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.
8 # Koha is free software; you can redistribute it and/or modify it
9 # under the terms of the GNU General Public License as published by
10 # the Free Software Foundation; either version 3 of the License, or
11 # (at your option) any later version.
13 # Koha is distributed in the hope that it will be useful, but
14 # WITHOUT ANY WARRANTY; without even the implied warranty of
15 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
16 # GNU General Public License for more details.
18 # You should have received a copy of the GNU General Public License
19 # along with Koha; if not, see <http://www.gnu.org/licenses>.
27 use List::MoreUtils qw( any );
30 use Koha::Cache::Memory::Lite;
32 our (@ISA, @EXPORT_OK);
38 getTranslatedLanguages
42 push @EXPORT_OK, qw(getFrameworkLanguages getTranslatedLanguages getAllLanguages getLanguages get_bidi regex_lang_subtags language_get_description accept_language getlanguage get_rfc4646_from_iso639);
47 C4::Languages - Perl Module containing language list functions for Koha
59 =head2 getFrameworkLanguages
61 Returns a reference to an array of hashes:
63 my $languages = getFrameworkLanguages();
64 for my $language(@$languages) {
65 print "$language->{language_code}\n"; # language code in iso 639-2
66 print "$language->{language_name}\n"; # language name in native script
67 print "$language->{language_locale_name}\n"; # language name in current locale
72 sub getFrameworkLanguages {
73 # get a hash with all language codes, names, and locale names
74 my $all_languages = getAllLanguages();
77 # find the available directory names
78 my $dir=C4::Context->config('intranetdir')."/installer/data/";
80 opendir ($dir_h,$dir);
81 my @listdir= grep { !/^\.|CVS/ && -d "$dir/$_"} readdir($dir_h);
84 # pull out all data for the dir names that exist
85 for my $dirname (@listdir) {
86 for my $language_set (@$all_languages) {
88 if ($dirname eq $language_set->{language_code}) {
90 'language_code'=>$dirname,
91 'language_description'=>$language_set->{language_description},
92 'native_descrition'=>$language_set->{language_native_description} }
99 =head2 getTranslatedLanguages
101 Returns a reference to an array of hashes:
103 my $languages = getTranslatedLanguages();
104 print "Available translated languages:\n";
105 for my $language(@$trlanguages) {
106 print "$language->{language_code}\n"; # language code in iso 639-2
107 print "$language->{language_name}\n"; # language name in native script
108 print "$language->{language_locale_name}\n"; # language name in current locale
113 sub getTranslatedLanguages {
114 my ($interface, $theme, $current_language, $which) = @_;
116 my @enabled_languages =
117 ( $interface && $interface eq 'intranet' )
118 ? split ",", C4::Context->preference('language')
119 : split ",", C4::Context->preference('OPACLanguages');
121 my $cache = Koha::Caches->get_instance;
122 my $cache_key = "languages_${interface}_${theme}";
123 if ($interface && $interface eq 'opac' ) {
124 my $htdocs = C4::Context->config('opachtdocs');
125 my $cached = $cache->get_from_cache($cache_key);
127 @languages = @{$cached};
129 @languages = _get_opac_language_dirs( $htdocs, $theme );
130 $cache->set_in_cache($cache_key, \@languages );
133 elsif ($interface && $interface eq 'intranet' ) {
134 my $htdocs = C4::Context->config('intrahtdocs');
135 my $cached = $cache->get_from_cache($cache_key);
137 @languages = @{$cached};
139 @languages = _get_intranet_language_dirs( $htdocs, $theme );
140 $cache->set_in_cache($cache_key, \@languages );
144 my $htdocs = C4::Context->config('intrahtdocs');
145 push @languages, _get_intranet_language_dirs( $htdocs );
147 $htdocs = C4::Context->config('opachtdocs');
148 push @languages, _get_opac_language_dirs( $htdocs );
151 $seen{$_}++ for @languages;
152 @languages = keys %seen;
154 return _build_languages_arrayref(\@languages,$current_language,\@enabled_languages);
157 =head2 getAllLanguages
159 Returns a reference to an array of hashes:
161 my $alllanguages = getAllLanguages();
162 print "Available translated languages:\n";
163 for my $language(@$alllanguages) {
164 print "$language->{language_code}\n";
165 print "$language->{language_name}\n";
166 print "$language->{language_locale_name}\n";
