3 perllocale - Perl locale handling (internationalization and localization)
7 Perl supports language-specific notions of data such as "is this
8 a letter", "what is the uppercase equivalent of this letter", and
9 "which of these letters comes first". These are important issues,
10 especially for languages other than English--but also for English: it
11 would be naE<iuml>ve to imagine that C<A-Za-z> defines all the "letters"
12 needed to write in English. Perl is also aware that some character other
13 than '.' may be preferred as a decimal point, and that output date
14 representations may be language-specific. The process of making an
15 application take account of its users' preferences in such matters is
16 called B<internationalization> (often abbreviated as B<i18n>); telling
17 such an application about a particular set of preferences is known as
18 B<localization> (B<l10n>).
20 Perl can understand language-specific data via the standardized (ISO C,
21 XPG4, POSIX 1.c) method called "the locale system". The locale system is
22 controlled per application using one pragma, one function call, and
23 several environment variables.
25 B<NOTE>: This feature is new in Perl 5.004, and does not apply unless an
26 application specifically requests it--see L<Backward compatibility>.
27 The one exception is that write() now B<always> uses the current locale
30 =head1 PREPARING TO USE LOCALES
32 If Perl applications are to understand and present your data
33 correctly according a locale of your choice, B<all> of the following
40 B<Your operating system must support the locale system>. If it does,
41 you should find that the setlocale() function is a documented part of
46 B<Definitions for locales that you use must be installed>. You, or
47 your system administrator, must make sure that this is the case. The
48 available locales, the location in which they are kept, and the manner
49 in which they are installed all vary from system to system. Some systems
50 provide only a few, hard-wired locales and do not allow more to be
51 added. Others allow you to add "canned" locales provided by the system
52 supplier. Still others allow you or the system administrator to define
53 and add arbitrary locales. (You may have to ask your supplier to
54 provide canned locales that are not delivered with your operating
55 system.) Read your system documentation for further illumination.
59 B<Perl must believe that the locale system is supported>. If it does,
60 C<perl -V:d_setlocale> will say that the value for C<d_setlocale> is
65 If you want a Perl application to process and present your data
66 according to a particular locale, the application code should include
67 the S<C<use locale>> pragma (see L<The use locale pragma>) where
68 appropriate, and B<at least one> of the following must be true:
74 B<The locale-determining environment variables (see L<"ENVIRONMENT">)
75 must be correctly set up> at the time the application is started, either
76 by yourself or by whoever set up your system account.
80 B<The application must set its own locale> using the method described in
81 L<The setlocale function>.
87 =head2 The use locale pragma
89 By default, Perl ignores the current locale. The S<C<use locale>>
90 pragma tells Perl to use the current locale for some operations:
96 B<The comparison operators> (C<lt>, C<le>, C<cmp>, C<ge>, and C<gt>) and
97 the POSIX string collation functions strcoll() and strxfrm() use
98 C<LC_COLLATE>. sort() is also affected if used without an
99 explicit comparison function, because it uses C<cmp> by default.
101 B<Note:> C<eq> and C<ne> are unaffected by locale: they always
102 perform a byte-by-byte comparison of their scalar operands. What's
103 more, if C<cmp> finds that its operands are equal according to the
104 collation sequence specified by the current locale, it goes on to
105 perform a byte-by-byte comparison, and only returns I<0> (equal) if the
106 operands are bit-for-bit identical. If you really want to know whether
107 two strings--which C<eq> and C<cmp> may consider different--are equal
108 as far as collation in the locale is concerned, see the discussion in
109 L<Category LC_COLLATE: Collation>.
113 B<Regular expressions and case-modification functions> (uc(), lc(),
114 ucfirst(), and lcfirst()) use C<LC_CTYPE>
118 B<The formatting functions> (printf(), sprintf() and write()) use
123 B<The POSIX date formatting function> (strftime()) uses C<LC_TIME>.
127 C<LC_COLLATE>, C<LC_CTYPE>, and so on, are discussed further in L<LOCALE
130 The default behavior is restored with the S<C<no locale>> pragma, or
131 upon reaching the end of block enclosing C<use locale>.
133 The string result of any operation that uses locale
134 information is tainted, as it is possible for a locale to be
135 untrustworthy. See L<"SECURITY">.
