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
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.
23 #use warnings; FIXME - Bug 2505
26 use vars qw($VERSION @ISA @EXPORT @EXPORT_OK %EXPORT_TAGS $DEBUG);
29 my $servers = C4::Context->config('memcached_servers');
31 require Memoize::Memcached;
32 import Memoize::Memcached qw(memoize_memcached);
35 servers => [$servers],
36 key_prefix => C4::Context->config('memcached_namespace') || 'koha',
38 }; # cache for 10 mins
40 memoize_memcached( 'getTranslatedLanguages', memcached => $memcached );
41 memoize_memcached( 'getFrameworkLanguages', memcached => $memcached );
42 memoize_memcached( 'getAllLanguages', memcached => $memcached );
51 &getFrameworkLanguages
52 &getTranslatedLanguages
55 @EXPORT_OK = qw(getFrameworkLanguages getTranslatedLanguages getAllLanguages get_bidi regex_lang_subtags language_get_description accept_language);
61 C4::Languages - Perl Module containing language list functions for Koha
73 =head2 getFrameworkLanguages
75 Returns a reference to an array of hashes:
77 my $languages = getFrameworkLanguages();
78 for my $language(@$languages) {
79 print "$language->{language_code}\n"; # language code in iso 639-2
80 print "$language->{language_name}\n"; # language name in native script
81 print "$language->{language_locale_name}\n"; # language name in current locale
86 sub getFrameworkLanguages {
87 # get a hash with all language codes, names, and locale names
88 my $all_languages = getAllLanguages();
91 # find the available directory names
92 my $dir=C4::Context->config('intranetdir')."/installer/data/";
94 my @listdir= grep { !/^\.|CVS/ && -d "$dir/$_"} readdir(MYDIR);
97 # pull out all data for the dir names that exist
98 for my $dirname (@listdir) {
99 for my $language_set (@$all_languages) {
101 if ($dirname eq $language_set->{language_code}) {
103 'language_code'=>$dirname,
104 'language_description'=>$language_set->{language_description},
105 'native_descrition'=>$language_set->{language_native_description} }
112 =head2 getTranslatedLanguages
114 Returns a reference to an array of hashes:
116 my $languages = getTranslatedLanguages();
117 print "Available translated languages:\n";
118 for my $language(@$trlanguages) {
119 print "$language->{language_code}\n"; # language code in iso 639-2
120 print "$language->{language_name}\n"; # language name in native script
121 print "$language->{language_locale_name}\n"; # language name in current locale
126 sub getTranslatedLanguages {
127 my ($interface, $theme, $current_language, $which) = @_;
129 my $all_languages = getAllLanguages();
131 my @enabled_languages;
133 if ($interface && $interface eq 'opac' ) {
134 @enabled_languages = split ",", C4::Context->preference('opaclanguages');
135 $htdocs = C4::Context->config('opachtdocs');
136 if ( $theme and -d "$htdocs/$theme" ) {
137 (@languages) = _get_language_dirs($htdocs,$theme);
140 for my $theme ( _get_themes('opac') ) {
141 push @languages, _get_language_dirs($htdocs,$theme);
145 elsif ($interface && $interface eq 'intranet' ) {
146 @enabled_languages = split ",", C4::Context->preference('language');
147 $htdocs = C4::Context->config('intrahtdocs');
148 if ( $theme and -d "$htdocs/$theme" ) {
149 @languages = _get_language_dirs($htdocs,$theme);
152 foreach my $theme ( _get_themes('intranet') ) {
153 push @languages, _get_language_dirs($htdocs,$theme);
158 @enabled_languages = split ",", C4::Context->preference('opaclanguages');
159 my $htdocs = C4::Context->config('intrahtdocs');
160 foreach my $theme ( _get_themes('intranet') ) {
161 push @languages, _get_language_dirs($htdocs,$theme);
163 $htdocs = C4::Context->config('opachtdocs');
164 foreach my $theme ( _get_themes('opac') ) {
165 push @languages, _get_language_dirs($htdocs,$theme);
168 $seen{$_}++ for @languages;
169 @languages = keys %seen;
171 return _build_languages_arrayref($all_languages,\@languages,$current_language,\@enabled_languages);
174 =head2 getAllLanguages
176 Returns a reference to an array of hashes:
178 my $alllanguages = getAllLanguages();
179 print "Available translated languages:\n";
180 for my $language(@$alllanguages) {
181 print "$language->{language_code}\n";
182 print "$language->{language_name}\n";
183 print "$language->{language_locale_name}\n";
188 sub getAllLanguages {
190 my $dbh=C4::Context->dbh;
191 my $current_language = shift || 'en';
192 my $sth = $dbh->prepare('SELECT * FROM language_subtag_registry WHERE type=\'language\'');
194 while (my $language_subtag_registry = $sth->fetchrow_hashref) {
196 # pull out all the script descriptions for each language
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 = ?");
198 $sth2->execute($language_subtag_registry->{subtag},$current_language);
200 my $sth3 = $dbh->prepare("SELECT description FROM language_descriptions WHERE type='language' AND subtag=? AND lang=?");
202 # add the correct description info
203 while (my $language_descriptions = $sth2->fetchrow_hashref) {
204 $sth3->execute($language_subtag_registry->{subtag},$language_subtag_registry->{subtag});
205 my $native_description;
206 while (my $description = $sth3->fetchrow_hashref) {
207 $native_description = $description->{description};
210 # fill in the ISO6329 code
211 $language_subtag_registry->{iso639_2_code} = $language_descriptions->{iso639_2_code};
212 # fill in the native description of the language, as well as the current language's translation of that if it exists
213 if ($native_description) {
214 $language_subtag_registry->{language_description} = $native_description;
215 $language_subtag_registry->{language_description}.=" ($language_descriptions->{description})" if $language_descriptions->{description};
218 $language_subtag_registry->{language_description} = $language_descriptions->{description};
221 push @languages_loop, $language_subtag_registry;
223 return \@languages_loop;
