C4/Languages.pm

   1 package C4::Languages;
   2
   3 # Copyright 2006 (C) LibLime
   4 # Joshua Ferraro <jmf@liblime.com>
   5 # Portions Copyright 2009 Chris Cormack and the Koha Dev Team
   6 # This file is part of Koha.
   7 #
   8 # Koha is free software; you can redistribute it and/or modify it under the
   9 # terms of the GNU General Public License as published by the Free Software
  10 # Foundation; either version 2 of the License, or (at your option) any later
  11 # version.
  12 #
  13 # Koha is distributed in the hope that it will be useful, but WITHOUT ANY
  14 # WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
  15 # A PARTICULAR PURPOSE.  See the GNU General Public License for more details.
  16 #
  17 # You should have received a copy of the GNU General Public License along
  18 # with Koha; if not, write to the Free Software Foundation, Inc.,
  19 # 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
  20
  21
  22 use strict;
  23 #use warnings; FIXME - Bug 2505
  24 use Carp;
  25 use C4::Context;
  26 use vars qw($VERSION @ISA @EXPORT @EXPORT_OK %EXPORT_TAGS $DEBUG);
  27
  28 eval {
  29     if (C4::Context->ismemcached) {
  30         require Memoize::Memcached;
  31         import Memoize::Memcached qw(memoize_memcached);
  32
  33         memoize_memcached('getTranslatedLanguages', memcached => C4::Context->memcached);
  34         memoize_memcached('getFrameworkLanguages' , memcached => C4::Context->memcached);
  35         memoize_memcached('getAllLanguages',        memcached => C4::Context->memcached);
  36     }
  37 };
  38
  39 BEGIN {
  40     $VERSION = 3.00;
  41     require Exporter;
  42     @ISA    = qw(Exporter);
  43     @EXPORT = qw(
  44         &getFrameworkLanguages
  45         &getTranslatedLanguages
  46         &getAllLanguages
  47     );
  48     @EXPORT_OK = qw(getFrameworkLanguages getTranslatedLanguages getAllLanguages get_bidi regex_lang_subtags language_get_description accept_language);
  49     $DEBUG = 0;
  50 }
  51
  52 =head1 NAME
  53
  54 C4::Languages - Perl Module containing language list functions for Koha
  55
  56 =head1 SYNOPSIS
  57
  58 use C4::Languages;
  59
  60 =head1 DESCRIPTION
  61
  62 =cut
  63
  64 =head1 FUNCTIONS
  65
  66 =head2 getFrameworkLanguages
  67
  68 Returns a reference to an array of hashes:
  69
  70  my $languages = getFrameworkLanguages();
  71  for my $language(@$languages) {
  72     print "$language->{language_code}\n"; # language code in iso 639-2
  73     print "$language->{language_name}\n"; # language name in native script
  74     print "$language->{language_locale_name}\n"; # language name in current locale
  75  }
  76
  77 =cut
  78
  79 sub getFrameworkLanguages {
  80     # get a hash with all language codes, names, and locale names
  81     my $all_languages = getAllLanguages();
  82     my @languages;
  83
  84     # find the available directory names
  85     my $dir=C4::Context->config('intranetdir')."/installer/data/";
  86     opendir (MYDIR,$dir);
  87     my @listdir= grep { !/^\.|CVS/ && -d "$dir/$_"} readdir(MYDIR);
  88     closedir MYDIR;
  89
  90     # pull out all data for the dir names that exist
  91     for my $dirname (@listdir) {
  92         for my $language_set (@$all_languages) {
  93
  94             if ($dirname eq $language_set->{language_code}) {
  95                 push @languages, {
  96                     'language_code'=>$dirname,
  97                     'language_description'=>$language_set->{language_description},
  98                     'native_descrition'=>$language_set->{language_native_description} }
  99             }
 100         }
 101     }
 102     return \@languages;
 103 }
 104
 105 =head2 getTranslatedLanguages
 106
 107 Returns a reference to an array of hashes:
 108
 109  my $languages = getTranslatedLanguages();
 110  print "Available translated languages:\n";
 111  for my $language(@$trlanguages) {
 112     print "$language->{language_code}\n"; # language code in iso 639-2
 113     print "$language->{language_name}\n"; # language name in native script
 114     print "$language->{language_locale_name}\n"; # language name in current locale
 115  }
 116
 117 =cut
 118
 119 sub getTranslatedLanguages {
 120     my ($interface, $theme, $current_language, $which) = @_;
 121     my $htdocs;
 122     my $all_languages = getAllLanguages();
 123     my @languages;
 124     my @enabled_languages;
