3 * Internationalisation code.
5 * This program is free software; you can redistribute it and/or modify
6 * it under the terms of the GNU General Public License as published by
7 * the Free Software Foundation; either version 2 of the License, or
8 * (at your option) any later version.
10 * This program is distributed in the hope that it will be useful,
11 * but WITHOUT ANY WARRANTY; without even the implied warranty of
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 * GNU General Public License for more details.
15 * You should have received a copy of the GNU General Public License along
16 * with this program; if not, write to the Free Software Foundation, Inc.,
17 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
18 * http://www.gnu.org/copyleft/gpl.html
25 * @defgroup Language Language
28 if ( !defined( 'MEDIAWIKI' ) ) {
29 echo "This file is part of MediaWiki, it is not a valid entry point.\n";
34 global $wgLanguageNames;
35 require_once( __DIR__
. '/Names.php' );
37 if ( function_exists( 'mb_strtoupper' ) ) {
38 mb_internal_encoding( 'UTF-8' );
42 * a fake language converter
51 function __construct( $langobj ) { $this->mLang
= $langobj; }
52 function autoConvertToAllVariants( $text ) { return array( $this->mLang
->getCode() => $text ); }
53 function convert( $t ) { return $t; }
54 function convertTo( $text, $variant ) { return $text; }
55 function convertTitle( $t ) { return $t->getPrefixedText(); }
56 function convertNamespace( $ns ) { return $this->mLang
->getFormattedNsText( $ns ); }
57 function getVariants() { return array( $this->mLang
->getCode() ); }
58 function getPreferredVariant() { return $this->mLang
->getCode(); }
59 function getDefaultVariant() { return $this->mLang
->getCode(); }
60 function getURLVariant() { return ''; }
61 function getConvRuleTitle() { return false; }
62 function findVariantLink( &$l, &$n, $ignoreOtherCond = false ) { }
63 function getExtraHashOptions() { return ''; }
64 function getParsedTitle() { return ''; }
65 function markNoConversion( $text, $noParse = false ) { return $text; }
66 function convertCategoryKey( $key ) { return $key; }
67 function convertLinkToAllVariants( $text ) { return $this->autoConvertToAllVariants( $text ); }
68 function armourMath( $text ) { return $text; }
72 * Internationalisation code
78 * @var LanguageConverter
82 public $mVariants, $mCode, $mLoaded = false;
83 public $mMagicExtensions = array(), $mMagicHookDone = false;
84 private $mHtmlCode = null;
86 public $dateFormatStrings = array();
87 public $mExtendedSpecialPageAliases;
89 protected $namespaceNames, $mNamespaceIds, $namespaceAliases;
92 * ReplacementArray object caches
94 public $transformData = array();
97 * @var LocalisationCache
99 static public $dataCache;
101 static public $mLangObjCache = array();
103 static public $mWeekdayMsgs = array(
104 'sunday', 'monday', 'tuesday', 'wednesday', 'thursday',
108 static public $mWeekdayAbbrevMsgs = array(
109 'sun', 'mon', 'tue', 'wed', 'thu', 'fri', 'sat'
112 static public $mMonthMsgs = array(
113 'january', 'february', 'march', 'april', 'may_long', 'june',
114 'july', 'august', 'september', 'october', 'november',
117 static public $mMonthGenMsgs = array(
118 'january-gen', 'february-gen', 'march-gen', 'april-gen', 'may-gen', 'june-gen',
119 'july-gen', 'august-gen', 'september-gen', 'october-gen', 'november-gen',
122 static public $mMonthAbbrevMsgs = array(
123 'jan', 'feb', 'mar', 'apr', 'may', 'jun', 'jul', 'aug',
124 'sep', 'oct', 'nov', 'dec'
127 static public $mIranianCalendarMonthMsgs = array(
128 'iranian-calendar-m1', 'iranian-calendar-m2', 'iranian-calendar-m3',
129 'iranian-calendar-m4', 'iranian-calendar-m5', 'iranian-calendar-m6',
130 'iranian-calendar-m7', 'iranian-calendar-m8', 'iranian-calendar-m9',
131 'iranian-calendar-m10', 'iranian-calendar-m11', 'iranian-calendar-m12'
134 static public $mHebrewCalendarMonthMsgs = array(
135 'hebrew-calendar-m1', 'hebrew-calendar-m2', 'hebrew-calendar-m3',
136 'hebrew-calendar-m4', 'hebrew-calendar-m5', 'hebrew-calendar-m6',
137 'hebrew-calendar-m7', 'hebrew-calendar-m8', 'hebrew-calendar-m9',
138 'hebrew-calendar-m10', 'hebrew-calendar-m11', 'hebrew-calendar-m12',
139 'hebrew-calendar-m6a', 'hebrew-calendar-m6b'
142 static public $mHebrewCalendarMonthGenMsgs = array(
143 'hebrew-calendar-m1-gen', 'hebrew-calendar-m2-gen', 'hebrew-calendar-m3-gen',
144 'hebrew-calendar-m4-gen', 'hebrew-calendar-m5-gen', 'hebrew-calendar-m6-gen',
145 'hebrew-calendar-m7-gen', 'hebrew-calendar-m8-gen', 'hebrew-calendar-m9-gen',
146 'hebrew-calendar-m10-gen', 'hebrew-calendar-m11-gen', 'hebrew-calendar-m12-gen',
147 'hebrew-calendar-m6a-gen', 'hebrew-calendar-m6b-gen'
150 static public $mHijriCalendarMonthMsgs = array(
151 'hijri-calendar-m1', 'hijri-calendar-m2', 'hijri-calendar-m3',
152 'hijri-calendar-m4', 'hijri-calendar-m5', 'hijri-calendar-m6',
153 'hijri-calendar-m7', 'hijri-calendar-m8', 'hijri-calendar-m9',
154 'hijri-calendar-m10', 'hijri-calendar-m11', 'hijri-calendar-m12'
161 static public $durationIntervals = array(
162 'millennia' => 31556952000,
163 'centuries' => 3155695200,
164 'decades' => 315569520,
165 'years' => 31556952, // 86400 * ( 365 + ( 24 * 3 + 25 ) / 400 )
174 * Get a cached or new language object for a given language code
175 * @param $code String
178 static function factory( $code ) {
179 global $wgDummyLanguageCodes, $wgLangObjCacheSize;
181 if ( isset( $wgDummyLanguageCodes[$code] ) ) {
182 $code = $wgDummyLanguageCodes[$code];
185 // get the language object to process
186 $langObj = isset( self
::$mLangObjCache[$code] )
187 ? self
::$mLangObjCache[$code]
188 : self
::newFromCode( $code );
190 // merge the language object in to get it up front in the cache
191 self
::$mLangObjCache = array_merge( array( $code => $langObj ), self
::$mLangObjCache );
192 // get rid of the oldest ones in case we have an overflow
193 self
::$mLangObjCache = array_slice( self
::$mLangObjCache, 0, $wgLangObjCacheSize, true );
199 * Create a language object for a given language code
200 * @param $code String
201 * @throws MWException
204 protected static function newFromCode( $code ) {
205 // Protect against path traversal below
206 if ( !Language
::isValidCode( $code )
207 ||
strcspn( $code, ":/\\\000" ) !== strlen( $code ) )
209 throw new MWException( "Invalid language code \"$code\"" );
212 if ( !Language
::isValidBuiltInCode( $code ) ) {
213 // It's not possible to customise this code with class files, so
214 // just return a Language object. This is to support uselang= hacks.
215 $lang = new Language
;
216 $lang->setCode( $code );
220 // Check if there is a language class for the code
221 $class = self
::classFromCode( $code );
222 self
::preloadLanguageClass( $class );
223 if ( MWInit
::classExists( $class ) ) {
228 // Keep trying the fallback list until we find an existing class
229 $fallbacks = Language
::getFallbacksFor( $code );
230 foreach ( $fallbacks as $fallbackCode ) {
231 if ( !Language
::isValidBuiltInCode( $fallbackCode ) ) {
232 throw new MWException( "Invalid fallback '$fallbackCode' in fallback sequence for '$code'" );
235 $class = self
::classFromCode( $fallbackCode );
236 self
::preloadLanguageClass( $class );
237 if ( MWInit
::classExists( $class ) ) {
238 $lang = Language
::newFromCode( $fallbackCode );
239 $lang->setCode( $code );
244 throw new MWException( "Invalid fallback sequence for language '$code'" );
248 * Checks whether any localisation is available for that language tag
249 * in MediaWiki (MessagesXx.php exists).
251 * @param string $code Language tag (in lower case)
252 * @return bool Whether language is supported
255 public static function isSupportedLanguage( $code ) {
256 return $code === strtolower( $code ) && is_readable( self
::getMessagesFileName( $code ) );
260 * Returns true if a language code string is a well-formed language tag
261 * according to RFC 5646.
262 * This function only checks well-formedness; it doesn't check that
263 * language, script or variant codes actually exist in the repositories.
265 * Based on regexes by Mark Davis of the Unicode Consortium:
266 * http://unicode.org/repos/cldr/trunk/tools/java/org/unicode/cldr/util/data/langtagRegex.txt
268 * @param $code string
269 * @param $lenient boolean Whether to allow '_' as separator. The default is only '-'.
274 public static function isWellFormedLanguageTag( $code, $lenient = false ) {
277 $alphanum = '[a-z0-9]';
278 $x = 'x' ; # private use singleton
279 $singleton = '[a-wy-z]'; # other singleton
280 $s = $lenient ?
'[-_]' : '-';
282 $language = "$alpha{2,8}|$alpha{2,3}$s$alpha{3}";
283 $script = "$alpha{4}"; # ISO 15924
284 $region = "(?:$alpha{2}|$digit{3})"; # ISO 3166-1 alpha-2 or UN M.49
285 $variant = "(?:$alphanum{5,8}|$digit$alphanum{3})";
286 $extension = "$singleton(?:$s$alphanum{2,8})+";
287 $privateUse = "$x(?:$s$alphanum{1,8})+";
289 # Define certain grandfathered codes, since otherwise the regex is pretty useless.
290 # Since these are limited, this is safe even later changes to the registry --
291 # the only oddity is that it might change the type of the tag, and thus
292 # the results from the capturing groups.
293 # http://www.iana.org/assignments/language-subtag-registry
295 $grandfathered = "en{$s}GB{$s}oed"
296 . "|i{$s}(?:ami|bnn|default|enochian|hak|klingon|lux|mingo|navajo|pwn|tao|tay|tsu)"
297 . "|no{$s}(?:bok|nyn)"
298 . "|sgn{$s}(?:BE{$s}(?:fr|nl)|CH{$s}de)"
299 . "|zh{$s}min{$s}nan";
301 $variantList = "$variant(?:$s$variant)*";
302 $extensionList = "$extension(?:$s$extension)*";
304 $langtag = "(?:($language)"
307 . "(?:$s$variantList)?"
308 . "(?:$s$extensionList)?"
309 . "(?:$s$privateUse)?)";
311 # The final breakdown, with capturing groups for each of these components
312 # The variants, extensions, grandfathered, and private-use may have interior '-'
314 $root = "^(?:$langtag|$privateUse|$grandfathered)$";
316 return (bool)preg_match( "/$root/", strtolower( $code ) );
320 * Returns true if a language code string is of a valid form, whether or
321 * not it exists. This includes codes which are used solely for
322 * customisation via the MediaWiki namespace.
324 * @param $code string
328 public static function isValidCode( $code ) {
330 // People think language codes are html safe, so enforce it.
331 // Ideally we should only allow a-zA-Z0-9-
332 // but, .+ and other chars are often used for {{int:}} hacks
333 // see bugs 37564, 37587, 36938
334 strcspn( $code, ":/\\\000&<>'\"" ) === strlen( $code )
335 && !preg_match( Title
::getTitleInvalidRegex(), $code );
339 * Returns true if a language code is of a valid form for the purposes of
340 * internal customisation of MediaWiki, via Messages*.php.
342 * @param $code string
344 * @throws MWException
348 public static function isValidBuiltInCode( $code ) {
350 if ( !is_string( $code ) ) {
351 if ( is_object( $code ) ) {
352 $addmsg = " of class " . get_class( $code );
356 $type = gettype( $code );
357 throw new MWException( __METHOD__
. " must be passed a string, $type given$addmsg" );
360 return (bool)preg_match( '/^[a-z0-9-]{2,}$/i', $code );
364 * Returns true if a language code is an IETF tag known to MediaWiki.
366 * @param $code string
371 public static function isKnownLanguageTag( $tag ) {
372 static $coreLanguageNames;
374 if ( $coreLanguageNames === null ) {
375 include( MWInit
::compiledPath( 'languages/Names.php' ) );
378 if ( isset( $coreLanguageNames[$tag] )
379 || self
::fetchLanguageName( $tag, $tag ) !== ''
389 * @return String Name of the language class
391 public static function classFromCode( $code ) {
392 if ( $code == 'en' ) {
395 return 'Language' . str_replace( '-', '_', ucfirst( $code ) );
400 * Includes language class files
402 * @param $class string Name of the language class
404 public static function preloadLanguageClass( $class ) {
407 if ( $class === 'Language' ) {
411 if ( !defined( 'MW_COMPILED' ) ) {
412 if ( file_exists( "$IP/languages/classes/$class.php" ) ) {
413 include_once( "$IP/languages/classes/$class.php" );
419 * Get the LocalisationCache instance
421 * @return LocalisationCache
423 public static function getLocalisationCache() {
424 if ( is_null( self
::$dataCache ) ) {
425 global $wgLocalisationCacheConf;
426 $class = $wgLocalisationCacheConf['class'];
427 self
::$dataCache = new $class( $wgLocalisationCacheConf );
429 return self
::$dataCache;
432 function __construct() {
433 $this->mConverter
= new FakeConverter( $this );
434 // Set the code to the name of the descendant
435 if ( get_class( $this ) == 'Language' ) {
438 $this->mCode
= str_replace( '_', '-', strtolower( substr( get_class( $this ), 8 ) ) );
440 self
::getLocalisationCache();
444 * Reduce memory usage
446 function __destruct() {
447 foreach ( $this as $name => $value ) {
448 unset( $this->$name );
453 * Hook which will be called if this is the content language.
454 * Descendants can use this to register hook functions or modify globals
456 function initContLang() { }
459 * Same as getFallbacksFor for current language.
461 * @deprecated in 1.19
463 function getFallbackLanguageCode() {
464 wfDeprecated( __METHOD__
, '1.19' );
465 return self
::getFallbackFor( $this->mCode
);
472 function getFallbackLanguages() {
473 return self
::getFallbacksFor( $this->mCode
);
477 * Exports $wgBookstoreListEn
480 function getBookstoreList() {
481 return self
::$dataCache->getItem( $this->mCode
, 'bookstoreList' );
485 * Returns an array of localised namespaces indexed by their numbers. If the namespace is not
486 * available in localised form, it will be included in English.
490 public function getNamespaces() {
491 if ( is_null( $this->namespaceNames
) ) {
492 global $wgMetaNamespace, $wgMetaNamespaceTalk, $wgExtraNamespaces;
494 $this->namespaceNames
= self
::$dataCache->getItem( $this->mCode
, 'namespaceNames' );
495 $validNamespaces = MWNamespace
::getCanonicalNamespaces();
497 $this->namespaceNames
= $wgExtraNamespaces +
$this->namespaceNames +
$validNamespaces;
499 $this->namespaceNames
[NS_PROJECT
] = $wgMetaNamespace;
500 if ( $wgMetaNamespaceTalk ) {
501 $this->namespaceNames
[NS_PROJECT_TALK
] = $wgMetaNamespaceTalk;
503 $talk = $this->namespaceNames
[NS_PROJECT_TALK
];
504 $this->namespaceNames
[NS_PROJECT_TALK
] =
505 $this->fixVariableInNamespace( $talk );
508 # Sometimes a language will be localised but not actually exist on this wiki.
