This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
lib/locale.t: Add some test names
[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
ebc3223b 64The nbsp below makes this look better (though not great)
66cbab2c
KW
65
66E<160>
67
68=item Category LC_TIME: Date/Time formatting
69
70=for comment
ebc3223b 71The nbsp below makes this look better (though not great)
66cbab2c
KW
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
ebc3223b 186effect at the start of the program.
dfcc8045 187If there is no valid environment, the current locale is whatever the
65ebb059
KW
188system default has been set to. On POSIX systems, it is likely, but
189not necessarily, the "C" locale. On Windows, the default is set via the
190computer's S<C<Control Panel-E<gt>Regional and Language Options>> (or its
191current equivalent).
c052850d
KW
192
193The operations that are affected by locale are:
5f05dabc
PP
194
195=over 4
196
4c9b78f4 197=item B<Not within the scope of any C<"use locale"> variant>
b960a36e 198
1d2ab946
KW
199Only operations originating outside Perl should be affected, as follows:
200
201=over 4
202
203=item *
2619d284
KW
204
205The variable L<$!|perlvar/$ERRNO> (and its synonyms C<$ERRNO> and
206C<$OS_ERROR>) when used as strings always are in terms of the current
207locale.
208
1d2ab946
KW
209=item *
210
2619d284
KW
211The current locale is also used when going outside of Perl with
212operations like L<system()|perlfunc/system LIST> or
213L<qxE<sol>E<sol>|perlop/qxE<sol>STRINGE<sol>>, if those operations are
214locale-sensitive.
215
1d2ab946
KW
216=item *
217
2619d284
KW
218Also Perl gives access to various C library functions through the
219L<POSIX> module. Some of those functions are always affected by the
220current locale. For example, C<POSIX::strftime()> uses C<LC_TIME>;
221C<POSIX::strtod()> uses C<LC_NUMERIC>; C<POSIX::strcoll()> and
222C<POSIX::strxfrm()> use C<LC_COLLATE>; and character classification
223functions like C<POSIX::isalnum()> use C<LC_CTYPE>. All such functions
224will behave according to the current underlying locale, even if that
1d2ab946 225locale isn't exposed to Perl space.
2619d284 226
1d2ab946
KW
227=item *
228
229Perl also provides lite wrappers for XS modules to use some C library
230C<printf> functions. These wrappers don't do anything with the locale,
231and the underlying C library function is affected by the locale in
232effect at the time of the wrapper call.
233The affected functions are
234L<perlapi/my_sprintf>,
235L<perlapi/my_snprintf>,
236and
237L<perlapi/my_vsnprintf>.
238
239=back
240
ebc3223b
KW
241=for comment
242The nbsp below makes this look better (though not great)
243
244E<160>
245
246=item B<Lingering effects of C<S<use locale>>>
1d2ab946
KW
247
248Certain Perl operations that are set-up within the scope of a
b960a36e
KW
249C<use locale> variant retain that effect even outside the scope.
250These include:
251
252=over 4
253
254=item *
255
256The output format of a L<write()|perlfunc/write> is determined by an
257earlier format declaration (L<perlfunc/format>), so whether or not the
258output is affected by locale is determined by if the C<format()> is
259within the scope of a C<use locale> variant, not whether the C<write()>
260is.
261
262=item *
263
264Regular expression patterns can be compiled using
265L<qrE<sol>E<sol>|perlop/qrE<sol>STRINGE<sol>msixpodual> with actual
266matching deferred to later. Again, it is whether or not the compilation
267was done within the scope of C<use locale> that determines the match
268behavior, not if the matches are done within such a scope or not.
269
270=back
271
ebc3223b
KW
272=for comment
273The nbsp below makes this look better (though not great)
274
275E<160>
276
4c9b78f4 277=item B<Under C<"use locale ':not_characters';">>
66cbab2c
KW
278
279=over 4
280
281=item *
282
b960a36e
KW
283All the non-Perl operations.
284
285=item *
286
287B<Format declarations> (L<perlfunc/format>) and hence any subsequent
288C<write()>s use C<LC_NUMERIC>.
66cbab2c
KW
289
290=item *
291
b960a36e
KW
292B<stringification and output> use C<LC_NUMERIC>.
293These include the results of
294C<print()>,
295C<printf()>,
296C<say()>,
297and
298C<sprintf()>.
66cbab2c
KW
299
300=back
301
302=for comment
ebc3223b 303The nbsp below makes this look better (though not great)
66cbab2c
KW
304
305E<160>
306
4c9b78f4 307=item B<Under just plain C<"use locale";>>
66cbab2c 308
66cbab2c
KW
309=over 4
310
5f05dabc
PP
311=item *
312
b960a36e
KW
313All the above operations
314
315=item *
316
317B<The comparison operators> (C<lt>, C<le>, C<cmp>, C<ge>, and C<gt>) use
39332f68 318C<LC_COLLATE>. C<sort()> is also affected if used without an
5a964f20 319explicit comparison function, because it uses C<cmp> by default.
14280422 320
5a964f20 321B<Note:> C<eq> and C<ne> are unaffected by locale: they always
de108802 322perform a char-by-char comparison of their scalar operands. What's
14280422
DD
323more, if C<cmp> finds that its operands are equal according to the
324collation sequence specified by the current locale, it goes on to
de108802
RGS
325perform a char-by-char comparison, and only returns I<0> (equal) if the
326operands are char-for-char identical. If you really want to know whether
5a964f20 327two strings--which C<eq> and C<cmp> may consider different--are equal
14280422
DD
328as far as collation in the locale is concerned, see the discussion in
329L<Category LC_COLLATE: Collation>.
5f05dabc
PP
330
331=item *
332
39332f68
KW
333B<Regular expressions and case-modification functions> (C<uc()>, C<lc()>,
334C<ucfirst()>, and C<lcfirst()>) use C<LC_CTYPE>
5f05dabc 335
5f05dabc
PP
336=back
337
66cbab2c 338=back
5f05dabc 339
5a964f20 340The default behavior is restored with the S<C<no locale>> pragma, or
ef3087ec 341upon reaching the end of the block enclosing C<use locale>.
66cbab2c
KW
342Note that C<use locale> and C<use locale ':not_characters'> may be
343nested, and that what is in effect within an inner scope will revert to
344the outer scope's rules at the end of the inner scope.
5f05dabc 345
5a964f20 346The string result of any operation that uses locale
14280422
DD
347information is tainted, as it is possible for a locale to be
348untrustworthy. See L<"SECURITY">.
5f05dabc
PP
349
350=head2 The setlocale function
351
14280422 352You can switch locales as often as you wish at run time with the
39332f68 353C<POSIX::setlocale()> function:
5f05dabc 354
5f05dabc
PP
355 # Import locale-handling tool set from POSIX module.
356 # This example uses: setlocale -- the function call
357 # LC_CTYPE -- explained below
2619d284
KW
358 # (Showing the testing for success/failure of operations is
359 # omitted in these examples to avoid distracting from the main
ebc3223b 360 # point)
6ea81ccf 361
5f05dabc 362 use POSIX qw(locale_h);
dfcc8045
KW
363 use locale;
364 my $old_locale;
5f05dabc 365
14280422 366 # query and save the old locale
5f05dabc
PP
367 $old_locale = setlocale(LC_CTYPE);
368
369 setlocale(LC_CTYPE, "fr_CA.ISO8859-1");
370 # LC_CTYPE now in locale "French, Canada, codeset ISO 8859-1"
371
372 setlocale(LC_CTYPE, "");
65ebb059
KW
373 # LC_CTYPE now reset to the default defined by the
374 # LC_ALL/LC_CTYPE/LANG environment variables, or to the system
375 # default. See below for documentation.
5f05dabc
PP
376
377 # restore the old locale
378 setlocale(LC_CTYPE, $old_locale);
379
39332f68 380The first argument of C<setlocale()> gives the B<category>, the second the
14280422
DD
381B<locale>. The category tells in what aspect of data processing you
382want to apply locale-specific rules. Category names are discussed in
66cbab2c 383L</LOCALE CATEGORIES> and L</"ENVIRONMENT">. The locale is the name of a
14280422
DD
384collection of customization information corresponding to a particular
385combination of language, country or territory, and codeset. Read on for
386hints on the naming of locales: not all systems name locales as in the
387example.
