Commit | Line | Data |
---|---|---|
66cbab2c KW |
1 | =encoding utf8 |
2 | ||
5f05dabc | 3 | =head1 NAME |
4 | ||
b0c42ed9 | 5 | perllocale - Perl locale handling (internationalization and localization) |
5f05dabc | 6 | |
7 | =head1 DESCRIPTION | |
8 | ||
66cbab2c KW |
9 | In the beginning there was ASCII, the "American Standard Code for |
10 | Information Interchange", which works quite well for Americans with | |
11 | their English alphabet and dollar-denominated currency. But it doesn't | |
12 | work so well even for other English speakers, who may use different | |
13 | currencies, such as the pound sterling (as the symbol for that currency | |
14 | is not in ASCII); and it's hopelessly inadequate for many of the | |
15 | thousands of the world's other languages. | |
16 | ||
17 | To address these deficiencies, the concept of locales was invented | |
18 | (formally the ISO C, XPG4, POSIX 1.c "locale system"). And applications | |
19 | were and are being written that use the locale mechanism. The process of | |
20 | making such an application take account of its users' preferences in | |
21 | these kinds of matters is called B<internationalization> (often | |
22 | abbreviated as B<i18n>); telling such an application about a particular | |
23 | set of preferences is known as B<localization> (B<l10n>). | |
24 | ||
39332f68 | 25 | Perl has been extended to support the locale system. This |
66cbab2c KW |
26 | is controlled per application by using one pragma, one function call, |
27 | and several environment variables. | |
28 | ||
29 | Unfortunately, there are quite a few deficiencies with the design (and | |
31f05a37 KW |
30 | often, the implementations) of locales. Unicode was invented (see |
31 | L<perlunitut> for an introduction to that) in part to address these | |
32 | design deficiencies, and nowadays, there is a series of "UTF-8 | |
33 | locales", based on Unicode. These are locales whose character set is | |
34 | Unicode, encoded in UTF-8. Starting in v5.20, Perl fully supports | |
5c25e937 | 35 | UTF-8 locales, except for sorting and string comparisons. (Use |
31f05a37 | 36 | L<Unicode::Collate> for these.) Perl continues to support the old |
84035de0 KW |
37 | non UTF-8 locales as well. There are currently no UTF-8 locales for |
38 | EBCDIC platforms. | |
31f05a37 | 39 | |
66cbab2c KW |
40 | (Unicode is also creating C<CLDR>, the "Common Locale Data Repository", |
41 | L<http://cldr.unicode.org/> which includes more types of information than | |
42 | are available in the POSIX locale system. At the time of this writing, | |
43 | there was no CPAN module that provides access to this XML-encoded data. | |
44 | However, many of its locales have the POSIX-only data extracted, and are | |
31f05a37 KW |
45 | available as UTF-8 locales at |
46 | L<http://unicode.org/Public/cldr/latest/>.) | |
66cbab2c KW |
47 | |
48 | =head1 WHAT IS A LOCALE | |
49 | ||
50 | A locale is a set of data that describes various aspects of how various | |
51 | communities in the world categorize their world. These categories are | |
52 | broken down into the following types (some of which include a brief | |
53 | note here): | |
54 | ||
55 | =over | |
56 | ||
cb88b78e | 57 | =item Category C<LC_NUMERIC>: Numeric formatting |
66cbab2c KW |
58 | |
59 | This indicates how numbers should be formatted for human readability, | |
60 | for example the character used as the decimal point. | |
61 | ||
cb88b78e | 62 | =item Category C<LC_MONETARY>: Formatting of monetary amounts |
66cbab2c KW |
63 | |
64 | =for comment | |
ebc3223b | 65 | The nbsp below makes this look better (though not great) |
66cbab2c KW |
66 | |
67 | E<160> | |
68 | ||
cb88b78e | 69 | =item Category C<LC_TIME>: Date/Time formatting |
66cbab2c KW |
70 | |
71 | =for comment | |
ebc3223b | 72 | The nbsp below makes this look better (though not great) |
66cbab2c KW |
73 | |
74 | E<160> | |
75 | ||
cb88b78e | 76 | =item Category C<LC_MESSAGES>: Error and other messages |
66cbab2c | 77 | |
2619d284 | 78 | This is used by Perl itself only for accessing operating system error |
03c702c5 | 79 | messages via L<$!|perlvar/$ERRNO> and L<$^E|perlvar/$EXTENDED_OS_ERROR>. |
66cbab2c | 80 | |
cb88b78e | 81 | =item Category C<LC_COLLATE>: Collation |
66cbab2c | 82 | |
76073c88 | 83 | This indicates the ordering of letters for comparison and sorting. |
66cbab2c KW |
84 | In Latin alphabets, for example, "b", generally follows "a". |
85 | ||
cb88b78e | 86 | =item Category C<LC_CTYPE>: Character Types |
66cbab2c KW |
87 | |
88 | This indicates, for example if a character is an uppercase letter. | |
89 | ||
2619d284 KW |
90 | =item Other categories |
91 | ||
92 | Some platforms have other categories, dealing with such things as | |
93 | measurement units and paper sizes. None of these are used directly by | |
94 | Perl, but outside operations that Perl interacts with may use | |
d6ded950 | 95 | these. See L</Not within the scope of "use locale"> below. |
2619d284 | 96 | |
66cbab2c KW |
97 | =back |
98 | ||
2619d284 KW |
99 | More details on the categories used by Perl are given below in L</LOCALE |
100 | CATEGORIES>. | |
66cbab2c KW |
101 | |
102 | Together, these categories go a long way towards being able to customize | |
103 | a single program to run in many different locations. But there are | |
104 | deficiencies, so keep reading. | |
5f05dabc | 105 | |
106 | =head1 PREPARING TO USE LOCALES | |
107 | ||
c5c88224 KW |
108 | Perl itself (outside the L<POSIX> module) will not use locales unless |
109 | specifically requested to (but | |
2619d284 KW |
110 | again note that Perl may interact with code that does use them). Even |
111 | if there is such a request, B<all> of the following must be true | |
b960a36e | 112 | for it to work properly: |
5f05dabc | 113 | |
114 | =over 4 | |
115 | ||
116 | =item * | |
117 | ||
118 | B<Your operating system must support the locale system>. If it does, | |
39332f68 | 119 | you should find that the C<setlocale()> function is a documented part of |
5f05dabc | 120 | its C library. |
121 | ||
122 | =item * | |
123 | ||
5a964f20 | 124 | B<Definitions for locales that you use must be installed>. You, or |
14280422 DD |
125 | your system administrator, must make sure that this is the case. The |
126 | available locales, the location in which they are kept, and the manner | |
5a964f20 TC |
127 | in which they are installed all vary from system to system. Some systems |
128 | provide only a few, hard-wired locales and do not allow more to be | |
129 | added. Others allow you to add "canned" locales provided by the system | |
130 | supplier. Still others allow you or the system administrator to define | |
14280422 | 131 | and add arbitrary locales. (You may have to ask your supplier to |
5a964f20 | 132 | provide canned locales that are not delivered with your operating |
14280422 | 133 | system.) Read your system documentation for further illumination. |
5f05dabc | 134 | |
135 | =item * | |
136 | ||
137 | B<Perl must believe that the locale system is supported>. If it does, | |
138 | C<perl -V:d_setlocale> will say that the value for C<d_setlocale> is | |
139 | C<define>. | |
140 | ||
141 | =back | |
142 | ||
143 | If you want a Perl application to process and present your data | |
144 | according to a particular locale, the application code should include | |
d6ded950 | 145 | the S<C<use locale>> pragma (see L<The "use locale" pragma>) where |
5f05dabc | 146 | appropriate, and B<at least one> of the following must be true: |
147 | ||
148 | =over 4 | |
149 | ||
c052850d | 150 | =item 1 |
5f05dabc | 151 | |
66cbab2c | 152 | B<The locale-determining environment variables (see L</"ENVIRONMENT">) |
5a964f20 | 153 | must be correctly set up> at the time the application is started, either |
ef3087ec | 154 | by yourself or by whomever set up your system account; or |
5f05dabc | 155 | |
c052850d | 156 | =item 2 |
5f05dabc | 157 | |
14280422 DD |
158 | B<The application must set its own locale> using the method described in |
159 | L<The setlocale function>. | |
5f05dabc | 160 | |
161 | =back | |
162 | ||
163 | =head1 USING LOCALES | |
164 | ||
d6ded950 | 165 | =head2 The C<"use locale"> pragma |
5f05dabc | 166 | |
c5c88224 KW |
167 | By default, Perl itself (outside the L<POSIX> module) |
168 | ignores the current locale. The S<C<use locale>> | |
66cbab2c | 169 | pragma tells Perl to use the current locale for some operations. |
d6ded950 KW |
170 | Starting in v5.16, there are optional parameters to this pragma, |
171 | described below, which restrict which operations are affected by it. | |
c052850d KW |
172 | |
173 | The current locale is set at execution time by | |
174 | L<setlocale()|/The setlocale function> described below. If that function | |
175 | hasn't yet been called in the course of the program's execution, the | |
66cbab2c | 176 | current locale is that which was determined by the L</"ENVIRONMENT"> in |
ebc3223b | 177 | effect at the start of the program. |
dfcc8045 | 178 | If there is no valid environment, the current locale is whatever the |
65ebb059 KW |
179 | system default has been set to. On POSIX systems, it is likely, but |
180 | not necessarily, the "C" locale. On Windows, the default is set via the | |
181 | computer's S<C<Control Panel-E<gt>Regional and Language Options>> (or its | |
182 | current equivalent). | |
c052850d KW |
183 | |
184 | The operations that are affected by locale are: | |
5f05dabc | 185 | |
186 | =over 4 | |
187 | ||
d6ded950 | 188 | =item B<Not within the scope of C<"use locale">> |
b960a36e | 189 | |
d6ded950 KW |
190 | Only certain operations originating outside Perl should be affected, as |
191 | follows: | |
1d2ab946 KW |
192 | |
193 | =over 4 | |
194 | ||
195 | =item * | |
2619d284 | 196 | |
663d437a | 197 | The current locale is used when going outside of Perl with |
2619d284 KW |
198 | operations like L<system()|perlfunc/system LIST> or |
199 | L<qxE<sol>E<sol>|perlop/qxE<sol>STRINGE<sol>>, if those operations are | |
200 | locale-sensitive. | |
201 | ||
1d2ab946 KW |
202 | =item * |
203 | ||
2619d284 KW |
204 | Also Perl gives access to various C library functions through the |
205 | L<POSIX> module. Some of those functions are always affected by the | |
206 | current locale. For example, C<POSIX::strftime()> uses C<LC_TIME>; | |
207 | C<POSIX::strtod()> uses C<LC_NUMERIC>; C<POSIX::strcoll()> and | |
208 | C<POSIX::strxfrm()> use C<LC_COLLATE>; and character classification | |
209 | functions like C<POSIX::isalnum()> use C<LC_CTYPE>. All such functions | |
210 | will behave according to the current underlying locale, even if that | |
1d2ab946 | 211 | locale isn't exposed to Perl space. |
2619d284 | 212 | |
1d2ab946 KW |
213 | =item * |
214 | ||
52686f2a KW |
215 | XS modules for all categories but C<LC_NUMERIC> get the underlying |
216 | locale, and hence any C library functions they call will use that | |
8a384d3a | 217 | underlying locale. For more discussion, see L<perlxs/CAVEATS>. |
