This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
perllocale: Wording/formatting nits
[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
0c880285 43releases or for better control, use L<Unicode::Collate>. Perl continues to
a4a439fb
KW
44support the old non UTF-8 locales as well. There are currently no UTF-8
45locales for EBCDIC platforms.
31f05a37 46
66cbab2c
KW
47(Unicode is also creating C<CLDR>, the "Common Locale Data Repository",
48L<http://cldr.unicode.org/> which includes more types of information than
49are available in the POSIX locale system. At the time of this writing,
50there was no CPAN module that provides access to this XML-encoded data.
e784ce68
KW
51However, it is possible to compute the POSIX locale data from them, and
52earlier CLDR versions had these already extracted for you as UTF-8 locales
53L<http://unicode.org/Public/cldr/2.0.1/>.)
66cbab2c
KW
54
55=head1 WHAT IS A LOCALE
56
57A locale is a set of data that describes various aspects of how various
58communities in the world categorize their world. These categories are
59broken down into the following types (some of which include a brief
60note here):
61
62=over
63
cb88b78e 64=item Category C<LC_NUMERIC>: Numeric formatting
66cbab2c
KW
65
66This indicates how numbers should be formatted for human readability,
67for example the character used as the decimal point.
68
cb88b78e 69=item Category C<LC_MONETARY>: Formatting of monetary amounts
66cbab2c
KW
70
71=for comment
ebc3223b 72The nbsp below makes this look better (though not great)
66cbab2c
KW
73
74E<160>
75
cb88b78e 76=item Category C<LC_TIME>: Date/Time formatting
66cbab2c
KW
77
78=for comment
ebc3223b 79The nbsp below makes this look better (though not great)
66cbab2c
KW
80
81E<160>
82
cb88b78e 83=item Category C<LC_MESSAGES>: Error and other messages
66cbab2c 84
2619d284 85This is used by Perl itself only for accessing operating system error
03c702c5 86messages via L<$!|perlvar/$ERRNO> and L<$^E|perlvar/$EXTENDED_OS_ERROR>.
66cbab2c 87
cb88b78e 88=item Category C<LC_COLLATE>: Collation
66cbab2c 89
76073c88 90This indicates the ordering of letters for comparison and sorting.
66cbab2c
KW
91In Latin alphabets, for example, "b", generally follows "a".
92
cb88b78e 93=item Category C<LC_CTYPE>: Character Types
66cbab2c
KW
94
95This indicates, for example if a character is an uppercase letter.
96
2619d284
KW
97=item Other categories
98
99Some platforms have other categories, dealing with such things as
100measurement units and paper sizes. None of these are used directly by
101Perl, but outside operations that Perl interacts with may use
d6ded950 102these. See L</Not within the scope of "use locale"> below.
2619d284 103
66cbab2c
KW
104=back
105
2619d284
KW
106More details on the categories used by Perl are given below in L</LOCALE
107CATEGORIES>.
66cbab2c
KW
108
109Together, these categories go a long way towards being able to customize
110a single program to run in many different locations. But there are
111deficiencies, so keep reading.
5f05dabc
PP
112
113=head1 PREPARING TO USE LOCALES
114
c5c88224
KW
115Perl itself (outside the L<POSIX> module) will not use locales unless
116specifically requested to (but
2619d284
KW
117again note that Perl may interact with code that does use them). Even
118if there is such a request, B<all> of the following must be true
b960a36e 119for it to work properly:
5f05dabc
PP
120
121=over 4
122
123=item *
124
125B<Your operating system must support the locale system>. If it does,
39332f68 126you should find that the C<setlocale()> function is a documented part of
5f05dabc
PP
127its C library.
128
129=item *
130
5a964f20 131B<Definitions for locales that you use must be installed>. You, or
14280422
DD
132your system administrator, must make sure that this is the case. The
133available locales, the location in which they are kept, and the manner
5a964f20
TC
134in which they are installed all vary from system to system. Some systems
135provide only a few, hard-wired locales and do not allow more to be
136added. Others allow you to add "canned" locales provided by the system
137supplier. Still others allow you or the system administrator to define
14280422 138and add arbitrary locales. (You may have to ask your supplier to
5a964f20 139provide canned locales that are not delivered with your operating
14280422 140system.) Read your system documentation for further illumination.
5f05dabc
PP
141
142=item *
143
144B<Perl must believe that the locale system is supported>. If it does,
145C<perl -V:d_setlocale> will say that the value for C<d_setlocale> is
146C<define>.
147
148=back
149
150If you want a Perl application to process and present your data
151according to a particular locale, the application code should include
5a0de581 152the S<C<use locale>> pragma (see L</The "use locale" pragma>) where
5f05dabc
PP
153appropriate, and B<at least one> of the following must be true:
154
155=over 4
156
c052850d 157=item 1
5f05dabc 158
66cbab2c 159B<The locale-determining environment variables (see L</"ENVIRONMENT">)
5a964f20 160must be correctly set up> at the time the application is started, either
ef3087ec 161by yourself or by whomever set up your system account; or
5f05dabc 162
c052850d 163=item 2
5f05dabc 164
14280422 165B<The application must set its own locale> using the method described in
5a0de581 166L</The setlocale function>.
5f05dabc
PP
167
168=back
169
170=head1 USING LOCALES
171
d6ded950 172=head2 The C<"use locale"> pragma
5f05dabc 173
fc82b82e
KW
174WARNING! Do NOT use this pragma in scripts that have multiple
175L<threads|threads> active. The locale is not local to a single thread.
176Another thread may change the locale at any time, which could cause at a
177minimum that a given thread is operating in a locale it isn't expecting
178to be in. On some platforms, segfaults can also occur. The locale
179change need not be explicit; some operations cause perl to change the
26a9b33a
KW
180locale itself. You are vulnerable simply by having done a S<C<"use
181locale">>.
fc82b82e 182
c5c88224
KW
183By default, Perl itself (outside the L<POSIX> module)
184ignores the current locale. The S<C<use locale>>
66cbab2c 185pragma tells Perl to use the current locale for some operations.
d6ded950
KW
186Starting in v5.16, there are optional parameters to this pragma,
187described below, which restrict which operations are affected by it.
c052850d
KW
188
189The current locale is set at execution time by
190L<setlocale()|/The setlocale function> described below. If that function
191hasn't yet been called in the course of the program's execution, the
66cbab2c 192current locale is that which was determined by the L</"ENVIRONMENT"> in
ebc3223b 193effect at the start of the program.
dfcc8045 194If there is no valid environment, the current locale is whatever the
65ebb059
KW
195system default has been set to. On POSIX systems, it is likely, but
196not necessarily, the "C" locale. On Windows, the default is set via the
197computer's S<C<Control Panel-E<gt>Regional and Language Options>> (or its
198current equivalent).
c052850d
KW
199
200The operations that are affected by locale are:
5f05dabc
PP
201
202=over 4
203
d6ded950 204=item B<Not within the scope of C<"use locale">>
b960a36e 205
d6ded950
KW
206Only certain operations originating outside Perl should be affected, as
207follows:
1d2ab946
KW
208
209=over 4
210
211=item *
2619d284 212
663d437a 213The current locale is used when going outside of Perl with
2619d284
KW
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
9accf6df 224C<POSIX::strxfrm()> use C<LC_COLLATE>. All such functions
2619d284 225will behave according to the current underlying locale, even if that
1d2ab946 226locale isn't exposed to Perl space.
2619d284 227
d9e2553c
KW
228This applies as well to L<I18N::Langinfo>.
229
1d2ab946
KW
230=item *
231
52686f2a
KW
232XS modules for all categories but C<LC_NUMERIC> get the underlying
233locale, and hence any C library functions they call will use that
8a384d3a 234underlying locale. For more discussion, see L<perlxs/CAVEATS>.
1d2ab946
KW
235
236=back
237
c5c88224
KW
238Note that all C programs (including the perl interpreter, which is
239written in C) always have an underlying locale. That locale is the "C"
240locale unless changed by a call to L<setlocale()|/The setlocale
241function>. When Perl starts up, it changes the underlying locale to the
242one which is indicated by the L</ENVIRONMENT>. When using the L<POSIX>
243module or writing XS code, it is important to keep in mind that the
244underlying locale may be something other than "C", even if the program
245hasn't explicitly changed it.
246
ebc3223b
KW
247=for comment
248The nbsp below makes this look better (though not great)
249
250E<160>
251
252=item B<Lingering effects of C<S<use locale>>>
1d2ab946
KW
253
254Certain Perl operations that are set-up within the scope of a
d6ded950 255C<use locale> retain that effect even outside the scope.
b960a36e
KW
256These include:
257
258=over 4
259
260=item *
261
262The output format of a L<write()|perlfunc/write> is determined by an
263earlier format declaration (L<perlfunc/format>), so whether or not the
264output is affected by locale is determined by if the C<format()> is
d6ded950 265within the scope of a C<use locale>, not whether the C<write()>
b960a36e
KW
266is.
267
268=item *
269
270Regular expression patterns can be compiled using
33be4c61 271L<qrE<sol>E<sol>|perlop/qrE<sol>STRINGE<sol>msixpodualn> with actual
b960a36e
KW
272matching deferred to later. Again, it is whether or not the compilation
273was done within the scope of C<use locale> that determines the match
274behavior, not if the matches are done within such a scope or not.
275
276=back
277
ebc3223b
KW
278=for comment
279The nbsp below makes this look better (though not great)
280
d6ded950 281
ebc3223b
KW
282E<160>
283
d6ded950 284=item B<Under C<"use locale";>>
66cbab2c
KW
285
286=over 4
287
288=item *
289
d6ded950 290All the above operations
b960a36e
KW
291
292=item *
293
294B<Format declarations> (L<perlfunc/format>) and hence any subsequent
295C<write()>s use C<LC_NUMERIC>.
