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 with
18 # Koha; if not, write to the Free Software Foundation, Inc., 59 Temple Place,
19 # Suite 330, Boston, MA 02111-1307 USA
23 #use warnings; #FIXME: turn off warnings before release
26 use vars qw($VERSION @ISA @EXPORT @EXPORT_OK %EXPORT_TAGS $DEBUG);
29 require Memoize::Memcached;
30 import Memoize::Memcached qw(memoize_memcached);
33 servers => [ C4::Context->config('memcached_servers') ],
34 key_prefix => C4::Context->config('memcached_namespace'),
37 memoize_memcached('getTranslatedLanguages', memcached => $memcached, expire_time => 600); #cache for 10 minutes
38 memoize_memcached('getFrameworkLanguages' , memcached => $memcached, expire_time => 600);
39 memoize_memcached('getAllLanguages', memcached => $memcached, expire_time => 600);
47 &getFrameworkLanguages
48 &getTranslatedLanguages
51 @EXPORT_OK = qw(getFrameworkLanguages getTranslatedLanguages getAllLanguages get_bidi regex_lang_subtags language_get_description accept_language);
57 C4::Languages - Perl Module containing language list functions for Koha
67 =head2 getFrameworkLanguages
69 Returns a reference to an array of hashes:
71 my $languages = getFrameworkLanguages();
72 for my $language(@$languages) {
73 print "$language->{language_code}\n"; # language code in iso 639-2
74 print "$language->{language_name}\n"; # language name in native script
75 print "$language->{language_locale_name}\n"; # language name in current locale
80 sub getFrameworkLanguages {
81 # get a hash with all language codes, names, and locale names
82 my $all_languages = getAllLanguages();
85 # find the available directory names
86 my $dir=C4::Context->config('intranetdir')."/installer/data/";
88 my @listdir= grep { !/^\.|CVS/ && -d "$dir/$_"} readdir(MYDIR);
91 # pull out all data for the dir names that exist
92 for my $dirname (@listdir) {
93 for my $language_set (@$all_languages) {
95 if ($dirname eq $language_set->{language_code}) {
97 'language_code'=>$dirname,
98 'language_description'=>$language_set->{language_description},
99 'native_descrition'=>$language_set->{language_native_description} }
106 =head2 getTranslatedLanguages
108 Returns a reference to an array of hashes:
110 my $languages = getTranslatedLanguages();
111 print "Available translated languages:\n";
112 for my $language(@$trlanguages) {
113 print "$language->{language_code}\n"; # language code in iso 639-2
114 print "$language->{language_name}\n"; # language name in native script
115 print "$language->{language_locale_name}\n"; # language name in current locale
120 sub getTranslatedLanguages {
121 my ($interface, $theme, $current_language, $which) = @_;
123 my $all_languages = getAllLanguages();
125 my @enabled_languages;
127 if ($interface && $interface eq 'opac' ) {
128 @enabled_languages = split ",", C4::Context->preference('opaclanguages');
129 $htdocs = C4::Context->config('opachtdocs');
130 if ( $theme and -d "$htdocs/$theme" ) {
131 (@languages) = _get_language_dirs($htdocs,$theme);
134 for my $theme ( _get_themes('opac') ) {
135 push @languages, _get_language_dirs($htdocs,$theme);
139 elsif ($interface && $interface eq 'intranet' ) {
140 @enabled_languages = split ",", C4::Context->preference('language');
141 $htdocs = C4::Context->config('intrahtdocs');
142 if ( $theme and -d "$htdocs/$theme" ) {
143 @languages = _get_language_dirs($htdocs,$theme);
146 foreach my $theme ( _get_themes('intranet') ) {
147 push @languages, _get_language_dirs($htdocs,$theme);
152 @enabled_languages = split ",", C4::Context->preference('opaclanguages');
153 my $htdocs = C4::Context->config('intrahtdocs');
154 foreach my $theme ( _get_themes('intranet') ) {
155 push @languages, _get_language_dirs($htdocs,$theme);
157 $htdocs = C4::Context->config('opachtdocs');
158 foreach my $theme ( _get_themes('opac') ) {
159 push @languages, _get_language_dirs($htdocs,$theme);
162 $seen{$_}++ for @languages;
163 @languages = keys %seen;
165 return _build_languages_arrayref($all_languages,\@languages,$current_language,\@enabled_languages);
168 =head2 getAllLanguages
170 Returns a reference to an array of hashes:
172 my $alllanguages = getAllLanguages();
173 print "Available translated languages:\n";
174 for my $language(@$alllanguages) {
175 print "$language->{language_code}\n";
176 print "$language->{language_name}\n";
177 print "$language->{language_locale_name}\n";
182 sub getAllLanguages {
184 my $dbh=C4::Context->dbh;
185 my $current_language = shift || 'en';
186 my $sth = $dbh->prepare('SELECT * FROM language_subtag_registry WHERE type=\'language\'');
188 while (my $language_subtag_registry = $sth->fetchrow_hashref) {
190 # pull out all the script descriptions for each language
191 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 = ?");
