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
25 use vars qw($VERSION @ISA @EXPORT @EXPORT_OK %EXPORT_TAGS $DEBUG);
32 &getFrameworkLanguages
33 &getTranslatedLanguages
36 @EXPORT_OK = qw(getFrameworkLanguages getTranslatedLanguages getAllLanguages get_bidi regex_lang_subtags language_get_description accept_language);
42 C4::Languages - Perl Module containing language list functions for Koha
52 =head2 getFrameworkLanguages
54 Returns a reference to an array of hashes:
56 my $languages = getFrameworkLanguages();
57 for my $language(@$languages) {
58 print "$language->{language_code}\n"; # language code in iso 639-2
59 print "$language->{language_name}\n"; # language name in native script
60 print "$language->{language_locale_name}\n"; # language name in current locale
65 sub getFrameworkLanguages {
66 # get a hash with all language codes, names, and locale names
67 my $all_languages = getAllLanguages();
70 # find the available directory names
71 my $dir=C4::Context->config('intranetdir')."/installer/data/";
73 my @listdir= grep { !/^\.|CVS/ && -d "$dir/$_"} readdir(MYDIR);
76 # pull out all data for the dir names that exist
77 for my $dirname (@listdir) {
78 for my $language_set (@$all_languages) {
80 if ($dirname eq $language_set->{language_code}) {
82 'language_code'=>$dirname,
83 'language_description'=>$language_set->{language_description},
84 'native_descrition'=>$language_set->{language_native_description} }
91 =head2 getTranslatedLanguages
93 Returns a reference to an array of hashes:
95 my $languages = getTranslatedLanguages();
96 print "Available translated langauges:\n";
97 for my $language(@$trlanguages) {
98 print "$language->{language_code}\n"; # language code in iso 639-2
99 print "$language->{language_name}\n"; # language name in native script
100 print "$language->{language_locale_name}\n"; # language name in current locale
105 sub getTranslatedLanguages {
106 my ($interface, $theme) = @_;
108 my $all_languages = getAllLanguages();
112 if ($interface && $interface eq 'opac' ) {
113 $htdocs = C4::Context->config('opachtdocs');
114 if ( $theme and -d "$htdocs/$theme" ) {
115 (@languages) = _get_language_dirs($htdocs,$theme);
116 return _build_languages_arrayref($all_languages,@languages);
119 for my $theme ( _get_themes('opac') ) {
120 push @languages, _get_language_dirs($htdocs,$theme);
122 return _build_languages_arrayref($all_languages,@languages);
125 elsif ($interface && $interface eq 'intranet' ) {
126 $htdocs = C4::Context->config('intrahtdocs');
127 if ( $theme and -d "$htdocs/$theme" ) {
128 @languages = _get_language_dirs($htdocs,$theme);
129 return _build_languages_arrayref($all_languages,@languages);
132 foreach my $theme ( _get_themes('opac') ) {
133 push @languages, _get_language_dirs($htdocs,$theme);
135 return _build_languages_arrayref($all_languages,@languages);
139 my $htdocs = C4::Context->config('intrahtdocs');
140 foreach my $theme ( _get_themes('intranet') ) {
141 push @languages, _get_language_dirs($htdocs,$theme);
143 $htdocs = C4::Context->config('opachtdocs');
144 foreach my $theme ( _get_themes('opac') ) {
145 push @languages, _get_language_dirs($htdocs,$theme);
147 return _build_languages_arrayref($all_languages,@languages);
151 =head2 getAllLanguages
153 Returns a reference to an array of hashes:
155 my $alllanguages = getAllLanguages();
156 print "Available translated langauges:\n";
157 for my $language(@$alllanguages) {
158 print "$language->{language_code}\n";
159 print "$language->{language_name}\n";
160 print "$language->{language_locale_name}\n";
165 sub getAllLanguages {
167 my $dbh=C4::Context->dbh;
168 my $current_language = 'en';
169 my $sth = $dbh->prepare('SELECT * FROM language_subtag_registry WHERE type=\'language\'');
171 while (my $language_subtag_registry = $sth->fetchrow_hashref) {
173 # pull out all the script descriptions for each language
174 my $sth2= $dbh->prepare('SELECT * FROM language_descriptions WHERE type=\'language\' AND subtag =?');
175 $sth2->execute($language_subtag_registry->{subtag});
177 # add the correct description info
178 while (my $language_descriptions = $sth2->fetchrow_hashref) {
