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($VERSION @ISA @EXPORT @EXPORT_OK %EXPORT_TAGS $DEBUG);
33 $VERSION = 3.07.00.049;
37 &getFrameworkLanguages
38 &getTranslatedLanguages
42 @EXPORT_OK = qw(getFrameworkLanguages getTranslatedLanguages getAllLanguages getLanguages get_bidi regex_lang_subtags language_get_description accept_language getlanguage);
48 C4::Languages - Perl Module containing language list functions for Koha
60 =head2 getFrameworkLanguages
62 Returns a reference to an array of hashes:
64 my $languages = getFrameworkLanguages();
65 for my $language(@$languages) {
66 print "$language->{language_code}\n"; # language code in iso 639-2
67 print "$language->{language_name}\n"; # language name in native script
68 print "$language->{language_locale_name}\n"; # language name in current locale
73 sub getFrameworkLanguages {
74 # get a hash with all language codes, names, and locale names
75 my $all_languages = getAllLanguages();
78 # find the available directory names
79 my $dir=C4::Context->config('intranetdir')."/installer/data/";
81 my @listdir= grep { !/^\.|CVS/ && -d "$dir/$_"} readdir(MYDIR);
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, $which) = @_;
117 my @enabled_languages;
119 if ($interface && $interface eq 'opac' ) {
120 @enabled_languages = split ",", C4::Context->preference('opaclanguages');
121 $htdocs = C4::Context->config('opachtdocs');
122 if ( $theme and -d "$htdocs/$theme" ) {
123 (@languages) = _get_language_dirs($htdocs,$theme);
126 for my $theme ( _get_themes('opac') ) {
127 push @languages, _get_language_dirs($htdocs,$theme);
131 elsif ($interface && $interface eq 'intranet' ) {
132 @enabled_languages = split ",", C4::Context->preference('language');
133 $htdocs = C4::Context->config('intrahtdocs');
134 if ( $theme and -d "$htdocs/$theme" ) {
135 @languages = _get_language_dirs($htdocs,$theme);
138 foreach my $theme ( _get_themes('intranet') ) {
139 push @languages, _get_language_dirs($htdocs,$theme);
144 @enabled_languages = split ",", C4::Context->preference('opaclanguages');
145 my $htdocs = C4::Context->config('intrahtdocs');
146 foreach my $theme ( _get_themes('intranet') ) {
147 push @languages, _get_language_dirs($htdocs,$theme);
149 $htdocs = C4::Context->config('opachtdocs');
150 foreach my $theme ( _get_themes('opac') ) {
151 push @languages, _get_language_dirs($htdocs,$theme);
154 $seen{$_}++ for @languages;
155 @languages = keys %seen;
157 return _build_languages_arrayref(\@languages,$current_language,\@enabled_languages);
160 =head2 getAllLanguages
162 Returns a reference to an array of hashes:
164 my $alllanguages = getAllLanguages();
165 print "Available translated languages:\n";
166 for my $language(@$alllanguages) {
167 print "$language->{language_code}\n";
168 print "$language->{language_name}\n";
169 print "$language->{language_locale_name}\n";
172 This routine is a wrapper for getLanguages().
176 sub getAllLanguages {
177 return getLanguages(shift);
182 my $lang_arrayref = getLanguages([$lang[, $isFiltered]]);
184 Returns a reference to an array of hashes of languages.
186 - If no parameter is passed to the function, it returns english languages names
187 - If a $lang parameter conforming to RFC4646 syntax is passed, the function returns languages names translated in $lang
188 If a language name is not translated in $lang in database, the function returns english language name
189 - If $isFiltered is set to true, only the detail of the languages selected in system preferences AdvanceSearchLanguages is returned.
