6 if( !defined( 'MEDIAWIKI' ) ) {
7 echo "This file is part of MediaWiki, it is not a valid entry point.\n";
12 global $wgLanguageNames;
13 require_once( dirname(__FILE__
) . '/Names.php' ) ;
15 global $wgInputEncoding, $wgOutputEncoding;
18 * These are always UTF-8, they exist only for backwards compatibility
20 $wgInputEncoding = "UTF-8";
21 $wgOutputEncoding = "UTF-8";
23 if( function_exists( 'mb_strtoupper' ) ) {
24 mb_internal_encoding('UTF-8');
27 /* a fake language converter */
30 function FakeConverter($langobj) {$this->mLang
= $langobj;}
31 function convert($t, $i) {return $t;}
32 function parserConvert($t, $p) {return $t;}
33 function getVariants() { return array( $this->mLang
->getCode() ); }
34 function getPreferredVariant() {return $this->mLang
->getCode(); }
35 function findVariantLink(&$l, &$n) {}
36 function getExtraHashOptions() {return '';}
37 function getParsedTitle() {return '';}
38 function markNoConversion($text, $noParse=false) {return $text;}
39 function convertCategoryKey( $key ) {return $key; }
40 function convertLinkToAllVariants($text){ return array( $this->mLang
->getCode() => $text); }
41 function armourMath($text){ return $text; }
44 #--------------------------------------------------------------------------
45 # Internationalisation code
46 #--------------------------------------------------------------------------
49 var $mConverter, $mVariants, $mCode, $mLoaded = false;
50 var $mMagicExtensions = array(), $mMagicHookDone = false;
52 static public $mLocalisationKeys = array( 'fallback', 'namespaceNames',
53 'skinNames', 'mathNames',
54 'bookstoreList', 'magicWords', 'messages', 'rtl', 'digitTransformTable',
55 'separatorTransformTable', 'fallback8bitEncoding', 'linkPrefixExtension',
56 'defaultUserOptionOverrides', 'linkTrail', 'namespaceAliases',
57 'dateFormats', 'datePreferences', 'datePreferenceMigrationMap',
58 'defaultDateFormat', 'extraUserToggles', 'specialPageAliases' );
60 static public $mMergeableMapKeys = array( 'messages', 'namespaceNames', 'mathNames',
61 'dateFormats', 'defaultUserOptionOverrides', 'magicWords' );
63 static public $mMergeableListKeys = array( 'extraUserToggles' );
65 static public $mMergeableAliasListKeys = array( 'specialPageAliases' );
67 static public $mLocalisationCache = array();
69 static public $mWeekdayMsgs = array(
70 'sunday', 'monday', 'tuesday', 'wednesday', 'thursday',
74 static public $mWeekdayAbbrevMsgs = array(
75 'sun', 'mon', 'tue', 'wed', 'thu', 'fri', 'sat'
78 static public $mMonthMsgs = array(
79 'january', 'february', 'march', 'april', 'may_long', 'june',
80 'july', 'august', 'september', 'october', 'november',
83 static public $mMonthGenMsgs = array(
84 'january-gen', 'february-gen', 'march-gen', 'april-gen', 'may-gen', 'june-gen',
85 'july-gen', 'august-gen', 'september-gen', 'october-gen', 'november-gen',
88 static public $mMonthAbbrevMsgs = array(
89 'jan', 'feb', 'mar', 'apr', 'may', 'jun', 'jul', 'aug',
90 'sep', 'oct', 'nov', 'dec'
93 static public $mIranianCalendarMonthMsgs = array(
94 'iranian-calendar-m1', 'iranian-calendar-m2', 'iranian-calendar-m3',
95 'iranian-calendar-m4', 'iranian-calendar-m5', 'iranian-calendar-m6',
96 'iranian-calendar-m7', 'iranian-calendar-m8', 'iranian-calendar-m9',
97 'iranian-calendar-m10', 'iranian-calendar-m11', 'iranian-calendar-m12'
100 static public $mHebrewCalendarMonthMsgs = array(
101 'hebrew-calendar-m1', 'hebrew-calendar-m2', 'hebrew-calendar-m3',
102 'hebrew-calendar-m4', 'hebrew-calendar-m5', 'hebrew-calendar-m6',
103 'hebrew-calendar-m7', 'hebrew-calendar-m8', 'hebrew-calendar-m9',
104 'hebrew-calendar-m10', 'hebrew-calendar-m11', 'hebrew-calendar-m12',
105 'hebrew-calendar-m6a', 'hebrew-calendar-m6b'
108 static public $mHebrewCalendarMonthGenMsgs = array(
109 'hebrew-calendar-m1-gen', 'hebrew-calendar-m2-gen', 'hebrew-calendar-m3-gen',
110 'hebrew-calendar-m4-gen', 'hebrew-calendar-m5-gen', 'hebrew-calendar-m6-gen',
111 'hebrew-calendar-m7-gen', 'hebrew-calendar-m8-gen', 'hebrew-calendar-m9-gen',
112 'hebrew-calendar-m10-gen', 'hebrew-calendar-m11-gen', 'hebrew-calendar-m12-gen',
113 'hebrew-calendar-m6a-gen', 'hebrew-calendar-m6b-gen'
117 * Create a language object for a given language code
119 static function factory( $code ) {
121 static $recursionLevel = 0;
123 if ( $code == 'en' ) {
126 $class = 'Language' . str_replace( '-', '_', ucfirst( $code ) );
127 // Preload base classes to work around APC/PHP5 bug
128 if ( file_exists( "$IP/languages/classes/$class.deps.php" ) ) {
129 include_once("$IP/languages/classes/$class.deps.php");
131 if ( file_exists( "$IP/languages/classes/$class.php" ) ) {
132 include_once("$IP/languages/classes/$class.php");
136 if ( $recursionLevel > 5 ) {
137 throw new MWException( "Language fallback loop detected when creating class $class\n" );
140 if( ! class_exists( $class ) ) {
141 $fallback = Language
::getFallbackFor( $code );
143 $lang = Language
::factory( $fallback );
145 $lang->setCode( $code );
153 function __construct() {
154 $this->mConverter
= new FakeConverter($this);
155 // Set the code to the name of the descendant
156 if ( get_class( $this ) == 'Language' ) {
159 $this->mCode
= str_replace( '_', '-', strtolower( substr( get_class( $this ), 8 ) ) );
164 * Hook which will be called if this is the content language.
165 * Descendants can use this to register hook functions or modify globals
167 function initContLang() {}
173 function getDefaultUserOptions() {
174 return User
::getDefaultOptions();
177 function getFallbackLanguageCode() {
179 return $this->fallback
;
183 * Exports $wgBookstoreListEn
186 function getBookstoreList() {
188 return $this->bookstoreList
;
194 function getNamespaces() {
196 return $this->namespaceNames
;
200 * A convenience function that returns the same thing as
201 * getNamespaces() except with the array values changed to ' '
202 * where it found '_', useful for producing output to be displayed
203 * e.g. in <select> forms.
207 function getFormattedNamespaces() {
208 $ns = $this->getNamespaces();
209 foreach($ns as $k => $v) {
210 $ns[$k] = strtr($v, '_', ' ');
216 * Get a namespace value by key
218 * $mw_ns = $wgContLang->getNsText( NS_MEDIAWIKI );
219 * echo $mw_ns; // prints 'MediaWiki'
222 * @param int $index the array key of the namespace to return
223 * @return mixed, string if the namespace value exists, otherwise false
225 function getNsText( $index ) {
226 $ns = $this->getNamespaces();
227 return isset( $ns[$index] ) ?
