This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
perllocale: Clarify text
[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
0c880285
KW
25Perl has been extended to support certain types of locales available in
26the locale system. This is controlled per application by using one
27pragma, one function call, and several environment variables.
28
29Perl supports single-byte locales that are supersets of ASCII, such as
30the ISO 8859 ones, and one multi-byte-type locale, UTF-8 ones, described
31in the next paragraph. Perl doesn't support any other multi-byte
32locales, such as the ones for East Asian languages.
66cbab2c
KW
33
34Unfortunately, there are quite a few deficiencies with the design (and
31f05a37
KW
35often, the implementations) of locales. Unicode was invented (see
36L<perlunitut> for an introduction to that) in part to address these
37design deficiencies, and nowadays, there is a series of "UTF-8
38locales", based on Unicode. These are locales whose character set is
39Unicode, encoded in UTF-8. Starting in v5.20, Perl fully supports
9accf6df 40UTF-8 locales, except for sorting and string comparisons like C<lt> and
a4a439fb
KW
41C<ge>. Starting in v5.26, Perl can handle these reasonably as well,
42depending on the platform's implementation. However, for earlier
0c8d0e83
KW
43releases or for better control, use L<Unicode::Collate>. There are
44actually two slightly different types of UTF-8 locales: one for Turkic
00872368
TC
45languages and one for everything else.
46
47Starting in Perl v5.30, Perl detects Turkic locales by their
48behaviour, and seamlessly handles both types; previously only the
49non-Turkic one was supported. The name of the locale is ignored, if
50your system has a C<tr_TR.UTF-8> locale and it doesn't behave like a
51Turkic locale, perl will treat it like a non-Turkic locale.
0c8d0e83
KW
52
53Perl continues to support the old non UTF-8 locales as well. There are
54currently no UTF-8 locales for EBCDIC platforms.
31f05a37 55
66cbab2c
KW
56(Unicode is also creating C<CLDR>, the "Common Locale Data Repository",
57L<http://cldr.unicode.org/> which includes more types of information than
58are available in the POSIX locale system. At the time of this writing,
59there was no CPAN module that provides access to this XML-encoded data.
e784ce68
KW
60However, it is possible to compute the POSIX locale data from them, and
61earlier CLDR versions had these already extracted for you as UTF-8 locales
62L<http://unicode.org/Public/cldr/2.0.1/>.)
66cbab2c
KW
63
64=head1 WHAT IS A LOCALE
65
66A locale is a set of data that describes various aspects of how various
67communities in the world categorize their world. These categories are
68broken down into the following types (some of which include a brief
69note here):
70
71=over
72
cb88b78e 73=item Category C<LC_NUMERIC>: Numeric formatting
66cbab2c
KW
74
75This indicates how numbers should be formatted for human readability,
76for example the character used as the decimal point.
77
cb88b78e 78=item Category C<LC_MONETARY>: Formatting of monetary amounts
66cbab2c 79
032639c4 80Z<>
66cbab2c 81
cb88b78e 82=item Category C<LC_TIME>: Date/Time formatting
66cbab2c 83
032639c4 84Z<>
66cbab2c 85
cb88b78e 86=item Category C<LC_MESSAGES>: Error and other messages
66cbab2c 87
2619d284 88This is used by Perl itself only for accessing operating system error
03c702c5 89messages via L<$!|perlvar/$ERRNO> and L<$^E|perlvar/$EXTENDED_OS_ERROR>.
66cbab2c 90
cb88b78e 91=item Category C<LC_COLLATE>: Collation
66cbab2c 92
76073c88 93This indicates the ordering of letters for comparison and sorting.
66cbab2c
KW
94In Latin alphabets, for example, "b", generally follows "a".
95
cb88b78e 96=item Category C<LC_CTYPE>: Character Types
66cbab2c
KW
97
98This indicates, for example if a character is an uppercase letter.
99
2619d284
KW
100=item Other categories
101
102Some platforms have other categories, dealing with such things as
103measurement units and paper sizes. None of these are used directly by
104Perl, but outside operations that Perl interacts with may use
d6ded950 105these. See L</Not within the scope of "use locale"> below.
2619d284 106
66cbab2c
KW
107=back
108
2619d284
KW
109More details on the categories used by Perl are given below in L</LOCALE
110CATEGORIES>.
66cbab2c
KW
111
112Together, these categories go a long way towards being able to customize
113a single program to run in many different locations. But there are
114deficiencies, so keep reading.
5f05dabc
PP
115
116=head1 PREPARING TO USE LOCALES
117
c5c88224
KW
118Perl itself (outside the L<POSIX> module) will not use locales unless
119specifically requested to (but
2619d284
KW
120again note that Perl may interact with code that does use them). Even
121if there is such a request, B<all> of the following must be true
b960a36e 122for it to work properly:
5f05dabc
PP
123
124=over 4
125
126=item *
127
128B<Your operating system must support the locale system>. If it does,
39332f68 129you should find that the C<setlocale()> function is a documented part of
5f05dabc
PP
130its C library.
131
132=item *
133
5a964f20 134B<Definitions for locales that you use must be installed>. You, or
14280422
DD
135your system administrator, must make sure that this is the case. The
136available locales, the location in which they are kept, and the manner
5a964f20
TC
137in which they are installed all vary from system to system. Some systems
138provide only a few, hard-wired locales and do not allow more to be
139added. Others allow you to add "canned" locales provided by the system
140supplier. Still others allow you or the system administrator to define
14280422 141and add arbitrary locales. (You may have to ask your supplier to
5a964f20 142provide canned locales that are not delivered with your operating
14280422 143system.) Read your system documentation for further illumination.
5f05dabc
PP
144
145=item *
146
147B<Perl must believe that the locale system is supported>. If it does,
148C<perl -V:d_setlocale> will say that the value for C<d_setlocale> is
149C<define>.
150
151=back
152
153If you want a Perl application to process and present your data
154according to a particular locale, the application code should include
5a0de581 155the S<C<use locale>> pragma (see L</The "use locale" pragma>) where
5f05dabc
PP
156appropriate, and B<at least one> of the following must be true:
157
158=over 4
159
c052850d 160=item 1
5f05dabc 161
66cbab2c 162B<The locale-determining environment variables (see L</"ENVIRONMENT">)
5a964f20 163must be correctly set up> at the time the application is started, either
ef3087ec 164by yourself or by whomever set up your system account; or
5f05dabc 165
c052850d 166=item 2
5f05dabc 167
14280422 168B<The application must set its own locale> using the method described in
5a0de581 169L</The setlocale function>.
5f05dabc
PP
170
171=back
172
173=head1 USING LOCALES
174
d6ded950 175=head2 The C<"use locale"> pragma
5f05dabc 176
e9bc6d6b
KW
177Starting in Perl 5.28, this pragma may be used in
178L<multi-threaded|threads> applications on systems that have thread-safe
179locale ability. Some caveats apply, see L</Multi-threaded> below. On
180systems without this capability, or in earlier Perls, do NOT use this
181pragma in scripts that have multiple L<threads|threads> active. The
182locale in these cases is not local to a single thread. Another thread
183may change the locale at any time, which could cause at a minimum that a
184given thread is operating in a locale it isn't expecting to be in. On
185some platforms, segfaults can also occur. The locale change need not be
186explicit; some operations cause perl to change the locale itself. You
187are vulnerable simply by having done a S<C<"use locale">>.
fc82b82e 188
c5c88224
KW
189By default, Perl itself (outside the L<POSIX> module)
190ignores the current locale. The S<C<use locale>>
66cbab2c 191pragma tells Perl to use the current locale for some operations.
d6ded950
KW
192Starting in v5.16, there are optional parameters to this pragma,
193described below, which restrict which operations are affected by it.
c052850d
KW
194
195The current locale is set at execution time by
196L<setlocale()|/The setlocale function> described below. If that function
197hasn't yet been called in the course of the program's execution, the
66cbab2c 198current locale is that which was determined by the L</"ENVIRONMENT"> in
ebc3223b 199effect at the start of the program.
dfcc8045 200If there is no valid environment, the current locale is whatever the
65ebb059
KW
201system default has been set to. On POSIX systems, it is likely, but
202not necessarily, the "C" locale. On Windows, the default is set via the
203computer's S<C<Control Panel-E<gt>Regional and Language Options>> (or its
204current equivalent).
c052850d
KW
205
206The operations that are affected by locale are:
5f05dabc
PP
207
208=over 4
209
d6ded950 210=item B<Not within the scope of C<"use locale">>
b960a36e 211
c0a087f2
KW
212Only certain operations (all originating outside Perl) should be
213affected, as follows:
1d2ab946
KW
214
215=over 4
216
217=item *
2619d284 218
663d437a 219The current locale is used when going outside of Perl with
2619d284
KW
220operations like L<system()|perlfunc/system LIST> or
221L<qxE<sol>E<sol>|perlop/qxE<sol>STRINGE<sol>>, if those operations are
222locale-sensitive.
223
1d2ab946
KW
224=item *
225
2619d284
KW
226Also Perl gives access to various C library functions through the
227L<POSIX> module. Some of those functions are always affected by the
228current locale. For example, C<POSIX::strftime()> uses C<LC_TIME>;
229C<POSIX::strtod()> uses C<LC_NUMERIC>; C<POSIX::strcoll()> and
9accf6df 230C<POSIX::strxfrm()> use C<LC_COLLATE>. All such functions
2619d284 231will behave according to the current underlying locale, even if that
1d2ab946 232locale isn't exposed to Perl space.
2619d284 233
d9e2553c
KW
234This applies as well to L<I18N::Langinfo>.
235
1d2ab946
KW
236=item *
237
52686f2a
KW
238XS modules for all categories but C<LC_NUMERIC> get the underlying
239locale, and hence any C library functions they call will use that
8a384d3a 240underlying locale. For more discussion, see L<perlxs/CAVEATS>.
1d2ab946
KW
241
242=back
243
c5c88224
KW
244Note that all C programs (including the perl interpreter, which is
245written in C) always have an underlying locale. That locale is the "C"
246locale unless changed by a call to L<setlocale()|/The setlocale
247function>. When Perl starts up, it changes the underlying locale to the
248one which is indicated by the L</ENVIRONMENT>. When using the L<POSIX>
249module or writing XS code, it is important to keep in mind that the
250underlying locale may be something other than "C", even if the program
251hasn't explicitly changed it.
252
032639c4 253Z<>
ebc3223b
KW
254
255=item B<Lingering effects of C<S<use locale>>>
1d2ab946
KW
256
257Certain Perl operations that are set-up within the scope of a
d6ded950 258C<use locale> retain that effect even outside the scope.
b960a36e
KW
259These include:
260
261=over 4
262
263=item *
264
265The output format of a L<write()|perlfunc/write> is determined by an
266earlier format declaration (L<perlfunc/format>), so whether or not the
267output is affected by locale is determined by if the C<format()> is
d6ded950 268within the scope of a C<use locale>, not whether the C<write()>
b960a36e
KW
269is.
