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 my $servers = C4::Context->config('memcached_servers');
31 require Memoize::Memcached;
32 import Memoize::Memcached qw(memoize_memcached);
35 servers
=> [ $servers ],
36 key_prefix
=> C4
::Context
->config('memcached_namespace') || 'koha',
39 memoize_memcached
('getTranslatedLanguages', memcached
=> $memcached, expire_time
=> 600); #cache for 10 minutes
40 memoize_memcached
('getFrameworkLanguages' , memcached
=> $memcached, expire_time
=> 600);
41 memoize_memcached
('getAllLanguages', memcached
=> $memcached, expire_time
=> 600);
50 &getFrameworkLanguages
51 &getTranslatedLanguages
54 @EXPORT_OK = qw(getFrameworkLanguages getTranslatedLanguages getAllLanguages get_bidi regex_lang_subtags language_get_description accept_language);
60 C4::Languages - Perl Module containing language list functions for Koha
70 =head2 getFrameworkLanguages
72 Returns a reference to an array of hashes:
74 my $languages = getFrameworkLanguages();
75 for my $language(@$languages) {
76 print "$language->{language_code}\n"; # language code in iso 639-2
77 print "$language->{language_name}\n"; # language name in native script
78 print "$language->{language_locale_name}\n"; # language name in current locale
83 sub getFrameworkLanguages
{
84 # get a hash with all language codes, names, and locale names
85 my $all_languages = getAllLanguages
();
88 # find the available directory names
89 my $dir=C4
::Context
->config('intranetdir')."/installer/data/";
91 my @listdir= grep { !/^\.|CVS/ && -d
"$dir/$_"} readdir(MYDIR
);
94 # pull out all data for the dir names that exist
95 for my $dirname (@listdir) {
96 for my $language_set (@
$all_languages) {
98 if ($dirname eq $language_set->{language_code
}) {
100 'language_code'=>$dirname,
101 'language_description'=>$language_set->{language_description
},
102 'native_descrition'=>$language_set->{language_native_description
} }
109 =head2 getTranslatedLanguages
111 Returns a reference to an array of hashes:
113 my $languages = getTranslatedLanguages();
114 print "Available translated languages:\n";
115 for my $language(@$trlanguages) {
116 print "$language->{language_code}\n"; # language code in iso 639-2
117 print "$language->{language_name}\n"; # language name in native script
118 print "$language->{language_locale_name}\n"; # language name in current locale
123 sub getTranslatedLanguages
{
124 my ($interface, $theme, $current_language, $which) = @_;
126 my $all_languages = getAllLanguages
();
128 my @enabled_languages;
130 if ($interface && $interface eq 'opac' ) {
131 @enabled_languages = split ",", C4
::Context
->preference('opaclanguages');
132 $htdocs = C4
::Context
->config('opachtdocs');
133 if ( $theme and -d
"$htdocs/$theme" ) {
134 (@languages) = _get_language_dirs
($htdocs,$theme);
137 for my $theme ( _get_themes
('opac') ) {
138 push @languages, _get_language_dirs
($htdocs,$theme);
142 elsif ($interface && $interface eq 'intranet' ) {
143 @enabled_languages = split ",", C4
::Context
->preference('language');
144 $htdocs = C4
::Context
->config('intrahtdocs');
145 if ( $theme and -d
"$htdocs/$theme" ) {
146 @languages = _get_language_dirs
($htdocs,$theme);
149 foreach my $theme ( _get_themes
('intranet') ) {
150 push @languages, _get_language_dirs
($htdocs,$theme);
155 @enabled_languages = split ",", C4
::Context
->preference('opaclanguages');
156 my $htdocs = C4
::Context
->config('intrahtdocs');
157 foreach my $theme ( _get_themes
('intranet') ) {
158 push @languages, _get_language_dirs
($htdocs,$theme);
160 $htdocs = C4
::Context
->config('opachtdocs');
161 foreach my $theme ( _get_themes
('opac') ) {
162 push @languages, _get_language_dirs
($htdocs,$theme);
165 $seen{$_}++ for @languages;
166 @languages = keys %seen;
168 return _build_languages_arrayref
($all_languages,\
@languages,$current_language,\
@enabled_languages);
171 =head2 getAllLanguages
173 Returns a reference to an array of hashes:
175 my $alllanguages = getAllLanguages();
