1PERLLOCALE(1) Perl Programmers Reference Guide PERLLOCALE(1)
2
6 perllocale - Perl locale handling (internationalization and localiza‐
7 tion)
8
10 Perl supports language-specific notions of data such as "is this a let‐
11 ter", "what is the uppercase equivalent of this letter", and "which of
12 these letters comes first". These are important issues, especially for
13 languages other than English--but also for English: it would be naieve
14 to imagine that "A-Za-z" defines all the "letters" needed to write in
15 English. Perl is also aware that some character other than '.' may be
16 preferred as a decimal point, and that output date representations may
17 be language-specific. The process of making an application take
18 account of its users' preferences in such matters is called interna‐
19 tionalization (often abbreviated as i18n); telling such an application
20 about a particular set of preferences is known as localization (l10n).
21 Perl can understand language-specific data via the standardized (ISO C,
23 XPG4, POSIX 1.c) method called "the locale system". The locale system
24 is controlled per application using one pragma, one function call, and
25 several environment variables.
26 NOTE: This feature is new in Perl 5.004, and does not apply unless an
28 application specifically requests it--see "Backward compatibility".
29 The one exception is that write() now always uses the current locale -
30 see "NOTES".
31
33 If Perl applications are to understand and present your data correctly
34 according a locale of your choice, all of the following must be true:
35 · Your operating system must support the locale system. If it does,
37 you should find that the setlocale() function is a documented part
38 of its C library.
39 · Definitions for locales that you use must be installed. You, or
41 your system administrator, must make sure that this is the case.
42 The available locales, the location in which they are kept, and the
43 manner in which they are installed all vary from system to system.
44 Some systems provide only a few, hard-wired locales and do not
45 allow more to be added. Others allow you to add "canned" locales
46 provided by the system supplier. Still others allow you or the
47 system administrator to define and add arbitrary locales. (You may
48 have to ask your supplier to provide canned locales that are not
49 delivered with your operating system.) Read your system documenta‐
50 tion for further illumination.
51 · Perl must believe that the locale system is supported. If it does,
53 "perl -V:d_setlocale" will say that the value for "d_setlocale" is
54 "define".
55 If you want a Perl application to process and present your data accord‐
57 ing to a particular locale, the application code should include the
58 "use locale" pragma (see "The use locale pragma") where appropriate,
59 and at least one of the following must be true:
60 · The locale-determining environment variables (see "ENVIRONMENT")
62 must be correctly set up at the time the application is started,
63 either by yourself or by whoever set up your system account.
64 · The application must set its own locale using the method described
66 in "The setlocale function".
67
69 The use locale pragma
70 By default, Perl ignores the current locale. The "use locale" pragma
72 tells Perl to use the current locale for some operations:
73 · The comparison operators ("lt", "le", "cmp", "ge", and "gt") and
75 the POSIX string collation functions strcoll() and strxfrm() use
76 "LC_COLLATE". sort() is also affected if used without an explicit
77 comparison function, because it uses "cmp" by default.
78 Note: "eq" and "ne" are unaffected by locale: they always perform a
80 char-by-char comparison of their scalar operands. What's more, if
81 "cmp" finds that its operands are equal according to the collation
82 sequence specified by the current locale, it goes on to perform a
83 char-by-char comparison, and only returns 0 (equal) if the operands
84 are char-for-char identical. If you really want to know whether
85 two strings--which "eq" and "cmp" may consider different--are equal
86 as far as collation in the locale is concerned, see the discussion
87 in "Category LC_COLLATE: Collation".
88 · Regular expressions and case-modification functions (uc(), lc(),
90 ucfirst(), and lcfirst()) use "LC_CTYPE"
91 · The formatting functions (printf(), sprintf() and write()) use
93 "LC_NUMERIC"
94 · The POSIX date formatting function (strftime()) uses "LC_TIME".
96 "LC_COLLATE", "LC_CTYPE", and so on, are discussed further in "LOCALE
98 CATEGORIES".
99 The default behavior is restored with the "no locale" pragma, or upon
101 reaching the end of block enclosing "use locale".
102 The string result of any operation that uses locale information is
104 tainted, as it is possible for a locale to be untrustworthy. See
105 "SECURITY".
106 The setlocale function
108 You can switch locales as often as you wish at run time with the
110 POSIX::setlocale() function:
111 # This functionality not usable prior to Perl 5.004
113 require 5.004;
114 # Import locale-handling tool set from POSIX module.
116 # This example uses: setlocale -- the function call
117 # LC_CTYPE -- explained below
118 use POSIX qw(locale_h);
119 # query and save the old locale
121 $old_locale = setlocale(LC_CTYPE);
122 setlocale(LC_CTYPE, "fr_CA.ISO8859-1");
124 # LC_CTYPE now in locale "French, Canada, codeset ISO 8859-1"
125 setlocale(LC_CTYPE, "");
127 # LC_CTYPE now reset to default defined by LC_ALL/LC_CTYPE/LANG
128 # environment variables. See below for documentation.