137 =head2 The setlocale function
139 You can switch locales as often as you wish at run time with the
140 POSIX::setlocale() function:
142 # This functionality not usable prior to Perl 5.004
145 # Import locale-handling tool set from POSIX module.
146 # This example uses: setlocale -- the function call
147 # LC_CTYPE -- explained below
148 use POSIX qw(locale_h);
150 # query and save the old locale
151 $old_locale = setlocale(LC_CTYPE);
153 setlocale(LC_CTYPE, "fr_CA.ISO8859-1");
154 # LC_CTYPE now in locale "French, Canada, codeset ISO 8859-1"
156 setlocale(LC_CTYPE, "");
157 # LC_CTYPE now reset to default defined by LC_ALL/LC_CTYPE/LANG
158 # environment variables. See below for documentation.
160 # restore the old locale
161 setlocale(LC_CTYPE, $old_locale);
163 The first argument of setlocale() gives the B<category>, the second the
164 B<locale>. The category tells in what aspect of data processing you
165 want to apply locale-specific rules. Category names are discussed in
166 L<LOCALE CATEGORIES> and L<"ENVIRONMENT">. The locale is the name of a
167 collection of customization information corresponding to a particular
168 combination of language, country or territory, and codeset. Read on for
169 hints on the naming of locales: not all systems name locales as in the
172 If no second argument is provided, the function returns a string naming
173 the current locale for the category. You can use this value as the
174 second argument in a subsequent call to setlocale(). If a second
175 argument is given and it corresponds to a valid locale, the locale for
176 the category is set to that value, and the function returns the
177 now-current locale value. You can then use this in yet another call to
178 setlocale(). (In some implementations, the return value may sometimes
179 differ from the value you gave as the second argument--think of it as
180 an alias for the value you gave.)
182 As the example shows, if the second argument is an empty string, the
183 category's locale is returned to the default specified by the
184 corresponding environment variables. Generally, this results in a
185 return to the default that was in force when Perl started up: changes
186 to the environment made by the application after startup may or may not
187 be noticed, depending on your system's C library.
189 If the second argument does not correspond to a valid locale, the locale
190 for the category is not changed, and the function returns I<undef>.
192 For further information about the categories, consult L<setlocale(3)>.
194 =head2 Finding locales
196 For locales available in your system, consult also L<setlocale(3)> to
197 see whether it leads to the list of available locales (search for the
198 I<SEE ALSO> section). If that fails, try the following command lines:
210 and see whether they list something resembling these
212 en_US.ISO8859-1 de_DE.ISO8859-1 ru_RU.ISO8859-5
215 english german russian
216 english.iso88591 german.iso88591 russian.iso88595
218 Sadly, even though the calling interface for setlocale() has
219 been standardized, names of locales and the directories where the
220 configuration resides have not been. The basic form of the name is
221 I<language_country/territory>B<.>I<codeset>, but the latter parts after
222 I<language> are not always present. The I<language> and I<country> are
223 usually from the standards B<ISO 3166> and B<ISO 639>, the two-letter
224 abbreviations for the countries and the languages of the world,
225 respectively. The I<codeset> part often mentions some B<ISO 8859>
226 character set, the Latin codesets. For example, C<ISO 8859-1> is the
227 so-called "Western codeset" that can be used to encode most Western
228 European languages. Again, there are several ways to write even the
229 name of that one standard. Lamentably.
231 Two special locales are worth particular mention: "C" and "POSIX".
232 Currently these are effectively the same locale: the difference is
233 mainly that the first one is defined by the C standard, the second by
234 the POSIX standard. They define the B<default locale> in which
235 every program starts in the absence of locale information in its
236 environment. (The I<default> default locale, if you will.) Its language
237 is (American) English and its character codeset ASCII.
239 B<NOTE>: Not all systems have the "POSIX" locale (not all systems are
240 POSIX-conformant), so use "C" when you need explicitly to specify this
243 =head2 LOCALE PROBLEMS
245 You may encounter the following warning message at Perl startup:
247 perl: warning: Setting locale failed.
248 perl: warning: Please check that your locale settings:
251 are supported and installed on your system.
252 perl: warning: Falling back to the standard locale ("C").
