3 * Internationalisation code.
5 * This program is free software; you can redistribute it and/or modify
6 * it under the terms of the GNU General Public License as published by
7 * the Free Software Foundation; either version 2 of the License, or
8 * (at your option) any later version.
10 * This program is distributed in the hope that it will be useful,
11 * but WITHOUT ANY WARRANTY; without even the implied warranty of
12 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 * GNU General Public License for more details.
15 * You should have received a copy of the GNU General Public License along
16 * with this program; if not, write to the Free Software Foundation, Inc.,
17 * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
18 * http://www.gnu.org/copyleft/gpl.html
25 * @defgroup Language Language
28 if ( !defined( 'MEDIAWIKI' ) ) {
29 echo "This file is part of MediaWiki, it is not a valid entry point.\n";
33 if ( function_exists( 'mb_strtoupper' ) ) {
34 mb_internal_encoding( 'UTF-8' );
38 * a fake language converter
47 function __construct( $langobj ) { $this->mLang
= $langobj; }
48 function autoConvert( $text, $variant = false ) { return $text; }
49 function autoConvertToAllVariants( $text ) { return array( $this->mLang
->getCode() => $text ); }
50 function convert( $t ) { return $t; }
51 function convertTo( $text, $variant ) { return $text; }
52 function convertTitle( $t ) { return $t->getPrefixedText(); }
53 function convertNamespace( $ns ) { return $this->mLang
->getFormattedNsText( $ns ); }
54 function getVariants() { return array( $this->mLang
->getCode() ); }
55 function getVariantFallbacks( $variant ) { return $this->mLang
->getCode(); }
56 function getPreferredVariant() { return $this->mLang
->getCode(); }
57 function getDefaultVariant() { return $this->mLang
->getCode(); }
58 function getURLVariant() { return ''; }
59 function getConvRuleTitle() { return false; }
60 function findVariantLink( &$l, &$n, $ignoreOtherCond = false ) { }
61 function getExtraHashOptions() { return ''; }
62 function getParsedTitle() { return ''; }
63 function markNoConversion( $text, $noParse = false ) { return $text; }
64 function convertCategoryKey( $key ) { return $key; }
65 function convertLinkToAllVariants( $text ) { return $this->autoConvertToAllVariants( $text ); }
66 /** @deprecated since 1.22 is no longer used */
67 function armourMath( $text ) { return $text; }
68 function validateVariant( $variant = null ) { return $variant === $this->mLang
->getCode() ?
$variant : null; }
69 function translate( $text, $variant ) { return $text; }
73 * Internationalisation code
79 * @var LanguageConverter
83 public $mVariants, $mCode, $mLoaded = false;
84 public $mMagicExtensions = array(), $mMagicHookDone = false;
85 private $mHtmlCode = null, $mParentLanguage = false;
87 public $dateFormatStrings = array();
88 public $mExtendedSpecialPageAliases;
90 protected $namespaceNames, $mNamespaceIds, $namespaceAliases;
93 * ReplacementArray object caches
95 public $transformData = array();
98 * @var LocalisationCache
100 static public $dataCache;
102 static public $mLangObjCache = array();
104 static public $mWeekdayMsgs = array(
105 'sunday', 'monday', 'tuesday', 'wednesday', 'thursday',
109 static public $mWeekdayAbbrevMsgs = array(
110 'sun', 'mon', 'tue', 'wed', 'thu', 'fri', 'sat'
113 static public $mMonthMsgs = array(
114 'january', 'february', 'march', 'april', 'may_long', 'june',
115 'july', 'august', 'september', 'october', 'november',
118 static public $mMonthGenMsgs = array(
119 'january-gen', 'february-gen', 'march-gen', 'april-gen', 'may-gen', 'june-gen',
120 'july-gen', 'august-gen', 'september-gen', 'october-gen', 'november-gen',
123 static public $mMonthAbbrevMsgs = array(
124 'jan', 'feb', 'mar', 'apr', 'may', 'jun', 'jul', 'aug',
125 'sep', 'oct', 'nov', 'dec'
128 static public $mIranianCalendarMonthMsgs = array(
129 'iranian-calendar-m1', 'iranian-calendar-m2', 'iranian-calendar-m3',
130 'iranian-calendar-m4', 'iranian-calendar-m5', 'iranian-calendar-m6',
131 'iranian-calendar-m7', 'iranian-calendar-m8', 'iranian-calendar-m9',
132 'iranian-calendar-m10', 'iranian-calendar-m11', 'iranian-calendar-m12'
135 static public $mHebrewCalendarMonthMsgs = array(
136 'hebrew-calendar-m1', 'hebrew-calendar-m2', 'hebrew-calendar-m3',
137 'hebrew-calendar-m4', 'hebrew-calendar-m5', 'hebrew-calendar-m6',
138 'hebrew-calendar-m7', 'hebrew-calendar-m8', 'hebrew-calendar-m9',
139 'hebrew-calendar-m10', 'hebrew-calendar-m11', 'hebrew-calendar-m12',
140 'hebrew-calendar-m6a', 'hebrew-calendar-m6b'
143 static public $mHebrewCalendarMonthGenMsgs = array(
144 'hebrew-calendar-m1-gen', 'hebrew-calendar-m2-gen', 'hebrew-calendar-m3-gen',
145 'hebrew-calendar-m4-gen', 'hebrew-calendar-m5-gen', 'hebrew-calendar-m6-gen',
146 'hebrew-calendar-m7-gen', 'hebrew-calendar-m8-gen', 'hebrew-calendar-m9-gen',
147 'hebrew-calendar-m10-gen', 'hebrew-calendar-m11-gen', 'hebrew-calendar-m12-gen',
148 'hebrew-calendar-m6a-gen', 'hebrew-calendar-m6b-gen'
151 static public $mHijriCalendarMonthMsgs = array(
152 'hijri-calendar-m1', 'hijri-calendar-m2', 'hijri-calendar-m3',
153 'hijri-calendar-m4', 'hijri-calendar-m5', 'hijri-calendar-m6',
154 'hijri-calendar-m7', 'hijri-calendar-m8', 'hijri-calendar-m9',
155 'hijri-calendar-m10', 'hijri-calendar-m11', 'hijri-calendar-m12'
162 static public $durationIntervals = array(
163 'millennia' => 31556952000,
164 'centuries' => 3155695200,
165 'decades' => 315569520,
166 'years' => 31556952, // 86400 * ( 365 + ( 24 * 3 + 25 ) / 400 )
175 * Cache for language fallbacks.
176 * @see Language::getFallbacksIncludingSiteLanguage
180 static private $fallbackLanguageCache = array();
183 * Get a cached or new language object for a given language code
184 * @param $code String
187 static function factory( $code ) {
188 global $wgDummyLanguageCodes, $wgLangObjCacheSize;
190 if ( isset( $wgDummyLanguageCodes[$code] ) ) {
191 $code = $wgDummyLanguageCodes[$code];
194 // get the language object to process
195 $langObj = isset( self
::$mLangObjCache[$code] )
196 ? self
::$mLangObjCache[$code]
197 : self
::newFromCode( $code );
199 // merge the language object in to get it up front in the cache
200 self
::$mLangObjCache = array_merge( array( $code => $langObj ), self
::$mLangObjCache );
201 // get rid of the oldest ones in case we have an overflow
202 self
::$mLangObjCache = array_slice( self
::$mLangObjCache, 0, $wgLangObjCacheSize, true );
208 * Create a language object for a given language code
209 * @param $code String
210 * @throws MWException
213 protected static function newFromCode( $code ) {
214 // Protect against path traversal below
215 if ( !Language
::isValidCode( $code )
216 ||
strcspn( $code, ":/\\\000" ) !== strlen( $code )
218 throw new MWException( "Invalid language code \"$code\"" );
221 if ( !Language
::isValidBuiltInCode( $code ) ) {
222 // It's not possible to customise this code with class files, so
223 // just return a Language object. This is to support uselang= hacks.
224 $lang = new Language
;
225 $lang->setCode( $code );
229 // Check if there is a language class for the code
230 $class = self
::classFromCode( $code );
231 self
::preloadLanguageClass( $class );
232 if ( class_exists( $class ) ) {
237 // Keep trying the fallback list until we find an existing class
238 $fallbacks = Language
::getFallbacksFor( $code );
239 foreach ( $fallbacks as $fallbackCode ) {
240 if ( !Language
::isValidBuiltInCode( $fallbackCode ) ) {
241 throw new MWException( "Invalid fallback '$fallbackCode' in fallback sequence for '$code'" );
244 $class = self
::classFromCode( $fallbackCode );
245 self
::preloadLanguageClass( $class );
246 if ( class_exists( $class ) ) {
247 $lang = Language
::newFromCode( $fallbackCode );
248 $lang->setCode( $code );
253 throw new MWException( "Invalid fallback sequence for language '$code'" );
257 * Checks whether any localisation is available for that language tag
258 * in MediaWiki (MessagesXx.php exists).
260 * @param string $code Language tag (in lower case)
261 * @return bool Whether language is supported
264 public static function isSupportedLanguage( $code ) {
265 return $code === strtolower( $code ) && is_readable( self
::getMessagesFileName( $code ) );
269 * Returns true if a language code string is a well-formed language tag
270 * according to RFC 5646.
271 * This function only checks well-formedness; it doesn't check that
272 * language, script or variant codes actually exist in the repositories.
274 * Based on regexes by Mark Davis of the Unicode Consortium:
275 * http://unicode.org/repos/cldr/trunk/tools/java/org/unicode/cldr/util/data/langtagRegex.txt
277 * @param $code string
278 * @param $lenient boolean Whether to allow '_' as separator. The default is only '-'.
283 public static function isWellFormedLanguageTag( $code, $lenient = false ) {
286 $alphanum = '[a-z0-9]';
287 $x = 'x'; # private use singleton
288 $singleton = '[a-wy-z]'; # other singleton
289 $s = $lenient ?
'[-_]' : '-';
291 $language = "$alpha{2,8}|$alpha{2,3}$s$alpha{3}";
292 $script = "$alpha{4}"; # ISO 15924
293 $region = "(?:$alpha{2}|$digit{3})"; # ISO 3166-1 alpha-2 or UN M.49
294 $variant = "(?:$alphanum{5,8}|$digit$alphanum{3})";
295 $extension = "$singleton(?:$s$alphanum{2,8})+";
296 $privateUse = "$x(?:$s$alphanum{1,8})+";
298 # Define certain grandfathered codes, since otherwise the regex is pretty useless.
299 # Since these are limited, this is safe even later changes to the registry --
300 # the only oddity is that it might change the type of the tag, and thus
301 # the results from the capturing groups.
302 # http://www.iana.org/assignments/language-subtag-registry
304 $grandfathered = "en{$s}GB{$s}oed"
305 . "|i{$s}(?:ami|bnn|default|enochian|hak|klingon|lux|mingo|navajo|pwn|tao|tay|tsu)"
306 . "|no{$s}(?:bok|nyn)"
307 . "|sgn{$s}(?:BE{$s}(?:fr|nl)|CH{$s}de)"
308 . "|zh{$s}min{$s}nan";
310 $variantList = "$variant(?:$s$variant)*";
311 $extensionList = "$extension(?:$s$extension)*";
313 $langtag = "(?:($language)"
316 . "(?:$s$variantList)?"
317 . "(?:$s$extensionList)?"
318 . "(?:$s$privateUse)?)";
320 # The final breakdown, with capturing groups for each of these components
321 # The variants, extensions, grandfathered, and private-use may have interior '-'
323 $root = "^(?:$langtag|$privateUse|$grandfathered)$";
325 return (bool)preg_match( "/$root/", strtolower( $code ) );
329 * Returns true if a language code string is of a valid form, whether or
330 * not it exists. This includes codes which are used solely for
331 * customisation via the MediaWiki namespace.
333 * @param $code string
337 public static function isValidCode( $code ) {
338 static $cache = array();
339 if ( isset( $cache[$code] ) ) {
340 return $cache[$code];
342 // People think language codes are html safe, so enforce it.
343 // Ideally we should only allow a-zA-Z0-9-
344 // but, .+ and other chars are often used for {{int:}} hacks
345 // see bugs 37564, 37587, 36938
347 strcspn( $code, ":/\\\000&<>'\"" ) === strlen( $code )
348 && !preg_match( Title
::getTitleInvalidRegex(), $code );
350 return $cache[$code];
354 * Returns true if a language code is of a valid form for the purposes of
355 * internal customisation of MediaWiki, via Messages*.php.
357 * @param $code string
359 * @throws MWException
363 public static function isValidBuiltInCode( $code ) {
365 if ( !is_string( $code ) ) {
366 if ( is_object( $code ) ) {
367 $addmsg = " of class " . get_class( $code );
371 $type = gettype( $code );
372 throw new MWException( __METHOD__
. " must be passed a string, $type given$addmsg" );
375 return (bool)preg_match( '/^[a-z0-9-]{2,}$/i', $code );
379 * Returns true if a language code is an IETF tag known to MediaWiki.
381 * @param $code string
386 public static function isKnownLanguageTag( $tag ) {
387 static $coreLanguageNames;
389 // Quick escape for invalid input to avoid exceptions down the line
390 // when code tries to process tags which are not valid at all.
391 if ( !self
::isValidBuiltInCode( $tag ) ) {
395 if ( $coreLanguageNames === null ) {
397 include "$IP/languages/Names.php";
400 if ( isset( $coreLanguageNames[$tag] )
401 || self
::fetchLanguageName( $tag, $tag ) !== ''
411 * @return String Name of the language class
413 public static function classFromCode( $code ) {
414 if ( $code == 'en' ) {
417 return 'Language' . str_replace( '-', '_', ucfirst( $code ) );
422 * Includes language class files
424 * @param $class string Name of the language class
426 public static function preloadLanguageClass( $class ) {
429 if ( $class === 'Language' ) {
433 if ( file_exists( "$IP/languages/classes/$class.php" ) ) {
434 include_once "$IP/languages/classes/$class.php";
439 * Get the LocalisationCache instance
441 * @return LocalisationCache
443 public static function getLocalisationCache() {
444 if ( is_null( self
::$dataCache ) ) {
445 global $wgLocalisationCacheConf;
446 $class = $wgLocalisationCacheConf['class'];
447 self
::$dataCache = new $class( $wgLocalisationCacheConf );
449 return self
::$dataCache;
452 function __construct() {
453 $this->mConverter
= new FakeConverter( $this );
454 // Set the code to the name of the descendant
455 if ( get_class( $this ) == 'Language' ) {
458 $this->mCode
= str_replace( '_', '-', strtolower( substr( get_class( $this ), 8 ) ) );
460 self
::getLocalisationCache();
464 * Reduce memory usage
466 function __destruct() {
467 foreach ( $this as $name => $value ) {
468 unset( $this->$name );
473 * Hook which will be called if this is the content language.
474 * Descendants can use this to register hook functions or modify globals
476 function initContLang() { }
479 * Same as getFallbacksFor for current language.
481 * @deprecated in 1.19
483 function getFallbackLanguageCode() {
484 wfDeprecated( __METHOD__
, '1.19' );
485 return self
::getFallbackFor( $this->mCode
);
492 function getFallbackLanguages() {
493 return self
::getFallbacksFor( $this->mCode
);
497 * Exports $wgBookstoreListEn
500 function getBookstoreList() {
501 return self
::$dataCache->getItem( $this->mCode
, 'bookstoreList' );
505 * Returns an array of localised namespaces indexed by their numbers. If the namespace is not
506 * available in localised form, it will be included in English.
510 public function getNamespaces() {
511 if ( is_null( $this->namespaceNames
) ) {
512 global $wgMetaNamespace, $wgMetaNamespaceTalk, $wgExtraNamespaces;
514 $this->namespaceNames
= self
::$dataCache->getItem( $this->mCode
, 'namespaceNames' );
515 $validNamespaces = MWNamespace
::getCanonicalNamespaces();
517 $this->namespaceNames
= $wgExtraNamespaces +
$this->namespaceNames +
$validNamespaces;
519 $this->namespaceNames
[NS_PROJECT
] = $wgMetaNamespace;
520 if ( $wgMetaNamespaceTalk ) {
521 $this->namespaceNames
[NS_PROJECT_TALK
] = $wgMetaNamespaceTalk;
523 $talk = $this->namespaceNames
[NS_PROJECT_TALK
];
524 $this->namespaceNames
[NS_PROJECT_TALK
] =
525 $this->fixVariableInNamespace( $talk );
528 # Sometimes a language will be localised but not actually exist on this wiki.
529 foreach ( $this->namespaceNames
as $key => $text ) {
530 if ( !isset( $validNamespaces[$key] ) ) {
531 unset( $this->namespaceNames
[$key] );
535 # The above mixing may leave namespaces out of canonical order.
536 # Re-order by namespace ID number...
