This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
perlfunc: Nit
[perl5.git] / pod / perllocale.pod
CommitLineData
66cbab2c
KW
1=encoding utf8
2
5f05dabc
PP
3=head1 NAME
4
b0c42ed9 5perllocale - Perl locale handling (internationalization and localization)
5f05dabc
PP
6
7=head1 DESCRIPTION
8
66cbab2c
KW
9In the beginning there was ASCII, the "American Standard Code for
10Information Interchange", which works quite well for Americans with
11their English alphabet and dollar-denominated currency. But it doesn't
12work so well even for other English speakers, who may use different
13currencies, such as the pound sterling (as the symbol for that currency
14is not in ASCII); and it's hopelessly inadequate for many of the
15thousands of the world's other languages.
16
17To address these deficiencies, the concept of locales was invented
18(formally the ISO C, XPG4, POSIX 1.c "locale system"). And applications
19were and are being written that use the locale mechanism. The process of
20making such an application take account of its users' preferences in
21these kinds of matters is called B<internationalization> (often
22abbreviated as B<i18n>); telling such an application about a particular
23set of preferences is known as B<localization> (B<l10n>).
24
39332f68 25Perl has been extended to support the locale system. This
66cbab2c
KW
26is controlled per application by using one pragma, one function call,
27and several environment variables.
28
29Unfortunately, there are quite a few deficiencies with the design (and
31f05a37
KW
30often, the implementations) of locales. Unicode was invented (see
31L<perlunitut> for an introduction to that) in part to address these
32design deficiencies, and nowadays, there is a series of "UTF-8
33locales", based on Unicode. These are locales whose character set is
34Unicode, encoded in UTF-8. Starting in v5.20, Perl fully supports
35UTF-8 locales, except for sorting and string comparisions. (Use
36L<Unicode::Collate> for these.) Perl continues to support the old
37non UTF-8 locales as well.
38
66cbab2c
KW
39(Unicode is also creating C<CLDR>, the "Common Locale Data Repository",
40L<http://cldr.unicode.org/> which includes more types of information than
41are available in the POSIX locale system. At the time of this writing,
42there was no CPAN module that provides access to this XML-encoded data.
43However, many of its locales have the POSIX-only data extracted, and are
31f05a37
KW
44available as UTF-8 locales at
45L<http://unicode.org/Public/cldr/latest/>.)
66cbab2c
KW
46
47=head1 WHAT IS A LOCALE
48
49A locale is a set of data that describes various aspects of how various
50communities in the world categorize their world. These categories are
51broken down into the following types (some of which include a brief
52note here):
53
54=over
55
56=item Category LC_NUMERIC: Numeric formatting
57
58This indicates how numbers should be formatted for human readability,
59for example the character used as the decimal point.
60
61=item Category LC_MONETARY: Formatting of monetary amounts
62
63=for comment
64The nbsp below makes this look better
65
66E<160>
67
68=item Category LC_TIME: Date/Time formatting
69
70=for comment
71The nbsp below makes this look better
72
73E<160>
74
75=item Category LC_MESSAGES: Error and other messages
76
2619d284 77This is used by Perl itself only for accessing operating system error
03c702c5 78messages via L<$!|perlvar/$ERRNO> and L<$^E|perlvar/$EXTENDED_OS_ERROR>.
66cbab2c
KW
79
80=item Category LC_COLLATE: Collation
81
76073c88 82This indicates the ordering of letters for comparison and sorting.
66cbab2c
KW
83In Latin alphabets, for example, "b", generally follows "a".
84
85=item Category LC_CTYPE: Character Types
86
87This indicates, for example if a character is an uppercase letter.
88
2619d284
KW
89=item Other categories
90
91Some platforms have other categories, dealing with such things as
92measurement units and paper sizes. None of these are used directly by
93Perl, but outside operations that Perl interacts with may use
4c9b78f4 94these. See L</Not within the scope of any "use locale" variant> below.
2619d284 95
66cbab2c
KW
96=back
97
2619d284
KW
98More details on the categories used by Perl are given below in L</LOCALE
99CATEGORIES>.
66cbab2c
KW
100
101Together, these categories go a long way towards being able to customize
102a single program to run in many different locations. But there are
103deficiencies, so keep reading.
5f05dabc
PP
104
105=head1 PREPARING TO USE LOCALES
106
2619d284
KW
107Perl itself will not use locales unless specifically requested to (but
108again note that Perl may interact with code that does use them). Even
109if there is such a request, B<all> of the following must be true
b960a36e 110for it to work properly:
5f05dabc
PP
111
112=over 4
113
114=item *
115
116B<Your operating system must support the locale system>. If it does,
39332f68 117you should find that the C<setlocale()> function is a documented part of
5f05dabc
PP
118its C library.
119
120=item *
121
5a964f20 122B<Definitions for locales that you use must be installed>. You, or
14280422
DD
123your system administrator, must make sure that this is the case. The
124available locales, the location in which they are kept, and the manner
5a964f20
TC
125in which they are installed all vary from system to system. Some systems
126provide only a few, hard-wired locales and do not allow more to be
127added. Others allow you to add "canned" locales provided by the system
128supplier. Still others allow you or the system administrator to define
14280422 129and add arbitrary locales. (You may have to ask your supplier to
5a964f20 130provide canned locales that are not delivered with your operating
14280422 131system.) Read your system documentation for further illumination.
5f05dabc
PP
132
133=item *
134
135B<Perl must believe that the locale system is supported>. If it does,
136C<perl -V:d_setlocale> will say that the value for C<d_setlocale> is
137C<define>.
138
139=back
140
141If you want a Perl application to process and present your data
142according to a particular locale, the application code should include
2ae324a7 143the S<C<use locale>> pragma (see L<The use locale pragma>) where
5f05dabc
PP
144appropriate, and B<at least one> of the following must be true:
145
146=over 4
147
c052850d 148=item 1
5f05dabc 149
66cbab2c 150B<The locale-determining environment variables (see L</"ENVIRONMENT">)
5a964f20 151must be correctly set up> at the time the application is started, either
ef3087ec 152by yourself or by whomever set up your system account; or
5f05dabc 153
c052850d 154=item 2
5f05dabc 155
14280422
DD
156B<The application must set its own locale> using the method described in
157L<The setlocale function>.
5f05dabc
PP
158
159=back
160
161=head1 USING LOCALES
162
163=head2 The use locale pragma
164
2619d284 165By default, Perl itself ignores the current locale. The S<C<use locale>>
66cbab2c 166pragma tells Perl to use the current locale for some operations.
7ee2ae1e 167Starting in v5.16, there is an optional parameter to this pragma:
66cbab2c
KW
168
169 use locale ':not_characters';
170
31f05a37
KW
171This parameter allows better mixing of locales and Unicode (less useful
172in v5.20 and later), and is
66cbab2c
KW
173described fully in L</Unicode and UTF-8>, but briefly, it tells Perl to
174not use the character portions of the locale definition, that is
175the C<LC_CTYPE> and C<LC_COLLATE> categories. Instead it will use the
2619d284 176native character set (extended by Unicode). When using this parameter,
66cbab2c
KW
177you are responsible for getting the external character set translated
178into the native/Unicode one (which it already will be if it is one of
179the increasingly popular UTF-8 locales). There are convenient ways of
180doing this, as described in L</Unicode and UTF-8>.
c052850d
KW
181
182The current locale is set at execution time by
183L<setlocale()|/The setlocale function> described below. If that function
184hasn't yet been called in the course of the program's execution, the
66cbab2c 185current locale is that which was determined by the L</"ENVIRONMENT"> in
c052850d
KW
186effect at the start of the program, except that
187C<L<LC_NUMERIC|/Category LC_NUMERIC: Numeric Formatting>> is always
3f7bac22
KW
188initialized to the C locale (the C locale is mentioned under L<Finding
189locales>).
dfcc8045 190If there is no valid environment, the current locale is whatever the
65ebb059
KW
191system default has been set to. On POSIX systems, it is likely, but
192not necessarily, the "C" locale. On Windows, the default is set via the
193computer's S<C<Control Panel-E<gt>Regional and Language Options>> (or its
194current equivalent).
c052850d
KW
195
196The operations that are affected by locale are:
5f05dabc
PP
197
198=over 4
199
4c9b78f4 200=item B<Not within the scope of any C<"use locale"> variant>
b960a36e 201
1d2ab946
KW
202Only operations originating outside Perl should be affected, as follows:
203
204=over 4
205
206=item *
2619d284
KW
207
208The variable L<$!|perlvar/$ERRNO> (and its synonyms C<$ERRNO> and
209C<$OS_ERROR>) when used as strings always are in terms of the current
210locale.
211
1d2ab946
KW
212=item *
213
2619d284
KW
214The current locale is also used when going outside of Perl with
215operations like L<system()|perlfunc/system LIST> or
216L<qxE<sol>E<sol>|perlop/qxE<sol>STRINGE<sol>>, if those operations are
217locale-sensitive.