$ns[$index] : false;
231 * A convenience function that returns the same thing as
232 * getNsText() except with '_' changed to ' ', useful for
237 function getFormattedNsText( $index ) {
238 $ns = $this->getNsText( $index );
239 return strtr($ns, '_', ' ');
243 * Get a namespace key by value, case insensitive.
244 * Only matches namespace names for the current language, not the
245 * canonical ones defined in Namespace.php.
247 * @param string $text
248 * @return mixed An integer if $text is a valid value otherwise false
250 function getLocalNsIndex( $text ) {
252 $lctext = $this->lc($text);
253 return isset( $this->mNamespaceIds
[$lctext] ) ?
$this->mNamespaceIds
[$lctext] : false;
257 * Get a namespace key by value, case insensitive. Canonical namespace
258 * names override custom ones defined for the current language.
260 * @param string $text
261 * @return mixed An integer if $text is a valid value otherwise false
263 function getNsIndex( $text ) {
265 $lctext = $this->lc($text);
266 if( ( $ns = Namespace::getCanonicalIndex( $lctext ) ) !== null ) return $ns;
267 return isset( $this->mNamespaceIds
[$lctext] ) ?
$this->mNamespaceIds
[$lctext] : false;
271 * short names for language variants used for language conversion links.
273 * @param string $code
276 function getVariantname( $code ) {
277 return $this->getMessageFromDB( "variantname-$code" );
280 function specialPage( $name ) {
281 $aliases = $this->getSpecialPageAliases();
282 if ( isset( $aliases[$name][0] ) ) {
283 $name = $aliases[$name][0];
285 return $this->getNsText(NS_SPECIAL
) . ':' . $name;
288 function getQuickbarSettings() {
290 $this->getMessage( 'qbsettings-none' ),
291 $this->getMessage( 'qbsettings-fixedleft' ),
292 $this->getMessage( 'qbsettings-fixedright' ),
293 $this->getMessage( 'qbsettings-floatingleft' ),
294 $this->getMessage( 'qbsettings-floatingright' )
298 function getSkinNames() {
300 return $this->skinNames
;
303 function getMathNames() {
305 return $this->mathNames
;
308 function getDatePreferences() {
310 return $this->datePreferences
;
313 function getDateFormats() {
315 return $this->dateFormats
;
318 function getDefaultDateFormat() {
320 return $this->defaultDateFormat
;
323 function getDatePreferenceMigrationMap() {
325 return $this->datePreferenceMigrationMap
;
328 function getDefaultUserOptionOverrides() {
330 # XXX - apparently some languageas get empty arrays, didn't get to it yet -- midom
331 if (is_array($this->defaultUserOptionOverrides
)) {
332 return $this->defaultUserOptionOverrides
;
338 function getExtraUserToggles() {
340 return $this->extraUserToggles
;
343 function getUserToggle( $tog ) {
344 return $this->getMessageFromDB( "tog-$tog" );
348 * Get language names, indexed by code.
349 * If $customisedOnly is true, only returns codes with a messages file
351 public static function getLanguageNames( $customisedOnly = false ) {
352 global $wgLanguageNames, $wgExtraLanguageNames;
353 $allNames = $wgExtraLanguageNames +
$wgLanguageNames;
354 if ( !$customisedOnly ) {
360 $dir = opendir( "$IP/languages/messages" );
361 while( false !== ( $file = readdir( $dir ) ) ) {
363 if( preg_match( '/Messages([A-Z][a-z_]+)\.php$/', $file, $m ) ) {
364 $code = str_replace( '_', '-', strtolower( $m[1] ) );
365 if ( isset( $allNames[$code] ) ) {
366 $names[$code] = $allNames[$code];
375 * Ugly hack to get a message maybe from the MediaWiki namespace, if this
376 * language object is the content or user language.
378 function getMessageFromDB( $msg ) {
379 global $wgContLang, $wgLang;
380 if ( $wgContLang->getCode() == $this->getCode() ) {
382 return wfMsgForContent( $msg );
383 } elseif ( $wgLang->getCode() == $this->getCode() ) {
385 return wfMsg( $msg );
387 # Neither, get from localisation
388 return $this->getMessage( $msg );
392 function getLanguageName( $code ) {
393 $names = self
::getLanguageNames();
394 if ( !array_key_exists( $code, $names ) ) {
397 return $names[$code];
400 function getMonthName( $key ) {
401 return $this->getMessageFromDB( self
::$mMonthMsgs[$key-1] );
404 function getMonthNameGen( $key ) {
405 return $this->getMessageFromDB( self
::$mMonthGenMsgs[$key-1] );
408 function getMonthAbbreviation( $key ) {
409 return $this->getMessageFromDB( self
::$mMonthAbbrevMsgs[$key-1] );
412 function getWeekdayName( $key ) {
413 return $this->getMessageFromDB( self
::$mWeekdayMsgs[$key-1] );
416 function getWeekdayAbbreviation( $key ) {
417 return $this->getMessageFromDB( self
::$mWeekdayAbbrevMsgs[$key-1] );
420 function getIranianCalendarMonthName( $key ) {
421 return $this->getMessageFromDB( self
::$mIranianCalendarMonthMsgs[$key-1] );
424 function getHebrewCalendarMonthName( $key ) {
425 return $this->getMessageFromDB( self
::$mHebrewCalendarMonthMsgs[$key-1] );
428 function getHebrewCalendarMonthNameGen( $key ) {
429 return $this->getMessageFromDB( self
::$mHebrewCalendarMonthGenMsgs[$key-1] );
434 * Used by date() and time() to adjust the time output.
436 * @param int $ts the time in date('YmdHis') format
437 * @param mixed $tz adjust the time by this amount (default false,
438 * mean we get user timecorrection setting)
441 function userAdjust( $ts, $tz = false ) {
442 global $wgUser, $wgLocalTZoffset;
445 $tz = $wgUser->getOption( 'timecorrection' );
448 # minutes and hours differences:
453 # Global offset in minutes.
454 if( isset($wgLocalTZoffset) ) {
455 if( $wgLocalTZoffset >= 0 ) {
456 $hrDiff = floor($wgLocalTZoffset / 60);
458 $hrDiff = ceil($wgLocalTZoffset / 60);
460 $minDiff = $wgLocalTZoffset %
60;
462 } elseif ( strpos( $tz, ':' ) !== false ) {
463 $tzArray = explode( ':', $tz );
464 $hrDiff = intval($tzArray[0]);
465 $minDiff = intval($hrDiff < 0 ?
