I18N::Langinfo(3perl)

✨ NAME ✨

I18N::Langinfo — it's the module that lets you query locale information, bestie 💖 (this is the I18N::Langinfo(3perl) page from the Perl Programmers Reference Guide, and honestly? iconic.)

💫 SYNOPSIS 💫

use I18N::Langinfo;

🌸 DESCRIPTION 🌸

Okay so. The langinfo() function goes and asks about all kinds of locale information — the stuff you'd use to localize output and user interfaces so everything doesn't look aggressively American. And here's the tea: it uses the current underlying locale, no matter whether or not you called it from inside the scope of "use locale". It said what it said 🙈

langinfo() needs one numeric argument that identifies which locale constant you wanna query. And if you hand it no argument at all? It just uses $_ like a chill little default. The numeric constants that are appropriate to pass as arguments are exportable from I18N::Langinfo.

So this example right here imports the langinfo() function itself plus three constants to use as arguments to langinfo(): a constant for the abbreviated first day of the week (and heads up, the numbering starts from Sunday = 1) and then two more constants for the affirmative and negative answers to a yes/no question in the current locale. 🗓️

use I18N::Langinfo qw(langinfo ABDAY_1 YESSTR NOSTR);

my ($abday_1, $yesstr, $nostr) =
    map { langinfo($_) } (ABDAY_1, YESSTR, NOSTR);

print "$abday_1? [$yesstr/$nostr] ";

In other words bestie, in the "C" (or English) locale the above will probably print something like:

Sun? [yes/no]

but under a French locale it's serving

dim? [oui/non]

The usually available constants are as follows, and yes I'm listing every single one, don't test me 💅

---

• For abbreviated and full length days of the week and months of the year:

ABDAY_1 ABDAY_2 ABDAY_3 ABDAY_4 ABDAY_5 ABDAY_6 ABDAY_7
ABMON_1 ABMON_2 ABMON_3 ABMON_4 ABMON_5 ABMON_6
ABMON_7 ABMON_8 ABMON_9 ABMON_10 ABMON_11 ABMON_12
DAY_1 DAY_2 DAY_3 DAY_4 DAY_5 DAY_6 DAY_7
MON_1 MON_2 MON_3 MON_4 MON_5 MON_6
MON_7 MON_8 MON_9 MON_10 MON_11 MON_12

• For the date-time, date, and time formats used by the strftime() function (see POSIX):

D_T_FMT D_FMT T_FMT

• For the locales where it actually makes sense to have ante meridiem and post meridiem time formats:

AM_STR PM_STR T_FMT_AMPM

• For the character code set being used (the vibes here are "ISO8859-1", "cp850", "koi8-r", "sjis", "utf8", etc.):

CODESET

• For the symbol or string of characters that says "hey, this is money":

CRNCYSTR

An example is the dollar sign "$". Some locales that aren't tied to a particular location might have an empty currency string. (The C locale is literally one of those.) Otherwise, the return of this is always prefixed by one of these three characters:

• "-" means that in this locale, the string precedes the numeric value, like a U.S. locale would do it: "$9.95".
• "+" means that in this locale, the string follows the numeric value, like "9.95USD".
• "." means that in this locale, the string replaces the radix character, like "9$95".
• For the radix character used between the integer and the fractional part of decimal numbers, and the group separator string for large-ish floating point numbers (yes, these are redundant with POSIX::localeconv()):

RADIXCHAR THOUSEP

• For any alternate digits used in this locale besides the standard 0..9:

ALT_DIGITS

This one returns a sequence of alternate numeric representations for the numbers 0 ... up to 99. The representations come back in a single string, with a semi-colon ";" used to separate the individual ones.