 125
 126     if ($interface && $interface eq 'opac' ) {
 127         @enabled_languages = split ",", C4::Context->preference('opaclanguages');
 128         $htdocs = C4::Context->config('opachtdocs');
 129         if ( $theme and -d "$htdocs/$theme" ) {
 130             (@languages) = _get_language_dirs($htdocs,$theme);
 131         }
 132         else {
 133             for my $theme ( _get_themes('opac') ) {
 134                 push @languages, _get_language_dirs($htdocs,$theme);
 135             }
 136         }
 137     }
 138     elsif ($interface && $interface eq 'intranet' ) {
 139         @enabled_languages = split ",", C4::Context->preference('language');
 140         $htdocs = C4::Context->config('intrahtdocs');
 141         if ( $theme and -d "$htdocs/$theme" ) {
 142             @languages = _get_language_dirs($htdocs,$theme);
 143         }
 144         else {
 145             foreach my $theme ( _get_themes('intranet') ) {
 146                 push @languages, _get_language_dirs($htdocs,$theme);
 147             }
 148         }
 149     }
 150     else {
 151         @enabled_languages = split ",", C4::Context->preference('opaclanguages');
 152         my $htdocs = C4::Context->config('intrahtdocs');
 153         foreach my $theme ( _get_themes('intranet') ) {
 154             push @languages, _get_language_dirs($htdocs,$theme);
 155         }
 156         $htdocs = C4::Context->config('opachtdocs');
 157         foreach my $theme ( _get_themes('opac') ) {
 158             push @languages, _get_language_dirs($htdocs,$theme);
 159         }
 160         my %seen;
 161         $seen{$_}++ for @languages;
 162         @languages = keys %seen;
 163     }
 164     return _build_languages_arrayref($all_languages,\@languages,$current_language,\@enabled_languages);
 165 }
 166
 167 =head2 getAllLanguages
 168
 169 Returns a reference to an array of hashes:
 170
 171  my $alllanguages = getAllLanguages();
 172  print "Available translated languages:\n";
 173  for my $language(@$alllanguages) {
 174     print "$language->{language_code}\n";
 175     print "$language->{language_name}\n";
 176     print "$language->{language_locale_name}\n";
 177  }
 178
 179 =cut
 180
 181 sub getAllLanguages {
 182     my @languages_loop;
 183     my $dbh=C4::Context->dbh;
 184     my $current_language = shift || 'en';
 185     my $sth = $dbh->prepare('SELECT * FROM language_subtag_registry WHERE type=\'language\'');
 186     $sth->execute();
 187     while (my $language_subtag_registry = $sth->fetchrow_hashref) {
 188
 189         # pull out all the script descriptions for each language
 190         my $sth2= $dbh->prepare("SELECT * FROM language_descriptions LEFT JOIN language_rfc4646_to_iso639 on language_rfc4646_to_iso639.rfc4646_subtag = language_descriptions.subtag WHERE type='language' AND subtag =? AND language_descriptions.lang = ?");
 191         $sth2->execute($language_subtag_registry->{subtag},$current_language);
 192
 193         my $sth3 = $dbh->prepare("SELECT description FROM language_descriptions WHERE type='language' AND subtag=? AND lang=?");
 194
 195         # add the correct description info
 196         while (my $language_descriptions = $sth2->fetchrow_hashref) {
 197             $sth3->execute($language_subtag_registry->{subtag},$language_subtag_registry->{subtag});
 198             my $native_description;
 199             while (my $description = $sth3->fetchrow_hashref) {
 200                 $native_description = $description->{description};
 201             }
 202
 203             # fill in the ISO6329 code
 204             $language_subtag_registry->{iso639_2_code} = $language_descriptions->{iso639_2_code};
 205             # fill in the native description of the language, as well as the current language's translation of that if it exists
 206             if ($native_description) {
 207                 $language_subtag_registry->{language_description} = $native_description;
 208                 $language_subtag_registry->{language_description}.=" ($language_descriptions->{description})" if $language_descriptions->{description};
 209             }
 210             else {
 211                 $language_subtag_registry->{language_description} = $language_descriptions->{description};
 212             }
 213         }
 214         push @languages_loop, $language_subtag_registry;
 215     }
 216     return \@languages_loop;
