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 );
29 use Koha::Cache::Memory::Lite;
30 use vars qw(@ISA @EXPORT @EXPORT_OK %EXPORT_TAGS $DEBUG);
36 &getFrameworkLanguages
37 &getTranslatedLanguages
41 @EXPORT_OK = qw(getFrameworkLanguages getTranslatedLanguages getAllLanguages getLanguages get_bidi regex_lang_subtags language_get_description accept_language getlanguage);
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 $dir=C4::Context->config('intranetdir')."/installer/data/";
80 my @listdir= grep { !/^\.|CVS/ && -d "$dir/$_"} readdir(MYDIR);
83 # pull out all data for the dir names that exist
84 for my $dirname (@listdir) {
85 for my $language_set (@$all_languages) {
87 if ($dirname eq $language_set->{language_code}) {
89 'language_code'=>$dirname,
90 'language_description'=>$language_set->{language_description},
91 'native_descrition'=>$language_set->{language_native_description} }
98 =head2 getTranslatedLanguages
100 Returns a reference to an array of hashes:
102 my $languages = getTranslatedLanguages();
103 print "Available translated languages:\n";
104 for my $language(@$trlanguages) {
105 print "$language->{language_code}\n"; # language code in iso 639-2
106 print "$language->{language_name}\n"; # language name in native script
107 print "$language->{language_locale_name}\n"; # language name in current locale
112 sub getTranslatedLanguages {
113 my ($interface, $theme, $current_language, $which) = @_;
115 my @enabled_languages =
116 ( $interface && $interface eq 'intranet' )
117 ? split ",", C4::Context->preference('language')
118 : split ",", C4::Context->preference('opaclanguages');
120 if ($interface && $interface eq 'opac' ) {
121 my $htdocs = C4::Context->config('opachtdocs');
122 @languages = _get_opac_language_dirs( $htdocs, $theme );
124 elsif ($interface && $interface eq 'intranet' ) {
125 my $htdocs = C4::Context->config('intrahtdocs');
126 @languages = _get_intranet_language_dirs( $htdocs, $theme );
129 my $htdocs = C4::Context->config('intrahtdocs');
130 push @languages, _get_intranet_language_dirs( $htdocs );
132 $htdocs = C4::Context->config('opachtdocs');
133 push @languages, _get_opac_language_dirs( $htdocs );
136 $seen{$_}++ for @languages;
137 @languages = keys %seen;
139 return _build_languages_arrayref(\@languages,$current_language,\@enabled_languages);
142 =head2 getAllLanguages
144 Returns a reference to an array of hashes:
146 my $alllanguages = getAllLanguages();
147 print "Available translated languages:\n";
148 for my $language(@$alllanguages) {
149 print "$language->{language_code}\n";
150 print "$language->{language_name}\n";
151 print "$language->{language_locale_name}\n";
154 This routine is a wrapper for getLanguages().
158 sub getAllLanguages {
159 return getLanguages(shift);
164 my $lang_arrayref = getLanguages([$lang[, $isFiltered]]);
166 Returns a reference to an array of hashes of languages.
168 - If no parameter is passed to the function, it returns english languages names
169 - If a $lang parameter conforming to RFC4646 syntax is passed, the function returns languages names translated in $lang
170 If a language name is not translated in $lang in database, the function returns english language name
171 - If $isFiltered is set to true, only the detail of the languages selected in system preferences AdvanceSearchLanguages is returned.