66cbab2c
KW
296
297=item *
298
b960a36e
KW
299B<stringification and output> use C<LC_NUMERIC>.
300These include the results of
301C<print()>,
302C<printf()>,
303C<say()>,
304and
305C<sprintf()>.
66cbab2c 306
b960a36e
KW
307=item *
308
309B<The comparison operators> (C<lt>, C<le>, C<cmp>, C<ge>, and C<gt>) use
39332f68 310C<LC_COLLATE>. C<sort()> is also affected if used without an
5a964f20 311explicit comparison function, because it uses C<cmp> by default.
14280422 312
5a964f20 313B<Note:> C<eq> and C<ne> are unaffected by locale: they always
de108802 314perform a char-by-char comparison of their scalar operands. What's
14280422
DD
315more, if C<cmp> finds that its operands are equal according to the
316collation sequence specified by the current locale, it goes on to
de108802
RGS
317perform a char-by-char comparison, and only returns I<0> (equal) if the
318operands are char-for-char identical. If you really want to know whether
5a964f20 319two strings--which C<eq> and C<cmp> may consider different--are equal
14280422 320as far as collation in the locale is concerned, see the discussion in
cb88b78e 321L<Category C<LC_COLLATE>: Collation>.
5f05dabc
PP
322
323=item *
324
39332f68
KW
325B<Regular expressions and case-modification functions> (C<uc()>, C<lc()>,
326C<ucfirst()>, and C<lcfirst()>) use C<LC_CTYPE>
5f05dabc 327
2c6ee1a7
KW
328=item *
329
dbf3c4d7
KW
330B<The variables L<$!|perlvar/$ERRNO>> (and its synonyms C<$ERRNO> and
331C<$OS_ERROR>) B<and L<$^E|perlvar/$EXTENDED_OS_ERROR>> (and its synonym
2c6ee1a7
KW
332C<$EXTENDED_OS_ERROR>) when used as strings use C<LC_MESSAGES>.
333
5f05dabc
PP
334=back
335
66cbab2c 336=back
5f05dabc 337
5a964f20 338The default behavior is restored with the S<C<no locale>> pragma, or
ef3087ec 339upon reaching the end of the block enclosing C<use locale>.
3cd61afa 340Note that C<use locale> calls may be
66cbab2c
KW
341nested, and that what is in effect within an inner scope will revert to
342the outer scope's rules at the end of the inner scope.
5f05dabc 343
5a964f20 344The string result of any operation that uses locale
14280422 345information is tainted, as it is possible for a locale to be
5a0de581 346untrustworthy. See L</"SECURITY">.
5f05dabc 347
d6ded950
KW
348Starting in Perl v5.16 in a very limited way, and more generally in
349v5.22, you can restrict which category or categories are enabled by this
350particular instance of the pragma by adding parameters to it. For
351example,
352
353 use locale qw(:ctype :numeric);
354
355enables locale awareness within its scope of only those operations
356(listed above) that are affected by C<LC_CTYPE> and C<LC_NUMERIC>.
357
358The possible categories are: C<:collate>, C<:ctype>, C<:messages>,
359C<:monetary>, C<:numeric>, C<:time>, and the pseudo category
360C<:characters> (described below).
361
362Thus you can say
363
364 use locale ':messages';
365
366and only L<$!|perlvar/$ERRNO> and L<$^E|perlvar/$EXTENDED_OS_ERROR>
367will be locale aware. Everything else is unaffected.
368
369Since Perl doesn't currently do anything with the C<LC_MONETARY>
370category, specifying C<:monetary> does effectively nothing. Some
22803c6a 371systems have other categories, such as C<LC_PAPER>, but Perl
d6ded950
KW
372also doesn't know anything about them, and there is no way to specify
373them in this pragma's arguments.
374
375You can also easily say to use all categories but one, by either, for
376example,
377
378 use locale ':!ctype';
379 use locale ':not_ctype';
380
381both of which mean to enable locale awarness of all categories but
382C<LC_CTYPE>. Only one category argument may be specified in a
383S<C<use locale>> if it is of the negated form.
384
385Prior to v5.22 only one form of the pragma with arguments is available:
386
387 use locale ':not_characters';
388
389(and you have to say C<not_>; you can't use the bang C<!> form). This
390pseudo category is a shorthand for specifying both C<:collate> and
391C<:ctype>. Hence, in the negated form, it is nearly the same thing as
392saying
393
394 use locale qw(:messages :monetary :numeric :time);
395
396We use the term "nearly", because C<:not_characters> also turns on
397S<C<use feature 'unicode_strings'>> within its scope. This form is
398less useful in v5.20 and later, and is described fully in
399L</Unicode and UTF-8>, but briefly, it tells Perl to not use the
400character portions of the locale definition, that is the C<LC_CTYPE> and
401C<LC_COLLATE> categories. Instead it will use the native character set
402(extended by Unicode). When using this parameter, you are responsible
403for getting the external character set translated into the
404native/Unicode one (which it already will be if it is one of the
405increasingly popular UTF-8 locales). There are convenient ways of doing
406this, as described in L</Unicode and UTF-8>.
407
5f05dabc
PP
408=head2 The setlocale function
409
fc82b82e
KW
410WARNING! Do NOT use this function in a L<thread|threads>. The locale
411will change in all other threads at the same time, and should your
412thread get paused by the operating system, and another started, that
413thread will not have the locale it is expecting. On some platforms,
414there can be a race leading to segfaults if two threads call this
415function nearly simultaneously.
416
14280422 417You can switch locales as often as you wish at run time with the
39332f68 418C<POSIX::setlocale()> function:
5f05dabc 419
5f05dabc
PP
420 # Import locale-handling tool set from POSIX module.
421 # This example uses: setlocale -- the function call
422 # LC_CTYPE -- explained below
2619d284
KW
423 # (Showing the testing for success/failure of operations is
424 # omitted in these examples to avoid distracting from the main
ebc3223b 425 # point)
6ea81ccf 426
5f05dabc 427 use POSIX qw(locale_h);
dfcc8045
KW
428 use locale;
429 my $old_locale;
5f05dabc 430
14280422 431 # query and save the old locale
5f05dabc
PP
432 $old_locale = setlocale(LC_CTYPE);
433
434 setlocale(LC_CTYPE, "fr_CA.ISO8859-1");
435 # LC_CTYPE now in locale "French, Canada, codeset ISO 8859-1"
436
437 setlocale(LC_CTYPE, "");
65ebb059
KW
438 # LC_CTYPE now reset to the default defined by the
439 # LC_ALL/LC_CTYPE/LANG environment variables, or to the system
440 # default. See below for documentation.
5f05dabc
PP
441
442 # restore the old locale
443 setlocale(LC_CTYPE, $old_locale);
444
39332f68 445The first argument of C<setlocale()> gives the B<category>, the second the
14280422
DD
446B<locale>. The category tells in what aspect of data processing you
447want to apply locale-specific rules. Category names are discussed in
66cbab2c 448L</LOCALE CATEGORIES> and L</"ENVIRONMENT">. The locale is the name of a
14280422
DD
449collection of customization information corresponding to a particular
450combination of language, country or territory, and codeset. Read on for
451hints on the naming of locales: not all systems name locales as in the
452example.
453
39332f68 454If no second argument is provided and the category is something other
cb88b78e 455than C<LC_ALL>, the function returns a string naming the current locale
502a173a 456for the category. You can use this value as the second argument in a
f170b852
KW
457subsequent call to C<setlocale()>, B<but> on some platforms the string
458is opaque, not something that most people would be able to decipher as
459to what locale it means.
502a173a 460
cb88b78e 461If no second argument is provided and the category is C<LC_ALL>, the
502a173a 462result is implementation-dependent. It may be a string of
c052850d 463concatenated locale names (separator also implementation-dependent)
39332f68 464or a single locale name. Please consult your L<setlocale(3)> man page for
502a173a
JH
465details.
466
467If a second argument is given and it corresponds to a valid locale,
468the locale for the category is set to that value, and the function
469returns the now-current locale value. You can then use this in yet
39332f68 470another call to C<setlocale()>. (In some implementations, the return
502a173a
JH
471value may sometimes differ from the value you gave as the second
472argument--think of it as an alias for the value you gave.)
5f05dabc
PP
473
474As the example shows, if the second argument is an empty string, the
475category's locale is returned to the default specified by the
476corresponding environment variables. Generally, this results in a
5a964f20 477return to the default that was in force when Perl started up: changes
54310121 478to the environment made by the application after startup may or may not
5a964f20 479be noticed, depending on your system's C library.
5f05dabc 480
d6ded950
KW
481Note that when a form of C<use locale> that doesn't include all
482categories is specified, Perl ignores the excluded categories.
66cbab2c 483
f170b852 484If C<set_locale()> fails for some reason (for example, an attempt to set
dfcc8045
KW
485to a locale unknown to the system), the locale for the category is not
486changed, and the function returns C<undef>.
487
2619d284 488
39332f68 489For further information about the categories, consult L<setlocale(3)>.