176 print "Available translated languages:\n";
177 for my $language(@$alllanguages) {
178 print "$language->{language_code}\n";
179 print "$language->{language_name}\n";
180 print "$language->{language_locale_name}\n";
185 sub getAllLanguages
{
187 my $dbh=C4
::Context
->dbh;
188 my $current_language = shift || 'en';
189 my $sth = $dbh->prepare('SELECT * FROM language_subtag_registry WHERE type=\'language\'');
191 while (my $language_subtag_registry = $sth->fetchrow_hashref) {
193 # pull out all the script descriptions for each language
194 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 = ?");
195 $sth2->execute($language_subtag_registry->{subtag
},$current_language);
197 my $sth3 = $dbh->prepare("SELECT description FROM language_descriptions WHERE type='language' AND subtag=? AND lang=?");
199 # add the correct description info
200 while (my $language_descriptions = $sth2->fetchrow_hashref) {
201 $sth3->execute($language_subtag_registry->{subtag
},$language_subtag_registry->{subtag
});
202 my $native_description;
203 while (my $description = $sth3->fetchrow_hashref) {
204 $native_description = $description->{description
};
207 # fill in the ISO6329 code
208 $language_subtag_registry->{iso639_2_code
} = $language_descriptions->{iso639_2_code
};
209 # fill in the native description of the language, as well as the current language's translation of that if it exists
210 if ($native_description) {
211 $language_subtag_registry->{language_description
} = $native_description;
212 $language_subtag_registry->{language_description
}.=" ($language_descriptions->{description})" if $language_descriptions->{description
};
215 $language_subtag_registry->{language_description
} = $language_descriptions->{description
};
218 push @languages_loop, $language_subtag_registry;
220 return \
@languages_loop;
225 Internal function, returns an array of all available themes.
227 (@themes) = &_get_themes('opac');
228 (@themes) = &_get_themes('intranet');
233 my $interface = shift;
236 if ( $interface eq 'intranet' ) {
237 $htdocs = C4
::Context
->config('intrahtdocs');
240 $htdocs = C4
::Context
->config('opachtdocs');
242 opendir D
, "$htdocs";
243 my @dirlist = readdir D
;
244 foreach my $directory (@dirlist) {
245 # if there's an en dir, it's a valid theme
246 -d
"$htdocs/$directory/en" and push @themes, $directory;
251 =head2 _get_language_dirs
253 Internal function, returns an array of directory names, excluding non-language directories
257 sub _get_language_dirs
{
258 my ($htdocs,$theme) = @_;
260 opendir D
, "$htdocs/$theme";
261 for my $lang_string ( readdir D
) {
262 next if $lang_string =~/^\./;
263 next if $lang_string eq 'all';
264 next if $lang_string =~/png$/;
265 next if $lang_string =~/css$/;
266 next if $lang_string =~/CVS$/;
267 next if $lang_string =~/\.txt$/i; #Don't read the readme.txt !
268 next if $lang_string =~/img|images|famfam|sound/;
269 push @lang_strings, $lang_string;
271 return (@lang_strings);
274 =head2 _build_languages_arrayref
276 Internal function for building the ref to array of hashes
278 FIXME: this could be rewritten and simplified using map
282 sub _build_languages_arrayref
{
283 my ($all_languages,$translated_languages,$current_language,$enabled_languages) = @_;
284 my @translated_languages = @
$translated_languages;
285 my @languages_loop; # the final reference to an array of hashrefs
286 my @enabled_languages = @
$enabled_languages;
287 # how many languages are enabled, if one, take note, some contexts won't need to display it
288 my %seen_languages; # the language tags we've seen
291 my $track_language_groups;
292 my $current_language_regex = regex_lang_subtags
($current_language);
293 # Loop through the translated languages
294 for my $translated_language (@translated_languages) {
295 # separate the language string into its subtag types
296 my $language_subtags_hashref = regex_lang_subtags
($translated_language);
298 # is this language string 'enabled'?