270
271=item *
272
273Regular expression patterns can be compiled using
33be4c61 274L<qrE<sol>E<sol>|perlop/qrE<sol>STRINGE<sol>msixpodualn> with actual
b960a36e
KW
275matching deferred to later. Again, it is whether or not the compilation
276was done within the scope of C<use locale> that determines the match
277behavior, not if the matches are done within such a scope or not.
278
279=back
280
032639c4 281Z<>
ebc3223b 282
d6ded950 283=item B<Under C<"use locale";>>
66cbab2c
KW
284
285=over 4
286
287=item *
288
d6ded950 289All the above operations
b960a36e
KW
290
291=item *
292
293B<Format declarations> (L<perlfunc/format>) and hence any subsequent
294C<write()>s use C<LC_NUMERIC>.
66cbab2c
KW
295
296=item *
297
b960a36e
KW
298B<stringification and output> use C<LC_NUMERIC>.
299These include the results of
300C<print()>,
301C<printf()>,
302C<say()>,
303and
304C<sprintf()>.
66cbab2c 305
b960a36e
KW
306=item *
307
308B<The comparison operators> (C<lt>, C<le>, C<cmp>, C<ge>, and C<gt>) use
39332f68 309C<LC_COLLATE>. C<sort()> is also affected if used without an
5a964f20 310explicit comparison function, because it uses C<cmp> by default.
14280422 311
5a964f20 312B<Note:> C<eq> and C<ne> are unaffected by locale: they always
de108802 313perform a char-by-char comparison of their scalar operands. What's
14280422
DD
314more, if C<cmp> finds that its operands are equal according to the
315collation sequence specified by the current locale, it goes on to
de108802
RGS
316perform a char-by-char comparison, and only returns I<0> (equal) if the
317operands are char-for-char identical. If you really want to know whether
5a964f20 318two strings--which C<eq> and C<cmp> may consider different--are equal
14280422 319as far as collation in the locale is concerned, see the discussion in
d38c72a5 320L</Category C<LC_COLLATE>: Collation>.
5f05dabc
PP
321
322=item *
323
39332f68
KW
324B<Regular expressions and case-modification functions> (C<uc()>, C<lc()>,
325C<ucfirst()>, and C<lcfirst()>) use C<LC_CTYPE>
5f05dabc 326
2c6ee1a7
KW
327=item *
328
dbf3c4d7
KW
329B<The variables L<$!|perlvar/$ERRNO>> (and its synonyms C<$ERRNO> and
330C<$OS_ERROR>) B<and L<$^E|perlvar/$EXTENDED_OS_ERROR>> (and its synonym
2c6ee1a7
KW
331C<$EXTENDED_OS_ERROR>) when used as strings use C<LC_MESSAGES>.
332
5f05dabc
PP
333=back
334
66cbab2c 335=back
5f05dabc 336
5a964f20 337The default behavior is restored with the S<C<no locale>> pragma, or
ef3087ec 338upon reaching the end of the block enclosing C<use locale>.
3cd61afa 339Note that C<use locale> calls may be
66cbab2c
KW
340nested, and that what is in effect within an inner scope will revert to
341the outer scope's rules at the end of the inner scope.
5f05dabc 342
5a964f20 343The string result of any operation that uses locale
14280422 344information is tainted, as it is possible for a locale to be
5a0de581 345untrustworthy. See L</"SECURITY">.
5f05dabc 346
d6ded950
KW
347Starting in Perl v5.16 in a very limited way, and more generally in
348v5.22, you can restrict which category or categories are enabled by this
349particular instance of the pragma by adding parameters to it. For
350example,
351
352 use locale qw(:ctype :numeric);
353
354enables locale awareness within its scope of only those operations
355(listed above) that are affected by C<LC_CTYPE> and C<LC_NUMERIC>.
356
357The possible categories are: C<:collate>, C<:ctype>, C<:messages>,
358C<:monetary>, C<:numeric>, C<:time>, and the pseudo category
359C<:characters> (described below).
360
361Thus you can say
362
363 use locale ':messages';
364
365and only L<$!|perlvar/$ERRNO> and L<$^E|perlvar/$EXTENDED_OS_ERROR>
366will be locale aware. Everything else is unaffected.
367
368Since Perl doesn't currently do anything with the C<LC_MONETARY>
369category, specifying C<:monetary> does effectively nothing. Some
22803c6a 370systems have other categories, such as C<LC_PAPER>, but Perl
e9bc6d6b 371also doesn't do anything with them, and there is no way to specify
d6ded950
KW
372them in this pragma's arguments.
373
374You can also easily say to use all categories but one, by either, for
375example,
376
377 use locale ':!ctype';
378 use locale ':not_ctype';
379
380both of which mean to enable locale awarness of all categories but
381C<LC_CTYPE>. Only one category argument may be specified in a
382S<C<use locale>> if it is of the negated form.
383
384Prior to v5.22 only one form of the pragma with arguments is available:
385
386 use locale ':not_characters';
387
388(and you have to say C<not_>; you can't use the bang C<!> form). This
389pseudo category is a shorthand for specifying both C<:collate> and
390C<:ctype>. Hence, in the negated form, it is nearly the same thing as
391saying
392
393 use locale qw(:messages :monetary :numeric :time);
394
395We use the term "nearly", because C<:not_characters> also turns on
396S<C<use feature 'unicode_strings'>> within its scope. This form is
397less useful in v5.20 and later, and is described fully in
398L</Unicode and UTF-8>, but briefly, it tells Perl to not use the
399character portions of the locale definition, that is the C<LC_CTYPE> and
400C<LC_COLLATE> categories. Instead it will use the native character set
401(extended by Unicode). When using this parameter, you are responsible
402for getting the external character set translated into the
403native/Unicode one (which it already will be if it is one of the
404increasingly popular UTF-8 locales). There are convenient ways of doing
405this, as described in L</Unicode and UTF-8>.
406
5f05dabc
PP
407=head2 The setlocale function
408
e9bc6d6b
KW
409WARNING! Prior to Perl 5.28 or on a system that does not support
410thread-safe locale operations, do NOT use this function in a
411L<thread|threads>. The locale will change in all other threads at the
412same time, and should your thread get paused by the operating system,
413and another started, that thread will not have the locale it is
414expecting. On some platforms, there can be a race leading to segfaults
355107d3
KW
415if two threads call this function nearly simultaneously. On unthreaded
416builds, or on Perl 5.28 and later on thread-safe systems, this warning
417does not apply.
fc82b82e 418
14280422 419You can switch locales as often as you wish at run time with the
39332f68 420C<POSIX::setlocale()> function:
5f05dabc 421
5f05dabc
PP
422 # Import locale-handling tool set from POSIX module.
423 # This example uses: setlocale -- the function call
424 # LC_CTYPE -- explained below
2619d284
KW
425 # (Showing the testing for success/failure of operations is
426 # omitted in these examples to avoid distracting from the main
ebc3223b 427 # point)
6ea81ccf 428
5f05dabc 429 use POSIX qw(locale_h);
dfcc8045
KW
430 use locale;
431 my $old_locale;
5f05dabc 432
14280422 433 # query and save the old locale
5f05dabc
PP
434 $old_locale = setlocale(LC_CTYPE);
435
436 setlocale(LC_CTYPE, "fr_CA.ISO8859-1");
437 # LC_CTYPE now in locale "French, Canada, codeset ISO 8859-1"
438
439 setlocale(LC_CTYPE, "");
65ebb059
KW
440 # LC_CTYPE now reset to the default defined by the
441 # LC_ALL/LC_CTYPE/LANG environment variables, or to the system
442 # default. See below for documentation.
5f05dabc
PP
443
444 # restore the old locale
445 setlocale(LC_CTYPE, $old_locale);
446
39332f68 447The first argument of C<setlocale()> gives the B<category>, the second the
14280422
DD
448B<locale>. The category tells in what aspect of data processing you
449want to apply locale-specific rules. Category names are discussed in
66cbab2c 450L</LOCALE CATEGORIES> and L</"ENVIRONMENT">. The locale is the name of a
14280422
DD
451collection of customization information corresponding to a particular
452combination of language, country or territory, and codeset. Read on for
453hints on the naming of locales: not all systems name locales as in the
454example.
455
39332f68 456If no second argument is provided and the category is something other
cb88b78e 457than C<LC_ALL>, the function returns a string naming the current locale
502a173a 458for the category. You can use this value as the second argument in a
f170b852
KW
459subsequent call to C<setlocale()>, B<but> on some platforms the string
460is opaque, not something that most people would be able to decipher as
461to what locale it means.
502a173a 462
cb88b78e 463If no second argument is provided and the category is C<LC_ALL>, the
502a173a 464result is implementation-dependent. It may be a string of
c052850d 465concatenated locale names (separator also implementation-dependent)
39332f68 466or a single locale name. Please consult your L<setlocale(3)> man page for
502a173a
JH
467details.
468
469If a second argument is given and it corresponds to a valid locale,
470the locale for the category is set to that value, and the function
471returns the now-current locale value. You can then use this in yet
39332f68 472another call to C<setlocale()>. (In some implementations, the return
502a173a
JH
473value may sometimes differ from the value you gave as the second
474argument--think of it as an alias for the value you gave.)
5f05dabc
PP
475
476As the example shows, if the second argument is an empty string, the
477category's locale is returned to the default specified by the
478corresponding environment variables. Generally, this results in a
5a964f20 479return to the default that was in force when Perl started up: changes
54310121 480to the environment made by the application after startup may or may not
5a964f20 481be noticed, depending on your system's C library.
5f05dabc 482
d6ded950
KW
483Note that when a form of C<use locale> that doesn't include all
484categories is specified, Perl ignores the excluded categories.
66cbab2c 485
f170b852 486If C<set_locale()> fails for some reason (for example, an attempt to set
dfcc8045
KW
487to a locale unknown to the system), the locale for the category is not
488changed, and the function returns C<undef>.
489
e9bc6d6b
KW
490Starting in Perl 5.28, on multi-threaded perls compiled on systems that
491implement POSIX 2008 thread-safe locale operations, this function
492doesn't actually call the system C<setlocale>. Instead those
493thread-safe operations are used to emulate the C<setlocale> function,
494but in a thread-safe manner.
2619d284 495
44289c0b
KW
496You can force the thread-safe locale operations to always be used (if
497available) by recompiling perl with
498
499 -Accflags='-DUSE_THREAD_SAFE_LOCALE'
500
501added to your call to F<Configure>.