195 my $isFiltered = shift;
198 my $dbh=C4::Context->dbh;
199 my $default_language = 'en';
200 my $current_language = $default_language;
201 my $language_list = $isFiltered ? C4::Context->preference("AdvancedSearchLanguages") : undef;
203 $current_language = regex_lang_subtags($lang)->{'language'};
205 my $sth = $dbh->prepare('SELECT * FROM language_subtag_registry WHERE type=\'language\'');
207 while (my $language_subtag_registry = $sth->fetchrow_hashref) {
209 # check if language name is stored in current language
210 my $sth4= $dbh->prepare("SELECT description FROM language_descriptions WHERE type='language' AND subtag =? AND lang = ?");
211 $sth4->execute($language_subtag_registry->{subtag},$current_language);
212 while (my $language_desc = $sth4->fetchrow_hashref) {
213 $desc=$language_desc->{description};
215 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 = ?");
217 $sth2->execute($language_subtag_registry->{subtag},$current_language);
220 $sth2->execute($language_subtag_registry->{subtag},$default_language);
222 my $sth3 = $dbh->prepare("SELECT description FROM language_descriptions WHERE type='language' AND subtag=? AND lang=?");
223 # add the correct description info
224 while (my $language_descriptions = $sth2->fetchrow_hashref) {
225 $sth3->execute($language_subtag_registry->{subtag},$language_subtag_registry->{subtag});
226 my $native_description;
227 while (my $description = $sth3->fetchrow_hashref) {
228 $native_description = $description->{description};
231 # fill in the ISO6329 code
232 $language_subtag_registry->{iso639_2_code} = $language_descriptions->{iso639_2_code};
233 # fill in the native description of the language, as well as the current language's translation of that if it exists
234 if ($native_description) {
235 $language_subtag_registry->{language_description} = $native_description;
236 $language_subtag_registry->{language_description}.=" ($language_descriptions->{description})" if $language_descriptions->{description};
239 $language_subtag_registry->{language_description} = $language_descriptions->{description};
242 # Do not push unless valid iso639-2 code
243 if ( $language_subtag_registry->{ iso639_2_code } and ( !$language_list || index ( $language_list, $language_subtag_registry->{ iso639_2_code } ) >= 0) ) {
244 push @languages_loop, $language_subtag_registry;
247 return \@languages_loop;
252 Internal function, returns an array of all available themes.
254 (@themes) = &_get_themes('opac');
255 (@themes) = &_get_themes('intranet');
260 my $interface = shift;
263 if ( $interface && $interface eq 'intranet' ) {
264 $htdocs = C4::Context->config('intrahtdocs');
267 $htdocs = C4::Context->config('opachtdocs');
269 opendir D, "$htdocs";
270 my @dirlist = readdir D;
271 foreach my $directory (@dirlist) {
272 # if there's an en dir, it's a valid theme
273 -d "$htdocs/$directory/en" and push @themes, $directory;
278 =head2 _get_language_dirs
280 Internal function, returns an array of directory names, excluding non-language directories
284 sub _get_language_dirs {
285 my ($htdocs,$theme) = @_;
289 opendir D, "$htdocs/$theme";
290 for my $lang_string ( readdir D ) {
291 next if $lang_string =~/^\./;
292 next if $lang_string eq 'all';
293 next if $lang_string =~/png$/;
294 next if $lang_string =~/js$/;
295 next if $lang_string =~/css$/;
296 next if $lang_string =~/CVS$/;
297 next if $lang_string =~/\.txt$/i; #Don't read the readme.txt !
298 next if $lang_string =~/img|images|famfam|js|less|lib|sound|pdf/;
299 push @lang_strings, $lang_string;
301 return (@lang_strings);
304 =head2 _build_languages_arrayref
306 Internal function for building the ref to array of hashes
308 FIXME: this could be rewritten and simplified using map
312 sub _build_languages_arrayref {
313 my ($translated_languages,$current_language,$enabled_languages) = @_;
314 $current_language //= '';
315 my @translated_languages = @$translated_languages;
316 my @languages_loop; # the final reference to an array of hashrefs
317 my @enabled_languages = @$enabled_languages;
318 # how many languages are enabled, if one, take note, some contexts won't need to display it
319 my %seen_languages; # the language tags we've seen
322 my $track_language_groups;
323 my $current_language_regex = regex_lang_subtags($current_language);
324 # Loop through the translated languages
325 for my $translated_language (@translated_languages) {
326 # separate the language string into its subtag types
327 my $language_subtags_hashref = regex_lang_subtags($translated_language);
329 # is this language string 'enabled'?