254 This means that your locale settings had LC_ALL set to "En_US" and
255 LANG exists but has no value. Perl tried to believe you but could not.
256 Instead, Perl gave up and fell back to the "C" locale, the default locale
257 that is supposed to work no matter what. This usually means your locale
258 settings were wrong, they mention locales your system has never heard
259 of, or the locale installation in your system has problems (for example,
260 some system files are broken or missing). There are quick and temporary
261 fixes to these problems, as well as more thorough and lasting fixes.
263 =head2 Temporarily fixing locale problems
265 The two quickest fixes are either to render Perl silent about any
266 locale inconsistencies or to run Perl under the default locale "C".
268 Perl's moaning about locale problems can be silenced by setting the
269 environment variable PERL_BADLANG to a non-zero value, for example
270 "1". This method really just sweeps the problem under the carpet: you
271 tell Perl to shut up even when Perl sees that something is wrong. Do
272 not be surprised if later something locale-dependent misbehaves.
274 Perl can be run under the "C" locale by setting the environment
275 variable LC_ALL to "C". This method is perhaps a bit more civilized
276 than the PERL_BADLANG approach, but setting LC_ALL (or
277 other locale variables) may affect other programs as well, not just
278 Perl. In particular, external programs run from within Perl will see
279 these changes. If you make the new settings permanent (read on), all
280 programs you run see the changes. See L<ENVIRONMENT> for for
281 the full list of relevant environment variables and L<USING LOCALES>
282 for their effects in Perl. Effects in other programs are
283 easily deducible. For example, the variable LC_COLLATE may well affect
284 your B<sort> program (or whatever the program that arranges `records'
285 alphabetically in your system is called).
287 You can test out changing these variables temporarily, and if the
288 new settings seem to help, put those settings into your shell startup
289 files. Consult your local documentation for the exact details. For in
290 Bourne-like shells (B<sh>, B<ksh>, B<bash>, B<zsh>):
292 LC_ALL=en_US.ISO8859-1
295 This assumes that we saw the locale "en_US.ISO8859-1" using the commands
296 discussed above. We decided to try that instead of the above faulty
297 locale "En_US"--and in Cshish shells (B<csh>, B<tcsh>)
299 setenv LC_ALL en_US.ISO8859-1
301 If you do not know what shell you have, consult your local
302 helpdesk or the equivalent.
304 =head2 Permanently fixing locale problems
306 The slower but superior fixes are when you may be able to yourself
307 fix the misconfiguration of your own environment variables. The
308 mis(sing)configuration of the whole system's locales usually requires
309 the help of your friendly system administrator.
311 First, see earlier in this document about L<Finding locales>. That tells
312 how to find which locales are really supported--and more importantly,
313 installed--on your system. In our example error message, environment
314 variables affecting the locale are listed in the order of decreasing
315 importance (and unset variables do not matter). Therefore, having
316 LC_ALL set to "En_US" must have been the bad choice, as shown by the
317 error message. First try fixing locale settings listed first.
319 Second, if using the listed commands you see something B<exactly>
320 (prefix matches do not count and case usually counts) like "En_US"
321 without the quotes, then you should be okay because you are using a
322 locale name that should be installed and available in your system.
323 In this case, see L<Fixing system locale configuration>.
325 =head2 Permanently fixing your locale configuration
327 This is when you see something like:
329 perl: warning: Please check that your locale settings:
332 are supported and installed on your system.
334 but then cannot see that "En_US" listed by the above-mentioned
335 commands. You may see things like "en_US.ISO8859-1", but that isn't
336 the same. In this case, try running under a locale
337 that you can list and which somehow matches what you tried. The
338 rules for matching locale names are a bit vague because
339 standardization is weak in this area. See again the L<Finding
340 locales> about general rules.
342 =head2 Permanently fixing system locale configuration
344 Contact a system administrator (preferably your own) and report the exact
345 error message you get, and ask them to read this same documentation you
346 are now reading. They should be able to check whether there is something
347 wrong with the locale configuration of the system. The L<Finding locales>
348 section is unfortunately a bit vague about the exact commands and places
349 because these things are not that standardized.
351 =head2 The localeconv function
353 The POSIX::localeconv() function allows you to get particulars of the
354 locale-dependent numeric formatting information specified by the current
355 C<LC_NUMERIC> and C<LC_MONETARY> locales. (If you just want the name of
356 the current locale for a particular category, use POSIX::setlocale()
357 with a single parameter--see L<The setlocale function>.)