177 my $isFiltered = shift;
180 my $dbh=C4::Context->dbh;
181 my $default_language = 'en';
182 my $current_language = $default_language;
183 my $language_list = $isFiltered ? C4::Context->preference("AdvancedSearchLanguages") : undef;
185 $current_language = regex_lang_subtags($lang)->{'language'};
187 my $sth = $dbh->prepare('SELECT * FROM language_subtag_registry WHERE type=\'language\'');
189 while (my $language_subtag_registry = $sth->fetchrow_hashref) {
191 # check if language name is stored in current language
192 my $sth4= $dbh->prepare("SELECT description FROM language_descriptions WHERE type='language' AND subtag =? AND lang = ?");
193 $sth4->execute($language_subtag_registry->{subtag},$current_language);
194 while (my $language_desc = $sth4->fetchrow_hashref) {
195 $desc=$language_desc->{description};
197 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 = ?");
199 $sth2->execute($language_subtag_registry->{subtag},$current_language);
202 $sth2->execute($language_subtag_registry->{subtag},$default_language);
204 my $sth3 = $dbh->prepare("SELECT description FROM language_descriptions WHERE type='language' AND subtag=? AND lang=?");
205 # add the correct description info
206 while (my $language_descriptions = $sth2->fetchrow_hashref) {
207 $sth3->execute($language_subtag_registry->{subtag},$language_subtag_registry->{subtag});
208 my $native_description;
209 while (my $description = $sth3->fetchrow_hashref) {
210 $native_description = $description->{description};
213 # fill in the ISO6329 code
214 $language_subtag_registry->{iso639_2_code} = $language_descriptions->{iso639_2_code};
215 # fill in the native description of the language, as well as the current language's translation of that if it exists
216 if ($native_description) {
217 $language_subtag_registry->{language_description} = $native_description;
218 $language_subtag_registry->{language_description}.=" ($language_descriptions->{description})" if $language_descriptions->{description};
221 $language_subtag_registry->{language_description} = $language_descriptions->{description};
224 # Do not push unless valid iso639-2 code
225 if ( $language_subtag_registry->{ iso639_2_code } and ( !$language_list || index ( $language_list, $language_subtag_registry->{ iso639_2_code } ) >= 0) ) {
226 push @languages_loop, $language_subtag_registry;
229 return \@languages_loop;
232 sub _get_opac_language_dirs {
233 my ( $htdocs, $theme ) = @_;
236 if ( $theme and -d "$htdocs/$theme" ) {
237 (@languages) = _get_language_dirs($htdocs,$theme);
240 for my $theme ( _get_themes('opac') ) {
241 push @languages, _get_language_dirs($htdocs,$theme);
248 sub _get_intranet_language_dirs {
249 my ( $htdocs, $theme ) = @_;
252 if ( $theme and -d "$htdocs/$theme" ) {
253 @languages = _get_language_dirs($htdocs,$theme);
256 foreach my $theme ( _get_themes('intranet') ) {
257 push @languages, _get_language_dirs($htdocs,$theme);
265 Internal function, returns an array of all available themes.
267 (@themes) = &_get_themes('opac');
268 (@themes) = &_get_themes('intranet');
273 my $interface = shift;
276 if ( $interface && $interface eq 'intranet' ) {
277 $htdocs = C4::Context->config('intrahtdocs');
280 $htdocs = C4::Context->config('opachtdocs');
282 opendir D, "$htdocs";
283 my @dirlist = readdir D;
284 foreach my $directory (@dirlist) {
285 # if there's an en dir, it's a valid theme
286 -d "$htdocs/$directory/en" and push @themes, $directory;
291 =head2 _get_language_dirs
293 Internal function, returns an array of directory names, excluding non-language directories
297 sub _get_language_dirs {
298 my ($htdocs,$theme) = @_;
302 opendir D, "$htdocs/$theme";
303 for my $lang_string ( readdir D ) {
304 next if $lang_string =~/^\./;
305 next if $lang_string eq 'all';
306 next if $lang_string =~/png$/;
307 next if $lang_string =~/js$/;
308 next if $lang_string =~/css$/;
309 next if $lang_string =~/CVS$/;
310 next if $lang_string =~/\.txt$/i; #Don't read the readme.txt !
311 next if $lang_string =~/img|images|famfam|js|less|lib|sound|pdf/;
312 push @lang_strings, $lang_string;
314 return (@lang_strings);
317 =head2 _build_languages_arrayref
319 Internal function for building the ref to array of hashes
321 FIXME: this could be rewritten and simplified using map
325 sub _build_languages_arrayref {
326 my ($translated_languages,$current_language,$enabled_languages) = @_;
327 $current_language //= '';
328 my @translated_languages = @$translated_languages;
329 my @languages_loop; # the final reference to an array of hashrefs
330 my @enabled_languages = @$enabled_languages;
331 # how many languages are enabled, if one, take note, some contexts won't need to display it
332 my %seen_languages; # the language tags we've seen
335 my $track_language_groups;
336 my $current_language_regex = regex_lang_subtags($current_language);
337 # Loop through the translated languages
338 for my $translated_language (@translated_languages) {
339 # separate the language string into its subtag types
340 my $language_subtags_hashref = regex_lang_subtags($translated_language);
342 # is this language string 'enabled'?
