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, $which) = @_;
116 my @enabled_languages;
118 if ($interface && $interface eq 'opac' ) {
119 @enabled_languages = split ",", C4
::Context
->preference('opaclanguages');
120 $htdocs = C4
::Context
->config('opachtdocs');
121 if ( $theme and -d
"$htdocs/$theme" ) {
122 (@languages) = _get_language_dirs
($htdocs,$theme);
125 for my $theme ( _get_themes
('opac') ) {
126 push @languages, _get_language_dirs
($htdocs,$theme);
130 elsif ($interface && $interface eq 'intranet' ) {
131 @enabled_languages = split ",", C4
::Context
->preference('language');
132 $htdocs = C4
::Context
->config('intrahtdocs');
133 if ( $theme and -d
"$htdocs/$theme" ) {
134 @languages = _get_language_dirs
($htdocs,$theme);
137 foreach my $theme ( _get_themes
('intranet') ) {
138 push @languages, _get_language_dirs
($htdocs,$theme);
143 @enabled_languages = split ",", C4
::Context
->preference('opaclanguages');
144 my $htdocs = C4
::Context
->config('intrahtdocs');
145 foreach my $theme ( _get_themes
('intranet') ) {
146 push @languages, _get_language_dirs
($htdocs,$theme);
148 $htdocs = C4
::Context
->config('opachtdocs');
149 foreach my $theme ( _get_themes
('opac') ) {
150 push @languages, _get_language_dirs
($htdocs,$theme);
153 $seen{$_}++ for @languages;
154 @languages = keys %seen;
156 return _build_languages_arrayref
(\
@languages,$current_language,\
@enabled_languages);
159 =head2 getAllLanguages
161 Returns a reference to an array of hashes:
163 my $alllanguages = getAllLanguages();
164 print "Available translated languages:\n";
165 for my $language(@$alllanguages) {
166 print "$language->{language_code}\n";
167 print "$language->{language_name}\n";
168 print "$language->{language_locale_name}\n";
171 This routine is a wrapper for getLanguages().
175 sub getAllLanguages
{
176 return getLanguages
(shift);
181 my $lang_arrayref = getLanguages([$lang[, $isFiltered]]);
183 Returns a reference to an array of hashes of languages.
185 - If no parameter is passed to the function, it returns english languages names
186 - If a $lang parameter conforming to RFC4646 syntax is passed, the function returns languages names translated in $lang
187 If a language name is not translated in $lang in database, the function returns english language name
188 - If $isFiltered is set to true, only the detail of the languages selected in system preferences AdvanceSearchLanguages is returned.
194 my $isFiltered = shift;
197 my $dbh=C4
::Context
->dbh;
198 my $default_language = 'en';
199 my $current_language = $default_language;
200 my $language_list = $isFiltered ? C4
::Context
->preference("AdvancedSearchLanguages") : undef;
202 $current_language = regex_lang_subtags
($lang)->{'language'};
204 my $sth = $dbh->prepare('SELECT * FROM language_subtag_registry WHERE type=\'language\'');
206 while (my $language_subtag_registry = $sth->fetchrow_hashref) {
208 # check if language name is stored in current language
209 my $sth4= $dbh->prepare("SELECT description FROM language_descriptions WHERE type='language' AND subtag =? AND lang = ?");
210 $sth4->execute($language_subtag_registry->{subtag
},$current_language);
211 while (my $language_desc = $sth4->fetchrow_hashref) {
212 $desc=$language_desc->{description
};
214 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 = ?");
216 $sth2->execute($language_subtag_registry->{subtag
},$current_language);
219 $sth2->execute($language_subtag_registry->{subtag
},$default_language);
221 my $sth3 = $dbh->prepare("SELECT description FROM language_descriptions WHERE type='language' AND subtag=? AND lang=?");
222 # add the correct description info
223 while (my $language_descriptions = $sth2->fetchrow_hashref) {
224 $sth3->execute($language_subtag_registry->{subtag
},$language_subtag_registry->{subtag
});
225 my $native_description;
226 while (my $description = $sth3->fetchrow_hashref) {
227 $native_description = $description->{description
};
230 # fill in the ISO6329 code
231 $language_subtag_registry->{iso639_2_code
} = $language_descriptions->{iso639_2_code
};
232 # fill in the native description of the language, as well as the current language's translation of that if it exists
233 if ($native_description) {
234 $language_subtag_registry->{language_description
} = $native_description;
235 $language_subtag_registry->{language_description
}.=" ($language_descriptions->{description})" if $language_descriptions->{description
};
238 $language_subtag_registry->{language_description
} = $language_descriptions->{description
};
241 # Do not push unless valid iso639-2 code
242 if ( $language_subtag_registry->{ iso639_2_code
} and ( !$language_list || index ( $language_list, $language_subtag_registry->{ iso639_2_code
} ) >= 0) ) {
243 push @languages_loop, $language_subtag_registry;
246 return \
@languages_loop;
