7 if( !defined( 'MEDIAWIKI' ) ) {
8 echo "This file is part of MediaWiki, it is not a valid entry point.\n";
13 # In general you should not make customizations in these language files
14 # directly, but should use the MediaWiki: special namespace to customize
15 # user interface messages through the wiki.
16 # See http://meta.wikipedia.org/wiki/MediaWiki_namespace
18 # NOTE TO TRANSLATORS: Do not copy this whole file when making translations!
19 # A lot of common constants and a base class with inheritable methods are
20 # defined here, which should not be redefined. See the other LanguageXx.php
25 global $wgLanguageNames;
26 require_once( 'Names.php' );
28 global $wgInputEncoding, $wgOutputEncoding;
31 * These are always UTF-8, they exist only for backwards compatibility
33 $wgInputEncoding = "UTF-8";
34 $wgOutputEncoding = "UTF-8";
36 if( function_exists( 'mb_strtoupper' ) ) {
37 mb_internal_encoding('UTF-8');
40 /* a fake language converter */
43 function FakeConverter($langobj) {$this->mLang
= $langobj;}
44 function convert($t, $i) {return $t;}
45 function parserConvert($t, $p) {return $t;}
46 function getVariants() { return array( $this->mLang
->getCode() ); }
47 function getPreferredVariant() {return $this->mLang
->getCode(); }
48 function findVariantLink(&$l, &$n) {}
49 function getExtraHashOptions() {return '';}
50 function getParsedTitle() {return '';}
51 function markNoConversion($text, $noParse=false) {return $text;}
52 function convertCategoryKey( $key ) {return $key; }
53 function convertLinkToAllVariants($text){ return array( $this->mLang
->getCode() => $text); }
54 function setNoTitleConvert(){}
57 #--------------------------------------------------------------------------
58 # Internationalisation code
59 #--------------------------------------------------------------------------
62 var $mConverter, $mVariants, $mCode, $mLoaded = false;
64 static public $mLocalisationKeys = array( 'fallback', 'namespaceNames',
65 'quickbarSettings', 'skinNames', 'mathNames',
66 'bookstoreList', 'magicWords', 'messages', 'rtl', 'digitTransformTable',
67 'separatorTransformTable', 'fallback8bitEncoding', 'linkPrefixExtension',
68 'defaultUserOptionOverrides', 'linkTrail', 'namespaceAliases',
69 'dateFormats', 'datePreferences', 'datePreferenceMigrationMap',
70 'defaultDateFormat', 'extraUserToggles', 'specialPageAliases' );
72 static public $mMergeableMapKeys = array( 'messages', 'namespaceNames', 'mathNames',
73 'dateFormats', 'defaultUserOptionOverrides', 'magicWords' );
75 static public $mMergeableListKeys = array( 'extraUserToggles' );
77 static public $mMergeableAliasListKeys = array( 'specialPageAliases' );
79 static public $mLocalisationCache = array();
81 static public $mWeekdayMsgs = array(
82 'sunday', 'monday', 'tuesday', 'wednesday', 'thursday',
86 static public $mWeekdayAbbrevMsgs = array(
87 'sun', 'mon', 'tue', 'wed', 'thu', 'fri', 'sat'
90 static public $mMonthMsgs = array(
91 'january', 'february', 'march', 'april', 'may_long', 'june',
92 'july', 'august', 'september', 'october', 'november',
95 static public $mMonthGenMsgs = array(
96 'january-gen', 'february-gen', 'march-gen', 'april-gen', 'may-gen', 'june-gen',
97 'july-gen', 'august-gen', 'september-gen', 'october-gen', 'november-gen',
100 static public $mMonthAbbrevMsgs = array(
101 'jan', 'feb', 'mar', 'apr', 'may', 'jun', 'jul', 'aug',
102 'sep', 'oct', 'nov', 'dec'
106 * Create a language object for a given language code
108 static function factory( $code ) {
110 static $recursionLevel = 0;
112 if ( $code == 'en' ) {
115 $class = 'Language' . str_replace( '-', '_', ucfirst( $code ) );
116 // Preload base classes to work around APC/PHP5 bug
117 if ( file_exists( "$IP/languages/classes/$class.deps.php" ) ) {
118 include_once("$IP/languages/classes/$class.deps.php");
120 if ( file_exists( "$IP/languages/classes/$class.php" ) ) {
121 include_once("$IP/languages/classes/$class.php");
125 if ( $recursionLevel > 5 ) {
126 throw new MWException( "Language fallback loop detected when creating class $class\n" );
129 if( ! class_exists( $class ) ) {
130 $fallback = Language
::getFallbackFor( $code );
132 $lang = Language
::factory( $fallback );
134 $lang->setCode( $code );
142 function __construct() {
143 $this->mConverter
= new FakeConverter($this);
144 // Set the code to the name of the descendant
145 if ( get_class( $this ) == 'Language' ) {
148 $this->mCode
= str_replace( '_', '-', strtolower( substr( get_class( $this ), 8 ) ) );
153 * Hook which will be called if this is the content language.
154 * Descendants can use this to register hook functions or modify globals
156 function initContLang() {}
162 function getDefaultUserOptions() {
163 return User
::getDefaultOptions();
167 * Exports $wgBookstoreListEn
170 function getBookstoreList() {
172 return $this->bookstoreList
;
178 function getNamespaces() {
180 return $this->namespaceNames
;
184 * A convenience function that returns the same thing as
185 * getNamespaces() except with the array values changed to ' '
186 * where it found '_', useful for producing output to be displayed
187 * e.g. in <select> forms.
191 function getFormattedNamespaces() {
192 $ns = $this->getNamespaces();
193 foreach($ns as $k => $v) {
194 $ns[$k] = strtr($v, '_', ' ');
200 * Get a namespace value by key
202 * $mw_ns = $wgContLang->getNsText( NS_MEDIAWIKI );
203 * echo $mw_ns; // prints 'MediaWiki'
206 * @param int $index the array key of the namespace to return
207 * @return mixed, string if the namespace value exists, otherwise false
209 function getNsText( $index ) {
210 $ns = $this->getNamespaces();
211 return isset( $ns[$index] ) ?