509 foreach ( $this->namespaceNames
as $key => $text ) {
510 if ( !isset( $validNamespaces[$key] ) ) {
511 unset( $this->namespaceNames
[$key] );
515 # The above mixing may leave namespaces out of canonical order.
516 # Re-order by namespace ID number...
517 ksort( $this->namespaceNames
);
519 wfRunHooks( 'LanguageGetNamespaces', array( &$this->namespaceNames
) );
521 return $this->namespaceNames
;
525 * Arbitrarily set all of the namespace names at once. Mainly used for testing
526 * @param $namespaces Array of namespaces (id => name)
528 public function setNamespaces( array $namespaces ) {
529 $this->namespaceNames
= $namespaces;
530 $this->mNamespaceIds
= null;
534 * Resets all of the namespace caches. Mainly used for testing
536 public function resetNamespaces() {
537 $this->namespaceNames
= null;
538 $this->mNamespaceIds
= null;
539 $this->namespaceAliases
= null;
543 * A convenience function that returns the same thing as
544 * getNamespaces() except with the array values changed to ' '
545 * where it found '_', useful for producing output to be displayed
546 * e.g. in <select> forms.
550 function getFormattedNamespaces() {
551 $ns = $this->getNamespaces();
552 foreach ( $ns as $k => $v ) {
553 $ns[$k] = strtr( $v, '_', ' ' );
559 * Get a namespace value by key
561 * $mw_ns = $wgContLang->getNsText( NS_MEDIAWIKI );
562 * echo $mw_ns; // prints 'MediaWiki'
565 * @param $index Int: the array key of the namespace to return
566 * @return mixed, string if the namespace value exists, otherwise false
568 function getNsText( $index ) {
569 $ns = $this->getNamespaces();
570 return isset( $ns[$index] ) ?
$ns[$index] : false;
574 * A convenience function that returns the same thing as
575 * getNsText() except with '_' changed to ' ', useful for
579 * $mw_ns = $wgContLang->getFormattedNsText( NS_MEDIAWIKI_TALK );
580 * echo $mw_ns; // prints 'MediaWiki talk'
583 * @param int $index The array key of the namespace to return
584 * @return string Namespace name without underscores (empty string if namespace does not exist)
586 function getFormattedNsText( $index ) {
587 $ns = $this->getNsText( $index );
588 return strtr( $ns, '_', ' ' );
592 * Returns gender-dependent namespace alias if available.
593 * @param $index Int: namespace index
594 * @param $gender String: gender key (male, female... )
598 function getGenderNsText( $index, $gender ) {
599 global $wgExtraGenderNamespaces;
601 $ns = $wgExtraGenderNamespaces + self
::$dataCache->getItem( $this->mCode
, 'namespaceGenderAliases' );
602 return isset( $ns[$index][$gender] ) ?
$ns[$index][$gender] : $this->getNsText( $index );
606 * Whether this language makes distinguishes genders for example in
611 function needsGenderDistinction() {
612 global $wgExtraGenderNamespaces, $wgExtraNamespaces;
613 if ( count( $wgExtraGenderNamespaces ) > 0 ) {
614 // $wgExtraGenderNamespaces overrides everything
616 } elseif ( isset( $wgExtraNamespaces[NS_USER
] ) && isset( $wgExtraNamespaces[NS_USER_TALK
] ) ) {
617 /// @todo There may be other gender namespace than NS_USER & NS_USER_TALK in the future
618 // $wgExtraNamespaces overrides any gender aliases specified in i18n files
621 // Check what is in i18n files
622 $aliases = self
::$dataCache->getItem( $this->mCode
, 'namespaceGenderAliases' );
623 return count( $aliases ) > 0;
628 * Get a namespace key by value, case insensitive.
629 * Only matches namespace names for the current language, not the
630 * canonical ones defined in Namespace.php.
632 * @param $text String
633 * @return mixed An integer if $text is a valid value otherwise false
635 function getLocalNsIndex( $text ) {
636 $lctext = $this->lc( $text );
637 $ids = $this->getNamespaceIds();
638 return isset( $ids[$lctext] ) ?
$ids[$lctext] : false;
644 function getNamespaceAliases() {
645 if ( is_null( $this->namespaceAliases
) ) {
646 $aliases = self
::$dataCache->getItem( $this->mCode
, 'namespaceAliases' );
650 foreach ( $aliases as $name => $index ) {
651 if ( $index === NS_PROJECT_TALK
) {
652 unset( $aliases[$name] );
653 $name = $this->fixVariableInNamespace( $name );
654 $aliases[$name] = $index;
659 global $wgExtraGenderNamespaces;
660 $genders = $wgExtraGenderNamespaces +
(array)self
::$dataCache->getItem( $this->mCode
, 'namespaceGenderAliases' );
661 foreach ( $genders as $index => $forms ) {
662 foreach ( $forms as $alias ) {
663 $aliases[$alias] = $index;
667 $this->namespaceAliases
= $aliases;
669 return $this->namespaceAliases
;
675 function getNamespaceIds() {
676 if ( is_null( $this->mNamespaceIds
) ) {
677 global $wgNamespaceAliases;
678 # Put namespace names and aliases into a hashtable.
679 # If this is too slow, then we should arrange it so that it is done
680 # before caching. The catch is that at pre-cache time, the above
681 # class-specific fixup hasn't been done.
682 $this->mNamespaceIds
= array();
683 foreach ( $this->getNamespaces() as $index => $name ) {
684 $this->mNamespaceIds
[$this->lc( $name )] = $index;
686 foreach ( $this->getNamespaceAliases() as $name => $index ) {
687 $this->mNamespaceIds
[$this->lc( $name )] = $index;
689 if ( $wgNamespaceAliases ) {
690 foreach ( $wgNamespaceAliases as $name => $index ) {
691 $this->mNamespaceIds
[$this->lc( $name )] = $index;
695 return $this->mNamespaceIds
;
699 * Get a namespace key by value, case insensitive. Canonical namespace
700 * names override custom ones defined for the current language.
702 * @param $text String
703 * @return mixed An integer if $text is a valid value otherwise false
705 function getNsIndex( $text ) {
706 $lctext = $this->lc( $text );
707 $ns = MWNamespace
::getCanonicalIndex( $lctext );
708 if ( $ns !== null ) {
711 $ids = $this->getNamespaceIds();
712 return isset( $ids[$lctext] ) ?
$ids[$lctext] : false;
716 * short names for language variants used for language conversion links.
718 * @param $code String
719 * @param $usemsg bool Use the "variantname-xyz" message if it exists
722 function getVariantname( $code, $usemsg = true ) {
723 $msg = "variantname-$code";
724 if ( $usemsg && wfMessage( $msg )->exists() ) {
725 return $this->getMessageFromDB( $msg );
727 $name = self
::fetchLanguageName( $code );
729 return $name; # if it's defined as a language name, show that
731 # otherwise, output the language code
737 * @param $name string
740 function specialPage( $name ) {
741 $aliases = $this->getSpecialPageAliases();
742 if ( isset( $aliases[$name][0] ) ) {
743 $name = $aliases[$name][0];
745 return $this->getNsText( NS_SPECIAL
) . ':' . $name;
751 function getDatePreferences() {
752 return self
::$dataCache->getItem( $this->mCode
, 'datePreferences' );
758 function getDateFormats() {
759 return self
::$dataCache->getItem( $this->mCode
, 'dateFormats' );
763 * @return array|string
765 function getDefaultDateFormat() {
766 $df = self
::$dataCache->getItem( $this->mCode
, 'defaultDateFormat' );
767 if ( $df === 'dmy or mdy' ) {
768 global $wgAmericanDates;
769 return $wgAmericanDates ?
'mdy' : 'dmy';
778 function getDatePreferenceMigrationMap() {
779 return self
::$dataCache->getItem( $this->mCode
, 'datePreferenceMigrationMap' );
786 function getImageFile( $image ) {
787 return self
::$dataCache->getSubitem( $this->mCode
, 'imageFiles', $image );
793 function getExtraUserToggles() {
794 return (array)self
::$dataCache->getItem( $this->mCode
, 'extraUserToggles' );
801 function getUserToggle( $tog ) {
802 return $this->getMessageFromDB( "tog-$tog" );
806 * Get native language names, indexed by code.
807 * Only those defined in MediaWiki, no other data like CLDR.
808 * If $customisedOnly is true, only returns codes with a messages file
810 * @param $customisedOnly bool
813 * @deprecated in 1.20, use fetchLanguageNames()
815 public static function getLanguageNames( $customisedOnly = false ) {
816 return self
::fetchLanguageNames( null, $customisedOnly ?
'mwfile' : 'mw' );
820 * Get translated language names. This is done on best effort and
821 * by default this is exactly the same as Language::getLanguageNames.
822 * The CLDR extension provides translated names.
823 * @param $code String Language code.
824 * @return Array language code => language name
826 * @deprecated in 1.20, use fetchLanguageNames()
828 public static function getTranslatedLanguageNames( $code ) {
829 return self
::fetchLanguageNames( $code, 'all' );
833 * Get an array of language names, indexed by code.
834 * @param $inLanguage null|string: Code of language in which to return the names
835 * Use null for autonyms (native names)
836 * @param $include string:
837 * 'all' all available languages
838 * 'mw' only if the language is defined in MediaWiki or wgExtraLanguageNames (default)
839 * 'mwfile' only if the language is in 'mw' *and* has a message file
840 * @return array: language code => language name
843 public static function fetchLanguageNames( $inLanguage = null, $include = 'mw' ) {
844 global $wgExtraLanguageNames;
845 static $coreLanguageNames;
847 if ( $coreLanguageNames === null ) {
848 include( MWInit
::compiledPath( 'languages/Names.php' ) );
854 # TODO: also include when $inLanguage is null, when this code is more efficient
855 wfRunHooks( 'LanguageGetTranslatedLanguageNames', array( &$names, $inLanguage ) );
858 $mwNames = $wgExtraLanguageNames +
$coreLanguageNames;
859 foreach ( $mwNames as $mwCode => $mwName ) {
860 # - Prefer own MediaWiki native name when not using the hook
861 # - For other names just add if not added through the hook
862 if ( $mwCode === $inLanguage ||
!isset( $names[$mwCode] ) ) {
863 $names[$mwCode] = $mwName;
867 if ( $include === 'all' ) {
872 $coreCodes = array_keys( $mwNames );
873 foreach ( $coreCodes as $coreCode ) {
874 $returnMw[$coreCode] = $names[$coreCode];
877 if ( $include === 'mwfile' ) {
878 $namesMwFile = array();
879 # We do this using a foreach over the codes instead of a directory
880 # loop so that messages files in extensions will work correctly.
881 foreach ( $returnMw as $code => $value ) {
882 if ( is_readable( self
::getMessagesFileName( $code ) ) ) {
883 $namesMwFile[$code] = $names[$code];
888 # 'mw' option; default if it's not one of the other two options (all/mwfile)
893 * @param $code string: The code of the language for which to get the name
894 * @param $inLanguage null|string: Code of language in which to return the name (null for autonyms)
895 * @param $include string: 'all', 'mw' or 'mwfile'; see fetchLanguageNames()
896 * @return string: Language name or empty
899 public static function fetchLanguageName( $code, $inLanguage = null, $include = 'all' ) {
900 $array = self
::fetchLanguageNames( $inLanguage, $include );
901 return !array_key_exists( $code, $array ) ?
'' : $array[$code];
905 * Get a message from the MediaWiki namespace.
907 * @param $msg String: message name
910 function getMessageFromDB( $msg ) {
911 return wfMessage( $msg )->inLanguage( $this )->text();
915 * Get the native language name of $code.
916 * Only if defined in MediaWiki, no other data like CLDR.
917 * @param $code string
919 * @deprecated in 1.20, use fetchLanguageName()
921 function getLanguageName( $code ) {
922 return self
::fetchLanguageName( $code );
929 function getMonthName( $key ) {
930 return $this->getMessageFromDB( self
::$mMonthMsgs[$key - 1] );
936 function getMonthNamesArray() {
937 $monthNames = array( '' );
938 for ( $i = 1; $i < 13; $i++
) {
939 $monthNames[] = $this->getMonthName( $i );
948 function getMonthNameGen( $key ) {
949 return $this->getMessageFromDB( self
::$mMonthGenMsgs[$key - 1] );
956 function getMonthAbbreviation( $key ) {
957 return $this->getMessageFromDB( self
::$mMonthAbbrevMsgs[$key - 1] );
963 function getMonthAbbreviationsArray() {
964 $monthNames = array( '' );
965 for ( $i = 1; $i < 13; $i++
) {
966 $monthNames[] = $this->getMonthAbbreviation( $i );
975 function getWeekdayName( $key ) {
976 return $this->getMessageFromDB( self
::$mWeekdayMsgs[$key - 1] );
983 function getWeekdayAbbreviation( $key ) {
984 return $this->getMessageFromDB( self
::$mWeekdayAbbrevMsgs[$key - 1] );
991 function getIranianCalendarMonthName( $key ) {
992 return $this->getMessageFromDB( self
::$mIranianCalendarMonthMsgs[$key - 1] );
999 function getHebrewCalendarMonthName( $key ) {
1000 return $this->getMessageFromDB( self
::$mHebrewCalendarMonthMsgs[$key - 1] );
1004 * @param $key string
1007 function getHebrewCalendarMonthNameGen( $key ) {
1008 return $this->getMessageFromDB( self
::$mHebrewCalendarMonthGenMsgs[$key - 1] );
1012 * @param $key string
1015 function getHijriCalendarMonthName( $key ) {
1016 return $this->getMessageFromDB( self
::$mHijriCalendarMonthMsgs[$key - 1] );
1020 * This is a workalike of PHP's date() function, but with better
1021 * internationalisation, a reduced set of format characters, and a better
1024 * Supported format characters are dDjlNwzWFmMntLoYyaAgGhHiscrUeIOPTZ. See
1025 * the PHP manual for definitions. There are a number of extensions, which
1028 * xn Do not translate digits of the next numeric format character
1029 * xN Toggle raw digit (xn) flag, stays set until explicitly unset
1030 * xr Use roman numerals for the next numeric format character
1031 * xh Use hebrew numerals for the next numeric format character
1033 * xg Genitive month name
1035 * xij j (day number) in Iranian calendar
1036 * xiF F (month name) in Iranian calendar
1037 * xin n (month number) in Iranian calendar
1038 * xiy y (two digit year) in Iranian calendar
1039 * xiY Y (full year) in Iranian calendar
1041 * xjj j (day number) in Hebrew calendar
1042 * xjF F (month name) in Hebrew calendar
1043 * xjt t (days in month) in Hebrew calendar
1044 * xjx xg (genitive month name) in Hebrew calendar
1045 * xjn n (month number) in Hebrew calendar
1046 * xjY Y (full year) in Hebrew calendar
1048 * xmj j (day number) in Hijri calendar
1049 * xmF F (month name) in Hijri calendar
1050 * xmn n (month number) in Hijri calendar
1051 * xmY Y (full year) in Hijri calendar
1053 * xkY Y (full year) in Thai solar calendar. Months and days are
1054 * identical to the Gregorian calendar
1055 * xoY Y (full year) in Minguo calendar or Juche year.
1056 * Months and days are identical to the
1057 * Gregorian calendar
1058 * xtY Y (full year) in Japanese nengo. Months and days are
1059 * identical to the Gregorian calendar
1061 * Characters enclosed in double quotes will be considered literal (with
1062 * the quotes themselves removed). Unmatched quotes will be considered
1063 * literal quotes. Example:
1065 * "The month is" F => The month is January
1068 * Backslash escaping is also supported.
1070 * Input timestamp is assumed to be pre-normalized to the desired local
1071 * time zone, if any. Note that the format characters crUeIOPTZ will assume
1072 * $ts is UTC if $zone is not given.
1074 * @param $format String
1075 * @param $ts String: 14-character timestamp
1078 * @param $zone DateTimeZone: Timezone of $ts
1079 * @todo handling of "o" format character for Iranian, Hebrew, Hijri & Thai?
