This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
perllocale: Wrap some text in C<...>
[perl5.git] / pod / perllocale.pod
CommitLineData
66cbab2c
KW
1=encoding utf8
2
5f05dabc 3=head1 NAME
4
b0c42ed9 5perllocale - Perl locale handling (internationalization and localization)
5f05dabc 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
cb88b78e 56=item Category C<LC_NUMERIC>: Numeric formatting
66cbab2c
KW
57
58This indicates how numbers should be formatted for human readability,
59for example the character used as the decimal point.
60
cb88b78e 61=item Category C<LC_MONETARY>: Formatting of monetary amounts
66cbab2c
KW
62
63=for comment
ebc3223b 64The nbsp below makes this look better (though not great)
66cbab2c
KW
65
66E<160>
67
cb88b78e 68=item Category C<LC_TIME>: Date/Time formatting
66cbab2c
KW
69
70=for comment
ebc3223b 71The nbsp below makes this look better (though not great)
66cbab2c
KW
72
73E<160>
74
cb88b78e 75=item Category C<LC_MESSAGES>: Error and other messages
66cbab2c 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 79
cb88b78e 80=item Category C<LC_COLLATE>: Collation
66cbab2c 81
76073c88 82This indicates the ordering of letters for comparison and sorting.
66cbab2c
KW
83In Latin alphabets, for example, "b", generally follows "a".
84
cb88b78e 85=item Category C<LC_CTYPE>: Character Types
66cbab2c
KW
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 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 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 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 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 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 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 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 204
b38d7779
KW
205The variables L<$!|perlvar/$ERRNO> (and its synonyms C<$ERRNO> and
206C<$OS_ERROR>) and L<$^E|perlvar/$EXTENDED_OS_ERROR> (and its synonym
c67edbcd
SH
207C<$EXTENDED_OS_ERROR>) when used as strings always are in terms of the
208current locale and as if within the scope of L<"use bytes"|bytes>. This is
b17e32ea 209likely to change in Perl v5.22.
2619d284 210
1d2ab946
KW
211=item *
212
2619d284
KW
213The current locale is also used when going outside of Perl with
214operations like L<system()|perlfunc/system LIST> or
215L<qxE<sol>E<sol>|perlop/qxE<sol>STRINGE<sol>>, if those operations are
216locale-sensitive.
217
1d2ab946
KW
218=item *
219
2619d284
KW
220Also Perl gives access to various C library functions through the
221L<POSIX> module. Some of those functions are always affected by the
222current locale. For example, C<POSIX::strftime()> uses C<LC_TIME>;
223C<POSIX::strtod()> uses C<LC_NUMERIC>; C<POSIX::strcoll()> and
224C<POSIX::strxfrm()> use C<LC_COLLATE>; and character classification
225functions like C<POSIX::isalnum()> use C<LC_CTYPE>. All such functions
226will behave according to the current underlying locale, even if that
1d2ab946 227locale isn't exposed to Perl space.
2619d284 228
1d2ab946
KW
229=item *
230
52686f2a
KW
231XS modules for all categories but C<LC_NUMERIC> get the underlying
232locale, and hence any C library functions they call will use that
233underlying locale. Perl always initializes C<LC_NUMERIC> to C<"C">
234because too many modules are unable to cope with the decimal point in a
235floating point number not being a dot (it's a comma in many locales).
236But note that these modules are vulnerable because C<LC_NUMERIC>
237currently can be changed at any time by a call to the C C<set_locale()>
238by XS code or by something XS code calls, or by C<POSIX::setlocale()> by
239Perl code. This is true also for the Perl-provided lite wrappers for XS
240modules to use some C library C<printf> functions:
241C<Gconvert>,
9fe6720f
KW
242L<my_sprintf|perlapi/my_sprintf>,
243L<my_snprintf|perlapi/my_snprintf>,
1d2ab946 244and
9fe6720f 245L<my_vsnprintf|perlapi/my_vsnprintf>.
1d2ab946
KW
246
247=back
248
ebc3223b
KW
249=for comment
250The nbsp below makes this look better (though not great)
251
252E<160>
253
254=item B<Lingering effects of C<S<use locale>>>
1d2ab946
KW
255
256Certain Perl operations that are set-up within the scope of a
b960a36e
KW
257C<use locale> variant retain that effect even outside the scope.
258These include:
259
260=over 4
261
262=item *
263
264The output format of a L<write()|perlfunc/write> is determined by an
265earlier format declaration (L<perlfunc/format>), so whether or not the
266output is affected by locale is determined by if the C<format()> is
267within the scope of a C<use locale> variant, not whether the C<write()>
268is.
269
270=item *
271
272Regular expression patterns can be compiled using
273L<qrE<sol>E<sol>|perlop/qrE<sol>STRINGE<sol>msixpodual> with actual
274matching deferred to later. Again, it is whether or not the compilation
275was done within the scope of C<use locale> that determines the match
276behavior, not if the matches are done within such a scope or not.
277
278=back
279
ebc3223b
KW
280=for comment
281The nbsp below makes this look better (though not great)
282
283E<160>
284
4c9b78f4 285=item B<Under C<"use locale ':not_characters';">>
66cbab2c
KW
286
287=over 4
288
289=item *
290
b960a36e
KW
291All the non-Perl operations.
292
293=item *
294
295B<Format declarations> (L<perlfunc/format>) and hence any subsequent
296C<write()>s use C<LC_NUMERIC>.
66cbab2c
KW
297
298=item *
299
b960a36e
KW
300B<stringification and output> use C<LC_NUMERIC>.
301These include the results of
302C<print()>,
303C<printf()>,
304C<say()>,
305and
306C<sprintf()>.
66cbab2c
KW
307
308=back
309
310=for comment
ebc3223b 311The nbsp below makes this look better (though not great)
66cbab2c
KW
312
313E<160>
314
4c9b78f4 315=item B<Under just plain C<"use locale";>>
66cbab2c 316
66cbab2c
KW
317=over 4
318
5f05dabc 319=item *
320
b960a36e
KW
321All the above operations
322
323=item *
324
325B<The comparison operators> (C<lt>, C<le>, C<cmp>, C<ge>, and C<gt>) use
39332f68 326C<LC_COLLATE>. C<sort()> is also affected if used without an
5a964f20 327explicit comparison function, because it uses C<cmp> by default.
14280422 328
5a964f20 329B<Note:> C<eq> and C<ne> are unaffected by locale: they always
de108802 330perform a char-by-char comparison of their scalar operands. What's
14280422
DD
331more, if C<cmp> finds that its operands are equal according to the
332collation sequence specified by the current locale, it goes on to
de108802
RGS
333perform a char-by-char comparison, and only returns I<0> (equal) if the
334operands are char-for-char identical. If you really want to know whether
5a964f20 335two strings--which C<eq> and C<cmp> may consider different--are equal
14280422 336as far as collation in the locale is concerned, see the discussion in
cb88b78e 337L<Category C<LC_COLLATE>: Collation>.
5f05dabc 338
339=item *
340
39332f68
KW
341B<Regular expressions and case-modification functions> (C<uc()>, C<lc()>,
342C<ucfirst()>, and C<lcfirst()>) use C<LC_CTYPE>
5f05dabc 343
5f05dabc 344=back
345
66cbab2c 346=back
5f05dabc 347
5a964f20 348The default behavior is restored with the S<C<no locale>> pragma, or
ef3087ec 349upon reaching the end of the block enclosing C<use locale>.
66cbab2c
KW
350Note that C<use locale> and C<use locale ':not_characters'> may be
351nested, and that what is in effect within an inner scope will revert to
352the outer scope's rules at the end of the inner scope.
5f05dabc 353
5a964f20 354The string result of any operation that uses locale
14280422
DD
355information is tainted, as it is possible for a locale to be
356untrustworthy. See L<"SECURITY">.
5f05dabc 357
358=head2 The setlocale function
359
14280422 360You can switch locales as often as you wish at run time with the
39332f68 361C<POSIX::setlocale()> function:
5f05dabc 362
5f05dabc 363 # Import locale-handling tool set from POSIX module.