251 Internal function, returns an array of all available themes.
253 (@themes) = &_get_themes('opac');
254 (@themes) = &_get_themes('intranet');
259 my $interface = shift;
262 if ( $interface && $interface eq 'intranet' ) {
263 $htdocs = C4
::Context
->config('intrahtdocs');
266 $htdocs = C4
::Context
->config('opachtdocs');
268 opendir D
, "$htdocs";
269 my @dirlist = readdir D
;
270 foreach my $directory (@dirlist) {
271 # if there's an en dir, it's a valid theme
272 -d
"$htdocs/$directory/en" and push @themes, $directory;
277 =head2 _get_language_dirs
279 Internal function, returns an array of directory names, excluding non-language directories
283 sub _get_language_dirs
{
284 my ($htdocs,$theme) = @_;
288 opendir D
, "$htdocs/$theme";
289 for my $lang_string ( readdir D
) {
290 next if $lang_string =~/^\./;
291 next if $lang_string eq 'all';
292 next if $lang_string =~/png$/;
293 next if $lang_string =~/js$/;
294 next if $lang_string =~/css$/;
295 next if $lang_string =~/CVS$/;
296 next if $lang_string =~/\.txt$/i; #Don't read the readme.txt !
297 next if $lang_string =~/img|images|famfam|js|less|lib|sound|pdf/;
298 push @lang_strings, $lang_string;
300 return (@lang_strings);
303 =head2 _build_languages_arrayref
305 Internal function for building the ref to array of hashes
307 FIXME: this could be rewritten and simplified using map
311 sub _build_languages_arrayref
{
312 my ($translated_languages,$current_language,$enabled_languages) = @_;
313 $current_language //= '';
314 my @translated_languages = @
$translated_languages;
315 my @languages_loop; # the final reference to an array of hashrefs
316 my @enabled_languages = @
$enabled_languages;
317 # how many languages are enabled, if one, take note, some contexts won't need to display it
318 my %seen_languages; # the language tags we've seen
321 my $track_language_groups;
322 my $current_language_regex = regex_lang_subtags
($current_language);
323 # Loop through the translated languages
324 for my $translated_language (@translated_languages) {
325 # separate the language string into its subtag types
326 my $language_subtags_hashref = regex_lang_subtags
($translated_language);
328 # is this language string 'enabled'?
329 for my $enabled_language (@enabled_languages) {
330 #warn "Checking out if $translated_language eq $enabled_language";
331 $language_subtags_hashref->{'enabled'} = 1 if $translated_language eq $enabled_language;
334 # group this language, key by langtag
335 $language_subtags_hashref->{'sublanguage_current'} = 1 if $translated_language eq $current_language;
336 $language_subtags_hashref->{'rfc4646_subtag'} = $translated_language;
337 $language_subtags_hashref->{'native_description'} = language_get_description
($language_subtags_hashref->{language
},$language_subtags_hashref->{language
},'language');
338 $language_subtags_hashref->{'script_description'} = language_get_description
($language_subtags_hashref->{script
},$language_subtags_hashref->{'language'},'script');
339 $language_subtags_hashref->{'region_description'} = language_get_description
($language_subtags_hashref->{region
},$language_subtags_hashref->{'language'},'region');
340 $language_subtags_hashref->{'variant_description'} = language_get_description
($language_subtags_hashref->{variant
},$language_subtags_hashref->{'language'},'variant');
341 $track_language_groups->{$language_subtags_hashref->{'language'}}++;
342 push ( @
{ $language_groups->{$language_subtags_hashref->{language
}} }, $language_subtags_hashref );
344 # $key is a language subtag like 'en'
345 while( my ($key, $value) = each %$language_groups) {
347 # is this language group enabled? are any of the languages within it enabled?