537 ksort( $this->namespaceNames
);
539 wfRunHooks( 'LanguageGetNamespaces', array( &$this->namespaceNames
) );
541 return $this->namespaceNames
;
545 * Arbitrarily set all of the namespace names at once. Mainly used for testing
546 * @param $namespaces Array of namespaces (id => name)
548 public function setNamespaces( array $namespaces ) {
549 $this->namespaceNames
= $namespaces;
550 $this->mNamespaceIds
= null;
554 * Resets all of the namespace caches. Mainly used for testing
556 public function resetNamespaces() {
557 $this->namespaceNames
= null;
558 $this->mNamespaceIds
= null;
559 $this->namespaceAliases
= null;
563 * A convenience function that returns the same thing as
564 * getNamespaces() except with the array values changed to ' '
565 * where it found '_', useful for producing output to be displayed
566 * e.g. in <select> forms.
570 function getFormattedNamespaces() {
571 $ns = $this->getNamespaces();
572 foreach ( $ns as $k => $v ) {
573 $ns[$k] = strtr( $v, '_', ' ' );
579 * Get a namespace value by key
581 * $mw_ns = $wgContLang->getNsText( NS_MEDIAWIKI );
582 * echo $mw_ns; // prints 'MediaWiki'
585 * @param $index Int: the array key of the namespace to return
586 * @return mixed, string if the namespace value exists, otherwise false
588 function getNsText( $index ) {
589 $ns = $this->getNamespaces();
590 return isset( $ns[$index] ) ?
$ns[$index] : false;
594 * A convenience function that returns the same thing as
595 * getNsText() except with '_' changed to ' ', useful for
599 * $mw_ns = $wgContLang->getFormattedNsText( NS_MEDIAWIKI_TALK );
600 * echo $mw_ns; // prints 'MediaWiki talk'
603 * @param int $index The array key of the namespace to return
604 * @return string Namespace name without underscores (empty string if namespace does not exist)
606 function getFormattedNsText( $index ) {
607 $ns = $this->getNsText( $index );
608 return strtr( $ns, '_', ' ' );
612 * Returns gender-dependent namespace alias if available.
613 * @param $index Int: namespace index
614 * @param $gender String: gender key (male, female... )
618 function getGenderNsText( $index, $gender ) {
619 global $wgExtraGenderNamespaces;
621 $ns = $wgExtraGenderNamespaces + self
::$dataCache->getItem( $this->mCode
, 'namespaceGenderAliases' );
622 return isset( $ns[$index][$gender] ) ?
$ns[$index][$gender] : $this->getNsText( $index );
626 * Whether this language makes distinguishes genders for example in
631 function needsGenderDistinction() {
632 global $wgExtraGenderNamespaces, $wgExtraNamespaces;
633 if ( count( $wgExtraGenderNamespaces ) > 0 ) {
634 // $wgExtraGenderNamespaces overrides everything
636 } elseif ( isset( $wgExtraNamespaces[NS_USER
] ) && isset( $wgExtraNamespaces[NS_USER_TALK
] ) ) {
637 /// @todo There may be other gender namespace than NS_USER & NS_USER_TALK in the future
638 // $wgExtraNamespaces overrides any gender aliases specified in i18n files
641 // Check what is in i18n files
642 $aliases = self
::$dataCache->getItem( $this->mCode
, 'namespaceGenderAliases' );
643 return count( $aliases ) > 0;
648 * Get a namespace key by value, case insensitive.
649 * Only matches namespace names for the current language, not the
650 * canonical ones defined in Namespace.php.
652 * @param $text String
653 * @return mixed An integer if $text is a valid value otherwise false
655 function getLocalNsIndex( $text ) {
656 $lctext = $this->lc( $text );
657 $ids = $this->getNamespaceIds();
658 return isset( $ids[$lctext] ) ?
$ids[$lctext] : false;
664 function getNamespaceAliases() {
665 if ( is_null( $this->namespaceAliases
) ) {
666 $aliases = self
::$dataCache->getItem( $this->mCode
, 'namespaceAliases' );
670 foreach ( $aliases as $name => $index ) {
671 if ( $index === NS_PROJECT_TALK
) {
672 unset( $aliases[$name] );
673 $name = $this->fixVariableInNamespace( $name );
674 $aliases[$name] = $index;
679 global $wgExtraGenderNamespaces;
680 $genders = $wgExtraGenderNamespaces +
(array)self
::$dataCache->getItem( $this->mCode
, 'namespaceGenderAliases' );
681 foreach ( $genders as $index => $forms ) {
682 foreach ( $forms as $alias ) {
683 $aliases[$alias] = $index;
687 # Also add converted namespace names as aliases, to avoid confusion.
688 $convertedNames = array();
689 foreach ( $this->getVariants() as $variant ) {
690 if ( $variant === $this->mCode
) {
693 foreach ( $this->getNamespaces() as $ns => $_ ) {
694 $convertedNames[$this->getConverter()->convertNamespace( $ns, $variant )] = $ns;
698 $this->namespaceAliases
= $aliases +
$convertedNames;
700 return $this->namespaceAliases
;
706 function getNamespaceIds() {
707 if ( is_null( $this->mNamespaceIds
) ) {
708 global $wgNamespaceAliases;
709 # Put namespace names and aliases into a hashtable.
710 # If this is too slow, then we should arrange it so that it is done
711 # before caching. The catch is that at pre-cache time, the above
712 # class-specific fixup hasn't been done.
713 $this->mNamespaceIds
= array();
714 foreach ( $this->getNamespaces() as $index => $name ) {
715 $this->mNamespaceIds
[$this->lc( $name )] = $index;
717 foreach ( $this->getNamespaceAliases() as $name => $index ) {
718 $this->mNamespaceIds
[$this->lc( $name )] = $index;
720 if ( $wgNamespaceAliases ) {
721 foreach ( $wgNamespaceAliases as $name => $index ) {
722 $this->mNamespaceIds
[$this->lc( $name )] = $index;
726 return $this->mNamespaceIds
;
730 * Get a namespace key by value, case insensitive. Canonical namespace
731 * names override custom ones defined for the current language.
733 * @param $text String
734 * @return mixed An integer if $text is a valid value otherwise false
736 function getNsIndex( $text ) {
737 $lctext = $this->lc( $text );
738 $ns = MWNamespace
::getCanonicalIndex( $lctext );
739 if ( $ns !== null ) {
742 $ids = $this->getNamespaceIds();
743 return isset( $ids[$lctext] ) ?
$ids[$lctext] : false;
747 * short names for language variants used for language conversion links.
749 * @param $code String
750 * @param $usemsg bool Use the "variantname-xyz" message if it exists
753 function getVariantname( $code, $usemsg = true ) {
754 $msg = "variantname-$code";
755 if ( $usemsg && wfMessage( $msg )->exists() ) {
756 return $this->getMessageFromDB( $msg );
758 $name = self
::fetchLanguageName( $code );
760 return $name; # if it's defined as a language name, show that
762 # otherwise, output the language code
768 * @param $name string
771 function specialPage( $name ) {
772 $aliases = $this->getSpecialPageAliases();
773 if ( isset( $aliases[$name][0] ) ) {
774 $name = $aliases[$name][0];
776 return $this->getNsText( NS_SPECIAL
) . ':' . $name;
782 function getDatePreferences() {
783 return self
::$dataCache->getItem( $this->mCode
, 'datePreferences' );
789 function getDateFormats() {
790 return self
::$dataCache->getItem( $this->mCode
, 'dateFormats' );
794 * @return array|string
796 function getDefaultDateFormat() {
797 $df = self
::$dataCache->getItem( $this->mCode
, 'defaultDateFormat' );
798 if ( $df === 'dmy or mdy' ) {
799 global $wgAmericanDates;
800 return $wgAmericanDates ?
'mdy' : 'dmy';
809 function getDatePreferenceMigrationMap() {
810 return self
::$dataCache->getItem( $this->mCode
, 'datePreferenceMigrationMap' );
817 function getImageFile( $image ) {
818 return self
::$dataCache->getSubitem( $this->mCode
, 'imageFiles', $image );
824 function getExtraUserToggles() {
825 return (array)self
::$dataCache->getItem( $this->mCode
, 'extraUserToggles' );
832 function getUserToggle( $tog ) {
833 return $this->getMessageFromDB( "tog-$tog" );
837 * Get native language names, indexed by code.
838 * Only those defined in MediaWiki, no other data like CLDR.
839 * If $customisedOnly is true, only returns codes with a messages file
841 * @param $customisedOnly bool
844 * @deprecated in 1.20, use fetchLanguageNames()
846 public static function getLanguageNames( $customisedOnly = false ) {
847 return self
::fetchLanguageNames( null, $customisedOnly ?
'mwfile' : 'mw' );
851 * Get translated language names. This is done on best effort and
852 * by default this is exactly the same as Language::getLanguageNames.
853 * The CLDR extension provides translated names.
854 * @param $code String Language code.
855 * @return Array language code => language name
857 * @deprecated in 1.20, use fetchLanguageNames()
859 public static function getTranslatedLanguageNames( $code ) {
860 return self
::fetchLanguageNames( $code, 'all' );
864 * Get an array of language names, indexed by code.
865 * @param $inLanguage null|string: Code of language in which to return the names
866 * Use null for autonyms (native names)
867 * @param $include string:
868 * 'all' all available languages
869 * 'mw' only if the language is defined in MediaWiki or wgExtraLanguageNames (default)
870 * 'mwfile' only if the language is in 'mw' *and* has a message file
871 * @return array: language code => language name
874 public static function fetchLanguageNames( $inLanguage = null, $include = 'mw' ) {
875 global $wgExtraLanguageNames;
876 static $coreLanguageNames;
878 if ( $coreLanguageNames === null ) {
880 include "$IP/languages/Names.php";
886 # TODO: also include when $inLanguage is null, when this code is more efficient
887 wfRunHooks( 'LanguageGetTranslatedLanguageNames', array( &$names, $inLanguage ) );
890 $mwNames = $wgExtraLanguageNames +
$coreLanguageNames;
891 foreach ( $mwNames as $mwCode => $mwName ) {
892 # - Prefer own MediaWiki native name when not using the hook
893 # - For other names just add if not added through the hook
894 if ( $mwCode === $inLanguage ||
!isset( $names[$mwCode] ) ) {
895 $names[$mwCode] = $mwName;
899 if ( $include === 'all' ) {
904 $coreCodes = array_keys( $mwNames );
905 foreach ( $coreCodes as $coreCode ) {
906 $returnMw[$coreCode] = $names[$coreCode];
909 if ( $include === 'mwfile' ) {
910 $namesMwFile = array();
911 # We do this using a foreach over the codes instead of a directory
912 # loop so that messages files in extensions will work correctly.
913 foreach ( $returnMw as $code => $value ) {
914 if ( is_readable( self
::getMessagesFileName( $code ) ) ) {
915 $namesMwFile[$code] = $names[$code];
920 # 'mw' option; default if it's not one of the other two options (all/mwfile)
925 * @param $code string: The code of the language for which to get the name
926 * @param $inLanguage null|string: Code of language in which to return the name (null for autonyms)
927 * @param $include string: 'all', 'mw' or 'mwfile'; see fetchLanguageNames()
928 * @return string: Language name or empty
931 public static function fetchLanguageName( $code, $inLanguage = null, $include = 'all' ) {
932 $code = strtolower( $code );
933 $array = self
::fetchLanguageNames( $inLanguage, $include );
934 return !array_key_exists( $code, $array ) ?
'' : $array[$code];
938 * Get a message from the MediaWiki namespace.
940 * @param $msg String: message name
943 function getMessageFromDB( $msg ) {
944 return wfMessage( $msg )->inLanguage( $this )->text();
948 * Get the native language name of $code.
949 * Only if defined in MediaWiki, no other data like CLDR.
950 * @param $code string
952 * @deprecated in 1.20, use fetchLanguageName()
954 function getLanguageName( $code ) {
955 return self
::fetchLanguageName( $code );
962 function getMonthName( $key ) {
963 return $this->getMessageFromDB( self
::$mMonthMsgs[$key - 1] );
969 function getMonthNamesArray() {
970 $monthNames = array( '' );
971 for ( $i = 1; $i < 13; $i++
) {
972 $monthNames[] = $this->getMonthName( $i );
981 function getMonthNameGen( $key ) {
982 return $this->getMessageFromDB( self
::$mMonthGenMsgs[$key - 1] );
989 function getMonthAbbreviation( $key ) {
990 return $this->getMessageFromDB( self
::$mMonthAbbrevMsgs[$key - 1] );
996 function getMonthAbbreviationsArray() {
997 $monthNames = array( '' );
998 for ( $i = 1; $i < 13; $i++
) {
999 $monthNames[] = $this->getMonthAbbreviation( $i );
1005 * @param $key string
1008 function getWeekdayName( $key ) {
1009 return $this->getMessageFromDB( self
::$mWeekdayMsgs[$key - 1] );
1013 * @param $key string
1016 function getWeekdayAbbreviation( $key ) {
1017 return $this->getMessageFromDB( self
::$mWeekdayAbbrevMsgs[$key - 1] );
1021 * @param $key string
1024 function getIranianCalendarMonthName( $key ) {
1025 return $this->getMessageFromDB( self
::$mIranianCalendarMonthMsgs[$key - 1] );
1029 * @param $key string
1032 function getHebrewCalendarMonthName( $key ) {
1033 return $this->getMessageFromDB( self
::$mHebrewCalendarMonthMsgs[$key - 1] );
1037 * @param $key string
1040 function getHebrewCalendarMonthNameGen( $key ) {
1041 return $this->getMessageFromDB( self
::$mHebrewCalendarMonthGenMsgs[$key - 1] );
1045 * @param $key string
1048 function getHijriCalendarMonthName( $key ) {
1049 return $this->getMessageFromDB( self
::$mHijriCalendarMonthMsgs[$key - 1] );
1053 * This is a workalike of PHP's date() function, but with better
1054 * internationalisation, a reduced set of format characters, and a better
1057 * Supported format characters are dDjlNwzWFmMntLoYyaAgGhHiscrUeIOPTZ. See
1058 * the PHP manual for definitions. There are a number of extensions, which
1061 * xn Do not translate digits of the next numeric format character
1062 * xN Toggle raw digit (xn) flag, stays set until explicitly unset
1063 * xr Use roman numerals for the next numeric format character
1064 * xh Use hebrew numerals for the next numeric format character
1066 * xg Genitive month name
1068 * xij j (day number) in Iranian calendar
1069 * xiF F (month name) in Iranian calendar
1070 * xin n (month number) in Iranian calendar
1071 * xiy y (two digit year) in Iranian calendar
1072 * xiY Y (full year) in Iranian calendar
1074 * xjj j (day number) in Hebrew calendar
1075 * xjF F (month name) in Hebrew calendar
1076 * xjt t (days in month) in Hebrew calendar
1077 * xjx xg (genitive month name) in Hebrew calendar
1078 * xjn n (month number) in Hebrew calendar
1079 * xjY Y (full year) in Hebrew calendar
1081 * xmj j (day number) in Hijri calendar
1082 * xmF F (month name) in Hijri calendar
1083 * xmn n (month number) in Hijri calendar
1084 * xmY Y (full year) in Hijri calendar
1086 * xkY Y (full year) in Thai solar calendar. Months and days are
1087 * identical to the Gregorian calendar
1088 * xoY Y (full year) in Minguo calendar or Juche year.
1089 * Months and days are identical to the
1090 * Gregorian calendar
1091 * xtY Y (full year) in Japanese nengo. Months and days are
1092 * identical to the Gregorian calendar
1094 * Characters enclosed in double quotes will be considered literal (with
1095 * the quotes themselves removed). Unmatched quotes will be considered
1096 * literal quotes. Example:
1098 * "The month is" F => The month is January
1101 * Backslash escaping is also supported.
1103 * Input timestamp is assumed to be pre-normalized to the desired local
1104 * time zone, if any. Note that the format characters crUeIOPTZ will assume
1105 * $ts is UTC if $zone is not given.
1107 * @param $format String
1108 * @param $ts String: 14-character timestamp
1111 * @param $zone DateTimeZone: Timezone of $ts
1112 * @todo handling of "o" format character for Iranian, Hebrew, Hijri & Thai?
1114 * @throws MWException
1117 function sprintfDate( $format, $ts, DateTimeZone
$zone = null ) {
1122 $dateTimeObj = false;
1131 if ( strlen( $ts ) !== 14 ) {
1132 throw new MWException( __METHOD__
. ": The timestamp $ts should have 14 characters" );
1135 if ( !ctype_digit( $ts ) ) {
1136 throw new MWException( __METHOD__
. ": The timestamp $ts should be a number" );
1139 for ( $p = 0; $p < strlen( $format ); $p++
) {
1141 $code = $format[$p];
1142 if ( $code == 'x' && $p < strlen( $format ) - 1 ) {
1143 $code .= $format[++
$p];
1146 if ( ( $code === 'xi' ||
$code == 'xj' ||
$code == 'xk' ||
$code == 'xm' ||
$code == 'xo' ||
$code == 'xt' ) && $p < strlen( $format ) - 1 ) {
1147 $code .= $format[++
$p];
1158 $rawToggle = !$rawToggle;
1167 $s .= $this->getMonthNameGen( substr( $ts, 4, 2 ) );
1171 $hebrew = self
::tsToHebrew( $ts );
1173 $s .= $this->getHebrewCalendarMonthNameGen( $hebrew[1] );
1176 $num = substr( $ts, 6, 2 );
1179 if ( !$dateTimeObj ) {
1180 $dateTimeObj = DateTime
::createFromFormat(
1181 'YmdHis', $ts, $zone ?