502
39332f68 503For further information about the categories, consult L<setlocale(3)>.
3e6e419a 504
e9bc6d6b
KW
505=head2 Multi-threaded operation
506
507Beginning in Perl 5.28, multi-threaded locale operation is supported on
508systems that implement either the POSIX 2008 or Windows-specific
509thread-safe locale operations. Many modern systems, such as various
510Unix variants and Darwin do have this.
511
512You can tell if using locales is safe on your system by looking at the
513read-only boolean variable C<${^SAFE_LOCALES}>. The value is 1 if the
514perl is not threaded, or if it is using thread-safe locale operations.
515
516Thread-safe operations are supported in Windows starting in Visual Studio
5172005, and in systems compatible with POSIX 2008. Some platforms claim
518to support POSIX 2008, but have buggy implementations, so that the hints
519files for compiling to run on them turn off attempting to use
520thread-safety. C<${^SAFE_LOCALES}> will be 0 on them.
521
522Be aware that writing a multi-threaded application will not be portable
523to a platform which lacks the native thread-safe locale support. On
524systems that do have it, you automatically get this behavior for
525threaded perls, without having to do anything. If for some reason, you
526don't want to use this capability (perhaps the POSIX 2008 support is
527buggy on your system), you can manually compile Perl to use the old
528non-thread-safe implementation by passing the argument
529C<-Accflags='-DNO_THREAD_SAFE_LOCALE'> to F<Configure>.
530Except on Windows, this will continue to use certain of the POSIX 2008
531functions in some situations. If these are buggy, you can pass the
532following to F<Configure> instead or additionally:
533C<-Accflags='-DNO_POSIX_2008_LOCALE'>. This will also keep the code
534from using thread-safe locales.
535C<${^SAFE_LOCALES}> will be 0 on systems that turn off the thread-safe
536operations.
537
e1895adc
KW
538Normally on unthreaded builds, the traditional C<setlocale()> is used
539and not the thread-safe locale functions. You can force the use of these
540on systems that have them by adding the
541C<-Accflags='-DUSE_THREAD_SAFE_LOCALE'> to F<Configure>.
542
e9bc6d6b
KW
543The initial program is started up using the locale specified from the
544environment, as currently, described in L</ENVIRONMENT>. All newly
545created threads start with C<LC_ALL> set to C<"C">>. Each thread may
546use C<POSIX::setlocale()> to query or switch its locale at any time,
547without affecting any other thread. All locale-dependent operations
548automatically use their thread's locale.
549
550This should be completely transparent to any applications written
551entirely in Perl (minus a few rarely encountered caveats given in the
552L</Multi-threaded> section). Information for XS module writers is given
553in L<perlxs/Locale-aware XS code>.
554
3e6e419a
JH
555=head2 Finding locales
556
39332f68 557For locales available in your system, consult also L<setlocale(3)> to
5a964f20
TC
558see whether it leads to the list of available locales (search for the
559I<SEE ALSO> section). If that fails, try the following command lines:
5f05dabc
PP
560
561 locale -a
562
563 nlsinfo
564
565 ls /usr/lib/nls/loc
566
567 ls /usr/lib/locale
568
569 ls /usr/lib/nls
570
b478f28d
JH
571 ls /usr/share/locale
572
5f05dabc
PP
573and see whether they list something resembling these
574
2bdf8add 575 en_US.ISO8859-1 de_DE.ISO8859-1 ru_RU.ISO8859-5
502a173a 576 en_US.iso88591 de_DE.iso88591 ru_RU.iso88595
2bdf8add 577 en_US de_DE ru_RU
14280422 578 en de ru
2bdf8add
JH
579 english german russian
580 english.iso88591 german.iso88591 russian.iso88595
502a173a 581 english.roman8 russian.koi8r
5f05dabc 582
39332f68 583Sadly, even though the calling interface for C<setlocale()> has been
528d65ad 584standardized, names of locales and the directories where the
5a964f20 585configuration resides have not been. The basic form of the name is
528d65ad
JH
586I<language_territory>B<.>I<codeset>, but the latter parts after
587I<language> are not always present. The I<language> and I<country>
588are usually from the standards B<ISO 3166> and B<ISO 639>, the
589two-letter abbreviations for the countries and the languages of the
590world, respectively. The I<codeset> part often mentions some B<ISO
5918859> character set, the Latin codesets. For example, C<ISO 8859-1>
592is the so-called "Western European codeset" that can be used to encode
593most Western European languages adequately. Again, there are several
594ways to write even the name of that one standard. Lamentably.
5f05dabc 595
14280422
DD
596Two special locales are worth particular mention: "C" and "POSIX".
597Currently these are effectively the same locale: the difference is
5a964f20
TC
598mainly that the first one is defined by the C standard, the second by
599the POSIX standard. They define the B<default locale> in which
14280422 600every program starts in the absence of locale information in its
5a964f20 601environment. (The I<default> default locale, if you will.) Its language
39332f68
KW
602is (American) English and its character codeset ASCII or, rarely, a
603superset thereof (such as the "DEC Multinational Character Set
604(DEC-MCS)"). B<Warning>. The C locale delivered by some vendors
605may not actually exactly match what the C standard calls for. So
606beware.
5f05dabc 607
14280422
DD
608B<NOTE>: Not all systems have the "POSIX" locale (not all systems are
609POSIX-conformant), so use "C" when you need explicitly to specify this
610default locale.
5f05dabc 611
3e6e419a
JH
612=head2 LOCALE PROBLEMS
613
5a964f20 614You may encounter the following warning message at Perl startup:
3e6e419a
JH
615
616 perl: warning: Setting locale failed.
617 perl: warning: Please check that your locale settings:
618 LC_ALL = "En_US",
619 LANG = (unset)
620 are supported and installed on your system.
621 perl: warning: Falling back to the standard locale ("C").
622
cb88b78e 623This means that your locale settings had C<LC_ALL> set to "En_US" and
5a964f20
TC
624LANG exists but has no value. Perl tried to believe you but could not.
625Instead, Perl gave up and fell back to the "C" locale, the default locale
65ebb059
KW
626that is supposed to work no matter what. (On Windows, it first tries
627falling back to the system default locale.) This usually means your
628locale settings were wrong, they mention locales your system has never
629heard of, or the locale installation in your system has problems (for
630example, some system files are broken or missing). There are quick and
631temporary fixes to these problems, as well as more thorough and lasting
632fixes.
3e6e419a 633
83fb1bf0
KW
634=head2 Testing for broken locales
635
636If you are building Perl from source, the Perl test suite file
637F<lib/locale.t> can be used to test the locales on your system.
638Setting the environment variable C<PERL_DEBUG_FULL_TEST> to 1
639will cause it to output detailed results. For example, on Linux, you
640could say
641
1d2ab946 642 PERL_DEBUG_FULL_TEST=1 ./perl -T -Ilib lib/locale.t > locale.log 2>&1
83fb1bf0
KW
643
644Besides many other tests, it will test every locale it finds on your
645system to see if they conform to the POSIX standard. If any have
646errors, it will include a summary near the end of the output of which
647locales passed all its tests, and which failed, and why.
648
3e6e419a
JH
649=head2 Temporarily fixing locale problems
650
5a964f20 651The two quickest fixes are either to render Perl silent about any
3e6e419a
JH
652locale inconsistencies or to run Perl under the default locale "C".
653
654Perl's moaning about locale problems can be silenced by setting the
22ff3130 655environment variable C<PERL_BADLANG> to "0" or "".
900bd440
JH
656This method really just sweeps the problem under the carpet: you tell
657Perl to shut up even when Perl sees that something is wrong. Do not
658be surprised if later something locale-dependent misbehaves.
3e6e419a
JH
659
660Perl can be run under the "C" locale by setting the environment
cb88b78e
KW
661variable C<LC_ALL> to "C". This method is perhaps a bit more civilized
662than the C<PERL_BADLANG> approach, but setting C<LC_ALL> (or
5a964f20
TC
663other locale variables) may affect other programs as well, not just
664Perl. In particular, external programs run from within Perl will see
3e6e419a 665these changes. If you make the new settings permanent (read on), all
5a0de581
LM
666programs you run see the changes. See L</"ENVIRONMENT"> for
667the full list of relevant environment variables and L</"USING LOCALES">
e05ffc7d 668for their effects in Perl. Effects in other programs are
cb88b78e 669easily deducible. For example, the variable C<LC_COLLATE> may well affect
b432a672 670your B<sort> program (or whatever the program that arranges "records"
3e6e419a
JH
671alphabetically in your system is called).
672
5a964f20
TC
673You can test out changing these variables temporarily, and if the
674new settings seem to help, put those settings into your shell startup
663d437a 675files. Consult your local documentation for the exact details. For
5a964f20 676Bourne-like shells (B<sh>, B<ksh>, B<bash>, B<zsh>):
3e6e419a
JH
677
678 LC_ALL=en_US.ISO8859-1
679 export LC_ALL
680
5a964f20
TC
681This assumes that we saw the locale "en_US.ISO8859-1" using the commands
682discussed above. We decided to try that instead of the above faulty
683locale "En_US"--and in Cshish shells (B<csh>, B<tcsh>)
3e6e419a
JH
684
685 setenv LC_ALL en_US.ISO8859-1
c47ff5f1 686
663d437a 687or if you have the "env" application you can do (in any shell)
c406981e
JH
688
689 env LC_ALL=en_US.ISO8859-1 perl ...
690
5a964f20 691If you do not know what shell you have, consult your local
3e6e419a
JH
692helpdesk or the equivalent.
693
694=head2 Permanently fixing locale problems
695
5a964f20
TC
696The slower but superior fixes are when you may be able to yourself
697fix the misconfiguration of your own environment variables. The
3e6e419a
JH
698mis(sing)configuration of the whole system's locales usually requires
699the help of your friendly system administrator.
700
5a0de581 701First, see earlier in this document about L</Finding locales>. That tells
5a964f20
TC
702how to find which locales are really supported--and more importantly,
703installed--on your system. In our example error message, environment
704variables affecting the locale are listed in the order of decreasing
705importance (and unset variables do not matter). Therefore, having
706LC_ALL set to "En_US" must have been the bad choice, as shown by the
707error message. First try fixing locale settings listed first.
3e6e419a 708
5a964f20
TC
709Second, if using the listed commands you see something B<exactly>
710(prefix matches do not count and case usually counts) like "En_US"
711without the quotes, then you should be okay because you are using a
712locale name that should be installed and available in your system.
5a0de581 713In this case, see L</Permanently fixing your system's locale configuration>.