228 Internal function, returns an array of all available themes.
230 (@themes) = &_get_themes('opac');
231 (@themes) = &_get_themes('intranet');
236 my $interface = shift;
239 if ( $interface eq 'intranet' ) {
240 $htdocs = C4::Context->config('intrahtdocs');
243 $htdocs = C4::Context->config('opachtdocs');
245 opendir D, "$htdocs";
246 my @dirlist = readdir D;
247 foreach my $directory (@dirlist) {
248 # if there's an en dir, it's a valid theme
249 -d "$htdocs/$directory/en" and push @themes, $directory;
254 =head2 _get_language_dirs
256 Internal function, returns an array of directory names, excluding non-language directories
260 sub _get_language_dirs {
261 my ($htdocs,$theme) = @_;
263 opendir D, "$htdocs/$theme";
264 for my $lang_string ( readdir D ) {
265 next if $lang_string =~/^\./;
266 next if $lang_string eq 'all';
267 next if $lang_string =~/png$/;
268 next if $lang_string =~/css$/;
269 next if $lang_string =~/CVS$/;
270 next if $lang_string =~/\.txt$/i; #Don't read the readme.txt !
271 next if $lang_string =~/img|images|famfam|sound|pdf/;
272 push @lang_strings, $lang_string;
274 return (@lang_strings);
277 =head2 _build_languages_arrayref
279 Internal function for building the ref to array of hashes
281 FIXME: this could be rewritten and simplified using map
285 sub _build_languages_arrayref {
286 my ($all_languages,$translated_languages,$current_language,$enabled_languages) = @_;
287 my @translated_languages = @$translated_languages;
288 my @languages_loop; # the final reference to an array of hashrefs
289 my @enabled_languages = @$enabled_languages;
290 # how many languages are enabled, if one, take note, some contexts won't need to display it
291 my %seen_languages; # the language tags we've seen
294 my $track_language_groups;
295 my $current_language_regex = regex_lang_subtags($current_language);
296 # Loop through the translated languages
297 for my $translated_language (@translated_languages) {
298 # separate the language string into its subtag types
299 my $language_subtags_hashref = regex_lang_subtags($translated_language);
301 # is this language string 'enabled'?
302 for my $enabled_language (@enabled_languages) {
303 #warn "Checking out if $translated_language eq $enabled_language";
304 $language_subtags_hashref->{'enabled'} = 1 if $translated_language eq $enabled_language;
307 # group this language, key by langtag
308 $language_subtags_hashref->{'sublanguage_current'} = 1 if $translated_language eq $current_language;
309 $language_subtags_hashref->{'rfc4646_subtag'} = $translated_language;
310 $language_subtags_hashref->{'native_description'} = language_get_description($language_subtags_hashref->{language},$language_subtags_hashref->{language},'language');
311 $language_subtags_hashref->{'script_description'} = language_get_description($language_subtags_hashref->{script},$language_subtags_hashref->{'language'},'script');
312 $language_subtags_hashref->{'region_description'} = language_get_description($language_subtags_hashref->{region},$language_subtags_hashref->{'language'},'region');
313 $language_subtags_hashref->{'variant_description'} = language_get_description($language_subtags_hashref->{variant},$language_subtags_hashref->{'language'},'variant');
314 $track_language_groups->{$language_subtags_hashref->{'language'}}++;
315 push ( @{ $language_groups->{$language_subtags_hashref->{language}} }, $language_subtags_hashref );
317 # $key is a language subtag like 'en'
318 while( my ($key, $value) = each %$language_groups) {
320 # is this language group enabled? are any of the languages within it enabled?