3e6e419a
JH
490
491=head2 Finding locales
492
39332f68 493For locales available in your system, consult also L<setlocale(3)> to
5a964f20
TC
494see whether it leads to the list of available locales (search for the
495I<SEE ALSO> section). If that fails, try the following command lines:
5f05dabc
PP
496
497 locale -a
498
499 nlsinfo
500
501 ls /usr/lib/nls/loc
502
503 ls /usr/lib/locale
504
505 ls /usr/lib/nls
506
b478f28d
JH
507 ls /usr/share/locale
508
5f05dabc
PP
509and see whether they list something resembling these
510
2bdf8add 511 en_US.ISO8859-1 de_DE.ISO8859-1 ru_RU.ISO8859-5
502a173a 512 en_US.iso88591 de_DE.iso88591 ru_RU.iso88595
2bdf8add 513 en_US de_DE ru_RU
14280422 514 en de ru
2bdf8add
JH
515 english german russian
516 english.iso88591 german.iso88591 russian.iso88595
502a173a 517 english.roman8 russian.koi8r
5f05dabc 518
39332f68 519Sadly, even though the calling interface for C<setlocale()> has been
528d65ad 520standardized, names of locales and the directories where the
5a964f20 521configuration resides have not been. The basic form of the name is
528d65ad
JH
522I<language_territory>B<.>I<codeset>, but the latter parts after
523I<language> are not always present. The I<language> and I<country>
524are usually from the standards B<ISO 3166> and B<ISO 639>, the
525two-letter abbreviations for the countries and the languages of the
526world, respectively. The I<codeset> part often mentions some B<ISO
5278859> character set, the Latin codesets. For example, C<ISO 8859-1>
528is the so-called "Western European codeset" that can be used to encode
529most Western European languages adequately. Again, there are several
530ways to write even the name of that one standard. Lamentably.
5f05dabc 531
14280422
DD
532Two special locales are worth particular mention: "C" and "POSIX".
533Currently these are effectively the same locale: the difference is
5a964f20
TC
534mainly that the first one is defined by the C standard, the second by
535the POSIX standard. They define the B<default locale> in which
14280422 536every program starts in the absence of locale information in its
5a964f20 537environment. (The I<default> default locale, if you will.) Its language
39332f68
KW
538is (American) English and its character codeset ASCII or, rarely, a
539superset thereof (such as the "DEC Multinational Character Set
540(DEC-MCS)"). B<Warning>. The C locale delivered by some vendors
541may not actually exactly match what the C standard calls for. So
542beware.
5f05dabc 543
14280422
DD
544B<NOTE>: Not all systems have the "POSIX" locale (not all systems are
545POSIX-conformant), so use "C" when you need explicitly to specify this
546default locale.
5f05dabc 547
3e6e419a
JH
548=head2 LOCALE PROBLEMS
549
5a964f20 550You may encounter the following warning message at Perl startup:
3e6e419a
JH
551
552 perl: warning: Setting locale failed.
553 perl: warning: Please check that your locale settings:
554 LC_ALL = "En_US",
555 LANG = (unset)
556 are supported and installed on your system.
557 perl: warning: Falling back to the standard locale ("C").
558
cb88b78e 559This means that your locale settings had C<LC_ALL> set to "En_US" and
5a964f20
TC
560LANG exists but has no value. Perl tried to believe you but could not.
561Instead, Perl gave up and fell back to the "C" locale, the default locale
65ebb059
KW
562that is supposed to work no matter what. (On Windows, it first tries
563falling back to the system default locale.) This usually means your
564locale settings were wrong, they mention locales your system has never
565heard of, or the locale installation in your system has problems (for
566example, some system files are broken or missing). There are quick and
567temporary fixes to these problems, as well as more thorough and lasting
568fixes.
3e6e419a 569
83fb1bf0
KW
570=head2 Testing for broken locales
571
572If you are building Perl from source, the Perl test suite file
573F<lib/locale.t> can be used to test the locales on your system.
574Setting the environment variable C<PERL_DEBUG_FULL_TEST> to 1
575will cause it to output detailed results. For example, on Linux, you
576could say
577
1d2ab946 578 PERL_DEBUG_FULL_TEST=1 ./perl -T -Ilib lib/locale.t > locale.log 2>&1
83fb1bf0
KW
579
580Besides many other tests, it will test every locale it finds on your
581system to see if they conform to the POSIX standard. If any have
582errors, it will include a summary near the end of the output of which
583locales passed all its tests, and which failed, and why.
584
3e6e419a
JH
585=head2 Temporarily fixing locale problems
586
5a964f20 587The two quickest fixes are either to render Perl silent about any
3e6e419a
JH
588locale inconsistencies or to run Perl under the default locale "C".
589
590Perl's moaning about locale problems can be silenced by setting the
22ff3130 591environment variable C<PERL_BADLANG> to "0" or "".
900bd440
JH
592This method really just sweeps the problem under the carpet: you tell
593Perl to shut up even when Perl sees that something is wrong. Do not
594be surprised if later something locale-dependent misbehaves.
3e6e419a
JH
595
596Perl can be run under the "C" locale by setting the environment
cb88b78e
KW
597variable C<LC_ALL> to "C". This method is perhaps a bit more civilized
598than the C<PERL_BADLANG> approach, but setting C<LC_ALL> (or
5a964f20
TC
599other locale variables) may affect other programs as well, not just
600Perl. In particular, external programs run from within Perl will see
3e6e419a 601these changes. If you make the new settings permanent (read on), all
5a0de581
LM
602programs you run see the changes. See L</"ENVIRONMENT"> for
603the full list of relevant environment variables and L</"USING LOCALES">
e05ffc7d 604for their effects in Perl. Effects in other programs are
cb88b78e 605easily deducible. For example, the variable C<LC_COLLATE> may well affect
b432a672 606your B<sort> program (or whatever the program that arranges "records"
3e6e419a
JH
607alphabetically in your system is called).
608
5a964f20
TC
609You can test out changing these variables temporarily, and if the
610new settings seem to help, put those settings into your shell startup
663d437a 611files. Consult your local documentation for the exact details. For
5a964f20 612Bourne-like shells (B<sh>, B<ksh>, B<bash>, B<zsh>):
3e6e419a
JH
613
614 LC_ALL=en_US.ISO8859-1
615 export LC_ALL
616
5a964f20
TC
617This assumes that we saw the locale "en_US.ISO8859-1" using the commands
618discussed above. We decided to try that instead of the above faulty
619locale "En_US"--and in Cshish shells (B<csh>, B<tcsh>)
3e6e419a
JH
620
621 setenv LC_ALL en_US.ISO8859-1
c47ff5f1 622
663d437a 623or if you have the "env" application you can do (in any shell)
c406981e
JH
624
625 env LC_ALL=en_US.ISO8859-1 perl ...
626
5a964f20 627If you do not know what shell you have, consult your local
3e6e419a
JH
628helpdesk or the equivalent.
629
630=head2 Permanently fixing locale problems
631
5a964f20
TC
632The slower but superior fixes are when you may be able to yourself
633fix the misconfiguration of your own environment variables. The
3e6e419a
JH
634mis(sing)configuration of the whole system's locales usually requires
635the help of your friendly system administrator.
636
5a0de581 637First, see earlier in this document about L</Finding locales>. That tells
5a964f20
TC
638how to find which locales are really supported--and more importantly,
639installed--on your system. In our example error message, environment
640variables affecting the locale are listed in the order of decreasing
641importance (and unset variables do not matter). Therefore, having
642LC_ALL set to "En_US" must have been the bad choice, as shown by the
643error message. First try fixing locale settings listed first.
3e6e419a 644
5a964f20
TC
645Second, if using the listed commands you see something B<exactly>
646(prefix matches do not count and case usually counts) like "En_US"
647without the quotes, then you should be okay because you are using a
648locale name that should be installed and available in your system.
5a0de581 649In this case, see L</Permanently fixing your system's locale configuration>.
3e6e419a 650
4a4eefd0 651=head2 Permanently fixing your system's locale configuration
3e6e419a 652
5a964f20 653This is when you see something like:
3e6e419a
JH
654
655 perl: warning: Please check that your locale settings:
656 LC_ALL = "En_US",
657 LANG = (unset)
658 are supported and installed on your system.
659
660but then cannot see that "En_US" listed by the above-mentioned
5a964f20
TC
661commands. You may see things like "en_US.ISO8859-1", but that isn't
662the same. In this case, try running under a locale
663that you can list and which somehow matches what you tried. The
3e6e419a 664rules for matching locale names are a bit vague because
e05ffc7d 665standardization is weak in this area. See again the
5a0de581 666L</Finding locales> about general rules.
3e6e419a 667
b687b08b 668=head2 Fixing system locale configuration
3e6e419a 669
5a964f20
TC
670Contact a system administrator (preferably your own) and report the exact
671error message you get, and ask them to read this same documentation you
672are now reading. They should be able to check whether there is something
5a0de581 673wrong with the locale configuration of the system. The L</Finding locales>
5a964f20
TC
674section is unfortunately a bit vague about the exact commands and places
675because these things are not that standardized.
3e6e419a 676
5f05dabc
PP
677=head2 The localeconv function
678
39332f68 679The C<POSIX::localeconv()> function allows you to get particulars of the
14280422 680locale-dependent numeric formatting information specified by the current
a835cd47
KW
681underlying C<LC_NUMERIC> and C<LC_MONETARY> locales (regardless of
682whether called from within the scope of C<S<use locale>> or not). (If
683you just want the name of
39332f68 684the current locale for a particular category, use C<POSIX::setlocale()>
5a0de581 685with a single parameter--see L</The setlocale function>.)
5f05dabc
PP
686
687 use POSIX qw(locale_h);
5f05dabc
PP
688
689 # Get a reference to a hash of locale-dependent info
690 $locale_values = localeconv();
691
692 # Output sorted list of the values
693 for (sort keys %$locale_values) {
14280422 694 printf "%-20s = %s\n", $_, $locale_values->{$_}
5f05dabc
PP
695 }
696
39332f68 697C<localeconv()> takes no arguments, and returns B<a reference to> a hash.
5a964f20 698The keys of this hash are variable names for formatting, such as
502a173a 699C<decimal_point> and C<thousands_sep>. The values are the
cea6626f 700corresponding, er, values. See L<POSIX/localeconv> for a longer
502a173a
JH
701example listing the categories an implementation might be expected to
702provide; some provide more and others fewer. You don't need an
39332f68 703explicit C<use locale>, because C<localeconv()> always observes the
502a173a 704current locale.
