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
22 use strict; use warnings; #FIXME: turn off warnings before release
25 use vars qw($VERSION @ISA @EXPORT @EXPORT_OK %EXPORT_TAGS);
29 C4::Languages - Perl Module containing language list functions for Koha
42 @EXPORT_OK = qw(getFrameworkLanguages getTranslatedLanguages getAllLanguages get_bidi regex_lang_subtags language_get_description accept_language);
45 =head2 getFrameworkLanguages
47 Returns a reference to an array of hashes:
49 my $languages = getFrameworkLanguages();
50 for my $language(@$languages) {
51 print "$language->{language_code}\n"; # language code in iso 639-2
52 print "$language->{language_name}\n"; # language name in native script
53 print "$language->{language_locale_name}\n"; # language name in current locale
58 sub getFrameworkLanguages {
59 # get a hash with all language codes, names, and locale names
60 my $all_languages = getAllLanguages();
63 # find the available directory names
64 my $dir=C4::Context->config('intranetdir')."/installer/data/";
66 my @listdir= grep { !/^\.|CVS/ && -d "$dir/$_"} readdir(MYDIR);
69 # pull out all data for the dir names that exist
70 for my $dirname (@listdir) {
71 for my $language_set (@$all_languages) {
73 if ($dirname eq $language_set->{language_code}) {
75 'language_code'=>$dirname,
76 'language_description'=>$language_set->{language_description},
77 'native_descrition'=>$language_set->{language_native_description} }
84 =head2 getTranslatedLanguages
86 Returns a reference to an array of hashes:
88 my $languages = getTranslatedLanguages();
89 print "Available translated langauges:\n";
90 for my $language(@$trlanguages) {
91 print "$language->{language_code}\n"; # language code in iso 639-2
92 print "$language->{language_name}\n"; # language name in native script
93 print "$language->{language_locale_name}\n"; # language name in current locale
98 sub getTranslatedLanguages {
99 my ($interface, $theme) = @_;
101 my $all_languages = getAllLanguages();
105 if ($interface && $interface eq 'opac' ) {
106 $htdocs = C4::Context->config('opachtdocs');
107 if ( $theme and -d "$htdocs/$theme" ) {
108 (@languages) = _get_language_dirs($htdocs,$theme);
109 return _build_languages_arrayref($all_languages,@languages);
112 for my $theme ( _get_themes('opac') ) {
113 push @languages, _get_language_dirs($htdocs,$theme);
115 return _build_languages_arrayref($all_languages,@languages);
118 elsif ($interface && $interface eq 'intranet' ) {
119 $htdocs = C4::Context->config('intrahtdocs');
120 if ( $theme and -d "$htdocs/$theme" ) {
121 @languages = _get_language_dirs($htdocs,$theme);
122 return _build_languages_arrayref($all_languages,@languages);
125 foreach my $theme ( _get_themes('opac') ) {
126 push @languages, _get_language_dirs($htdocs,$theme);
128 return _build_languages_arrayref($all_languages,@languages);
132 my $htdocs = C4::Context->config('intrahtdocs');
133 foreach my $theme ( _get_themes('intranet') ) {
134 push @languages, _get_language_dirs($htdocs,$theme);
136 $htdocs = C4::Context->config('opachtdocs');
137 foreach my $theme ( _get_themes('opac') ) {
138 push @languages, _get_language_dirs($htdocs,$theme);
140 return _build_languages_arrayref($all_languages,@languages);
144 =head2 getAllLanguages
146 Returns a reference to an array of hashes:
148 my $alllanguages = getAllLanguages();
149 print "Available translated langauges:\n";
150 for my $language(@$alllanguages) {
151 print "$language->{language_code}\n";
152 print "$language->{language_name}\n";
153 print "$language->{language_locale_name}\n";
158 sub getAllLanguages {
160 my $dbh=C4::Context->dbh;
161 my $current_language = 'en';
162 my $sth = $dbh->prepare('SELECT * FROM language_subtag_registry WHERE type=\'language\'');
164 while (my $language_subtag_registry = $sth->fetchrow_hashref) {
166 # pull out all the script descriptions for each language
167 my $sth2= $dbh->prepare('SELECT * FROM language_descriptions WHERE type=\'language\' AND subtag =?');
168 $sth2->execute($language_subtag_registry->{subtag});
170 # add the correct description info
171 while (my $language_descriptions = $sth2->fetchrow_hashref) {