$ns[$index] : false;
215 * A convenience function that returns the same thing as
216 * getNsText() except with '_' changed to ' ', useful for
221 function getFormattedNsText( $index ) {
222 $ns = $this->getNsText( $index );
223 return strtr($ns, '_', ' ');
227 * Get a namespace key by value, case insensetive.
229 * @param string $text
230 * @return mixed An integer if $text is a valid value otherwise false
232 function getNsIndex( $text ) {
234 $lctext = $this->lc($text);
235 return isset( $this->mNamespaceIds
[$lctext] ) ?
$this->mNamespaceIds
[$lctext] : false;
239 * short names for language variants used for language conversion links.
241 * @param string $code
244 function getVariantname( $code ) {
245 return $this->getMessageFromDB( "variantname-$code" );
248 function specialPage( $name ) {
249 $aliases = $this->getSpecialPageAliases();
250 if ( isset( $aliases[$name][0] ) ) {
251 $name = $aliases[$name][0];
253 return $this->getNsText(NS_SPECIAL
) . ':' . $name;
256 function getQuickbarSettings() {
258 return $this->quickbarSettings
;
261 function getSkinNames() {
263 return $this->skinNames
;
266 function getMathNames() {
268 return $this->mathNames
;
271 function getDatePreferences() {
273 return $this->datePreferences
;
276 function getDateFormats() {
278 return $this->dateFormats
;
281 function getDefaultDateFormat() {
283 return $this->defaultDateFormat
;
286 function getDatePreferenceMigrationMap() {
288 return $this->datePreferenceMigrationMap
;
291 function getDefaultUserOptionOverrides() {
293 return $this->defaultUserOptionOverrides
;
296 function getExtraUserToggles() {
298 return $this->extraUserToggles
;
301 function getUserToggle( $tog ) {
302 return $this->getMessageFromDB( "tog-$tog" );
306 * Get language names, indexed by code.
307 * If $customisedOnly is true, only returns codes with a messages file
309 function getLanguageNames( $customisedOnly = false ) {
310 global $wgLanguageNames;
311 if ( !$customisedOnly ) {
312 return $wgLanguageNames;
316 $messageFiles = glob( "$IP/languages/messages/Messages*.php" );
318 foreach ( $messageFiles as $file ) {
319 if( preg_match( '/Messages([A-Z][a-z_]+)\.php$/', $file, $m ) ) {
320 $code = str_replace( '_', '-', strtolower( $m[1] ) );
321 if ( isset( $wgLanguageNames[$code] ) ) {
322 $names[$code] = $wgLanguageNames[$code];
330 * Ugly hack to get a message maybe from the MediaWiki namespace, if this
331 * language object is the content or user language.
333 function getMessageFromDB( $msg ) {
334 global $wgContLang, $wgLang;
335 if ( $wgContLang->getCode() == $this->getCode() ) {
337 return wfMsgForContent( $msg );
338 } elseif ( $wgLang->getCode() == $this->getCode() ) {
340 return wfMsg( $msg );
342 # Neither, get from localisation
343 return $this->getMessage( $msg );
347 function getLanguageName( $code ) {
348 global $wgLanguageNames;
349 if ( ! array_key_exists( $code, $wgLanguageNames ) ) {
352 return $wgLanguageNames[$code];
355 function getMonthName( $key ) {
356 return $this->getMessageFromDB( self
::$mMonthMsgs[$key-1] );
359 function getMonthNameGen( $key ) {
360 return $this->getMessageFromDB( self
::$mMonthGenMsgs[$key-1] );
363 function getMonthAbbreviation( $key ) {
364 return $this->getMessageFromDB( self
::$mMonthAbbrevMsgs[$key-1] );
367 function getWeekdayName( $key ) {
368 return $this->getMessageFromDB( self
::$mWeekdayMsgs[$key-1] );
371 function getWeekdayAbbreviation( $key ) {
372 return $this->getMessageFromDB( self
::$mWeekdayAbbrevMsgs[$key-1] );
376 * Used by date() and time() to adjust the time output.
378 * @param int $ts the time in date('YmdHis') format
379 * @param mixed $tz adjust the time by this amount (default false,
380 * mean we get user timecorrection setting)
383 function userAdjust( $ts, $tz = false ) {
384 global $wgUser, $wgLocalTZoffset;
387 $tz = $wgUser->getOption( 'timecorrection' );
390 # minutes and hours differences:
395 # Global offset in minutes.
396 if( isset($wgLocalTZoffset) ) {
397 $hrDiff = $wgLocalTZoffset %
60;
398 $minDiff = $wgLocalTZoffset - ($hrDiff * 60);
400 } elseif ( strpos( $tz, ':' ) !== false ) {
401 $tzArray = explode( ':', $tz );
402 $hrDiff = intval($tzArray[0]);
403 $minDiff = intval($hrDiff < 0 ?
-$tzArray[1] : $tzArray[1]);
405 $hrDiff = intval( $tz );
408 # No difference ? Return time unchanged
409 if ( 0 == $hrDiff && 0 == $minDiff ) { return $ts; }
411 # Generate an adjusted date
413 (int)substr( $ts, 8, 2) ) +
$hrDiff, # Hours
414 (int)substr( $ts, 10, 2 ) +
$minDiff, # Minutes
415 (int)substr( $ts, 12, 2 ), # Seconds
416 (int)substr( $ts, 4, 2 ), # Month
417 (int)substr( $ts, 6, 2 ), # Day
418 (int)substr( $ts, 0, 4 ) ); #Year
419 return date( 'YmdHis', $t );
423 * This is a workalike of PHP's date() function, but with better
424 * internationalisation, a reduced set of format characters, and a better
427 * Supported format characters are dDjlNwzWFmMntLYyaAgGhHiscrU. See the
428 * PHP manual for definitions. There are a number of extensions, which
431 * xn Do not translate digits of the next numeric format character
432 * xN Toggle raw digit (xn) flag, stays set until explicitly unset
433 * xr Use roman numerals for the next numeric format character
435 * xg Genitive month name
437 * Characters enclosed in double quotes will be considered literal (with
438 * the quotes themselves removed). Unmatched quotes will be considered
439 * literal quotes. Example:
441 * "The month is" F => The month is January
444 * Backslash escaping is also supported.