129 # restore the old locale
131 setlocale(LC_CTYPE, $old_locale);
132 The first argument of setlocale() gives the category, the second the
134 locale. The category tells in what aspect of data processing you want
135 to apply locale-specific rules. Category names are discussed in
136 "LOCALE CATEGORIES" and "ENVIRONMENT". The locale is the name of a
137 collection of customization information corresponding to a particular
138 combination of language, country or territory, and codeset. Read on
139 for hints on the naming of locales: not all systems name locales as in
140 the example.
141 If no second argument is provided and the category is something else
143 than LC_ALL, the function returns a string naming the current locale
144 for the category. You can use this value as the second argument in a
145 subsequent call to setlocale().
146 If no second argument is provided and the category is LC_ALL, the
148 result is implementation-dependent. It may be a string of concatenated
149 locales names (separator also implementation-dependent) or a single
150 locale name. Please consult your setlocale(3) for details.
151 If a second argument is given and it corresponds to a valid locale, the
153 locale for the category is set to that value, and the function returns
154 the now-current locale value. You can then use this in yet another
155 call to setlocale(). (In some implementations, the return value may
156 sometimes differ from the value you gave as the second argument--think
157 of it as an alias for the value you gave.)
158 As the example shows, if the second argument is an empty string, the
160 category's locale is returned to the default specified by the corre‐
161 sponding environment variables. Generally, this results in a return to
162 the default that was in force when Perl started up: changes to the
163 environment made by the application after startup may or may not be
164 noticed, depending on your system's C library.
165 If the second argument does not correspond to a valid locale, the
167 locale for the category is not changed, and the function returns undef.
168 For further information about the categories, consult setlocale(3).
170 Finding locales
172 For locales available in your system, consult also setlocale(3) to see
174 whether it leads to the list of available locales (search for the SEE
175 ALSO section). If that fails, try the following command lines:
176 locale -a
178 nlsinfo
180 ls /usr/lib/nls/loc
182 ls /usr/lib/locale
184 ls /usr/lib/nls
186 ls /usr/share/locale
188 and see whether they list something resembling these
190 en_US.ISO8859-1 de_DE.ISO8859-1 ru_RU.ISO8859-5
192 en_US.iso88591 de_DE.iso88591 ru_RU.iso88595
193 en_US de_DE ru_RU
194 en de ru
195 english german russian
196 english.iso88591 german.iso88591 russian.iso88595
197 english.roman8 russian.koi8r
198 Sadly, even though the calling interface for setlocale() has been stan‐
200 dardized, names of locales and the directories where the configuration
201 resides have not been. The basic form of the name is language_terri‐
202 tory.codeset, but the latter parts after language are not always
203 present. The language and country are usually from the standards ISO
204 3166 and ISO 639, the two-letter abbreviations for the countries and
205 the languages of the world, respectively. The codeset part often men‐
206 tions some ISO 8859 character set, the Latin codesets. For example,
207 "ISO 8859-1" is the so-called "Western European codeset" that can be
208 used to encode most Western European languages adequately. Again,
209 there are several ways to write even the name of that one standard.
210 Lamentably.
211 Two special locales are worth particular mention: "C" and "POSIX".
213 Currently these are effectively the same locale: the difference is
214 mainly that the first one is defined by the C standard, the second by
215 the POSIX standard. They define the default locale in which every pro‐
216 gram starts in the absence of locale information in its environment.
217 (The default default locale, if you will.) Its language is (American)
218 English and its character codeset ASCII.
219 NOTE: Not all systems have the "POSIX" locale (not all systems are
221 POSIX-conformant), so use "C" when you need explicitly to specify this
222 default locale.
223 LOCALE PROBLEMS
225 You may encounter the following warning message at Perl startup:
227 perl: warning: Setting locale failed.
229 perl: warning: Please check that your locale settings:
230 LC_ALL = "En_US",
231 LANG = (unset)
232 are supported and installed on your system.
233 perl: warning: Falling back to the standard locale ("C").
234 This means that your locale settings had LC_ALL set to "En_US" and LANG
236 exists but has no value. Perl tried to believe you but could not.
237 Instead, Perl gave up and fell back to the "C" locale, the default
238 locale that is supposed to work no matter what. This usually means
239 your locale settings were wrong, they mention locales your system has
240 never heard of, or the locale installation in your system has problems
241 (for example, some system files are broken or missing). There are
242 quick and temporary fixes to these problems, as well as more thorough
243 and lasting fixes.
244 Temporarily fixing locale problems
246 The two quickest fixes are either to render Perl silent about any
248 locale inconsistencies or to run Perl under the default locale "C".