359 use POSIX qw(locale_h);
361 # Get a reference to a hash of locale-dependent info
362 $locale_values = localeconv();
364 # Output sorted list of the values
365 for (sort keys %$locale_values) {
366 printf "%-20s = %s\n", $_, $locale_values->{$_}
369 localeconv() takes no arguments, and returns B<a reference to> a hash.
370 The keys of this hash are variable names for formatting, such as
371 C<decimal_point> and C<thousands_sep>. The values are the corresponding,
372 er, values. See L<POSIX (3)/localeconv> for a longer example listing
373 the categories an implementation might be expected to provide; some
374 provide more and others fewer, however. You don't need an explicit C<use
375 locale>, because localeconv() always observes the current locale.
377 Here's a simple-minded example program that rewrites its command-line
378 parameters as integers correctly formatted in the current locale:
380 # See comments in previous example
382 use POSIX qw(locale_h);
384 # Get some of locale's numeric formatting parameters
385 my ($thousands_sep, $grouping) =
386 @{localeconv()}{'thousands_sep', 'grouping'};
388 # Apply defaults if values are missing
389 $thousands_sep = ',' unless $thousands_sep;
390 $grouping = 3 unless $grouping;
392 # Format command line params for current locale
394 $_ = int; # Chop non-integer part
396 s/(\d)(\d{$grouping}($|$thousands_sep))/$1$thousands_sep$2/;
401 =head1 LOCALE CATEGORIES
403 The following subsections describe basic locale categories. Beyond these,
404 some combination categories allow manipulation of more than one
405 basic category at a time. See L<"ENVIRONMENT"> for a discussion of these.
407 =head2 Category LC_COLLATE: Collation
409 In the scope of S<C<use locale>>, Perl looks to the C<LC_COLLATE>
410 environment variable to determine the application's notions on collation
411 (ordering) of characters. For example, 'b' follows 'a' in Latin
412 alphabets, but where do 'E<aacute>' and 'E<aring>' belong? And while
413 'color' follows 'chocolate' in English, what about in Spanish?
415 The following collations all make sense and you may meet any of them
423 Here is a code snippet to tell what alphanumeric
424 characters are in the current locale, in that locale's order:
427 print +(sort grep /\w/, map { chr() } 0..255), "\n";
429 Compare this with the characters that you see and their order if you
430 state explicitly that the locale should be ignored:
433 print +(sort grep /\w/, map { chr() } 0..255), "\n";
435 This machine-native collation (which is what you get unless S<C<use
436 locale>> has appeared earlier in the same block) must be used for
437 sorting raw binary data, whereas the locale-dependent collation of the
438 first example is useful for natural text.
440 As noted in L<USING LOCALES>, C<cmp> compares according to the current
441 collation locale when C<use locale> is in effect, but falls back to a
442 byte-by-byte comparison for strings that the locale says are equal. You
443 can use POSIX::strcoll() if you don't want this fall-back:
445 use POSIX qw(strcoll);
447 !strcoll("space and case ignored", "SpaceAndCaseIgnored");
449 $equal_in_locale will be true if the collation locale specifies a
450 dictionary-like ordering that ignores space characters completely and
453 If you have a single string that you want to check for "equality in
454 locale" against several others, you might think you could gain a little
455 efficiency by using POSIX::strxfrm() in conjunction with C<eq>:
457 use POSIX qw(strxfrm);
458 $xfrm_string = strxfrm("Mixed-case string");
459 print "locale collation ignores spaces\n"
460 if $xfrm_string eq strxfrm("Mixed-casestring");
461 print "locale collation ignores hyphens\n"
462 if $xfrm_string eq strxfrm("Mixedcase string");
463 print "locale collation ignores case\n"
464 if $xfrm_string eq strxfrm("mixed-case string");
466 strxfrm() takes a string and maps it into a transformed string for use
467 in byte-by-byte comparisons against other transformed strings during
468 collation. "Under the hood", locale-affected Perl comparison operators
469 call strxfrm() for both operands, then do a byte-by-byte
470 comparison of the transformed strings. By calling strxfrm() explicitly
471 and using a non locale-affected comparison, the example attempts to save
472 a couple of transformations. But in fact, it doesn't save anything: Perl
473 magic (see L<perlguts/Magic Variables>) creates the transformed version of a
474 string the first time it's needed in a comparison, then keeps this version around
475 in case it's needed again. An example rewritten the easy way with
476 C<cmp> runs just about as fast. It also copes with null characters
477 embedded in strings; if you call strxfrm() directly, it treats the first
478 null it finds as a terminator. don't expect the transformed strings
479 it produces to be portable across systems--or even from one revision
480 of your operating system to the next. In short, don't call strxfrm()
481 directly: let Perl do it for you.