 217 }
 218
 219 =head2 _get_themes
 220
 221 Internal function, returns an array of all available themes.
 222
 223   (@themes) = &_get_themes('opac');
 224   (@themes) = &_get_themes('intranet');
 225
 226 =cut
 227
 228 sub _get_themes {
 229     my $interface = shift;
 230     my $htdocs;
 231     my @themes;
 232     if ( $interface eq 'intranet' ) {
 233         $htdocs = C4::Context->config('intrahtdocs');
 234     }
 235     else {
 236         $htdocs = C4::Context->config('opachtdocs');
 237     }
 238     opendir D, "$htdocs";
 239     my @dirlist = readdir D;
 240     foreach my $directory (@dirlist) {
 241         # if there's an en dir, it's a valid theme
 242         -d "$htdocs/$directory/en" and push @themes, $directory;
 243     }
 244     return @themes;
 245 }
 246
 247 =head2 _get_language_dirs
 248
 249 Internal function, returns an array of directory names, excluding non-language directories
 250
 251 =cut
 252
 253 sub _get_language_dirs {
 254     my ($htdocs,$theme) = @_;
 255     my @lang_strings;
 256     opendir D, "$htdocs/$theme";
 257     for my $lang_string ( readdir D ) {
 258         next if $lang_string =~/^\./;
 259         next if $lang_string eq 'all';
 260         next if $lang_string =~/png$/;
 261         next if $lang_string =~/css$/;
 262         next if $lang_string =~/CVS$/;
 263         next if $lang_string =~/\.txt$/i;     #Don't read the readme.txt !
 264         next if $lang_string =~/img|images|famfam|sound|pdf/;
 265         push @lang_strings, $lang_string;
 266     }
 267         return (@lang_strings);
 268 }
 269
 270 =head2 _build_languages_arrayref
 271
 272 Internal function for building the ref to array of hashes
 273
 274 FIXME: this could be rewritten and simplified using map
 275
 276 =cut
 277
 278 sub _build_languages_arrayref {
 279         my ($all_languages,$translated_languages,$current_language,$enabled_languages) = @_;
 280         my @translated_languages = @$translated_languages;
 281         my @languages_loop; # the final reference to an array of hashrefs
 282         my @enabled_languages = @$enabled_languages;
 283         # how many languages are enabled, if one, take note, some contexts won't need to display it
 284         my %seen_languages; # the language tags we've seen
 285         my %found_languages;
 286         my $language_groups;
 287         my $track_language_groups;
 288         my $current_language_regex = regex_lang_subtags($current_language);
 289         # Loop through the translated languages
 290         for my $translated_language (@translated_languages) {
 291             # separate the language string into its subtag types
 292             my $language_subtags_hashref = regex_lang_subtags($translated_language);
 293
 294             # is this language string 'enabled'?