343 for my $enabled_language (@enabled_languages) {
344 #warn "Checking out if $translated_language eq $enabled_language";
345 $language_subtags_hashref->{'enabled'} = 1 if $translated_language eq $enabled_language;
348 # group this language, key by langtag
349 $language_subtags_hashref->{'sublanguage_current'} = 1 if $translated_language eq $current_language;
350 $language_subtags_hashref->{'rfc4646_subtag'} = $translated_language;
351 $language_subtags_hashref->{'native_description'} = language_get_description($language_subtags_hashref->{language},$language_subtags_hashref->{language},'language');
352 $language_subtags_hashref->{'script_description'} = language_get_description($language_subtags_hashref->{script},$language_subtags_hashref->{'language'},'script');
353 $language_subtags_hashref->{'region_description'} = language_get_description($language_subtags_hashref->{region},$language_subtags_hashref->{'language'},'region');
354 $language_subtags_hashref->{'variant_description'} = language_get_description($language_subtags_hashref->{variant},$language_subtags_hashref->{'language'},'variant');
355 $track_language_groups->{$language_subtags_hashref->{'language'}}++;
356 push ( @{ $language_groups->{$language_subtags_hashref->{language}} }, $language_subtags_hashref );
358 # $key is a language subtag like 'en'
360 my %idx = map { $enabled_languages->[$_] => $_ } reverse 0 .. @$enabled_languages-1;
361 my @ordered_keys = sort {
362 my $aa = $language_groups->{$a}->[0]->{rfc4646_subtag};
363 my $bb = $language_groups->{$b}->[0]->{rfc4646_subtag};
364 ( exists $idx{$aa} and exists $idx{$bb} and ( $idx{$aa} cmp $idx{$bb} ) )
365 || ( exists $idx{$aa} and exists $idx{$bb} )
367 } keys %$language_groups;
369 for my $key ( @ordered_keys ) {
370 my $value = $language_groups->{$key};
371 # is this language group enabled? are any of the languages within it enabled?
373 for my $enabled_language (@enabled_languages) {
374 my $regex_enabled_language = regex_lang_subtags($enabled_language);
375 $enabled = 1 if $key eq ($regex_enabled_language->{language} // '');
377 push @languages_loop, {
378 # this is only use if there is one
379 rfc4646_subtag => @$value[0]->{rfc4646_subtag},
380 native_description => language_get_description($key,$key,'language'),
382 sublanguages_loop => $value,
383 plural => $track_language_groups->{$key} >1 ? 1 : 0,
384 current => ($current_language_regex->{language} // '') eq $key ? 1 : 0,
385 group_enabled => $enabled,
388 return \@languages_loop;
391 sub language_get_description {
392 my ($script,$lang,$type) = @_;
393 my $dbh = C4::Context->dbh;
395 my $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
396 #warn "QUERY: SELECT description FROM language_descriptions WHERE subtag=$script AND lang=$lang AND type=$type";
397 $sth->execute($script,$lang,$type);
398 while (my $descriptions = $sth->fetchrow_hashref) {
399 $desc = $descriptions->{'description'};
402 $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
403 $sth->execute($script,'en',$type);
404 while (my $descriptions = $sth->fetchrow_hashref) {
405 $desc = $descriptions->{'description'};
410 =head2 regex_lang_subtags
412 This internal sub takes a string composed according to RFC 4646 as
413 an input and returns a reference to a hash containing keys and values
414 for ( language, script, region, variant, extension, privateuse )
418 sub regex_lang_subtags {
421 # Regex for recognizing RFC 4646 well-formed tags
422 # http://www.rfc-editor.org/rfc/rfc4646.txt
424 # regexes based on : http://unicode.org/cldr/data/tools/java/org/unicode/cldr/util/data/langtagRegex.txt
425 # The structure requires no forward references, so it reverses the order.
426 # The uppercase comments are fragments copied from RFC 4646
428 # Note: the tool requires that any real "=" or "#" or ";" in the regex be escaped.
430 my $alpha = qr/[a-zA-Z]/ ; # ALPHA
431 my $digit = qr/[0-9]/ ; # DIGIT
432 my $alphanum = qr/[a-zA-Z0-9]/ ; # ALPHA / DIGIT
433 my $x = qr/[xX]/ ; # private use singleton
434 my $singleton = qr/[a-w y-z A-W Y-Z]/ ; # other singleton
435 my $s = qr/[-]/ ; # separator -- lenient parsers will use [-_]
437 # Now do the components. The structure is slightly different to allow for capturing the right components.
438 # The notation (?:....) is a non-capturing version of (...): so the "?:" can be deleted if someone doesn't care about capturing.