169 This routine is a wrapper for getLanguages().
173 sub getAllLanguages {
174 return getLanguages(shift);
179 my $lang_arrayref = getLanguages([$lang[, $isFiltered]]);
181 Returns a reference to an array of hashes of languages.
183 - If no parameter is passed to the function, it returns english languages names
184 - If a $lang parameter conforming to RFC4646 syntax is passed, the function returns languages names translated in $lang
185 If a language name is not translated in $lang in database, the function returns english language name
186 - If $isFiltered is set to true, only the detail of the languages selected in system preferences AdvanceSearchLanguages is returned.
192 my $isFiltered = shift;
195 my $dbh=C4::Context->dbh;
196 my $default_language = 'en';
197 my $current_language = $default_language;
198 my $language_list = $isFiltered ? C4::Context->preference("AdvancedSearchLanguages") : undef;
200 $current_language = regex_lang_subtags($lang)->{'language'};
202 my $sth = $dbh->prepare('SELECT * FROM language_subtag_registry WHERE type=\'language\' ORDER BY description');
204 while (my $language_subtag_registry = $sth->fetchrow_hashref) {
206 # check if language name is stored in current language
207 my $sth4= $dbh->prepare("SELECT description FROM language_descriptions WHERE type='language' AND subtag =? AND lang = ?");
208 $sth4->execute($language_subtag_registry->{subtag},$current_language);
209 while (my $language_desc = $sth4->fetchrow_hashref) {
210 $desc=$language_desc->{description};
212 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 = ?");
214 $sth2->execute($language_subtag_registry->{subtag},$current_language);
217 $sth2->execute($language_subtag_registry->{subtag},$default_language);
219 my $sth3 = $dbh->prepare("SELECT description FROM language_descriptions WHERE type='language' AND subtag=? AND lang=?");
220 # add the correct description info
221 while (my $language_descriptions = $sth2->fetchrow_hashref) {
222 $sth3->execute($language_subtag_registry->{subtag},$language_subtag_registry->{subtag});
223 my $native_description;
224 while (my $description = $sth3->fetchrow_hashref) {
225 $native_description = $description->{description};
228 # fill in the ISO6329 code
229 $language_subtag_registry->{iso639_2_code} = $language_descriptions->{iso639_2_code};
230 # fill in the current selected ui language's translation, as well as the native description of the language
231 if ($native_description) {
232 $language_subtag_registry->{language_description} = $language_descriptions->{description}
233 if $language_descriptions->{description};
234 $language_subtag_registry->{language_description} .= " ($native_description)";
237 $language_subtag_registry->{language_description} = $language_descriptions->{description};
240 # Do not push unless valid iso639-2 code
241 if ( $language_subtag_registry->{ iso639_2_code } and ( !$language_list || index ( $language_list, $language_subtag_registry->{ iso639_2_code } ) >= 0) ) {
242 push @languages_loop, $language_subtag_registry;
245 return \@languages_loop;
248 sub _get_opac_language_dirs {
249 my ( $htdocs, $theme ) = @_;
252 if ( $theme and -d "$htdocs/$theme" ) {
253 (@languages) = _get_language_dirs($htdocs,$theme);
256 for my $theme ( _get_themes('opac') ) {
257 push @languages, _get_language_dirs($htdocs,$theme);
264 sub _get_intranet_language_dirs {
265 my ( $htdocs, $theme ) = @_;
268 if ( $theme and -d "$htdocs/$theme" ) {
269 @languages = _get_language_dirs($htdocs,$theme);
272 foreach my $theme ( _get_themes('intranet') ) {
273 push @languages, _get_language_dirs($htdocs,$theme);
