3 * Internationalisation code
10 * @defgroup Language Language
13 if ( !defined( 'MEDIAWIKI' ) ) {
14 echo "This file is part of MediaWiki, it is not a valid entry point.\n";
19 global $wgLanguageNames;
20 require_once( dirname( __FILE__
) . '/Names.php' );
22 if ( function_exists( 'mb_strtoupper' ) ) {
23 mb_internal_encoding( 'UTF-8' );
27 * a fake language converter
37 function __construct( $langobj ) { $this->mLang
= $langobj; }
38 function autoConvertToAllVariants( $text ) { return array( $this->mLang
->getCode() => $text ); }
39 function convert( $t ) { return $t; }
40 function convertTo( $text, $variant ) { return $text; }
41 function convertTitle( $t ) { return $t->getPrefixedText(); }
42 function getVariants() { return array( $this->mLang
->getCode() ); }
43 function getPreferredVariant() { return $this->mLang
->getCode(); }
44 function getDefaultVariant() { return $this->mLang
->getCode(); }
45 function getURLVariant() { return ''; }
46 function getConvRuleTitle() { return false; }
47 function findVariantLink( &$l, &$n, $ignoreOtherCond = false ) { }
48 function getExtraHashOptions() { return ''; }
49 function getParsedTitle() { return ''; }
50 function markNoConversion( $text, $noParse = false ) { return $text; }
51 function convertCategoryKey( $key ) { return $key; }
52 function convertLinkToAllVariants( $text ) { return $this->autoConvertToAllVariants( $text ); }
53 function armourMath( $text ) { return $text; }
57 * Internationalisation code
63 * @var LanguageConverter
67 var $mVariants, $mCode, $mLoaded = false;
68 var $mMagicExtensions = array(), $mMagicHookDone = false;
69 private $mHtmlCode = null;
71 var $dateFormatStrings = array();
72 var $mExtendedSpecialPageAliases;
74 protected $namespaceNames, $mNamespaceIds, $namespaceAliases;
77 * ReplacementArray object caches
79 var $transformData = array();
82 * @var LocalisationCache
84 static public $dataCache;
86 static public $mLangObjCache = array();
88 static public $mWeekdayMsgs = array(
89 'sunday', 'monday', 'tuesday', 'wednesday', 'thursday',
93 static public $mWeekdayAbbrevMsgs = array(
94 'sun', 'mon', 'tue', 'wed', 'thu', 'fri', 'sat'
97 static public $mMonthMsgs = array(
98 'january', 'february', 'march', 'april', 'may_long', 'june',
99 'july', 'august', 'september', 'october', 'november',
102 static public $mMonthGenMsgs = array(
103 'january-gen', 'february-gen', 'march-gen', 'april-gen', 'may-gen', 'june-gen',
104 'july-gen', 'august-gen', 'september-gen', 'october-gen', 'november-gen',
107 static public $mMonthAbbrevMsgs = array(
108 'jan', 'feb', 'mar', 'apr', 'may', 'jun', 'jul', 'aug',
109 'sep', 'oct', 'nov', 'dec'
112 static public $mIranianCalendarMonthMsgs = array(
113 'iranian-calendar-m1', 'iranian-calendar-m2', 'iranian-calendar-m3',
114 'iranian-calendar-m4', 'iranian-calendar-m5', 'iranian-calendar-m6',
115 'iranian-calendar-m7', 'iranian-calendar-m8', 'iranian-calendar-m9',
116 'iranian-calendar-m10', 'iranian-calendar-m11', 'iranian-calendar-m12'
119 static public $mHebrewCalendarMonthMsgs = array(
120 'hebrew-calendar-m1', 'hebrew-calendar-m2', 'hebrew-calendar-m3',
121 'hebrew-calendar-m4', 'hebrew-calendar-m5', 'hebrew-calendar-m6',
122 'hebrew-calendar-m7', 'hebrew-calendar-m8', 'hebrew-calendar-m9',
123 'hebrew-calendar-m10', 'hebrew-calendar-m11', 'hebrew-calendar-m12',
124 'hebrew-calendar-m6a', 'hebrew-calendar-m6b'
127 static public $mHebrewCalendarMonthGenMsgs = array(
128 'hebrew-calendar-m1-gen', 'hebrew-calendar-m2-gen', 'hebrew-calendar-m3-gen',
129 'hebrew-calendar-m4-gen', 'hebrew-calendar-m5-gen', 'hebrew-calendar-m6-gen',
130 'hebrew-calendar-m7-gen', 'hebrew-calendar-m8-gen', 'hebrew-calendar-m9-gen',
131 'hebrew-calendar-m10-gen', 'hebrew-calendar-m11-gen', 'hebrew-calendar-m12-gen',
132 'hebrew-calendar-m6a-gen', 'hebrew-calendar-m6b-gen'
135 static public $mHijriCalendarMonthMsgs = array(
136 'hijri-calendar-m1', 'hijri-calendar-m2', 'hijri-calendar-m3',
137 'hijri-calendar-m4', 'hijri-calendar-m5', 'hijri-calendar-m6',
138 'hijri-calendar-m7', 'hijri-calendar-m8', 'hijri-calendar-m9',
139 'hijri-calendar-m10', 'hijri-calendar-m11', 'hijri-calendar-m12'
146 static public $durationIntervals = array(
147 'millennia' => 31557600000,
148 'centuries' => 3155760000,
149 'decades' => 315576000,
150 'years' => 31557600, // 86400 * 365.25
159 * Get a cached language object for a given language code
160 * @param $code String
163 static function factory( $code ) {
164 if ( !isset( self
::$mLangObjCache[$code] ) ) {
165 if ( count( self
::$mLangObjCache ) > 10 ) {
166 // Don't keep a billion objects around, that's stupid.
167 self
::$mLangObjCache = array();
169 self
::$mLangObjCache[$code] = self
::newFromCode( $code );
171 return self
::$mLangObjCache[$code];
175 * Create a language object for a given language code
176 * @param $code String
177 * @throws MWException
180 protected static function newFromCode( $code ) {
181 // Protect against path traversal below
182 if ( !Language
::isValidCode( $code )
183 ||
strcspn( $code, ":/\\\000" ) !== strlen( $code ) )
185 throw new MWException( "Invalid language code \"$code\"" );
188 if ( !Language
::isValidBuiltInCode( $code ) ) {
189 // It's not possible to customise this code with class files, so
190 // just return a Language object. This is to support uselang= hacks.
191 $lang = new Language
;
192 $lang->setCode( $code );
196 // Check if there is a language class for the code
197 $class = self
::classFromCode( $code );
198 self
::preloadLanguageClass( $class );
199 if ( MWInit
::classExists( $class ) ) {
204 // Keep trying the fallback list until we find an existing class
205 $fallbacks = Language
::getFallbacksFor( $code );
206 foreach ( $fallbacks as $fallbackCode ) {
207 if ( !Language
::isValidBuiltInCode( $fallbackCode ) ) {
208 throw new MWException( "Invalid fallback '$fallbackCode' in fallback sequence for '$code'" );
211 $class = self
::classFromCode( $fallbackCode );
212 self
::preloadLanguageClass( $class );
213 if ( MWInit
::classExists( $class ) ) {
214 $lang = Language
::newFromCode( $fallbackCode );
215 $lang->setCode( $code );
220 throw new MWException( "Invalid fallback sequence for language '$code'" );
224 * Returns true if a language code string is of a valid form, whether or
225 * not it exists. This includes codes which are used solely for
226 * customisation via the MediaWiki namespace.
228 * @param $code string
232 public static function isValidCode( $code ) {
234 strcspn( $code, ":/\\\000" ) === strlen( $code )
235 && !preg_match( Title
::getTitleInvalidRegex(), $code );
239 * Returns true if a language code is of a valid form for the purposes of
240 * internal customisation of MediaWiki, via Messages*.php.
242 * @param $code string
247 public static function isValidBuiltInCode( $code ) {
249 if( !is_string($code) ) {
250 $type = gettype( $code );
251 if( $type === 'object' ) {
252 $addmsg = " of class " . get_class( $code );
256 throw new MWException( __METHOD__
. " must be passed a string, $type given$addmsg" );
259 return preg_match( '/^[a-z0-9-]+$/i', $code );
264 * @return String Name of the language class
266 public static function classFromCode( $code ) {
267 if ( $code == 'en' ) {
270 return 'Language' . str_replace( '-', '_', ucfirst( $code ) );
275 * Includes language class files
277 * @param $class string Name of the language class
279 public static function preloadLanguageClass( $class ) {
282 if ( $class === 'Language' ) {
286 if ( !defined( 'MW_COMPILED' ) ) {
287 if ( file_exists( "$IP/languages/classes/$class.php" ) ) {
288 include_once( "$IP/languages/classes/$class.php" );
294 * Get the LocalisationCache instance
296 * @return LocalisationCache
298 public static function getLocalisationCache() {
299 if ( is_null( self
::$dataCache ) ) {
300 global $wgLocalisationCacheConf;
301 $class = $wgLocalisationCacheConf['class'];
302 self
::$dataCache = new $class( $wgLocalisationCacheConf );
304 return self
::$dataCache;
307 function __construct() {
308 $this->mConverter
= new FakeConverter( $this );
309 // Set the code to the name of the descendant
310 if ( get_class( $this ) == 'Language' ) {
313 $this->mCode
= str_replace( '_', '-', strtolower( substr( get_class( $this ), 8 ) ) );
315 self
::getLocalisationCache();
319 * Reduce memory usage
321 function __destruct() {
322 foreach ( $this as $name => $value ) {
323 unset( $this->$name );
328 * Hook which will be called if this is the content language.
329 * Descendants can use this to register hook functions or modify globals
331 function initContLang() { }
334 * Same as getFallbacksFor for current language.
336 * @deprecated in 1.19
338 function getFallbackLanguageCode() {
339 wfDeprecated( __METHOD__
);
340 return self
::getFallbackFor( $this->mCode
);
347 function getFallbackLanguages() {
348 return self
::getFallbacksFor( $this->mCode
);
352 * Exports $wgBookstoreListEn
355 function getBookstoreList() {
356 return self
::$dataCache->getItem( $this->mCode
, 'bookstoreList' );
362 public function getNamespaces() {
363 if ( is_null( $this->namespaceNames
) ) {
364 global $wgMetaNamespace, $wgMetaNamespaceTalk, $wgExtraNamespaces;
366 $this->namespaceNames
= self
::$dataCache->getItem( $this->mCode
, 'namespaceNames' );
367 $validNamespaces = MWNamespace
::getCanonicalNamespaces();
369 $this->namespaceNames
= $wgExtraNamespaces +
$this->namespaceNames +
$validNamespaces;
371 $this->namespaceNames
[NS_PROJECT
] = $wgMetaNamespace;
372 if ( $wgMetaNamespaceTalk ) {
373 $this->namespaceNames
[NS_PROJECT_TALK
] = $wgMetaNamespaceTalk;
375 $talk = $this->namespaceNames
[NS_PROJECT_TALK
];
376 $this->namespaceNames
[NS_PROJECT_TALK
] =
377 $this->fixVariableInNamespace( $talk );
380 # Sometimes a language will be localised but not actually exist on this wiki.
381 foreach ( $this->namespaceNames
as $key => $text ) {
382 if ( !isset( $validNamespaces[$key] ) ) {
383 unset( $this->namespaceNames
[$key] );
387 # The above mixing may leave namespaces out of canonical order.
388 # Re-order by namespace ID number...
389 ksort( $this->namespaceNames
);
391 wfRunHooks( 'LanguageGetNamespaces', array( &$this->namespaceNames
) );
393 return $this->namespaceNames
;
397 * Arbitrarily set all of the namespace names at once. Mainly used for testing
398 * @param $namespaces Array of namespaces (id => name)
400 public function setNamespaces( array $namespaces ) {
401 $this->namespaceNames
= $namespaces;
405 * A convenience function that returns the same thing as
406 * getNamespaces() except with the array values changed to ' '
407 * where it found '_', useful for producing output to be displayed
408 * e.g. in <select> forms.
412 function getFormattedNamespaces() {
413 $ns = $this->getNamespaces();
414 foreach ( $ns as $k => $v ) {
415 $ns[$k] = strtr( $v, '_', ' ' );
421 * Get a namespace value by key
423 * $mw_ns = $wgContLang->getNsText( NS_MEDIAWIKI );
424 * echo $mw_ns; // prints 'MediaWiki'
427 * @param $index Int: the array key of the namespace to return
428 * @return mixed, string if the namespace value exists, otherwise false
430 function getNsText( $index ) {
431 $ns = $this->getNamespaces();
432 return isset( $ns[$index] ) ?
$ns[$index] : false;
436 * A convenience function that returns the same thing as
437 * getNsText() except with '_' changed to ' ', useful for
440 * @param $index string
444 function getFormattedNsText( $index ) {
445 $ns = $this->getNsText( $index );
446 return strtr( $ns, '_', ' ' );
450 * Returns gender-dependent namespace alias if available.
451 * @param $index Int: namespace index
452 * @param $gender String: gender key (male, female... )
456 function getGenderNsText( $index, $gender ) {
457 global $wgExtraGenderNamespaces;
459 $ns = $wgExtraGenderNamespaces + self
::$dataCache->getItem( $this->mCode
, 'namespaceGenderAliases' );
460 return isset( $ns[$index][$gender] ) ?
$ns[$index][$gender] : $this->getNsText( $index );
464 * Whether this language makes distinguishes genders for example in
469 function needsGenderDistinction() {
470 global $wgExtraGenderNamespaces, $wgExtraNamespaces;
471 if ( count( $wgExtraGenderNamespaces ) > 0 ) {
472 // $wgExtraGenderNamespaces overrides everything
474 } elseif ( isset( $wgExtraNamespaces[NS_USER
] ) && isset( $wgExtraNamespaces[NS_USER_TALK
] ) ) {
475 /// @todo There may be other gender namespace than NS_USER & NS_USER_TALK in the future
476 // $wgExtraNamespaces overrides any gender aliases specified in i18n files
479 // Check what is in i18n files
480 $aliases = self
::$dataCache->getItem( $this->mCode
, 'namespaceGenderAliases' );
481 return count( $aliases ) > 0;
486 * Get a namespace key by value, case insensitive.
487 * Only matches namespace names for the current language, not the
488 * canonical ones defined in Namespace.php.
490 * @param $text String
491 * @return mixed An integer if $text is a valid value otherwise false
493 function getLocalNsIndex( $text ) {
494 $lctext = $this->lc( $text );
495 $ids = $this->getNamespaceIds();
496 return isset( $ids[$lctext] ) ?