446 * @param string $format
447 * @param string $ts 14-character timestamp
451 function sprintfDate( $format, $ts ) {
457 for ( $p = 0; $p < strlen( $format ); $p++
) {
460 if ( $code == 'x' && $p < strlen( $format ) - 1 ) {
461 $code .= $format[++
$p];
472 $rawToggle = !$rawToggle;
478 $s .= $this->getMonthNameGen( substr( $ts, 4, 2 ) );
481 $num = substr( $ts, 6, 2 );
484 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
485 $s .= $this->getWeekdayAbbreviation( date( 'w', $unix ) +
1 );
488 $num = intval( substr( $ts, 6, 2 ) );
491 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
492 $s .= $this->getWeekdayName( date( 'w', $unix ) +
1 );
495 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
496 $w = date( 'w', $unix );
500 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
501 $num = date( 'w', $unix );
504 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
505 $num = date( 'z', $unix );
508 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
509 $num = date( 'W', $unix );
512 $s .= $this->getMonthName( substr( $ts, 4, 2 ) );
515 $num = substr( $ts, 4, 2 );
518 $s .= $this->getMonthAbbreviation( substr( $ts, 4, 2 ) );
521 $num = intval( substr( $ts, 4, 2 ) );
524 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
525 $num = date( 't', $unix );
528 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
529 $num = date( 'L', $unix );
532 $num = substr( $ts, 0, 4 );
535 $num = substr( $ts, 2, 2 );
538 $s .= intval( substr( $ts, 8, 2 ) ) < 12 ?
'am' : 'pm';
541 $s .= intval( substr( $ts, 8, 2 ) ) < 12 ?
'AM' : 'PM';
544 $h = substr( $ts, 8, 2 );
545 $num = $h %
12 ?
$h %
12 : 12;
548 $num = intval( substr( $ts, 8, 2 ) );
551 $h = substr( $ts, 8, 2 );
552 $num = sprintf( '%02d', $h %
12 ?
$h %
12 : 12 );
555 $num = substr( $ts, 8, 2 );
558 $num = substr( $ts, 10, 2 );
561 $num = substr( $ts, 12, 2 );
564 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
565 $s .= date( 'c', $unix );
568 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
569 $s .= date( 'r', $unix );
572 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
577 if ( $p < strlen( $format ) - 1 ) {
585 if ( $p < strlen( $format ) - 1 ) {
586 $endQuote = strpos( $format, '"', $p +
1 );
587 if ( $endQuote === false ) {
588 # No terminating quote, assume literal "
591 $s .= substr( $format, $p +
1, $endQuote - $p - 1 );
595 # Quote at end of string, assume literal "
602 if ( $num !== false ) {
603 if ( $rawToggle ||
$raw ) {
606 } elseif ( $roman ) {
607 $s .= self
::romanNumeral( $num );
610 $s .= $this->formatNum( $num, true );
619 * Roman number formatting up to 3000
621 static function romanNumeral( $num ) {
622 static $table = array(
623 array( '', 'I', 'II', 'III', 'IV', 'V', 'VI', 'VII', 'VIII', 'IX', 'X' ),
624 array( '', 'X', 'XX', 'XXX', 'XL', 'L', 'LX', 'LXX', 'LXXX', 'XC', 'C' ),
625 array( '', 'C', 'CC', 'CCC', 'CD', 'D', 'DC', 'DCC', 'DCCC', 'CM', 'M' ),
626 array( '', 'M', 'MM', 'MMM' )
629 $num = intval( $num );
630 if ( $num > 3000 ||
$num <= 0 ) {
635 for ( $pow10 = 1000, $i = 3; $i >= 0; $pow10 /= 10, $i-- ) {
636 if ( $num >= $pow10 ) {
637 $s .= $table[$i][floor($num / $pow10)];
639 $num = $num %
$pow10;
645 * This is meant to be used by time(), date(), and timeanddate() to get
646 * the date preference they're supposed to use, it should be used in
650 * function timeanddate([...], $format = true) {
651 * $datePreference = $this->dateFormat($format);
656 * @param mixed $usePrefs: if true, the user's preference is used
657 * if false, the site/language default is used
658 * if int/string, assumed to be a format.
661 function dateFormat( $usePrefs = true ) {
664 if( is_bool( $usePrefs ) ) {
666 $datePreference = $wgUser->getDatePreference();
668 $options = User
::getDefaultOptions();
669 $datePreference = (string)$options['date'];
672 $datePreference = (string)$usePrefs;
676 if( $datePreference == '' ) {
680 return $datePreference;
685 * @param mixed $ts the time format which needs to be turned into a
686 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
687 * @param bool $adj whether to adjust the time output according to the
688 * user configured offset ($timecorrection)
689 * @param mixed $format true to use user's date format preference
690 * @param string $timecorrection the time offset as returned by
691 * validateTimeZone() in Special:Preferences
694 function date( $ts, $adj = false, $format = true, $timecorrection = false ) {
697 $ts = $this->userAdjust( $ts, $timecorrection );
700 $pref = $this->dateFormat( $format );
701 if( $pref == 'default' ||
!isset( $this->dateFormats
["$pref date"] ) ) {
702 $pref = $this->defaultDateFormat
;
704 return $this->sprintfDate( $this->dateFormats
["$pref date"], $ts );
709 * @param mixed $ts the time format which needs to be turned into a
710 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
711 * @param bool $adj whether to adjust the time output according to the
712 * user configured offset ($timecorrection)
713 * @param mixed $format true to use user's date format preference
714 * @param string $timecorrection the time offset as returned by
715 * validateTimeZone() in Special:Preferences
718 function time( $ts, $adj = false, $format = true, $timecorrection = false ) {
721 $ts = $this->userAdjust( $ts, $timecorrection );
724 $pref = $this->dateFormat( $format );
725 if( $pref == 'default' ||
!isset( $this->dateFormats
["$pref time"] ) ) {
726 $pref = $this->defaultDateFormat
;
728 return $this->sprintfDate( $this->dateFormats
["$pref time"], $ts );
733 * @param mixed $ts the time format which needs to be turned into a
734 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
735 * @param bool $adj whether to adjust the time output according to the
736 * user configured offset ($timecorrection)
738 * @param mixed $format what format to return, if it's false output the
739 * default one (default true)
740 * @param string $timecorrection the time offset as returned by
741 * validateTimeZone() in Special:Preferences
744 function timeanddate( $ts, $adj = false, $format = true, $timecorrection = false) {
747 $ts = $this->userAdjust( $ts, $timecorrection );
750 $pref = $this->dateFormat( $format );
751 if( $pref == 'default' ||
!isset( $this->dateFormats
["$pref both"] ) ) {
752 $pref = $this->defaultDateFormat
;
755 return $this->sprintfDate( $this->dateFormats
["$pref both"], $ts );
758 function getMessage( $key ) {
760 return isset( $this->messages
[$key] ) ?
