Debian Codegraph @

upstream/debian-unstable/man1/perlrecharclass.1

raw
1 .\" -*- mode: troff; coding: utf-8 -*-
2 .\" Automatically generated by Pod::Man 5.0102 (Pod::Simple 3.45)
3 .\"
4 .\" Standard preamble:
5 .\" ========================================================================
6 .de Sp \" Vertical space (when we can't use .PP)
7 .if t .sp .5v
8 .if n .sp
9 ..
10 .de Vb \" Begin verbatim text
11 .ft CW
12 .nf
13 .ne \\$1
14 ..
15 .de Ve \" End verbatim text
16 .ft R
17 .fi
18 ..
19 .\" \*(C` and \*(C' are quotes in nroff, nothing in troff, for use with C<>.
20 .ie n \{\
21 . ds C` ""
22 . ds C' ""
23 'br\}
24 .el\{\
25 . ds C`
26 . ds C'
27 'br\}
28 .\"
29 .\" Escape single quotes in literal strings from groff's Unicode transform.
30 .ie \n(.g .ds Aq \(aq
31 .el .ds Aq '
32 .\"
33 .\" If the F register is >0, we'll generate index entries on stderr for
34 .\" titles (.TH), headers (.SH), subsections (.SS), items (.Ip), and index
35 .\" entries marked with X<> in POD. Of course, you'll have to process the
36 .\" output yourself in some meaningful fashion.
37 .\"
38 .\" Avoid warning from groff about undefined register 'F'.
39 .de IX
40 ..
41 .nr rF 0
42 .if \n(.g .if rF .nr rF 1
43 .if (\n(rF:(\n(.g==0)) \{\
44 . if \nF \{\
45 . de IX
46 . tm Index:\\$1\t\\n%\t"\\$2"
47 ..
48 . if !\nF==2 \{\
49 . nr % 0
50 . nr F 2
51 . \}
52 . \}
53 .\}
54 .rr rF
55 .\" ========================================================================
56 .\"
57 .IX Title "PERLRECHARCLASS 1"
58 .TH PERLRECHARCLASS 1 2025-11-16 "perl v5.40.1" "Perl Programmers Reference Guide"
59 .\" For nroff, turn off justification. Always turn off hyphenation; it makes
60 .\" way too many mistakes in technical documents.
61 .if n .ad l
62 .nh
63 .SH NAME
64 perlrecharclass \- Perl Regular Expression Character Classes
65 .IX Xref "character class"
66 .SH DESCRIPTION
67 .IX Header "DESCRIPTION"
68 The top level documentation about Perl regular expressions
69 is found in perlre.
70 .PP
71 This manual page discusses the syntax and use of character
72 classes in Perl regular expressions.
73 .PP
74 A character class is a way of denoting a set of characters
75 in such a way that one character of the set is matched.
76 It's important to remember that: matching a character class
77 consumes exactly one character in the source string. (The source
78 string is the string the regular expression is matched against.)
79 .PP
80 There are three types of character classes in Perl regular
81 expressions: the dot, backslash sequences, and the form enclosed in square
82 brackets. Keep in mind, though, that often the term "character class" is used
83 to mean just the bracketed form. Certainly, most Perl documentation does that.
84 .SS "The dot"
85 .IX Subsection "The dot"
86 The dot (or period), \f(CW\*(C`.\*(C'\fR is probably the most used, and certainly
87 the most well-known character class. By default, a dot matches any
88 character, except for the newline. That default can be changed to
89 add matching the newline by using the \fIsingle line\fR modifier:
90 for the entire regular expression with the \f(CW\*(C`/s\*(C'\fR modifier, or
91 locally with \f(CW\*(C`(?s)\*(C'\fR (and even globally within the scope of
92 \&\f(CW\*(C`use re \*(Aq/s\*(Aq\*(C'\fR). (The \f(CW"\eN"\fR backslash
93 sequence, described
94 below, matches any character except newline without regard to the
95 \&\fIsingle line\fR modifier.)
96 .PP
97 Here are some examples:
98 .PP
99 .Vb 7
100 \& "a" =~ /./ # Match
101 \& "." =~ /./ # Match
102 \& "" =~ /./ # No match (dot has to match a character)
103 \& "\en" =~ /./ # No match (dot does not match a newline)
104 \& "\en" =~ /./s # Match (global \*(Aqsingle line\*(Aq modifier)
105 \& "\en" =~ /(?s:.)/ # Match (local \*(Aqsingle line\*(Aq modifier)
106 \& "ab" =~ /^.$/ # No match (dot matches one character)
107 .Ve
108 .SS "Backslash sequences"
109 .IX Xref "\\w \\W \\s \\S \\d \\D \\p \\P \\N \\v \\V \\h \\H word whitespace"
110 .IX Subsection "Backslash sequences"
111 A backslash sequence is a sequence of characters, the first one of which is a
112 backslash. Perl ascribes special meaning to many such sequences, and some of
113 these are character classes. That is, they match a single character each,
114 provided that the character belongs to the specific set of characters defined
115 by the sequence.
116 .PP
117 Here's a list of the backslash sequences that are character classes. They
118 are discussed in more detail below. (For the backslash sequences that aren't
119 character classes, see perlrebackslash.)
120 .PP
121 .Vb 10
122 \& \ed Match a decimal digit character.
123 \& \eD Match a non\-decimal\-digit character.
124 \& \ew Match a "word" character.
125 \& \eW Match a non\-"word" character.
126 \& \es Match a whitespace character.
127 \& \eS Match a non\-whitespace character.
128 \& \eh Match a horizontal whitespace character.
129 \& \eH Match a character that isn\*(Aqt horizontal whitespace.
130 \& \ev Match a vertical whitespace character.
131 \& \eV Match a character that isn\*(Aqt vertical whitespace.
132 \& \eN Match a character that isn\*(Aqt a newline.
133 \& \epP, \ep{Prop} Match a character that has the given Unicode property.
134 \& \ePP, \eP{Prop} Match a character that doesn\*(Aqt have the Unicode property
135 .Ve
136 .PP
137 \fI\eN\fR
138 .IX Subsection "N"
139 .PP
140 \&\f(CW\*(C`\eN\*(C'\fR, available starting in v5.12, like the dot, matches any
141 character that is not a newline. The difference is that \f(CW\*(C`\eN\*(C'\fR is not influenced
142 by the \fIsingle line\fR regular expression modifier (see "The dot" above). Note
143 that the form \f(CW\*(C`\eN{...}\*(C'\fR may mean something completely different. When the
144 \&\f(CW\*(C`{...}\*(C'\fR is a quantifier, it means to match a non-newline
145 character that many times. For example, \f(CW\*(C`\eN{3}\*(C'\fR means to match 3
146 non-newlines; \f(CW\*(C`\eN{5,}\*(C'\fR means to match 5 or more non-newlines. But if \f(CW\*(C`{...}\*(C'\fR
147 is not a legal quantifier, it is presumed to be a named character. See
148 charnames for those. For example, none of \f(CW\*(C`\eN{COLON}\*(C'\fR, \f(CW\*(C`\eN{4F}\*(C'\fR, and
149 \&\f(CW\*(C`\eN{F4}\*(C'\fR contain legal quantifiers, so Perl will try to find characters whose
150 names are respectively \f(CW\*(C`COLON\*(C'\fR, \f(CW\*(C`4F\*(C'\fR, and \f(CW\*(C`F4\*(C'\fR.
151 .PP
152 \fIDigits\fR
153 .IX Subsection "Digits"
154 .PP
155 \&\f(CW\*(C`\ed\*(C'\fR matches a single character considered to be a decimal \fIdigit\fR.
156 If the \f(CW\*(C`/a\*(C'\fR regular expression modifier is in effect, it matches [0\-9].
157 Otherwise, it
158 matches anything that is matched by \f(CW\*(C`\ep{Digit}\*(C'\fR, which includes [0\-9].
159 (An unlikely possible exception is that under locale matching rules, the
160 current locale might not have \f(CW\*(C`[0\-9]\*(C'\fR matched by \f(CW\*(C`\ed\*(C'\fR, and/or might match
161 other characters whose code point is less than 256. The only such locale
162 definitions that are legal would be to match \f(CW\*(C`[0\-9]\*(C'\fR plus another set of
163 10 consecutive digit characters; anything else would be in violation of
164 the C language standard, but Perl doesn't currently assume anything in
165 regard to this.)
166 .PP
167 What this means is that unless the \f(CW\*(C`/a\*(C'\fR modifier is in effect \f(CW\*(C`\ed\*(C'\fR not
168 only matches the digits '0' \- '9', but also Arabic, Devanagari, and
169 digits from other languages. This may cause some confusion, and some
170 security issues.
171 .PP
172 Some digits that \f(CW\*(C`\ed\*(C'\fR matches look like some of the [0\-9] ones, but
173 have different values. For example, BENGALI DIGIT FOUR (U+09EA) looks
174 very much like an ASCII DIGIT EIGHT (U+0038), and LEPCHA DIGIT SIX
175 (U+1C46) looks very much like an ASCII DIGIT FIVE (U+0035). An
176 application that
177 is expecting only the ASCII digits might be misled, or if the match is
178 \&\f(CW\*(C`\ed+\*(C'\fR, the matched string might contain a mixture of digits from
179 different writing systems that look like they signify a number different
180 than they actually do. "\fBnum()\fR" in Unicode::UCD can
181 be used to safely
182 calculate the value, returning \f(CW\*(C`undef\*(C'\fR if the input string contains
183 such a mixture. Otherwise, for example, a displayed price might be
184 deliberately different than it appears.