180 # Insert the language description using the current language script
181 #if ( $language_subtag_registry->{subtag}
182 if ( $current_language eq $language_descriptions->{lang} ) {
183 $language_subtag_registry->{language_description} = $language_descriptions->{description};
184 #warn "CUR:".$language_subtag_registry->{description};
187 # Insert the language name using the script native to the language (FIXME: should really be based on script)
188 if ($language_subtag_registry->{subtag} eq $language_descriptions->{lang}) {
189 $language_subtag_registry->{language_native_description} = $language_descriptions->{description};
190 #warn "NAT: Desc:$language_descriptions->{description} SubtagDesc: $language_subtag_registry->{language_description}";
193 push @languages_loop, $language_subtag_registry;
195 return \@languages_loop;
200 Internal function, returns an array of all available themes.
202 (@themes) = &_get_themes('opac');
203 (@themes) = &_get_themes('intranet');
208 my $interface = shift;
211 if ( $interface eq 'intranet' ) {
212 $htdocs = C4::Context->config('intrahtdocs');
215 $htdocs = C4::Context->config('opachtdocs');
217 opendir D, "$htdocs";
218 my @dirlist = readdir D;
219 foreach my $directory (@dirlist) {
220 # if there's an en dir, it's a valid theme
221 -d "$htdocs/$directory/en" and push @themes, $directory;
226 =head2 _get_language_dirs
228 Internal function, returns an array of directory names, excluding non-language directories
232 sub _get_language_dirs {
233 my ($htdocs,$theme) = @_;
235 opendir D, "$htdocs/$theme";
236 for my $language ( readdir D ) {
237 next if $language =~/^\./;
238 next if $language eq 'all';
239 next if $language =~/png$/;
240 next if $language =~/css$/;
241 next if $language =~/CVS$/;
242 next if $language =~/\.txt$/i; #Don't read the readme.txt !
243 next if $language =~/img|images/;
244 push @languages, $language;
249 =head2 _build_languages_arrayref
251 Internal function for building the ref to array of hashes
253 FIXME: this could be rewritten and simplified using map
257 sub _build_languages_arrayref {
258 my ($all_languages,@languages) = @_;
262 # Loop through the languages, pick the ones that are translated
263 for my $language (@languages) {
265 # separate the language string into its subtag types
266 my $language_subtags_hashref = regex_lang_subtags($language);
267 unless ($seen_languages{$language}) {
268 for my $language_code (@$all_languages) {
269 if ($language_subtags_hashref->{language} eq $language_code->{'subtag'}) {
270 $language_code->{'language_lang'} = $language;
271 $language_code->{'language_code'} = $language_subtags_hashref->{'language'};
272 $language_code->{'script_code'} = $language_subtags_hashref->{'script'};
273 $language_code->{'region_code'} = $language_subtags_hashref->{'region'};
274 $language_code->{'variant_code'} = $language_subtags_hashref->{'variant'};
275 push @final_languages, $language_code;
276 $found_languages{$language}++;
279 $seen_languages{$language}++;
281 # Handle languages not in our database with their code
282 unless ($found_languages{$language}) {
284 $language_code->{'language_lang'} = $language;
285 $language_code->{'language_code'} = $language;
286 push @final_languages, $language_code;
290 return \@final_languages;
293 sub language_get_description {
294 my ($script,$lang,$type) = @_;
295 my $dbh = C4::Context->dbh;
297 my $sth = $dbh->prepare('SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?');
298 $sth->execute($script,$lang,$type);
299 while (my $descriptions = $sth->fetchrow_hashref) {
300 $desc = $descriptions->{'description'};
304 =head2 regex_lang_subtags
306 This internal sub takes a string composed according to RFC 4646 as
307 an input and returns a reference to a hash containing keys and values
308 for ( language, script, region, variant, extension, privateuse )
312 sub regex_lang_subtags {
315 # Regex for recognizing RFC 4646 well-formed tags
316 # http://www.rfc-editor.org/rfc/rfc4646.txt
318 # regexes based on : http://unicode.org/cldr/data/tools/java/org/unicode/cldr/util/data/langtagRegex.txt
319 # The structure requires no forward references, so it reverses the order.