440 my $extlang = qr{(?: $s $alpha{3} )}x ; # *3("-" 3ALPHA)
441 my $language = qr{(?: $alpha{2,3} | $alpha{4,8} )}x ;
442 #my $language = qr{(?: $alpha{2,3}$extlang{0,3} | $alpha{4,8} )}x ; # (2*3ALPHA [ extlang ]) / 4ALPHA / 5*8ALPHA
444 my $script = qr{(?: $alpha{4} )}x ; # 4ALPHA
446 my $region = qr{(?: $alpha{2} | $digit{3} )}x ; # 2ALPHA / 3DIGIT
448 my $variantSub = qr{(?: $digit$alphanum{3} | $alphanum{5,8} )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
449 my $variant = qr{(?: $variantSub (?: $s$variantSub )* )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
451 my $extensionSub = qr{(?: $singleton (?: $s$alphanum{2,8} )+ )}x ; # singleton 1*("-" (2*8alphanum))
452 my $extension = qr{(?: $extensionSub (?: $s$extensionSub )* )}x ; # singleton 1*("-" (2*8alphanum))
454 my $privateuse = qr{(?: $x (?: $s$alphanum{1,8} )+ )}x ; # ("x"/"X") 1*("-" (1*8alphanum))
456 # Define certain grandfathered codes, since otherwise the regex is pretty useless.
457 # Since these are limited, this is safe even later changes to the registry --
458 # the only oddity is that it might change the type of the tag, and thus
459 # the results from the capturing groups.
460 # http://www.iana.org/assignments/language-subtag-registry
461 # Note that these have to be compared case insensitively, requiring (?i) below.
463 my $grandfathered = qr{(?: (?i)
465 | i $s (?: ami | bnn | default | enochian | hak | klingon | lux | mingo | navajo | pwn | tao | tay | tsu )
466 | sgn $s (?: BE $s fr | BE $s nl | CH $s de)
469 # For well-formedness, we don't need the ones that would otherwise pass, so they are commented out here
473 # | en $s (?: boont | GB $s oed | scouse )
474 # | no $s (?: bok | nyn)
475 # | zh $s (?: cmn | cmn $s Hans | cmn $s Hant | gan | guoyu | hakka | min | min $s nan | wuu | xiang | yue)
477 # Here is the final breakdown, with capturing groups for each of these components
478 # The language, variants, extensions, grandfathered, and private-use may have interior '-'
480 #my $root = qr{(?: ($language) (?: $s ($script) )? 40% (?: $s ($region) )? 40% (?: $s ($variant) )? 10% (?: $s ($extension) )? 5% (?: $s ($privateuse) )? 5% ) 90% | ($grandfathered) 5% | ($privateuse) 5% };
482 $string =~ qr{^ (?:($language)) (?:$s($script))? (?:$s($region))? (?:$s($variant))? (?:$s($extension))? (?:$s($privateuse))? $}xi; # |($grandfathered) | ($privateuse) $}xi;
484 'rfc4646_subtag' => $string,
495 # Script Direction Resources:
496 # http://www.w3.org/International/questions/qa-scripts
498 my ($language_script)= @_;
499 my $dbh = C4::Context->dbh;
501 my $sth = $dbh->prepare('SELECT bidi FROM language_script_bidi WHERE rfc4646_subtag=?');
502 $sth->execute($language_script);
503 while (my $result = $sth->fetchrow_hashref) {
504 $bidi = $result->{'bidi'};
509 sub accept_language {
510 # referenced http://search.cpan.org/src/CGILMORE/I18N-AcceptLanguage-1.04/lib/I18N/AcceptLanguage.pm
511 my ($clientPreferences,$supportedLanguages) = @_;
513 if ($clientPreferences) {
514 # There should be no whitespace anways, but a cleanliness/sanity check
515 $clientPreferences =~ s/\s//g;
516 # Prepare the list of client-acceptable languages
517 foreach my $tag (split(/,/, $clientPreferences)) {
518 my ($language, $quality) = split(/\;/, $tag);
519 $quality =~ s/^q=//i if $quality;
520 $quality = 1 unless $quality;
521 next if $quality <= 0;
522 # We want to force the wildcard to be last
523 $quality = 0 if ($language eq '*');
524 # Pushing lowercase language here saves processing later
525 push(@languages, { quality => $quality,
526 language => $language,
527 lclanguage => lc($language) });
530 carp "accept_language(x,y) called with no clientPreferences (x).";
532 # Prepare the list of server-supported languages
533 my %supportedLanguages = ();
534 my %secondaryLanguages = ();