322 for my $enabled_language (@enabled_languages) {
323 my $regex_enabled_language = regex_lang_subtags($enabled_language);
324 $enabled = 1 if $key eq $regex_enabled_language->{language};
326 push @languages_loop, {
327 # this is only use if there is one
328 rfc4646_subtag => @$value[0]->{rfc4646_subtag},
329 native_description => language_get_description($key,$key,'language'),
331 sublanguages_loop => $value,
332 plural => $track_language_groups->{$key} >1 ? 1 : 0,
333 current => $current_language_regex->{language} eq $key ? 1 : 0,
334 group_enabled => $enabled,
337 return \@languages_loop;
340 sub language_get_description {
341 my ($script,$lang,$type) = @_;
342 my $dbh = C4::Context->dbh;
344 my $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
345 #warn "QUERY: SELECT description FROM language_descriptions WHERE subtag=$script AND lang=$lang AND type=$type";
346 $sth->execute($script,$lang,$type);
347 while (my $descriptions = $sth->fetchrow_hashref) {
348 $desc = $descriptions->{'description'};
351 $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
352 $sth->execute($script,'en',$type);
353 while (my $descriptions = $sth->fetchrow_hashref) {
354 $desc = $descriptions->{'description'};
359 =head2 regex_lang_subtags
361 This internal sub takes a string composed according to RFC 4646 as
362 an input and returns a reference to a hash containing keys and values
363 for ( language, script, region, variant, extension, privateuse )
367 sub regex_lang_subtags {
370 # Regex for recognizing RFC 4646 well-formed tags
371 # http://www.rfc-editor.org/rfc/rfc4646.txt
373 # regexes based on : http://unicode.org/cldr/data/tools/java/org/unicode/cldr/util/data/langtagRegex.txt
374 # The structure requires no forward references, so it reverses the order.
375 # The uppercase comments are fragments copied from RFC 4646
377 # Note: the tool requires that any real "=" or "#" or ";" in the regex be escaped.
379 my $alpha = qr/[a-zA-Z]/ ; # ALPHA
380 my $digit = qr/[0-9]/ ; # DIGIT
381 my $alphanum = qr/[a-zA-Z0-9]/ ; # ALPHA / DIGIT
382 my $x = qr/[xX]/ ; # private use singleton
383 my $singleton = qr/[a-w y-z A-W Y-Z]/ ; # other singleton
384 my $s = qr/[-]/ ; # separator -- lenient parsers will use [-_]
386 # Now do the components. The structure is slightly different to allow for capturing the right components.
387 # The notation (?:....) is a non-capturing version of (...): so the "?:" can be deleted if someone doesn't care about capturing.
389 my $extlang = qr{(?: $s $alpha{3} )}x ; # *3("-" 3ALPHA)
390 my $language = qr{(?: $alpha{2,3} | $alpha{4,8} )}x ;
391 #my $language = qr{(?: $alpha{2,3}$extlang{0,3} | $alpha{4,8} )}x ; # (2*3ALPHA [ extlang ]) / 4ALPHA / 5*8ALPHA
393 my $script = qr{(?: $alpha{4} )}x ; # 4ALPHA
395 my $region = qr{(?: $alpha{2} | $digit{3} )}x ; # 2ALPHA / 3DIGIT
397 my $variantSub = qr{(?: $digit$alphanum{3} | $alphanum{5,8} )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
398 my $variant = qr{(?: $variantSub (?: $s$variantSub )* )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
400 my $extensionSub = qr{(?: $singleton (?: $s$alphanum{2,8} )+ )}x ; # singleton 1*("-" (2*8alphanum))
401 my $extension = qr{(?: $extensionSub (?: $s$extensionSub )* )}x ; # singleton 1*("-" (2*8alphanum))
403 my $privateuse = qr{(?: $x (?: $s$alphanum{1,8} )+ )}x ; # ("x"/"X") 1*("-" (1*8alphanum))
405 # Define certain grandfathered codes, since otherwise the regex is pretty useless.
406 # Since these are limited, this is safe even later changes to the registry --
407 # the only oddity is that it might change the type of the tag, and thus
408 # the results from the capturing groups.
409 # http://www.iana.org/assignments/language-subtag-registry
410 # Note that these have to be compared case insensitively, requiring (?i) below.