249 Perl's moaning about locale problems can be silenced by setting the
251 environment variable PERL_BADLANG to a zero value, for example "0".
252 This method really just sweeps the problem under the carpet: you tell
253 Perl to shut up even when Perl sees that something is wrong. Do not be
254 surprised if later something locale-dependent misbehaves.
255 Perl can be run under the "C" locale by setting the environment vari‐
257 able LC_ALL to "C". This method is perhaps a bit more civilized than
258 the PERL_BADLANG approach, but setting LC_ALL (or other locale vari‐
259 ables) may affect other programs as well, not just Perl. In particu‐
260 lar, external programs run from within Perl will see these changes. If
261 you make the new settings permanent (read on), all programs you run see
262 the changes. See ENVIRONMENT for the full list of relevant environment
263 variables and "USING LOCALES" for their effects in Perl. Effects in
264 other programs are easily deducible. For example, the variable LC_COL‐
265 LATE may well affect your sort program (or whatever the program that
266 arranges "records" alphabetically in your system is called).
267 You can test out changing these variables temporarily, and if the new
269 settings seem to help, put those settings into your shell startup
270 files. Consult your local documentation for the exact details. For in
271 Bourne-like shells (sh, ksh, bash, zsh):
272 LC_ALL=en_US.ISO8859-1
274 export LC_ALL
275 This assumes that we saw the locale "en_US.ISO8859-1" using the com‐
277 mands discussed above. We decided to try that instead of the above
278 faulty locale "En_US"--and in Cshish shells (csh, tcsh)
279 setenv LC_ALL en_US.ISO8859-1
281 or if you have the "env" application you can do in any shell
283 env LC_ALL=en_US.ISO8859-1 perl ...
285 If you do not know what shell you have, consult your local helpdesk or
287 the equivalent.
288 Permanently fixing locale problems
290 The slower but superior fixes are when you may be able to yourself fix
292 the misconfiguration of your own environment variables. The
293 mis(sing)configuration of the whole system's locales usually requires
294 the help of your friendly system administrator.
295 First, see earlier in this document about "Finding locales". That
297 tells how to find which locales are really supported--and more impor‐
298 tantly, installed--on your system. In our example error message, envi‐
299 ronment variables affecting the locale are listed in the order of
300 decreasing importance (and unset variables do not matter). Therefore,
301 having LC_ALL set to "En_US" must have been the bad choice, as shown by
302 the error message. First try fixing locale settings listed first.
303 Second, if using the listed commands you see something exactly (prefix
305 matches do not count and case usually counts) like "En_US" without the
306 quotes, then you should be okay because you are using a locale name
307 that should be installed and available in your system. In this case,
308 see "Permanently fixing your system's locale configuration".
309 Permanently fixing your system's locale configuration
311 This is when you see something like:
313 perl: warning: Please check that your locale settings:
315 LC_ALL = "En_US",
316 LANG = (unset)
317 are supported and installed on your system.
318 but then cannot see that "En_US" listed by the above-mentioned com‐
320 mands. You may see things like "en_US.ISO8859-1", but that isn't the
321 same. In this case, try running under a locale that you can list and
322 which somehow matches what you tried. The rules for matching locale
323 names are a bit vague because standardization is weak in this area.
324 See again the "Finding locales" about general rules.
325 Fixing system locale configuration
327 Contact a system administrator (preferably your own) and report the
329 exact error message you get, and ask them to read this same documenta‐
330 tion you are now reading. They should be able to check whether there
331 is something wrong with the locale configuration of the system. The
332 "Finding locales" section is unfortunately a bit vague about the exact
333 commands and places because these things are not that standardized.
334 The localeconv function
336 The POSIX::localeconv() function allows you to get particulars of the
338 locale-dependent numeric formatting information specified by the cur‐
339 rent "LC_NUMERIC" and "LC_MONETARY" locales. (If you just want the
340 name of the current locale for a particular category, use POSIX::setlo‐
341 cale() with a single parameter--see "The setlocale function".)
342 use POSIX qw(locale_h);
344 # Get a reference to a hash of locale-dependent info
346 $locale_values = localeconv();
347 # Output sorted list of the values
349 for (sort keys %$locale_values) {
350 printf "%-20s = %s\n", $_, $locale_values->{$_}
351 }
352 localeconv() takes no arguments, and returns a reference to a hash.
354 The keys of this hash are variable names for formatting, such as "deci‐
355 mal_point" and "thousands_sep". The values are the corresponding, er,
356 values. See "localeconv" in POSIX for a longer example listing the
357 categories an implementation might be expected to provide; some provide
358 more and others fewer. You don't need an explicit "use locale",
359 because localeconv() always observes the current locale.