: new DateTimeZone( 'UTC' )
1184 $s .= $this->getWeekdayAbbreviation( $dateTimeObj->format( 'w' ) +
1 );
1187 $num = intval( substr( $ts, 6, 2 ) );
1191 $iranian = self
::tsToIranian( $ts );
1197 $hijri = self
::tsToHijri( $ts );
1203 $hebrew = self
::tsToHebrew( $ts );
1208 if ( !$dateTimeObj ) {
1209 $dateTimeObj = DateTime
::createFromFormat(
1210 'YmdHis', $ts, $zone ?
: new DateTimeZone( 'UTC' )
1213 $s .= $this->getWeekdayName( $dateTimeObj->format( 'w' ) +
1 );
1216 $s .= $this->getMonthName( substr( $ts, 4, 2 ) );
1220 $iranian = self
::tsToIranian( $ts );
1222 $s .= $this->getIranianCalendarMonthName( $iranian[1] );
1226 $hijri = self
::tsToHijri( $ts );
1228 $s .= $this->getHijriCalendarMonthName( $hijri[1] );
1232 $hebrew = self
::tsToHebrew( $ts );
1234 $s .= $this->getHebrewCalendarMonthName( $hebrew[1] );
1237 $num = substr( $ts, 4, 2 );
1240 $s .= $this->getMonthAbbreviation( substr( $ts, 4, 2 ) );
1243 $num = intval( substr( $ts, 4, 2 ) );
1247 $iranian = self
::tsToIranian( $ts );
1253 $hijri = self
::tsToHijri ( $ts );
1259 $hebrew = self
::tsToHebrew( $ts );
1265 $hebrew = self
::tsToHebrew( $ts );
1270 $num = substr( $ts, 0, 4 );
1274 $iranian = self
::tsToIranian( $ts );
1280 $hijri = self
::tsToHijri( $ts );
1286 $hebrew = self
::tsToHebrew( $ts );
1292 $thai = self
::tsToYear( $ts, 'thai' );
1298 $minguo = self
::tsToYear( $ts, 'minguo' );
1304 $tenno = self
::tsToYear( $ts, 'tenno' );
1309 $num = substr( $ts, 2, 2 );
1313 $iranian = self
::tsToIranian( $ts );
1315 $num = substr( $iranian[0], -2 );
1318 $s .= intval( substr( $ts, 8, 2 ) ) < 12 ?
'am' : 'pm';
1321 $s .= intval( substr( $ts, 8, 2 ) ) < 12 ?
'AM' : 'PM';
1324 $h = substr( $ts, 8, 2 );
1325 $num = $h %
12 ?
$h %
12 : 12;
1328 $num = intval( substr( $ts, 8, 2 ) );
1331 $h = substr( $ts, 8, 2 );
1332 $num = sprintf( '%02d', $h %
12 ?
$h %
12 : 12 );
1335 $num = substr( $ts, 8, 2 );
1338 $num = substr( $ts, 10, 2 );
1341 $num = substr( $ts, 12, 2 );
1349 // Pass through string from $dateTimeObj->format()
1350 if ( !$dateTimeObj ) {
1351 $dateTimeObj = DateTime
::createFromFormat(
1352 'YmdHis', $ts, $zone ?
: new DateTimeZone( 'UTC' )
1355 $s .= $dateTimeObj->format( $code );
1367 // Pass through number from $dateTimeObj->format()
1368 if ( !$dateTimeObj ) {
1369 $dateTimeObj = DateTime
::createFromFormat(
1370 'YmdHis', $ts, $zone ?
: new DateTimeZone( 'UTC' )
1373 $num = $dateTimeObj->format( $code );
1376 # Backslash escaping
1377 if ( $p < strlen( $format ) - 1 ) {
1378 $s .= $format[++
$p];
1385 if ( $p < strlen( $format ) - 1 ) {
1386 $endQuote = strpos( $format, '"', $p +
1 );
1387 if ( $endQuote === false ) {
1388 # No terminating quote, assume literal "
1391 $s .= substr( $format, $p +
1, $endQuote - $p - 1 );
1395 # Quote at end of string, assume literal "
1402 if ( $num !== false ) {
1403 if ( $rawToggle ||
$raw ) {
1406 } elseif ( $roman ) {
1407 $s .= Language
::romanNumeral( $num );
1409 } elseif ( $hebrewNum ) {
1410 $s .= self
::hebrewNumeral( $num );
1413 $s .= $this->formatNum( $num, true );
1420 private static $GREG_DAYS = array( 31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31 );
1421 private static $IRANIAN_DAYS = array( 31, 31, 31, 31, 31, 31, 30, 30, 30, 30, 30, 29 );
1424 * Algorithm by Roozbeh Pournader and Mohammad Toossi to convert
1425 * Gregorian dates to Iranian dates. Originally written in C, it
1426 * is released under the terms of GNU Lesser General Public
1427 * License. Conversion to PHP was performed by Niklas Laxström.
1429 * Link: http://www.farsiweb.info/jalali/jalali.c
1435 private static function tsToIranian( $ts ) {
1436 $gy = substr( $ts, 0, 4 ) -1600;
1437 $gm = substr( $ts, 4, 2 ) -1;
1438 $gd = substr( $ts, 6, 2 ) -1;
1440 # Days passed from the beginning (including leap years)
1442 +
floor( ( $gy +
3 ) / 4 )
1443 - floor( ( $gy +
99 ) / 100 )
1444 +
floor( ( $gy +
399 ) / 400 );
1446 // Add days of the past months of this year
1447 for ( $i = 0; $i < $gm; $i++
) {
1448 $gDayNo +
= self
::$GREG_DAYS[$i];
1452 if ( $gm > 1 && ( ( $gy %
4 === 0 && $gy %
100 !== 0 ||
( $gy %
400 == 0 ) ) ) ) {
1456 // Days passed in current month
1457 $gDayNo +
= (int)$gd;
1459 $jDayNo = $gDayNo - 79;
1461 $jNp = floor( $jDayNo / 12053 );
1464 $jy = 979 +
33 * $jNp +
4 * floor( $jDayNo / 1461 );
1467 if ( $jDayNo >= 366 ) {
1468 $jy +
= floor( ( $jDayNo - 1 ) / 365 );
1469 $jDayNo = floor( ( $jDayNo - 1 ) %
365 );
1472 for ( $i = 0; $i < 11 && $jDayNo >= self
::$IRANIAN_DAYS[$i]; $i++
) {
1473 $jDayNo -= self
::$IRANIAN_DAYS[$i];
1479 return array( $jy, $jm, $jd );
1483 * Converting Gregorian dates to Hijri dates.
1485 * Based on a PHP-Nuke block by Sharjeel which is released under GNU/GPL license
1487 * @see http://phpnuke.org/modules.php?name=News&file=article&sid=8234&mode=thread&order=0&thold=0
1493 private static function tsToHijri( $ts ) {
1494 $year = substr( $ts, 0, 4 );
1495 $month = substr( $ts, 4, 2 );
1496 $day = substr( $ts, 6, 2 );
1504 ( $zy > 1582 ) ||
( ( $zy == 1582 ) && ( $zm > 10 ) ) ||
1505 ( ( $zy == 1582 ) && ( $zm == 10 ) && ( $zd > 14 ) )
1507 $zjd = (int)( ( 1461 * ( $zy +
4800 +
(int)( ( $zm - 14 ) / 12 ) ) ) / 4 ) +
1508 (int)( ( 367 * ( $zm - 2 - 12 * ( (int)( ( $zm - 14 ) / 12 ) ) ) ) / 12 ) -
1509 (int)( ( 3 * (int)( ( ( $zy +
4900 +
(int)( ( $zm - 14 ) / 12 ) ) / 100 ) ) ) / 4 ) +
1512 $zjd = 367 * $zy - (int)( ( 7 * ( $zy +
5001 +
(int)( ( $zm - 9 ) / 7 ) ) ) / 4 ) +
1513 (int)( ( 275 * $zm ) / 9 ) +
$zd +
1729777;
1516 $zl = $zjd -1948440 +
10632;
1517 $zn = (int)( ( $zl - 1 ) / 10631 );
1518 $zl = $zl - 10631 * $zn +
354;
1519 $zj = ( (int)( ( 10985 - $zl ) / 5316 ) ) * ( (int)( ( 50 * $zl ) / 17719 ) ) +
( (int)( $zl / 5670 ) ) * ( (int)( ( 43 * $zl ) / 15238 ) );
1520 $zl = $zl - ( (int)( ( 30 - $zj ) / 15 ) ) * ( (int)( ( 17719 * $zj ) / 50 ) ) - ( (int)( $zj / 16 ) ) * ( (int)( ( 15238 * $zj ) / 43 ) ) +
29;
1521 $zm = (int)( ( 24 * $zl ) / 709 );
1522 $zd = $zl - (int)( ( 709 * $zm ) / 24 );
1523 $zy = 30 * $zn +
$zj - 30;
1525 return array( $zy, $zm, $zd );
1529 * Converting Gregorian dates to Hebrew dates.
1531 * Based on a JavaScript code by Abu Mami and Yisrael Hersch
1532 * (abu-mami@kaluach.net, http://www.kaluach.net), who permitted
1533 * to translate the relevant functions into PHP and release them under
1536 * The months are counted from Tishrei = 1. In a leap year, Adar I is 13
1537 * and Adar II is 14. In a non-leap year, Adar is 6.
1543 private static function tsToHebrew( $ts ) {
1545 $year = substr( $ts, 0, 4 );
1546 $month = substr( $ts, 4, 2 );
1547 $day = substr( $ts, 6, 2 );
1549 # Calculate Hebrew year
1550 $hebrewYear = $year +
3760;
1552 # Month number when September = 1, August = 12
1554 if ( $month > 12 ) {
1561 # Calculate day of year from 1 September
1563 for ( $i = 1; $i < $month; $i++
) {
1567 # Check if the year is leap
1568 if ( $year %
400 == 0 ||
( $year %
4 == 0 && $year %
100 > 0 ) ) {
1571 } elseif ( $i == 8 ||
$i == 10 ||
$i == 1 ||
$i == 3 ) {
1578 # Calculate the start of the Hebrew year
1579 $start = self
::hebrewYearStart( $hebrewYear );
1581 # Calculate next year's start
1582 if ( $dayOfYear <= $start ) {
1583 # Day is before the start of the year - it is the previous year
1585 $nextStart = $start;
1589 # Add days since previous year's 1 September
1591 if ( ( $year %
400 == 0 ) ||
( $year %
100 != 0 && $year %
4 == 0 ) ) {
1595 # Start of the new (previous) year
1596 $start = self
::hebrewYearStart( $hebrewYear );
1599 $nextStart = self
::hebrewYearStart( $hebrewYear +
1 );
1602 # Calculate Hebrew day of year
1603 $hebrewDayOfYear = $dayOfYear - $start;
1605 # Difference between year's days
1606 $diff = $nextStart - $start;
1607 # Add 12 (or 13 for leap years) days to ignore the difference between
1608 # Hebrew and Gregorian year (353 at least vs. 365/6) - now the
1609 # difference is only about the year type
1610 if ( ( $year %
400 == 0 ) ||
( $year %
100 != 0 && $year %
4 == 0 ) ) {
1616 # Check the year pattern, and is leap year
1617 # 0 means an incomplete year, 1 means a regular year, 2 means a complete year
1618 # This is mod 30, to work on both leap years (which add 30 days of Adar I)
1619 # and non-leap years
1620 $yearPattern = $diff %
30;
1621 # Check if leap year
1622 $isLeap = $diff >= 30;
1624 # Calculate day in the month from number of day in the Hebrew year
1625 # Don't check Adar - if the day is not in Adar, we will stop before;
1626 # if it is in Adar, we will use it to check if it is Adar I or Adar II
1627 $hebrewDay = $hebrewDayOfYear;
1630 while ( $hebrewMonth <= 12 ) {
1631 # Calculate days in this month
1632 if ( $isLeap && $hebrewMonth == 6 ) {
1633 # Adar in a leap year
1635 # Leap year - has Adar I, with 30 days, and Adar II, with 29 days
1637 if ( $hebrewDay <= $days ) {
1641 # Subtract the days of Adar I
1642 $hebrewDay -= $days;
1645 if ( $hebrewDay <= $days ) {
1651 } elseif ( $hebrewMonth == 2 && $yearPattern == 2 ) {
1652 # Cheshvan in a complete year (otherwise as the rule below)
1654 } elseif ( $hebrewMonth == 3 && $yearPattern == 0 ) {
1655 # Kislev in an incomplete year (otherwise as the rule below)
1658 # Odd months have 30 days, even have 29
1659 $days = 30 - ( $hebrewMonth - 1 ) %
2;
1661 if ( $hebrewDay <= $days ) {
1662 # In the current month
1665 # Subtract the days of the current month
1666 $hebrewDay -= $days;
1667 # Try in the next month
1672 return array( $hebrewYear, $hebrewMonth, $hebrewDay, $days );
1676 * This calculates the Hebrew year start, as days since 1 September.
1677 * Based on Carl Friedrich Gauss algorithm for finding Easter date.
1678 * Used for Hebrew date.
1684 private static function hebrewYearStart( $year ) {
1685 $a = intval( ( 12 * ( $year - 1 ) +
17 ) %
19 );
1686 $b = intval( ( $year - 1 ) %
4 );
1687 $m = 32.044093161144 +
1.5542417966212 * $a +
$b / 4.0 - 0.0031777940220923 * ( $year - 1 );
1691 $Mar = intval( $m );
1697 $c = intval( ( $Mar +
3 * ( $year - 1 ) +
5 * $b +
5 ) %
7 );
1698 if ( $c == 0 && $a > 11 && $m >= 0.89772376543210 ) {
1700 } elseif ( $c == 1 && $a > 6 && $m >= 0.63287037037037 ) {
1702 } elseif ( $c == 2 ||
$c == 4 ||
$c == 6 ) {
1706 $Mar +
= intval( ( $year - 3761 ) / 100 ) - intval( ( $year - 3761 ) / 400 ) - 24;
1711 * Algorithm to convert Gregorian dates to Thai solar dates,
1712 * Minguo dates or Minguo dates.