$ids[$lctext] : false;
502 function getNamespaceAliases() {
503 if ( is_null( $this->namespaceAliases
) ) {
504 $aliases = self
::$dataCache->getItem( $this->mCode
, 'namespaceAliases' );
508 foreach ( $aliases as $name => $index ) {
509 if ( $index === NS_PROJECT_TALK
) {
510 unset( $aliases[$name] );
511 $name = $this->fixVariableInNamespace( $name );
512 $aliases[$name] = $index;
517 global $wgExtraGenderNamespaces;
518 $genders = $wgExtraGenderNamespaces +
(array)self
::$dataCache->getItem( $this->mCode
, 'namespaceGenderAliases' );
519 foreach ( $genders as $index => $forms ) {
520 foreach ( $forms as $alias ) {
521 $aliases[$alias] = $index;
525 $this->namespaceAliases
= $aliases;
527 return $this->namespaceAliases
;
533 function getNamespaceIds() {
534 if ( is_null( $this->mNamespaceIds
) ) {
535 global $wgNamespaceAliases;
536 # Put namespace names and aliases into a hashtable.
537 # If this is too slow, then we should arrange it so that it is done
538 # before caching. The catch is that at pre-cache time, the above
539 # class-specific fixup hasn't been done.
540 $this->mNamespaceIds
= array();
541 foreach ( $this->getNamespaces() as $index => $name ) {
542 $this->mNamespaceIds
[$this->lc( $name )] = $index;
544 foreach ( $this->getNamespaceAliases() as $name => $index ) {
545 $this->mNamespaceIds
[$this->lc( $name )] = $index;
547 if ( $wgNamespaceAliases ) {
548 foreach ( $wgNamespaceAliases as $name => $index ) {
549 $this->mNamespaceIds
[$this->lc( $name )] = $index;
553 return $this->mNamespaceIds
;
557 * Get a namespace key by value, case insensitive. Canonical namespace
558 * names override custom ones defined for the current language.
560 * @param $text String
561 * @return mixed An integer if $text is a valid value otherwise false
563 function getNsIndex( $text ) {
564 $lctext = $this->lc( $text );
565 $ns = MWNamespace
::getCanonicalIndex( $lctext );
566 if ( $ns !== null ) {
569 $ids = $this->getNamespaceIds();
570 return isset( $ids[$lctext] ) ?
$ids[$lctext] : false;
574 * short names for language variants used for language conversion links.
576 * @param $code String
577 * @param $usemsg bool Use the "variantname-xyz" message if it exists
580 function getVariantname( $code, $usemsg = true ) {
581 $msg = "variantname-$code";
582 if ( $usemsg && wfMessage( $msg )->exists() ) {
583 return $this->getMessageFromDB( $msg );
585 $name = self
::fetchLanguageName( $code );
587 return $name; # if it's defined as a language name, show that
589 # otherwise, output the language code
595 * @param $name string
598 function specialPage( $name ) {
599 $aliases = $this->getSpecialPageAliases();
600 if ( isset( $aliases[$name][0] ) ) {
601 $name = $aliases[$name][0];
603 return $this->getNsText( NS_SPECIAL
) . ':' . $name;
609 function getQuickbarSettings() {
611 $this->getMessage( 'qbsettings-none' ),
612 $this->getMessage( 'qbsettings-fixedleft' ),
613 $this->getMessage( 'qbsettings-fixedright' ),
614 $this->getMessage( 'qbsettings-floatingleft' ),
615 $this->getMessage( 'qbsettings-floatingright' ),
616 $this->getMessage( 'qbsettings-directionality' )
623 function getDatePreferences() {
624 return self
::$dataCache->getItem( $this->mCode
, 'datePreferences' );
630 function getDateFormats() {
631 return self
::$dataCache->getItem( $this->mCode
, 'dateFormats' );
635 * @return array|string
637 function getDefaultDateFormat() {
638 $df = self
::$dataCache->getItem( $this->mCode
, 'defaultDateFormat' );
639 if ( $df === 'dmy or mdy' ) {
640 global $wgAmericanDates;
641 return $wgAmericanDates ?
'mdy' : 'dmy';
650 function getDatePreferenceMigrationMap() {
651 return self
::$dataCache->getItem( $this->mCode
, 'datePreferenceMigrationMap' );
658 function getImageFile( $image ) {
659 return self
::$dataCache->getSubitem( $this->mCode
, 'imageFiles', $image );
665 function getExtraUserToggles() {
666 return (array)self
::$dataCache->getItem( $this->mCode
, 'extraUserToggles' );
673 function getUserToggle( $tog ) {
674 return $this->getMessageFromDB( "tog-$tog" );
678 * Get native language names, indexed by code.
679 * Only those defined in MediaWiki, no other data like CLDR.
680 * If $customisedOnly is true, only returns codes with a messages file
682 * @param $customisedOnly bool
685 * @deprecated in 1.20, use fetchLanguageNames()
687 public static function getLanguageNames( $customisedOnly = false ) {
688 return self
::fetchLanguageNames( null, $customisedOnly ?
'mwfile' : 'mw' );
692 * Get translated language names. This is done on best effort and
693 * by default this is exactly the same as Language::getLanguageNames.
694 * The CLDR extension provides translated names.
695 * @param $code String Language code.
696 * @return Array language code => language name
698 * @deprecated in 1.20, use fetchLanguageNames()
700 public static function getTranslatedLanguageNames( $code ) {
701 return self
::fetchLanguageNames( $code, 'all' );
705 * Get an array of language names, indexed by code.
706 * @param $inLanguage null|string: Code of language in which to return the names
707 * Use null for autonyms (native names)
708 * @param $include string:
709 * 'all' all available languages
710 * 'mw' only if the language is defined in MediaWiki or wgExtraLanguageNames (default)
711 * 'mwfile' only if the language is in 'mw' *and* has a message file
712 * @return array: language code => language name
715 public static function fetchLanguageNames( $inLanguage = null, $include = 'mw' ) {
716 global $wgExtraLanguageNames;
717 static $coreLanguageNames;
719 if ( $coreLanguageNames === null ) {
720 include( MWInit
::compiledPath( 'languages/Names.php' ) );
726 # TODO: also include when $inLanguage is null, when this code is more efficient
727 wfRunHooks( 'LanguageGetTranslatedLanguageNames', array( &$names, $inLanguage ) );
730 $mwNames = $wgExtraLanguageNames +
$coreLanguageNames;
731 foreach ( $mwNames as $mwCode => $mwName ) {
732 # - Prefer own MediaWiki native name when not using the hook
733 # - For other names just add if not added through the hook
734 if ( $mwCode === $inLanguage ||
!isset( $names[$mwCode] ) ) {
735 $names[$mwCode] = $mwName;
739 if ( $include === 'all' ) {
744 $coreCodes = array_keys( $mwNames );
745 foreach( $coreCodes as $coreCode ) {
746 $returnMw[$coreCode] = $names[$coreCode];
749 if( $include === 'mwfile' ) {
750 $namesMwFile = array();
751 # We do this using a foreach over the codes instead of a directory
752 # loop so that messages files in extensions will work correctly.
753 foreach ( $returnMw as $code => $value ) {
754 if ( is_readable( self
::getMessagesFileName( $code ) ) ) {
755 $namesMwFile[$code] = $names[$code];
760 # 'mw' option; default if it's not one of the other two options (all/mwfile)
765 * @param $code string: The code of the language for which to get the name
766 * @param $inLanguage null|string: Code of language in which to return the name (null for autonyms)
767 * @param $include string: 'all', 'mw' or 'mwfile'; see fetchLanguageNames()
768 * @return string: Language name or empty
771 public static function fetchLanguageName( $code, $inLanguage = null, $include = 'all' ) {
772 $array = self
::fetchLanguageNames( $inLanguage, $include );
773 return !array_key_exists( $code, $array ) ?
'' : $array[$code];
777 * Get a message from the MediaWiki namespace.
779 * @param $msg String: message name
782 function getMessageFromDB( $msg ) {
783 return wfMsgExt( $msg, array( 'parsemag', 'language' => $this ) );
787 * Get the native language name of $code.
788 * Only if defined in MediaWiki, no other data like CLDR.
789 * @param $code string
791 * @deprecated in 1.20, use fetchLanguageName()
793 function getLanguageName( $code ) {
794 return self
::fetchLanguageName( $code );
801 function getMonthName( $key ) {
802 return $this->getMessageFromDB( self
::$mMonthMsgs[$key - 1] );
808 function getMonthNamesArray() {
809 $monthNames = array( '' );
810 for ( $i = 1; $i < 13; $i++
) {
811 $monthNames[] = $this->getMonthName( $i );
820 function getMonthNameGen( $key ) {
821 return $this->getMessageFromDB( self
::$mMonthGenMsgs[$key - 1] );
828 function getMonthAbbreviation( $key ) {
829 return $this->getMessageFromDB( self
::$mMonthAbbrevMsgs[$key - 1] );
835 function getMonthAbbreviationsArray() {
836 $monthNames = array( '' );
837 for ( $i = 1; $i < 13; $i++
) {
838 $monthNames[] = $this->getMonthAbbreviation( $i );
847 function getWeekdayName( $key ) {
848 return $this->getMessageFromDB( self
::$mWeekdayMsgs[$key - 1] );
855 function getWeekdayAbbreviation( $key ) {
856 return $this->getMessageFromDB( self
::$mWeekdayAbbrevMsgs[$key - 1] );
863 function getIranianCalendarMonthName( $key ) {
864 return $this->getMessageFromDB( self
::$mIranianCalendarMonthMsgs[$key - 1] );
871 function getHebrewCalendarMonthName( $key ) {
872 return $this->getMessageFromDB( self
::$mHebrewCalendarMonthMsgs[$key - 1] );
879 function getHebrewCalendarMonthNameGen( $key ) {
880 return $this->getMessageFromDB( self
::$mHebrewCalendarMonthGenMsgs[$key - 1] );
887 function getHijriCalendarMonthName( $key ) {
888 return $this->getMessageFromDB( self
::$mHijriCalendarMonthMsgs[$key - 1] );
892 * This is a workalike of PHP's date() function, but with better
893 * internationalisation, a reduced set of format characters, and a better
896 * Supported format characters are dDjlNwzWFmMntLoYyaAgGhHiscrU. See the
897 * PHP manual for definitions. There are a number of extensions, which
900 * xn Do not translate digits of the next numeric format character
901 * xN Toggle raw digit (xn) flag, stays set until explicitly unset
902 * xr Use roman numerals for the next numeric format character
903 * xh Use hebrew numerals for the next numeric format character
905 * xg Genitive month name
907 * xij j (day number) in Iranian calendar
908 * xiF F (month name) in Iranian calendar
909 * xin n (month number) in Iranian calendar
910 * xiy y (two digit year) in Iranian calendar
911 * xiY Y (full year) in Iranian calendar
913 * xjj j (day number) in Hebrew calendar
914 * xjF F (month name) in Hebrew calendar
915 * xjt t (days in month) in Hebrew calendar
916 * xjx xg (genitive month name) in Hebrew calendar
917 * xjn n (month number) in Hebrew calendar
918 * xjY Y (full year) in Hebrew calendar
920 * xmj j (day number) in Hijri calendar
921 * xmF F (month name) in Hijri calendar
922 * xmn n (month number) in Hijri calendar
923 * xmY Y (full year) in Hijri calendar
925 * xkY Y (full year) in Thai solar calendar. Months and days are
926 * identical to the Gregorian calendar
927 * xoY Y (full year) in Minguo calendar or Juche year.
928 * Months and days are identical to the
930 * xtY Y (full year) in Japanese nengo. Months and days are
931 * identical to the Gregorian calendar
933 * Characters enclosed in double quotes will be considered literal (with
934 * the quotes themselves removed). Unmatched quotes will be considered
935 * literal quotes. Example:
937 * "The month is" F => The month is January
940 * Backslash escaping is also supported.
942 * Input timestamp is assumed to be pre-normalized to the desired local
945 * @param $format String
946 * @param $ts String: 14-character timestamp
949 * @todo handling of "o" format character for Iranian, Hebrew, Hijri & Thai?
953 function sprintfDate( $format, $ts ) {
966 for ( $p = 0; $p < strlen( $format ); $p++
) {
969 if ( $code == 'x' && $p < strlen( $format ) - 1 ) {
970 $code .= $format[++
$p];
973 if ( ( $code === 'xi' ||
$code == 'xj' ||
$code == 'xk' ||
$code == 'xm' ||
$code == 'xo' ||
$code == 'xt' ) && $p < strlen( $format ) - 1 ) {
974 $code .= $format[++
$p];
985 $rawToggle = !$rawToggle;
994 $s .= $this->getMonthNameGen( substr( $ts, 4, 2 ) );
997 if ( !$hebrew ) $hebrew = self
::tsToHebrew( $ts );
998 $s .= $this->getHebrewCalendarMonthNameGen( $hebrew[1] );
1001 $num = substr( $ts, 6, 2 );
1004 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
1005 $s .= $this->getWeekdayAbbreviation( gmdate( 'w', $unix ) +
1 );
1008 $num = intval( substr( $ts, 6, 2 ) );
1012 $iranian = self
::tsToIranian( $ts );
1018 $hijri = self
::tsToHijri( $ts );
1024 $hebrew = self
::tsToHebrew( $ts );
1030 $unix = wfTimestamp( TS_UNIX
, $ts );
1032 $s .= $this->getWeekdayName( gmdate( 'w', $unix ) +
1 );
1036 $unix = wfTimestamp( TS_UNIX
, $ts );
1038 $w = gmdate( 'w', $unix );
1043 $unix = wfTimestamp( TS_UNIX
, $ts );
1045 $num = gmdate( 'w', $unix );
1049 $unix = wfTimestamp( TS_UNIX
, $ts );
1051 $num = gmdate( 'z', $unix );
1055 $unix = wfTimestamp( TS_UNIX
, $ts );
1057 $num = gmdate( 'W', $unix );
1060 $s .= $this->getMonthName( substr( $ts, 4, 2 ) );
1064 $iranian = self
::tsToIranian( $ts );
1066 $s .= $this->getIranianCalendarMonthName( $iranian[1] );
1070 $hijri = self
::tsToHijri( $ts );
1072 $s .= $this->getHijriCalendarMonthName( $hijri[1] );
1076 $hebrew = self
::tsToHebrew( $ts );
1078 $s .= $this->getHebrewCalendarMonthName( $hebrew[1] );
1081 $num = substr( $ts, 4, 2 );
1084 $s .= $this->getMonthAbbreviation( substr( $ts, 4, 2 ) );
1087 $num = intval( substr( $ts, 4, 2 ) );
1091 $iranian = self
::tsToIranian( $ts );
1097 $hijri = self
::tsToHijri ( $ts );
1103 $hebrew = self
::tsToHebrew( $ts );
1109 $unix = wfTimestamp( TS_UNIX
, $ts );
1111 $num = gmdate( 't', $unix );
1115 $hebrew = self
::tsToHebrew( $ts );
1121 $unix = wfTimestamp( TS_UNIX
, $ts );
1123 $num = gmdate( 'L', $unix );
1127 $unix = wfTimestamp( TS_UNIX
, $ts );
1129 $num = gmdate( 'o', $unix );
1132 $num = substr( $ts, 0, 4 );
1136 $iranian = self
::tsToIranian( $ts );
1142 $hijri = self
::tsToHijri( $ts );
1148 $hebrew = self
::tsToHebrew( $ts );
1154 $thai = self
::tsToYear( $ts, 'thai' );
1160 $minguo = self
::tsToYear( $ts, 'minguo' );
1166 $tenno = self
::tsToYear( $ts, 'tenno' );
1171 $num = substr( $ts, 2, 2 );
1175 $iranian = self
::tsToIranian( $ts );
1177 $num = substr( $iranian[0], -2 );
1180 $s .= intval( substr( $ts, 8, 2 ) ) < 12 ?