$this->messages
[$key] : null;
763 function getAllMessages() {
765 return $this->messages
;
768 function iconv( $in, $out, $string ) {
769 # For most languages, this is a wrapper for iconv
770 return iconv( $in, $out, $string );
773 // callback functions for uc(), lc(), ucwords(), ucwordbreaks()
774 function ucwordbreaksCallbackAscii($matches){
775 return $this->ucfirst($matches[1]);
778 function ucwordbreaksCallbackMB($matches){
779 return mb_strtoupper($matches[0]);
782 function ucCallback($matches){
783 list( $wikiUpperChars ) = self
::getCaseMaps();
784 return strtr( $matches[1], $wikiUpperChars );
787 function lcCallback($matches){
788 list( , $wikiLowerChars ) = self
::getCaseMaps();
789 return strtr( $matches[1], $wikiLowerChars );
792 function ucwordsCallbackMB($matches){
793 return mb_strtoupper($matches[0]);
796 function ucwordsCallbackWiki($matches){
797 list( $wikiUpperChars ) = self
::getCaseMaps();
798 return strtr( $matches[0], $wikiUpperChars );
801 function ucfirst( $str ) {
802 return self
::uc( $str, true );
805 function uc( $str, $first = false ) {
806 if ( function_exists( 'mb_strtoupper' ) )
808 if ( self
::isMultibyte( $str ) )
809 return mb_strtoupper( mb_substr( $str, 0, 1 ) ) . mb_substr( $str, 1 );
811 return ucfirst( $str );
813 return self
::isMultibyte( $str ) ?
mb_strtoupper( $str ) : strtoupper( $str );
815 if ( self
::isMultibyte( $str ) ) {
816 list( $wikiUpperChars ) = $this->getCaseMaps();
817 $x = $first ?
'^' : '';
818 return preg_replace_callback(
819 "/$x([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/",
820 array($this,"ucCallback"),
824 return $first ?
ucfirst( $str ) : strtoupper( $str );
827 function lcfirst( $str ) {
828 return self
::lc( $str, true );
831 function lc( $str, $first = false ) {
832 if ( function_exists( 'mb_strtolower' ) )
834 if ( self
::isMultibyte( $str ) )
835 return mb_strtolower( mb_substr( $str, 0, 1 ) ) . mb_substr( $str, 1 );
837 return strtolower( substr( $str, 0, 1 ) ) . substr( $str, 1 );
839 return self
::isMultibyte( $str ) ?
mb_strtolower( $str ) : strtolower( $str );
841 if ( self
::isMultibyte( $str ) ) {
842 list( , $wikiLowerChars ) = self
::getCaseMaps();
843 $x = $first ?
'^' : '';
844 return preg_replace_callback(
845 "/$x([A-Z]|[\\xc0-\\xff][\\x80-\\xbf]*)/",
846 array($this,"lcCallback"),
850 return $first ?
strtolower( substr( $str, 0, 1 ) ) . substr( $str, 1 ) : strtolower( $str );
853 function isMultibyte( $str ) {
854 return (bool)preg_match( '/[\x80-\xff]/', $str );
857 function ucwords($str) {
858 if ( self
::isMultibyte( $str ) ) {
859 $str = self
::lc($str);
861 // regexp to find first letter in each word (i.e. after each space)
862 $replaceRegexp = "/^([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)| ([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/";
864 // function to use to capitalize a single char
865 if ( function_exists( 'mb_strtoupper' ) )
866 return preg_replace_callback(
868 array($this,"ucwordsCallbackMB"),
872 return preg_replace_callback(
874 array($this,"ucwordsCallbackWiki"),
879 return ucwords( strtolower( $str ) );
882 # capitalize words at word breaks
883 function ucwordbreaks($str){
884 if (self
::isMultibyte( $str ) ) {
885 $str = self
::lc($str);
887 // since \b doesn't work for UTF-8, we explicitely define word break chars
888 $breaks= "[ \-\(\)\}\{\.,\?!]";
890 // find first letter after word break
891 $replaceRegexp = "/^([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)|$breaks([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/";
893 if ( function_exists( 'mb_strtoupper' ) )
894 return preg_replace_callback(
896 array($this,"ucwordbreaksCallbackMB"),
900 return preg_replace_callback(
902 array($this,"ucwordsCallbackWiki"),
907 return preg_replace_callback(
908 '/\b([\w\x80-\xff]+)\b/',
909 array($this,"ucwordbreaksCallbackAscii"),
914 * Return a case-folded representation of $s
916 * This is a representation such that caseFold($s1)==caseFold($s2) if $s1
917 * and $s2 are the same except for the case of their characters. It is not
918 * necessary for the value returned to make sense when displayed.
920 * Do *not* perform any other normalisation in this function. If a caller
921 * uses this function when it should be using a more general normalisation
922 * function, then fix the caller.
924 function caseFold( $s ) {
925 return $this->uc( $s );
928 function checkTitleEncoding( $s ) {
929 if( is_array( $s ) ) {
930 wfDebugDieBacktrace( 'Given array to checkTitleEncoding.' );
932 # Check for non-UTF-8 URLs
933 $ishigh = preg_match( '/[\x80-\xff]/', $s);
934 if(!$ishigh) return $s;
936 $isutf8 = preg_match( '/^([\x00-\x7f]|[\xc0-\xdf][\x80-\xbf]|' .