218
1d2ab946
KW
219=item *
220
2619d284
KW
221Also Perl gives access to various C library functions through the
222L<POSIX> module. Some of those functions are always affected by the
223current locale. For example, C<POSIX::strftime()> uses C<LC_TIME>;
224C<POSIX::strtod()> uses C<LC_NUMERIC>; C<POSIX::strcoll()> and
225C<POSIX::strxfrm()> use C<LC_COLLATE>; and character classification
226functions like C<POSIX::isalnum()> use C<LC_CTYPE>. All such functions
227will behave according to the current underlying locale, even if that
1d2ab946 228locale isn't exposed to Perl space.
2619d284 229
1d2ab946
KW
230=item *
231
232Perl also provides lite wrappers for XS modules to use some C library
233C<printf> functions. These wrappers don't do anything with the locale,
234and the underlying C library function is affected by the locale in
235effect at the time of the wrapper call.
236The affected functions are
237L<perlapi/my_sprintf>,
238L<perlapi/my_snprintf>,
239and
240L<perlapi/my_vsnprintf>.
241
242=back
243
244=item Lingering effects of C<S<use locale>>
245
246Certain Perl operations that are set-up within the scope of a
b960a36e
KW
247C<use locale> variant retain that effect even outside the scope.
248These include:
249
250=over 4
251
252=item *
253
254The output format of a L<write()|perlfunc/write> is determined by an
255earlier format declaration (L<perlfunc/format>), so whether or not the
256output is affected by locale is determined by if the C<format()> is
257within the scope of a C<use locale> variant, not whether the C<write()>
258is.
259
260=item *
261
262Regular expression patterns can be compiled using
263L<qrE<sol>E<sol>|perlop/qrE<sol>STRINGE<sol>msixpodual> with actual
264matching deferred to later. Again, it is whether or not the compilation
265was done within the scope of C<use locale> that determines the match
266behavior, not if the matches are done within such a scope or not.
267
268=back
269
4c9b78f4 270=item B<Under C<"use locale ':not_characters';">>
66cbab2c
KW
271
272=over 4
273
274=item *
275
b960a36e
KW
276All the non-Perl operations.
277
278=item *
279
280B<Format declarations> (L<perlfunc/format>) and hence any subsequent
281C<write()>s use C<LC_NUMERIC>.
66cbab2c
KW
282
283=item *
284
b960a36e
KW
285B<stringification and output> use C<LC_NUMERIC>.
286These include the results of
287C<print()>,
288C<printf()>,
289C<say()>,
290and
291C<sprintf()>.
66cbab2c
KW
292
293=back
294
295=for comment
296The nbsp below makes this look better
297
298E<160>
299
4c9b78f4 300=item B<Under just plain C<"use locale";>>
66cbab2c 301
66cbab2c
KW
302=over 4
303
5f05dabc
PP
304=item *
305
b960a36e
KW
306All the above operations
307
308=item *
309
310B<The comparison operators> (C<lt>, C<le>, C<cmp>, C<ge>, and C<gt>) use
39332f68 311C<LC_COLLATE>. C<sort()> is also affected if used without an
5a964f20 312explicit comparison function, because it uses C<cmp> by default.
14280422 313
5a964f20 314B<Note:> C<eq> and C<ne> are unaffected by locale: they always
de108802 315perform a char-by-char comparison of their scalar operands. What's
14280422
DD
316more, if C<cmp> finds that its operands are equal according to the
317collation sequence specified by the current locale, it goes on to
de108802
RGS
318perform a char-by-char comparison, and only returns I<0> (equal) if the
319operands are char-for-char identical. If you really want to know whether
5a964f20 320two strings--which C<eq> and C<cmp> may consider different--are equal
14280422
DD
321as far as collation in the locale is concerned, see the discussion in
322L<Category LC_COLLATE: Collation>.
5f05dabc
PP
323
324=item *
325
39332f68
KW
326B<Regular expressions and case-modification functions> (C<uc()>, C<lc()>,
327C<ucfirst()>, and C<lcfirst()>) use C<LC_CTYPE>
5f05dabc 328
5f05dabc
PP
329=back
330
66cbab2c 331=back
5f05dabc 332
5a964f20 333The default behavior is restored with the S<C<no locale>> pragma, or
ef3087ec 334upon reaching the end of the block enclosing C<use locale>.
66cbab2c
KW
335Note that C<use locale> and C<use locale ':not_characters'> may be
336nested, and that what is in effect within an inner scope will revert to
337the outer scope's rules at the end of the inner scope.
5f05dabc 338
5a964f20 339The string result of any operation that uses locale
14280422
DD
340information is tainted, as it is possible for a locale to be
341untrustworthy. See L<"SECURITY">.
5f05dabc
PP
342
343=head2 The setlocale function
344
14280422 345You can switch locales as often as you wish at run time with the
39332f68 346C<POSIX::setlocale()> function:
5f05dabc 347
5f05dabc
PP
348 # Import locale-handling tool set from POSIX module.
349 # This example uses: setlocale -- the function call
350 # LC_CTYPE -- explained below
2619d284
KW
351 # (Showing the testing for success/failure of operations is
352 # omitted in these examples to avoid distracting from the main
dfcc8045 353 # point
6ea81ccf 354
5f05dabc 355 use POSIX qw(locale_h);
dfcc8045
KW
356 use locale;
357 my $old_locale;
5f05dabc 358
14280422 359 # query and save the old locale
5f05dabc
PP
360 $old_locale = setlocale(LC_CTYPE);
361
362 setlocale(LC_CTYPE, "fr_CA.ISO8859-1");
363 # LC_CTYPE now in locale "French, Canada, codeset ISO 8859-1"
364
365 setlocale(LC_CTYPE, "");
65ebb059
KW
366 # LC_CTYPE now reset to the default defined by the
367 # LC_ALL/LC_CTYPE/LANG environment variables, or to the system
368 # default. See below for documentation.
5f05dabc
PP
369
370 # restore the old locale
371 setlocale(LC_CTYPE, $old_locale);
372
39332f68 373The first argument of C<setlocale()> gives the B<category>, the second the
14280422
DD
374B<locale>. The category tells in what aspect of data processing you
375want to apply locale-specific rules. Category names are discussed in
66cbab2c 376L</LOCALE CATEGORIES> and L</"ENVIRONMENT">. The locale is the name of a
14280422
DD
377collection of customization information corresponding to a particular
378combination of language, country or territory, and codeset. Read on for
379hints on the naming of locales: not all systems name locales as in the
380example.
381
39332f68 382If no second argument is provided and the category is something other
502a173a
JH
383than LC_ALL, the function returns a string naming the current locale
384for the category. You can use this value as the second argument in a
f170b852
KW
385subsequent call to C<setlocale()>, B<but> on some platforms the string
386is opaque, not something that most people would be able to decipher as
387to what locale it means.
502a173a
JH
388
389If no second argument is provided and the category is LC_ALL, the
390result is implementation-dependent. It may be a string of
c052850d 391concatenated locale names (separator also implementation-dependent)
39332f68 392or a single locale name. Please consult your L<setlocale(3)> man page for
502a173a
JH
393details.
394
395If a second argument is given and it corresponds to a valid locale,
396the locale for the category is set to that value, and the function
397returns the now-current locale value. You can then use this in yet
39332f68 398another call to C<setlocale()>. (In some implementations, the return
502a173a
JH
399value may sometimes differ from the value you gave as the second
400argument--think of it as an alias for the value you gave.)
5f05dabc
PP
401
402As the example shows, if the second argument is an empty string, the
403category's locale is returned to the default specified by the
404corresponding environment variables. Generally, this results in a
5a964f20 405return to the default that was in force when Perl started up: changes
54310121 406to the environment made by the application after startup may or may not
5a964f20 407be noticed, depending on your system's C library.
5f05dabc 408
66cbab2c
KW
409Note that Perl ignores the current C<LC_CTYPE> and C<LC_COLLATE> locales
410within the scope of a C<use locale ':not_characters'>.
411
f170b852 412If C<set_locale()> fails for some reason (for example, an attempt to set
dfcc8045
KW
413to a locale unknown to the system), the locale for the category is not
414changed, and the function returns C<undef>.
415
2619d284 416
39332f68 417For further information about the categories, consult L<setlocale(3)>.