360 Here's a simple-minded example program that rewrites its command-line
362 parameters as integers correctly formatted in the current locale:
363 # See comments in previous example
365 require 5.004;
366 use POSIX qw(locale_h);
367 # Get some of locale's numeric formatting parameters
369 my ($thousands_sep, $grouping) =
370 @{localeconv()}{'thousands_sep', 'grouping'};
371 # Apply defaults if values are missing
373 $thousands_sep = ',' unless $thousands_sep;
374 # grouping and mon_grouping are packed lists
376 # of small integers (characters) telling the
377 # grouping (thousand_seps and mon_thousand_seps
378 # being the group dividers) of numbers and
379 # monetary quantities. The integers' meanings:
380 # 255 means no more grouping, 0 means repeat
381 # the previous grouping, 1-254 means use that
382 # as the current grouping. Grouping goes from
383 # right to left (low to high digits). In the
384 # below we cheat slightly by never using anything
385 # else than the first grouping (whatever that is).
386 if ($grouping) {
387 @grouping = unpack("C*", $grouping);
388 } else {
389 @grouping = (3);
390 }
391 # Format command line params for current locale
393 for (@ARGV) {
394 $_ = int; # Chop non-integer part
395 1 while
396 s/(\d)(\d{$grouping[0]}($⎪$thousands_sep))/$1$thousands_sep$2/;
397 print "$_";
398 }
399 print "\n";
400 I18N::Langinfo
402 Another interface for querying locale-dependent information is the
404 I18N::Langinfo::langinfo() function, available at least in UNIX-like
405 systems and VMS.
406 The following example will import the langinfo() function itself and
408 three constants to be used as arguments to langinfo(): a constant for
409 the abbreviated first day of the week (the numbering starts from Sunday
410 = 1) and two more constants for the affirmative and negative answers
411 for a yes/no question in the current locale.
412 use I18N::Langinfo qw(langinfo ABDAY_1 YESSTR NOSTR);
414 my ($abday_1, $yesstr, $nostr) = map { langinfo } qw(ABDAY_1 YESSTR NOSTR);
416 print "$abday_1? [$yesstr/$nostr] ";
418 In other words, in the "C" (or English) locale the above will probably
420 print something like:
421 Sun? [yes/no]
423 See I18N::Langinfo for more information.
425
427 The following subsections describe basic locale categories. Beyond
428 these, some combination categories allow manipulation of more than one
429 basic category at a time. See "ENVIRONMENT" for a discussion of these.
430 Category LC_COLLATE: Collation
432 In the scope of "use locale", Perl looks to the "LC_COLLATE" environ‐
434 ment variable to determine the application's notions on collation
435 (ordering) of characters. For example, 'b' follows 'a' in Latin alpha‐
436 bets, but where do 'a' and 'aa' belong? And while 'color' follows
437 'chocolate' in English, what about in Spanish?
438 The following collations all make sense and you may meet any of them if
440 you "use locale".
441 A B C D E a b c d e
443 A a B b C c D d E e
444 a A b B c C d D e E
445 a b c d e A B C D E
446 Here is a code snippet to tell what "word" characters are in the cur‐
448 rent locale, in that locale's order:
449 use locale;
451 print +(sort grep /\w/, map { chr } 0..255), "\n";
452 Compare this with the characters that you see and their order if you
454 state explicitly that the locale should be ignored:
455 no locale;
457 print +(sort grep /\w/, map { chr } 0..255), "\n";
458 This machine-native collation (which is what you get unless
460 "use locale" has appeared earlier in the same block) must be used for
461 sorting raw binary data, whereas the locale-dependent collation of the
462 first example is useful for natural text.
463 As noted in "USING LOCALES", "cmp" compares according to the current
465 collation locale when "use locale" is in effect, but falls back to a
466 char-by-char comparison for strings that the locale says are equal. You
467 can use POSIX::strcoll() if you don't want this fall-back:
468 use POSIX qw(strcoll);
470 $equal_in_locale =
471 !strcoll("space and case ignored", "SpaceAndCaseIgnored");
472 $equal_in_locale will be true if the collation locale specifies a dic‐
474 tionary-like ordering that ignores space characters completely and
475 which folds case.