185 .PP
186 What \f(CW\*(C`\ep{Digit}\*(C'\fR means (and hence \f(CW\*(C`\ed\*(C'\fR except under the \f(CW\*(C`/a\*(C'\fR
187 modifier) is \f(CW\*(C`\ep{General_Category=Decimal_Number}\*(C'\fR, or synonymously,
188 \&\f(CW\*(C`\ep{General_Category=Digit}\*(C'\fR. Starting with Unicode version 4.1, this
189 is the same set of characters matched by \f(CW\*(C`\ep{Numeric_Type=Decimal}\*(C'\fR.
190 But Unicode also has a different property with a similar name,
191 \&\f(CW\*(C`\ep{Numeric_Type=Digit}\*(C'\fR, which matches a completely different set of
192 characters. These characters are things such as \f(CW\*(C`CIRCLED DIGIT ONE\*(C'\fR
193 or subscripts, or are from writing systems that lack all ten digits.
194 .PP
195 The design intent is for \f(CW\*(C`\ed\*(C'\fR to exactly match the set of characters
196 that can safely be used with "normal" big-endian positional decimal
197 syntax, where, for example 123 means one 'hundred', plus two 'tens',
198 plus three 'ones'. This positional notation does not necessarily apply
199 to characters that match the other type of "digit",
200 \&\f(CW\*(C`\ep{Numeric_Type=Digit}\*(C'\fR, and so \f(CW\*(C`\ed\*(C'\fR doesn't match them.
201 .PP
202 The Tamil digits (U+0BE6 \- U+0BEF) can also legally be
203 used in old-style Tamil numbers in which they would appear no more than
204 one in a row, separated by characters that mean "times 10", "times 100",
205 etc. (See <https://www.unicode.org/notes/tn21>.)
206 .PP
207 Any character not matched by \f(CW\*(C`\ed\*(C'\fR is matched by \f(CW\*(C`\eD\*(C'\fR.
208 .PP
209 \fIWord characters\fR
210 .IX Subsection "Word characters"
211 .PP
212 A \f(CW\*(C`\ew\*(C'\fR matches a single alphanumeric character (an alphabetic character, or a
213 decimal digit); or a connecting punctuation character, such as an
214 underscore ("_"); or a "mark" character (like some sort of accent) that
215 attaches to one of those. It does not match a whole word. To match a
216 whole word, use \f(CW\*(C`\ew+\*(C'\fR. This isn't the same thing as matching an
217 English word, but in the ASCII range it is the same as a string of
218 Perl-identifier characters.
219 .ie n .IP "If the ""/a"" modifier is in effect ..." 4
220 .el .IP "If the \f(CW/a\fR modifier is in effect ..." 4
221 .IX Item "If the /a modifier is in effect ..."
222 \&\f(CW\*(C`\ew\*(C'\fR matches the 63 characters [a\-zA\-Z0\-9_].
223 .IP "otherwise ..." 4
224 .IX Item "otherwise ..."
225 .RS 4
226 .PD 0
227 .IP "For code points above 255 ..." 4
228 .IX Item "For code points above 255 ..."
229 .PD
230 \&\f(CW\*(C`\ew\*(C'\fR matches the same as \f(CW\*(C`\ep{Word}\*(C'\fR matches in this range. That is,
231 it matches Thai letters, Greek letters, etc. This includes connector
232 punctuation (like the underscore) which connect two words together, or
233 diacritics, such as a \f(CW\*(C`COMBINING TILDE\*(C'\fR and the modifier letters, which
234 are generally used to add auxiliary markings to letters.
235 .IP "For code points below 256 ..." 4
236 .IX Item "For code points below 256 ..."
237 .RS 4
238 .PD 0
239 .IP "if locale rules are in effect ..." 4
240 .IX Item "if locale rules are in effect ..."
241 .PD
242 \&\f(CW\*(C`\ew\*(C'\fR matches the platform's native underscore character plus whatever
243 the locale considers to be alphanumeric.
244 .IP "if, instead, Unicode rules are in effect ..." 4
245 .IX Item "if, instead, Unicode rules are in effect ..."
246 \&\f(CW\*(C`\ew\*(C'\fR matches exactly what \f(CW\*(C`\ep{Word}\*(C'\fR matches.
247 .IP "otherwise ..." 4
248 .IX Item "otherwise ..."
249 \&\f(CW\*(C`\ew\*(C'\fR matches [a\-zA\-Z0\-9_].
250 .RE
251 .RS 4
252 .RE
253 .RE
254 .RS 4
255 .RE
256 .PP
257 Which rules apply are determined as described in "Which character set modifier is in effect?" in perlre.
258 .PP
259 There are a number of security issues with the full Unicode list of word
260 characters. See <https://unicode.org/reports/tr36>.
261 .PP
262 Also, for a somewhat finer-grained set of characters that are in programming
263 language identifiers beyond the ASCII range, you may wish to instead use the
264 more customized "Unicode Properties", \f(CW\*(C`\ep{ID_Start}\*(C'\fR,
265 \&\f(CW\*(C`\ep{ID_Continue}\*(C'\fR, \f(CW\*(C`\ep{XID_Start}\*(C'\fR, and \f(CW\*(C`\ep{XID_Continue}\*(C'\fR. See
266 <https://unicode.org/reports/tr31>.
267 .PP
268 Any character not matched by \f(CW\*(C`\ew\*(C'\fR is matched by \f(CW\*(C`\eW\*(C'\fR.
269 .PP
270 \fIWhitespace\fR
271 .IX Subsection "Whitespace"
272 .PP
273 \&\f(CW\*(C`\es\*(C'\fR matches any single character considered whitespace.
274 .ie n .IP "If the ""/a"" modifier is in effect ..." 4
275 .el .IP "If the \f(CW/a\fR modifier is in effect ..." 4
276 .IX Item "If the /a modifier is in effect ..."
277 In all Perl versions, \f(CW\*(C`\es\*(C'\fR matches the 5 characters [\et\en\ef\er ]; that
278 is, the horizontal tab,
279 the newline, the form feed, the carriage return, and the space.
280 Starting in Perl v5.18, it also matches the vertical tab, \f(CW\*(C`\ecK\*(C'\fR.
281 See note \f(CW\*(C`[1]\*(C'\fR below for a discussion of this.
282 .IP "otherwise ..." 4
283 .IX Item "otherwise ..."
284 .RS 4
285 .PD 0
286 .IP "For code points above 255 ..." 4
287 .IX Item "For code points above 255 ..."
288 .PD
289 \&\f(CW\*(C`\es\*(C'\fR matches exactly the code points above 255 shown with an "s" column
290 in the table below.
291 .IP "For code points below 256 ..." 4
292 .IX Item "For code points below 256 ..."
293 .RS 4
294 .PD 0
295 .IP "if locale rules are in effect ..." 4
296 .IX Item "if locale rules are in effect ..."
297 .PD
298 \&\f(CW\*(C`\es\*(C'\fR matches whatever the locale considers to be whitespace.
299 .IP "if, instead, Unicode rules are in effect ..." 4
300 .IX Item "if, instead, Unicode rules are in effect ..."
301 \&\f(CW\*(C`\es\*(C'\fR matches exactly the characters shown with an "s" column in the
302 table below.
303 .IP "otherwise ..." 4
304 .IX Item "otherwise ..."
305 \&\f(CW\*(C`\es\*(C'\fR matches [\et\en\ef\er ] and, starting in Perl
306 v5.18, the vertical tab, \f(CW\*(C`\ecK\*(C'\fR.
307 (See note \f(CW\*(C`[1]\*(C'\fR below for a discussion of this.)
308 Note that this list doesn't include the non-breaking space.
309 .RE
310 .RS 4
311 .RE
312 .RE
313 .RS 4
314 .RE
315 .PP
316 Which rules apply are determined as described in "Which character set modifier is in effect?" in perlre.
317 .PP
318 Any character not matched by \f(CW\*(C`\es\*(C'\fR is matched by \f(CW\*(C`\eS\*(C'\fR.
319 .PP
320 \&\f(CW\*(C`\eh\*(C'\fR matches any character considered horizontal whitespace;
321 this includes the platform's space and tab characters and several others
322 listed in the table below. \f(CW\*(C`\eH\*(C'\fR matches any character
323 not considered horizontal whitespace. They use the platform's native
324 character set, and do not consider any locale that may otherwise be in
325 use.
326 .PP
327 \&\f(CW\*(C`\ev\*(C'\fR matches any character considered vertical whitespace;
328 this includes the platform's carriage return and line feed characters (newline)
329 plus several other characters, all listed in the table below.
330 \&\f(CW\*(C`\eV\*(C'\fR matches any character not considered vertical whitespace.
331 They use the platform's native character set, and do not consider any
332 locale that may otherwise be in use.
333 .PP
334 \&\f(CW\*(C`\eR\*(C'\fR matches anything that can be considered a newline under Unicode
335 rules. It can match a multi-character sequence. It cannot be used inside
336 a bracketed character class; use \f(CW\*(C`\ev\*(C'\fR instead (vertical whitespace).
337 It uses the platform's
338 native character set, and does not consider any locale that may
339 otherwise be in use.