173 # Insert the language description using the current language script
174 #if ( $language_subtag_registry->{subtag}
175 if ( $current_language eq $language_descriptions->{lang} ) {
176 $language_subtag_registry->{language_description} = $language_descriptions->{description};
177 #warn "CUR:".$language_subtag_registry->{description};
180 # Insert the language name using the script native to the language (FIXME: should really be based on script)
181 if ($language_subtag_registry->{subtag} eq $language_descriptions->{lang}) {
182 $language_subtag_registry->{language_native_description} = $language_descriptions->{description};
183 #warn "NAT: Desc:$language_descriptions->{description} SubtagDesc: $language_subtag_registry->{language_description}";
186 push @languages_loop, $language_subtag_registry;
188 return \@languages_loop;
193 Internal function, returns an array of all available themes.
195 (@themes) = &_get_themes('opac');
196 (@themes) = &_get_themes('intranet');
201 my $interface = shift;
204 if ( $interface eq 'intranet' ) {
205 $htdocs = C4::Context->config('intrahtdocs');
208 $htdocs = C4::Context->config('opachtdocs');
210 opendir D, "$htdocs";
211 my @dirlist = readdir D;
212 foreach my $directory (@dirlist) {
213 # if there's an en dir, it's a valid theme
214 -d "$htdocs/$directory/en" and push @themes, $directory;
219 =head2 _get_language_dirs
221 Internal function, returns an array of directory names, excluding non-language directories
225 sub _get_language_dirs {
226 my ($htdocs,$theme) = @_;
228 opendir D, "$htdocs/$theme";
229 for my $language ( readdir D ) {
230 next if $language =~/^\./;
231 next if $language eq 'all';
232 next if $language =~/png$/;
233 next if $language =~/css$/;
234 next if $language =~/CVS$/;
235 next if $language =~/\.txt$/i; #Don't read the readme.txt !
236 next if $language =~/img/;
237 push @languages, $language;
242 =head2 _build_languages_arrayref
244 Internal function for building the ref to array of hashes
246 FIXME: this could be rewritten and simplified using map
250 sub _build_languages_arrayref {
251 my ($all_languages,@languages) = @_;
255 # Loop through the languages, pick the ones that are translated
256 for my $language (@languages) {
258 # separate the language string into its subtag types
259 my $language_subtags_hashref = regex_lang_subtags($language);
260 unless ($seen_languages{$language}) {
261 for my $language_code (@$all_languages) {
262 if ($language_subtags_hashref->{language} eq $language_code->{'subtag'}) {
263 $language_code->{'language_lang'} = $language;
264 $language_code->{'language_code'} = $language_subtags_hashref->{'language'};
265 $language_code->{'script_code'} = $language_subtags_hashref->{'script'};
266 $language_code->{'region_code'} = $language_subtags_hashref->{'region'};
267 $language_code->{'variant_code'} = $language_subtags_hashref->{'variant'};
268 push @final_languages, $language_code;
269 $found_languages{$language}++;
272 $seen_languages{$language}++;
274 # Handle languages not in our database with their code
275 unless ($found_languages{$language}) {
277 $language_code->{'language_lang'} = $language;
278 $language_code->{'language_code'} = $language;
279 push @final_languages, $language_code;
283 return \@final_languages;
286 sub language_get_description {
287 my ($script,$lang,$type) = @_;
288 my $dbh = C4::Context->dbh;
290 my $sth = $dbh->prepare('SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?');
291 $sth->execute($script,$lang,$type);
292 while (my $descriptions = $sth->fetchrow_hashref) {
293 $desc = $descriptions->{'description'};
297 =head2 regex_lang_subtags
299 This internal sub takes a string composed according to RFC 4646 as
300 an input and returns a reference to a hash containing keys and values
301 for ( language, script, region, variant, extension, privateuse )
305 sub regex_lang_subtags {
308 # Regex for recognizing RFC 4646 well-formed tags
309 # http://www.rfc-editor.org/rfc/rfc4646.txt
311 # regexes based on : http://unicode.org/cldr/data/tools/java/org/unicode/cldr/util/data/langtagRegex.txt
312 # The structure requires no forward references, so it reverses the order.