364 # This example uses: setlocale -- the function call
365 # LC_CTYPE -- explained below
2619d284
KW
366 # (Showing the testing for success/failure of operations is
367 # omitted in these examples to avoid distracting from the main
ebc3223b 368 # point)
6ea81ccf 369
5f05dabc 370 use POSIX qw(locale_h);
dfcc8045
KW
371 use locale;
372 my $old_locale;
5f05dabc 373
14280422 374 # query and save the old locale
5f05dabc 375 $old_locale = setlocale(LC_CTYPE);
376
377 setlocale(LC_CTYPE, "fr_CA.ISO8859-1");
378 # LC_CTYPE now in locale "French, Canada, codeset ISO 8859-1"
379
380 setlocale(LC_CTYPE, "");
65ebb059
KW
381 # LC_CTYPE now reset to the default defined by the
382 # LC_ALL/LC_CTYPE/LANG environment variables, or to the system
383 # default. See below for documentation.
5f05dabc 384
385 # restore the old locale
386 setlocale(LC_CTYPE, $old_locale);
387
39332f68 388The first argument of C<setlocale()> gives the B<category>, the second the
14280422
DD
389B<locale>. The category tells in what aspect of data processing you
390want to apply locale-specific rules. Category names are discussed in
66cbab2c 391L</LOCALE CATEGORIES> and L</"ENVIRONMENT">. The locale is the name of a
14280422
DD
392collection of customization information corresponding to a particular
393combination of language, country or territory, and codeset. Read on for
394hints on the naming of locales: not all systems name locales as in the
395example.
396
39332f68 397If no second argument is provided and the category is something other
cb88b78e 398than C<LC_ALL>, the function returns a string naming the current locale
502a173a 399for the category. You can use this value as the second argument in a
f170b852
KW
400subsequent call to C<setlocale()>, B<but> on some platforms the string
401is opaque, not something that most people would be able to decipher as
402to what locale it means.
502a173a 403
cb88b78e 404If no second argument is provided and the category is C<LC_ALL>, the
502a173a 405result is implementation-dependent. It may be a string of
c052850d 406concatenated locale names (separator also implementation-dependent)
39332f68 407or a single locale name. Please consult your L<setlocale(3)> man page for
502a173a
JH
408details.
409
410If a second argument is given and it corresponds to a valid locale,
411the locale for the category is set to that value, and the function
412returns the now-current locale value. You can then use this in yet
39332f68 413another call to C<setlocale()>. (In some implementations, the return
502a173a
JH
414value may sometimes differ from the value you gave as the second
415argument--think of it as an alias for the value you gave.)
5f05dabc 416
417As the example shows, if the second argument is an empty string, the
418category's locale is returned to the default specified by the
419corresponding environment variables. Generally, this results in a
5a964f20 420return to the default that was in force when Perl started up: changes
54310121 421to the environment made by the application after startup may or may not
5a964f20 422be noticed, depending on your system's C library.
5f05dabc 423
66cbab2c
KW
424Note that Perl ignores the current C<LC_CTYPE> and C<LC_COLLATE> locales
425within the scope of a C<use locale ':not_characters'>.
426
f170b852 427If C<set_locale()> fails for some reason (for example, an attempt to set
dfcc8045
KW
428to a locale unknown to the system), the locale for the category is not
429changed, and the function returns C<undef>.
430
2619d284 431
39332f68 432For further information about the categories, consult L<setlocale(3)>.
3e6e419a
JH
433
434=head2 Finding locales
435
39332f68 436For locales available in your system, consult also L<setlocale(3)> to
5a964f20
TC
437see whether it leads to the list of available locales (search for the
438I<SEE ALSO> section). If that fails, try the following command lines:
5f05dabc 439
440 locale -a
441
442 nlsinfo
443
444 ls /usr/lib/nls/loc
445
446 ls /usr/lib/locale
447
448 ls /usr/lib/nls
449
b478f28d
JH
450 ls /usr/share/locale
451
5f05dabc 452and see whether they list something resembling these
453
2bdf8add 454 en_US.ISO8859-1 de_DE.ISO8859-1 ru_RU.ISO8859-5
502a173a 455 en_US.iso88591 de_DE.iso88591 ru_RU.iso88595
2bdf8add 456 en_US de_DE ru_RU
14280422 457 en de ru
2bdf8add
JH
458 english german russian
459 english.iso88591 german.iso88591 russian.iso88595
502a173a 460 english.roman8 russian.koi8r
5f05dabc 461
39332f68 462Sadly, even though the calling interface for C<setlocale()> has been
528d65ad 463standardized, names of locales and the directories where the
5a964f20 464configuration resides have not been. The basic form of the name is
528d65ad
JH
465I<language_territory>B<.>I<codeset>, but the latter parts after
466I<language> are not always present. The I<language> and I<country>
467are usually from the standards B<ISO 3166> and B<ISO 639>, the
468two-letter abbreviations for the countries and the languages of the
469world, respectively. The I<codeset> part often mentions some B<ISO
4708859> character set, the Latin codesets. For example, C<ISO 8859-1>
471is the so-called "Western European codeset" that can be used to encode
472most Western European languages adequately. Again, there are several
473ways to write even the name of that one standard. Lamentably.
5f05dabc 474
14280422
DD
475Two special locales are worth particular mention: "C" and "POSIX".
476Currently these are effectively the same locale: the difference is
5a964f20
TC
477mainly that the first one is defined by the C standard, the second by
478the POSIX standard. They define the B<default locale> in which
14280422 479every program starts in the absence of locale information in its
5a964f20 480environment. (The I<default> default locale, if you will.) Its language
39332f68
KW
481is (American) English and its character codeset ASCII or, rarely, a
482superset thereof (such as the "DEC Multinational Character Set
483(DEC-MCS)"). B<Warning>. The C locale delivered by some vendors
484may not actually exactly match what the C standard calls for. So
485beware.
5f05dabc 486
14280422
DD
487B<NOTE>: Not all systems have the "POSIX" locale (not all systems are
488POSIX-conformant), so use "C" when you need explicitly to specify this
489default locale.
5f05dabc 490
3e6e419a
JH
491=head2 LOCALE PROBLEMS
492
5a964f20 493You may encounter the following warning message at Perl startup:
3e6e419a
JH
494
495 perl: warning: Setting locale failed.
496 perl: warning: Please check that your locale settings:
497 LC_ALL = "En_US",
498 LANG = (unset)
499 are supported and installed on your system.
500 perl: warning: Falling back to the standard locale ("C").
501
cb88b78e 502This means that your locale settings had C<LC_ALL> set to "En_US" and
5a964f20
TC
503LANG exists but has no value. Perl tried to believe you but could not.
504Instead, Perl gave up and fell back to the "C" locale, the default locale
65ebb059
KW
505that is supposed to work no matter what. (On Windows, it first tries
506falling back to the system default locale.) This usually means your
507locale settings were wrong, they mention locales your system has never
508heard of, or the locale installation in your system has problems (for
509example, some system files are broken or missing). There are quick and
510temporary fixes to these problems, as well as more thorough and lasting
511fixes.
3e6e419a 512
83fb1bf0
KW
513=head2 Testing for broken locales
514
515If you are building Perl from source, the Perl test suite file
516F<lib/locale.t> can be used to test the locales on your system.
517Setting the environment variable C<PERL_DEBUG_FULL_TEST> to 1
518will cause it to output detailed results. For example, on Linux, you
519could say
520
1d2ab946 521 PERL_DEBUG_FULL_TEST=1 ./perl -T -Ilib lib/locale.t > locale.log 2>&1
83fb1bf0
KW
522
523Besides many other tests, it will test every locale it finds on your
524system to see if they conform to the POSIX standard. If any have
525errors, it will include a summary near the end of the output of which
526locales passed all its tests, and which failed, and why.
527
3e6e419a
JH
528=head2 Temporarily fixing locale problems
529
5a964f20 530The two quickest fixes are either to render Perl silent about any
3e6e419a
JH
531locale inconsistencies or to run Perl under the default locale "C".
532
533Perl's moaning about locale problems can be silenced by setting the
cb88b78e 534environment variable C<PERL_BADLANG> to a zero value, for example "0".
900bd440
JH
535This method really just sweeps the problem under the carpet: you tell
536Perl to shut up even when Perl sees that something is wrong. Do not
537be surprised if later something locale-dependent misbehaves.
3e6e419a
JH
538
539Perl can be run under the "C" locale by setting the environment
cb88b78e
KW
540variable C<LC_ALL> to "C". This method is perhaps a bit more civilized
541than the C<PERL_BADLANG> approach, but setting C<LC_ALL> (or
5a964f20
TC
542other locale variables) may affect other programs as well, not just
543Perl. In particular, external programs run from within Perl will see
3e6e419a 544these changes. If you make the new settings permanent (read on), all
f979aebc 545programs you run see the changes. See L<"ENVIRONMENT"> for
5a964f20 546the full list of relevant environment variables and L<USING LOCALES>
e05ffc7d 547for their effects in Perl. Effects in other programs are
cb88b78e 548easily deducible. For example, the variable C<LC_COLLATE> may well affect
b432a672 549your B<sort> program (or whatever the program that arranges "records"
3e6e419a
JH
550alphabetically in your system is called).
