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,$theme);
152 return _build_languages_arrayref
($all_languages,\
@languages,$current_language,\
@enabled_languages);
156 =head2 getAllLanguages
158 Returns a reference to an array of hashes:
160 my $alllanguages = getAllLanguages();
161 print "Available translated langauges:\n";
162 for my $language(@$alllanguages) {
163 print "$language->{language_code}\n";
164 print "$language->{language_name}\n";
165 print "$language->{language_locale_name}\n";
170 sub getAllLanguages
{
172 my $dbh=C4
::Context
->dbh;
173 my $current_language = 'en';
174 my $sth = $dbh->prepare('SELECT * FROM language_subtag_registry WHERE type=\'language\'');
176 while (my $language_subtag_registry = $sth->fetchrow_hashref) {
178 # pull out all the script descriptions for each language
179 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 =?");
180 $sth2->execute($language_subtag_registry->{subtag
});
182 # add the correct description info
183 while (my $language_descriptions = $sth2->fetchrow_hashref) {
184 # fill in the ISO6329 code
185 $language_subtag_registry->{iso639_2_code
} = $language_descriptions->{iso639_2_code
};
186 $language_subtag_registry->{language_description
} = $language_descriptions->{description
};
188 push @languages_loop, $language_subtag_registry;
190 return \
@languages_loop;
195 Internal function, returns an array of all available themes.
197 (@themes) = &_get_themes('opac');
198 (@themes) = &_get_themes('intranet');
203 my $interface = shift;
206 if ( $interface eq 'intranet' ) {
207 $htdocs = C4
::Context
->config('intrahtdocs');
210 $htdocs = C4
::Context
->config('opachtdocs');
212 opendir D
, "$htdocs";
213 my @dirlist = readdir D
;
214 foreach my $directory (@dirlist) {
215 # if there's an en dir, it's a valid theme
216 -d
"$htdocs/$directory/en" and push @themes, $directory;
221 =head2 _get_language_dirs
223 Internal function, returns an array of directory names, excluding non-language directories
227 sub _get_language_dirs
{
228 my ($htdocs,$theme) = @_;
230 opendir D
, "$htdocs/$theme";
231 for my $lang_string ( readdir D
) {
232 next if $lang_string =~/^\./;
233 next if $lang_string eq 'all';
234 next if $lang_string =~/png$/;
235 next if $lang_string =~/css$/;
236 next if $lang_string =~/CVS$/;
237 next if $lang_string =~/\.txt$/i; #Don't read the readme.txt !
238 next if $lang_string =~/img|images/;
239 push @lang_strings, $lang_string;
241 return (@lang_strings);
244 =head2 _build_languages_arrayref
246 Internal function for building the ref to array of hashes
248 FIXME: this could be rewritten and simplified using map
252 sub _build_languages_arrayref
{
253 my ($all_languages,$translated_languages,$current_language,$enabled_languages) = @_;
254 my @translated_languages = @
$translated_languages;
255 my @languages_loop; # the final reference to an array of hashrefs
256 my @enabled_languages = @
$enabled_languages;
257 my %seen_languages; # the language tags we've seen
260 my $track_language_groups;
261 my $current_language_regex = regex_lang_subtags
($current_language);
262 # Loop through the translated languages
263 for my $translated_language (@translated_languages) {
264 # separate the language string into its subtag types
265 my $language_subtags_hashref = regex_lang_subtags
($translated_language);
267 # is this language string 'enabled'?
268 for my $enabled_language (@enabled_languages) {
269 #warn "Checking out if $translated_language eq $enabled_language";
270 $language_subtags_hashref->{'enabled'} = 1 if $translated_language eq $enabled_language;
273 # group this language, key by langtag
274 $language_subtags_hashref->{'sublanguage_current'} = 1 if $translated_language eq $current_language;
275 $language_subtags_hashref->{'rfc4646_subtag'} = $translated_language;
276 $language_subtags_hashref->{'native_description'} = language_get_description
($language_subtags_hashref->{language
},$language_subtags_hashref->{language
},'language');
277 $language_subtags_hashref->{'script_description'} = language_get_description
($language_subtags_hashref->{script
},$language_subtags_hashref->{'language'},'script');
278 $language_subtags_hashref->{'region_description'} = language_get_description
($language_subtags_hashref->{region
},$language_subtags_hashref->{'language'},'region');
279 $language_subtags_hashref->{'variant_description'} = language_get_description
($language_subtags_hashref->{variant
},$language_subtags_hashref->{'language'},'variant');
280 $track_language_groups->{$language_subtags_hashref->{'language'}}++;
281 push ( @
{ $language_groups->{$language_subtags_hashref->{language
}} }, $language_subtags_hashref );
283 # $key is a language subtag like 'en'
284 while( my ($key, $value) = each %$language_groups) {
286 # is this language group enabled? are any of the languages within it enabled?