1083 function sprintfDate( $format, $ts, DateTimeZone
$zone = null ) {
1088 $dateTimeObj = false;
1096 for ( $p = 0; $p < strlen( $format ); $p++
) {
1098 $code = $format[$p];
1099 if ( $code == 'x' && $p < strlen( $format ) - 1 ) {
1100 $code .= $format[++
$p];
1103 if ( ( $code === 'xi' ||
$code == 'xj' ||
$code == 'xk' ||
$code == 'xm' ||
$code == 'xo' ||
$code == 'xt' ) && $p < strlen( $format ) - 1 ) {
1104 $code .= $format[++
$p];
1115 $rawToggle = !$rawToggle;
1124 $s .= $this->getMonthNameGen( substr( $ts, 4, 2 ) );
1127 if ( !$hebrew ) $hebrew = self
::tsToHebrew( $ts );
1128 $s .= $this->getHebrewCalendarMonthNameGen( $hebrew[1] );
1131 $num = substr( $ts, 6, 2 );
1134 if ( !$dateTimeObj ) {
1135 $dateTimeObj = DateTime
::createFromFormat(
1136 'YmdHis', $ts, $zone ?
: new DateTimeZone( 'UTC' )
1139 $s .= $this->getWeekdayAbbreviation( $dateTimeObj->format( 'w' ) +
1 );
1142 $num = intval( substr( $ts, 6, 2 ) );
1146 $iranian = self
::tsToIranian( $ts );
1152 $hijri = self
::tsToHijri( $ts );
1158 $hebrew = self
::tsToHebrew( $ts );
1163 if ( !$dateTimeObj ) {
1164 $dateTimeObj = DateTime
::createFromFormat(
1165 'YmdHis', $ts, $zone ?
: new DateTimeZone( 'UTC' )
1168 $s .= $this->getWeekdayName( $dateTimeObj->format( 'w' ) +
1 );
1171 $s .= $this->getMonthName( substr( $ts, 4, 2 ) );
1175 $iranian = self
::tsToIranian( $ts );
1177 $s .= $this->getIranianCalendarMonthName( $iranian[1] );
1181 $hijri = self
::tsToHijri( $ts );
1183 $s .= $this->getHijriCalendarMonthName( $hijri[1] );
1187 $hebrew = self
::tsToHebrew( $ts );
1189 $s .= $this->getHebrewCalendarMonthName( $hebrew[1] );
1192 $num = substr( $ts, 4, 2 );
1195 $s .= $this->getMonthAbbreviation( substr( $ts, 4, 2 ) );
1198 $num = intval( substr( $ts, 4, 2 ) );
1202 $iranian = self
::tsToIranian( $ts );
1208 $hijri = self
::tsToHijri ( $ts );
1214 $hebrew = self
::tsToHebrew( $ts );
1220 $hebrew = self
::tsToHebrew( $ts );
1225 $num = substr( $ts, 0, 4 );
1229 $iranian = self
::tsToIranian( $ts );
1235 $hijri = self
::tsToHijri( $ts );
1241 $hebrew = self
::tsToHebrew( $ts );
1247 $thai = self
::tsToYear( $ts, 'thai' );
1253 $minguo = self
::tsToYear( $ts, 'minguo' );
1259 $tenno = self
::tsToYear( $ts, 'tenno' );
1264 $num = substr( $ts, 2, 2 );
1268 $iranian = self
::tsToIranian( $ts );
1270 $num = substr( $iranian[0], -2 );
1273 $s .= intval( substr( $ts, 8, 2 ) ) < 12 ?
'am' : 'pm';
1276 $s .= intval( substr( $ts, 8, 2 ) ) < 12 ?
'AM' : 'PM';
1279 $h = substr( $ts, 8, 2 );
1280 $num = $h %
12 ?
$h %
12 : 12;
1283 $num = intval( substr( $ts, 8, 2 ) );
1286 $h = substr( $ts, 8, 2 );
1287 $num = sprintf( '%02d', $h %
12 ?
$h %
12 : 12 );
1290 $num = substr( $ts, 8, 2 );
1293 $num = substr( $ts, 10, 2 );
1296 $num = substr( $ts, 12, 2 );
1304 // Pass through string from $dateTimeObj->format()
1305 if ( !$dateTimeObj ) {
1306 $dateTimeObj = DateTime
::createFromFormat(
1307 'YmdHis', $ts, $zone ?
: new DateTimeZone( 'UTC' )
1310 $s .= $dateTimeObj->format( $code );
1322 // Pass through number from $dateTimeObj->format()
1323 if ( !$dateTimeObj ) {
1324 $dateTimeObj = DateTime
::createFromFormat(
1325 'YmdHis', $ts, $zone ?
: new DateTimeZone( 'UTC' )
1328 $num = $dateTimeObj->format( $code );
1331 # Backslash escaping
1332 if ( $p < strlen( $format ) - 1 ) {
1333 $s .= $format[++
$p];
1340 if ( $p < strlen( $format ) - 1 ) {
1341 $endQuote = strpos( $format, '"', $p +
1 );
1342 if ( $endQuote === false ) {
1343 # No terminating quote, assume literal "
1346 $s .= substr( $format, $p +
1, $endQuote - $p - 1 );
1350 # Quote at end of string, assume literal "
1357 if ( $num !== false ) {
1358 if ( $rawToggle ||
$raw ) {
1361 } elseif ( $roman ) {
1362 $s .= Language
::romanNumeral( $num );
1364 } elseif ( $hebrewNum ) {
1365 $s .= self
::hebrewNumeral( $num );
1368 $s .= $this->formatNum( $num, true );
1375 private static $GREG_DAYS = array( 31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31 );
1376 private static $IRANIAN_DAYS = array( 31, 31, 31, 31, 31, 31, 30, 30, 30, 30, 30, 29 );
1379 * Algorithm by Roozbeh Pournader and Mohammad Toossi to convert
1380 * Gregorian dates to Iranian dates. Originally written in C, it
1381 * is released under the terms of GNU Lesser General Public
1382 * License. Conversion to PHP was performed by Niklas Laxström.
1384 * Link: http://www.farsiweb.info/jalali/jalali.c
1390 private static function tsToIranian( $ts ) {
1391 $gy = substr( $ts, 0, 4 ) -1600;
1392 $gm = substr( $ts, 4, 2 ) -1;
1393 $gd = substr( $ts, 6, 2 ) -1;
1395 # Days passed from the beginning (including leap years)
1397 +
floor( ( $gy +
3 ) / 4 )
1398 - floor( ( $gy +
99 ) / 100 )
1399 +
floor( ( $gy +
399 ) / 400 );
1401 // Add days of the past months of this year
1402 for ( $i = 0; $i < $gm; $i++
) {
1403 $gDayNo +
= self
::$GREG_DAYS[$i];
1407 if ( $gm > 1 && ( ( $gy %
4 === 0 && $gy %
100 !== 0 ||
( $gy %
400 == 0 ) ) ) ) {
1411 // Days passed in current month
1412 $gDayNo +
= (int)$gd;
1414 $jDayNo = $gDayNo - 79;
1416 $jNp = floor( $jDayNo / 12053 );
1419 $jy = 979 +
33 * $jNp +
4 * floor( $jDayNo / 1461 );
1422 if ( $jDayNo >= 366 ) {
1423 $jy +
= floor( ( $jDayNo - 1 ) / 365 );
1424 $jDayNo = floor( ( $jDayNo - 1 ) %
365 );
1427 for ( $i = 0; $i < 11 && $jDayNo >= self
::$IRANIAN_DAYS[$i]; $i++
) {
1428 $jDayNo -= self
::$IRANIAN_DAYS[$i];
1434 return array( $jy, $jm, $jd );
1438 * Converting Gregorian dates to Hijri dates.
1440 * Based on a PHP-Nuke block by Sharjeel which is released under GNU/GPL license
1442 * @see http://phpnuke.org/modules.php?name=News&file=article&sid=8234&mode=thread&order=0&thold=0
1448 private static function tsToHijri( $ts ) {
1449 $year = substr( $ts, 0, 4 );
1450 $month = substr( $ts, 4, 2 );
1451 $day = substr( $ts, 6, 2 );
1459 ( $zy > 1582 ) ||
( ( $zy == 1582 ) && ( $zm > 10 ) ) ||
1460 ( ( $zy == 1582 ) && ( $zm == 10 ) && ( $zd > 14 ) )
1463 $zjd = (int)( ( 1461 * ( $zy +
4800 +
(int)( ( $zm - 14 ) / 12 ) ) ) / 4 ) +
1464 (int)( ( 367 * ( $zm - 2 - 12 * ( (int)( ( $zm - 14 ) / 12 ) ) ) ) / 12 ) -
1465 (int)( ( 3 * (int)( ( ( $zy +
4900 +
(int)( ( $zm - 14 ) / 12 ) ) / 100 ) ) ) / 4 ) +
1468 $zjd = 367 * $zy - (int)( ( 7 * ( $zy +
5001 +
(int)( ( $zm - 9 ) / 7 ) ) ) / 4 ) +
1469 (int)( ( 275 * $zm ) / 9 ) +
$zd +
1729777;
1472 $zl = $zjd -1948440 +
10632;
1473 $zn = (int)( ( $zl - 1 ) / 10631 );
1474 $zl = $zl - 10631 * $zn +
354;
1475 $zj = ( (int)( ( 10985 - $zl ) / 5316 ) ) * ( (int)( ( 50 * $zl ) / 17719 ) ) +
( (int)( $zl / 5670 ) ) * ( (int)( ( 43 * $zl ) / 15238 ) );
1476 $zl = $zl - ( (int)( ( 30 - $zj ) / 15 ) ) * ( (int)( ( 17719 * $zj ) / 50 ) ) - ( (int)( $zj / 16 ) ) * ( (int)( ( 15238 * $zj ) / 43 ) ) +
29;
1477 $zm = (int)( ( 24 * $zl ) / 709 );
1478 $zd = $zl - (int)( ( 709 * $zm ) / 24 );
1479 $zy = 30 * $zn +
$zj - 30;
1481 return array( $zy, $zm, $zd );
1485 * Converting Gregorian dates to Hebrew dates.
1487 * Based on a JavaScript code by Abu Mami and Yisrael Hersch
1488 * (abu-mami@kaluach.net, http://www.kaluach.net), who permitted
1489 * to translate the relevant functions into PHP and release them under
1492 * The months are counted from Tishrei = 1. In a leap year, Adar I is 13
1493 * and Adar II is 14. In a non-leap year, Adar is 6.
1499 private static function tsToHebrew( $ts ) {
1501 $year = substr( $ts, 0, 4 );
1502 $month = substr( $ts, 4, 2 );
1503 $day = substr( $ts, 6, 2 );
1505 # Calculate Hebrew year
1506 $hebrewYear = $year +
3760;
1508 # Month number when September = 1, August = 12
1510 if ( $month > 12 ) {
1517 # Calculate day of year from 1 September
1519 for ( $i = 1; $i < $month; $i++
) {
1523 # Check if the year is leap
1524 if ( $year %
400 == 0 ||
( $year %
4 == 0 && $year %
100 > 0 ) ) {
1527 } elseif ( $i == 8 ||
$i == 10 ||
$i == 1 ||
$i == 3 ) {
1534 # Calculate the start of the Hebrew year
1535 $start = self
::hebrewYearStart( $hebrewYear );
1537 # Calculate next year's start
1538 if ( $dayOfYear <= $start ) {
1539 # Day is before the start of the year - it is the previous year
1541 $nextStart = $start;
1545 # Add days since previous year's 1 September
1547 if ( ( $year %
400 == 0 ) ||
( $year %
100 != 0 && $year %
4 == 0 ) ) {
1551 # Start of the new (previous) year
1552 $start = self
::hebrewYearStart( $hebrewYear );
1555 $nextStart = self
::hebrewYearStart( $hebrewYear +
1 );
1558 # Calculate Hebrew day of year
1559 $hebrewDayOfYear = $dayOfYear - $start;
1561 # Difference between year's days
1562 $diff = $nextStart - $start;
1563 # Add 12 (or 13 for leap years) days to ignore the difference between
1564 # Hebrew and Gregorian year (353 at least vs. 365/6) - now the
1565 # difference is only about the year type
1566 if ( ( $year %
400 == 0 ) ||
( $year %
100 != 0 && $year %
4 == 0 ) ) {
1572 # Check the year pattern, and is leap year
1573 # 0 means an incomplete year, 1 means a regular year, 2 means a complete year
1574 # This is mod 30, to work on both leap years (which add 30 days of Adar I)
1575 # and non-leap years
1576 $yearPattern = $diff %
30;
1577 # Check if leap year
1578 $isLeap = $diff >= 30;
1580 # Calculate day in the month from number of day in the Hebrew year
1581 # Don't check Adar - if the day is not in Adar, we will stop before;
1582 # if it is in Adar, we will use it to check if it is Adar I or Adar II
1583 $hebrewDay = $hebrewDayOfYear;
1586 while ( $hebrewMonth <= 12 ) {
1587 # Calculate days in this month
1588 if ( $isLeap && $hebrewMonth == 6 ) {
1589 # Adar in a leap year
1591 # Leap year - has Adar I, with 30 days, and Adar II, with 29 days
1593 if ( $hebrewDay <= $days ) {
1597 # Subtract the days of Adar I
1598 $hebrewDay -= $days;
1601 if ( $hebrewDay <= $days ) {
1607 } elseif ( $hebrewMonth == 2 && $yearPattern == 2 ) {
1608 # Cheshvan in a complete year (otherwise as the rule below)
1610 } elseif ( $hebrewMonth == 3 && $yearPattern == 0 ) {
1611 # Kislev in an incomplete year (otherwise as the rule below)
1614 # Odd months have 30 days, even have 29
1615 $days = 30 - ( $hebrewMonth - 1 ) %
2;
1617 if ( $hebrewDay <= $days ) {
1618 # In the current month
1621 # Subtract the days of the current month
1622 $hebrewDay -= $days;
1623 # Try in the next month
1628 return array( $hebrewYear, $hebrewMonth, $hebrewDay, $days );
1632 * This calculates the Hebrew year start, as days since 1 September.
1633 * Based on Carl Friedrich Gauss algorithm for finding Easter date.
1634 * Used for Hebrew date.
1640 private static function hebrewYearStart( $year ) {
1641 $a = intval( ( 12 * ( $year - 1 ) +
17 ) %
19 );
1642 $b = intval( ( $year - 1 ) %
4 );
1643 $m = 32.044093161144 +
1.5542417966212 * $a +
$b / 4.0 - 0.0031777940220923 * ( $year - 1 );
1647 $Mar = intval( $m );
1653 $c = intval( ( $Mar +
3 * ( $year - 1 ) +
5 * $b +
5 ) %
7 );
1654 if ( $c == 0 && $a > 11 && $m >= 0.89772376543210 ) {
1656 } elseif ( $c == 1 && $a > 6 && $m >= 0.63287037037037 ) {
1658 } elseif ( $c == 2 ||
$c == 4 ||
$c == 6 ) {
1662 $Mar +
= intval( ( $year - 3761 ) / 100 ) - intval( ( $year - 3761 ) / 400 ) - 24;
1667 * Algorithm to convert Gregorian dates to Thai solar dates,
1668 * Minguo dates or Minguo dates.