-$tzArray[1] : $tzArray[1]);
467 $hrDiff = intval( $tz );
470 # No difference ? Return time unchanged
471 if ( 0 == $hrDiff && 0 == $minDiff ) { return $ts; }
473 wfSuppressWarnings(); // E_STRICT system time bitching
474 # Generate an adjusted date
476 (int)substr( $ts, 8, 2) ) +
$hrDiff, # Hours
477 (int)substr( $ts, 10, 2 ) +
$minDiff, # Minutes
478 (int)substr( $ts, 12, 2 ), # Seconds
479 (int)substr( $ts, 4, 2 ), # Month
480 (int)substr( $ts, 6, 2 ), # Day
481 (int)substr( $ts, 0, 4 ) ); #Year
483 $date = date( 'YmdHis', $t );
490 * This is a workalike of PHP's date() function, but with better
491 * internationalisation, a reduced set of format characters, and a better
494 * Supported format characters are dDjlNwzWFmMntLYyaAgGhHiscrU. See the
495 * PHP manual for definitions. There are a number of extensions, which
498 * xn Do not translate digits of the next numeric format character
499 * xN Toggle raw digit (xn) flag, stays set until explicitly unset
500 * xr Use roman numerals for the next numeric format character
501 * xh Use hebrew numerals for the next numeric format character
503 * xg Genitive month name
505 * xij j (day number) in Iranian calendar
506 * xiF F (month name) in Iranian calendar
507 * xin n (month number) in Iranian calendar
508 * xiY Y (full year) in Iranian calendar
510 * xjj j (day number) in Hebrew calendar
511 * xjF F (month name) in Hebrew calendar
512 * xjt t (days in month) in Hebrew calendar
513 * xjx xg (genitive month name) in Hebrew calendar
514 * xjn n (month number) in Hebrew calendar
515 * xjY Y (full year) in Hebrew calendar
517 * xkY Y (full year) in Thai solar calendar. Months and days are
518 * identical to the Gregorian calendar
520 * Characters enclosed in double quotes will be considered literal (with
521 * the quotes themselves removed). Unmatched quotes will be considered
522 * literal quotes. Example:
524 * "The month is" F => The month is January
527 * Backslash escaping is also supported.
529 * Input timestamp is assumed to be pre-normalized to the desired local
532 * @param string $format
533 * @param string $ts 14-character timestamp
537 function sprintfDate( $format, $ts ) {
547 for ( $p = 0; $p < strlen( $format ); $p++
) {
550 if ( $code == 'x' && $p < strlen( $format ) - 1 ) {
551 $code .= $format[++
$p];
554 if ( ( $code === 'xi' ||
$code == 'xj' ||
$code == 'xk' ) && $p < strlen( $format ) - 1 ) {
555 $code .= $format[++
$p];
566 $rawToggle = !$rawToggle;
575 $s .= $this->getMonthNameGen( substr( $ts, 4, 2 ) );
578 if ( !$hebrew ) $hebrew = self
::tsToHebrew( $ts );
579 $s .= $this->getHebrewCalendarMonthNameGen( $hebrew[1] );
582 $num = substr( $ts, 6, 2 );
585 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
586 $s .= $this->getWeekdayAbbreviation( gmdate( 'w', $unix ) +
1 );
589 $num = intval( substr( $ts, 6, 2 ) );
592 if ( !$iranian ) $iranian = self
::tsToIranian( $ts );
596 if ( !$hebrew ) $hebrew = self
::tsToHebrew( $ts );
600 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
601 $s .= $this->getWeekdayName( gmdate( 'w', $unix ) +
1 );
604 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
605 $w = gmdate( 'w', $unix );
609 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
610 $num = gmdate( 'w', $unix );
613 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
614 $num = gmdate( 'z', $unix );
617 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
618 $num = gmdate( 'W', $unix );
621 $s .= $this->getMonthName( substr( $ts, 4, 2 ) );
624 if ( !$iranian ) $iranian = self
::tsToIranian( $ts );
625 $s .= $this->getIranianCalendarMonthName( $iranian[1] );
628 if ( !$hebrew ) $hebrew = self
::tsToHebrew( $ts );
629 $s .= $this->getHebrewCalendarMonthName( $hebrew[1] );
632 $num = substr( $ts, 4, 2 );
635 $s .= $this->getMonthAbbreviation( substr( $ts, 4, 2 ) );
638 $num = intval( substr( $ts, 4, 2 ) );
641 if ( !$iranian ) $iranian = self
::tsToIranian( $ts );
645 if ( !$hebrew ) $hebrew = self
::tsToHebrew( $ts );
649 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
650 $num = gmdate( 't', $unix );
653 if ( !$hebrew ) $hebrew = self
::tsToHebrew( $ts );
657 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
658 $num = gmdate( 'L', $unix );
661 $num = substr( $ts, 0, 4 );
664 if ( !$iranian ) $iranian = self
::tsToIranian( $ts );
668 if ( !$hebrew ) $hebrew = self
::tsToHebrew( $ts );
672 if ( !$thai ) $thai = self
::tsToThai( $ts );
676 $num = substr( $ts, 2, 2 );
679 $s .= intval( substr( $ts, 8, 2 ) ) < 12 ?
'am' : 'pm';
682 $s .= intval( substr( $ts, 8, 2 ) ) < 12 ?
'AM' : 'PM';
685 $h = substr( $ts, 8, 2 );
686 $num = $h %
12 ?
$h %
12 : 12;
689 $num = intval( substr( $ts, 8, 2 ) );
692 $h = substr( $ts, 8, 2 );
693 $num = sprintf( '%02d', $h %
12 ?
$h %
12 : 12 );
696 $num = substr( $ts, 8, 2 );
699 $num = substr( $ts, 10, 2 );
702 $num = substr( $ts, 12, 2 );
705 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
706 $s .= gmdate( 'c', $unix );
709 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
710 $s .= gmdate( 'r', $unix );
713 if ( !$unix ) $unix = wfTimestamp( TS_UNIX
, $ts );
718 if ( $p < strlen( $format ) - 1 ) {
726 if ( $p < strlen( $format ) - 1 ) {
727 $endQuote = strpos( $format, '"', $p +
1 );
728 if ( $endQuote === false ) {
729 # No terminating quote, assume literal "
732 $s .= substr( $format, $p +
1, $endQuote - $p - 1 );
736 # Quote at end of string, assume literal "
743 if ( $num !== false ) {
744 if ( $rawToggle ||
$raw ) {
747 } elseif ( $roman ) {
748 $s .= self
::romanNumeral( $num );
750 } elseif( $hebrewNum ) {
751 $s .= self
::hebrewNumeral( $num );
754 $s .= $this->formatNum( $num, true );
762 private static $GREG_DAYS = array( 31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31 );
763 private static $IRANIAN_DAYS = array( 31, 31, 31, 31, 31, 31, 30, 30, 30, 30, 30, 29 );
765 * Algorithm by Roozbeh Pournader and Mohammad Toossi to convert
766 * Gregorian dates to Iranian dates. Originally written in C, it
767 * is released under the terms of GNU Lesser General Public
768 * License. Conversion to PHP was performed by Niklas Laxström.
770 * Link: http://www.farsiweb.info/jalali/jalali.c
772 private static function tsToIranian( $ts ) {
773 $gy = substr( $ts, 0, 4 ) -1600;
774 $gm = substr( $ts, 4, 2 ) -1;
775 $gd = substr( $ts, 6, 2 ) -1;
777 # Days passed from the beginning (including leap years)
780 - floor(($gy+
99) / 100)
781 +
floor(($gy+
399) / 400);
784 // Add days of the past months of this year
785 for( $i = 0; $i < $gm; $i++
) {
786 $gDayNo +
= self
::$GREG_DAYS[$i];
790 if ( $gm > 1 && (($gy%4
===0 && $gy%100
!==0 ||
($gy%400
==0)))) {
794 // Days passed in current month
797 $jDayNo = $gDayNo - 79;
799 $jNp = floor($jDayNo / 12053);
802 $jy = 979 +
33*$jNp +
4*floor($jDayNo/1461);
805 if ( $jDayNo >= 366 ) {
806 $jy +
= floor(($jDayNo-1)/365);
807 $jDayNo = floor(($jDayNo-1)%365
);
810 for ( $i = 0; $i < 11 && $jDayNo >= self
::$IRANIAN_DAYS[$i]; $i++
) {
811 $jDayNo -= self
::$IRANIAN_DAYS[$i];
817 return array($jy, $jm, $jd);
821 * Converting Gregorian dates to Hebrew dates.
