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 if (C4::Context->ismemcached) {
30 require Memoize::Memcached;
31 import Memoize::Memcached qw(memoize_memcached);
33 memoize_memcached('getTranslatedLanguages', memcached => C4::Context->memcached);
34 memoize_memcached('getFrameworkLanguages' , memcached => C4::Context->memcached);
35 memoize_memcached('getAllLanguages', memcached => C4::Context->memcached);
40 $VERSION = 3.07.00.049;
44 &getFrameworkLanguages
45 &getTranslatedLanguages
48 @EXPORT_OK = qw(getFrameworkLanguages getTranslatedLanguages getAllLanguages get_bidi regex_lang_subtags language_get_description accept_language);
54 C4::Languages - Perl Module containing language list functions for Koha
66 =head2 getFrameworkLanguages
68 Returns a reference to an array of hashes:
70 my $languages = getFrameworkLanguages();
71 for my $language(@$languages) {
72 print "$language->{language_code}\n"; # language code in iso 639-2
73 print "$language->{language_name}\n"; # language name in native script
74 print "$language->{language_locale_name}\n"; # language name in current locale
79 sub getFrameworkLanguages {
80 # get a hash with all language codes, names, and locale names
81 my $all_languages = getAllLanguages();
84 # find the available directory names
85 my $dir=C4::Context->config('intranetdir')."/installer/data/";
87 my @listdir= grep { !/^\.|CVS/ && -d "$dir/$_"} readdir(MYDIR);
90 # pull out all data for the dir names that exist
91 for my $dirname (@listdir) {
92 for my $language_set (@$all_languages) {
94 if ($dirname eq $language_set->{language_code}) {
96 'language_code'=>$dirname,
97 'language_description'=>$language_set->{language_description},
98 'native_descrition'=>$language_set->{language_native_description} }
105 =head2 getTranslatedLanguages
107 Returns a reference to an array of hashes:
109 my $languages = getTranslatedLanguages();
110 print "Available translated languages:\n";
111 for my $language(@$trlanguages) {
112 print "$language->{language_code}\n"; # language code in iso 639-2
113 print "$language->{language_name}\n"; # language name in native script
114 print "$language->{language_locale_name}\n"; # language name in current locale
119 sub getTranslatedLanguages {
120 my ($interface, $theme, $current_language, $which) = @_;
123 my @enabled_languages;
125 if ($interface && $interface eq 'opac' ) {
126 @enabled_languages = split ",", C4::Context->preference('opaclanguages');
127 $htdocs = C4::Context->config('opachtdocs');
128 if ( $theme and -d "$htdocs/$theme" ) {
129 (@languages) = _get_language_dirs($htdocs,$theme);
132 for my $theme ( _get_themes('opac') ) {
133 push @languages, _get_language_dirs($htdocs,$theme);
137 elsif ($interface && $interface eq 'intranet' ) {
138 @enabled_languages = split ",", C4::Context->preference('language');
139 $htdocs = C4::Context->config('intrahtdocs');
140 if ( $theme and -d "$htdocs/$theme" ) {
141 @languages = _get_language_dirs($htdocs,$theme);
144 foreach my $theme ( _get_themes('intranet') ) {
145 push @languages, _get_language_dirs($htdocs,$theme);
150 @enabled_languages = split ",", C4::Context->preference('opaclanguages');
151 my $htdocs = C4::Context->config('intrahtdocs');
152 foreach my $theme ( _get_themes('intranet') ) {
153 push @languages, _get_language_dirs($htdocs,$theme);
155 $htdocs = C4::Context->config('opachtdocs');
156 foreach my $theme ( _get_themes('opac') ) {
157 push @languages, _get_language_dirs($htdocs,$theme);
160 $seen{$_}++ for @languages;
161 @languages = keys %seen;
163 return _build_languages_arrayref(\@languages,$current_language,\@enabled_languages);
166 =head2 getAllLanguages
168 Returns a reference to an array of hashes:
170 my $alllanguages = getAllLanguages();
171 print "Available translated languages:\n";
172 for my $language(@$alllanguages) {
173 print "$language->{language_code}\n";
174 print "$language->{language_name}\n";
175 print "$language->{language_locale_name}\n";
180 sub getAllLanguages {
182 my $dbh=C4::Context->dbh;
183 my $current_language = shift || 'en';
184 my $sth = $dbh->prepare('SELECT * FROM language_subtag_registry WHERE type=\'language\'');
186 while (my $language_subtag_registry = $sth->fetchrow_hashref) {
188 # pull out all the script descriptions for each language
189 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 = ?");
190 $sth2->execute($language_subtag_registry->{subtag},$current_language);
192 my $sth3 = $dbh->prepare("SELECT description FROM language_descriptions WHERE type='language' AND subtag=? AND lang=?");
194 # add the correct description info
195 while (my $language_descriptions = $sth2->fetchrow_hashref) {
196 $sth3->execute($language_subtag_registry->{subtag},$language_subtag_registry->{subtag});
197 my $native_description;
198 while (my $description = $sth3->fetchrow_hashref) {
199 $native_description = $description->{description};
202 # fill in the ISO6329 code
203 $language_subtag_registry->{iso639_2_code} = $language_descriptions->{iso639_2_code};
204 # fill in the native description of the language, as well as the current language's translation of that if it exists
205 if ($native_description) {
206 $language_subtag_registry->{language_description} = $native_description;
207 $language_subtag_registry->{language_description}.=" ($language_descriptions->{description})" if $language_descriptions->{description};
210 $language_subtag_registry->{language_description} = $language_descriptions->{description};
213 push @languages_loop, $language_subtag_registry;
215 return \@languages_loop;