1714 * Link: http://en.wikipedia.org/wiki/Thai_solar_calendar
1715 * http://en.wikipedia.org/wiki/Minguo_calendar
1716 * http://en.wikipedia.org/wiki/Japanese_era_name
1718 * @param $ts String: 14-character timestamp
1719 * @param $cName String: calender name
1720 * @return Array: converted year, month, day
1722 private static function tsToYear( $ts, $cName ) {
1723 $gy = substr( $ts, 0, 4 );
1724 $gm = substr( $ts, 4, 2 );
1725 $gd = substr( $ts, 6, 2 );
1727 if ( !strcmp( $cName, 'thai' ) ) {
1729 # Add 543 years to the Gregorian calendar
1730 # Months and days are identical
1731 $gy_offset = $gy +
543;
1732 } elseif ( ( !strcmp( $cName, 'minguo' ) ) ||
!strcmp( $cName, 'juche' ) ) {
1734 # Deduct 1911 years from the Gregorian calendar
1735 # Months and days are identical
1736 $gy_offset = $gy - 1911;
1737 } elseif ( !strcmp( $cName, 'tenno' ) ) {
1738 # Nengō dates up to Meiji period
1739 # Deduct years from the Gregorian calendar
1740 # depending on the nengo periods
1741 # Months and days are identical
1742 if ( ( $gy < 1912 ) ||
( ( $gy == 1912 ) && ( $gm < 7 ) ) ||
( ( $gy == 1912 ) && ( $gm == 7 ) && ( $gd < 31 ) ) ) {
1744 $gy_gannen = $gy - 1868 +
1;
1745 $gy_offset = $gy_gannen;
1746 if ( $gy_gannen == 1 ) {
1749 $gy_offset = '明治' . $gy_offset;
1751 ( ( $gy == 1912 ) && ( $gm == 7 ) && ( $gd == 31 ) ) ||
1752 ( ( $gy == 1912 ) && ( $gm >= 8 ) ) ||
1753 ( ( $gy > 1912 ) && ( $gy < 1926 ) ) ||
1754 ( ( $gy == 1926 ) && ( $gm < 12 ) ) ||
1755 ( ( $gy == 1926 ) && ( $gm == 12 ) && ( $gd < 26 ) )
1758 $gy_gannen = $gy - 1912 +
1;
1759 $gy_offset = $gy_gannen;
1760 if ( $gy_gannen == 1 ) {
1763 $gy_offset = '大正' . $gy_offset;
1765 ( ( $gy == 1926 ) && ( $gm == 12 ) && ( $gd >= 26 ) ) ||
1766 ( ( $gy > 1926 ) && ( $gy < 1989 ) ) ||
1767 ( ( $gy == 1989 ) && ( $gm == 1 ) && ( $gd < 8 ) )
1770 $gy_gannen = $gy - 1926 +
1;
1771 $gy_offset = $gy_gannen;
1772 if ( $gy_gannen == 1 ) {
1775 $gy_offset = '昭和' . $gy_offset;
1778 $gy_gannen = $gy - 1989 +
1;
1779 $gy_offset = $gy_gannen;
1780 if ( $gy_gannen == 1 ) {
1783 $gy_offset = '平成' . $gy_offset;
1789 return array( $gy_offset, $gm, $gd );
1793 * Roman number formatting up to 10000
1799 static function romanNumeral( $num ) {
1800 static $table = array(
1801 array( '', 'I', 'II', 'III', 'IV', 'V', 'VI', 'VII', 'VIII', 'IX', 'X' ),
1802 array( '', 'X', 'XX', 'XXX', 'XL', 'L', 'LX', 'LXX', 'LXXX', 'XC', 'C' ),
1803 array( '', 'C', 'CC', 'CCC', 'CD', 'D', 'DC', 'DCC', 'DCCC', 'CM', 'M' ),
1804 array( '', 'M', 'MM', 'MMM', 'MMMM', 'MMMMM', 'MMMMMM', 'MMMMMMM', 'MMMMMMMM', 'MMMMMMMMM', 'MMMMMMMMMM' )
1807 $num = intval( $num );
1808 if ( $num > 10000 ||
$num <= 0 ) {
1813 for ( $pow10 = 1000, $i = 3; $i >= 0; $pow10 /= 10, $i-- ) {
1814 if ( $num >= $pow10 ) {
1815 $s .= $table[$i][(int)floor( $num / $pow10 )];
1817 $num = $num %
$pow10;
1823 * Hebrew Gematria number formatting up to 9999
1829 static function hebrewNumeral( $num ) {
1830 static $table = array(
1831 array( '', 'א', 'ב', 'ג', 'ד', 'ה', 'ו', 'ז', 'ח', 'ט', 'י' ),
1832 array( '', 'י', 'כ', 'ל', 'מ', 'נ', 'ס', 'ע', 'פ', 'צ', 'ק' ),
1833 array( '', 'ק', 'ר', 'ש', 'ת', 'תק', 'תר', 'תש', 'תת', 'תתק', 'תתר' ),
1834 array( '', 'א', 'ב', 'ג', 'ד', 'ה', 'ו', 'ז', 'ח', 'ט', 'י' )
1837 $num = intval( $num );
1838 if ( $num > 9999 ||
$num <= 0 ) {
1843 for ( $pow10 = 1000, $i = 3; $i >= 0; $pow10 /= 10, $i-- ) {
1844 if ( $num >= $pow10 ) {
1845 if ( $num == 15 ||
$num == 16 ) {
1846 $s .= $table[0][9] . $table[0][$num - 9];
1849 $s .= $table[$i][intval( ( $num / $pow10 ) )];
1850 if ( $pow10 == 1000 ) {
1855 $num = $num %
$pow10;
1857 if ( strlen( $s ) == 2 ) {
1860 $str = substr( $s, 0, strlen( $s ) - 2 ) . '"';
1861 $str .= substr( $s, strlen( $s ) - 2, 2 );
1863 $start = substr( $str, 0, strlen( $str ) - 2 );
1864 $end = substr( $str, strlen( $str ) - 2 );
1867 $str = $start . 'ך';
1870 $str = $start . 'ם';
1873 $str = $start . 'ן';
1876 $str = $start . 'ף';
1879 $str = $start . 'ץ';
1886 * Used by date() and time() to adjust the time output.
1888 * @param $ts Int the time in date('YmdHis') format
1889 * @param $tz Mixed: adjust the time by this amount (default false, mean we
1890 * get user timecorrection setting)
1893 function userAdjust( $ts, $tz = false ) {
1894 global $wgUser, $wgLocalTZoffset;
1896 if ( $tz === false ) {
1897 $tz = $wgUser->getOption( 'timecorrection' );
1900 $data = explode( '|', $tz, 3 );
1902 if ( $data[0] == 'ZoneInfo' ) {
1903 wfSuppressWarnings();
1904 $userTZ = timezone_open( $data[2] );
1905 wfRestoreWarnings();
1906 if ( $userTZ !== false ) {
1907 $date = date_create( $ts, timezone_open( 'UTC' ) );
1908 date_timezone_set( $date, $userTZ );
1909 $date = date_format( $date, 'YmdHis' );
1912 # Unrecognized timezone, default to 'Offset' with the stored offset.
1913 $data[0] = 'Offset';
1917 if ( $data[0] == 'System' ||
$tz == '' ) {
1918 # Global offset in minutes.
1919 if ( isset( $wgLocalTZoffset ) ) {
1920 $minDiff = $wgLocalTZoffset;
1922 } elseif ( $data[0] == 'Offset' ) {
1923 $minDiff = intval( $data[1] );
1925 $data = explode( ':', $tz );
1926 if ( count( $data ) == 2 ) {
1927 $data[0] = intval( $data[0] );
1928 $data[1] = intval( $data[1] );
1929 $minDiff = abs( $data[0] ) * 60 +
$data[1];
1930 if ( $data[0] < 0 ) {
1931 $minDiff = -$minDiff;
1934 $minDiff = intval( $data[0] ) * 60;
1938 # No difference ? Return time unchanged
1939 if ( 0 == $minDiff ) {
1943 wfSuppressWarnings(); // E_STRICT system time bitching
1944 # Generate an adjusted date; take advantage of the fact that mktime
1945 # will normalize out-of-range values so we don't have to split $minDiff
1946 # into hours and minutes.
1948 (int)substr( $ts, 8, 2 ) ), # Hours
1949 (int)substr( $ts, 10, 2 ) +
$minDiff, # Minutes
1950 (int)substr( $ts, 12, 2 ), # Seconds
1951 (int)substr( $ts, 4, 2 ), # Month
1952 (int)substr( $ts, 6, 2 ), # Day
1953 (int)substr( $ts, 0, 4 ) ); # Year
1955 $date = date( 'YmdHis', $t );
1956 wfRestoreWarnings();
1962 * This is meant to be used by time(), date(), and timeanddate() to get
1963 * the date preference they're supposed to use, it should be used in
1967 * function timeanddate([...], $format = true) {
1968 * $datePreference = $this->dateFormat($format);
1973 * @param $usePrefs Mixed: if true, the user's preference is used
1974 * if false, the site/language default is used
1975 * if int/string, assumed to be a format.
1978 function dateFormat( $usePrefs = true ) {
1981 if ( is_bool( $usePrefs ) ) {
1983 $datePreference = $wgUser->getDatePreference();
1985 $datePreference = (string)User
::getDefaultOption( 'date' );
1988 $datePreference = (string)$usePrefs;
1992 if ( $datePreference == '' ) {
1996 return $datePreference;
2000 * Get a format string for a given type and preference
2001 * @param $type string May be date, time or both
2002 * @param $pref string The format name as it appears in Messages*.php
2004 * @since 1.22 New type 'pretty' that provides a more readable timestamp format
2008 function getDateFormatString( $type, $pref ) {
2009 if ( !isset( $this->dateFormatStrings
[$type][$pref] ) ) {
2010 if ( $pref == 'default' ) {
2011 $pref = $this->getDefaultDateFormat();
2012 $df = self
::$dataCache->getSubitem( $this->mCode
, 'dateFormats', "$pref $type" );
2014 $df = self
::$dataCache->getSubitem( $this->mCode
, 'dateFormats', "$pref $type" );
2016 if ( $type === 'pretty' && $df === null ) {
2017 $df = $this->getDateFormatString( 'date', $pref );
2020 if ( $df === null ) {
2021 $pref = $this->getDefaultDateFormat();
2022 $df = self
::$dataCache->getSubitem( $this->mCode
, 'dateFormats', "$pref $type" );
2025 $this->dateFormatStrings
[$type][$pref] = $df;
2027 return $this->dateFormatStrings
[$type][$pref];
2031 * @param $ts Mixed: the time format which needs to be turned into a
2032 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2033 * @param $adj Bool: whether to adjust the time output according to the
2034 * user configured offset ($timecorrection)
2035 * @param $format Mixed: true to use user's date format preference
2036 * @param $timecorrection String|bool the time offset as returned by
2037 * validateTimeZone() in Special:Preferences
2040 function date( $ts, $adj = false, $format = true, $timecorrection = false ) {
2041 $ts = wfTimestamp( TS_MW
, $ts );
2043 $ts = $this->userAdjust( $ts, $timecorrection );
2045 $df = $this->getDateFormatString( 'date', $this->dateFormat( $format ) );
2046 return $this->sprintfDate( $df, $ts );
2050 * @param $ts Mixed: the time format which needs to be turned into a
2051 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2052 * @param $adj Bool: whether to adjust the time output according to the
2053 * user configured offset ($timecorrection)
2054 * @param $format Mixed: true to use user's date format preference
2055 * @param $timecorrection String|bool the time offset as returned by
2056 * validateTimeZone() in Special:Preferences
2059 function time( $ts, $adj = false, $format = true, $timecorrection = false ) {
2060 $ts = wfTimestamp( TS_MW
, $ts );
2062 $ts = $this->userAdjust( $ts, $timecorrection );
2064 $df = $this->getDateFormatString( 'time', $this->dateFormat( $format ) );
2065 return $this->sprintfDate( $df, $ts );
2069 * @param $ts Mixed: the time format which needs to be turned into a
2070 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2071 * @param $adj Bool: whether to adjust the time output according to the
2072 * user configured offset ($timecorrection)
2073 * @param $format Mixed: what format to return, if it's false output the
2074 * default one (default true)
2075 * @param $timecorrection String|bool the time offset as returned by
2076 * validateTimeZone() in Special:Preferences
2079 function timeanddate( $ts, $adj = false, $format = true, $timecorrection = false ) {
2080 $ts = wfTimestamp( TS_MW
, $ts );
2082 $ts = $this->userAdjust( $ts, $timecorrection );
2084 $df = $this->getDateFormatString( 'both', $this->dateFormat( $format ) );
2085 return $this->sprintfDate( $df, $ts );
2089 * Takes a number of seconds and turns it into a text using values such as hours and minutes.
2093 * @param integer $seconds The amount of seconds.
2094 * @param array $chosenIntervals The intervals to enable.
2098 public function formatDuration( $seconds, array $chosenIntervals = array() ) {
2099 $intervals = $this->getDurationIntervals( $seconds, $chosenIntervals );
2101 $segments = array();
2103 foreach ( $intervals as $intervalName => $intervalValue ) {
2104 // Messages: duration-seconds, duration-minutes, duration-hours, duration-days, duration-weeks,
2105 // duration-years, duration-decades, duration-centuries, duration-millennia
2106 $message = wfMessage( 'duration-' . $intervalName )->numParams( $intervalValue );
2107 $segments[] = $message->inLanguage( $this )->escaped();
2110 return $this->listToText( $segments );
2114 * Takes a number of seconds and returns an array with a set of corresponding intervals.
2115 * For example 65 will be turned into array( minutes => 1, seconds => 5 ).
2119 * @param integer $seconds The amount of seconds.
2120 * @param array $chosenIntervals The intervals to enable.
2124 public function getDurationIntervals( $seconds, array $chosenIntervals = array() ) {
2125 if ( empty( $chosenIntervals ) ) {
2126 $chosenIntervals = array( 'millennia', 'centuries', 'decades', 'years', 'days', 'hours', 'minutes', 'seconds' );
2129 $intervals = array_intersect_key( self
::$durationIntervals, array_flip( $chosenIntervals ) );
2130 $sortedNames = array_keys( $intervals );
2131 $smallestInterval = array_pop( $sortedNames );
2133 $segments = array();
2135 foreach ( $intervals as $name => $length ) {
2136 $value = floor( $seconds / $length );
2138 if ( $value > 0 ||
( $name == $smallestInterval && empty( $segments ) ) ) {
2139 $seconds -= $value * $length;
2140 $segments[$name] = $value;
2148 * Internal helper function for userDate(), userTime() and userTimeAndDate()
2150 * @param $type String: can be 'date', 'time' or 'both'
2151 * @param $ts Mixed: the time format which needs to be turned into a
2152 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2153 * @param $user User object used to get preferences for timezone and format
2154 * @param $options Array, can contain the following keys:
2155 * - 'timecorrection': time correction, can have the following values:
2156 * - true: use user's preference
2157 * - false: don't use time correction
2158 * - integer: value of time correction in minutes
2159 * - 'format': format to use, can have the following values:
2160 * - true: use user's preference
2161 * - false: use default preference
2162 * - string: format to use
2166 private function internalUserTimeAndDate( $type, $ts, User
$user, array $options ) {
2167 $ts = wfTimestamp( TS_MW
, $ts );
2168 $options +
= array( 'timecorrection' => true, 'format' => true );
2169 if ( $options['timecorrection'] !== false ) {
2170 if ( $options['timecorrection'] === true ) {
2171 $offset = $user->getOption( 'timecorrection' );
2173 $offset = $options['timecorrection'];
2175 $ts = $this->userAdjust( $ts, $offset );
2177 if ( $options['format'] === true ) {
2178 $format = $user->getDatePreference();
2180 $format = $options['format'];
2182 $df = $this->getDateFormatString( $type, $this->dateFormat( $format ) );
2183 return $this->sprintfDate( $df, $ts );
2187 * Get the formatted date for the given timestamp and formatted for
2190 * @param $ts Mixed: the time format which needs to be turned into a
2191 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2192 * @param $user User object used to get preferences for timezone and format
2193 * @param $options Array, can contain the following keys:
2194 * - 'timecorrection': time correction, can have the following values:
2195 * - true: use user's preference
2196 * - false: don't use time correction
2197 * - integer: value of time correction in minutes
2198 * - 'format': format to use, can have the following values:
2199 * - true: use user's preference
2200 * - false: use default preference
2201 * - string: format to use
2205 public function userDate( $ts, User
$user, array $options = array() ) {
2206 return $this->internalUserTimeAndDate( 'date', $ts, $user, $options );
2210 * Get the formatted time for the given timestamp and formatted for
2213 * @param $ts Mixed: the time format which needs to be turned into a
2214 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2215 * @param $user User object used to get preferences for timezone and format
2216 * @param $options Array, can contain the following keys:
2217 * - 'timecorrection': time correction, can have the following values:
2218 * - true: use user's preference
2219 * - false: don't use time correction
2220 * - integer: value of time correction in minutes
2221 * - 'format': format to use, can have the following values:
2222 * - true: use user's preference
2223 * - false: use default preference
2224 * - string: format to use
2228 public function userTime( $ts, User
$user, array $options = array() ) {
2229 return $this->internalUserTimeAndDate( 'time', $ts, $user, $options );
2233 * Get the formatted date and time for the given timestamp and formatted for
2236 * @param $ts Mixed: the time format which needs to be turned into a
2237 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
2238 * @param $user User object used to get preferences for timezone and format
2239 * @param $options Array, can contain the following keys:
2240 * - 'timecorrection': time correction, can have the following values:
2241 * - true: use user's preference
2242 * - false: don't use time correction
2243 * - integer: value of time correction in minutes
2244 * - 'format': format to use, can have the following values:
2245 * - true: use user's preference
2246 * - false: use default preference
2247 * - string: format to use
2251 public function userTimeAndDate( $ts, User
$user, array $options = array() ) {
2252 return $this->internalUserTimeAndDate( 'both', $ts, $user, $options );
2256 * Convert an MWTimestamp into a pretty human-readable timestamp using
2257 * the given user preferences and relative base time.
2259 * DO NOT USE THIS FUNCTION DIRECTLY. Instead, call MWTimestamp::getHumanTimestamp
2260 * on your timestamp object, which will then call this function. Calling
2261 * this function directly will cause hooks to be skipped over.
2263 * @see MWTimestamp::getHumanTimestamp
2264 * @param MWTimestamp $ts Timestamp to prettify
2265 * @param MWTimestamp $relativeTo Base timestamp
2266 * @param User $user User preferences to use
2267 * @return string Human timestamp
2270 public function getHumanTimestamp( MWTimestamp
$ts, MWTimestamp
$relativeTo, User
$user ) {
2271 $diff = $ts->diff( $relativeTo );
2272 $diffDay = (bool)( (int)$ts->timestamp
->format( 'w' ) - (int)$relativeTo->timestamp
->format( 'w' ) );
2273 $days = $diff->days ?
: (int)$diffDay;
2274 if ( $diff->invert ||
$days > 5 && $ts->timestamp
->format( 'Y' ) !== $relativeTo->timestamp
->format( 'Y' ) ) {
2275 // Timestamps are in different years: use full timestamp
2276 // Also do full timestamp for future dates
2278 * @FIXME Add better handling of future timestamps.
2280 $format = $this->getDateFormatString( 'both', $user->getDatePreference() ?
: 'default' );
2281 $ts = $this->sprintfDate( $format, $ts->getTimestamp( TS_MW
) );
2282 } elseif ( $days > 5 ) {
2283 // Timestamps are in same year, but more than 5 days ago: show day and month only.
2284 $format = $this->getDateFormatString( 'pretty', $user->getDatePreference() ?