340 Details are discussed in perlrebackslash.
341 .PP
342 Note that unlike \f(CW\*(C`\es\*(C'\fR (and \f(CW\*(C`\ed\*(C'\fR and \f(CW\*(C`\ew\*(C'\fR), \f(CW\*(C`\eh\*(C'\fR and \f(CW\*(C`\ev\*(C'\fR always match
343 the same characters, without regard to other factors, such as the active
344 locale or whether the source string is in UTF\-8 format.
345 .PP
346 One might think that \f(CW\*(C`\es\*(C'\fR is equivalent to \f(CW\*(C`[\eh\ev]\*(C'\fR. This is indeed true
347 starting in Perl v5.18, but prior to that, the sole difference was that the
348 vertical tab (\f(CW"\ecK"\fR) was not matched by \f(CW\*(C`\es\*(C'\fR.
349 .PP
350 The following table is a complete listing of characters matched by
351 \&\f(CW\*(C`\es\*(C'\fR, \f(CW\*(C`\eh\*(C'\fR and \f(CW\*(C`\ev\*(C'\fR as of Unicode 14.0.
352 .PP
353 The first column gives the Unicode code point of the character (in hex format),
354 the second column gives the (Unicode) name. The third column indicates
355 by which class(es) the character is matched (assuming no locale is in
356 effect that changes the \f(CW\*(C`\es\*(C'\fR matching).
357 .PP
358 .Vb 10
359 \& 0x0009 CHARACTER TABULATION h s
360 \& 0x000a LINE FEED (LF) vs
361 \& 0x000b LINE TABULATION vs [1]
362 \& 0x000c FORM FEED (FF) vs
363 \& 0x000d CARRIAGE RETURN (CR) vs
364 \& 0x0020 SPACE h s
365 \& 0x0085 NEXT LINE (NEL) vs [2]
366 \& 0x00a0 NO\-BREAK SPACE h s [2]
367 \& 0x1680 OGHAM SPACE MARK h s
368 \& 0x2000 EN QUAD h s
369 \& 0x2001 EM QUAD h s
370 \& 0x2002 EN SPACE h s
371 \& 0x2003 EM SPACE h s
372 \& 0x2004 THREE\-PER\-EM SPACE h s
373 \& 0x2005 FOUR\-PER\-EM SPACE h s
374 \& 0x2006 SIX\-PER\-EM SPACE h s
375 \& 0x2007 FIGURE SPACE h s
376 \& 0x2008 PUNCTUATION SPACE h s
377 \& 0x2009 THIN SPACE h s
378 \& 0x200a HAIR SPACE h s
379 \& 0x2028 LINE SEPARATOR vs
380 \& 0x2029 PARAGRAPH SEPARATOR vs
381 \& 0x202f NARROW NO\-BREAK SPACE h s
382 \& 0x205f MEDIUM MATHEMATICAL SPACE h s
383 \& 0x3000 IDEOGRAPHIC SPACE h s
384 .Ve
385 .IP [1] 4
386 .IX Item "[1]"
387 Prior to Perl v5.18, \f(CW\*(C`\es\*(C'\fR did not match the vertical tab.
388 \&\f(CW\*(C`[^\eS\ecK]\*(C'\fR (obscurely) matches what \f(CW\*(C`\es\*(C'\fR traditionally did.
389 .IP [2] 4
390 .IX Item "[2]"
391 NEXT LINE and NO-BREAK SPACE may or may not match \f(CW\*(C`\es\*(C'\fR depending
392 on the rules in effect. See
393 the beginning of this section.
394 .PP
395 \fIUnicode Properties\fR
396 .IX Subsection "Unicode Properties"
397 .PP
398 \&\f(CW\*(C`\epP\*(C'\fR and \f(CW\*(C`\ep{Prop}\*(C'\fR are character classes to match characters that fit given
399 Unicode properties. One letter property names can be used in the \f(CW\*(C`\epP\*(C'\fR form,
400 with the property name following the \f(CW\*(C`\ep\*(C'\fR, otherwise, braces are required.
401 When using braces, there is a single form, which is just the property name
402 enclosed in the braces, and a compound form which looks like \f(CW\*(C`\ep{name=value}\*(C'\fR,
403 which means to match if the property "name" for the character has that particular
404 "value".
405 For instance, a match for a number can be written as \f(CW\*(C`/\epN/\*(C'\fR or as
406 \&\f(CW\*(C`/\ep{Number}/\*(C'\fR, or as \f(CW\*(C`/\ep{Number=True}/\*(C'\fR.
407 Lowercase letters are matched by the property \fILowercase_Letter\fR which
408 has the short form \fILl\fR. They need the braces, so are written as \f(CW\*(C`/\ep{Ll}/\*(C'\fR or
409 \&\f(CW\*(C`/\ep{Lowercase_Letter}/\*(C'\fR, or \f(CW\*(C`/\ep{General_Category=Lowercase_Letter}/\*(C'\fR
410 (the underscores are optional).
411 \&\f(CW\*(C`/\epLl/\*(C'\fR is valid, but means something different.
412 It matches a two character string: a letter (Unicode property \f(CW\*(C`\epL\*(C'\fR),
413 followed by a lowercase \f(CW\*(C`l\*(C'\fR.
414 .PP
415 What a Unicode property matches is never subject to locale rules, and
416 if locale rules are not otherwise in effect, the use of a Unicode
417 property will force the regular expression into using Unicode rules, if
418 it isn't already.
419 .PP
420 Note that almost all properties are immune to case-insensitive matching.
421 That is, adding a \f(CW\*(C`/i\*(C'\fR regular expression modifier does not change what
422 they match. But there are two sets that are affected. The first set is
423 \&\f(CW\*(C`Uppercase_Letter\*(C'\fR,
424 \&\f(CW\*(C`Lowercase_Letter\*(C'\fR,
425 and \f(CW\*(C`Titlecase_Letter\*(C'\fR,
426 all of which match \f(CW\*(C`Cased_Letter\*(C'\fR under \f(CW\*(C`/i\*(C'\fR matching.
427 The second set is
428 \&\f(CW\*(C`Uppercase\*(C'\fR,
429 \&\f(CW\*(C`Lowercase\*(C'\fR,
430 and \f(CW\*(C`Titlecase\*(C'\fR,
431 all of which match \f(CW\*(C`Cased\*(C'\fR under \f(CW\*(C`/i\*(C'\fR matching.
432 (The difference between these sets is that some things, such as Roman
433 numerals, come in both upper and lower case, so they are \f(CW\*(C`Cased\*(C'\fR, but
434 aren't considered to be letters, so they aren't \f(CW\*(C`Cased_Letter\*(C'\fRs. They're
435 actually \f(CW\*(C`Letter_Number\*(C'\fRs.)
436 This set also includes its subsets \f(CW\*(C`PosixUpper\*(C'\fR and \f(CW\*(C`PosixLower\*(C'\fR, both
437 of which under \f(CW\*(C`/i\*(C'\fR match \f(CW\*(C`PosixAlpha\*(C'\fR.
438 .PP
439 For more details on Unicode properties, see "Unicode
440 Character Properties" in perlunicode; for a
441 complete list of possible properties, see
442 "Properties accessible through \ep{} and \eP{}" in perluniprops,
443 which notes all forms that have \f(CW\*(C`/i\*(C'\fR differences.
444 It is also possible to define your own properties. This is discussed in
445 "User-Defined Character Properties" in perlunicode.
446 .PP
447 Unicode properties are defined (surprise!) only on Unicode code points.
448 Starting in v5.20, when matching against \f(CW\*(C`\ep\*(C'\fR and \f(CW\*(C`\eP\*(C'\fR, Perl treats
449 non-Unicode code points (those above the legal Unicode maximum of
450 0x10FFFF) as if they were typical unassigned Unicode code points.
451 .PP
452 Prior to v5.20, Perl raised a warning and made all matches fail on
453 non-Unicode code points. This could be somewhat surprising:
454 .PP
455 .Vb 3
456 \& chr(0x110000) =~ \ep{ASCII_Hex_Digit=True} # Fails on Perls < v5.20.
457 \& chr(0x110000) =~ \ep{ASCII_Hex_Digit=False} # Also fails on Perls
458 \& # < v5.20
459 .Ve
460 .PP
461 Even though these two matches might be thought of as complements, until
462 v5.20 they were so only on Unicode code points.
463 .PP
464 Starting in perl v5.30, wildcards are allowed in Unicode property
465 values. See "Wildcards in Property Values" in perlunicode.
466 .PP
467 Examples
468 .IX Subsection "Examples"
469 .PP
470 .Vb 8
471 \& "a" =~ /\ew/ # Match, "a" is a \*(Aqword\*(Aq character.
472 \& "7" =~ /\ew/ # Match, "7" is a \*(Aqword\*(Aq character as well.
473 \& "a" =~ /\ed/ # No match, "a" isn\*(Aqt a digit.
474 \& "7" =~ /\ed/ # Match, "7" is a digit.
475 \& " " =~ /\es/ # Match, a space is whitespace.
476 \& "a" =~ /\eD/ # Match, "a" is a non\-digit.
477 \& "7" =~ /\eD/ # No match, "7" is not a non\-digit.
478 \& " " =~ /\eS/ # No match, a space is not non\-whitespace.
479 \&
480 \& " " =~ /\eh/ # Match, space is horizontal whitespace.
481 \& " " =~ /\ev/ # No match, space is not vertical whitespace.