 295             for my $enabled_language (@enabled_languages) {
 296                 #warn "Checking out if $translated_language eq $enabled_language";
 297                 $language_subtags_hashref->{'enabled'} = 1 if $translated_language eq $enabled_language;
 298             }
 299
 300             # group this language, key by langtag
 301             $language_subtags_hashref->{'sublanguage_current'} = 1 if $translated_language eq $current_language;
 302             $language_subtags_hashref->{'rfc4646_subtag'} = $translated_language;
 303             $language_subtags_hashref->{'native_description'} = language_get_description($language_subtags_hashref->{language},$language_subtags_hashref->{language},'language');
 304             $language_subtags_hashref->{'script_description'} = language_get_description($language_subtags_hashref->{script},$language_subtags_hashref->{'language'},'script');
 305             $language_subtags_hashref->{'region_description'} = language_get_description($language_subtags_hashref->{region},$language_subtags_hashref->{'language'},'region');
 306             $language_subtags_hashref->{'variant_description'} = language_get_description($language_subtags_hashref->{variant},$language_subtags_hashref->{'language'},'variant');
 307             $track_language_groups->{$language_subtags_hashref->{'language'}}++;
 308             push ( @{ $language_groups->{$language_subtags_hashref->{language}} }, $language_subtags_hashref );
 309         }
 310         # $key is a language subtag like 'en'
 311         while( my ($key, $value) = each %$language_groups) {
 312
 313             # is this language group enabled? are any of the languages within it enabled?
 314             my $enabled;
 315             for my $enabled_language (@enabled_languages) {
 316                 my $regex_enabled_language = regex_lang_subtags($enabled_language);
 317                 $enabled = 1 if $key eq $regex_enabled_language->{language};
 318             }
 319             push @languages_loop,  {
 320                             # this is only use if there is one
 321                             rfc4646_subtag => @$value[0]->{rfc4646_subtag},
 322                             native_description => language_get_description($key,$key,'language'),
 323                             language => $key,
 324                             sublanguages_loop => $value,
 325                             plural => $track_language_groups->{$key} >1 ? 1 : 0,
 326                             current => $current_language_regex->{language} eq $key ? 1 : 0,
 327                             group_enabled => $enabled,
 328                            };
 329         }
 330         return \@languages_loop;
 331 }
 332
 333 sub language_get_description {
 334     my ($script,$lang,$type) = @_;
 335     my $dbh = C4::Context->dbh;
 336     my $desc;
 337     my $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
 338     #warn "QUERY: SELECT description FROM language_descriptions WHERE subtag=$script AND lang=$lang AND type=$type";
 339     $sth->execute($script,$lang,$type);
 340     while (my $descriptions = $sth->fetchrow_hashref) {
 341         $desc = $descriptions->{'description'};
 342     }
 343     unless ($desc) {
 344         $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
 345         $sth->execute($script,'en',$type);
 346         while (my $descriptions = $sth->fetchrow_hashref) {
 347             $desc = $descriptions->{'description'};
 348         }
 349     }
 350     return $desc;
 351 }
 352 =head2 regex_lang_subtags
 353
 354 This internal sub takes a string composed according to RFC 4646 as
 355 an input and returns a reference to a hash containing keys and values
 356 for ( language, script, region, variant, extension, privateuse )
 357
 358 =cut
 359
 360 sub regex_lang_subtags {
 361     my $string = shift;
 362
 363     # Regex for recognizing RFC 4646 well-formed tags
 364     # http://www.rfc-editor.org/rfc/rfc4646.txt
 365
 366     # regexes based on : http://unicode.org/cldr/data/tools/java/org/unicode/cldr/util/data/langtagRegex.txt
 367     # The structure requires no forward references, so it reverses the order.
 368     # The uppercase comments are fragments copied from RFC 4646
 369     #
 370     # Note: the tool requires that any real "=" or "#" or ";" in the regex be escaped.
 371
 372     my $alpha   = qr/[a-zA-Z]/ ;    # ALPHA
 373     my $digit   = qr/[0-9]/ ;   # DIGIT
 374     my $alphanum    = qr/[a-zA-Z0-9]/ ; # ALPHA / DIGIT
 375     my $x   = qr/[xX]/ ;    # private use singleton
 376     my $singleton = qr/[a-w y-z A-W Y-Z]/ ; # other singleton
 377     my $s   = qr/[-]/ ; # separator -- lenient parsers will use [-_]
 378
 379     # Now do the components. The structure is slightly different to allow for capturing the right components.