412 my $grandfathered = qr{(?: (?i)
414 | i $s (?: ami | bnn | default | enochian | hak | klingon | lux | mingo | navajo | pwn | tao | tay | tsu )
415 | sgn $s (?: BE $s fr | BE $s nl | CH $s de)
418 # For well-formedness, we don't need the ones that would otherwise pass, so they are commented out here
422 # | en $s (?: boont | GB $s oed | scouse )
423 # | no $s (?: bok | nyn)
424 # | zh $s (?: cmn | cmn $s Hans | cmn $s Hant | gan | guoyu | hakka | min | min $s nan | wuu | xiang | yue)
426 # Here is the final breakdown, with capturing groups for each of these components
427 # The language, variants, extensions, grandfathered, and private-use may have interior '-'
429 #my $root = qr{(?: ($language) (?: $s ($script) )? 40% (?: $s ($region) )? 40% (?: $s ($variant) )? 10% (?: $s ($extension) )? 5% (?: $s ($privateuse) )? 5% ) 90% | ($grandfathered) 5% | ($privateuse) 5% };
431 $string =~ qr{^ (?:($language)) (?:$s($script))? (?:$s($region))? (?:$s($variant))? (?:$s($extension))? (?:$s($privateuse))? $}xi; # |($grandfathered) | ($privateuse) $}xi;
433 'rfc4646_subtag' => $string,
444 # Script Direction Resources:
445 # http://www.w3.org/International/questions/qa-scripts
447 my ($language_script)= @_;
448 my $dbh = C4::Context->dbh;
450 my $sth = $dbh->prepare('SELECT bidi FROM language_script_bidi WHERE rfc4646_subtag=?');
451 $sth->execute($language_script);
452 while (my $result = $sth->fetchrow_hashref) {
453 $bidi = $result->{'bidi'};
458 sub accept_language {
459 # referenced http://search.cpan.org/src/CGILMORE/I18N-AcceptLanguage-1.04/lib/I18N/AcceptLanguage.pm
460 my ($clientPreferences,$supportedLanguages) = @_;
462 if ($clientPreferences) {
463 # There should be no whitespace anways, but a cleanliness/sanity check
464 $clientPreferences =~ s/\s//g;
465 # Prepare the list of client-acceptable languages
466 foreach my $tag (split(/,/, $clientPreferences)) {
467 my ($language, $quality) = split(/\;/, $tag);
468 $quality =~ s/^q=//i if $quality;
469 $quality = 1 unless $quality;
470 next if $quality <= 0;
471 # We want to force the wildcard to be last
472 $quality = 0 if ($language eq '*');
473 # Pushing lowercase language here saves processing later
474 push(@languages, { quality => $quality,
475 language => $language,
476 lclanguage => lc($language) });
479 carp "accept_language(x,y) called with no clientPreferences (x).";
481 # Prepare the list of server-supported languages
482 my %supportedLanguages = ();
483 my %secondaryLanguages = ();
484 foreach my $language (@$supportedLanguages) {
485 # warn "Language supported: " . $language->{language};
486 my $subtag = $language->{rfc4646_subtag};
487 $supportedLanguages{lc($subtag)} = $subtag;
488 if ( $subtag =~ /^([^-]+)-?/ ) {
489 $secondaryLanguages{lc($1)} = $subtag;
493 # Reverse sort the list, making best quality at the front of the array
494 @languages = sort { $b->{quality} <=> $a->{quality} } @languages;
495 my $secondaryMatch = '';
496 foreach my $tag (@languages) {
497 if (exists($supportedLanguages{$tag->{lclanguage}})) {
498 # Client en-us eq server en-us
499 return $supportedLanguages{$tag->{language}} if exists($supportedLanguages{$tag->{language}});
500 return $supportedLanguages{$tag->{lclanguage}};
501 } elsif (exists($secondaryLanguages{$tag->{lclanguage}})) {
502 # Client en eq server en-us
503 return $secondaryLanguages{$tag->{language}} if exists($secondaryLanguages{$tag->{language}});
504 return $supportedLanguages{$tag->{lclanguage}};
505 } elsif ($tag->{lclanguage} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
506 # Client en-gb eq server en-us
507 $secondaryMatch = $secondaryLanguages{$1};
508 } elsif ($tag->{lclanguage} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
509 # FIXME: We just checked the exact same conditional!
510 # Client en-us eq server en
511 $secondaryMatch = $supportedLanguages{$1};
512 } elsif ($tag->{lclanguage} eq '*') {
513 # * matches every language not already specified.
514 # It doesn't care which we pick, so let's pick the default,
515 # if available, then the first in the array.
516 #return $acceptor->defaultLanguage() if $acceptor->defaultLanguage();
517 return $supportedLanguages->[0];
520 # No primary matches. Secondary? (ie, en-us requested and en supported)
521 return $secondaryMatch if $secondaryMatch;
522 return undef; # else, we got nothing.