330 for my $enabled_language (@enabled_languages) {
331 #warn "Checking out if $translated_language eq $enabled_language";
332 $language_subtags_hashref->{'enabled'} = 1 if $translated_language eq $enabled_language;
335 # group this language, key by langtag
336 $language_subtags_hashref->{'sublanguage_current'} = 1 if $translated_language eq $current_language;
337 $language_subtags_hashref->{'rfc4646_subtag'} = $translated_language;
338 $language_subtags_hashref->{'native_description'} = language_get_description($language_subtags_hashref->{language},$language_subtags_hashref->{language},'language');
339 $language_subtags_hashref->{'script_description'} = language_get_description($language_subtags_hashref->{script},$language_subtags_hashref->{'language'},'script');
340 $language_subtags_hashref->{'region_description'} = language_get_description($language_subtags_hashref->{region},$language_subtags_hashref->{'language'},'region');
341 $language_subtags_hashref->{'variant_description'} = language_get_description($language_subtags_hashref->{variant},$language_subtags_hashref->{'language'},'variant');
342 $track_language_groups->{$language_subtags_hashref->{'language'}}++;
343 push ( @{ $language_groups->{$language_subtags_hashref->{language}} }, $language_subtags_hashref );
345 # $key is a language subtag like 'en'
346 while( my ($key, $value) = each %$language_groups) {
348 # is this language group enabled? are any of the languages within it enabled?
350 for my $enabled_language (@enabled_languages) {
351 my $regex_enabled_language = regex_lang_subtags($enabled_language);
352 $enabled = 1 if $key eq ($regex_enabled_language->{language} // '');
354 push @languages_loop, {
355 # this is only use if there is one
356 rfc4646_subtag => @$value[0]->{rfc4646_subtag},
357 native_description => language_get_description($key,$key,'language'),
359 sublanguages_loop => $value,
360 plural => $track_language_groups->{$key} >1 ? 1 : 0,
361 current => ($current_language_regex->{language} // '') eq $key ? 1 : 0,
362 group_enabled => $enabled,
365 return \@languages_loop;
368 sub language_get_description {
369 my ($script,$lang,$type) = @_;
370 my $dbh = C4::Context->dbh;
372 my $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
373 #warn "QUERY: SELECT description FROM language_descriptions WHERE subtag=$script AND lang=$lang AND type=$type";
374 $sth->execute($script,$lang,$type);
375 while (my $descriptions = $sth->fetchrow_hashref) {
376 $desc = $descriptions->{'description'};
379 $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
380 $sth->execute($script,'en',$type);
381 while (my $descriptions = $sth->fetchrow_hashref) {
382 $desc = $descriptions->{'description'};
387 =head2 regex_lang_subtags
389 This internal sub takes a string composed according to RFC 4646 as
390 an input and returns a reference to a hash containing keys and values
391 for ( language, script, region, variant, extension, privateuse )
395 sub regex_lang_subtags {
398 # Regex for recognizing RFC 4646 well-formed tags
399 # http://www.rfc-editor.org/rfc/rfc4646.txt
401 # regexes based on : http://unicode.org/cldr/data/tools/java/org/unicode/cldr/util/data/langtagRegex.txt
402 # The structure requires no forward references, so it reverses the order.
403 # The uppercase comments are fragments copied from RFC 4646
405 # Note: the tool requires that any real "=" or "#" or ";" in the regex be escaped.
407 my $alpha = qr/[a-zA-Z]/ ; # ALPHA
408 my $digit = qr/[0-9]/ ; # DIGIT
409 my $alphanum = qr/[a-zA-Z0-9]/ ; # ALPHA / DIGIT
410 my $x = qr/[xX]/ ; # private use singleton
411 my $singleton = qr/[a-w y-z A-W Y-Z]/ ; # other singleton
412 my $s = qr/[-]/ ; # separator -- lenient parsers will use [-_]
414 # Now do the components. The structure is slightly different to allow for capturing the right components.
415 # The notation (?:....) is a non-capturing version of (...): so the "?:" can be deleted if someone doesn't care about capturing.