1d2ab946 KW |
218 | |
219 | =back | |
220 | ||
c5c88224 KW |
221 | Note that all C programs (including the perl interpreter, which is |
222 | written in C) always have an underlying locale. That locale is the "C" | |
223 | locale unless changed by a call to L<setlocale()|/The setlocale | |
224 | function>. When Perl starts up, it changes the underlying locale to the | |
225 | one which is indicated by the L</ENVIRONMENT>. When using the L<POSIX> | |
226 | module or writing XS code, it is important to keep in mind that the | |
227 | underlying locale may be something other than "C", even if the program | |
228 | hasn't explicitly changed it. | |
229 | ||
ebc3223b KW |
230 | =for comment |
231 | The nbsp below makes this look better (though not great) | |
232 | ||
233 | E<160> | |
234 | ||
235 | =item B<Lingering effects of C<S<use locale>>> | |
1d2ab946 KW |
236 | |
237 | Certain Perl operations that are set-up within the scope of a | |
d6ded950 | 238 | C<use locale> retain that effect even outside the scope. |
b960a36e KW |
239 | These include: |
240 | ||
241 | =over 4 | |
242 | ||
243 | =item * | |
244 | ||
245 | The output format of a L<write()|perlfunc/write> is determined by an | |
246 | earlier format declaration (L<perlfunc/format>), so whether or not the | |
247 | output is affected by locale is determined by if the C<format()> is | |
d6ded950 | 248 | within the scope of a C<use locale>, not whether the C<write()> |
b960a36e KW |
249 | is. |
250 | ||
251 | =item * | |
252 | ||
253 | Regular expression patterns can be compiled using | |
33be4c61 | 254 | L<qrE<sol>E<sol>|perlop/qrE<sol>STRINGE<sol>msixpodualn> with actual |
b960a36e KW |
255 | matching deferred to later. Again, it is whether or not the compilation |
256 | was done within the scope of C<use locale> that determines the match | |
257 | behavior, not if the matches are done within such a scope or not. | |
258 | ||
259 | =back | |
260 | ||
ebc3223b KW |
261 | =for comment |
262 | The nbsp below makes this look better (though not great) | |
263 | ||
d6ded950 | 264 | |
ebc3223b KW |
265 | E<160> |
266 | ||
d6ded950 | 267 | =item B<Under C<"use locale";>> |
66cbab2c KW |
268 | |
269 | =over 4 | |
270 | ||
271 | =item * | |
272 | ||
d6ded950 | 273 | All the above operations |
b960a36e KW |
274 | |
275 | =item * | |
276 | ||
277 | B<Format declarations> (L<perlfunc/format>) and hence any subsequent | |
278 | C<write()>s use C<LC_NUMERIC>. | |
66cbab2c KW |
279 | |
280 | =item * | |
281 | ||
b960a36e KW |
282 | B<stringification and output> use C<LC_NUMERIC>. |
283 | These include the results of | |
284 | C<print()>, | |
285 | C<printf()>, | |
286 | C<say()>, | |
287 | and | |
288 | C<sprintf()>. | |
66cbab2c | 289 | |
b960a36e KW |
290 | =item * |
291 | ||
292 | B<The comparison operators> (C<lt>, C<le>, C<cmp>, C<ge>, and C<gt>) use | |
39332f68 | 293 | C<LC_COLLATE>. C<sort()> is also affected if used without an |
5a964f20 | 294 | explicit comparison function, because it uses C<cmp> by default. |
14280422 | 295 | |
5a964f20 | 296 | B<Note:> C<eq> and C<ne> are unaffected by locale: they always |
de108802 | 297 | perform a char-by-char comparison of their scalar operands. What's |
14280422 DD |
298 | more, if C<cmp> finds that its operands are equal according to the |
299 | collation sequence specified by the current locale, it goes on to | |
de108802 RGS |
300 | perform a char-by-char comparison, and only returns I<0> (equal) if the |
301 | operands are char-for-char identical. If you really want to know whether | |
5a964f20 | 302 | two strings--which C<eq> and C<cmp> may consider different--are equal |
14280422 | 303 | as far as collation in the locale is concerned, see the discussion in |
cb88b78e | 304 | L<Category C<LC_COLLATE>: Collation>. |
5f05dabc | 305 | |
306 | =item * | |
307 | ||
39332f68 KW |
308 | B<Regular expressions and case-modification functions> (C<uc()>, C<lc()>, |
309 | C<ucfirst()>, and C<lcfirst()>) use C<LC_CTYPE> | |
5f05dabc | 310 | |
2c6ee1a7 KW |
311 | =item * |
312 | ||
dbf3c4d7 KW |
313 | B<The variables L<$!|perlvar/$ERRNO>> (and its synonyms C<$ERRNO> and |
314 | C<$OS_ERROR>) B<and L<$^E|perlvar/$EXTENDED_OS_ERROR>> (and its synonym | |
2c6ee1a7 KW |
315 | C<$EXTENDED_OS_ERROR>) when used as strings use C<LC_MESSAGES>. |
316 | ||
5f05dabc | 317 | =back |
318 | ||
66cbab2c | 319 | =back |
5f05dabc | 320 | |
5a964f20 | 321 | The default behavior is restored with the S<C<no locale>> pragma, or |
ef3087ec | 322 | upon reaching the end of the block enclosing C<use locale>. |
3cd61afa | 323 | Note that C<use locale> calls may be |
66cbab2c KW |
324 | nested, and that what is in effect within an inner scope will revert to |
325 | the outer scope's rules at the end of the inner scope. | |
5f05dabc | 326 | |
5a964f20 | 327 | The string result of any operation that uses locale |
14280422 DD |
328 | information is tainted, as it is possible for a locale to be |
329 | untrustworthy. See L<"SECURITY">. | |
5f05dabc | 330 | |
d6ded950 KW |
331 | Starting in Perl v5.16 in a very limited way, and more generally in |
332 | v5.22, you can restrict which category or categories are enabled by this | |
333 | particular instance of the pragma by adding parameters to it. For | |
334 | example, | |
335 | ||
336 | use locale qw(:ctype :numeric); | |
337 | ||
338 | enables locale awareness within its scope of only those operations | |
339 | (listed above) that are affected by C<LC_CTYPE> and C<LC_NUMERIC>. | |
340 | ||
341 | The possible categories are: C<:collate>, C<:ctype>, C<:messages>, | |
342 | C<:monetary>, C<:numeric>, C<:time>, and the pseudo category | |
343 | C<:characters> (described below). | |
344 | ||
345 | Thus you can say | |
346 | ||
347 | use locale ':messages'; | |
348 | ||
349 | and only L<$!|perlvar/$ERRNO> and L<$^E|perlvar/$EXTENDED_OS_ERROR> | |
350 | will be locale aware. Everything else is unaffected. | |
351 | ||
352 | Since Perl doesn't currently do anything with the C<LC_MONETARY> | |
353 | category, specifying C<:monetary> does effectively nothing. Some | |
354 | systems have other categories, such as C<LC_PAPER_SIZE>, but Perl | |
355 | also doesn't know anything about them, and there is no way to specify | |
356 | them in this pragma's arguments. | |
357 | ||
358 | You can also easily say to use all categories but one, by either, for | |
359 | example, | |
360 | ||
361 | use locale ':!ctype'; | |
362 | use locale ':not_ctype'; | |
363 | ||
364 | both of which mean to enable locale awarness of all categories but | |
365 | C<LC_CTYPE>. Only one category argument may be specified in a | |
366 | S<C<use locale>> if it is of the negated form. | |
367 | ||
368 | Prior to v5.22 only one form of the pragma with arguments is available: | |
369 | ||
370 | use locale ':not_characters'; | |
371 | ||
372 | (and you have to say C<not_>; you can't use the bang C<!> form). This | |
373 | pseudo category is a shorthand for specifying both C<:collate> and | |
374 | C<:ctype>. Hence, in the negated form, it is nearly the same thing as | |
375 | saying | |
376 | ||
377 | use locale qw(:messages :monetary :numeric :time); | |
378 | ||
379 | We use the term "nearly", because C<:not_characters> also turns on | |
380 | S<C<use feature 'unicode_strings'>> within its scope. This form is | |
381 | less useful in v5.20 and later, and is described fully in | |
382 | L</Unicode and UTF-8>, but briefly, it tells Perl to not use the | |
383 | character portions of the locale definition, that is the C<LC_CTYPE> and | |
384 | C<LC_COLLATE> categories. Instead it will use the native character set | |
385 | (extended by Unicode). When using this parameter, you are responsible | |
386 | for getting the external character set translated into the | |
387 | native/Unicode one (which it already will be if it is one of the | |
388 | increasingly popular UTF-8 locales). There are convenient ways of doing | |
389 | this, as described in L</Unicode and UTF-8>. | |
390 | ||
5f05dabc | 391 | =head2 The setlocale function |
392 | ||
14280422 | 393 | You can switch locales as often as you wish at run time with the |
39332f68 | 394 | C<POSIX::setlocale()> function: |
5f05dabc | 395 | |
5f05dabc | 396 | # Import locale-handling tool set from POSIX module. |
397 | # This example uses: setlocale -- the function call | |
398 | # LC_CTYPE -- explained below | |
2619d284 KW |
399 | # (Showing the testing for success/failure of operations is |
400 | # omitted in these examples to avoid distracting from the main | |
ebc3223b | 401 | # point) |
6ea81ccf | 402 | |
5f05dabc | 403 | use POSIX qw(locale_h); |
dfcc8045 KW |
404 | use locale; |
405 | my $old_locale; | |
5f05dabc | 406 | |
14280422 | 407 | # query and save the old locale |
5f05dabc | 408 | $old_locale = setlocale(LC_CTYPE); |
409 | ||
410 | setlocale(LC_CTYPE, "fr_CA.ISO8859-1"); | |
411 | # LC_CTYPE now in locale "French, Canada, codeset ISO 8859-1" | |
412 | ||
413 | setlocale(LC_CTYPE, ""); | |
65ebb059 KW |
414 | # LC_CTYPE now reset to the default defined by the |
415 | # LC_ALL/LC_CTYPE/LANG environment variables, or to the system | |
416 | # default. See below for documentation. | |
5f05dabc | 417 | |
418 | # restore the old locale | |
419 | setlocale(LC_CTYPE, $old_locale); | |
420 | ||
663d437a KW |
421 | This simultaneously affects all threads of the program, so it may be |
422 | problematic to use locales in threaded applications except where there | |
423 | is a single locale applicable to all threads. | |
424 | ||
39332f68 | 425 | The first argument of C<setlocale()> gives the B<category>, the second the |
14280422 DD |
426 | B<locale>. The category tells in what aspect of data processing you |
427 | want to apply locale-specific rules. Category names are discussed in | |
66cbab2c | 428 | L</LOCALE CATEGORIES> and L</"ENVIRONMENT">. The locale is the name of a |
14280422 DD |
429 | collection of customization information corresponding to a particular |
430 | combination of language, country or territory, and codeset. Read on for | |
431 | hints on the naming of locales: not all systems name locales as in the | |
432 | example. | |
433 | ||
39332f68 | 434 | If no second argument is provided and the category is something other |
cb88b78e | 435 | than C<LC_ALL>, the function returns a string naming the current locale |
502a173a | 436 | for the category. You can use this value as the second argument in a |
f170b852 KW |
437 | subsequent call to C<setlocale()>, B<but> on some platforms the string |
438 | is opaque, not something that most people would be able to decipher as | |
439 | to what locale it means. | |
502a173a | 440 | |
cb88b78e | 441 | If no second argument is provided and the category is C<LC_ALL>, the |
502a173a | 442 | result is implementation-dependent. It may be a string of |
c052850d | 443 | concatenated locale names (separator also implementation-dependent) |
39332f68 | 444 | or a single locale name. Please consult your L<setlocale(3)> man page for |