823 * Based on a JavaScript code by Abu Mami and Yisrael Hersch
824 * (abu-mami@kaluach.net, http://www.kaluach.net), who permitted
825 * to translate the relevant functions into PHP and release them under
828 private static function tsToHebrew( $ts ) {
830 $year = substr( $ts, 0, 4 );
831 $month = substr( $ts, 4, 2 );
832 $day = substr( $ts, 6, 2 );
834 # Month number when March = 1, February = 12
837 # January or February
842 # Days since 1 March - calculating 30 days a month,
843 # and then adding the missing number of days
844 $dayOfYear = $day +
intval( 7 * $month / 12 +
30 * ( $month - 1 ) );
845 # Calculate Hebrew year for days after 1 Nisan
846 $hebrewYear = $year +
3760;
847 # Passover date for this year (as days since 1 March)
848 $passover = self
::passoverDate( $hebrewYear );
849 if( $dayOfYear <= $passover - 15 ) {
850 # Day is before 1 Nisan (passover is 15 Nisan) - it is the previous year
851 # Next year's passover (as days since 1 March)
853 # Add days since previous year's 1 March
855 if( ( $year %
400 == 0 ) ||
( $year %
100 != 0 && $year %
4 == 0 ) ) {
862 # Passover date for the new year (as days since 1 March)
863 $passover = self
::passoverDate( $hebrewYear );
865 # Next year's passover (as days since 1 March)
866 $anchor = self
::passoverDate( $hebrewYear +
1 );
870 $dayOfYear -= $passover - 15;
871 # Difference between this year's passover date by gregorian calendar,
872 # and the next year's one + 12 days. This should be 1 days for a regular year,
873 # but 0 for incomplete one, 2 for complete, and those + 30 days of Adar I
875 $anchor -= $passover - 12;
876 $nextYear = $year +
1;
877 if( ( $nextYear %
400 == 0 ) ||
( $nextYear %
100 != 0 && $nextYear %
4 == 0 ) ) {
878 # Next year is a leap year - difference is growing
882 # Check the year pattern
883 # 0 means an incomplete year, 1 means a regular year, 2 means a complete year
884 # This is mod 30, to work on both leap years (which add 30 days of Adar I)
886 $yearPattern = $anchor %
30;
888 # Calculate day in the month from number of days sine 1 Nisan
889 # Don't check Adar - if the day is not in Adar, we will stop before;
890 # if it is in Adar, we will use it to check if it is Adar I or Adar II
892 for( $month = 0; $month < 11; $month++
) {
893 # Calculate days in this month
894 if( $month == 7 && $yearPattern == 2 ) {
895 # Cheshvan in a complete year (otherwise as the rule below)
897 } else if( $month == 8 && $yearPattern == 0 ) {
898 # Kislev in an incomplete year (otherwise as the rule below)
901 # Even months have 30 days, odd have 29
902 $days = 30 - $month %
2;
904 if( $day <= $days ) {
912 # Now we move to a year from Tishrei
914 # After Tishrei, use next year
917 # Recalculate month number so that we start from Tishrei
918 $month = ( $month +
6 ) %
12 +
1;
920 # Get Adar type and number of days
923 if( $anchor >= 30 ) {
924 # Leap year - check type and set number of days
936 # Non-leap year - just set number of days (29)
941 return array( $hebrewYear, $month, $day, $days );
945 * Based on Carl Friedrich Gauss algorithm for finding Easter date.
946 * Used for Hebrew date.
948 private static function passoverDate( $year ) {
949 $a = intval( ( 12 * $year +
17 ) %
19 );
950 $b = intval( $year %
4 );
951 $m = 32.044093161144 +
1.5542417966212 * $a +
$b / 4.0 - 0.0031777940220923 * $year;
961 $c = intval( ( $Mar +
3 * $year +
5 * $b +
5 ) %
7);
962 if( $c == 0 && $a > 11 && $m >= 0.89772376543210 ) {
964 } else if( $c == 1 && $a > 6 && $m >= 0.63287037037037 ) {
966 } else if( $c == 2 ||
$c == 4 ||
$c == 6 ) {
970 $Mar +
= intval( ( $year - 3760 ) / 100 ) - intval( ( $year - 3760 ) / 400 ) - 2;
975 * Algorithm to convert Gregorian dates to Thai solar dates.
977 * Link: http://en.wikipedia.org/wiki/Thai_solar_calendar
979 * @param string $ts 14-character timestamp
980 * @return array converted year, month, day
982 private static function tsToThai( $ts ) {
983 $gy = substr( $ts, 0, 4 );
984 $gm = substr( $ts, 4, 2 );
985 $gd = substr( $ts, 6, 2 );
987 # Add 543 years to the Gregorian calendar
988 # Months and days are identical
989 $gy_thai = $gy +
543;
991 return array( $gy_thai, $gm, $gd );
996 * Roman number formatting up to 3000
998 static function romanNumeral( $num ) {
999 static $table = array(
1000 array( '', 'I', 'II', 'III', 'IV', 'V', 'VI', 'VII', 'VIII', 'IX', 'X' ),
1001 array( '', 'X', 'XX', 'XXX', 'XL', 'L', 'LX', 'LXX', 'LXXX', 'XC', 'C' ),
1002 array( '', 'C', 'CC', 'CCC', 'CD', 'D', 'DC', 'DCC', 'DCCC', 'CM', 'M' ),
1003 array( '', 'M', 'MM', 'MMM' )
1006 $num = intval( $num );
1007 if ( $num > 3000 ||
$num <= 0 ) {
1012 for ( $pow10 = 1000, $i = 3; $i >= 0; $pow10 /= 10, $i-- ) {
1013 if ( $num >= $pow10 ) {
1014 $s .= $table[$i][floor($num / $pow10)];
1016 $num = $num %
$pow10;
1022 * Hebrew Gematria number formatting up to 9999
1024 static function hebrewNumeral( $num ) {
1025 static $table = array(
1026 array( '', 'א', 'ב', 'ג', 'ד', 'ה', 'ו', 'ז', 'ח', 'ט', 'י' ),
1027 array( '', 'י', 'כ', 'ל', 'מ', 'נ', 'ס', 'ע', 'פ', 'צ', 'ק' ),
1028 array( '', 'ק', 'ר', 'ש', 'ת', 'תק', 'תר', 'תש', 'תת', 'תתק', 'תתר' ),
1029 array( '', 'א', 'ב', 'ג', 'ד', 'ה', 'ו', 'ז', 'ח', 'ט', 'י' )
1032 $num = intval( $num );
1033 if ( $num > 9999 ||
$num <= 0 ) {
1038 for ( $pow10 = 1000, $i = 3; $i >= 0; $pow10 /= 10, $i-- ) {
1039 if ( $num >= $pow10 ) {
1040 if ( $num == 15 ||
$num == 16 ) {
1041 $s .= $table[0][9] . $table[0][$num - 9];
1044 $s .= $table[$i][intval( ( $num / $pow10 ) )];
1045 if( $pow10 == 1000 ) {
1050 $num = $num %
$pow10;
1052 if( strlen( $s ) == 2 ) {
1055 $str = substr( $s, 0, strlen( $s ) - 2 ) . '"';
1056 $str .= substr( $s, strlen( $s ) - 2, 2 );
1058 $start = substr( $str, 0, strlen( $str ) - 2 );
1059 $end = substr( $str, strlen( $str ) - 2 );
1062 $str = $start . 'ך';
1065 $str = $start . 'ם';
1068 $str = $start . 'ן';
1071 $str = $start . 'ף';
1074 $str = $start . 'ץ';
1081 * This is meant to be used by time(), date(), and timeanddate() to get
1082 * the date preference they're supposed to use, it should be used in
1086 * function timeanddate([...], $format = true) {
1087 * $datePreference = $this->dateFormat($format);
1092 * @param mixed $usePrefs: if true, the user's preference is used
1093 * if false, the site/language default is used
1094 * if int/string, assumed to be a format.