417 my $extlang = qr{(?: $s $alpha{3} )}x ; # *3("-" 3ALPHA)
418 my $language = qr{(?: $alpha{2,3} | $alpha{4,8} )}x ;
419 #my $language = qr{(?: $alpha{2,3}$extlang{0,3} | $alpha{4,8} )}x ; # (2*3ALPHA [ extlang ]) / 4ALPHA / 5*8ALPHA
421 my $script = qr{(?: $alpha{4} )}x ; # 4ALPHA
423 my $region = qr{(?: $alpha{2} | $digit{3} )}x ; # 2ALPHA / 3DIGIT
425 my $variantSub = qr{(?: $digit$alphanum{3} | $alphanum{5,8} )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
426 my $variant = qr{(?: $variantSub (?: $s$variantSub )* )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
428 my $extensionSub = qr{(?: $singleton (?: $s$alphanum{2,8} )+ )}x ; # singleton 1*("-" (2*8alphanum))
429 my $extension = qr{(?: $extensionSub (?: $s$extensionSub )* )}x ; # singleton 1*("-" (2*8alphanum))
431 my $privateuse = qr{(?: $x (?: $s$alphanum{1,8} )+ )}x ; # ("x"/"X") 1*("-" (1*8alphanum))
433 # Define certain grandfathered codes, since otherwise the regex is pretty useless.
434 # Since these are limited, this is safe even later changes to the registry --
435 # the only oddity is that it might change the type of the tag, and thus
436 # the results from the capturing groups.
437 # http://www.iana.org/assignments/language-subtag-registry
438 # Note that these have to be compared case insensitively, requiring (?i) below.
440 my $grandfathered = qr{(?: (?i)
442 | i $s (?: ami | bnn | default | enochian | hak | klingon | lux | mingo | navajo | pwn | tao | tay | tsu )
443 | sgn $s (?: BE $s fr | BE $s nl | CH $s de)
446 # For well-formedness, we don't need the ones that would otherwise pass, so they are commented out here
450 # | en $s (?: boont | GB $s oed | scouse )
451 # | no $s (?: bok | nyn)
452 # | zh $s (?: cmn | cmn $s Hans | cmn $s Hant | gan | guoyu | hakka | min | min $s nan | wuu | xiang | yue)
454 # Here is the final breakdown, with capturing groups for each of these components
455 # The language, variants, extensions, grandfathered, and private-use may have interior '-'
457 #my $root = qr{(?: ($language) (?: $s ($script) )? 40% (?: $s ($region) )? 40% (?: $s ($variant) )? 10% (?: $s ($extension) )? 5% (?: $s ($privateuse) )? 5% ) 90% | ($grandfathered) 5% | ($privateuse) 5% };
459 $string =~ qr{^ (?:($language)) (?:$s($script))? (?:$s($region))? (?:$s($variant))? (?:$s($extension))? (?:$s($privateuse))? $}xi; # |($grandfathered) | ($privateuse) $}xi;
461 'rfc4646_subtag' => $string,
472 # Script Direction Resources:
473 # http://www.w3.org/International/questions/qa-scripts
475 my ($language_script)= @_;
476 my $dbh = C4::Context->dbh;
478 my $sth = $dbh->prepare('SELECT bidi FROM language_script_bidi WHERE rfc4646_subtag=?');
479 $sth->execute($language_script);
480 while (my $result = $sth->fetchrow_hashref) {
481 $bidi = $result->{'bidi'};
486 sub accept_language {
487 # referenced http://search.cpan.org/src/CGILMORE/I18N-AcceptLanguage-1.04/lib/I18N/AcceptLanguage.pm
488 my ($clientPreferences,$supportedLanguages) = @_;
490 if ($clientPreferences) {
491 # There should be no whitespace anways, but a cleanliness/sanity check
492 $clientPreferences =~ s/\s//g;
493 # Prepare the list of client-acceptable languages
494 foreach my $tag (split(/,/, $clientPreferences)) {
495 my ($language, $quality) = split(/\;/, $tag);
496 $quality =~ s/^q=//i if $quality;
497 $quality = 1 unless $quality;
498 next if $quality <= 0;
499 # We want to force the wildcard to be last
500 $quality = 0 if ($language eq '*');
501 # Pushing lowercase language here saves processing later
502 push(@languages, { quality => $quality,
503 language => $language,
504 lclanguage => lc($language) });
507 carp "accept_language(x,y) called with no clientPreferences (x).";