937 '[\xe0-\xef][\x80-\xbf]{2}|[\xf0-\xf7][\x80-\xbf]{3})+$/', $s );
938 if( $isutf8 ) return $s;
940 return $this->iconv( $this->fallback8bitEncoding(), "utf-8", $s );
943 function fallback8bitEncoding() {
945 return $this->fallback8bitEncoding
;
949 * Some languages have special punctuation to strip out
950 * or characters which need to be converted for MySQL's
951 * indexing to grok it correctly. Make such changes here.
956 function stripForSearch( $string ) {
957 # MySQL fulltext index doesn't grok utf-8, so we
958 # need to fold cases and convert to hex
960 wfProfileIn( __METHOD__
);
961 if( function_exists( 'mb_strtolower' ) ) {
963 "/([\\xc0-\\xff][\\x80-\\xbf]*)/e",
964 "'U8' . bin2hex( \"$1\" )",
965 mb_strtolower( $string ) );
967 list( , $wikiLowerChars ) = self
::getCaseMaps();
969 "/([\\xc0-\\xff][\\x80-\\xbf]*)/e",
970 "'U8' . bin2hex( strtr( \"\$1\", \$wikiLowerChars ) )",
973 wfProfileOut( __METHOD__
);
977 function convertForSearchResult( $termsArray ) {
978 # some languages, e.g. Chinese, need to do a conversion
979 # in order for search results to be displayed correctly
984 * Get the first character of a string.
989 function firstChar( $s ) {
990 preg_match( '/^([\x00-\x7f]|[\xc0-\xdf][\x80-\xbf]|' .
991 '[\xe0-\xef][\x80-\xbf]{2}|[\xf0-\xf7][\x80-\xbf]{3})/', $s, $matches);
993 return isset( $matches[1] ) ?
$matches[1] : "";
996 function initEncoding() {
997 # Some languages may have an alternate char encoding option
998 # (Esperanto X-coding, Japanese furigana conversion, etc)
999 # If this language is used as the primary content language,
1000 # an override to the defaults can be set here on startup.
1003 function recodeForEdit( $s ) {
1004 # For some languages we'll want to explicitly specify
1005 # which characters make it into the edit box raw
1006 # or are converted in some way or another.
1007 # Note that if wgOutputEncoding is different from
1008 # wgInputEncoding, this text will be further converted
1009 # to wgOutputEncoding.
1010 global $wgEditEncoding;
1011 if( $wgEditEncoding == '' or
1012 $wgEditEncoding == 'UTF-8' ) {
1015 return $this->iconv( 'UTF-8', $wgEditEncoding, $s );
1019 function recodeInput( $s ) {
1020 # Take the previous into account.
1021 global $wgEditEncoding;
1022 if($wgEditEncoding != "") {
1023 $enc = $wgEditEncoding;
1027 if( $enc == 'UTF-8' ) {
1030 return $this->iconv( $enc, 'UTF-8', $s );
1035 * For right-to-left language support
1045 * A hidden direction mark (LRM or RLM), depending on the language direction
1049 function getDirMark() {
1050 return $this->isRTL() ?
"\xE2\x80\x8F" : "\xE2\x80\x8E";
1054 * An arrow, depending on the language direction
1058 function getArrow() {
1059 return $this->isRTL() ?
'←' : '→';
1063 * To allow "foo[[bar]]" to extend the link over the whole word "foobar"
1067 function linkPrefixExtension() {
1069 return $this->linkPrefixExtension
;
1072 function &getMagicWords() {
1074 return $this->magicWords
;
1077 # Fill a MagicWord object with data from here
1078 function getMagic( &$mw ) {
1079 if ( !isset( $this->mMagicExtensions
) ) {
1080 $this->mMagicExtensions
= array();
1081 wfRunHooks( 'LanguageGetMagic', array( &$this->mMagicExtensions
, $this->getCode() ) );
1083 if ( isset( $this->mMagicExtensions
[$mw->mId
] ) ) {
1084 $rawEntry = $this->mMagicExtensions
[$mw->mId
];
1086 $magicWords =& $this->getMagicWords();
1087 if ( isset( $magicWords[$mw->mId
] ) ) {
1088 $rawEntry = $magicWords[$mw->mId
];
1090 # Fall back to English if local list is incomplete
1091 $magicWords =& Language
::getMagicWords();
1092 $rawEntry = $magicWords[$mw->mId
];
1096 if( !is_array( $rawEntry ) ) {
1097 error_log( "\"$rawEntry\" is not a valid magic thingie for \"$mw->mId\"" );
1099 $mw->mCaseSensitive
= $rawEntry[0];
1100 $mw->mSynonyms
= array_slice( $rawEntry, 1 );
1104 * Get special page names, as an associative array
1105 * case folded alias => real name
1107 function getSpecialPageAliases() {
1109 if ( !isset( $this->mExtendedSpecialPageAliases
) ) {
1110 $this->mExtendedSpecialPageAliases
= $this->specialPageAliases
;
1111 wfRunHooks( 'LangugeGetSpecialPageAliases',
1112 array( &$this->mExtendedSpecialPageAliases
, $this->getCode() ) );
1114 return $this->mExtendedSpecialPageAliases
;
1118 * Italic is unsuitable for some languages
1122 * @param string $text The text to be emphasized.
1125 function emphasize( $text ) {
1126 return "<em>$text</em>";
1130 * Normally we output all numbers in plain en_US style, that is
1131 * 293,291.235 for twohundredninetythreethousand-twohundredninetyone
1132 * point twohundredthirtyfive. However this is not sutable for all
1133 * languages, some such as Pakaran want ੨੯੩,੨੯੫.੨੩੫ and others such as
1134 * Icelandic just want to use commas instead of dots, and dots instead
1135 * of commas like "293.291,235".
1137 * An example of this function being called:
1139 * wfMsg( 'message', $wgLang->formatNum( $num ) )
1142 * See LanguageGu.php for the Gujarati implementation and
1143 * LanguageIs.php for the , => . and . => , implementation.