349 for my $enabled_language (@enabled_languages) {
350 my $regex_enabled_language = regex_lang_subtags
($enabled_language);
351 $enabled = 1 if $key eq ($regex_enabled_language->{language
} // '');
353 push @languages_loop, {
354 # this is only use if there is one
355 rfc4646_subtag
=> @
$value[0]->{rfc4646_subtag
},
356 native_description
=> language_get_description
($key,$key,'language'),
358 sublanguages_loop
=> $value,
359 plural
=> $track_language_groups->{$key} >1 ?
1 : 0,
360 current
=> ($current_language_regex->{language
} // '') eq $key ?
1 : 0,
361 group_enabled
=> $enabled,
364 return \
@languages_loop;
367 sub language_get_description
{
368 my ($script,$lang,$type) = @_;
369 my $dbh = C4
::Context
->dbh;
371 my $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
372 #warn "QUERY: SELECT description FROM language_descriptions WHERE subtag=$script AND lang=$lang AND type=$type";
373 $sth->execute($script,$lang,$type);
374 while (my $descriptions = $sth->fetchrow_hashref) {
375 $desc = $descriptions->{'description'};
378 $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
379 $sth->execute($script,'en',$type);
380 while (my $descriptions = $sth->fetchrow_hashref) {
381 $desc = $descriptions->{'description'};
386 =head2 regex_lang_subtags
388 This internal sub takes a string composed according to RFC 4646 as
389 an input and returns a reference to a hash containing keys and values
390 for ( language, script, region, variant, extension, privateuse )
394 sub regex_lang_subtags
{
397 # Regex for recognizing RFC 4646 well-formed tags
398 # http://www.rfc-editor.org/rfc/rfc4646.txt
400 # regexes based on : http://unicode.org/cldr/data/tools/java/org/unicode/cldr/util/data/langtagRegex.txt
401 # The structure requires no forward references, so it reverses the order.
402 # The uppercase comments are fragments copied from RFC 4646
404 # Note: the tool requires that any real "=" or "#" or ";" in the regex be escaped.
406 my $alpha = qr/[a-zA-Z]/ ; # ALPHA
407 my $digit = qr/[0-9]/ ; # DIGIT
408 my $alphanum = qr/[a-zA-Z0-9]/ ; # ALPHA / DIGIT
409 my $x = qr/[xX]/ ; # private use singleton
410 my $singleton = qr/[a-w y-z A-W Y-Z]/ ; # other singleton
411 my $s = qr/[-]/ ; # separator -- lenient parsers will use [-_]
413 # Now do the components. The structure is slightly different to allow for capturing the right components.
414 # The notation (?:....) is a non-capturing version of (...): so the "?:" can be deleted if someone doesn't care about capturing.
