Bug 8909: MARC21 240$a (Uniform title) is not a URL
[koha.git] / C4 / Languages.pm
blob57bef47465f111f773dd0176476cc2471ed48323
1 package C4::Languages;
3 # Copyright 2006 (C) LibLime
4 # Joshua Ferraro <jmf@liblime.com>
5 # Portions Copyright 2009 Chris Cormack and the Koha Dev Team
6 # This file is part of Koha.
8 # Koha is free software; you can redistribute it and/or modify it under the
9 # terms of the GNU General Public License as published by the Free Software
10 # Foundation; either version 2 of the License, or (at your option) any later
11 # version.
13 # Koha is distributed in the hope that it will be useful, but WITHOUT ANY
14 # WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
15 # A PARTICULAR PURPOSE. See the GNU General Public License for more details.
17 # You should have received a copy of the GNU General Public License along
18 # with Koha; if not, write to the Free Software Foundation, Inc.,
19 # 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
22 use strict;
23 use warnings;
25 use Carp;
26 use CGI;
27 use List::MoreUtils qw( any );
28 use C4::Context;
29 use vars qw($VERSION @ISA @EXPORT @EXPORT_OK %EXPORT_TAGS $DEBUG);
31 eval {
32 if (C4::Context->ismemcached) {
33 require Memoize::Memcached;
34 import Memoize::Memcached qw(memoize_memcached);
36 memoize_memcached('getTranslatedLanguages', memcached => C4::Context->memcached);
37 memoize_memcached('getFrameworkLanguages' , memcached => C4::Context->memcached);
38 memoize_memcached('getAllLanguages', memcached => C4::Context->memcached);
42 BEGIN {
43 $VERSION = 3.07.00.049;
44 require Exporter;
45 @ISA = qw(Exporter);
46 @EXPORT = qw(
47 &getFrameworkLanguages
48 &getTranslatedLanguages
49 &getLanguages
50 &getAllLanguages
52 @EXPORT_OK = qw(getFrameworkLanguages getTranslatedLanguages getAllLanguages getLanguages get_bidi regex_lang_subtags language_get_description accept_language getlanguage);
53 $DEBUG = 0;
56 =head1 NAME
58 C4::Languages - Perl Module containing language list functions for Koha
60 =head1 SYNOPSIS
62 use C4::Languages;
64 =head1 DESCRIPTION
66 =cut
68 =head1 FUNCTIONS
70 =head2 getFrameworkLanguages
72 Returns a reference to an array of hashes:
74 my $languages = getFrameworkLanguages();
75 for my $language(@$languages) {
76 print "$language->{language_code}\n"; # language code in iso 639-2
77 print "$language->{language_name}\n"; # language name in native script
78 print "$language->{language_locale_name}\n"; # language name in current locale
81 =cut
83 sub getFrameworkLanguages {
84 # get a hash with all language codes, names, and locale names
85 my $all_languages = getAllLanguages();
86 my @languages;
88 # find the available directory names
89 my $dir=C4::Context->config('intranetdir')."/installer/data/";
90 opendir (MYDIR,$dir);
91 my @listdir= grep { !/^\.|CVS/ && -d "$dir/$_"} readdir(MYDIR);
92 closedir MYDIR;
94 # pull out all data for the dir names that exist
95 for my $dirname (@listdir) {
96 for my $language_set (@$all_languages) {
98 if ($dirname eq $language_set->{language_code}) {
99 push @languages, {
100 'language_code'=>$dirname,
101 'language_description'=>$language_set->{language_description},
102 'native_descrition'=>$language_set->{language_native_description} }
106 return \@languages;
109 =head2 getTranslatedLanguages
111 Returns a reference to an array of hashes:
113 my $languages = getTranslatedLanguages();
114 print "Available translated languages:\n";
115 for my $language(@$trlanguages) {
116 print "$language->{language_code}\n"; # language code in iso 639-2
117 print "$language->{language_name}\n"; # language name in native script
118 print "$language->{language_locale_name}\n"; # language name in current locale
121 =cut
123 sub getTranslatedLanguages {
124 my ($interface, $theme, $current_language, $which) = @_;
125 my $htdocs;
126 my @languages;
127 my @enabled_languages;
129 if ($interface && $interface eq 'opac' ) {
130 @enabled_languages = split ",", C4::Context->preference('opaclanguages');
131 $htdocs = C4::Context->config('opachtdocs');
132 if ( $theme and -d "$htdocs/$theme" ) {
133 (@languages) = _get_language_dirs($htdocs,$theme);
135 else {
136 for my $theme ( _get_themes('opac') ) {
137 push @languages, _get_language_dirs($htdocs,$theme);
141 elsif ($interface && $interface eq 'intranet' ) {
142 @enabled_languages = split ",", C4::Context->preference('language');
143 $htdocs = C4::Context->config('intrahtdocs');