3e6e419a
JH
418
419=head2 Finding locales
420
39332f68 421For locales available in your system, consult also L<setlocale(3)> to
5a964f20
TC
422see whether it leads to the list of available locales (search for the
423I<SEE ALSO> section). If that fails, try the following command lines:
5f05dabc
PP
424
425 locale -a
426
427 nlsinfo
428
429 ls /usr/lib/nls/loc
430
431 ls /usr/lib/locale
432
433 ls /usr/lib/nls
434
b478f28d
JH
435 ls /usr/share/locale
436
5f05dabc
PP
437and see whether they list something resembling these
438
2bdf8add 439 en_US.ISO8859-1 de_DE.ISO8859-1 ru_RU.ISO8859-5
502a173a 440 en_US.iso88591 de_DE.iso88591 ru_RU.iso88595
2bdf8add 441 en_US de_DE ru_RU
14280422 442 en de ru
2bdf8add
JH
443 english german russian
444 english.iso88591 german.iso88591 russian.iso88595
502a173a 445 english.roman8 russian.koi8r
5f05dabc 446
39332f68 447Sadly, even though the calling interface for C<setlocale()> has been
528d65ad 448standardized, names of locales and the directories where the
5a964f20 449configuration resides have not been. The basic form of the name is
528d65ad
JH
450I<language_territory>B<.>I<codeset>, but the latter parts after
451I<language> are not always present. The I<language> and I<country>
452are usually from the standards B<ISO 3166> and B<ISO 639>, the
453two-letter abbreviations for the countries and the languages of the
454world, respectively. The I<codeset> part often mentions some B<ISO
4558859> character set, the Latin codesets. For example, C<ISO 8859-1>
456is the so-called "Western European codeset" that can be used to encode
457most Western European languages adequately. Again, there are several
458ways to write even the name of that one standard. Lamentably.
5f05dabc 459
14280422
DD
460Two special locales are worth particular mention: "C" and "POSIX".
461Currently these are effectively the same locale: the difference is
5a964f20
TC
462mainly that the first one is defined by the C standard, the second by
463the POSIX standard. They define the B<default locale> in which
14280422 464every program starts in the absence of locale information in its
5a964f20 465environment. (The I<default> default locale, if you will.) Its language
39332f68
KW
466is (American) English and its character codeset ASCII or, rarely, a
467superset thereof (such as the "DEC Multinational Character Set
468(DEC-MCS)"). B<Warning>. The C locale delivered by some vendors
469may not actually exactly match what the C standard calls for. So
470beware.
5f05dabc 471
14280422
DD
472B<NOTE>: Not all systems have the "POSIX" locale (not all systems are
473POSIX-conformant), so use "C" when you need explicitly to specify this
474default locale.
5f05dabc 475
3e6e419a
JH
476=head2 LOCALE PROBLEMS
477
5a964f20 478You may encounter the following warning message at Perl startup:
3e6e419a
JH
479
480 perl: warning: Setting locale failed.
481 perl: warning: Please check that your locale settings:
482 LC_ALL = "En_US",
483 LANG = (unset)
484 are supported and installed on your system.
485 perl: warning: Falling back to the standard locale ("C").
486
5a964f20
TC
487This means that your locale settings had LC_ALL set to "En_US" and
488LANG exists but has no value. Perl tried to believe you but could not.
489Instead, Perl gave up and fell back to the "C" locale, the default locale
65ebb059
KW
490that is supposed to work no matter what. (On Windows, it first tries
491falling back to the system default locale.) This usually means your
492locale settings were wrong, they mention locales your system has never
493heard of, or the locale installation in your system has problems (for
494example, some system files are broken or missing). There are quick and
495temporary fixes to these problems, as well as more thorough and lasting
496fixes.
3e6e419a 497
83fb1bf0
KW
498=head2 Testing for broken locales
499
500If you are building Perl from source, the Perl test suite file
501F<lib/locale.t> can be used to test the locales on your system.
502Setting the environment variable C<PERL_DEBUG_FULL_TEST> to 1
503will cause it to output detailed results. For example, on Linux, you
504could say
505
1d2ab946 506 PERL_DEBUG_FULL_TEST=1 ./perl -T -Ilib lib/locale.t > locale.log 2>&1
83fb1bf0
KW
507
508Besides many other tests, it will test every locale it finds on your
509system to see if they conform to the POSIX standard. If any have
510errors, it will include a summary near the end of the output of which
511locales passed all its tests, and which failed, and why.
512
3e6e419a
JH
513=head2 Temporarily fixing locale problems
514
5a964f20 515The two quickest fixes are either to render Perl silent about any
3e6e419a
JH
516locale inconsistencies or to run Perl under the default locale "C".
517
518Perl's moaning about locale problems can be silenced by setting the
900bd440
JH
519environment variable PERL_BADLANG to a zero value, for example "0".
520This method really just sweeps the problem under the carpet: you tell
521Perl to shut up even when Perl sees that something is wrong. Do not
522be surprised if later something locale-dependent misbehaves.
3e6e419a
JH
523
524Perl can be run under the "C" locale by setting the environment
5a964f20
TC
525variable LC_ALL to "C". This method is perhaps a bit more civilized
526than the PERL_BADLANG approach, but setting LC_ALL (or
527other locale variables) may affect other programs as well, not just
528Perl. In particular, external programs run from within Perl will see
3e6e419a 529these changes. If you make the new settings permanent (read on), all
f979aebc 530programs you run see the changes. See L<"ENVIRONMENT"> for
5a964f20 531the full list of relevant environment variables and L<USING LOCALES>
e05ffc7d 532for their effects in Perl. Effects in other programs are
5a964f20 533easily deducible. For example, the variable LC_COLLATE may well affect
b432a672 534your B<sort> program (or whatever the program that arranges "records"
3e6e419a
JH
535alphabetically in your system is called).
536
5a964f20
TC
537You can test out changing these variables temporarily, and if the
538new settings seem to help, put those settings into your shell startup
539files. Consult your local documentation for the exact details. For in
540Bourne-like shells (B<sh>, B<ksh>, B<bash>, B<zsh>):
3e6e419a
JH
541
542 LC_ALL=en_US.ISO8859-1
543 export LC_ALL
544
5a964f20
TC
545This assumes that we saw the locale "en_US.ISO8859-1" using the commands
546discussed above. We decided to try that instead of the above faulty
547locale "En_US"--and in Cshish shells (B<csh>, B<tcsh>)
3e6e419a
JH
548
549 setenv LC_ALL en_US.ISO8859-1
c47ff5f1 550
c406981e
JH
551or if you have the "env" application you can do in any shell
552
553 env LC_ALL=en_US.ISO8859-1 perl ...
554
5a964f20 555If you do not know what shell you have, consult your local
3e6e419a
JH
556helpdesk or the equivalent.
557
558=head2 Permanently fixing locale problems
559
5a964f20
TC
560The slower but superior fixes are when you may be able to yourself
561fix the misconfiguration of your own environment variables. The
3e6e419a
JH
562mis(sing)configuration of the whole system's locales usually requires
563the help of your friendly system administrator.
564
5a964f20
TC
565First, see earlier in this document about L<Finding locales>. That tells
566how to find which locales are really supported--and more importantly,
567installed--on your system. In our example error message, environment
568variables affecting the locale are listed in the order of decreasing
569importance (and unset variables do not matter). Therefore, having
570LC_ALL set to "En_US" must have been the bad choice, as shown by the
571error message. First try fixing locale settings listed first.
3e6e419a 572
5a964f20
TC
573Second, if using the listed commands you see something B<exactly>
574(prefix matches do not count and case usually counts) like "En_US"
575without the quotes, then you should be okay because you are using a
576locale name that should be installed and available in your system.
4a4eefd0 577In this case, see L<Permanently fixing your system's locale configuration>.
3e6e419a 578
4a4eefd0 579=head2 Permanently fixing your system's locale configuration
3e6e419a 580
5a964f20 581This is when you see something like:
3e6e419a
JH
582
583 perl: warning: Please check that your locale settings:
584 LC_ALL = "En_US",
585 LANG = (unset)
586 are supported and installed on your system.
587
588but then cannot see that "En_US" listed by the above-mentioned
5a964f20
TC
589commands. You may see things like "en_US.ISO8859-1", but that isn't
590the same. In this case, try running under a locale
591that you can list and which somehow matches what you tried. The
3e6e419a 592rules for matching locale names are a bit vague because
e05ffc7d 593standardization is weak in this area. See again the
13a2d996 594L<Finding locales> about general rules.
3e6e419a 595
b687b08b 596=head2 Fixing system locale configuration
3e6e419a 597
5a964f20
TC
598Contact a system administrator (preferably your own) and report the exact
599error message you get, and ask them to read this same documentation you
600are now reading. They should be able to check whether there is something
601wrong with the locale configuration of the system. The L<Finding locales>
602section is unfortunately a bit vague about the exact commands and places
603because these things are not that standardized.
3e6e419a 604
5f05dabc
PP
605=head2 The localeconv function
606
39332f68 607The C<POSIX::localeconv()> function allows you to get particulars of the
14280422
DD
608locale-dependent numeric formatting information specified by the current
609C<LC_NUMERIC> and C<LC_MONETARY> locales. (If you just want the name of
39332f68 610the current locale for a particular category, use C<POSIX::setlocale()>
5a964f20 611with a single parameter--see L<The setlocale function>.)
5f05dabc
PP
612
613 use POSIX qw(locale_h);
5f05dabc
PP
614
615 # Get a reference to a hash of locale-dependent info
616 $locale_values = localeconv();
617
618 # Output sorted list of the values
619 for (sort keys %$locale_values) {
14280422 620 printf "%-20s = %s\n", $_, $locale_values->{$_}
5f05dabc
PP
621 }
622
39332f68 623C<localeconv()> takes no arguments, and returns B<a reference to> a hash.
