subject
also before internal newline in multiline mode
\Z          end of subject
also before newline at end of subject
\z          end of subject
\G          first matching position in subject

MATCH POINT RESET

\K          reset start of match

ALTERNATION

expr|expr|expr…

CAPTURING

(…)          capturing group
(?<name>…)   named capturing group (Perl)
(?’name’…)   named capturing group (Perl)
(?P<name>…)  named capturing group (Python)
(?:…)        non-capturing group
(?|…)        non-capturing group; reset group numbers for
capturing groups in each alternative

ATOMIC GROUPS

(?>…)        atomic, non-capturing group

COMMENT

(?#….)       comment (not nestable)

OPTION SETTING

(?i)           caseless
(?J)           allow duplicate names
(?m)           multiline
(?s)           single line (dotall)
(?U)           default ungreedy (lazy)
(?x)           extended (ignore white space)
(?-…)        unset option(s)

LOOKAHEAD AND LOOKBEHIND ASSERTIONS

(?=…)        positive look ahead
(?!…)        negative look ahead
(?<=…)       positive look behind
(?<!…)       negative look behind

Each top-level branch of a look behind must be of a fixed length.

BACKREFERENCES

\n             reference by number (can be ambiguous)
\gn            reference by number
\g{n}          reference by number
\g{-n}         relative reference by number
\k<name>       reference by name (Perl)
\k’name’       reference by name (Perl)
\g{name}       reference by name (Perl)
\k{name}       reference by name (.NET)
(?P=name)      reference by name (Python)

SUBROUTINE REFERENCES (POSSIBLY RECURSIVE)

(?R)           recurse whole pattern
(?n)           call subpattern by absolute number
(?+n)          call subpattern by relative number
(?-n)          call subpattern by relative number
(?&name)       call subpattern by name (Perl)
(?P>name)      call subpattern by name (Python)

CONDITIONAL PATTERNS

(?(condition)yes-pattern)
(?(condition)yes-pattern|no-pattern)

(?(n)…       absolute reference condition
(?(+n)…      relative reference condition
(?(-n)…      relative reference condition
(?(<name>)…  named reference condition (Perl)
(?(‘name’)…  named reference condition (Perl)
(?(name)…    named reference condition (PCRE)
(?(R)…       overall recursion condition
(?(Rn)…      specific group recursion condition
(?(R&name)…  specific recursion condition
(?(DEFINE)…  define subpattern for reference
(?(assert)…  assertion condition

BACKTRACKING CONTROL

The following act immediately they are reached:

(*ACCEPT)      force successful match
(*FAIL)        force backtrack; synonym (*F)

The following act only when a subsequent match failure causes a backtrack to reach them. They all force a match failure, but they differ in what happens afterwards. Those that advance the start-of-match point do so only if the pattern is not anchored.

(*COMMIT)      overall failure, no advance of starting point
(*PRUNE)       advance to next starting character
(*SKIP)        advance start to current matching position
(*THEN)        local failure, backtrack to next alternation

NEWLINE CONVENTIONS

These are recognized only at the very start of the pattern or after a (*BSR_…) option.

(*CR)
(*LF)
(*CRLF)
(*ANYCRLF)
(*ANY)

WHAT \R MATCHES

These are recognized only at the very start of the pattern or after a (*…) option that sets the newline convention.

(*BSR_ANYCRLF)
(*BSR_UNICODE)

CALLOUTS

(?C)      callout
(?Cn)     callout with data n

SEE ALSO

pcrepattern(3), pcreapi(3), pcrecallout(3), pcrematching(3), pcre(3).

AUTHOR

Philip Hazel
University Computing Service
Cambridge CB2 3QH, England.

REVISION

Last updated: 21 September 2007
Copyright © 1997-2007 University of Cambridge.