'am' : 'pm';
1183 $s .= intval( substr( $ts, 8, 2 ) ) < 12 ?
'AM' : 'PM';
1186 $h = substr( $ts, 8, 2 );
1187 $num = $h %
12 ?
$h %
12 : 12;
1190 $num = intval( substr( $ts, 8, 2 ) );
1193 $h = substr( $ts, 8, 2 );
1194 $num = sprintf( '%02d', $h %
12 ?
$h %
12 : 12 );
1197 $num = substr( $ts, 8, 2 );
1200 $num = substr( $ts, 10, 2 );
1203 $num = substr( $ts, 12, 2 );
1207 $unix = wfTimestamp( TS_UNIX
, $ts );
1209 $s .= gmdate( 'c', $unix );
1213 $unix = wfTimestamp( TS_UNIX
, $ts );
1215 $s .= gmdate( 'r', $unix );
1219 $unix = wfTimestamp( TS_UNIX
, $ts );
1224 # Backslash escaping
1225 if ( $p < strlen( $format ) - 1 ) {
1226 $s .= $format[++
$p];
1233 if ( $p < strlen( $format ) - 1 ) {
1234 $endQuote = strpos( $format, '"', $p +
1 );
1235 if ( $endQuote === false ) {
1236 # No terminating quote, assume literal "
1239 $s .= substr( $format, $p +
1, $endQuote - $p - 1 );
1243 # Quote at end of string, assume literal "
1250 if ( $num !== false ) {
1251 if ( $rawToggle ||
$raw ) {
1254 } elseif ( $roman ) {
1255 $s .= self
::romanNumeral( $num );
1257 } elseif ( $hebrewNum ) {
1258 $s .= self
::hebrewNumeral( $num );
1261 $s .= $this->formatNum( $num, true );
1268 private static $GREG_DAYS = array( 31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31 );
1269 private static $IRANIAN_DAYS = array( 31, 31, 31, 31, 31, 31, 30, 30, 30, 30, 30, 29 );
1272 * Algorithm by Roozbeh Pournader and Mohammad Toossi to convert
1273 * Gregorian dates to Iranian dates. Originally written in C, it
1274 * is released under the terms of GNU Lesser General Public
1275 * License. Conversion to PHP was performed by Niklas Laxström.
1277 * Link: http://www.farsiweb.info/jalali/jalali.c
1283 private static function tsToIranian( $ts ) {
1284 $gy = substr( $ts, 0, 4 ) -1600;
1285 $gm = substr( $ts, 4, 2 ) -1;
1286 $gd = substr( $ts, 6, 2 ) -1;
1288 # Days passed from the beginning (including leap years)
1290 +
floor( ( $gy +
3 ) / 4 )
1291 - floor( ( $gy +
99 ) / 100 )
1292 +
floor( ( $gy +
399 ) / 400 );
1294 // Add days of the past months of this year
1295 for ( $i = 0; $i < $gm; $i++
) {
1296 $gDayNo +
= self
::$GREG_DAYS[$i];
1300 if ( $gm > 1 && ( ( $gy %
4 === 0 && $gy %
100 !== 0 ||
( $gy %
400 == 0 ) ) ) ) {
1304 // Days passed in current month
1305 $gDayNo +
= (int)$gd;
1307 $jDayNo = $gDayNo - 79;
1309 $jNp = floor( $jDayNo / 12053 );
1312 $jy = 979 +
33 * $jNp +
4 * floor( $jDayNo / 1461 );
1315 if ( $jDayNo >= 366 ) {
1316 $jy +
= floor( ( $jDayNo - 1 ) / 365 );
1317 $jDayNo = floor( ( $jDayNo - 1 ) %
365 );
1320 for ( $i = 0; $i < 11 && $jDayNo >= self
::$IRANIAN_DAYS[$i]; $i++
) {
1321 $jDayNo -= self
::$IRANIAN_DAYS[$i];
1327 return array( $jy, $jm, $jd );
1331 * Converting Gregorian dates to Hijri dates.
1333 * Based on a PHP-Nuke block by Sharjeel which is released under GNU/GPL license
1335 * @see http://phpnuke.org/modules.php?name=News&file=article&sid=8234&mode=thread&order=0&thold=0
1341 private static function tsToHijri( $ts ) {
1342 $year = substr( $ts, 0, 4 );
1343 $month = substr( $ts, 4, 2 );
1344 $day = substr( $ts, 6, 2 );
1352 ( $zy > 1582 ) ||
( ( $zy == 1582 ) && ( $zm > 10 ) ) ||
1353 ( ( $zy == 1582 ) && ( $zm == 10 ) && ( $zd > 14 ) )
1356 $zjd = (int)( ( 1461 * ( $zy +
4800 +
(int)( ( $zm - 14 ) / 12 ) ) ) / 4 ) +
1357 (int)( ( 367 * ( $zm - 2 - 12 * ( (int)( ( $zm - 14 ) / 12 ) ) ) ) / 12 ) -
1358 (int)( ( 3 * (int)( ( ( $zy +
4900 +
(int)( ( $zm - 14 ) / 12 ) ) / 100 ) ) ) / 4 ) +
1361 $zjd = 367 * $zy - (int)( ( 7 * ( $zy +
5001 +
(int)( ( $zm - 9 ) / 7 ) ) ) / 4 ) +
1362 (int)( ( 275 * $zm ) / 9 ) +
$zd +
1729777;
1365 $zl = $zjd -1948440 +
10632;
1366 $zn = (int)( ( $zl - 1 ) / 10631 );
1367 $zl = $zl - 10631 * $zn +
354;
1368 $zj = ( (int)( ( 10985 - $zl ) / 5316 ) ) * ( (int)( ( 50 * $zl ) / 17719 ) ) +
( (int)( $zl / 5670 ) ) * ( (int)( ( 43 * $zl ) / 15238 ) );
1369 $zl = $zl - ( (int)( ( 30 - $zj ) / 15 ) ) * ( (int)( ( 17719 * $zj ) / 50 ) ) - ( (int)( $zj / 16 ) ) * ( (int)( ( 15238 * $zj ) / 43 ) ) +
29;
1370 $zm = (int)( ( 24 * $zl ) / 709 );
1371 $zd = $zl - (int)( ( 709 * $zm ) / 24 );
1372 $zy = 30 * $zn +
$zj - 30;
1374 return array( $zy, $zm, $zd );
1378 * Converting Gregorian dates to Hebrew dates.
1380 * Based on a JavaScript code by Abu Mami and Yisrael Hersch
1381 * (abu-mami@kaluach.net, http://www.kaluach.net), who permitted
1382 * to translate the relevant functions into PHP and release them under
1385 * The months are counted from Tishrei = 1. In a leap year, Adar I is 13
1386 * and Adar II is 14. In a non-leap year, Adar is 6.
1392 private static function tsToHebrew( $ts ) {
1394 $year = substr( $ts, 0, 4 );
1395 $month = substr( $ts, 4, 2 );
1396 $day = substr( $ts, 6, 2 );
1398 # Calculate Hebrew year
1399 $hebrewYear = $year +
3760;
1401 # Month number when September = 1, August = 12
1403 if ( $month > 12 ) {
1410 # Calculate day of year from 1 September
1412 for ( $i = 1; $i < $month; $i++
) {
1416 # Check if the year is leap
1417 if ( $year %
400 == 0 ||
( $year %
4 == 0 && $year %
100 > 0 ) ) {
1420 } elseif ( $i == 8 ||
$i == 10 ||
$i == 1 ||
$i == 3 ) {
1427 # Calculate the start of the Hebrew year
1428 $start = self
::hebrewYearStart( $hebrewYear );
1430 # Calculate next year's start
1431 if ( $dayOfYear <= $start ) {
1432 # Day is before the start of the year - it is the previous year
1434 $nextStart = $start;
1438 # Add days since previous year's 1 September
1440 if ( ( $year %
400 == 0 ) ||
( $year %
100 != 0 && $year %
4 == 0 ) ) {
1444 # Start of the new (previous) year
1445 $start = self
::hebrewYearStart( $hebrewYear );
1448 $nextStart = self
::hebrewYearStart( $hebrewYear +
1 );
1451 # Calculate Hebrew day of year
1452 $hebrewDayOfYear = $dayOfYear - $start;
1454 # Difference between year's days
1455 $diff = $nextStart - $start;
1456 # Add 12 (or 13 for leap years) days to ignore the difference between
1457 # Hebrew and Gregorian year (353 at least vs. 365/6) - now the
1458 # difference is only about the year type
1459 if ( ( $year %
400 == 0 ) ||
( $year %
100 != 0 && $year %
4 == 0 ) ) {
1465 # Check the year pattern, and is leap year
1466 # 0 means an incomplete year, 1 means a regular year, 2 means a complete year
1467 # This is mod 30, to work on both leap years (which add 30 days of Adar I)
1468 # and non-leap years
1469 $yearPattern = $diff %
30;
1470 # Check if leap year
1471 $isLeap = $diff >= 30;
1473 # Calculate day in the month from number of day in the Hebrew year
1474 # Don't check Adar - if the day is not in Adar, we will stop before;
1475 # if it is in Adar, we will use it to check if it is Adar I or Adar II
1476 $hebrewDay = $hebrewDayOfYear;
1479 while ( $hebrewMonth <= 12 ) {
1480 # Calculate days in this month
1481 if ( $isLeap && $hebrewMonth == 6 ) {
1482 # Adar in a leap year
1484 # Leap year - has Adar I, with 30 days, and Adar II, with 29 days
1486 if ( $hebrewDay <= $days ) {
1490 # Subtract the days of Adar I
1491 $hebrewDay -= $days;
1494 if ( $hebrewDay <= $days ) {
1500 } elseif ( $hebrewMonth == 2 && $yearPattern == 2 ) {
1501 # Cheshvan in a complete year (otherwise as the rule below)
1503 } elseif ( $hebrewMonth == 3 && $yearPattern == 0 ) {
1504 # Kislev in an incomplete year (otherwise as the rule below)
1507 # Odd months have 30 days, even have 29
1508 $days = 30 - ( $hebrewMonth - 1 ) %
2;
1510 if ( $hebrewDay <= $days ) {
1511 # In the current month
1514 # Subtract the days of the current month
1515 $hebrewDay -= $days;
1516 # Try in the next month
1521 return array( $hebrewYear, $hebrewMonth, $hebrewDay, $days );
1525 * This calculates the Hebrew year start, as days since 1 September.
1526 * Based on Carl Friedrich Gauss algorithm for finding Easter date.
1527 * Used for Hebrew date.
1533 private static function hebrewYearStart( $year ) {
1534 $a = intval( ( 12 * ( $year - 1 ) +
17 ) %
19 );
1535 $b = intval( ( $year - 1 ) %
4 );
1536 $m = 32.044093161144 +
1.5542417966212 * $a +
$b / 4.0 - 0.0031777940220923 * ( $year - 1 );
1540 $Mar = intval( $m );
1546 $c = intval( ( $Mar +
3 * ( $year - 1 ) +
5 * $b +
5 ) %
7 );
1547 if ( $c == 0 && $a > 11 && $m >= 0.89772376543210 ) {
1549 } elseif ( $c == 1 && $a > 6 && $m >= 0.63287037037037 ) {
1551 } elseif ( $c == 2 ||
$c == 4 ||
$c == 6 ) {
1555 $Mar +
= intval( ( $year - 3761 ) / 100 ) - intval( ( $year - 3761 ) / 400 ) - 24;
1560 * Algorithm to convert Gregorian dates to Thai solar dates,
1561 * Minguo dates or Minguo dates.