476 If you have a single string that you want to check for "equality in
478 locale" against several others, you might think you could gain a little
479 efficiency by using POSIX::strxfrm() in conjunction with "eq":
480 use POSIX qw(strxfrm);
482 $xfrm_string = strxfrm("Mixed-case string");
483 print "locale collation ignores spaces\n"
484 if $xfrm_string eq strxfrm("Mixed-casestring");
485 print "locale collation ignores hyphens\n"
486 if $xfrm_string eq strxfrm("Mixedcase string");
487 print "locale collation ignores case\n"
488 if $xfrm_string eq strxfrm("mixed-case string");
489 strxfrm() takes a string and maps it into a transformed string for use
491 in char-by-char comparisons against other transformed strings during
492 collation. "Under the hood", locale-affected Perl comparison operators
493 call strxfrm() for both operands, then do a char-by-char comparison of
494 the transformed strings. By calling strxfrm() explicitly and using a
495 non locale-affected comparison, the example attempts to save a couple
496 of transformations. But in fact, it doesn't save anything: Perl magic
497 (see "Magic Variables" in perlguts) creates the transformed version of
498 a string the first time it's needed in a comparison, then keeps this
499 version around in case it's needed again. An example rewritten the
500 easy way with "cmp" runs just about as fast. It also copes with null
501 characters embedded in strings; if you call strxfrm() directly, it
502 treats the first null it finds as a terminator. don't expect the
503 transformed strings it produces to be portable across systems--or even
504 from one revision of your operating system to the next. In short,
505 don't call strxfrm() directly: let Perl do it for you.
506 Note: "use locale" isn't shown in some of these examples because it
508 isn't needed: strcoll() and strxfrm() exist only to generate locale-
509 dependent results, and so always obey the current "LC_COLLATE" locale.
510 Category LC_CTYPE: Character Types
512 In the scope of "use locale", Perl obeys the "LC_CTYPE" locale setting.
514 This controls the application's notion of which characters are alpha‐
515 betic. This affects Perl's "\w" regular expression metanotation, which
516 stands for alphanumeric characters--that is, alphabetic, numeric, and
517 including other special characters such as the underscore or hyphen.
518 (Consult perlre for more information about regular expressions.)
519 Thanks to "LC_CTYPE", depending on your locale setting, characters like
520 'ae', 'd`', 'ss', and 'o' may be understood as "\w" characters.
521 The "LC_CTYPE" locale also provides the map used in transliterating
523 characters between lower and uppercase. This affects the case-mapping
524 functions--lc(), lcfirst, uc(), and ucfirst(); case-mapping interpola‐
525 tion with "\l", "\L", "\u", or "\U" in double-quoted strings and "s///"
526 substitutions; and case-independent regular expression pattern matching
527 using the "i" modifier.
528 Finally, "LC_CTYPE" affects the POSIX character-class test func‐
530 tions--isalpha(), islower(), and so on. For example, if you move from
531 the "C" locale to a 7-bit Scandinavian one, you may find--possibly to
532 your surprise--that "⎪" moves from the ispunct() class to isalpha().
533 Note: A broken or malicious "LC_CTYPE" locale definition may result in
535 clearly ineligible characters being considered to be alphanumeric by
536 your application. For strict matching of (mundane) letters and dig‐
537 its--for example, in command strings--locale-aware applications should
538 use "\w" inside a "no locale" block. See "SECURITY".
539 Category LC_NUMERIC: Numeric Formatting
541 In the scope of "use locale", Perl obeys the "LC_NUMERIC" locale infor‐
543 mation, which controls an application's idea of how numbers should be
544 formatted for human readability by the printf(), sprintf(), and write()
545 functions. String-to-numeric conversion by the POSIX::strtod() func‐
546 tion is also affected. In most implementations the only effect is to
547 change the character used for the decimal point--perhaps from '.' to
548 ','. These functions aren't aware of such niceties as thousands sepa‐
549 ration and so on. (See "The localeconv function" if you care about
550 these things.)
551 Output produced by print() is also affected by the current locale: it
553 depends on whether "use locale" or "no locale" is in effect, and corre‐
554 sponds to what you'd get from printf() in the "C" locale. The same is
555 true for Perl's internal conversions between numeric and string for‐
556 mats:
557 use POSIX qw(strtod);
559 use locale;
560 $n = 5/2; # Assign numeric 2.5 to $n
562 $a = " $n"; # Locale-dependent conversion to string
564 print "half five is $n\n"; # Locale-dependent output
566 printf "half five is %g\n", $n; # Locale-dependent output
568 print "DECIMAL POINT IS COMMA\n"
570 if $n == (strtod("2,5"))[0]; # Locale-dependent conversion
571 See also I18N::Langinfo and "RADIXCHAR".
573 Category LC_MONETARY: Formatting of monetary amounts
575 The C standard defines the "LC_MONETARY" category, but no function that
577 is affected by its contents. (Those with experience of standards com‐
578 mittees will recognize that the working group decided to punt on the
579 issue.) Consequently, Perl takes no notice of it. If you really want
580 to use "LC_MONETARY", you can query its contents--see "The localeconv
581 function"--and use the information that it returns in your applica‐
582 tion's own formatting of currency amounts. However, you may well find
583 that the information, voluminous and complex though it may be, still
584 does not quite meet your requirements: currency formatting is a hard
585 nut to crack.
586 See also I18N::Langinfo and "CRNCYSTR".
588 LC_TIME
590 Output produced by POSIX::strftime(), which builds a formatted human-
592 readable date/time string, is affected by the current "LC_TIME" locale.