5a964f20 624The keys of this hash are variable names for formatting, such as
502a173a 625C<decimal_point> and C<thousands_sep>. The values are the
cea6626f 626corresponding, er, values. See L<POSIX/localeconv> for a longer
502a173a
JH
627example listing the categories an implementation might be expected to
628provide; some provide more and others fewer. You don't need an
39332f68 629explicit C<use locale>, because C<localeconv()> always observes the
502a173a 630current locale.
5f05dabc 631
5a964f20
TC
632Here's a simple-minded example program that rewrites its command-line
633parameters as integers correctly formatted in the current locale:
5f05dabc 634
ef3087ec
KW
635 use POSIX qw(locale_h);
636
637 # Get some of locale's numeric formatting parameters
638 my ($thousands_sep, $grouping) =
639 @{localeconv()}{'thousands_sep', 'grouping'};
640
641 # Apply defaults if values are missing
642 $thousands_sep = ',' unless $thousands_sep;
643
644 # grouping and mon_grouping are packed lists
645 # of small integers (characters) telling the
646 # grouping (thousand_seps and mon_thousand_seps
647 # being the group dividers) of numbers and
648 # monetary quantities. The integers' meanings:
649 # 255 means no more grouping, 0 means repeat
650 # the previous grouping, 1-254 means use that
651 # as the current grouping. Grouping goes from
652 # right to left (low to high digits). In the
653 # below we cheat slightly by never using anything
654 # else than the first grouping (whatever that is).
655 if ($grouping) {
656 @grouping = unpack("C*", $grouping);
657 } else {
658 @grouping = (3);
659 }
660
661 # Format command line params for current locale
662 for (@ARGV) {
663 $_ = int; # Chop non-integer part
664 1 while
665 s/(\d)(\d{$grouping[0]}($|$thousands_sep))/$1$thousands_sep$2/;
666 print "$_";
667 }
668 print "\n";
5f05dabc 669
74c76037 670=head2 I18N::Langinfo
4bbcc6e8
JH
671
672Another interface for querying locale-dependent information is the
39332f68 673C<I18N::Langinfo::langinfo()> function, available at least in Unix-like
4bbcc6e8
JH
674systems and VMS.
675
39332f68
KW
676The following example will import the C<langinfo()> function itself and
677three constants to be used as arguments to C<langinfo()>: a constant for
74c76037
JH
678the abbreviated first day of the week (the numbering starts from
679Sunday = 1) and two more constants for the affirmative and negative
680answers for a yes/no question in the current locale.
4bbcc6e8 681
74c76037 682 use I18N::Langinfo qw(langinfo ABDAY_1 YESSTR NOSTR);
4bbcc6e8 683
ef3087ec
KW
684 my ($abday_1, $yesstr, $nostr)
685 = map { langinfo } qw(ABDAY_1 YESSTR NOSTR);
4bbcc6e8 686
74c76037 687 print "$abday_1? [$yesstr/$nostr] ";
4bbcc6e8 688
74c76037
JH
689In other words, in the "C" (or English) locale the above will probably
690print something like:
691
e05ffc7d 692 Sun? [yes/no]
4bbcc6e8
JH
693
694See L<I18N::Langinfo> for more information.
695
5f05dabc
PP
696=head1 LOCALE CATEGORIES
697
5a964f20
TC
698The following subsections describe basic locale categories. Beyond these,
699some combination categories allow manipulation of more than one
700basic category at a time. See L<"ENVIRONMENT"> for a discussion of these.
5f05dabc
PP
701
702=head2 Category LC_COLLATE: Collation
703
66cbab2c
KW
704In the scope of S<C<use locale>> (but not a
705C<use locale ':not_characters'>), Perl looks to the C<LC_COLLATE>
5a964f20 706environment variable to determine the application's notions on collation
b4ffc3db
TC
707(ordering) of characters. For example, "b" follows "a" in Latin
708alphabets, but where do "E<aacute>" and "E<aring>" belong? And while
f87fa335 709"color" follows "chocolate" in English, what about in traditional Spanish?
5f05dabc 710
60f0fa02
JH
711The following collations all make sense and you may meet any of them
712if you "use locale".
713
714 A B C D E a b c d e
35316ca3 715 A a B b C c D d E e
60f0fa02
JH
716 a A b B c C d D e E
717 a b c d e A B C D E
718
f1cbbd6e 719Here is a code snippet to tell what "word"
5a964f20 720characters are in the current locale, in that locale's order:
5f05dabc
PP
721
722 use locale;
35316ca3 723 print +(sort grep /\w/, map { chr } 0..255), "\n";
5f05dabc 724
14280422
DD
725Compare this with the characters that you see and their order if you
726state explicitly that the locale should be ignored:
5f05dabc
PP
727
728 no locale;
35316ca3 729 print +(sort grep /\w/, map { chr } 0..255), "\n";
5f05dabc
PP
730
731This machine-native collation (which is what you get unless S<C<use
732locale>> has appeared earlier in the same block) must be used for
733sorting raw binary data, whereas the locale-dependent collation of the
b0c42ed9 734first example is useful for natural text.
5f05dabc 735
14280422
DD
736As noted in L<USING LOCALES>, C<cmp> compares according to the current
737collation locale when C<use locale> is in effect, but falls back to a
de108802 738char-by-char comparison for strings that the locale says are equal. You
39332f68 739can use C<POSIX::strcoll()> if you don't want this fall-back:
14280422
DD
740
741 use POSIX qw(strcoll);
742 $equal_in_locale =
743 !strcoll("space and case ignored", "SpaceAndCaseIgnored");
744
39332f68 745C<$equal_in_locale> will be true if the collation locale specifies a
5a964f20 746dictionary-like ordering that ignores space characters completely and
9e3a2af8 747which folds case.
14280422 748
31f05a37
KW
749Perl only supports single-byte locales for C<LC_COLLATE>. This means
750that a UTF-8 locale likely will just give you machine-native ordering.
751Use L<Unicode::Collate> for the full implementation of the Unicode
752Collation Algorithm.
753
5a964f20 754If you have a single string that you want to check for "equality in
14280422 755locale" against several others, you might think you could gain a little
39332f68 756efficiency by using C<POSIX::strxfrm()> in conjunction with C<eq>:
14280422
DD
757
758 use POSIX qw(strxfrm);
759 $xfrm_string = strxfrm("Mixed-case string");
760 print "locale collation ignores spaces\n"
761 if $xfrm_string eq strxfrm("Mixed-casestring");
762 print "locale collation ignores hyphens\n"
763 if $xfrm_string eq strxfrm("Mixedcase string");
764 print "locale collation ignores case\n"
765 if $xfrm_string eq strxfrm("mixed-case string");
766
39332f68 767C<strxfrm()> takes a string and maps it into a transformed string for use
de108802 768in char-by-char comparisons against other transformed strings during
14280422 769collation. "Under the hood", locale-affected Perl comparison operators
39332f68
KW
770call C<strxfrm()> for both operands, then do a char-by-char
771comparison of the transformed strings. By calling C<strxfrm()> explicitly
14280422 772and using a non locale-affected comparison, the example attempts to save
5a964f20 773a couple of transformations. But in fact, it doesn't save anything: Perl
2ae324a7 774magic (see L<perlguts/Magic Variables>) creates the transformed version of a
5a964f20 775string the first time it's needed in a comparison, then keeps this version around
14280422 776in case it's needed again. An example rewritten the easy way with
e38874e2 777C<cmp> runs just about as fast. It also copes with null characters
39332f68 778embedded in strings; if you call C<strxfrm()> directly, it treats the first
5a964f20
TC
779null it finds as a terminator. don't expect the transformed strings
780it produces to be portable across systems--or even from one revision
39332f68 781of your operating system to the next. In short, don't call C<strxfrm()>
e38874e2 782directly: let Perl do it for you.
14280422 783
5a964f20 784Note: C<use locale> isn't shown in some of these examples because it isn't
dfcc8045
KW
785needed: C<strcoll()> and C<strxfrm()> are POSIX functions
786which use the standard system-supplied C<libc> functions that
787always obey the current C<LC_COLLATE> locale.
5f05dabc
PP
788
789=head2 Category LC_CTYPE: Character Types
790
66cbab2c
KW
791In the scope of S<C<use locale>> (but not a
792C<use locale ':not_characters'>), Perl obeys the C<LC_CTYPE> locale
14280422
DD
793setting. This controls the application's notion of which characters are
794alphabetic. This affects Perl's C<\w> regular expression metanotation,
f1cbbd6e
GS
795which stands for alphanumeric characters--that is, alphabetic,
796numeric, and including other special characters such as the underscore or
797hyphen. (Consult L<perlre> for more information about
14280422 798regular expressions.) Thanks to C<LC_CTYPE>, depending on your locale
b4ffc3db
TC
799setting, characters like "E<aelig>", "E<eth>", "E<szlig>", and
800"E<oslash>" may be understood as C<\w> characters.