3e6e419a 714
4a4eefd0 715=head2 Permanently fixing your system's locale configuration
3e6e419a 716
5a964f20 717This is when you see something like:
3e6e419a
JH
718
719 perl: warning: Please check that your locale settings:
720 LC_ALL = "En_US",
721 LANG = (unset)
722 are supported and installed on your system.
723
724but then cannot see that "En_US" listed by the above-mentioned
5a964f20
TC
725commands. You may see things like "en_US.ISO8859-1", but that isn't
726the same. In this case, try running under a locale
727that you can list and which somehow matches what you tried. The
3e6e419a 728rules for matching locale names are a bit vague because
e05ffc7d 729standardization is weak in this area. See again the
5a0de581 730L</Finding locales> about general rules.
3e6e419a 731
b687b08b 732=head2 Fixing system locale configuration
3e6e419a 733
5a964f20
TC
734Contact a system administrator (preferably your own) and report the exact
735error message you get, and ask them to read this same documentation you
736are now reading. They should be able to check whether there is something
5a0de581 737wrong with the locale configuration of the system. The L</Finding locales>
5a964f20
TC
738section is unfortunately a bit vague about the exact commands and places
739because these things are not that standardized.
3e6e419a 740
5f05dabc
PP
741=head2 The localeconv function
742
39332f68 743The C<POSIX::localeconv()> function allows you to get particulars of the
14280422 744locale-dependent numeric formatting information specified by the current
a835cd47
KW
745underlying C<LC_NUMERIC> and C<LC_MONETARY> locales (regardless of
746whether called from within the scope of C<S<use locale>> or not). (If
747you just want the name of
39332f68 748the current locale for a particular category, use C<POSIX::setlocale()>
5a0de581 749with a single parameter--see L</The setlocale function>.)
5f05dabc
PP
750
751 use POSIX qw(locale_h);
5f05dabc
PP
752
753 # Get a reference to a hash of locale-dependent info
754 $locale_values = localeconv();
755
756 # Output sorted list of the values
757 for (sort keys %$locale_values) {
14280422 758 printf "%-20s = %s\n", $_, $locale_values->{$_}
5f05dabc
PP
759 }
760
39332f68 761C<localeconv()> takes no arguments, and returns B<a reference to> a hash.
5a964f20 762The keys of this hash are variable names for formatting, such as
502a173a 763C<decimal_point> and C<thousands_sep>. The values are the
cea6626f 764corresponding, er, values. See L<POSIX/localeconv> for a longer
502a173a
JH
765example listing the categories an implementation might be expected to
766provide; some provide more and others fewer. You don't need an
39332f68 767explicit C<use locale>, because C<localeconv()> always observes the
502a173a 768current locale.
5f05dabc 769
5a964f20
TC
770Here's a simple-minded example program that rewrites its command-line
771parameters as integers correctly formatted in the current locale:
5f05dabc 772
ef3087ec
KW
773 use POSIX qw(locale_h);
774
775 # Get some of locale's numeric formatting parameters
776 my ($thousands_sep, $grouping) =
777 @{localeconv()}{'thousands_sep', 'grouping'};
778
779 # Apply defaults if values are missing
780 $thousands_sep = ',' unless $thousands_sep;
781
782 # grouping and mon_grouping are packed lists
783 # of small integers (characters) telling the
784 # grouping (thousand_seps and mon_thousand_seps
785 # being the group dividers) of numbers and
786 # monetary quantities. The integers' meanings:
787 # 255 means no more grouping, 0 means repeat
788 # the previous grouping, 1-254 means use that
789 # as the current grouping. Grouping goes from
790 # right to left (low to high digits). In the
791 # below we cheat slightly by never using anything
792 # else than the first grouping (whatever that is).
793 if ($grouping) {
794 @grouping = unpack("C*", $grouping);
795 } else {
796 @grouping = (3);
797 }
798
799 # Format command line params for current locale
800 for (@ARGV) {
801 $_ = int; # Chop non-integer part
802 1 while
803 s/(\d)(\d{$grouping[0]}($|$thousands_sep))/$1$thousands_sep$2/;
804 print "$_";
805 }
806 print "\n";
5f05dabc 807
03ceeedf
KW
808Note that if the platform doesn't have C<LC_NUMERIC> and/or
809C<LC_MONETARY> available or enabled, the corresponding elements of the
810hash will be missing.
811
74c76037 812=head2 I18N::Langinfo
4bbcc6e8
JH
813
814Another interface for querying locale-dependent information is the
c0a087f2 815C<I18N::Langinfo::langinfo()> function.
4bbcc6e8 816
39332f68
KW
817The following example will import the C<langinfo()> function itself and
818three constants to be used as arguments to C<langinfo()>: a constant for
74c76037
JH
819the abbreviated first day of the week (the numbering starts from
820Sunday = 1) and two more constants for the affirmative and negative
821answers for a yes/no question in the current locale.
4bbcc6e8 822
74c76037 823 use I18N::Langinfo qw(langinfo ABDAY_1 YESSTR NOSTR);
4bbcc6e8 824
ef3087ec
KW
825 my ($abday_1, $yesstr, $nostr)
826 = map { langinfo } qw(ABDAY_1 YESSTR NOSTR);
4bbcc6e8 827
74c76037 828 print "$abday_1? [$yesstr/$nostr] ";
4bbcc6e8 829
74c76037
JH
830In other words, in the "C" (or English) locale the above will probably
831print something like:
832
e05ffc7d 833 Sun? [yes/no]
4bbcc6e8
JH
834
835See L<I18N::Langinfo> for more information.
836
5f05dabc
PP
837=head1 LOCALE CATEGORIES
838
5a964f20
TC
839The following subsections describe basic locale categories. Beyond these,
840some combination categories allow manipulation of more than one
5a0de581 841basic category at a time. See L</"ENVIRONMENT"> for a discussion of these.
5f05dabc 842
ff52fcf1 843=head2 Category C<LC_COLLATE>: Collation: Text Comparisons and Sorting
5f05dabc 844
d6ded950
KW
845In the scope of a S<C<use locale>> form that includes collation, Perl
846looks to the C<LC_COLLATE>
5a964f20 847environment variable to determine the application's notions on collation
b4ffc3db
TC
848(ordering) of characters. For example, "b" follows "a" in Latin
849alphabets, but where do "E<aacute>" and "E<aring>" belong? And while
f87fa335 850"color" follows "chocolate" in English, what about in traditional Spanish?
5f05dabc 851
60f0fa02 852The following collations all make sense and you may meet any of them
dbf3c4d7 853if you C<"use locale">.
60f0fa02
JH
854
855 A B C D E a b c d e
35316ca3 856 A a B b C c D d E e
60f0fa02
JH
857 a A b B c C d D e E
858 a b c d e A B C D E
859
f1cbbd6e 860Here is a code snippet to tell what "word"
5a964f20 861characters are in the current locale, in that locale's order:
5f05dabc
PP
862
863 use locale;
35316ca3 864 print +(sort grep /\w/, map { chr } 0..255), "\n";
5f05dabc 865
14280422
DD
866Compare this with the characters that you see and their order if you
867state explicitly that the locale should be ignored:
5f05dabc
PP
868
869 no locale;
35316ca3 870 print +(sort grep /\w/, map { chr } 0..255), "\n";
5f05dabc
PP
871
872This machine-native collation (which is what you get unless S<C<use
873locale>> has appeared earlier in the same block) must be used for
874sorting raw binary data, whereas the locale-dependent collation of the
b0c42ed9 875first example is useful for natural text.
5f05dabc 876
5a0de581 877As noted in L</USING LOCALES>, C<cmp> compares according to the current
14280422 878collation locale when C<use locale> is in effect, but falls back to a
de108802 879char-by-char comparison for strings that the locale says are equal. You
39332f68 880can use C<POSIX::strcoll()> if you don't want this fall-back:
14280422
DD
881
882 use POSIX qw(strcoll);
883 $equal_in_locale =
884 !strcoll("space and case ignored", "SpaceAndCaseIgnored");
885
39332f68 886C<$equal_in_locale> will be true if the collation locale specifies a
5a964f20 887dictionary-like ordering that ignores space characters completely and
9e3a2af8 888which folds case.
14280422 889
a4a439fb
KW
890Perl uses the platform's C library collation functions C<strcoll()> and
891C<strxfrm()>. That means you get whatever they give. On some
892platforms, these functions work well on UTF-8 locales, giving
893a reasonable default collation for the code points that are important in
894that locale. (And if they aren't working well, the problem may only be
895that the locale definition is deficient, so can be fixed by using a
896better definition file. Unicode's definitions (see L</Freely available
897locale definitions>) provide reasonable UTF-8 locale collation
898definitions.) Starting in Perl v5.26, Perl's use of these functions has
899been made more seamless. This may be sufficient for your needs. For
900more control, and to make sure strings containing any code point (not
901just the ones important in the locale) collate properly, the
902L<Unicode::Collate> module is suggested.
903
904In non-UTF-8 locales (hence single byte), code points above 0xFF are
905technically invalid. But if present, again starting in v5.26, they will
906collate to the same position as the highest valid code point does. This
907generally gives good results, but the collation order may be skewed if
908the valid code point gets special treatment when it forms particular
909sequences with other characters as defined by the locale.
910When two strings collate identically, the code point order is used as a
911tie breaker.
912
913If Perl detects that there are problems with the locale collation order,
914it reverts to using non-locale collation rules for that locale.
31f05a37 915
5a964f20 916If you have a single string that you want to check for "equality in
14280422 917locale" against several others, you might think you could gain a little
39332f68 918efficiency by using C<POSIX::strxfrm()> in conjunction with C<eq>:
14280422
DD
919
920 use POSIX qw(strxfrm);
921 $xfrm_string = strxfrm("Mixed-case string");
922 print "locale collation ignores spaces\n"
923 if $xfrm_string eq strxfrm("Mixed-casestring");
924 print "locale collation ignores hyphens\n"
925 if $xfrm_string eq strxfrm("Mixedcase string");
926 print "locale collation ignores case\n"
927 if $xfrm_string eq strxfrm("mixed-case string");
928
39332f68 929C<strxfrm()> takes a string and maps it into a transformed string for use
de108802 930in char-by-char comparisons against other transformed strings during
14280422 931collation. "Under the hood", locale-affected Perl comparison operators
39332f68
KW
932call C<strxfrm()> for both operands, then do a char-by-char
933comparison of the transformed strings. By calling C<strxfrm()> explicitly
14280422 934and using a non locale-affected comparison, the example attempts to save
5a964f20 935a couple of transformations. But in fact, it doesn't save anything: Perl
2ae324a7 936magic (see L<perlguts/Magic Variables>) creates the transformed version of a
5a964f20 937string the first time it's needed in a comparison, then keeps this version around
14280422 938in case it's needed again. An example rewritten the easy way with
e38874e2 939C<cmp> runs just about as fast. It also copes with null characters
39332f68 940embedded in strings; if you call C<strxfrm()> directly, it treats the first
0c880285 941null it finds as a terminator. Don't expect the transformed strings
5a964f20 942it produces to be portable across systems--or even from one revision
39332f68 943of your operating system to the next. In short, don't call C<strxfrm()>
e38874e2 944directly: let Perl do it for you.