313 # The uppercase comments are fragments copied from RFC 4646
315 # Note: the tool requires that any real "=" or "#" or ";" in the regex be escaped.
317 my $alpha = qr/[a-zA-Z]/ ; # ALPHA
318 my $digit = qr/[0-9]/ ; # DIGIT
319 my $alphanum = qr/[a-zA-Z0-9]/ ; # ALPHA / DIGIT
320 my $x = qr/[xX]/ ; # private use singleton
321 my $singleton = qr/[a-w y-z A-W Y-Z]/ ; # other singleton
322 my $s = qr/[-]/ ; # separator -- lenient parsers will use [-_]
324 # Now do the components. The structure is slightly different to allow for capturing the right components.
325 # The notation (?:....) is a non-capturing version of (...): so the "?:" can be deleted if someone doesn't care about capturing.
327 my $extlang = qr{(?: $s $alpha{3} )}x ; # *3("-" 3ALPHA)
328 my $language = qr{(?: $alpha{2,3} | $alpha{4,8} )}x ;
329 #my $language = qr{(?: $alpha{2,3}$extlang{0,3} | $alpha{4,8} )}x ; # (2*3ALPHA [ extlang ]) / 4ALPHA / 5*8ALPHA
331 my $script = qr{(?: $alpha{4} )}x ; # 4ALPHA
333 my $region = qr{(?: $alpha{2} | $digit{3} )}x ; # 2ALPHA / 3DIGIT
335 my $variantSub = qr{(?: $digit$alphanum{3} | $alphanum{5,8} )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
336 my $variant = qr{(?: $variantSub (?: $s$variantSub )* )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
338 my $extensionSub = qr{(?: $singleton (?: $s$alphanum{2,8} )+ )}x ; # singleton 1*("-" (2*8alphanum))
339 my $extension = qr{(?: $extensionSub (?: $s$extensionSub )* )}x ; # singleton 1*("-" (2*8alphanum))
341 my $privateuse = qr{(?: $x (?: $s$alphanum{1,8} )+ )}x ; # ("x"/"X") 1*("-" (1*8alphanum))
343 # Define certain grandfathered codes, since otherwise the regex is pretty useless.
344 # Since these are limited, this is safe even later changes to the registry --
345 # the only oddity is that it might change the type of the tag, and thus
346 # the results from the capturing groups.
347 # http://www.iana.org/assignments/language-subtag-registry
348 # Note that these have to be compared case insensitively, requiring (?i) below.
