SYNOPSIS

use Lingua::Preferred qw(which_lang acceptable_lang);
my @wanted = qw(en de fr it de_CH);
my @available = qw(fr it de);
my $which = which_lang(\@wanted, \@available);
print "language $which is the best of those available\n";
foreach (qw(en_US fr nl de_DE)) {
print "language $_ is acceptable\n"
if acceptable_lang(\@wanted, $_);
}

DESCRIPTION

Often human-readable information is available in more than one language. Which should you use? This module provides a way for the user to specify possible languages in order of preference, and then to pick the best language of those available. Different 'dialects' given by the 'territory' part of the language specifier (such as en, en_GB, and en_US) are also supported.

The routine "which_lang()" picks the best language from a list of alternatives. The arguments are:

• a reference to a list of preferred languages (first is best). Here, a language is a string like 'en' or 'fr_CA'. ('fr_*' can also be given - see below.) 'C' (named for the Unix 'C' locale) matches any language.
• a reference to non-empty list of available languages. Here, a language can be like 'en', 'en_CA', or "undef" meaning 'unknown'.

The return code is which language to use. This will always be an element of the available languages list.

The cleverness of this module (if you can call it that) comes from inferring implicit language preferences based on the explicit list passed in. For example, if you say that en is acceptable, then en_IE and en_DK will presumably be acceptable too (but not as good as just plain en). If you give your language as en_US, then en is almost as good, with the other dialects of en following soon afterwards.

If there is a tie between two choices, as when two dialects of the same language are available and neither is explicitly preferred, or when none of the available languages appears in the user's list, then the choice appearing earlier in the available list is preferred.

Sometimes, the automatic inferring of related dialects is not what you want, because a language dialect may be very different to the 'main' language, for example Swiss German or some forms of English. For this case, the special form 'XX_*' is available. If you dislike Mexican Spanish (as a completely arbitrary example), then "[ 'es', 'es_*', 'es_MX' ]" would rank this dialect below any other dialect of es (but still acceptable). You don't have to explicitly list every other dialect of Spanish before es_MX.

So for example, supposing @avail contains the languages available:

• You know English and prefer US English:

      $which = which_lang([ 'en_US' ], \@avail);
  

• You know English and German, German/Germany is preferred:

      $which = which_lang([ 'en', 'de_DE' ], \@avail);
  

• You know English and German, but preferably not Swiss German:

      $which = which_lang([ 'en', 'de', 'de_*', 'de_CH' ], \@avail);
  

  Here any dialect of German (eg de_DE, de_AT) is preferable to de_CH.

  Whereas "which_lang()" picks the best language from a list of alternatives, "acceptable_lang()" answers whether a single language is included (explicitly or implicitly) in the list of wanted languages. It adds the implicit dialects in the same way.

AUTHOR

Ed Avis, [email protected]

POD ERRORS

Hey! The above document had some coding errors, which are explained below:

Around line 258:

You forgot a '=back' before '=head1'