1563 * Link: http://en.wikipedia.org/wiki/Thai_solar_calendar
1564 * http://en.wikipedia.org/wiki/Minguo_calendar
1565 * http://en.wikipedia.org/wiki/Japanese_era_name
1567 * @param $ts String: 14-character timestamp
1568 * @param $cName String: calender name
1569 * @return Array: converted year, month, day
1571 private static function tsToYear( $ts, $cName ) {
1572 $gy = substr( $ts, 0, 4 );
1573 $gm = substr( $ts, 4, 2 );
1574 $gd = substr( $ts, 6, 2 );
1576 if ( !strcmp( $cName, 'thai' ) ) {
1578 # Add 543 years to the Gregorian calendar
1579 # Months and days are identical
1580 $gy_offset = $gy +
543;
1581 } elseif ( ( !strcmp( $cName, 'minguo' ) ) ||
!strcmp( $cName, 'juche' ) ) {
1583 # Deduct 1911 years from the Gregorian calendar
1584 # Months and days are identical
1585 $gy_offset = $gy - 1911;
1586 } elseif ( !strcmp( $cName, 'tenno' ) ) {
1587 # Nengō dates up to Meiji period
1588 # Deduct years from the Gregorian calendar
1589 # depending on the nengo periods
1590 # Months and days are identical
1591 if ( ( $gy < 1912 ) ||
( ( $gy == 1912 ) && ( $gm < 7 ) ) ||
( ( $gy == 1912 ) && ( $gm == 7 ) && ( $gd < 31 ) ) ) {
1593 $gy_gannen = $gy - 1868 +
1;
1594 $gy_offset = $gy_gannen;
1595 if ( $gy_gannen == 1 ) {
1598 $gy_offset = '明治' . $gy_offset;
1600 ( ( $gy == 1912 ) && ( $gm == 7 ) && ( $gd == 31 ) ) ||
1601 ( ( $gy == 1912 ) && ( $gm >= 8 ) ) ||
1602 ( ( $gy > 1912 ) && ( $gy < 1926 ) ) ||
1603 ( ( $gy == 1926 ) && ( $gm < 12 ) ) ||
1604 ( ( $gy == 1926 ) && ( $gm == 12 ) && ( $gd < 26 ) )
1608 $gy_gannen = $gy - 1912 +
1;
1609 $gy_offset = $gy_gannen;
1610 if ( $gy_gannen == 1 ) {
1613 $gy_offset = '大正' . $gy_offset;
1615 ( ( $gy == 1926 ) && ( $gm == 12 ) && ( $gd >= 26 ) ) ||
1616 ( ( $gy > 1926 ) && ( $gy < 1989 ) ) ||
1617 ( ( $gy == 1989 ) && ( $gm == 1 ) && ( $gd < 8 ) )
1621 $gy_gannen = $gy - 1926 +
1;
1622 $gy_offset = $gy_gannen;
1623 if ( $gy_gannen == 1 ) {
1626 $gy_offset = '昭和' . $gy_offset;
1629 $gy_gannen = $gy - 1989 +
1;
1630 $gy_offset = $gy_gannen;
1631 if ( $gy_gannen == 1 ) {
1634 $gy_offset = '平成' . $gy_offset;
1640 return array( $gy_offset, $gm, $gd );
1644 * Roman number formatting up to 3000
1650 static function romanNumeral( $num ) {
1651 static $table = array(
1652 array( '', 'I', 'II', 'III', 'IV', 'V', 'VI', 'VII', 'VIII', 'IX', 'X' ),
1653 array( '', 'X', 'XX', 'XXX', 'XL', 'L', 'LX', 'LXX', 'LXXX', 'XC', 'C' ),
1654 array( '', 'C', 'CC', 'CCC', 'CD', 'D', 'DC', 'DCC', 'DCCC', 'CM', 'M' ),
1655 array( '', 'M', 'MM', 'MMM' )
1658 $num = intval( $num );
1659 if ( $num > 3000 ||
$num <= 0 ) {
1664 for ( $pow10 = 1000, $i = 3; $i >= 0; $pow10 /= 10, $i-- ) {
1665 if ( $num >= $pow10 ) {
1666 $s .= $table[$i][(int)floor( $num / $pow10 )];
1668 $num = $num %
$pow10;
1674 * Hebrew Gematria number formatting up to 9999
1680 static function hebrewNumeral( $num ) {
1681 static $table = array(
1682 array( '', 'א', 'ב', 'ג', 'ד', 'ה', 'ו', 'ז', 'ח', 'ט', 'י' ),
1683 array( '', 'י', 'כ', 'ל', 'מ', 'נ', 'ס', 'ע', 'פ', 'צ', 'ק' ),
1684 array( '', 'ק', 'ר', 'ש', 'ת', 'תק', 'תר', 'תש', 'תת', 'תתק', 'תתר' ),
1685 array( '', 'א', 'ב', 'ג', 'ד', 'ה', 'ו', 'ז', 'ח', 'ט', 'י' )
1688 $num = intval( $num );
1689 if ( $num > 9999 ||
$num <= 0 ) {
1694 for ( $pow10 = 1000, $i = 3; $i >= 0; $pow10 /= 10, $i-- ) {
1695 if ( $num >= $pow10 ) {
1696 if ( $num == 15 ||
$num == 16 ) {
1697 $s .= $table[0][9] . $table[0][$num - 9];
1700 $s .= $table[$i][intval( ( $num / $pow10 ) )];
1701 if ( $pow10 == 1000 ) {
1706 $num = $num %
$pow10;
1708 if ( strlen( $s ) == 2 ) {
1711 $str = substr( $s, 0, strlen( $s ) - 2 ) . '"';
1712 $str .= substr( $s, strlen( $s ) - 2, 2 );
1714 $start = substr( $str, 0, strlen( $str ) - 2 );
1715 $end = substr( $str, strlen( $str ) - 2 );
1718 $str = $start . 'ך';
1721 $str = $start . 'ם';
1724 $str = $start . 'ן';
1727 $str = $start . 'ף';
1730 $str = $start . 'ץ';
1737 * Used by date() and time() to adjust the time output.
1739 * @param $ts Int the time in date('YmdHis') format
1740 * @param $tz Mixed: adjust the time by this amount (default false, mean we
1741 * get user timecorrection setting)
1744 function userAdjust( $ts, $tz = false ) {
1745 global $wgUser, $wgLocalTZoffset;
1747 if ( $tz === false ) {
1748 $tz = $wgUser->getOption( 'timecorrection' );
1751 $data = explode( '|', $tz, 3 );
1753 if ( $data[0] == 'ZoneInfo' ) {
1754 wfSuppressWarnings();
1755 $userTZ = timezone_open( $data[2] );
1756 wfRestoreWarnings();
1757 if ( $userTZ !== false ) {
1758 $date = date_create( $ts, timezone_open( 'UTC' ) );
1759 date_timezone_set( $date, $userTZ );
1760 $date = date_format( $date, 'YmdHis' );
1763 # Unrecognized timezone, default to 'Offset' with the stored offset.
1764 $data[0] = 'Offset';
1768 if ( $data[0] == 'System' ||
$tz == '' ) {
1769 # Global offset in minutes.
1770 if ( isset( $wgLocalTZoffset ) ) {
1771 $minDiff = $wgLocalTZoffset;
1773 } elseif ( $data[0] == 'Offset' ) {
1774 $minDiff = intval( $data[1] );
1776 $data = explode( ':', $tz );
1777 if ( count( $data ) == 2 ) {
1778 $data[0] = intval( $data[0] );
1779 $data[1] = intval( $data[1] );
1780 $minDiff = abs( $data[0] ) * 60 +
$data[1];
1781 if ( $data[0] < 0 ) {
1782 $minDiff = -$minDiff;
1785 $minDiff = intval( $data[0] ) * 60;
1789 # No difference ? Return time unchanged
1790 if ( 0 == $minDiff ) {
1794 wfSuppressWarnings(); // E_STRICT system time bitching
1795 # Generate an adjusted date; take advantage of the fact that mktime
1796 # will normalize out-of-range values so we don't have to split $minDiff
1797 # into hours and minutes.
1799 (int)substr( $ts, 8, 2 ) ), # Hours
1800 (int)substr( $ts, 10, 2 ) +
$minDiff, # Minutes
1801 (int)substr( $ts, 12, 2 ), # Seconds
1802 (int)substr( $ts, 4, 2 ), # Month
1803 (int)substr( $ts, 6, 2 ), # Day
1804 (int)substr( $ts, 0, 4 ) ); # Year
1806 $date = date( 'YmdHis', $t );
1807 wfRestoreWarnings();
1813 * This is meant to be used by time(), date(), and timeanddate() to get
1814 * the date preference they're supposed to use, it should be used in
1818 * function timeanddate([...], $format = true) {
1819 * $datePreference = $this->dateFormat($format);
1824 * @param $usePrefs Mixed: if true, the user's preference is used
1825 * if false, the site/language default is used
1826 * if int/string, assumed to be a format.
1829 function dateFormat( $usePrefs = true ) {
1832 if ( is_bool( $usePrefs ) ) {
1834 $datePreference = $wgUser->getDatePreference();
1836 $datePreference = (string)User
::getDefaultOption( 'date' );
1839 $datePreference = (string)$usePrefs;
1843 if ( $datePreference == '' ) {
1847 return $datePreference;
1851 * Get a format string for a given type and preference
1852 * @param $type string May be date, time or both
1853 * @param $pref string The format name as it appears in Messages*.php
1857 function getDateFormatString( $type, $pref ) {
1858 if ( !isset( $this->dateFormatStrings
[$type][$pref] ) ) {
1859 if ( $pref == 'default' ) {
1860 $pref = $this->getDefaultDateFormat();
1861 $df = self
::$dataCache->getSubitem( $this->mCode
, 'dateFormats', "$pref $type" );
1863 $df = self
::$dataCache->getSubitem( $this->mCode
, 'dateFormats', "$pref $type" );
1864 if ( is_null( $df ) ) {
1865 $pref = $this->getDefaultDateFormat();
1866 $df = self
::$dataCache->getSubitem( $this->mCode
, 'dateFormats', "$pref $type" );
1869 $this->dateFormatStrings
[$type][$pref] = $df;
1871 return $this->dateFormatStrings
[$type][$pref];
1875 * @param $ts Mixed: the time format which needs to be turned into a
1876 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
1877 * @param $adj Bool: whether to adjust the time output according to the
1878 * user configured offset ($timecorrection)
1879 * @param $format Mixed: true to use user's date format preference
1880 * @param $timecorrection String|bool the time offset as returned by
1881 * validateTimeZone() in Special:Preferences
1884 function date( $ts, $adj = false, $format = true, $timecorrection = false ) {
1885 $ts = wfTimestamp( TS_MW
, $ts );
1887 $ts = $this->userAdjust( $ts, $timecorrection );
1889 $df = $this->getDateFormatString( 'date', $this->dateFormat( $format ) );
1890 return $this->sprintfDate( $df, $ts );
1894 * @param $ts Mixed: the time format which needs to be turned into a
1895 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
1896 * @param $adj Bool: whether to adjust the time output according to the
1897 * user configured offset ($timecorrection)
1898 * @param $format Mixed: true to use user's date format preference
1899 * @param $timecorrection String|bool the time offset as returned by
1900 * validateTimeZone() in Special:Preferences
1903 function time( $ts, $adj = false, $format = true, $timecorrection = false ) {
1904 $ts = wfTimestamp( TS_MW
, $ts );
1906 $ts = $this->userAdjust( $ts, $timecorrection );
1908 $df = $this->getDateFormatString( 'time', $this->dateFormat( $format ) );
1909 return $this->sprintfDate( $df, $ts );
1913 * @param $ts Mixed: the time format which needs to be turned into a
1914 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
1915 * @param $adj Bool: whether to adjust the time output according to the
1916 * user configured offset ($timecorrection)
1917 * @param $format Mixed: what format to return, if it's false output the
1918 * default one (default true)
1919 * @param $timecorrection String|bool the time offset as returned by
1920 * validateTimeZone() in Special:Preferences
1923 function timeanddate( $ts, $adj = false, $format = true, $timecorrection = false ) {
1924 $ts = wfTimestamp( TS_MW
, $ts );
1926 $ts = $this->userAdjust( $ts, $timecorrection );
1928 $df = $this->getDateFormatString( 'both', $this->dateFormat( $format ) );
1929 return $this->sprintfDate( $df, $ts );
1933 * Takes a number of seconds and turns it into a text using values such as hours and minutes.
1937 * @param integer $seconds The amount of seconds.
1938 * @param array $chosenIntervals The intervals to enable.
1942 public function formatDuration( $seconds, array $chosenIntervals = array() ) {
1943 $intervals = $this->getDurationIntervals( $seconds, $chosenIntervals );
1945 $segments = array();
1947 foreach ( $intervals as $intervalName => $intervalValue ) {
1948 $message = new Message( 'duration-' . $intervalName, array( $intervalValue ) );
1949 $segments[] = $message->inLanguage( $this )->escaped();
1952 return $this->listToText( $segments );
1956 * Takes a number of seconds and returns an array with a set of corresponding intervals.
1957 * For example 65 will be turned into array( minutes => 1, seconds => 5 ).
1961 * @param integer $seconds The amount of seconds.
1962 * @param array $chosenIntervals The intervals to enable.
1966 public function getDurationIntervals( $seconds, array $chosenIntervals = array() ) {
1967 if ( empty( $chosenIntervals ) ) {
1968 $chosenIntervals = array( 'millennia', 'centuries', 'decades', 'years', 'days', 'hours', 'minutes', 'seconds' );
1971 $intervals = array_intersect_key( self
::$durationIntervals, array_flip( $chosenIntervals ) );
1972 $sortedNames = array_keys( $intervals );
1973 $smallestInterval = array_pop( $sortedNames );
1975 $segments = array();
1977 foreach ( $intervals as $name => $length ) {
1978 $value = floor( $seconds / $length );
1980 if ( $value > 0 ||
( $name == $smallestInterval && empty( $segments ) ) ) {
1981 $seconds -= $value * $length;
1982 $segments[$name] = $value;
1990 * Internal helper function for userDate(), userTime() and userTimeAndDate()
1992 * @param $type String: can be 'date', 'time' or 'both'
1993 * @param $ts Mixed: the time format which needs to be turned into a
1994 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
1995 * @param $user User object used to get preferences for timezone and format
1996 * @param $options Array, can contain the following keys:
1997 * - 'timecorrection': time correction, can have the following values:
1998 * - true: use user's preference
1999 * - false: don't use time correction
2000 * - integer: value of time correction in minutes
2001 * - 'format': format to use, can have the following values:
2002 * - true: use user's preference
2003 * - false: use default preference
2004 * - string: format to use
2008 private function internalUserTimeAndDate( $type, $ts, User
$user, array $options ) {
2009 $ts = wfTimestamp( TS_MW
, $ts );
2010 $options +
= array( 'timecorrection' => true, 'format' => true );
2011 if ( $options['timecorrection'] !== false ) {
2012 if ( $options['timecorrection'] === true ) {
2013 $offset = $user->getOption( 'timecorrection' );
2015 $offset = $options['timecorrection'];
2017 $ts = $this->userAdjust( $ts, $offset );
2019 if ( $options['format'] === true ) {
2020 $format = $user->getDatePreference();
2022 $format = $options['format'];
2024 $df = $this->getDateFormatString( $type, $this->dateFormat( $format ) );
2025 return $this->sprintfDate( $df, $ts );
2029 * Get the formatted date for the given timestamp and formatted for
2032 * @param $ts Mixed: the time format which needs to be turned into a
2033 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2034 * @param $user User object used to get preferences for timezone and format
2035 * @param $options Array, can contain the following keys:
2036 * - 'timecorrection': time correction, can have the following values:
2037 * - true: use user's preference
2038 * - false: don't use time correction
2039 * - integer: value of time correction in minutes
2040 * - 'format': format to use, can have the following values:
2041 * - true: use user's preference
2042 * - false: use default preference
2043 * - string: format to use
2047 public function userDate( $ts, User
$user, array $options = array() ) {
2048 return $this->internalUserTimeAndDate( 'date', $ts, $user, $options );
2052 * Get the formatted time for the given timestamp and formatted for
2055 * @param $ts Mixed: the time format which needs to be turned into a
2056 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2057 * @param $user User object used to get preferences for timezone and format
2058 * @param $options Array, can contain the following keys:
2059 * - 'timecorrection': time correction, can have the following values:
2060 * - true: use user's preference
2061 * - false: don't use time correction
2062 * - integer: value of time correction in minutes
2063 * - 'format': format to use, can have the following values:
2064 * - true: use user's preference
2065 * - false: use default preference
2066 * - string: format to use
2070 public function userTime( $ts, User
$user, array $options = array() ) {
2071 return $this->internalUserTimeAndDate( 'time', $ts, $user, $options );
2075 * Get the formatted date and time for the given timestamp and formatted for
2078 * @param $ts Mixed: the time format which needs to be turned into a
2079 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2080 * @param $user User object used to get preferences for timezone and format
2081 * @param $options Array, can contain the following keys:
2082 * - 'timecorrection': time correction, can have the following values:
2083 * - true: use user's preference
2084 * - false: don't use time correction
2085 * - integer: value of time correction in minutes
2086 * - 'format': format to use, can have the following values:
2087 * - true: use user's preference
2088 * - false: use default preference
2089 * - string: format to use
2093 public function userTimeAndDate( $ts, User
$user, array $options = array() ) {
2094 return $this->internalUserTimeAndDate( 'both', $ts, $user, $options );
2098 * @param $key string
2099 * @return array|null
2101 function getMessage( $key ) {
2102 return self
::$dataCache->getSubitem( $this->mCode
, 'messages', $key );
2108 function getAllMessages() {
2109 return self
::$dataCache->getItem( $this->mCode
, 'messages' );
2118 function iconv( $in, $out, $string ) {
2119 # This is a wrapper for iconv in all languages except esperanto,
2120 # which does some nasty x-conversions beforehand
2122 # Even with //IGNORE iconv can whine about illegal characters in
2123 # *input* string. We just ignore those too.