483 Note: C<use locale> isn't shown in some of these examples because it isn't
484 needed: strcoll() and strxfrm() exist only to generate locale-dependent
485 results, and so always obey the current C<LC_COLLATE> locale.
487 =head2 Category LC_CTYPE: Character Types
489 In the scope of S<C<use locale>>, Perl obeys the C<LC_CTYPE> locale
490 setting. This controls the application's notion of which characters are
491 alphabetic. This affects Perl's C<\w> regular expression metanotation,
492 which stands for alphanumeric characters--that is, alphabetic and
493 numeric characters. (Consult L<perlre> for more information about
494 regular expressions.) Thanks to C<LC_CTYPE>, depending on your locale
495 setting, characters like 'E<aelig>', 'E<eth>', 'E<szlig>', and
496 'E<oslash>' may be understood as C<\w> characters.
498 The C<LC_CTYPE> locale also provides the map used in transliterating
499 characters between lower and uppercase. This affects the case-mapping
500 functions--lc(), lcfirst, uc(), and ucfirst(); case-mapping
501 interpolation with C<\l>, C<\L>, C<\u>, or C<\U> in double-quoted strings
502 and C<s///> substitutions; and case-independent regular expression
503 pattern matching using the C<i> modifier.
505 Finally, C<LC_CTYPE> affects the POSIX character-class test
506 functions--isalpha(), islower(), and so on. For example, if you move
507 from the "C" locale to a 7-bit Scandinavian one, you may find--possibly
508 to your surprise--that "|" moves from the ispunct() class to isalpha().
510 B<Note:> A broken or malicious C<LC_CTYPE> locale definition may result
511 in clearly ineligible characters being considered to be alphanumeric by
512 your application. For strict matching of (mundane) letters and
513 digits--for example, in command strings--locale-aware applications
514 should use C<\w> inside a C<no locale> block. See L<"SECURITY">.
516 =head2 Category LC_NUMERIC: Numeric Formatting
518 In the scope of S<C<use locale>>, Perl obeys the C<LC_NUMERIC> locale
519 information, which controls an application's idea of how numbers should
520 be formatted for human readability by the printf(), sprintf(), and
521 write() functions. String-to-numeric conversion by the POSIX::strtod()
522 function is also affected. In most implementations the only effect is to
523 change the character used for the decimal point--perhaps from '.' to ','.
524 These functions aren't aware of such niceties as thousands separation and
525 so on. (See L<The localeconv function> if you care about these things.)
527 Output produced by print() is B<never> affected by the
528 current locale: it is independent of whether C<use locale> or C<no
529 locale> is in effect, and corresponds to what you'd get from printf()
530 in the "C" locale. The same is true for Perl's internal conversions
531 between numeric and string formats:
533 use POSIX qw(strtod);
536 $n = 5/2; # Assign numeric 2.5 to $n
538 $a = " $n"; # Locale-independent conversion to string
540 print "half five is $n\n"; # Locale-independent output
542 printf "half five is %g\n", $n; # Locale-dependent output
544 print "DECIMAL POINT IS COMMA\n"
545 if $n == (strtod("2,5"))[0]; # Locale-dependent conversion
547 =head2 Category LC_MONETARY: Formatting of monetary amounts
549 The C standard defines the C<LC_MONETARY> category, but no function
550 that is affected by its contents. (Those with experience of standards
551 committees will recognize that the working group decided to punt on the
552 issue.) Consequently, Perl takes no notice of it. If you really want
553 to use C<LC_MONETARY>, you can query its contents--see L<The localeconv
554 function>--and use the information that it returns in your application's
555 own formatting of currency amounts. However, you may well find that
556 the information, voluminous and complex though it may be, still does not
557 quite meet your requirements: currency formatting is a hard nut to crack.