416 my $extlang = qr{(?: $s $alpha{3} )}x
; # *3("-" 3ALPHA)
417 my $language = qr{(?: $alpha{2,3} | $alpha{4,8} )}x
;
418 #my $language = qr{(?: $alpha{2,3}$extlang{0,3} | $alpha{4,8} )}x ; # (2*3ALPHA [ extlang ]) / 4ALPHA / 5*8ALPHA
420 my $script = qr{(?: $alpha{4} )}x
; # 4ALPHA
422 my $region = qr{(?: $alpha{2} | $digit{3} )}x
; # 2ALPHA / 3DIGIT
424 my $variantSub = qr{(?: $digit$alphanum{3} | $alphanum{5,8} )}x
; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
425 my $variant = qr{(?: $variantSub (?: $s$variantSub )* )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
427 my $extensionSub = qr{(?: $singleton (?: $s$alphanum{2,8} )+ )}x
; # singleton 1*("-" (2*8alphanum))
428 my $extension = qr{(?: $extensionSub (?: $s$extensionSub )* )}x ; # singleton 1*("-" (2*8alphanum))
430 my $privateuse = qr{(?: $x (?: $s$alphanum{1,8} )+ )}x
; # ("x"/"X") 1*("-" (1*8alphanum))
432 # Define certain grandfathered codes, since otherwise the regex is pretty useless.
433 # Since these are limited, this is safe even later changes to the registry --
434 # the only oddity is that it might change the type of the tag, and thus
435 # the results from the capturing groups.
436 # http://www.iana.org/assignments/language-subtag-registry
437 # Note that these have to be compared case insensitively, requiring (?i) below.
439 my $grandfathered = qr{(?
: (?i
)
441 | i
$s (?
: ami
| bnn
| default | enochian
| hak
| klingon
| lux
| mingo
| navajo
| pwn
| tao
| tay
| tsu
)
442 | sgn
$s (?
: BE
$s fr
| BE
$s nl
| CH
$s de
)
445 # For well-formedness, we don't need the ones that would otherwise pass, so they are commented out here
449 # | en $s (?: boont | GB $s oed | scouse )
450 # | no $s (?: bok | nyn)
451 # | zh $s (?: cmn | cmn $s Hans | cmn $s Hant | gan | guoyu | hakka | min | min $s nan | wuu | xiang | yue)
453 # Here is the final breakdown, with capturing groups for each of these components
454 # The language, variants, extensions, grandfathered, and private-use may have interior '-'
456 #my $root = qr{(?: ($language) (?: $s ($script) )? 40% (?: $s ($region) )? 40% (?: $s ($variant) )? 10% (?: $s ($extension) )? 5% (?: $s ($privateuse) )? 5% ) 90% | ($grandfathered) 5% | ($privateuse) 5% };
458 $string =~ qr{^ (?:($language)) (?:$s($script))? (?:$s($region))? (?:$s($variant))? (?:$s($extension))? (?:$s($privateuse))? $}xi; # |($grandfathered) | ($privateuse) $}xi;
460 'rfc4646_subtag' => $string,
471 # Script Direction Resources:
472 # http://www.w3.org/International/questions/qa-scripts
474 my ($language_script)= @_;
475 my $dbh = C4
::Context
->dbh;
477 my $sth = $dbh->prepare('SELECT bidi FROM language_script_bidi WHERE rfc4646_subtag=?');
478 $sth->execute($language_script);
479 while (my $result = $sth->fetchrow_hashref) {
480 $bidi = $result->{'bidi'};
485 sub accept_language
{
486 # referenced http://search.cpan.org/src/CGILMORE/I18N-AcceptLanguage-1.04/lib/I18N/AcceptLanguage.pm
487 my ($clientPreferences,$supportedLanguages) = @_;
489 if ($clientPreferences) {
490 # There should be no whitespace anways, but a cleanliness/sanity check
491 $clientPreferences =~ s/\s//g;
492 # Prepare the list of client-acceptable languages
493 foreach my $tag (split(/,/, $clientPreferences)) {
494 my ($language, $quality) = split(/\;/, $tag);
495 $quality =~ s/^q=//i if $quality;
496 $quality = 1 unless $quality;
497 next if $quality <= 0;
498 # We want to force the wildcard to be last
499 $quality = 0 if ($language eq '*');
500 # Pushing lowercase language here saves processing later
501 push(@languages, { quality
=> $quality,
502 language
=> $language,
503 lclanguage
=> lc($language) });
506 carp
"accept_language(x,y) called with no clientPreferences (x).";