1145 * @todo check if it's viable to use localeconv() for the decimal
1148 * @param mixed $number the string to be formatted, should be an integer or
1149 * a floating point number.
1150 * @param bool $nocommafy Set to true for special numbers like dates
1153 function formatNum( $number, $nocommafy = false ) {
1154 global $wgTranslateNumerals;
1156 $number = $this->commafy($number);
1157 $s = $this->separatorTransformTable();
1158 if (!is_null($s)) { $number = strtr($number, $s); }
1161 if ($wgTranslateNumerals) {
1162 $s = $this->digitTransformTable();
1163 if (!is_null($s)) { $number = strtr($number, $s); }
1170 * Adds commas to a given number
1175 function commafy($_) {
1176 return strrev((string)preg_replace('/(\d{3})(?=\d)(?!\d*\.)/','$1,',strrev($_)));
1179 function digitTransformTable() {
1181 return $this->digitTransformTable
;
1184 function separatorTransformTable() {
1186 return $this->separatorTransformTable
;
1191 * For the credit list in includes/Credits.php (action=credits)
1196 function listToText( $l ) {
1199 for ($i = $m; $i >= 0; $i--) {
1202 } else if ($i == $m - 1) {
1203 $s = $l[$i] . ' ' . $this->getMessageFromDB( 'and' ) . ' ' . $s;
1205 $s = $l[$i] . ', ' . $s;
1211 # Crop a string from the beginning or end to a certain number of bytes.
1212 # (Bytes are used because our storage has limited byte lengths for some
1213 # columns in the database.) Multibyte charsets will need to make sure that
1214 # only whole characters are included!
1216 # $length does not include the optional ellipsis.
1217 # If $length is negative, snip from the beginning
1218 function truncate( $string, $length, $ellipsis = "" ) {
1219 if( $length == 0 ) {
1222 if ( strlen( $string ) <= abs( $length ) ) {
1226 $string = substr( $string, 0, $length );
1227 $char = ord( $string[strlen( $string ) - 1] );
1228 if ($char >= 0xc0) {
1229 # We got the first byte only of a multibyte char; remove it.
1230 $string = substr( $string, 0, -1 );
1231 } elseif( $char >= 0x80 &&
1232 preg_match( '/^(.*)(?:[\xe0-\xef][\x80-\xbf]|' .
1233 '[\xf0-\xf7][\x80-\xbf]{1,2})$/', $string, $m ) ) {
1234 # We chopped in the middle of a character; remove it
1237 return $string . $ellipsis;
1239 $string = substr( $string, $length );
1240 $char = ord( $string[0] );
1241 if( $char >= 0x80 && $char < 0xc0 ) {
1242 # We chopped in the middle of a character; remove the whole thing
1243 $string = preg_replace( '/^[\x80-\xbf]+/', '', $string );
1245 return $ellipsis . $string;
1250 * Grammatical transformations, needed for inflected languages
1251 * Invoked by putting {{grammar:case|word}} in a message
1253 * @param string $word
1254 * @param string $case
1257 function convertGrammar( $word, $case ) {
1258 global $wgGrammarForms;
1259 if ( isset($wgGrammarForms['en'][$case][$word]) ) {
1260 return $wgGrammarForms['en'][$case][$word];
1266 * Plural form transformations, needed for some languages.
1267 * For example, where are 3 form of plural in Russian and Polish,
1268 * depending on "count mod 10". See [[w:Plural]]
1269 * For English it is pretty simple.
1271 * Invoked by putting {{plural:count|wordform1|wordform2}}
1272 * or {{plural:count|wordform1|wordform2|wordform3}}
1274 * Example: {{plural:{{NUMBEROFARTICLES}}|article|articles}}
1276 * @param integer $count
1277 * @param string $wordform1
1278 * @param string $wordform2
1279 * @param string $wordform3 (optional)
1282 function convertPlural( $count, $w1, $w2, $w3) {
1283 return $count == '1' ?
$w1 : $w2;
1287 * For translaing of expiry times
1288 * @param string The validated block time in English
1289 * @return Somehow translated block time
1290 * @see LanguageFi.php for example implementation
1292 function translateBlockExpiry( $str ) {
1294 $scBlockExpiryOptions = $this->getMessageFromDB( 'ipboptions' );
1296 if ( $scBlockExpiryOptions == '-') {
1300 foreach (explode(',', $scBlockExpiryOptions) as $option) {
1301 if ( strpos($option, ":") === false )
1303 list($show, $value) = explode(":", $option);
1304 if ( strcmp ( $str, $value) == 0 )
1305 return '<span title="' . htmlspecialchars($str). '">' .
1306 htmlspecialchars( trim( $show ) ) . '</span>';
1313 * languages like Chinese need to be segmented in order for the diff
1316 * @param string $text
1319 function segmentForDiff( $text ) {
1324 * and unsegment to show the result
1326 * @param string $text
1329 function unsegmentForDiff( $text ) {
1333 # convert text to different variants of a language.
1334 function convert( $text, $isTitle = false) {
1335 return $this->mConverter
->convert($text, $isTitle);
1338 # Convert text from within Parser
1339 function parserConvert( $text, &$parser ) {
1340 return $this->mConverter
->parserConvert( $text, $parser );
1343 # Tell the converter that it shouldn't convert titles
1344 function setNoTitleConvert(){
1345 $this->mConverter
->setNotitleConvert();
1348 # Check if this is a language with variants
1349 function hasVariants(){
1350 return sizeof($this->getVariants())>1;
1355 * Perform output conversion on a string, and encode for safe HTML output.
1356 * @param string $text
1357 * @param bool $isTitle -- wtf?