 380     # The notation (?:....) is a non-capturing version of (...): so the "?:" can be deleted if someone doesn't care about capturing.
 381
 382     my $extlang = qr{(?: $s $alpha{3} )}x ; # *3("-" 3ALPHA)
 383     my $language    = qr{(?: $alpha{2,3} | $alpha{4,8} )}x ;
 384     #my $language   = qr{(?: $alpha{2,3}$extlang{0,3} | $alpha{4,8} )}x ;   # (2*3ALPHA [ extlang ]) / 4ALPHA / 5*8ALPHA
 385
 386     my $script  = qr{(?: $alpha{4} )}x ;    # 4ALPHA
 387
 388     my $region  = qr{(?: $alpha{2} | $digit{3} )}x ;     # 2ALPHA / 3DIGIT
 389
 390     my $variantSub  = qr{(?: $digit$alphanum{3} | $alphanum{5,8} )}x ;  # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
 391     my $variant = qr{(?: $variantSub (?: $s$variantSub )* )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
 392
 393     my $extensionSub    = qr{(?: $singleton (?: $s$alphanum{2,8} )+ )}x ;   # singleton 1*("-" (2*8alphanum))
 394     my $extension   = qr{(?: $extensionSub (?: $s$extensionSub )* )}x ; # singleton 1*("-" (2*8alphanum))
 395
 396     my $privateuse  = qr{(?: $x (?: $s$alphanum{1,8} )+ )}x ;   # ("x"/"X") 1*("-" (1*8alphanum))
 397
 398     # Define certain grandfathered codes, since otherwise the regex is pretty useless.
 399     # Since these are limited, this is safe even later changes to the registry --
 400     # the only oddity is that it might change the type of the tag, and thus
 401     # the results from the capturing groups.
 402     # http://www.iana.org/assignments/language-subtag-registry
 403     # Note that these have to be compared case insensitively, requiring (?i) below.
 404
 405     my $grandfathered   = qr{(?: (?i)
 406         en $s GB $s oed
 407     |   i $s (?: ami | bnn | default | enochian | hak | klingon | lux | mingo | navajo | pwn | tao | tay | tsu )
 408     |   sgn $s (?: BE $s fr | BE $s nl | CH $s de)
 409 )}x;
 410
 411     # For well-formedness, we don't need the ones that would otherwise pass, so they are commented out here
 412
 413     #   |   art $s lojban
 414     #   |   cel $s gaulish
 415     #   |   en $s (?: boont | GB $s oed | scouse )
 416     #   |   no $s (?: bok | nyn)
 417     #   |   zh $s (?: cmn | cmn $s Hans | cmn $s Hant | gan | guoyu | hakka | min | min $s nan | wuu | xiang | yue)
 418
 419     # Here is the final breakdown, with capturing groups for each of these components
 420     # The language, variants, extensions, grandfathered, and private-use may have interior '-'
 421
 422     #my $root = qr{(?: ($language) (?: $s ($script) )? 40% (?: $s ($region) )? 40% (?: $s ($variant) )? 10% (?: $s ($extension) )? 5% (?: $s ($privateuse) )? 5% ) 90% | ($grandfathered) 5% | ($privateuse) 5% };
 423
 424     $string =~  qr{^ (?:($language)) (?:$s($script))? (?:$s($region))?  (?:$s($variant))?  (?:$s($extension))?  (?:$s($privateuse))? $}xi;  # |($grandfathered) | ($privateuse) $}xi;
 425     my %subtag = (
 426         'rfc4646_subtag' => $string,
 427         'language' => $1,
 428         'script' => $2,
 429         'region' => $3,
 430         'variant' => $4,
 431         'extension' => $5,
 432         'privateuse' => $6,
 433     );
 434     return \%subtag;
 435 }
 436
 437 # Script Direction Resources:
 438 # http://www.w3.org/International/questions/qa-scripts
 439 sub get_bidi {
 440     my ($language_script)= @_;
 441     my $dbh = C4::Context->dbh;
 442     my $bidi;
 443     my $sth = $dbh->prepare('SELECT bidi FROM language_script_bidi WHERE rfc4646_subtag=?');
 444     $sth->execute($language_script);
 445     while (my $result = $sth->fetchrow_hashref) {
 446         $bidi = $result->{'bidi'};
 447     }
 448     return $bidi;