502a173a JH |
445 | details. |
446 | ||
447 | If a second argument is given and it corresponds to a valid locale, | |
448 | the locale for the category is set to that value, and the function | |
449 | returns the now-current locale value. You can then use this in yet | |
39332f68 | 450 | another call to C<setlocale()>. (In some implementations, the return |
502a173a JH |
451 | value may sometimes differ from the value you gave as the second |
452 | argument--think of it as an alias for the value you gave.) | |
5f05dabc | 453 | |
454 | As the example shows, if the second argument is an empty string, the | |
455 | category's locale is returned to the default specified by the | |
456 | corresponding environment variables. Generally, this results in a | |
5a964f20 | 457 | return to the default that was in force when Perl started up: changes |
54310121 | 458 | to the environment made by the application after startup may or may not |
5a964f20 | 459 | be noticed, depending on your system's C library. |
5f05dabc | 460 | |
d6ded950 KW |
461 | Note that when a form of C<use locale> that doesn't include all |
462 | categories is specified, Perl ignores the excluded categories. | |
66cbab2c | 463 | |
f170b852 | 464 | If C<set_locale()> fails for some reason (for example, an attempt to set |
dfcc8045 KW |
465 | to a locale unknown to the system), the locale for the category is not |
466 | changed, and the function returns C<undef>. | |
467 | ||
2619d284 | 468 | |
39332f68 | 469 | For further information about the categories, consult L<setlocale(3)>. |
3e6e419a JH |
470 | |
471 | =head2 Finding locales | |
472 | ||
39332f68 | 473 | For locales available in your system, consult also L<setlocale(3)> to |
5a964f20 TC |
474 | see whether it leads to the list of available locales (search for the |
475 | I<SEE ALSO> section). If that fails, try the following command lines: | |
5f05dabc | 476 | |
477 | locale -a | |
478 | ||
479 | nlsinfo | |
480 | ||
481 | ls /usr/lib/nls/loc | |
482 | ||
483 | ls /usr/lib/locale | |
484 | ||
485 | ls /usr/lib/nls | |
486 | ||
b478f28d JH |
487 | ls /usr/share/locale |
488 | ||
5f05dabc | 489 | and see whether they list something resembling these |
490 | ||
2bdf8add | 491 | en_US.ISO8859-1 de_DE.ISO8859-1 ru_RU.ISO8859-5 |
502a173a | 492 | en_US.iso88591 de_DE.iso88591 ru_RU.iso88595 |
2bdf8add | 493 | en_US de_DE ru_RU |
14280422 | 494 | en de ru |
2bdf8add JH |
495 | english german russian |
496 | english.iso88591 german.iso88591 russian.iso88595 | |
502a173a | 497 | english.roman8 russian.koi8r |
5f05dabc | 498 | |
39332f68 | 499 | Sadly, even though the calling interface for C<setlocale()> has been |
528d65ad | 500 | standardized, names of locales and the directories where the |
5a964f20 | 501 | configuration resides have not been. The basic form of the name is |
528d65ad JH |
502 | I<language_territory>B<.>I<codeset>, but the latter parts after |
503 | I<language> are not always present. The I<language> and I<country> | |
504 | are usually from the standards B<ISO 3166> and B<ISO 639>, the | |
505 | two-letter abbreviations for the countries and the languages of the | |
506 | world, respectively. The I<codeset> part often mentions some B<ISO | |
507 | 8859> character set, the Latin codesets. For example, C<ISO 8859-1> | |
508 | is the so-called "Western European codeset" that can be used to encode | |
509 | most Western European languages adequately. Again, there are several | |
510 | ways to write even the name of that one standard. Lamentably. | |
5f05dabc | 511 | |
14280422 DD |
512 | Two special locales are worth particular mention: "C" and "POSIX". |
513 | Currently these are effectively the same locale: the difference is | |
5a964f20 TC |
514 | mainly that the first one is defined by the C standard, the second by |
515 | the POSIX standard. They define the B<default locale> in which | |
14280422 | 516 | every program starts in the absence of locale information in its |
5a964f20 | 517 | environment. (The I<default> default locale, if you will.) Its language |
39332f68 KW |
518 | is (American) English and its character codeset ASCII or, rarely, a |
519 | superset thereof (such as the "DEC Multinational Character Set | |
520 | (DEC-MCS)"). B<Warning>. The C locale delivered by some vendors | |
521 | may not actually exactly match what the C standard calls for. So | |
522 | beware. | |
5f05dabc | 523 | |
14280422 DD |
524 | B<NOTE>: Not all systems have the "POSIX" locale (not all systems are |
525 | POSIX-conformant), so use "C" when you need explicitly to specify this | |
526 | default locale. | |
5f05dabc | 527 | |
3e6e419a JH |
528 | =head2 LOCALE PROBLEMS |
529 | ||
5a964f20 | 530 | You may encounter the following warning message at Perl startup: |
3e6e419a JH |
531 | |
532 | perl: warning: Setting locale failed. | |
533 | perl: warning: Please check that your locale settings: | |
534 | LC_ALL = "En_US", | |
535 | LANG = (unset) | |
536 | are supported and installed on your system. | |
537 | perl: warning: Falling back to the standard locale ("C"). | |
538 | ||
cb88b78e | 539 | This means that your locale settings had C<LC_ALL> set to "En_US" and |
5a964f20 TC |
540 | LANG exists but has no value. Perl tried to believe you but could not. |
541 | Instead, Perl gave up and fell back to the "C" locale, the default locale | |
65ebb059 KW |
542 | that is supposed to work no matter what. (On Windows, it first tries |
543 | falling back to the system default locale.) This usually means your | |
544 | locale settings were wrong, they mention locales your system has never | |
545 | heard of, or the locale installation in your system has problems (for | |
546 | example, some system files are broken or missing). There are quick and | |
547 | temporary fixes to these problems, as well as more thorough and lasting | |
548 | fixes. | |
3e6e419a | 549 | |
83fb1bf0 KW |
550 | =head2 Testing for broken locales |
551 | ||
552 | If you are building Perl from source, the Perl test suite file | |
553 | F<lib/locale.t> can be used to test the locales on your system. | |
554 | Setting the environment variable C<PERL_DEBUG_FULL_TEST> to 1 | |
555 | will cause it to output detailed results. For example, on Linux, you | |
556 | could say | |
557 | ||
1d2ab946 | 558 | PERL_DEBUG_FULL_TEST=1 ./perl -T -Ilib lib/locale.t > locale.log 2>&1 |
83fb1bf0 KW |
559 | |
560 | Besides many other tests, it will test every locale it finds on your | |
561 | system to see if they conform to the POSIX standard. If any have | |
562 | errors, it will include a summary near the end of the output of which | |
563 | locales passed all its tests, and which failed, and why. | |
564 | ||
3e6e419a JH |
565 | =head2 Temporarily fixing locale problems |
566 | ||
5a964f20 | 567 | The two quickest fixes are either to render Perl silent about any |
3e6e419a JH |
568 | locale inconsistencies or to run Perl under the default locale "C". |
569 | ||
570 | Perl's moaning about locale problems can be silenced by setting the | |
22ff3130 | 571 | environment variable C<PERL_BADLANG> to "0" or "". |
900bd440 JH |
572 | This method really just sweeps the problem under the carpet: you tell |
573 | Perl to shut up even when Perl sees that something is wrong. Do not | |
574 | be surprised if later something locale-dependent misbehaves. | |
3e6e419a JH |
575 | |
576 | Perl can be run under the "C" locale by setting the environment | |
cb88b78e KW |
577 | variable C<LC_ALL> to "C". This method is perhaps a bit more civilized |
578 | than the C<PERL_BADLANG> approach, but setting C<LC_ALL> (or | |
5a964f20 TC |
579 | other locale variables) may affect other programs as well, not just |
580 | Perl. In particular, external programs run from within Perl will see | |
3e6e419a | 581 | these changes. If you make the new settings permanent (read on), all |
f979aebc | 582 | programs you run see the changes. See L<"ENVIRONMENT"> for |
5a964f20 | 583 | the full list of relevant environment variables and L<USING LOCALES> |
e05ffc7d | 584 | for their effects in Perl. Effects in other programs are |
cb88b78e | 585 | easily deducible. For example, the variable C<LC_COLLATE> may well affect |
b432a672 | 586 | your B<sort> program (or whatever the program that arranges "records" |
3e6e419a JH |
587 | alphabetically in your system is called). |
588 | ||
5a964f20 TC |
589 | You can test out changing these variables temporarily, and if the |
590 | new settings seem to help, put those settings into your shell startup | |
663d437a | 591 | files. Consult your local documentation for the exact details. For |
5a964f20 | 592 | Bourne-like shells (B<sh>, B<ksh>, B<bash>, B<zsh>): |
3e6e419a JH |
593 | |
594 | LC_ALL=en_US.ISO8859-1 | |
595 | export LC_ALL | |
596 | ||
5a964f20 TC |
597 | This assumes that we saw the locale "en_US.ISO8859-1" using the commands |
598 | discussed above. We decided to try that instead of the above faulty | |
599 | locale "En_US"--and in Cshish shells (B<csh>, B<tcsh>) | |
3e6e419a JH |
600 | |
601 | setenv LC_ALL en_US.ISO8859-1 | |
c47ff5f1 | 602 | |
663d437a | 603 | or if you have the "env" application you can do (in any shell) |
c406981e JH |
604 | |
605 | env LC_ALL=en_US.ISO8859-1 perl ... | |
606 | ||
5a964f20 | 607 | If you do not know what shell you have, consult your local |
3e6e419a JH |
608 | helpdesk or the equivalent. |
609 | ||
610 | =head2 Permanently fixing locale problems | |
611 | ||
5a964f20 TC |
612 | The slower but superior fixes are when you may be able to yourself |
613 | fix the misconfiguration of your own environment variables. The | |
3e6e419a JH |
614 | mis(sing)configuration of the whole system's locales usually requires |
615 | the help of your friendly system administrator. | |
616 | ||
5a964f20 TC |
617 | First, see earlier in this document about L<Finding locales>. That tells |
618 | how to find which locales are really supported--and more importantly, | |
619 | installed--on your system. In our example error message, environment | |
620 | variables affecting the locale are listed in the order of decreasing | |
621 | importance (and unset variables do not matter). Therefore, having | |
622 | LC_ALL set to "En_US" must have been the bad choice, as shown by the | |
623 | error message. First try fixing locale settings listed first. | |
3e6e419a | 624 | |
5a964f20 TC |
625 | Second, if using the listed commands you see something B<exactly> |
626 | (prefix matches do not count and case usually counts) like "En_US" | |
627 | without the quotes, then you should be okay because you are using a | |
628 | locale name that should be installed and available in your system. | |
4a4eefd0 | 629 | In this case, see L<Permanently fixing your system's locale configuration>. |
3e6e419a | 630 | |
4a4eefd0 | 631 | =head2 Permanently fixing your system's locale configuration |
3e6e419a | 632 | |
5a964f20 | 633 | This is when you see something like: |