593 Thus, in a French locale, the output produced by the %B format element
594 (full month name) for the first month of the year would be "janvier".
595 Here's how to get a list of long month names in the current locale:
596 use POSIX qw(strftime);
598 for (0..11) {
599 $long_month_name[$_] =
600 strftime("%B", 0, 0, 0, 1, $_, 96);
601 }
602 Note: "use locale" isn't needed in this example: as a function that
604 exists only to generate locale-dependent results, strftime() always
605 obeys the current "LC_TIME" locale.
606 See also I18N::Langinfo and "ABDAY_1".."ABDAY_7", "DAY_1".."DAY_7",
608 "ABMON_1".."ABMON_12", and "ABMON_1".."ABMON_12".
609 Other categories
611 The remaining locale category, "LC_MESSAGES" (possibly supplemented by
613 others in particular implementations) is not currently used by
614 Perl--except possibly to affect the behavior of library functions
615 called by extensions outside the standard Perl distribution and by the
616 operating system and its utilities. Note especially that the string
617 value of $! and the error messages given by external utilities may be
618 changed by "LC_MESSAGES". If you want to have portable error codes,
619 use "%!". See Errno.
620
622 Although the main discussion of Perl security issues can be found in
623 perlsec, a discussion of Perl's locale handling would be incomplete if
624 it did not draw your attention to locale-dependent security issues.
625 Locales--particularly on systems that allow unprivileged users to build
626 their own locales--are untrustworthy. A malicious (or just plain bro‐
627 ken) locale can make a locale-aware application give unexpected
628 results. Here are a few possibilities:
629 · Regular expression checks for safe file names or mail addresses
631 using "\w" may be spoofed by an "LC_CTYPE" locale that claims that
632 characters such as ">" and "⎪" are alphanumeric.
633 · String interpolation with case-mapping, as in, say, "$dest =
635 "C:\U$name.$ext"", may produce dangerous results if a bogus
636 LC_CTYPE case-mapping table is in effect.
637 · A sneaky "LC_COLLATE" locale could result in the names of students
639 with "D" grades appearing ahead of those with "A"s.
640 · An application that takes the trouble to use information in
642 "LC_MONETARY" may format debits as if they were credits and vice
643 versa if that locale has been subverted. Or it might make payments
644 in US dollars instead of Hong Kong dollars.
645 · The date and day names in dates formatted by strftime() could be
647 manipulated to advantage by a malicious user able to subvert the
648 "LC_DATE" locale. ("Look--it says I wasn't in the building on Sun‐
649 day.")
650 Such dangers are not peculiar to the locale system: any aspect of an
652 application's environment which may be modified maliciously presents
653 similar challenges. Similarly, they are not specific to Perl: any pro‐
654 gramming language that allows you to write programs that take account
655 of their environment exposes you to these issues.
656 Perl cannot protect you from all possibilities shown in the exam‐
658 ples--there is no substitute for your own vigilance--but, when "use
659 locale" is in effect, Perl uses the tainting mechanism (see perlsec) to
660 mark string results that become locale-dependent, and which may be
661 untrustworthy in consequence. Here is a summary of the tainting behav‐
662 ior of operators and functions that may be affected by the locale:
663 · Comparison operators ("lt", "le", "ge", "gt" and "cmp"):
665 Scalar true/false (or less/equal/greater) result is never tainted.
667 · Case-mapping interpolation (with "\l", "\L", "\u" or "\U")
669 Result string containing interpolated material is tainted if "use
671 locale" is in effect.
672 · Matching operator ("m//"):
674 Scalar true/false result never tainted.
676 Subpatterns, either delivered as a list-context result or as $1
678 etc. are tainted if "use locale" is in effect, and the subpattern
679 regular expression contains "\w" (to match an alphanumeric charac‐
680 ter), "\W" (non-alphanumeric character), "\s" (whitespace charac‐
681 ter), or "\S" (non whitespace character). The matched-pattern
682 variable, $&, $` (pre-match), $' (post-match), and $+ (last match)
683 are also tainted if "use locale" is in effect and the regular
684 expression contains "\w", "\W", "\s", or "\S".
685 · Substitution operator ("s///"):
687 Has the same behavior as the match operator. Also, the left oper‐
689 and of "=~" becomes tainted when "use locale" in effect if modified
690 as a result of a substitution based on a regular expression match
691 involving "\w", "\W", "\s", or "\S"; or of case-mapping with "\l",
692 "\L","\u" or "\U".
693 · Output formatting functions (printf() and write()):
695 Results are never tainted because otherwise even output from print,
697 for example "print(1/7)", should be tainted if "use locale" is in
698 effect.