14280422 945
5a964f20 946Note: C<use locale> isn't shown in some of these examples because it isn't
dfcc8045
KW
947needed: C<strcoll()> and C<strxfrm()> are POSIX functions
948which use the standard system-supplied C<libc> functions that
949always obey the current C<LC_COLLATE> locale.
5f05dabc 950
cb88b78e 951=head2 Category C<LC_CTYPE>: Character Types
5f05dabc 952
d6ded950
KW
953In the scope of a S<C<use locale>> form that includes C<LC_CTYPE>, Perl
954obeys the C<LC_CTYPE> locale
14280422 955setting. This controls the application's notion of which characters are
ebc3223b
KW
956alphabetic, numeric, punctuation, I<etc>. This affects Perl's C<\w>
957regular expression metanotation,
f1cbbd6e 958which stands for alphanumeric characters--that is, alphabetic,
ebc3223b
KW
959numeric, and the platform's native underscore.
960(Consult L<perlre> for more information about
14280422 961regular expressions.) Thanks to C<LC_CTYPE>, depending on your locale
b4ffc3db
TC
962setting, characters like "E<aelig>", "E<eth>", "E<szlig>", and
963"E<oslash>" may be understood as C<\w> characters.
ebc3223b
KW
964It also affects things like C<\s>, C<\D>, and the POSIX character
965classes, like C<[[:graph:]]>. (See L<perlrecharclass> for more
966information on all these.)
5f05dabc 967
2c268ad5 968The C<LC_CTYPE> locale also provides the map used in transliterating
68dc0745 969characters between lower and uppercase. This affects the case-mapping
663d437a
KW
970functions--C<fc()>, C<lc()>, C<lcfirst()>, C<uc()>, and C<ucfirst()>;
971case-mapping
b9cc4f69 972interpolation with C<\F>, C<\l>, C<\L>, C<\u>, or C<\U> in double-quoted
26a9b33a 973strings and C<s///> substitutions; and case-insensitive regular expression
e38874e2
DD
974pattern matching using the C<i> modifier.
975
31f05a37
KW
976Starting in v5.20, Perl supports UTF-8 locales for C<LC_CTYPE>, but
977otherwise Perl only supports single-byte locales, such as the ISO 8859
978series. This means that wide character locales, for example for Asian
b5eea289
KW
979languages, are not well-supported. Use of these locales may cause core
980dumps. If the platform has the capability for Perl to detect such a
981locale, starting in Perl v5.22, L<Perl will warn, default
982enabled|warnings/Category Hierarchy>, using the C<locale> warning
983category, whenever such a locale is switched into. The UTF-8 locale
984support is actually a
31f05a37 985superset of POSIX locales, because it is really full Unicode behavior
13af4fd9
KW
986as if no C<LC_CTYPE> locale were in effect at all (except for tainting;
987see L</SECURITY>). POSIX locales, even UTF-8 ones,
31f05a37
KW
988are lacking certain concepts in Unicode, such as the idea that changing
989the case of a character could expand to be more than one character.
990Perl in a UTF-8 locale, will give you that expansion. Prior to v5.20,
991Perl treated a UTF-8 locale on some platforms like an ISO 8859-1 one,
992with some restrictions, and on other platforms more like the "C" locale.
993For releases v5.16 and v5.18, C<S<use locale 'not_characters>> could be
994used as a workaround for this (see L</Unicode and UTF-8>).
995
5d63e270 996Note that there are quite a few things that are unaffected by the
663d437a
KW
997current locale. Any literal character is the native character for the
998given platform. Hence 'A' means the character at code point 65 on ASCII
999platforms, and 193 on EBCDIC. That may or may not be an 'A' in the
1000current locale, if that locale even has an 'A'.
1001Similarly, all the escape sequences for particular characters,
5d63e270
KW
1002C<\n> for example, always mean the platform's native one. This means,
1003for example, that C<\N> in regular expressions (every character
1d2ab946 1004but new-line) works on the platform character set.
5d63e270 1005
8c6180a9
KW
1006Starting in v5.22, Perl will by default warn when switching into a
1007locale that redefines any ASCII printable character (plus C<\t> and
84035de0
KW
1008C<\n>) into a different class than expected. This is likely to
1009happen on modern locales only on EBCDIC platforms, where, for example,
1010a CCSID 0037 locale on a CCSID 1047 machine moves C<"[">, but it can
1011happen on ASCII platforms with the ISO 646 and other
8c6180a9
KW
10127-bit locales that are essentially obsolete. Things may still work,
1013depending on what features of Perl are used by the program. For
1014example, in the example from above where C<"|"> becomes a C<\w>, and
1015there are no regular expressions where this matters, the program may
1016still work properly. The warning lists all the characters that
1017it can determine could be adversely affected.
1018
14280422
DD
1019B<Note:> A broken or malicious C<LC_CTYPE> locale definition may result
1020in clearly ineligible characters being considered to be alphanumeric by
e199995e 1021your application. For strict matching of (mundane) ASCII letters and
5a964f20 1022digits--for example, in command strings--locale-aware applications
5a0de581 1023should use C<\w> with the C</a> regular expression modifier. See L</"SECURITY">.
5f05dabc 1024
fee33030 1025=head2 Category C<LC_NUMERIC>: Numeric Formatting
5f05dabc 1026
d6ded950
KW
1027After a proper C<POSIX::setlocale()> call, and within the scope of
1028of a C<use locale> form that includes numerics, Perl obeys the
1029C<LC_NUMERIC> locale information, which controls an application's idea
1030of how numbers should be formatted for human readability.
b960a36e 1031In most implementations the only effect is to
b4ffc3db 1032change the character used for the decimal point--perhaps from "." to ",".
b960a36e 1033The functions aren't aware of such niceties as thousands separation and
5a0de581 1034so on. (See L</The localeconv function> if you care about these things.)
5a964f20 1035
b960a36e
KW
1036 use POSIX qw(strtod setlocale LC_NUMERIC);
1037 use locale;
5f05dabc 1038
b960a36e 1039 setlocale LC_NUMERIC, "";
14280422 1040
b960a36e 1041 $n = 5/2; # Assign numeric 2.5 to $n
5f05dabc 1042
b960a36e 1043 $a = " $n"; # Locale-dependent conversion to string
5f05dabc 1044
b960a36e 1045 print "half five is $n\n"; # Locale-dependent output
5f05dabc 1046
b960a36e 1047 printf "half five is %g\n", $n; # Locale-dependent output
5f05dabc 1048
b960a36e
KW
1049 print "DECIMAL POINT IS COMMA\n"
1050 if $n == (strtod("2,5"))[0]; # Locale-dependent conversion
5f05dabc 1051
4bbcc6e8
JH
1052See also L<I18N::Langinfo> and C<RADIXCHAR>.
1053
cb88b78e 1054=head2 Category C<LC_MONETARY>: Formatting of monetary amounts
5f05dabc 1055
e199995e 1056The C standard defines the C<LC_MONETARY> category, but not a function
5a964f20 1057that is affected by its contents. (Those with experience of standards
b0c42ed9 1058committees will recognize that the working group decided to punt on the
fa9b773e
KW
1059issue.) Consequently, Perl essentially takes no notice of it. If you
1060really want to use C<LC_MONETARY>, you can query its contents--see
5a0de581 1061L</The localeconv function>--and use the information that it returns in your
e05ffc7d
KW
1062application's own formatting of currency amounts. However, you may well
1063find that the information, voluminous and complex though it may be, still
1064does not quite meet your requirements: currency formatting is a hard nut
13a2d996 1065to crack.
5f05dabc 1066
4bbcc6e8
JH
1067See also L<I18N::Langinfo> and C<CRNCYSTR>.
1068
ff52fcf1 1069=head2 Category C<LC_TIME>: Respresentation of time
5f05dabc 1070
39332f68 1071Output produced by C<POSIX::strftime()>, which builds a formatted
5f05dabc
PP
1072human-readable date/time string, is affected by the current C<LC_TIME>
1073locale. Thus, in a French locale, the output produced by the C<%B>
1074format element (full month name) for the first month of the year would
5a964f20 1075be "janvier". Here's how to get a list of long month names in the
5f05dabc
PP
1076current locale:
1077
1078 use POSIX qw(strftime);
14280422
DD
1079 for (0..11) {
1080 $long_month_name[$_] =
1081 strftime("%B", 0, 0, 0, 1, $_, 96);
5f05dabc
PP
1082 }
1083
2619d284
KW
1084Note: C<use locale> isn't needed in this example: C<strftime()> is a POSIX
1085function which uses the standard system-supplied C<libc> function that
1086always obeys the current C<LC_TIME> locale.
5f05dabc 1087
4bbcc6e8 1088See also L<I18N::Langinfo> and C<ABDAY_1>..C<ABDAY_7>, C<DAY_1>..C<DAY_7>,
2a2bf5f4 1089C<ABMON_1>..C<ABMON_12>, and C<ABMON_1>..C<ABMON_12>.
4bbcc6e8 1090
5f05dabc
PP
1091=head2 Other categories
1092
2619d284
KW
1093The remaining locale categories are not currently used by Perl itself.
1094But again note that things Perl interacts with may use these, including
1095extensions outside the standard Perl distribution, and by the
98a6f11e 1096operating system and its utilities. Note especially that the string
1097value of C<$!> and the error messages given by external utilities may
1098be changed by C<LC_MESSAGES>. If you want to have portable error
265f5c4a 1099codes, use C<%!>. See L<Errno>.
14280422
DD
1100
1101=head1 SECURITY
1102
5a964f20 1103Although the main discussion of Perl security issues can be found in
14280422
DD
1104L<perlsec>, a discussion of Perl's locale handling would be incomplete
1105if it did not draw your attention to locale-dependent security issues.
5a964f20
TC
1106Locales--particularly on systems that allow unprivileged users to
1107build their own locales--are untrustworthy. A malicious (or just plain
14280422
DD
1108broken) locale can make a locale-aware application give unexpected
1109results. Here are a few possibilities:
1110
1111=over 4
1112
1113=item *
1114
1115Regular expression checks for safe file names or mail addresses using
5a964f20 1116C<\w> may be spoofed by an C<LC_CTYPE> locale that claims that
dbf3c4d7 1117characters such as C<"E<gt>"> and C<"|"> are alphanumeric.