5f05dabc 801
2c268ad5 802The C<LC_CTYPE> locale also provides the map used in transliterating
68dc0745 803characters between lower and uppercase. This affects the case-mapping
39332f68 804functions--C<fc()>, C<lc()>, C<lcfirst()>, C<uc()>, and C<ucfirst()>; case-mapping
b9cc4f69
KW
805interpolation with C<\F>, C<\l>, C<\L>, C<\u>, or C<\U> in double-quoted
806strings and C<s///> substitutions; and case-independent regular expression
e38874e2
DD
807pattern matching using the C<i> modifier.
808
2da736a2 809Finally, C<LC_CTYPE> affects the (deprecated) POSIX character-class test
1d2ab946
KW
810functions--C<POSIX::isalpha()>, C<POSIX::islower()>, and so on. For
811example, if you move from the "C" locale to a 7-bit Scandinavian one,
812you may find--possibly to your surprise--that "|" moves from the
813C<POSIX::ispunct()> class to C<POSIX::isalpha()>.
ef3087ec
KW
814Unfortunately, this creates big problems for regular expressions. "|" still
815means alternation even though it matches C<\w>.
5f05dabc 816
31f05a37
KW
817Starting in v5.20, Perl supports UTF-8 locales for C<LC_CTYPE>, but
818otherwise Perl only supports single-byte locales, such as the ISO 8859
819series. This means that wide character locales, for example for Asian
820languages, are not supported. The UTF-8 locale support is actually a
821superset of POSIX locales, because it is really full Unicode behavior
822as if no locale were in effect at all (except for tainting; see
823L</SECURITY>). POSIX locales, even UTF-8 ones,
824are lacking certain concepts in Unicode, such as the idea that changing
825the case of a character could expand to be more than one character.
826Perl in a UTF-8 locale, will give you that expansion. Prior to v5.20,
827Perl treated a UTF-8 locale on some platforms like an ISO 8859-1 one,
828with some restrictions, and on other platforms more like the "C" locale.
829For releases v5.16 and v5.18, C<S<use locale 'not_characters>> could be
830used as a workaround for this (see L</Unicode and UTF-8>).
831
5d63e270
KW
832Note that there are quite a few things that are unaffected by the
833current locale. All the escape sequences for particular characters,
834C<\n> for example, always mean the platform's native one. This means,
835for example, that C<\N> in regular expressions (every character
1d2ab946 836but new-line) works on the platform character set.
5d63e270 837
14280422
DD
838B<Note:> A broken or malicious C<LC_CTYPE> locale definition may result
839in clearly ineligible characters being considered to be alphanumeric by
e199995e 840your application. For strict matching of (mundane) ASCII letters and
5a964f20 841digits--for example, in command strings--locale-aware applications
e199995e 842should use C<\w> with the C</a> regular expression modifier. See L<"SECURITY">.
5f05dabc
PP
843
844=head2 Category LC_NUMERIC: Numeric Formatting
845
b960a36e
KW
846After a proper C<POSIX::setlocale()> call, and within the scope of one
847of the C<use locale> variants, Perl obeys the C<LC_NUMERIC>
2095dafa 848locale information, which controls an application's idea of how numbers
b960a36e
KW
849should be formatted for human readability.
850In most implementations the only effect is to
b4ffc3db 851change the character used for the decimal point--perhaps from "." to ",".
b960a36e 852The functions aren't aware of such niceties as thousands separation and
2095dafa 853so on. (See L<The localeconv function> if you care about these things.)
5a964f20 854
b960a36e
KW
855 use POSIX qw(strtod setlocale LC_NUMERIC);
856 use locale;
5f05dabc 857
b960a36e 858 setlocale LC_NUMERIC, "";
14280422 859
b960a36e 860 $n = 5/2; # Assign numeric 2.5 to $n
5f05dabc 861
b960a36e 862 $a = " $n"; # Locale-dependent conversion to string
5f05dabc 863
b960a36e 864 print "half five is $n\n"; # Locale-dependent output
5f05dabc 865
b960a36e 866 printf "half five is %g\n", $n; # Locale-dependent output
5f05dabc 867
b960a36e
KW
868 print "DECIMAL POINT IS COMMA\n"
869 if $n == (strtod("2,5"))[0]; # Locale-dependent conversion
5f05dabc 870
4bbcc6e8
JH
871See also L<I18N::Langinfo> and C<RADIXCHAR>.
872
5f05dabc
PP
873=head2 Category LC_MONETARY: Formatting of monetary amounts
874
e199995e 875The C standard defines the C<LC_MONETARY> category, but not a function
5a964f20 876that is affected by its contents. (Those with experience of standards
b0c42ed9 877committees will recognize that the working group decided to punt on the
fa9b773e
KW
878issue.) Consequently, Perl essentially takes no notice of it. If you
879really want to use C<LC_MONETARY>, you can query its contents--see
e05ffc7d
KW
880L<The localeconv function>--and use the information that it returns in your
881application's own formatting of currency amounts. However, you may well
882find that the information, voluminous and complex though it may be, still
883does not quite meet your requirements: currency formatting is a hard nut
13a2d996 884to crack.
5f05dabc 885
4bbcc6e8
JH
886See also L<I18N::Langinfo> and C<CRNCYSTR>.
887
5f05dabc
PP
888=head2 LC_TIME
889
39332f68 890Output produced by C<POSIX::strftime()>, which builds a formatted
5f05dabc
PP
891human-readable date/time string, is affected by the current C<LC_TIME>
892locale. Thus, in a French locale, the output produced by the C<%B>
893format element (full month name) for the first month of the year would
5a964f20 894be "janvier". Here's how to get a list of long month names in the
5f05dabc
PP
895current locale:
896
897 use POSIX qw(strftime);
14280422
DD
898 for (0..11) {
899 $long_month_name[$_] =
900 strftime("%B", 0, 0, 0, 1, $_, 96);
5f05dabc
PP
901 }
902
2619d284
KW
903Note: C<use locale> isn't needed in this example: C<strftime()> is a POSIX
904function which uses the standard system-supplied C<libc> function that
905always obeys the current C<LC_TIME> locale.
5f05dabc 906
4bbcc6e8 907See also L<I18N::Langinfo> and C<ABDAY_1>..C<ABDAY_7>, C<DAY_1>..C<DAY_7>,
2a2bf5f4 908C<ABMON_1>..C<ABMON_12>, and C<ABMON_1>..C<ABMON_12>.
4bbcc6e8 909
5f05dabc
PP
910=head2 Other categories
911
2619d284
KW
912The remaining locale categories are not currently used by Perl itself.
913But again note that things Perl interacts with may use these, including
914extensions outside the standard Perl distribution, and by the
98a6f11e 915operating system and its utilities. Note especially that the string
916value of C<$!> and the error messages given by external utilities may
917be changed by C<LC_MESSAGES>. If you want to have portable error
265f5c4a 918codes, use C<%!>. See L<Errno>.
14280422
DD
919
920=head1 SECURITY
921
5a964f20 922Although the main discussion of Perl security issues can be found in
14280422
DD
923L<perlsec>, a discussion of Perl's locale handling would be incomplete
924if it did not draw your attention to locale-dependent security issues.
5a964f20
TC
925Locales--particularly on systems that allow unprivileged users to
926build their own locales--are untrustworthy. A malicious (or just plain
14280422
DD
927broken) locale can make a locale-aware application give unexpected
928results. Here are a few possibilities:
929
930=over 4
931
932=item *
933
934Regular expression checks for safe file names or mail addresses using
5a964f20 935C<\w> may be spoofed by an C<LC_CTYPE> locale that claims that
14280422
DD
936characters such as "E<gt>" and "|" are alphanumeric.
937
938=item *
939
e38874e2
DD
940String interpolation with case-mapping, as in, say, C<$dest =
941"C:\U$name.$ext">, may produce dangerous results if a bogus LC_CTYPE
942case-mapping table is in effect.
943
944=item *
945
14280422
DD
946A sneaky C<LC_COLLATE> locale could result in the names of students with
947"D" grades appearing ahead of those with "A"s.
948
949=item *
950
5a964f20 951An application that takes the trouble to use information in
14280422 952C<LC_MONETARY> may format debits as if they were credits and vice versa
5a964f20 953if that locale has been subverted. Or it might make payments in US
14280422
DD
954dollars instead of Hong Kong dollars.
955
956=item *
957
39332f68 958The date and day names in dates formatted by C<strftime()> could be
14280422 959manipulated to advantage by a malicious user able to subvert the
5a964f20 960C<LC_DATE> locale. ("Look--it says I wasn't in the building on
14280422
DD
961Sunday.")
