1 .\" $NetBSD: re_format.7,v 1.9 2009/04/21 14:46:02 joerg Exp $
3 .\" Copyright (c) 1992, 1993, 1994
4 .\" The Regents of the University of California. All rights reserved.
6 .\" This code is derived from software contributed to Berkeley by
9 .\" Redistribution and use in source and binary forms, with or without
10 .\" modification, are permitted provided that the following conditions
12 .\" 1. Redistributions of source code must retain the above copyright
13 .\" notice, this list of conditions and the following disclaimer.
14 .\" 2. Redistributions in binary form must reproduce the above copyright
15 .\" notice, this list of conditions and the following disclaimer in the
16 .\" documentation and/or other materials provided with the distribution.
17 .\" 3. Neither the name of the University nor the names of its contributors
18 .\" may be used to endorse or promote products derived from this software
19 .\" without specific prior written permission.
21 .\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
22 .\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
23 .\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
24 .\" ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
25 .\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
26 .\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
27 .\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
28 .\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
29 .\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
30 .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
33 .\" Copyright (c) 1992, 1993, 1994 Henry Spencer.
35 .\" This code is derived from software contributed to Berkeley by
38 .\" Redistribution and use in source and binary forms, with or without
39 .\" modification, are permitted provided that the following conditions
41 .\" 1. Redistributions of source code must retain the above copyright
42 .\" notice, this list of conditions and the following disclaimer.
43 .\" 2. Redistributions in binary form must reproduce the above copyright
44 .\" notice, this list of conditions and the following disclaimer in the
45 .\" documentation and/or other materials provided with the distribution.
46 .\" 3. All advertising materials mentioning features or use of this software
47 .\" must display the following acknowledgement:
48 .\" This product includes software developed by the University of
49 .\" California, Berkeley and its contributors.
50 .\" 4. Neither the name of the University nor the names of its contributors
51 .\" may be used to endorse or promote products derived from this software
52 .\" without specific prior written permission.
54 .\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
55 .\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
56 .\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
57 .\" ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
58 .\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
59 .\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
60 .\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
61 .\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
62 .\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
63 .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
66 .\" @(#)re_format.7 8.3 (Berkeley) 3/20/94
73 .Nd POSIX 1003.2 regular expressions
75 Regular expressions (``RE''s),
76 as defined in POSIX 1003.2, come in two forms:
77 modern REs (roughly those of
79 1003.2 calls these ``extended'' REs)
80 and obsolete REs (roughly those of
82 1003.2 ``basic'' REs).
83 Obsolete REs mostly exist for backward compatibility in some old programs;
84 they will be discussed at the end.
85 1003.2 leaves some aspects of RE syntax and semantics open;
86 `\(dg' marks decisions on these aspects that
87 may not be fully portable to other 1003.2 implementations.
89 A (modern) RE is one\(dg or more non-empty\(dg
92 It matches anything that matches one of the branches.
94 A branch is one\(dg or more
97 It matches a match for the first, followed by a match for the second, etc.
102 by a single\(dg `*', `+', `?', or
104 An atom followed by `*' matches a sequence of 0 or more matches of the atom.
105 An atom followed by `+' matches a sequence of 1 or more matches of the atom.
106 An atom followed by `?' matches a sequence of 0 or 1 matches of the atom.
110 is `{' followed by an unsigned decimal integer, possibly followed by `,'
111 possibly followed by another unsigned decimal integer,
112 always followed by `}'.
113 The integers must lie between 0 and RE_DUP_MAX (255\(dg) inclusive,
114 and if there are two of them, the first may not exceed the second.
115 An atom followed by a bound containing one integer
117 and no comma matches a sequence of exactly
120 An atom followed by a bound containing one integer
122 and a comma matches a sequence of
124 or more matches of the atom.
125 An atom followed by a bound containing two integers
129 matches a sequence of
133 (inclusive) matches of the atom.
135 An atom is a regular expression enclosed in `()' (matching a match for the
136 regular expression), an empty set of `()' (matching the null string)\(dg, a
137 .Em bracket expression
138 (see below), `.' (matching any single character),
139 `^' (matching the null string at the beginning of a line),
140 `$' (matching the null string at the end of a line),
141 a `\e' followed by one of the characters `^.[$()|*+?{\e'
142 (matching that character taken as an ordinary character),
143 a `\e' followed by any other character\(dg
144 (matching that character taken as an ordinary character,
145 as if the `\e' had not been present\(dg),
146 or a single character with no other significance (matching that character).
147 A `{' followed by a character other than a digit is an ordinary
148 character, not the beginning of a bound\(dg.
149 It is illegal to end an RE with `\e'.
152 .Em bracket expression
153 is a list of characters enclosed in `[]'.
154 It normally matches any single character from the list (but see below).
155 If the list begins with `^',
156 it matches any single character (but see below)
158 from the rest of the list.
159 If two characters in the list are separated by `\-', this is shorthand
162 of characters between those two (inclusive) in the collating sequence,
163 e.g. `[0-9]' in ASCII matches any decimal digit.
164 It is illegal\(dg for two ranges to share an endpoint, e.g. `a-c-e'.
165 Ranges are very collating-sequence-dependent,
166 and portable programs should avoid relying on them.
168 To include a literal `]' in the list, make it the first character
169 (following a possible `^').
170 To include a literal `\-', make it the first or last character,
171 or the second endpoint of a range.
172 To use a literal `\-' as the first endpoint of a range,
173 enclose it in `[.' and `.]' to make it a collating element (see below).