388
39332f68 389If no second argument is provided and the category is something other
502a173a
JH
390than LC_ALL, the function returns a string naming the current locale
391for the category. You can use this value as the second argument in a
f170b852
KW
392subsequent call to C<setlocale()>, B<but> on some platforms the string
393is opaque, not something that most people would be able to decipher as
394to what locale it means.
502a173a
JH
395
396If no second argument is provided and the category is LC_ALL, the
397result is implementation-dependent. It may be a string of
c052850d 398concatenated locale names (separator also implementation-dependent)
39332f68 399or a single locale name. Please consult your L<setlocale(3)> man page for
502a173a
JH
400details.
401
402If a second argument is given and it corresponds to a valid locale,
403the locale for the category is set to that value, and the function
404returns the now-current locale value. You can then use this in yet
39332f68 405another call to C<setlocale()>. (In some implementations, the return
502a173a
JH
406value may sometimes differ from the value you gave as the second
407argument--think of it as an alias for the value you gave.)
5f05dabc
PP
408
409As the example shows, if the second argument is an empty string, the
410category's locale is returned to the default specified by the
411corresponding environment variables. Generally, this results in a
5a964f20 412return to the default that was in force when Perl started up: changes
54310121 413to the environment made by the application after startup may or may not
5a964f20 414be noticed, depending on your system's C library.
5f05dabc 415
66cbab2c
KW
416Note that Perl ignores the current C<LC_CTYPE> and C<LC_COLLATE> locales
417within the scope of a C<use locale ':not_characters'>.
418
f170b852 419If C<set_locale()> fails for some reason (for example, an attempt to set
dfcc8045
KW
420to a locale unknown to the system), the locale for the category is not
421changed, and the function returns C<undef>.
422
2619d284 423
39332f68 424For further information about the categories, consult L<setlocale(3)>.
3e6e419a
JH
425
426=head2 Finding locales
427
39332f68 428For locales available in your system, consult also L<setlocale(3)> to
5a964f20
TC
429see whether it leads to the list of available locales (search for the
430I<SEE ALSO> section). If that fails, try the following command lines:
5f05dabc
PP
431
432 locale -a
433
434 nlsinfo
435
436 ls /usr/lib/nls/loc
437
438 ls /usr/lib/locale
439
440 ls /usr/lib/nls
441
b478f28d
JH
442 ls /usr/share/locale
443
5f05dabc
PP
444and see whether they list something resembling these
445
2bdf8add 446 en_US.ISO8859-1 de_DE.ISO8859-1 ru_RU.ISO8859-5
502a173a 447 en_US.iso88591 de_DE.iso88591 ru_RU.iso88595
2bdf8add 448 en_US de_DE ru_RU
14280422 449 en de ru
2bdf8add
JH
450 english german russian
451 english.iso88591 german.iso88591 russian.iso88595
502a173a 452 english.roman8 russian.koi8r
5f05dabc 453
39332f68 454Sadly, even though the calling interface for C<setlocale()> has been
528d65ad 455standardized, names of locales and the directories where the
5a964f20 456configuration resides have not been. The basic form of the name is
528d65ad
JH
457I<language_territory>B<.>I<codeset>, but the latter parts after
458I<language> are not always present. The I<language> and I<country>
459are usually from the standards B<ISO 3166> and B<ISO 639>, the
460two-letter abbreviations for the countries and the languages of the
461world, respectively. The I<codeset> part often mentions some B<ISO
4628859> character set, the Latin codesets. For example, C<ISO 8859-1>
463is the so-called "Western European codeset" that can be used to encode
464most Western European languages adequately. Again, there are several
465ways to write even the name of that one standard. Lamentably.
5f05dabc 466
14280422
DD
467Two special locales are worth particular mention: "C" and "POSIX".
468Currently these are effectively the same locale: the difference is
5a964f20
TC
469mainly that the first one is defined by the C standard, the second by
470the POSIX standard. They define the B<default locale> in which
14280422 471every program starts in the absence of locale information in its
5a964f20 472environment. (The I<default> default locale, if you will.) Its language
39332f68
KW
473is (American) English and its character codeset ASCII or, rarely, a
474superset thereof (such as the "DEC Multinational Character Set
475(DEC-MCS)"). B<Warning>. The C locale delivered by some vendors
476may not actually exactly match what the C standard calls for. So
477beware.
5f05dabc 478
14280422
DD
479B<NOTE>: Not all systems have the "POSIX" locale (not all systems are
480POSIX-conformant), so use "C" when you need explicitly to specify this
481default locale.
5f05dabc 482
3e6e419a
JH
483=head2 LOCALE PROBLEMS
484
5a964f20 485You may encounter the following warning message at Perl startup:
3e6e419a
JH
486
487 perl: warning: Setting locale failed.
488 perl: warning: Please check that your locale settings:
489 LC_ALL = "En_US",
490 LANG = (unset)
491 are supported and installed on your system.
492 perl: warning: Falling back to the standard locale ("C").
493
5a964f20
TC
494This means that your locale settings had LC_ALL set to "En_US" and
495LANG exists but has no value. Perl tried to believe you but could not.
496Instead, Perl gave up and fell back to the "C" locale, the default locale
65ebb059
KW
497that is supposed to work no matter what. (On Windows, it first tries
498falling back to the system default locale.) This usually means your
499locale settings were wrong, they mention locales your system has never
500heard of, or the locale installation in your system has problems (for
501example, some system files are broken or missing). There are quick and
502temporary fixes to these problems, as well as more thorough and lasting
503fixes.
3e6e419a 504
83fb1bf0
KW
505=head2 Testing for broken locales
506
507If you are building Perl from source, the Perl test suite file
508F<lib/locale.t> can be used to test the locales on your system.
509Setting the environment variable C<PERL_DEBUG_FULL_TEST> to 1
510will cause it to output detailed results. For example, on Linux, you
511could say
512
1d2ab946 513 PERL_DEBUG_FULL_TEST=1 ./perl -T -Ilib lib/locale.t > locale.log 2>&1
83fb1bf0
KW
514
515Besides many other tests, it will test every locale it finds on your
516system to see if they conform to the POSIX standard. If any have
517errors, it will include a summary near the end of the output of which
518locales passed all its tests, and which failed, and why.
519
3e6e419a
JH
520=head2 Temporarily fixing locale problems
521
5a964f20 522The two quickest fixes are either to render Perl silent about any
3e6e419a
JH
523locale inconsistencies or to run Perl under the default locale "C".
524
525Perl's moaning about locale problems can be silenced by setting the
900bd440
JH
526environment variable PERL_BADLANG to a zero value, for example "0".
527This method really just sweeps the problem under the carpet: you tell
528Perl to shut up even when Perl sees that something is wrong. Do not
529be surprised if later something locale-dependent misbehaves.
3e6e419a
JH
530
531Perl can be run under the "C" locale by setting the environment
5a964f20
TC
532variable LC_ALL to "C". This method is perhaps a bit more civilized
533than the PERL_BADLANG approach, but setting LC_ALL (or
534other locale variables) may affect other programs as well, not just
535Perl. In particular, external programs run from within Perl will see
3e6e419a 536these changes. If you make the new settings permanent (read on), all
f979aebc 537programs you run see the changes. See L<"ENVIRONMENT"> for
5a964f20 538the full list of relevant environment variables and L<USING LOCALES>
e05ffc7d 539for their effects in Perl. Effects in other programs are
5a964f20 540easily deducible. For example, the variable LC_COLLATE may well affect
b432a672 541your B<sort> program (or whatever the program that arranges "records"
3e6e419a
JH
542alphabetically in your system is called).
543
5a964f20
TC
544You can test out changing these variables temporarily, and if the
545new settings seem to help, put those settings into your shell startup
546files. Consult your local documentation for the exact details. For in
547Bourne-like shells (B<sh>, B<ksh>, B<bash>, B<zsh>):
3e6e419a
JH
548
549 LC_ALL=en_US.ISO8859-1
550 export LC_ALL
551
5a964f20
TC
552This assumes that we saw the locale "en_US.ISO8859-1" using the commands
553discussed above. We decided to try that instead of the above faulty
554locale "En_US"--and in Cshish shells (B<csh>, B<tcsh>)
3e6e419a
JH
555
556 setenv LC_ALL en_US.ISO8859-1
c47ff5f1 557
c406981e
JH
558or if you have the "env" application you can do in any shell
559
560 env LC_ALL=en_US.ISO8859-1 perl ...
