3 # Copyright 2006 (C) LibLime
4 # Joshua Ferraro <jmf@liblime.com>
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);
33 &getFrameworkLanguages
34 &getTranslatedLanguages
37 @EXPORT_OK = qw(getFrameworkLanguages getTranslatedLanguages getAllLanguages get_bidi regex_lang_subtags language_get_description accept_language);
43 C4::Languages - Perl Module containing language list functions for Koha
53 =head2 getFrameworkLanguages
55 Returns a reference to an array of hashes:
57 my $languages = getFrameworkLanguages();
58 for my $language(@$languages) {
59 print "$language->{language_code}\n"; # language code in iso 639-2
60 print "$language->{language_name}\n"; # language name in native script
61 print "$language->{language_locale_name}\n"; # language name in current locale
66 sub getFrameworkLanguages {
67 # get a hash with all language codes, names, and locale names
68 my $all_languages = getAllLanguages();
71 # find the available directory names
72 my $dir=C4::Context->config('intranetdir')."/installer/data/";
74 my @listdir= grep { !/^\.|CVS/ && -d "$dir/$_"} readdir(MYDIR);
77 # pull out all data for the dir names that exist
78 for my $dirname (@listdir) {
79 for my $language_set (@$all_languages) {
81 if ($dirname eq $language_set->{language_code}) {
83 'language_code'=>$dirname,
84 'language_description'=>$language_set->{language_description},
85 'native_descrition'=>$language_set->{language_native_description} }
92 =head2 getTranslatedLanguages
94 Returns a reference to an array of hashes:
96 my $languages = getTranslatedLanguages();
97 print "Available translated langauges:\n";
98 for my $language(@$trlanguages) {
99 print "$language->{language_code}\n"; # language code in iso 639-2
100 print "$language->{language_name}\n"; # language name in native script
101 print "$language->{language_locale_name}\n"; # language name in current locale
106 sub getTranslatedLanguages {
107 my ($interface, $theme, $current_language, $which) = @_;
109 my $all_languages = getAllLanguages();
112 my @enabled_languages;
114 if ($interface && $interface eq 'opac' ) {
115 @enabled_languages = split ",", C4::Context->preference('opaclanguages');
116 $htdocs = C4::Context->config('opachtdocs');
117 if ( $theme and -d "$htdocs/$theme" ) {
118 (@languages) = _get_language_dirs($htdocs,$theme);
119 return _build_languages_arrayref($all_languages,\@languages,$current_language,\@enabled_languages);
122 for my $theme ( _get_themes('opac') ) {
123 push @languages, _get_language_dirs($htdocs,$theme);
125 return _build_languages_arrayref($all_languages,\@languages,$current_language,\@enabled_languages);
128 elsif ($interface && $interface eq 'intranet' ) {
129 @enabled_languages = split ",", C4::Context->preference('language');
130 $htdocs = C4::Context->config('intrahtdocs');
131 if ( $theme and -d "$htdocs/$theme" ) {
132 @languages = _get_language_dirs($htdocs,$theme);
133 return _build_languages_arrayref($all_languages,\@languages,$current_language,\@enabled_languages);
136 foreach my $theme ( _get_themes('intranet') ) {
137 push @languages, _get_language_dirs($htdocs,$theme);
139 return _build_languages_arrayref($all_languages,\@languages,$current_language,\@enabled_languages);
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;
155 return _build_languages_arrayref($all_languages,\@languages,$current_language,\@enabled_languages);
159 =head2 getAllLanguages
161 Returns a reference to an array of hashes:
163 my $alllanguages = getAllLanguages();
164 print "Available translated langauges:\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";
173 sub getAllLanguages {
175 my $dbh=C4::Context->dbh;
176 my $current_language = 'en';
177 my $sth = $dbh->prepare('SELECT * FROM language_subtag_registry WHERE type=\'language\'');
179 while (my $language_subtag_registry = $sth->fetchrow_hashref) {
181 # pull out all the script descriptions for each language
182 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 =?");
183 $sth2->execute($language_subtag_registry->{subtag});
185 # add the correct description info
186 while (my $language_descriptions = $sth2->fetchrow_hashref) {
187 # fill in the ISO6329 code
188 $language_subtag_registry->{iso639_2_code} = $language_descriptions->{iso639_2_code};
189 $language_subtag_registry->{language_description} = $language_descriptions->{description};
191 push @languages_loop, $language_subtag_registry;
193 return \@languages_loop;