220 Internal function, returns an array of all available themes.
222 (@themes) = &_get_themes('opac');
223 (@themes) = &_get_themes('intranet');
228 my $interface = shift;
231 if ( $interface eq 'intranet' ) {
232 $htdocs = C4::Context->config('intrahtdocs');
235 $htdocs = C4::Context->config('opachtdocs');
237 opendir D, "$htdocs";
238 my @dirlist = readdir D;
239 foreach my $directory (@dirlist) {
240 # if there's an en dir, it's a valid theme
241 -d "$htdocs/$directory/en" and push @themes, $directory;
246 =head2 _get_language_dirs
248 Internal function, returns an array of directory names, excluding non-language directories
252 sub _get_language_dirs {
253 my ($htdocs,$theme) = @_;
255 opendir D, "$htdocs/$theme";
256 for my $lang_string ( readdir D ) {
257 next if $lang_string =~/^\./;
258 next if $lang_string eq 'all';
259 next if $lang_string =~/png$/;
260 next if $lang_string =~/css$/;
261 next if $lang_string =~/CVS$/;
262 next if $lang_string =~/\.txt$/i; #Don't read the readme.txt !
263 next if $lang_string =~/img|images|famfam|sound|pdf/;
264 push @lang_strings, $lang_string;
266 return (@lang_strings);
269 =head2 _build_languages_arrayref
271 Internal function for building the ref to array of hashes
273 FIXME: this could be rewritten and simplified using map
277 sub _build_languages_arrayref {
278 my ($translated_languages,$current_language,$enabled_languages) = @_;
279 my @translated_languages = @$translated_languages;
280 my @languages_loop; # the final reference to an array of hashrefs
281 my @enabled_languages = @$enabled_languages;
282 # how many languages are enabled, if one, take note, some contexts won't need to display it
283 my %seen_languages; # the language tags we've seen
286 my $track_language_groups;
287 my $current_language_regex = regex_lang_subtags($current_language);
288 # Loop through the translated languages
289 for my $translated_language (@translated_languages) {
290 # separate the language string into its subtag types
291 my $language_subtags_hashref = regex_lang_subtags($translated_language);
293 # is this language string 'enabled'?
294 for my $enabled_language (@enabled_languages) {
295 #warn "Checking out if $translated_language eq $enabled_language";
296 $language_subtags_hashref->{'enabled'} = 1 if $translated_language eq $enabled_language;
299 # group this language, key by langtag
300 $language_subtags_hashref->{'sublanguage_current'} = 1 if $translated_language eq $current_language;
301 $language_subtags_hashref->{'rfc4646_subtag'} = $translated_language;
302 $language_subtags_hashref->{'native_description'} = language_get_description($language_subtags_hashref->{language},$language_subtags_hashref->{language},'language');
303 $language_subtags_hashref->{'script_description'} = language_get_description($language_subtags_hashref->{script},$language_subtags_hashref->{'language'},'script');
304 $language_subtags_hashref->{'region_description'} = language_get_description($language_subtags_hashref->{region},$language_subtags_hashref->{'language'},'region');
305 $language_subtags_hashref->{'variant_description'} = language_get_description($language_subtags_hashref->{variant},$language_subtags_hashref->{'language'},'variant');
306 $track_language_groups->{$language_subtags_hashref->{'language'}}++;
307 push ( @{ $language_groups->{$language_subtags_hashref->{language}} }, $language_subtags_hashref );
309 # $key is a language subtag like 'en'
310 while( my ($key, $value) = each %$language_groups) {
312 # is this language group enabled? are any of the languages within it enabled?