281 Internal function, returns an array of all available themes.
283 (@themes) = &_get_themes('opac');
284 (@themes) = &_get_themes('intranet');
289 my $interface = shift;
292 if ( $interface && $interface eq 'intranet' ) {
293 $htdocs = C4::Context->config('intrahtdocs');
296 $htdocs = C4::Context->config('opachtdocs');
299 opendir $dir_h, "$htdocs";
300 my @dirlist = readdir $dir_h;
301 foreach my $directory (@dirlist) {
302 # if there's an en dir, it's a valid theme
303 -d "$htdocs/$directory/en" and push @themes, $directory;
309 =head2 _get_language_dirs
311 Internal function, returns an array of directory names, excluding non-language directories
315 sub _get_language_dirs {
316 my ($htdocs,$theme) = @_;
321 opendir $dir_h, "$htdocs/$theme";
322 for my $lang_string ( sort readdir $dir_h ) {
323 next if $lang_string =~/^\./;
324 next if $lang_string eq 'all';
325 next if $lang_string =~/png$/;
326 next if $lang_string =~/js$/;
327 next if $lang_string =~/css$/;
328 next if $lang_string =~/CVS$/;
329 next if $lang_string =~/\.txt$/i; #Don't read the readme.txt !
330 next if $lang_string =~/img|images|famfam|js|less|lib|sound|pdf/;
331 push @lang_strings, $lang_string;
334 return (@lang_strings);
337 =head2 _build_languages_arrayref
339 Internal function for building the ref to array of hashes
341 FIXME: this could be rewritten and simplified using map
345 sub _build_languages_arrayref {
346 my ($translated_languages,$current_language,$enabled_languages) = @_;
347 $current_language //= '';
348 my @translated_languages = @$translated_languages;
349 my @languages_loop; # the final reference to an array of hashrefs
350 my @enabled_languages = @$enabled_languages;
351 # how many languages are enabled, if one, take note, some contexts won't need to display it
353 my $track_language_groups;
354 my $current_language_regex = regex_lang_subtags($current_language);
355 # Loop through the translated languages
356 for my $translated_language (@translated_languages) {
357 # separate the language string into its subtag types
358 my $language_subtags_hashref = regex_lang_subtags($translated_language);
360 # is this language string 'enabled'?
361 for my $enabled_language (@enabled_languages) {
362 #warn "Checking out if $translated_language eq $enabled_language";
363 $language_subtags_hashref->{'enabled'} = 1 if $translated_language eq $enabled_language;
366 # group this language, key by langtag
367 $language_subtags_hashref->{'sublanguage_current'} = 1 if $translated_language eq $current_language;
368 $language_subtags_hashref->{'rfc4646_subtag'} = $translated_language;
369 $language_subtags_hashref->{'native_description'} = language_get_description($language_subtags_hashref->{language},$language_subtags_hashref->{language},'language');
370 $language_subtags_hashref->{'script_description'} = language_get_description($language_subtags_hashref->{script},$language_subtags_hashref->{'language'},'script');
371 $language_subtags_hashref->{'region_description'} = language_get_description($language_subtags_hashref->{region},$language_subtags_hashref->{'language'},'region');
372 $language_subtags_hashref->{'variant_description'} = language_get_description($language_subtags_hashref->{variant},$language_subtags_hashref->{'language'},'variant');
373 $track_language_groups->{$language_subtags_hashref->{'language'}}++;
374 push ( @{ $language_groups->{$language_subtags_hashref->{language}} }, $language_subtags_hashref );
376 # $key is a language subtag like 'en'
378 my %idx = map { $enabled_languages->[$_] => $_ } reverse 0 .. @$enabled_languages-1;
379 my @ordered_keys = sort {
382 my $acount = @{ $language_groups->{$a} };
383 my $bcount = @{ $language_groups->{$b} };
384 if ( $language_groups->{$a}->[0]->{enabled} ) {
385 $aa = $language_groups->{$a}->[0]->{rfc4646_subtag};
387 elsif ( $acount > 1 ) {
388 for ( my $i = 1 ; $i < $acount ; $i++ ) {
389 if ( $language_groups->{$a}->[$i]->{enabled} ) {
390 $aa = $language_groups->{$a}->[$i]->{rfc4646_subtag};
395 if ( $language_groups->{$b}->[0]->{enabled} ) {
396 $bb = $language_groups->{$b}->[0]->{rfc4646_subtag};
398 elsif ( $bcount > 1 ) {
399 for ( my $i = 1 ; $i < $bcount ; $i++ ) {
400 if ( $language_groups->{$b}->[$i]->{enabled} ) {
401 $bb = $language_groups->{$b}->[$i]->{rfc4646_subtag};
408 and ( $idx{$aa} cmp $idx{$bb} ) )
409 || ( exists $idx{$aa} and exists $idx{$bb} )
411 } keys %$language_groups;
413 for my $key ( @ordered_keys ) {
414 my $value = $language_groups->{$key};
415 # is this language group enabled? are any of the languages within it enabled?