198 Internal function, returns an array of all available themes.
200 (@themes) = &_get_themes('opac');
201 (@themes) = &_get_themes('intranet');
206 my $interface = shift;
209 if ( $interface eq 'intranet' ) {
210 $htdocs = C4::Context->config('intrahtdocs');
213 $htdocs = C4::Context->config('opachtdocs');
215 opendir D, "$htdocs";
216 my @dirlist = readdir D;
217 foreach my $directory (@dirlist) {
218 # if there's an en dir, it's a valid theme
219 -d "$htdocs/$directory/en" and push @themes, $directory;
224 =head2 _get_language_dirs
226 Internal function, returns an array of directory names, excluding non-language directories
230 sub _get_language_dirs {
231 my ($htdocs,$theme) = @_;
233 opendir D, "$htdocs/$theme";
234 for my $lang_string ( readdir D ) {
235 next if $lang_string =~/^\./;
236 next if $lang_string eq 'all';
237 next if $lang_string =~/png$/;
238 next if $lang_string =~/css$/;
239 next if $lang_string =~/CVS$/;
240 next if $lang_string =~/\.txt$/i; #Don't read the readme.txt !
241 next if $lang_string =~/fam|img|images/;
242 push @lang_strings, $lang_string;
244 return (@lang_strings);
247 =head2 _build_languages_arrayref
249 Internal function for building the ref to array of hashes
251 FIXME: this could be rewritten and simplified using map
255 sub _build_languages_arrayref {
256 my ($all_languages,$translated_languages,$current_language,$enabled_languages) = @_;
257 my @translated_languages = @$translated_languages;
258 my @languages_loop; # the final reference to an array of hashrefs
259 my @enabled_languages = @$enabled_languages;
260 # how many languages are enabled, if one, take note, some contexts won't need to display it
261 my $one_language_enabled = 1 unless @enabled_languages > 1;
262 my %seen_languages; # the language tags we've seen
265 my $track_language_groups;
266 my $current_language_regex = regex_lang_subtags($current_language);
267 # Loop through the translated languages
268 for my $translated_language (@translated_languages) {
269 # separate the language string into its subtag types
270 my $language_subtags_hashref = regex_lang_subtags($translated_language);
272 # is this language string 'enabled'?
273 for my $enabled_language (@enabled_languages) {
274 #warn "Checking out if $translated_language eq $enabled_language";
275 $language_subtags_hashref->{'enabled'} = 1 if $translated_language eq $enabled_language;
278 # group this language, key by langtag
279 $language_subtags_hashref->{'sublanguage_current'} = 1 if $translated_language eq $current_language;
280 $language_subtags_hashref->{'rfc4646_subtag'} = $translated_language;
281 $language_subtags_hashref->{'native_description'} = language_get_description($language_subtags_hashref->{language},$language_subtags_hashref->{language},'language');
282 $language_subtags_hashref->{'script_description'} = language_get_description($language_subtags_hashref->{script},$language_subtags_hashref->{'language'},'script');
283 $language_subtags_hashref->{'region_description'} = language_get_description($language_subtags_hashref->{region},$language_subtags_hashref->{'language'},'region');
284 $language_subtags_hashref->{'variant_description'} = language_get_description($language_subtags_hashref->{variant},$language_subtags_hashref->{'language'},'variant');
285 $track_language_groups->{$language_subtags_hashref->{'language'}}++;
286 push ( @{ $language_groups->{$language_subtags_hashref->{language}} }, $language_subtags_hashref );
288 # $key is a language subtag like 'en'
289 while( my ($key, $value) = each %$language_groups) {
291 # is this language group enabled? are any of the languages within it enabled?