192 $sth2->execute($language_subtag_registry->{subtag},$current_language);
194 my $sth3 = $dbh->prepare("SELECT description FROM language_descriptions WHERE type='language' AND subtag=? AND lang=?");
196 # add the correct description info
197 while (my $language_descriptions = $sth2->fetchrow_hashref) {
198 $sth3->execute($language_subtag_registry->{subtag},$language_subtag_registry->{subtag});
199 my $native_description;
200 while (my $description = $sth3->fetchrow_hashref) {
201 $native_description = $description->{description};
204 # fill in the ISO6329 code
205 $language_subtag_registry->{iso639_2_code} = $language_descriptions->{iso639_2_code};
206 # fill in the native description of the language, as well as the current language's translation of that if it exists
207 if ($native_description) {
208 $language_subtag_registry->{language_description} = $native_description;
209 $language_subtag_registry->{language_description}.=" ($language_descriptions->{description})" if $language_descriptions->{description};
212 $language_subtag_registry->{language_description} = $language_descriptions->{description};
215 push @languages_loop, $language_subtag_registry;
217 return \@languages_loop;
222 Internal function, returns an array of all available themes.
224 (@themes) = &_get_themes('opac');
225 (@themes) = &_get_themes('intranet');
230 my $interface = shift;
233 if ( $interface eq 'intranet' ) {
234 $htdocs = C4::Context->config('intrahtdocs');
237 $htdocs = C4::Context->config('opachtdocs');
239 opendir D, "$htdocs";
240 my @dirlist = readdir D;
241 foreach my $directory (@dirlist) {
242 # if there's an en dir, it's a valid theme
243 -d "$htdocs/$directory/en" and push @themes, $directory;
248 =head2 _get_language_dirs
250 Internal function, returns an array of directory names, excluding non-language directories
254 sub _get_language_dirs {
255 my ($htdocs,$theme) = @_;
257 opendir D, "$htdocs/$theme";
258 for my $lang_string ( readdir D ) {
259 next if $lang_string =~/^\./;
260 next if $lang_string eq 'all';
261 next if $lang_string =~/png$/;
262 next if $lang_string =~/css$/;
263 next if $lang_string =~/CVS$/;
264 next if $lang_string =~/\.txt$/i; #Don't read the readme.txt !
265 next if $lang_string =~/img|images|famfam/;
266 push @lang_strings, $lang_string;
268 return (@lang_strings);
271 =head2 _build_languages_arrayref
273 Internal function for building the ref to array of hashes
275 FIXME: this could be rewritten and simplified using map
279 sub _build_languages_arrayref {
280 my ($all_languages,$translated_languages,$current_language,$enabled_languages) = @_;
281 my @translated_languages = @$translated_languages;
282 my @languages_loop; # the final reference to an array of hashrefs
283 my @enabled_languages = @$enabled_languages;
284 # how many languages are enabled, if one, take note, some contexts won't need to display it
285 my %seen_languages; # the language tags we've seen
288 my $track_language_groups;
289 my $current_language_regex = regex_lang_subtags($current_language);
290 # Loop through the translated languages
291 for my $translated_language (@translated_languages) {
292 # separate the language string into its subtag types
293 my $language_subtags_hashref = regex_lang_subtags($translated_language);
295 # is this language string 'enabled'?
296 for my $enabled_language (@enabled_languages) {
297 #warn "Checking out if $translated_language eq $enabled_language";
298 $language_subtags_hashref->{'enabled'} = 1 if $translated_language eq $enabled_language;
301 # group this language, key by langtag
302 $language_subtags_hashref->{'sublanguage_current'} = 1 if $translated_language eq $current_language;
303 $language_subtags_hashref->{'rfc4646_subtag'} = $translated_language;
304 $language_subtags_hashref->{'native_description'} = language_get_description($language_subtags_hashref->{language},$language_subtags_hashref->{language},'language');
305 $language_subtags_hashref->{'script_description'} = language_get_description($language_subtags_hashref->{script},$language_subtags_hashref->{'language'},'script');
306 $language_subtags_hashref->{'region_description'} = language_get_description($language_subtags_hashref->{region},$language_subtags_hashref->{'language'},'region');
307 $language_subtags_hashref->{'variant_description'} = language_get_description($language_subtags_hashref->{variant},$language_subtags_hashref->{'language'},'variant');
308 $track_language_groups->{$language_subtags_hashref->{'language'}}++;
309 push ( @{ $language_groups->{$language_subtags_hashref->{language}} }, $language_subtags_hashref );
311 # $key is a language subtag like 'en'
312 while( my ($key, $value) = each %$language_groups) {
314 # is this language group enabled? are any of the languages within it enabled?