417 for my $enabled_language (@enabled_languages) {
418 my $regex_enabled_language = regex_lang_subtags($enabled_language);
419 $enabled = 1 if $key eq ($regex_enabled_language->{language} // '');
421 push @languages_loop, {
422 # this is only use if there is one
423 rfc4646_subtag => @$value[0]->{rfc4646_subtag},
424 native_description => language_get_description($key,$key,'language'),
426 sublanguages_loop => $value,
427 plural => $track_language_groups->{$key} >1 ? 1 : 0,
428 current => ($current_language_regex->{language} // '') eq $key ? 1 : 0,
429 group_enabled => $enabled,
432 return \@languages_loop;
435 sub language_get_description {
436 my ($script,$lang,$type) = @_;
437 my $dbh = C4::Context->dbh;
439 my $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
440 #warn "QUERY: SELECT description FROM language_descriptions WHERE subtag=$script AND lang=$lang AND type=$type";
441 $sth->execute($script,$lang,$type);
442 while (my $descriptions = $sth->fetchrow_hashref) {
443 $desc = $descriptions->{'description'};
446 $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
447 $sth->execute($script,'en',$type);
448 while (my $descriptions = $sth->fetchrow_hashref) {
449 $desc = $descriptions->{'description'};
454 =head2 regex_lang_subtags
456 This internal sub takes a string composed according to RFC 4646 as
457 an input and returns a reference to a hash containing keys and values
458 for ( language, script, region, variant, extension, privateuse )
462 sub regex_lang_subtags {
465 # Regex for recognizing RFC 4646 well-formed tags
466 # http://www.rfc-editor.org/rfc/rfc4646.txt
468 # regexes based on : http://unicode.org/cldr/data/tools/java/org/unicode/cldr/util/data/langtagRegex.txt
469 # The structure requires no forward references, so it reverses the order.
470 # The uppercase comments are fragments copied from RFC 4646
472 # Note: the tool requires that any real "=" or "#" or ";" in the regex be escaped.
474 my $alpha = qr/[a-zA-Z]/ ; # ALPHA
475 my $digit = qr/[0-9]/ ; # DIGIT
476 my $alphanum = qr/[a-zA-Z0-9]/ ; # ALPHA / DIGIT
477 my $x = qr/[xX]/ ; # private use singleton
478 my $singleton = qr/[a-w y-z A-W Y-Z]/ ; # other singleton
479 my $s = qr/[-]/ ; # separator -- lenient parsers will use [-_]
481 # Now do the components. The structure is slightly different to allow for capturing the right components.
482 # The notation (?:....) is a non-capturing version of (...): so the "?:" can be deleted if someone doesn't care about capturing.
