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
;
31 use vars
qw(@ISA @EXPORT @EXPORT_OK %EXPORT_TAGS $DEBUG);
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) = @_;
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\'');
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
} = $language_descriptions->{iso639_2_code
};
230 # fill in the native description of the language, as well as the current language's translation of that if it exists
231 if ($native_description) {
232 $language_subtag_registry->{language_description
} = $native_description;
233 $language_subtag_registry->{language_description
}.=" ($language_descriptions->{description})" if $language_descriptions->{description
};
236 $language_subtag_registry->{language_description
} = $language_descriptions->{description
};
239 # Do not push unless valid iso639-2 code
240 if ( $language_subtag_registry->{ iso639_2_code
} and ( !$language_list || index ( $language_list, $language_subtag_registry->{ iso639_2_code
} ) >= 0) ) {
241 push @languages_loop, $language_subtag_registry;
244 return \
@languages_loop;
247 sub _get_opac_language_dirs
{
248 my ( $htdocs, $theme ) = @_;
251 if ( $theme and -d
"$htdocs/$theme" ) {
252 (@languages) = _get_language_dirs
($htdocs,$theme);
255 for my $theme ( _get_themes
('opac') ) {
256 push @languages, _get_language_dirs
($htdocs,$theme);
263 sub _get_intranet_language_dirs
{
264 my ( $htdocs, $theme ) = @_;
267 if ( $theme and -d
"$htdocs/$theme" ) {
268 @languages = _get_language_dirs
($htdocs,$theme);
271 foreach my $theme ( _get_themes
('intranet') ) {
272 push @languages, _get_language_dirs
($htdocs,$theme);
280 Internal function, returns an array of all available themes.
282 (@themes) = &_get_themes('opac');
283 (@themes) = &_get_themes('intranet');
288 my $interface = shift;
291 if ( $interface && $interface eq 'intranet' ) {
292 $htdocs = C4
::Context
->config('intrahtdocs');
295 $htdocs = C4
::Context
->config('opachtdocs');
297 opendir D
, "$htdocs";
298 my @dirlist = readdir D
;
299 foreach my $directory (@dirlist) {
300 # if there's an en dir, it's a valid theme
301 -d
"$htdocs/$directory/en" and push @themes, $directory;
306 =head2 _get_language_dirs
308 Internal function, returns an array of directory names, excluding non-language directories
312 sub _get_language_dirs
{
313 my ($htdocs,$theme) = @_;
317 opendir D
, "$htdocs/$theme";
318 for my $lang_string ( readdir D
) {
319 next if $lang_string =~/^\./;
320 next if $lang_string eq 'all';
321 next if $lang_string =~/png$/;
322 next if $lang_string =~/js$/;
323 next if $lang_string =~/css$/;
324 next if $lang_string =~/CVS$/;
325 next if $lang_string =~/\.txt$/i; #Don't read the readme.txt !
326 next if $lang_string =~/img|images|famfam|js|less|lib|sound|pdf/;
327 push @lang_strings, $lang_string;
329 return (@lang_strings);
332 =head2 _build_languages_arrayref
334 Internal function for building the ref to array of hashes
336 FIXME: this could be rewritten and simplified using map
340 sub _build_languages_arrayref
{
341 my ($translated_languages,$current_language,$enabled_languages) = @_;
342 $current_language //= '';
343 my @translated_languages = @
$translated_languages;
344 my @languages_loop; # the final reference to an array of hashrefs
345 my @enabled_languages = @
$enabled_languages;
346 # how many languages are enabled, if one, take note, some contexts won't need to display it
347 my %seen_languages; # the language tags we've seen
350 my $track_language_groups;
351 my $current_language_regex = regex_lang_subtags
($current_language);
352 # Loop through the translated languages
353 for my $translated_language (@translated_languages) {
354 # separate the language string into its subtag types
355 my $language_subtags_hashref = regex_lang_subtags
($translated_language);
357 # is this language string 'enabled'?