14280422
DD
1118
1119=item *
1120
e38874e2 1121String interpolation with case-mapping, as in, say, C<$dest =
cb88b78e 1122"C:\U$name.$ext">, may produce dangerous results if a bogus C<LC_CTYPE>
e38874e2
DD
1123case-mapping table is in effect.
1124
1125=item *
1126
14280422
DD
1127A sneaky C<LC_COLLATE> locale could result in the names of students with
1128"D" grades appearing ahead of those with "A"s.
1129
1130=item *
1131
5a964f20 1132An application that takes the trouble to use information in
14280422 1133C<LC_MONETARY> may format debits as if they were credits and vice versa
5a964f20 1134if that locale has been subverted. Or it might make payments in US
14280422
DD
1135dollars instead of Hong Kong dollars.
1136
1137=item *
1138
39332f68 1139The date and day names in dates formatted by C<strftime()> could be
14280422 1140manipulated to advantage by a malicious user able to subvert the
5a964f20 1141C<LC_DATE> locale. ("Look--it says I wasn't in the building on
14280422
DD
1142Sunday.")
1143
1144=back
1145
1146Such dangers are not peculiar to the locale system: any aspect of an
5a964f20 1147application's environment which may be modified maliciously presents
14280422 1148similar challenges. Similarly, they are not specific to Perl: any
5a964f20 1149programming language that allows you to write programs that take
14280422
DD
1150account of their environment exposes you to these issues.
1151
5a964f20
TC
1152Perl cannot protect you from all possibilities shown in the
1153examples--there is no substitute for your own vigilance--but, when
14280422 1154C<use locale> is in effect, Perl uses the tainting mechanism (see
5a964f20 1155L<perlsec>) to mark string results that become locale-dependent, and
14280422 1156which may be untrustworthy in consequence. Here is a summary of the
5a964f20 1157tainting behavior of operators and functions that may be affected by
14280422
DD
1158the locale:
1159
1160=over 4
1161
551e1d92
RB
1162=item *
1163
1164B<Comparison operators> (C<lt>, C<le>, C<ge>, C<gt> and C<cmp>):
14280422
DD
1165
1166Scalar true/false (or less/equal/greater) result is never tainted.
1167
551e1d92
RB
1168=item *
1169
1d2ab946 1170B<Case-mapping interpolation> (with C<\l>, C<\L>, C<\u>, C<\U>, or C<\F>)
e38874e2 1171
d6ded950
KW
1172The result string containing interpolated material is tainted if
1173a C<use locale> form that includes C<LC_CTYPE> is in effect.
e38874e2 1174
551e1d92
RB
1175=item *
1176
1177B<Matching operator> (C<m//>):
14280422
DD
1178
1179Scalar true/false result never tainted.
1180
1d2ab946 1181All subpatterns, either delivered as a list-context result or as C<$1>
d6ded950
KW
1182I<etc>., are tainted if a C<use locale> form that includes
1183C<LC_CTYPE> is in effect, and the subpattern
63baef57
KW
1184regular expression contains a locale-dependent construct. These
1185constructs include C<\w> (to match an alphanumeric character), C<\W>
1186(non-alphanumeric character), C<\b> and C<\B> (word-boundary and
1187non-boundardy, which depend on what C<\w> and C<\W> match), C<\s>
1188(whitespace character), C<\S> (non whitespace character), C<\d> and
1189C<\D> (digits and non-digits), and the POSIX character classes, such as
1190C<[:alpha:]> (see L<perlrecharclass/POSIX Character Classes>).
1191
1192Tainting is also likely if the pattern is to be matched
1193case-insensitively (via C</i>). The exception is if all the code points
1194to be matched this way are above 255 and do not have folds under Unicode
1195rules to below 256. Tainting is not done for these because Perl
1196only uses Unicode rules for such code points, and those rules are the
1197same no matter what the current locale.
1198
1d2ab946
KW
1199The matched-pattern variables, C<$&>, C<$`> (pre-match), C<$'>
1200(post-match), and C<$+> (last match) also are tainted.
14280422 1201
551e1d92
RB
1202=item *
1203
1204B<Substitution operator> (C<s///>):
14280422 1205
e38874e2 1206Has the same behavior as the match operator. Also, the left
d6ded950
KW
1207operand of C<=~> becomes tainted when a C<use locale>
1208form that includes C<LC_CTYPE> is in effect, if modified as
66cbab2c 1209a result of a substitution based on a regular
1d2ab946
KW
1210expression match involving any of the things mentioned in the previous
1211item, or of case-mapping, such as C<\l>, C<\L>,C<\u>, C<\U>, or C<\F>.
14280422 1212
551e1d92
RB
1213=item *
1214
39332f68 1215B<Output formatting functions> (C<printf()> and C<write()>):
14280422 1216
3cf03d68
JH
1217Results are never tainted because otherwise even output from print,
1218for example C<print(1/7)>, should be tainted if C<use locale> is in
1219effect.
14280422 1220
551e1d92
RB
1221=item *
1222
39332f68 1223B<Case-mapping functions> (C<lc()>, C<lcfirst()>, C<uc()>, C<ucfirst()>):
14280422 1224
d6ded950
KW
1225Results are tainted if a C<use locale> form that includes C<LC_CTYPE> is
1226in effect.
14280422 1227
551e1d92
RB
1228=item *
1229
39332f68
KW
1230B<POSIX locale-dependent functions> (C<localeconv()>, C<strcoll()>,
1231C<strftime()>, C<strxfrm()>):
14280422
DD
1232
1233Results are never tainted.
1234
14280422
DD
1235=back
1236
1237Three examples illustrate locale-dependent tainting.
1238The first program, which ignores its locale, won't run: a value taken
54310121 1239directly from the command line may not be used to name an output file
14280422
DD
1240when taint checks are enabled.
1241
1242 #/usr/local/bin/perl -T
1243 # Run with taint checking
1244
54310121 1245 # Command line sanity check omitted...
14280422
DD
1246 $tainted_output_file = shift;
1247
1248 open(F, ">$tainted_output_file")
3183d96c 1249 or warn "Open of $tainted_output_file failed: $!\n";
14280422
DD
1250
1251The program can be made to run by "laundering" the tainted value through
5a964f20
TC
1252a regular expression: the second example--which still ignores locale
1253information--runs, creating the file named on its command line
14280422
DD
1254if it can.
1255
1256 #/usr/local/bin/perl -T
1257
1258 $tainted_output_file = shift;
1259 $tainted_output_file =~ m%[\w/]+%;
1260 $untainted_output_file = $&;
1261
1262 open(F, ">$untainted_output_file")
1263 or warn "Open of $untainted_output_file failed: $!\n";
1264
5a964f20 1265Compare this with a similar but locale-aware program:
14280422
DD
1266
1267 #/usr/local/bin/perl -T
1268
1269 $tainted_output_file = shift;
1270 use locale;
1271 $tainted_output_file =~ m%[\w/]+%;
1272 $localized_output_file = $&;
1273
1274 open(F, ">$localized_output_file")
1275 or warn "Open of $localized_output_file failed: $!\n";
1276
1d2ab946 1277This third program fails to run because C<$&> is tainted: it is the result
5a964f20 1278of a match involving C<\w> while C<use locale> is in effect.
5f05dabc
PP
1279
1280=head1 ENVIRONMENT
1281
1282=over 12
1283
ee1ec05f
KW
1284=item PERL_SKIP_LOCALE_INIT
1285
c5c88224
KW
1286This environment variable, available starting in Perl v5.20, if set
1287(to any value), tells Perl to not use the rest of the
ee1ec05f
KW
1288environment variables to initialize with. Instead, Perl uses whatever
1289the current locale settings are. This is particularly useful in
1290embedded environments, see
1291L<perlembed/Using embedded Perl with POSIX locales>.
1292
5f05dabc
PP
1293=item PERL_BADLANG
1294
14280422 1295A string that can suppress Perl's warning about failed locale settings
54310121 1296at startup. Failure can occur if the locale support in the operating
5a964f20 1297system is lacking (broken) in some way--or if you mistyped the name of
900bd440 1298a locale when you set up your environment. If this environment
22ff3130
HS
1299variable is absent, or has a value other than "0" or "", Perl will
1300complain about locale setting failures.
5f05dabc 1301
cb88b78e 1302B<NOTE>: C<PERL_BADLANG> only gives you a way to hide the warning message.
14280422
DD
1303The message tells about some problem in your system's locale support,
1304and you should investigate what the problem is.
5f05dabc
PP
1305
1306=back
1307
1308The following environment variables are not specific to Perl: They are
39332f68 1309part of the standardized (ISO C, XPG4, POSIX 1.c) C<setlocale()> method
b385bb4d
KW
1310for controlling an application's opinion on data. Windows is non-POSIX,
1311but Perl arranges for the following to work as described anyway.
65ebb059
KW
1312If the locale given by an environment variable is not valid, Perl tries
1313the next lower one in priority. If none are valid, on Windows, the
1314system default locale is then tried. If all else fails, the C<"C">
1315locale is used. If even that doesn't work, something is badly broken,
c5e9a8e7 1316but Perl tries to forge ahead with whatever the locale settings might
65ebb059 1317be.
5f05dabc
PP
1318
1319=over 12
1320
cb88b78e 1321=item C<LC_ALL>
5f05dabc 1322
5a964f20 1323C<LC_ALL> is the "override-all" locale environment variable. If
5f05dabc
PP
1324set, it overrides all the rest of the locale environment variables.
1325
cb88b78e 1326=item C<LANGUAGE>
528d65ad
JH
1327
1328B<NOTE>: C<LANGUAGE> is a GNU extension, it affects you only if you
1329are using the GNU libc. This is the case if you are using e.g. Linux.
e1020413 1330If you are using "commercial" Unixes you are most probably I<not>
22b6f60d
JH
1331using GNU libc and you can ignore C<LANGUAGE>.
1332
1333However, in the case you are using C<LANGUAGE>: it affects the
1334language of informational, warning, and error messages output by
1335commands (in other words, it's like C<LC_MESSAGES>) but it has higher
96090e4f 1336priority than C<LC_ALL>. Moreover, it's not a single value but
22b6f60d
JH
1337instead a "path" (":"-separated list) of I<languages> (not locales).
1338See the GNU C<gettext> library documentation for more information.
528d65ad 1339
3ee1a09c 1340=item C<LC_CTYPE>
5f05dabc
PP
1341
1342In the absence of C<LC_ALL>, C<LC_CTYPE> chooses the character type
1343locale. In the absence of both C<LC_ALL> and C<LC_CTYPE>, C<LANG>
1344chooses the character type locale.
1345
cb88b78e 1346=item C<LC_COLLATE>
5f05dabc 1347
14280422
DD
1348In the absence of C<LC_ALL>, C<LC_COLLATE> chooses the collation
1349(sorting) locale. In the absence of both C<LC_ALL> and C<LC_COLLATE>,
1350C<LANG> chooses the collation locale.