561
5a964f20 562If you do not know what shell you have, consult your local
3e6e419a
JH
563helpdesk or the equivalent.
564
565=head2 Permanently fixing locale problems
566
5a964f20
TC
567The slower but superior fixes are when you may be able to yourself
568fix the misconfiguration of your own environment variables. The
3e6e419a
JH
569mis(sing)configuration of the whole system's locales usually requires
570the help of your friendly system administrator.
571
5a964f20
TC
572First, see earlier in this document about L<Finding locales>. That tells
573how to find which locales are really supported--and more importantly,
574installed--on your system. In our example error message, environment
575variables affecting the locale are listed in the order of decreasing
576importance (and unset variables do not matter). Therefore, having
577LC_ALL set to "En_US" must have been the bad choice, as shown by the
578error message. First try fixing locale settings listed first.
3e6e419a 579
5a964f20
TC
580Second, if using the listed commands you see something B<exactly>
581(prefix matches do not count and case usually counts) like "En_US"
582without the quotes, then you should be okay because you are using a
583locale name that should be installed and available in your system.
4a4eefd0 584In this case, see L<Permanently fixing your system's locale configuration>.
3e6e419a 585
4a4eefd0 586=head2 Permanently fixing your system's locale configuration
3e6e419a 587
5a964f20 588This is when you see something like:
3e6e419a
JH
589
590 perl: warning: Please check that your locale settings:
591 LC_ALL = "En_US",
592 LANG = (unset)
593 are supported and installed on your system.
594
595but then cannot see that "En_US" listed by the above-mentioned
5a964f20
TC
596commands. You may see things like "en_US.ISO8859-1", but that isn't
597the same. In this case, try running under a locale
598that you can list and which somehow matches what you tried. The
3e6e419a 599rules for matching locale names are a bit vague because
e05ffc7d 600standardization is weak in this area. See again the
13a2d996 601L<Finding locales> about general rules.
3e6e419a 602
b687b08b 603=head2 Fixing system locale configuration
3e6e419a 604
5a964f20
TC
605Contact a system administrator (preferably your own) and report the exact
606error message you get, and ask them to read this same documentation you
607are now reading. They should be able to check whether there is something
608wrong with the locale configuration of the system. The L<Finding locales>
609section is unfortunately a bit vague about the exact commands and places
610because these things are not that standardized.
3e6e419a 611
5f05dabc
PP
612=head2 The localeconv function
613
39332f68 614The C<POSIX::localeconv()> function allows you to get particulars of the
14280422
DD
615locale-dependent numeric formatting information specified by the current
616C<LC_NUMERIC> and C<LC_MONETARY> locales. (If you just want the name of
39332f68 617the current locale for a particular category, use C<POSIX::setlocale()>
5a964f20 618with a single parameter--see L<The setlocale function>.)
5f05dabc
PP
619
620 use POSIX qw(locale_h);
5f05dabc
PP
621
622 # Get a reference to a hash of locale-dependent info
623 $locale_values = localeconv();
624
625 # Output sorted list of the values
626 for (sort keys %$locale_values) {
14280422 627 printf "%-20s = %s\n", $_, $locale_values->{$_}
5f05dabc
PP
628 }
629
39332f68 630C<localeconv()> takes no arguments, and returns B<a reference to> a hash.
5a964f20 631The keys of this hash are variable names for formatting, such as
502a173a 632C<decimal_point> and C<thousands_sep>. The values are the
cea6626f 633corresponding, er, values. See L<POSIX/localeconv> for a longer
502a173a
JH
634example listing the categories an implementation might be expected to
635provide; some provide more and others fewer. You don't need an
39332f68 636explicit C<use locale>, because C<localeconv()> always observes the
502a173a 637current locale.
5f05dabc 638
5a964f20
TC
639Here's a simple-minded example program that rewrites its command-line
640parameters as integers correctly formatted in the current locale:
5f05dabc 641
ef3087ec
KW
642 use POSIX qw(locale_h);
643
644 # Get some of locale's numeric formatting parameters
645 my ($thousands_sep, $grouping) =
646 @{localeconv()}{'thousands_sep', 'grouping'};
647
648 # Apply defaults if values are missing
649 $thousands_sep = ',' unless $thousands_sep;
650
651 # grouping and mon_grouping are packed lists
652 # of small integers (characters) telling the
653 # grouping (thousand_seps and mon_thousand_seps
654 # being the group dividers) of numbers and
655 # monetary quantities. The integers' meanings:
656 # 255 means no more grouping, 0 means repeat
657 # the previous grouping, 1-254 means use that
658 # as the current grouping. Grouping goes from
659 # right to left (low to high digits). In the
660 # below we cheat slightly by never using anything
661 # else than the first grouping (whatever that is).
662 if ($grouping) {
663 @grouping = unpack("C*", $grouping);
664 } else {
665 @grouping = (3);
666 }
667
668 # Format command line params for current locale
669 for (@ARGV) {
670 $_ = int; # Chop non-integer part
671 1 while
672 s/(\d)(\d{$grouping[0]}($|$thousands_sep))/$1$thousands_sep$2/;
673 print "$_";
674 }
675 print "\n";
5f05dabc 676
74c76037 677=head2 I18N::Langinfo
4bbcc6e8
JH
678
679Another interface for querying locale-dependent information is the
39332f68 680C<I18N::Langinfo::langinfo()> function, available at least in Unix-like
4bbcc6e8
JH
681systems and VMS.
682
39332f68
KW
683The following example will import the C<langinfo()> function itself and
684three constants to be used as arguments to C<langinfo()>: a constant for
74c76037
JH
685the abbreviated first day of the week (the numbering starts from
686Sunday = 1) and two more constants for the affirmative and negative
687answers for a yes/no question in the current locale.
4bbcc6e8 688
74c76037 689 use I18N::Langinfo qw(langinfo ABDAY_1 YESSTR NOSTR);
4bbcc6e8 690
ef3087ec
KW
691 my ($abday_1, $yesstr, $nostr)
692 = map { langinfo } qw(ABDAY_1 YESSTR NOSTR);
4bbcc6e8 693
74c76037 694 print "$abday_1? [$yesstr/$nostr] ";
4bbcc6e8 695
74c76037
JH
696In other words, in the "C" (or English) locale the above will probably
697print something like:
698
e05ffc7d 699 Sun? [yes/no]
4bbcc6e8
JH
700
701See L<I18N::Langinfo> for more information.
702
5f05dabc
PP
703=head1 LOCALE CATEGORIES
704
5a964f20
TC
705The following subsections describe basic locale categories. Beyond these,
706some combination categories allow manipulation of more than one
707basic category at a time. See L<"ENVIRONMENT"> for a discussion of these.
5f05dabc
PP
708
709=head2 Category LC_COLLATE: Collation
710
66cbab2c
KW
711In the scope of S<C<use locale>> (but not a
712C<use locale ':not_characters'>), Perl looks to the C<LC_COLLATE>
5a964f20 713environment variable to determine the application's notions on collation
b4ffc3db
TC
714(ordering) of characters. For example, "b" follows "a" in Latin
715alphabets, but where do "E<aacute>" and "E<aring>" belong? And while
f87fa335 716"color" follows "chocolate" in English, what about in traditional Spanish?
5f05dabc 717
60f0fa02
JH
718The following collations all make sense and you may meet any of them
719if you "use locale".