2124 # REF: http://bugs.php.net/bug.php?id=37166
2125 # REF: https://bugzilla.wikimedia.org/show_bug.cgi?id=16885
2126 wfSuppressWarnings();
2127 $text = iconv( $in, $out . '//IGNORE', $string );
2128 wfRestoreWarnings();
2132 // callback functions for uc(), lc(), ucwords(), ucwordbreaks()
2135 * @param $matches array
2136 * @return mixed|string
2138 function ucwordbreaksCallbackAscii( $matches ) {
2139 return $this->ucfirst( $matches[1] );
2143 * @param $matches array
2146 function ucwordbreaksCallbackMB( $matches ) {
2147 return mb_strtoupper( $matches[0] );
2151 * @param $matches array
2154 function ucCallback( $matches ) {
2155 list( $wikiUpperChars ) = self
::getCaseMaps();
2156 return strtr( $matches[1], $wikiUpperChars );
2160 * @param $matches array
2163 function lcCallback( $matches ) {
2164 list( , $wikiLowerChars ) = self
::getCaseMaps();
2165 return strtr( $matches[1], $wikiLowerChars );
2169 * @param $matches array
2172 function ucwordsCallbackMB( $matches ) {
2173 return mb_strtoupper( $matches[0] );
2177 * @param $matches array
2180 function ucwordsCallbackWiki( $matches ) {
2181 list( $wikiUpperChars ) = self
::getCaseMaps();
2182 return strtr( $matches[0], $wikiUpperChars );
2186 * Make a string's first character uppercase
2188 * @param $str string
2192 function ucfirst( $str ) {
2194 if ( $o < 96 ) { // if already uppercase...
2196 } elseif ( $o < 128 ) {
2197 return ucfirst( $str ); // use PHP's ucfirst()
2199 // fall back to more complex logic in case of multibyte strings
2200 return $this->uc( $str, true );
2205 * Convert a string to uppercase
2207 * @param $str string
2208 * @param $first bool
2212 function uc( $str, $first = false ) {
2213 if ( function_exists( 'mb_strtoupper' ) ) {
2215 if ( $this->isMultibyte( $str ) ) {
2216 return mb_strtoupper( mb_substr( $str, 0, 1 ) ) . mb_substr( $str, 1 );
2218 return ucfirst( $str );
2221 return $this->isMultibyte( $str ) ?
mb_strtoupper( $str ) : strtoupper( $str );
2224 if ( $this->isMultibyte( $str ) ) {
2225 $x = $first ?
'^' : '';
2226 return preg_replace_callback(
2227 "/$x([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/",
2228 array( $this, 'ucCallback' ),
2232 return $first ?
ucfirst( $str ) : strtoupper( $str );
2238 * @param $str string
2239 * @return mixed|string
2241 function lcfirst( $str ) {
2244 return strval( $str );
2245 } elseif ( $o >= 128 ) {
2246 return $this->lc( $str, true );
2247 } elseif ( $o > 96 ) {
2250 $str[0] = strtolower( $str[0] );
2256 * @param $str string
2257 * @param $first bool
2258 * @return mixed|string
2260 function lc( $str, $first = false ) {
2261 if ( function_exists( 'mb_strtolower' ) ) {
2263 if ( $this->isMultibyte( $str ) ) {
2264 return mb_strtolower( mb_substr( $str, 0, 1 ) ) . mb_substr( $str, 1 );
2266 return strtolower( substr( $str, 0, 1 ) ) . substr( $str, 1 );
2269 return $this->isMultibyte( $str ) ?
mb_strtolower( $str ) : strtolower( $str );
2272 if ( $this->isMultibyte( $str ) ) {
2273 $x = $first ?
'^' : '';
2274 return preg_replace_callback(
2275 "/$x([A-Z]|[\\xc0-\\xff][\\x80-\\xbf]*)/",
2276 array( $this, 'lcCallback' ),
2280 return $first ?
strtolower( substr( $str, 0, 1 ) ) . substr( $str, 1 ) : strtolower( $str );
2286 * @param $str string
2289 function isMultibyte( $str ) {
2290 return (bool)preg_match( '/[\x80-\xff]/', $str );
2294 * @param $str string
2295 * @return mixed|string
2297 function ucwords( $str ) {
2298 if ( $this->isMultibyte( $str ) ) {
2299 $str = $this->lc( $str );
2301 // regexp to find first letter in each word (i.e. after each space)
2302 $replaceRegexp = "/^([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)| ([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/";
2304 // function to use to capitalize a single char
2305 if ( function_exists( 'mb_strtoupper' ) ) {
2306 return preg_replace_callback(
2308 array( $this, 'ucwordsCallbackMB' ),
2312 return preg_replace_callback(
2314 array( $this, 'ucwordsCallbackWiki' ),
2319 return ucwords( strtolower( $str ) );
2324 * capitalize words at word breaks
2326 * @param $str string
2329 function ucwordbreaks( $str ) {
2330 if ( $this->isMultibyte( $str ) ) {
2331 $str = $this->lc( $str );
2333 // since \b doesn't work for UTF-8, we explicitely define word break chars
2334 $breaks = "[ \-\(\)\}\{\.,\?!]";
2336 // find first letter after word break
2337 $replaceRegexp = "/^([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)|$breaks([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/";
2339 if ( function_exists( 'mb_strtoupper' ) ) {
2340 return preg_replace_callback(
2342 array( $this, 'ucwordbreaksCallbackMB' ),
2346 return preg_replace_callback(
2348 array( $this, 'ucwordsCallbackWiki' ),
2353 return preg_replace_callback(
2354 '/\b([\w\x80-\xff]+)\b/',
2355 array( $this, 'ucwordbreaksCallbackAscii' ),
2362 * Return a case-folded representation of $s
2364 * This is a representation such that caseFold($s1)==caseFold($s2) if $s1
2365 * and $s2 are the same except for the case of their characters. It is not
2366 * necessary for the value returned to make sense when displayed.
2368 * Do *not* perform any other normalisation in this function. If a caller
2369 * uses this function when it should be using a more general normalisation
2370 * function, then fix the caller.
2376 function caseFold( $s ) {
2377 return $this->uc( $s );
2384 function checkTitleEncoding( $s ) {
2385 if ( is_array( $s ) ) {
2386 wfDebugDieBacktrace( 'Given array to checkTitleEncoding.' );
2388 # Check for non-UTF-8 URLs
2389 $ishigh = preg_match( '/[\x80-\xff]/', $s );
2394 if ( function_exists( 'mb_check_encoding' ) ) {
2395 $isutf8 = mb_check_encoding( $s, 'UTF-8' );
2397 $isutf8 = preg_match( '/^(?>[\x00-\x7f]|[\xc0-\xdf][\x80-\xbf]|' .
2398 '[\xe0-\xef][\x80-\xbf]{2}|[\xf0-\xf7][\x80-\xbf]{3})+$/', $s );
2404 return $this->iconv( $this->fallback8bitEncoding(), 'utf-8', $s );
2410 function fallback8bitEncoding() {
2411 return self
::$dataCache->getItem( $this->mCode
, 'fallback8bitEncoding' );
2415 * Most writing systems use whitespace to break up words.
2416 * Some languages such as Chinese don't conventionally do this,
2417 * which requires special handling when breaking up words for
2422 function hasWordBreaks() {
2427 * Some languages such as Chinese require word segmentation,
2428 * Specify such segmentation when overridden in derived class.
2430 * @param $string String
2433 function segmentByWord( $string ) {
2438 * Some languages have special punctuation need to be normalized.
2439 * Make such changes here.
2441 * @param $string String
2444 function normalizeForSearch( $string ) {
2445 return self
::convertDoubleWidth( $string );
2449 * convert double-width roman characters to single-width.
2450 * range: ff00-ff5f ~= 0020-007f
2452 * @param $string string
2456 protected static function convertDoubleWidth( $string ) {
2457 static $full = null;
2458 static $half = null;
2460 if ( $full === null ) {
2461 $fullWidth = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz";
2462 $halfWidth = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz";
2463 $full = str_split( $fullWidth, 3 );
2464 $half = str_split( $halfWidth );
2467 $string = str_replace( $full, $half, $string );
2472 * @param $string string
2473 * @param $pattern string
2476 protected static function insertSpace( $string, $pattern ) {
2477 $string = preg_replace( $pattern, " $1 ", $string );
2478 $string = preg_replace( '/ +/', ' ', $string );
2483 * @param $termsArray array
2486 function convertForSearchResult( $termsArray ) {
2487 # some languages, e.g. Chinese, need to do a conversion
2488 # in order for search results to be displayed correctly
2493 * Get the first character of a string.
2498 function firstChar( $s ) {
2501 '/^([\x00-\x7f]|[\xc0-\xdf][\x80-\xbf]|' .
2502 '[\xe0-\xef][\x80-\xbf]{2}|[\xf0-\xf7][\x80-\xbf]{3})/',
2507 if ( isset( $matches[1] ) ) {
2508 if ( strlen( $matches[1] ) != 3 ) {
2512 // Break down Hangul syllables to grab the first jamo
2513 $code = utf8ToCodepoint( $matches[1] );
2514 if ( $code < 0xac00 ||
0xd7a4 <= $code ) {
2516 } elseif ( $code < 0xb098 ) {
2517 return "\xe3\x84\xb1";
2518 } elseif ( $code < 0xb2e4 ) {
2519 return "\xe3\x84\xb4";
2520 } elseif ( $code < 0xb77c ) {
2521 return "\xe3\x84\xb7";
2522 } elseif ( $code < 0xb9c8 ) {
2523 return "\xe3\x84\xb9";
2524 } elseif ( $code < 0xbc14 ) {
2525 return "\xe3\x85\x81";
2526 } elseif ( $code < 0xc0ac ) {
2527 return "\xe3\x85\x82";
2528 } elseif ( $code < 0xc544 ) {
2529 return "\xe3\x85\x85";
2530 } elseif ( $code < 0xc790 ) {
2531 return "\xe3\x85\x87";
2532 } elseif ( $code < 0xcc28 ) {
2533 return "\xe3\x85\x88";
2534 } elseif ( $code < 0xce74 ) {
2535 return "\xe3\x85\x8a";
2536 } elseif ( $code < 0xd0c0 ) {
2537 return "\xe3\x85\x8b";
2538 } elseif ( $code < 0xd30c ) {
2539 return "\xe3\x85\x8c";
2540 } elseif ( $code < 0xd558 ) {
2541 return "\xe3\x85\x8d";
2543 return "\xe3\x85\x8e";
2550 function initEncoding() {
2551 # Some languages may have an alternate char encoding option
2552 # (Esperanto X-coding, Japanese furigana conversion, etc)
2553 # If this language is used as the primary content language,
2554 # an override to the defaults can be set here on startup.
2561 function recodeForEdit( $s ) {
2562 # For some languages we'll want to explicitly specify
2563 # which characters make it into the edit box raw
2564 # or are converted in some way or another.
2565 global $wgEditEncoding;
2566 if ( $wgEditEncoding == '' ||
$wgEditEncoding == 'UTF-8' ) {
2569 return $this->iconv( 'UTF-8', $wgEditEncoding, $s );
2577 function recodeInput( $s ) {
2578 # Take the previous into account.
2579 global $wgEditEncoding;
2580 if ( $wgEditEncoding != '' ) {
2581 $enc = $wgEditEncoding;
2585 if ( $enc == 'UTF-8' ) {
2588 return $this->iconv( $enc, 'UTF-8', $s );
2593 * Convert a UTF-8 string to normal form C. In Malayalam and Arabic, this
2594 * also cleans up certain backwards-compatible sequences, converting them
2595 * to the modern Unicode equivalent.
2597 * This is language-specific for performance reasons only.
2603 function normalize( $s ) {
2604 global $wgAllUnicodeFixes;
2605 $s = UtfNormal
::cleanUp( $s );
2606 if ( $wgAllUnicodeFixes ) {
2607 $s = $this->transformUsingPairFile( 'normalize-ar.ser', $s );
2608 $s = $this->transformUsingPairFile( 'normalize-ml.ser', $s );
2615 * Transform a string using serialized data stored in the given file (which
2616 * must be in the serialized subdirectory of $IP). The file contains pairs
2617 * mapping source characters to destination characters.
2619 * The data is cached in process memory. This will go faster if you have the
2620 * FastStringSearch extension.
2622 * @param $file string
2623 * @param $string string
2625 * @throws MWException
2628 function transformUsingPairFile( $file, $string ) {
2629 if ( !isset( $this->transformData
[$file] ) ) {
2630 $data = wfGetPrecompiledData( $file );
2631 if ( $data === false ) {
2632 throw new MWException( __METHOD__
. ": The transformation file $file is missing" );
2634 $this->transformData
[$file] = new ReplacementArray( $data );
2636 return $this->transformData
[$file]->replace( $string );
2640 * For right-to-left language support
2645 return self
::$dataCache->getItem( $this->mCode
, 'rtl' );
2649 * Return the correct HTML 'dir' attribute value for this language.
2653 return $this->isRTL() ?
'rtl' : 'ltr';
2657 * Return 'left' or 'right' as appropriate alignment for line-start
2658 * for this language's text direction.
2660 * Should be equivalent to CSS3 'start' text-align value....
2664 function alignStart() {
2665 return $this->isRTL() ?
'right' : 'left';
2669 * Return 'right' or 'left' as appropriate alignment for line-end
2670 * for this language's text direction.
2672 * Should be equivalent to CSS3 'end' text-align value....
2676 function alignEnd() {
2677 return $this->isRTL() ?
'left' : 'right';
2681 * A hidden direction mark (LRM or RLM), depending on the language direction.
2682 * Unlike getDirMark(), this function returns the character as an HTML entity.
2683 * This function should be used when the output is guaranteed to be HTML,
2684 * because it makes the output HTML source code more readable. When
2685 * the output is plain text or can be escaped, getDirMark() should be used.