Most locales don't have alternate digits at all, so the string will just be empty. 🫥

To access this data conveniently, you could do something like this:

use I18N::Langinfo qw(langinfo ALT_DIGITS);
my @alt_digits = split ';', langinfo(ALT_DIGITS);

The array @alt_digits will contain 0 elements if the current locale doesn't have alternate digits specified for it. Otherwise, it'll have as many elements as the locale defines, with "[0]" holding the alternate digit for zero; "[1]" for one; and so forth, all the way up to potentially "[99]" for the alternate representation of ninety-nine. 🔢

Be aware (fr, be aware) that the alternate representation in some locales for the numbers 0..9 will have a leading alternate-zero, so it would look like the equivalent of 00..09.

Running this program

use I18N::Langinfo qw(langinfo ALT_DIGITS);
my @alt_digits = split ';', langinfo(ALT_DIGITS);
splice @alt_digits, 15;
print join " ", @alt_digits, "\n";

on a Japanese locale yields

"〇 一 二 三 四 五 六 七 八 九 十 十一 十二 十三 十四"

on some platforms. ✨

• For the affirmative and negative responses and expressions:

YESSTR YESEXPR NOSTR NOEXPR

• For the eras based on typically some ruler, such as the Japanese Emperor (naturally only defined in the appropriate locales, bestie):

ERA ERA_D_FMT ERA_D_T_FMT ERA_T_FMT

---

In addition, Linux boxes have extra items, listed as follows. (And when you call these from other platform types, they return a stub value that's honestly not much use. 💀)

• _NL_ADDRESS_POSTAL_FMT, _NL_ADDRESS_COUNTRY_NAME, _NL_ADDRESS_COUNTRY_POST, _NL_ADDRESS_COUNTRY_AB2, _NL_ADDRESS_COUNTRY_AB3, _NL_ADDRESS_COUNTRY_CAR, _NL_ADDRESS_COUNTRY_NUM, _NL_ADDRESS_COUNTRY_ISBN, _NL_ADDRESS_LANG_NAME, _NL_ADDRESS_LANG_AB, _NL_ADDRESS_LANG_TERM, _NL_ADDRESS_LANG_LIB — on Linux boxes, these return information about the country for the current locale. Further information lives in langinfo.h.
• _NL_IDENTIFICATION_TITLE, _NL_IDENTIFICATION_SOURCE, _NL_IDENTIFICATION_ADDRESS, _NL_IDENTIFICATION_CONTACT, _NL_IDENTIFICATION_EMAIL, _NL_IDENTIFICATION_TEL, _NL_IDENTIFICATION_FAX, _NL_IDENTIFICATION_LANGUAGE, _NL_IDENTIFICATION_TERRITORY, _NL_IDENTIFICATION_AUDIENCE, _NL_IDENTIFICATION_APPLICATION, _NL_IDENTIFICATION_ABBREVIATION, _NL_IDENTIFICATION_REVISION, _NL_IDENTIFICATION_DATE, _NL_IDENTIFICATION_CATEGORY — on Linux boxes, these return meta information about the current locale, like how to get in touch with its maintainers. Further information lives in langinfo.h.
• _NL_MEASUREMENT_MEASUREMENT — on Linux boxes, it returns 1 if the metric system of measurement prevails in the locale; or 2 if US customary units prevail.
• _NL_NAME_NAME_FMT, _NL_NAME_NAME_GEN, _NL_NAME_NAME_MR, _NL_NAME_NAME_MRS, _NL_NAME_NAME_MISS, _NL_NAME_NAME_MS — on Linux boxes, these return information about how names are formatted and the personal salutations used in the current locale. Further information lives in locale(7) and langinfo.h.
• _NL_PAPER_HEIGHT, _NL_PAPER_WIDTH — on Linux boxes, these return the standard size of sheets of paper (in millimeters) in the current locale. 📄
• _NL_TELEPHONE_TEL_INT_FMT, _NL_TELEPHONE_TEL_DOM_FMT, _NL_TELEPHONE_INT_SELECT, _NL_TELEPHONE_INT_PREFIX — on Linux boxes, these return information about how telephone numbers are formatted (both domestic and international calling) in the current locale. Further information lives in langinfo.h.