1359 * @todo this should get integrated somewhere sane
1361 function convertHtml( $text, $isTitle = false ) {
1362 return htmlspecialchars( $this->convert( $text, $isTitle ) );
1365 function convertCategoryKey( $key ) {
1366 return $this->mConverter
->convertCategoryKey( $key );
1370 * get the list of variants supported by this langauge
1371 * see sample implementation in LanguageZh.php
1373 * @return array an array of language codes
1375 function getVariants() {
1376 return $this->mConverter
->getVariants();
1380 function getPreferredVariant( $fromUser = true ) {
1381 return $this->mConverter
->getPreferredVariant( $fromUser );
1385 * if a language supports multiple variants, it is
1386 * possible that non-existing link in one variant
1387 * actually exists in another variant. this function
1388 * tries to find it. See e.g. LanguageZh.php
1390 * @param string $link the name of the link
1391 * @param mixed $nt the title object of the link
1392 * @return null the input parameters may be modified upon return
1394 function findVariantLink( &$link, &$nt ) {
1395 $this->mConverter
->findVariantLink($link, $nt);
1399 * If a language supports multiple variants, converts text
1400 * into an array of all possible variants of the text:
1401 * 'variant' => text in that variant
1404 function convertLinkToAllVariants($text){
1405 return $this->mConverter
->convertLinkToAllVariants($text);
1410 * returns language specific options used by User::getPageRenderHash()
1411 * for example, the preferred language variant
1416 function getExtraHashOptions() {
1417 return $this->mConverter
->getExtraHashOptions();
1421 * for languages that support multiple variants, the title of an
1422 * article may be displayed differently in different variants. this
1423 * function returns the apporiate title defined in the body of the article.
1427 function getParsedTitle() {
1428 return $this->mConverter
->getParsedTitle();
1432 * Enclose a string with the "no conversion" tag. This is used by
1433 * various functions in the Parser
1435 * @param string $text text to be tagged for no conversion
1436 * @return string the tagged text
1438 function markNoConversion( $text, $noParse=false ) {
1439 return $this->mConverter
->markNoConversion( $text, $noParse );
1443 * A regular expression to match legal word-trailing characters
1444 * which should be merged onto a link of the form [[foo]]bar.
1449 function linkTrail() {
1451 return $this->linkTrail
;
1454 function getLangObj() {
1459 * Get the RFC 3066 code for this language object
1461 function getCode() {
1462 return $this->mCode
;
1465 function setCode( $code ) {
1466 $this->mCode
= $code;
1469 static function getFileName( $prefix = 'Language', $code, $suffix = '.php' ) {
1470 return $prefix . str_replace( '-', '_', ucfirst( $code ) ) . $suffix;
1473 static function getMessagesFileName( $code ) {
1475 return self
::getFileName( "$IP/languages/messages/Messages", $code, '.php' );
1478 static function getClassFileName( $code ) {
1480 return self
::getFileName( "$IP/languages/classes/Language", $code, '.php' );
1483 static function getLocalisationArray( $code, $disableCache = false ) {
1484 self
::loadLocalisation( $code, $disableCache );
1485 return self
::$mLocalisationCache[$code];
1489 * Load localisation data for a given code into the static cache
1491 * @return array Dependencies, map of filenames to mtimes
1493 static function loadLocalisation( $code, $disableCache = false ) {
1494 static $recursionGuard = array();
1498 throw new MWException( "Invalid language code requested" );
1501 if ( !$disableCache ) {
1502 # Try the per-process cache
1503 if ( isset( self
::$mLocalisationCache[$code] ) ) {
1504 return self
::$mLocalisationCache[$code]['deps'];
1507 wfProfileIn( __METHOD__
);
1509 # Try the serialized directory
1510 $cache = wfGetPrecompiledData( self
::getFileName( "Messages", $code, '.ser' ) );
1512 self
::$mLocalisationCache[$code] = $cache;
1513 wfDebug( "Got localisation for $code from precompiled data file\n" );
1514 wfProfileOut( __METHOD__
);
1515 return self
::$mLocalisationCache[$code]['deps'];
1518 # Try the global cache
1519 $memcKey = wfMemcKey('localisation', $code );
1520 $cache = $wgMemc->get( $memcKey );
1523 # Check file modification times
1524 foreach ( $cache['deps'] as $file => $mtime ) {
1525 if ( !file_exists( $file ) ||
filemtime( $file ) > $mtime ) {
1530 if ( self
::isLocalisationOutOfDate( $cache ) ) {
1531 $wgMemc->delete( $memcKey );
1533 wfDebug( "Localisation cache for $code had expired due to update of $file\n" );
1535 self
::$mLocalisationCache[$code] = $cache;
1536 wfDebug( "Got localisation for $code from cache\n" );
1537 wfProfileOut( __METHOD__
);
1538 return $cache['deps'];
1542 wfProfileIn( __METHOD__
);
1545 # Default fallback, may be overridden when the messages file is included
1546 if ( $code != 'en' ) {
1552 # Load the primary localisation from the source file
1553 $filename = self
::getMessagesFileName( $code );
1554 if ( !file_exists( $filename ) ) {
1555 wfDebug( "No localisation file for $code, using implicit fallback to en\n" );
1559 $deps = array( $filename => filemtime( $filename ) );
1560 require( $filename );
1561 $cache = compact( self
::$mLocalisationKeys );
1562 wfDebug( "Got localisation for $code from source\n" );
1565 if ( !empty( $fallback ) ) {
1566 # Load the fallback localisation, with a circular reference guard
1567 if ( isset( $recursionGuard[$code] ) ) {
1568 throw new MWException( "Error: Circular fallback reference in language code $code" );
1570 $recursionGuard[$code] = true;
1571 $newDeps = self
::loadLocalisation( $fallback, $disableCache );
1572 unset( $recursionGuard[$code] );
1574 $secondary = self