720
721 A B C D E a b c d e
35316ca3 722 A a B b C c D d E e
60f0fa02
JH
723 a A b B c C d D e E
724 a b c d e A B C D E
725
f1cbbd6e 726Here is a code snippet to tell what "word"
5a964f20 727characters are in the current locale, in that locale's order:
5f05dabc
PP
728
729 use locale;
35316ca3 730 print +(sort grep /\w/, map { chr } 0..255), "\n";
5f05dabc 731
14280422
DD
732Compare this with the characters that you see and their order if you
733state explicitly that the locale should be ignored:
5f05dabc
PP
734
735 no locale;
35316ca3 736 print +(sort grep /\w/, map { chr } 0..255), "\n";
5f05dabc
PP
737
738This machine-native collation (which is what you get unless S<C<use
739locale>> has appeared earlier in the same block) must be used for
740sorting raw binary data, whereas the locale-dependent collation of the
b0c42ed9 741first example is useful for natural text.
5f05dabc 742
14280422
DD
743As noted in L<USING LOCALES>, C<cmp> compares according to the current
744collation locale when C<use locale> is in effect, but falls back to a
de108802 745char-by-char comparison for strings that the locale says are equal. You
39332f68 746can use C<POSIX::strcoll()> if you don't want this fall-back:
14280422
DD
747
748 use POSIX qw(strcoll);
749 $equal_in_locale =
750 !strcoll("space and case ignored", "SpaceAndCaseIgnored");
751
39332f68 752C<$equal_in_locale> will be true if the collation locale specifies a
5a964f20 753dictionary-like ordering that ignores space characters completely and
9e3a2af8 754which folds case.
14280422 755
31f05a37
KW
756Perl only supports single-byte locales for C<LC_COLLATE>. This means
757that a UTF-8 locale likely will just give you machine-native ordering.
758Use L<Unicode::Collate> for the full implementation of the Unicode
759Collation Algorithm.
760
5a964f20 761If you have a single string that you want to check for "equality in
14280422 762locale" against several others, you might think you could gain a little
39332f68 763efficiency by using C<POSIX::strxfrm()> in conjunction with C<eq>:
14280422
DD
764
765 use POSIX qw(strxfrm);
766 $xfrm_string = strxfrm("Mixed-case string");
767 print "locale collation ignores spaces\n"
768 if $xfrm_string eq strxfrm("Mixed-casestring");
769 print "locale collation ignores hyphens\n"
770 if $xfrm_string eq strxfrm("Mixedcase string");
771 print "locale collation ignores case\n"
772 if $xfrm_string eq strxfrm("mixed-case string");
773
39332f68 774C<strxfrm()> takes a string and maps it into a transformed string for use
de108802 775in char-by-char comparisons against other transformed strings during
14280422 776collation. "Under the hood", locale-affected Perl comparison operators
39332f68
KW
777call C<strxfrm()> for both operands, then do a char-by-char
778comparison of the transformed strings. By calling C<strxfrm()> explicitly
14280422 779and using a non locale-affected comparison, the example attempts to save
5a964f20 780a couple of transformations. But in fact, it doesn't save anything: Perl
2ae324a7 781magic (see L<perlguts/Magic Variables>) creates the transformed version of a
5a964f20 782string the first time it's needed in a comparison, then keeps this version around
14280422 783in case it's needed again. An example rewritten the easy way with
e38874e2 784C<cmp> runs just about as fast. It also copes with null characters
39332f68 785embedded in strings; if you call C<strxfrm()> directly, it treats the first
5a964f20
TC
786null it finds as a terminator. don't expect the transformed strings
787it produces to be portable across systems--or even from one revision
39332f68 788of your operating system to the next. In short, don't call C<strxfrm()>
e38874e2 789directly: let Perl do it for you.
14280422 790
5a964f20 791Note: C<use locale> isn't shown in some of these examples because it isn't
dfcc8045
KW
792needed: C<strcoll()> and C<strxfrm()> are POSIX functions
793which use the standard system-supplied C<libc> functions that
794always obey the current C<LC_COLLATE> locale.
5f05dabc
PP
795
796=head2 Category LC_CTYPE: Character Types
797
66cbab2c
KW
798In the scope of S<C<use locale>> (but not a
799C<use locale ':not_characters'>), Perl obeys the C<LC_CTYPE> locale
14280422 800setting. This controls the application's notion of which characters are
ebc3223b
KW
801alphabetic, numeric, punctuation, I<etc>. This affects Perl's C<\w>
802regular expression metanotation,
f1cbbd6e 803which stands for alphanumeric characters--that is, alphabetic,
ebc3223b
KW
804numeric, and the platform's native underscore.
805(Consult L<perlre> for more information about
14280422 806regular expressions.) Thanks to C<LC_CTYPE>, depending on your locale
b4ffc3db
TC
807setting, characters like "E<aelig>", "E<eth>", "E<szlig>", and
808"E<oslash>" may be understood as C<\w> characters.
ebc3223b
KW
809It also affects things like C<\s>, C<\D>, and the POSIX character
810classes, like C<[[:graph:]]>. (See L<perlrecharclass> for more
811information on all these.)
5f05dabc 812
2c268ad5 813The C<LC_CTYPE> locale also provides the map used in transliterating
68dc0745 814characters between lower and uppercase. This affects the case-mapping
39332f68 815functions--C<fc()>, C<lc()>, C<lcfirst()>, C<uc()>, and C<ucfirst()>; case-mapping
b9cc4f69
KW
816interpolation with C<\F>, C<\l>, C<\L>, C<\u>, or C<\U> in double-quoted
817strings and C<s///> substitutions; and case-independent regular expression
e38874e2
DD
818pattern matching using the C<i> modifier.
819
2da736a2 820Finally, C<LC_CTYPE> affects the (deprecated) POSIX character-class test
1d2ab946
KW
821functions--C<POSIX::isalpha()>, C<POSIX::islower()>, and so on. For
822example, if you move from the "C" locale to a 7-bit Scandinavian one,
823you may find--possibly to your surprise--that "|" moves from the
824C<POSIX::ispunct()> class to C<POSIX::isalpha()>.
ef3087ec
KW
825Unfortunately, this creates big problems for regular expressions. "|" still
826means alternation even though it matches C<\w>.
5f05dabc 827
31f05a37
KW
828Starting in v5.20, Perl supports UTF-8 locales for C<LC_CTYPE>, but
829otherwise Perl only supports single-byte locales, such as the ISO 8859
830series. This means that wide character locales, for example for Asian
831languages, are not supported. The UTF-8 locale support is actually a
832superset of POSIX locales, because it is really full Unicode behavior
833as if no locale were in effect at all (except for tainting; see
834L</SECURITY>). POSIX locales, even UTF-8 ones,
835are lacking certain concepts in Unicode, such as the idea that changing
836the case of a character could expand to be more than one character.
837Perl in a UTF-8 locale, will give you that expansion. Prior to v5.20,
838Perl treated a UTF-8 locale on some platforms like an ISO 8859-1 one,
839with some restrictions, and on other platforms more like the "C" locale.
840For releases v5.16 and v5.18, C<S<use locale 'not_characters>> could be
841used as a workaround for this (see L</Unicode and UTF-8>).
842
5d63e270
KW
843Note that there are quite a few things that are unaffected by the
844current locale. All the escape sequences for particular characters,
845C<\n> for example, always mean the platform's native one. This means,
846for example, that C<\N> in regular expressions (every character
1d2ab946 847but new-line) works on the platform character set.
5d63e270 848
14280422
DD
849B<Note:> A broken or malicious C<LC_CTYPE> locale definition may result
850in clearly ineligible characters being considered to be alphanumeric by
e199995e 851your application. For strict matching of (mundane) ASCII letters and
5a964f20 852digits--for example, in command strings--locale-aware applications
e199995e 853should use C<\w> with the C</a> regular expression modifier. See L<"SECURITY">.
5f05dabc
PP
854
855=head2 Category LC_NUMERIC: Numeric Formatting
856
b960a36e
KW
857After a proper C<POSIX::setlocale()> call, and within the scope of one
858of the C<use locale> variants, Perl obeys the C<LC_NUMERIC>
2095dafa 859locale information, which controls an application's idea of how numbers
b960a36e
KW
860should be formatted for human readability.
861In most implementations the only effect is to
b4ffc3db 862change the character used for the decimal point--perhaps from "." to ",".
b960a36e 863The functions aren't aware of such niceties as thousands separation and
2095dafa 864so on. (See L<The localeconv function> if you care about these things.)