174 With the exception of these and some combinations using `[' (see next
175 paragraphs), all other special characters, including `\e', lose their
176 special significance within a bracket expression.
178 Within a bracket expression, a collating element (a character,
179 a multi-character sequence that collates as if it were a single character,
180 or a collating-sequence name for either)
181 enclosed in `[.' and `.]' stands for the
182 sequence of characters of that collating element.
183 The sequence is a single element of the bracket expression's list.
184 A bracket expression containing a multi-character collating element
185 can thus match more than one character,
186 e.g. if the collating sequence includes a `ch' collating element,
187 then the RE `[[.ch.]]*c' matches the first five characters
190 Within a bracket expression, a collating element enclosed in `[=' and
191 `=]' is an equivalence class, standing for the sequences of characters
192 of all collating elements equivalent to that one, including itself.
193 (If there are no other equivalent collating elements,
194 the treatment is as if the enclosing delimiters were `[.' and `.]'.)
195 For example, if o and '\(^o' are the members of an equivalence class,
196 then `[[=o=]]', `[[=\(^o'=]]', and `[o\(^o']' are all synonymous.
197 An equivalence class may not\(dg be an endpoint
200 Within a bracket expression, the name of a
202 enclosed in `[:' and `:]' stands for the list of all characters
203 belonging to that class.
204 Standard character class names are:
205 .Bl -column "alnum" "digit" "xdigit"
206 .It alnum digit punct
207 .It alpha graph space
208 .It blank lower upper
209 .It cntrl print xdigit
212 These stand for the character classes defined in
214 A locale may provide others.
215 A character class may not be used as an endpoint of a range.
217 There are two special cases\(dg of bracket expressions:
218 the bracket expressions `[[:\*[Lt]:]]' and `[[:\*[Gt]:]]' match
219 the null string at the beginning and end of a word respectively.
220 A word is defined as a sequence of word characters
221 which is neither preceded nor followed by word characters.
222 A word character is an
224 character (as defined by
227 This is an extension, compatible with but not specified by POSIX 1003.2,
228 and should be used with caution in software intended to be portable
231 In the event that an RE could match more than one substring of a given
232 string, the RE matches the one starting earliest in the string.
233 If the RE could match more than one substring starting at that point,
234 it matches the longest.
235 Subexpressions also match the longest possible substrings, subject to
236 the constraint that the whole match be as long as possible,
237 with subexpressions starting earlier in the RE taking priority over
239 Note that higher-level subexpressions thus take priority over
240 their lower-level component subexpressions.
242 Match lengths are measured in characters, not collating elements.
243 A null string is considered longer than no match at all.
245 `bb*' matches the three middle characters of `abbbc',
246 `(wee|week)(knights|nights)' matches all ten characters of `weeknights',
247 when `(.*).*' is matched against `abc' the parenthesized subexpression
248 matches all three characters, and
249 when `(a*)*' is matched against `bc' both the whole RE and the parenthesized
250 subexpression match the null string.
252 If case-independent matching is specified,
253 the effect is much as if all case distinctions had vanished from the
255 When an alphabetic that exists in multiple cases appears as an
256 ordinary character outside a bracket expression, it is effectively
257 transformed into a bracket expression containing both cases,
258 e.g. `x' becomes `[xX]'.
259 When it appears inside a bracket expression, all case counterparts
260 of it are added to the bracket expression, so that (e.g.) `[x]'
261 becomes `[xX]' and `[^x]' becomes `[^xX]'.
263 No particular limit is imposed on the length of REs\(dg.
264 Programs intended to be portable should not employ REs longer
266 as an implementation can refuse to accept such REs and remain
269 Obsolete (``basic'') regular expressions differ in several respects.
270 `|', `+', and `?' are ordinary characters and there is no equivalent
271 for their functionality.
272 The delimiters for bounds are `\e{' and `\e}',
273 with `{' and `}' by themselves ordinary characters.
274 The parentheses for nested subexpressions are `\e(' and `\e)',
275 with `(' and `)' by themselves ordinary characters.
276 `^' is an ordinary character except at the beginning of the
277 RE or\(dg the beginning of a parenthesized subexpression,
278 `$' is an ordinary character except at the end of the
279 RE or\(dg the end of a parenthesized subexpression,
280 and `*' is an ordinary character if it appears at the beginning of the
281 RE or the beginning of a parenthesized subexpression
282 (after a possible leading `^').
283 Finally, there is one new type of atom, a
285 `\e' followed by a non-zero decimal digit
287 matches the same sequence of characters
289 .Em d Ns th parenthesized subexpression
290 (numbering subexpressions by the positions of their opening parentheses,
292 so that (e.g.) `\e([bc]\e)\e1' matches `bb' or `cc' but not `bc'.
296 POSIX 1003.2, section 2.8 (Regular Expression Notation).
298 Having two kinds of REs is a botch.
300 The current 1003.2 spec says that `)' is an ordinary character in
301 the absence of an unmatched `(';
302 this was an unintentional result of a wording error, and change is likely.
305 Back references are a dreadful botch,
306 posing major problems for efficient implementations.
307 They are also somewhat vaguely defined
308 (does `a\e(\e(b\e)*\e2\e)*d' match `abbbd'?).
311 1003.2's specification of case-independent matching is vague.
312 The ``one case implies all cases'' definition given above
313 is current consensus among implementors as to the right interpretation.
315 The syntax for word boundaries is incredibly ugly.