509 # Prepare the list of server-supported languages
510 my %supportedLanguages = ();
511 my %secondaryLanguages = ();
512 foreach my $language (@$supportedLanguages) {
513 # warn "Language supported: " . $language->{language};
514 my $subtag = $language->{rfc4646_subtag};
515 $supportedLanguages{lc($subtag)} = $subtag;
516 if ( $subtag =~ /^([^-]+)-?/ ) {
517 $secondaryLanguages{lc($1)} = $subtag;
521 # Reverse sort the list, making best quality at the front of the array
522 @languages = sort { $b->{quality} <=> $a->{quality} } @languages;
523 my $secondaryMatch = '';
524 foreach my $tag (@languages) {
525 if (exists($supportedLanguages{$tag->{lclanguage}})) {
526 # Client en-us eq server en-us
527 return $supportedLanguages{$tag->{language}} if exists($supportedLanguages{$tag->{language}});
528 return $supportedLanguages{$tag->{lclanguage}};
529 } elsif (exists($secondaryLanguages{$tag->{lclanguage}})) {
530 # Client en eq server en-us
531 return $secondaryLanguages{$tag->{language}} if exists($secondaryLanguages{$tag->{language}});
532 return $supportedLanguages{$tag->{lclanguage}};
533 } elsif ($tag->{lclanguage} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
534 # Client en-gb eq server en-us
535 $secondaryMatch = $secondaryLanguages{$1};
536 } elsif ($tag->{lclanguage} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
537 # FIXME: We just checked the exact same conditional!
538 # Client en-us eq server en
539 $secondaryMatch = $supportedLanguages{$1};
540 } elsif ($tag->{lclanguage} eq '*') {
541 # * matches every language not already specified.
542 # It doesn't care which we pick, so let's pick the default,
543 # if available, then the first in the array.
544 #return $acceptor->defaultLanguage() if $acceptor->defaultLanguage();
545 return $supportedLanguages->[0];
548 # No primary matches. Secondary? (ie, en-us requested and en supported)
549 return $secondaryMatch if $secondaryMatch;
550 return undef; # else, we got nothing.
555 Select a language based on the URL parameter 'language', a cookie,
556 syspref available languages & browser
563 my $memory_cache = Koha::Cache::Memory::Lite->get_instance();
564 my $cache_key = "getlanguage";
565 unless ( $cgi and $cgi->param('language') ) {
566 my $cached = $memory_cache->get_from_cache($cache_key);
567 return $cached if $cached;
571 my $interface = C4::Context->interface;
572 my $theme = C4::Context->preference( ( $interface eq 'opac' ) ? 'opacthemes' : 'template' );
575 my $preference_to_check =
576 $interface eq 'intranet' ? 'language' : 'opaclanguages';
577 # Get the available/valid languages list
579 my $preference_value = C4::Context->preference($preference_to_check);
580 if ($preference_value) {
581 @languages = split /,/, $preference_value;
584 # Chose language from the URL
585 my $cgi_param_language = $cgi->param( 'language' );
586 if ( defined $cgi_param_language && any { $_ eq $cgi_param_language } @languages) {
587 $language = $cgi_param_language;
591 if (not $language and my $cgi_cookie_language = $cgi->cookie('KohaOpacLanguage') ) {
592 ( $language = $cgi_cookie_language ) =~ s/[^a-zA-Z_-]*//; # sanitize cookie
595 # HTTP_ACCEPT_LANGUAGE
596 if ( !$language && $ENV{HTTP_ACCEPT_LANGUAGE} ) {
597 $language = accept_language( $ENV{HTTP_ACCEPT_LANGUAGE},
598 getTranslatedLanguages( $interface, $theme ) );
601 # Ignore a lang not selected in sysprefs
602 if ( $language && not any { $_ eq $language } @languages ) {
606 # Pick the first selected syspref language
607 $language = shift @languages unless $language;
609 # Fall back to English if necessary
612 $memory_cache->set_in_cache( $cache_key, $language );