288 for my $enabled_language (@enabled_languages) {
289 my $regex_enabled_language = regex_lang_subtags
($enabled_language);
290 $enabled = 1 if $key eq $regex_enabled_language->{language
};
292 push @languages_loop, {
293 # this is only use if there is one
294 rfc4646_subtag
=> @
$value[0]->{rfc4646_subtag
},
295 native_description
=> language_get_description
($key,$key,'language'),
297 sublanguages_loop
=> $value,
298 plural
=> $track_language_groups->{$key} >1 ?
1 : 0,
299 current
=> $current_language_regex->{language
} eq $key ?
1 : 0,
300 group_enabled
=> $enabled,
303 return \
@languages_loop;
306 sub language_get_description
{
307 my ($script,$lang,$type) = @_;
308 my $dbh = C4
::Context
->dbh;
310 my $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
311 #warn "QUERY: SELECT description FROM language_descriptions WHERE subtag=$script AND lang=$lang AND type=$type";
312 $sth->execute($script,$lang,$type);
313 while (my $descriptions = $sth->fetchrow_hashref) {
314 $desc = $descriptions->{'description'};
317 $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
318 $sth->execute($script,'en',$type);
319 while (my $descriptions = $sth->fetchrow_hashref) {
320 $desc = $descriptions->{'description'};
325 =head2 regex_lang_subtags
327 This internal sub takes a string composed according to RFC 4646 as
328 an input and returns a reference to a hash containing keys and values
329 for ( language, script, region, variant, extension, privateuse )
333 sub regex_lang_subtags
{
336 # Regex for recognizing RFC 4646 well-formed tags
337 # http://www.rfc-editor.org/rfc/rfc4646.txt
339 # regexes based on : http://unicode.org/cldr/data/tools/java/org/unicode/cldr/util/data/langtagRegex.txt
340 # The structure requires no forward references, so it reverses the order.
341 # The uppercase comments are fragments copied from RFC 4646
343 # Note: the tool requires that any real "=" or "#" or ";" in the regex be escaped.
345 my $alpha = qr/[a-zA-Z]/ ; # ALPHA
346 my $digit = qr/[0-9]/ ; # DIGIT
347 my $alphanum = qr/[a-zA-Z0-9]/ ; # ALPHA / DIGIT
348 my $x = qr/[xX]/ ; # private use singleton
349 my $singleton = qr/[a-w y-z A-W Y-Z]/ ; # other singleton
350 my $s = qr/[-]/ ; # separator -- lenient parsers will use [-_]
352 # Now do the components. The structure is slightly different to allow for capturing the right components.
353 # The notation (?:....) is a non-capturing version of (...): so the "?:" can be deleted if someone doesn't care about capturing.
355 my $extlang = qr{(?: $s $alpha{3} )}x
; # *3("-" 3ALPHA)
356 my $language = qr{(?: $alpha{2,3} | $alpha{4,8} )}x
;
357 #my $language = qr{(?: $alpha{2,3}$extlang{0,3} | $alpha{4,8} )}x ; # (2*3ALPHA [ extlang ]) / 4ALPHA / 5*8ALPHA
359 my $script = qr{(?: $alpha{4} )}x
; # 4ALPHA
361 my $region = qr{(?: $alpha{2} | $digit{3} )}x
; # 2ALPHA / 3DIGIT
363 my $variantSub = qr{(?: $digit$alphanum{3} | $alphanum{5,8} )}x
; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
364 my $variant = qr{(?: $variantSub (?: $s$variantSub )* )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
366 my $extensionSub = qr{(?: $singleton (?: $s$alphanum{2,8} )+ )}x
; # singleton 1*("-" (2*8alphanum))
367 my $extension = qr{(?: $extensionSub (?: $s$extensionSub )* )}x ; # singleton 1*("-" (2*8alphanum))
369 my $privateuse = qr{(?: $x (?: $s$alphanum{1,8} )+ )}x
; # ("x"/"X") 1*("-" (1*8alphanum))
371 # Define certain grandfathered codes, since otherwise the regex is pretty useless.
372 # Since these are limited, this is safe even later changes to the registry --
373 # the only oddity is that it might change the type of the tag, and thus
374 # the results from the capturing groups.
375 # http://www.iana.org/assignments/language-subtag-registry
376 # Note that these have to be compared case insensitively, requiring (?i) below.
378 my $grandfathered = qr{(?
: (?i
)
380 | i
$s (?
: ami
| bnn
| default | enochian
| hak
| klingon
| lux
| mingo
| navajo
| pwn
| tao
| tay
| tsu
)
381 | sgn
$s (?