1670 * Link: http://en.wikipedia.org/wiki/Thai_solar_calendar
1671 * http://en.wikipedia.org/wiki/Minguo_calendar
1672 * http://en.wikipedia.org/wiki/Japanese_era_name
1674 * @param $ts String: 14-character timestamp
1675 * @param $cName String: calender name
1676 * @return Array: converted year, month, day
1678 private static function tsToYear( $ts, $cName ) {
1679 $gy = substr( $ts, 0, 4 );
1680 $gm = substr( $ts, 4, 2 );
1681 $gd = substr( $ts, 6, 2 );
1683 if ( !strcmp( $cName, 'thai' ) ) {
1685 # Add 543 years to the Gregorian calendar
1686 # Months and days are identical
1687 $gy_offset = $gy +
543;
1688 } elseif ( ( !strcmp( $cName, 'minguo' ) ) ||
!strcmp( $cName, 'juche' ) ) {
1690 # Deduct 1911 years from the Gregorian calendar
1691 # Months and days are identical
1692 $gy_offset = $gy - 1911;
1693 } elseif ( !strcmp( $cName, 'tenno' ) ) {
1694 # Nengō dates up to Meiji period
1695 # Deduct years from the Gregorian calendar
1696 # depending on the nengo periods
1697 # Months and days are identical
1698 if ( ( $gy < 1912 ) ||
( ( $gy == 1912 ) && ( $gm < 7 ) ) ||
( ( $gy == 1912 ) && ( $gm == 7 ) && ( $gd < 31 ) ) ) {
1700 $gy_gannen = $gy - 1868 +
1;
1701 $gy_offset = $gy_gannen;
1702 if ( $gy_gannen == 1 ) {
1705 $gy_offset = '明治' . $gy_offset;
1707 ( ( $gy == 1912 ) && ( $gm == 7 ) && ( $gd == 31 ) ) ||
1708 ( ( $gy == 1912 ) && ( $gm >= 8 ) ) ||
1709 ( ( $gy > 1912 ) && ( $gy < 1926 ) ) ||
1710 ( ( $gy == 1926 ) && ( $gm < 12 ) ) ||
1711 ( ( $gy == 1926 ) && ( $gm == 12 ) && ( $gd < 26 ) )
1715 $gy_gannen = $gy - 1912 +
1;
1716 $gy_offset = $gy_gannen;
1717 if ( $gy_gannen == 1 ) {
1720 $gy_offset = '大正' . $gy_offset;
1722 ( ( $gy == 1926 ) && ( $gm == 12 ) && ( $gd >= 26 ) ) ||
1723 ( ( $gy > 1926 ) && ( $gy < 1989 ) ) ||
1724 ( ( $gy == 1989 ) && ( $gm == 1 ) && ( $gd < 8 ) )
1728 $gy_gannen = $gy - 1926 +
1;
1729 $gy_offset = $gy_gannen;
1730 if ( $gy_gannen == 1 ) {
1733 $gy_offset = '昭和' . $gy_offset;
1736 $gy_gannen = $gy - 1989 +
1;
1737 $gy_offset = $gy_gannen;
1738 if ( $gy_gannen == 1 ) {
1741 $gy_offset = '平成' . $gy_offset;
1747 return array( $gy_offset, $gm, $gd );
1751 * Roman number formatting up to 10000
1757 static function romanNumeral( $num ) {
1758 static $table = array(
1759 array( '', 'I', 'II', 'III', 'IV', 'V', 'VI', 'VII', 'VIII', 'IX', 'X' ),
1760 array( '', 'X', 'XX', 'XXX', 'XL', 'L', 'LX', 'LXX', 'LXXX', 'XC', 'C' ),
1761 array( '', 'C', 'CC', 'CCC', 'CD', 'D', 'DC', 'DCC', 'DCCC', 'CM', 'M' ),
1762 array( '', 'M', 'MM', 'MMM', 'MMMM', 'MMMMM', 'MMMMMM', 'MMMMMMM', 'MMMMMMMM', 'MMMMMMMMM', 'MMMMMMMMMM' )
1765 $num = intval( $num );
1766 if ( $num > 10000 ||
$num <= 0 ) {
1771 for ( $pow10 = 1000, $i = 3; $i >= 0; $pow10 /= 10, $i-- ) {
1772 if ( $num >= $pow10 ) {
1773 $s .= $table[$i][(int)floor( $num / $pow10 )];
1775 $num = $num %
$pow10;
1781 * Hebrew Gematria number formatting up to 9999
1787 static function hebrewNumeral( $num ) {
1788 static $table = array(
1789 array( '', 'א', 'ב', 'ג', 'ד', 'ה', 'ו', 'ז', 'ח', 'ט', 'י' ),
1790 array( '', 'י', 'כ', 'ל', 'מ', 'נ', 'ס', 'ע', 'פ', 'צ', 'ק' ),
1791 array( '', 'ק', 'ר', 'ש', 'ת', 'תק', 'תר', 'תש', 'תת', 'תתק', 'תתר' ),
1792 array( '', 'א', 'ב', 'ג', 'ד', 'ה', 'ו', 'ז', 'ח', 'ט', 'י' )
1795 $num = intval( $num );
1796 if ( $num > 9999 ||
$num <= 0 ) {
1801 for ( $pow10 = 1000, $i = 3; $i >= 0; $pow10 /= 10, $i-- ) {
1802 if ( $num >= $pow10 ) {
1803 if ( $num == 15 ||
$num == 16 ) {
1804 $s .= $table[0][9] . $table[0][$num - 9];
1807 $s .= $table[$i][intval( ( $num / $pow10 ) )];
1808 if ( $pow10 == 1000 ) {
1813 $num = $num %
$pow10;
1815 if ( strlen( $s ) == 2 ) {
1818 $str = substr( $s, 0, strlen( $s ) - 2 ) . '"';
1819 $str .= substr( $s, strlen( $s ) - 2, 2 );
1821 $start = substr( $str, 0, strlen( $str ) - 2 );
1822 $end = substr( $str, strlen( $str ) - 2 );
1825 $str = $start . 'ך';
1828 $str = $start . 'ם';
1831 $str = $start . 'ן';
1834 $str = $start . 'ף';
1837 $str = $start . 'ץ';
1844 * Used by date() and time() to adjust the time output.
1846 * @param $ts Int the time in date('YmdHis') format
1847 * @param $tz Mixed: adjust the time by this amount (default false, mean we
1848 * get user timecorrection setting)
1851 function userAdjust( $ts, $tz = false ) {
1852 global $wgUser, $wgLocalTZoffset;
1854 if ( $tz === false ) {
1855 $tz = $wgUser->getOption( 'timecorrection' );
1858 $data = explode( '|', $tz, 3 );
1860 if ( $data[0] == 'ZoneInfo' ) {
1861 wfSuppressWarnings();
1862 $userTZ = timezone_open( $data[2] );
1863 wfRestoreWarnings();
1864 if ( $userTZ !== false ) {
1865 $date = date_create( $ts, timezone_open( 'UTC' ) );
1866 date_timezone_set( $date, $userTZ );
1867 $date = date_format( $date, 'YmdHis' );
1870 # Unrecognized timezone, default to 'Offset' with the stored offset.
1871 $data[0] = 'Offset';
1875 if ( $data[0] == 'System' ||
$tz == '' ) {
1876 # Global offset in minutes.
1877 if ( isset( $wgLocalTZoffset ) ) {
1878 $minDiff = $wgLocalTZoffset;
1880 } elseif ( $data[0] == 'Offset' ) {
1881 $minDiff = intval( $data[1] );
1883 $data = explode( ':', $tz );
1884 if ( count( $data ) == 2 ) {
1885 $data[0] = intval( $data[0] );
1886 $data[1] = intval( $data[1] );
1887 $minDiff = abs( $data[0] ) * 60 +
$data[1];
1888 if ( $data[0] < 0 ) {
1889 $minDiff = -$minDiff;
1892 $minDiff = intval( $data[0] ) * 60;
1896 # No difference ? Return time unchanged
1897 if ( 0 == $minDiff ) {
1901 wfSuppressWarnings(); // E_STRICT system time bitching
1902 # Generate an adjusted date; take advantage of the fact that mktime
1903 # will normalize out-of-range values so we don't have to split $minDiff
1904 # into hours and minutes.
1906 (int)substr( $ts, 8, 2 ) ), # Hours
1907 (int)substr( $ts, 10, 2 ) +
$minDiff, # Minutes
1908 (int)substr( $ts, 12, 2 ), # Seconds
1909 (int)substr( $ts, 4, 2 ), # Month
1910 (int)substr( $ts, 6, 2 ), # Day
1911 (int)substr( $ts, 0, 4 ) ); # Year
1913 $date = date( 'YmdHis', $t );
1914 wfRestoreWarnings();
1920 * This is meant to be used by time(), date(), and timeanddate() to get
1921 * the date preference they're supposed to use, it should be used in
1925 * function timeanddate([...], $format = true) {
1926 * $datePreference = $this->dateFormat($format);
1931 * @param $usePrefs Mixed: if true, the user's preference is used
1932 * if false, the site/language default is used
1933 * if int/string, assumed to be a format.
1936 function dateFormat( $usePrefs = true ) {
1939 if ( is_bool( $usePrefs ) ) {
1941 $datePreference = $wgUser->getDatePreference();
1943 $datePreference = (string)User
::getDefaultOption( 'date' );
1946 $datePreference = (string)$usePrefs;
1950 if ( $datePreference == '' ) {
1954 return $datePreference;
1958 * Get a format string for a given type and preference
1959 * @param $type string May be date, time or both
1960 * @param $pref string The format name as it appears in Messages*.php
1964 function getDateFormatString( $type, $pref ) {
1965 if ( !isset( $this->dateFormatStrings
[$type][$pref] ) ) {
1966 if ( $pref == 'default' ) {
1967 $pref = $this->getDefaultDateFormat();
1968 $df = self
::$dataCache->getSubitem( $this->mCode
, 'dateFormats', "$pref $type" );
1970 $df = self
::$dataCache->getSubitem( $this->mCode
, 'dateFormats', "$pref $type" );
1971 if ( is_null( $df ) ) {
1972 $pref = $this->getDefaultDateFormat();
1973 $df = self
::$dataCache->getSubitem( $this->mCode
, 'dateFormats', "$pref $type" );
1976 $this->dateFormatStrings
[$type][$pref] = $df;
1978 return $this->dateFormatStrings
[$type][$pref];
1982 * @param $ts Mixed: the time format which needs to be turned into a
1983 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
1984 * @param $adj Bool: whether to adjust the time output according to the
1985 * user configured offset ($timecorrection)
1986 * @param $format Mixed: true to use user's date format preference
1987 * @param $timecorrection String|bool the time offset as returned by
1988 * validateTimeZone() in Special:Preferences
1991 function date( $ts, $adj = false, $format = true, $timecorrection = false ) {
1992 $ts = wfTimestamp( TS_MW
, $ts );
1994 $ts = $this->userAdjust( $ts, $timecorrection );
1996 $df = $this->getDateFormatString( 'date', $this->dateFormat( $format ) );
1997 return $this->sprintfDate( $df, $ts );
2001 * @param $ts Mixed: the time format which needs to be turned into a
2002 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2003 * @param $adj Bool: whether to adjust the time output according to the
2004 * user configured offset ($timecorrection)
2005 * @param $format Mixed: true to use user's date format preference
2006 * @param $timecorrection String|bool the time offset as returned by
2007 * validateTimeZone() in Special:Preferences
2010 function time( $ts, $adj = false, $format = true, $timecorrection = false ) {
2011 $ts = wfTimestamp( TS_MW
, $ts );
2013 $ts = $this->userAdjust( $ts, $timecorrection );
2015 $df = $this->getDateFormatString( 'time', $this->dateFormat( $format ) );
2016 return $this->sprintfDate( $df, $ts );
2020 * @param $ts Mixed: the time format which needs to be turned into a
2021 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2022 * @param $adj Bool: whether to adjust the time output according to the
2023 * user configured offset ($timecorrection)
2024 * @param $format Mixed: what format to return, if it's false output the
2025 * default one (default true)
2026 * @param $timecorrection String|bool the time offset as returned by
2027 * validateTimeZone() in Special:Preferences
2030 function timeanddate( $ts, $adj = false, $format = true, $timecorrection = false ) {
2031 $ts = wfTimestamp( TS_MW
, $ts );
2033 $ts = $this->userAdjust( $ts, $timecorrection );
2035 $df = $this->getDateFormatString( 'both', $this->dateFormat( $format ) );
2036 return $this->sprintfDate( $df, $ts );
2040 * Takes a number of seconds and turns it into a text using values such as hours and minutes.
2044 * @param integer $seconds The amount of seconds.
2045 * @param array $chosenIntervals The intervals to enable.
2049 public function formatDuration( $seconds, array $chosenIntervals = array() ) {
2050 $intervals = $this->getDurationIntervals( $seconds, $chosenIntervals );
2052 $segments = array();
2054 foreach ( $intervals as $intervalName => $intervalValue ) {
2055 $message = wfMessage( 'duration-' . $intervalName )->numParams( $intervalValue );
2056 $segments[] = $message->inLanguage( $this )->escaped();
2059 return $this->listToText( $segments );
2063 * Takes a number of seconds and returns an array with a set of corresponding intervals.
2064 * For example 65 will be turned into array( minutes => 1, seconds => 5 ).
2068 * @param integer $seconds The amount of seconds.
2069 * @param array $chosenIntervals The intervals to enable.
2073 public function getDurationIntervals( $seconds, array $chosenIntervals = array() ) {
2074 if ( empty( $chosenIntervals ) ) {
2075 $chosenIntervals = array( 'millennia', 'centuries', 'decades', 'years', 'days', 'hours', 'minutes', 'seconds' );
2078 $intervals = array_intersect_key( self
::$durationIntervals, array_flip( $chosenIntervals ) );
2079 $sortedNames = array_keys( $intervals );
2080 $smallestInterval = array_pop( $sortedNames );
2082 $segments = array();
2084 foreach ( $intervals as $name => $length ) {
2085 $value = floor( $seconds / $length );
2087 if ( $value > 0 ||
( $name == $smallestInterval && empty( $segments ) ) ) {
2088 $seconds -= $value * $length;
2089 $segments[$name] = $value;
2097 * Internal helper function for userDate(), userTime() and userTimeAndDate()
2099 * @param $type String: can be 'date', 'time' or 'both'
2100 * @param $ts Mixed: the time format which needs to be turned into a
2101 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2102 * @param $user User object used to get preferences for timezone and format
2103 * @param $options Array, can contain the following keys:
2104 * - 'timecorrection': time correction, can have the following values:
2105 * - true: use user's preference
2106 * - false: don't use time correction
2107 * - integer: value of time correction in minutes
2108 * - 'format': format to use, can have the following values:
2109 * - true: use user's preference
2110 * - false: use default preference
2111 * - string: format to use
2115 private function internalUserTimeAndDate( $type, $ts, User
$user, array $options ) {
2116 $ts = wfTimestamp( TS_MW
, $ts );
2117 $options +
= array( 'timecorrection' => true, 'format' => true );
2118 if ( $options['timecorrection'] !== false ) {
2119 if ( $options['timecorrection'] === true ) {
2120 $offset = $user->getOption( 'timecorrection' );
2122 $offset = $options['timecorrection'];
2124 $ts = $this->userAdjust( $ts, $offset );
2126 if ( $options['format'] === true ) {
2127 $format = $user->getDatePreference();
2129 $format = $options['format'];
2131 $df = $this->getDateFormatString( $type, $this->dateFormat( $format ) );
2132 return $this->sprintfDate( $df, $ts );
2136 * Get the formatted date for the given timestamp and formatted for
2139 * @param $ts Mixed: the time format which needs to be turned into a
2140 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2141 * @param $user User object used to get preferences for timezone and format
2142 * @param $options Array, can contain the following keys:
2143 * - 'timecorrection': time correction, can have the following values:
2144 * - true: use user's preference
2145 * - false: don't use time correction
2146 * - integer: value of time correction in minutes
2147 * - 'format': format to use, can have the following values:
2148 * - true: use user's preference
2149 * - false: use default preference
2150 * - string: format to use
2154 public function userDate( $ts, User
$user, array $options = array() ) {
2155 return $this->internalUserTimeAndDate( 'date', $ts, $user, $options );
2159 * Get the formatted time for the given timestamp and formatted for
2162 * @param $ts Mixed: the time format which needs to be turned into a
2163 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2164 * @param $user User object used to get preferences for timezone and format
2165 * @param $options Array, can contain the following keys:
2166 * - 'timecorrection': time correction, can have the following values:
2167 * - true: use user's preference
2168 * - false: don't use time correction
2169 * - integer: value of time correction in minutes
2170 * - 'format': format to use, can have the following values:
2171 * - true: use user's preference
2172 * - false: use default preference
2173 * - string: format to use
2177 public function userTime( $ts, User
$user, array $options = array() ) {
2178 return $this->internalUserTimeAndDate( 'time', $ts, $user, $options );
2182 * Get the formatted date and time for the given timestamp and formatted for
2185 * @param $ts Mixed: the time format which needs to be turned into a
2186 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2187 * @param $user User object used to get preferences for timezone and format
2188 * @param $options Array, can contain the following keys:
2189 * - 'timecorrection': time correction, can have the following values:
2190 * - true: use user's preference
2191 * - false: don't use time correction
2192 * - integer: value of time correction in minutes
2193 * - 'format': format to use, can have the following values:
2194 * - true: use user's preference
2195 * - false: use default preference
2196 * - string: format to use
2200 public function userTimeAndDate( $ts, User
$user, array $options = array() ) {
2201 return $this->internalUserTimeAndDate( 'both', $ts, $user, $options );
2205 * @param $key string
2206 * @return array|null
2208 function getMessage( $key ) {
2209 return self
::$dataCache->getSubitem( $this->mCode
, 'messages', $key );
2215 function getAllMessages() {
2216 return self
::$dataCache->getItem( $this->mCode
, 'messages' );
2225 function iconv( $in, $out, $string ) {
2226 # This is a wrapper for iconv in all languages except esperanto,
2227 # which does some nasty x-conversions beforehand
2229 # Even with //IGNORE iconv can whine about illegal characters in
2230 # *input* string. We just ignore those too.