5f05dabc 705
5a964f20
TC
706Here's a simple-minded example program that rewrites its command-line
707parameters as integers correctly formatted in the current locale:
5f05dabc 708
ef3087ec
KW
709 use POSIX qw(locale_h);
710
711 # Get some of locale's numeric formatting parameters
712 my ($thousands_sep, $grouping) =
713 @{localeconv()}{'thousands_sep', 'grouping'};
714
715 # Apply defaults if values are missing
716 $thousands_sep = ',' unless $thousands_sep;
717
718 # grouping and mon_grouping are packed lists
719 # of small integers (characters) telling the
720 # grouping (thousand_seps and mon_thousand_seps
721 # being the group dividers) of numbers and
722 # monetary quantities. The integers' meanings:
723 # 255 means no more grouping, 0 means repeat
724 # the previous grouping, 1-254 means use that
725 # as the current grouping. Grouping goes from
726 # right to left (low to high digits). In the
727 # below we cheat slightly by never using anything
728 # else than the first grouping (whatever that is).
729 if ($grouping) {
730 @grouping = unpack("C*", $grouping);
731 } else {
732 @grouping = (3);
733 }
734
735 # Format command line params for current locale
736 for (@ARGV) {
737 $_ = int; # Chop non-integer part
738 1 while
739 s/(\d)(\d{$grouping[0]}($|$thousands_sep))/$1$thousands_sep$2/;
740 print "$_";
741 }
742 print "\n";
5f05dabc 743
03ceeedf
KW
744Note that if the platform doesn't have C<LC_NUMERIC> and/or
745C<LC_MONETARY> available or enabled, the corresponding elements of the
746hash will be missing.
747
74c76037 748=head2 I18N::Langinfo
4bbcc6e8
JH
749
750Another interface for querying locale-dependent information is the
39332f68 751C<I18N::Langinfo::langinfo()> function, available at least in Unix-like
4bbcc6e8
JH
752systems and VMS.
753
39332f68
KW
754The following example will import the C<langinfo()> function itself and
755three constants to be used as arguments to C<langinfo()>: a constant for
74c76037
JH
756the abbreviated first day of the week (the numbering starts from
757Sunday = 1) and two more constants for the affirmative and negative
758answers for a yes/no question in the current locale.
4bbcc6e8 759
74c76037 760 use I18N::Langinfo qw(langinfo ABDAY_1 YESSTR NOSTR);
4bbcc6e8 761
ef3087ec
KW
762 my ($abday_1, $yesstr, $nostr)
763 = map { langinfo } qw(ABDAY_1 YESSTR NOSTR);
4bbcc6e8 764
74c76037 765 print "$abday_1? [$yesstr/$nostr] ";
4bbcc6e8 766
74c76037
JH
767In other words, in the "C" (or English) locale the above will probably
768print something like:
769
e05ffc7d 770 Sun? [yes/no]
4bbcc6e8
JH
771
772See L<I18N::Langinfo> for more information.
773
5f05dabc
PP
774=head1 LOCALE CATEGORIES
775
5a964f20
TC
776The following subsections describe basic locale categories. Beyond these,
777some combination categories allow manipulation of more than one
5a0de581 778basic category at a time. See L</"ENVIRONMENT"> for a discussion of these.
5f05dabc 779
ff52fcf1 780=head2 Category C<LC_COLLATE>: Collation: Text Comparisons and Sorting
5f05dabc 781
d6ded950
KW
782In the scope of a S<C<use locale>> form that includes collation, Perl
783looks to the C<LC_COLLATE>
5a964f20 784environment variable to determine the application's notions on collation
b4ffc3db
TC
785(ordering) of characters. For example, "b" follows "a" in Latin
786alphabets, but where do "E<aacute>" and "E<aring>" belong? And while
f87fa335 787"color" follows "chocolate" in English, what about in traditional Spanish?
5f05dabc 788
60f0fa02 789The following collations all make sense and you may meet any of them
dbf3c4d7 790if you C<"use locale">.
60f0fa02
JH
791
792 A B C D E a b c d e
35316ca3 793 A a B b C c D d E e
60f0fa02
JH
794 a A b B c C d D e E
795 a b c d e A B C D E
796
f1cbbd6e 797Here is a code snippet to tell what "word"
5a964f20 798characters are in the current locale, in that locale's order:
5f05dabc
PP
799
800 use locale;
35316ca3 801 print +(sort grep /\w/, map { chr } 0..255), "\n";
5f05dabc 802
14280422
DD
803Compare this with the characters that you see and their order if you
804state explicitly that the locale should be ignored:
5f05dabc
PP
805
806 no locale;
35316ca3 807 print +(sort grep /\w/, map { chr } 0..255), "\n";
5f05dabc
PP
808
809This machine-native collation (which is what you get unless S<C<use
810locale>> has appeared earlier in the same block) must be used for
811sorting raw binary data, whereas the locale-dependent collation of the
b0c42ed9 812first example is useful for natural text.
5f05dabc 813
5a0de581 814As noted in L</USING LOCALES>, C<cmp> compares according to the current
14280422 815collation locale when C<use locale> is in effect, but falls back to a
de108802 816char-by-char comparison for strings that the locale says are equal. You
39332f68 817can use C<POSIX::strcoll()> if you don't want this fall-back:
14280422
DD
818
819 use POSIX qw(strcoll);
820 $equal_in_locale =
821 !strcoll("space and case ignored", "SpaceAndCaseIgnored");
822
39332f68 823C<$equal_in_locale> will be true if the collation locale specifies a
5a964f20 824dictionary-like ordering that ignores space characters completely and
9e3a2af8 825which folds case.
14280422 826
a4a439fb
KW
827Perl uses the platform's C library collation functions C<strcoll()> and
828C<strxfrm()>. That means you get whatever they give. On some
829platforms, these functions work well on UTF-8 locales, giving
830a reasonable default collation for the code points that are important in
831that locale. (And if they aren't working well, the problem may only be
832that the locale definition is deficient, so can be fixed by using a
833better definition file. Unicode's definitions (see L</Freely available
834locale definitions>) provide reasonable UTF-8 locale collation
835definitions.) Starting in Perl v5.26, Perl's use of these functions has
836been made more seamless. This may be sufficient for your needs. For
837more control, and to make sure strings containing any code point (not
838just the ones important in the locale) collate properly, the
839L<Unicode::Collate> module is suggested.
840
841In non-UTF-8 locales (hence single byte), code points above 0xFF are
842technically invalid. But if present, again starting in v5.26, they will
843collate to the same position as the highest valid code point does. This
844generally gives good results, but the collation order may be skewed if
845the valid code point gets special treatment when it forms particular
846sequences with other characters as defined by the locale.
847When two strings collate identically, the code point order is used as a
848tie breaker.
849
850If Perl detects that there are problems with the locale collation order,
851it reverts to using non-locale collation rules for that locale.
31f05a37 852
5a964f20 853If you have a single string that you want to check for "equality in
14280422 854locale" against several others, you might think you could gain a little
39332f68 855efficiency by using C<POSIX::strxfrm()> in conjunction with C<eq>:
14280422
DD
856
857 use POSIX qw(strxfrm);
858 $xfrm_string = strxfrm("Mixed-case string");
859 print "locale collation ignores spaces\n"
860 if $xfrm_string eq strxfrm("Mixed-casestring");
861 print "locale collation ignores hyphens\n"
862 if $xfrm_string eq strxfrm("Mixedcase string");
863 print "locale collation ignores case\n"
864 if $xfrm_string eq strxfrm("mixed-case string");
865
39332f68 866C<strxfrm()> takes a string and maps it into a transformed string for use
de108802 867in char-by-char comparisons against other transformed strings during
14280422 868collation. "Under the hood", locale-affected Perl comparison operators
39332f68
KW
869call C<strxfrm()> for both operands, then do a char-by-char
870comparison of the transformed strings. By calling C<strxfrm()> explicitly
14280422 871and using a non locale-affected comparison, the example attempts to save
5a964f20 872a couple of transformations. But in fact, it doesn't save anything: Perl
2ae324a7 873magic (see L<perlguts/Magic Variables>) creates the transformed version of a
5a964f20 874string the first time it's needed in a comparison, then keeps this version around
14280422 875in case it's needed again. An example rewritten the easy way with
e38874e2 876C<cmp> runs just about as fast. It also copes with null characters
39332f68 877embedded in strings; if you call C<strxfrm()> directly, it treats the first
0c880285 878null it finds as a terminator. Don't expect the transformed strings
5a964f20 879it produces to be portable across systems--or even from one revision
39332f68 880of your operating system to the next. In short, don't call C<strxfrm()>
e38874e2 881directly: let Perl do it for you.
14280422 882
5a964f20 883Note: C<use locale> isn't shown in some of these examples because it isn't
dfcc8045
KW
884needed: C<strcoll()> and C<strxfrm()> are POSIX functions
885which use the standard system-supplied C<libc> functions that
886always obey the current C<LC_COLLATE> locale.
5f05dabc 887
cb88b78e 888=head2 Category C<LC_CTYPE>: Character Types
5f05dabc 889
d6ded950
KW
890In the scope of a S<C<use locale>> form that includes C<LC_CTYPE>, Perl
891obeys the C<LC_CTYPE> locale
14280422 892setting. This controls the application's notion of which characters are
ebc3223b
KW
893alphabetic, numeric, punctuation, I<etc>. This affects Perl's C<\w>
894regular expression metanotation,
f1cbbd6e 895which stands for alphanumeric characters--that is, alphabetic,
ebc3223b
KW
896numeric, and the platform's native underscore.
897(Consult L<perlre> for more information about
14280422 898regular expressions.) Thanks to C<LC_CTYPE>, depending on your locale
b4ffc3db
TC
899setting, characters like "E<aelig>", "E<eth>", "E<szlig>", and
900"E<oslash>" may be understood as C<\w> characters.
ebc3223b
KW
901It also affects things like C<\s>, C<\D>, and the POSIX character
902classes, like C<[[:graph:]]>. (See L<perlrecharclass> for more
903information on all these.)