See your nl_langinfo(3) for more info about the available constants. (And often that means you're just gonna have to look directly at the langinfo.h C header file. I know. I know. 😭)

🧚 For systems without nl_langinfo 🧚

This module originally was just a wrapper for the libc nl_langinfo function, and it did not work on systems lacking it — like Windows. Sad. But starting in Perl 5.28, this module works on all platforms. 🎉

When nl_langinfo isn't available, it uses various methods to construct what that function, if present, would return. But there are potential glitches, bestie. These are the items that could be different:

• ERA — unimplemented, so it returns "".
• CODESET — this should work properly for Windows platforms. On almost all other modern platforms, it will reliably return "UTF-8" if that is the code set. Otherwise, it depends on the locale's name. If that's of the form foo.bar, it will assume bar is the code set; and it also knows about the two locales C and POSIX. If none of those apply, it returns "".
• YESEXPR, YESSTR, NOEXPR, NOSTR — only the values for English are returned. YESSTR and NOSTR have been removed from POSIX 2008, and are retained here for backwards compatibility. Your platform's nl_langinfo may not support them.
• ALT_DIGITS — on systems with a strftime(3) that recognizes the POSIX-defined %O format modifier (not Windows), perl tries hard to return these. The result likely will go as high as what nl_langinfo() would return, but not necessarily; and the numbers from 0..9 will always be stripped of leading zeros. Without %O, an empty string is always returned.
• D_FMT — always evaluates to %x, the locale's appropriate date representation.
• T_FMT — always evaluates to %X, the locale's appropriate time representation.
• D_T_FMT — always evaluates to %c, the locale's appropriate date and time representation.
• CRNCYSTR — the return may be incorrect for those rare locales where the currency symbol replaces the radix character. If you have examples of it needing to work differently, please file a report at https://github.com/Perl/perl5/issues.
• ERA_D_FMT, ERA_T_FMT, ERA_D_T_FMT, T_FMT_AMPM — these are derived by using strftime(), and not all versions of that function know about them. "" is returned for these on such systems.
• All _NL_foo items — these return the same values as they do on boxes that don't have the appropriate underlying locale categories.

🫶 EXPORT 🫶

By default, only the langinfo() function is exported.

💀 BUGS 💀

Before Perl 5.28, the returned values are unreliable for the "RADIXCHAR" and "THOUSEP" locale constants. Not me being dramatic, that's just literally true.

Starting in 5.28, changing locales on threaded builds is supported on systems that offer thread-safe locale functions. These include POSIX 2008 systems and Windows starting with Visual Studio 2005, and this module will work properly in such situations. slay

However — and this is the important part — on threaded builds on Windows prior to Visual Studio 2015, retrieving the items "CRNCYSTR" and "THOUSEP" can result in a race with a thread that has converted to use the global locale. It is quite uncommon for a thread to have done this. It would be possible to construct a workaround for this; patches welcome: see "switch_to_global_locale" in perlapi.

🔥 SEE ALSO 🔥

perllocale, "localeconv" in POSIX, "setlocale" in POSIX, nl_langinfo(3).

👑 AUTHOR 👑

So who do we have to thank for this whole masterpiece? None other than Jarkko Hietaniemi, reachable at jhi@hut.fi. Iconic behavior honestly. 🫶

And these days? This baby is maintained by the Perl 5 porters — the whole group chat, keeping it alive and thriving. No cap. 💅

---

💖 COPYRIGHT AND LICENSE 💖

Okay bestie, legal tea time, but make it cute: ☕✨

Copyright 2001 by Jarkko Hietaniemi

And the vibe here is free software, as it should be:

This library is free software; you can redistribute it and/or modify it under the same terms as Perl itself.

That's it, that's the whole license. Redistribute it, remix it, pass it around — same terms as Perl itself, periodt. 🎀

---

And that's a wrap on the whole page, bestie!! 🌸 Go forth and nl_langinfo responsibly, okay? 💖

Small print for the girlies who like receipts: this rendering goes out with perl v5.44.0, dated 2026-10-17, and lives at I18N::Langinfo(3perl). 🦋 Mwah, love u, bye!! 💅✨