5a964f20 865
b960a36e
KW
866 use POSIX qw(strtod setlocale LC_NUMERIC);
867 use locale;
5f05dabc 868
b960a36e 869 setlocale LC_NUMERIC, "";
14280422 870
b960a36e 871 $n = 5/2; # Assign numeric 2.5 to $n
5f05dabc 872
b960a36e 873 $a = " $n"; # Locale-dependent conversion to string
5f05dabc 874
b960a36e 875 print "half five is $n\n"; # Locale-dependent output
5f05dabc 876
b960a36e 877 printf "half five is %g\n", $n; # Locale-dependent output
5f05dabc 878
b960a36e
KW
879 print "DECIMAL POINT IS COMMA\n"
880 if $n == (strtod("2,5"))[0]; # Locale-dependent conversion
5f05dabc 881
4bbcc6e8
JH
882See also L<I18N::Langinfo> and C<RADIXCHAR>.
883
5f05dabc
PP
884=head2 Category LC_MONETARY: Formatting of monetary amounts
885
e199995e 886The C standard defines the C<LC_MONETARY> category, but not a function
5a964f20 887that is affected by its contents. (Those with experience of standards
b0c42ed9 888committees will recognize that the working group decided to punt on the
fa9b773e
KW
889issue.) Consequently, Perl essentially takes no notice of it. If you
890really want to use C<LC_MONETARY>, you can query its contents--see
e05ffc7d
KW
891L<The localeconv function>--and use the information that it returns in your
892application's own formatting of currency amounts. However, you may well
893find that the information, voluminous and complex though it may be, still
894does not quite meet your requirements: currency formatting is a hard nut
13a2d996 895to crack.
5f05dabc 896
4bbcc6e8
JH
897See also L<I18N::Langinfo> and C<CRNCYSTR>.
898
5f05dabc
PP
899=head2 LC_TIME
900
39332f68 901Output produced by C<POSIX::strftime()>, which builds a formatted
5f05dabc
PP
902human-readable date/time string, is affected by the current C<LC_TIME>
903locale. Thus, in a French locale, the output produced by the C<%B>
904format element (full month name) for the first month of the year would
5a964f20 905be "janvier". Here's how to get a list of long month names in the
5f05dabc
PP
906current locale:
907
908 use POSIX qw(strftime);
14280422
DD
909 for (0..11) {
910 $long_month_name[$_] =
911 strftime("%B", 0, 0, 0, 1, $_, 96);
5f05dabc
PP
912 }
913
2619d284
KW
914Note: C<use locale> isn't needed in this example: C<strftime()> is a POSIX
915function which uses the standard system-supplied C<libc> function that
916always obeys the current C<LC_TIME> locale.
5f05dabc 917
4bbcc6e8 918See also L<I18N::Langinfo> and C<ABDAY_1>..C<ABDAY_7>, C<DAY_1>..C<DAY_7>,
2a2bf5f4 919C<ABMON_1>..C<ABMON_12>, and C<ABMON_1>..C<ABMON_12>.
4bbcc6e8 920
5f05dabc
PP
921=head2 Other categories
922
2619d284
KW
923The remaining locale categories are not currently used by Perl itself.
924But again note that things Perl interacts with may use these, including
925extensions outside the standard Perl distribution, and by the
98a6f11e 926operating system and its utilities. Note especially that the string
927value of C<$!> and the error messages given by external utilities may
928be changed by C<LC_MESSAGES>. If you want to have portable error
265f5c4a 929codes, use C<%!>. See L<Errno>.
14280422
DD
930
931=head1 SECURITY
932
5a964f20 933Although the main discussion of Perl security issues can be found in
14280422
DD
934L<perlsec>, a discussion of Perl's locale handling would be incomplete
935if it did not draw your attention to locale-dependent security issues.
5a964f20
TC
936Locales--particularly on systems that allow unprivileged users to
937build their own locales--are untrustworthy. A malicious (or just plain
14280422
DD
938broken) locale can make a locale-aware application give unexpected
939results. Here are a few possibilities:
940
941=over 4
942
943=item *
944
945Regular expression checks for safe file names or mail addresses using
5a964f20 946C<\w> may be spoofed by an C<LC_CTYPE> locale that claims that
14280422
DD
947characters such as "E<gt>" and "|" are alphanumeric.
948
949=item *
950
e38874e2
DD
951String interpolation with case-mapping, as in, say, C<$dest =
952"C:\U$name.$ext">, may produce dangerous results if a bogus LC_CTYPE
953case-mapping table is in effect.
954
955=item *
956
14280422
DD
957A sneaky C<LC_COLLATE> locale could result in the names of students with
958"D" grades appearing ahead of those with "A"s.
959
960=item *
961
5a964f20 962An application that takes the trouble to use information in
14280422 963C<LC_MONETARY> may format debits as if they were credits and vice versa
5a964f20 964if that locale has been subverted. Or it might make payments in US
14280422
DD
965dollars instead of Hong Kong dollars.
966
967=item *
968
39332f68 969The date and day names in dates formatted by C<strftime()> could be
14280422 970manipulated to advantage by a malicious user able to subvert the
5a964f20 971C<LC_DATE> locale. ("Look--it says I wasn't in the building on
14280422
DD
972Sunday.")
973
974=back
975
976Such dangers are not peculiar to the locale system: any aspect of an
5a964f20 977application's environment which may be modified maliciously presents
14280422 978similar challenges. Similarly, they are not specific to Perl: any
5a964f20 979programming language that allows you to write programs that take
14280422
DD
980account of their environment exposes you to these issues.
981
5a964f20
TC
982Perl cannot protect you from all possibilities shown in the
983examples--there is no substitute for your own vigilance--but, when
14280422 984C<use locale> is in effect, Perl uses the tainting mechanism (see
5a964f20 985L<perlsec>) to mark string results that become locale-dependent, and
14280422 986which may be untrustworthy in consequence. Here is a summary of the
5a964f20 987tainting behavior of operators and functions that may be affected by
14280422
DD
988the locale:
989
990=over 4
991
551e1d92
RB
992=item *
993
994B<Comparison operators> (C<lt>, C<le>, C<ge>, C<gt> and C<cmp>):
14280422
DD
995
996Scalar true/false (or less/equal/greater) result is never tainted.
997
551e1d92
RB
998=item *
999
1d2ab946 1000B<Case-mapping interpolation> (with C<\l>, C<\L>, C<\u>, C<\U>, or C<\F>)
e38874e2
DD
1001
1002Result string containing interpolated material is tainted if
66cbab2c 1003C<use locale> (but not S<C<use locale ':not_characters'>>) is in effect.
e38874e2 1004
551e1d92
RB
1005=item *
1006
1007B<Matching operator> (C<m//>):
14280422
DD
1008
1009Scalar true/false result never tainted.
1010
1d2ab946
KW
1011All subpatterns, either delivered as a list-context result or as C<$1>
1012I<etc>., are tainted if C<use locale> (but not
1013S<C<use locale ':not_characters'>>) is in effect, and the subpattern
1014regular expression is matched case-insensitively (C</i>) or contains a
1015locale-dependent construct. These constructs include C<\w>
1016(to match an alphanumeric character), C<\W> (non-alphanumeric
1017character), C<\s> (whitespace character), C<\S> (non whitespace
1018character), and the POSIX character classes, such as C<[:alpha:]> (see
1019L<perlrecharclass/POSIX Character Classes>).
1020The matched-pattern variables, C<$&>, C<$`> (pre-match), C<$'>
1021(post-match), and C<$+> (last match) also are tainted.
1022(Note that currently there are some bugs where not everything that
1023should be tainted gets tainted in all circumstances.)
14280422 1024
551e1d92
RB
1025=item *
1026
1027B<Substitution operator> (C<s///>):
14280422 1028
e38874e2 1029Has the same behavior as the match operator. Also, the left
66cbab2c
KW
1030operand of C<=~> becomes tainted when C<use locale>
1031(but not S<C<use locale ':not_characters'>>) is in effect if modified as
1032a result of a substitution based on a regular
1d2ab946
KW
1033expression match involving any of the things mentioned in the previous
1034item, or of case-mapping, such as C<\l>, C<\L>,C<\u>, C<\U>, or C<\F>.