551
5a964f20
TC
552You can test out changing these variables temporarily, and if the
553new settings seem to help, put those settings into your shell startup
554files. Consult your local documentation for the exact details. For in
555Bourne-like shells (B<sh>, B<ksh>, B<bash>, B<zsh>):
3e6e419a
JH
556
557 LC_ALL=en_US.ISO8859-1
558 export LC_ALL
559
5a964f20
TC
560This assumes that we saw the locale "en_US.ISO8859-1" using the commands
561discussed above. We decided to try that instead of the above faulty
562locale "En_US"--and in Cshish shells (B<csh>, B<tcsh>)
3e6e419a
JH
563
564 setenv LC_ALL en_US.ISO8859-1
c47ff5f1 565
c406981e
JH
566or if you have the "env" application you can do in any shell
567
568 env LC_ALL=en_US.ISO8859-1 perl ...
569
5a964f20 570If you do not know what shell you have, consult your local
3e6e419a
JH
571helpdesk or the equivalent.
572
573=head2 Permanently fixing locale problems
574
5a964f20
TC
575The slower but superior fixes are when you may be able to yourself
576fix the misconfiguration of your own environment variables. The
3e6e419a
JH
577mis(sing)configuration of the whole system's locales usually requires
578the help of your friendly system administrator.
579
5a964f20
TC
580First, see earlier in this document about L<Finding locales>. That tells
581how to find which locales are really supported--and more importantly,
582installed--on your system. In our example error message, environment
583variables affecting the locale are listed in the order of decreasing
584importance (and unset variables do not matter). Therefore, having
585LC_ALL set to "En_US" must have been the bad choice, as shown by the
586error message. First try fixing locale settings listed first.
3e6e419a 587
5a964f20
TC
588Second, if using the listed commands you see something B<exactly>
589(prefix matches do not count and case usually counts) like "En_US"
590without the quotes, then you should be okay because you are using a
591locale name that should be installed and available in your system.
4a4eefd0 592In this case, see L<Permanently fixing your system's locale configuration>.
3e6e419a 593
4a4eefd0 594=head2 Permanently fixing your system's locale configuration
3e6e419a 595
5a964f20 596This is when you see something like:
3e6e419a
JH
597
598 perl: warning: Please check that your locale settings:
599 LC_ALL = "En_US",
600 LANG = (unset)
601 are supported and installed on your system.
602
603but then cannot see that "En_US" listed by the above-mentioned
5a964f20
TC
604commands. You may see things like "en_US.ISO8859-1", but that isn't
605the same. In this case, try running under a locale
606that you can list and which somehow matches what you tried. The
3e6e419a 607rules for matching locale names are a bit vague because
e05ffc7d 608standardization is weak in this area. See again the
13a2d996 609L<Finding locales> about general rules.
3e6e419a 610
b687b08b 611=head2 Fixing system locale configuration
3e6e419a 612
5a964f20
TC
613Contact a system administrator (preferably your own) and report the exact
614error message you get, and ask them to read this same documentation you
615are now reading. They should be able to check whether there is something
616wrong with the locale configuration of the system. The L<Finding locales>
617section is unfortunately a bit vague about the exact commands and places
618because these things are not that standardized.
3e6e419a 619
5f05dabc 620=head2 The localeconv function
621
39332f68 622The C<POSIX::localeconv()> function allows you to get particulars of the
14280422 623locale-dependent numeric formatting information specified by the current
a835cd47
KW
624underlying C<LC_NUMERIC> and C<LC_MONETARY> locales (regardless of
625whether called from within the scope of C<S<use locale>> or not). (If
626you just want the name of
39332f68 627the current locale for a particular category, use C<POSIX::setlocale()>
5a964f20 628with a single parameter--see L<The setlocale function>.)
5f05dabc 629
630 use POSIX qw(locale_h);
5f05dabc 631
632 # Get a reference to a hash of locale-dependent info
633 $locale_values = localeconv();
634
635 # Output sorted list of the values
636 for (sort keys %$locale_values) {
14280422 637 printf "%-20s = %s\n", $_, $locale_values->{$_}
5f05dabc 638 }
639
39332f68 640C<localeconv()> takes no arguments, and returns B<a reference to> a hash.
5a964f20 641The keys of this hash are variable names for formatting, such as
502a173a 642C<decimal_point> and C<thousands_sep>. The values are the
cea6626f 643corresponding, er, values. See L<POSIX/localeconv> for a longer
502a173a
JH
644example listing the categories an implementation might be expected to
645provide; some provide more and others fewer. You don't need an
39332f68 646explicit C<use locale>, because C<localeconv()> always observes the
502a173a 647current locale.
5f05dabc 648
5a964f20
TC
649Here's a simple-minded example program that rewrites its command-line
650parameters as integers correctly formatted in the current locale:
5f05dabc 651
ef3087ec
KW
652 use POSIX qw(locale_h);
653
654 # Get some of locale's numeric formatting parameters
655 my ($thousands_sep, $grouping) =
656 @{localeconv()}{'thousands_sep', 'grouping'};
657
658 # Apply defaults if values are missing
659 $thousands_sep = ',' unless $thousands_sep;
660
661 # grouping and mon_grouping are packed lists
662 # of small integers (characters) telling the
663 # grouping (thousand_seps and mon_thousand_seps
664 # being the group dividers) of numbers and
665 # monetary quantities. The integers' meanings:
666 # 255 means no more grouping, 0 means repeat
667 # the previous grouping, 1-254 means use that
668 # as the current grouping. Grouping goes from
669 # right to left (low to high digits). In the
670 # below we cheat slightly by never using anything
671 # else than the first grouping (whatever that is).
672 if ($grouping) {
673 @grouping = unpack("C*", $grouping);
674 } else {
675 @grouping = (3);
676 }
677
678 # Format command line params for current locale
679 for (@ARGV) {
680 $_ = int; # Chop non-integer part
681 1 while
682 s/(\d)(\d{$grouping[0]}($|$thousands_sep))/$1$thousands_sep$2/;
683 print "$_";
684 }
685 print "\n";
5f05dabc 686
03ceeedf
KW
687Note that if the platform doesn't have C<LC_NUMERIC> and/or
688C<LC_MONETARY> available or enabled, the corresponding elements of the
689hash will be missing.
690
74c76037 691=head2 I18N::Langinfo
4bbcc6e8
JH
692
693Another interface for querying locale-dependent information is the
39332f68 694C<I18N::Langinfo::langinfo()> function, available at least in Unix-like
4bbcc6e8
JH
695systems and VMS.
696
39332f68
KW
697The following example will import the C<langinfo()> function itself and
698three constants to be used as arguments to C<langinfo()>: a constant for
74c76037
JH
699the abbreviated first day of the week (the numbering starts from
700Sunday = 1) and two more constants for the affirmative and negative
701answers for a yes/no question in the current locale.
4bbcc6e8 702
74c76037 703 use I18N::Langinfo qw(langinfo ABDAY_1 YESSTR NOSTR);
4bbcc6e8 704
ef3087ec
KW
705 my ($abday_1, $yesstr, $nostr)
706 = map { langinfo } qw(ABDAY_1 YESSTR NOSTR);
4bbcc6e8 707
74c76037 708 print "$abday_1? [$yesstr/$nostr] ";
4bbcc6e8 709
74c76037
JH
710In other words, in the "C" (or English) locale the above will probably
711print something like:
712
e05ffc7d 713 Sun? [yes/no]
4bbcc6e8
JH
714
715See L<I18N::Langinfo> for more information.
716
5f05dabc 717=head1 LOCALE CATEGORIES
718
5a964f20
TC
719The following subsections describe basic locale categories. Beyond these,
720some combination categories allow manipulation of more than one
721basic category at a time. See L<"ENVIRONMENT"> for a discussion of these.
5f05dabc 722
cb88b78e 723=head2 Category C<LC_COLLATE>: Collation
5f05dabc 724
66cbab2c
KW
725In the scope of S<C<use locale>> (but not a
726C<use locale ':not_characters'>), Perl looks to the C<LC_COLLATE>
5a964f20 727environment variable to determine the application's notions on collation
b4ffc3db
TC
728(ordering) of characters. For example, "b" follows "a" in Latin
729alphabets, but where do "E<aacute>" and "E<aring>" belong? And while
f87fa335 730"color" follows "chocolate" in English, what about in traditional Spanish?