962
963=back
964
965Such dangers are not peculiar to the locale system: any aspect of an
5a964f20 966application's environment which may be modified maliciously presents
14280422 967similar challenges. Similarly, they are not specific to Perl: any
5a964f20 968programming language that allows you to write programs that take
14280422
DD
969account of their environment exposes you to these issues.
970
5a964f20
TC
971Perl cannot protect you from all possibilities shown in the
972examples--there is no substitute for your own vigilance--but, when
14280422 973C<use locale> is in effect, Perl uses the tainting mechanism (see
5a964f20 974L<perlsec>) to mark string results that become locale-dependent, and
14280422 975which may be untrustworthy in consequence. Here is a summary of the
5a964f20 976tainting behavior of operators and functions that may be affected by
14280422
DD
977the locale:
978
979=over 4
980
551e1d92
RB
981=item *
982
983B<Comparison operators> (C<lt>, C<le>, C<ge>, C<gt> and C<cmp>):
14280422
DD
984
985Scalar true/false (or less/equal/greater) result is never tainted.
986
551e1d92
RB
987=item *
988
1d2ab946 989B<Case-mapping interpolation> (with C<\l>, C<\L>, C<\u>, C<\U>, or C<\F>)
e38874e2
DD
990
991Result string containing interpolated material is tainted if
66cbab2c 992C<use locale> (but not S<C<use locale ':not_characters'>>) is in effect.
e38874e2 993
551e1d92
RB
994=item *
995
996B<Matching operator> (C<m//>):
14280422
DD
997
998Scalar true/false result never tainted.
999
1d2ab946
KW
1000All subpatterns, either delivered as a list-context result or as C<$1>
1001I<etc>., are tainted if C<use locale> (but not
1002S<C<use locale ':not_characters'>>) is in effect, and the subpattern
1003regular expression is matched case-insensitively (C</i>) or contains a
1004locale-dependent construct. These constructs include C<\w>
1005(to match an alphanumeric character), C<\W> (non-alphanumeric
1006character), C<\s> (whitespace character), C<\S> (non whitespace
1007character), and the POSIX character classes, such as C<[:alpha:]> (see
1008L<perlrecharclass/POSIX Character Classes>).
1009The matched-pattern variables, C<$&>, C<$`> (pre-match), C<$'>
1010(post-match), and C<$+> (last match) also are tainted.
1011(Note that currently there are some bugs where not everything that
1012should be tainted gets tainted in all circumstances.)
14280422 1013
551e1d92
RB
1014=item *
1015
1016B<Substitution operator> (C<s///>):
14280422 1017
e38874e2 1018Has the same behavior as the match operator. Also, the left
66cbab2c
KW
1019operand of C<=~> becomes tainted when C<use locale>
1020(but not S<C<use locale ':not_characters'>>) is in effect if modified as
1021a result of a substitution based on a regular
1d2ab946
KW
1022expression match involving any of the things mentioned in the previous
1023item, or of case-mapping, such as C<\l>, C<\L>,C<\u>, C<\U>, or C<\F>.
14280422 1024
551e1d92
RB
1025=item *
1026
39332f68 1027B<Output formatting functions> (C<printf()> and C<write()>):
14280422 1028
3cf03d68
JH
1029Results are never tainted because otherwise even output from print,
1030for example C<print(1/7)>, should be tainted if C<use locale> is in
1031effect.
14280422 1032
551e1d92
RB
1033=item *
1034
39332f68 1035B<Case-mapping functions> (C<lc()>, C<lcfirst()>, C<uc()>, C<ucfirst()>):
14280422 1036
66cbab2c
KW
1037Results are tainted if C<use locale> (but not
1038S<C<use locale ':not_characters'>>) is in effect.
14280422 1039
551e1d92
RB
1040=item *
1041
39332f68
KW
1042B<POSIX locale-dependent functions> (C<localeconv()>, C<strcoll()>,
1043C<strftime()>, C<strxfrm()>):
14280422
DD
1044
1045Results are never tainted.
1046
551e1d92
RB
1047=item *
1048
1d2ab946
KW
1049B<POSIX character class tests> (C<POSIX::isalnum()>,
1050C<POSIX::isalpha()>, C<POSIX::isdigit()>, C<POSIX::isgraph()>,
1051C<POSIX::islower()>, C<POSIX::isprint()>, C<POSIX::ispunct()>,
1052C<POSIX::isspace()>, C<POSIX::isupper()>, C<POSIX::isxdigit()>):
14280422
DD
1053
1054True/false results are never tainted.
1055
1056=back
1057
1058Three examples illustrate locale-dependent tainting.
1059The first program, which ignores its locale, won't run: a value taken
54310121 1060directly from the command line may not be used to name an output file
14280422
DD
1061when taint checks are enabled.
1062
1063 #/usr/local/bin/perl -T
1064 # Run with taint checking
1065
54310121 1066 # Command line sanity check omitted...
14280422
DD
1067 $tainted_output_file = shift;
1068
1069 open(F, ">$tainted_output_file")
3183d96c 1070 or warn "Open of $tainted_output_file failed: $!\n";
14280422
DD
1071
1072The program can be made to run by "laundering" the tainted value through
5a964f20
TC
1073a regular expression: the second example--which still ignores locale
1074information--runs, creating the file named on its command line
14280422
DD
1075if it can.
1076
1077 #/usr/local/bin/perl -T
1078
1079 $tainted_output_file = shift;
1080 $tainted_output_file =~ m%[\w/]+%;
1081 $untainted_output_file = $&;
1082
1083 open(F, ">$untainted_output_file")
1084 or warn "Open of $untainted_output_file failed: $!\n";
1085
5a964f20 1086Compare this with a similar but locale-aware program:
14280422
DD
1087
1088 #/usr/local/bin/perl -T
1089
1090 $tainted_output_file = shift;
1091 use locale;
1092 $tainted_output_file =~ m%[\w/]+%;
1093 $localized_output_file = $&;
1094
1095 open(F, ">$localized_output_file")
1096 or warn "Open of $localized_output_file failed: $!\n";
1097
1d2ab946 1098This third program fails to run because C<$&> is tainted: it is the result
5a964f20 1099of a match involving C<\w> while C<use locale> is in effect.
5f05dabc
PP
1100
1101=head1 ENVIRONMENT
1102
1103=over 12
1104
1105=item PERL_BADLANG
1106
14280422 1107A string that can suppress Perl's warning about failed locale settings
54310121 1108at startup. Failure can occur if the locale support in the operating
5a964f20 1109system is lacking (broken) in some way--or if you mistyped the name of
900bd440
JH
1110a locale when you set up your environment. If this environment
1111variable is absent, or has a value that does not evaluate to integer
1112zero--that is, "0" or ""-- Perl will complain about locale setting
1113failures.
5f05dabc 1114
14280422
DD
1115B<NOTE>: PERL_BADLANG only gives you a way to hide the warning message.
1116The message tells about some problem in your system's locale support,
1117and you should investigate what the problem is.
5f05dabc
PP
1118
1119=back
1120
1121The following environment variables are not specific to Perl: They are
39332f68 1122part of the standardized (ISO C, XPG4, POSIX 1.c) C<setlocale()> method
b385bb4d
KW
1123for controlling an application's opinion on data. Windows is non-POSIX,
1124but Perl arranges for the following to work as described anyway.
65ebb059
KW
1125If the locale given by an environment variable is not valid, Perl tries
1126the next lower one in priority. If none are valid, on Windows, the
1127system default locale is then tried. If all else fails, the C<"C">
1128locale is used. If even that doesn't work, something is badly broken,
1129but Perl tries to forge ahead with whatever the locale settinga might
1130be.
5f05dabc
PP
1131
1132=over 12
1133
1134=item LC_ALL
1135
5a964f20 1136C<LC_ALL> is the "override-all" locale environment variable. If
5f05dabc
PP
1137set, it overrides all the rest of the locale environment variables.
1138
528d65ad
JH
1139=item LANGUAGE
1140
1141B<NOTE>: C<LANGUAGE> is a GNU extension, it affects you only if you
1142are using the GNU libc. This is the case if you are using e.g. Linux.
e1020413 1143If you are using "commercial" Unixes you are most probably I<not>
22b6f60d
JH
1144using GNU libc and you can ignore C<LANGUAGE>.
1145
1146However, in the case you are using C<LANGUAGE>: it affects the
1147language of informational, warning, and error messages output by
1148commands (in other words, it's like C<LC_MESSAGES>) but it has higher
96090e4f 1149priority than C<LC_ALL>. Moreover, it's not a single value but
22b6f60d
JH
1150instead a "path" (":"-separated list) of I<languages> (not locales).
1151See the GNU C<gettext> library documentation for more information.
528d65ad 1152
5f05dabc
PP
1153=item LC_CTYPE
1154
1155In the absence of C<LC_ALL>, C<LC_CTYPE> chooses the character type
1156locale. In the absence of both C<LC_ALL> and C<LC_CTYPE>, C<LANG>
1157chooses the character type locale.