144 if ( $theme and -d "$htdocs/$theme" ) {
145 @languages = _get_language_dirs($htdocs,$theme);
147 else {
148 foreach my $theme ( _get_themes('intranet') ) {
149 push @languages, _get_language_dirs($htdocs,$theme);
153 else {
154 @enabled_languages = split ",", C4::Context->preference('opaclanguages');
155 my $htdocs = C4::Context->config('intrahtdocs');
156 foreach my $theme ( _get_themes('intranet') ) {
157 push @languages, _get_language_dirs($htdocs,$theme);
159 $htdocs = C4::Context->config('opachtdocs');
160 foreach my $theme ( _get_themes('opac') ) {
161 push @languages, _get_language_dirs($htdocs,$theme);
163 my %seen;
164 $seen{$_}++ for @languages;
165 @languages = keys %seen;
167 return _build_languages_arrayref(\@languages,$current_language,\@enabled_languages);
170 =head2 getAllLanguages
172 Returns a reference to an array of hashes:
174 my $alllanguages = getAllLanguages();
175 print "Available translated languages:\n";
176 for my $language(@$alllanguages) {
177 print "$language->{language_code}\n";
178 print "$language->{language_name}\n";
179 print "$language->{language_locale_name}\n";
182 This routine is a wrapper for getLanguages().
184 =cut
186 sub getAllLanguages {
187 return getLanguages(shift);
190 =head2 getLanguages
192 my $lang_arrayref = getLanguages([$lang[, $isFiltered]]);
194 Returns a reference to an array of hashes of languages.
196 - If no parameter is passed to the function, it returns english languages names
197 - If a $lang parameter conforming to RFC4646 syntax is passed, the function returns languages names translated in $lang
198 If a language name is not translated in $lang in database, the function returns english language name
199 - If $isFiltered is set to true, only the detail of the languages selected in system preferences AdvanceSearchLanguages is returned.
201 =cut
203 sub getLanguages {
204 my $lang = shift;
205 my $isFiltered = shift;
207 my @languages_loop;
208 my $dbh=C4::Context->dbh;
209 my $default_language = 'en';
210 my $current_language = $default_language;
211 my $language_list = $isFiltered ? C4::Context->preference("AdvancedSearchLanguages") : undef;
212 if ($lang) {
213 $current_language = regex_lang_subtags($lang)->{'language'};
215 my $sth = $dbh->prepare('SELECT * FROM language_subtag_registry WHERE type=\'language\'');
216 $sth->execute();
217 while (my $language_subtag_registry = $sth->fetchrow_hashref) {
218 my $desc;
219 # check if language name is stored in current language
220 my $sth4= $dbh->prepare("SELECT description FROM language_descriptions WHERE type='language' AND subtag =? AND lang = ?");
221 $sth4->execute($language_subtag_registry->{subtag},$current_language);
222 while (my $language_desc = $sth4->fetchrow_hashref) {
223 $desc=$language_desc->{description};
225 my $sth2= $dbh->prepare("SELECT * FROM language_descriptions LEFT JOIN language_rfc4646_to_iso639 on language_rfc4646_to_iso639.rfc4646_subtag = language_descriptions.subtag WHERE type='language' AND subtag =? AND language_descriptions.lang = ?");
226 if ($desc) {
227 $sth2->execute($language_subtag_registry->{subtag},$current_language);
229 else {
230 $sth2->execute($language_subtag_registry->{subtag},$default_language);
232 my $sth3 = $dbh->prepare("SELECT description FROM language_descriptions WHERE type='language' AND subtag=? AND lang=?");
233 # add the correct description info
234 while (my $language_descriptions = $sth2->fetchrow_hashref) {
235 $sth3->execute($language_subtag_registry->{subtag},$language_subtag_registry->{subtag});
236 my $native_description;
237 while (my $description = $sth3->fetchrow_hashref) {
238 $native_description = $description->{description};
241 # fill in the ISO6329 code
242 $language_subtag_registry->{iso639_2_code} = $language_descriptions->{iso639_2_code};
243 # fill in the native description of the language, as well as the current language's translation of that if it exists
244 if ($native_description) {
245 $language_subtag_registry->{language_description} = $native_description;
246 $language_subtag_registry->{language_description}.=" ($language_descriptions->{description})" if $language_descriptions->{description};
248 else {
249 $language_subtag_registry->{language_description} = $language_descriptions->{description};
252 if ( !$language_list || index ( $language_list, $language_subtag_registry->{ iso639_2_code } ) >= 0) {
253 push @languages_loop, $language_subtag_registry;
256 return \@languages_loop;