5f05dabc 731
60f0fa02
JH
732The following collations all make sense and you may meet any of them
733if you "use locale".
734
735 A B C D E a b c d e
35316ca3 736 A a B b C c D d E e
60f0fa02
JH
737 a A b B c C d D e E
738 a b c d e A B C D E
739
f1cbbd6e 740Here is a code snippet to tell what "word"
5a964f20 741characters are in the current locale, in that locale's order:
5f05dabc 742
743 use locale;
35316ca3 744 print +(sort grep /\w/, map { chr } 0..255), "\n";
5f05dabc 745
14280422
DD
746Compare this with the characters that you see and their order if you
747state explicitly that the locale should be ignored:
5f05dabc 748
749 no locale;
35316ca3 750 print +(sort grep /\w/, map { chr } 0..255), "\n";
5f05dabc 751
752This machine-native collation (which is what you get unless S<C<use
753locale>> has appeared earlier in the same block) must be used for
754sorting raw binary data, whereas the locale-dependent collation of the
b0c42ed9 755first example is useful for natural text.
5f05dabc 756
14280422
DD
757As noted in L<USING LOCALES>, C<cmp> compares according to the current
758collation locale when C<use locale> is in effect, but falls back to a
de108802 759char-by-char comparison for strings that the locale says are equal. You
39332f68 760can use C<POSIX::strcoll()> if you don't want this fall-back:
14280422
DD
761
762 use POSIX qw(strcoll);
763 $equal_in_locale =
764 !strcoll("space and case ignored", "SpaceAndCaseIgnored");
765
39332f68 766C<$equal_in_locale> will be true if the collation locale specifies a
5a964f20 767dictionary-like ordering that ignores space characters completely and
9e3a2af8 768which folds case.
14280422 769
31f05a37
KW
770Perl only supports single-byte locales for C<LC_COLLATE>. This means
771that a UTF-8 locale likely will just give you machine-native ordering.
772Use L<Unicode::Collate> for the full implementation of the Unicode
773Collation Algorithm.
774
5a964f20 775If you have a single string that you want to check for "equality in
14280422 776locale" against several others, you might think you could gain a little
39332f68 777efficiency by using C<POSIX::strxfrm()> in conjunction with C<eq>:
14280422
DD
778
779 use POSIX qw(strxfrm);
780 $xfrm_string = strxfrm("Mixed-case string");
781 print "locale collation ignores spaces\n"
782 if $xfrm_string eq strxfrm("Mixed-casestring");
783 print "locale collation ignores hyphens\n"
784 if $xfrm_string eq strxfrm("Mixedcase string");
785 print "locale collation ignores case\n"
786 if $xfrm_string eq strxfrm("mixed-case string");
787
39332f68 788C<strxfrm()> takes a string and maps it into a transformed string for use
de108802 789in char-by-char comparisons against other transformed strings during
14280422 790collation. "Under the hood", locale-affected Perl comparison operators
39332f68
KW
791call C<strxfrm()> for both operands, then do a char-by-char
792comparison of the transformed strings. By calling C<strxfrm()> explicitly
14280422 793and using a non locale-affected comparison, the example attempts to save
5a964f20 794a couple of transformations. But in fact, it doesn't save anything: Perl
2ae324a7 795magic (see L<perlguts/Magic Variables>) creates the transformed version of a
5a964f20 796string the first time it's needed in a comparison, then keeps this version around
14280422 797in case it's needed again. An example rewritten the easy way with
e38874e2 798C<cmp> runs just about as fast. It also copes with null characters
39332f68 799embedded in strings; if you call C<strxfrm()> directly, it treats the first
5a964f20
TC
800null it finds as a terminator. don't expect the transformed strings
801it produces to be portable across systems--or even from one revision
39332f68 802of your operating system to the next. In short, don't call C<strxfrm()>
e38874e2 803directly: let Perl do it for you.
14280422 804
5a964f20 805Note: C<use locale> isn't shown in some of these examples because it isn't
dfcc8045
KW
806needed: C<strcoll()> and C<strxfrm()> are POSIX functions
807which use the standard system-supplied C<libc> functions that
808always obey the current C<LC_COLLATE> locale.
5f05dabc 809
cb88b78e 810=head2 Category C<LC_CTYPE>: Character Types
5f05dabc 811
66cbab2c
KW
812In the scope of S<C<use locale>> (but not a
813C<use locale ':not_characters'>), Perl obeys the C<LC_CTYPE> locale
14280422 814setting. This controls the application's notion of which characters are
ebc3223b
KW
815alphabetic, numeric, punctuation, I<etc>. This affects Perl's C<\w>
816regular expression metanotation,
f1cbbd6e 817which stands for alphanumeric characters--that is, alphabetic,
ebc3223b
KW
818numeric, and the platform's native underscore.
819(Consult L<perlre> for more information about
14280422 820regular expressions.) Thanks to C<LC_CTYPE>, depending on your locale
b4ffc3db
TC
821setting, characters like "E<aelig>", "E<eth>", "E<szlig>", and
822"E<oslash>" may be understood as C<\w> characters.
ebc3223b
KW
823It also affects things like C<\s>, C<\D>, and the POSIX character
824classes, like C<[[:graph:]]>. (See L<perlrecharclass> for more
825information on all these.)
5f05dabc 826
2c268ad5 827The C<LC_CTYPE> locale also provides the map used in transliterating
68dc0745 828characters between lower and uppercase. This affects the case-mapping
39332f68 829functions--C<fc()>, C<lc()>, C<lcfirst()>, C<uc()>, and C<ucfirst()>; case-mapping
b9cc4f69
KW
830interpolation with C<\F>, C<\l>, C<\L>, C<\u>, or C<\U> in double-quoted
831strings and C<s///> substitutions; and case-independent regular expression
e38874e2
DD
832pattern matching using the C<i> modifier.
833
2da736a2 834Finally, C<LC_CTYPE> affects the (deprecated) POSIX character-class test
1d2ab946
KW
835functions--C<POSIX::isalpha()>, C<POSIX::islower()>, and so on. For
836example, if you move from the "C" locale to a 7-bit Scandinavian one,
837you may find--possibly to your surprise--that "|" moves from the
838C<POSIX::ispunct()> class to C<POSIX::isalpha()>.
ef3087ec
KW
839Unfortunately, this creates big problems for regular expressions. "|" still
840means alternation even though it matches C<\w>.
5f05dabc 841
31f05a37
KW
842Starting in v5.20, Perl supports UTF-8 locales for C<LC_CTYPE>, but
843otherwise Perl only supports single-byte locales, such as the ISO 8859
844series. This means that wide character locales, for example for Asian
845languages, are not supported. The UTF-8 locale support is actually a
846superset of POSIX locales, because it is really full Unicode behavior
847as if no locale were in effect at all (except for tainting; see
848L</SECURITY>). POSIX locales, even UTF-8 ones,
849are lacking certain concepts in Unicode, such as the idea that changing
850the case of a character could expand to be more than one character.
851Perl in a UTF-8 locale, will give you that expansion. Prior to v5.20,
852Perl treated a UTF-8 locale on some platforms like an ISO 8859-1 one,
853with some restrictions, and on other platforms more like the "C" locale.
854For releases v5.16 and v5.18, C<S<use locale 'not_characters>> could be
855used as a workaround for this (see L</Unicode and UTF-8>).
856
5d63e270
KW
857Note that there are quite a few things that are unaffected by the
858current locale. All the escape sequences for particular characters,
859C<\n> for example, always mean the platform's native one. This means,
860for example, that C<\N> in regular expressions (every character
1d2ab946 861but new-line) works on the platform character set.
5d63e270 862
14280422
DD
863B<Note:> A broken or malicious C<LC_CTYPE> locale definition may result
864in clearly ineligible characters being considered to be alphanumeric by
e199995e 865your application. For strict matching of (mundane) ASCII letters and
5a964f20 866digits--for example, in command strings--locale-aware applications
e199995e 867should use C<\w> with the C</a> regular expression modifier. See L<"SECURITY">.
5f05dabc 868
fee33030 869=head2 Category C<LC_NUMERIC>: Numeric Formatting
5f05dabc 870
b960a36e
KW
871After a proper C<POSIX::setlocale()> call, and within the scope of one
872of the C<use locale> variants, Perl obeys the C<LC_NUMERIC>
2095dafa 873locale information, which controls an application's idea of how numbers
b960a36e
KW
874should be formatted for human readability.