314 for my $enabled_language (@enabled_languages) {
315 my $regex_enabled_language = regex_lang_subtags($enabled_language);
316 $enabled = 1 if $key eq $regex_enabled_language->{language};
318 push @languages_loop, {
319 # this is only use if there is one
320 rfc4646_subtag => @$value[0]->{rfc4646_subtag},
321 native_description => language_get_description($key,$key,'language'),
323 sublanguages_loop => $value,
324 plural => $track_language_groups->{$key} >1 ? 1 : 0,
325 current => $current_language_regex->{language} eq $key ? 1 : 0,
326 group_enabled => $enabled,
329 return \@languages_loop;
332 sub language_get_description {
333 my ($script,$lang,$type) = @_;
334 my $dbh = C4::Context->dbh;
336 my $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
337 #warn "QUERY: SELECT description FROM language_descriptions WHERE subtag=$script AND lang=$lang AND type=$type";
338 $sth->execute($script,$lang,$type);
339 while (my $descriptions = $sth->fetchrow_hashref) {
340 $desc = $descriptions->{'description'};
343 $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
344 $sth->execute($script,'en',$type);
345 while (my $descriptions = $sth->fetchrow_hashref) {
346 $desc = $descriptions->{'description'};
351 =head2 regex_lang_subtags
353 This internal sub takes a string composed according to RFC 4646 as
354 an input and returns a reference to a hash containing keys and values
355 for ( language, script, region, variant, extension, privateuse )
359 sub regex_lang_subtags {
362 # Regex for recognizing RFC 4646 well-formed tags
363 # http://www.rfc-editor.org/rfc/rfc4646.txt
365 # regexes based on : http://unicode.org/cldr/data/tools/java/org/unicode/cldr/util/data/langtagRegex.txt
366 # The structure requires no forward references, so it reverses the order.
367 # The uppercase comments are fragments copied from RFC 4646
369 # Note: the tool requires that any real "=" or "#" or ";" in the regex be escaped.
371 my $alpha = qr/[a-zA-Z]/ ; # ALPHA
372 my $digit = qr/[0-9]/ ; # DIGIT
373 my $alphanum = qr/[a-zA-Z0-9]/ ; # ALPHA / DIGIT
374 my $x = qr/[xX]/ ; # private use singleton
375 my $singleton = qr/[a-w y-z A-W Y-Z]/ ; # other singleton
376 my $s = qr/[-]/ ; # separator -- lenient parsers will use [-_]
378 # Now do the components. The structure is slightly different to allow for capturing the right components.
379 # The notation (?:....) is a non-capturing version of (...): so the "?:" can be deleted if someone doesn't care about capturing.
381 my $extlang = qr{(?: $s $alpha{3} )}x ; # *3("-" 3ALPHA)
382 my $language = qr{(?: $alpha{2,3} | $alpha{4,8} )}x ;
383 #my $language = qr{(?: $alpha{2,3}$extlang{0,3} | $alpha{4,8} )}x ; # (2*3ALPHA [ extlang ]) / 4ALPHA / 5*8ALPHA
385 my $script = qr{(?: $alpha{4} )}x ; # 4ALPHA
387 my $region = qr{(?: $alpha{2} | $digit{3} )}x ; # 2ALPHA / 3DIGIT
389 my $variantSub = qr{(?: $digit$alphanum{3} | $alphanum{5,8} )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
390 my $variant = qr{(?: $variantSub (?: $s$variantSub )* )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
392 my $extensionSub = qr{(?: $singleton (?: $s$alphanum{2,8} )+ )}x ; # singleton 1*("-" (2*8alphanum))
393 my $extension = qr{(?: $extensionSub (?: $s$extensionSub )* )}x ; # singleton 1*("-" (2*8alphanum))
395 my $privateuse = qr{(?: $x (?: $s$alphanum{1,8} )+ )}x ; # ("x"/"X") 1*("-" (1*8alphanum))
397 # Define certain grandfathered codes, since otherwise the regex is pretty useless.
398 # Since these are limited, this is safe even later changes to the registry --
399 # the only oddity is that it might change the type of the tag, and thus
400 # the results from the capturing groups.
401 # http://www.iana.org/assignments/language-subtag-registry
402 # Note that these have to be compared case insensitively, requiring (?i) below.