320 # The uppercase comments are fragments copied from RFC 4646
322 # Note: the tool requires that any real "=" or "#" or ";" in the regex be escaped.
324 my $alpha = qr/[a-zA-Z]/ ; # ALPHA
325 my $digit = qr/[0-9]/ ; # DIGIT
326 my $alphanum = qr/[a-zA-Z0-9]/ ; # ALPHA / DIGIT
327 my $x = qr/[xX]/ ; # private use singleton
328 my $singleton = qr/[a-w y-z A-W Y-Z]/ ; # other singleton
329 my $s = qr/[-]/ ; # separator -- lenient parsers will use [-_]
331 # Now do the components. The structure is slightly different to allow for capturing the right components.
332 # The notation (?:....) is a non-capturing version of (...): so the "?:" can be deleted if someone doesn't care about capturing.
334 my $extlang = qr{(?: $s $alpha{3} )}x ; # *3("-" 3ALPHA)
335 my $language = qr{(?: $alpha{2,3} | $alpha{4,8} )}x ;
336 #my $language = qr{(?: $alpha{2,3}$extlang{0,3} | $alpha{4,8} )}x ; # (2*3ALPHA [ extlang ]) / 4ALPHA / 5*8ALPHA
338 my $script = qr{(?: $alpha{4} )}x ; # 4ALPHA
340 my $region = qr{(?: $alpha{2} | $digit{3} )}x ; # 2ALPHA / 3DIGIT
342 my $variantSub = qr{(?: $digit$alphanum{3} | $alphanum{5,8} )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
343 my $variant = qr{(?: $variantSub (?: $s$variantSub )* )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
345 my $extensionSub = qr{(?: $singleton (?: $s$alphanum{2,8} )+ )}x ; # singleton 1*("-" (2*8alphanum))
346 my $extension = qr{(?: $extensionSub (?: $s$extensionSub )* )}x ; # singleton 1*("-" (2*8alphanum))
348 my $privateuse = qr{(?: $x (?: $s$alphanum{1,8} )+ )}x ; # ("x"/"X") 1*("-" (1*8alphanum))
350 # Define certain grandfathered codes, since otherwise the regex is pretty useless.
351 # Since these are limited, this is safe even later changes to the registry --
352 # the only oddity is that it might change the type of the tag, and thus
353 # the results from the capturing groups.
354 # http://www.iana.org/assignments/language-subtag-registry
355 # Note that these have to be compared case insensitively, requiring (?i) below.