358 for my $enabled_language (@enabled_languages) {
359 #warn "Checking out if $translated_language eq $enabled_language";
360 $language_subtags_hashref->{'enabled'} = 1 if $translated_language eq $enabled_language;
363 # group this language, key by langtag
364 $language_subtags_hashref->{'sublanguage_current'} = 1 if $translated_language eq $current_language;
365 $language_subtags_hashref->{'rfc4646_subtag'} = $translated_language;
366 $language_subtags_hashref->{'native_description'} = language_get_description
($language_subtags_hashref->{language
},$language_subtags_hashref->{language
},'language');
367 $language_subtags_hashref->{'script_description'} = language_get_description
($language_subtags_hashref->{script
},$language_subtags_hashref->{'language'},'script');
368 $language_subtags_hashref->{'region_description'} = language_get_description
($language_subtags_hashref->{region
},$language_subtags_hashref->{'language'},'region');
369 $language_subtags_hashref->{'variant_description'} = language_get_description
($language_subtags_hashref->{variant
},$language_subtags_hashref->{'language'},'variant');
370 $track_language_groups->{$language_subtags_hashref->{'language'}}++;
371 push ( @
{ $language_groups->{$language_subtags_hashref->{language
}} }, $language_subtags_hashref );
373 # $key is a language subtag like 'en'
375 my %idx = map { $enabled_languages->[$_] => $_ } reverse 0 .. @
$enabled_languages-1;
376 my @ordered_keys = sort {
377 my $aa = $language_groups->{$a}->[0]->{rfc4646_subtag
};
378 my $bb = $language_groups->{$b}->[0]->{rfc4646_subtag
};
379 ( exists $idx{$aa} and exists $idx{$bb} and ( $idx{$aa} cmp $idx{$bb} ) )
380 || ( exists $idx{$aa} and exists $idx{$bb} )
382 } keys %$language_groups;
384 for my $key ( @ordered_keys ) {
385 my $value = $language_groups->{$key};
386 # is this language group enabled? are any of the languages within it enabled?
388 for my $enabled_language (@enabled_languages) {
389 my $regex_enabled_language = regex_lang_subtags
($enabled_language);
390 $enabled = 1 if $key eq ($regex_enabled_language->{language
} // '');
392 push @languages_loop, {
393 # this is only use if there is one
394 rfc4646_subtag
=> @
$value[0]->{rfc4646_subtag
},
395 native_description
=> language_get_description
($key,$key,'language'),
397 sublanguages_loop
=> $value,
398 plural
=> $track_language_groups->{$key} >1 ?
1 : 0,
399 current
=> ($current_language_regex->{language
} // '') eq $key ?
1 : 0,
400 group_enabled
=> $enabled,
403 return \
@languages_loop;
406 sub language_get_description
{
407 my ($script,$lang,$type) = @_;
408 my $dbh = C4
::Context
->dbh;
410 my $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
411 #warn "QUERY: SELECT description FROM language_descriptions WHERE subtag=$script AND lang=$lang AND type=$type";
412 $sth->execute($script,$lang,$type);
413 while (my $descriptions = $sth->fetchrow_hashref) {
414 $desc = $descriptions->{'description'};
417 $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
418 $sth->execute($script,'en',$type);
419 while (my $descriptions = $sth->fetchrow_hashref) {
420 $desc = $descriptions->{'description'};
425 =head2 regex_lang_subtags
427 This internal sub takes a string composed according to RFC 4646 as
428 an input and returns a reference to a hash containing keys and values
429 for ( language, script, region, variant, extension, privateuse )
433 sub regex_lang_subtags
{
436 # Regex for recognizing RFC 4646 well-formed tags
437 # http://www.rfc-editor.org/rfc/rfc4646.txt
439 # regexes based on : http://unicode.org/cldr/data/tools/java/org/unicode/cldr/util/data/langtagRegex.txt
440 # The structure requires no forward references, so it reverses the order.
441 # The uppercase comments are fragments copied from RFC 4646
443 # Note: the tool requires that any real "=" or "#" or ";" in the regex be escaped.
445 my $alpha = qr/[a-zA-Z]/ ; # ALPHA
446 my $digit = qr/[0-9]/ ; # DIGIT
447 my $alphanum = qr/[a-zA-Z0-9]/ ; # ALPHA / DIGIT
448 my $x = qr/[xX]/ ; # private use singleton
449 my $singleton = qr/[a-w y-z A-W Y-Z]/ ; # other singleton
450 my $s = qr/[-]/ ; # separator -- lenient parsers will use [-_]
452 # Now do the components. The structure is slightly different to allow for capturing the right components.
453 # The notation (?:....) is a non-capturing version of (...): so the "?:" can be deleted if someone doesn't care about capturing.