404 my $grandfathered = qr{(?: (?i)
406 | i $s (?: ami | bnn | default | enochian | hak | klingon | lux | mingo | navajo | pwn | tao | tay | tsu )
407 | sgn $s (?: BE $s fr | BE $s nl | CH $s de)
410 # For well-formedness, we don't need the ones that would otherwise pass, so they are commented out here
414 # | en $s (?: boont | GB $s oed | scouse )
415 # | no $s (?: bok | nyn)
416 # | zh $s (?: cmn | cmn $s Hans | cmn $s Hant | gan | guoyu | hakka | min | min $s nan | wuu | xiang | yue)
418 # Here is the final breakdown, with capturing groups for each of these components
419 # The language, variants, extensions, grandfathered, and private-use may have interior '-'
421 #my $root = qr{(?: ($language) (?: $s ($script) )? 40% (?: $s ($region) )? 40% (?: $s ($variant) )? 10% (?: $s ($extension) )? 5% (?: $s ($privateuse) )? 5% ) 90% | ($grandfathered) 5% | ($privateuse) 5% };
423 $string =~ qr{^ (?:($language)) (?:$s($script))? (?:$s($region))? (?:$s($variant))? (?:$s($extension))? (?:$s($privateuse))? $}xi; # |($grandfathered) | ($privateuse) $}xi;
425 'rfc4646_subtag' => $string,
436 # Script Direction Resources:
437 # http://www.w3.org/International/questions/qa-scripts
439 my ($language_script)= @_;
440 my $dbh = C4::Context->dbh;
442 my $sth = $dbh->prepare('SELECT bidi FROM language_script_bidi WHERE rfc4646_subtag=?');
443 $sth->execute($language_script);
444 while (my $result = $sth->fetchrow_hashref) {
445 $bidi = $result->{'bidi'};
450 sub accept_language {
451 # referenced http://search.cpan.org/src/CGILMORE/I18N-AcceptLanguage-1.04/lib/I18N/AcceptLanguage.pm
452 my ($clientPreferences,$supportedLanguages) = @_;
454 if ($clientPreferences) {
455 # There should be no whitespace anways, but a cleanliness/sanity check
456 $clientPreferences =~ s/\s//g;
457 # Prepare the list of client-acceptable languages
458 foreach my $tag (split(/,/, $clientPreferences)) {
459 my ($language, $quality) = split(/\;/, $tag);
460 $quality =~ s/^q=//i if $quality;
461 $quality = 1 unless $quality;
462 next if $quality <= 0;
463 # We want to force the wildcard to be last
464 $quality = 0 if ($language eq '*');
465 # Pushing lowercase language here saves processing later
466 push(@languages, { quality => $quality,
467 language => $language,
468 lclanguage => lc($language) });
471 carp "accept_language(x,y) called with no clientPreferences (x).";
473 # Prepare the list of server-supported languages
474 my %supportedLanguages = ();
475 my %secondaryLanguages = ();
476 foreach my $language (@$supportedLanguages) {
477 # warn "Language supported: " . $language->{language};
478 my $subtag = $language->{rfc4646_subtag};
479 $supportedLanguages{lc($subtag)} = $subtag;
480 if ( $subtag =~ /^([^-]+)-?/ ) {
481 $secondaryLanguages{lc($1)} = $subtag;
485 # Reverse sort the list, making best quality at the front of the array
486 @languages = sort { $b->{quality} <=> $a->{quality} } @languages;
487 my $secondaryMatch = '';
488 foreach my $tag (@languages) {
489 if (exists($supportedLanguages{$tag->{lclanguage}})) {
490 # Client en-us eq server en-us
491 return $supportedLanguages{$tag->{language}} if exists($supportedLanguages{$tag->{language}});
492 return $supportedLanguages{$tag->{lclanguage}};
493 } elsif (exists($secondaryLanguages{$tag->{lclanguage}})) {
494 # Client en eq server en-us
495 return $secondaryLanguages{$tag->{language}} if exists($secondaryLanguages{$tag->{language}});
496 return $supportedLanguages{$tag->{lclanguage}};
497 } elsif ($tag->{lclanguage} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
498 # Client en-gb eq server en-us
499 $secondaryMatch = $secondaryLanguages{$1};
500 } elsif ($tag->{lclanguage} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
501 # FIXME: We just checked the exact same conditional!
502 # Client en-us eq server en
503 $secondaryMatch = $supportedLanguages{$1};
504 } elsif ($tag->{lclanguage} eq '*') {
505 # * matches every language not already specified.
506 # It doesn't care which we pick, so let's pick the default,
507 # if available, then the first in the array.
508 #return $acceptor->defaultLanguage() if $acceptor->defaultLanguage();
509 return $supportedLanguages->[0];
512 # No primary matches. Secondary? (ie, en-us requested and en supported)
513 return $secondaryMatch if $secondaryMatch;
514 return undef; # else, we got nothing.