484 my $extlang = qr{(?: $s $alpha{3} )}x ; # *3("-" 3ALPHA)
485 my $language = qr{(?: $alpha{2,3} | $alpha{4,8} )}x ;
486 #my $language = qr{(?: $alpha{2,3}$extlang{0,3} | $alpha{4,8} )}x ; # (2*3ALPHA [ extlang ]) / 4ALPHA / 5*8ALPHA
488 my $script = qr{(?: $alpha{4} )}x ; # 4ALPHA
490 my $region = qr{(?: $alpha{2} | $digit{3} )}x ; # 2ALPHA / 3DIGIT
492 my $variantSub = qr{(?: $digit$alphanum{3} | $alphanum{5,8} )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
493 my $variant = qr{(?: $variantSub (?: $s$variantSub )* )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
495 my $extensionSub = qr{(?: $singleton (?: $s$alphanum{2,8} )+ )}x ; # singleton 1*("-" (2*8alphanum))
496 my $extension = qr{(?: $extensionSub (?: $s$extensionSub )* )}x ; # singleton 1*("-" (2*8alphanum))
498 my $privateuse = qr{(?: $x (?: $s$alphanum{1,8} )+ )}x ; # ("x"/"X") 1*("-" (1*8alphanum))
500 # Define certain grandfathered codes, since otherwise the regex is pretty useless.
501 # Since these are limited, this is safe even later changes to the registry --
502 # the only oddity is that it might change the type of the tag, and thus
503 # the results from the capturing groups.
504 # http://www.iana.org/assignments/language-subtag-registry
505 # Note that these have to be compared case insensitively, requiring (?i) below.
507 my $grandfathered = qr{(?: (?i)
509 | i $s (?: ami | bnn | default | enochian | hak | klingon | lux | mingo | navajo | pwn | tao | tay | tsu )
510 | sgn $s (?: BE $s fr | BE $s nl | CH $s de)
513 # For well-formedness, we don't need the ones that would otherwise pass, so they are commented out here
517 # | en $s (?: boont | GB $s oed | scouse )
518 # | no $s (?: bok | nyn)
519 # | zh $s (?: cmn | cmn $s Hans | cmn $s Hant | gan | guoyu | hakka | min | min $s nan | wuu | xiang | yue)
521 # Here is the final breakdown, with capturing groups for each of these components
522 # The language, variants, extensions, grandfathered, and private-use may have interior '-'
524 #my $root = qr{(?: ($language) (?: $s ($script) )? 40% (?: $s ($region) )? 40% (?: $s ($variant) )? 10% (?: $s ($extension) )? 5% (?: $s ($privateuse) )? 5% ) 90% | ($grandfathered) 5% | ($privateuse) 5% };
526 $string =~ qr{^ (?:($language)) (?:$s($script))? (?:$s($region))? (?:$s($variant))? (?:$s($extension))? (?:$s($privateuse))? $}xi; # |($grandfathered) | ($privateuse) $}xi;
528 'rfc4646_subtag' => $string,
539 # Script Direction Resources:
540 # http://www.w3.org/International/questions/qa-scripts
542 my ($language_script)= @_;
543 my $dbh = C4::Context->dbh;
545 my $sth = $dbh->prepare('SELECT bidi FROM language_script_bidi WHERE rfc4646_subtag=?');
546 $sth->execute($language_script);
547 while (my $result = $sth->fetchrow_hashref) {
548 $bidi = $result->{'bidi'};
553 sub accept_language {
554 # referenced http://search.cpan.org/src/CGILMORE/I18N-AcceptLanguage-1.04/lib/I18N/AcceptLanguage.pm
555 my ($clientPreferences,$supportedLanguages) = @_;
557 if ($clientPreferences) {
558 # There should be no whitespace anways, but a cleanliness/sanity check
559 $clientPreferences =~ s/\s//g;
560 # Prepare the list of client-acceptable languages
561 foreach my $tag (split(/,/, $clientPreferences)) {
562 my ($language, $quality) = split(/\;/, $tag);
563 $quality =~ s/^q=//i if $quality;
564 $quality = 1 unless $quality;
565 next if $quality <= 0;
566 # We want to force the wildcard to be last
567 $quality = 0 if ($language eq '*');
568 # Pushing lowercase language here saves processing later
569 push(@languages, { quality => $quality,
570 language => $language,
571 lclanguage => lc($language) });
574 carp "accept_language(x,y) called with no clientPreferences (x).";
576 # Prepare the list of server-supported languages
577 my %supportedLanguages = ();
578 my %secondaryLanguages = ();