5f05dabc 904
2c268ad5 905The C<LC_CTYPE> locale also provides the map used in transliterating
68dc0745 906characters between lower and uppercase. This affects the case-mapping
663d437a
KW
907functions--C<fc()>, C<lc()>, C<lcfirst()>, C<uc()>, and C<ucfirst()>;
908case-mapping
b9cc4f69 909interpolation with C<\F>, C<\l>, C<\L>, C<\u>, or C<\U> in double-quoted
26a9b33a 910strings and C<s///> substitutions; and case-insensitive regular expression
e38874e2
DD
911pattern matching using the C<i> modifier.
912
31f05a37
KW
913Starting in v5.20, Perl supports UTF-8 locales for C<LC_CTYPE>, but
914otherwise Perl only supports single-byte locales, such as the ISO 8859
915series. This means that wide character locales, for example for Asian
b5eea289
KW
916languages, are not well-supported. Use of these locales may cause core
917dumps. If the platform has the capability for Perl to detect such a
918locale, starting in Perl v5.22, L<Perl will warn, default
919enabled|warnings/Category Hierarchy>, using the C<locale> warning
920category, whenever such a locale is switched into. The UTF-8 locale
921support is actually a
31f05a37 922superset of POSIX locales, because it is really full Unicode behavior
13af4fd9
KW
923as if no C<LC_CTYPE> locale were in effect at all (except for tainting;
924see L</SECURITY>). POSIX locales, even UTF-8 ones,
31f05a37
KW
925are lacking certain concepts in Unicode, such as the idea that changing
926the case of a character could expand to be more than one character.
927Perl in a UTF-8 locale, will give you that expansion. Prior to v5.20,
928Perl treated a UTF-8 locale on some platforms like an ISO 8859-1 one,
929with some restrictions, and on other platforms more like the "C" locale.
930For releases v5.16 and v5.18, C<S<use locale 'not_characters>> could be
931used as a workaround for this (see L</Unicode and UTF-8>).
932
5d63e270 933Note that there are quite a few things that are unaffected by the
663d437a
KW
934current locale. Any literal character is the native character for the
935given platform. Hence 'A' means the character at code point 65 on ASCII
936platforms, and 193 on EBCDIC. That may or may not be an 'A' in the
937current locale, if that locale even has an 'A'.
938Similarly, all the escape sequences for particular characters,
5d63e270
KW
939C<\n> for example, always mean the platform's native one. This means,
940for example, that C<\N> in regular expressions (every character
1d2ab946 941but new-line) works on the platform character set.
5d63e270 942
8c6180a9
KW
943Starting in v5.22, Perl will by default warn when switching into a
944locale that redefines any ASCII printable character (plus C<\t> and
84035de0
KW
945C<\n>) into a different class than expected. This is likely to
946happen on modern locales only on EBCDIC platforms, where, for example,
947a CCSID 0037 locale on a CCSID 1047 machine moves C<"[">, but it can
948happen on ASCII platforms with the ISO 646 and other
8c6180a9
KW
9497-bit locales that are essentially obsolete. Things may still work,
950depending on what features of Perl are used by the program. For
951example, in the example from above where C<"|"> becomes a C<\w>, and
952there are no regular expressions where this matters, the program may
953still work properly. The warning lists all the characters that
954it can determine could be adversely affected.
955
14280422
DD
956B<Note:> A broken or malicious C<LC_CTYPE> locale definition may result
957in clearly ineligible characters being considered to be alphanumeric by
e199995e 958your application. For strict matching of (mundane) ASCII letters and
5a964f20 959digits--for example, in command strings--locale-aware applications
5a0de581 960should use C<\w> with the C</a> regular expression modifier. See L</"SECURITY">.
5f05dabc 961
fee33030 962=head2 Category C<LC_NUMERIC>: Numeric Formatting
5f05dabc 963
d6ded950
KW
964After a proper C<POSIX::setlocale()> call, and within the scope of
965of a C<use locale> form that includes numerics, Perl obeys the
966C<LC_NUMERIC> locale information, which controls an application's idea
967of how numbers should be formatted for human readability.
b960a36e 968In most implementations the only effect is to
b4ffc3db 969change the character used for the decimal point--perhaps from "." to ",".
b960a36e 970The functions aren't aware of such niceties as thousands separation and
5a0de581 971so on. (See L</The localeconv function> if you care about these things.)
5a964f20 972
b960a36e
KW
973 use POSIX qw(strtod setlocale LC_NUMERIC);
974 use locale;
5f05dabc 975
b960a36e 976 setlocale LC_NUMERIC, "";
14280422 977
b960a36e 978 $n = 5/2; # Assign numeric 2.5 to $n
5f05dabc 979
b960a36e 980 $a = " $n"; # Locale-dependent conversion to string
5f05dabc 981
b960a36e 982 print "half five is $n\n"; # Locale-dependent output
5f05dabc 983
b960a36e 984 printf "half five is %g\n", $n; # Locale-dependent output
5f05dabc 985
b960a36e
KW
986 print "DECIMAL POINT IS COMMA\n"
987 if $n == (strtod("2,5"))[0]; # Locale-dependent conversion
5f05dabc 988
4bbcc6e8
JH
989See also L<I18N::Langinfo> and C<RADIXCHAR>.
990
cb88b78e 991=head2 Category C<LC_MONETARY>: Formatting of monetary amounts
5f05dabc 992
e199995e 993The C standard defines the C<LC_MONETARY> category, but not a function
5a964f20 994that is affected by its contents. (Those with experience of standards
b0c42ed9 995committees will recognize that the working group decided to punt on the
fa9b773e
KW
996issue.) Consequently, Perl essentially takes no notice of it. If you
997really want to use C<LC_MONETARY>, you can query its contents--see
5a0de581 998L</The localeconv function>--and use the information that it returns in your
e05ffc7d
KW
999application's own formatting of currency amounts. However, you may well
1000find that the information, voluminous and complex though it may be, still
1001does not quite meet your requirements: currency formatting is a hard nut
13a2d996 1002to crack.
5f05dabc 1003
4bbcc6e8
JH
1004See also L<I18N::Langinfo> and C<CRNCYSTR>.
1005
ff52fcf1 1006=head2 Category C<LC_TIME>: Respresentation of time
5f05dabc 1007
39332f68 1008Output produced by C<POSIX::strftime()>, which builds a formatted
5f05dabc
PP
1009human-readable date/time string, is affected by the current C<LC_TIME>
1010locale. Thus, in a French locale, the output produced by the C<%B>
1011format element (full month name) for the first month of the year would
5a964f20 1012be "janvier". Here's how to get a list of long month names in the
5f05dabc
PP
1013current locale:
1014
1015 use POSIX qw(strftime);
14280422
DD
1016 for (0..11) {
1017 $long_month_name[$_] =
1018 strftime("%B", 0, 0, 0, 1, $_, 96);
5f05dabc
PP
1019 }
1020
2619d284
KW
1021Note: C<use locale> isn't needed in this example: C<strftime()> is a POSIX
1022function which uses the standard system-supplied C<libc> function that
1023always obeys the current C<LC_TIME> locale.
5f05dabc 1024
4bbcc6e8 1025See also L<I18N::Langinfo> and C<ABDAY_1>..C<ABDAY_7>, C<DAY_1>..C<DAY_7>,
2a2bf5f4 1026C<ABMON_1>..C<ABMON_12>, and C<ABMON_1>..C<ABMON_12>.
4bbcc6e8 1027
5f05dabc
PP
1028=head2 Other categories
1029
2619d284
KW
1030The remaining locale categories are not currently used by Perl itself.
1031But again note that things Perl interacts with may use these, including
1032extensions outside the standard Perl distribution, and by the
98a6f11e 1033operating system and its utilities. Note especially that the string
1034value of C<$!> and the error messages given by external utilities may
1035be changed by C<LC_MESSAGES>. If you want to have portable error
265f5c4a 1036codes, use C<%!>. See L<Errno>.
14280422
DD
1037
1038=head1 SECURITY
1039
5a964f20 1040Although the main discussion of Perl security issues can be found in
14280422
DD
1041L<perlsec>, a discussion of Perl's locale handling would be incomplete
1042if it did not draw your attention to locale-dependent security issues.
5a964f20
TC
1043Locales--particularly on systems that allow unprivileged users to
1044build their own locales--are untrustworthy. A malicious (or just plain
14280422
DD
1045broken) locale can make a locale-aware application give unexpected
1046results. Here are a few possibilities:
1047
1048=over 4
1049
1050=item *
1051
1052Regular expression checks for safe file names or mail addresses using
5a964f20 1053C<\w> may be spoofed by an C<LC_CTYPE> locale that claims that
dbf3c4d7 1054characters such as C<"E<gt>"> and C<"|"> are alphanumeric.
14280422
DD
1055
1056=item *
1057
e38874e2 1058String interpolation with case-mapping, as in, say, C<$dest =
cb88b78e 1059"C:\U$name.$ext">, may produce dangerous results if a bogus C<LC_CTYPE>
e38874e2
DD
1060case-mapping table is in effect.
1061
1062=item *
1063
14280422
DD
1064A sneaky C<LC_COLLATE> locale could result in the names of students with
1065"D" grades appearing ahead of those with "A"s.
1066
1067=item *
1068
5a964f20 1069An application that takes the trouble to use information in
14280422 1070C<LC_MONETARY> may format debits as if they were credits and vice versa
5a964f20 1071if that locale has been subverted. Or it might make payments in US
14280422
DD
1072dollars instead of Hong Kong dollars.
1073
1074=item *
1075
39332f68 1076The date and day names in dates formatted by C<strftime()> could be
14280422 1077manipulated to advantage by a malicious user able to subvert the
5a964f20 1078C<LC_DATE> locale. ("Look--it says I wasn't in the building on
14280422
DD
1079Sunday.")