875In most implementations the only effect is to
b4ffc3db 876change the character used for the decimal point--perhaps from "." to ",".
b960a36e 877The functions aren't aware of such niceties as thousands separation and
2095dafa 878so on. (See L<The localeconv function> if you care about these things.)
5a964f20 879
b960a36e
KW
880 use POSIX qw(strtod setlocale LC_NUMERIC);
881 use locale;
5f05dabc 882
b960a36e 883 setlocale LC_NUMERIC, "";
14280422 884
b960a36e 885 $n = 5/2; # Assign numeric 2.5 to $n
5f05dabc 886
b960a36e 887 $a = " $n"; # Locale-dependent conversion to string
5f05dabc 888
b960a36e 889 print "half five is $n\n"; # Locale-dependent output
5f05dabc 890
b960a36e 891 printf "half five is %g\n", $n; # Locale-dependent output
5f05dabc 892
b960a36e
KW
893 print "DECIMAL POINT IS COMMA\n"
894 if $n == (strtod("2,5"))[0]; # Locale-dependent conversion
5f05dabc 895
4bbcc6e8
JH
896See also L<I18N::Langinfo> and C<RADIXCHAR>.
897
cb88b78e 898=head2 Category C<LC_MONETARY>: Formatting of monetary amounts
5f05dabc 899
e199995e 900The C standard defines the C<LC_MONETARY> category, but not a function
5a964f20 901that is affected by its contents. (Those with experience of standards
b0c42ed9 902committees will recognize that the working group decided to punt on the
fa9b773e
KW
903issue.) Consequently, Perl essentially takes no notice of it. If you
904really want to use C<LC_MONETARY>, you can query its contents--see
e05ffc7d
KW
905L<The localeconv function>--and use the information that it returns in your
906application's own formatting of currency amounts. However, you may well
907find that the information, voluminous and complex though it may be, still
908does not quite meet your requirements: currency formatting is a hard nut
13a2d996 909to crack.
5f05dabc 910
4bbcc6e8
JH
911See also L<I18N::Langinfo> and C<CRNCYSTR>.
912
cb88b78e 913=head2 C<LC_TIME>
5f05dabc 914
39332f68 915Output produced by C<POSIX::strftime()>, which builds a formatted
5f05dabc 916human-readable date/time string, is affected by the current C<LC_TIME>
917locale. Thus, in a French locale, the output produced by the C<%B>
918format element (full month name) for the first month of the year would
5a964f20 919be "janvier". Here's how to get a list of long month names in the
5f05dabc 920current locale:
921
922 use POSIX qw(strftime);
14280422
DD
923 for (0..11) {
924 $long_month_name[$_] =
925 strftime("%B", 0, 0, 0, 1, $_, 96);
5f05dabc 926 }
927
2619d284
KW
928Note: C<use locale> isn't needed in this example: C<strftime()> is a POSIX
929function which uses the standard system-supplied C<libc> function that
930always obeys the current C<LC_TIME> locale.
5f05dabc 931
4bbcc6e8 932See also L<I18N::Langinfo> and C<ABDAY_1>..C<ABDAY_7>, C<DAY_1>..C<DAY_7>,
2a2bf5f4 933C<ABMON_1>..C<ABMON_12>, and C<ABMON_1>..C<ABMON_12>.
4bbcc6e8 934
5f05dabc 935=head2 Other categories
936
2619d284
KW
937The remaining locale categories are not currently used by Perl itself.
938But again note that things Perl interacts with may use these, including
939extensions outside the standard Perl distribution, and by the
98a6f11e 940operating system and its utilities. Note especially that the string
941value of C<$!> and the error messages given by external utilities may
942be changed by C<LC_MESSAGES>. If you want to have portable error
265f5c4a 943codes, use C<%!>. See L<Errno>.
14280422
DD
944
945=head1 SECURITY
946
5a964f20 947Although the main discussion of Perl security issues can be found in
14280422
DD
948L<perlsec>, a discussion of Perl's locale handling would be incomplete
949if it did not draw your attention to locale-dependent security issues.
5a964f20
TC
950Locales--particularly on systems that allow unprivileged users to
951build their own locales--are untrustworthy. A malicious (or just plain
14280422
DD
952broken) locale can make a locale-aware application give unexpected
953results. Here are a few possibilities:
954
955=over 4
956
957=item *
958
959Regular expression checks for safe file names or mail addresses using
5a964f20 960C<\w> may be spoofed by an C<LC_CTYPE> locale that claims that
14280422
DD
961characters such as "E<gt>" and "|" are alphanumeric.
962
963=item *
964
e38874e2 965String interpolation with case-mapping, as in, say, C<$dest =
cb88b78e 966"C:\U$name.$ext">, may produce dangerous results if a bogus C<LC_CTYPE>
e38874e2
DD
967case-mapping table is in effect.
968
969=item *
970
14280422
DD
971A sneaky C<LC_COLLATE> locale could result in the names of students with
972"D" grades appearing ahead of those with "A"s.
973
974=item *
975
5a964f20 976An application that takes the trouble to use information in
14280422 977C<LC_MONETARY> may format debits as if they were credits and vice versa
5a964f20 978if that locale has been subverted. Or it might make payments in US
14280422
DD
979dollars instead of Hong Kong dollars.
980
981=item *
982
39332f68 983The date and day names in dates formatted by C<strftime()> could be
14280422 984manipulated to advantage by a malicious user able to subvert the
5a964f20 985C<LC_DATE> locale. ("Look--it says I wasn't in the building on
14280422
DD
986Sunday.")
987
988=back
989
990Such dangers are not peculiar to the locale system: any aspect of an
5a964f20 991application's environment which may be modified maliciously presents
14280422 992similar challenges. Similarly, they are not specific to Perl: any
5a964f20 993programming language that allows you to write programs that take
14280422
DD
994account of their environment exposes you to these issues.
995
5a964f20
TC
996Perl cannot protect you from all possibilities shown in the
997examples--there is no substitute for your own vigilance--but, when
14280422 998C<use locale> is in effect, Perl uses the tainting mechanism (see
5a964f20 999L<perlsec>) to mark string results that become locale-dependent, and
14280422 1000which may be untrustworthy in consequence. Here is a summary of the
5a964f20 1001tainting behavior of operators and functions that may be affected by
14280422
DD
1002the locale:
1003
1004=over 4
1005
551e1d92
RB
1006=item *
1007
1008B<Comparison operators> (C<lt>, C<le>, C<ge>, C<gt> and C<cmp>):
14280422
DD
1009
1010Scalar true/false (or less/equal/greater) result is never tainted.
1011
551e1d92
RB
1012=item *
1013
1d2ab946 1014B<Case-mapping interpolation> (with C<\l>, C<\L>, C<\u>, C<\U>, or C<\F>)
e38874e2
DD
1015
1016Result string containing interpolated material is tainted if
66cbab2c 1017C<use locale> (but not S<C<use locale ':not_characters'>>) is in effect.
e38874e2 1018
551e1d92
RB
1019=item *
1020
1021B<Matching operator> (C<m//>):
14280422
DD
1022
1023Scalar true/false result never tainted.
1024
1d2ab946
KW
1025All subpatterns, either delivered as a list-context result or as C<$1>
1026I<etc>., are tainted if C<use locale> (but not
1027S<C<use locale ':not_characters'>>) is in effect, and the subpattern
63baef57
KW
1028regular expression contains a locale-dependent construct. These
1029constructs include C<\w> (to match an alphanumeric character), C<\W>
1030(non-alphanumeric character), C<\b> and C<\B> (word-boundary and
1031non-boundardy, which depend on what C<\w> and C<\W> match), C<\s>
1032(whitespace character), C<\S> (non whitespace character), C<\d> and
1033C<\D> (digits and non-digits), and the POSIX character classes, such as
1034C<[:alpha:]> (see L<perlrecharclass/POSIX Character Classes>).
1035
1036Tainting is also likely if the pattern is to be matched
1037case-insensitively (via C</i>). The exception is if all the code points
1038to be matched this way are above 255 and do not have folds under Unicode
1039rules to below 256. Tainting is not done for these because Perl
1040only uses Unicode rules for such code points, and those rules are the
1041same no matter what the current locale.
1042
1d2ab946
KW
1043The matched-pattern variables, C<$&>, C<$`> (pre-match), C<$'>
1044(post-match), and C<$+> (last match) also are tainted.