561 Output produced by POSIX::strftime(), which builds a formatted
562 human-readable date/time string, is affected by the current C<LC_TIME>
563 locale. Thus, in a French locale, the output produced by the C<%B>
564 format element (full month name) for the first month of the year would
565 be "janvier". Here's how to get a list of long month names in the
568 use POSIX qw(strftime);
570 $long_month_name[$_] =
571 strftime("%B", 0, 0, 0, 1, $_, 96);
574 Note: C<use locale> isn't needed in this example: as a function that
575 exists only to generate locale-dependent results, strftime() always
576 obeys the current C<LC_TIME> locale.
578 =head2 Other categories
580 The remaining locale category, C<LC_MESSAGES> (possibly supplemented
581 by others in particular implementations) is not currently used by
582 Perl--except possibly to affect the behavior of library functions called
583 by extensions outside the standard Perl distribution.
587 Although the main discussion of Perl security issues can be found in
588 L<perlsec>, a discussion of Perl's locale handling would be incomplete
589 if it did not draw your attention to locale-dependent security issues.
590 Locales--particularly on systems that allow unprivileged users to
591 build their own locales--are untrustworthy. A malicious (or just plain
592 broken) locale can make a locale-aware application give unexpected
593 results. Here are a few possibilities:
599 Regular expression checks for safe file names or mail addresses using
600 C<\w> may be spoofed by an C<LC_CTYPE> locale that claims that
601 characters such as "E<gt>" and "|" are alphanumeric.
605 String interpolation with case-mapping, as in, say, C<$dest =
606 "C:\U$name.$ext">, may produce dangerous results if a bogus LC_CTYPE
607 case-mapping table is in effect.
611 If the decimal point character in the C<LC_NUMERIC> locale is
612 surreptitiously changed from a dot to a comma, C<sprintf("%g",
613 0.123456e3)> produces a string result of "123,456". Many people would
614 interpret this as one hundred and twenty-three thousand, four hundred
619 A sneaky C<LC_COLLATE> locale could result in the names of students with
620 "D" grades appearing ahead of those with "A"s.
624 An application that takes the trouble to use information in
625 C<LC_MONETARY> may format debits as if they were credits and vice versa
626 if that locale has been subverted. Or it might make payments in US
627 dollars instead of Hong Kong dollars.
631 The date and day names in dates formatted by strftime() could be
632 manipulated to advantage by a malicious user able to subvert the
633 C<LC_DATE> locale. ("Look--it says I wasn't in the building on
638 Such dangers are not peculiar to the locale system: any aspect of an
639 application's environment which may be modified maliciously presents
640 similar challenges. Similarly, they are not specific to Perl: any
641 programming language that allows you to write programs that take
642 account of their environment exposes you to these issues.
644 Perl cannot protect you from all possibilities shown in the
645 examples--there is no substitute for your own vigilance--but, when
646 C<use locale> is in effect, Perl uses the tainting mechanism (see
647 L<perlsec>) to mark string results that become locale-dependent, and
648 which may be untrustworthy in consequence. Here is a summary of the
649 tainting behavior of operators and functions that may be affected by
654 =item B<Comparison operators> (C<lt>, C<le>, C<ge>, C<gt> and C<cmp>):
656 Scalar true/false (or less/equal/greater) result is never tainted.
658 =item B<Case-mapping interpolation> (with C<\l>, C<\L>, C<\u> or C<\U>)
660 Result string containing interpolated material is tainted if
661 C<use locale> is in effect.
663 =item B<Matching operator> (C<m//>):
665 Scalar true/false result never tainted.
667 Subpatterns, either delivered as a list-context result or as $1 etc.
668 are tainted if C<use locale> is in effect, and the subpattern regular
669 expression contains C<\w> (to match an alphanumeric character), C<\W>
670 (non-alphanumeric character), C<\s> (white-space character), or C<\S>
671 (non white-space character). The matched-pattern variable, $&, $`
672 (pre-match), $' (post-match), and $+ (last match) are also tainted if
673 C<use locale> is in effect and the regular expression contains C<\w>,
674 C<\W>, C<\s>, or C<\S>.
676 =item B<Substitution operator> (C<s///>):
678 Has the same behavior as the match operator. Also, the left
679 operand of C<=~> becomes tainted when C<use locale> in effect
680 if modified as a result of a substitution based on a regular
681 expression match involving C<\w>, C<\W>, C<\s>, or C<\S>; or of
682 case-mapping with C<\l>, C<\L>,C<\u> or C<\U>.