259 =head2 _get_themes
261 Internal function, returns an array of all available themes.
263 (@themes) = &_get_themes('opac');
264 (@themes) = &_get_themes('intranet');
266 =cut
268 sub _get_themes {
269 my $interface = shift;
270 my $htdocs;
271 my @themes;
272 if ( $interface && $interface eq 'intranet' ) {
273 $htdocs = C4::Context->config('intrahtdocs');
275 else {
276 $htdocs = C4::Context->config('opachtdocs');
278 opendir D, "$htdocs";
279 my @dirlist = readdir D;
280 foreach my $directory (@dirlist) {
281 # if there's an en dir, it's a valid theme
282 -d "$htdocs/$directory/en" and push @themes, $directory;
284 return @themes;
287 =head2 _get_language_dirs
289 Internal function, returns an array of directory names, excluding non-language directories
291 =cut
293 sub _get_language_dirs {
294 my ($htdocs,$theme) = @_;
295 $htdocs //= '';
296 $theme //= '';
297 my @lang_strings;
298 opendir D, "$htdocs/$theme";
299 for my $lang_string ( readdir D ) {
300 next if $lang_string =~/^\./;
301 next if $lang_string eq 'all';
302 next if $lang_string =~/png$/;
303 next if $lang_string =~/js$/;
304 next if $lang_string =~/css$/;
305 next if $lang_string =~/CVS$/;
306 next if $lang_string =~/\.txt$/i; #Don't read the readme.txt !
307 next if $lang_string =~/img|images|famfam|js|less|lib|sound|pdf/;
308 push @lang_strings, $lang_string;
310 return (@lang_strings);
313 =head2 _build_languages_arrayref
315 Internal function for building the ref to array of hashes
317 FIXME: this could be rewritten and simplified using map
319 =cut
321 sub _build_languages_arrayref {
322 my ($translated_languages,$current_language,$enabled_languages) = @_;
323 $current_language //= '';
324 my @translated_languages = @$translated_languages;
325 my @languages_loop; # the final reference to an array of hashrefs
326 my @enabled_languages = @$enabled_languages;
327 # how many languages are enabled, if one, take note, some contexts won't need to display it
328 my %seen_languages; # the language tags we've seen
329 my %found_languages;
330 my $language_groups;
331 my $track_language_groups;
332 my $current_language_regex = regex_lang_subtags($current_language);
333 # Loop through the translated languages
334 for my $translated_language (@translated_languages) {
335 # separate the language string into its subtag types
336 my $language_subtags_hashref = regex_lang_subtags($translated_language);
338 # is this language string 'enabled'?