579 foreach my $language (@$supportedLanguages) {
580 # warn "Language supported: " . $language->{language};
581 my $subtag = $language->{rfc4646_subtag};
582 $supportedLanguages{lc($subtag)} = $subtag;
583 if ( $subtag =~ /^([^-]+)-?/ ) {
584 $secondaryLanguages{lc($1)} = $subtag;
588 # Reverse sort the list, making best quality at the front of the array
589 @languages = sort { $b->{quality} <=> $a->{quality} } @languages;
590 my $secondaryMatch = '';
591 foreach my $tag (@languages) {
592 if (exists($supportedLanguages{$tag->{lclanguage}})) {
593 # Client en-us eq server en-us
594 return $supportedLanguages{$tag->{language}} if exists($supportedLanguages{$tag->{language}});
595 return $supportedLanguages{$tag->{lclanguage}};
596 } elsif (exists($secondaryLanguages{$tag->{lclanguage}})) {
597 # Client en eq server en-us
598 return $secondaryLanguages{$tag->{language}} if exists($secondaryLanguages{$tag->{language}});
599 return $supportedLanguages{$tag->{lclanguage}};
600 } elsif ($tag->{lclanguage} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
601 # Client en-gb eq server en-us
602 $secondaryMatch = $secondaryLanguages{$1};
603 } elsif ($tag->{lclanguage} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
604 # FIXME: We just checked the exact same conditional!
605 # Client en-us eq server en
606 $secondaryMatch = $supportedLanguages{$1};
607 } elsif ($tag->{lclanguage} eq '*') {
608 # * matches every language not already specified.
609 # It doesn't care which we pick, so let's pick the default,
610 # if available, then the first in the array.
611 #return $acceptor->defaultLanguage() if $acceptor->defaultLanguage();
612 return $supportedLanguages->[0];
615 # No primary matches. Secondary? (ie, en-us requested and en supported)
616 return $secondaryMatch if $secondaryMatch;
617 return; # else, we got nothing.
622 Select a language based on the URL parameter 'language', a cookie,
623 syspref available languages & browser
630 my $memory_cache = Koha::Cache::Memory::Lite->get_instance();
631 my $cache_key = "getlanguage";
632 unless ( $cgi and $cgi->param('language') ) {
633 my $cached = $memory_cache->get_from_cache($cache_key);
634 return $cached if $cached;
638 my $interface = C4::Context->interface;
639 my $theme = C4::Context->preference( ( $interface eq 'opac' ) ? 'opacthemes' : 'template' );
642 my $preference_to_check =
643 $interface eq 'intranet' ? 'language' : 'OPACLanguages';
644 # Get the available/valid languages list
646 my $preference_value = C4::Context->preference($preference_to_check);
647 if ($preference_value) {
648 @languages = split /,/, $preference_value;
651 # Chose language from the URL
652 my $cgi_param_language = $cgi->param( 'language' );
653 if ( defined $cgi_param_language && any { $_ eq $cgi_param_language } @languages) {
654 $language = $cgi_param_language;
658 if (not $language and my $cgi_cookie_language = $cgi->cookie('KohaOpacLanguage') ) {
659 ( $language = $cgi_cookie_language ) =~ s/[^a-zA-Z_-]*//; # sanitize cookie
662 # HTTP_ACCEPT_LANGUAGE
663 if ( !$language && $ENV{HTTP_ACCEPT_LANGUAGE} ) {
664 $language = accept_language( $ENV{HTTP_ACCEPT_LANGUAGE},
665 getTranslatedLanguages( $interface, $theme ) );
668 # Ignore a lang not selected in sysprefs
669 if ( $language && not any { $_ eq $language } @languages ) {
673 # Pick the first selected syspref language
674 $language = shift @languages unless $language;
676 # Fall back to English if necessary
679 $memory_cache->set_in_cache( $cache_key, $language );
683 =head2 get_rfc4646_from_iso639
685 Select a language rfc4646 code given an iso639 code
689 sub get_rfc4646_from_iso639 {
691 my $iso_code = shift;
692 my $rfc_subtag = Koha::Database->new()->schema->resultset('LanguageRfc4646ToIso639')->find({iso639_2_code=>$iso_code});
694 return $rfc_subtag->rfc4646_subtag;