699 · Case-mapping functions (lc(), lcfirst(), uc(), ucfirst()):
701 Results are tainted if "use locale" is in effect.
703 · POSIX locale-dependent functions (localeconv(), strcoll(), strf‐
705 time(), strxfrm()):
706 Results are never tainted.
708 · POSIX character class tests (isalnum(), isalpha(), isdigit(),
710 isgraph(), islower(), isprint(), ispunct(), isspace(), isupper(),
711 isxdigit()):
712 True/false results are never tainted.
714 Three examples illustrate locale-dependent tainting. The first pro‐
716 gram, which ignores its locale, won't run: a value taken directly from
717 the command line may not be used to name an output file when taint
718 checks are enabled.
719 #/usr/local/bin/perl -T
721 # Run with taint checking
722 # Command line sanity check omitted...
724 $tainted_output_file = shift;
725 open(F, ">$tainted_output_file")
727 or warn "Open of $untainted_output_file failed: $!\n";
728 The program can be made to run by "laundering" the tainted value
730 through a regular expression: the second example--which still ignores
731 locale information--runs, creating the file named on its command line
732 if it can.
733 #/usr/local/bin/perl -T
735 $tainted_output_file = shift;
737 $tainted_output_file =~ m%[\w/]+%;
738 $untainted_output_file = $&;
739 open(F, ">$untainted_output_file")
741 or warn "Open of $untainted_output_file failed: $!\n";
742 Compare this with a similar but locale-aware program:
744 #/usr/local/bin/perl -T
746 $tainted_output_file = shift;
748 use locale;
749 $tainted_output_file =~ m%[\w/]+%;
750 $localized_output_file = $&;
751 open(F, ">$localized_output_file")
753 or warn "Open of $localized_output_file failed: $!\n";
754 This third program fails to run because $& is tainted: it is the result
756 of a match involving "\w" while "use locale" is in effect.
757
759 PERL_BADLANG
760 A string that can suppress Perl's warning about failed
761 locale settings at startup. Failure can occur if the
762 locale support in the operating system is lacking (broken)
763 in some way--or if you mistyped the name of a locale when
764 you set up your environment. If this environment variable
765 is absent, or has a value that does not evaluate to integer
766 zero--that is, "0" or ""-- Perl will complain about locale
767 setting failures.
768 NOTE: PERL_BADLANG only gives you a way to hide the warning
770 message. The message tells about some problem in your sys‐
771 tem's locale support, and you should investigate what the
772 problem is.
773 The following environment variables are not specific to Perl: They are
775 part of the standardized (ISO C, XPG4, POSIX 1.c) setlocale() method
776 for controlling an application's opinion on data.
777 LC_ALL "LC_ALL" is the "override-all" locale environment variable.
779 If set, it overrides all the rest of the locale environment
780 variables.
781 LANGUAGE NOTE: "LANGUAGE" is a GNU extension, it affects you only if
783 you are using the GNU libc. This is the case if you are
784 using e.g. Linux. If you are using "commercial" UNIXes you
785 are most probably not using GNU libc and you can ignore
786 "LANGUAGE".
787 However, in the case you are using "LANGUAGE": it affects
789 the language of informational, warning, and error messages
790 output by commands (in other words, it's like "LC_MES‐
791 SAGES") but it has higher priority than LC_ALL. Moreover,
792 it's not a single value but instead a "path" (":"-separated
793 list) of languages (not locales). See the GNU "gettext"
794 library documentation for more information.
795 LC_CTYPE In the absence of "LC_ALL", "LC_CTYPE" chooses the charac‐
797 ter type locale. In the absence of both "LC_ALL" and
798 "LC_CTYPE", "LANG" chooses the character type locale.
799 LC_COLLATE In the absence of "LC_ALL", "LC_COLLATE" chooses the colla‐
801 tion (sorting) locale. In the absence of both "LC_ALL" and
802 "LC_COLLATE", "LANG" chooses the collation locale.
803 LC_MONETARY In the absence of "LC_ALL", "LC_MONETARY" chooses the mone‐
805 tary formatting locale. In the absence of both "LC_ALL"
806 and "LC_MONETARY", "LANG" chooses the monetary formatting
807 locale.
808 LC_NUMERIC In the absence of "LC_ALL", "LC_NUMERIC" chooses the
810 numeric format locale. In the absence of both "LC_ALL" and
811 "LC_NUMERIC", "LANG" chooses the numeric format.
812 LC_TIME In the absence of "LC_ALL", "LC_TIME" chooses the date and
814 time formatting locale. In the absence of both "LC_ALL"
815 and "LC_TIME", "LANG" chooses the date and time formatting
816 locale.
817 LANG "LANG" is the "catch-all" locale environment variable. If
819 it is set, it is used as the last resort after the overall
820 "LC_ALL" and the category-specific "LC_...".