339 for my $enabled_language (@enabled_languages) {
340 #warn "Checking out if $translated_language eq $enabled_language";
341 $language_subtags_hashref->{'enabled'} = 1 if $translated_language eq $enabled_language;
344 # group this language, key by langtag
345 $language_subtags_hashref->{'sublanguage_current'} = 1 if $translated_language eq $current_language;
346 $language_subtags_hashref->{'rfc4646_subtag'} = $translated_language;
347 $language_subtags_hashref->{'native_description'} = language_get_description($language_subtags_hashref->{language},$language_subtags_hashref->{language},'language');
348 $language_subtags_hashref->{'script_description'} = language_get_description($language_subtags_hashref->{script},$language_subtags_hashref->{'language'},'script');
349 $language_subtags_hashref->{'region_description'} = language_get_description($language_subtags_hashref->{region},$language_subtags_hashref->{'language'},'region');
350 $language_subtags_hashref->{'variant_description'} = language_get_description($language_subtags_hashref->{variant},$language_subtags_hashref->{'language'},'variant');
351 $track_language_groups->{$language_subtags_hashref->{'language'}}++;
352 push ( @{ $language_groups->{$language_subtags_hashref->{language}} }, $language_subtags_hashref );
354 # $key is a language subtag like 'en'
355 while( my ($key, $value) = each %$language_groups) {
357 # is this language group enabled? are any of the languages within it enabled?
358 my $enabled;
359 for my $enabled_language (@enabled_languages) {
360 my $regex_enabled_language = regex_lang_subtags($enabled_language);
361 $enabled = 1 if $key eq ($regex_enabled_language->{language} // '');
363 push @languages_loop, {
364 # this is only use if there is one
365 rfc4646_subtag => @$value[0]->{rfc4646_subtag},
366 native_description => language_get_description($key,$key,'language'),
367 language => $key,
368 sublanguages_loop => $value,
369 plural => $track_language_groups->{$key} >1 ? 1 : 0,
370 current => ($current_language_regex->{language} // '') eq $key ? 1 : 0,
371 group_enabled => $enabled,
374 return \@languages_loop;
377 sub language_get_description {
378 my ($script,$lang,$type) = @_;
379 my $dbh = C4::Context->dbh;
380 my $desc;
381 my $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
382 #warn "QUERY: SELECT description FROM language_descriptions WHERE subtag=$script AND lang=$lang AND type=$type";
383 $sth->execute($script,$lang,$type);
384 while (my $descriptions = $sth->fetchrow_hashref) {
385 $desc = $descriptions->{'description'};
387 unless ($desc) {
388 $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
389 $sth->execute($script,'en',$type);
390 while (my $descriptions = $sth->fetchrow_hashref) {
391 $desc = $descriptions->{'description'};
394 return $desc;
396 =head2 regex_lang_subtags
398 This internal sub takes a string composed according to RFC 4646 as
399 an input and returns a reference to a hash containing keys and values
400 for ( language, script, region, variant, extension, privateuse )
402 =cut
404 sub regex_lang_subtags {
405 my $string = shift;
407 # Regex for recognizing RFC 4646 well-formed tags
408 # http://www.rfc-editor.org/rfc/rfc4646.txt
410 # regexes based on : http://unicode.org/cldr/data/tools/java/org/unicode/cldr/util/data/langtagRegex.txt
411 # The structure requires no forward references, so it reverses the order.
412 # The uppercase comments are fragments copied from RFC 4646
414 # Note: the tool requires that any real "=" or "#" or ";" in the regex be escaped.
416 my $alpha = qr/[a-zA-Z]/ ; # ALPHA
417 my $digit = qr/[0-9]/ ; # DIGIT
418 my $alphanum = qr/[a-zA-Z0-9]/ ; # ALPHA / DIGIT
419 my $x = qr/[xX]/ ; # private use singleton
420 my $singleton = qr/[a-w y-z A-W Y-Z]/ ; # other singleton
421 my $s = qr/[-]/ ; # separator -- lenient parsers will use [-_]
423 # Now do the components. The structure is slightly different to allow for capturing the right components.
424 # The notation (?:....) is a non-capturing version of (...): so the "?:" can be deleted if someone doesn't care about capturing.