1080
1081=back
1082
1083Such dangers are not peculiar to the locale system: any aspect of an
5a964f20 1084application's environment which may be modified maliciously presents
14280422 1085similar challenges. Similarly, they are not specific to Perl: any
5a964f20 1086programming language that allows you to write programs that take
14280422
DD
1087account of their environment exposes you to these issues.
1088
5a964f20
TC
1089Perl cannot protect you from all possibilities shown in the
1090examples--there is no substitute for your own vigilance--but, when
14280422 1091C<use locale> is in effect, Perl uses the tainting mechanism (see
5a964f20 1092L<perlsec>) to mark string results that become locale-dependent, and
14280422 1093which may be untrustworthy in consequence. Here is a summary of the
5a964f20 1094tainting behavior of operators and functions that may be affected by
14280422
DD
1095the locale:
1096
1097=over 4
1098
551e1d92
RB
1099=item *
1100
1101B<Comparison operators> (C<lt>, C<le>, C<ge>, C<gt> and C<cmp>):
14280422
DD
1102
1103Scalar true/false (or less/equal/greater) result is never tainted.
1104
551e1d92
RB
1105=item *
1106
1d2ab946 1107B<Case-mapping interpolation> (with C<\l>, C<\L>, C<\u>, C<\U>, or C<\F>)
e38874e2 1108
d6ded950
KW
1109The result string containing interpolated material is tainted if
1110a C<use locale> form that includes C<LC_CTYPE> is in effect.
e38874e2 1111
551e1d92
RB
1112=item *
1113
1114B<Matching operator> (C<m//>):
14280422
DD
1115
1116Scalar true/false result never tainted.
1117
1d2ab946 1118All subpatterns, either delivered as a list-context result or as C<$1>
d6ded950
KW
1119I<etc>., are tainted if a C<use locale> form that includes
1120C<LC_CTYPE> is in effect, and the subpattern
63baef57
KW
1121regular expression contains a locale-dependent construct. These
1122constructs include C<\w> (to match an alphanumeric character), C<\W>
1123(non-alphanumeric character), C<\b> and C<\B> (word-boundary and
1124non-boundardy, which depend on what C<\w> and C<\W> match), C<\s>
1125(whitespace character), C<\S> (non whitespace character), C<\d> and
1126C<\D> (digits and non-digits), and the POSIX character classes, such as
1127C<[:alpha:]> (see L<perlrecharclass/POSIX Character Classes>).
1128
1129Tainting is also likely if the pattern is to be matched
1130case-insensitively (via C</i>). The exception is if all the code points
1131to be matched this way are above 255 and do not have folds under Unicode
1132rules to below 256. Tainting is not done for these because Perl
1133only uses Unicode rules for such code points, and those rules are the
1134same no matter what the current locale.
1135
1d2ab946
KW
1136The matched-pattern variables, C<$&>, C<$`> (pre-match), C<$'>
1137(post-match), and C<$+> (last match) also are tainted.
14280422 1138
551e1d92
RB
1139=item *
1140
1141B<Substitution operator> (C<s///>):
14280422 1142
e38874e2 1143Has the same behavior as the match operator. Also, the left
d6ded950
KW
1144operand of C<=~> becomes tainted when a C<use locale>
1145form that includes C<LC_CTYPE> is in effect, if modified as
66cbab2c 1146a result of a substitution based on a regular
1d2ab946
KW
1147expression match involving any of the things mentioned in the previous
1148item, or of case-mapping, such as C<\l>, C<\L>,C<\u>, C<\U>, or C<\F>.
14280422 1149
551e1d92
RB
1150=item *
1151
39332f68 1152B<Output formatting functions> (C<printf()> and C<write()>):
14280422 1153
3cf03d68
JH
1154Results are never tainted because otherwise even output from print,
1155for example C<print(1/7)>, should be tainted if C<use locale> is in
1156effect.
14280422 1157
551e1d92
RB
1158=item *
1159
39332f68 1160B<Case-mapping functions> (C<lc()>, C<lcfirst()>, C<uc()>, C<ucfirst()>):
14280422 1161
d6ded950
KW
1162Results are tainted if a C<use locale> form that includes C<LC_CTYPE> is
1163in effect.
14280422 1164
551e1d92
RB
1165=item *
1166
39332f68
KW
1167B<POSIX locale-dependent functions> (C<localeconv()>, C<strcoll()>,
1168C<strftime()>, C<strxfrm()>):
14280422
DD
1169
1170Results are never tainted.
1171
14280422
DD
1172=back
1173
1174Three examples illustrate locale-dependent tainting.
1175The first program, which ignores its locale, won't run: a value taken
54310121 1176directly from the command line may not be used to name an output file
14280422
DD
1177when taint checks are enabled.
1178
1179 #/usr/local/bin/perl -T
1180 # Run with taint checking
1181
54310121 1182 # Command line sanity check omitted...
14280422
DD
1183 $tainted_output_file = shift;
1184
1185 open(F, ">$tainted_output_file")
3183d96c 1186 or warn "Open of $tainted_output_file failed: $!\n";
14280422
DD
1187
1188The program can be made to run by "laundering" the tainted value through
5a964f20
TC
1189a regular expression: the second example--which still ignores locale
1190information--runs, creating the file named on its command line
14280422
DD
1191if it can.
1192
1193 #/usr/local/bin/perl -T
1194
1195 $tainted_output_file = shift;
1196 $tainted_output_file =~ m%[\w/]+%;
1197 $untainted_output_file = $&;
1198
1199 open(F, ">$untainted_output_file")
1200 or warn "Open of $untainted_output_file failed: $!\n";
1201
5a964f20 1202Compare this with a similar but locale-aware program:
14280422
DD
1203
1204 #/usr/local/bin/perl -T
1205
1206 $tainted_output_file = shift;
1207 use locale;
1208 $tainted_output_file =~ m%[\w/]+%;
1209 $localized_output_file = $&;
1210
1211 open(F, ">$localized_output_file")
1212 or warn "Open of $localized_output_file failed: $!\n";
1213
1d2ab946 1214This third program fails to run because C<$&> is tainted: it is the result
5a964f20 1215of a match involving C<\w> while C<use locale> is in effect.
5f05dabc
PP
1216
1217=head1 ENVIRONMENT
1218
1219=over 12
1220
ee1ec05f
KW
1221=item PERL_SKIP_LOCALE_INIT
1222
c5c88224
KW
1223This environment variable, available starting in Perl v5.20, if set
1224(to any value), tells Perl to not use the rest of the
ee1ec05f
KW
1225environment variables to initialize with. Instead, Perl uses whatever
1226the current locale settings are. This is particularly useful in
1227embedded environments, see
1228L<perlembed/Using embedded Perl with POSIX locales>.
1229
5f05dabc
PP
1230=item PERL_BADLANG
1231
14280422 1232A string that can suppress Perl's warning about failed locale settings
54310121 1233at startup. Failure can occur if the locale support in the operating
5a964f20 1234system is lacking (broken) in some way--or if you mistyped the name of
900bd440 1235a locale when you set up your environment. If this environment
22ff3130
HS
1236variable is absent, or has a value other than "0" or "", Perl will
1237complain about locale setting failures.
5f05dabc 1238
cb88b78e 1239B<NOTE>: C<PERL_BADLANG> only gives you a way to hide the warning message.
14280422
DD
1240The message tells about some problem in your system's locale support,
1241and you should investigate what the problem is.
5f05dabc
PP
1242
1243=back
1244
1245The following environment variables are not specific to Perl: They are
39332f68 1246part of the standardized (ISO C, XPG4, POSIX 1.c) C<setlocale()> method
b385bb4d
KW
1247for controlling an application's opinion on data. Windows is non-POSIX,
1248but Perl arranges for the following to work as described anyway.
65ebb059
KW
1249If the locale given by an environment variable is not valid, Perl tries
1250the next lower one in priority. If none are valid, on Windows, the
1251system default locale is then tried. If all else fails, the C<"C">
1252locale is used. If even that doesn't work, something is badly broken,
c5e9a8e7 1253but Perl tries to forge ahead with whatever the locale settings might
65ebb059 1254be.
5f05dabc
PP
1255
1256=over 12
1257
cb88b78e 1258=item C<LC_ALL>
5f05dabc 1259
5a964f20 1260C<LC_ALL> is the "override-all" locale environment variable. If
5f05dabc
PP
1261set, it overrides all the rest of the locale environment variables.
1262
cb88b78e 1263=item C<LANGUAGE>
528d65ad
JH
1264
1265B<NOTE>: C<LANGUAGE> is a GNU extension, it affects you only if you
1266are using the GNU libc. This is the case if you are using e.g. Linux.
e1020413 1267If you are using "commercial" Unixes you are most probably I<not>
22b6f60d
JH
1268using GNU libc and you can ignore C<LANGUAGE>.
1269
1270However, in the case you are using C<LANGUAGE>: it affects the
1271language of informational, warning, and error messages output by
1272commands (in other words, it's like C<LC_MESSAGES>) but it has higher
96090e4f 1273priority than C<LC_ALL>. Moreover, it's not a single value but
22b6f60d
JH
1274instead a "path" (":"-separated list) of I<languages> (not locales).
1275See the GNU C<gettext> library documentation for more information.
528d65ad 1276
3ee1a09c 1277=item C<LC_CTYPE>
5f05dabc
PP
1278
1279In the absence of C<LC_ALL>, C<LC_CTYPE> chooses the character type
1280locale. In the absence of both C<LC_ALL> and C<LC_CTYPE>, C<LANG>
1281chooses the character type locale.
1282
cb88b78e 1283=item C<LC_COLLATE>
5f05dabc 1284
14280422
DD
1285In the absence of C<LC_ALL>, C<LC_COLLATE> chooses the collation
1286(sorting) locale. In the absence of both C<LC_ALL> and C<LC_COLLATE>,
1287C<LANG> chooses the collation locale.