2687 * @param $opposite Boolean Get the direction mark opposite to your language
2690 function getDirMarkEntity( $opposite = false ) {
2691 if ( $opposite ) { return $this->isRTL() ?
'‎' : '‏'; }
2692 return $this->isRTL() ?
'‏' : '‎';
2696 * A hidden direction mark (LRM or RLM), depending on the language direction.
2697 * This function produces them as invisible Unicode characters and
2698 * the output may be hard to read and debug, so it should only be used
2699 * when the output is plain text or can be escaped. When the output is
2700 * HTML, use getDirMarkEntity() instead.
2702 * @param $opposite Boolean Get the direction mark opposite to your language
2705 function getDirMark( $opposite = false ) {
2706 $lrm = "\xE2\x80\x8E"; # LEFT-TO-RIGHT MARK, commonly abbreviated LRM
2707 $rlm = "\xE2\x80\x8F"; # RIGHT-TO-LEFT MARK, commonly abbreviated RLM
2708 if ( $opposite ) { return $this->isRTL() ?
$lrm : $rlm; }
2709 return $this->isRTL() ?
$rlm : $lrm;
2715 function capitalizeAllNouns() {
2716 return self
::$dataCache->getItem( $this->mCode
, 'capitalizeAllNouns' );
2720 * An arrow, depending on the language direction.
2722 * @param $direction String: the direction of the arrow: forwards (default), backwards, left, right, up, down.
2725 function getArrow( $direction = 'forwards' ) {
2726 switch ( $direction ) {
2728 return $this->isRTL() ?
'←' : '→';
2730 return $this->isRTL() ?
'→' : '←';
2743 * To allow "foo[[bar]]" to extend the link over the whole word "foobar"
2747 function linkPrefixExtension() {
2748 return self
::$dataCache->getItem( $this->mCode
, 'linkPrefixExtension' );
2754 function getMagicWords() {
2755 return self
::$dataCache->getItem( $this->mCode
, 'magicWords' );
2758 protected function doMagicHook() {
2759 if ( $this->mMagicHookDone
) {
2762 $this->mMagicHookDone
= true;
2763 wfProfileIn( 'LanguageGetMagic' );
2764 wfRunHooks( 'LanguageGetMagic', array( &$this->mMagicExtensions
, $this->getCode() ) );
2765 wfProfileOut( 'LanguageGetMagic' );
2769 * Fill a MagicWord object with data from here
2773 function getMagic( $mw ) {
2774 $this->doMagicHook();
2776 if ( isset( $this->mMagicExtensions
[$mw->mId
] ) ) {
2777 $rawEntry = $this->mMagicExtensions
[$mw->mId
];
2779 $magicWords = $this->getMagicWords();
2780 if ( isset( $magicWords[$mw->mId
] ) ) {
2781 $rawEntry = $magicWords[$mw->mId
];
2787 if ( !is_array( $rawEntry ) ) {
2788 error_log( "\"$rawEntry\" is not a valid magic word for \"$mw->mId\"" );
2790 $mw->mCaseSensitive
= $rawEntry[0];
2791 $mw->mSynonyms
= array_slice( $rawEntry, 1 );
2796 * Add magic words to the extension array
2798 * @param $newWords array
2800 function addMagicWordsByLang( $newWords ) {
2801 $fallbackChain = $this->getFallbackLanguages();
2802 $fallbackChain = array_reverse( $fallbackChain );
2803 foreach ( $fallbackChain as $code ) {
2804 if ( isset( $newWords[$code] ) ) {
2805 $this->mMagicExtensions
= $newWords[$code] +
$this->mMagicExtensions
;
2811 * Get special page names, as an associative array
2812 * case folded alias => real name
2814 function getSpecialPageAliases() {
2815 // Cache aliases because it may be slow to load them
2816 if ( is_null( $this->mExtendedSpecialPageAliases
) ) {
2818 $this->mExtendedSpecialPageAliases
=
2819 self
::$dataCache->getItem( $this->mCode
, 'specialPageAliases' );
2820 wfRunHooks( 'LanguageGetSpecialPageAliases',
2821 array( &$this->mExtendedSpecialPageAliases
, $this->getCode() ) );
2824 return $this->mExtendedSpecialPageAliases
;
2828 * Italic is unsuitable for some languages
2830 * @param $text String: the text to be emphasized.
2833 function emphasize( $text ) {
2834 return "<em>$text</em>";
2838 * Normally we output all numbers in plain en_US style, that is
2839 * 293,291.235 for twohundredninetythreethousand-twohundredninetyone
2840 * point twohundredthirtyfive. However this is not suitable for all
2841 * languages, some such as Pakaran want ੨੯੩,੨੯੫.੨੩੫ and others such as
2842 * Icelandic just want to use commas instead of dots, and dots instead
2843 * of commas like "293.291,235".
2845 * An example of this function being called:
2847 * wfMsg( 'message', $wgLang->formatNum( $num ) )
2850 * See LanguageGu.php for the Gujarati implementation and
2851 * $separatorTransformTable on MessageIs.php for
2852 * the , => . and . => , implementation.
2854 * @todo check if it's viable to use localeconv() for the decimal
2856 * @param $number Mixed: the string to be formatted, should be an integer
2857 * or a floating point number.
2858 * @param $nocommafy Bool: set to true for special numbers like dates
2861 public function formatNum( $number, $nocommafy = false ) {
2862 global $wgTranslateNumerals;
2863 if ( !$nocommafy ) {
2864 $number = $this->commafy( $number );
2865 $s = $this->separatorTransformTable();
2867 $number = strtr( $number, $s );
2871 if ( $wgTranslateNumerals ) {
2872 $s = $this->digitTransformTable();
2874 $number = strtr( $number, $s );
2882 * @param $number string
2885 function parseFormattedNumber( $number ) {
2886 $s = $this->digitTransformTable();
2888 $number = strtr( $number, array_flip( $s ) );
2891 $s = $this->separatorTransformTable();
2893 $number = strtr( $number, array_flip( $s ) );
2896 $number = strtr( $number, array( ',' => '' ) );
2901 * Adds commas to a given number
2906 function commafy( $_ ) {
2907 $digitGroupingPattern = $this->digitGroupingPattern();
2908 if ( $_ === null ) {
2912 if ( !$digitGroupingPattern ||
$digitGroupingPattern === "###,###,###" ) {
2913 // default grouping is at thousands, use the same for ###,###,### pattern too.
2914 return strrev( (string)preg_replace( '/(\d{3})(?=\d)(?!\d*\.)/', '$1,', strrev( $_ ) ) );
2916 // Ref: http://cldr.unicode.org/translation/number-patterns
2918 if ( intval( $_ ) < 0 ) {
2919 // For negative numbers apply the algorithm like positive number and add sign.
2921 $_ = substr( $_, 1 );
2923 $numberpart = array();
2924 $decimalpart = array();
2925 $numMatches = preg_match_all( "/(#+)/", $digitGroupingPattern, $matches );
2926 preg_match( "/\d+/", $_, $numberpart );
2927 preg_match( "/\.\d*/", $_, $decimalpart );
2928 $groupedNumber = ( count( $decimalpart ) > 0 ) ?
$decimalpart[0]:"";
2929 if ( $groupedNumber === $_ ) {
2930 // the string does not have any number part. Eg: .12345
2931 return $sign . $groupedNumber;
2933 $start = $end = strlen( $numberpart[0] );
2934 while ( $start > 0 ) {
2935 $match = $matches[0][$numMatches -1] ;
2936 $matchLen = strlen( $match );
2937 $start = $end - $matchLen;
2941 $groupedNumber = substr( $_ , $start, $end -$start ) . $groupedNumber ;
2943 if ( $numMatches > 1 ) {
2944 // use the last pattern for the rest of the number
2948 $groupedNumber = "," . $groupedNumber;
2951 return $sign . $groupedNumber;
2957 function digitGroupingPattern() {
2958 return self
::$dataCache->getItem( $this->mCode
, 'digitGroupingPattern' );
2964 function digitTransformTable() {
2965 return self
::$dataCache->getItem( $this->mCode
, 'digitTransformTable' );
2971 function separatorTransformTable() {
2972 return self
::$dataCache->getItem( $this->mCode
, 'separatorTransformTable' );
2976 * Take a list of strings and build a locale-friendly comma-separated
2977 * list, using the local comma-separator message.
2978 * The last two strings are chained with an "and".
2983 function listToText( array $l ) {
2985 $m = count( $l ) - 1;
2987 return $l[0] . $this->getMessageFromDB( 'and' ) . $this->getMessageFromDB( 'word-separator' ) . $l[1];
2989 for ( $i = $m; $i >= 0; $i-- ) {
2992 } elseif ( $i == $m - 1 ) {
2993 $s = $l[$i] . $this->getMessageFromDB( 'and' ) . $this->getMessageFromDB( 'word-separator' ) . $s;
2995 $s = $l[$i] . $this->getMessageFromDB( 'comma-separator' ) . $s;
3003 * Take a list of strings and build a locale-friendly comma-separated
3004 * list, using the local comma-separator message.
3005 * @param $list array of strings to put in a comma list
3008 function commaList( array $list ) {
3012 array( 'parsemag', 'escapenoentities', 'language' => $this )
3019 * Take a list of strings and build a locale-friendly semicolon-separated
3020 * list, using the local semicolon-separator message.
3021 * @param $list array of strings to put in a semicolon list
3024 function semicolonList( array $list ) {
3027 'semicolon-separator',
3028 array( 'parsemag', 'escapenoentities', 'language' => $this )
3035 * Same as commaList, but separate it with the pipe instead.
3036 * @param $list array of strings to put in a pipe list
3039 function pipeList( array $list ) {
3043 array( 'escapenoentities', 'language' => $this )
3050 * Truncate a string to a specified length in bytes, appending an optional
3051 * string (e.g. for ellipses)
3053 * The database offers limited byte lengths for some columns in the database;
3054 * multi-byte character sets mean we need to ensure that only whole characters
3055 * are included, otherwise broken characters can be passed to the user
3057 * If $length is negative, the string will be truncated from the beginning
3059 * @param $string String to truncate
3060 * @param $length Int: maximum length (including ellipses)
3061 * @param $ellipsis String to append to the truncated text
3062 * @param $adjustLength Boolean: Subtract length of ellipsis from $length.
3063 * $adjustLength was introduced in 1.18, before that behaved as if false.
3066 function truncate( $string, $length, $ellipsis = '...', $adjustLength = true ) {
3067 # Use the localized ellipsis character
3068 if ( $ellipsis == '...' ) {
3069 $ellipsis = wfMsgExt( 'ellipsis', array( 'escapenoentities', 'language' => $this ) );
3071 # Check if there is no need to truncate
3072 if ( $length == 0 ) {
3073 return $ellipsis; // convention
3074 } elseif ( strlen( $string ) <= abs( $length ) ) {
3075 return $string; // no need to truncate
3077 $stringOriginal = $string;
3078 # If ellipsis length is >= $length then we can't apply $adjustLength
3079 if ( $adjustLength && strlen( $ellipsis ) >= abs( $length ) ) {
3080 $string = $ellipsis; // this can be slightly unexpected
3081 # Otherwise, truncate and add ellipsis...
3083 $eLength = $adjustLength ?
strlen( $ellipsis ) : 0;
3084 if ( $length > 0 ) {
3085 $length -= $eLength;
3086 $string = substr( $string, 0, $length ); // xyz...
3087 $string = $this->removeBadCharLast( $string );
3088 $string = $string . $ellipsis;
3090 $length +
= $eLength;
3091 $string = substr( $string, $length ); // ...xyz
3092 $string = $this->removeBadCharFirst( $string );
3093 $string = $ellipsis . $string;
3096 # Do not truncate if the ellipsis makes the string longer/equal (bug 22181).
3097 # This check is *not* redundant if $adjustLength, due to the single case where
3098 # LEN($ellipsis) > ABS($limit arg); $stringOriginal could be shorter than $string.
3099 if ( strlen( $string ) < strlen( $stringOriginal ) ) {
3102 return $stringOriginal;
3107 * Remove bytes that represent an incomplete Unicode character
3108 * at the end of string (e.g. bytes of the char are missing)
3110 * @param $string String
3113 protected function removeBadCharLast( $string ) {
3114 if ( $string != '' ) {
3115 $char = ord( $string[strlen( $string ) - 1] );
3117 if ( $char >= 0xc0 ) {
3118 # We got the first byte only of a multibyte char; remove it.
3119 $string = substr( $string, 0, -1 );
3120 } elseif ( $char >= 0x80 &&
3121 preg_match( '/^(.*)(?:[\xe0-\xef][\x80-\xbf]|' .
3122 '[\xf0-\xf7][\x80-\xbf]{1,2})$/', $string, $m ) )
3124 # We chopped in the middle of a character; remove it
3132 * Remove bytes that represent an incomplete Unicode character
3133 * at the start of string (e.g. bytes of the char are missing)
3135 * @param $string String
3138 protected function removeBadCharFirst( $string ) {
3139 if ( $string != '' ) {
3140 $char = ord( $string[0] );
3141 if ( $char >= 0x80 && $char < 0xc0 ) {
3142 # We chopped in the middle of a character; remove the whole thing
3143 $string = preg_replace( '/^[\x80-\xbf]+/', '', $string );
3150 * Truncate a string of valid HTML to a specified length in bytes,
3151 * appending an optional string (e.g. for ellipses), and return valid HTML
3153 * This is only intended for styled/linked text, such as HTML with
3154 * tags like <span> and <a>, were the tags are self-contained (valid HTML).
3155 * Also, this will not detect things like "display:none" CSS.
3157 * Note: since 1.18 you do not need to leave extra room in $length for ellipses.
3159 * @param string $text HTML string to truncate
3160 * @param int $length (zero/positive) Maximum length (including ellipses)
3161 * @param string $ellipsis String to append to the truncated text
3164 function truncateHtml( $text, $length, $ellipsis = '...' ) {
3165 # Use the localized ellipsis character
3166 if ( $ellipsis == '...' ) {
3167 $ellipsis = wfMsgExt( 'ellipsis', array( 'escapenoentities', 'language' => $this ) );
3169 # Check if there is clearly no need to truncate
3170 if ( $length <= 0 ) {
3171 return $ellipsis; // no text shown, nothing to format (convention)
3172 } elseif ( strlen( $text ) <= $length ) {
3173 return $text; // string short enough even *with* HTML (short-circuit)
3176 $dispLen = 0; // innerHTML legth so far
3177 $testingEllipsis = false; // checking if ellipses will make string longer/equal?