299 for my $enabled_language (@enabled_languages) {
300 #warn "Checking out if $translated_language eq $enabled_language";
301 $language_subtags_hashref->{'enabled'} = 1 if $translated_language eq $enabled_language;
304 # group this language, key by langtag
305 $language_subtags_hashref->{'sublanguage_current'} = 1 if $translated_language eq $current_language;
306 $language_subtags_hashref->{'rfc4646_subtag'} = $translated_language;
307 $language_subtags_hashref->{'native_description'} = language_get_description
($language_subtags_hashref->{language
},$language_subtags_hashref->{language
},'language');
308 $language_subtags_hashref->{'script_description'} = language_get_description
($language_subtags_hashref->{script
},$language_subtags_hashref->{'language'},'script');
309 $language_subtags_hashref->{'region_description'} = language_get_description
($language_subtags_hashref->{region
},$language_subtags_hashref->{'language'},'region');
310 $language_subtags_hashref->{'variant_description'} = language_get_description
($language_subtags_hashref->{variant
},$language_subtags_hashref->{'language'},'variant');
311 $track_language_groups->{$language_subtags_hashref->{'language'}}++;
312 push ( @
{ $language_groups->{$language_subtags_hashref->{language
}} }, $language_subtags_hashref );
314 # $key is a language subtag like 'en'
315 while( my ($key, $value) = each %$language_groups) {
317 # is this language group enabled? are any of the languages within it enabled?
319 for my $enabled_language (@enabled_languages) {
320 my $regex_enabled_language = regex_lang_subtags
($enabled_language);
321 $enabled = 1 if $key eq $regex_enabled_language->{language
};
323 push @languages_loop, {
324 # this is only use if there is one
325 rfc4646_subtag
=> @
$value[0]->{rfc4646_subtag
},
326 native_description
=> language_get_description
($key,$key,'language'),
328 sublanguages_loop
=> $value,
329 plural
=> $track_language_groups->{$key} >1 ?
1 : 0,
330 current
=> $current_language_regex->{language
} eq $key ?
1 : 0,
331 group_enabled
=> $enabled,
334 return \
@languages_loop;
337 sub language_get_description
{
338 my ($script,$lang,$type) = @_;
339 my $dbh = C4
::Context
->dbh;
341 my $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
342 #warn "QUERY: SELECT description FROM language_descriptions WHERE subtag=$script AND lang=$lang AND type=$type";
343 $sth->execute($script,$lang,$type);
344 while (my $descriptions = $sth->fetchrow_hashref) {
345 $desc = $descriptions->{'description'};
348 $sth = $dbh->prepare("SELECT description FROM language_descriptions WHERE subtag=? AND lang=? AND type=?");
349 $sth->execute($script,'en',$type);
350 while (my $descriptions = $sth->fetchrow_hashref) {
351 $desc = $descriptions->{'description'};
356 =head2 regex_lang_subtags
358 This internal sub takes a string composed according to RFC 4646 as
359 an input and returns a reference to a hash containing keys and values
360 for ( language, script, region, variant, extension, privateuse )
364 sub regex_lang_subtags
{
367 # Regex for recognizing RFC 4646 well-formed tags
368 # http://www.rfc-editor.org/rfc/rfc4646.txt
370 # regexes based on : http://unicode.org/cldr/data/tools/java/org/unicode/cldr/util/data/langtagRegex.txt
371 # The structure requires no forward references, so it reverses the order.
