Merge branch 'master' of ssh://git.debian.org/git/webwml/packages

[deb/packages.git] / lib / Packages / I18N / Languages.pm

# Taken from the webwml CVS tree (english/templates/languages.wml)

package Packages::I18N::Languages;

use strict;
use warnings;

use Exporter;

our @ISA = qw( Exporter );
our @EXPORT = qw( langcmp get_transliteration get_selfname );

# language directory name => ISO 639 two-letter code for the language name
my %langs = (
             english    => "en",
             arabic     => "ar",
             armenian   => "hy",
             bulgarian  => "bg",
             catalan    => "ca",
             chinese    => "zh",
             croatian   => "hr",
             czech      => "cs",
             danish     => "da",
             dutch      => "nl",
             german     => "de",
             greek      => "el",
             esperanto  => "eo",
             spanish    => "es",
             finnish    => "fi",
             french     => "fr",
             hungarian  => "hu",
             indonesian => "id",
             italian    => "it",
             japanese   => "ja",
             korean     => "ko",
             lithuanian => "lt",
             norwegian  => "no",
             persian    => "fa",
             polish     => "pl",
             portuguese => "pt",
             romanian   => "ro",
             russian    => "ru",
             swedish    => "sv",
             slovene    => "sl",
             slovak     => "sk",
             turkish    => "tr",
             );

# language directory name => native name of the language
# non-ASCII letters must be escaped (using entities)!
my %selflang = (
                ar     => 'عربية',
                bg     => 'Български',
                ca     => 'català',
                cs     => 'česky',
                da     => 'dansk',
                de     => 'Deutsch',
                el     => 'Ελληνικά',
                en     => 'English',
                eo     => 'Esperanto',
                es     => 'español',
                fa     => 'فارسی',
                fi     => 'suomi',
                fr     => 'français',
                hu     => 'magyar',
                hr     => 'hrvatski',
                hy     => 'Հայերեն',
                id     => 'Indonesia',
                it     => 'Italiano',
                ja     => '日本語',
                ko     => '한국어',
                lt     => 'Lietuvių',
                nl     => 'Nederlands',
                "no"   => 'norsk (bokmål)',
                pl     => 'polski',
                pt     => 'Português (pt)',
                'pt-pt'=> 'Português (pt)',
                'pt-br'=> 'Português (br)',
                ro     => 'română',
                ru     => 'Русский',
                sk     => 'slovensky',
                sv     => 'svenska',
                'sv-se'=> 'svenska',
                sl     => 'slovenščina',
                tr     => 'Türkçe',
                uk     => 'українська',
                zh     => '中文',
                'zh-cn'=> '中文',
                'zh-hk'=> '正體中文',
                'zh-tw'=> '中文',
                );

# language directory name => Latin transliteration of the language name
# This is used for language names which consist entirely of non-Latin
# characters, to aid those that have browsers which cannot show different
# character sets at once.
my %translit = (
                ar => "Arabiya",
                bg => "Bəlgarski",
                el => "Ellinika",
                fa => "Farsi",
                hy => "hayeren",
                ja => "Nihongo",
                ko => "Hangul", # Not sure. "Hanguk-Mal" (=Spoken Korean)?
                ru => "Russkij",
                uk => "ukrajins'ka",
                zh => "Zhongzu",
                'zh-cn'=> "Zhongzu,简",
                'zh-hk'=> "Zhongzu,HK",
                'zh-tw'=> "Zhongzu,繁",
                );

# second transliteration table, used for languages starting with a latin
# diacritic letter
my %translit2 = (
                 cs    => "cesky",
);

sub langcmp ($$) {
  my ($first, $second) = @_;

  # Handle sorting of non-latin characters
  # If there is a transliteration for this language available, use it
  $first = $translit{$first} if defined $translit{$first};
  $second = $translit{$second} if defined $translit{$second};

  # Then handle special cases (initial latin letters with diacritics)
  $first = $translit2{$first} if defined $translit2{$first};
  $second = $translit2{$second} if defined $translit2{$second};

  # Put remaining entity-only names last in the list
  if (substr($first,0,1) eq '&')
  {
    $first =~ s/^&/ZZZ&/;
  }
  if (substr($second,0,1) eq '&')
  {
    $second =~ s/^&/ZZZ&/;
  }
  #    There seems to be a bug with localization in
  #    Perl 5.005 so we need those extra variables.
  my ($ufirst, $usecond) = (uc($first), uc($second));
  return $ufirst cmp $usecond;
}

sub get_selfname {
    return $selflang{$_[0]} if exists $selflang{$_[0]};
    return undef;
}

sub get_transliteration {
    return $translit{$_[0]} if exists $translit{$_[0]};
    return undef;
}

1;