482 \& "\er" =~ /\ev/ # Match, a return is vertical whitespace.
483 \&
484 \& "a" =~ /\epL/ # Match, "a" is a letter.
485 \& "a" =~ /\ep{Lu}/ # No match, /\ep{Lu}/ matches upper case letters.
486 \&
487 \& "\ex{0e0b}" =~ /\ep{Thai}/ # Match, \ex{0e0b} is the character
488 \& # \*(AqTHAI CHARACTER SO SO\*(Aq, and that\*(Aqs in
489 \& # Thai Unicode class.
490 \& "a" =~ /\eP{Lao}/ # Match, as "a" is not a Laotian character.
491 .Ve
492 .PP
493 It is worth emphasizing that \f(CW\*(C`\ed\*(C'\fR, \f(CW\*(C`\ew\*(C'\fR, etc, match single characters, not
494 complete numbers or words. To match a number (that consists of digits),
495 use \f(CW\*(C`\ed+\*(C'\fR; to match a word, use \f(CW\*(C`\ew+\*(C'\fR. But be aware of the security
496 considerations in doing so, as mentioned above.
497 .SS "Bracketed Character Classes"
498 .IX Subsection "Bracketed Character Classes"
499 The third form of character class you can use in Perl regular expressions
500 is the bracketed character class. In its simplest form, it lists the characters
501 that may be matched, surrounded by square brackets, like this: \f(CW\*(C`[aeiou]\*(C'\fR.
502 This matches one of \f(CW\*(C`a\*(C'\fR, \f(CW\*(C`e\*(C'\fR, \f(CW\*(C`i\*(C'\fR, \f(CW\*(C`o\*(C'\fR or \f(CW\*(C`u\*(C'\fR. Like the other
503 character classes, exactly one character is matched.* To match
504 a longer string consisting of characters mentioned in the character
505 class, follow the character class with a quantifier. For
506 instance, \f(CW\*(C`[aeiou]+\*(C'\fR matches one or more lowercase English vowels.
507 .PP
508 Repeating a character in a character class has no
509 effect; it's considered to be in the set only once.
510 .PP
511 Examples:
512 .PP
513 .Vb 5
514 \& "e" =~ /[aeiou]/ # Match, as "e" is listed in the class.
515 \& "p" =~ /[aeiou]/ # No match, "p" is not listed in the class.
516 \& "ae" =~ /^[aeiou]$/ # No match, a character class only matches
517 \& # a single character.
518 \& "ae" =~ /^[aeiou]+$/ # Match, due to the quantifier.
519 \&
520 \& \-\-\-\-\-\-\-
521 .Ve
522 .PP
523 * There are two exceptions to a bracketed character class matching a
524 single character only. Each requires special handling by Perl to make
525 things work:
526 .IP \(bu 4
527 When the class is to match caselessly under \f(CW\*(C`/i\*(C'\fR matching rules, and a
528 character that is explicitly mentioned inside the class matches a
529 multiple-character sequence caselessly under Unicode rules, the class
530 will also match that sequence. For example, Unicode says that the
531 letter \f(CW\*(C`LATIN SMALL LETTER SHARP S\*(C'\fR should match the sequence \f(CW\*(C`ss\*(C'\fR
532 under \f(CW\*(C`/i\*(C'\fR rules. Thus,
533 .Sp
534 .Vb 2
535 \& \*(Aqss\*(Aq =~ /\eA\eN{LATIN SMALL LETTER SHARP S}\ez/i # Matches
536 \& \*(Aqss\*(Aq =~ /\eA[aeioust\eN{LATIN SMALL LETTER SHARP S}]\ez/i # Matches
537 .Ve
538 .Sp
539 For this to happen, the class must not be inverted (see "Negation")
540 and the character must be explicitly specified, and not be part of a
541 multi-character range (not even as one of its endpoints). ("Character
542 Ranges" will be explained shortly.) Therefore,
543 .Sp
544 .Vb 6
545 \& \*(Aqss\*(Aq =~ /\eA[\e0\-\ex{ff}]\ez/ui # Doesn\*(Aqt match
546 \& \*(Aqss\*(Aq =~ /\eA[\e0\-\eN{LATIN SMALL LETTER SHARP S}]\ez/ui # No match
547 \& \*(Aqss\*(Aq =~ /\eA[\exDF\-\exDF]\ez/ui # Matches on ASCII platforms, since
548 \& # \exDF is LATIN SMALL LETTER SHARP S,
549 \& # and the range is just a single
550 \& # element
551 .Ve
552 .Sp
553 Note that it isn't a good idea to specify these types of ranges anyway.
554 .IP \(bu 4
555 Some names known to \f(CW\*(C`\eN{...}\*(C'\fR refer to a sequence of multiple characters,
556 instead of the usual single character. When one of these is included in
557 the class, the entire sequence is matched. For example,
558 .Sp
559 .Vb 2
560 \& "\eN{TAMIL LETTER KA}\eN{TAMIL VOWEL SIGN AU}"
561 \& =~ / ^ [\eN{TAMIL SYLLABLE KAU}] $ /x;
562 .Ve
563 .Sp
564 matches, because \f(CW\*(C`\eN{TAMIL SYLLABLE KAU}\*(C'\fR is a named sequence
565 consisting of the two characters matched against. Like the other
566 instance where a bracketed class can match multiple characters, and for
567 similar reasons, the class must not be inverted, and the named sequence
568 may not appear in a range, even one where it is both endpoints. If
569 these happen, it is a fatal error if the character class is within the
570 scope of \f(CW\*(C`use re \*(Aqstrict\*(C'\fR, or within an extended
571 \&\f(CW\*(C`(?[...])\*(C'\fR class; otherwise
572 only the first code point is used (with a \f(CW\*(C`regexp\*(C'\fR\-type warning
573 raised).
574 .PP
575 \fISpecial Characters Inside a Bracketed Character Class\fR
576 .IX Subsection "Special Characters Inside a Bracketed Character Class"
577 .PP
578 Most characters that are meta characters in regular expressions (that
579 is, characters that carry a special meaning like \f(CW\*(C`.\*(C'\fR, \f(CW\*(C`*\*(C'\fR, or \f(CW\*(C`(\*(C'\fR) lose
580 their special meaning and can be used inside a character class without
581 the need to escape them. For instance, \f(CW\*(C`[()]\*(C'\fR matches either an opening
582 parenthesis, or a closing parenthesis, and the parens inside the character
583 class don't group or capture. Be aware that, unless the pattern is
584 evaluated in single-quotish context, variable interpolation will take
585 place before the bracketed class is parsed:
586 .PP
587 .Vb 6
588 \& $, = "\et| ";
589 \& $x =~ m\*(Aq[$,]\*(Aq; # single\-quotish: matches \*(Aq$\*(Aq or \*(Aq,\*(Aq
590 \& $x =~ q{[$,]}\*(Aq # same
591 \& $x =~ m/[$,]/; # double\-quotish: Because we made an
592 \& # assignment to $, above, this now
593 \& # matches "\et", "|", or " "
594 .Ve
595 .PP
596 Characters that may carry a special meaning inside a character class are:
597 \&\f(CW\*(C`\e\*(C'\fR, \f(CW\*(C`^\*(C'\fR, \f(CW\*(C`\-\*(C'\fR, \f(CW\*(C`[\*(C'\fR and \f(CW\*(C`]\*(C'\fR, and are discussed below. They can be
598 escaped with a backslash, although this is sometimes not needed, in which
599 case the backslash may be omitted.
600 .PP
601 The sequence \f(CW\*(C`\eb\*(C'\fR is special inside a bracketed character class. While
602 outside the character class, \f(CW\*(C`\eb\*(C'\fR is an assertion indicating a point
603 that does not have either two word characters or two non-word characters
604 on either side, inside a bracketed character class, \f(CW\*(C`\eb\*(C'\fR matches a
605 backspace character.
606 .PP
607 The sequences
608 \&\f(CW\*(C`\ea\*(C'\fR,
609 \&\f(CW\*(C`\ec\*(C'\fR,
610 \&\f(CW\*(C`\ee\*(C'\fR,
611 \&\f(CW\*(C`\ef\*(C'\fR,
612 \&\f(CW\*(C`\en\*(C'\fR,
613 \&\f(CW\*(C`\eN{\fR\f(CINAME\fR\f(CW}\*(C'\fR,
614 \&\f(CW\*(C`\eN{U+\fR\f(CIhex char\fR\f(CW}\*(C'\fR,
615 \&\f(CW\*(C`\er\*(C'\fR,
616 \&\f(CW\*(C`\et\*(C'\fR,
617 and
618 \&\f(CW\*(C`\ex\*(C'\fR
619 are also special and have the same meanings as they do outside a
620 bracketed character class.
621 .PP
622 Also, a backslash followed by two or three octal digits is considered an octal
623 number.
624 .PP
625 A \f(CW\*(C`[\*(C'\fR is not special inside a character class, unless it's the start of a
626 POSIX character class (see "POSIX Character Classes" below). It normally does
627 not need escaping.
628 .PP
629 A \f(CW\*(C`]\*(C'\fR is normally either the end of a POSIX character class (see
630 "POSIX Character Classes" below), or it signals the end of the bracketed
631 character class. If you want to include a \f(CW\*(C`]\*(C'\fR in the set of characters, you
632 must generally escape it.