455 my $extlang = qr{(?: $s $alpha{3} )}x
; # *3("-" 3ALPHA)
456 my $language = qr{(?: $alpha{2,3} | $alpha{4,8} )}x
;
457 #my $language = qr{(?: $alpha{2,3}$extlang{0,3} | $alpha{4,8} )}x ; # (2*3ALPHA [ extlang ]) / 4ALPHA / 5*8ALPHA
459 my $script = qr{(?: $alpha{4} )}x
; # 4ALPHA
461 my $region = qr{(?: $alpha{2} | $digit{3} )}x
; # 2ALPHA / 3DIGIT
463 my $variantSub = qr{(?: $digit$alphanum{3} | $alphanum{5,8} )}x
; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
464 my $variant = qr{(?: $variantSub (?: $s$variantSub )* )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
466 my $extensionSub = qr{(?: $singleton (?: $s$alphanum{2,8} )+ )}x
; # singleton 1*("-" (2*8alphanum))
467 my $extension = qr{(?: $extensionSub (?: $s$extensionSub )* )}x ; # singleton 1*("-" (2*8alphanum))
469 my $privateuse = qr{(?: $x (?: $s$alphanum{1,8} )+ )}x
; # ("x"/"X") 1*("-" (1*8alphanum))
471 # Define certain grandfathered codes, since otherwise the regex is pretty useless.
472 # Since these are limited, this is safe even later changes to the registry --
473 # the only oddity is that it might change the type of the tag, and thus
474 # the results from the capturing groups.
475 # http://www.iana.org/assignments/language-subtag-registry
476 # Note that these have to be compared case insensitively, requiring (?i) below.
478 my $grandfathered = qr{(?
: (?i
)
480 | i
$s (?
: ami
| bnn
| default | enochian
| hak
| klingon
| lux
| mingo
| navajo
| pwn
| tao
| tay
| tsu
)
481 | sgn
$s (?
: BE
$s fr
| BE
$s nl
| CH
$s de
)
484 # For well-formedness, we don't need the ones that would otherwise pass, so they are commented out here
488 # | en $s (?: boont | GB $s oed | scouse )
489 # | no $s (?: bok | nyn)
490 # | zh $s (?: cmn | cmn $s Hans | cmn $s Hant | gan | guoyu | hakka | min | min $s nan | wuu | xiang | yue)
492 # Here is the final breakdown, with capturing groups for each of these components
493 # The language, variants, extensions, grandfathered, and private-use may have interior '-'
495 #my $root = qr{(?: ($language) (?: $s ($script) )? 40% (?: $s ($region) )? 40% (?: $s ($variant) )? 10% (?: $s ($extension) )? 5% (?: $s ($privateuse) )? 5% ) 90% | ($grandfathered) 5% | ($privateuse) 5% };
497 $string =~ qr{^ (?:($language)) (?:$s($script))? (?:$s($region))? (?:$s($variant))? (?:$s($extension))? (?:$s($privateuse))? $}xi; # |($grandfathered) | ($privateuse) $}xi;
499 'rfc4646_subtag' => $string,
510 # Script Direction Resources:
511 # http://www.w3.org/International/questions/qa-scripts
513 my ($language_script)= @_;
514 my $dbh = C4
::Context
->dbh;
516 my $sth = $dbh->prepare('SELECT bidi FROM language_script_bidi WHERE rfc4646_subtag=?');
517 $sth->execute($language_script);
518 while (my $result = $sth->fetchrow_hashref) {
519 $bidi = $result->{'bidi'};
524 sub accept_language
{
525 # referenced http://search.cpan.org/src/CGILMORE/I18N-AcceptLanguage-1.04/lib/I18N/AcceptLanguage.pm
526 my ($clientPreferences,$supportedLanguages) = @_;
528 if ($clientPreferences) {
529 # There should be no whitespace anways, but a cleanliness/sanity check
530 $clientPreferences =~ s/\s//g;
531 # Prepare the list of client-acceptable languages
532 foreach my $tag (split(/,/, $clientPreferences)) {
533 my ($language, $quality) = split(/\;/, $tag);
534 $quality =~ s/^q=//i if $quality;
535 $quality = 1 unless $quality;
536 next if $quality <= 0;
537 # We want to force the wildcard to be last
538 $quality = 0 if ($language eq '*');
539 # Pushing lowercase language here saves processing later
540 push(@languages, { quality
=> $quality,
541 language
=> $language,
542 lclanguage
=> lc($language) });
545 carp
"accept_language(x,y) called with no clientPreferences (x).";
547 # Prepare the list of server-supported languages
548 my %supportedLanguages = ();
549 my %secondaryLanguages = ();
550 foreach my $language (@