316 for my $enabled_language (@enabled_languages) {
317 my $regex_enabled_language = regex_lang_subtags($enabled_language);
318 $enabled = 1 if $key eq $regex_enabled_language->{language};
320 push @languages_loop, {
321 # this is only use if there is one
322 rfc4646_subtag => @$value[0]->{rfc4646_subtag},
323 native_description => language_get_description($key,$key,'language'),
325 sublanguages_loop => $value,
326 plural => $track_language_groups->{$key} >1 ? 1 : 0,
327 current => $current_language_regex->{language} eq $key ? 1 : 0,
328 group_enabled => $enabled,
331 return \@languages_loop;
334 sub language_get_description {
335 my ($script,$lang,$type) = @_;
336 my $dbh = C4::Context->dbh;
338 my $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
339 #warn "QUERY: SELECT description FROM language_descriptions WHERE subtag=$script AND lang=$lang AND type=$type";
340 $sth->execute($script,$lang,$type);
341 while (my $descriptions = $sth->fetchrow_hashref) {
342 $desc = $descriptions->{'description'};
345 $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
346 $sth->execute($script,'en',$type);
347 while (my $descriptions = $sth->fetchrow_hashref) {
348 $desc = $descriptions->{'description'};
353 =head2 regex_lang_subtags
355 This internal sub takes a string composed according to RFC 4646 as
356 an input and returns a reference to a hash containing keys and values
357 for ( language, script, region, variant, extension, privateuse )
361 sub regex_lang_subtags {
364 # Regex for recognizing RFC 4646 well-formed tags
365 # http://www.rfc-editor.org/rfc/rfc4646.txt
367 # regexes based on : http://unicode.org/cldr/data/tools/java/org/unicode/cldr/util/data/langtagRegex.txt
368 # The structure requires no forward references, so it reverses the order.
369 # The uppercase comments are fragments copied from RFC 4646
371 # Note: the tool requires that any real "=" or "#" or ";" in the regex be escaped.
373 my $alpha = qr/[a-zA-Z]/ ; # ALPHA
374 my $digit = qr/[0-9]/ ; # DIGIT
375 my $alphanum = qr/[a-zA-Z0-9]/ ; # ALPHA / DIGIT
376 my $x = qr/[xX]/ ; # private use singleton
377 my $singleton = qr/[a-w y-z A-W Y-Z]/ ; # other singleton
378 my $s = qr/[-]/ ; # separator -- lenient parsers will use [-_]
380 # Now do the components. The structure is slightly different to allow for capturing the right components.
381 # The notation (?:....) is a non-capturing version of (...): so the "?:" can be deleted if someone doesn't care about capturing.
383 my $extlang = qr{(?: $s $alpha{3} )}x ; # *3("-" 3ALPHA)
384 my $language = qr{(?: $alpha{2,3} | $alpha{4,8} )}x ;
385 #my $language = qr{(?: $alpha{2,3}$extlang{0,3} | $alpha{4,8} )}x ; # (2*3ALPHA [ extlang ]) / 4ALPHA / 5*8ALPHA
387 my $script = qr{(?: $alpha{4} )}x ; # 4ALPHA
389 my $region = qr{(?: $alpha{2} | $digit{3} )}x ; # 2ALPHA / 3DIGIT
391 my $variantSub = qr{(?: $digit$alphanum{3} | $alphanum{5,8} )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
392 my $variant = qr{(?: $variantSub (?: $s$variantSub )* )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
394 my $extensionSub = qr{(?: $singleton (?: $s$alphanum{2,8} )+ )}x ; # singleton 1*("-" (2*8alphanum))
395 my $extension = qr{(?: $extensionSub (?: $s$extensionSub )* )}x ; # singleton 1*("-" (2*8alphanum))
397 my $privateuse = qr{(?: $x (?: $s$alphanum{1,8} )+ )}x ; # ("x"/"X") 1*("-" (1*8alphanum))
399 # Define certain grandfathered codes, since otherwise the regex is pretty useless.
400 # Since these are limited, this is safe even later changes to the registry --
401 # the only oddity is that it might change the type of the tag, and thus
402 # the results from the capturing groups.
403 # http://www.iana.org/assignments/language-subtag-registry
404 # Note that these have to be compared case insensitively, requiring (?i) below.