508 # Prepare the list of server-supported languages
509 my %supportedLanguages = ();
510 my %secondaryLanguages = ();
511 foreach my $language (@
$supportedLanguages) {
512 # warn "Language supported: " . $language->{language};
513 my $subtag = $language->{rfc4646_subtag
};
514 $supportedLanguages{lc($subtag)} = $subtag;
515 if ( $subtag =~ /^([^-]+)-?/ ) {
516 $secondaryLanguages{lc($1)} = $subtag;
520 # Reverse sort the list, making best quality at the front of the array
521 @languages = sort { $b->{quality
} <=> $a->{quality
} } @languages;
522 my $secondaryMatch = '';
523 foreach my $tag (@languages) {
524 if (exists($supportedLanguages{$tag->{lclanguage
}})) {
525 # Client en-us eq server en-us
526 return $supportedLanguages{$tag->{language
}} if exists($supportedLanguages{$tag->{language
}});
527 return $supportedLanguages{$tag->{lclanguage
}};
528 } elsif (exists($secondaryLanguages{$tag->{lclanguage
}})) {
529 # Client en eq server en-us
530 return $secondaryLanguages{$tag->{language
}} if exists($secondaryLanguages{$tag->{language
}});
531 return $supportedLanguages{$tag->{lclanguage
}};
532 } elsif ($tag->{lclanguage
} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
533 # Client en-gb eq server en-us
534 $secondaryMatch = $secondaryLanguages{$1};
535 } elsif ($tag->{lclanguage
} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
536 # FIXME: We just checked the exact same conditional!
537 # Client en-us eq server en
538 $secondaryMatch = $supportedLanguages{$1};
539 } elsif ($tag->{lclanguage
} eq '*') {
540 # * matches every language not already specified.
541 # It doesn't care which we pick, so let's pick the default,
542 # if available, then the first in the array.
543 #return $acceptor->defaultLanguage() if $acceptor->defaultLanguage();
544 return $supportedLanguages->[0];
547 # No primary matches. Secondary? (ie, en-us requested and en supported)
548 return $secondaryMatch if $secondaryMatch;
549 return undef; # else, we got nothing.
554 Select a language based on the URL parameter 'language', a cookie,
555 syspref available languages & browser
562 my $memory_cache = Koha
::Cache
::Memory
::Lite
->get_instance();
563 my $cache_key = "getlanguage";
564 unless ( $cgi and $cgi->param('language') ) {
565 my $cached = $memory_cache->get_from_cache($cache_key);
566 return $cached if $cached;
570 my $interface = C4
::Context
->interface;
571 my $theme = C4
::Context
->preference( ( $interface eq 'opac' ) ?
'opacthemes' : 'template' );
574 my $preference_to_check =
575 $interface eq 'intranet' ?
'language' : 'opaclanguages';
576 # Get the available/valid languages list
578 my $preference_value = C4
::Context
->preference($preference_to_check);
579 if ($preference_value) {
580 @languages = split /,/, $preference_value;
583 # Chose language from the URL
584 my $cgi_param_language = $cgi->param( 'language' );
585 if ( defined $cgi_param_language && any
{ $_ eq $cgi_param_language } @languages) {
586 $language = $cgi_param_language;
590 if (not $language and my $cgi_cookie_language = $cgi->cookie('KohaOpacLanguage') ) {
591 ( $language = $cgi_cookie_language ) =~ s/[^a-zA-Z_-]*//; # sanitize cookie
594 # HTTP_ACCEPT_LANGUAGE
595 if ( !$language && $ENV{HTTP_ACCEPT_LANGUAGE
} ) {
596 $language = accept_language
( $ENV{HTTP_ACCEPT_LANGUAGE
},
597 getTranslatedLanguages
( $interface, $theme ) );
600 # Ignore a lang not selected in sysprefs
601 if ( $language && not any
{ $_ eq $language } @languages ) {
605 # Pick the first selected syspref language
606 $language = shift @languages unless $language;
608 # Fall back to English if necessary
611 $memory_cache->set_in_cache( $cache_key, $language );