14280422 1035
551e1d92
RB
1036=item *
1037
39332f68 1038B<Output formatting functions> (C<printf()> and C<write()>):
14280422 1039
3cf03d68
JH
1040Results are never tainted because otherwise even output from print,
1041for example C<print(1/7)>, should be tainted if C<use locale> is in
1042effect.
14280422 1043
551e1d92
RB
1044=item *
1045
39332f68 1046B<Case-mapping functions> (C<lc()>, C<lcfirst()>, C<uc()>, C<ucfirst()>):
14280422 1047
66cbab2c
KW
1048Results are tainted if C<use locale> (but not
1049S<C<use locale ':not_characters'>>) is in effect.
14280422 1050
551e1d92
RB
1051=item *
1052
39332f68
KW
1053B<POSIX locale-dependent functions> (C<localeconv()>, C<strcoll()>,
1054C<strftime()>, C<strxfrm()>):
14280422
DD
1055
1056Results are never tainted.
1057
551e1d92
RB
1058=item *
1059
1d2ab946
KW
1060B<POSIX character class tests> (C<POSIX::isalnum()>,
1061C<POSIX::isalpha()>, C<POSIX::isdigit()>, C<POSIX::isgraph()>,
1062C<POSIX::islower()>, C<POSIX::isprint()>, C<POSIX::ispunct()>,
1063C<POSIX::isspace()>, C<POSIX::isupper()>, C<POSIX::isxdigit()>):
14280422
DD
1064
1065True/false results are never tainted.
1066
1067=back
1068
1069Three examples illustrate locale-dependent tainting.
1070The first program, which ignores its locale, won't run: a value taken
54310121 1071directly from the command line may not be used to name an output file
14280422
DD
1072when taint checks are enabled.
1073
1074 #/usr/local/bin/perl -T
1075 # Run with taint checking
1076
54310121 1077 # Command line sanity check omitted...
14280422
DD
1078 $tainted_output_file = shift;
1079
1080 open(F, ">$tainted_output_file")
3183d96c 1081 or warn "Open of $tainted_output_file failed: $!\n";
14280422
DD
1082
1083The program can be made to run by "laundering" the tainted value through
5a964f20
TC
1084a regular expression: the second example--which still ignores locale
1085information--runs, creating the file named on its command line
14280422
DD
1086if it can.
1087
1088 #/usr/local/bin/perl -T
1089
1090 $tainted_output_file = shift;
1091 $tainted_output_file =~ m%[\w/]+%;
1092 $untainted_output_file = $&;
1093
1094 open(F, ">$untainted_output_file")
1095 or warn "Open of $untainted_output_file failed: $!\n";
1096
5a964f20 1097Compare this with a similar but locale-aware program:
14280422
DD
1098
1099 #/usr/local/bin/perl -T
1100
1101 $tainted_output_file = shift;
1102 use locale;
1103 $tainted_output_file =~ m%[\w/]+%;
1104 $localized_output_file = $&;
1105
1106 open(F, ">$localized_output_file")
1107 or warn "Open of $localized_output_file failed: $!\n";
1108
1d2ab946 1109This third program fails to run because C<$&> is tainted: it is the result
5a964f20 1110of a match involving C<\w> while C<use locale> is in effect.
5f05dabc
PP
1111
1112=head1 ENVIRONMENT
1113
1114=over 12
1115
ee1ec05f
KW
1116=item PERL_SKIP_LOCALE_INIT
1117
1118This environment variable, available starting in Perl v5.20, and if it
1119evaluates to a TRUE value, tells Perl to not use the rest of the
1120environment variables to initialize with. Instead, Perl uses whatever
1121the current locale settings are. This is particularly useful in
1122embedded environments, see
1123L<perlembed/Using embedded Perl with POSIX locales>.
1124
5f05dabc
PP
1125=item PERL_BADLANG
1126
14280422 1127A string that can suppress Perl's warning about failed locale settings
54310121 1128at startup. Failure can occur if the locale support in the operating
5a964f20 1129system is lacking (broken) in some way--or if you mistyped the name of
900bd440
JH
1130a locale when you set up your environment. If this environment
1131variable is absent, or has a value that does not evaluate to integer
1132zero--that is, "0" or ""-- Perl will complain about locale setting
1133failures.
5f05dabc 1134
14280422
DD
1135B<NOTE>: PERL_BADLANG only gives you a way to hide the warning message.
1136The message tells about some problem in your system's locale support,
1137and you should investigate what the problem is.
5f05dabc
PP
1138
1139=back
1140
1141The following environment variables are not specific to Perl: They are
39332f68 1142part of the standardized (ISO C, XPG4, POSIX 1.c) C<setlocale()> method
b385bb4d
KW
1143for controlling an application's opinion on data. Windows is non-POSIX,
1144but Perl arranges for the following to work as described anyway.
65ebb059
KW
1145If the locale given by an environment variable is not valid, Perl tries
1146the next lower one in priority. If none are valid, on Windows, the
1147system default locale is then tried. If all else fails, the C<"C">
1148locale is used. If even that doesn't work, something is badly broken,
1149but Perl tries to forge ahead with whatever the locale settinga might
1150be.
5f05dabc
PP
1151
1152=over 12
1153
1154=item LC_ALL
1155
5a964f20 1156C<LC_ALL> is the "override-all" locale environment variable. If
5f05dabc
PP
1157set, it overrides all the rest of the locale environment variables.
1158
528d65ad
JH
1159=item LANGUAGE
1160
1161B<NOTE>: C<LANGUAGE> is a GNU extension, it affects you only if you
1162are using the GNU libc. This is the case if you are using e.g. Linux.
e1020413 1163If you are using "commercial" Unixes you are most probably I<not>
22b6f60d
JH
1164using GNU libc and you can ignore C<LANGUAGE>.
1165
1166However, in the case you are using C<LANGUAGE>: it affects the
1167language of informational, warning, and error messages output by
1168commands (in other words, it's like C<LC_MESSAGES>) but it has higher
96090e4f 1169priority than C<LC_ALL>. Moreover, it's not a single value but
22b6f60d
JH
1170instead a "path" (":"-separated list) of I<languages> (not locales).
1171See the GNU C<gettext> library documentation for more information.
528d65ad 1172
5f05dabc
PP
1173=item LC_CTYPE
1174
1175In the absence of C<LC_ALL>, C<LC_CTYPE> chooses the character type
1176locale. In the absence of both C<LC_ALL> and C<LC_CTYPE>, C<LANG>
1177chooses the character type locale.
1178
1179=item LC_COLLATE
1180
14280422
DD
1181In the absence of C<LC_ALL>, C<LC_COLLATE> chooses the collation
1182(sorting) locale. In the absence of both C<LC_ALL> and C<LC_COLLATE>,
1183C<LANG> chooses the collation locale.
5f05dabc
PP
1184
1185=item LC_MONETARY
1186
14280422
DD
1187In the absence of C<LC_ALL>, C<LC_MONETARY> chooses the monetary
1188formatting locale. In the absence of both C<LC_ALL> and C<LC_MONETARY>,
1189C<LANG> chooses the monetary formatting locale.
5f05dabc
PP
1190
1191=item LC_NUMERIC
1192
1193In the absence of C<LC_ALL>, C<LC_NUMERIC> chooses the numeric format
1194locale. In the absence of both C<LC_ALL> and C<LC_NUMERIC>, C<LANG>
1195chooses the numeric format.
1196
1197=item LC_TIME
1198
14280422
DD
1199In the absence of C<LC_ALL>, C<LC_TIME> chooses the date and time
1200formatting locale. In the absence of both C<LC_ALL> and C<LC_TIME>,
1201C<LANG> chooses the date and time formatting locale.
5f05dabc
PP
1202
1203=item LANG
1204
14280422
DD
1205C<LANG> is the "catch-all" locale environment variable. If it is set, it
1206is used as the last resort after the overall C<LC_ALL> and the
5f05dabc
PP
1207category-specific C<LC_...>.
1208
1209=back
1210
7e4353e9
RGS
1211=head2 Examples
1212
1213The LC_NUMERIC controls the numeric output:
1214
ef3087ec
KW
1215 use locale;
1216 use POSIX qw(locale_h); # Imports setlocale() and the LC_ constants.