$supportedLanguages) {
551 # warn "Language supported: " . $language->{language};
552 my $subtag = $language->{rfc4646_subtag
};
553 $supportedLanguages{lc($subtag)} = $subtag;
554 if ( $subtag =~ /^([^-]+)-?/ ) {
555 $secondaryLanguages{lc($1)} = $subtag;
559 # Reverse sort the list, making best quality at the front of the array
560 @languages = sort { $b->{quality
} <=> $a->{quality
} } @languages;
561 my $secondaryMatch = '';
562 foreach my $tag (@languages) {
563 if (exists($supportedLanguages{$tag->{lclanguage
}})) {
564 # Client en-us eq server en-us
565 return $supportedLanguages{$tag->{language
}} if exists($supportedLanguages{$tag->{language
}});
566 return $supportedLanguages{$tag->{lclanguage
}};
567 } elsif (exists($secondaryLanguages{$tag->{lclanguage
}})) {
568 # Client en eq server en-us
569 return $secondaryLanguages{$tag->{language
}} if exists($secondaryLanguages{$tag->{language
}});
570 return $supportedLanguages{$tag->{lclanguage
}};
571 } elsif ($tag->{lclanguage
} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
572 # Client en-gb eq server en-us
573 $secondaryMatch = $secondaryLanguages{$1};
574 } elsif ($tag->{lclanguage
} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
575 # FIXME: We just checked the exact same conditional!
576 # Client en-us eq server en
577 $secondaryMatch = $supportedLanguages{$1};
578 } elsif ($tag->{lclanguage
} eq '*') {
579 # * matches every language not already specified.
580 # It doesn't care which we pick, so let's pick the default,
581 # if available, then the first in the array.
582 #return $acceptor->defaultLanguage() if $acceptor->defaultLanguage();
583 return $supportedLanguages->[0];
586 # No primary matches. Secondary? (ie, en-us requested and en supported)
587 return $secondaryMatch if $secondaryMatch;
588 return undef; # else, we got nothing.
593 Select a language based on the URL parameter 'language', a cookie,
594 syspref available languages & browser
601 my $memory_cache = Koha
::Cache
::Memory
::Lite
->get_instance();
602 my $cache_key = "getlanguage";
603 unless ( $cgi and $cgi->param('language') ) {
604 my $cached = $memory_cache->get_from_cache($cache_key);
605 return $cached if $cached;
609 my $interface = C4
::Context
->interface;
610 my $theme = C4
::Context
->preference( ( $interface eq 'opac' ) ?
'opacthemes' : 'template' );
613 my $preference_to_check =
614 $interface eq 'intranet' ?
'language' : 'opaclanguages';
615 # Get the available/valid languages list
617 my $preference_value = C4
::Context
->preference($preference_to_check);
618 if ($preference_value) {
619 @languages = split /,/, $preference_value;
622 # Chose language from the URL
623 my $cgi_param_language = $cgi->param( 'language' );
624 if ( defined $cgi_param_language && any
{ $_ eq $cgi_param_language } @languages) {
625 $language = $cgi_param_language;
629 if (not $language and my $cgi_cookie_language = $cgi->cookie('KohaOpacLanguage') ) {
630 ( $language = $cgi_cookie_language ) =~ s/[^a-zA-Z_-]*//; # sanitize cookie
633 # HTTP_ACCEPT_LANGUAGE
634 if ( !$language && $ENV{HTTP_ACCEPT_LANGUAGE
} ) {
635 $language = accept_language
( $ENV{HTTP_ACCEPT_LANGUAGE
},
636 getTranslatedLanguages
( $interface, $theme ) );
639 # Ignore a lang not selected in sysprefs
640 if ( $language && not any
{ $_ eq $language } @languages ) {
644 # Pick the first selected syspref language
645 $language = shift @languages unless $language;
647 # Fall back to English if necessary
650 $memory_cache->set_in_cache( $cache_key, $language );
654 =head2 get_rfc4646_from_iso639
656 Select a language rfc4646 code given an iso639 code
660 sub get_rfc4646_from_iso639
{
662 my $iso_code = shift;
663 my $rfc_subtag = Koha
::Database
->new()->schema->resultset('LanguageRfc4646ToIso639')->find({iso639_2_code
=>$iso_code});
665 return $rfc_subtag->rfc4646_subtag;