2231 # REF: http://bugs.php.net/bug.php?id=37166
2232 # REF: https://bugzilla.wikimedia.org/show_bug.cgi?id=16885
2233 wfSuppressWarnings();
2234 $text = iconv( $in, $out . '//IGNORE', $string );
2235 wfRestoreWarnings();
2239 // callback functions for uc(), lc(), ucwords(), ucwordbreaks()
2242 * @param $matches array
2243 * @return mixed|string
2245 function ucwordbreaksCallbackAscii( $matches ) {
2246 return $this->ucfirst( $matches[1] );
2250 * @param $matches array
2253 function ucwordbreaksCallbackMB( $matches ) {
2254 return mb_strtoupper( $matches[0] );
2258 * @param $matches array
2261 function ucCallback( $matches ) {
2262 list( $wikiUpperChars ) = self
::getCaseMaps();
2263 return strtr( $matches[1], $wikiUpperChars );
2267 * @param $matches array
2270 function lcCallback( $matches ) {
2271 list( , $wikiLowerChars ) = self
::getCaseMaps();
2272 return strtr( $matches[1], $wikiLowerChars );
2276 * @param $matches array
2279 function ucwordsCallbackMB( $matches ) {
2280 return mb_strtoupper( $matches[0] );
2284 * @param $matches array
2287 function ucwordsCallbackWiki( $matches ) {
2288 list( $wikiUpperChars ) = self
::getCaseMaps();
2289 return strtr( $matches[0], $wikiUpperChars );
2293 * Make a string's first character uppercase
2295 * @param $str string
2299 function ucfirst( $str ) {
2301 if ( $o < 96 ) { // if already uppercase...
2303 } elseif ( $o < 128 ) {
2304 return ucfirst( $str ); // use PHP's ucfirst()
2306 // fall back to more complex logic in case of multibyte strings
2307 return $this->uc( $str, true );
2312 * Convert a string to uppercase
2314 * @param $str string
2315 * @param $first bool
2319 function uc( $str, $first = false ) {
2320 if ( function_exists( 'mb_strtoupper' ) ) {
2322 if ( $this->isMultibyte( $str ) ) {
2323 return mb_strtoupper( mb_substr( $str, 0, 1 ) ) . mb_substr( $str, 1 );
2325 return ucfirst( $str );
2328 return $this->isMultibyte( $str ) ?
mb_strtoupper( $str ) : strtoupper( $str );
2331 if ( $this->isMultibyte( $str ) ) {
2332 $x = $first ?
'^' : '';
2333 return preg_replace_callback(
2334 "/$x([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/",
2335 array( $this, 'ucCallback' ),
2339 return $first ?
ucfirst( $str ) : strtoupper( $str );
2345 * @param $str string
2346 * @return mixed|string
2348 function lcfirst( $str ) {
2351 return strval( $str );
2352 } elseif ( $o >= 128 ) {
2353 return $this->lc( $str, true );
2354 } elseif ( $o > 96 ) {
2357 $str[0] = strtolower( $str[0] );
2363 * @param $str string
2364 * @param $first bool
2365 * @return mixed|string
2367 function lc( $str, $first = false ) {
2368 if ( function_exists( 'mb_strtolower' ) ) {
2370 if ( $this->isMultibyte( $str ) ) {
2371 return mb_strtolower( mb_substr( $str, 0, 1 ) ) . mb_substr( $str, 1 );
2373 return strtolower( substr( $str, 0, 1 ) ) . substr( $str, 1 );
2376 return $this->isMultibyte( $str ) ?
mb_strtolower( $str ) : strtolower( $str );
2379 if ( $this->isMultibyte( $str ) ) {
2380 $x = $first ?
'^' : '';
2381 return preg_replace_callback(
2382 "/$x([A-Z]|[\\xc0-\\xff][\\x80-\\xbf]*)/",
2383 array( $this, 'lcCallback' ),
2387 return $first ?
strtolower( substr( $str, 0, 1 ) ) . substr( $str, 1 ) : strtolower( $str );
2393 * @param $str string
2396 function isMultibyte( $str ) {
2397 return (bool)preg_match( '/[\x80-\xff]/', $str );
2401 * @param $str string
2402 * @return mixed|string
2404 function ucwords( $str ) {
2405 if ( $this->isMultibyte( $str ) ) {
2406 $str = $this->lc( $str );
2408 // regexp to find first letter in each word (i.e. after each space)
2409 $replaceRegexp = "/^([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)| ([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/";
2411 // function to use to capitalize a single char
2412 if ( function_exists( 'mb_strtoupper' ) ) {
2413 return preg_replace_callback(
2415 array( $this, 'ucwordsCallbackMB' ),
2419 return preg_replace_callback(
2421 array( $this, 'ucwordsCallbackWiki' ),
2426 return ucwords( strtolower( $str ) );
2431 * capitalize words at word breaks
2433 * @param $str string
2436 function ucwordbreaks( $str ) {
2437 if ( $this->isMultibyte( $str ) ) {
2438 $str = $this->lc( $str );
2440 // since \b doesn't work for UTF-8, we explicitely define word break chars
2441 $breaks = "[ \-\(\)\}\{\.,\?!]";
2443 // find first letter after word break
2444 $replaceRegexp = "/^([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)|$breaks([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/";
2446 if ( function_exists( 'mb_strtoupper' ) ) {
2447 return preg_replace_callback(
2449 array( $this, 'ucwordbreaksCallbackMB' ),
2453 return preg_replace_callback(
2455 array( $this, 'ucwordsCallbackWiki' ),
2460 return preg_replace_callback(
2461 '/\b([\w\x80-\xff]+)\b/',
2462 array( $this, 'ucwordbreaksCallbackAscii' ),
2469 * Return a case-folded representation of $s
2471 * This is a representation such that caseFold($s1)==caseFold($s2) if $s1
2472 * and $s2 are the same except for the case of their characters. It is not
2473 * necessary for the value returned to make sense when displayed.
2475 * Do *not* perform any other normalisation in this function. If a caller
2476 * uses this function when it should be using a more general normalisation
2477 * function, then fix the caller.
2483 function caseFold( $s ) {
2484 return $this->uc( $s );
2491 function checkTitleEncoding( $s ) {
2492 if ( is_array( $s ) ) {
2493 wfDebugDieBacktrace( 'Given array to checkTitleEncoding.' );
2495 if ( StringUtils
::isUtf8( $s ) ) {
2499 return $this->iconv( $this->fallback8bitEncoding(), 'utf-8', $s );
2505 function fallback8bitEncoding() {
2506 return self
::$dataCache->getItem( $this->mCode
, 'fallback8bitEncoding' );
2510 * Most writing systems use whitespace to break up words.
2511 * Some languages such as Chinese don't conventionally do this,
2512 * which requires special handling when breaking up words for
2517 function hasWordBreaks() {
2522 * Some languages such as Chinese require word segmentation,
2523 * Specify such segmentation when overridden in derived class.
2525 * @param $string String
2528 function segmentByWord( $string ) {
2533 * Some languages have special punctuation need to be normalized.
2534 * Make such changes here.
2536 * @param $string String
2539 function normalizeForSearch( $string ) {
2540 return self
::convertDoubleWidth( $string );
2544 * convert double-width roman characters to single-width.
2545 * range: ff00-ff5f ~= 0020-007f
2547 * @param $string string
2551 protected static function convertDoubleWidth( $string ) {
2552 static $full = null;
2553 static $half = null;
2555 if ( $full === null ) {
2556 $fullWidth = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz";
2557 $halfWidth = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz";
2558 $full = str_split( $fullWidth, 3 );
2559 $half = str_split( $halfWidth );
2562 $string = str_replace( $full, $half, $string );
2567 * @param $string string
2568 * @param $pattern string
2571 protected static function insertSpace( $string, $pattern ) {
2572 $string = preg_replace( $pattern, " $1 ", $string );
2573 $string = preg_replace( '/ +/', ' ', $string );
2578 * @param $termsArray array
2581 function convertForSearchResult( $termsArray ) {
2582 # some languages, e.g. Chinese, need to do a conversion
2583 # in order for search results to be displayed correctly
2588 * Get the first character of a string.
2593 function firstChar( $s ) {
2596 '/^([\x00-\x7f]|[\xc0-\xdf][\x80-\xbf]|' .
2597 '[\xe0-\xef][\x80-\xbf]{2}|[\xf0-\xf7][\x80-\xbf]{3})/',
2602 if ( isset( $matches[1] ) ) {
2603 if ( strlen( $matches[1] ) != 3 ) {
2607 // Break down Hangul syllables to grab the first jamo
2608 $code = utf8ToCodepoint( $matches[1] );
2609 if ( $code < 0xac00 ||
0xd7a4 <= $code ) {
2611 } elseif ( $code < 0xb098 ) {
2612 return "\xe3\x84\xb1";
2613 } elseif ( $code < 0xb2e4 ) {
2614 return "\xe3\x84\xb4";
2615 } elseif ( $code < 0xb77c ) {
2616 return "\xe3\x84\xb7";
2617 } elseif ( $code < 0xb9c8 ) {
2618 return "\xe3\x84\xb9";
2619 } elseif ( $code < 0xbc14 ) {
2620 return "\xe3\x85\x81";
2621 } elseif ( $code < 0xc0ac ) {
2622 return "\xe3\x85\x82";
2623 } elseif ( $code < 0xc544 ) {
2624 return "\xe3\x85\x85";
2625 } elseif ( $code < 0xc790 ) {
2626 return "\xe3\x85\x87";
2627 } elseif ( $code < 0xcc28 ) {
2628 return "\xe3\x85\x88";
2629 } elseif ( $code < 0xce74 ) {
2630 return "\xe3\x85\x8a";
2631 } elseif ( $code < 0xd0c0 ) {
2632 return "\xe3\x85\x8b";
2633 } elseif ( $code < 0xd30c ) {
2634 return "\xe3\x85\x8c";
2635 } elseif ( $code < 0xd558 ) {
2636 return "\xe3\x85\x8d";
2638 return "\xe3\x85\x8e";
2645 function initEncoding() {
2646 # Some languages may have an alternate char encoding option
2647 # (Esperanto X-coding, Japanese furigana conversion, etc)
2648 # If this language is used as the primary content language,
2649 # an override to the defaults can be set here on startup.
2656 function recodeForEdit( $s ) {
2657 # For some languages we'll want to explicitly specify
2658 # which characters make it into the edit box raw
2659 # or are converted in some way or another.
2660 global $wgEditEncoding;
2661 if ( $wgEditEncoding == '' ||
$wgEditEncoding == 'UTF-8' ) {
2664 return $this->iconv( 'UTF-8', $wgEditEncoding, $s );
2672 function recodeInput( $s ) {
2673 # Take the previous into account.
2674 global $wgEditEncoding;
2675 if ( $wgEditEncoding != '' ) {
2676 $enc = $wgEditEncoding;
2680 if ( $enc == 'UTF-8' ) {
2683 return $this->iconv( $enc, 'UTF-8', $s );
2688 * Convert a UTF-8 string to normal form C. In Malayalam and Arabic, this
2689 * also cleans up certain backwards-compatible sequences, converting them
2690 * to the modern Unicode equivalent.
2692 * This is language-specific for performance reasons only.
2698 function normalize( $s ) {
2699 global $wgAllUnicodeFixes;
2700 $s = UtfNormal
::cleanUp( $s );
2701 if ( $wgAllUnicodeFixes ) {
2702 $s = $this->transformUsingPairFile( 'normalize-ar.ser', $s );
2703 $s = $this->transformUsingPairFile( 'normalize-ml.ser', $s );
2710 * Transform a string using serialized data stored in the given file (which
2711 * must be in the serialized subdirectory of $IP). The file contains pairs
2712 * mapping source characters to destination characters.
2714 * The data is cached in process memory. This will go faster if you have the
2715 * FastStringSearch extension.
2717 * @param $file string
2718 * @param $string string
2720 * @throws MWException
2723 function transformUsingPairFile( $file, $string ) {
2724 if ( !isset( $this->transformData
[$file] ) ) {
2725 $data = wfGetPrecompiledData( $file );
2726 if ( $data === false ) {
2727 throw new MWException( __METHOD__
. ": The transformation file $file is missing" );
2729 $this->transformData
[$file] = new ReplacementArray( $data );
2731 return $this->transformData
[$file]->replace( $string );
2735 * For right-to-left language support
2740 return self
::$dataCache->getItem( $this->mCode
, 'rtl' );
2744 * Return the correct HTML 'dir' attribute value for this language.
2748 return $this->isRTL() ?
'rtl' : 'ltr';
2752 * Return 'left' or 'right' as appropriate alignment for line-start
2753 * for this language's text direction.
2755 * Should be equivalent to CSS3 'start' text-align value....
2759 function alignStart() {
2760 return $this->isRTL() ?
'right' : 'left';
2764 * Return 'right' or 'left' as appropriate alignment for line-end
2765 * for this language's text direction.
2767 * Should be equivalent to CSS3 'end' text-align value....
2771 function alignEnd() {
2772 return $this->isRTL() ?
'left' : 'right';
2776 * A hidden direction mark (LRM or RLM), depending on the language direction.
2777 * Unlike getDirMark(), this function returns the character as an HTML entity.
2778 * This function should be used when the output is guaranteed to be HTML,
2779 * because it makes the output HTML source code more readable. When
2780 * the output is plain text or can be escaped, getDirMark() should be used.
2782 * @param $opposite Boolean Get the direction mark opposite to your language
2786 function getDirMarkEntity( $opposite = false ) {
2787 if ( $opposite ) { return $this->isRTL() ?
'‎' : '‏'; }
2788 return $this->isRTL() ?
'‏' : '‎';
2792 * A hidden direction mark (LRM or RLM), depending on the language direction.
2793 * This function produces them as invisible Unicode characters and
2794 * the output may be hard to read and debug, so it should only be used
2795 * when the output is plain text or can be escaped. When the output is
2796 * HTML, use getDirMarkEntity() instead.
2798 * @param $opposite Boolean Get the direction mark opposite to your language
2801 function getDirMark( $opposite = false ) {
2802 $lrm = "\xE2\x80\x8E"; # LEFT-TO-RIGHT MARK, commonly abbreviated LRM
2803 $rlm = "\xE2\x80\x8F"; # RIGHT-TO-LEFT MARK, commonly abbreviated RLM
2804 if ( $opposite ) { return $this->isRTL() ?
$lrm : $rlm; }
2805 return $this->isRTL() ?