633 .PP
634 However, if the \f(CW\*(C`]\*(C'\fR is the \fIfirst\fR (or the second if the first
635 character is a caret) character of a bracketed character class, it
636 does not denote the end of the class (as you cannot have an empty class)
637 and is considered part of the set of characters that can be matched without
638 escaping.
639 .PP
640 Examples:
641 .PP
642 .Vb 8
643 \& "+" =~ /[+?*]/ # Match, "+" in a character class is not special.
644 \& "\ecH" =~ /[\eb]/ # Match, \eb inside in a character class
645 \& # is equivalent to a backspace.
646 \& "]" =~ /[][]/ # Match, as the character class contains
647 \& # both [ and ].
648 \& "[]" =~ /[[]]/ # Match, the pattern contains a character class
649 \& # containing just [, and the character class is
650 \& # followed by a ].
651 .Ve
652 .PP
653 \fIBracketed Character Classes and the \fR\f(CI\*(C`/xx\*(C'\fR\fI pattern modifier\fR
654 .IX Subsection "Bracketed Character Classes and the /xx pattern modifier"
655 .PP
656 Normally SPACE and TAB characters have no special meaning inside a
657 bracketed character class; they are just added to the list of characters
658 matched by the class. But if the \f(CW\*(C`/xx\*(C'\fR
659 pattern modifier is in effect, they are generally ignored and can be
660 added to improve readability. They can't be added in the middle of a
661 single construct:
662 .PP
663 .Vb 1
664 \& / [ \ex{10 FFFF} ] /xx # WRONG!
665 .Ve
666 .PP
667 The SPACE in the middle of the hex constant is illegal.
668 .PP
669 To specify a literal SPACE character, you can escape it with a
670 backslash, like:
671 .PP
672 .Vb 1
673 \& /[ a e i o u \e ]/xx
674 .Ve
675 .PP
676 This matches the English vowels plus the SPACE character.
677 .PP
678 For clarity, you should already have been using \f(CW\*(C`\et\*(C'\fR to specify a
679 literal tab, and \f(CW\*(C`\et\*(C'\fR is unaffected by \f(CW\*(C`/xx\*(C'\fR.
680 .PP
681 \fICharacter Ranges\fR
682 .IX Subsection "Character Ranges"
683 .PP
684 It is not uncommon to want to match a range of characters. Luckily, instead
685 of listing all characters in the range, one may use the hyphen (\f(CW\*(C`\-\*(C'\fR).
686 If inside a bracketed character class you have two characters separated
687 by a hyphen, it's treated as if all characters between the two were in
688 the class. For instance, \f(CW\*(C`[0\-9]\*(C'\fR matches any ASCII digit, and \f(CW\*(C`[a\-m]\*(C'\fR
689 matches any lowercase letter from the first half of the ASCII alphabet.
690 .PP
691 Note that the two characters on either side of the hyphen are not
692 necessarily both letters or both digits. Any character is possible,
693 although not advisable. \f(CW\*(C`[\*(Aq\-?]\*(C'\fR contains a range of characters, but
694 most people will not know which characters that means. Furthermore,
695 such ranges may lead to portability problems if the code has to run on
696 a platform that uses a different character set, such as EBCDIC.
697 .PP
698 If a hyphen in a character class cannot syntactically be part of a range, for
699 instance because it is the first or the last character of the character class,
700 or if it immediately follows a range, the hyphen isn't special, and so is
701 considered a character to be matched literally. If you want a hyphen in
702 your set of characters to be matched and its position in the class is such
703 that it could be considered part of a range, you must escape that hyphen
704 with a backslash.
705 .PP
706 Examples:
707 .PP
708 .Vb 12
709 \& [a\-z] # Matches a character that is a lower case ASCII letter.
710 \& [a\-fz] # Matches any letter between \*(Aqa\*(Aq and \*(Aqf\*(Aq (inclusive) or
711 \& # the letter \*(Aqz\*(Aq.
712 \& [\-z] # Matches either a hyphen (\*(Aq\-\*(Aq) or the letter \*(Aqz\*(Aq.
713 \& [a\-f\-m] # Matches any letter between \*(Aqa\*(Aq and \*(Aqf\*(Aq (inclusive), the
714 \& # hyphen (\*(Aq\-\*(Aq), or the letter \*(Aqm\*(Aq.
715 \& [\*(Aq\-?] # Matches any of the characters \*(Aq()*+,\-./0123456789:;<=>?
716 \& # (But not on an EBCDIC platform).
717 \& [\eN{APOSTROPHE}\-\eN{QUESTION MARK}]
718 \& # Matches any of the characters \*(Aq()*+,\-./0123456789:;<=>?
719 \& # even on an EBCDIC platform.
720 \& [\eN{U+27}\-\eN{U+3F}] # Same. (U+27 is "\*(Aq", and U+3F is "?")
721 .Ve
722 .PP
723 As the final two examples above show, you can achieve portability to
724 non-ASCII platforms by using the \f(CW\*(C`\eN{...}\*(C'\fR form for the range
725 endpoints. These indicate that the specified range is to be interpreted
726 using Unicode values, so \f(CW\*(C`[\eN{U+27}\-\eN{U+3F}]\*(C'\fR means to match
727 \&\f(CW\*(C`\eN{U+27}\*(C'\fR, \f(CW\*(C`\eN{U+28}\*(C'\fR, \f(CW\*(C`\eN{U+29}\*(C'\fR, ..., \f(CW\*(C`\eN{U+3D}\*(C'\fR, \f(CW\*(C`\eN{U+3E}\*(C'\fR,
728 and \f(CW\*(C`\eN{U+3F}\*(C'\fR, whatever the native code point versions for those are.
729 These are called "Unicode" ranges. If either end is of the \f(CW\*(C`\eN{...}\*(C'\fR
730 form, the range is considered Unicode. A \f(CW\*(C`regexp\*(C'\fR warning is raised
731 under \f(CW"use\ re\ \*(Aqstrict\*(Aq"\fR if the other endpoint is specified
732 non-portably:
733 .PP
734 .Vb 2
735 \& [\eN{U+00}\-\ex09] # Warning under re \*(Aqstrict\*(Aq; \ex09 is non\-portable
736 \& [\eN{U+00}\-\et] # No warning;
737 .Ve
738 .PP
739 Both of the above match the characters \f(CW\*(C`\eN{U+00}\*(C'\fR \f(CW\*(C`\eN{U+01}\*(C'\fR, ...
740 \&\f(CW\*(C`\eN{U+08}\*(C'\fR, \f(CW\*(C`\eN{U+09}\*(C'\fR, but the \f(CW\*(C`\ex09\*(C'\fR looks like it could be a
741 mistake so the warning is raised (under \f(CW\*(C`re \*(Aqstrict\*(Aq\*(C'\fR) for it.
742 .PP
743 Perl also guarantees that the ranges \f(CW\*(C`A\-Z\*(C'\fR, \f(CW\*(C`a\-z\*(C'\fR, \f(CW\*(C`0\-9\*(C'\fR, and any
744 subranges of these match what an English-only speaker would expect them
745 to match on any platform. That is, \f(CW\*(C`[A\-Z]\*(C'\fR matches the 26 ASCII
746 uppercase letters;
747 \&\f(CW\*(C`[a\-z]\*(C'\fR matches the 26 lowercase letters; and \f(CW\*(C`[0\-9]\*(C'\fR matches the 10
748 digits. Subranges, like \f(CW\*(C`[h\-k]\*(C'\fR, match correspondingly, in this case
749 just the four letters \f(CW"h"\fR, \f(CW"i"\fR, \f(CW"j"\fR, and \f(CW"k"\fR. This is the
750 natural behavior on ASCII platforms where the code points (ordinal
751 values) for \f(CW"h"\fR through \f(CW"k"\fR are consecutive integers (0x68 through
752 0x6B). But special handling to achieve this may be needed on platforms
753 with a non-ASCII native character set. For example, on EBCDIC
754 platforms, the code point for \f(CW"h"\fR is 0x88, \f(CW"i"\fR is 0x89, \f(CW"j"\fR is
755 0x91, and \f(CW"k"\fR is 0x92. Perl specially treats \f(CW\*(C`[h\-k]\*(C'\fR to exclude the
756 seven code points in the gap: 0x8A through 0x90. This special handling is
757 only invoked when the range is a subrange of one of the ASCII uppercase,
758 lowercase, and digit ranges, AND each end of the range is expressed
759 either as a literal, like \f(CW"A"\fR, or as a named character (\f(CW\*(C`\eN{...}\*(C'\fR,
760 including the \f(CW\*(C`\eN{U+...\*(C'\fR form).
761 .PP
762 EBCDIC Examples:
763 .PP
764 .Vb 10
765 \& [i\-j] # Matches either "i" or "j"
766 \& [i\-\eN{LATIN SMALL LETTER J}] # Same
767 \& [i\-\eN{U+6A}] # Same
768 \& [\eN{U+69}\-\eN{U+6A}] # Same
769 \& [\ex{89}\-\ex{91}] # Matches 0x89 ("i"), 0x8A .. 0x90, 0x91 ("j")
770 \& [i\-\ex{91}] # Same
771 \& [\ex{89}\-j] # Same
772 \& [i\-J] # Matches, 0x89 ("i") .. 0xC1 ("J"); special
773 \& # handling doesn\*(Aqt apply because range is mixed
774 \& # case
775 .Ve
776 .PP
777 \fINegation\fR
778 .IX Subsection "Negation"
779 .PP
780 It is also possible to instead list the characters you do not want to
781 match. You can do so by using a caret (\f(CW\*(C`^\*(C'\fR) as the first character in the
782 character class. For instance, \f(CW\*(C`[^a\-z]\*(C'\fR matches any character that is not a
783 lowercase ASCII letter, which therefore includes more than a million
784 Unicode code points. The class is said to be "negated" or "inverted".
785 .PP
786 This syntax make the caret a special character inside a bracketed character
787 class, but only if it is the first character of the class. So if you want
788 the caret as one of the characters to match, either escape the caret or
789 else don't list it first.