3e6e419a JH |
634 | |
635 | perl: warning: Please check that your locale settings: | |
636 | LC_ALL = "En_US", | |
637 | LANG = (unset) | |
638 | are supported and installed on your system. | |
639 | ||
640 | but then cannot see that "En_US" listed by the above-mentioned | |
5a964f20 TC |
641 | commands. You may see things like "en_US.ISO8859-1", but that isn't |
642 | the same. In this case, try running under a locale | |
643 | that you can list and which somehow matches what you tried. The | |
3e6e419a | 644 | rules for matching locale names are a bit vague because |
e05ffc7d | 645 | standardization is weak in this area. See again the |
13a2d996 | 646 | L<Finding locales> about general rules. |
3e6e419a | 647 | |
b687b08b | 648 | =head2 Fixing system locale configuration |
3e6e419a | 649 | |
5a964f20 TC |
650 | Contact a system administrator (preferably your own) and report the exact |
651 | error message you get, and ask them to read this same documentation you | |
652 | are now reading. They should be able to check whether there is something | |
653 | wrong with the locale configuration of the system. The L<Finding locales> | |
654 | section is unfortunately a bit vague about the exact commands and places | |
655 | because these things are not that standardized. | |
3e6e419a | 656 | |
5f05dabc | 657 | =head2 The localeconv function |
658 | ||
39332f68 | 659 | The C<POSIX::localeconv()> function allows you to get particulars of the |
14280422 | 660 | locale-dependent numeric formatting information specified by the current |
a835cd47 KW |
661 | underlying C<LC_NUMERIC> and C<LC_MONETARY> locales (regardless of |
662 | whether called from within the scope of C<S<use locale>> or not). (If | |
663 | you just want the name of | |
39332f68 | 664 | the current locale for a particular category, use C<POSIX::setlocale()> |
5a964f20 | 665 | with a single parameter--see L<The setlocale function>.) |
5f05dabc | 666 | |
667 | use POSIX qw(locale_h); | |
5f05dabc | 668 | |
669 | # Get a reference to a hash of locale-dependent info | |
670 | $locale_values = localeconv(); | |
671 | ||
672 | # Output sorted list of the values | |
673 | for (sort keys %$locale_values) { | |
14280422 | 674 | printf "%-20s = %s\n", $_, $locale_values->{$_} |
5f05dabc | 675 | } |
676 | ||
39332f68 | 677 | C<localeconv()> takes no arguments, and returns B<a reference to> a hash. |
5a964f20 | 678 | The keys of this hash are variable names for formatting, such as |
502a173a | 679 | C<decimal_point> and C<thousands_sep>. The values are the |
cea6626f | 680 | corresponding, er, values. See L<POSIX/localeconv> for a longer |
502a173a JH |
681 | example listing the categories an implementation might be expected to |
682 | provide; some provide more and others fewer. You don't need an | |
39332f68 | 683 | explicit C<use locale>, because C<localeconv()> always observes the |
502a173a | 684 | current locale. |
5f05dabc | 685 | |
5a964f20 TC |
686 | Here's a simple-minded example program that rewrites its command-line |
687 | parameters as integers correctly formatted in the current locale: | |
5f05dabc | 688 | |
ef3087ec KW |
689 | use POSIX qw(locale_h); |
690 | ||
691 | # Get some of locale's numeric formatting parameters | |
692 | my ($thousands_sep, $grouping) = | |
693 | @{localeconv()}{'thousands_sep', 'grouping'}; | |
694 | ||
695 | # Apply defaults if values are missing | |
696 | $thousands_sep = ',' unless $thousands_sep; | |
697 | ||
698 | # grouping and mon_grouping are packed lists | |
699 | # of small integers (characters) telling the | |
700 | # grouping (thousand_seps and mon_thousand_seps | |
701 | # being the group dividers) of numbers and | |
702 | # monetary quantities. The integers' meanings: | |
703 | # 255 means no more grouping, 0 means repeat | |
704 | # the previous grouping, 1-254 means use that | |
705 | # as the current grouping. Grouping goes from | |
706 | # right to left (low to high digits). In the | |
707 | # below we cheat slightly by never using anything | |
708 | # else than the first grouping (whatever that is). | |
709 | if ($grouping) { | |
710 | @grouping = unpack("C*", $grouping); | |
711 | } else { | |
712 | @grouping = (3); | |
713 | } | |
714 | ||
715 | # Format command line params for current locale | |
716 | for (@ARGV) { | |
717 | $_ = int; # Chop non-integer part | |
718 | 1 while | |
719 | s/(\d)(\d{$grouping[0]}($|$thousands_sep))/$1$thousands_sep$2/; | |
720 | print "$_"; | |
721 | } | |
722 | print "\n"; | |
5f05dabc | 723 | |
03ceeedf KW |
724 | Note that if the platform doesn't have C<LC_NUMERIC> and/or |
725 | C<LC_MONETARY> available or enabled, the corresponding elements of the | |
726 | hash will be missing. | |
727 | ||
74c76037 | 728 | =head2 I18N::Langinfo |
4bbcc6e8 JH |
729 | |
730 | Another interface for querying locale-dependent information is the | |
39332f68 | 731 | C<I18N::Langinfo::langinfo()> function, available at least in Unix-like |
4bbcc6e8 JH |
732 | systems and VMS. |
733 | ||
39332f68 KW |
734 | The following example will import the C<langinfo()> function itself and |
735 | three constants to be used as arguments to C<langinfo()>: a constant for | |
74c76037 JH |
736 | the abbreviated first day of the week (the numbering starts from |
737 | Sunday = 1) and two more constants for the affirmative and negative | |
738 | answers for a yes/no question in the current locale. | |
4bbcc6e8 | 739 | |
74c76037 | 740 | use I18N::Langinfo qw(langinfo ABDAY_1 YESSTR NOSTR); |
4bbcc6e8 | 741 | |
ef3087ec KW |
742 | my ($abday_1, $yesstr, $nostr) |
743 | = map { langinfo } qw(ABDAY_1 YESSTR NOSTR); | |
4bbcc6e8 | 744 | |
74c76037 | 745 | print "$abday_1? [$yesstr/$nostr] "; |
4bbcc6e8 | 746 | |
74c76037 JH |
747 | In other words, in the "C" (or English) locale the above will probably |
748 | print something like: | |
749 | ||
e05ffc7d | 750 | Sun? [yes/no] |
4bbcc6e8 JH |
751 | |
752 | See L<I18N::Langinfo> for more information. | |
753 | ||
5f05dabc | 754 | =head1 LOCALE CATEGORIES |
755 | ||
5a964f20 TC |
756 | The following subsections describe basic locale categories. Beyond these, |
757 | some combination categories allow manipulation of more than one | |
758 | basic category at a time. See L<"ENVIRONMENT"> for a discussion of these. | |
5f05dabc | 759 | |
cb88b78e | 760 | =head2 Category C<LC_COLLATE>: Collation |
5f05dabc | 761 | |
d6ded950 KW |
762 | In the scope of a S<C<use locale>> form that includes collation, Perl |
763 | looks to the C<LC_COLLATE> | |
5a964f20 | 764 | environment variable to determine the application's notions on collation |
b4ffc3db TC |
765 | (ordering) of characters. For example, "b" follows "a" in Latin |
766 | alphabets, but where do "E<aacute>" and "E<aring>" belong? And while | |
f87fa335 | 767 | "color" follows "chocolate" in English, what about in traditional Spanish? |
5f05dabc | 768 | |
60f0fa02 | 769 | The following collations all make sense and you may meet any of them |
dbf3c4d7 | 770 | if you C<"use locale">. |
60f0fa02 JH |
771 | |
772 | A B C D E a b c d e | |
35316ca3 | 773 | A a B b C c D d E e |
60f0fa02 JH |
774 | a A b B c C d D e E |
775 | a b c d e A B C D E | |
776 | ||
f1cbbd6e | 777 | Here is a code snippet to tell what "word" |
5a964f20 | 778 | characters are in the current locale, in that locale's order: |
5f05dabc | 779 | |
780 | use locale; | |
35316ca3 | 781 | print +(sort grep /\w/, map { chr } 0..255), "\n"; |
5f05dabc | 782 | |
14280422 DD |
783 | Compare this with the characters that you see and their order if you |
784 | state explicitly that the locale should be ignored: | |
5f05dabc | 785 | |
786 | no locale; | |
35316ca3 | 787 | print +(sort grep /\w/, map { chr } 0..255), "\n"; |
5f05dabc | 788 | |
789 | This machine-native collation (which is what you get unless S<C<use | |
790 | locale>> has appeared earlier in the same block) must be used for | |
791 | sorting raw binary data, whereas the locale-dependent collation of the | |
b0c42ed9 | 792 | first example is useful for natural text. |
5f05dabc | 793 | |
14280422 DD |
794 | As noted in L<USING LOCALES>, C<cmp> compares according to the current |
795 | collation locale when C<use locale> is in effect, but falls back to a | |
de108802 | 796 | char-by-char comparison for strings that the locale says are equal. You |
39332f68 | 797 | can use C<POSIX::strcoll()> if you don't want this fall-back: |
14280422 DD |
798 | |
799 | use POSIX qw(strcoll); | |
800 | $equal_in_locale = | |
801 | !strcoll("space and case ignored", "SpaceAndCaseIgnored"); | |
802 | ||
39332f68 | 803 | C<$equal_in_locale> will be true if the collation locale specifies a |
5a964f20 | 804 | dictionary-like ordering that ignores space characters completely and |
9e3a2af8 | 805 | which folds case. |
14280422 | 806 | |
dbf3c4d7 | 807 | Perl currently only supports single-byte locales for C<LC_COLLATE>. This means |
31f05a37 KW |
808 | that a UTF-8 locale likely will just give you machine-native ordering. |
809 | Use L<Unicode::Collate> for the full implementation of the Unicode | |
810 | Collation Algorithm. | |
811 | ||
5a964f20 | 812 | If you have a single string that you want to check for "equality in |
14280422 | 813 | locale" against several others, you might think you could gain a little |
39332f68 | 814 | efficiency by using C<POSIX::strxfrm()> in conjunction with C<eq>: |
14280422 DD |
815 | |
816 | use POSIX qw(strxfrm); | |
817 | $xfrm_string = strxfrm("Mixed-case string"); | |
818 | print "locale collation ignores spaces\n" | |
819 | if $xfrm_string eq strxfrm("Mixed-casestring"); | |
820 | print "locale collation ignores hyphens\n" | |
821 | if $xfrm_string eq strxfrm("Mixedcase string"); | |
822 | print "locale collation ignores case\n" | |
823 | if $xfrm_string eq strxfrm("mixed-case string"); | |
824 | ||
39332f68 | 825 | C<strxfrm()> takes a string and maps it into a transformed string for use |
de108802 | 826 | in char-by-char comparisons against other transformed strings during |
14280422 | 827 | collation. "Under the hood", locale-affected Perl comparison operators |
39332f68 KW |
828 | call C<strxfrm()> for both operands, then do a char-by-char |
829 | comparison of the transformed strings. By calling C<strxfrm()> explicitly | |
14280422 | 830 | and using a non locale-affected comparison, the example attempts to save |
5a964f20 | 831 | a couple of transformations. But in fact, it doesn't save anything: Perl |
2ae324a7 | 832 | magic (see L<perlguts/Magic Variables>) creates the transformed version of a |
5a964f20 | 833 | string the first time it's needed in a comparison, then keeps this version around |
14280422 | 834 | in case it's needed again. An example rewritten the easy way with |
e38874e2 | 835 | C<cmp> runs just about as fast. It also copes with null characters |
39332f68 | 836 | embedded in strings; if you call C<strxfrm()> directly, it treats the first |
5a964f20 TC |
837 | null it finds as a terminator. don't expect the transformed strings |