5f05dabc 1351
cb88b78e 1352=item C<LC_MONETARY>
5f05dabc 1353
14280422
DD
1354In the absence of C<LC_ALL>, C<LC_MONETARY> chooses the monetary
1355formatting locale. In the absence of both C<LC_ALL> and C<LC_MONETARY>,
1356C<LANG> chooses the monetary formatting locale.
5f05dabc 1357
cb88b78e 1358=item C<LC_NUMERIC>
5f05dabc
PP
1359
1360In the absence of C<LC_ALL>, C<LC_NUMERIC> chooses the numeric format
1361locale. In the absence of both C<LC_ALL> and C<LC_NUMERIC>, C<LANG>
1362chooses the numeric format.
1363
cb88b78e 1364=item C<LC_TIME>
5f05dabc 1365
14280422
DD
1366In the absence of C<LC_ALL>, C<LC_TIME> chooses the date and time
1367formatting locale. In the absence of both C<LC_ALL> and C<LC_TIME>,
1368C<LANG> chooses the date and time formatting locale.
5f05dabc 1369
cb88b78e 1370=item C<LANG>
5f05dabc 1371
14280422
DD
1372C<LANG> is the "catch-all" locale environment variable. If it is set, it
1373is used as the last resort after the overall C<LC_ALL> and the
3ee1a09c 1374category-specific C<LC_I<foo>>.
5f05dabc
PP
1375
1376=back
1377
7e4353e9
RGS
1378=head2 Examples
1379
cb88b78e 1380The C<LC_NUMERIC> controls the numeric output:
7e4353e9 1381
ef3087ec
KW
1382 use locale;
1383 use POSIX qw(locale_h); # Imports setlocale() and the LC_ constants.
1384 setlocale(LC_NUMERIC, "fr_FR") or die "Pardon";
1385 printf "%g\n", 1.23; # If the "fr_FR" succeeded, probably shows 1,23.
7e4353e9 1386
39332f68 1387and also how strings are parsed by C<POSIX::strtod()> as numbers:
7e4353e9 1388
ef3087ec
KW
1389 use locale;
1390 use POSIX qw(locale_h strtod);
1391 setlocale(LC_NUMERIC, "de_DE") or die "Entschuldigung";
1392 my $x = strtod("2,34") + 5;
1393 print $x, "\n"; # Probably shows 7,34.
7e4353e9 1394
5f05dabc
PP
1395=head1 NOTES
1396
b960a36e
KW
1397=head2 String C<eval> and C<LC_NUMERIC>
1398
1399A string L<eval|perlfunc/eval EXPR> parses its expression as standard
1400Perl. It is therefore expecting the decimal point to be a dot. If
1401C<LC_NUMERIC> is set to have this be a comma instead, the parsing will
1402be confused, perhaps silently.
1403
1404 use locale;
1405 use POSIX qw(locale_h);
1406 setlocale(LC_NUMERIC, "fr_FR") or die "Pardon";
1407 my $a = 1.2;
1408 print eval "$a + 1.5";
1409 print "\n";
1410
1411prints C<13,5>. This is because in that locale, the comma is the
1412decimal point character. The C<eval> thus expands to:
1413
1414 eval "1,2 + 1.5"
1415
1416and the result is not what you likely expected. No warnings are
1417generated. If you do string C<eval>'s within the scope of
1418S<C<use locale>>, you should instead change the C<eval> line to do
1419something like:
1420
1421 print eval "no locale; $a + 1.5";
1422
1423This prints C<2.7>.
1424
d6ded950
KW
1425You could also exclude C<LC_NUMERIC>, if you don't need it, by
1426
1427 use locale ':!numeric';
1428
5f05dabc
PP
1429=head2 Backward compatibility
1430
b0c42ed9 1431Versions of Perl prior to 5.004 B<mostly> ignored locale information,
5a964f20
TC
1432generally behaving as if something similar to the C<"C"> locale were
1433always in force, even if the program environment suggested otherwise
5a0de581 1434(see L</The setlocale function>). By default, Perl still behaves this
5a964f20
TC
1435way for backward compatibility. If you want a Perl application to pay
1436attention to locale information, you B<must> use the S<C<use locale>>
5a0de581 1437pragma (see L</The "use locale" pragma>) or, in the unlikely event
062ca197 1438that you want to do so for just pattern matching, the
70709c68
KW
1439C</l> regular expression modifier (see L<perlre/Character set
1440modifiers>) to instruct it to do so.
b0c42ed9
JH
1441
1442Versions of Perl from 5.002 to 5.003 did use the C<LC_CTYPE>
5a964f20
TC
1443information if available; that is, C<\w> did understand what
1444were the letters according to the locale environment variables.
b0c42ed9
JH
1445The problem was that the user had no control over the feature:
1446if the C library supported locales, Perl used them.
1447
1448=head2 I18N:Collate obsolete
1449
5a964f20 1450In versions of Perl prior to 5.004, per-locale collation was possible
b0c42ed9
JH
1451using the C<I18N::Collate> library module. This module is now mildly
1452obsolete and should be avoided in new applications. The C<LC_COLLATE>
1453functionality is now integrated into the Perl core language: One can
1454use locale-specific scalar data completely normally with C<use locale>,
1455so there is no longer any need to juggle with the scalar references of
1456C<I18N::Collate>.
5f05dabc 1457
14280422 1458=head2 Sort speed and memory use impacts
5f05dabc
PP
1459
1460Comparing and sorting by locale is usually slower than the default
14280422
DD
1461sorting; slow-downs of two to four times have been observed. It will
1462also consume more memory: once a Perl scalar variable has participated
1463in any string comparison or sorting operation obeying the locale
1464collation rules, it will take 3-15 times more memory than before. (The
1465exact multiplier depends on the string's contents, the operating system
1466and the locale.) These downsides are dictated more by the operating
1467system's implementation of the locale system than by Perl.
5f05dabc 1468
5f05dabc
PP
1469=head2 Freely available locale definitions
1470
66cbab2c
KW
1471The Unicode CLDR project extracts the POSIX portion of many of its
1472locales, available at
1473
30659cfd 1474 https://unicode.org/Public/cldr/2.0.1/
e784ce68
KW
1475
1476(Newer versions of CLDR require you to compute the POSIX data yourself.
1477See L<http://unicode.org/Public/cldr/latest/>.)
66cbab2c 1478
08d7a6b2
LB
1479There is a large collection of locale definitions at:
1480
1481 http://std.dkuug.dk/i18n/WG15-collection/locales/
1482
1483You should be aware that it is
14280422 1484unsupported, and is not claimed to be fit for any purpose. If your
5a964f20 1485system allows installation of arbitrary locales, you may find the
14280422
DD
1486definitions useful as they are, or as a basis for the development of
1487your own locales.
5f05dabc 1488
14280422 1489=head2 I18n and l10n
5f05dabc 1490
b0c42ed9
JH
1491"Internationalization" is often abbreviated as B<i18n> because its first
1492and last letters are separated by eighteen others. (You may guess why
1493the internalin ... internaliti ... i18n tends to get abbreviated.) In
1494the same way, "localization" is often abbreviated to B<l10n>.
14280422
DD
1495
1496=head2 An imperfect standard
1497
1498Internationalization, as defined in the C and POSIX standards, can be
e9bc6d6b
KW
1499criticized as incomplete and ungainly. They also have a tendency, like
1500standards groups, to divide the world into nations, when we all know
1501that the world can equally well be divided into bankers, bikers, gamers,
1502and so on.
5f05dabc 1503
b310b053
JH
1504=head1 Unicode and UTF-8
1505
7ee2ae1e 1506The support of Unicode is new starting from Perl version v5.6, and more fully
31f05a37
KW
1507implemented in versions v5.8 and later. See L<perluniintro>.
1508
a4a439fb
KW
1509Starting in Perl v5.20, UTF-8 locales are supported in Perl, except
1510C<LC_COLLATE> is only partially supported; collation support is improved
1511in Perl v5.26 to a level that may be sufficient for your needs
1512(see L</Category C<LC_COLLATE>: Collation: Text Comparisons and Sorting>).
1513
1514If you have Perl v5.16 or v5.18 and can't upgrade, you can use
66cbab2c
KW
1515
1516 use locale ':not_characters';
1517
1518When this form of the pragma is used, only the non-character portions of
1519locales are used by Perl, for example C<LC_NUMERIC>. Perl assumes that
1520you have translated all the characters it is to operate on into Unicode
1521(actually the platform's native character set (ASCII or EBCDIC) plus
1522Unicode). For data in files, this can conveniently be done by also
1523specifying
1524
1525 use open ':locale';
1526
1527This pragma arranges for all inputs from files to be translated into
1528Unicode from the current locale as specified in the environment (see
1529L</ENVIRONMENT>), and all outputs to files to be translated back
1530into the locale. (See L<open>). On a per-filehandle basis, you can
1531instead use the L<PerlIO::locale> module, or the L<Encode::Locale>
1532module, both available from CPAN. The latter module also has methods to
1533ease the handling of C<ARGV> and environment variables, and can be used
31f05a37 1534on individual strings. If you know that all your locales will be
66cbab2c
KW
1535UTF-8, as many are these days, you can use the L<B<-C>|perlrun/-C>
1536command line switch.
1537
1538This form of the pragma allows essentially seamless handling of locales
31f05a37 1539with Unicode. The collation order will be by Unicode code point order.
a4a439fb 1540L<Unicode::Collate> can be used to get Unicode rules collation.
66cbab2c 1541
31f05a37
KW
1542All the modules and switches just described can be used in v5.20 with
1543just plain C<use locale>, and, should the input locales not be UTF-8,
1544you'll get the less than ideal behavior, described below, that you get
1545with pre-v5.16 Perls, or when you use the locale pragma without the
1546C<:not_characters> parameter in v5.16 and v5.18. If you are using
1547exclusively UTF-8 locales in v5.20 and higher, the rest of this section
1548does not apply to you.
1549
1550There are two cases, multi-byte and single-byte locales. First
1551multi-byte:
1552
1553The only multi-byte (or wide character) locale that Perl is ever likely
1554to support is UTF-8. This is due to the difficulty of implementation,
1555the fact that high quality UTF-8 locales are now published for every
30659cfd 1556area of the world (L<https://unicode.org/Public/cldr/2.0.1/> for
e784ce68 1557ones that are already set-up, but from an earlier version;
30659cfd 1558L<https://unicode.org/Public/cldr/latest/> for the most up-to-date, but
e784ce68 1559you have to extract the POSIX information yourself), and that
31f05a37
KW
1560failing all that you can use the L<Encode> module to translate to/from
1561your locale. So, you'll have to do one of those things if you're using
1562one of these locales, such as Big5 or Shift JIS. For UTF-8 locales, in
1563Perls (pre v5.20) that don't have full UTF-8 locale support, they may
1564work reasonably well (depending on your C library implementation)
1565simply because both
dc4bfc4b
KW
1566they and Perl store characters that take up multiple bytes the same way.