 449 };
 450
 451 sub accept_language {
 452     # referenced http://search.cpan.org/src/CGILMORE/I18N-AcceptLanguage-1.04/lib/I18N/AcceptLanguage.pm
 453     my ($clientPreferences,$supportedLanguages) = @_;
 454     my @languages = ();
 455     if ($clientPreferences) {
 456         # There should be no whitespace anways, but a cleanliness/sanity check
 457         $clientPreferences =~ s/\s//g;
 458         # Prepare the list of client-acceptable languages
 459         foreach my $tag (split(/,/, $clientPreferences)) {
 460             my ($language, $quality) = split(/\;/, $tag);
 461             $quality =~ s/^q=//i if $quality;
 462             $quality = 1 unless $quality;
 463             next if $quality <= 0;
 464             # We want to force the wildcard to be last
 465             $quality = 0 if ($language eq '*');
 466             # Pushing lowercase language here saves processing later
 467             push(@languages, { quality => $quality,
 468                language => $language,
 469                lclanguage => lc($language) });
 470         }
 471     } else {
 472         carp "accept_language(x,y) called with no clientPreferences (x).";
 473     }
 474     # Prepare the list of server-supported languages
 475     my %supportedLanguages = ();
 476     my %secondaryLanguages = ();
 477     foreach my $language (@$supportedLanguages) {
 478         # warn "Language supported: " . $language->{language};
 479         my $subtag = $language->{rfc4646_subtag};
 480         $supportedLanguages{lc($subtag)} = $subtag;
 481         if ( $subtag =~ /^([^-]+)-?/ ) {
 482             $secondaryLanguages{lc($1)} = $subtag;
 483         }
 484     }
 485
 486     # Reverse sort the list, making best quality at the front of the array
 487     @languages = sort { $b->{quality} <=> $a->{quality} } @languages;
 488     my $secondaryMatch = '';
 489     foreach my $tag (@languages) {
 490         if (exists($supportedLanguages{$tag->{lclanguage}})) {
 491             # Client en-us eq server en-us
 492             return $supportedLanguages{$tag->{language}} if exists($supportedLanguages{$tag->{language}});
 493             return $supportedLanguages{$tag->{lclanguage}};
 494         } elsif (exists($secondaryLanguages{$tag->{lclanguage}})) {
 495             # Client en eq server en-us
 496             return $secondaryLanguages{$tag->{language}} if exists($secondaryLanguages{$tag->{language}});
 497             return $supportedLanguages{$tag->{lclanguage}};
 498         } elsif ($tag->{lclanguage} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
 499             # Client en-gb eq server en-us
 500             $secondaryMatch = $secondaryLanguages{$1};
 501         } elsif ($tag->{lclanguage} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
 502             # FIXME: We just checked the exact same conditional!
 503             # Client en-us eq server en
 504             $secondaryMatch = $supportedLanguages{$1};
 505         } elsif ($tag->{lclanguage} eq '*') {
 506         # * matches every language not already specified.
 507         # It doesn't care which we pick, so let's pick the default,
 508         # if available, then the first in the array.
 509         #return $acceptor->defaultLanguage() if $acceptor->defaultLanguage();
 510         return $supportedLanguages->[0];
 511         }
 512     }
 513     # No primary matches. Secondary? (ie, en-us requested and en supported)
 514     return $secondaryMatch if $secondaryMatch;
 515     return undef;   # else, we got nothing.
 516 }
 517 1;
 518
 519 __END__
 520
 521 =head1 AUTHOR
 522
 523 Joshua Ferraro
 524
 525 =cut