3178 $tagType = 0; // 0-open, 1-close
3179 $bracketState = 0; // 1-tag start, 2-tag name, 0-neither
3180 $entityState = 0; // 0-not entity, 1-entity
3181 $tag = $ret = ''; // accumulated tag name, accumulated result string
3182 $openTags = array(); // open tag stack
3183 $maybeState = null; // possible truncation state
3185 $textLen = strlen( $text );
3186 $neLength = max( 0, $length - strlen( $ellipsis ) ); // non-ellipsis len if truncated
3187 for ( $pos = 0; true; ++
$pos ) {
3188 # Consider truncation once the display length has reached the maximim.
3189 # We check if $dispLen > 0 to grab tags for the $neLength = 0 case.
3190 # Check that we're not in the middle of a bracket/entity...
3191 if ( $dispLen && $dispLen >= $neLength && $bracketState == 0 && !$entityState ) {
3192 if ( !$testingEllipsis ) {
3193 $testingEllipsis = true;
3194 # Save where we are; we will truncate here unless there turn out to
3195 # be so few remaining characters that truncation is not necessary.
3196 if ( !$maybeState ) { // already saved? ($neLength = 0 case)
3197 $maybeState = array( $ret, $openTags ); // save state
3199 } elseif ( $dispLen > $length && $dispLen > strlen( $ellipsis ) ) {
3200 # String in fact does need truncation, the truncation point was OK.
3201 list( $ret, $openTags ) = $maybeState; // reload state
3202 $ret = $this->removeBadCharLast( $ret ); // multi-byte char fix
3203 $ret .= $ellipsis; // add ellipsis
3207 if ( $pos >= $textLen ) break; // extra iteration just for above checks
3209 # Read the next char...
3211 $lastCh = $pos ?
$text[$pos - 1] : '';
3212 $ret .= $ch; // add to result string
3214 $this->truncate_endBracket( $tag, $tagType, $lastCh, $openTags ); // for bad HTML
3215 $entityState = 0; // for bad HTML
3216 $bracketState = 1; // tag started (checking for backslash)
3217 } elseif ( $ch == '>' ) {
3218 $this->truncate_endBracket( $tag, $tagType, $lastCh, $openTags );
3219 $entityState = 0; // for bad HTML
3220 $bracketState = 0; // out of brackets
3221 } elseif ( $bracketState == 1 ) {
3223 $tagType = 1; // close tag (e.g. "</span>")
3225 $tagType = 0; // open tag (e.g. "<span>")
3228 $bracketState = 2; // building tag name
3229 } elseif ( $bracketState == 2 ) {
3233 // Name found (e.g. "<a href=..."), add on tag attributes...
3234 $pos +
= $this->truncate_skip( $ret, $text, "<>", $pos +
1 );
3236 } elseif ( $bracketState == 0 ) {
3237 if ( $entityState ) {
3240 $dispLen++
; // entity is one displayed char
3243 if ( $neLength == 0 && !$maybeState ) {
3244 // Save state without $ch. We want to *hit* the first
3245 // display char (to get tags) but not *use* it if truncating.
3246 $maybeState = array( substr( $ret, 0, -1 ), $openTags );
3249 $entityState = 1; // entity found, (e.g. " ")
3251 $dispLen++
; // this char is displayed
3252 // Add the next $max display text chars after this in one swoop...
3253 $max = ( $testingEllipsis ?
$length : $neLength ) - $dispLen;
3254 $skipped = $this->truncate_skip( $ret, $text, "<>&", $pos +
1, $max );
3255 $dispLen +
= $skipped;
3261 // Close the last tag if left unclosed by bad HTML
3262 $this->truncate_endBracket( $tag, $text[$textLen - 1], $tagType, $openTags );
3263 while ( count( $openTags ) > 0 ) {
3264 $ret .= '</' . array_pop( $openTags ) . '>'; // close open tags
3270 * truncateHtml() helper function
3271 * like strcspn() but adds the skipped chars to $ret
3280 private function truncate_skip( &$ret, $text, $search, $start, $len = null ) {
3281 if ( $len === null ) {
3282 $len = -1; // -1 means "no limit" for strcspn
3283 } elseif ( $len < 0 ) {
3287 if ( $start < strlen( $text ) ) {
3288 $skipCount = strcspn( $text, $search, $start, $len );
3289 $ret .= substr( $text, $start, $skipCount );
3295 * truncateHtml() helper function
3296 * (a) push or pop $tag from $openTags as needed
3297 * (b) clear $tag value
3298 * @param &$tag string Current HTML tag name we are looking at
3299 * @param $tagType int (0-open tag, 1-close tag)
3300 * @param $lastCh string Character before the '>' that ended this tag
3301 * @param &$openTags array Open tag stack (not accounting for $tag)
3303 private function truncate_endBracket( &$tag, $tagType, $lastCh, &$openTags ) {
3304 $tag = ltrim( $tag );
3306 if ( $tagType == 0 && $lastCh != '/' ) {
3307 $openTags[] = $tag; // tag opened (didn't close itself)
3308 } elseif ( $tagType == 1 ) {
3309 if ( $openTags && $tag == $openTags[count( $openTags ) - 1] ) {
3310 array_pop( $openTags ); // tag closed
3318 * Grammatical transformations, needed for inflected languages
3319 * Invoked by putting {{grammar:case|word}} in a message
3321 * @param $word string
3322 * @param $case string
3325 function convertGrammar( $word, $case ) {
3326 global $wgGrammarForms;
3327 if ( isset( $wgGrammarForms[$this->getCode()][$case][$word] ) ) {
3328 return $wgGrammarForms[$this->getCode()][$case][$word];
3333 * Get the grammar forms for the content language
3334 * @return array of grammar forms
3337 function getGrammarForms() {
3338 global $wgGrammarForms;
3339 if ( isset( $wgGrammarForms[$this->getCode()] ) && is_array( $wgGrammarForms[$this->getCode()] ) ) {
3340 return $wgGrammarForms[$this->getCode()];
3345 * Provides an alternative text depending on specified gender.
3346 * Usage {{gender:username|masculine|feminine|neutral}}.
3347 * username is optional, in which case the gender of current user is used,
3348 * but only in (some) interface messages; otherwise default gender is used.
3350 * If no forms are given, an empty string is returned. If only one form is
3351 * given, it will be returned unconditionally. These details are implied by
3352 * the caller and cannot be overridden in subclasses.
3354 * If more than one form is given, the default is to use the neutral one
3355 * if it is specified, and to use the masculine one otherwise. These
3356 * details can be overridden in subclasses.
3358 * @param $gender string
3359 * @param $forms array
3363 function gender( $gender, $forms ) {
3364 if ( !count( $forms ) ) {
3367 $forms = $this->preConvertPlural( $forms, 2 );
3368 if ( $gender === 'male' ) {
3371 if ( $gender === 'female' ) {
3374 return isset( $forms[2] ) ?
$forms[2] : $forms[0];
3378 * Plural form transformations, needed for some languages.
3379 * For example, there are 3 form of plural in Russian and Polish,
3380 * depending on "count mod 10". See [[w:Plural]]
3381 * For English it is pretty simple.
3383 * Invoked by putting {{plural:count|wordform1|wordform2}}
3384 * or {{plural:count|wordform1|wordform2|wordform3}}
3386 * Example: {{plural:{{NUMBEROFARTICLES}}|article|articles}}
3388 * @param $count Integer: non-localized number
3389 * @param $forms Array: different plural forms
3390 * @return string Correct form of plural for $count in this language
3392 function convertPlural( $count, $forms ) {
3393 if ( !count( $forms ) ) {
3396 $forms = $this->preConvertPlural( $forms, 2 );
3398 return ( $count == 1 ) ?
$forms[0] : $forms[1];
3402 * Checks that convertPlural was given an array and pads it to requested
3403 * amount of forms by copying the last one.
3405 * @param $count Integer: How many forms should there be at least
3406 * @param $forms Array of forms given to convertPlural
3407 * @return array Padded array of forms or an exception if not an array
3409 protected function preConvertPlural( /* Array */ $forms, $count ) {
3410 while ( count( $forms ) < $count ) {
3411 $forms[] = $forms[count( $forms ) - 1];
3417 * @todo Maybe translate block durations. Note that this function is somewhat misnamed: it
3418 * deals with translating the *duration* ("1 week", "4 days", etc), not the expiry time
3419 * (which is an absolute timestamp). Please note: do NOT add this blindly, as it is used
3420 * on old expiry lengths recorded in log entries. You'd need to provide the start date to
3423 * @param $str String: the validated block duration in English
3424 * @return string Somehow translated block duration
3425 * @see LanguageFi.php for example implementation
3427 function translateBlockExpiry( $str ) {
3428 $duration = SpecialBlock
::getSuggestedDurations( $this );
3429 foreach ( $duration as $show => $value ) {
3430 if ( strcmp( $str, $value ) == 0 ) {
3431 return htmlspecialchars( trim( $show ) );
3435 // Since usually only infinite or indefinite is only on list, so try
3436 // equivalents if still here.
3437 $indefs = array( 'infinite', 'infinity', 'indefinite' );
3438 if ( in_array( $str, $indefs ) ) {
3439 foreach ( $indefs as $val ) {
3440 $show = array_search( $val, $duration, true );
3441 if ( $show !== false ) {
3442 return htmlspecialchars( trim( $show ) );
3446 // If all else fails, return the original string.
3451 * languages like Chinese need to be segmented in order for the diff
3454 * @param $text String
3457 public function segmentForDiff( $text ) {
3462 * and unsegment to show the result
3464 * @param $text String
3467 public function unsegmentForDiff( $text ) {
3472 * Return the LanguageConverter used in the Language
3475 * @return LanguageConverter
3477 public function getConverter() {
3478 return $this->mConverter
;
3482 * convert text to all supported variants
3484 * @param $text string
3487 public function autoConvertToAllVariants( $text ) {
3488 return $this->mConverter
->autoConvertToAllVariants( $text );
3492 * convert text to different variants of a language.
3494 * @param $text string
3497 public function convert( $text ) {
3498 return $this->mConverter
->convert( $text );
3502 * Convert a Title object to a string in the preferred variant
3504 * @param $title Title
3507 public function convertTitle( $title ) {
3508 return $this->mConverter
->convertTitle( $title );
3512 * Check if this is a language with variants
3516 public function hasVariants() {
3517 return sizeof( $this->getVariants() ) > 1;
3521 * Check if the language has the specific variant
3524 * @param $variant string
3527 public function hasVariant( $variant ) {
3528 return (bool)$this->mConverter
->validateVariant( $variant );
3532 * Put custom tags (e.g. -{ }-) around math to prevent conversion
3534 * @param $text string
3537 public function armourMath( $text ) {
3538 return $this->mConverter
->armourMath( $text );
3542 * Perform output conversion on a string, and encode for safe HTML output.
3543 * @param $text String text to be converted
3544 * @param $isTitle Bool whether this conversion is for the article title
3546 * @todo this should get integrated somewhere sane
3548 public function convertHtml( $text, $isTitle = false ) {
3549 return htmlspecialchars( $this->convert( $text, $isTitle ) );
3553 * @param $key string
3556 public function convertCategoryKey( $key ) {
3557 return $this->mConverter
->convertCategoryKey( $key );
3561 * Get the list of variants supported by this language
3562 * see sample implementation in LanguageZh.php
3564 * @return array an array of language codes
3566 public function getVariants() {
3567 return $this->mConverter
->getVariants();
3573 public function getPreferredVariant() {
3574 return $this->mConverter
->getPreferredVariant();
3580 public function getDefaultVariant() {
3581 return $this->mConverter
->getDefaultVariant();
3587 public function getURLVariant() {
3588 return $this->mConverter
->getURLVariant();
3592 * If a language supports multiple variants, it is
3593 * possible that non-existing link in one variant
3594 * actually exists in another variant. this function
3595 * tries to find it. See e.g. LanguageZh.php
3597 * @param $link String: the name of the link
3598 * @param $nt Mixed: the title object of the link
3599 * @param $ignoreOtherCond Boolean: to disable other conditions when
3600 * we need to transclude a template or update a category's link
3601 * @return null the input parameters may be modified upon return
3603 public function findVariantLink( &$link, &$nt, $ignoreOtherCond = false ) {
3604 $this->mConverter
->findVariantLink( $link, $nt, $ignoreOtherCond );
3608 * If a language supports multiple variants, converts text
3609 * into an array of all possible variants of the text:
3610 * 'variant' => text in that variant
3612 * @deprecated since 1.17 Use autoConvertToAllVariants()
3614 * @param $text string
3618 public function convertLinkToAllVariants( $text ) {
3619 return $this->mConverter
->convertLinkToAllVariants( $text );
3623 * returns language specific options used by User::getPageRenderHash()
3624 * for example, the preferred language variant
3628 function getExtraHashOptions() {
3629 return $this->mConverter
->getExtraHashOptions();
3633 * For languages that support multiple variants, the title of an
3634 * article may be displayed differently in different variants. this
3635 * function returns the apporiate title defined in the body of the article.
3639 public function getParsedTitle() {
3640 return $this->mConverter
->getParsedTitle();
3644 * Enclose a string with the "no conversion" tag. This is used by
3645 * various functions in the Parser
3647 * @param $text String: text to be tagged for no conversion
3648 * @param $noParse bool
3649 * @return string the tagged text
3651 public function markNoConversion( $text, $noParse = false ) {
3652 return $this->mConverter
->markNoConversion( $text, $noParse );
3656 * A regular expression to match legal word-trailing characters
3657 * which should be merged onto a link of the form [[foo]]bar.
3661 public function linkTrail() {
3662 return self
::$dataCache->getItem( $this->mCode
, 'linkTrail' );
3668 function getLangObj() {
3673 * Get the RFC 3066 code for this language object
3677 public function getCode() {
3678 return $this->mCode
;
3682 * Get the code in Bcp47 format which we can use
3683 * inside of html lang="" tags.
3687 public function getHtmlCode() {
3688 if ( is_null( $this->mHtmlCode
) ) {
3689 $this->mHtmlCode
= wfBCP47( $this->getCode() );
3691 return $this->mHtmlCode
;
3695 * @param $code string
3697 public function setCode( $code ) {
3698 $this->mCode
= $code;
3699 // Ensure we don't leave an incorrect html code lying around
3700 $this->mHtmlCode
= null;
3704 * Get the name of a file for a certain language code
3705 * @param $prefix string Prepend this to the filename
3706 * @param $code string Language code
3707 * @param $suffix string Append this to the filename
3708 * @throws MWException
3709 * @return string $prefix . $mangledCode . $suffix
3711 public static function getFileName( $prefix = 'Language', $code, $suffix = '.php' ) {
3712 // Protect against path traversal
3713 if ( !Language
::isValidCode( $code )
3714 ||
strcspn( $code, ":/\\\000" ) !== strlen( $code ) )
3716 throw new MWException( "Invalid language code \"$code\"" );
3719 return $prefix . str_replace( '-', '_', ucfirst( $code ) ) . $suffix;
3723 * Get the language code from a file name. Inverse of getFileName()
3724 * @param $filename string $prefix . $languageCode . $suffix
3725 * @param $prefix string Prefix before the language code
3726 * @param $suffix string Suffix after the language code
3727 * @return string Language code, or false if $prefix or $suffix isn't found
3729 public static function getCodeFromFileName( $filename, $prefix = 'Language', $suffix = '.php' ) {
3731 preg_match( '/' . preg_quote( $prefix, '/' ) . '([A-Z][a-z_]+)' .