1217 setlocale(LC_NUMERIC, "fr_FR") or die "Pardon";
1218 printf "%g\n", 1.23; # If the "fr_FR" succeeded, probably shows 1,23.
7e4353e9 1219
39332f68 1220and also how strings are parsed by C<POSIX::strtod()> as numbers:
7e4353e9 1221
ef3087ec
KW
1222 use locale;
1223 use POSIX qw(locale_h strtod);
1224 setlocale(LC_NUMERIC, "de_DE") or die "Entschuldigung";
1225 my $x = strtod("2,34") + 5;
1226 print $x, "\n"; # Probably shows 7,34.
7e4353e9 1227
5f05dabc
PP
1228=head1 NOTES
1229
b960a36e
KW
1230=head2 String C<eval> and C<LC_NUMERIC>
1231
1232A string L<eval|perlfunc/eval EXPR> parses its expression as standard
1233Perl. It is therefore expecting the decimal point to be a dot. If
1234C<LC_NUMERIC> is set to have this be a comma instead, the parsing will
1235be confused, perhaps silently.
1236
1237 use locale;
1238 use POSIX qw(locale_h);
1239 setlocale(LC_NUMERIC, "fr_FR") or die "Pardon";
1240 my $a = 1.2;
1241 print eval "$a + 1.5";
1242 print "\n";
1243
1244prints C<13,5>. This is because in that locale, the comma is the
1245decimal point character. The C<eval> thus expands to:
1246
1247 eval "1,2 + 1.5"
1248
1249and the result is not what you likely expected. No warnings are
1250generated. If you do string C<eval>'s within the scope of
1251S<C<use locale>>, you should instead change the C<eval> line to do
1252something like:
1253
1254 print eval "no locale; $a + 1.5";
1255
1256This prints C<2.7>.
1257
5f05dabc
PP
1258=head2 Backward compatibility
1259
b0c42ed9 1260Versions of Perl prior to 5.004 B<mostly> ignored locale information,
5a964f20
TC
1261generally behaving as if something similar to the C<"C"> locale were
1262always in force, even if the program environment suggested otherwise
1263(see L<The setlocale function>). By default, Perl still behaves this
1264way for backward compatibility. If you want a Perl application to pay
1265attention to locale information, you B<must> use the S<C<use locale>>
062ca197
KW
1266pragma (see L<The use locale pragma>) or, in the unlikely event
1267that you want to do so for just pattern matching, the
70709c68
KW
1268C</l> regular expression modifier (see L<perlre/Character set
1269modifiers>) to instruct it to do so.
b0c42ed9
JH
1270
1271Versions of Perl from 5.002 to 5.003 did use the C<LC_CTYPE>
5a964f20
TC
1272information if available; that is, C<\w> did understand what
1273were the letters according to the locale environment variables.
b0c42ed9
JH
1274The problem was that the user had no control over the feature:
1275if the C library supported locales, Perl used them.
1276
1277=head2 I18N:Collate obsolete
1278
5a964f20 1279In versions of Perl prior to 5.004, per-locale collation was possible
b0c42ed9
JH
1280using the C<I18N::Collate> library module. This module is now mildly
1281obsolete and should be avoided in new applications. The C<LC_COLLATE>
1282functionality is now integrated into the Perl core language: One can
1283use locale-specific scalar data completely normally with C<use locale>,
1284so there is no longer any need to juggle with the scalar references of
1285C<I18N::Collate>.
5f05dabc 1286
14280422 1287=head2 Sort speed and memory use impacts
5f05dabc
PP
1288
1289Comparing and sorting by locale is usually slower than the default
14280422
DD
1290sorting; slow-downs of two to four times have been observed. It will
1291also consume more memory: once a Perl scalar variable has participated
1292in any string comparison or sorting operation obeying the locale
1293collation rules, it will take 3-15 times more memory than before. (The
1294exact multiplier depends on the string's contents, the operating system
1295and the locale.) These downsides are dictated more by the operating
1296system's implementation of the locale system than by Perl.
5f05dabc 1297
5f05dabc
PP
1298=head2 Freely available locale definitions
1299
66cbab2c
KW
1300The Unicode CLDR project extracts the POSIX portion of many of its
1301locales, available at
1302
1303 http://unicode.org/Public/cldr/latest/
1304
08d7a6b2
LB
1305There is a large collection of locale definitions at:
1306
1307 http://std.dkuug.dk/i18n/WG15-collection/locales/
1308
1309You should be aware that it is
14280422 1310unsupported, and is not claimed to be fit for any purpose. If your
5a964f20 1311system allows installation of arbitrary locales, you may find the
14280422
DD
1312definitions useful as they are, or as a basis for the development of
1313your own locales.
5f05dabc 1314
14280422 1315=head2 I18n and l10n
5f05dabc 1316
b0c42ed9
JH
1317"Internationalization" is often abbreviated as B<i18n> because its first
1318and last letters are separated by eighteen others. (You may guess why
1319the internalin ... internaliti ... i18n tends to get abbreviated.) In
1320the same way, "localization" is often abbreviated to B<l10n>.
14280422
DD
1321
1322=head2 An imperfect standard
1323
1324Internationalization, as defined in the C and POSIX standards, can be
1325criticized as incomplete, ungainly, and having too large a granularity.
1326(Locales apply to a whole process, when it would arguably be more useful
1327to have them apply to a single thread, window group, or whatever.) They
1328also have a tendency, like standards groups, to divide the world into
1329nations, when we all know that the world can equally well be divided
e199995e 1330into bankers, bikers, gamers, and so on.
5f05dabc 1331
b310b053
JH
1332=head1 Unicode and UTF-8
1333
7ee2ae1e 1334The support of Unicode is new starting from Perl version v5.6, and more fully
31f05a37
KW
1335implemented in versions v5.8 and later. See L<perluniintro>.
1336
1337Starting in Perl v5.20, UTF-8 locales are supported in Perl, except for
1338C<LC_COLLATE> (use L<Unicode::Collate> instead). If you have Perl v5.16
1339or v5.18 and can't upgrade, you can use
66cbab2c
KW
1340
1341 use locale ':not_characters';
1342
1343When this form of the pragma is used, only the non-character portions of
1344locales are used by Perl, for example C<LC_NUMERIC>. Perl assumes that
1345you have translated all the characters it is to operate on into Unicode
1346(actually the platform's native character set (ASCII or EBCDIC) plus
1347Unicode). For data in files, this can conveniently be done by also
1348specifying
1349
1350 use open ':locale';
1351
1352This pragma arranges for all inputs from files to be translated into
1353Unicode from the current locale as specified in the environment (see
1354L</ENVIRONMENT>), and all outputs to files to be translated back
1355into the locale. (See L<open>). On a per-filehandle basis, you can
1356instead use the L<PerlIO::locale> module, or the L<Encode::Locale>
1357module, both available from CPAN. The latter module also has methods to
1358ease the handling of C<ARGV> and environment variables, and can be used
31f05a37 1359on individual strings. If you know that all your locales will be
66cbab2c
KW
1360UTF-8, as many are these days, you can use the L<B<-C>|perlrun/-C>
1361command line switch.
1362
1363This form of the pragma allows essentially seamless handling of locales
31f05a37
KW
1364with Unicode. The collation order will be by Unicode code point order.
1365It is strongly
66cbab2c
KW
1366recommended that when you need to order and sort strings that you use
1367the standard module L<Unicode::Collate> which gives much better results
1368in many instances than you can get with the old-style locale handling.
1369
31f05a37
KW
1370All the modules and switches just described can be used in v5.20 with
1371just plain C<use locale>, and, should the input locales not be UTF-8,
1372you'll get the less than ideal behavior, described below, that you get
1373with pre-v5.16 Perls, or when you use the locale pragma without the
1374C<:not_characters> parameter in v5.16 and v5.18. If you are using
1375exclusively UTF-8 locales in v5.20 and higher, the rest of this section
1376does not apply to you.