684 =item B<In-memory formatting function> (sprintf()):
686 Result is tainted if "use locale" is in effect.
688 =item B<Output formatting functions> (printf() and write()):
690 Success/failure result is never tainted.
692 =item B<Case-mapping functions> (lc(), lcfirst(), uc(), ucfirst()):
694 Results are tainted if C<use locale> is in effect.
696 =item B<POSIX locale-dependent functions> (localeconv(), strcoll(),
697 strftime(), strxfrm()):
699 Results are never tainted.
701 =item B<POSIX character class tests> (isalnum(), isalpha(), isdigit(),
702 isgraph(), islower(), isprint(), ispunct(), isspace(), isupper(),
705 True/false results are never tainted.
709 Three examples illustrate locale-dependent tainting.
710 The first program, which ignores its locale, won't run: a value taken
711 directly from the command line may not be used to name an output file
712 when taint checks are enabled.
714 #/usr/local/bin/perl -T
715 # Run with taint checking
717 # Command line sanity check omitted...
718 $tainted_output_file = shift;
720 open(F, ">$tainted_output_file")
721 or warn "Open of $untainted_output_file failed: $!\n";
723 The program can be made to run by "laundering" the tainted value through
724 a regular expression: the second example--which still ignores locale
725 information--runs, creating the file named on its command line
728 #/usr/local/bin/perl -T
730 $tainted_output_file = shift;
731 $tainted_output_file =~ m%[\w/]+%;
732 $untainted_output_file = $&;
734 open(F, ">$untainted_output_file")
735 or warn "Open of $untainted_output_file failed: $!\n";
737 Compare this with a similar but locale-aware program:
739 #/usr/local/bin/perl -T
741 $tainted_output_file = shift;
743 $tainted_output_file =~ m%[\w/]+%;
744 $localized_output_file = $&;
746 open(F, ">$localized_output_file")
747 or warn "Open of $localized_output_file failed: $!\n";
749 This third program fails to run because $& is tainted: it is the result
750 of a match involving C<\w> while C<use locale> is in effect.
758 A string that can suppress Perl's warning about failed locale settings
759 at startup. Failure can occur if the locale support in the operating
760 system is lacking (broken) in some way--or if you mistyped the name of
761 a locale when you set up your environment. If this environment variable
762 is absent, or has a value that does not evaluate to integer zero--that
763 is, "0" or ""--Perl will complain about locale setting failures.
765 B<NOTE>: PERL_BADLANG only gives you a way to hide the warning message.
766 The message tells about some problem in your system's locale support,
767 and you should investigate what the problem is.
771 The following environment variables are not specific to Perl: They are
772 part of the standardized (ISO C, XPG4, POSIX 1.c) setlocale() method
773 for controlling an application's opinion on data.
779 C<LC_ALL> is the "override-all" locale environment variable. If
780 set, it overrides all the rest of the locale environment variables.
784 In the absence of C<LC_ALL>, C<LC_CTYPE> chooses the character type
785 locale. In the absence of both C<LC_ALL> and C<LC_CTYPE>, C<LANG>
786 chooses the character type locale.
790 In the absence of C<LC_ALL>, C<LC_COLLATE> chooses the collation
791 (sorting) locale. In the absence of both C<LC_ALL> and C<LC_COLLATE>,
792 C<LANG> chooses the collation locale.
796 In the absence of C<LC_ALL>, C<LC_MONETARY> chooses the monetary
797 formatting locale. In the absence of both C<LC_ALL> and C<LC_MONETARY>,
798 C<LANG> chooses the monetary formatting locale.
802 In the absence of C<LC_ALL>, C<LC_NUMERIC> chooses the numeric format
803 locale. In the absence of both C<LC_ALL> and C<LC_NUMERIC>, C<LANG>
804 chooses the numeric format.
808 In the absence of C<LC_ALL>, C<LC_TIME> chooses the date and time
809 formatting locale. In the absence of both C<LC_ALL> and C<LC_TIME>,
810 C<LANG> chooses the date and time formatting locale.
814 C<LANG> is the "catch-all" locale environment variable. If it is set, it
815 is used as the last resort after the overall C<LC_ALL> and the
816 category-specific C<LC_...>.