5f05dabc 1288
cb88b78e 1289=item C<LC_MONETARY>
5f05dabc 1290
14280422
DD
1291In the absence of C<LC_ALL>, C<LC_MONETARY> chooses the monetary
1292formatting locale. In the absence of both C<LC_ALL> and C<LC_MONETARY>,
1293C<LANG> chooses the monetary formatting locale.
5f05dabc 1294
cb88b78e 1295=item C<LC_NUMERIC>
5f05dabc
PP
1296
1297In the absence of C<LC_ALL>, C<LC_NUMERIC> chooses the numeric format
1298locale. In the absence of both C<LC_ALL> and C<LC_NUMERIC>, C<LANG>
1299chooses the numeric format.
1300
cb88b78e 1301=item C<LC_TIME>
5f05dabc 1302
14280422
DD
1303In the absence of C<LC_ALL>, C<LC_TIME> chooses the date and time
1304formatting locale. In the absence of both C<LC_ALL> and C<LC_TIME>,
1305C<LANG> chooses the date and time formatting locale.
5f05dabc 1306
cb88b78e 1307=item C<LANG>
5f05dabc 1308
14280422
DD
1309C<LANG> is the "catch-all" locale environment variable. If it is set, it
1310is used as the last resort after the overall C<LC_ALL> and the
3ee1a09c 1311category-specific C<LC_I<foo>>.
5f05dabc
PP
1312
1313=back
1314
7e4353e9
RGS
1315=head2 Examples
1316
cb88b78e 1317The C<LC_NUMERIC> controls the numeric output:
7e4353e9 1318
ef3087ec
KW
1319 use locale;
1320 use POSIX qw(locale_h); # Imports setlocale() and the LC_ constants.
1321 setlocale(LC_NUMERIC, "fr_FR") or die "Pardon";
1322 printf "%g\n", 1.23; # If the "fr_FR" succeeded, probably shows 1,23.
7e4353e9 1323
39332f68 1324and also how strings are parsed by C<POSIX::strtod()> as numbers:
7e4353e9 1325
ef3087ec
KW
1326 use locale;
1327 use POSIX qw(locale_h strtod);
1328 setlocale(LC_NUMERIC, "de_DE") or die "Entschuldigung";
1329 my $x = strtod("2,34") + 5;
1330 print $x, "\n"; # Probably shows 7,34.
7e4353e9 1331
5f05dabc
PP
1332=head1 NOTES
1333
b960a36e
KW
1334=head2 String C<eval> and C<LC_NUMERIC>
1335
1336A string L<eval|perlfunc/eval EXPR> parses its expression as standard
1337Perl. It is therefore expecting the decimal point to be a dot. If
1338C<LC_NUMERIC> is set to have this be a comma instead, the parsing will
1339be confused, perhaps silently.
1340
1341 use locale;
1342 use POSIX qw(locale_h);
1343 setlocale(LC_NUMERIC, "fr_FR") or die "Pardon";
1344 my $a = 1.2;
1345 print eval "$a + 1.5";
1346 print "\n";
1347
1348prints C<13,5>. This is because in that locale, the comma is the
1349decimal point character. The C<eval> thus expands to:
1350
1351 eval "1,2 + 1.5"
1352
1353and the result is not what you likely expected. No warnings are
1354generated. If you do string C<eval>'s within the scope of
1355S<C<use locale>>, you should instead change the C<eval> line to do
1356something like:
1357
1358 print eval "no locale; $a + 1.5";
1359
1360This prints C<2.7>.
1361
d6ded950
KW
1362You could also exclude C<LC_NUMERIC>, if you don't need it, by
1363
1364 use locale ':!numeric';
1365
5f05dabc
PP
1366=head2 Backward compatibility
1367
b0c42ed9 1368Versions of Perl prior to 5.004 B<mostly> ignored locale information,
5a964f20
TC
1369generally behaving as if something similar to the C<"C"> locale were
1370always in force, even if the program environment suggested otherwise
5a0de581 1371(see L</The setlocale function>). By default, Perl still behaves this
5a964f20
TC
1372way for backward compatibility. If you want a Perl application to pay
1373attention to locale information, you B<must> use the S<C<use locale>>
5a0de581 1374pragma (see L</The "use locale" pragma>) or, in the unlikely event
062ca197 1375that you want to do so for just pattern matching, the
70709c68
KW
1376C</l> regular expression modifier (see L<perlre/Character set
1377modifiers>) to instruct it to do so.
b0c42ed9
JH
1378
1379Versions of Perl from 5.002 to 5.003 did use the C<LC_CTYPE>
5a964f20
TC
1380information if available; that is, C<\w> did understand what
1381were the letters according to the locale environment variables.
b0c42ed9
JH
1382The problem was that the user had no control over the feature:
1383if the C library supported locales, Perl used them.
1384
1385=head2 I18N:Collate obsolete
1386
5a964f20 1387In versions of Perl prior to 5.004, per-locale collation was possible
b0c42ed9
JH
1388using the C<I18N::Collate> library module. This module is now mildly
1389obsolete and should be avoided in new applications. The C<LC_COLLATE>
1390functionality is now integrated into the Perl core language: One can
1391use locale-specific scalar data completely normally with C<use locale>,
1392so there is no longer any need to juggle with the scalar references of
1393C<I18N::Collate>.
5f05dabc 1394
14280422 1395=head2 Sort speed and memory use impacts
5f05dabc
PP
1396
1397Comparing and sorting by locale is usually slower than the default
14280422
DD
1398sorting; slow-downs of two to four times have been observed. It will
1399also consume more memory: once a Perl scalar variable has participated
1400in any string comparison or sorting operation obeying the locale
1401collation rules, it will take 3-15 times more memory than before. (The
1402exact multiplier depends on the string's contents, the operating system
1403and the locale.) These downsides are dictated more by the operating
1404system's implementation of the locale system than by Perl.
5f05dabc 1405
5f05dabc
PP
1406=head2 Freely available locale definitions
1407
66cbab2c
KW
1408The Unicode CLDR project extracts the POSIX portion of many of its
1409locales, available at
1410
e784ce68
KW
1411 http://unicode.org/Public/cldr/2.0.1/
1412
1413(Newer versions of CLDR require you to compute the POSIX data yourself.
1414See L<http://unicode.org/Public/cldr/latest/>.)
66cbab2c 1415
08d7a6b2
LB
1416There is a large collection of locale definitions at:
1417
1418 http://std.dkuug.dk/i18n/WG15-collection/locales/
1419
1420You should be aware that it is
14280422 1421unsupported, and is not claimed to be fit for any purpose. If your
5a964f20 1422system allows installation of arbitrary locales, you may find the
14280422
DD
1423definitions useful as they are, or as a basis for the development of
1424your own locales.
5f05dabc 1425
14280422 1426=head2 I18n and l10n
5f05dabc 1427
b0c42ed9
JH
1428"Internationalization" is often abbreviated as B<i18n> because its first
1429and last letters are separated by eighteen others. (You may guess why
1430the internalin ... internaliti ... i18n tends to get abbreviated.) In
1431the same way, "localization" is often abbreviated to B<l10n>.
14280422
DD
1432
1433=head2 An imperfect standard
1434
1435Internationalization, as defined in the C and POSIX standards, can be
1436criticized as incomplete, ungainly, and having too large a granularity.
1437(Locales apply to a whole process, when it would arguably be more useful
1438to have them apply to a single thread, window group, or whatever.) They
1439also have a tendency, like standards groups, to divide the world into
1440nations, when we all know that the world can equally well be divided
e199995e 1441into bankers, bikers, gamers, and so on.
5f05dabc 1442
b310b053
JH
1443=head1 Unicode and UTF-8
1444
7ee2ae1e 1445The support of Unicode is new starting from Perl version v5.6, and more fully
31f05a37
KW
1446implemented in versions v5.8 and later. See L<perluniintro>.
1447
a4a439fb
KW
1448Starting in Perl v5.20, UTF-8 locales are supported in Perl, except
1449C<LC_COLLATE> is only partially supported; collation support is improved
1450in Perl v5.26 to a level that may be sufficient for your needs
1451(see L</Category C<LC_COLLATE>: Collation: Text Comparisons and Sorting>).
1452
1453If you have Perl v5.16 or v5.18 and can't upgrade, you can use
66cbab2c
KW
1454
1455 use locale ':not_characters';
1456
1457When this form of the pragma is used, only the non-character portions of
1458locales are used by Perl, for example C<LC_NUMERIC>. Perl assumes that
1459you have translated all the characters it is to operate on into Unicode
1460(actually the platform's native character set (ASCII or EBCDIC) plus
1461Unicode). For data in files, this can conveniently be done by also
1462specifying
1463
1464 use open ':locale';
1465
1466This pragma arranges for all inputs from files to be translated into
1467Unicode from the current locale as specified in the environment (see
1468L</ENVIRONMENT>), and all outputs to files to be translated back
1469into the locale. (See L<open>). On a per-filehandle basis, you can
1470instead use the L<PerlIO::locale> module, or the L<Encode::Locale>
1471module, both available from CPAN. The latter module also has methods to
1472ease the handling of C<ARGV> and environment variables, and can be used
31f05a37 1473on individual strings. If you know that all your locales will be
66cbab2c
KW
1474UTF-8, as many are these days, you can use the L<B<-C>|perlrun/-C>
1475command line switch.
1476
1477This form of the pragma allows essentially seamless handling of locales
31f05a37 1478with Unicode. The collation order will be by Unicode code point order.
a4a439fb 1479L<Unicode::Collate> can be used to get Unicode rules collation.