: 'default' );
2285 $ts = $this->sprintfDate( $format, $ts->getTimestamp( TS_MW
) );
2286 } elseif ( $days > 1 ) {
2287 // Timestamp within the past week: show the day of the week and time
2288 $format = $this->getDateFormatString( 'time', $user->getDatePreference() ?
: 'default' );
2289 $weekday = self
::$mWeekdayMsgs[$ts->timestamp
->format( 'w' )];
2291 // sunday-at, monday-at, tuesday-at, wednesday-at, thursday-at, friday-at, saturday-at
2292 $ts = wfMessage( "$weekday-at" )
2293 ->inLanguage( $this )
2294 ->params( $this->sprintfDate( $format, $ts->getTimestamp( TS_MW
) ) )
2296 } elseif ( $days == 1 ) {
2297 // Timestamp was yesterday: say 'yesterday' and the time.
2298 $format = $this->getDateFormatString( 'time', $user->getDatePreference() ?
: 'default' );
2299 $ts = wfMessage( 'yesterday-at' )
2300 ->inLanguage( $this )
2301 ->params( $this->sprintfDate( $format, $ts->getTimestamp( TS_MW
) ) )
2303 } elseif ( $diff->h
> 1 ||
$diff->h
== 1 && $diff->i
> 30 ) {
2304 // Timestamp was today, but more than 90 minutes ago: say 'today' and the time.
2305 $format = $this->getDateFormatString( 'time', $user->getDatePreference() ?
: 'default' );
2306 $ts = wfMessage( 'today-at' )
2307 ->inLanguage( $this )
2308 ->params( $this->sprintfDate( $format, $ts->getTimestamp( TS_MW
) ) )
2311 // From here on in, the timestamp was soon enough ago so that we can simply say
2312 // XX units ago, e.g., "2 hours ago" or "5 minutes ago"
2313 } elseif ( $diff->h
== 1 ) {
2314 // Less than 90 minutes, but more than an hour ago.
2315 $ts = wfMessage( 'hours-ago' )->inLanguage( $this )->numParams( 1 )->text();
2316 } elseif ( $diff->i
>= 1 ) {
2317 // A few minutes ago.
2318 $ts = wfMessage( 'minutes-ago' )->inLanguage( $this )->numParams( $diff->i
)->text();
2319 } elseif ( $diff->s
>= 30 ) {
2320 // Less than a minute, but more than 30 sec ago.
2321 $ts = wfMessage( 'seconds-ago' )->inLanguage( $this )->numParams( $diff->s
)->text();
2323 // Less than 30 seconds ago.
2324 $ts = wfMessage( 'just-now' )->text();
2331 * @param $key string
2332 * @return array|null
2334 function getMessage( $key ) {
2335 return self
::$dataCache->getSubitem( $this->mCode
, 'messages', $key );
2341 function getAllMessages() {
2342 return self
::$dataCache->getItem( $this->mCode
, 'messages' );
2351 function iconv( $in, $out, $string ) {
2352 # This is a wrapper for iconv in all languages except esperanto,
2353 # which does some nasty x-conversions beforehand
2355 # Even with //IGNORE iconv can whine about illegal characters in
2356 # *input* string. We just ignore those too.
2357 # REF: http://bugs.php.net/bug.php?id=37166
2358 # REF: https://bugzilla.wikimedia.org/show_bug.cgi?id=16885
2359 wfSuppressWarnings();
2360 $text = iconv( $in, $out . '//IGNORE', $string );
2361 wfRestoreWarnings();
2365 // callback functions for uc(), lc(), ucwords(), ucwordbreaks()
2368 * @param $matches array
2369 * @return mixed|string
2371 function ucwordbreaksCallbackAscii( $matches ) {
2372 return $this->ucfirst( $matches[1] );
2376 * @param $matches array
2379 function ucwordbreaksCallbackMB( $matches ) {
2380 return mb_strtoupper( $matches[0] );
2384 * @param $matches array
2387 function ucCallback( $matches ) {
2388 list( $wikiUpperChars ) = self
::getCaseMaps();
2389 return strtr( $matches[1], $wikiUpperChars );
2393 * @param $matches array
2396 function lcCallback( $matches ) {
2397 list( , $wikiLowerChars ) = self
::getCaseMaps();
2398 return strtr( $matches[1], $wikiLowerChars );
2402 * @param $matches array
2405 function ucwordsCallbackMB( $matches ) {
2406 return mb_strtoupper( $matches[0] );
2410 * @param $matches array
2413 function ucwordsCallbackWiki( $matches ) {
2414 list( $wikiUpperChars ) = self
::getCaseMaps();
2415 return strtr( $matches[0], $wikiUpperChars );
2419 * Make a string's first character uppercase
2421 * @param $str string
2425 function ucfirst( $str ) {
2427 if ( $o < 96 ) { // if already uppercase...
2429 } elseif ( $o < 128 ) {
2430 return ucfirst( $str ); // use PHP's ucfirst()
2432 // fall back to more complex logic in case of multibyte strings
2433 return $this->uc( $str, true );
2438 * Convert a string to uppercase
2440 * @param $str string
2441 * @param $first bool
2445 function uc( $str, $first = false ) {
2446 if ( function_exists( 'mb_strtoupper' ) ) {
2448 if ( $this->isMultibyte( $str ) ) {
2449 return mb_strtoupper( mb_substr( $str, 0, 1 ) ) . mb_substr( $str, 1 );
2451 return ucfirst( $str );
2454 return $this->isMultibyte( $str ) ?
mb_strtoupper( $str ) : strtoupper( $str );
2457 if ( $this->isMultibyte( $str ) ) {
2458 $x = $first ?
'^' : '';
2459 return preg_replace_callback(
2460 "/$x([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/",
2461 array( $this, 'ucCallback' ),
2465 return $first ?
ucfirst( $str ) : strtoupper( $str );
2471 * @param $str string
2472 * @return mixed|string
2474 function lcfirst( $str ) {
2477 return strval( $str );
2478 } elseif ( $o >= 128 ) {
2479 return $this->lc( $str, true );
2480 } elseif ( $o > 96 ) {
2483 $str[0] = strtolower( $str[0] );
2489 * @param $str string
2490 * @param $first bool
2491 * @return mixed|string
2493 function lc( $str, $first = false ) {
2494 if ( function_exists( 'mb_strtolower' ) ) {
2496 if ( $this->isMultibyte( $str ) ) {
2497 return mb_strtolower( mb_substr( $str, 0, 1 ) ) . mb_substr( $str, 1 );
2499 return strtolower( substr( $str, 0, 1 ) ) . substr( $str, 1 );
2502 return $this->isMultibyte( $str ) ?
mb_strtolower( $str ) : strtolower( $str );
2505 if ( $this->isMultibyte( $str ) ) {
2506 $x = $first ?
'^' : '';
2507 return preg_replace_callback(
2508 "/$x([A-Z]|[\\xc0-\\xff][\\x80-\\xbf]*)/",
2509 array( $this, 'lcCallback' ),
2513 return $first ?
strtolower( substr( $str, 0, 1 ) ) . substr( $str, 1 ) : strtolower( $str );
2519 * @param $str string
2522 function isMultibyte( $str ) {
2523 return (bool)preg_match( '/[\x80-\xff]/', $str );
2527 * @param $str string
2528 * @return mixed|string
2530 function ucwords( $str ) {
2531 if ( $this->isMultibyte( $str ) ) {
2532 $str = $this->lc( $str );
2534 // regexp to find first letter in each word (i.e. after each space)
2535 $replaceRegexp = "/^([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)| ([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/";
2537 // function to use to capitalize a single char
2538 if ( function_exists( 'mb_strtoupper' ) ) {
2539 return preg_replace_callback(
2541 array( $this, 'ucwordsCallbackMB' ),
2545 return preg_replace_callback(
2547 array( $this, 'ucwordsCallbackWiki' ),
2552 return ucwords( strtolower( $str ) );
2557 * capitalize words at word breaks
2559 * @param $str string
2562 function ucwordbreaks( $str ) {
2563 if ( $this->isMultibyte( $str ) ) {
2564 $str = $this->lc( $str );
2566 // since \b doesn't work for UTF-8, we explicitely define word break chars
2567 $breaks = "[ \-\(\)\}\{\.,\?!]";
2569 // find first letter after word break
2570 $replaceRegexp = "/^([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)|$breaks([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/";
2572 if ( function_exists( 'mb_strtoupper' ) ) {
2573 return preg_replace_callback(
2575 array( $this, 'ucwordbreaksCallbackMB' ),
2579 return preg_replace_callback(
2581 array( $this, 'ucwordsCallbackWiki' ),
2586 return preg_replace_callback(
2587 '/\b([\w\x80-\xff]+)\b/',
2588 array( $this, 'ucwordbreaksCallbackAscii' ),
2595 * Return a case-folded representation of $s
2597 * This is a representation such that caseFold($s1)==caseFold($s2) if $s1
2598 * and $s2 are the same except for the case of their characters. It is not
2599 * necessary for the value returned to make sense when displayed.
2601 * Do *not* perform any other normalisation in this function. If a caller
2602 * uses this function when it should be using a more general normalisation
2603 * function, then fix the caller.
2609 function caseFold( $s ) {
2610 return $this->uc( $s );
2617 function checkTitleEncoding( $s ) {
2618 if ( is_array( $s ) ) {
2619 throw new MWException( 'Given array to checkTitleEncoding.' );
2621 if ( StringUtils
::isUtf8( $s ) ) {
2625 return $this->iconv( $this->fallback8bitEncoding(), 'utf-8', $s );
2631 function fallback8bitEncoding() {
2632 return self
::$dataCache->getItem( $this->mCode
, 'fallback8bitEncoding' );
2636 * Most writing systems use whitespace to break up words.
2637 * Some languages such as Chinese don't conventionally do this,
2638 * which requires special handling when breaking up words for
2643 function hasWordBreaks() {
2648 * Some languages such as Chinese require word segmentation,
2649 * Specify such segmentation when overridden in derived class.
2651 * @param $string String
2654 function segmentByWord( $string ) {
2659 * Some languages have special punctuation need to be normalized.
2660 * Make such changes here.
2662 * @param $string String
2665 function normalizeForSearch( $string ) {
2666 return self
::convertDoubleWidth( $string );
2670 * convert double-width roman characters to single-width.
2671 * range: ff00-ff5f ~= 0020-007f
2673 * @param $string string
2677 protected static function convertDoubleWidth( $string ) {
2678 static $full = null;
2679 static $half = null;
2681 if ( $full === null ) {
2682 $fullWidth = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz";
2683 $halfWidth = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz";
2684 $full = str_split( $fullWidth, 3 );
2685 $half = str_split( $halfWidth );
2688 $string = str_replace( $full, $half, $string );
2693 * @param $string string
2694 * @param $pattern string
2697 protected static function insertSpace( $string, $pattern ) {
2698 $string = preg_replace( $pattern, " $1 ", $string );
2699 $string = preg_replace( '/ +/', ' ', $string );
2704 * @param $termsArray array
2707 function convertForSearchResult( $termsArray ) {
2708 # some languages, e.g. Chinese, need to do a conversion
2709 # in order for search results to be displayed correctly
2714 * Get the first character of a string.
2719 function firstChar( $s ) {
2722 '/^([\x00-\x7f]|[\xc0-\xdf][\x80-\xbf]|' .
2723 '[\xe0-\xef][\x80-\xbf]{2}|[\xf0-\xf7][\x80-\xbf]{3})/',
2728 if ( isset( $matches[1] ) ) {
2729 if ( strlen( $matches[1] ) != 3 ) {
2733 // Break down Hangul syllables to grab the first jamo
2734 $code = utf8ToCodepoint( $matches[1] );
2735 if ( $code < 0xac00 ||
0xd7a4 <= $code ) {
2737 } elseif ( $code < 0xb098 ) {
2738 return "\xe3\x84\xb1";
2739 } elseif ( $code < 0xb2e4 ) {
2740 return "\xe3\x84\xb4";
2741 } elseif ( $code < 0xb77c ) {
2742 return "\xe3\x84\xb7";
2743 } elseif ( $code < 0xb9c8 ) {
2744 return "\xe3\x84\xb9";
2745 } elseif ( $code < 0xbc14 ) {
2746 return "\xe3\x85\x81";
2747 } elseif ( $code < 0xc0ac ) {
2748 return "\xe3\x85\x82";
2749 } elseif ( $code < 0xc544 ) {
2750 return "\xe3\x85\x85";
2751 } elseif ( $code < 0xc790 ) {
2752 return "\xe3\x85\x87";
2753 } elseif ( $code < 0xcc28 ) {
2754 return "\xe3\x85\x88";
2755 } elseif ( $code < 0xce74 ) {
2756 return "\xe3\x85\x8a";
2757 } elseif ( $code < 0xd0c0 ) {
2758 return "\xe3\x85\x8b";
2759 } elseif ( $code < 0xd30c ) {
2760 return "\xe3\x85\x8c";
2761 } elseif ( $code < 0xd558 ) {
2762 return "\xe3\x85\x8d";
2764 return "\xe3\x85\x8e";
2771 function initEncoding() {
2772 # Some languages may have an alternate char encoding option
2773 # (Esperanto X-coding, Japanese furigana conversion, etc)
2774 # If this language is used as the primary content language,
2775 # an override to the defaults can be set here on startup.
2782 function recodeForEdit( $s ) {
2783 # For some languages we'll want to explicitly specify
2784 # which characters make it into the edit box raw
2785 # or are converted in some way or another.
2786 global $wgEditEncoding;
2787 if ( $wgEditEncoding == '' ||
$wgEditEncoding == 'UTF-8' ) {
2790 return $this->iconv( 'UTF-8', $wgEditEncoding, $s );
2798 function recodeInput( $s ) {
2799 # Take the previous into account.
2800 global $wgEditEncoding;
2801 if ( $wgEditEncoding != '' ) {
2802 $enc = $wgEditEncoding;
2806 if ( $enc == 'UTF-8' ) {
2809 return $this->iconv( $enc, 'UTF-8', $s );
2814 * Convert a UTF-8 string to normal form C. In Malayalam and Arabic, this
2815 * also cleans up certain backwards-compatible sequences, converting them
2816 * to the modern Unicode equivalent.
2818 * This is language-specific for performance reasons only.
2824 function normalize( $s ) {
2825 global $wgAllUnicodeFixes;
2826 $s = UtfNormal
::cleanUp( $s );
2827 if ( $wgAllUnicodeFixes ) {
2828 $s = $this->transformUsingPairFile( 'normalize-ar.ser', $s );
2829 $s = $this->transformUsingPairFile( 'normalize-ml.ser', $s );
2836 * Transform a string using serialized data stored in the given file (which
2837 * must be in the serialized subdirectory of $IP). The file contains pairs
2838 * mapping source characters to destination characters.
2840 * The data is cached in process memory. This will go faster if you have the
2841 * FastStringSearch extension.
2843 * @param $file string
2844 * @param $string string
2846 * @throws MWException
2849 function transformUsingPairFile( $file, $string ) {
2850 if ( !isset( $this->transformData
[$file] ) ) {
2851 $data = wfGetPrecompiledData( $file );
2852 if ( $data === false ) {
2853 throw new MWException( __METHOD__
. ": The transformation file $file is missing" );
2855 $this->transformData
[$file] = new ReplacementArray( $data );
2857 return $this->transformData
[$file]->replace( $string );
2861 * For right-to-left language support
2866 return self
::$dataCache->getItem( $this->mCode
, 'rtl' );
2870 * Return the correct HTML 'dir' attribute value for this language.
2874 return $this->isRTL() ?
'rtl' : 'ltr';
2878 * Return 'left' or 'right' as appropriate alignment for line-start
2879 * for this language's text direction.
2881 * Should be equivalent to CSS3 'start' text-align value....
2885 function alignStart() {
2886 return $this->isRTL() ?
'right' : 'left';
2890 * Return 'right' or 'left' as appropriate alignment for line-end
2891 * for this language's text direction.
2893 * Should be equivalent to CSS3 'end' text-align value....
2897 function alignEnd() {
2898 return $this->isRTL() ?
'left' : 'right';
2902 * A hidden direction mark (LRM or RLM), depending on the language direction.
2903 * Unlike getDirMark(), this function returns the character as an HTML entity.
2904 * This function should be used when the output is guaranteed to be HTML,
2905 * because it makes the output HTML source code more readable. When
2906 * the output is plain text or can be escaped, getDirMark() should be used.
2908 * @param $opposite Boolean Get the direction mark opposite to your language
2912 function getDirMarkEntity( $opposite = false ) {
2914 return $this->isRTL() ?
'‎' : '‏';
2916 return $this->isRTL() ?
'‏' : '‎';
2920 * A hidden direction mark (LRM or RLM), depending on the language direction.
2921 * This function produces them as invisible Unicode characters and
2922 * the output may be hard to read and debug, so it should only be used
2923 * when the output is plain text or can be escaped. When the output is
2924 * HTML, use getDirMarkEntity() instead.