838 | it produces to be portable across systems--or even from one revision | |
39332f68 | 839 | of your operating system to the next. In short, don't call C<strxfrm()> |
e38874e2 | 840 | directly: let Perl do it for you. |
14280422 | 841 | |
5a964f20 | 842 | Note: C<use locale> isn't shown in some of these examples because it isn't |
dfcc8045 KW |
843 | needed: C<strcoll()> and C<strxfrm()> are POSIX functions |
844 | which use the standard system-supplied C<libc> functions that | |
845 | always obey the current C<LC_COLLATE> locale. | |
5f05dabc | 846 | |
cb88b78e | 847 | =head2 Category C<LC_CTYPE>: Character Types |
5f05dabc | 848 | |
d6ded950 KW |
849 | In the scope of a S<C<use locale>> form that includes C<LC_CTYPE>, Perl |
850 | obeys the C<LC_CTYPE> locale | |
14280422 | 851 | setting. This controls the application's notion of which characters are |
ebc3223b KW |
852 | alphabetic, numeric, punctuation, I<etc>. This affects Perl's C<\w> |
853 | regular expression metanotation, | |
f1cbbd6e | 854 | which stands for alphanumeric characters--that is, alphabetic, |
ebc3223b KW |
855 | numeric, and the platform's native underscore. |
856 | (Consult L<perlre> for more information about | |
14280422 | 857 | regular expressions.) Thanks to C<LC_CTYPE>, depending on your locale |
b4ffc3db TC |
858 | setting, characters like "E<aelig>", "E<eth>", "E<szlig>", and |
859 | "E<oslash>" may be understood as C<\w> characters. | |
ebc3223b KW |
860 | It also affects things like C<\s>, C<\D>, and the POSIX character |
861 | classes, like C<[[:graph:]]>. (See L<perlrecharclass> for more | |
862 | information on all these.) | |
5f05dabc | 863 | |
2c268ad5 | 864 | The C<LC_CTYPE> locale also provides the map used in transliterating |
68dc0745 | 865 | characters between lower and uppercase. This affects the case-mapping |
663d437a KW |
866 | functions--C<fc()>, C<lc()>, C<lcfirst()>, C<uc()>, and C<ucfirst()>; |
867 | case-mapping | |
b9cc4f69 KW |
868 | interpolation with C<\F>, C<\l>, C<\L>, C<\u>, or C<\U> in double-quoted |
869 | strings and C<s///> substitutions; and case-independent regular expression | |
e38874e2 DD |
870 | pattern matching using the C<i> modifier. |
871 | ||
2da736a2 | 872 | Finally, C<LC_CTYPE> affects the (deprecated) POSIX character-class test |
1d2ab946 | 873 | functions--C<POSIX::isalpha()>, C<POSIX::islower()>, and so on. For |
663d437a KW |
874 | example, if you move from the "C" locale to a 7-bit ISO 646 one, |
875 | you may find--possibly to your surprise--that C<"|"> moves from the | |
1d2ab946 | 876 | C<POSIX::ispunct()> class to C<POSIX::isalpha()>. |
ef3087ec | 877 | Unfortunately, this creates big problems for regular expressions. "|" still |
8c6180a9 KW |
878 | means alternation even though it matches C<\w>. Starting in v5.22, a |
879 | warning will be raised when such a locale is switched into. More | |
880 | details are given several paragraphs further down. | |
5f05dabc | 881 | |
31f05a37 KW |
882 | Starting in v5.20, Perl supports UTF-8 locales for C<LC_CTYPE>, but |
883 | otherwise Perl only supports single-byte locales, such as the ISO 8859 | |
884 | series. This means that wide character locales, for example for Asian | |
8c6180a9 KW |
885 | languages, are not well-supported. (If the platform has the capability |
886 | for Perl to detect such a locale, starting in Perl v5.22, | |
887 | L<Perl will warn, default enabled|warnings/Category Hierarchy>, | |
888 | using the C<locale> warning category, whenever such a locale is switched | |
889 | into.) The UTF-8 locale support is actually a | |
31f05a37 | 890 | superset of POSIX locales, because it is really full Unicode behavior |
13af4fd9 KW |
891 | as if no C<LC_CTYPE> locale were in effect at all (except for tainting; |
892 | see L</SECURITY>). POSIX locales, even UTF-8 ones, | |
31f05a37 KW |
893 | are lacking certain concepts in Unicode, such as the idea that changing |
894 | the case of a character could expand to be more than one character. | |
895 | Perl in a UTF-8 locale, will give you that expansion. Prior to v5.20, | |
896 | Perl treated a UTF-8 locale on some platforms like an ISO 8859-1 one, | |
897 | with some restrictions, and on other platforms more like the "C" locale. | |
898 | For releases v5.16 and v5.18, C<S<use locale 'not_characters>> could be | |
899 | used as a workaround for this (see L</Unicode and UTF-8>). | |
900 | ||
5d63e270 | 901 | Note that there are quite a few things that are unaffected by the |
663d437a KW |
902 | current locale. Any literal character is the native character for the |
903 | given platform. Hence 'A' means the character at code point 65 on ASCII | |
904 | platforms, and 193 on EBCDIC. That may or may not be an 'A' in the | |
905 | current locale, if that locale even has an 'A'. | |
906 | Similarly, all the escape sequences for particular characters, | |
5d63e270 KW |
907 | C<\n> for example, always mean the platform's native one. This means, |
908 | for example, that C<\N> in regular expressions (every character | |
1d2ab946 | 909 | but new-line) works on the platform character set. |
5d63e270 | 910 | |
8c6180a9 KW |
911 | Starting in v5.22, Perl will by default warn when switching into a |
912 | locale that redefines any ASCII printable character (plus C<\t> and | |
84035de0 KW |
913 | C<\n>) into a different class than expected. This is likely to |
914 | happen on modern locales only on EBCDIC platforms, where, for example, | |
915 | a CCSID 0037 locale on a CCSID 1047 machine moves C<"[">, but it can | |
916 | happen on ASCII platforms with the ISO 646 and other | |
8c6180a9 KW |
917 | 7-bit locales that are essentially obsolete. Things may still work, |
918 | depending on what features of Perl are used by the program. For | |
919 | example, in the example from above where C<"|"> becomes a C<\w>, and | |
920 | there are no regular expressions where this matters, the program may | |
921 | still work properly. The warning lists all the characters that | |
922 | it can determine could be adversely affected. | |
923 | ||
14280422 DD |
924 | B<Note:> A broken or malicious C<LC_CTYPE> locale definition may result |
925 | in clearly ineligible characters being considered to be alphanumeric by | |
e199995e | 926 | your application. For strict matching of (mundane) ASCII letters and |
5a964f20 | 927 | digits--for example, in command strings--locale-aware applications |
e199995e | 928 | should use C<\w> with the C</a> regular expression modifier. See L<"SECURITY">. |
5f05dabc | 929 | |
fee33030 | 930 | =head2 Category C<LC_NUMERIC>: Numeric Formatting |
5f05dabc | 931 | |
d6ded950 KW |
932 | After a proper C<POSIX::setlocale()> call, and within the scope of |
933 | of a C<use locale> form that includes numerics, Perl obeys the | |
934 | C<LC_NUMERIC> locale information, which controls an application's idea | |
935 | of how numbers should be formatted for human readability. | |
b960a36e | 936 | In most implementations the only effect is to |
b4ffc3db | 937 | change the character used for the decimal point--perhaps from "." to ",". |
b960a36e | 938 | The functions aren't aware of such niceties as thousands separation and |
2095dafa | 939 | so on. (See L<The localeconv function> if you care about these things.) |
5a964f20 | 940 | |
b960a36e KW |
941 | use POSIX qw(strtod setlocale LC_NUMERIC); |
942 | use locale; | |
5f05dabc | 943 | |
b960a36e | 944 | setlocale LC_NUMERIC, ""; |
14280422 | 945 | |
b960a36e | 946 | $n = 5/2; # Assign numeric 2.5 to $n |
5f05dabc | 947 | |
b960a36e | 948 | $a = " $n"; # Locale-dependent conversion to string |
5f05dabc | 949 | |
b960a36e | 950 | print "half five is $n\n"; # Locale-dependent output |
5f05dabc | 951 | |
b960a36e | 952 | printf "half five is %g\n", $n; # Locale-dependent output |
5f05dabc | 953 | |
b960a36e KW |
954 | print "DECIMAL POINT IS COMMA\n" |
955 | if $n == (strtod("2,5"))[0]; # Locale-dependent conversion | |
5f05dabc | 956 | |
4bbcc6e8 JH |
957 | See also L<I18N::Langinfo> and C<RADIXCHAR>. |
958 | ||
cb88b78e | 959 | =head2 Category C<LC_MONETARY>: Formatting of monetary amounts |
5f05dabc | 960 | |
e199995e | 961 | The C standard defines the C<LC_MONETARY> category, but not a function |
5a964f20 | 962 | that is affected by its contents. (Those with experience of standards |
b0c42ed9 | 963 | committees will recognize that the working group decided to punt on the |
fa9b773e KW |
964 | issue.) Consequently, Perl essentially takes no notice of it. If you |
965 | really want to use C<LC_MONETARY>, you can query its contents--see | |
e05ffc7d KW |
966 | L<The localeconv function>--and use the information that it returns in your |
967 | application's own formatting of currency amounts. However, you may well | |
968 | find that the information, voluminous and complex though it may be, still | |
969 | does not quite meet your requirements: currency formatting is a hard nut | |
13a2d996 | 970 | to crack. |
5f05dabc | 971 | |
4bbcc6e8 JH |
972 | See also L<I18N::Langinfo> and C<CRNCYSTR>. |
973 | ||
cb88b78e | 974 | =head2 C<LC_TIME> |
5f05dabc | 975 | |
39332f68 | 976 | Output produced by C<POSIX::strftime()>, which builds a formatted |
5f05dabc | 977 | human-readable date/time string, is affected by the current C<LC_TIME> |
978 | locale. Thus, in a French locale, the output produced by the C<%B> | |
979 | format element (full month name) for the first month of the year would | |
5a964f20 | 980 | be "janvier". Here's how to get a list of long month names in the |
5f05dabc | 981 | current locale: |
982 | ||
983 | use POSIX qw(strftime); | |
14280422 DD |
984 | for (0..11) { |
985 | $long_month_name[$_] = | |
986 | strftime("%B", 0, 0, 0, 1, $_, 96); | |
5f05dabc | 987 | } |
988 | ||
2619d284 KW |
989 | Note: C<use locale> isn't needed in this example: C<strftime()> is a POSIX |
990 | function which uses the standard system-supplied C<libc> function that | |
991 | always obeys the current C<LC_TIME> locale. | |
5f05dabc | 992 | |
4bbcc6e8 | 993 | See also L<I18N::Langinfo> and C<ABDAY_1>..C<ABDAY_7>, C<DAY_1>..C<DAY_7>, |
2a2bf5f4 | 994 | C<ABMON_1>..C<ABMON_12>, and C<ABMON_1>..C<ABMON_12>. |
4bbcc6e8 | 995 | |
5f05dabc | 996 | =head2 Other categories |
997 | ||
2619d284 KW |
998 | The remaining locale categories are not currently used by Perl itself. |
999 | But again note that things Perl interacts with may use these, including | |
1000 | extensions outside the standard Perl distribution, and by the | |
98a6f11e | 1001 | operating system and its utilities. Note especially that the string |
1002 | value of C<$!> and the error messages given by external utilities may | |
1003 | be changed by C<LC_MESSAGES>. If you want to have portable error | |
265f5c4a | 1004 | codes, use C<%!>. See L<Errno>. |
14280422 DD |
1005 | |
1006 | =head1 SECURITY | |
1007 | ||
5a964f20 | 1008 | Although the main discussion of Perl security issues can be found in |
14280422 DD |
1009 | L<perlsec>, a discussion of Perl's locale handling would be incomplete |