790 .PP
791 In inverted bracketed character classes, Perl ignores the Unicode rules
792 that normally say that named sequence, and certain characters should
793 match a sequence of multiple characters use under caseless \f(CW\*(C`/i\*(C'\fR
794 matching. Following those rules could lead to highly confusing
795 situations:
796 .PP
797 .Vb 1
798 \& "ss" =~ /^[^\exDF]+$/ui; # Matches!
799 .Ve
800 .PP
801 This should match any sequences of characters that aren't \f(CW\*(C`\exDF\*(C'\fR nor
802 what \f(CW\*(C`\exDF\*(C'\fR matches under \f(CW\*(C`/i\*(C'\fR. \f(CW"s"\fR isn't \f(CW\*(C`\exDF\*(C'\fR, but Unicode
803 says that \f(CW"ss"\fR is what \f(CW\*(C`\exDF\*(C'\fR matches under \f(CW\*(C`/i\*(C'\fR. So which one
804 "wins"? Do you fail the match because the string has \f(CW\*(C`ss\*(C'\fR or accept it
805 because it has an \f(CW\*(C`s\*(C'\fR followed by another \f(CW\*(C`s\*(C'\fR? Perl has chosen the
806 latter. (See note in "Bracketed Character Classes" above.)
807 .PP
808 Examples:
809 .PP
810 .Vb 4
811 \& "e" =~ /[^aeiou]/ # No match, the \*(Aqe\*(Aq is listed.
812 \& "x" =~ /[^aeiou]/ # Match, as \*(Aqx\*(Aq isn\*(Aqt a lowercase vowel.
813 \& "^" =~ /[^^]/ # No match, matches anything that isn\*(Aqt a caret.
814 \& "^" =~ /[x^]/ # Match, caret is not special here.
815 .Ve
816 .PP
817 \fIBackslash Sequences\fR
818 .IX Subsection "Backslash Sequences"
819 .PP
820 You can put any backslash sequence character class (with the exception of
821 \&\f(CW\*(C`\eN\*(C'\fR and \f(CW\*(C`\eR\*(C'\fR) inside a bracketed character class, and it will act just
822 as if you had put all characters matched by the backslash sequence inside the
823 character class. For instance, \f(CW\*(C`[a\-f\ed]\*(C'\fR matches any decimal digit, or any
824 of the lowercase letters between 'a' and 'f' inclusive.
825 .PP
826 \&\f(CW\*(C`\eN\*(C'\fR within a bracketed character class must be of the forms \f(CW\*(C`\eN{\fR\f(CIname\fR\f(CW}\*(C'\fR
827 or \f(CW\*(C`\eN{U+\fR\f(CIhex char\fR\f(CW}\*(C'\fR, and NOT be the form that matches non-newlines,
828 for the same reason that a dot \f(CW\*(C`.\*(C'\fR inside a bracketed character class loses
829 its special meaning: it matches nearly anything, which generally isn't what you
830 want to happen.
831 .PP
832 Examples:
833 .PP
834 .Vb 4
835 \& /[\ep{Thai}\ed]/ # Matches a character that is either a Thai
836 \& # character, or a digit.
837 \& /[^\ep{Arabic}()]/ # Matches a character that is neither an Arabic
838 \& # character, nor a parenthesis.
839 .Ve
840 .PP
841 Backslash sequence character classes cannot form one of the endpoints
842 of a range. Thus, you can't say:
843 .PP
844 .Vb 1
845 \& /[\ep{Thai}\-\ed]/ # Wrong!
846 .Ve
847 .PP
848 \fIPOSIX Character Classes\fR
849 .IX Xref "character class \\p \\p{} alpha alnum ascii blank cntrl digit graph lower print punct space upper word xdigit"
850 .IX Subsection "POSIX Character Classes"
851 .PP
852 POSIX character classes have the form \f(CW\*(C`[:class:]\*(C'\fR, where \fIclass\fR is the
853 name, and the \f(CW\*(C`[:\*(C'\fR and \f(CW\*(C`:]\*(C'\fR delimiters. POSIX character classes only appear
854 \&\fIinside\fR bracketed character classes, and are a convenient and descriptive
855 way of listing a group of characters.
856 .PP
857 Be careful about the syntax,
858 .PP
859 .Vb 2
860 \& # Correct:
861 \& $string =~ /[[:alpha:]]/
862 \&
863 \& # Incorrect (will warn):
864 \& $string =~ /[:alpha:]/
865 .Ve
866 .PP
867 The latter pattern would be a character class consisting of a colon,
868 and the letters \f(CW\*(C`a\*(C'\fR, \f(CW\*(C`l\*(C'\fR, \f(CW\*(C`p\*(C'\fR and \f(CW\*(C`h\*(C'\fR.
869 .PP
870 POSIX character classes can be part of a larger bracketed character class.
871 For example,
872 .PP
873 .Vb 1
874 \& [01[:alpha:]%]
875 .Ve
876 .PP
877 is valid and matches '0', '1', any alphabetic character, and the percent sign.
878 .PP
879 Perl recognizes the following POSIX character classes:
880 .PP
881 .Vb 10
882 \& alpha Any alphabetical character (e.g., [A\-Za\-z]).
883 \& alnum Any alphanumeric character (e.g., [A\-Za\-z0\-9]).
884 \& ascii Any character in the ASCII character set.
885 \& blank Any horizontal whitespace character (e.g. space or horizontal
886 \& tab ("\et")).
887 \& cntrl Any control character. See Note [2] below.
888 \& digit Any decimal digit (e.g., [0\-9]), equivalent to "\ed".
889 \& graph Any printable character, excluding a space. See Note [3] below.
890 \& lower Any lowercase character (e.g., [a\-z]).
891 \& print Any printable character, including a space. See Note [4] below.
892 \& punct Any graphical character excluding "word" characters. Note [5].
893 \& space Any whitespace character. "\es" including the vertical tab
894 \& ("\ecK").
895 \& upper Any uppercase character (e.g., [A\-Z]).
896 \& word A Perl extension (e.g., [A\-Za\-z0\-9_]), equivalent to "\ew".
897 \& xdigit Any hexadecimal digit (e.g., [0\-9a\-fA\-F]). Note [7].
898 .Ve
899 .PP
900 Like the Unicode properties, most of the POSIX
901 properties match the same regardless of whether case-insensitive (\f(CW\*(C`/i\*(C'\fR)
902 matching is in effect or not. The two exceptions are \f(CW\*(C`[:upper:]\*(C'\fR and
903 \&\f(CW\*(C`[:lower:]\*(C'\fR. Under \f(CW\*(C`/i\*(C'\fR, they each match the union of \f(CW\*(C`[:upper:]\*(C'\fR and
904 \&\f(CW\*(C`[:lower:]\*(C'\fR.
905 .PP
906 Most POSIX character classes have two Unicode-style \f(CW\*(C`\ep\*(C'\fR property
907 counterparts. (They are not official Unicode properties, but Perl extensions
908 derived from official Unicode properties.) The table below shows the relation
909 between POSIX character classes and these counterparts.
910 .PP
911 One counterpart, in the column labelled "ASCII-range Unicode" in
912 the table, matches only characters in the ASCII character set.
913 .PP
914 The other counterpart, in the column labelled "Full-range Unicode", matches any
915 appropriate characters in the full Unicode character set. For example,
916 \&\f(CW\*(C`\ep{Alpha}\*(C'\fR matches not just the ASCII alphabetic characters, but any
917 character in the entire Unicode character set considered alphabetic.
918 An entry in the column labelled "backslash sequence" is a (short)
919 equivalent.
920 .PP
921 .Vb 10
922 \& [[:...:]] ASCII\-range Full\-range backslash Note
923 \& Unicode Unicode sequence
924 \& \-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-
925 \& alpha \ep{PosixAlpha} \ep{XPosixAlpha}
926 \& alnum \ep{PosixAlnum} \ep{XPosixAlnum}
927 \& ascii \ep{ASCII}
928 \& blank \ep{PosixBlank} \ep{XPosixBlank} \eh [1]
929 \& or \ep{HorizSpace} [1]
930 \& cntrl \ep{PosixCntrl} \ep{XPosixCntrl} [2]
931 \& digit \ep{PosixDigit} \ep{XPosixDigit} \ed
932 \& graph \ep{PosixGraph} \ep{XPosixGraph} [3]
933 \& lower \ep{PosixLower} \ep{XPosixLower}
934 \& print \ep{PosixPrint} \ep{XPosixPrint} [4]
935 \& punct \ep{PosixPunct} \ep{XPosixPunct} [5]
936 \& \ep{PerlSpace} \ep{XPerlSpace} \es [6]
937 \& space \ep{PosixSpace} \ep{XPosixSpace} [6]
938 \& upper \ep{PosixUpper} \ep{XPosixUpper}
939 \& word \ep{PosixWord} \ep{XPosixWord} \ew
940 \& xdigit \ep{PosixXDigit} \ep{XPosixXDigit} [7]
941 .Ve
942 .IP [1] 4
943 .IX Item "[1]"
944 \&\f(CW\*(C`\ep{Blank}\*(C'\fR and \f(CW\*(C`\ep{HorizSpace}\*(C'\fR are synonyms.