14280422 1045
551e1d92
RB
1046=item *
1047
1048B<Substitution operator> (C<s///>):
14280422 1049
e38874e2 1050Has the same behavior as the match operator. Also, the left
66cbab2c
KW
1051operand of C<=~> becomes tainted when C<use locale>
1052(but not S<C<use locale ':not_characters'>>) is in effect if modified as
1053a result of a substitution based on a regular
1d2ab946
KW
1054expression match involving any of the things mentioned in the previous
1055item, or of case-mapping, such as C<\l>, C<\L>,C<\u>, C<\U>, or C<\F>.
14280422 1056
551e1d92
RB
1057=item *
1058
39332f68 1059B<Output formatting functions> (C<printf()> and C<write()>):
14280422 1060
3cf03d68
JH
1061Results are never tainted because otherwise even output from print,
1062for example C<print(1/7)>, should be tainted if C<use locale> is in
1063effect.
14280422 1064
551e1d92
RB
1065=item *
1066
39332f68 1067B<Case-mapping functions> (C<lc()>, C<lcfirst()>, C<uc()>, C<ucfirst()>):
14280422 1068
66cbab2c
KW
1069Results are tainted if C<use locale> (but not
1070S<C<use locale ':not_characters'>>) is in effect.
14280422 1071
551e1d92
RB
1072=item *
1073
39332f68
KW
1074B<POSIX locale-dependent functions> (C<localeconv()>, C<strcoll()>,
1075C<strftime()>, C<strxfrm()>):
14280422
DD
1076
1077Results are never tainted.
1078
551e1d92
RB
1079=item *
1080
1d2ab946
KW
1081B<POSIX character class tests> (C<POSIX::isalnum()>,
1082C<POSIX::isalpha()>, C<POSIX::isdigit()>, C<POSIX::isgraph()>,
1083C<POSIX::islower()>, C<POSIX::isprint()>, C<POSIX::ispunct()>,
1084C<POSIX::isspace()>, C<POSIX::isupper()>, C<POSIX::isxdigit()>):
14280422
DD
1085
1086True/false results are never tainted.
1087
1088=back
1089
1090Three examples illustrate locale-dependent tainting.
1091The first program, which ignores its locale, won't run: a value taken
54310121 1092directly from the command line may not be used to name an output file
14280422
DD
1093when taint checks are enabled.
1094
1095 #/usr/local/bin/perl -T
1096 # Run with taint checking
1097
54310121 1098 # Command line sanity check omitted...
14280422
DD
1099 $tainted_output_file = shift;
1100
1101 open(F, ">$tainted_output_file")
3183d96c 1102 or warn "Open of $tainted_output_file failed: $!\n";
14280422
DD
1103
1104The program can be made to run by "laundering" the tainted value through
5a964f20
TC
1105a regular expression: the second example--which still ignores locale
1106information--runs, creating the file named on its command line
14280422
DD
1107if it can.
1108
1109 #/usr/local/bin/perl -T
1110
1111 $tainted_output_file = shift;
1112 $tainted_output_file =~ m%[\w/]+%;
1113 $untainted_output_file = $&;
1114
1115 open(F, ">$untainted_output_file")
1116 or warn "Open of $untainted_output_file failed: $!\n";
1117
5a964f20 1118Compare this with a similar but locale-aware program:
14280422
DD
1119
1120 #/usr/local/bin/perl -T
1121
1122 $tainted_output_file = shift;
1123 use locale;
1124 $tainted_output_file =~ m%[\w/]+%;
1125 $localized_output_file = $&;
1126
1127 open(F, ">$localized_output_file")
1128 or warn "Open of $localized_output_file failed: $!\n";
1129
1d2ab946 1130This third program fails to run because C<$&> is tainted: it is the result
5a964f20 1131of a match involving C<\w> while C<use locale> is in effect.
5f05dabc 1132
1133=head1 ENVIRONMENT
1134
1135=over 12
1136
ee1ec05f
KW
1137=item PERL_SKIP_LOCALE_INIT
1138
1139This environment variable, available starting in Perl v5.20, and if it
1140evaluates to a TRUE value, tells Perl to not use the rest of the
1141environment variables to initialize with. Instead, Perl uses whatever
1142the current locale settings are. This is particularly useful in
1143embedded environments, see
1144L<perlembed/Using embedded Perl with POSIX locales>.
1145
5f05dabc 1146=item PERL_BADLANG
1147
14280422 1148A string that can suppress Perl's warning about failed locale settings
54310121 1149at startup. Failure can occur if the locale support in the operating
5a964f20 1150system is lacking (broken) in some way--or if you mistyped the name of
900bd440
JH
1151a locale when you set up your environment. If this environment
1152variable is absent, or has a value that does not evaluate to integer
1153zero--that is, "0" or ""-- Perl will complain about locale setting
1154failures.
5f05dabc 1155
cb88b78e 1156B<NOTE>: C<PERL_BADLANG> only gives you a way to hide the warning message.
14280422
DD
1157The message tells about some problem in your system's locale support,
1158and you should investigate what the problem is.
5f05dabc 1159
1160=back
1161
1162The following environment variables are not specific to Perl: They are
39332f68 1163part of the standardized (ISO C, XPG4, POSIX 1.c) C<setlocale()> method
b385bb4d
KW
1164for controlling an application's opinion on data. Windows is non-POSIX,
1165but Perl arranges for the following to work as described anyway.
65ebb059
KW
1166If the locale given by an environment variable is not valid, Perl tries
1167the next lower one in priority. If none are valid, on Windows, the
1168system default locale is then tried. If all else fails, the C<"C">
1169locale is used. If even that doesn't work, something is badly broken,
c5e9a8e7 1170but Perl tries to forge ahead with whatever the locale settings might
65ebb059 1171be.
5f05dabc 1172
1173=over 12
1174
cb88b78e 1175=item C<LC_ALL>
5f05dabc 1176
5a964f20 1177C<LC_ALL> is the "override-all" locale environment variable. If
5f05dabc 1178set, it overrides all the rest of the locale environment variables.
1179
cb88b78e 1180=item C<LANGUAGE>
528d65ad
JH
1181
1182B<NOTE>: C<LANGUAGE> is a GNU extension, it affects you only if you
1183are using the GNU libc. This is the case if you are using e.g. Linux.
e1020413 1184If you are using "commercial" Unixes you are most probably I<not>
22b6f60d
JH
1185using GNU libc and you can ignore C<LANGUAGE>.
1186
1187However, in the case you are using C<LANGUAGE>: it affects the
1188language of informational, warning, and error messages output by
1189commands (in other words, it's like C<LC_MESSAGES>) but it has higher
96090e4f 1190priority than C<LC_ALL>. Moreover, it's not a single value but
22b6f60d
JH
1191instead a "path" (":"-separated list) of I<languages> (not locales).
1192See the GNU C<gettext> library documentation for more information.
528d65ad 1193
cb88b78e 1194=item C<LC_CTYPE>.
5f05dabc 1195
1196In the absence of C<LC_ALL>, C<LC_CTYPE> chooses the character type
1197locale. In the absence of both C<LC_ALL> and C<LC_CTYPE>, C<LANG>
1198chooses the character type locale.
1199
cb88b78e 1200=item C<LC_COLLATE>
5f05dabc 1201
14280422
DD
1202In the absence of C<LC_ALL>, C<LC_COLLATE> chooses the collation
1203(sorting) locale. In the absence of both C<LC_ALL> and C<LC_COLLATE>,
1204C<LANG> chooses the collation locale.
5f05dabc 1205
cb88b78e 1206=item C<LC_MONETARY>
5f05dabc 1207
14280422
DD
1208In the absence of C<LC_ALL>, C<LC_MONETARY> chooses the monetary
1209formatting locale. In the absence of both C<LC_ALL> and C<LC_MONETARY>,
1210C<LANG> chooses the monetary formatting locale.
5f05dabc 1211
cb88b78e 1212=item C<LC_NUMERIC>
5f05dabc 1213
1214In the absence of C<LC_ALL>, C<LC_NUMERIC> chooses the numeric format
1215locale. In the absence of both C<LC_ALL> and C<LC_NUMERIC>, C<LANG>
1216chooses the numeric format.
1217
cb88b78e 1218=item C<LC_TIME>
5f05dabc 1219
14280422
DD
1220In the absence of C<LC_ALL>, C<LC_TIME> chooses the date and time
1221formatting locale. In the absence of both C<LC_ALL> and C<LC_TIME>,
1222C<LANG> chooses the date and time formatting locale.