372 # The uppercase comments are fragments copied from RFC 4646
374 # Note: the tool requires that any real "=" or "#" or ";" in the regex be escaped.
376 my $alpha = qr/[a-zA-Z]/ ; # ALPHA
377 my $digit = qr/[0-9]/ ; # DIGIT
378 my $alphanum = qr/[a-zA-Z0-9]/ ; # ALPHA / DIGIT
379 my $x = qr/[xX]/ ; # private use singleton
380 my $singleton = qr/[a-w y-z A-W Y-Z]/ ; # other singleton
381 my $s = qr/[-]/ ; # separator -- lenient parsers will use [-_]
383 # Now do the components. The structure is slightly different to allow for capturing the right components.
384 # The notation (?:....) is a non-capturing version of (...): so the "?:" can be deleted if someone doesn't care about capturing.
386 my $extlang = qr{(?: $s $alpha{3} )}x
; # *3("-" 3ALPHA)
387 my $language = qr{(?: $alpha{2,3} | $alpha{4,8} )}x
;
388 #my $language = qr{(?: $alpha{2,3}$extlang{0,3} | $alpha{4,8} )}x ; # (2*3ALPHA [ extlang ]) / 4ALPHA / 5*8ALPHA
390 my $script = qr{(?: $alpha{4} )}x
; # 4ALPHA
392 my $region = qr{(?: $alpha{2} | $digit{3} )}x
; # 2ALPHA / 3DIGIT
394 my $variantSub = qr{(?: $digit$alphanum{3} | $alphanum{5,8} )}x
; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
395 my $variant = qr{(?: $variantSub (?: $s$variantSub )* )}x ; # *("-" variant), 5*8alphanum / (DIGIT 3alphanum)
397 my $extensionSub = qr{(?: $singleton (?: $s$alphanum{2,8} )+ )}x
; # singleton 1*("-" (2*8alphanum))
398 my $extension = qr{(?: $extensionSub (?: $s$extensionSub )* )}x ; # singleton 1*("-" (2*8alphanum))
400 my $privateuse = qr{(?: $x (?: $s$alphanum{1,8} )+ )}x
; # ("x"/"X") 1*("-" (1*8alphanum))
402 # Define certain grandfathered codes, since otherwise the regex is pretty useless.
403 # Since these are limited, this is safe even later changes to the registry --
404 # the only oddity is that it might change the type of the tag, and thus
405 # the results from the capturing groups.
406 # http://www.iana.org/assignments/language-subtag-registry
407 # Note that these have to be compared case insensitively, requiring (?i) below.
409 my $grandfathered = qr{(?
: (?i
)
411 | i
$s (?
: ami
| bnn
| default | enochian
| hak
| klingon
| lux
| mingo
| navajo
| pwn
| tao
| tay
| tsu
)
412 | sgn
$s (?