1377
1378There are two cases, multi-byte and single-byte locales. First
1379multi-byte:
1380
1381The only multi-byte (or wide character) locale that Perl is ever likely
1382to support is UTF-8. This is due to the difficulty of implementation,
1383the fact that high quality UTF-8 locales are now published for every
1384area of the world (L<http://unicode.org/Public/cldr/latest/>), and that
1385failing all that you can use the L<Encode> module to translate to/from
1386your locale. So, you'll have to do one of those things if you're using
1387one of these locales, such as Big5 or Shift JIS. For UTF-8 locales, in
1388Perls (pre v5.20) that don't have full UTF-8 locale support, they may
1389work reasonably well (depending on your C library implementation)
1390simply because both
dc4bfc4b
KW
1391they and Perl store characters that take up multiple bytes the same way.
1392However, some, if not most, C library implementations may not process
1393the characters in the upper half of the Latin-1 range (128 - 255)
1394properly under LC_CTYPE. To see if a character is a particular type
1395under a locale, Perl uses the functions like C<isalnum()>. Your C
1396library may not work for UTF-8 locales with those functions, instead
1397only working under the newer wide library functions like C<iswalnum()>.
31f05a37
KW
1398However, they are treated like single-byte locales, and will have the
1399restrictions described below.
e199995e 1400
31f05a37 1401For single-byte locales,
e199995e 1402Perl generally takes the tack to use locale rules on code points that can fit
66cbab2c
KW
1403in a single byte, and Unicode rules for those that can't (though this
1404isn't uniformly applied, see the note at the end of this section). This
1405prevents many problems in locales that aren't UTF-8. Suppose the locale
1406is ISO8859-7, Greek. The character at 0xD7 there is a capital Chi. But
1407in the ISO8859-1 locale, Latin1, it is a multiplication sign. The POSIX
1408regular expression character class C<[[:alpha:]]> will magically match
14090xD7 in the Greek locale but not in the Latin one.
e199995e 1410
1d2ab946 1411However, there are places where this breaks down. Certain Perl constructs are
b4ffc3db
TC
1412for Unicode only, such as C<\p{Alpha}>. They assume that 0xD7 always has its
1413Unicode meaning (or the equivalent on EBCDIC platforms). Since Latin1 is a
1414subset of Unicode and 0xD7 is the multiplication sign in both Latin1 and
1415Unicode, C<\p{Alpha}> will never match it, regardless of locale. A similar
31f05a37
KW
1416issue occurs with C<\N{...}>. Prior to v5.20, It is therefore a bad
1417idea to use C<\p{}> or
66cbab2c
KW
1418C<\N{}> under plain C<use locale>--I<unless> you can guarantee that the
1419locale will be a ISO8859-1. Use POSIX character classes instead.
1420
1421Another problem with this approach is that operations that cross the
1422single byte/multiple byte boundary are not well-defined, and so are
1423disallowed. (This boundary is between the codepoints at 255/256.).
1424For example, lower casing LATIN CAPITAL LETTER Y WITH DIAERESIS (U+0178)
1425should return LATIN SMALL LETTER Y WITH DIAERESIS (U+00FF). But in the
1426Greek locale, for example, there is no character at 0xFF, and Perl
1427has no way of knowing what the character at 0xFF is really supposed to
1428represent. Thus it disallows the operation. In this mode, the
1429lowercase of U+0178 is itself.
1430
1431The same problems ensue if you enable automatic UTF-8-ification of your
e199995e 1432standard file handles, default C<open()> layer, and C<@ARGV> on non-ISO8859-1,
b4ffc3db
TC
1433non-UTF-8 locales (by using either the B<-C> command line switch or the
1434C<PERL_UNICODE> environment variable; see L<perlrun>).
1435Things are read in as UTF-8, which would normally imply a Unicode
1436interpretation, but the presence of a locale causes them to be interpreted
1437in that locale instead. For example, a 0xD7 code point in the Unicode
1438input, which should mean the multiplication sign, won't be interpreted by
66cbab2c 1439Perl that way under the Greek locale. This is not a problem
b4ffc3db 1440I<provided> you make certain that all locales will always and only be either
66cbab2c 1441an ISO8859-1, or, if you don't have a deficient C library, a UTF-8 locale.
b4ffc3db 1442
1d2ab946
KW
1443Still another problem is that this approach can lead to two code
1444points meaning the same character. Thus in a Greek locale, both U+03A7
1445and U+00D7 are GREEK CAPITAL LETTER CHI.
1446
b4ffc3db
TC
1447Vendor locales are notoriously buggy, and it is difficult for Perl to test
1448its locale-handling code because this interacts with code that Perl has no
1449control over; therefore the locale-handling code in Perl may be buggy as
66cbab2c
KW
1450well. (However, the Unicode-supplied locales should be better, and
1451there is a feed back mechanism to correct any problems. See
1452L</Freely available locale definitions>.)
1453
7ee2ae1e 1454If you have Perl v5.16, the problems mentioned above go away if you use
66cbab2c 1455the C<:not_characters> parameter to the locale pragma (except for vendor
7ee2ae1e 1456bugs in the non-character portions). If you don't have v5.16, and you
66cbab2c
KW
1457I<do> have locales that work, using them may be worthwhile for certain
1458specific purposes, as long as you keep in mind the gotchas already
1459mentioned. For example, if the collation for your locales works, it
1460runs faster under locales than under L<Unicode::Collate>; and you gain
1461access to such things as the local currency symbol and the names of the
7ee2ae1e 1462months and days of the week. (But to hammer home the point, in v5.16,
66cbab2c
KW
1463you get this access without the downsides of locales by using the
1464C<:not_characters> form of the pragma.)
1465
1466Note: The policy of using locale rules for code points that can fit in a
1467byte, and Unicode rules for those that can't is not uniformly applied.
7ee2ae1e 1468Pre-v5.12, it was somewhat haphazard; in v5.12 it was applied fairly
66cbab2c 1469consistently to regular expression matching except for bracketed
7ee2ae1e
KW
1470character classes; in v5.14 it was extended to all regex matches; and in
1471v5.16 to the casing operations such as C<"\L"> and C<uc()>. For
66cbab2c
KW
1472collation, in all releases, the system's C<strxfrm()> function is called,
1473and whatever it does is what you get.
b310b053 1474
5f05dabc
PP
1475=head1 BUGS
1476
1477=head2 Broken systems
1478
5a964f20 1479In certain systems, the operating system's locale support
2bdf8add 1480is broken and cannot be fixed or used by Perl. Such deficiencies can
b4ffc3db 1481and will result in mysterious hangs and/or Perl core dumps when
2bdf8add 1482C<use locale> is in effect. When confronted with such a system,
7f2de2d2 1483please report in excruciating detail to <F<perlbug@perl.org>>, and
b4ffc3db 1484also contact your vendor: bug fixes may exist for these problems
2bdf8add 1485in your operating system. Sometimes such bug fixes are called an
83fb1bf0
KW
1486operating system upgrade. If you have the source for Perl, include in
1487the perlbug email the output of the test described above in L</Testing
1488for broken locales>.
5f05dabc
PP
1489
1490=head1 SEE ALSO
1491
b310b053
JH
1492L<I18N::Langinfo>, L<perluniintro>, L<perlunicode>, L<open>,
1493L<POSIX/isalnum>, L<POSIX/isalpha>,
4bbcc6e8
JH
1494L<POSIX/isdigit>, L<POSIX/isgraph>, L<POSIX/islower>,
1495L<POSIX/isprint>, L<POSIX/ispunct>, L<POSIX/isspace>,
1496L<POSIX/isupper>, L<POSIX/isxdigit>, L<POSIX/localeconv>,
1497L<POSIX/setlocale>, L<POSIX/strcoll>, L<POSIX/strftime>,
1498L<POSIX/strtod>, L<POSIX/strxfrm>.
5f05dabc 1499
ccd65d51
KW
1500For special considerations when Perl is embedded in a C program,
1501see L<perlembed/Using embedded Perl with POSIX locales>.
1502
5f05dabc
PP
1503=head1 HISTORY
1504
b0c42ed9 1505Jarkko Hietaniemi's original F<perli18n.pod> heavily hacked by Dominic
5a964f20 1506Dunlop, assisted by the perl5-porters. Prose worked over a bit by
c052850d 1507Tom Christiansen, and updated by Perl 5 porters.