$rlm : $lrm;
2811 function capitalizeAllNouns() {
2812 return self
::$dataCache->getItem( $this->mCode
, 'capitalizeAllNouns' );
2816 * An arrow, depending on the language direction.
2818 * @param $direction String: the direction of the arrow: forwards (default), backwards, left, right, up, down.
2821 function getArrow( $direction = 'forwards' ) {
2822 switch ( $direction ) {
2824 return $this->isRTL() ?
'←' : '→';
2826 return $this->isRTL() ?
'→' : '←';
2839 * To allow "foo[[bar]]" to extend the link over the whole word "foobar"
2843 function linkPrefixExtension() {
2844 return self
::$dataCache->getItem( $this->mCode
, 'linkPrefixExtension' );
2850 function getMagicWords() {
2851 return self
::$dataCache->getItem( $this->mCode
, 'magicWords' );
2854 protected function doMagicHook() {
2855 if ( $this->mMagicHookDone
) {
2858 $this->mMagicHookDone
= true;
2859 wfProfileIn( 'LanguageGetMagic' );
2860 wfRunHooks( 'LanguageGetMagic', array( &$this->mMagicExtensions
, $this->getCode() ) );
2861 wfProfileOut( 'LanguageGetMagic' );
2865 * Fill a MagicWord object with data from here
2869 function getMagic( $mw ) {
2870 $this->doMagicHook();
2872 if ( isset( $this->mMagicExtensions
[$mw->mId
] ) ) {
2873 $rawEntry = $this->mMagicExtensions
[$mw->mId
];
2875 $magicWords = $this->getMagicWords();
2876 if ( isset( $magicWords[$mw->mId
] ) ) {
2877 $rawEntry = $magicWords[$mw->mId
];
2883 if ( !is_array( $rawEntry ) ) {
2884 error_log( "\"$rawEntry\" is not a valid magic word for \"$mw->mId\"" );
2886 $mw->mCaseSensitive
= $rawEntry[0];
2887 $mw->mSynonyms
= array_slice( $rawEntry, 1 );
2892 * Add magic words to the extension array
2894 * @param $newWords array
2896 function addMagicWordsByLang( $newWords ) {
2897 $fallbackChain = $this->getFallbackLanguages();
2898 $fallbackChain = array_reverse( $fallbackChain );
2899 foreach ( $fallbackChain as $code ) {
2900 if ( isset( $newWords[$code] ) ) {
2901 $this->mMagicExtensions
= $newWords[$code] +
$this->mMagicExtensions
;
2907 * Get special page names, as an associative array
2908 * case folded alias => real name
2910 function getSpecialPageAliases() {
2911 // Cache aliases because it may be slow to load them
2912 if ( is_null( $this->mExtendedSpecialPageAliases
) ) {
2914 $this->mExtendedSpecialPageAliases
=
2915 self
::$dataCache->getItem( $this->mCode
, 'specialPageAliases' );
2916 wfRunHooks( 'LanguageGetSpecialPageAliases',
2917 array( &$this->mExtendedSpecialPageAliases
, $this->getCode() ) );
2920 return $this->mExtendedSpecialPageAliases
;
2924 * Italic is unsuitable for some languages
2926 * @param $text String: the text to be emphasized.
2929 function emphasize( $text ) {
2930 return "<em>$text</em>";
2934 * Normally we output all numbers in plain en_US style, that is
2935 * 293,291.235 for twohundredninetythreethousand-twohundredninetyone
2936 * point twohundredthirtyfive. However this is not suitable for all
2937 * languages, some such as Pakaran want ੨੯੩,੨੯੫.੨੩੫ and others such as
2938 * Icelandic just want to use commas instead of dots, and dots instead
2939 * of commas like "293.291,235".
2941 * An example of this function being called:
2943 * wfMessage( 'message' )->numParams( $num )->text()
2946 * See LanguageGu.php for the Gujarati implementation and
2947 * $separatorTransformTable on MessageIs.php for
2948 * the , => . and . => , implementation.
2950 * @todo check if it's viable to use localeconv() for the decimal
2952 * @param $number Mixed: the string to be formatted, should be an integer
2953 * or a floating point number.
2954 * @param $nocommafy Bool: set to true for special numbers like dates
2957 public function formatNum( $number, $nocommafy = false ) {
2958 global $wgTranslateNumerals;
2959 if ( !$nocommafy ) {
2960 $number = $this->commafy( $number );
2961 $s = $this->separatorTransformTable();
2963 $number = strtr( $number, $s );
2967 if ( $wgTranslateNumerals ) {
2968 $s = $this->digitTransformTable();
2970 $number = strtr( $number, $s );
2978 * Front-end for non-commafied formatNum
2980 * @param mixed $number the string to be formatted, should be an integer
2981 * or a floating point number.
2985 public function formatNumNoSeparators( $number ) {
2986 return $this->formatNum( $number, true );
2990 * @param $number string
2993 function parseFormattedNumber( $number ) {
2994 $s = $this->digitTransformTable();
2996 $number = strtr( $number, array_flip( $s ) );
2999 $s = $this->separatorTransformTable();
3001 $number = strtr( $number, array_flip( $s ) );
3004 $number = strtr( $number, array( ',' => '' ) );
3009 * Adds commas to a given number
3011 * @param $number mixed
3014 function commafy( $number ) {
3015 $digitGroupingPattern = $this->digitGroupingPattern();
3016 if ( $number === null ) {
3020 if ( !$digitGroupingPattern ||
$digitGroupingPattern === "###,###,###" ) {
3021 // default grouping is at thousands, use the same for ###,###,### pattern too.
3022 return strrev( (string)preg_replace( '/(\d{3})(?=\d)(?!\d*\.)/', '$1,', strrev( $number ) ) );
3024 // Ref: http://cldr.unicode.org/translation/number-patterns
3026 if ( intval( $number ) < 0 ) {
3027 // For negative numbers apply the algorithm like positive number and add sign.
3029 $number = substr( $number, 1 );
3031 $integerPart = array();
3032 $decimalPart = array();
3033 $numMatches = preg_match_all( "/(#+)/", $digitGroupingPattern, $matches );
3034 preg_match( "/\d+/", $number, $integerPart );
3035 preg_match( "/\.\d*/", $number, $decimalPart );
3036 $groupedNumber = ( count( $decimalPart ) > 0 ) ?
$decimalPart[0]:"";
3037 if ( $groupedNumber === $number ) {
3038 // the string does not have any number part. Eg: .12345
3039 return $sign . $groupedNumber;
3041 $start = $end = strlen( $integerPart[0] );
3042 while ( $start > 0 ) {
3043 $match = $matches[0][$numMatches -1] ;
3044 $matchLen = strlen( $match );
3045 $start = $end - $matchLen;
3049 $groupedNumber = substr( $number, $start, $end -$start ) . $groupedNumber ;
3051 if ( $numMatches > 1 ) {
3052 // use the last pattern for the rest of the number
3056 $groupedNumber = "," . $groupedNumber;
3059 return $sign . $groupedNumber;
3066 function digitGroupingPattern() {
3067 return self
::$dataCache->getItem( $this->mCode
, 'digitGroupingPattern' );
3073 function digitTransformTable() {
3074 return self
::$dataCache->getItem( $this->mCode
, 'digitTransformTable' );
3080 function separatorTransformTable() {
3081 return self
::$dataCache->getItem( $this->mCode
, 'separatorTransformTable' );
3085 * Take a list of strings and build a locale-friendly comma-separated
3086 * list, using the local comma-separator message.
3087 * The last two strings are chained with an "and".
3088 * NOTE: This function will only work with standard numeric array keys (0, 1, 2…)
3093 function listToText( array $l ) {
3094 $m = count( $l ) - 1;
3099 $and = $this->getMessageFromDB( 'and' );
3100 $space = $this->getMessageFromDB( 'word-separator' );
3102 $comma = $this->getMessageFromDB( 'comma-separator' );
3106 for ( $i = $m - 1; $i >= 0; $i-- ) {
3107 if ( $i == $m - 1 ) {
3108 $s = $l[$i] . $and . $space . $s;
3110 $s = $l[$i] . $comma . $s;
3117 * Take a list of strings and build a locale-friendly comma-separated
3118 * list, using the local comma-separator message.
3119 * @param $list array of strings to put in a comma list
3122 function commaList( array $list ) {
3124 wfMessage( 'comma-separator' )->inLanguage( $this )->escaped(),
3130 * Take a list of strings and build a locale-friendly semicolon-separated
3131 * list, using the local semicolon-separator message.
3132 * @param $list array of strings to put in a semicolon list
3135 function semicolonList( array $list ) {
3137 wfMessage( 'semicolon-separator' )->inLanguage( $this )->escaped(),
3143 * Same as commaList, but separate it with the pipe instead.
3144 * @param $list array of strings to put in a pipe list
3147 function pipeList( array $list ) {
3149 wfMessage( 'pipe-separator' )->inLanguage( $this )->escaped(),
3155 * Truncate a string to a specified length in bytes, appending an optional
3156 * string (e.g. for ellipses)
3158 * The database offers limited byte lengths for some columns in the database;
3159 * multi-byte character sets mean we need to ensure that only whole characters
3160 * are included, otherwise broken characters can be passed to the user
3162 * If $length is negative, the string will be truncated from the beginning
3164 * @param $string String to truncate
3165 * @param $length Int: maximum length (including ellipses)
3166 * @param $ellipsis String to append to the truncated text
3167 * @param $adjustLength Boolean: Subtract length of ellipsis from $length.
3168 * $adjustLength was introduced in 1.18, before that behaved as if false.
3171 function truncate( $string, $length, $ellipsis = '...', $adjustLength = true ) {
3172 # Use the localized ellipsis character
3173 if ( $ellipsis == '...' ) {
3174 $ellipsis = wfMessage( 'ellipsis' )->inLanguage( $this )->escaped();
3176 # Check if there is no need to truncate
3177 if ( $length == 0 ) {
3178 return $ellipsis; // convention
3179 } elseif ( strlen( $string ) <= abs( $length ) ) {
3180 return $string; // no need to truncate
3182 $stringOriginal = $string;
3183 # If ellipsis length is >= $length then we can't apply $adjustLength
3184 if ( $adjustLength && strlen( $ellipsis ) >= abs( $length ) ) {
3185 $string = $ellipsis; // this can be slightly unexpected
3186 # Otherwise, truncate and add ellipsis...
3188 $eLength = $adjustLength ?
strlen( $ellipsis ) : 0;
3189 if ( $length > 0 ) {
3190 $length -= $eLength;
3191 $string = substr( $string, 0, $length ); // xyz...
3192 $string = $this->removeBadCharLast( $string );
3193 $string = $string . $ellipsis;
3195 $length +
= $eLength;
3196 $string = substr( $string, $length ); // ...xyz
3197 $string = $this->removeBadCharFirst( $string );
3198 $string = $ellipsis . $string;
3201 # Do not truncate if the ellipsis makes the string longer/equal (bug 22181).
3202 # This check is *not* redundant if $adjustLength, due to the single case where
3203 # LEN($ellipsis) > ABS($limit arg); $stringOriginal could be shorter than $string.
3204 if ( strlen( $string ) < strlen( $stringOriginal ) ) {
3207 return $stringOriginal;
3212 * Remove bytes that represent an incomplete Unicode character
3213 * at the end of string (e.g. bytes of the char are missing)
3215 * @param $string String
3218 protected function removeBadCharLast( $string ) {
3219 if ( $string != '' ) {
3220 $char = ord( $string[strlen( $string ) - 1] );
3222 if ( $char >= 0xc0 ) {
3223 # We got the first byte only of a multibyte char; remove it.
3224 $string = substr( $string, 0, -1 );
3225 } elseif ( $char >= 0x80 &&
3226 preg_match( '/^(.*)(?:[\xe0-\xef][\x80-\xbf]|' .
3227 '[\xf0-\xf7][\x80-\xbf]{1,2})$/', $string, $m ) )
3229 # We chopped in the middle of a character; remove it
3237 * Remove bytes that represent an incomplete Unicode character
3238 * at the start of string (e.g. bytes of the char are missing)
3240 * @param $string String
3243 protected function removeBadCharFirst( $string ) {
3244 if ( $string != '' ) {
3245 $char = ord( $string[0] );
3246 if ( $char >= 0x80 && $char < 0xc0 ) {
3247 # We chopped in the middle of a character; remove the whole thing
3248 $string = preg_replace( '/^[\x80-\xbf]+/', '', $string );
3255 * Truncate a string of valid HTML to a specified length in bytes,
3256 * appending an optional string (e.g. for ellipses), and return valid HTML
3258 * This is only intended for styled/linked text, such as HTML with
3259 * tags like <span> and <a>, were the tags are self-contained (valid HTML).
3260 * Also, this will not detect things like "display:none" CSS.
3262 * Note: since 1.18 you do not need to leave extra room in $length for ellipses.
3264 * @param string $text HTML string to truncate
3265 * @param int $length (zero/positive) Maximum length (including ellipses)
3266 * @param string $ellipsis String to append to the truncated text
3269 function truncateHtml( $text, $length, $ellipsis = '...' ) {
3270 # Use the localized ellipsis character
3271 if ( $ellipsis == '...' ) {
3272 $ellipsis = wfMessage( 'ellipsis' )->inLanguage( $this )->escaped();
3274 # Check if there is clearly no need to truncate
3275 if ( $length <= 0 ) {
3276 return $ellipsis; // no text shown, nothing to format (convention)
3277 } elseif ( strlen( $text ) <= $length ) {
3278 return $text; // string short enough even *with* HTML (short-circuit)
3281 $dispLen = 0; // innerHTML legth so far
3282 $testingEllipsis = false; // checking if ellipses will make string longer/equal?
3283 $tagType = 0; // 0-open, 1-close
3284 $bracketState = 0; // 1-tag start, 2-tag name, 0-neither
3285 $entityState = 0; // 0-not entity, 1-entity
3286 $tag = $ret = ''; // accumulated tag name, accumulated result string
3287 $openTags = array(); // open tag stack
3288 $maybeState = null; // possible truncation state
3290 $textLen = strlen( $text );
3291 $neLength = max( 0, $length - strlen( $ellipsis ) ); // non-ellipsis len if truncated
3292 for ( $pos = 0; true; ++
$pos ) {
3293 # Consider truncation once the display length has reached the maximim.
3294 # We check if $dispLen > 0 to grab tags for the $neLength = 0 case.
3295 # Check that we're not in the middle of a bracket/entity...
3296 if ( $dispLen && $dispLen >= $neLength && $bracketState == 0 && !$entityState ) {
3297 if ( !$testingEllipsis ) {
3298 $testingEllipsis = true;
3299 # Save where we are; we will truncate here unless there turn out to
3300 # be so few remaining characters that truncation is not necessary.
3301 if ( !$maybeState ) { // already saved? ($neLength = 0 case)
3302 $maybeState = array( $ret, $openTags ); // save state
3304 } elseif ( $dispLen > $length && $dispLen > strlen( $ellipsis ) ) {
3305 # String in fact does need truncation, the truncation point was OK.
3306 list( $ret, $openTags ) = $maybeState; // reload state
3307 $ret = $this->removeBadCharLast( $ret ); // multi-byte char fix
3308 $ret .= $ellipsis; // add ellipsis
3312 if ( $pos >= $textLen ) break; // extra iteration just for above checks
3314 # Read the next char...
3316 $lastCh = $pos ?
$text[$pos - 1] : '';
3317 $ret .= $ch; // add to result string
3319 $this->truncate_endBracket( $tag, $tagType, $lastCh, $openTags ); // for bad HTML
3320 $entityState = 0; // for bad HTML
3321 $bracketState = 1; // tag started (checking for backslash)
3322 } elseif ( $ch == '>' ) {
3323 $this->truncate_endBracket( $tag, $tagType, $lastCh, $openTags );
3324 $entityState = 0; // for bad HTML
3325 $bracketState = 0; // out of brackets
3326 } elseif ( $bracketState == 1 ) {
3328 $tagType = 1; // close tag (e.g. "</span>")
3330 $tagType = 0; // open tag (e.g. "<span>")
3333 $bracketState = 2; // building tag name
3334 } elseif ( $bracketState == 2 ) {
3338 // Name found (e.g. "<a href=..."), add on tag attributes...