1097 function dateFormat( $usePrefs = true ) {
1100 if( is_bool( $usePrefs ) ) {
1102 $datePreference = $wgUser->getDatePreference();
1104 $options = User
::getDefaultOptions();
1105 $datePreference = (string)$options['date'];
1108 $datePreference = (string)$usePrefs;
1112 if( $datePreference == '' ) {
1116 return $datePreference;
1121 * @param mixed $ts the time format which needs to be turned into a
1122 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
1123 * @param bool $adj whether to adjust the time output according to the
1124 * user configured offset ($timecorrection)
1125 * @param mixed $format true to use user's date format preference
1126 * @param string $timecorrection the time offset as returned by
1127 * validateTimeZone() in Special:Preferences
1130 function date( $ts, $adj = false, $format = true, $timecorrection = false ) {
1133 $ts = $this->userAdjust( $ts, $timecorrection );
1136 $pref = $this->dateFormat( $format );
1137 if( $pref == 'default' ||
!isset( $this->dateFormats
["$pref date"] ) ) {
1138 $pref = $this->defaultDateFormat
;
1140 return $this->sprintfDate( $this->dateFormats
["$pref date"], $ts );
1145 * @param mixed $ts the time format which needs to be turned into a
1146 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
1147 * @param bool $adj whether to adjust the time output according to the
1148 * user configured offset ($timecorrection)
1149 * @param mixed $format true to use user's date format preference
1150 * @param string $timecorrection the time offset as returned by
1151 * validateTimeZone() in Special:Preferences
1154 function time( $ts, $adj = false, $format = true, $timecorrection = false ) {
1157 $ts = $this->userAdjust( $ts, $timecorrection );
1160 $pref = $this->dateFormat( $format );
1161 if( $pref == 'default' ||
!isset( $this->dateFormats
["$pref time"] ) ) {
1162 $pref = $this->defaultDateFormat
;
1164 return $this->sprintfDate( $this->dateFormats
["$pref time"], $ts );
1169 * @param mixed $ts the time format which needs to be turned into a
1170 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
1171 * @param bool $adj whether to adjust the time output according to the
1172 * user configured offset ($timecorrection)
1174 * @param mixed $format what format to return, if it's false output the
1175 * default one (default true)
1176 * @param string $timecorrection the time offset as returned by
1177 * validateTimeZone() in Special:Preferences
1180 function timeanddate( $ts, $adj = false, $format = true, $timecorrection = false) {
1183 $ts = wfTimestamp( TS_MW
, $ts );
1186 $ts = $this->userAdjust( $ts, $timecorrection );
1189 $pref = $this->dateFormat( $format );
1190 if( $pref == 'default' ||
!isset( $this->dateFormats
["$pref both"] ) ) {
1191 $pref = $this->defaultDateFormat
;
1194 return $this->sprintfDate( $this->dateFormats
["$pref both"], $ts );
1197 function getMessage( $key ) {
1199 return isset( $this->messages
[$key] ) ?
$this->messages
[$key] : null;
1202 function getAllMessages() {
1204 return $this->messages
;
1207 function iconv( $in, $out, $string ) {
1208 # For most languages, this is a wrapper for iconv
1209 return iconv( $in, $out . '//IGNORE', $string );
1212 // callback functions for uc(), lc(), ucwords(), ucwordbreaks()
1213 function ucwordbreaksCallbackAscii($matches){
1214 return $this->ucfirst($matches[1]);
1217 function ucwordbreaksCallbackMB($matches){
1218 return mb_strtoupper($matches[0]);
1221 function ucCallback($matches){
1222 list( $wikiUpperChars ) = self
::getCaseMaps();
1223 return strtr( $matches[1], $wikiUpperChars );
1226 function lcCallback($matches){
1227 list( , $wikiLowerChars ) = self
::getCaseMaps();
1228 return strtr( $matches[1], $wikiLowerChars );
1231 function ucwordsCallbackMB($matches){
1232 return mb_strtoupper($matches[0]);
1235 function ucwordsCallbackWiki($matches){
1236 list( $wikiUpperChars ) = self
::getCaseMaps();
1237 return strtr( $matches[0], $wikiUpperChars );
1240 function ucfirst( $str ) {
1241 if ( empty($str) ) return $str;
1242 if ( ord($str[0]) < 128 ) return ucfirst($str);
1243 else return self
::uc($str,true); // fall back to more complex logic in case of multibyte strings
1246 function uc( $str, $first = false ) {
1247 if ( function_exists( 'mb_strtoupper' ) ) {
1249 if ( self
::isMultibyte( $str ) ) {
1250 return mb_strtoupper( mb_substr( $str, 0, 1 ) ) . mb_substr( $str, 1 );
1252 return ucfirst( $str );
1255 return self
::isMultibyte( $str ) ?
mb_strtoupper( $str ) : strtoupper( $str );
1258 if ( self
::isMultibyte( $str ) ) {
1259 list( $wikiUpperChars ) = $this->getCaseMaps();
1260 $x = $first ?
'^' : '';
1261 return preg_replace_callback(
1262 "/$x([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/",
1263 array($this,"ucCallback"),
1267 return $first ?
ucfirst( $str ) : strtoupper( $str );
1272 function lcfirst( $str ) {
1273 if ( empty($str) ) return $str;
1274 if ( is_string( $str ) && ord($str[0]) < 128 ) {
1275 // editing string in place = cool
1276 $str[0]=strtolower($str[0]);
1279 else return self
::lc( $str, true );
1282 function lc( $str, $first = false ) {
1283 if ( function_exists( 'mb_strtolower' ) )
1285 if ( self
::isMultibyte( $str ) )
1286 return mb_strtolower( mb_substr( $str, 0, 1 ) ) . mb_substr( $str, 1 );
1288 return strtolower( substr( $str, 0, 1 ) ) . substr( $str, 1 );
1290 return self
::isMultibyte( $str ) ?
mb_strtolower( $str ) : strtolower( $str );
1292 if ( self
::isMultibyte( $str ) ) {
1293 list( , $wikiLowerChars ) = self
::getCaseMaps();
1294 $x = $first ?
'^' : '';
1295 return preg_replace_callback(
1296 "/$x([A-Z]|[\\xc0-\\xff][\\x80-\\xbf]*)/",
1297 array($this,"lcCallback"),
1301 return $first ?
strtolower( substr( $str, 0, 1 ) ) . substr( $str, 1 ) : strtolower( $str );
1304 function isMultibyte( $str ) {
1305 return (bool)preg_match( '/[\x80-\xff]/', $str );
1308 function ucwords($str) {
1309 if ( self
::isMultibyte( $str ) ) {
1310 $str = self
::lc($str);
1312 // regexp to find first letter in each word (i.e. after each space)
1313 $replaceRegexp = "/^([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)| ([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/";
1315 // function to use to capitalize a single char
1316 if ( function_exists( 'mb_strtoupper' ) )
1317 return preg_replace_callback(
1319 array($this,"ucwordsCallbackMB"),
1323 return preg_replace_callback(
1325 array($this,"ucwordsCallbackWiki"),
1330 return ucwords( strtolower( $str ) );
1333 # capitalize words at word breaks
1334 function ucwordbreaks($str){
1335 if (self
::isMultibyte( $str ) ) {
1336 $str = self
::lc($str);
1338 // since \b doesn't work for UTF-8, we explicitely define word break chars
1339 $breaks= "[ \-\(\)\}\{\.,\?!]";
1341 // find first letter after word break
1342 $replaceRegexp = "/^([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)|$breaks([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/";
1344 if ( function_exists( 'mb_strtoupper' ) )
1345 return preg_replace_callback(
1347 array($this,"ucwordbreaksCallbackMB"),
1351 return preg_replace_callback(
1353 array($this,"ucwordsCallbackWiki"),
1358 return preg_replace_callback(
1359 '/\b([\w\x80-\xff]+)\b/',
1360 array($this,"ucwordbreaksCallbackAscii"),
1365 * Return a case-folded representation of $s
1367 * This is a representation such that caseFold($s1)==caseFold($s2) if $s1
1368 * and $s2 are the same except for the case of their characters. It is not
1369 * necessary for the value returned to make sense when displayed.