350 my $grandfathered = qr{(?: (?i)
352 | i $s (?: ami | bnn | default | enochian | hak | klingon | lux | mingo | navajo | pwn | tao | tay | tsu )
353 | sgn $s (?: BE $s fr | BE $s nl | CH $s de)
356 # For well-formedness, we don't need the ones that would otherwise pass, so they are commented out here
360 # | en $s (?: boont | GB $s oed | scouse )
361 # | no $s (?: bok | nyn)
362 # | zh $s (?: cmn | cmn $s Hans | cmn $s Hant | gan | guoyu | hakka | min | min $s nan | wuu | xiang | yue)
364 # Here is the final breakdown, with capturing groups for each of these components
365 # The language, variants, extensions, grandfathered, and private-use may have interior '-'
367 #my $root = qr{(?: ($language) (?: $s ($script) )? 40% (?: $s ($region) )? 40% (?: $s ($variant) )? 10% (?: $s ($extension) )? 5% (?: $s ($privateuse) )? 5% ) 90% | ($grandfathered) 5% | ($privateuse) 5% };
369 $string =~ qr{^ (?:($language)) (?:$s($script))? (?:$s($region))? (?:$s($variant))? (?:$s($extension))? (?:$s($privateuse))? $}xi; # |($grandfathered) | ($privateuse) $}xi;
381 # Script Direction Resources:
382 # http://www.w3.org/International/questions/qa-scripts
384 my ($language_script)= @_;
385 my $dbh = C4::Context->dbh;
387 my $sth = $dbh->prepare('SELECT bidi FROM language_bidi WHERE rfc4646_subtag=?');
388 $sth->execute($language_script);
389 while (my $result = $sth->fetchrow_hashref) {
390 $bidi = $result->{'bidi'};
395 sub accept_language {
396 # referenced http://search.cpan.org/src/CGILMORE/I18N-AcceptLanguage-1.04/lib/I18N/AcceptLanguage.pm
397 my ($clientPreferences,$supportedLanguages) = @_;
398 # There should be no whitespace anways, but a cleanliness/sanity check
399 $clientPreferences =~ s/\s//g;
401 # Prepare the list of client-acceptable languages
403 foreach my $tag (split(/,/, $clientPreferences)) {
404 my ($language, $quality) = split(/\;/, $tag);
405 $quality =~ s/^q=//i if $quality;
406 $quality = 1 unless $quality;
407 next if $quality <= 0;
408 # We want to force the wildcard to be last
409 $quality = 0 if ($language eq '*');
410 # Pushing lowercase language here saves processing later
411 push(@languages, { quality => $quality,
412 language => $language,
413 lclanguage => lc($language) });
415 # Prepare the list of server-supported languages
416 my %supportedLanguages = ();
417 my %secondaryLanguages = ();
418 foreach my $language (@$supportedLanguages) {
419 warn "SUP: ".$language->{language_code};
420 $supportedLanguages{lc($language->{language_code})} = $language->{language_code};
421 if ($language->{language_code} =~ /^([^-]+)-?/) {
422 $secondaryLanguages{lc($1)} = $language->{language_code};
426 # Reverse sort the list, making best quality at the front of the array
427 @languages = sort { $b->{quality} <=> $a->{quality} } @languages;
428 my $secondaryMatch = '';
429 foreach my $tag (@languages) {
430 if (exists($supportedLanguages{$tag->{lclanguage}})) {
431 # Client en-us eq server en-us
432 return $supportedLanguages{$tag->{language}} if exists($supportedLanguages{$tag->{language}});
433 return $supportedLanguages{$tag->{lclanguage}};
434 } elsif (exists($secondaryLanguages{$tag->{lclanguage}})) {
435 # Client en eq server en-us
436 return $secondaryLanguages{$tag->{language}} if exists($secondaryLanguages{$tag->{language}});
437 return $supportedLanguages{$tag->{lclanguage}};
438 } elsif ($tag->{lclanguage} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
439 # Client en-gb eq server en-us
440 $secondaryMatch = $secondaryLanguages{$1};
441 } elsif ($tag->{lclanguage} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
442 # Client en-us eq server en
443 $secondaryMatch = $supportedLanguages{$1};
444 } elsif ($tag->{lclanguage} eq '*') {
445 # * matches every language not already specified.
446 # It doesn't care which we pick, so let's pick the default,
447 # if available, then the first in the array.
448 #return $acceptor->defaultLanguage() if $acceptor->defaultLanguage();
449 return $supportedLanguages->[0];
452 # No primary matches. Secondary? (ie, en-us requested and en supported)
453 return $secondaryMatch if $secondaryMatch;