1158
1159=item LC_COLLATE
1160
14280422
DD
1161In the absence of C<LC_ALL>, C<LC_COLLATE> chooses the collation
1162(sorting) locale. In the absence of both C<LC_ALL> and C<LC_COLLATE>,
1163C<LANG> chooses the collation locale.
5f05dabc
PP
1164
1165=item LC_MONETARY
1166
14280422
DD
1167In the absence of C<LC_ALL>, C<LC_MONETARY> chooses the monetary
1168formatting locale. In the absence of both C<LC_ALL> and C<LC_MONETARY>,
1169C<LANG> chooses the monetary formatting locale.
5f05dabc
PP
1170
1171=item LC_NUMERIC
1172
1173In the absence of C<LC_ALL>, C<LC_NUMERIC> chooses the numeric format
1174locale. In the absence of both C<LC_ALL> and C<LC_NUMERIC>, C<LANG>
1175chooses the numeric format.
1176
1177=item LC_TIME
1178
14280422
DD
1179In the absence of C<LC_ALL>, C<LC_TIME> chooses the date and time
1180formatting locale. In the absence of both C<LC_ALL> and C<LC_TIME>,
1181C<LANG> chooses the date and time formatting locale.
5f05dabc
PP
1182
1183=item LANG
1184
14280422
DD
1185C<LANG> is the "catch-all" locale environment variable. If it is set, it
1186is used as the last resort after the overall C<LC_ALL> and the
5f05dabc
PP
1187category-specific C<LC_...>.
1188
1189=back
1190
7e4353e9
RGS
1191=head2 Examples
1192
1193The LC_NUMERIC controls the numeric output:
1194
ef3087ec
KW
1195 use locale;
1196 use POSIX qw(locale_h); # Imports setlocale() and the LC_ constants.
1197 setlocale(LC_NUMERIC, "fr_FR") or die "Pardon";
1198 printf "%g\n", 1.23; # If the "fr_FR" succeeded, probably shows 1,23.
7e4353e9 1199
39332f68 1200and also how strings are parsed by C<POSIX::strtod()> as numbers:
7e4353e9 1201
ef3087ec
KW
1202 use locale;
1203 use POSIX qw(locale_h strtod);
1204 setlocale(LC_NUMERIC, "de_DE") or die "Entschuldigung";
1205 my $x = strtod("2,34") + 5;
1206 print $x, "\n"; # Probably shows 7,34.
7e4353e9 1207
5f05dabc
PP
1208=head1 NOTES
1209
b960a36e
KW
1210=head2 String C<eval> and C<LC_NUMERIC>
1211
1212A string L<eval|perlfunc/eval EXPR> parses its expression as standard
1213Perl. It is therefore expecting the decimal point to be a dot. If
1214C<LC_NUMERIC> is set to have this be a comma instead, the parsing will
1215be confused, perhaps silently.
1216
1217 use locale;
1218 use POSIX qw(locale_h);
1219 setlocale(LC_NUMERIC, "fr_FR") or die "Pardon";
1220 my $a = 1.2;
1221 print eval "$a + 1.5";
1222 print "\n";
1223
1224prints C<13,5>. This is because in that locale, the comma is the
1225decimal point character. The C<eval> thus expands to:
1226
1227 eval "1,2 + 1.5"
1228
1229and the result is not what you likely expected. No warnings are
1230generated. If you do string C<eval>'s within the scope of
1231S<C<use locale>>, you should instead change the C<eval> line to do
1232something like:
1233
1234 print eval "no locale; $a + 1.5";
1235
1236This prints C<2.7>.
1237
5f05dabc
PP
1238=head2 Backward compatibility
1239
b0c42ed9 1240Versions of Perl prior to 5.004 B<mostly> ignored locale information,
5a964f20
TC
1241generally behaving as if something similar to the C<"C"> locale were
1242always in force, even if the program environment suggested otherwise
1243(see L<The setlocale function>). By default, Perl still behaves this
1244way for backward compatibility. If you want a Perl application to pay
1245attention to locale information, you B<must> use the S<C<use locale>>
062ca197
KW
1246pragma (see L<The use locale pragma>) or, in the unlikely event
1247that you want to do so for just pattern matching, the
70709c68
KW
1248C</l> regular expression modifier (see L<perlre/Character set
1249modifiers>) to instruct it to do so.
b0c42ed9
JH
1250
1251Versions of Perl from 5.002 to 5.003 did use the C<LC_CTYPE>
5a964f20
TC
1252information if available; that is, C<\w> did understand what
1253were the letters according to the locale environment variables.
b0c42ed9
JH
1254The problem was that the user had no control over the feature:
1255if the C library supported locales, Perl used them.
1256
1257=head2 I18N:Collate obsolete
1258
5a964f20 1259In versions of Perl prior to 5.004, per-locale collation was possible
b0c42ed9
JH
1260using the C<I18N::Collate> library module. This module is now mildly
1261obsolete and should be avoided in new applications. The C<LC_COLLATE>
1262functionality is now integrated into the Perl core language: One can
1263use locale-specific scalar data completely normally with C<use locale>,
1264so there is no longer any need to juggle with the scalar references of
1265C<I18N::Collate>.
5f05dabc 1266
14280422 1267=head2 Sort speed and memory use impacts
5f05dabc
PP
1268
1269Comparing and sorting by locale is usually slower than the default
14280422
DD
1270sorting; slow-downs of two to four times have been observed. It will
1271also consume more memory: once a Perl scalar variable has participated
1272in any string comparison or sorting operation obeying the locale
1273collation rules, it will take 3-15 times more memory than before. (The
1274exact multiplier depends on the string's contents, the operating system
1275and the locale.) These downsides are dictated more by the operating
1276system's implementation of the locale system than by Perl.
5f05dabc 1277
5f05dabc
PP
1278=head2 Freely available locale definitions
1279
66cbab2c
KW
1280The Unicode CLDR project extracts the POSIX portion of many of its
1281locales, available at
1282
1283 http://unicode.org/Public/cldr/latest/
1284
08d7a6b2
LB
1285There is a large collection of locale definitions at:
1286
1287 http://std.dkuug.dk/i18n/WG15-collection/locales/
1288
1289You should be aware that it is
14280422 1290unsupported, and is not claimed to be fit for any purpose. If your
5a964f20 1291system allows installation of arbitrary locales, you may find the
14280422
DD
1292definitions useful as they are, or as a basis for the development of
1293your own locales.
5f05dabc 1294
14280422 1295=head2 I18n and l10n
5f05dabc 1296
b0c42ed9
JH
1297"Internationalization" is often abbreviated as B<i18n> because its first
1298and last letters are separated by eighteen others. (You may guess why
1299the internalin ... internaliti ... i18n tends to get abbreviated.) In
1300the same way, "localization" is often abbreviated to B<l10n>.
14280422
DD
1301
1302=head2 An imperfect standard
1303
1304Internationalization, as defined in the C and POSIX standards, can be
1305criticized as incomplete, ungainly, and having too large a granularity.
1306(Locales apply to a whole process, when it would arguably be more useful
1307to have them apply to a single thread, window group, or whatever.) They
1308also have a tendency, like standards groups, to divide the world into
1309nations, when we all know that the world can equally well be divided
e199995e 1310into bankers, bikers, gamers, and so on.
5f05dabc 1311
b310b053
JH
1312=head1 Unicode and UTF-8
1313
7ee2ae1e 1314The support of Unicode is new starting from Perl version v5.6, and more fully
31f05a37
KW
1315implemented in versions v5.8 and later. See L<perluniintro>.
1316
1317Starting in Perl v5.20, UTF-8 locales are supported in Perl, except for
1318C<LC_COLLATE> (use L<Unicode::Collate> instead). If you have Perl v5.16
1319or v5.18 and can't upgrade, you can use
66cbab2c
KW
1320
1321 use locale ':not_characters';
1322
1323When this form of the pragma is used, only the non-character portions of
1324locales are used by Perl, for example C<LC_NUMERIC>. Perl assumes that
1325you have translated all the characters it is to operate on into Unicode
1326(actually the platform's native character set (ASCII or EBCDIC) plus
1327Unicode). For data in files, this can conveniently be done by also
1328specifying
1329
1330 use open ':locale';
1331
1332This pragma arranges for all inputs from files to be translated into
1333Unicode from the current locale as specified in the environment (see
1334L</ENVIRONMENT>), and all outputs to files to be translated back
1335into the locale. (See L<open>). On a per-filehandle basis, you can
1336instead use the L<PerlIO::locale> module, or the L<Encode::Locale>
1337module, both available from CPAN. The latter module also has methods to
1338ease the handling of C<ARGV> and environment variables, and can be used
31f05a37 1339on individual strings. If you know that all your locales will be
66cbab2c
KW
1340UTF-8, as many are these days, you can use the L<B<-C>|perlrun/-C>
1341command line switch.
1342
1343This form of the pragma allows essentially seamless handling of locales
31f05a37
KW
1344with Unicode. The collation order will be by Unicode code point order.