1371 * Do *not* perform any other normalisation in this function. If a caller
1372 * uses this function when it should be using a more general normalisation
1373 * function, then fix the caller.
1375 function caseFold( $s ) {
1376 return $this->uc( $s );
1379 function checkTitleEncoding( $s ) {
1380 if( is_array( $s ) ) {
1381 wfDebugDieBacktrace( 'Given array to checkTitleEncoding.' );
1383 # Check for non-UTF-8 URLs
1384 $ishigh = preg_match( '/[\x80-\xff]/', $s);
1385 if(!$ishigh) return $s;
1387 $isutf8 = preg_match( '/^([\x00-\x7f]|[\xc0-\xdf][\x80-\xbf]|' .
1388 '[\xe0-\xef][\x80-\xbf]{2}|[\xf0-\xf7][\x80-\xbf]{3})+$/', $s );
1389 if( $isutf8 ) return $s;
1391 return $this->iconv( $this->fallback8bitEncoding(), "utf-8", $s );
1394 function fallback8bitEncoding() {
1396 return $this->fallback8bitEncoding
;
1400 * Some languages have special punctuation to strip out
1401 * or characters which need to be converted for MySQL's
1402 * indexing to grok it correctly. Make such changes here.
1407 function stripForSearch( $string ) {
1409 if ( $wgDBtype != 'mysql' ) {
1413 # MySQL fulltext index doesn't grok utf-8, so we
1414 # need to fold cases and convert to hex
1416 wfProfileIn( __METHOD__
);
1417 if( function_exists( 'mb_strtolower' ) ) {
1418 $out = preg_replace(
1419 "/([\\xc0-\\xff][\\x80-\\xbf]*)/e",
1420 "'U8' . bin2hex( \"$1\" )",
1421 mb_strtolower( $string ) );
1423 list( , $wikiLowerChars ) = self
::getCaseMaps();
1424 $out = preg_replace(
1425 "/([\\xc0-\\xff][\\x80-\\xbf]*)/e",
1426 "'U8' . bin2hex( strtr( \"\$1\", \$wikiLowerChars ) )",
1429 wfProfileOut( __METHOD__
);
1433 function convertForSearchResult( $termsArray ) {
1434 # some languages, e.g. Chinese, need to do a conversion
1435 # in order for search results to be displayed correctly
1440 * Get the first character of a string.
1445 function firstChar( $s ) {
1447 preg_match( '/^([\x00-\x7f]|[\xc0-\xdf][\x80-\xbf]|' .
1448 '[\xe0-\xef][\x80-\xbf]{2}|[\xf0-\xf7][\x80-\xbf]{3})/', $s, $matches);
1450 return isset( $matches[1] ) ?
$matches[1] : "";
1453 function initEncoding() {
1454 # Some languages may have an alternate char encoding option
1455 # (Esperanto X-coding, Japanese furigana conversion, etc)
1456 # If this language is used as the primary content language,
1457 # an override to the defaults can be set here on startup.
1460 function recodeForEdit( $s ) {
1461 # For some languages we'll want to explicitly specify
1462 # which characters make it into the edit box raw
1463 # or are converted in some way or another.
1464 # Note that if wgOutputEncoding is different from
1465 # wgInputEncoding, this text will be further converted
1466 # to wgOutputEncoding.
1467 global $wgEditEncoding;
1468 if( $wgEditEncoding == '' or
1469 $wgEditEncoding == 'UTF-8' ) {
1472 return $this->iconv( 'UTF-8', $wgEditEncoding, $s );
1476 function recodeInput( $s ) {
1477 # Take the previous into account.
1478 global $wgEditEncoding;
1479 if($wgEditEncoding != "") {
1480 $enc = $wgEditEncoding;
1484 if( $enc == 'UTF-8' ) {
1487 return $this->iconv( $enc, 'UTF-8', $s );
1492 * For right-to-left language support
1502 * A hidden direction mark (LRM or RLM), depending on the language direction
1506 function getDirMark() {
1507 return $this->isRTL() ?
"\xE2\x80\x8F" : "\xE2\x80\x8E";
1511 * An arrow, depending on the language direction
1515 function getArrow() {
1516 return $this->isRTL() ?
'←' : '→';
1520 * To allow "foo[[bar]]" to extend the link over the whole word "foobar"
1524 function linkPrefixExtension() {
1526 return $this->linkPrefixExtension
;
1529 function &getMagicWords() {
1531 return $this->magicWords
;
1534 # Fill a MagicWord object with data from here
1535 function getMagic( &$mw ) {
1536 if ( !$this->mMagicHookDone
) {
1537 $this->mMagicHookDone
= true;
1538 wfRunHooks( 'LanguageGetMagic', array( &$this->mMagicExtensions
, $this->getCode() ) );
1540 if ( isset( $this->mMagicExtensions
[$mw->mId
] ) ) {
1541 $rawEntry = $this->mMagicExtensions
[$mw->mId
];
1543 $magicWords =& $this->getMagicWords();
1544 if ( isset( $magicWords[$mw->mId
] ) ) {
1545 $rawEntry = $magicWords[$mw->mId
];
1547 # Fall back to English if local list is incomplete
1548 $magicWords =& Language
::getMagicWords();
1549 $rawEntry = $magicWords[$mw->mId
];
1553 if( !is_array( $rawEntry ) ) {
1554 error_log( "\"$rawEntry\" is not a valid magic thingie for \"$mw->mId\"" );
1556 $mw->mCaseSensitive
= $rawEntry[0];
1557 $mw->mSynonyms
= array_slice( $rawEntry, 1 );
1562 * Add magic words to the extension array
1564 function addMagicWordsByLang( $newWords ) {
1565 $code = $this->getCode();
1566 $fallbackChain = array();
1567 while ( $code && !in_array( $code, $fallbackChain ) ) {
1568 $fallbackChain[] = $code;
1569 $code = self
::getFallbackFor( $code );
1571 if ( !in_array( 'en', $fallbackChain ) ) {
1572 $fallbackChain[] = 'en';
1574 $fallbackChain = array_reverse( $fallbackChain );
1575 foreach ( $fallbackChain as $code ) {
1576 if ( isset( $newWords[$code] ) ) {
1577 $this->mMagicExtensions
= $newWords[$code] +
$this->mMagicExtensions
;
1583 * Get special page names, as an associative array
1584 * case folded alias => real name
1586 function getSpecialPageAliases() {
1588 if ( !isset( $this->mExtendedSpecialPageAliases
) ) {
1589 $this->mExtendedSpecialPageAliases
= $this->specialPageAliases
;
1590 wfRunHooks( 'LanguageGetSpecialPageAliases',
1591 array( &$this->mExtendedSpecialPageAliases
, $this->getCode() ) );
1593 return $this->mExtendedSpecialPageAliases
;
1597 * Italic is unsuitable for some languages
1601 * @param string $text The text to be emphasized.