945 .IP [2] 4
946 .IX Item "[2]"
947 Control characters don't produce output as such, but instead usually control
948 the terminal somehow: for example, newline and backspace are control characters.
949 On ASCII platforms, in the ASCII range, characters whose code points are
950 between 0 and 31 inclusive, plus 127 (\f(CW\*(C`DEL\*(C'\fR) are control characters; on
951 EBCDIC platforms, their counterparts are control characters.
952 .IP [3] 4
953 .IX Item "[3]"
954 Any character that is \fIgraphical\fR, that is, visible. This class consists
955 of all alphanumeric characters and all punctuation characters.
956 .IP [4] 4
957 .IX Item "[4]"
958 All printable characters, which is the set of all graphical characters
959 plus those whitespace characters which are not also controls.
960 .IP [5] 4
961 .IX Item "[5]"
962 \&\f(CW\*(C`\ep{PosixPunct}\*(C'\fR and \f(CW\*(C`[[:punct:]]\*(C'\fR in the ASCII range match all
963 non-controls, non-alphanumeric, non-space characters:
964 \&\f(CW\*(C`[\-!"#$%&\*(Aq()*+,./:;<=>?@[\e\e\e]^_\`{|}~]\*(C'\fR (although if a locale is in effect,
965 it could alter the behavior of \f(CW\*(C`[[:punct:]]\*(C'\fR).
966 .Sp
967 The similarly named property, \f(CW\*(C`\ep{Punct}\*(C'\fR, matches a somewhat different
968 set in the ASCII range, namely
969 \&\f(CW\*(C`[\-!"#%&\*(Aq()*,./:;?@[\e\e\e]_{}]\*(C'\fR. That is, it is missing the nine
970 characters \f(CW\*(C`[$+<=>^\`|~]\*(C'\fR.
971 This is because Unicode splits what POSIX considers to be punctuation into two
972 categories, Punctuation and Symbols.
973 .Sp
974 \&\f(CW\*(C`\ep{XPosixPunct}\*(C'\fR and (under Unicode rules) \f(CW\*(C`[[:punct:]]\*(C'\fR, match what
975 \&\f(CW\*(C`\ep{PosixPunct}\*(C'\fR matches in the ASCII range, plus what \f(CW\*(C`\ep{Punct}\*(C'\fR
976 matches. This is different than strictly matching according to
977 \&\f(CW\*(C`\ep{Punct}\*(C'\fR. Another way to say it is that
978 if Unicode rules are in effect, \f(CW\*(C`[[:punct:]]\*(C'\fR matches all characters
979 that Unicode considers punctuation, plus all ASCII-range characters that
980 Unicode considers symbols.
981 .IP [6] 4
982 .IX Item "[6]"
983 \&\f(CW\*(C`\ep{XPerlSpace}\*(C'\fR and \f(CW\*(C`\ep{Space}\*(C'\fR match identically starting with Perl
984 v5.18. In earlier versions, these differ only in that in non-locale
985 matching, \f(CW\*(C`\ep{XPerlSpace}\*(C'\fR did not match the vertical tab, \f(CW\*(C`\ecK\*(C'\fR.
986 Same for the two ASCII-only range forms.
987 .IP [7] 4
988 .IX Item "[7]"
989 Unlike \f(CW\*(C`[[:digit:]]\*(C'\fR which matches digits in many writing systems, such
990 as Thai and Devanagari, there are currently only two sets of hexadecimal
991 digits, and it is unlikely that more will be added. This is because you
992 not only need the ten digits, but also the six \f(CW\*(C`[A\-F]\*(C'\fR (and \f(CW\*(C`[a\-f]\*(C'\fR)
993 to correspond. That means only the Latin script is suitable for these,
994 and Unicode has only two sets of these, the familiar ASCII set, and the
995 fullwidth forms starting at U+FF10 (FULLWIDTH DIGIT ZERO).
996 .PP
997 There are various other synonyms that can be used besides the names
998 listed in the table. For example, \f(CW\*(C`\ep{XPosixAlpha}\*(C'\fR can be written as
999 \&\f(CW\*(C`\ep{Alpha}\*(C'\fR. All are listed in
1000 "Properties accessible through \ep{} and \eP{}" in perluniprops.
1001 .PP
1002 Both the \f(CW\*(C`\ep\*(C'\fR counterparts always assume Unicode rules are in effect.
1003 On ASCII platforms, this means they assume that the code points from 128
1004 to 255 are Latin\-1, and that means that using them under locale rules is
1005 unwise unless the locale is guaranteed to be Latin\-1 or UTF\-8. In contrast, the
1006 POSIX character classes are useful under locale rules. They are
1007 affected by the actual rules in effect, as follows:
1008 .ie n .IP "If the ""/a"" modifier, is in effect ..." 4
1009 .el .IP "If the \f(CW/a\fR modifier, is in effect ..." 4
1010 .IX Item "If the /a modifier, is in effect ..."
1011 Each of the POSIX classes matches exactly the same as their ASCII-range
1012 counterparts.
1013 .IP "otherwise ..." 4
1014 .IX Item "otherwise ..."
1015 .RS 4
1016 .PD 0
1017 .IP "For code points above 255 ..." 4
1018 .IX Item "For code points above 255 ..."
1019 .PD
1020 The POSIX class matches the same as its Full-range counterpart.
1021 .IP "For code points below 256 ..." 4
1022 .IX Item "For code points below 256 ..."
1023 .RS 4
1024 .PD 0
1025 .IP "if locale rules are in effect ..." 4
1026 .IX Item "if locale rules are in effect ..."
1027 .PD
1028 The POSIX class matches according to the locale, except:
1029 .RS 4
1030 .ie n .IP """word""" 4
1031 .el .IP \f(CWword\fR 4
1032 .IX Item "word"
1033 also includes the platform's native underscore character, no matter what
1034 the locale is.
1035 .ie n .IP """ascii""" 4
1036 .el .IP \f(CWascii\fR 4
1037 .IX Item "ascii"
1038 on platforms that don't have the POSIX \f(CW\*(C`ascii\*(C'\fR extension, this matches
1039 just the platform's native ASCII-range characters.
1040 .ie n .IP """blank""" 4
1041 .el .IP \f(CWblank\fR 4
1042 .IX Item "blank"
1043 on platforms that don't have the POSIX \f(CW\*(C`blank\*(C'\fR extension, this matches
1044 just the platform's native tab and space characters.
1045 .RE
1046 .RS 4
1047 .RE
1048 .IP "if, instead, Unicode rules are in effect ..." 4
1049 .IX Item "if, instead, Unicode rules are in effect ..."
1050 The POSIX class matches the same as the Full-range counterpart.
1051 .IP "otherwise ..." 4
1052 .IX Item "otherwise ..."
1053 The POSIX class matches the same as the ASCII range counterpart.
1054 .RE
1055 .RS 4
1056 .RE
1057 .RE
1058 .RS 4
1059 .RE
1060 .PP
1061 Which rules apply are determined as described in
1062 "Which character set modifier is in effect?" in perlre.
1063 .PP
1064 Negation of POSIX character classes
1065 .IX Xref "character class, negation"
1066 .IX Subsection "Negation of POSIX character classes"
1067 .PP
1068 A Perl extension to the POSIX character class is the ability to
1069 negate it. This is done by prefixing the class name with a caret (\f(CW\*(C`^\*(C'\fR).
1070 Some examples:
1071 .PP
1072 .Vb 7
1073 \& POSIX ASCII\-range Full\-range backslash
1074 \& Unicode Unicode sequence
1075 \& \-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-\-
1076 \& [[:^digit:]] \eP{PosixDigit} \eP{XPosixDigit} \eD
1077 \& [[:^space:]] \eP{PosixSpace} \eP{XPosixSpace}
1078 \& \eP{PerlSpace} \eP{XPerlSpace} \eS
1079 \& [[:^word:]] \eP{PerlWord} \eP{XPosixWord} \eW
1080 .Ve
1081 .PP
1082 The backslash sequence can mean either ASCII\- or Full-range Unicode,
1083 depending on various factors as described in "Which character set modifier is in effect?" in perlre.
1084 .PP
1085 [= =] and [. .]
1086 .IX Subsection "[= =] and [. .]"
1087 .PP
1088 Perl recognizes the POSIX character classes \f(CW\*(C`[=class=]\*(C'\fR and
1089 \&\f(CW\*(C`[.class.]\*(C'\fR, but does not (yet?) support them. Any attempt to use
1090 either construct raises an exception.
1091 .PP
1092 Examples
1093 .IX Subsection "Examples"
1094 .PP
1095 .Vb 12
1096 \& /[[:digit:]]/ # Matches a character that is a digit.
1097 \& /[01[:lower:]]/ # Matches a character that is either a
1098 \& # lowercase letter, or \*(Aq0\*(Aq or \*(Aq1\*(Aq.