3339 $pos +
= $this->truncate_skip( $ret, $text, "<>", $pos +
1 );
3341 } elseif ( $bracketState == 0 ) {
3342 if ( $entityState ) {
3345 $dispLen++
; // entity is one displayed char
3348 if ( $neLength == 0 && !$maybeState ) {
3349 // Save state without $ch. We want to *hit* the first
3350 // display char (to get tags) but not *use* it if truncating.
3351 $maybeState = array( substr( $ret, 0, -1 ), $openTags );
3354 $entityState = 1; // entity found, (e.g. " ")
3356 $dispLen++
; // this char is displayed
3357 // Add the next $max display text chars after this in one swoop...
3358 $max = ( $testingEllipsis ?
$length : $neLength ) - $dispLen;
3359 $skipped = $this->truncate_skip( $ret, $text, "<>&", $pos +
1, $max );
3360 $dispLen +
= $skipped;
3366 // Close the last tag if left unclosed by bad HTML
3367 $this->truncate_endBracket( $tag, $text[$textLen - 1], $tagType, $openTags );
3368 while ( count( $openTags ) > 0 ) {
3369 $ret .= '</' . array_pop( $openTags ) . '>'; // close open tags
3375 * truncateHtml() helper function
3376 * like strcspn() but adds the skipped chars to $ret
3385 private function truncate_skip( &$ret, $text, $search, $start, $len = null ) {
3386 if ( $len === null ) {
3387 $len = -1; // -1 means "no limit" for strcspn
3388 } elseif ( $len < 0 ) {
3392 if ( $start < strlen( $text ) ) {
3393 $skipCount = strcspn( $text, $search, $start, $len );
3394 $ret .= substr( $text, $start, $skipCount );
3400 * truncateHtml() helper function
3401 * (a) push or pop $tag from $openTags as needed
3402 * (b) clear $tag value
3403 * @param &$tag string Current HTML tag name we are looking at
3404 * @param $tagType int (0-open tag, 1-close tag)
3405 * @param $lastCh string Character before the '>' that ended this tag
3406 * @param &$openTags array Open tag stack (not accounting for $tag)
3408 private function truncate_endBracket( &$tag, $tagType, $lastCh, &$openTags ) {
3409 $tag = ltrim( $tag );
3411 if ( $tagType == 0 && $lastCh != '/' ) {
3412 $openTags[] = $tag; // tag opened (didn't close itself)
3413 } elseif ( $tagType == 1 ) {
3414 if ( $openTags && $tag == $openTags[count( $openTags ) - 1] ) {
3415 array_pop( $openTags ); // tag closed
3423 * Grammatical transformations, needed for inflected languages
3424 * Invoked by putting {{grammar:case|word}} in a message
3426 * @param $word string
3427 * @param $case string
3430 function convertGrammar( $word, $case ) {
3431 global $wgGrammarForms;
3432 if ( isset( $wgGrammarForms[$this->getCode()][$case][$word] ) ) {
3433 return $wgGrammarForms[$this->getCode()][$case][$word];
3438 * Get the grammar forms for the content language
3439 * @return array of grammar forms
3442 function getGrammarForms() {
3443 global $wgGrammarForms;
3444 if ( isset( $wgGrammarForms[$this->getCode()] ) && is_array( $wgGrammarForms[$this->getCode()] ) ) {
3445 return $wgGrammarForms[$this->getCode()];
3450 * Provides an alternative text depending on specified gender.
3451 * Usage {{gender:username|masculine|feminine|neutral}}.
3452 * username is optional, in which case the gender of current user is used,
3453 * but only in (some) interface messages; otherwise default gender is used.
3455 * If no forms are given, an empty string is returned. If only one form is
3456 * given, it will be returned unconditionally. These details are implied by
3457 * the caller and cannot be overridden in subclasses.
3459 * If more than one form is given, the default is to use the neutral one
3460 * if it is specified, and to use the masculine one otherwise. These
3461 * details can be overridden in subclasses.
3463 * @param $gender string
3464 * @param $forms array
3468 function gender( $gender, $forms ) {
3469 if ( !count( $forms ) ) {
3472 $forms = $this->preConvertPlural( $forms, 2 );
3473 if ( $gender === 'male' ) {
3476 if ( $gender === 'female' ) {
3479 return isset( $forms[2] ) ?
$forms[2] : $forms[0];
3483 * Plural form transformations, needed for some languages.
3484 * For example, there are 3 form of plural in Russian and Polish,
3485 * depending on "count mod 10". See [[w:Plural]]
3486 * For English it is pretty simple.
3488 * Invoked by putting {{plural:count|wordform1|wordform2}}
3489 * or {{plural:count|wordform1|wordform2|wordform3}}
3491 * Example: {{plural:{{NUMBEROFARTICLES}}|article|articles}}
3493 * @param $count Integer: non-localized number
3494 * @param $forms Array: different plural forms
3495 * @return string Correct form of plural for $count in this language
3497 function convertPlural( $count, $forms ) {
3498 if ( !count( $forms ) ) {
3502 // Handle explicit n=pluralform cases
3503 foreach ( $forms as $index => $form ) {
3504 if ( preg_match( '/\d+=/i', $form ) ) {
3505 $pos = strpos( $form, '=' );
3506 if ( substr( $form, 0, $pos ) === (string) $count ) {
3507 return substr( $form, $pos +
1 );
3509 unset( $forms[$index] );
3512 $forms = array_values( $forms );
3514 $pluralForm = $this->getPluralRuleIndexNumber( $count );
3515 $pluralForm = min( $pluralForm, count( $forms ) - 1 );
3516 return $forms[$pluralForm];
3520 * Checks that convertPlural was given an array and pads it to requested
3521 * amount of forms by copying the last one.
3523 * @param $count Integer: How many forms should there be at least
3524 * @param $forms Array of forms given to convertPlural
3525 * @return array Padded array of forms or an exception if not an array
3527 protected function preConvertPlural( /* Array */ $forms, $count ) {
3528 while ( count( $forms ) < $count ) {
3529 $forms[] = $forms[count( $forms ) - 1];
3535 * @todo Maybe translate block durations. Note that this function is somewhat misnamed: it
3536 * deals with translating the *duration* ("1 week", "4 days", etc), not the expiry time
3537 * (which is an absolute timestamp). Please note: do NOT add this blindly, as it is used
3538 * on old expiry lengths recorded in log entries. You'd need to provide the start date to
3541 * @param $str String: the validated block duration in English
3542 * @return string Somehow translated block duration
3543 * @see LanguageFi.php for example implementation
3545 function translateBlockExpiry( $str ) {
3546 $duration = SpecialBlock
::getSuggestedDurations( $this );
3547 foreach ( $duration as $show => $value ) {
3548 if ( strcmp( $str, $value ) == 0 ) {
3549 return htmlspecialchars( trim( $show ) );
3553 // Since usually only infinite or indefinite is only on list, so try
3554 // equivalents if still here.
3555 $indefs = array( 'infinite', 'infinity', 'indefinite' );
3556 if ( in_array( $str, $indefs ) ) {
3557 foreach ( $indefs as $val ) {
3558 $show = array_search( $val, $duration, true );
3559 if ( $show !== false ) {
3560 return htmlspecialchars( trim( $show ) );
3565 // If all else fails, return a standard duration or timestamp description.
3566 $time = strtotime( $str, 0 );
3567 if ( $time === false ) { // Unknown format. Return it as-is in case.
3569 } elseif ( $time !== strtotime( $str, 1 ) ) { // It's a relative timestamp.
3570 // $time is relative to 0 so it's a duration length.
3571 return $this->formatDuration( $time );
3572 } else { // It's an absolute timestamp.
3573 if ( $time === 0 ) {
3574 // wfTimestamp() handles 0 as current time instead of epoch.
3575 return $this->timeanddate( '19700101000000' );
3577 return $this->timeanddate( $time );
3583 * languages like Chinese need to be segmented in order for the diff
3586 * @param $text String
3589 public function segmentForDiff( $text ) {
3594 * and unsegment to show the result
3596 * @param $text String
3599 public function unsegmentForDiff( $text ) {
3604 * Return the LanguageConverter used in the Language
3607 * @return LanguageConverter
3609 public function getConverter() {
3610 return $this->mConverter
;
3614 * convert text to all supported variants
3616 * @param $text string
3619 public function autoConvertToAllVariants( $text ) {
3620 return $this->mConverter
->autoConvertToAllVariants( $text );
3624 * convert text to different variants of a language.
3626 * @param $text string
3629 public function convert( $text ) {
3630 return $this->mConverter
->convert( $text );
3634 * Convert a Title object to a string in the preferred variant
3636 * @param $title Title
3639 public function convertTitle( $title ) {
3640 return $this->mConverter
->convertTitle( $title );
3644 * Convert a namespace index to a string in the preferred variant
3649 public function convertNamespace( $ns ) {
3650 return $this->mConverter
->convertNamespace( $ns );
3654 * Check if this is a language with variants
3658 public function hasVariants() {
3659 return count( $this->getVariants() ) > 1;
3663 * Check if the language has the specific variant
3666 * @param $variant string
3669 public function hasVariant( $variant ) {
3670 return (bool)$this->mConverter
->validateVariant( $variant );
3674 * Put custom tags (e.g. -{ }-) around math to prevent conversion
3676 * @param $text string
3679 public function armourMath( $text ) {
3680 return $this->mConverter
->armourMath( $text );
3684 * Perform output conversion on a string, and encode for safe HTML output.
3685 * @param $text String text to be converted
3686 * @param $isTitle Bool whether this conversion is for the article title
3688 * @todo this should get integrated somewhere sane
3690 public function convertHtml( $text, $isTitle = false ) {
3691 return htmlspecialchars( $this->convert( $text, $isTitle ) );
3695 * @param $key string
3698 public function convertCategoryKey( $key ) {
3699 return $this->mConverter
->convertCategoryKey( $key );
3703 * Get the list of variants supported by this language
3704 * see sample implementation in LanguageZh.php
3706 * @return array an array of language codes
3708 public function getVariants() {
3709 return $this->mConverter
->getVariants();
3715 public function getPreferredVariant() {
3716 return $this->mConverter
->getPreferredVariant();
3722 public function getDefaultVariant() {
3723 return $this->mConverter
->getDefaultVariant();
3729 public function getURLVariant() {
3730 return $this->mConverter
->getURLVariant();
3734 * If a language supports multiple variants, it is
3735 * possible that non-existing link in one variant
3736 * actually exists in another variant. this function
3737 * tries to find it. See e.g. LanguageZh.php
3739 * @param $link String: the name of the link
3740 * @param $nt Mixed: the title object of the link
3741 * @param $ignoreOtherCond Boolean: to disable other conditions when
3742 * we need to transclude a template or update a category's link
3743 * @return null the input parameters may be modified upon return
3745 public function findVariantLink( &$link, &$nt, $ignoreOtherCond = false ) {
3746 $this->mConverter
->findVariantLink( $link, $nt, $ignoreOtherCond );
3750 * If a language supports multiple variants, converts text
3751 * into an array of all possible variants of the text:
3752 * 'variant' => text in that variant
3754 * @deprecated since 1.17 Use autoConvertToAllVariants()
3756 * @param $text string
3760 public function convertLinkToAllVariants( $text ) {
3761 return $this->mConverter
->convertLinkToAllVariants( $text );
3765 * returns language specific options used by User::getPageRenderHash()
3766 * for example, the preferred language variant
3770 function getExtraHashOptions() {
3771 return $this->mConverter
->getExtraHashOptions();
3775 * For languages that support multiple variants, the title of an
3776 * article may be displayed differently in different variants. this
3777 * function returns the apporiate title defined in the body of the article.
3781 public function getParsedTitle() {
3782 return $this->mConverter
->getParsedTitle();
3786 * Prepare external link text for conversion. When the text is
3787 * a URL, it shouldn't be converted, and it'll be wrapped in
3788 * the "raw" tag (-{R| }-) to prevent conversion.
3790 * This function is called "markNoConversion" for historical
3793 * @param $text String: text to be used for external link
3794 * @param $noParse bool: wrap it without confirming it's a real URL first
3795 * @return string the tagged text
3797 public function markNoConversion( $text, $noParse = false ) {
3798 // Excluding protocal-relative URLs may avoid many false positives.
3799 if ( $noParse ||
preg_match( '/^(?:' . wfUrlProtocolsWithoutProtRel() . ')/', $text ) ) {
3800 return $this->mConverter
->markNoConversion( $text );
3807 * A regular expression to match legal word-trailing characters
3808 * which should be merged onto a link of the form [[foo]]bar.
3812 public function linkTrail() {
3813 return self
::$dataCache->getItem( $this->mCode
, 'linkTrail' );
3819 function getLangObj() {
3824 * Get the RFC 3066 code for this language object
3826 * NOTE: The return value of this function is NOT HTML-safe and must be escaped with
3827 * htmlspecialchars() or similar
3831 public function getCode() {
3832 return $this->mCode
;
3836 * Get the code in Bcp47 format which we can use
3837 * inside of html lang="" tags.
3839 * NOTE: The return value of this function is NOT HTML-safe and must be escaped with
3840 * htmlspecialchars() or similar.
3845 public function getHtmlCode() {
3846 if ( is_null( $this->mHtmlCode
) ) {
3847 $this->mHtmlCode
= wfBCP47( $this->getCode() );
3849 return $this->mHtmlCode
;
3853 * @param $code string
3855 public function setCode( $code ) {
3856 $this->mCode
= $code;
3857 // Ensure we don't leave an incorrect html code lying around
3858 $this->mHtmlCode
= null;
3862 * Get the name of a file for a certain language code
3863 * @param $prefix string Prepend this to the filename
3864 * @param $code string Language code
3865 * @param $suffix string Append this to the filename
3866 * @throws MWException
3867 * @return string $prefix . $mangledCode . $suffix
3869 public static function getFileName( $prefix = 'Language', $code, $suffix = '.php' ) {
3870 // Protect against path traversal
3871 if ( !Language
::isValidCode( $code )
3872 ||
strcspn( $code, ":/\\\000" ) !== strlen( $code ) )
3874 throw new MWException( "Invalid language code \"$code\"" );
3877 return $prefix . str_replace( '-', '_', ucfirst( $code ) ) . $suffix;
3881 * Get the language code from a file name. Inverse of getFileName()
3882 * @param $filename string $prefix . $languageCode . $suffix
3883 * @param $prefix string Prefix before the language code
3884 * @param $suffix string Suffix after the language code
3885 * @return string Language code, or false if $prefix or $suffix isn't found
3887 public static function getCodeFromFileName( $filename, $prefix = 'Language', $suffix = '.php' ) {
3889 preg_match( '/' . preg_quote( $prefix, '/' ) . '([A-Z][a-z_]+)' .
3890 preg_quote( $suffix, '/' ) . '/', $filename, $m );
3891 if ( !count( $m ) ) {
3894 return str_replace( '_', '-', strtolower( $m[1] ) );
3898 * @param $code string
3901 public static function getMessagesFileName( $code ) {
3903 $file = self
::getFileName( "$IP/languages/messages/Messages", $code, '.php' );
3904 wfRunHooks( 'Language::getMessagesFileName', array( $code, &$file ) );
3909 * @param $code string
3912 public static function getClassFileName( $code ) {
3914 return self
::getFileName( "$IP/languages/classes/Language", $code, '.php' );
3918 * Get the first fallback for a given language.
3920 * @param $code string
3922 * @return bool|string
3924 public static function getFallbackFor( $code ) {
3925 if ( $code === 'en' ||
!Language
::isValidBuiltInCode( $code ) ) {
3928 $fallbacks = self
::getFallbacksFor( $code );
3929 $first = array_shift( $fallbacks );
3935 * Get the ordered list of fallback languages.