2926 * @param $opposite Boolean Get the direction mark opposite to your language
2929 function getDirMark( $opposite = false ) {
2930 $lrm = "\xE2\x80\x8E"; # LEFT-TO-RIGHT MARK, commonly abbreviated LRM
2931 $rlm = "\xE2\x80\x8F"; # RIGHT-TO-LEFT MARK, commonly abbreviated RLM
2933 return $this->isRTL() ?
$lrm : $rlm;
2935 return $this->isRTL() ?
$rlm : $lrm;
2941 function capitalizeAllNouns() {
2942 return self
::$dataCache->getItem( $this->mCode
, 'capitalizeAllNouns' );
2946 * An arrow, depending on the language direction.
2948 * @param $direction String: the direction of the arrow: forwards (default), backwards, left, right, up, down.
2951 function getArrow( $direction = 'forwards' ) {
2952 switch ( $direction ) {
2954 return $this->isRTL() ?
'←' : '→';
2956 return $this->isRTL() ?
'→' : '←';
2969 * To allow "foo[[bar]]" to extend the link over the whole word "foobar"
2973 function linkPrefixExtension() {
2974 return self
::$dataCache->getItem( $this->mCode
, 'linkPrefixExtension' );
2978 * Get all magic words from cache.
2981 function getMagicWords() {
2982 return self
::$dataCache->getItem( $this->mCode
, 'magicWords' );
2986 * Run the LanguageGetMagic hook once.
2988 protected function doMagicHook() {
2989 if ( $this->mMagicHookDone
) {
2992 $this->mMagicHookDone
= true;
2993 wfProfileIn( 'LanguageGetMagic' );
2994 wfRunHooks( 'LanguageGetMagic', array( &$this->mMagicExtensions
, $this->getCode() ) );
2995 wfProfileOut( 'LanguageGetMagic' );
2999 * Fill a MagicWord object with data from here
3003 function getMagic( $mw ) {
3004 // Saves a function call
3005 if ( ! $this->mMagicHookDone
) {
3006 $this->doMagicHook();
3009 if ( isset( $this->mMagicExtensions
[$mw->mId
] ) ) {
3010 $rawEntry = $this->mMagicExtensions
[$mw->mId
];
3012 $rawEntry = self
::$dataCache->getSubitem(
3013 $this->mCode
, 'magicWords', $mw->mId
);
3016 if ( !is_array( $rawEntry ) ) {
3017 error_log( "\"$rawEntry\" is not a valid magic word for \"$mw->mId\"" );
3019 $mw->mCaseSensitive
= $rawEntry[0];
3020 $mw->mSynonyms
= array_slice( $rawEntry, 1 );
3025 * Add magic words to the extension array
3027 * @param $newWords array
3029 function addMagicWordsByLang( $newWords ) {
3030 $fallbackChain = $this->getFallbackLanguages();
3031 $fallbackChain = array_reverse( $fallbackChain );
3032 foreach ( $fallbackChain as $code ) {
3033 if ( isset( $newWords[$code] ) ) {
3034 $this->mMagicExtensions
= $newWords[$code] +
$this->mMagicExtensions
;
3040 * Get special page names, as an associative array
3041 * case folded alias => real name
3043 function getSpecialPageAliases() {
3044 // Cache aliases because it may be slow to load them
3045 if ( is_null( $this->mExtendedSpecialPageAliases
) ) {
3047 $this->mExtendedSpecialPageAliases
=
3048 self
::$dataCache->getItem( $this->mCode
, 'specialPageAliases' );
3049 wfRunHooks( 'LanguageGetSpecialPageAliases',
3050 array( &$this->mExtendedSpecialPageAliases
, $this->getCode() ) );
3053 return $this->mExtendedSpecialPageAliases
;
3057 * Italic is unsuitable for some languages
3059 * @param $text String: the text to be emphasized.
3062 function emphasize( $text ) {
3063 return "<em>$text</em>";
3067 * Normally we output all numbers in plain en_US style, that is
3068 * 293,291.235 for twohundredninetythreethousand-twohundredninetyone
3069 * point twohundredthirtyfive. However this is not suitable for all
3070 * languages, some such as Punjabi want ੨੯੩,੨੯੫.੨੩੫ and others such as
3071 * Icelandic just want to use commas instead of dots, and dots instead
3072 * of commas like "293.291,235".
3074 * An example of this function being called:
3076 * wfMessage( 'message' )->numParams( $num )->text()
3079 * See LanguageGu.php for the Gujarati implementation and
3080 * $separatorTransformTable on MessageIs.php for
3081 * the , => . and . => , implementation.
3083 * @todo check if it's viable to use localeconv() for the decimal
3085 * @param $number Mixed: the string to be formatted, should be an integer
3086 * or a floating point number.
3087 * @param $nocommafy Bool: set to true for special numbers like dates
3090 public function formatNum( $number, $nocommafy = false ) {
3091 global $wgTranslateNumerals;
3092 if ( !$nocommafy ) {
3093 $number = $this->commafy( $number );
3094 $s = $this->separatorTransformTable();
3096 $number = strtr( $number, $s );
3100 if ( $wgTranslateNumerals ) {
3101 $s = $this->digitTransformTable();
3103 $number = strtr( $number, $s );
3111 * Front-end for non-commafied formatNum
3113 * @param mixed $number the string to be formatted, should be an integer
3114 * or a floating point number.
3118 public function formatNumNoSeparators( $number ) {
3119 return $this->formatNum( $number, true );
3123 * @param $number string
3126 function parseFormattedNumber( $number ) {
3127 $s = $this->digitTransformTable();
3129 $number = strtr( $number, array_flip( $s ) );
3132 $s = $this->separatorTransformTable();
3134 $number = strtr( $number, array_flip( $s ) );
3137 $number = strtr( $number, array( ',' => '' ) );
3142 * Adds commas to a given number
3144 * @param $number mixed
3147 function commafy( $number ) {
3148 $digitGroupingPattern = $this->digitGroupingPattern();
3149 if ( $number === null ) {
3153 if ( !$digitGroupingPattern ||
$digitGroupingPattern === "###,###,###" ) {
3154 // default grouping is at thousands, use the same for ###,###,### pattern too.
3155 return strrev( (string)preg_replace( '/(\d{3})(?=\d)(?!\d*\.)/', '$1,', strrev( $number ) ) );
3157 // Ref: http://cldr.unicode.org/translation/number-patterns
3159 if ( intval( $number ) < 0 ) {
3160 // For negative numbers apply the algorithm like positive number and add sign.
3162 $number = substr( $number, 1 );
3164 $integerPart = array();
3165 $decimalPart = array();
3166 $numMatches = preg_match_all( "/(#+)/", $digitGroupingPattern, $matches );
3167 preg_match( "/\d+/", $number, $integerPart );
3168 preg_match( "/\.\d*/", $number, $decimalPart );
3169 $groupedNumber = ( count( $decimalPart ) > 0 ) ?
$decimalPart[0] : "";
3170 if ( $groupedNumber === $number ) {
3171 // the string does not have any number part. Eg: .12345
3172 return $sign . $groupedNumber;
3174 $start = $end = strlen( $integerPart[0] );
3175 while ( $start > 0 ) {
3176 $match = $matches[0][$numMatches - 1];
3177 $matchLen = strlen( $match );
3178 $start = $end - $matchLen;
3182 $groupedNumber = substr( $number, $start, $end -$start ) . $groupedNumber;
3184 if ( $numMatches > 1 ) {
3185 // use the last pattern for the rest of the number
3189 $groupedNumber = "," . $groupedNumber;
3192 return $sign . $groupedNumber;
3199 function digitGroupingPattern() {
3200 return self
::$dataCache->getItem( $this->mCode
, 'digitGroupingPattern' );
3206 function digitTransformTable() {
3207 return self
::$dataCache->getItem( $this->mCode
, 'digitTransformTable' );
3213 function separatorTransformTable() {
3214 return self
::$dataCache->getItem( $this->mCode
, 'separatorTransformTable' );
3218 * Take a list of strings and build a locale-friendly comma-separated
3219 * list, using the local comma-separator message.
3220 * The last two strings are chained with an "and".
3221 * NOTE: This function will only work with standard numeric array keys (0, 1, 2…)
3226 function listToText( array $l ) {
3227 $m = count( $l ) - 1;
3232 $and = $this->getMessageFromDB( 'and' );
3233 $space = $this->getMessageFromDB( 'word-separator' );
3235 $comma = $this->getMessageFromDB( 'comma-separator' );
3239 for ( $i = $m - 1; $i >= 0; $i-- ) {
3240 if ( $i == $m - 1 ) {
3241 $s = $l[$i] . $and . $space . $s;
3243 $s = $l[$i] . $comma . $s;
3250 * Take a list of strings and build a locale-friendly comma-separated
3251 * list, using the local comma-separator message.
3252 * @param $list array of strings to put in a comma list
3255 function commaList( array $list ) {
3257 wfMessage( 'comma-separator' )->inLanguage( $this )->escaped(),
3263 * Take a list of strings and build a locale-friendly semicolon-separated
3264 * list, using the local semicolon-separator message.
3265 * @param $list array of strings to put in a semicolon list
3268 function semicolonList( array $list ) {
3270 wfMessage( 'semicolon-separator' )->inLanguage( $this )->escaped(),
3276 * Same as commaList, but separate it with the pipe instead.
3277 * @param $list array of strings to put in a pipe list
3280 function pipeList( array $list ) {
3282 wfMessage( 'pipe-separator' )->inLanguage( $this )->escaped(),
3288 * Truncate a string to a specified length in bytes, appending an optional
3289 * string (e.g. for ellipses)
3291 * The database offers limited byte lengths for some columns in the database;
3292 * multi-byte character sets mean we need to ensure that only whole characters
3293 * are included, otherwise broken characters can be passed to the user
3295 * If $length is negative, the string will be truncated from the beginning
3297 * @param $string String to truncate
3298 * @param $length Int: maximum length (including ellipses)
3299 * @param $ellipsis String to append to the truncated text
3300 * @param $adjustLength Boolean: Subtract length of ellipsis from $length.
3301 * $adjustLength was introduced in 1.18, before that behaved as if false.
3304 function truncate( $string, $length, $ellipsis = '...', $adjustLength = true ) {
3305 # Use the localized ellipsis character
3306 if ( $ellipsis == '...' ) {
3307 $ellipsis = wfMessage( 'ellipsis' )->inLanguage( $this )->escaped();
3309 # Check if there is no need to truncate
3310 if ( $length == 0 ) {
3311 return $ellipsis; // convention
3312 } elseif ( strlen( $string ) <= abs( $length ) ) {
3313 return $string; // no need to truncate
3315 $stringOriginal = $string;
3316 # If ellipsis length is >= $length then we can't apply $adjustLength
3317 if ( $adjustLength && strlen( $ellipsis ) >= abs( $length ) ) {
3318 $string = $ellipsis; // this can be slightly unexpected
3319 # Otherwise, truncate and add ellipsis...
3321 $eLength = $adjustLength ?
strlen( $ellipsis ) : 0;
3322 if ( $length > 0 ) {
3323 $length -= $eLength;
3324 $string = substr( $string, 0, $length ); // xyz...
3325 $string = $this->removeBadCharLast( $string );
3326 $string = rtrim( $string );
3327 $string = $string . $ellipsis;
3329 $length +
= $eLength;
3330 $string = substr( $string, $length ); // ...xyz
3331 $string = $this->removeBadCharFirst( $string );
3332 $string = ltrim( $string );
3333 $string = $ellipsis . $string;
3336 # Do not truncate if the ellipsis makes the string longer/equal (bug 22181).
3337 # This check is *not* redundant if $adjustLength, due to the single case where
3338 # LEN($ellipsis) > ABS($limit arg); $stringOriginal could be shorter than $string.
3339 if ( strlen( $string ) < strlen( $stringOriginal ) ) {
3342 return $stringOriginal;
3347 * Remove bytes that represent an incomplete Unicode character
3348 * at the end of string (e.g. bytes of the char are missing)
3350 * @param $string String
3353 protected function removeBadCharLast( $string ) {
3354 if ( $string != '' ) {
3355 $char = ord( $string[strlen( $string ) - 1] );
3357 if ( $char >= 0xc0 ) {
3358 # We got the first byte only of a multibyte char; remove it.
3359 $string = substr( $string, 0, -1 );
3360 } elseif ( $char >= 0x80 &&
3361 preg_match( '/^(.*)(?:[\xe0-\xef][\x80-\xbf]|' .
3362 '[\xf0-\xf7][\x80-\xbf]{1,2})$/', $string, $m )
3364 # We chopped in the middle of a character; remove it
3372 * Remove bytes that represent an incomplete Unicode character
3373 * at the start of string (e.g. bytes of the char are missing)
3375 * @param $string String
3378 protected function removeBadCharFirst( $string ) {
3379 if ( $string != '' ) {
3380 $char = ord( $string[0] );
3381 if ( $char >= 0x80 && $char < 0xc0 ) {
3382 # We chopped in the middle of a character; remove the whole thing
3383 $string = preg_replace( '/^[\x80-\xbf]+/', '', $string );
3390 * Truncate a string of valid HTML to a specified length in bytes,
3391 * appending an optional string (e.g. for ellipses), and return valid HTML
3393 * This is only intended for styled/linked text, such as HTML with
3394 * tags like <span> and <a>, were the tags are self-contained (valid HTML).
3395 * Also, this will not detect things like "display:none" CSS.
3397 * Note: since 1.18 you do not need to leave extra room in $length for ellipses.
3399 * @param string $text HTML string to truncate
3400 * @param int $length (zero/positive) Maximum length (including ellipses)
3401 * @param string $ellipsis String to append to the truncated text
3404 function truncateHtml( $text, $length, $ellipsis = '...' ) {
3405 # Use the localized ellipsis character
3406 if ( $ellipsis == '...' ) {
3407 $ellipsis = wfMessage( 'ellipsis' )->inLanguage( $this )->escaped();
3409 # Check if there is clearly no need to truncate
3410 if ( $length <= 0 ) {
3411 return $ellipsis; // no text shown, nothing to format (convention)
3412 } elseif ( strlen( $text ) <= $length ) {
3413 return $text; // string short enough even *with* HTML (short-circuit)
3416 $dispLen = 0; // innerHTML legth so far
3417 $testingEllipsis = false; // checking if ellipses will make string longer/equal?
3418 $tagType = 0; // 0-open, 1-close
3419 $bracketState = 0; // 1-tag start, 2-tag name, 0-neither
3420 $entityState = 0; // 0-not entity, 1-entity
3421 $tag = $ret = ''; // accumulated tag name, accumulated result string
3422 $openTags = array(); // open tag stack
3423 $maybeState = null; // possible truncation state
3425 $textLen = strlen( $text );
3426 $neLength = max( 0, $length - strlen( $ellipsis ) ); // non-ellipsis len if truncated
3427 for ( $pos = 0; true; ++
$pos ) {
3428 # Consider truncation once the display length has reached the maximim.
3429 # We check if $dispLen > 0 to grab tags for the $neLength = 0 case.
3430 # Check that we're not in the middle of a bracket/entity...
3431 if ( $dispLen && $dispLen >= $neLength && $bracketState == 0 && !$entityState ) {
3432 if ( !$testingEllipsis ) {
3433 $testingEllipsis = true;
3434 # Save where we are; we will truncate here unless there turn out to
3435 # be so few remaining characters that truncation is not necessary.
3436 if ( !$maybeState ) { // already saved? ($neLength = 0 case)
3437 $maybeState = array( $ret, $openTags ); // save state
3439 } elseif ( $dispLen > $length && $dispLen > strlen( $ellipsis ) ) {
3440 # String in fact does need truncation, the truncation point was OK.
3441 list( $ret, $openTags ) = $maybeState; // reload state
3442 $ret = $this->removeBadCharLast( $ret ); // multi-byte char fix
3443 $ret .= $ellipsis; // add ellipsis
3447 if ( $pos >= $textLen ) {
3448 break; // extra iteration just for above checks
3451 # Read the next char...
3453 $lastCh = $pos ?
$text[$pos - 1] : '';
3454 $ret .= $ch; // add to result string
3456 $this->truncate_endBracket( $tag, $tagType, $lastCh, $openTags ); // for bad HTML
3457 $entityState = 0; // for bad HTML
3458 $bracketState = 1; // tag started (checking for backslash)
3459 } elseif ( $ch == '>' ) {
3460 $this->truncate_endBracket( $tag, $tagType, $lastCh, $openTags );
3461 $entityState = 0; // for bad HTML
3462 $bracketState = 0; // out of brackets
3463 } elseif ( $bracketState == 1 ) {
3465 $tagType = 1; // close tag (e.g. "</span>")
3467 $tagType = 0; // open tag (e.g. "<span>")
3470 $bracketState = 2; // building tag name
3471 } elseif ( $bracketState == 2 ) {
3475 // Name found (e.g. "<a href=..."), add on tag attributes...
3476 $pos +
= $this->truncate_skip( $ret, $text, "<>", $pos +
1 );
3478 } elseif ( $bracketState == 0 ) {
3479 if ( $entityState ) {
3482 $dispLen++
; // entity is one displayed char
3485 if ( $neLength == 0 && !$maybeState ) {
3486 // Save state without $ch. We want to *hit* the first
3487 // display char (to get tags) but not *use* it if truncating.