1345It is strongly
66cbab2c
KW
1346recommended that when you need to order and sort strings that you use
1347the standard module L<Unicode::Collate> which gives much better results
1348in many instances than you can get with the old-style locale handling.
1349
31f05a37
KW
1350All the modules and switches just described can be used in v5.20 with
1351just plain C<use locale>, and, should the input locales not be UTF-8,
1352you'll get the less than ideal behavior, described below, that you get
1353with pre-v5.16 Perls, or when you use the locale pragma without the
1354C<:not_characters> parameter in v5.16 and v5.18. If you are using
1355exclusively UTF-8 locales in v5.20 and higher, the rest of this section
1356does not apply to you.
1357
1358There are two cases, multi-byte and single-byte locales. First
1359multi-byte:
1360
1361The only multi-byte (or wide character) locale that Perl is ever likely
1362to support is UTF-8. This is due to the difficulty of implementation,
1363the fact that high quality UTF-8 locales are now published for every
1364area of the world (L<http://unicode.org/Public/cldr/latest/>), and that
1365failing all that you can use the L<Encode> module to translate to/from
1366your locale. So, you'll have to do one of those things if you're using
1367one of these locales, such as Big5 or Shift JIS. For UTF-8 locales, in
1368Perls (pre v5.20) that don't have full UTF-8 locale support, they may
1369work reasonably well (depending on your C library implementation)
1370simply because both
dc4bfc4b
KW
1371they and Perl store characters that take up multiple bytes the same way.
1372However, some, if not most, C library implementations may not process
1373the characters in the upper half of the Latin-1 range (128 - 255)
1374properly under LC_CTYPE. To see if a character is a particular type
1375under a locale, Perl uses the functions like C<isalnum()>. Your C
1376library may not work for UTF-8 locales with those functions, instead
1377only working under the newer wide library functions like C<iswalnum()>.
31f05a37
KW
1378However, they are treated like single-byte locales, and will have the
1379restrictions described below.
e199995e 1380
31f05a37 1381For single-byte locales,
e199995e 1382Perl generally takes the tack to use locale rules on code points that can fit
66cbab2c
KW
1383in a single byte, and Unicode rules for those that can't (though this
1384isn't uniformly applied, see the note at the end of this section). This
1385prevents many problems in locales that aren't UTF-8. Suppose the locale
1386is ISO8859-7, Greek. The character at 0xD7 there is a capital Chi. But
1387in the ISO8859-1 locale, Latin1, it is a multiplication sign. The POSIX
1388regular expression character class C<[[:alpha:]]> will magically match
13890xD7 in the Greek locale but not in the Latin one.
e199995e 1390
1d2ab946 1391However, there are places where this breaks down. Certain Perl constructs are
b4ffc3db
TC
1392for Unicode only, such as C<\p{Alpha}>. They assume that 0xD7 always has its
1393Unicode meaning (or the equivalent on EBCDIC platforms). Since Latin1 is a
1394subset of Unicode and 0xD7 is the multiplication sign in both Latin1 and
1395Unicode, C<\p{Alpha}> will never match it, regardless of locale. A similar
31f05a37
KW
1396issue occurs with C<\N{...}>. Prior to v5.20, It is therefore a bad
1397idea to use C<\p{}> or
66cbab2c
KW
1398C<\N{}> under plain C<use locale>--I<unless> you can guarantee that the
1399locale will be a ISO8859-1. Use POSIX character classes instead.
1400
1401Another problem with this approach is that operations that cross the
1402single byte/multiple byte boundary are not well-defined, and so are
1403disallowed. (This boundary is between the codepoints at 255/256.).
1404For example, lower casing LATIN CAPITAL LETTER Y WITH DIAERESIS (U+0178)
1405should return LATIN SMALL LETTER Y WITH DIAERESIS (U+00FF). But in the
1406Greek locale, for example, there is no character at 0xFF, and Perl
1407has no way of knowing what the character at 0xFF is really supposed to
1408represent. Thus it disallows the operation. In this mode, the
1409lowercase of U+0178 is itself.
1410
1411The same problems ensue if you enable automatic UTF-8-ification of your
e199995e 1412standard file handles, default C<open()> layer, and C<@ARGV> on non-ISO8859-1,
b4ffc3db
TC
1413non-UTF-8 locales (by using either the B<-C> command line switch or the
1414C<PERL_UNICODE> environment variable; see L<perlrun>).
1415Things are read in as UTF-8, which would normally imply a Unicode
1416interpretation, but the presence of a locale causes them to be interpreted
1417in that locale instead. For example, a 0xD7 code point in the Unicode
1418input, which should mean the multiplication sign, won't be interpreted by
66cbab2c 1419Perl that way under the Greek locale. This is not a problem
b4ffc3db 1420I<provided> you make certain that all locales will always and only be either
66cbab2c 1421an ISO8859-1, or, if you don't have a deficient C library, a UTF-8 locale.
b4ffc3db 1422
1d2ab946
KW
1423Still another problem is that this approach can lead to two code
1424points meaning the same character. Thus in a Greek locale, both U+03A7
1425and U+00D7 are GREEK CAPITAL LETTER CHI.
1426
b4ffc3db
TC
1427Vendor locales are notoriously buggy, and it is difficult for Perl to test
1428its locale-handling code because this interacts with code that Perl has no
1429control over; therefore the locale-handling code in Perl may be buggy as
66cbab2c
KW
1430well. (However, the Unicode-supplied locales should be better, and
1431there is a feed back mechanism to correct any problems. See
1432L</Freely available locale definitions>.)
1433
7ee2ae1e 1434If you have Perl v5.16, the problems mentioned above go away if you use
66cbab2c 1435the C<:not_characters> parameter to the locale pragma (except for vendor
7ee2ae1e 1436bugs in the non-character portions). If you don't have v5.16, and you
66cbab2c
KW
1437I<do> have locales that work, using them may be worthwhile for certain
1438specific purposes, as long as you keep in mind the gotchas already
1439mentioned. For example, if the collation for your locales works, it
1440runs faster under locales than under L<Unicode::Collate>; and you gain
1441access to such things as the local currency symbol and the names of the
7ee2ae1e 1442months and days of the week. (But to hammer home the point, in v5.16,
66cbab2c
KW
1443you get this access without the downsides of locales by using the
1444C<:not_characters> form of the pragma.)
1445
1446Note: The policy of using locale rules for code points that can fit in a
1447byte, and Unicode rules for those that can't is not uniformly applied.
7ee2ae1e 1448Pre-v5.12, it was somewhat haphazard; in v5.12 it was applied fairly
66cbab2c 1449consistently to regular expression matching except for bracketed
7ee2ae1e
KW
1450character classes; in v5.14 it was extended to all regex matches; and in
1451v5.16 to the casing operations such as C<"\L"> and C<uc()>. For
66cbab2c
KW
1452collation, in all releases, the system's C<strxfrm()> function is called,
1453and whatever it does is what you get.
b310b053 1454
5f05dabc
PP
1455=head1 BUGS
1456
1457=head2 Broken systems
1458
5a964f20 1459In certain systems, the operating system's locale support
2bdf8add 1460is broken and cannot be fixed or used by Perl. Such deficiencies can
b4ffc3db 1461and will result in mysterious hangs and/or Perl core dumps when
2bdf8add 1462C<use locale> is in effect. When confronted with such a system,
7f2de2d2 1463please report in excruciating detail to <F<perlbug@perl.org>>, and
b4ffc3db 1464also contact your vendor: bug fixes may exist for these problems
2bdf8add 1465in your operating system. Sometimes such bug fixes are called an
83fb1bf0
KW
1466operating system upgrade. If you have the source for Perl, include in
1467the perlbug email the output of the test described above in L</Testing
1468for broken locales>.
5f05dabc
PP
1469
1470=head1 SEE ALSO
1471
b310b053
JH
1472L<I18N::Langinfo>, L<perluniintro>, L<perlunicode>, L<open>,
1473L<POSIX/isalnum>, L<POSIX/isalpha>,
4bbcc6e8
JH
1474L<POSIX/isdigit>, L<POSIX/isgraph>, L<POSIX/islower>,
1475L<POSIX/isprint>, L<POSIX/ispunct>, L<POSIX/isspace>,
1476L<POSIX/isupper>, L<POSIX/isxdigit>, L<POSIX/localeconv>,
1477L<POSIX/setlocale>, L<POSIX/strcoll>, L<POSIX/strftime>,
1478L<POSIX/strtod>, L<POSIX/strxfrm>.
5f05dabc 1479
ccd65d51
KW
1480For special considerations when Perl is embedded in a C program,
1481see L<perlembed/Using embedded Perl with POSIX locales>.
1482
5f05dabc
PP
1483=head1 HISTORY
1484
b0c42ed9 1485Jarkko Hietaniemi's original F<perli18n.pod> heavily hacked by Dominic
5a964f20 1486Dunlop, assisted by the perl5-porters. Prose worked over a bit by
c052850d 1487Tom Christiansen, and updated by Perl 5 porters.