1010 | if it did not draw your attention to locale-dependent security issues. | |
5a964f20 TC |
1011 | Locales--particularly on systems that allow unprivileged users to |
1012 | build their own locales--are untrustworthy. A malicious (or just plain | |
14280422 DD |
1013 | broken) locale can make a locale-aware application give unexpected |
1014 | results. Here are a few possibilities: | |
1015 | ||
1016 | =over 4 | |
1017 | ||
1018 | =item * | |
1019 | ||
1020 | Regular expression checks for safe file names or mail addresses using | |
5a964f20 | 1021 | C<\w> may be spoofed by an C<LC_CTYPE> locale that claims that |
dbf3c4d7 | 1022 | characters such as C<"E<gt>"> and C<"|"> are alphanumeric. |
14280422 DD |
1023 | |
1024 | =item * | |
1025 | ||
e38874e2 | 1026 | String interpolation with case-mapping, as in, say, C<$dest = |
cb88b78e | 1027 | "C:\U$name.$ext">, may produce dangerous results if a bogus C<LC_CTYPE> |
e38874e2 DD |
1028 | case-mapping table is in effect. |
1029 | ||
1030 | =item * | |
1031 | ||
14280422 DD |
1032 | A sneaky C<LC_COLLATE> locale could result in the names of students with |
1033 | "D" grades appearing ahead of those with "A"s. | |
1034 | ||
1035 | =item * | |
1036 | ||
5a964f20 | 1037 | An application that takes the trouble to use information in |
14280422 | 1038 | C<LC_MONETARY> may format debits as if they were credits and vice versa |
5a964f20 | 1039 | if that locale has been subverted. Or it might make payments in US |
14280422 DD |
1040 | dollars instead of Hong Kong dollars. |
1041 | ||
1042 | =item * | |
1043 | ||
39332f68 | 1044 | The date and day names in dates formatted by C<strftime()> could be |
14280422 | 1045 | manipulated to advantage by a malicious user able to subvert the |
5a964f20 | 1046 | C<LC_DATE> locale. ("Look--it says I wasn't in the building on |
14280422 DD |
1047 | Sunday.") |
1048 | ||
1049 | =back | |
1050 | ||
1051 | Such dangers are not peculiar to the locale system: any aspect of an | |
5a964f20 | 1052 | application's environment which may be modified maliciously presents |
14280422 | 1053 | similar challenges. Similarly, they are not specific to Perl: any |
5a964f20 | 1054 | programming language that allows you to write programs that take |
14280422 DD |
1055 | account of their environment exposes you to these issues. |
1056 | ||
5a964f20 TC |
1057 | Perl cannot protect you from all possibilities shown in the |
1058 | examples--there is no substitute for your own vigilance--but, when | |
14280422 | 1059 | C<use locale> is in effect, Perl uses the tainting mechanism (see |
5a964f20 | 1060 | L<perlsec>) to mark string results that become locale-dependent, and |
14280422 | 1061 | which may be untrustworthy in consequence. Here is a summary of the |
5a964f20 | 1062 | tainting behavior of operators and functions that may be affected by |
14280422 DD |
1063 | the locale: |
1064 | ||
1065 | =over 4 | |
1066 | ||
551e1d92 RB |
1067 | =item * |
1068 | ||
1069 | B<Comparison operators> (C<lt>, C<le>, C<ge>, C<gt> and C<cmp>): | |
14280422 DD |
1070 | |
1071 | Scalar true/false (or less/equal/greater) result is never tainted. | |
1072 | ||
551e1d92 RB |
1073 | =item * |
1074 | ||
1d2ab946 | 1075 | B<Case-mapping interpolation> (with C<\l>, C<\L>, C<\u>, C<\U>, or C<\F>) |
e38874e2 | 1076 | |
d6ded950 KW |
1077 | The result string containing interpolated material is tainted if |
1078 | a C<use locale> form that includes C<LC_CTYPE> is in effect. | |
e38874e2 | 1079 | |
551e1d92 RB |
1080 | =item * |
1081 | ||
1082 | B<Matching operator> (C<m//>): | |
14280422 DD |
1083 | |
1084 | Scalar true/false result never tainted. | |
1085 | ||
1d2ab946 | 1086 | All subpatterns, either delivered as a list-context result or as C<$1> |
d6ded950 KW |
1087 | I<etc>., are tainted if a C<use locale> form that includes |
1088 | C<LC_CTYPE> is in effect, and the subpattern | |
63baef57 KW |
1089 | regular expression contains a locale-dependent construct. These |
1090 | constructs include C<\w> (to match an alphanumeric character), C<\W> | |
1091 | (non-alphanumeric character), C<\b> and C<\B> (word-boundary and | |
1092 | non-boundardy, which depend on what C<\w> and C<\W> match), C<\s> | |
1093 | (whitespace character), C<\S> (non whitespace character), C<\d> and | |
1094 | C<\D> (digits and non-digits), and the POSIX character classes, such as | |
1095 | C<[:alpha:]> (see L<perlrecharclass/POSIX Character Classes>). | |
1096 | ||
1097 | Tainting is also likely if the pattern is to be matched | |
1098 | case-insensitively (via C</i>). The exception is if all the code points | |
1099 | to be matched this way are above 255 and do not have folds under Unicode | |
1100 | rules to below 256. Tainting is not done for these because Perl | |
1101 | only uses Unicode rules for such code points, and those rules are the | |
1102 | same no matter what the current locale. | |
1103 | ||
1d2ab946 KW |
1104 | The matched-pattern variables, C<$&>, C<$`> (pre-match), C<$'> |
1105 | (post-match), and C<$+> (last match) also are tainted. | |
14280422 | 1106 | |
551e1d92 RB |
1107 | =item * |
1108 | ||
1109 | B<Substitution operator> (C<s///>): | |
14280422 | 1110 | |
e38874e2 | 1111 | Has the same behavior as the match operator. Also, the left |
d6ded950 KW |
1112 | operand of C<=~> becomes tainted when a C<use locale> |
1113 | form that includes C<LC_CTYPE> is in effect, if modified as | |
66cbab2c | 1114 | a result of a substitution based on a regular |
1d2ab946 KW |
1115 | expression match involving any of the things mentioned in the previous |
1116 | item, or of case-mapping, such as C<\l>, C<\L>,C<\u>, C<\U>, or C<\F>. | |
14280422 | 1117 | |
551e1d92 RB |
1118 | =item * |
1119 | ||
39332f68 | 1120 | B<Output formatting functions> (C<printf()> and C<write()>): |
14280422 | 1121 | |
3cf03d68 JH |
1122 | Results are never tainted because otherwise even output from print, |
1123 | for example C<print(1/7)>, should be tainted if C<use locale> is in | |
1124 | effect. | |
14280422 | 1125 | |
551e1d92 RB |
1126 | =item * |
1127 | ||
39332f68 | 1128 | B<Case-mapping functions> (C<lc()>, C<lcfirst()>, C<uc()>, C<ucfirst()>): |
14280422 | 1129 | |
d6ded950 KW |
1130 | Results are tainted if a C<use locale> form that includes C<LC_CTYPE> is |
1131 | in effect. | |
14280422 | 1132 | |
551e1d92 RB |
1133 | =item * |
1134 | ||
39332f68 KW |
1135 | B<POSIX locale-dependent functions> (C<localeconv()>, C<strcoll()>, |
1136 | C<strftime()>, C<strxfrm()>): | |
14280422 DD |
1137 | |
1138 | Results are never tainted. | |
1139 | ||
551e1d92 RB |
1140 | =item * |
1141 | ||
1d2ab946 KW |
1142 | B<POSIX character class tests> (C<POSIX::isalnum()>, |
1143 | C<POSIX::isalpha()>, C<POSIX::isdigit()>, C<POSIX::isgraph()>, | |
1144 | C<POSIX::islower()>, C<POSIX::isprint()>, C<POSIX::ispunct()>, | |
1145 | C<POSIX::isspace()>, C<POSIX::isupper()>, C<POSIX::isxdigit()>): | |
14280422 DD |
1146 | |
1147 | True/false results are never tainted. | |
1148 | ||
1149 | =back | |
1150 | ||
1151 | Three examples illustrate locale-dependent tainting. | |
1152 | The first program, which ignores its locale, won't run: a value taken | |
54310121 | 1153 | directly from the command line may not be used to name an output file |
14280422 DD |
1154 | when taint checks are enabled. |
1155 | ||
1156 | #/usr/local/bin/perl -T | |
1157 | # Run with taint checking | |
1158 | ||
54310121 | 1159 | # Command line sanity check omitted... |
14280422 DD |
1160 | $tainted_output_file = shift; |
1161 | ||
1162 | open(F, ">$tainted_output_file") | |
3183d96c | 1163 | or warn "Open of $tainted_output_file failed: $!\n"; |
14280422 DD |
1164 | |
1165 | The program can be made to run by "laundering" the tainted value through | |
5a964f20 TC |
1166 | a regular expression: the second example--which still ignores locale |
1167 | information--runs, creating the file named on its command line | |
14280422 DD |
1168 | if it can. |
1169 | ||
1170 | #/usr/local/bin/perl -T | |
1171 | ||
1172 | $tainted_output_file = shift; | |
1173 | $tainted_output_file =~ m%[\w/]+%; | |
1174 | $untainted_output_file = $&; | |
1175 | ||
1176 | open(F, ">$untainted_output_file") | |
1177 | or warn "Open of $untainted_output_file failed: $!\n"; | |
1178 | ||
5a964f20 | 1179 | Compare this with a similar but locale-aware program: |
14280422 DD |
1180 | |
1181 | #/usr/local/bin/perl -T | |
1182 | ||
1183 | $tainted_output_file = shift; | |
1184 | use locale; | |
1185 | $tainted_output_file =~ m%[\w/]+%; | |
1186 | $localized_output_file = $&; | |
1187 | ||
1188 | open(F, ">$localized_output_file") | |
1189 | or warn "Open of $localized_output_file failed: $!\n"; | |
1190 | ||
1d2ab946 | 1191 | This third program fails to run because C<$&> is tainted: it is the result |
5a964f20 | 1192 | of a match involving C<\w> while C<use locale> is in effect. |
5f05dabc | 1193 | |
1194 | =head1 ENVIRONMENT | |
1195 | ||
1196 | =over 12 | |
1197 | ||
ee1ec05f KW |
1198 | =item PERL_SKIP_LOCALE_INIT |
1199 | ||
c5c88224 KW |
1200 | This environment variable, available starting in Perl v5.20, if set |
1201 | (to any value), tells Perl to not use the rest of the | |
ee1ec05f KW |
1202 | environment variables to initialize with. Instead, Perl uses whatever |
1203 | the current locale settings are. This is particularly useful in | |
1204 | embedded environments, see | |
1205 | L<perlembed/Using embedded Perl with POSIX locales>. | |
1206 | ||
5f05dabc | 1207 | =item PERL_BADLANG |
1208 | ||
14280422 | 1209 | A string that can suppress Perl's warning about failed locale settings |
54310121 | 1210 | at startup. Failure can occur if the locale support in the operating |
5a964f20 | 1211 | system is lacking (broken) in some way--or if you mistyped the name of |
900bd440 | 1212 | a locale when you set up your environment. If this environment |
22ff3130 HS |
1213 | variable is absent, or has a value other than "0" or "", Perl will |
1214 | complain about locale setting failures. | |
5f05dabc | 1215 | |
cb88b78e | 1216 | B<NOTE>: C<PERL_BADLANG> only gives you a way to hide the warning message. |
14280422 DD |
1217 | The message tells about some problem in your system's locale support, |
1218 | and you should investigate what the problem is. | |
5f05dabc | 1219 | |
1220 | =back | |
1221 | ||
1222 | The following environment variables are not specific to Perl: They are | |
39332f68 | 1223 | part of the standardized (ISO C, XPG4, POSIX 1.c) C<setlocale()> method |
b385bb4d KW |
1224 | for controlling an application's opinion on data. Windows is non-POSIX, |
1225 | but Perl arranges for the following to work as described anyway. | |
65ebb059 KW |
1226 | If the locale given by an environment variable is not valid, Perl tries |
1227 | the next lower one in priority. If none are valid, on Windows, the | |
1228 | system default locale is then tried. If all else fails, the C<"C"> | |
1229 | locale is used. If even that doesn't work, something is badly broken, | |
