lib/Packages/I18N/Languages.pm

   1 # $Id$
   2 #
   3 # Taken from the webwml CVS tree (english/templates/languages.wml)
   4
   5 package Packages::I18N::Languages;
   6
   7 use strict;
   8 use warnings;
   9
  10 use Exporter;
  11
  12 our @ISA = qw( Exporter );
  13 our @EXPORT = qw( langcmp get_transliteration get_selfname );
  14
  15 # language directory name => ISO 639 two-letter code for the language name
  16 my %langs = (
  17              english    => "en",
  18              arabic     => "ar",
  19              bulgarian  => "bg",
  20              catalan    => "ca",
  21              czech      => "cs",
  22              danish     => "da",
  23              german     => "de",
  24              greek      => "el",
  25              esperanto  => "eo",
  26              spanish    => "es",
  27              finnish    => "fi",
  28              french     => "fr",
  29              croatian   => "hr",
  30              hungarian  => "hu",
  31              indonesian => "id",
  32              italian    => "it",
  33              japanese   => "ja",
  34              korean     => "ko",
  35              lithuanian => "lt",
  36              dutch      => "nl",
  37              norwegian  => "no",
  38              polish     => "pl",
  39              portuguese => "pt",
  40              romanian   => "ro",
  41              russian    => "ru",
  42              swedish    => "sv",
  43              slovene    => "sl",
  44              slovak     => "sk",
  45              turkish    => "tr",
  46              chinese    => "zh",
  47              );
  48
  49 # language directory name => native name of the language
  50 # non-ASCII letters must be escaped (using entities)!
  51 my %selflang = (
  52                 ar     => '&#1593;&#1585;&#1576;&#1610;&#1577;',
  53                 bg     => '&#1041;&#1098;&#1083;&#1075;&#1072;&#1088;&#1089;&#1082;&#1080;',
  54                 ca     => 'catal&agrave;',
  55                 zh     => '&#20013;&#25991;',
  56                 hr     => 'hrvatski',
  57                 cs     => '&#269;esky',
  58                 da     => 'dansk',
  59                 nl     => 'Nederlands',
  60                 en     => 'English',
  61                 eo     => 'Esperanto',
  62                 fi     => 'suomi',
  63                 fr     => 'fran&ccedil;ais',
  64                 de     => 'Deutsch',
  65                 el     => '&#917;&#955;&#955;&#951;&#957;&#953;&#954;&#940;',
  66                 hu     => 'magyar',
  67                 id     => 'Indonesia',
  68                 it     => 'Italiano',
  69                 ja     => '&#26085;&#26412;&#35486;',
  70                 ko     => '&#54620;&#44397;&#50612;',
  71                 lt     => 'Lietuvi&#371;',
  72                 "no"   => 'norsk&nbsp;(bokm&aring;l)',
  73                 pl     => 'polski',
  74                 pt_PT  => 'Portugu&ecirc;s (pt)',
  75                 pt_BR  => 'Portugu&ecirc;s (br)',
  76                 ro     => 'rom&acirc;n&#259;',
  77                 ru     => '&#1056;&#1091;&#1089;&#1089;&#1082;&#1080;&#1081;',
  78                 es     => 'espa&ntilde;ol',
  79                 sk     => 'slovak', #FIXME
  80                 sv_SE  => 'svenska',
  81                 sl     => 'sloven&#353;&#269;ina',
  82                 tr     => 'T&uuml;rk&ccedil;e',
  83                 uk     => 'ukrainian', #FIXME
  84                 );
  85
  86 # language directory name => Latin transliteration of the language name
  87 # This is used for language names which consist entirely of non-Latin
  88 # characters, to aid those that have browsers which cannot show different
  89 # character sets at once.
  90 my %translit = (
  91                 ar => "Arabiya",
  92                 bg => "B&#601;lgarski",
  93                 zh => "Zhongzu", # Not printed due to Chinese-specific code; kept for sort order
  94                 el => "Ellinika",
  95                 ja => "Nihongo",
  96                 ko => "Hangul", # Not sure. "Hanguk-Mal" (=Spoken Korean)?
  97                 ru => "Russkij",
  98                 );
  99
 100 # second transliteration table, used for languages starting with a latin
 101 # diacritic letter
 102 my %translit2 = (
 103                  cs    => "cesky",
 104 );
 105
 106 sub langcmp ($$) {
 107   my ($first, $second) = @_;
 108
 109   # Handle sorting of non-latin characters
 110   # If there is a transliteration for this language available, use it
 111   $first = $translit{$first} if defined $translit{$first};
 112   $second = $translit{$second} if defined $translit{$second};
 113
 114   # Then handle special cases (initial latin letters with diacritics)
 115   $first = $translit2{$first} if defined $translit2{$first};
 116   $second = $translit2{$second} if defined $translit2{$second};
 117
 118   # Put remaining entity-only names last in the list
 119   if (substr($first,0,1) eq '&')
 120   {
 121     $first =~ s/^&/ZZZ&/;
 122   }
 123   if (substr($second,0,1) eq '&')
 124   {
 125     $second =~ s/^&/ZZZ&/;
 126   }
 127   #    There seems to be a bug with localization in
 128   #    Perl 5.005 so we need those extra variables.
 129   my ($ufirst, $usecond) = (uc($first), uc($second));
 130   return $ufirst cmp $usecond;
 131 }
 132
 133 sub get_selfname {
 134     return $selflang{$_[0]} if exists $selflang{$_[0]};
 135     return undef;
 136 }
 137
 138 sub get_transliteration {
 139     return $translit{$_[0]} if exists $translit{$_[0]};
 140     return undef;
 141 }
 142
 143 1;