: BE
$s fr
| BE
$s nl
| CH
$s de
)
384 # For well-formedness, we don't need the ones that would otherwise pass, so they are commented out here
388 # | en $s (?: boont | GB $s oed | scouse )
389 # | no $s (?: bok | nyn)
390 # | zh $s (?: cmn | cmn $s Hans | cmn $s Hant | gan | guoyu | hakka | min | min $s nan | wuu | xiang | yue)
392 # Here is the final breakdown, with capturing groups for each of these components
393 # The language, variants, extensions, grandfathered, and private-use may have interior '-'
395 #my $root = qr{(?: ($language) (?: $s ($script) )? 40% (?: $s ($region) )? 40% (?: $s ($variant) )? 10% (?: $s ($extension) )? 5% (?: $s ($privateuse) )? 5% ) 90% | ($grandfathered) 5% | ($privateuse) 5% };
397 $string =~ qr{^ (?:($language)) (?:$s($script))? (?:$s($region))? (?:$s($variant))? (?:$s($extension))? (?:$s($privateuse))? $}xi; # |($grandfathered) | ($privateuse) $}xi;
399 'rfc4646_subtag' => $string,
410 # Script Direction Resources:
411 # http://www.w3.org/International/questions/qa-scripts
413 my ($language_script)= @_;
414 my $dbh = C4
::Context
->dbh;
416 my $sth = $dbh->prepare('SELECT bidi FROM language_script_bidi WHERE rfc4646_subtag=?');
417 $sth->execute($language_script);
418 while (my $result = $sth->fetchrow_hashref) {
419 $bidi = $result->{'bidi'};
424 sub accept_language
{
425 # referenced http://search.cpan.org/src/CGILMORE/I18N-AcceptLanguage-1.04/lib/I18N/AcceptLanguage.pm
426 # FIXME: since this is only used in Output.pm as of Jan 8 2008, maybe it should be IN Output.pm
427 my ($clientPreferences,$supportedLanguages) = @_;
429 if ($clientPreferences) {
430 # There should be no whitespace anways, but a cleanliness/sanity check
431 $clientPreferences =~ s/\s//g;
433 # Prepare the list of client-acceptable languages
434 foreach my $tag (split(/,/, $clientPreferences)) {
435 my ($language, $quality) = split(/\;/, $tag);
436 $quality =~ s/^q=//i if $quality;
437 $quality = 1 unless $quality;
438 next if $quality <= 0;
439 # We want to force the wildcard to be last
440 $quality = 0 if ($language eq '*');
441 # Pushing lowercase language here saves processing later
442 push(@languages, { quality
=> $quality,
443 language
=> $language,
444 lclanguage
=> lc($language) });
447 carp
"accept_language(x,y) called with no clientPreferences (x).";
449 # Prepare the list of server-supported languages
450 my %supportedLanguages = ();
451 my %secondaryLanguages = ();
452 foreach my $language (@
$supportedLanguages) {
453 # warn "Language supported: " . $language->{language_code};
454 $supportedLanguages{lc($language->{language_code
})} = $language->{language_code
};
455 if ($language->{language_code
} =~ /^([^-]+)-?/) {
456 $secondaryLanguages{lc($1)} = $language->{language_code
};
460 # Reverse sort the list, making best quality at the front of the array
461 @languages = sort { $b->{quality
} <=> $a->{quality
} } @languages;
462 my $secondaryMatch = '';
463 foreach my $tag (@languages) {
464 if (exists($supportedLanguages{$tag->{lclanguage
}})) {
465 # Client en-us eq server en-us
466 return $supportedLanguages{$tag->{language
}} if exists($supportedLanguages{$tag->{language
}});
467 return $supportedLanguages{$tag->{lclanguage
}};
468 } elsif (exists($secondaryLanguages{$tag->{lclanguage
}})) {
469 # Client en eq server en-us
470 return $secondaryLanguages{$tag->{language
}} if exists($secondaryLanguages{$tag->{language
}});
471 return $supportedLanguages{$tag->{lclanguage
}};
472 } elsif ($tag->{lclanguage
} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
473 # Client en-gb eq server en-us
474 $secondaryMatch = $secondaryLanguages{$1};
475 } elsif ($tag->{lclanguage
} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
476 # FIXME: We just checked the exact same conditional!
477 # Client en-us eq server en
478 $secondaryMatch = $supportedLanguages{$1};
479 } elsif ($tag->{lclanguage
} eq '*') {
480 # * matches every language not already specified.
481 # It doesn't care which we pick, so let's pick the default,
482 # if available, then the first in the array.
483 #return $acceptor->defaultLanguage() if $acceptor->defaultLanguage();
484 return $supportedLanguages->[0];
487 # No primary matches. Secondary? (ie, en-us requested and en supported)
488 return $secondaryMatch if $secondaryMatch;
489 return undef; # else, we got nothing.