822 =head2 Backward compatibility
824 Versions of Perl prior to 5.004 B<mostly> ignored locale information,
825 generally behaving as if something similar to the C<"C"> locale were
826 always in force, even if the program environment suggested otherwise
827 (see L<The setlocale function>). By default, Perl still behaves this
828 way for backward compatibility. If you want a Perl application to pay
829 attention to locale information, you B<must> use the S<C<use locale>>
830 pragma (see L<The use locale Pragma>) to instruct it to do so.
832 Versions of Perl from 5.002 to 5.003 did use the C<LC_CTYPE>
833 information if available; that is, C<\w> did understand what
834 were the letters according to the locale environment variables.
835 The problem was that the user had no control over the feature:
836 if the C library supported locales, Perl used them.
838 =head2 I18N:Collate obsolete
840 In versions of Perl prior to 5.004, per-locale collation was possible
841 using the C<I18N::Collate> library module. This module is now mildly
842 obsolete and should be avoided in new applications. The C<LC_COLLATE>
843 functionality is now integrated into the Perl core language: One can
844 use locale-specific scalar data completely normally with C<use locale>,
845 so there is no longer any need to juggle with the scalar references of
848 =head2 Sort speed and memory use impacts
850 Comparing and sorting by locale is usually slower than the default
851 sorting; slow-downs of two to four times have been observed. It will
852 also consume more memory: once a Perl scalar variable has participated
853 in any string comparison or sorting operation obeying the locale
854 collation rules, it will take 3-15 times more memory than before. (The
855 exact multiplier depends on the string's contents, the operating system
856 and the locale.) These downsides are dictated more by the operating
857 system's implementation of the locale system than by Perl.
859 =head2 write() and LC_NUMERIC
861 Formats are the only part of Perl that unconditionally use information
862 from a program's locale; if a program's environment specifies an
863 LC_NUMERIC locale, it is always used to specify the decimal point
864 character in formatted output. Formatted output cannot be controlled by
865 C<use locale> because the pragma is tied to the block structure of the
866 program, and, for historical reasons, formats exist outside that block
869 =head2 Freely available locale definitions
871 There is a large collection of locale definitions at
872 C<ftp://dkuug.dk/i18n/WG15-collection>. You should be aware that it is
873 unsupported, and is not claimed to be fit for any purpose. If your
874 system allows installation of arbitrary locales, you may find the
875 definitions useful as they are, or as a basis for the development of
880 "Internationalization" is often abbreviated as B<i18n> because its first
881 and last letters are separated by eighteen others. (You may guess why
882 the internalin ... internaliti ... i18n tends to get abbreviated.) In
883 the same way, "localization" is often abbreviated to B<l10n>.
885 =head2 An imperfect standard
887 Internationalization, as defined in the C and POSIX standards, can be
888 criticized as incomplete, ungainly, and having too large a granularity.
889 (Locales apply to a whole process, when it would arguably be more useful
890 to have them apply to a single thread, window group, or whatever.) They
891 also have a tendency, like standards groups, to divide the world into
892 nations, when we all know that the world can equally well be divided
893 into bankers, bikers, gamers, and so on. But, for now, it's the only
894 standard we've got. This may be construed as a bug.
898 =head2 Broken systems
900 In certain systems, the operating system's locale support
901 is broken and cannot be fixed or used by Perl. Such deficiencies can
902 and will result in mysterious hangs and/or Perl core dumps when the
903 C<use locale> is in effect. When confronted with such a system,
904 please report in excruciating detail to <F<perlbug@perl.com>>, and
905 complain to your vendor: bug fixes may exist for these problems
906 in your operating system. Sometimes such bug fixes are called an
907 operating system upgrade.
921 L<POSIX (3)/isprint>,
927 L<POSIX (3)/isupper>,
929 L<POSIX (3)/isxdigit>
931 L<POSIX (3)/localeconv>
933 L<POSIX (3)/setlocale>,
937 L<POSIX (3)/strftime>
945 Jarkko Hietaniemi's original F<perli18n.pod> heavily hacked by Dominic
946 Dunlop, assisted by the perl5-porters. Prose worked over a bit by
949 Last update: Thu Jun 11 08:44:13 MDT 1998