3488 $maybeState = array( substr( $ret, 0, -1 ), $openTags );
3491 $entityState = 1; // entity found, (e.g. " ")
3493 $dispLen++
; // this char is displayed
3494 // Add the next $max display text chars after this in one swoop...
3495 $max = ( $testingEllipsis ?
$length : $neLength ) - $dispLen;
3496 $skipped = $this->truncate_skip( $ret, $text, "<>&", $pos +
1, $max );
3497 $dispLen +
= $skipped;
3503 // Close the last tag if left unclosed by bad HTML
3504 $this->truncate_endBracket( $tag, $text[$textLen - 1], $tagType, $openTags );
3505 while ( count( $openTags ) > 0 ) {
3506 $ret .= '</' . array_pop( $openTags ) . '>'; // close open tags
3512 * truncateHtml() helper function
3513 * like strcspn() but adds the skipped chars to $ret
3522 private function truncate_skip( &$ret, $text, $search, $start, $len = null ) {
3523 if ( $len === null ) {
3524 $len = -1; // -1 means "no limit" for strcspn
3525 } elseif ( $len < 0 ) {
3529 if ( $start < strlen( $text ) ) {
3530 $skipCount = strcspn( $text, $search, $start, $len );
3531 $ret .= substr( $text, $start, $skipCount );
3537 * truncateHtml() helper function
3538 * (a) push or pop $tag from $openTags as needed
3539 * (b) clear $tag value
3540 * @param &$tag string Current HTML tag name we are looking at
3541 * @param $tagType int (0-open tag, 1-close tag)
3542 * @param $lastCh string Character before the '>' that ended this tag
3543 * @param &$openTags array Open tag stack (not accounting for $tag)
3545 private function truncate_endBracket( &$tag, $tagType, $lastCh, &$openTags ) {
3546 $tag = ltrim( $tag );
3548 if ( $tagType == 0 && $lastCh != '/' ) {
3549 $openTags[] = $tag; // tag opened (didn't close itself)
3550 } elseif ( $tagType == 1 ) {
3551 if ( $openTags && $tag == $openTags[count( $openTags ) - 1] ) {
3552 array_pop( $openTags ); // tag closed
3560 * Grammatical transformations, needed for inflected languages
3561 * Invoked by putting {{grammar:case|word}} in a message
3563 * @param $word string
3564 * @param $case string
3567 function convertGrammar( $word, $case ) {
3568 global $wgGrammarForms;
3569 if ( isset( $wgGrammarForms[$this->getCode()][$case][$word] ) ) {
3570 return $wgGrammarForms[$this->getCode()][$case][$word];
3575 * Get the grammar forms for the content language
3576 * @return array of grammar forms
3579 function getGrammarForms() {
3580 global $wgGrammarForms;
3581 if ( isset( $wgGrammarForms[$this->getCode()] ) && is_array( $wgGrammarForms[$this->getCode()] ) ) {
3582 return $wgGrammarForms[$this->getCode()];
3587 * Provides an alternative text depending on specified gender.
3588 * Usage {{gender:username|masculine|feminine|neutral}}.
3589 * username is optional, in which case the gender of current user is used,
3590 * but only in (some) interface messages; otherwise default gender is used.
3592 * If no forms are given, an empty string is returned. If only one form is
3593 * given, it will be returned unconditionally. These details are implied by
3594 * the caller and cannot be overridden in subclasses.
3596 * If more than one form is given, the default is to use the neutral one
3597 * if it is specified, and to use the masculine one otherwise. These
3598 * details can be overridden in subclasses.
3600 * @param $gender string
3601 * @param $forms array
3605 function gender( $gender, $forms ) {
3606 if ( !count( $forms ) ) {
3609 $forms = $this->preConvertPlural( $forms, 2 );
3610 if ( $gender === 'male' ) {
3613 if ( $gender === 'female' ) {
3616 return isset( $forms[2] ) ?
$forms[2] : $forms[0];
3620 * Plural form transformations, needed for some languages.
3621 * For example, there are 3 form of plural in Russian and Polish,
3622 * depending on "count mod 10". See [[w:Plural]]
3623 * For English it is pretty simple.
3625 * Invoked by putting {{plural:count|wordform1|wordform2}}
3626 * or {{plural:count|wordform1|wordform2|wordform3}}
3628 * Example: {{plural:{{NUMBEROFARTICLES}}|article|articles}}
3630 * @param $count Integer: non-localized number
3631 * @param $forms Array: different plural forms
3632 * @return string Correct form of plural for $count in this language
3634 function convertPlural( $count, $forms ) {
3635 // Handle explicit n=pluralform cases
3636 $forms = $this->handleExplicitPluralForms( $count, $forms );
3637 if ( is_string( $forms ) ) {
3640 if ( !count( $forms ) ) {
3644 $pluralForm = $this->getPluralRuleIndexNumber( $count );
3645 $pluralForm = min( $pluralForm, count( $forms ) - 1 );
3646 return $forms[$pluralForm];
3650 * Handles explicit plural forms for Language::convertPlural()
3652 * In {{PLURAL:$1|0=nothing|one|many}}, 0=nothing will be returned if $1 equals zero.
3653 * If an explicitly defined plural form matches the $count, then
3654 * string value returned, otherwise array returned for further consideration
3655 * by CLDR rules or overridden convertPlural().
3659 * @param int $count non-localized number
3660 * @param array $forms different plural forms
3662 * @return array|string
3664 protected function handleExplicitPluralForms( $count, array $forms ) {
3665 foreach ( $forms as $index => $form ) {
3666 if ( preg_match( '/\d+=/i', $form ) ) {
3667 $pos = strpos( $form, '=' );
3668 if ( substr( $form, 0, $pos ) === (string) $count ) {
3669 return substr( $form, $pos +
1 );
3671 unset( $forms[$index] );
3674 return array_values( $forms );
3678 * Checks that convertPlural was given an array and pads it to requested
3679 * amount of forms by copying the last one.
3681 * @param $count Integer: How many forms should there be at least
3682 * @param $forms Array of forms given to convertPlural
3683 * @return array Padded array of forms or an exception if not an array
3685 protected function preConvertPlural( /* Array */ $forms, $count ) {
3686 while ( count( $forms ) < $count ) {
3687 $forms[] = $forms[count( $forms ) - 1];
3693 * @todo Maybe translate block durations. Note that this function is somewhat misnamed: it
3694 * deals with translating the *duration* ("1 week", "4 days", etc), not the expiry time
3695 * (which is an absolute timestamp). Please note: do NOT add this blindly, as it is used
3696 * on old expiry lengths recorded in log entries. You'd need to provide the start date to
3699 * @param $str String: the validated block duration in English
3700 * @return string Somehow translated block duration
3701 * @see LanguageFi.php for example implementation
3703 function translateBlockExpiry( $str ) {
3704 $duration = SpecialBlock
::getSuggestedDurations( $this );
3705 foreach ( $duration as $show => $value ) {
3706 if ( strcmp( $str, $value ) == 0 ) {
3707 return htmlspecialchars( trim( $show ) );
3711 // Since usually only infinite or indefinite is only on list, so try
3712 // equivalents if still here.
3713 $indefs = array( 'infinite', 'infinity', 'indefinite' );
3714 if ( in_array( $str, $indefs ) ) {
3715 foreach ( $indefs as $val ) {
3716 $show = array_search( $val, $duration, true );
3717 if ( $show !== false ) {
3718 return htmlspecialchars( trim( $show ) );
3723 // If all else fails, return a standard duration or timestamp description.
3724 $time = strtotime( $str, 0 );
3725 if ( $time === false ) { // Unknown format. Return it as-is in case.
3727 } elseif ( $time !== strtotime( $str, 1 ) ) { // It's a relative timestamp.
3728 // $time is relative to 0 so it's a duration length.
3729 return $this->formatDuration( $time );
3730 } else { // It's an absolute timestamp.
3731 if ( $time === 0 ) {
3732 // wfTimestamp() handles 0 as current time instead of epoch.
3733 return $this->timeanddate( '19700101000000' );
3735 return $this->timeanddate( $time );
3741 * languages like Chinese need to be segmented in order for the diff
3744 * @param $text String
3747 public function segmentForDiff( $text ) {
3752 * and unsegment to show the result
3754 * @param $text String
3757 public function unsegmentForDiff( $text ) {
3762 * Return the LanguageConverter used in the Language
3765 * @return LanguageConverter
3767 public function getConverter() {
3768 return $this->mConverter
;
3772 * convert text to all supported variants
3774 * @param $text string
3777 public function autoConvertToAllVariants( $text ) {
3778 return $this->mConverter
->autoConvertToAllVariants( $text );
3782 * convert text to different variants of a language.
3784 * @param $text string
3787 public function convert( $text ) {
3788 return $this->mConverter
->convert( $text );
3792 * Convert a Title object to a string in the preferred variant
3794 * @param $title Title
3797 public function convertTitle( $title ) {
3798 return $this->mConverter
->convertTitle( $title );
3802 * Convert a namespace index to a string in the preferred variant
3807 public function convertNamespace( $ns ) {
3808 return $this->mConverter
->convertNamespace( $ns );
3812 * Check if this is a language with variants
3816 public function hasVariants() {
3817 return count( $this->getVariants() ) > 1;
3821 * Check if the language has the specific variant
3824 * @param $variant string
3827 public function hasVariant( $variant ) {
3828 return (bool)$this->mConverter
->validateVariant( $variant );
3832 * Put custom tags (e.g. -{ }-) around math to prevent conversion
3834 * @param $text string
3836 * @deprecated since 1.22 is no longer used
3838 public function armourMath( $text ) {
3839 return $this->mConverter
->armourMath( $text );
3843 * Perform output conversion on a string, and encode for safe HTML output.
3844 * @param $text String text to be converted
3845 * @param $isTitle Bool whether this conversion is for the article title
3847 * @todo this should get integrated somewhere sane
3849 public function convertHtml( $text, $isTitle = false ) {
3850 return htmlspecialchars( $this->convert( $text, $isTitle ) );
3854 * @param $key string
3857 public function convertCategoryKey( $key ) {
3858 return $this->mConverter
->convertCategoryKey( $key );
3862 * Get the list of variants supported by this language
3863 * see sample implementation in LanguageZh.php
3865 * @return array an array of language codes
3867 public function getVariants() {
3868 return $this->mConverter
->getVariants();
3874 public function getPreferredVariant() {
3875 return $this->mConverter
->getPreferredVariant();
3881 public function getDefaultVariant() {
3882 return $this->mConverter
->getDefaultVariant();
3888 public function getURLVariant() {
3889 return $this->mConverter
->getURLVariant();
3893 * If a language supports multiple variants, it is
3894 * possible that non-existing link in one variant
3895 * actually exists in another variant. this function
3896 * tries to find it. See e.g. LanguageZh.php
3898 * @param $link String: the name of the link
3899 * @param $nt Mixed: the title object of the link
3900 * @param $ignoreOtherCond Boolean: to disable other conditions when
3901 * we need to transclude a template or update a category's link
3902 * @return null the input parameters may be modified upon return
3904 public function findVariantLink( &$link, &$nt, $ignoreOtherCond = false ) {
3905 $this->mConverter
->findVariantLink( $link, $nt, $ignoreOtherCond );
3909 * If a language supports multiple variants, converts text
3910 * into an array of all possible variants of the text:
3911 * 'variant' => text in that variant
3913 * @deprecated since 1.17 Use autoConvertToAllVariants()
3915 * @param $text string
3919 public function convertLinkToAllVariants( $text ) {
3920 return $this->mConverter
->convertLinkToAllVariants( $text );
3924 * returns language specific options used by User::getPageRenderHash()
3925 * for example, the preferred language variant
3929 function getExtraHashOptions() {
3930 return $this->mConverter
->getExtraHashOptions();
3934 * For languages that support multiple variants, the title of an
3935 * article may be displayed differently in different variants. this
3936 * function returns the apporiate title defined in the body of the article.
3940 public function getParsedTitle() {
3941 return $this->mConverter
->getParsedTitle();
3945 * Prepare external link text for conversion. When the text is
3946 * a URL, it shouldn't be converted, and it'll be wrapped in
3947 * the "raw" tag (-{R| }-) to prevent conversion.
3949 * This function is called "markNoConversion" for historical
3952 * @param $text String: text to be used for external link
3953 * @param $noParse bool: wrap it without confirming it's a real URL first
3954 * @return string the tagged text
3956 public function markNoConversion( $text, $noParse = false ) {
3957 // Excluding protocal-relative URLs may avoid many false positives.
3958 if ( $noParse ||
preg_match( '/^(?:' . wfUrlProtocolsWithoutProtRel() . ')/', $text ) ) {
3959 return $this->mConverter
->markNoConversion( $text );
3966 * A regular expression to match legal word-trailing characters
3967 * which should be merged onto a link of the form [[foo]]bar.
3971 public function linkTrail() {
3972 return self
::$dataCache->getItem( $this->mCode
, 'linkTrail' );
3976 * A regular expression character set to match legal word-prefixing
3977 * characters which should be merged onto a link of the form foo[[bar]].
3981 public function linkPrefixCharset() {
3982 return self
::$dataCache->getItem( $this->mCode
, 'linkPrefixCharset' );
3988 function getLangObj() {
3993 * Get the "parent" language which has a converter to convert a "compatible" language
3994 * (in another variant) to this language (eg. zh for zh-cn, but not en for en-gb).
3996 * @return Language|null
3999 public function getParentLanguage() {
4000 if ( $this->mParentLanguage
!== false ) {
4001 return $this->mParentLanguage
;
4004 $pieces = explode( '-', $this->getCode() );
4006 if ( !in_array( $code, LanguageConverter
::$languagesWithVariants ) ) {
4007 $this->mParentLanguage
= null;
4010 $lang = Language
::factory( $code );
4011 if ( !$lang->hasVariant( $this->getCode() ) ) {
4012 $this->mParentLanguage
= null;
4016 $this->mParentLanguage
= $lang;
4021 * Get the RFC 3066 code for this language object
4023 * NOTE: The return value of this function is NOT HTML-safe and must be escaped with
4024 * htmlspecialchars() or similar
4028 public function getCode() {
4029 return $this->mCode
;
4033 * Get the code in Bcp47 format which we can use
4034 * inside of html lang="" tags.
4036 * NOTE: The return value of this function is NOT HTML-safe and must be escaped with
4037 * htmlspecialchars() or similar.
4042 public function getHtmlCode() {
4043 if ( is_null( $this->mHtmlCode
) ) {
4044 $this->mHtmlCode
= wfBCP47( $this->getCode() );
4046 return $this->mHtmlCode
;
4050 * @param $code string
4052 public function setCode( $code ) {
4053 $this->mCode
= $code;
4054 // Ensure we don't leave incorrect cached data lying around
4055 $this->mHtmlCode
= null;
4056 $this->mParentLanguage
= false;
4060 * Get the name of a file for a certain language code
4061 * @param $prefix string Prepend this to the filename
4062 * @param $code string Language code
4063 * @param $suffix string Append this to the filename
4064 * @throws MWException
4065 * @return string $prefix . $mangledCode . $suffix
4067 public static function getFileName( $prefix = 'Language', $code, $suffix = '.php' ) {
4068 // Protect against path traversal
4069 if ( !Language
::isValidCode( $code )
4070 ||
strcspn( $code, ":/\\\000" ) !== strlen( $code )
4072 throw new MWException( "Invalid language code \"$code\"" );
4075 return $prefix . str_replace( '-', '_', ucfirst( $code ) ) . $suffix;
4079 * Get the language code from a file name. Inverse of getFileName()
4080 * @param $filename string $prefix . $languageCode . $suffix
4081 * @param $prefix string Prefix before the language code
4082 * @param $suffix string Suffix after the language code
4083 * @return string Language code, or false if $prefix or $suffix isn't found
4085 public static function getCodeFromFileName( $filename, $prefix = 'Language', $suffix = '.php' ) {
4087 preg_match( '/' . preg_quote( $prefix, '/' ) . '([A-Z][a-z_]+)' .
4088 preg_quote( $suffix, '/' ) . '/', $filename, $m );
4089 if ( !count( $m ) ) {
4092 return str_replace( '_', '-', strtolower( $m[1] ) );
4096 * @param $code string
4099 public static function getMessagesFileName( $code ) {
4101 $file = self
::getFileName( "$IP/languages/messages/Messages", $code, '.php' );
4102 wfRunHooks( 'Language::getMessagesFileName', array( $code, &$file ) );
4107 * @param $code string
4110 public static function getClassFileName( $code ) {
4112 return self
::getFileName( "$IP/languages/classes/Language", $code, '.php' );
4116 * Get the first fallback for a given language.
4118 * @param $code string
4120 * @return bool|string
4122 public static function getFallbackFor( $code ) {
4123 if ( $code === 'en' ||
!Language
::isValidBuiltInCode( $code ) ) {
4126 $fallbacks = self
::getFallbacksFor( $code );
4127 $first = array_shift( $fallbacks );
4133 * Get the ordered list of fallback languages.