293 for my $enabled_language (@enabled_languages) {
294 my $regex_enabled_language = regex_lang_subtags($enabled_language);
295 $enabled = 1 if $key eq $regex_enabled_language->{language};
297 push @languages_loop, {
298 # this is only use if there is one
299 rfc4646_subtag => @$value[0]->{rfc4646_subtag},
300 native_description => language_get_description($key,$key,'language'),
302 sublanguages_loop => $value,
303 plural => $track_language_groups->{$key} >1 ? 1 : 0,
304 current => $current_language_regex->{language} eq $key ? 1 : 0,
305 group_enabled => $enabled,
306 one_language_enabled => $one_language_enabled,
309 return \@languages_loop;
312 sub language_get_description {
313 my ($script,$lang,$type) = @_;
314 my $dbh = C4::Context->dbh;
316 my $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
317 #warn "QUERY: SELECT description FROM language_descriptions WHERE subtag=$script AND lang=$lang AND type=$type";
318 $sth->execute($script,$lang,$type);
319 while (my $descriptions = $sth->fetchrow_hashref) {
320 $desc = $descriptions->{'description'};
323 $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
324 $sth->execute($script,'en',$type);
325 while (my $descriptions = $sth->fetchrow_hashref) {
326 $desc = $descriptions->{'description'};
331 =head2 regex_lang_subtags
333 This internal sub takes a string composed according to RFC 4646 as
334 an input and returns a reference to a hash containing keys and values
335 for ( language, script, region, variant, extension, privateuse )
339 sub regex_lang_subtags {
342 # Regex for recognizing RFC 4646 well-formed tags
343 # http://www.rfc-editor.org/rfc/rfc4646.txt
345 # regexes based on : http://unicode.org/cldr/data/tools/java/org/unicode/cldr/util/data/langtagRegex.txt
346 # The structure requires no forward references, so it reverses the order.
347 # The uppercase comments are fragments copied from RFC 4646
349 # Note: the tool requires that any real "=" or "#" or ";" in the regex be escaped.
351 my $alpha = qr/[a-zA-Z]/ ; # ALPHA
352 my $digit = qr/[0-9]/ ; # DIGIT
353 my $alphanum = qr/[a-zA-Z0-9]/ ; # ALPHA / DIGIT
354 my $x = qr/[xX]/ ; # private use singleton
355 my $singleton = qr/[a-w y-z A-W Y-Z]/ ; # other singleton
356 my $s = qr/[-]/ ; # separator -- lenient parsers will use [-_]
358 # Now do the components. The structure is slightly different to allow for capturing the right components.
359 # The notation (?:....) is a non-capturing version of (...): so the "?:" can be deleted if someone doesn't care about capturing.
361 my $extlang = qr{(?: $s $alpha{3} )}x ; # *3("-" 3ALPHA)
362 my $language = qr{(?: $alpha{2,3} | $alpha{4,8} )}x ;
363 #my $language = qr{(?: $alpha{2,3}$extlang{0,3} | $alpha{4,8} )}x ; # (2*3ALPHA [ extlang ]) / 4ALPHA / 5*8ALPHA
365 my $script = qr{(?: $alpha{4} )}x ; # 4ALPHA
367 my $region = qr{(?: $alpha{2} | $digit{3} )}x ; # 2ALPHA / 3DIGIT
369 my $variantSub = qr{(?: $digit$alphanum{3} | $alphanum{5,8} )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
370 my $variant = qr{(?: $variantSub (?: $s$variantSub )* )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
372 my $extensionSub = qr{(?: $singleton (?: $s$alphanum{2,8} )+ )}x ; # singleton 1*("-" (2*8alphanum))
373 my $extension = qr{(?: $extensionSub (?: $s$extensionSub )* )}x ; # singleton 1*("-" (2*8alphanum))
375 my $privateuse = qr{(?: $x (?: $s$alphanum{1,8} )+ )}x ; # ("x"/"X") 1*("-" (1*8alphanum))
377 # Define certain grandfathered codes, since otherwise the regex is pretty useless.
378 # Since these are limited, this is safe even later changes to the registry --
379 # the only oddity is that it might change the type of the tag, and thus
380 # the results from the capturing groups.
381 # http://www.iana.org/assignments/language-subtag-registry
382 # Note that these have to be compared case insensitively, requiring (?i) below.