c5e9a8e7 | 1230 | but Perl tries to forge ahead with whatever the locale settings might |
65ebb059 | 1231 | be. |
5f05dabc | 1232 | |
1233 | =over 12 | |
1234 | ||
cb88b78e | 1235 | =item C<LC_ALL> |
5f05dabc | 1236 | |
5a964f20 | 1237 | C<LC_ALL> is the "override-all" locale environment variable. If |
5f05dabc | 1238 | set, it overrides all the rest of the locale environment variables. |
1239 | ||
cb88b78e | 1240 | =item C<LANGUAGE> |
528d65ad JH |
1241 | |
1242 | B<NOTE>: C<LANGUAGE> is a GNU extension, it affects you only if you | |
1243 | are using the GNU libc. This is the case if you are using e.g. Linux. | |
e1020413 | 1244 | If you are using "commercial" Unixes you are most probably I<not> |
22b6f60d JH |
1245 | using GNU libc and you can ignore C<LANGUAGE>. |
1246 | ||
1247 | However, in the case you are using C<LANGUAGE>: it affects the | |
1248 | language of informational, warning, and error messages output by | |
1249 | commands (in other words, it's like C<LC_MESSAGES>) but it has higher | |
96090e4f | 1250 | priority than C<LC_ALL>. Moreover, it's not a single value but |
22b6f60d JH |
1251 | instead a "path" (":"-separated list) of I<languages> (not locales). |
1252 | See the GNU C<gettext> library documentation for more information. | |
528d65ad | 1253 | |
3ee1a09c | 1254 | =item C<LC_CTYPE> |
5f05dabc | 1255 | |
1256 | In the absence of C<LC_ALL>, C<LC_CTYPE> chooses the character type | |
1257 | locale. In the absence of both C<LC_ALL> and C<LC_CTYPE>, C<LANG> | |
1258 | chooses the character type locale. | |
1259 | ||
cb88b78e | 1260 | =item C<LC_COLLATE> |
5f05dabc | 1261 | |
14280422 DD |
1262 | In the absence of C<LC_ALL>, C<LC_COLLATE> chooses the collation |
1263 | (sorting) locale. In the absence of both C<LC_ALL> and C<LC_COLLATE>, | |
1264 | C<LANG> chooses the collation locale. | |
5f05dabc | 1265 | |
cb88b78e | 1266 | =item C<LC_MONETARY> |
5f05dabc | 1267 | |
14280422 DD |
1268 | In the absence of C<LC_ALL>, C<LC_MONETARY> chooses the monetary |
1269 | formatting locale. In the absence of both C<LC_ALL> and C<LC_MONETARY>, | |
1270 | C<LANG> chooses the monetary formatting locale. | |
5f05dabc | 1271 | |
cb88b78e | 1272 | =item C<LC_NUMERIC> |
5f05dabc | 1273 | |
1274 | In the absence of C<LC_ALL>, C<LC_NUMERIC> chooses the numeric format | |
1275 | locale. In the absence of both C<LC_ALL> and C<LC_NUMERIC>, C<LANG> | |
1276 | chooses the numeric format. | |
1277 | ||
cb88b78e | 1278 | =item C<LC_TIME> |
5f05dabc | 1279 | |
14280422 DD |
1280 | In the absence of C<LC_ALL>, C<LC_TIME> chooses the date and time |
1281 | formatting locale. In the absence of both C<LC_ALL> and C<LC_TIME>, | |
1282 | C<LANG> chooses the date and time formatting locale. | |
5f05dabc | 1283 | |
cb88b78e | 1284 | =item C<LANG> |
5f05dabc | 1285 | |
14280422 DD |
1286 | C<LANG> is the "catch-all" locale environment variable. If it is set, it |
1287 | is used as the last resort after the overall C<LC_ALL> and the | |
3ee1a09c | 1288 | category-specific C<LC_I<foo>>. |
5f05dabc | 1289 | |
1290 | =back | |
1291 | ||
7e4353e9 RGS |
1292 | =head2 Examples |
1293 | ||
cb88b78e | 1294 | The C<LC_NUMERIC> controls the numeric output: |
7e4353e9 | 1295 | |
ef3087ec KW |
1296 | use locale; |
1297 | use POSIX qw(locale_h); # Imports setlocale() and the LC_ constants. | |
1298 | setlocale(LC_NUMERIC, "fr_FR") or die "Pardon"; | |
1299 | printf "%g\n", 1.23; # If the "fr_FR" succeeded, probably shows 1,23. | |
7e4353e9 | 1300 | |
39332f68 | 1301 | and also how strings are parsed by C<POSIX::strtod()> as numbers: |
7e4353e9 | 1302 | |
ef3087ec KW |
1303 | use locale; |
1304 | use POSIX qw(locale_h strtod); | |
1305 | setlocale(LC_NUMERIC, "de_DE") or die "Entschuldigung"; | |
1306 | my $x = strtod("2,34") + 5; | |
1307 | print $x, "\n"; # Probably shows 7,34. | |
7e4353e9 | 1308 | |
5f05dabc | 1309 | =head1 NOTES |
1310 | ||
b960a36e KW |
1311 | =head2 String C<eval> and C<LC_NUMERIC> |
1312 | ||
1313 | A string L<eval|perlfunc/eval EXPR> parses its expression as standard | |
1314 | Perl. It is therefore expecting the decimal point to be a dot. If | |
1315 | C<LC_NUMERIC> is set to have this be a comma instead, the parsing will | |
1316 | be confused, perhaps silently. | |
1317 | ||
1318 | use locale; | |
1319 | use POSIX qw(locale_h); | |
1320 | setlocale(LC_NUMERIC, "fr_FR") or die "Pardon"; | |
1321 | my $a = 1.2; | |
1322 | print eval "$a + 1.5"; | |
1323 | print "\n"; | |
1324 | ||
1325 | prints C<13,5>. This is because in that locale, the comma is the | |
1326 | decimal point character. The C<eval> thus expands to: | |
1327 | ||
1328 | eval "1,2 + 1.5" | |
1329 | ||
1330 | and the result is not what you likely expected. No warnings are | |
1331 | generated. If you do string C<eval>'s within the scope of | |
1332 | S<C<use locale>>, you should instead change the C<eval> line to do | |
1333 | something like: | |
1334 | ||
1335 | print eval "no locale; $a + 1.5"; | |
1336 | ||
1337 | This prints C<2.7>. | |
1338 | ||
d6ded950 KW |
1339 | You could also exclude C<LC_NUMERIC>, if you don't need it, by |
1340 | ||
1341 | use locale ':!numeric'; | |
1342 | ||
5f05dabc | 1343 | =head2 Backward compatibility |
1344 | ||
b0c42ed9 | 1345 | Versions of Perl prior to 5.004 B<mostly> ignored locale information, |
5a964f20 TC |
1346 | generally behaving as if something similar to the C<"C"> locale were |
1347 | always in force, even if the program environment suggested otherwise | |
1348 | (see L<The setlocale function>). By default, Perl still behaves this | |
1349 | way for backward compatibility. If you want a Perl application to pay | |
1350 | attention to locale information, you B<must> use the S<C<use locale>> | |
d6ded950 | 1351 | pragma (see L<The "use locale" pragma>) or, in the unlikely event |
062ca197 | 1352 | that you want to do so for just pattern matching, the |
70709c68 KW |
1353 | C</l> regular expression modifier (see L<perlre/Character set |
1354 | modifiers>) to instruct it to do so. | |
b0c42ed9 JH |
1355 | |
1356 | Versions of Perl from 5.002 to 5.003 did use the C<LC_CTYPE> | |
5a964f20 TC |
1357 | information if available; that is, C<\w> did understand what |
1358 | were the letters according to the locale environment variables. | |
b0c42ed9 JH |
1359 | The problem was that the user had no control over the feature: |
1360 | if the C library supported locales, Perl used them. | |
1361 | ||
1362 | =head2 I18N:Collate obsolete | |
1363 | ||
5a964f20 | 1364 | In versions of Perl prior to 5.004, per-locale collation was possible |
b0c42ed9 JH |
1365 | using the C<I18N::Collate> library module. This module is now mildly |
1366 | obsolete and should be avoided in new applications. The C<LC_COLLATE> | |
1367 | functionality is now integrated into the Perl core language: One can | |
1368 | use locale-specific scalar data completely normally with C<use locale>, | |
1369 | so there is no longer any need to juggle with the scalar references of | |
1370 | C<I18N::Collate>. | |
5f05dabc | 1371 | |
14280422 | 1372 | =head2 Sort speed and memory use impacts |
5f05dabc | 1373 | |
1374 | Comparing and sorting by locale is usually slower than the default | |
14280422 DD |
1375 | sorting; slow-downs of two to four times have been observed. It will |
1376 | also consume more memory: once a Perl scalar variable has participated | |
1377 | in any string comparison or sorting operation obeying the locale | |
1378 | collation rules, it will take 3-15 times more memory than before. (The | |
1379 | exact multiplier depends on the string's contents, the operating system | |
1380 | and the locale.) These downsides are dictated more by the operating | |
1381 | system's implementation of the locale system than by Perl. | |
5f05dabc | 1382 | |
5f05dabc | 1383 | =head2 Freely available locale definitions |
1384 | ||
66cbab2c KW |
1385 | The Unicode CLDR project extracts the POSIX portion of many of its |
1386 | locales, available at | |
1387 | ||
1388 | http://unicode.org/Public/cldr/latest/ | |
1389 | ||
08d7a6b2 LB |
1390 | There is a large collection of locale definitions at: |
1391 | ||
1392 | http://std.dkuug.dk/i18n/WG15-collection/locales/ | |
1393 | ||
1394 | You should be aware that it is | |
14280422 | 1395 | unsupported, and is not claimed to be fit for any purpose. If your |
5a964f20 | 1396 | system allows installation of arbitrary locales, you may find the |
14280422 DD |
1397 | definitions useful as they are, or as a basis for the development of |
1398 | your own locales. | |
5f05dabc | 1399 | |
14280422 | 1400 | =head2 I18n and l10n |
5f05dabc | 1401 | |
b0c42ed9 JH |
1402 | "Internationalization" is often abbreviated as B<i18n> because its first |
1403 | and last letters are separated by eighteen others. (You may guess why | |
1404 | the internalin ... internaliti ... i18n tends to get abbreviated.) In | |
1405 | the same way, "localization" is often abbreviated to B<l10n>. | |
14280422 DD |
1406 | |
1407 | =head2 An imperfect standard | |
1408 | ||
1409 | Internationalization, as defined in the C and POSIX standards, can be | |
1410 | criticized as incomplete, ungainly, and having too large a granularity. | |
1411 | (Locales apply to a whole process, when it would arguably be more useful | |
1412 | to have them apply to a single thread, window group, or whatever.) They | |
1413 | also have a tendency, like standards groups, to divide the world into | |
1414 | nations, when we all know that the world can equally well be divided | |
e199995e | 1415 | into bankers, bikers, gamers, and so on. |
5f05dabc | 1416 | |
b310b053 JH |
1417 | =head1 Unicode and UTF-8 |
1418 | ||
7ee2ae1e | 1419 | The support of Unicode is new starting from Perl version v5.6, and more fully |
31f05a37 KW |
1420 | implemented in versions v5.8 and later. See L<perluniintro>. |
1421 | ||
1422 | Starting in Perl v5.20, UTF-8 locales are supported in Perl, except for | |
1423 | C<LC_COLLATE> (use L<Unicode::Collate> instead). If you have Perl v5.16 | |
1424 | or v5.18 and can't upgrade, you can use | |
66cbab2c KW |
1425 | |
1426 | use locale ':not_characters'; | |
1427 | ||
1428 | When this form of the pragma is used, only the non-character portions of | |
1429 | locales are used by Perl, for example C<LC_NUMERIC>. Perl assumes that | |
1430 | you have translated all the characters it is to operate on into Unicode | |
1431 | (actually the platform's native character set (ASCII or EBCDIC) plus | |
1432 | Unicode). For data in files, this can conveniently be done by also | |
1433 | specifying | |
1434 | ||
1435 | use open ':locale'; | |
1436 | ||
1437 | This pragma arranges for all inputs from files to be translated into | |
1438 | Unicode from the current locale as specified in the environment (see | |
1439 | L</ENVIRONMENT>), and all outputs to files to be translated back | |
1440 | into the locale. (See L<open>). On a per-filehandle basis, you can | |