: BE
$s fr
| BE
$s nl
| CH
$s de
)
415 # For well-formedness, we don't need the ones that would otherwise pass, so they are commented out here
419 # | en $s (?: boont | GB $s oed | scouse )
420 # | no $s (?: bok | nyn)
421 # | zh $s (?: cmn | cmn $s Hans | cmn $s Hant | gan | guoyu | hakka | min | min $s nan | wuu | xiang | yue)
423 # Here is the final breakdown, with capturing groups for each of these components
424 # The language, variants, extensions, grandfathered, and private-use may have interior '-'
426 #my $root = qr{(?: ($language) (?: $s ($script) )? 40% (?: $s ($region) )? 40% (?: $s ($variant) )? 10% (?: $s ($extension) )? 5% (?: $s ($privateuse) )? 5% ) 90% | ($grandfathered) 5% | ($privateuse) 5% };
428 $string =~ qr{^ (?:($language)) (?:$s($script))? (?:$s($region))? (?:$s($variant))? (?:$s($extension))? (?:$s($privateuse))? $}xi; # |($grandfathered) | ($privateuse) $}xi;
430 'rfc4646_subtag' => $string,
441 # Script Direction Resources:
442 # http://www.w3.org/International/questions/qa-scripts
444 my ($language_script)= @_;
445 my $dbh = C4
::Context
->dbh;
447 my $sth = $dbh->prepare('SELECT bidi FROM language_script_bidi WHERE rfc4646_subtag=?');
448 $sth->execute($language_script);
449 while (my $result = $sth->fetchrow_hashref) {
450 $bidi = $result->{'bidi'};
455 sub accept_language
{
456 # referenced http://search.cpan.org/src/CGILMORE/I18N-AcceptLanguage-1.04/lib/I18N/AcceptLanguage.pm
457 # FIXME: since this is only used in Output.pm as of Jan 8 2008, maybe it should be IN Output.pm
458 my ($clientPreferences,$supportedLanguages) = @_;
460 if ($clientPreferences) {
461 # There should be no whitespace anways, but a cleanliness/sanity check
462 $clientPreferences =~ s/\s//g;
463 # Prepare the list of client-acceptable languages
464 foreach my $tag (split(/,/, $clientPreferences)) {
465 my ($language, $quality) = split(/\;/, $tag);
466 $quality =~ s/^q=//i if $quality;
467 $quality = 1 unless $quality;
468 next if $quality <= 0;
469 # We want to force the wildcard to be last
470 $quality = 0 if ($language eq '*');
471 # Pushing lowercase language here saves processing later
472 push(@languages, { quality
=> $quality,
473 language
=> $language,
474 lclanguage
=> lc($language) });
477 carp
"accept_language(x,y) called with no clientPreferences (x).";
479 # Prepare the list of server-supported languages
480 my %supportedLanguages = ();
481 my %secondaryLanguages = ();
482 foreach my $language (@
$supportedLanguages) {
483 # warn "Language supported: " . $language->{language};
484 my $subtag = $language->{rfc4646_subtag
};
485 $supportedLanguages{lc($subtag)} = $subtag;
486 if ( $subtag =~ /^([^-]+)-?/ ) {
487 $secondaryLanguages{lc($1)} = $subtag;
491 # Reverse sort the list, making best quality at the front of the array
492 @languages = sort { $b->{quality
} <=> $a->{quality
} } @languages;
493 my $secondaryMatch = '';
494 foreach my $tag (@languages) {
495 if (exists($supportedLanguages{$tag->{lclanguage
}})) {
496 # Client en-us eq server en-us
497 return $supportedLanguages{$tag->{language
}} if exists($supportedLanguages{$tag->{language
}});
498 return $supportedLanguages{$tag->{lclanguage
}};
499 } elsif (exists($secondaryLanguages{$tag->{lclanguage
}})) {
500 # Client en eq server en-us
501 return $secondaryLanguages{$tag->{language
}} if exists($secondaryLanguages{$tag->{language
}});
502 return $supportedLanguages{$tag->{lclanguage
}};
503 } elsif ($tag->{lclanguage
} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
504 # Client en-gb eq server en-us
505 $secondaryMatch = $secondaryLanguages{$1};
506 } elsif ($tag->{lclanguage
} =~ /^([^-]+)-/ && exists($secondaryLanguages{$1}) && $secondaryMatch eq '') {
507 # FIXME: We just checked the exact same conditional!
508 # Client en-us eq server en
509 $secondaryMatch = $supportedLanguages{$1};
510 } elsif ($tag->{lclanguage
} eq '*') {
511 # * matches every language not already specified.
512 # It doesn't care which we pick, so let's pick the default,
513 # if available, then the first in the array.
514 #return $acceptor->defaultLanguage() if $acceptor->defaultLanguage();
515 return $supportedLanguages->[0];
518 # No primary matches. Secondary? (ie, en-us requested and en supported)
519 return $secondaryMatch if $secondaryMatch;
520 return undef; # else, we got nothing.