384 my $grandfathered = qr{(?: (?i)
386 | i $s (?: ami | bnn | default | enochian | hak | klingon | lux | mingo | navajo | pwn | tao | tay | tsu )
387 | sgn $s (?: BE $s fr | BE $s nl | CH $s de)
390 # For well-formedness, we don't need the ones that would otherwise pass, so they are commented out here
394 # | en $s (?: boont | GB $s oed | scouse )
395 # | no $s (?: bok | nyn)
396 # | zh $s (?: cmn | cmn $s Hans | cmn $s Hant | gan | guoyu | hakka | min | min $s nan | wuu | xiang | yue)
398 # Here is the final breakdown, with capturing groups for each of these components
399 # The language, variants, extensions, grandfathered, and private-use may have interior '-'
401 #my $root = qr{(?: ($language) (?: $s ($script) )? 40% (?: $s ($region) )? 40% (?: $s ($variant) )? 10% (?: $s ($extension) )? 5% (?: $s ($privateuse) )? 5% ) 90% | ($grandfathered) 5% | ($privateuse) 5% };
403 $string =~ qr{^ (?:($language)) (?:$s($script))? (?:$s($region))? (?:$s($variant))? (?:$s($extension))? (?:$s($privateuse))? $}xi; # |($grandfathered) | ($privateuse) $}xi;
405 'rfc4646_subtag' => $string,
416 # Script Direction Resources:
417 # http://www.w3.org/International/questions/qa-scripts
419 my ($language_script)= @_;
420 my $dbh = C4::Context->dbh;
422 my $sth = $dbh->prepare('SELECT bidi FROM language_script_bidi WHERE rfc4646_subtag=?');
423 $sth->execute($language_script);
424 while (my $result = $sth->fetchrow_hashref) {
425 $bidi = $result->{'bidi'};
430 sub accept_language {
431 # referenced http://search.cpan.org/src/CGILMORE/I18N-AcceptLanguage-1.04/lib/I18N/AcceptLanguage.pm
432 # FIXME: since this is only used in Output.pm as of Jan 8 2008, maybe it should be IN Output.pm
433 my ($clientPreferences,$supportedLanguages) = @_;
435 if ($clientPreferences) {
436 # There should be no whitespace anways, but a cleanliness/sanity check
437 $clientPreferences =~ s/\s//g;
439 # Prepare the list of client-acceptable languages
440 foreach my $tag (split(/,/, $clientPreferences)) {
441 my ($language, $quality) = split(/\;/, $tag);
442 $quality =~ s/^q=//i if $quality;
443 $quality = 1 unless $quality;
444 next if $quality <= 0;
445 # We want to force the wildcard to be last
446 $quality = 0 if ($language eq '*');
447 # Pushing lowercase language here saves processing later
448 push(@languages, { quality => $quality,
449 language => $language,
450 lclanguage => lc($language) });
453 carp "accept_language(x,y) called with no clientPreferences (x).";
455 # Prepare the list of server-supported languages
456 my %supportedLanguages = ();
457 my %secondaryLanguages = ();
458 foreach my $language (@$supportedLanguages) {
459 # warn "Language supported: " . $language->{language_code};
460 $supportedLanguages{lc($language->{language_code})} = $language->{language_code};
461 if ($language->{language_code} =~ /^([^-]+)-?/) {
462 $secondaryLanguages{lc($1)} = $language->{language_code};
466 # Reverse sort the list, making best quality at the front of the array
467 @languages = sort { $b->{quality} <=> $a->{quality} } @languages;
468 my $secondaryMatch = '';
469 foreach my $tag (@languages) {
470 if (exists($supportedLanguages{$tag->{lclanguage}})) {
471 # Client en-us eq server en-us
472 return $supportedLanguages{$tag->{language}} if exists($supportedLanguages{$tag->{language}});
473 return $supportedLanguages{$tag->{lclanguage}};
474 } elsif (exists($secondaryLanguages{$tag->{lclanguage}})) {
475 # Client en eq server en-us
476 return $secondaryLanguages{$tag->{language}} if exists($secondaryLanguages{$tag->{language}});
477 return $supportedLanguages{$tag->{lclanguage}};
478 } elsif ($tag->{lclanguage} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
479 # Client en-gb eq server en-us
480 $secondaryMatch = $secondaryLanguages{$1};
481 } elsif ($tag->{lclanguage} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
482 # FIXME: We just checked the exact same conditional!
483 # Client en-us eq server en
484 $secondaryMatch = $supportedLanguages{$1};
485 } elsif ($tag->{lclanguage} eq '*') {
486 # * matches every language not already specified.
487 # It doesn't care which we pick, so let's pick the default,
488 # if available, then the first in the array.
489 #return $acceptor->defaultLanguage() if $acceptor->defaultLanguage();
490 return $supportedLanguages->[0];
493 # No primary matches. Secondary? (ie, en-us requested and en supported)
494 return $secondaryMatch if $secondaryMatch;
495 return undef; # else, we got nothing.