821
823 Backward compatibility
824 Versions of Perl prior to 5.004 mostly ignored locale information, gen‐
826 erally behaving as if something similar to the "C" locale were always
827 in force, even if the program environment suggested otherwise (see "The
828 setlocale function"). By default, Perl still behaves this way for
829 backward compatibility. If you want a Perl application to pay atten‐
830 tion to locale information, you must use the "use locale" pragma (see
831 "The use locale pragma") to instruct it to do so.
832 Versions of Perl from 5.002 to 5.003 did use the "LC_CTYPE" information
834 if available; that is, "\w" did understand what were the letters
835 according to the locale environment variables. The problem was that
836 the user had no control over the feature: if the C library supported
837 locales, Perl used them.
838 I18N:Collate obsolete
840 In versions of Perl prior to 5.004, per-locale collation was possible
842 using the "I18N::Collate" library module. This module is now mildly
843 obsolete and should be avoided in new applications. The "LC_COLLATE"
844 functionality is now integrated into the Perl core language: One can
845 use locale-specific scalar data completely normally with "use locale",
846 so there is no longer any need to juggle with the scalar references of
847 "I18N::Collate".
848 Sort speed and memory use impacts
850 Comparing and sorting by locale is usually slower than the default
852 sorting; slow-downs of two to four times have been observed. It will
853 also consume more memory: once a Perl scalar variable has participated
854 in any string comparison or sorting operation obeying the locale colla‐
855 tion rules, it will take 3-15 times more memory than before. (The
856 exact multiplier depends on the string's contents, the operating system
857 and the locale.) These downsides are dictated more by the operating
858 system's implementation of the locale system than by Perl.
859 write() and LC_NUMERIC
861 Formats are the only part of Perl that unconditionally use information
863 from a program's locale; if a program's environment specifies an
864 LC_NUMERIC locale, it is always used to specify the decimal point char‐
865 acter in formatted output. Formatted output cannot be controlled by
866 "use locale" because the pragma is tied to the block structure of the
867 program, and, for historical reasons, formats exist outside that block
868 structure.
869 Freely available locale definitions
871 There is a large collection of locale definitions at
873 ftp://dkuug.dk/i18n/WG15-collection . You should be aware that it is
874 unsupported, and is not claimed to be fit for any purpose. If your
875 system allows installation of arbitrary locales, you may find the defi‐
876 nitions useful as they are, or as a basis for the development of your
877 own locales.
878 I18n and l10n
880 "Internationalization" is often abbreviated as i18n because its first
882 and last letters are separated by eighteen others. (You may guess why
883 the internalin ... internaliti ... i18n tends to get abbreviated.) In
884 the same way, "localization" is often abbreviated to l10n.
885 An imperfect standard
887 Internationalization, as defined in the C and POSIX standards, can be
889 criticized as incomplete, ungainly, and having too large a granularity.
890 (Locales apply to a whole process, when it would arguably be more use‐
891 ful to have them apply to a single thread, window group, or whatever.)
892 They also have a tendency, like standards groups, to divide the world
893 into nations, when we all know that the world can equally well be
894 divided into bankers, bikers, gamers, and so on. But, for now, it's
895 the only standard we've got. This may be construed as a bug.
896
898 The support of Unicode is new starting from Perl version 5.6, and more
899 fully implemented in the version 5.8. See perluniintro and perlunicode
900 for more details.
901 Usually locale settings and Unicode do not affect each other, but there
903 are exceptions, see "Locales" in perlunicode for examples.
904
906 Broken systems
907 In certain systems, the operating system's locale support is broken and
909 cannot be fixed or used by Perl. Such deficiencies can and will result
910 in mysterious hangs and/or Perl core dumps when the "use locale" is in
911 effect. When confronted with such a system, please report in excruci‐
912 ating detail to <perlbug@perl.org>, and complain to your vendor: bug
913 fixes may exist for these problems in your operating system. Sometimes
914 such bug fixes are called an operating system upgrade.
915
917 I18N::Langinfo, perluniintro, perlunicode, open, "isalnum" in POSIX,
918 "isalpha" in POSIX, "isdigit" in POSIX, "isgraph" in POSIX, "islower"
919 in POSIX, "isprint" in POSIX, "ispunct" in POSIX, "isspace" in POSIX,
920 "isupper" in POSIX, "isxdigit" in POSIX, "localeconv" in POSIX, "setlo‐
921 cale" in POSIX, "strcoll" in POSIX, "strftime" in POSIX, "strtod" in
922 POSIX, "strxfrm" in POSIX.
923
925 Jarkko Hietaniemi's original perli18n.pod heavily hacked by Dominic
926 Dunlop, assisted by the perl5-porters. Prose worked over a bit by Tom
927 Christiansen.
928 Last update: Thu Jun 11 08:44:13 MDT 1998
930perl v5.8.8 2006-01-07 PERLLOCALE(1)