5f05dabc 1223
cb88b78e 1224=item C<LANG>
5f05dabc 1225
14280422
DD
1226C<LANG> is the "catch-all" locale environment variable. If it is set, it
1227is used as the last resort after the overall C<LC_ALL> and the
cb88b78e 1228category-specific C<LC_I<foo>>
5f05dabc 1229
1230=back
1231
7e4353e9
RGS
1232=head2 Examples
1233
cb88b78e 1234The C<LC_NUMERIC> controls the numeric output:
7e4353e9 1235
ef3087ec
KW
1236 use locale;
1237 use POSIX qw(locale_h); # Imports setlocale() and the LC_ constants.
1238 setlocale(LC_NUMERIC, "fr_FR") or die "Pardon";
1239 printf "%g\n", 1.23; # If the "fr_FR" succeeded, probably shows 1,23.
7e4353e9 1240
39332f68 1241and also how strings are parsed by C<POSIX::strtod()> as numbers:
7e4353e9 1242
ef3087ec
KW
1243 use locale;
1244 use POSIX qw(locale_h strtod);
1245 setlocale(LC_NUMERIC, "de_DE") or die "Entschuldigung";
1246 my $x = strtod("2,34") + 5;
1247 print $x, "\n"; # Probably shows 7,34.
7e4353e9 1248
5f05dabc 1249=head1 NOTES
1250
b960a36e
KW
1251=head2 String C<eval> and C<LC_NUMERIC>
1252
1253A string L<eval|perlfunc/eval EXPR> parses its expression as standard
1254Perl. It is therefore expecting the decimal point to be a dot. If
1255C<LC_NUMERIC> is set to have this be a comma instead, the parsing will
1256be confused, perhaps silently.
1257
1258 use locale;
1259 use POSIX qw(locale_h);
1260 setlocale(LC_NUMERIC, "fr_FR") or die "Pardon";
1261 my $a = 1.2;
1262 print eval "$a + 1.5";
1263 print "\n";
1264
1265prints C<13,5>. This is because in that locale, the comma is the
1266decimal point character. The C<eval> thus expands to:
1267
1268 eval "1,2 + 1.5"
1269
1270and the result is not what you likely expected. No warnings are
1271generated. If you do string C<eval>'s within the scope of
1272S<C<use locale>>, you should instead change the C<eval> line to do
1273something like:
1274
1275 print eval "no locale; $a + 1.5";
1276
1277This prints C<2.7>.
1278
5f05dabc 1279=head2 Backward compatibility
1280
b0c42ed9 1281Versions of Perl prior to 5.004 B<mostly> ignored locale information,
5a964f20
TC
1282generally behaving as if something similar to the C<"C"> locale were
1283always in force, even if the program environment suggested otherwise
1284(see L<The setlocale function>). By default, Perl still behaves this
1285way for backward compatibility. If you want a Perl application to pay
1286attention to locale information, you B<must> use the S<C<use locale>>
062ca197
KW
1287pragma (see L<The use locale pragma>) or, in the unlikely event
1288that you want to do so for just pattern matching, the
70709c68
KW
1289C</l> regular expression modifier (see L<perlre/Character set
1290modifiers>) to instruct it to do so.
b0c42ed9
JH
1291
1292Versions of Perl from 5.002 to 5.003 did use the C<LC_CTYPE>
5a964f20
TC
1293information if available; that is, C<\w> did understand what
1294were the letters according to the locale environment variables.
b0c42ed9
JH
1295The problem was that the user had no control over the feature:
1296if the C library supported locales, Perl used them.
1297
1298=head2 I18N:Collate obsolete
1299
5a964f20 1300In versions of Perl prior to 5.004, per-locale collation was possible
b0c42ed9
JH
1301using the C<I18N::Collate> library module. This module is now mildly
1302obsolete and should be avoided in new applications. The C<LC_COLLATE>
1303functionality is now integrated into the Perl core language: One can
1304use locale-specific scalar data completely normally with C<use locale>,
1305so there is no longer any need to juggle with the scalar references of
1306C<I18N::Collate>.
5f05dabc 1307
14280422 1308=head2 Sort speed and memory use impacts
5f05dabc 1309
1310Comparing and sorting by locale is usually slower than the default
14280422
DD
1311sorting; slow-downs of two to four times have been observed. It will
1312also consume more memory: once a Perl scalar variable has participated
1313in any string comparison or sorting operation obeying the locale
1314collation rules, it will take 3-15 times more memory than before. (The
1315exact multiplier depends on the string's contents, the operating system
1316and the locale.) These downsides are dictated more by the operating
1317system's implementation of the locale system than by Perl.
5f05dabc 1318
5f05dabc 1319=head2 Freely available locale definitions
1320
66cbab2c
KW
1321The Unicode CLDR project extracts the POSIX portion of many of its
1322locales, available at
1323
1324 http://unicode.org/Public/cldr/latest/
1325
08d7a6b2
LB
1326There is a large collection of locale definitions at:
1327
1328 http://std.dkuug.dk/i18n/WG15-collection/locales/
1329
1330You should be aware that it is
14280422 1331unsupported, and is not claimed to be fit for any purpose. If your
5a964f20 1332system allows installation of arbitrary locales, you may find the
14280422
DD
1333definitions useful as they are, or as a basis for the development of
1334your own locales.
5f05dabc 1335
14280422 1336=head2 I18n and l10n
5f05dabc 1337
b0c42ed9
JH
1338"Internationalization" is often abbreviated as B<i18n> because its first
1339and last letters are separated by eighteen others. (You may guess why
1340the internalin ... internaliti ... i18n tends to get abbreviated.) In
1341the same way, "localization" is often abbreviated to B<l10n>.
14280422
DD
1342
1343=head2 An imperfect standard
1344
1345Internationalization, as defined in the C and POSIX standards, can be
1346criticized as incomplete, ungainly, and having too large a granularity.
1347(Locales apply to a whole process, when it would arguably be more useful
1348to have them apply to a single thread, window group, or whatever.) They
1349also have a tendency, like standards groups, to divide the world into
1350nations, when we all know that the world can equally well be divided
e199995e 1351into bankers, bikers, gamers, and so on.
5f05dabc 1352
b310b053
JH
1353=head1 Unicode and UTF-8
1354
7ee2ae1e 1355The support of Unicode is new starting from Perl version v5.6, and more fully
31f05a37
KW
1356implemented in versions v5.8 and later. See L<perluniintro>.
1357
1358Starting in Perl v5.20, UTF-8 locales are supported in Perl, except for
1359C<LC_COLLATE> (use L<Unicode::Collate> instead). If you have Perl v5.16
1360or v5.18 and can't upgrade, you can use
66cbab2c
KW
1361
1362 use locale ':not_characters';
1363
1364When this form of the pragma is used, only the non-character portions of
1365locales are used by Perl, for example C<LC_NUMERIC>. Perl assumes that
1366you have translated all the characters it is to operate on into Unicode
1367(actually the platform's native character set (ASCII or EBCDIC) plus
1368Unicode). For data in files, this can conveniently be done by also
1369specifying
1370
1371 use open ':locale';
1372
1373This pragma arranges for all inputs from files to be translated into
1374Unicode from the current locale as specified in the environment (see
1375L</ENVIRONMENT>), and all outputs to files to be translated back
1376into the locale. (See L<open>). On a per-filehandle basis, you can
1377instead use the L<PerlIO::locale> module, or the L<Encode::Locale>
1378module, both available from CPAN. The latter module also has methods to
1379ease the handling of C<ARGV> and environment variables, and can be used
31f05a37 1380on individual strings. If you know that all your locales will be
66cbab2c
KW
1381UTF-8, as many are these days, you can use the L<B<-C>|perlrun/-C>
1382command line switch.
1383
1384This form of the pragma allows essentially seamless handling of locales
31f05a37
KW
1385with Unicode. The collation order will be by Unicode code point order.
1386It is strongly
66cbab2c
KW
1387recommended that when you need to order and sort strings that you use
1388the standard module L<Unicode::Collate> which gives much better results
1389in many instances than you can get with the old-style locale handling.
1390
31f05a37
KW
1391All the modules and switches just described can be used in v5.20 with
1392just plain C<use locale>, and, should the input locales not be UTF-8,
1393you'll get the less than ideal behavior, described below, that you get
1394with pre-v5.16 Perls, or when you use the locale pragma without the
1395C<:not_characters> parameter in v5.16 and v5.18. If you are using
1396exclusively UTF-8 locales in v5.20 and higher, the rest of this section
1397does not apply to you.