426 my $extlang = qr{(?: $s $alpha{3} )}x ; # *3("-" 3ALPHA)
427 my $language = qr{(?: $alpha{2,3} | $alpha{4,8} )}x ;
428 #my $language = qr{(?: $alpha{2,3}$extlang{0,3} | $alpha{4,8} )}x ; # (2*3ALPHA [ extlang ]) / 4ALPHA / 5*8ALPHA
430 my $script = qr{(?: $alpha{4} )}x ; # 4ALPHA
432 my $region = qr{(?: $alpha{2} | $digit{3} )}x ; # 2ALPHA / 3DIGIT
434 my $variantSub = qr{(?: $digit$alphanum{3} | $alphanum{5,8} )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
435 my $variant = qr{(?: $variantSub (?: $s$variantSub )* )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
437 my $extensionSub = qr{(?: $singleton (?: $s$alphanum{2,8} )+ )}x ; # singleton 1*("-" (2*8alphanum))
438 my $extension = qr{(?: $extensionSub (?: $s$extensionSub )* )}x ; # singleton 1*("-" (2*8alphanum))
440 my $privateuse = qr{(?: $x (?: $s$alphanum{1,8} )+ )}x ; # ("x"/"X") 1*("-" (1*8alphanum))
442 # Define certain grandfathered codes, since otherwise the regex is pretty useless.
443 # Since these are limited, this is safe even later changes to the registry --
444 # the only oddity is that it might change the type of the tag, and thus
445 # the results from the capturing groups.
446 # http://www.iana.org/assignments/language-subtag-registry
447 # Note that these have to be compared case insensitively, requiring (?i) below.
449 my $grandfathered = qr{(?: (?i)
450 en $s GB $s oed
451 | i $s (?: ami | bnn | default | enochian | hak | klingon | lux | mingo | navajo | pwn | tao | tay | tsu )
452 | sgn $s (?: BE $s fr | BE $s nl | CH $s de)
453 )}x;
455 # For well-formedness, we don't need the ones that would otherwise pass, so they are commented out here
457 # | art $s lojban
458 # | cel $s gaulish
459 # | en $s (?: boont | GB $s oed | scouse )
460 # | no $s (?: bok | nyn)
461 # | zh $s (?: cmn | cmn $s Hans | cmn $s Hant | gan | guoyu | hakka | min | min $s nan | wuu | xiang | yue)
463 # Here is the final breakdown, with capturing groups for each of these components
464 # The language, variants, extensions, grandfathered, and private-use may have interior '-'
466 #my $root = qr{(?: ($language) (?: $s ($script) )? 40% (?: $s ($region) )? 40% (?: $s ($variant) )? 10% (?: $s ($extension) )? 5% (?: $s ($privateuse) )? 5% ) 90% | ($grandfathered) 5% | ($privateuse) 5% };
468 $string =~ qr{^ (?:($language)) (?:$s($script))? (?:$s($region))? (?:$s($variant))? (?:$s($extension))? (?:$s($privateuse))? $}xi; # |($grandfathered) | ($privateuse) $}xi;
469 my %subtag = (
470 'rfc4646_subtag' => $string,
471 'language' => $1,
472 'script' => $2,
473 'region' => $3,
474 'variant' => $4,
475 'extension' => $5,
476 'privateuse' => $6,
478 return \%subtag;
481 # Script Direction Resources:
482 # http://www.w3.org/International/questions/qa-scripts
483 sub get_bidi {
484 my ($language_script)= @_;
485 my $dbh = C4::Context->dbh;
486 my $bidi;
487 my $sth = $dbh->prepare('SELECT bidi FROM language_script_bidi WHERE rfc4646_subtag=?');
488 $sth->execute($language_script);
489 while (my $result = $sth->fetchrow_hashref) {
490 $bidi = $result->{'bidi'};
492 return $bidi;
495 sub accept_language {
496 # referenced http://search.cpan.org/src/CGILMORE/I18N-AcceptLanguage-1.04/lib/I18N/AcceptLanguage.pm
497 my ($clientPreferences,$supportedLanguages) = @_;
498 my @languages = ();
499 if ($clientPreferences) {
500 # There should be no whitespace anways, but a cleanliness/sanity check
501 $clientPreferences =~ s/\s//g;
502 # Prepare the list of client-acceptable languages
503 foreach my $tag (split(/,/, $clientPreferences)) {
504 my ($language, $quality) = split(/\;/, $tag);
505 $quality =~ s/^q=//i if $quality;
506 $quality = 1 unless $quality;
507 next if $quality <= 0;
508 # We want to force the wildcard to be last