1604 function emphasize( $text ) {
1605 return "<em>$text</em>";
1609 * Normally we output all numbers in plain en_US style, that is
1610 * 293,291.235 for twohundredninetythreethousand-twohundredninetyone
1611 * point twohundredthirtyfive. However this is not sutable for all
1612 * languages, some such as Pakaran want ੨੯੩,੨੯੫.੨੩੫ and others such as
1613 * Icelandic just want to use commas instead of dots, and dots instead
1614 * of commas like "293.291,235".
1616 * An example of this function being called:
1618 * wfMsg( 'message', $wgLang->formatNum( $num ) )
1621 * See LanguageGu.php for the Gujarati implementation and
1622 * LanguageIs.php for the , => . and . => , implementation.
1624 * @todo check if it's viable to use localeconv() for the decimal
1627 * @param mixed $number the string to be formatted, should be an integer or
1628 * a floating point number.
1629 * @param bool $nocommafy Set to true for special numbers like dates
1632 function formatNum( $number, $nocommafy = false ) {
1633 global $wgTranslateNumerals;
1635 $number = $this->commafy($number);
1636 $s = $this->separatorTransformTable();
1637 if (!is_null($s)) { $number = strtr($number, $s); }
1640 if ($wgTranslateNumerals) {
1641 $s = $this->digitTransformTable();
1642 if (!is_null($s)) { $number = strtr($number, $s); }
1648 function parseFormattedNumber( $number ) {
1649 $s = $this->digitTransformTable();
1650 if (!is_null($s)) { $number = strtr($number, array_flip($s)); }
1652 $s = $this->separatorTransformTable();
1653 if (!is_null($s)) { $number = strtr($number, array_flip($s)); }
1655 $number = strtr( $number, array (',' => '') );
1660 * Adds commas to a given number
1665 function commafy($_) {
1666 return strrev((string)preg_replace('/(\d{3})(?=\d)(?!\d*\.)/','$1,',strrev($_)));
1669 function digitTransformTable() {
1671 return $this->digitTransformTable
;
1674 function separatorTransformTable() {
1676 return $this->separatorTransformTable
;
1681 * For the credit list in includes/Credits.php (action=credits)
1686 function listToText( $l ) {
1689 for ($i = $m; $i >= 0; $i--) {
1692 } else if ($i == $m - 1) {
1693 $s = $l[$i] . ' ' . $this->getMessageFromDB( 'and' ) . ' ' . $s;
1695 $s = $l[$i] . ', ' . $s;
1702 * Truncate a string to a specified length in bytes, appending an optional
1703 * string (e.g. for ellipses)
1705 * The database offers limited byte lengths for some columns in the database;
1706 * multi-byte character sets mean we need to ensure that only whole characters
1707 * are included, otherwise broken characters can be passed to the user
1709 * If $length is negative, the string will be truncated from the beginning
1711 * @param string $string String to truncate
1712 * @param int $length Maximum length (excluding ellipses)
1713 * @param string $ellipses String to append to the truncated text
1716 function truncate( $string, $length, $ellipsis = "" ) {
1717 if( $length == 0 ) {
1720 if ( strlen( $string ) <= abs( $length ) ) {
1724 $string = substr( $string, 0, $length );
1725 $char = ord( $string[strlen( $string ) - 1] );
1727 if ($char >= 0xc0) {
1728 # We got the first byte only of a multibyte char; remove it.
1729 $string = substr( $string, 0, -1 );
1730 } elseif( $char >= 0x80 &&
1731 preg_match( '/^(.*)(?:[\xe0-\xef][\x80-\xbf]|' .
1732 '[\xf0-\xf7][\x80-\xbf]{1,2})$/', $string, $m ) ) {
1733 # We chopped in the middle of a character; remove it
1736 return $string . $ellipsis;
1738 $string = substr( $string, $length );
1739 $char = ord( $string[0] );
1740 if( $char >= 0x80 && $char < 0xc0 ) {
1741 # We chopped in the middle of a character; remove the whole thing
1742 $string = preg_replace( '/^[\x80-\xbf]+/', '', $string );
1744 return $ellipsis . $string;
1749 * Grammatical transformations, needed for inflected languages
1750 * Invoked by putting {{grammar:case|word}} in a message
1752 * @param string $word
1753 * @param string $case
1756 function convertGrammar( $word, $case ) {
1757 global $wgGrammarForms;
1758 if ( isset($wgGrammarForms['en'][$case][$word]) ) {
1759 return $wgGrammarForms['en'][$case][$word];
1765 * Plural form transformations, needed for some languages.
1766 * For example, there are 3 form of plural in Russian and Polish,
1767 * depending on "count mod 10". See [[w:Plural]]
1768 * For English it is pretty simple.
1770 * Invoked by putting {{plural:count|wordform1|wordform2}}
1771 * or {{plural:count|wordform1|wordform2|wordform3}}
1773 * Example: {{plural:{{NUMBEROFARTICLES}}|article|articles}}
1775 * @param integer $count Non-localized number
1776 * @param array $forms Different plural forms
1777 * @return string Correct form of plural for $count in this language
1779 function convertPlural( $count, $forms ) {
1780 if ( !count($forms) ) { return ''; }
1781 $forms = $this->preConvertPlural( $forms, 2 );
1783 return ( abs($count) == 1 ) ?
$forms[0] : $forms[1];
1787 * Checks that convertPlural was given an array and pads it to requested
1788 * amound of forms by copying the last one.
1790 * @param integer $count How many forms should there be at least
1791 * @param array $forms Array of forms given to convertPlural
1792 * @return array Padded array of forms or an exception if not an array
1794 protected function preConvertPlural( Array $forms, $count ) {
1795 while ( count($forms) < $count ) {
1796 $forms[] = $forms[count($forms)-1];
1802 * For translaing of expiry times
1803 * @param string The validated block time in English
1804 * @return Somehow translated block time
1805 * @see LanguageFi.php for example implementation
1807 function translateBlockExpiry( $str ) {
1809 $scBlockExpiryOptions = $this->getMessageFromDB( 'ipboptions' );
1811 if ( $scBlockExpiryOptions == '-') {
1815 foreach (explode(',', $scBlockExpiryOptions) as $option) {
1816 if ( strpos($option, ":") === false )
1818 list($show, $value) = explode(":", $option);
1819 if ( strcmp ( $str, $value) == 0 ) {
1820 return htmlspecialchars( trim( $show ) );
1828 * languages like Chinese need to be segmented in order for the diff
1831 * @param string $text
1834 function segmentForDiff( $text ) {
1839 * and unsegment to show the result
1841 * @param string $text
1844 function unsegmentForDiff( $text ) {
1848 # convert text to different variants of a language.
1849 function convert( $text, $isTitle = false) {
1850 return $this->mConverter
->convert($text, $isTitle);
1853 # Convert text from within Parser
1854 function parserConvert( $text, &$parser ) {
1855 return $this->mConverter
->parserConvert( $text, $parser );
1858 # Check if this is a language with variants
1859 function hasVariants(){
1860 return sizeof($this->getVariants())>1;
1863 # Put custom tags (e.g. -{ }-) around math to prevent conversion
1864 function armourMath($text){
1865 return $this->mConverter
->armourMath($text);
1870 * Perform output conversion on a string, and encode for safe HTML output.
1871 * @param string $text
1872 * @param bool $isTitle -- wtf?