1398
1399There are two cases, multi-byte and single-byte locales. First
1400multi-byte:
1401
1402The only multi-byte (or wide character) locale that Perl is ever likely
1403to support is UTF-8. This is due to the difficulty of implementation,
1404the fact that high quality UTF-8 locales are now published for every
1405area of the world (L<http://unicode.org/Public/cldr/latest/>), and that
1406failing all that you can use the L<Encode> module to translate to/from
1407your locale. So, you'll have to do one of those things if you're using
1408one of these locales, such as Big5 or Shift JIS. For UTF-8 locales, in
1409Perls (pre v5.20) that don't have full UTF-8 locale support, they may
1410work reasonably well (depending on your C library implementation)
1411simply because both
dc4bfc4b
KW
1412they and Perl store characters that take up multiple bytes the same way.
1413However, some, if not most, C library implementations may not process
1414the characters in the upper half of the Latin-1 range (128 - 255)
cb88b78e 1415properly under C<LC_CTYPE>. To see if a character is a particular type
dc4bfc4b
KW
1416under a locale, Perl uses the functions like C<isalnum()>. Your C
1417library may not work for UTF-8 locales with those functions, instead
1418only working under the newer wide library functions like C<iswalnum()>.
31f05a37
KW
1419However, they are treated like single-byte locales, and will have the
1420restrictions described below.
e199995e 1421
31f05a37 1422For single-byte locales,
e199995e 1423Perl generally takes the tack to use locale rules on code points that can fit
66cbab2c
KW
1424in a single byte, and Unicode rules for those that can't (though this
1425isn't uniformly applied, see the note at the end of this section). This
1426prevents many problems in locales that aren't UTF-8. Suppose the locale
1427is ISO8859-7, Greek. The character at 0xD7 there is a capital Chi. But
1428in the ISO8859-1 locale, Latin1, it is a multiplication sign. The POSIX
1429regular expression character class C<[[:alpha:]]> will magically match
14300xD7 in the Greek locale but not in the Latin one.
e199995e 1431
1d2ab946 1432However, there are places where this breaks down. Certain Perl constructs are
b4ffc3db
TC
1433for Unicode only, such as C<\p{Alpha}>. They assume that 0xD7 always has its
1434Unicode meaning (or the equivalent on EBCDIC platforms). Since Latin1 is a
1435subset of Unicode and 0xD7 is the multiplication sign in both Latin1 and
1436Unicode, C<\p{Alpha}> will never match it, regardless of locale. A similar
31f05a37
KW
1437issue occurs with C<\N{...}>. Prior to v5.20, It is therefore a bad
1438idea to use C<\p{}> or
66cbab2c
KW
1439C<\N{}> under plain C<use locale>--I<unless> you can guarantee that the
1440locale will be a ISO8859-1. Use POSIX character classes instead.
1441
1442Another problem with this approach is that operations that cross the
1443single byte/multiple byte boundary are not well-defined, and so are
4a70680a 1444disallowed. (This boundary is between the codepoints at 255/256.)
66cbab2c
KW
1445For example, lower casing LATIN CAPITAL LETTER Y WITH DIAERESIS (U+0178)
1446should return LATIN SMALL LETTER Y WITH DIAERESIS (U+00FF). But in the
1447Greek locale, for example, there is no character at 0xFF, and Perl
1448has no way of knowing what the character at 0xFF is really supposed to
1449represent. Thus it disallows the operation. In this mode, the
1450lowercase of U+0178 is itself.
1451
1452The same problems ensue if you enable automatic UTF-8-ification of your
e199995e 1453standard file handles, default C<open()> layer, and C<@ARGV> on non-ISO8859-1,
b4ffc3db
TC
1454non-UTF-8 locales (by using either the B<-C> command line switch or the
1455C<PERL_UNICODE> environment variable; see L<perlrun>).
1456Things are read in as UTF-8, which would normally imply a Unicode
1457interpretation, but the presence of a locale causes them to be interpreted
1458in that locale instead. For example, a 0xD7 code point in the Unicode
1459input, which should mean the multiplication sign, won't be interpreted by
66cbab2c 1460Perl that way under the Greek locale. This is not a problem
b4ffc3db 1461I<provided> you make certain that all locales will always and only be either
66cbab2c 1462an ISO8859-1, or, if you don't have a deficient C library, a UTF-8 locale.
b4ffc3db 1463
1d2ab946
KW
1464Still another problem is that this approach can lead to two code
1465points meaning the same character. Thus in a Greek locale, both U+03A7
1466and U+00D7 are GREEK CAPITAL LETTER CHI.
1467
b4ffc3db
TC
1468Vendor locales are notoriously buggy, and it is difficult for Perl to test
1469its locale-handling code because this interacts with code that Perl has no
1470control over; therefore the locale-handling code in Perl may be buggy as
66cbab2c
KW
1471well. (However, the Unicode-supplied locales should be better, and
1472there is a feed back mechanism to correct any problems. See
1473L</Freely available locale definitions>.)
1474
7ee2ae1e 1475If you have Perl v5.16, the problems mentioned above go away if you use
66cbab2c 1476the C<:not_characters> parameter to the locale pragma (except for vendor
7ee2ae1e 1477bugs in the non-character portions). If you don't have v5.16, and you
66cbab2c
KW
1478I<do> have locales that work, using them may be worthwhile for certain
1479specific purposes, as long as you keep in mind the gotchas already
1480mentioned. For example, if the collation for your locales works, it
1481runs faster under locales than under L<Unicode::Collate>; and you gain
1482access to such things as the local currency symbol and the names of the
7ee2ae1e 1483months and days of the week. (But to hammer home the point, in v5.16,
66cbab2c
KW
1484you get this access without the downsides of locales by using the
1485C<:not_characters> form of the pragma.)
1486
1487Note: The policy of using locale rules for code points that can fit in a
1488byte, and Unicode rules for those that can't is not uniformly applied.
7ee2ae1e 1489Pre-v5.12, it was somewhat haphazard; in v5.12 it was applied fairly
66cbab2c 1490consistently to regular expression matching except for bracketed
7ee2ae1e
KW
1491character classes; in v5.14 it was extended to all regex matches; and in
1492v5.16 to the casing operations such as C<"\L"> and C<uc()>. For
66cbab2c
KW
1493collation, in all releases, the system's C<strxfrm()> function is called,
1494and whatever it does is what you get.
b310b053 1495
5f05dabc 1496=head1 BUGS
1497
1498=head2 Broken systems
1499
5a964f20 1500In certain systems, the operating system's locale support
2bdf8add 1501is broken and cannot be fixed or used by Perl. Such deficiencies can
b4ffc3db 1502and will result in mysterious hangs and/or Perl core dumps when
2bdf8add 1503C<use locale> is in effect. When confronted with such a system,
7f2de2d2 1504please report in excruciating detail to <F<perlbug@perl.org>>, and
b4ffc3db 1505also contact your vendor: bug fixes may exist for these problems
2bdf8add 1506in your operating system. Sometimes such bug fixes are called an
83fb1bf0
KW
1507operating system upgrade. If you have the source for Perl, include in
1508the perlbug email the output of the test described above in L</Testing
1509for broken locales>.
5f05dabc 1510
1511=head1 SEE ALSO
1512
b310b053
JH
1513L<I18N::Langinfo>, L<perluniintro>, L<perlunicode>, L<open>,
1514L<POSIX/isalnum>, L<POSIX/isalpha>,
4bbcc6e8
JH
1515L<POSIX/isdigit>, L<POSIX/isgraph>, L<POSIX/islower>,
1516L<POSIX/isprint>, L<POSIX/ispunct>, L<POSIX/isspace>,
1517L<POSIX/isupper>, L<POSIX/isxdigit>, L<POSIX/localeconv>,
1518L<POSIX/setlocale>, L<POSIX/strcoll>, L<POSIX/strftime>,
1519L<POSIX/strtod>, L<POSIX/strxfrm>.
5f05dabc 1520
ccd65d51
KW
1521For special considerations when Perl is embedded in a C program,
1522see L<perlembed/Using embedded Perl with POSIX locales>.
1523
5f05dabc 1524=head1 HISTORY
1525
b0c42ed9 1526Jarkko Hietaniemi's original F<perli18n.pod> heavily hacked by Dominic
5a964f20 1527Dunlop, assisted by the perl5-porters. Prose worked over a bit by
c052850d 1528Tom Christiansen, and updated by Perl 5 porters.