4136 * @param $code string Language code
4139 public static function getFallbacksFor( $code ) {
4140 if ( $code === 'en' ||
!Language
::isValidBuiltInCode( $code ) ) {
4143 $v = self
::getLocalisationCache()->getItem( $code, 'fallback' );
4144 $v = array_map( 'trim', explode( ',', $v ) );
4145 if ( $v[count( $v ) - 1] !== 'en' ) {
4153 * Get the ordered list of fallback languages, ending with the fallback
4154 * language chain for the site language.
4157 * @param string $code Language code
4158 * @return array array( fallbacks, site fallbacks )
4160 public static function getFallbacksIncludingSiteLanguage( $code ) {
4161 global $wgLanguageCode;
4163 // Usually, we will only store a tiny number of fallback chains, so we
4164 // keep them in static memory.
4165 $cacheKey = "{$code}-{$wgLanguageCode}";
4167 if ( !array_key_exists( $cacheKey, self
::$fallbackLanguageCache ) ) {
4168 $fallbacks = self
::getFallbacksFor( $code );
4170 // Append the site's fallback chain, including the site language itself
4171 $siteFallbacks = self
::getFallbacksFor( $wgLanguageCode );
4172 array_unshift( $siteFallbacks, $wgLanguageCode );
4174 // Eliminate any languages already included in the chain
4175 $siteFallbacks = array_diff( $siteFallbacks, $fallbacks );
4177 self
::$fallbackLanguageCache[$cacheKey] = array( $fallbacks, $siteFallbacks );
4179 return self
::$fallbackLanguageCache[$cacheKey];
4183 * Get all messages for a given language
4184 * WARNING: this may take a long time. If you just need all message *keys*
4185 * but need the *contents* of only a few messages, consider using getMessageKeysFor().
4187 * @param $code string
4191 public static function getMessagesFor( $code ) {
4192 return self
::getLocalisationCache()->getItem( $code, 'messages' );
4196 * Get a message for a given language
4198 * @param $key string
4199 * @param $code string
4203 public static function getMessageFor( $key, $code ) {
4204 return self
::getLocalisationCache()->getSubitem( $code, 'messages', $key );
4208 * Get all message keys for a given language. This is a faster alternative to
4209 * array_keys( Language::getMessagesFor( $code ) )
4212 * @param $code string Language code
4213 * @return array of message keys (strings)
4215 public static function getMessageKeysFor( $code ) {
4216 return self
::getLocalisationCache()->getSubItemList( $code, 'messages' );
4223 function fixVariableInNamespace( $talk ) {
4224 if ( strpos( $talk, '$1' ) === false ) {
4228 global $wgMetaNamespace;
4229 $talk = str_replace( '$1', $wgMetaNamespace, $talk );
4231 # Allow grammar transformations
4232 # Allowing full message-style parsing would make simple requests
4233 # such as action=raw much more expensive than they need to be.
4234 # This will hopefully cover most cases.
4235 $talk = preg_replace_callback( '/{{grammar:(.*?)\|(.*?)}}/i',
4236 array( &$this, 'replaceGrammarInNamespace' ), $talk );
4237 return str_replace( ' ', '_', $talk );
4244 function replaceGrammarInNamespace( $m ) {
4245 return $this->convertGrammar( trim( $m[2] ), trim( $m[1] ) );
4249 * @throws MWException
4252 static function getCaseMaps() {
4253 static $wikiUpperChars, $wikiLowerChars;
4254 if ( isset( $wikiUpperChars ) ) {
4255 return array( $wikiUpperChars, $wikiLowerChars );
4258 wfProfileIn( __METHOD__
);
4259 $arr = wfGetPrecompiledData( 'Utf8Case.ser' );
4260 if ( $arr === false ) {
4261 throw new MWException(
4262 "Utf8Case.ser is missing, please run \"make\" in the serialized directory\n" );
4264 $wikiUpperChars = $arr['wikiUpperChars'];
4265 $wikiLowerChars = $arr['wikiLowerChars'];
4266 wfProfileOut( __METHOD__
);
4267 return array( $wikiUpperChars, $wikiLowerChars );
4271 * Decode an expiry (block, protection, etc) which has come from the DB
4273 * @todo FIXME: why are we returnings DBMS-dependent strings???
4275 * @param $expiry String: Database expiry String
4276 * @param $format Bool|Int true to process using language functions, or TS_ constant
4277 * to return the expiry in a given timestamp
4281 public function formatExpiry( $expiry, $format = true ) {
4283 if ( $infinity === null ) {
4284 $infinity = wfGetDB( DB_SLAVE
)->getInfinity();
4287 if ( $expiry == '' ||
$expiry == $infinity ) {
4288 return $format === true
4289 ?
$this->getMessageFromDB( 'infiniteblock' )
4292 return $format === true
4293 ?
$this->timeanddate( $expiry, /* User preference timezone */ true )
4294 : wfTimestamp( $format, $expiry );
4300 * @param $seconds int|float
4301 * @param $format Array Optional
4302 * If $format['avoid'] == 'avoidseconds' - don't mention seconds if $seconds >= 1 hour
4303 * If $format['avoid'] == 'avoidminutes' - don't mention seconds/minutes if $seconds > 48 hours
4304 * If $format['noabbrevs'] is true - use 'seconds' and friends instead of 'seconds-abbrev' and friends
4305 * For backwards compatibility, $format may also be one of the strings 'avoidseconds' or 'avoidminutes'
4308 function formatTimePeriod( $seconds, $format = array() ) {
4309 if ( !is_array( $format ) ) {
4310 $format = array( 'avoid' => $format ); // For backwards compatibility
4312 if ( !isset( $format['avoid'] ) ) {
4313 $format['avoid'] = false;
4315 if ( !isset( $format['noabbrevs' ] ) ) {
4316 $format['noabbrevs'] = false;
4318 $secondsMsg = wfMessage(
4319 $format['noabbrevs'] ?
'seconds' : 'seconds-abbrev' )->inLanguage( $this );
4320 $minutesMsg = wfMessage(
4321 $format['noabbrevs'] ?
'minutes' : 'minutes-abbrev' )->inLanguage( $this );
4322 $hoursMsg = wfMessage(
4323 $format['noabbrevs'] ?
'hours' : 'hours-abbrev' )->inLanguage( $this );
4324 $daysMsg = wfMessage(
4325 $format['noabbrevs'] ?
'days' : 'days-abbrev' )->inLanguage( $this );
4327 if ( round( $seconds * 10 ) < 100 ) {
4328 $s = $this->formatNum( sprintf( "%.1f", round( $seconds * 10 ) / 10 ) );
4329 $s = $secondsMsg->params( $s )->text();
4330 } elseif ( round( $seconds ) < 60 ) {
4331 $s = $this->formatNum( round( $seconds ) );
4332 $s = $secondsMsg->params( $s )->text();
4333 } elseif ( round( $seconds ) < 3600 ) {
4334 $minutes = floor( $seconds / 60 );
4335 $secondsPart = round( fmod( $seconds, 60 ) );
4336 if ( $secondsPart == 60 ) {
4340 $s = $minutesMsg->params( $this->formatNum( $minutes ) )->text();
4342 $s .= $secondsMsg->params( $this->formatNum( $secondsPart ) )->text();
4343 } elseif ( round( $seconds ) <= 2 * 86400 ) {
4344 $hours = floor( $seconds / 3600 );
4345 $minutes = floor( ( $seconds - $hours * 3600 ) / 60 );
4346 $secondsPart = round( $seconds - $hours * 3600 - $minutes * 60 );
4347 if ( $secondsPart == 60 ) {
4351 if ( $minutes == 60 ) {
4355 $s = $hoursMsg->params( $this->formatNum( $hours ) )->text();
4357 $s .= $minutesMsg->params( $this->formatNum( $minutes ) )->text();
4358 if ( !in_array( $format['avoid'], array( 'avoidseconds', 'avoidminutes' ) ) ) {
4359 $s .= ' ' . $secondsMsg->params( $this->formatNum( $secondsPart ) )->text();
4362 $days = floor( $seconds / 86400 );
4363 if ( $format['avoid'] === 'avoidminutes' ) {
4364 $hours = round( ( $seconds - $days * 86400 ) / 3600 );
4365 if ( $hours == 24 ) {
4369 $s = $daysMsg->params( $this->formatNum( $days ) )->text();
4371 $s .= $hoursMsg->params( $this->formatNum( $hours ) )->text();
4372 } elseif ( $format['avoid'] === 'avoidseconds' ) {
4373 $hours = floor( ( $seconds - $days * 86400 ) / 3600 );
4374 $minutes = round( ( $seconds - $days * 86400 - $hours * 3600 ) / 60 );
4375 if ( $minutes == 60 ) {
4379 if ( $hours == 24 ) {
4383 $s = $daysMsg->params( $this->formatNum( $days ) )->text();
4385 $s .= $hoursMsg->params( $this->formatNum( $hours ) )->text();
4387 $s .= $minutesMsg->params( $this->formatNum( $minutes ) )->text();
4389 $s = $daysMsg->params( $this->formatNum( $days ) )->text();
4391 $s .= $this->formatTimePeriod( $seconds - $days * 86400, $format );
4398 * Format a bitrate for output, using an appropriate
4399 * unit (bps, kbps, Mbps, Gbps, Tbps, Pbps, Ebps, Zbps or Ybps) according to the magnitude in question
4401 * This use base 1000. For base 1024 use formatSize(), for another base
4402 * see formatComputingNumbers()
4407 function formatBitrate( $bps ) {
4408 return $this->formatComputingNumbers( $bps, 1000, "bitrate-$1bits" );
4412 * @param $size int Size of the unit
4413 * @param $boundary int Size boundary (1000, or 1024 in most cases)
4414 * @param $messageKey string Message key to be uesd
4417 function formatComputingNumbers( $size, $boundary, $messageKey ) {
4419 return str_replace( '$1', $this->formatNum( $size ),
4420 $this->getMessageFromDB( str_replace( '$1', '', $messageKey ) )
4423 $sizes = array( '', 'kilo', 'mega', 'giga', 'tera', 'peta', 'exa', 'zeta', 'yotta' );
4426 $maxIndex = count( $sizes ) - 1;
4427 while ( $size >= $boundary && $index < $maxIndex ) {
4432 // For small sizes no decimal places necessary
4435 // For MB and bigger two decimal places are smarter
4438 $msg = str_replace( '$1', $sizes[$index], $messageKey );
4440 $size = round( $size, $round );
4441 $text = $this->getMessageFromDB( $msg );
4442 return str_replace( '$1', $this->formatNum( $size ), $text );
4446 * Format a size in bytes for output, using an appropriate
4447 * unit (B, KB, MB, GB, TB, PB, EB, ZB or YB) according to the magnitude in question
4449 * This method use base 1024. For base 1000 use formatBitrate(), for
4450 * another base see formatComputingNumbers()
4452 * @param $size int Size to format
4453 * @return string Plain text (not HTML)
4455 function formatSize( $size ) {
4456 return $this->formatComputingNumbers( $size, 1024, "size-$1bytes" );
4460 * Make a list item, used by various special pages
4462 * @param $page String Page link
4463 * @param $details String Text between brackets
4464 * @param $oppositedm Boolean Add the direction mark opposite to your
4465 * language, to display text properly
4468 function specialList( $page, $details, $oppositedm = true ) {
4469 $dirmark = ( $oppositedm ?
$this->getDirMark( true ) : '' ) .
4470 $this->getDirMark();
4471 $details = $details ?
$dirmark . $this->getMessageFromDB( 'word-separator' ) .
4472 wfMessage( 'parentheses' )->rawParams( $details )->inLanguage( $this )->escaped() : '';
4473 return $page . $details;
4477 * Generate (prev x| next x) (20|50|100...) type links for paging
4479 * @param $title Title object to link
4480 * @param $offset Integer offset parameter
4481 * @param $limit Integer limit parameter
4482 * @param $query array|String optional URL query parameter string
4483 * @param $atend Bool optional param for specified if this is the last page
4486 public function viewPrevNext( Title
$title, $offset, $limit, array $query = array(), $atend = false ) {
4487 // @todo FIXME: Why on earth this needs one message for the text and another one for tooltip?
4489 # Make 'previous' link
4490 $prev = wfMessage( 'prevn' )->inLanguage( $this )->title( $title )->numParams( $limit )->text();
4491 if ( $offset > 0 ) {
4492 $plink = $this->numLink( $title, max( $offset - $limit, 0 ), $limit,
4493 $query, $prev, 'prevn-title', 'mw-prevlink' );
4495 $plink = htmlspecialchars( $prev );
4499 $next = wfMessage( 'nextn' )->inLanguage( $this )->title( $title )->numParams( $limit )->text();
4501 $nlink = htmlspecialchars( $next );
4503 $nlink = $this->numLink( $title, $offset +
$limit, $limit,
4504 $query, $next, 'nextn-title', 'mw-nextlink' );
4507 # Make links to set number of items per page
4508 $numLinks = array();
4509 foreach ( array( 20, 50, 100, 250, 500 ) as $num ) {
4510 $numLinks[] = $this->numLink( $title, $offset, $num,
4511 $query, $this->formatNum( $num ), 'shown-title', 'mw-numlink' );
4514 return wfMessage( 'viewprevnext' )->inLanguage( $this )->title( $title
4515 )->rawParams( $plink, $nlink, $this->pipeList( $numLinks ) )->escaped();
4519 * Helper function for viewPrevNext() that generates links
4521 * @param $title Title object to link
4522 * @param $offset Integer offset parameter
4523 * @param $limit Integer limit parameter
4524 * @param $query Array extra query parameters
4525 * @param $link String text to use for the link; will be escaped
4526 * @param $tooltipMsg String name of the message to use as tooltip
4527 * @param $class String value of the "class" attribute of the link
4528 * @return String HTML fragment
4530 private function numLink( Title
$title, $offset, $limit, array $query, $link, $tooltipMsg, $class ) {
4531 $query = array( 'limit' => $limit, 'offset' => $offset ) +
$query;
4532 $tooltip = wfMessage( $tooltipMsg )->inLanguage( $this )->title( $title )->numParams( $limit )->text();
4533 return Html
::element( 'a', array( 'href' => $title->getLocalURL( $query ),
4534 'title' => $tooltip, 'class' => $class ), $link );
4538 * Get the conversion rule title, if any.
4542 public function getConvRuleTitle() {
4543 return $this->mConverter
->getConvRuleTitle();
4547 * Get the compiled plural rules for the language
4549 * @return array Associative array with plural form, and plural rule as key-value pairs
4551 public function getCompiledPluralRules() {
4552 $pluralRules = self
::$dataCache->getItem( strtolower( $this->mCode
), 'compiledPluralRules' );
4553 $fallbacks = Language
::getFallbacksFor( $this->mCode
);
4554 if ( !$pluralRules ) {
4555 foreach ( $fallbacks as $fallbackCode ) {
4556 $pluralRules = self
::$dataCache->getItem( strtolower( $fallbackCode ), 'compiledPluralRules' );
4557 if ( $pluralRules ) {
4562 return $pluralRules;
4566 * Get the plural rules for the language
4568 * @return array Associative array with plural form number and plural rule as key-value pairs
4570 public function getPluralRules() {
4571 $pluralRules = self
::$dataCache->getItem( strtolower( $this->mCode
), 'pluralRules' );
4572 $fallbacks = Language
::getFallbacksFor( $this->mCode
);
4573 if ( !$pluralRules ) {
4574 foreach ( $fallbacks as $fallbackCode ) {
4575 $pluralRules = self
::$dataCache->getItem( strtolower( $fallbackCode ), 'pluralRules' );
4576 if ( $pluralRules ) {
4581 return $pluralRules;
4585 * Get the plural rule types for the language
4587 * @return array Associative array with plural form number and plural rule type as key-value pairs
4589 public function getPluralRuleTypes() {
4590 $pluralRuleTypes = self
::$dataCache->getItem( strtolower( $this->mCode
), 'pluralRuleTypes' );
4591 $fallbacks = Language
::getFallbacksFor( $this->mCode
);
4592 if ( !$pluralRuleTypes ) {
4593 foreach ( $fallbacks as $fallbackCode ) {
4594 $pluralRuleTypes = self
::$dataCache->getItem( strtolower( $fallbackCode ), 'pluralRuleTypes' );
4595 if ( $pluralRuleTypes ) {
4600 return $pluralRuleTypes;
4604 * Find the index number of the plural rule appropriate for the given number
4605 * @return int The index number of the plural rule
4607 public function getPluralRuleIndexNumber( $number ) {
4608 $pluralRules = $this->getCompiledPluralRules();
4609 $form = CLDRPluralRuleEvaluator
::evaluateCompiled( $number, $pluralRules );
4614 * Find the plural rule type appropriate for the given number
4615 * For example, if the language is set to Arabic, getPluralType(5) should
4618 * @return string The name of the plural rule type, e.g. one, two, few, many
4620 public function getPluralRuleType( $number ) {
4621 $index = $this->getPluralRuleIndexNumber( $number );
4622 $pluralRuleTypes = $this->getPluralRuleTypes();
4623 if ( isset( $pluralRuleTypes[$index] ) ) {
4624 return $pluralRuleTypes[$index];