509 $quality = 0 if ($language eq '*');
510 # Pushing lowercase language here saves processing later
511 push(@languages, { quality => $quality,
512 language => $language,
513 lclanguage => lc($language) });
515 } else {
516 carp "accept_language(x,y) called with no clientPreferences (x).";
518 # Prepare the list of server-supported languages
519 my %supportedLanguages = ();
520 my %secondaryLanguages = ();
521 foreach my $language (@$supportedLanguages) {
522 # warn "Language supported: " . $language->{language};
523 my $subtag = $language->{rfc4646_subtag};
524 $supportedLanguages{lc($subtag)} = $subtag;
525 if ( $subtag =~ /^([^-]+)-?/ ) {
526 $secondaryLanguages{lc($1)} = $subtag;
530 # Reverse sort the list, making best quality at the front of the array
531 @languages = sort { $b->{quality} <=> $a->{quality} } @languages;
532 my $secondaryMatch = '';
533 foreach my $tag (@languages) {
534 if (exists($supportedLanguages{$tag->{lclanguage}})) {
535 # Client en-us eq server en-us
536 return $supportedLanguages{$tag->{language}} if exists($supportedLanguages{$tag->{language}});
537 return $supportedLanguages{$tag->{lclanguage}};
538 } elsif (exists($secondaryLanguages{$tag->{lclanguage}})) {
539 # Client en eq server en-us
540 return $secondaryLanguages{$tag->{language}} if exists($secondaryLanguages{$tag->{language}});
541 return $supportedLanguages{$tag->{lclanguage}};
542 } elsif ($tag->{lclanguage} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
543 # Client en-gb eq server en-us
544 $secondaryMatch = $secondaryLanguages{$1};
545 } elsif ($tag->{lclanguage} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
546 # FIXME: We just checked the exact same conditional!
547 # Client en-us eq server en
548 $secondaryMatch = $supportedLanguages{$1};
549 } elsif ($tag->{lclanguage} eq '*') {
550 # * matches every language not already specified.
551 # It doesn't care which we pick, so let's pick the default,
552 # if available, then the first in the array.
553 #return $acceptor->defaultLanguage() if $acceptor->defaultLanguage();
554 return $supportedLanguages->[0];
557 # No primary matches. Secondary? (ie, en-us requested and en supported)
558 return $secondaryMatch if $secondaryMatch;
559 return undef; # else, we got nothing.
562 =head2 getlanguage
564 Select a language based on the URL parameter 'language', a cookie,
565 syspref available languages & browser
567 =cut
569 sub getlanguage {
570 my ($cgi) = @_;
572 $cgi //= new CGI;
573 my $interface = C4::Context->interface;
574 my $theme = C4::Context->preference( ( $interface eq 'opac' ) ? 'opacthemes' : 'template' );
575 my $language;
577 my $preference_to_check =
578 $interface eq 'intranet' ? 'language' : 'opaclanguages';
579 # Get the available/valid languages list
580 my @languages = split /,/, C4::Context->preference($preference_to_check);
582 # Chose language from the URL
583 $language = $cgi->param( 'language' );
584 if ( defined $language && any { $_ eq $language } @languages) {
585 return $language;
588 # cookie
589 if ($language = $cgi->cookie('KohaOpacLanguage') ) {
590 $language =~ s/[^a-zA-Z_-]*//; # sanitize cookie
593 # HTTP_ACCEPT_LANGUAGE
594 if ( !$language && $ENV{HTTP_ACCEPT_LANGUAGE} ) {
595 $language = accept_language( $ENV{HTTP_ACCEPT_LANGUAGE},
596 getTranslatedLanguages( $interface, $theme ) );
599 # Ignore a lang not selected in sysprefs
600 if ( $language && any { $_ eq $language } @languages ) {
601 return $language;
604 # Pick the first selected syspref language
605 $language = shift @languages;
606 return $language if $language;
608 # Fall back to English if necessary
609 return 'en';
614 __END__
616 =head1 AUTHOR
618 Joshua Ferraro
620 =cut