3732 preg_quote( $suffix, '/' ) . '/', $filename, $m );
3733 if ( !count( $m ) ) {
3736 return str_replace( '_', '-', strtolower( $m[1] ) );
3740 * @param $code string
3743 public static function getMessagesFileName( $code ) {
3745 $file = self
::getFileName( "$IP/languages/messages/Messages", $code, '.php' );
3746 wfRunHooks( 'Language::getMessagesFileName', array( $code, &$file ) );
3751 * @param $code string
3754 public static function getClassFileName( $code ) {
3756 return self
::getFileName( "$IP/languages/classes/Language", $code, '.php' );
3760 * Get the first fallback for a given language.
3762 * @param $code string
3764 * @return bool|string
3766 public static function getFallbackFor( $code ) {
3767 if ( $code === 'en' ||
!Language
::isValidBuiltInCode( $code ) ) {
3770 $fallbacks = self
::getFallbacksFor( $code );
3771 $first = array_shift( $fallbacks );
3777 * Get the ordered list of fallback languages.
3780 * @param $code string Language code
3783 public static function getFallbacksFor( $code ) {
3784 if ( $code === 'en' ||
!Language
::isValidBuiltInCode( $code ) ) {
3787 $v = self
::getLocalisationCache()->getItem( $code, 'fallback' );
3788 $v = array_map( 'trim', explode( ',', $v ) );
3789 if ( $v[count( $v ) - 1] !== 'en' ) {
3797 * Get all messages for a given language
3798 * WARNING: this may take a long time. If you just need all message *keys*
3799 * but need the *contents* of only a few messages, consider using getMessageKeysFor().
3801 * @param $code string
3805 public static function getMessagesFor( $code ) {
3806 return self
::getLocalisationCache()->getItem( $code, 'messages' );
3810 * Get a message for a given language
3812 * @param $key string
3813 * @param $code string
3817 public static function getMessageFor( $key, $code ) {
3818 return self
::getLocalisationCache()->getSubitem( $code, 'messages', $key );
3822 * Get all message keys for a given language. This is a faster alternative to
3823 * array_keys( Language::getMessagesFor( $code ) )
3826 * @param $code string Language code
3827 * @return array of message keys (strings)
3829 public static function getMessageKeysFor( $code ) {
3830 return self
::getLocalisationCache()->getSubItemList( $code, 'messages' );
3837 function fixVariableInNamespace( $talk ) {
3838 if ( strpos( $talk, '$1' ) === false ) {
3842 global $wgMetaNamespace;
3843 $talk = str_replace( '$1', $wgMetaNamespace, $talk );
3845 # Allow grammar transformations
3846 # Allowing full message-style parsing would make simple requests
3847 # such as action=raw much more expensive than they need to be.
3848 # This will hopefully cover most cases.
3849 $talk = preg_replace_callback( '/{{grammar:(.*?)\|(.*?)}}/i',
3850 array( &$this, 'replaceGrammarInNamespace' ), $talk );
3851 return str_replace( ' ', '_', $talk );
3858 function replaceGrammarInNamespace( $m ) {
3859 return $this->convertGrammar( trim( $m[2] ), trim( $m[1] ) );
3863 * @throws MWException
3866 static function getCaseMaps() {
3867 static $wikiUpperChars, $wikiLowerChars;
3868 if ( isset( $wikiUpperChars ) ) {
3869 return array( $wikiUpperChars, $wikiLowerChars );
3872 wfProfileIn( __METHOD__
);
3873 $arr = wfGetPrecompiledData( 'Utf8Case.ser' );
3874 if ( $arr === false ) {
3875 throw new MWException(
3876 "Utf8Case.ser is missing, please run \"make\" in the serialized directory\n" );
3878 $wikiUpperChars = $arr['wikiUpperChars'];
3879 $wikiLowerChars = $arr['wikiLowerChars'];
3880 wfProfileOut( __METHOD__
);
3881 return array( $wikiUpperChars, $wikiLowerChars );
3885 * Decode an expiry (block, protection, etc) which has come from the DB
3887 * @FIXME: why are we returnings DBMS-dependent strings???
3889 * @param $expiry String: Database expiry String
3890 * @param $format Bool|Int true to process using language functions, or TS_ constant
3891 * to return the expiry in a given timestamp
3894 public function formatExpiry( $expiry, $format = true ) {
3895 static $infinity, $infinityMsg;
3896 if ( $infinity === null ) {
3897 $infinityMsg = wfMessage( 'infiniteblock' );
3898 $infinity = wfGetDB( DB_SLAVE
)->getInfinity();
3901 if ( $expiry == '' ||
$expiry == $infinity ) {
3902 return $format === true
3906 return $format === true
3907 ?
$this->timeanddate( $expiry, /* User preference timezone */ true )
3908 : wfTimestamp( $format, $expiry );
3914 * @param $seconds int|float
3915 * @param $format Array Optional
3916 * If $format['avoid'] == 'avoidseconds' - don't mention seconds if $seconds >= 1 hour
3917 * If $format['avoid'] == 'avoidminutes' - don't mention seconds/minutes if $seconds > 48 hours
3918 * If $format['noabbrevs'] is true - use 'seconds' and friends instead of 'seconds-abbrev' and friends
3919 * For backwards compatibility, $format may also be one of the strings 'avoidseconds' or 'avoidminutes'
3922 function formatTimePeriod( $seconds, $format = array() ) {
3923 if ( !is_array( $format ) ) {
3924 $format = array( 'avoid' => $format ); // For backwards compatibility
3926 if ( !isset( $format['avoid'] ) ) {
3927 $format['avoid'] = false;
3929 if ( !isset( $format['noabbrevs' ] ) ) {
3930 $format['noabbrevs'] = false;
3932 $secondsMsg = wfMessage(
3933 $format['noabbrevs'] ?
'seconds' : 'seconds-abbrev' )->inLanguage( $this );
3934 $minutesMsg = wfMessage(
3935 $format['noabbrevs'] ?
'minutes' : 'minutes-abbrev' )->inLanguage( $this );
3936 $hoursMsg = wfMessage(
3937 $format['noabbrevs'] ?
'hours' : 'hours-abbrev' )->inLanguage( $this );
3938 $daysMsg = wfMessage(
3939 $format['noabbrevs'] ?
'days' : 'days-abbrev' )->inLanguage( $this );
3941 if ( round( $seconds * 10 ) < 100 ) {
3942 $s = $this->formatNum( sprintf( "%.1f", round( $seconds * 10 ) / 10 ) );
3943 $s = $secondsMsg->params( $s )->text();
3944 } elseif ( round( $seconds ) < 60 ) {
3945 $s = $this->formatNum( round( $seconds ) );
3946 $s = $secondsMsg->params( $s )->text();
3947 } elseif ( round( $seconds ) < 3600 ) {
3948 $minutes = floor( $seconds / 60 );
3949 $secondsPart = round( fmod( $seconds, 60 ) );
3950 if ( $secondsPart == 60 ) {
3954 $s = $minutesMsg->params( $this->formatNum( $minutes ) )->text();
3956 $s .= $secondsMsg->params( $this->formatNum( $secondsPart ) )->text();
3957 } elseif ( round( $seconds ) <= 2 * 86400 ) {
3958 $hours = floor( $seconds / 3600 );
3959 $minutes = floor( ( $seconds - $hours * 3600 ) / 60 );
3960 $secondsPart = round( $seconds - $hours * 3600 - $minutes * 60 );
3961 if ( $secondsPart == 60 ) {
3965 if ( $minutes == 60 ) {
3969 $s = $hoursMsg->params( $this->formatNum( $hours ) )->text();
3971 $s .= $minutesMsg->params( $this->formatNum( $minutes ) )->text();
3972 if ( !in_array( $format['avoid'], array( 'avoidseconds', 'avoidminutes' ) ) ) {
3973 $s .= ' ' . $secondsMsg->params( $this->formatNum( $secondsPart ) )->text();
3976 $days = floor( $seconds / 86400 );
3977 if ( $format['avoid'] === 'avoidminutes' ) {
3978 $hours = round( ( $seconds - $days * 86400 ) / 3600 );
3979 if ( $hours == 24 ) {
3983 $s = $daysMsg->params( $this->formatNum( $days ) )->text();
3985 $s .= $hoursMsg->params( $this->formatNum( $hours ) )->text();
3986 } elseif ( $format['avoid'] === 'avoidseconds' ) {
3987 $hours = floor( ( $seconds - $days * 86400 ) / 3600 );
3988 $minutes = round( ( $seconds - $days * 86400 - $hours * 3600 ) / 60 );
3989 if ( $minutes == 60 ) {
3993 if ( $hours == 24 ) {
3997 $s = $daysMsg->params( $this->formatNum( $days ) )->text();
3999 $s .= $hoursMsg->params( $this->formatNum( $hours ) )->text();
4001 $s .= $minutesMsg->params( $this->formatNum( $minutes ) )->text();
4003 $s = $daysMsg->params( $this->formatNum( $days ) )->text();
4005 $s .= $this->formatTimePeriod( $seconds - $days * 86400, $format );
4012 * Format a bitrate for output, using an appropriate
4013 * unit (bps, kbps, Mbps, Gbps, Tbps, Pbps, Ebps, Zbps or Ybps) according to the magnitude in question
4015 * This use base 1000. For base 1024 use formatSize(), for another base
4016 * see formatComputingNumbers()
4021 function formatBitrate( $bps ) {
4022 return $this->formatComputingNumbers( $bps, 1000, "bitrate-$1bits" );
4026 * @param $size int Size of the unit
4027 * @param $boundary int Size boundary (1000, or 1024 in most cases)
4028 * @param $messageKey string Message key to be uesd
4031 function formatComputingNumbers( $size, $boundary, $messageKey ) {
4033 return str_replace( '$1', $this->formatNum( $size ),
4034 $this->getMessageFromDB( str_replace( '$1', '', $messageKey ) )
4037 $sizes = array( '', 'kilo', 'mega', 'giga', 'tera', 'peta', 'exa', 'zeta', 'yotta' );
4040 $maxIndex = count( $sizes ) - 1;
4041 while ( $size >= $boundary && $index < $maxIndex ) {
4046 // For small sizes no decimal places necessary
4049 // For MB and bigger two decimal places are smarter
4052 $msg = str_replace( '$1', $sizes[$index], $messageKey );
4054 $size = round( $size, $round );
4055 $text = $this->getMessageFromDB( $msg );
4056 return str_replace( '$1', $this->formatNum( $size ), $text );
4060 * Format a size in bytes for output, using an appropriate
4061 * unit (B, KB, MB, GB, TB, PB, EB, ZB or YB) according to the magnitude in question
4063 * This method use base 1024. For base 1000 use formatBitrate(), for
4064 * another base see formatComputingNumbers()
4066 * @param $size int Size to format
4067 * @return string Plain text (not HTML)
4069 function formatSize( $size ) {
4070 return $this->formatComputingNumbers( $size, 1024, "size-$1bytes" );
4074 * Make a list item, used by various special pages
4076 * @param $page String Page link
4077 * @param $details String Text between brackets
4078 * @param $oppositedm Boolean Add the direction mark opposite to your
4079 * language, to display text properly
4082 function specialList( $page, $details, $oppositedm = true ) {
4083 $dirmark = ( $oppositedm ?
$this->getDirMark( true ) : '' ) .
4084 $this->getDirMark();
4085 $details = $details ?
$dirmark . $this->getMessageFromDB( 'word-separator' ) .
4086 wfMsgExt( 'parentheses', array( 'escape', 'replaceafter', 'language' => $this ), $details ) : '';
4087 return $page . $details;
4091 * Generate (prev x| next x) (20|50|100...) type links for paging
4093 * @param $title Title object to link
4094 * @param $offset Integer offset parameter
4095 * @param $limit Integer limit parameter
4096 * @param $query String optional URL query parameter string
4097 * @param $atend Bool optional param for specified if this is the last page
4100 public function viewPrevNext( Title
$title, $offset, $limit, array $query = array(), $atend = false ) {
4101 // @todo FIXME: Why on earth this needs one message for the text and another one for tooltip?
4103 # Make 'previous' link
4104 $prev = wfMessage( 'prevn' )->inLanguage( $this )->title( $title )->numParams( $limit )->text();
4105 if ( $offset > 0 ) {
4106 $plink = $this->numLink( $title, max( $offset - $limit, 0 ), $limit,
4107 $query, $prev, 'prevn-title', 'mw-prevlink' );
4109 $plink = htmlspecialchars( $prev );
4113 $next = wfMessage( 'nextn' )->inLanguage( $this )->title( $title )->numParams( $limit )->text();
4115 $nlink = htmlspecialchars( $next );
4117 $nlink = $this->numLink( $title, $offset +
$limit, $limit,
4118 $query, $next, 'prevn-title', 'mw-nextlink' );
4121 # Make links to set number of items per page
4122 $numLinks = array();
4123 foreach ( array( 20, 50, 100, 250, 500 ) as $num ) {
4124 $numLinks[] = $this->numLink( $title, $offset, $num,
4125 $query, $this->formatNum( $num ), 'shown-title', 'mw-numlink' );
4128 return wfMessage( 'viewprevnext' )->inLanguage( $this )->title( $title
4129 )->rawParams( $plink, $nlink, $this->pipeList( $numLinks ) )->escaped();
4133 * Helper function for viewPrevNext() that generates links
4135 * @param $title Title object to link
4136 * @param $offset Integer offset parameter
4137 * @param $limit Integer limit parameter
4138 * @param $query Array extra query parameters
4139 * @param $link String text to use for the link; will be escaped
4140 * @param $tooltipMsg String name of the message to use as tooltip
4141 * @param $class String value of the "class" attribute of the link
4142 * @return String HTML fragment
4144 private function numLink( Title
$title, $offset, $limit, array $query, $link, $tooltipMsg, $class ) {
4145 $query = array( 'limit' => $limit, 'offset' => $offset ) +
$query;
4146 $tooltip = wfMessage( $tooltipMsg )->inLanguage( $this )->title( $title )->numParams( $limit )->text();
4147 return Html
::element( 'a', array( 'href' => $title->getLocalURL( $query ),
4148 'title' => $tooltip, 'class' => $class ), $link );
4152 * Get the conversion rule title, if any.
4156 public function getConvRuleTitle() {
4157 return $this->mConverter
->getConvRuleTitle();