1567However, some, if not most, C library implementations may not process
1568the characters in the upper half of the Latin-1 range (128 - 255)
cb88b78e 1569properly under C<LC_CTYPE>. To see if a character is a particular type
dc4bfc4b
KW
1570under a locale, Perl uses the functions like C<isalnum()>. Your C
1571library may not work for UTF-8 locales with those functions, instead
dbf3c4d7
KW
1572only working under the newer wide library functions like C<iswalnum()>,
1573which Perl does not use.
1574These multi-byte locales are treated like single-byte locales, and will
1575have the restrictions described below. Starting in Perl v5.22 a warning
1576message is raised when Perl detects a multi-byte locale that it doesn't
1577fully support.
e199995e 1578
31f05a37 1579For single-byte locales,
e199995e 1580Perl generally takes the tack to use locale rules on code points that can fit
66cbab2c
KW
1581in a single byte, and Unicode rules for those that can't (though this
1582isn't uniformly applied, see the note at the end of this section). This
1583prevents many problems in locales that aren't UTF-8. Suppose the locale
1584is ISO8859-7, Greek. The character at 0xD7 there is a capital Chi. But
1585in the ISO8859-1 locale, Latin1, it is a multiplication sign. The POSIX
1586regular expression character class C<[[:alpha:]]> will magically match
15870xD7 in the Greek locale but not in the Latin one.
e199995e 1588
1d2ab946 1589However, there are places where this breaks down. Certain Perl constructs are
b4ffc3db
TC
1590for Unicode only, such as C<\p{Alpha}>. They assume that 0xD7 always has its
1591Unicode meaning (or the equivalent on EBCDIC platforms). Since Latin1 is a
1592subset of Unicode and 0xD7 is the multiplication sign in both Latin1 and
1593Unicode, C<\p{Alpha}> will never match it, regardless of locale. A similar
0c880285 1594issue occurs with C<\N{...}>. Prior to v5.20, it is therefore a bad
31f05a37 1595idea to use C<\p{}> or
66cbab2c 1596C<\N{}> under plain C<use locale>--I<unless> you can guarantee that the
dbf3c4d7 1597locale will be ISO8859-1. Use POSIX character classes instead.
66cbab2c
KW
1598
1599Another problem with this approach is that operations that cross the
1600single byte/multiple byte boundary are not well-defined, and so are
4a70680a 1601disallowed. (This boundary is between the codepoints at 255/256.)
66cbab2c
KW
1602For example, lower casing LATIN CAPITAL LETTER Y WITH DIAERESIS (U+0178)
1603should return LATIN SMALL LETTER Y WITH DIAERESIS (U+00FF). But in the
1604Greek locale, for example, there is no character at 0xFF, and Perl
1605has no way of knowing what the character at 0xFF is really supposed to
1606represent. Thus it disallows the operation. In this mode, the
1607lowercase of U+0178 is itself.
1608
1609The same problems ensue if you enable automatic UTF-8-ification of your
e199995e 1610standard file handles, default C<open()> layer, and C<@ARGV> on non-ISO8859-1,
b4ffc3db
TC
1611non-UTF-8 locales (by using either the B<-C> command line switch or the
1612C<PERL_UNICODE> environment variable; see L<perlrun>).
1613Things are read in as UTF-8, which would normally imply a Unicode
1614interpretation, but the presence of a locale causes them to be interpreted
1615in that locale instead. For example, a 0xD7 code point in the Unicode
1616input, which should mean the multiplication sign, won't be interpreted by
66cbab2c 1617Perl that way under the Greek locale. This is not a problem
b4ffc3db 1618I<provided> you make certain that all locales will always and only be either
66cbab2c 1619an ISO8859-1, or, if you don't have a deficient C library, a UTF-8 locale.
b4ffc3db 1620
1d2ab946
KW
1621Still another problem is that this approach can lead to two code
1622points meaning the same character. Thus in a Greek locale, both U+03A7
1623and U+00D7 are GREEK CAPITAL LETTER CHI.
1624
613abc6d
KW
1625Because of all these problems, starting in v5.22, Perl will raise a
1626warning if a multi-byte (hence Unicode) code point is used when a
1627single-byte locale is in effect. (Although it doesn't check for this if
1628doing so would unreasonably slow execution down.)
1629
b4ffc3db
TC
1630Vendor locales are notoriously buggy, and it is difficult for Perl to test
1631its locale-handling code because this interacts with code that Perl has no
1632control over; therefore the locale-handling code in Perl may be buggy as
66cbab2c
KW
1633well. (However, the Unicode-supplied locales should be better, and
1634there is a feed back mechanism to correct any problems. See
1635L</Freely available locale definitions>.)
1636
7ee2ae1e 1637If you have Perl v5.16, the problems mentioned above go away if you use
66cbab2c 1638the C<:not_characters> parameter to the locale pragma (except for vendor
7ee2ae1e 1639bugs in the non-character portions). If you don't have v5.16, and you
66cbab2c
KW
1640I<do> have locales that work, using them may be worthwhile for certain
1641specific purposes, as long as you keep in mind the gotchas already
1642mentioned. For example, if the collation for your locales works, it
1643runs faster under locales than under L<Unicode::Collate>; and you gain
1644access to such things as the local currency symbol and the names of the
7ee2ae1e 1645months and days of the week. (But to hammer home the point, in v5.16,
66cbab2c
KW
1646you get this access without the downsides of locales by using the
1647C<:not_characters> form of the pragma.)
1648
1649Note: The policy of using locale rules for code points that can fit in a
1650byte, and Unicode rules for those that can't is not uniformly applied.
7ee2ae1e 1651Pre-v5.12, it was somewhat haphazard; in v5.12 it was applied fairly
66cbab2c 1652consistently to regular expression matching except for bracketed
7ee2ae1e 1653character classes; in v5.14 it was extended to all regex matches; and in
663d437a 1654v5.16 to the casing operations such as C<\L> and C<uc()>. For
dbf3c4d7 1655collation, in all releases so far, the system's C<strxfrm()> function is
a4a439fb
KW
1656called, and whatever it does is what you get. Starting in v5.26, various
1657bugs are fixed with the way perl uses this function.
b310b053 1658
5f05dabc
PP
1659=head1 BUGS
1660
4e615abd
KW
1661=head2 Collation of strings containing embedded C<NUL> characters
1662
6696cfa7
KW
1663C<NUL> characters will sort the same as the lowest collating control
1664character does, or to C<"\001"> in the unlikely event that there are no
1665control characters at all in the locale. In cases where the strings
1666don't contain this non-C<NUL> control, the results will be correct, and
1667in many locales, this control, whatever it might be, will rarely be
1668encountered. But there are cases where a C<NUL> should sort before this
1669control, but doesn't. If two strings do collate identically, the one
0c880285
KW
1670containing the C<NUL> will sort to earlier. Prior to 5.26, there were
1671more bugs.
4e615abd 1672
e9bc6d6b
KW
1673=head2 Multi-threaded
1674
1675XS code or C-language libraries called from it that use the system
1676L<C<setlocale(3)>> function (except on Windows) likely will not work
1677from a multi-threaded application without changes. See
1678L<perlxs/Locale-aware XS code>.
1679
1680An XS module that is locale-dependent could have been written under the
1681assumption that it will never be called in a multi-threaded environment,
1682and so uses other non-locale constructs that aren't multi-thread-safe.
1683See L<perlxs/Thread-aware system interfaces>.
1684
1685POSIX does not define a way to get the name of the current per-thread
1686locale. Some systems, such as Darwin and NetBSD do implement a
1687function, L<querylocale(3)> to do this. On non-Windows systems without
1688it, such as Linux, there are some additional caveats:
1689
1690=over
1691
1692=item *
1693
1694An embedded perl needs to be started up while the global locale is in
1695effect. See L<perlembed/Using embedded Perl with POSIX locales>.
1696
1697=item *
1698
1699It becomes more important for perl to know about all the possible
1700locale categories on the platform, even if they aren't apparently used
1701in your program. Perl knows all of the Linux ones. If your platform
1702has others, you can send email to L<mailto:perlbug@perl.org> for
1703inclusion of it in the next release. In the meantime, it is possible to
1704edit the Perl source to teach it about the category, and then recompile.
1705Search for instances of, say, C<LC_PAPER> in the source, and use that as
1706a template to add the omitted one.
1707
1708=item *
1709
1710It is possible, though hard to do, to call C<POSIX::setlocale> with a
1711locale that it doesn't recognize as syntactically legal, but actually is
1712legal on that system. This should happen only with embedded perls, or
1713if you hand-craft a locale name yourself.
1714
1715=back
1716
5f05dabc
PP
1717=head2 Broken systems
1718
5a964f20 1719In certain systems, the operating system's locale support
2bdf8add 1720is broken and cannot be fixed or used by Perl. Such deficiencies can
b4ffc3db 1721and will result in mysterious hangs and/or Perl core dumps when
2bdf8add 1722C<use locale> is in effect. When confronted with such a system,
7f2de2d2 1723please report in excruciating detail to <F<perlbug@perl.org>>, and
b4ffc3db 1724also contact your vendor: bug fixes may exist for these problems
2bdf8add 1725in your operating system. Sometimes such bug fixes are called an
83fb1bf0
KW
1726operating system upgrade. If you have the source for Perl, include in
1727the perlbug email the output of the test described above in L</Testing
1728for broken locales>.
5f05dabc
PP
1729
1730=head1 SEE ALSO
1731
b310b053 1732L<I18N::Langinfo>, L<perluniintro>, L<perlunicode>, L<open>,
106ab961 1733L<POSIX/localeconv>,
4bbcc6e8
JH
1734L<POSIX/setlocale>, L<POSIX/strcoll>, L<POSIX/strftime>,
1735L<POSIX/strtod>, L<POSIX/strxfrm>.
5f05dabc 1736
ccd65d51
KW
1737For special considerations when Perl is embedded in a C program,
1738see L<perlembed/Using embedded Perl with POSIX locales>.
1739
5f05dabc
PP
1740=head1 HISTORY
1741
b0c42ed9 1742Jarkko Hietaniemi's original F<perli18n.pod> heavily hacked by Dominic
5a964f20 1743Dunlop, assisted by the perl5-porters. Prose worked over a bit by
106ab961 1744Tom Christiansen, and now maintained by Perl 5 porters.