1099 \& /[[:digit:][:^xdigit:]]/ # Matches a character that can be anything
1100 \& # except the letters \*(Aqa\*(Aq to \*(Aqf\*(Aq and \*(AqA\*(Aq to
1101 \& # \*(AqF\*(Aq. This is because the main character
1102 \& # class is composed of two POSIX character
1103 \& # classes that are ORed together, one that
1104 \& # matches any digit, and the other that
1105 \& # matches anything that isn\*(Aqt a hex digit.
1106 \& # The OR adds the digits, leaving only the
1107 \& # letters \*(Aqa\*(Aq to \*(Aqf\*(Aq and \*(AqA\*(Aq to \*(AqF\*(Aq excluded.
1108 .Ve
1109 .PP
1110 \fIExtended Bracketed Character Classes\fR
1111 .IX Xref "character class set operations"
1112 .IX Subsection "Extended Bracketed Character Classes"
1113 .PP
1114 This is a fancy bracketed character class that can be used for more
1115 readable and less error-prone classes, and to perform set operations,
1116 such as intersection. An example is
1117 .PP
1118 .Vb 1
1119 \& /(?[ \ep{Thai} & \ep{Digit} ])/
1120 .Ve
1121 .PP
1122 This will match all the digit characters that are in the Thai script.
1123 .PP
1124 This feature became available in Perl 5.18, as experimental; accepted in
1125 5.36.
1126 .PP
1127 The rules used by \f(CW\*(C`use re \*(Aqstrict\*(C'\fR apply to this
1128 construct.
1129 .PP
1130 We can extend the example above:
1131 .PP
1132 .Vb 1
1133 \& /(?[ ( \ep{Thai} + \ep{Lao} ) & \ep{Digit} ])/
1134 .Ve
1135 .PP
1136 This matches digits that are in either the Thai or Laotian scripts.
1137 .PP
1138 Notice the white space in these examples. This construct always has
1139 the \f(CW\*(C`/xx\*(C'\fR modifier turned on within it.
1140 .PP
1141 The available binary operators are:
1142 .PP
1143 .Vb 10
1144 \& & intersection
1145 \& + union
1146 \& | another name for \*(Aq+\*(Aq, hence means union
1147 \& \- subtraction (the result matches the set consisting of those
1148 \& code points matched by the first operand, excluding any that
1149 \& are also matched by the second operand)
1150 \& ^ symmetric difference (the union minus the intersection). This
1151 \& is like an exclusive or, in that the result is the set of code
1152 \& points that are matched by either, but not both, of the
1153 \& operands.
1154 .Ve
1155 .PP
1156 There is one unary operator:
1157 .PP
1158 .Vb 1
1159 \& ! complement
1160 .Ve
1161 .PP
1162 All the binary operators left associate; \f(CW"&"\fR is higher precedence
1163 than the others, which all have equal precedence. The unary operator
1164 right associates, and has highest precedence. Thus this follows the
1165 normal Perl precedence rules for logical operators. Use parentheses to
1166 override the default precedence and associativity.
1167 .PP
1168 The main restriction is that everything is a metacharacter. Thus,
1169 you cannot refer to single characters by doing something like this:
1170 .PP
1171 .Vb 1
1172 \& /(?[ a + b ])/ # Syntax error!
1173 .Ve
1174 .PP
1175 The easiest way to specify an individual typable character is to enclose
1176 it in brackets:
1177 .PP
1178 .Vb 1
1179 \& /(?[ [a] + [b] ])/
1180 .Ve
1181 .PP
1182 (This is the same thing as \f(CW\*(C`[ab]\*(C'\fR.) You could also have said the
1183 equivalent:
1184 .PP
1185 .Vb 1
1186 \& /(?[[ a b ]])/
1187 .Ve
1188 .PP
1189 (You can, of course, specify single characters by using, \f(CW\*(C`\ex{...}\*(C'\fR,
1190 \&\f(CW\*(C`\eN{...}\*(C'\fR, etc.)
1191 .PP
1192 This last example shows the use of this construct to specify an ordinary
1193 bracketed character class without additional set operations. Note the
1194 white space within it. This is allowed because \f(CW\*(C`/xx\*(C'\fR is
1195 automatically turned on within this construct.
1196 .PP
1197 All the other escapes accepted by normal bracketed character classes are
1198 accepted here as well.
1199 .PP
1200 Because this construct compiles under
1201 \&\f(CW\*(C`use re \*(Aqstrict\*(C'\fR, unrecognized escapes that
1202 generate warnings in normal classes are fatal errors here, as well as
1203 all other warnings from these class elements, as well as some
1204 practices that don't currently warn outside \f(CW\*(C`re \*(Aqstrict\*(Aq\*(C'\fR. For example
1205 you cannot say
1206 .PP
1207 .Vb 1
1208 \& /(?[ [ \exF ] ])/ # Syntax error!
1209 .Ve
1210 .PP
1211 You have to have two hex digits after a braceless \f(CW\*(C`\ex\*(C'\fR (use a leading
1212 zero to make two). These restrictions are to lower the incidence of
1213 typos causing the class to not match what you thought it would.
1214 .PP
1215 If a regular bracketed character class contains a \f(CW\*(C`\ep{}\*(C'\fR or \f(CW\*(C`\eP{}\*(C'\fR and
1216 is matched against a non-Unicode code point, a warning may be
1217 raised, as the result is not Unicode-defined. No such warning will come
1218 when using this extended form.
1219 .PP
1220 The final difference between regular bracketed character classes and
1221 these, is that it is not possible to get these to match a
1222 multi-character fold. Thus,
1223 .PP
1224 .Vb 1
1225 \& /(?[ [\exDF] ])/iu
1226 .Ve
1227 .PP
1228 does not match the string \f(CW\*(C`ss\*(C'\fR.
1229 .PP
1230 You don't have to enclose POSIX class names inside double brackets,
1231 hence both of the following work:
1232 .PP
1233 .Vb 2
1234 \& /(?[ [:word:] \- [:lower:] ])/
1235 \& /(?[ [[:word:]] \- [[:lower:]] ])/
1236 .Ve
1237 .PP
1238 Any contained POSIX character classes, including things like \f(CW\*(C`\ew\*(C'\fR and \f(CW\*(C`\eD\*(C'\fR
1239 respect the \f(CW\*(C`/a\*(C'\fR (and \f(CW\*(C`/aa\*(C'\fR) modifiers.
1240 .PP
1241 Note that \f(CW\*(C`(?[ ])\*(C'\fR is a regex-compile-time construct. Any attempt
1242 to use something which isn't knowable at the time the containing regular
1243 expression is compiled is a fatal error. In practice, this means
1244 just three limitations:
1245 .IP 1. 4
1246 When compiled within the scope of \f(CW\*(C`use locale\*(C'\fR (or the \f(CW\*(C`/l\*(C'\fR regex
1247 modifier), this construct assumes that the execution-time locale will be
1248 a UTF\-8 one, and the generated pattern always uses Unicode rules. What
1249 gets matched or not thus isn't dependent on the actual runtime locale, so
1250 tainting is not enabled. But a \f(CW\*(C`locale\*(C'\fR category warning is raised
1251 if the runtime locale turns out to not be UTF\-8.
1252 .IP 2. 4
1253 Any
1254 user-defined property
1255 used must be already defined by the time the regular expression is
1256 compiled (but note that this construct can be used instead of such
1257 properties).
1258 .IP 3. 4
1259 A regular expression that otherwise would compile
1260 using \f(CW\*(C`/d\*(C'\fR rules, and which uses this construct will instead
1261 use \f(CW\*(C`/u\*(C'\fR. Thus this construct tells Perl that you don't want
1262 \&\f(CW\*(C`/d\*(C'\fR rules for the entire regular expression containing it.
1263 .PP
1264 Note that skipping white space applies only to the interior of this
1265 construct. There must not be any space between any of the characters
1266 that form the initial \f(CW\*(C`(?[\*(C'\fR. Nor may there be space between the
1267 closing \f(CW\*(C`])\*(C'\fR characters.
1268 .PP
1269 Just as in all regular expressions, the pattern can be built up by
1270 including variables that are interpolated at regex compilation time.
1271 But currently each such sub-component should be an already-compiled
1272 extended bracketed character class.
1273 .PP
1274 .Vb 3
1275 \& my $thai_or_lao = qr/(?[ \ep{Thai} + \ep{Lao} ])/;
1276 \& ...
1277 \& qr/(?[ \ep{Digit} & $thai_or_lao ])/;
1278 .Ve
1279 .PP
1280 If you interpolate something else, the pattern may still compile (or it
1281 may die), but if it compiles, it very well may not behave as you would
1282 expect:
1283 .PP
1284 .Vb 2
1285 \& my $thai_or_lao = \*(Aq\ep{Thai} + \ep{Lao}\*(Aq;
1286 \& qr/(?[ \ep{Digit} & $thai_or_lao ])/;
1287 .Ve
1288 .PP
1289 compiles to
1290 .PP
1291 .Vb 1
1292 \& qr/(?[ \ep{Digit} & \ep{Thai} + \ep{Lao} ])/;
1293 .Ve
1294 .PP
1295 This does not have the effect that someone reading the source code
1296 would likely expect, as the intersection applies just to \f(CW\*(C`\ep{Thai}\*(C'\fR,
1297 excluding the Laotian.
1298 .PP
1299 Due to the way that Perl parses things, your parentheses and brackets
1300 may need to be balanced, even including comments. If you run into any
1301 examples, please submit them to <https://github.com/Perl/perl5/issues>,
1302 so that we can have a concrete example for this man page.