1441 | instead use the L<PerlIO::locale> module, or the L<Encode::Locale> | |
1442 | module, both available from CPAN. The latter module also has methods to | |
1443 | ease the handling of C<ARGV> and environment variables, and can be used | |
31f05a37 | 1444 | on individual strings. If you know that all your locales will be |
66cbab2c KW |
1445 | UTF-8, as many are these days, you can use the L<B<-C>|perlrun/-C> |
1446 | command line switch. | |
1447 | ||
1448 | This form of the pragma allows essentially seamless handling of locales | |
31f05a37 KW |
1449 | with Unicode. The collation order will be by Unicode code point order. |
1450 | It is strongly | |
66cbab2c KW |
1451 | recommended that when you need to order and sort strings that you use |
1452 | the standard module L<Unicode::Collate> which gives much better results | |
1453 | in many instances than you can get with the old-style locale handling. | |
1454 | ||
31f05a37 KW |
1455 | All the modules and switches just described can be used in v5.20 with |
1456 | just plain C<use locale>, and, should the input locales not be UTF-8, | |
1457 | you'll get the less than ideal behavior, described below, that you get | |
1458 | with pre-v5.16 Perls, or when you use the locale pragma without the | |
1459 | C<:not_characters> parameter in v5.16 and v5.18. If you are using | |
1460 | exclusively UTF-8 locales in v5.20 and higher, the rest of this section | |
1461 | does not apply to you. | |
1462 | ||
1463 | There are two cases, multi-byte and single-byte locales. First | |
1464 | multi-byte: | |
1465 | ||
1466 | The only multi-byte (or wide character) locale that Perl is ever likely | |
1467 | to support is UTF-8. This is due to the difficulty of implementation, | |
1468 | the fact that high quality UTF-8 locales are now published for every | |
1469 | area of the world (L<http://unicode.org/Public/cldr/latest/>), and that | |
1470 | failing all that you can use the L<Encode> module to translate to/from | |
1471 | your locale. So, you'll have to do one of those things if you're using | |
1472 | one of these locales, such as Big5 or Shift JIS. For UTF-8 locales, in | |
1473 | Perls (pre v5.20) that don't have full UTF-8 locale support, they may | |
1474 | work reasonably well (depending on your C library implementation) | |
1475 | simply because both | |
dc4bfc4b KW |
1476 | they and Perl store characters that take up multiple bytes the same way. |
1477 | However, some, if not most, C library implementations may not process | |
1478 | the characters in the upper half of the Latin-1 range (128 - 255) | |
cb88b78e | 1479 | properly under C<LC_CTYPE>. To see if a character is a particular type |
dc4bfc4b KW |
1480 | under a locale, Perl uses the functions like C<isalnum()>. Your C |
1481 | library may not work for UTF-8 locales with those functions, instead | |
dbf3c4d7 KW |
1482 | only working under the newer wide library functions like C<iswalnum()>, |
1483 | which Perl does not use. | |
1484 | These multi-byte locales are treated like single-byte locales, and will | |
1485 | have the restrictions described below. Starting in Perl v5.22 a warning | |
1486 | message is raised when Perl detects a multi-byte locale that it doesn't | |
1487 | fully support. | |
e199995e | 1488 | |
31f05a37 | 1489 | For single-byte locales, |
e199995e | 1490 | Perl generally takes the tack to use locale rules on code points that can fit |
66cbab2c KW |
1491 | in a single byte, and Unicode rules for those that can't (though this |
1492 | isn't uniformly applied, see the note at the end of this section). This | |
1493 | prevents many problems in locales that aren't UTF-8. Suppose the locale | |
1494 | is ISO8859-7, Greek. The character at 0xD7 there is a capital Chi. But | |
1495 | in the ISO8859-1 locale, Latin1, it is a multiplication sign. The POSIX | |
1496 | regular expression character class C<[[:alpha:]]> will magically match | |
1497 | 0xD7 in the Greek locale but not in the Latin one. | |
e199995e | 1498 | |
1d2ab946 | 1499 | However, there are places where this breaks down. Certain Perl constructs are |
b4ffc3db TC |
1500 | for Unicode only, such as C<\p{Alpha}>. They assume that 0xD7 always has its |
1501 | Unicode meaning (or the equivalent on EBCDIC platforms). Since Latin1 is a | |
1502 | subset of Unicode and 0xD7 is the multiplication sign in both Latin1 and | |
1503 | Unicode, C<\p{Alpha}> will never match it, regardless of locale. A similar | |
31f05a37 KW |
1504 | issue occurs with C<\N{...}>. Prior to v5.20, It is therefore a bad |
1505 | idea to use C<\p{}> or | |
66cbab2c | 1506 | C<\N{}> under plain C<use locale>--I<unless> you can guarantee that the |
dbf3c4d7 | 1507 | locale will be ISO8859-1. Use POSIX character classes instead. |
66cbab2c KW |
1508 | |
1509 | Another problem with this approach is that operations that cross the | |
1510 | single byte/multiple byte boundary are not well-defined, and so are | |
4a70680a | 1511 | disallowed. (This boundary is between the codepoints at 255/256.) |
66cbab2c KW |
1512 | For example, lower casing LATIN CAPITAL LETTER Y WITH DIAERESIS (U+0178) |
1513 | should return LATIN SMALL LETTER Y WITH DIAERESIS (U+00FF). But in the | |
1514 | Greek locale, for example, there is no character at 0xFF, and Perl | |
1515 | has no way of knowing what the character at 0xFF is really supposed to | |
1516 | represent. Thus it disallows the operation. In this mode, the | |
1517 | lowercase of U+0178 is itself. | |
1518 | ||
1519 | The same problems ensue if you enable automatic UTF-8-ification of your | |
e199995e | 1520 | standard file handles, default C<open()> layer, and C<@ARGV> on non-ISO8859-1, |
b4ffc3db TC |
1521 | non-UTF-8 locales (by using either the B<-C> command line switch or the |
1522 | C<PERL_UNICODE> environment variable; see L<perlrun>). | |
1523 | Things are read in as UTF-8, which would normally imply a Unicode | |
1524 | interpretation, but the presence of a locale causes them to be interpreted | |
1525 | in that locale instead. For example, a 0xD7 code point in the Unicode | |
1526 | input, which should mean the multiplication sign, won't be interpreted by | |
66cbab2c | 1527 | Perl that way under the Greek locale. This is not a problem |
b4ffc3db | 1528 | I<provided> you make certain that all locales will always and only be either |
66cbab2c | 1529 | an ISO8859-1, or, if you don't have a deficient C library, a UTF-8 locale. |
b4ffc3db | 1530 | |
1d2ab946 KW |
1531 | Still another problem is that this approach can lead to two code |
1532 | points meaning the same character. Thus in a Greek locale, both U+03A7 | |
1533 | and U+00D7 are GREEK CAPITAL LETTER CHI. | |
1534 | ||
613abc6d KW |
1535 | Because of all these problems, starting in v5.22, Perl will raise a |
1536 | warning if a multi-byte (hence Unicode) code point is used when a | |
1537 | single-byte locale is in effect. (Although it doesn't check for this if | |
1538 | doing so would unreasonably slow execution down.) | |
1539 | ||
b4ffc3db TC |
1540 | Vendor locales are notoriously buggy, and it is difficult for Perl to test |
1541 | its locale-handling code because this interacts with code that Perl has no | |
1542 | control over; therefore the locale-handling code in Perl may be buggy as | |
66cbab2c KW |
1543 | well. (However, the Unicode-supplied locales should be better, and |
1544 | there is a feed back mechanism to correct any problems. See | |
1545 | L</Freely available locale definitions>.) | |
1546 | ||
7ee2ae1e | 1547 | If you have Perl v5.16, the problems mentioned above go away if you use |
66cbab2c | 1548 | the C<:not_characters> parameter to the locale pragma (except for vendor |
7ee2ae1e | 1549 | bugs in the non-character portions). If you don't have v5.16, and you |
66cbab2c KW |
1550 | I<do> have locales that work, using them may be worthwhile for certain |
1551 | specific purposes, as long as you keep in mind the gotchas already | |
1552 | mentioned. For example, if the collation for your locales works, it | |
1553 | runs faster under locales than under L<Unicode::Collate>; and you gain | |
1554 | access to such things as the local currency symbol and the names of the | |
7ee2ae1e | 1555 | months and days of the week. (But to hammer home the point, in v5.16, |
66cbab2c KW |
1556 | you get this access without the downsides of locales by using the |
1557 | C<:not_characters> form of the pragma.) | |
1558 | ||
1559 | Note: The policy of using locale rules for code points that can fit in a | |
1560 | byte, and Unicode rules for those that can't is not uniformly applied. | |
7ee2ae1e | 1561 | Pre-v5.12, it was somewhat haphazard; in v5.12 it was applied fairly |
66cbab2c | 1562 | consistently to regular expression matching except for bracketed |
7ee2ae1e | 1563 | character classes; in v5.14 it was extended to all regex matches; and in |
663d437a | 1564 | v5.16 to the casing operations such as C<\L> and C<uc()>. For |
dbf3c4d7 KW |
1565 | collation, in all releases so far, the system's C<strxfrm()> function is |
1566 | called, and whatever it does is what you get. | |
b310b053 | 1567 | |
5f05dabc | 1568 | =head1 BUGS |
1569 | ||
1570 | =head2 Broken systems | |
1571 | ||
5a964f20 | 1572 | In certain systems, the operating system's locale support |
2bdf8add | 1573 | is broken and cannot be fixed or used by Perl. Such deficiencies can |
b4ffc3db | 1574 | and will result in mysterious hangs and/or Perl core dumps when |
2bdf8add | 1575 | C<use locale> is in effect. When confronted with such a system, |
7f2de2d2 | 1576 | please report in excruciating detail to <F<perlbug@perl.org>>, and |
b4ffc3db | 1577 | also contact your vendor: bug fixes may exist for these problems |
2bdf8add | 1578 | in your operating system. Sometimes such bug fixes are called an |
83fb1bf0 KW |
1579 | operating system upgrade. If you have the source for Perl, include in |
1580 | the perlbug email the output of the test described above in L</Testing | |
1581 | for broken locales>. | |
5f05dabc | 1582 | |
1583 | =head1 SEE ALSO | |
1584 | ||
b310b053 JH |
1585 | L<I18N::Langinfo>, L<perluniintro>, L<perlunicode>, L<open>, |
1586 | L<POSIX/isalnum>, L<POSIX/isalpha>, | |
4bbcc6e8 JH |
1587 | L<POSIX/isdigit>, L<POSIX/isgraph>, L<POSIX/islower>, |
1588 | L<POSIX/isprint>, L<POSIX/ispunct>, L<POSIX/isspace>, | |
1589 | L<POSIX/isupper>, L<POSIX/isxdigit>, L<POSIX/localeconv>, | |
1590 | L<POSIX/setlocale>, L<POSIX/strcoll>, L<POSIX/strftime>, | |
1591 | L<POSIX/strtod>, L<POSIX/strxfrm>. | |
5f05dabc | 1592 | |
ccd65d51 KW |
1593 | For special considerations when Perl is embedded in a C program, |
1594 | see L<perlembed/Using embedded Perl with POSIX locales>. | |
1595 | ||
5f05dabc | 1596 | =head1 HISTORY |
1597 | ||
b0c42ed9 | 1598 | Jarkko Hietaniemi's original F<perli18n.pod> heavily hacked by Dominic |
5a964f20 | 1599 | Dunlop, assisted by the perl5-porters. Prose worked over a bit by |
c052850d | 1600 | Tom Christiansen, and updated by Perl 5 porters. |