535 foreach my $language (@$supportedLanguages) {
536 # warn "Language supported: " . $language->{language};
537 my $subtag = $language->{rfc4646_subtag};
538 $supportedLanguages{lc($subtag)} = $subtag;
539 if ( $subtag =~ /^([^-]+)-?/ ) {
540 $secondaryLanguages{lc($1)} = $subtag;
544 # Reverse sort the list, making best quality at the front of the array
545 @languages = sort { $b->{quality} <=> $a->{quality} } @languages;
546 my $secondaryMatch = '';
547 foreach my $tag (@languages) {
548 if (exists($supportedLanguages{$tag->{lclanguage}})) {
549 # Client en-us eq server en-us
550 return $supportedLanguages{$tag->{language}} if exists($supportedLanguages{$tag->{language}});
551 return $supportedLanguages{$tag->{lclanguage}};
552 } elsif (exists($secondaryLanguages{$tag->{lclanguage}})) {
553 # Client en eq server en-us
554 return $secondaryLanguages{$tag->{language}} if exists($secondaryLanguages{$tag->{language}});
555 return $supportedLanguages{$tag->{lclanguage}};
556 } elsif ($tag->{lclanguage} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
557 # Client en-gb eq server en-us
558 $secondaryMatch = $secondaryLanguages{$1};
559 } elsif ($tag->{lclanguage} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
560 # FIXME: We just checked the exact same conditional!
561 # Client en-us eq server en
562 $secondaryMatch = $supportedLanguages{$1};
563 } elsif ($tag->{lclanguage} eq '*') {
564 # * matches every language not already specified.
565 # It doesn't care which we pick, so let's pick the default,
566 # if available, then the first in the array.
567 #return $acceptor->defaultLanguage() if $acceptor->defaultLanguage();
568 return $supportedLanguages->[0];
571 # No primary matches. Secondary? (ie, en-us requested and en supported)
572 return $secondaryMatch if $secondaryMatch;
573 return undef; # else, we got nothing.
578 Select a language based on the URL parameter 'language', a cookie,
579 syspref available languages & browser
586 my $memory_cache = Koha::Cache::Memory::Lite->get_instance();
587 my $cache_key = "getlanguage";
588 unless ( $cgi and $cgi->param('language') ) {
589 my $cached = $memory_cache->get_from_cache($cache_key);
590 return $cached if $cached;
594 my $interface = C4::Context->interface;
595 my $theme = C4::Context->preference( ( $interface eq 'opac' ) ? 'opacthemes' : 'template' );
598 my $preference_to_check =
599 $interface eq 'intranet' ? 'language' : 'opaclanguages';
600 # Get the available/valid languages list
602 my $preference_value = C4::Context->preference($preference_to_check);
603 if ($preference_value) {
604 @languages = split /,/, $preference_value;
607 # Chose language from the URL
608 my $cgi_param_language = $cgi->param( 'language' );
609 if ( defined $cgi_param_language && any { $_ eq $cgi_param_language } @languages) {
610 $language = $cgi_param_language;
614 if (not $language and my $cgi_cookie_language = $cgi->cookie('KohaOpacLanguage') ) {
615 ( $language = $cgi_cookie_language ) =~ s/[^a-zA-Z_-]*//; # sanitize cookie
618 # HTTP_ACCEPT_LANGUAGE
619 if ( !$language && $ENV{HTTP_ACCEPT_LANGUAGE} ) {
620 $language = accept_language( $ENV{HTTP_ACCEPT_LANGUAGE},
621 getTranslatedLanguages( $interface, $theme ) );
624 # Ignore a lang not selected in sysprefs
625 if ( $language && not any { $_ eq $language } @languages ) {
629 # Pick the first selected syspref language
630 $language = shift @languages unless $language;
632 # Fall back to English if necessary
635 $memory_cache->set_in_cache( $cache_key, $language );
639 =head2 get_rfc4646_from_iso639
641 Select a language rfc4646 code given an iso639 code
645 sub get_rfc4646_from_iso639 {
647 my $iso_code = shift;
648 my $rfc_subtag = Koha::Database->new()->schema->resultset('LanguageRfc4646ToIso639')->find({iso639_2_code=>$iso_code});
650 return $rfc_subtag->rfc4646_subtag;