::$mLocalisationCache[$fallback];
1575 $deps = array_merge( $deps, $newDeps );
1577 # Merge the fallback localisation with the current localisation
1578 foreach ( self
::$mLocalisationKeys as $key ) {
1579 if ( isset( $cache[$key] ) ) {
1580 if ( isset( $secondary[$key] ) ) {
1581 if ( in_array( $key, self
::$mMergeableMapKeys ) ) {
1582 $cache[$key] = $cache[$key] +
$secondary[$key];
1583 } elseif ( in_array( $key, self
::$mMergeableListKeys ) ) {
1584 $cache[$key] = array_merge( $secondary[$key], $cache[$key] );
1585 } elseif ( in_array( $key, self
::$mMergeableAliasListKeys ) ) {
1586 $cache[$key] = array_merge_recursive( $cache[$key], $secondary[$key] );
1590 $cache[$key] = $secondary[$key];
1594 # Merge bookstore lists if requested
1595 if ( !empty( $cache['bookstoreList']['inherit'] ) ) {
1596 $cache['bookstoreList'] = array_merge( $cache['bookstoreList'], $secondary['bookstoreList'] );
1598 if ( isset( $cache['bookstoreList']['inherit'] ) ) {
1599 unset( $cache['bookstoreList']['inherit'] );
1603 # Add dependencies to the cache entry
1604 $cache['deps'] = $deps;
1606 # Replace spaces with underscores in namespace names
1607 $cache['namespaceNames'] = str_replace( ' ', '_', $cache['namespaceNames'] );
1609 # Save to both caches
1610 self
::$mLocalisationCache[$code] = $cache;
1611 if ( !$disableCache ) {
1612 $wgMemc->set( $memcKey, $cache );
1615 wfProfileOut( __METHOD__
);
1620 * Test if a given localisation cache is out of date with respect to the
1621 * source Messages files. This is done automatically for the global cache
1622 * in $wgMemc, but is only done on certain occasions for the serialized
1625 * @param $cache mixed Either a language code or a cache array
1627 static function isLocalisationOutOfDate( $cache ) {
1628 if ( !is_array( $cache ) ) {
1629 self
::loadLocalisation( $cache );
1630 $cache = self
::$mLocalisationCache[$cache];
1633 foreach ( $cache['deps'] as $file => $mtime ) {
1634 if ( !file_exists( $file ) ||
filemtime( $file ) > $mtime ) {
1643 * Get the fallback for a given language
1645 static function getFallbackFor( $code ) {
1646 self
::loadLocalisation( $code );
1647 return self
::$mLocalisationCache[$code]['fallback'];
1651 * Get all messages for a given language
1653 static function getMessagesFor( $code ) {
1654 self
::loadLocalisation( $code );
1655 return self
::$mLocalisationCache[$code]['messages'];
1659 * Get a message for a given language
1661 static function getMessageFor( $key, $code ) {
1662 self
::loadLocalisation( $code );
1663 return isset( self
::$mLocalisationCache[$code]['messages'][$key] ) ? self
::$mLocalisationCache[$code]['messages'][$key] : null;
1667 * Load localisation data for this object
1670 if ( !$this->mLoaded
) {
1671 self
::loadLocalisation( $this->getCode() );
1672 $cache =& self
::$mLocalisationCache[$this->getCode()];
1673 foreach ( self
::$mLocalisationKeys as $key ) {
1674 $this->$key = $cache[$key];
1676 $this->mLoaded
= true;
1678 $this->fixUpSettings();
1683 * Do any necessary post-cache-load settings adjustment
1685 function fixUpSettings() {
1686 global $wgExtraNamespaces, $wgMetaNamespace, $wgMetaNamespaceTalk, $wgMessageCache,
1687 $wgNamespaceAliases, $wgAmericanDates;
1688 wfProfileIn( __METHOD__
);
1689 if ( $wgExtraNamespaces ) {
1690 $this->namespaceNames
= $wgExtraNamespaces +
$this->namespaceNames
;
1693 $this->namespaceNames
[NS_PROJECT
] = $wgMetaNamespace;
1694 if ( $wgMetaNamespaceTalk ) {
1695 $this->namespaceNames
[NS_PROJECT_TALK
] = $wgMetaNamespaceTalk;
1697 $talk = $this->namespaceNames
[NS_PROJECT_TALK
];
1698 $talk = str_replace( '$1', $wgMetaNamespace, $talk );
1700 # Allow grammar transformations
1701 # Allowing full message-style parsing would make simple requests
1702 # such as action=raw much more expensive than they need to be.
1703 # This will hopefully cover most cases.
1704 $talk = preg_replace_callback( '/{{grammar:(.*?)\|(.*?)}}/i',
1705 array( &$this, 'replaceGrammarInNamespace' ), $talk );
1706 $talk = str_replace( ' ', '_', $talk );
1707 $this->namespaceNames
[NS_PROJECT_TALK
] = $talk;
1710 # The above mixing may leave namespaces out of canonical order.
1711 # Re-order by namespace ID number...
1712 ksort( $this->namespaceNames
);
1714 # Put namespace names and aliases into a hashtable.
1715 # If this is too slow, then we should arrange it so that it is done
1716 # before caching. The catch is that at pre-cache time, the above
1717 # class-specific fixup hasn't been done.
1718 $this->mNamespaceIds
= array();
1719 foreach ( $this->namespaceNames
as $index => $name ) {
1720 $this->mNamespaceIds
[$this->lc($name)] = $index;
1722 if ( $this->namespaceAliases
) {
1723 foreach ( $this->namespaceAliases
as $name => $index ) {
1724 $this->mNamespaceIds
[$this->lc($name)] = $index;
1727 if ( $wgNamespaceAliases ) {
1728 foreach ( $wgNamespaceAliases as $name => $index ) {
1729 $this->mNamespaceIds
[$this->lc($name)] = $index;
1733 if ( $this->defaultDateFormat
== 'dmy or mdy' ) {
1734 $this->defaultDateFormat
= $wgAmericanDates ?
'mdy' : 'dmy';
1736 wfProfileOut( __METHOD__
);
1739 function replaceGrammarInNamespace( $m ) {
1740 return $this->convertGrammar( trim( $m[2] ), trim( $m[1] ) );
1743 static function getCaseMaps() {
1744 static $wikiUpperChars, $wikiLowerChars;
1745 if ( isset( $wikiUpperChars ) ) {
1746 return array( $wikiUpperChars, $wikiLowerChars );
1749 wfProfileIn( __METHOD__
);
1750 $arr = wfGetPrecompiledData( 'Utf8Case.ser' );
1751 if ( $arr === false ) {
1752 throw new MWException(
1753 "Utf8Case.ser is missing, please run \"make\" in the serialized directory\n" );
1756 wfProfileOut( __METHOD__
);
1757 return array( $wikiUpperChars, $wikiLowerChars );