406 my $grandfathered = qr{(?: (?i)
408 | i $s (?: ami | bnn | default | enochian | hak | klingon | lux | mingo | navajo | pwn | tao | tay | tsu )
409 | sgn $s (?: BE $s fr | BE $s nl | CH $s de)
412 # For well-formedness, we don't need the ones that would otherwise pass, so they are commented out here
416 # | en $s (?: boont | GB $s oed | scouse )
417 # | no $s (?: bok | nyn)
418 # | zh $s (?: cmn | cmn $s Hans | cmn $s Hant | gan | guoyu | hakka | min | min $s nan | wuu | xiang | yue)
420 # Here is the final breakdown, with capturing groups for each of these components
421 # The language, variants, extensions, grandfathered, and private-use may have interior '-'
423 #my $root = qr{(?: ($language) (?: $s ($script) )? 40% (?: $s ($region) )? 40% (?: $s ($variant) )? 10% (?: $s ($extension) )? 5% (?: $s ($privateuse) )? 5% ) 90% | ($grandfathered) 5% | ($privateuse) 5% };
425 $string =~ qr{^ (?:($language)) (?:$s($script))? (?:$s($region))? (?:$s($variant))? (?:$s($extension))? (?:$s($privateuse))? $}xi; # |($grandfathered) | ($privateuse) $}xi;
427 'rfc4646_subtag' => $string,
438 # Script Direction Resources:
439 # http://www.w3.org/International/questions/qa-scripts
441 my ($language_script)= @_;
442 my $dbh = C4::Context->dbh;
444 my $sth = $dbh->prepare('SELECT bidi FROM language_script_bidi WHERE rfc4646_subtag=?');
445 $sth->execute($language_script);
446 while (my $result = $sth->fetchrow_hashref) {
447 $bidi = $result->{'bidi'};
452 sub accept_language {
453 # referenced http://search.cpan.org/src/CGILMORE/I18N-AcceptLanguage-1.04/lib/I18N/AcceptLanguage.pm
454 # FIXME: since this is only used in Output.pm as of Jan 8 2008, maybe it should be IN Output.pm
455 my ($clientPreferences,$supportedLanguages) = @_;
457 if ($clientPreferences) {
458 # There should be no whitespace anways, but a cleanliness/sanity check
459 $clientPreferences =~ s/\s//g;
460 # Prepare the list of client-acceptable languages
461 foreach my $tag (split(/,/, $clientPreferences)) {
462 my ($language, $quality) = split(/\;/, $tag);
463 $quality =~ s/^q=//i if $quality;
464 $quality = 1 unless $quality;
465 next if $quality <= 0;
466 # We want to force the wildcard to be last
467 $quality = 0 if ($language eq '*');
468 # Pushing lowercase language here saves processing later
469 push(@languages, { quality => $quality,
470 language => $language,
471 lclanguage => lc($language) });
474 carp "accept_language(x,y) called with no clientPreferences (x).";
476 # Prepare the list of server-supported languages
477 my %supportedLanguages = ();
478 my %secondaryLanguages = ();
479 foreach my $language (@$supportedLanguages) {
480 # warn "Language supported: " . $language->{language};
481 my $subtag = $language->{rfc4646_subtag};
482 $supportedLanguages{lc($subtag)} = $subtag;
483 if ( $subtag =~ /^([^-]+)-?/ ) {
484 $secondaryLanguages{lc($1)} = $subtag;
488 # Reverse sort the list, making best quality at the front of the array
489 @languages = sort { $b->{quality} <=> $a->{quality} } @languages;
490 my $secondaryMatch = '';
491 foreach my $tag (@languages) {
492 if (exists($supportedLanguages{$tag->{lclanguage}})) {
493 # Client en-us eq server en-us
494 return $supportedLanguages{$tag->{language}} if exists($supportedLanguages{$tag->{language}});
495 return $supportedLanguages{$tag->{lclanguage}};
496 } elsif (exists($secondaryLanguages{$tag->{lclanguage}})) {
497 # Client en eq server en-us
498 return $secondaryLanguages{$tag->{language}} if exists($secondaryLanguages{$tag->{language}});
499 return $supportedLanguages{$tag->{lclanguage}};
500 } elsif ($tag->{lclanguage} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
501 # Client en-gb eq server en-us
502 $secondaryMatch = $secondaryLanguages{$1};
503 } elsif ($tag->{lclanguage} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
504 # FIXME: We just checked the exact same conditional!
505 # Client en-us eq server en
506 $secondaryMatch = $supportedLanguages{$1};
507 } elsif ($tag->{lclanguage} eq '*') {
508 # * matches every language not already specified.
509 # It doesn't care which we pick, so let's pick the default,
510 # if available, then the first in the array.
511 #return $acceptor->defaultLanguage() if $acceptor->defaultLanguage();
512 return $supportedLanguages->[0];
515 # No primary matches. Secondary? (ie, en-us requested and en supported)
516 return $secondaryMatch if $secondaryMatch;
517 return undef; # else, we got nothing.