66cbab2c 1480
31f05a37
KW
1481All the modules and switches just described can be used in v5.20 with
1482just plain C<use locale>, and, should the input locales not be UTF-8,
1483you'll get the less than ideal behavior, described below, that you get
1484with pre-v5.16 Perls, or when you use the locale pragma without the
1485C<:not_characters> parameter in v5.16 and v5.18. If you are using
1486exclusively UTF-8 locales in v5.20 and higher, the rest of this section
1487does not apply to you.
1488
1489There are two cases, multi-byte and single-byte locales. First
1490multi-byte:
1491
1492The only multi-byte (or wide character) locale that Perl is ever likely
1493to support is UTF-8. This is due to the difficulty of implementation,
1494the fact that high quality UTF-8 locales are now published for every
e784ce68
KW
1495area of the world (L<http://unicode.org/Public/cldr/2.0.1/> for
1496ones that are already set-up, but from an earlier version;
1497L<http://unicode.org/Public/cldr/latest/> for the most up-to-date, but
1498you have to extract the POSIX information yourself), and that
31f05a37
KW
1499failing all that you can use the L<Encode> module to translate to/from
1500your locale. So, you'll have to do one of those things if you're using
1501one of these locales, such as Big5 or Shift JIS. For UTF-8 locales, in
1502Perls (pre v5.20) that don't have full UTF-8 locale support, they may
1503work reasonably well (depending on your C library implementation)
1504simply because both
dc4bfc4b
KW
1505they and Perl store characters that take up multiple bytes the same way.
1506However, some, if not most, C library implementations may not process
1507the characters in the upper half of the Latin-1 range (128 - 255)
cb88b78e 1508properly under C<LC_CTYPE>. To see if a character is a particular type
dc4bfc4b
KW
1509under a locale, Perl uses the functions like C<isalnum()>. Your C
1510library may not work for UTF-8 locales with those functions, instead
dbf3c4d7
KW
1511only working under the newer wide library functions like C<iswalnum()>,
1512which Perl does not use.
1513These multi-byte locales are treated like single-byte locales, and will
1514have the restrictions described below. Starting in Perl v5.22 a warning
1515message is raised when Perl detects a multi-byte locale that it doesn't
1516fully support.
e199995e 1517
31f05a37 1518For single-byte locales,
e199995e 1519Perl generally takes the tack to use locale rules on code points that can fit
66cbab2c
KW
1520in a single byte, and Unicode rules for those that can't (though this
1521isn't uniformly applied, see the note at the end of this section). This
1522prevents many problems in locales that aren't UTF-8. Suppose the locale
1523is ISO8859-7, Greek. The character at 0xD7 there is a capital Chi. But
1524in the ISO8859-1 locale, Latin1, it is a multiplication sign. The POSIX
1525regular expression character class C<[[:alpha:]]> will magically match
15260xD7 in the Greek locale but not in the Latin one.
e199995e 1527
1d2ab946 1528However, there are places where this breaks down. Certain Perl constructs are
b4ffc3db
TC
1529for Unicode only, such as C<\p{Alpha}>. They assume that 0xD7 always has its
1530Unicode meaning (or the equivalent on EBCDIC platforms). Since Latin1 is a
1531subset of Unicode and 0xD7 is the multiplication sign in both Latin1 and
1532Unicode, C<\p{Alpha}> will never match it, regardless of locale. A similar
0c880285 1533issue occurs with C<\N{...}>. Prior to v5.20, it is therefore a bad
31f05a37 1534idea to use C<\p{}> or
66cbab2c 1535C<\N{}> under plain C<use locale>--I<unless> you can guarantee that the
dbf3c4d7 1536locale will be ISO8859-1. Use POSIX character classes instead.
66cbab2c
KW
1537
1538Another problem with this approach is that operations that cross the
1539single byte/multiple byte boundary are not well-defined, and so are
4a70680a 1540disallowed. (This boundary is between the codepoints at 255/256.)
66cbab2c
KW
1541For example, lower casing LATIN CAPITAL LETTER Y WITH DIAERESIS (U+0178)
1542should return LATIN SMALL LETTER Y WITH DIAERESIS (U+00FF). But in the
1543Greek locale, for example, there is no character at 0xFF, and Perl
1544has no way of knowing what the character at 0xFF is really supposed to
1545represent. Thus it disallows the operation. In this mode, the
1546lowercase of U+0178 is itself.
1547
1548The same problems ensue if you enable automatic UTF-8-ification of your
e199995e 1549standard file handles, default C<open()> layer, and C<@ARGV> on non-ISO8859-1,
b4ffc3db
TC
1550non-UTF-8 locales (by using either the B<-C> command line switch or the
1551C<PERL_UNICODE> environment variable; see L<perlrun>).
1552Things are read in as UTF-8, which would normally imply a Unicode
1553interpretation, but the presence of a locale causes them to be interpreted
1554in that locale instead. For example, a 0xD7 code point in the Unicode
1555input, which should mean the multiplication sign, won't be interpreted by
66cbab2c 1556Perl that way under the Greek locale. This is not a problem
b4ffc3db 1557I<provided> you make certain that all locales will always and only be either
66cbab2c 1558an ISO8859-1, or, if you don't have a deficient C library, a UTF-8 locale.
b4ffc3db 1559
1d2ab946
KW
1560Still another problem is that this approach can lead to two code
1561points meaning the same character. Thus in a Greek locale, both U+03A7
1562and U+00D7 are GREEK CAPITAL LETTER CHI.
1563
613abc6d
KW
1564Because of all these problems, starting in v5.22, Perl will raise a
1565warning if a multi-byte (hence Unicode) code point is used when a
1566single-byte locale is in effect. (Although it doesn't check for this if
1567doing so would unreasonably slow execution down.)
1568
b4ffc3db
TC
1569Vendor locales are notoriously buggy, and it is difficult for Perl to test
1570its locale-handling code because this interacts with code that Perl has no
1571control over; therefore the locale-handling code in Perl may be buggy as
66cbab2c
KW
1572well. (However, the Unicode-supplied locales should be better, and
1573there is a feed back mechanism to correct any problems. See
1574L</Freely available locale definitions>.)
1575
7ee2ae1e 1576If you have Perl v5.16, the problems mentioned above go away if you use
66cbab2c 1577the C<:not_characters> parameter to the locale pragma (except for vendor
7ee2ae1e 1578bugs in the non-character portions). If you don't have v5.16, and you
66cbab2c
KW
1579I<do> have locales that work, using them may be worthwhile for certain
1580specific purposes, as long as you keep in mind the gotchas already
1581mentioned. For example, if the collation for your locales works, it
1582runs faster under locales than under L<Unicode::Collate>; and you gain
1583access to such things as the local currency symbol and the names of the
7ee2ae1e 1584months and days of the week. (But to hammer home the point, in v5.16,
66cbab2c
KW
1585you get this access without the downsides of locales by using the
1586C<:not_characters> form of the pragma.)
1587
1588Note: The policy of using locale rules for code points that can fit in a
1589byte, and Unicode rules for those that can't is not uniformly applied.
7ee2ae1e 1590Pre-v5.12, it was somewhat haphazard; in v5.12 it was applied fairly
66cbab2c 1591consistently to regular expression matching except for bracketed
7ee2ae1e 1592character classes; in v5.14 it was extended to all regex matches; and in
663d437a 1593v5.16 to the casing operations such as C<\L> and C<uc()>. For
dbf3c4d7 1594collation, in all releases so far, the system's C<strxfrm()> function is
a4a439fb
KW
1595called, and whatever it does is what you get. Starting in v5.26, various
1596bugs are fixed with the way perl uses this function.
b310b053 1597
5f05dabc
PP
1598=head1 BUGS
1599
4e615abd
KW
1600=head2 Collation of strings containing embedded C<NUL> characters
1601
6696cfa7
KW
1602C<NUL> characters will sort the same as the lowest collating control
1603character does, or to C<"\001"> in the unlikely event that there are no
1604control characters at all in the locale. In cases where the strings
1605don't contain this non-C<NUL> control, the results will be correct, and
1606in many locales, this control, whatever it might be, will rarely be
1607encountered. But there are cases where a C<NUL> should sort before this
1608control, but doesn't. If two strings do collate identically, the one
0c880285
KW
1609containing the C<NUL> will sort to earlier. Prior to 5.26, there were
1610more bugs.
4e615abd 1611
5f05dabc
PP
1612=head2 Broken systems
1613
5a964f20 1614In certain systems, the operating system's locale support
2bdf8add 1615is broken and cannot be fixed or used by Perl. Such deficiencies can
b4ffc3db 1616and will result in mysterious hangs and/or Perl core dumps when
2bdf8add 1617C<use locale> is in effect. When confronted with such a system,
7f2de2d2 1618please report in excruciating detail to <F<perlbug@perl.org>>, and
b4ffc3db 1619also contact your vendor: bug fixes may exist for these problems
2bdf8add 1620in your operating system. Sometimes such bug fixes are called an
83fb1bf0
KW
1621operating system upgrade. If you have the source for Perl, include in
1622the perlbug email the output of the test described above in L</Testing
1623for broken locales>.
5f05dabc
PP
1624
1625=head1 SEE ALSO
1626
b310b053 1627L<I18N::Langinfo>, L<perluniintro>, L<perlunicode>, L<open>,
106ab961 1628L<POSIX/localeconv>,
4bbcc6e8
JH
1629L<POSIX/setlocale>, L<POSIX/strcoll>, L<POSIX/strftime>,
1630L<POSIX/strtod>, L<POSIX/strxfrm>.
5f05dabc 1631
ccd65d51
KW
1632For special considerations when Perl is embedded in a C program,
1633see L<perlembed/Using embedded Perl with POSIX locales>.
1634
5f05dabc
PP
1635=head1 HISTORY
1636
b0c42ed9 1637Jarkko Hietaniemi's original F<perli18n.pod> heavily hacked by Dominic
5a964f20 1638Dunlop, assisted by the perl5-porters. Prose worked over a bit by
106ab961 1639Tom Christiansen, and now maintained by Perl 5 porters.