3938 * @param $code string Language code
3941 public static function getFallbacksFor( $code ) {
3942 if ( $code === 'en' ||
!Language
::isValidBuiltInCode( $code ) ) {
3945 $v = self
::getLocalisationCache()->getItem( $code, 'fallback' );
3946 $v = array_map( 'trim', explode( ',', $v ) );
3947 if ( $v[count( $v ) - 1] !== 'en' ) {
3955 * Get all messages for a given language
3956 * WARNING: this may take a long time. If you just need all message *keys*
3957 * but need the *contents* of only a few messages, consider using getMessageKeysFor().
3959 * @param $code string
3963 public static function getMessagesFor( $code ) {
3964 return self
::getLocalisationCache()->getItem( $code, 'messages' );
3968 * Get a message for a given language
3970 * @param $key string
3971 * @param $code string
3975 public static function getMessageFor( $key, $code ) {
3976 return self
::getLocalisationCache()->getSubitem( $code, 'messages', $key );
3980 * Get all message keys for a given language. This is a faster alternative to
3981 * array_keys( Language::getMessagesFor( $code ) )
3984 * @param $code string Language code
3985 * @return array of message keys (strings)
3987 public static function getMessageKeysFor( $code ) {
3988 return self
::getLocalisationCache()->getSubItemList( $code, 'messages' );
3995 function fixVariableInNamespace( $talk ) {
3996 if ( strpos( $talk, '$1' ) === false ) {
4000 global $wgMetaNamespace;
4001 $talk = str_replace( '$1', $wgMetaNamespace, $talk );
4003 # Allow grammar transformations
4004 # Allowing full message-style parsing would make simple requests
4005 # such as action=raw much more expensive than they need to be.
4006 # This will hopefully cover most cases.
4007 $talk = preg_replace_callback( '/{{grammar:(.*?)\|(.*?)}}/i',
4008 array( &$this, 'replaceGrammarInNamespace' ), $talk );
4009 return str_replace( ' ', '_', $talk );
4016 function replaceGrammarInNamespace( $m ) {
4017 return $this->convertGrammar( trim( $m[2] ), trim( $m[1] ) );
4021 * @throws MWException
4024 static function getCaseMaps() {
4025 static $wikiUpperChars, $wikiLowerChars;
4026 if ( isset( $wikiUpperChars ) ) {
4027 return array( $wikiUpperChars, $wikiLowerChars );
4030 wfProfileIn( __METHOD__
);
4031 $arr = wfGetPrecompiledData( 'Utf8Case.ser' );
4032 if ( $arr === false ) {
4033 throw new MWException(
4034 "Utf8Case.ser is missing, please run \"make\" in the serialized directory\n" );
4036 $wikiUpperChars = $arr['wikiUpperChars'];
4037 $wikiLowerChars = $arr['wikiLowerChars'];
4038 wfProfileOut( __METHOD__
);
4039 return array( $wikiUpperChars, $wikiLowerChars );
4043 * Decode an expiry (block, protection, etc) which has come from the DB
4045 * @todo FIXME: why are we returnings DBMS-dependent strings???
4047 * @param $expiry String: Database expiry String
4048 * @param $format Bool|Int true to process using language functions, or TS_ constant
4049 * to return the expiry in a given timestamp
4053 public function formatExpiry( $expiry, $format = true ) {
4054 static $infinity, $infinityMsg;
4055 if ( $infinity === null ) {
4056 $infinityMsg = wfMessage( 'infiniteblock' );
4057 $infinity = wfGetDB( DB_SLAVE
)->getInfinity();
4060 if ( $expiry == '' ||
$expiry == $infinity ) {
4061 return $format === true
4065 return $format === true
4066 ?
$this->timeanddate( $expiry, /* User preference timezone */ true )
4067 : wfTimestamp( $format, $expiry );
4073 * @param $seconds int|float
4074 * @param $format Array Optional
4075 * If $format['avoid'] == 'avoidseconds' - don't mention seconds if $seconds >= 1 hour
4076 * If $format['avoid'] == 'avoidminutes' - don't mention seconds/minutes if $seconds > 48 hours
4077 * If $format['noabbrevs'] is true - use 'seconds' and friends instead of 'seconds-abbrev' and friends
4078 * For backwards compatibility, $format may also be one of the strings 'avoidseconds' or 'avoidminutes'
4081 function formatTimePeriod( $seconds, $format = array() ) {
4082 if ( !is_array( $format ) ) {
4083 $format = array( 'avoid' => $format ); // For backwards compatibility
4085 if ( !isset( $format['avoid'] ) ) {
4086 $format['avoid'] = false;
4088 if ( !isset( $format['noabbrevs' ] ) ) {
4089 $format['noabbrevs'] = false;
4091 $secondsMsg = wfMessage(
4092 $format['noabbrevs'] ?
'seconds' : 'seconds-abbrev' )->inLanguage( $this );
4093 $minutesMsg = wfMessage(
4094 $format['noabbrevs'] ?
'minutes' : 'minutes-abbrev' )->inLanguage( $this );
4095 $hoursMsg = wfMessage(
4096 $format['noabbrevs'] ?
'hours' : 'hours-abbrev' )->inLanguage( $this );
4097 $daysMsg = wfMessage(
4098 $format['noabbrevs'] ?
'days' : 'days-abbrev' )->inLanguage( $this );
4100 if ( round( $seconds * 10 ) < 100 ) {
4101 $s = $this->formatNum( sprintf( "%.1f", round( $seconds * 10 ) / 10 ) );
4102 $s = $secondsMsg->params( $s )->text();
4103 } elseif ( round( $seconds ) < 60 ) {
4104 $s = $this->formatNum( round( $seconds ) );
4105 $s = $secondsMsg->params( $s )->text();
4106 } elseif ( round( $seconds ) < 3600 ) {
4107 $minutes = floor( $seconds / 60 );
4108 $secondsPart = round( fmod( $seconds, 60 ) );
4109 if ( $secondsPart == 60 ) {
4113 $s = $minutesMsg->params( $this->formatNum( $minutes ) )->text();
4115 $s .= $secondsMsg->params( $this->formatNum( $secondsPart ) )->text();
4116 } elseif ( round( $seconds ) <= 2 * 86400 ) {
4117 $hours = floor( $seconds / 3600 );
4118 $minutes = floor( ( $seconds - $hours * 3600 ) / 60 );
4119 $secondsPart = round( $seconds - $hours * 3600 - $minutes * 60 );
4120 if ( $secondsPart == 60 ) {
4124 if ( $minutes == 60 ) {
4128 $s = $hoursMsg->params( $this->formatNum( $hours ) )->text();
4130 $s .= $minutesMsg->params( $this->formatNum( $minutes ) )->text();
4131 if ( !in_array( $format['avoid'], array( 'avoidseconds', 'avoidminutes' ) ) ) {
4132 $s .= ' ' . $secondsMsg->params( $this->formatNum( $secondsPart ) )->text();
4135 $days = floor( $seconds / 86400 );
4136 if ( $format['avoid'] === 'avoidminutes' ) {
4137 $hours = round( ( $seconds - $days * 86400 ) / 3600 );
4138 if ( $hours == 24 ) {
4142 $s = $daysMsg->params( $this->formatNum( $days ) )->text();
4144 $s .= $hoursMsg->params( $this->formatNum( $hours ) )->text();
4145 } elseif ( $format['avoid'] === 'avoidseconds' ) {
4146 $hours = floor( ( $seconds - $days * 86400 ) / 3600 );
4147 $minutes = round( ( $seconds - $days * 86400 - $hours * 3600 ) / 60 );
4148 if ( $minutes == 60 ) {
4152 if ( $hours == 24 ) {
4156 $s = $daysMsg->params( $this->formatNum( $days ) )->text();
4158 $s .= $hoursMsg->params( $this->formatNum( $hours ) )->text();
4160 $s .= $minutesMsg->params( $this->formatNum( $minutes ) )->text();
4162 $s = $daysMsg->params( $this->formatNum( $days ) )->text();
4164 $s .= $this->formatTimePeriod( $seconds - $days * 86400, $format );
4171 * Format a bitrate for output, using an appropriate
4172 * unit (bps, kbps, Mbps, Gbps, Tbps, Pbps, Ebps, Zbps or Ybps) according to the magnitude in question
4174 * This use base 1000. For base 1024 use formatSize(), for another base
4175 * see formatComputingNumbers()
4180 function formatBitrate( $bps ) {
4181 return $this->formatComputingNumbers( $bps, 1000, "bitrate-$1bits" );
4185 * @param $size int Size of the unit
4186 * @param $boundary int Size boundary (1000, or 1024 in most cases)
4187 * @param $messageKey string Message key to be uesd
4190 function formatComputingNumbers( $size, $boundary, $messageKey ) {
4192 return str_replace( '$1', $this->formatNum( $size ),
4193 $this->getMessageFromDB( str_replace( '$1', '', $messageKey ) )
4196 $sizes = array( '', 'kilo', 'mega', 'giga', 'tera', 'peta', 'exa', 'zeta', 'yotta' );
4199 $maxIndex = count( $sizes ) - 1;
4200 while ( $size >= $boundary && $index < $maxIndex ) {
4205 // For small sizes no decimal places necessary
4208 // For MB and bigger two decimal places are smarter
4211 $msg = str_replace( '$1', $sizes[$index], $messageKey );
4213 $size = round( $size, $round );
4214 $text = $this->getMessageFromDB( $msg );
4215 return str_replace( '$1', $this->formatNum( $size ), $text );
4219 * Format a size in bytes for output, using an appropriate
4220 * unit (B, KB, MB, GB, TB, PB, EB, ZB or YB) according to the magnitude in question
4222 * This method use base 1024. For base 1000 use formatBitrate(), for
4223 * another base see formatComputingNumbers()
4225 * @param $size int Size to format
4226 * @return string Plain text (not HTML)
4228 function formatSize( $size ) {
4229 return $this->formatComputingNumbers( $size, 1024, "size-$1bytes" );
4233 * Make a list item, used by various special pages
4235 * @param $page String Page link
4236 * @param $details String Text between brackets
4237 * @param $oppositedm Boolean Add the direction mark opposite to your
4238 * language, to display text properly
4241 function specialList( $page, $details, $oppositedm = true ) {
4242 $dirmark = ( $oppositedm ?
$this->getDirMark( true ) : '' ) .
4243 $this->getDirMark();
4244 $details = $details ?
$dirmark . $this->getMessageFromDB( 'word-separator' ) .
4245 wfMessage( 'parentheses' )->rawParams( $details )->inLanguage( $this )->escaped() : '';
4246 return $page . $details;
4250 * Generate (prev x| next x) (20|50|100...) type links for paging
4252 * @param $title Title object to link
4253 * @param $offset Integer offset parameter
4254 * @param $limit Integer limit parameter
4255 * @param $query array|String optional URL query parameter string
4256 * @param $atend Bool optional param for specified if this is the last page
4259 public function viewPrevNext( Title
$title, $offset, $limit, array $query = array(), $atend = false ) {
4260 // @todo FIXME: Why on earth this needs one message for the text and another one for tooltip?
4262 # Make 'previous' link
4263 $prev = wfMessage( 'prevn' )->inLanguage( $this )->title( $title )->numParams( $limit )->text();
4264 if ( $offset > 0 ) {
4265 $plink = $this->numLink( $title, max( $offset - $limit, 0 ), $limit,
4266 $query, $prev, 'prevn-title', 'mw-prevlink' );
4268 $plink = htmlspecialchars( $prev );
4272 $next = wfMessage( 'nextn' )->inLanguage( $this )->title( $title )->numParams( $limit )->text();
4274 $nlink = htmlspecialchars( $next );
4276 $nlink = $this->numLink( $title, $offset +
$limit, $limit,
4277 $query, $next, 'prevn-title', 'mw-nextlink' );
4280 # Make links to set number of items per page
4281 $numLinks = array();
4282 foreach ( array( 20, 50, 100, 250, 500 ) as $num ) {
4283 $numLinks[] = $this->numLink( $title, $offset, $num,
4284 $query, $this->formatNum( $num ), 'shown-title', 'mw-numlink' );
4287 return wfMessage( 'viewprevnext' )->inLanguage( $this )->title( $title
4288 )->rawParams( $plink, $nlink, $this->pipeList( $numLinks ) )->escaped();
4292 * Helper function for viewPrevNext() that generates links
4294 * @param $title Title object to link
4295 * @param $offset Integer offset parameter
4296 * @param $limit Integer limit parameter
4297 * @param $query Array extra query parameters
4298 * @param $link String text to use for the link; will be escaped
4299 * @param $tooltipMsg String name of the message to use as tooltip
4300 * @param $class String value of the "class" attribute of the link
4301 * @return String HTML fragment
4303 private function numLink( Title
$title, $offset, $limit, array $query, $link, $tooltipMsg, $class ) {
4304 $query = array( 'limit' => $limit, 'offset' => $offset ) +
$query;
4305 $tooltip = wfMessage( $tooltipMsg )->inLanguage( $this )->title( $title )->numParams( $limit )->text();
4306 return Html
::element( 'a', array( 'href' => $title->getLocalURL( $query ),
4307 'title' => $tooltip, 'class' => $class ), $link );
4311 * Get the conversion rule title, if any.
4315 public function getConvRuleTitle() {
4316 return $this->mConverter
->getConvRuleTitle();
4320 * Get the compiled plural rules for the language
4322 * @return array Associative array with plural form, and plural rule as key-value pairs
4324 public function getCompiledPluralRules() {
4325 $pluralRules = self
::$dataCache->getItem( strtolower( $this->mCode
), 'compiledPluralRules' );
4326 $fallbacks = Language
::getFallbacksFor( $this->mCode
);
4327 if ( !$pluralRules ) {
4328 foreach ( $fallbacks as $fallbackCode ) {
4329 $pluralRules = self
::$dataCache->getItem( strtolower( $fallbackCode ), 'compiledPluralRules' );
4330 if ( $pluralRules ) {
4335 return $pluralRules;
4339 * Get the plural rules for the language
4341 * @return array Associative array with plural form number and plural rule as key-value pairs
4343 public function getPluralRules() {
4344 $pluralRules = self
::$dataCache->getItem( strtolower( $this->mCode
), 'pluralRules' );
4345 $fallbacks = Language
::getFallbacksFor( $this->mCode
);
4346 if ( !$pluralRules ) {
4347 foreach ( $fallbacks as $fallbackCode ) {
4348 $pluralRules = self
::$dataCache->getItem( strtolower( $fallbackCode ), 'pluralRules' );
4349 if ( $pluralRules ) {
4354 return $pluralRules;
4358 * Get the plural rule types for the language
4360 * @return array Associative array with plural form number and plural rule type as key-value pairs
4362 public function getPluralRuleTypes() {
4363 $pluralRuleTypes = self
::$dataCache->getItem( strtolower( $this->mCode
), 'pluralRuleTypes' );
4364 $fallbacks = Language
::getFallbacksFor( $this->mCode
);
4365 if ( !$pluralRuleTypes ) {
4366 foreach ( $fallbacks as $fallbackCode ) {
4367 $pluralRuleTypes = self
::$dataCache->getItem( strtolower( $fallbackCode ), 'pluralRuleTypes' );
4368 if ( $pluralRuleTypes ) {
4373 return $pluralRuleTypes;
4377 * Find the index number of the plural rule appropriate for the given number
4378 * @return int The index number of the plural rule
4380 public function getPluralRuleIndexNumber( $number ) {
4381 $pluralRules = $this->getCompiledPluralRules();
4382 $form = CLDRPluralRuleEvaluator
::evaluateCompiled( $number, $pluralRules );
4387 * Find the plural rule type appropriate for the given number
4388 * For example, if the language is set to Arabic, getPluralType(5) should
4391 * @return string The name of the plural rule type, e.g. one, two, few, many
4393 public function getPluralRuleType( $number ) {
4394 $index = $this->getPluralRuleIndexNumber( $number );
4395 $pluralRuleTypes = $this->getPluralRuleTypes();
4396 if ( isset( $pluralRuleTypes[$index] ) ) {
4397 return $pluralRuleTypes[$index];