357 my $grandfathered = qr{(?: (?i)
359 | i $s (?: ami | bnn | default | enochian | hak | klingon | lux | mingo | navajo | pwn | tao | tay | tsu )
360 | sgn $s (?: BE $s fr | BE $s nl | CH $s de)
363 # For well-formedness, we don't need the ones that would otherwise pass, so they are commented out here
367 # | en $s (?: boont | GB $s oed | scouse )
368 # | no $s (?: bok | nyn)
369 # | zh $s (?: cmn | cmn $s Hans | cmn $s Hant | gan | guoyu | hakka | min | min $s nan | wuu | xiang | yue)
371 # Here is the final breakdown, with capturing groups for each of these components
372 # The language, variants, extensions, grandfathered, and private-use may have interior '-'
374 #my $root = qr{(?: ($language) (?: $s ($script) )? 40% (?: $s ($region) )? 40% (?: $s ($variant) )? 10% (?: $s ($extension) )? 5% (?: $s ($privateuse) )? 5% ) 90% | ($grandfathered) 5% | ($privateuse) 5% };
376 $string =~ qr{^ (?:($language)) (?:$s($script))? (?:$s($region))? (?:$s($variant))? (?:$s($extension))? (?:$s($privateuse))? $}xi; # |($grandfathered) | ($privateuse) $}xi;
388 # Script Direction Resources:
389 # http://www.w3.org/International/questions/qa-scripts
391 my ($language_script)= @_;
392 my $dbh = C4::Context->dbh;
394 my $sth = $dbh->prepare('SELECT bidi FROM language_script_bidi WHERE rfc4646_subtag=?');
395 $sth->execute($language_script);
396 while (my $result = $sth->fetchrow_hashref) {
397 $bidi = $result->{'bidi'};
402 sub accept_language {
403 # referenced http://search.cpan.org/src/CGILMORE/I18N-AcceptLanguage-1.04/lib/I18N/AcceptLanguage.pm
404 my ($clientPreferences,$supportedLanguages) = @_;
405 # There should be no whitespace anways, but a cleanliness/sanity check
406 $clientPreferences =~ s/\s//g;
408 # Prepare the list of client-acceptable languages
410 foreach my $tag (split(/,/, $clientPreferences)) {
411 my ($language, $quality) = split(/\;/, $tag);
412 $quality =~ s/^q=//i if $quality;
413 $quality = 1 unless $quality;
414 next if $quality <= 0;
415 # We want to force the wildcard to be last
416 $quality = 0 if ($language eq '*');
417 # Pushing lowercase language here saves processing later
418 push(@languages, { quality => $quality,
419 language => $language,
420 lclanguage => lc($language) });
422 # Prepare the list of server-supported languages
423 my %supportedLanguages = ();
424 my %secondaryLanguages = ();
425 foreach my $language (@$supportedLanguages) {
426 warn "SUP: ".$language->{language_code};
427 $supportedLanguages{lc($language->{language_code})} = $language->{language_code};
428 if ($language->{language_code} =~ /^([^-]+)-?/) {
429 $secondaryLanguages{lc($1)} = $language->{language_code};
433 # Reverse sort the list, making best quality at the front of the array
434 @languages = sort { $b->{quality} <=> $a->{quality} } @languages;
435 my $secondaryMatch = '';
436 foreach my $tag (@languages) {
437 if (exists($supportedLanguages{$tag->{lclanguage}})) {
438 # Client en-us eq server en-us
439 return $supportedLanguages{$tag->{language}} if exists($supportedLanguages{$tag->{language}});
440 return $supportedLanguages{$tag->{lclanguage}};
441 } elsif (exists($secondaryLanguages{$tag->{lclanguage}})) {
442 # Client en eq server en-us
443 return $secondaryLanguages{$tag->{language}} if exists($secondaryLanguages{$tag->{language}});
444 return $supportedLanguages{$tag->{lclanguage}};
445 } elsif ($tag->{lclanguage} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
446 # Client en-gb eq server en-us
447 $secondaryMatch = $secondaryLanguages{$1};
448 } elsif ($tag->{lclanguage} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
449 # Client en-us eq server en
450 $secondaryMatch = $supportedLanguages{$1};
451 } elsif ($tag->{lclanguage} eq '*') {
452 # * matches every language not already specified.
453 # It doesn't care which we pick, so let's pick the default,
454 # if available, then the first in the array.
455 #return $acceptor->defaultLanguage() if $acceptor->defaultLanguage();
456 return $supportedLanguages->[0];
459 # No primary matches. Secondary? (ie, en-us requested and en supported)
460 return $secondaryMatch if $secondaryMatch;