1874 * @todo this should get integrated somewhere sane
1876 function convertHtml( $text, $isTitle = false ) {
1877 return htmlspecialchars( $this->convert( $text, $isTitle ) );
1880 function convertCategoryKey( $key ) {
1881 return $this->mConverter
->convertCategoryKey( $key );
1885 * get the list of variants supported by this langauge
1886 * see sample implementation in LanguageZh.php
1888 * @return array an array of language codes
1890 function getVariants() {
1891 return $this->mConverter
->getVariants();
1895 function getPreferredVariant( $fromUser = true ) {
1896 return $this->mConverter
->getPreferredVariant( $fromUser );
1900 * if a language supports multiple variants, it is
1901 * possible that non-existing link in one variant
1902 * actually exists in another variant. this function
1903 * tries to find it. See e.g. LanguageZh.php
1905 * @param string $link the name of the link
1906 * @param mixed $nt the title object of the link
1907 * @return null the input parameters may be modified upon return
1909 function findVariantLink( &$link, &$nt ) {
1910 $this->mConverter
->findVariantLink($link, $nt);
1914 * If a language supports multiple variants, converts text
1915 * into an array of all possible variants of the text:
1916 * 'variant' => text in that variant
1919 function convertLinkToAllVariants($text){
1920 return $this->mConverter
->convertLinkToAllVariants($text);
1925 * returns language specific options used by User::getPageRenderHash()
1926 * for example, the preferred language variant
1931 function getExtraHashOptions() {
1932 return $this->mConverter
->getExtraHashOptions();
1936 * for languages that support multiple variants, the title of an
1937 * article may be displayed differently in different variants. this
1938 * function returns the apporiate title defined in the body of the article.
1942 function getParsedTitle() {
1943 return $this->mConverter
->getParsedTitle();
1947 * Enclose a string with the "no conversion" tag. This is used by
1948 * various functions in the Parser
1950 * @param string $text text to be tagged for no conversion
1951 * @return string the tagged text
1953 function markNoConversion( $text, $noParse=false ) {
1954 return $this->mConverter
->markNoConversion( $text, $noParse );
1958 * A regular expression to match legal word-trailing characters
1959 * which should be merged onto a link of the form [[foo]]bar.
1964 function linkTrail() {
1966 return $this->linkTrail
;
1969 function getLangObj() {
1974 * Get the RFC 3066 code for this language object
1976 function getCode() {
1977 return $this->mCode
;
1980 function setCode( $code ) {
1981 $this->mCode
= $code;
1984 static function getFileName( $prefix = 'Language', $code, $suffix = '.php' ) {
1985 return $prefix . str_replace( '-', '_', ucfirst( $code ) ) . $suffix;
1988 static function getMessagesFileName( $code ) {
1990 return self
::getFileName( "$IP/languages/messages/Messages", $code, '.php' );
1993 static function getClassFileName( $code ) {
1995 return self
::getFileName( "$IP/languages/classes/Language", $code, '.php' );
1998 static function getLocalisationArray( $code, $disableCache = false ) {
1999 self
::loadLocalisation( $code, $disableCache );
2000 return self
::$mLocalisationCache[$code];
2004 * Load localisation data for a given code into the static cache
2006 * @return array Dependencies, map of filenames to mtimes
2008 static function loadLocalisation( $code, $disableCache = false ) {
2009 static $recursionGuard = array();
2010 global $wgMemc, $wgCheckSerialized;
2013 throw new MWException( "Invalid language code requested" );
2016 if ( !$disableCache ) {
2017 # Try the per-process cache
2018 if ( isset( self
::$mLocalisationCache[$code] ) ) {
2019 return self
::$mLocalisationCache[$code]['deps'];
2022 wfProfileIn( __METHOD__
);
2024 # Try the serialized directory
2025 $cache = wfGetPrecompiledData( self
::getFileName( "Messages", $code, '.ser' ) );
2027 if ( $wgCheckSerialized && self
::isLocalisationOutOfDate( $cache ) ) {
2029 wfDebug( "Language::loadLocalisation(): precompiled data file for $code is out of date\n" );
2031 self
::$mLocalisationCache[$code] = $cache;
2032 wfDebug( "Language::loadLocalisation(): got localisation for $code from precompiled data file\n" );
2033 wfProfileOut( __METHOD__
);
2034 return self
::$mLocalisationCache[$code]['deps'];
2038 # Try the global cache
2039 $memcKey = wfMemcKey('localisation', $code );
2040 $cache = $wgMemc->get( $memcKey );
2042 if ( self
::isLocalisationOutOfDate( $cache ) ) {
2043 $wgMemc->delete( $memcKey );
2045 wfDebug( "Language::loadLocalisation(): localisation cache for $code had expired\n" );
2047 self
::$mLocalisationCache[$code] = $cache;
2048 wfDebug( "Language::loadLocalisation(): got localisation for $code from cache\n" );
2049 wfProfileOut( __METHOD__
);
2050 return $cache['deps'];
2054 wfProfileIn( __METHOD__
);
2057 # Default fallback, may be overridden when the messages file is included
2058 if ( $code != 'en' ) {
2064 # Load the primary localisation from the source file
2065 $filename = self
::getMessagesFileName( $code );
2066 if ( !file_exists( $filename ) ) {
2067 wfDebug( "Language::loadLocalisation(): no localisation file for $code, using implicit fallback to en\n" );
2071 $deps = array( $filename => filemtime( $filename ) );
2072 require( $filename );
2073 $cache = compact( self
::$mLocalisationKeys );
2074 wfDebug( "Language::loadLocalisation(): got localisation for $code from source\n" );
2077 if ( !empty( $fallback ) ) {
2078 # Load the fallback localisation, with a circular reference guard
2079 if ( isset( $recursionGuard[$code] ) ) {
2080 throw new MWException( "Error: Circular fallback reference in language code $code" );
2082 $recursionGuard[$code] = true;
2083 $newDeps = self
::loadLocalisation( $fallback, $disableCache );
2084 unset( $recursionGuard[$code] );
2086 $secondary = self
::$mLocalisationCache[$fallback];
2087 $deps = array_merge( $deps, $newDeps );
2089 # Merge the fallback localisation with the current localisation
2090 foreach ( self
::$mLocalisationKeys as $key ) {
2091 if ( isset( $cache[$key] ) ) {
2092 if ( isset( $secondary[$key] ) ) {
2093 if ( in_array( $key, self
::$mMergeableMapKeys ) ) {
2094 $cache[$key] = $cache[$key] +
$secondary[$key];
2095 } elseif ( in_array( $key, self
::$mMergeableListKeys ) ) {
2096 $cache[$key] = array_merge( $secondary[$key], $cache[$key] );
2097 } elseif ( in_array( $key, self
::$mMergeableAliasListKeys ) ) {
2098 $cache[$key] = array_merge_recursive( $cache[$key], $secondary[$key] );
2102 $cache[$key] = $secondary[$key];
2106 # Merge bookstore lists if requested
2107 if ( !empty( $cache['bookstoreList']['inherit'] ) ) {
2108 $cache['bookstoreList'] = array_merge( $cache['bookstoreList'], $secondary['bookstoreList'] );
2110 if ( isset( $cache['bookstoreList']['inherit'] ) ) {
2111 unset( $cache['bookstoreList']['inherit'] );
2115 # Add dependencies to the cache entry
2116 $cache['deps'] = $deps;
2118 # Replace spaces with underscores in namespace names
2119 $cache['namespaceNames'] = str_replace( ' ', '_', $cache['namespaceNames'] );
2121 # Save to both caches
2122 self
::$mLocalisationCache[$code] = $cache;
2123 if ( !$disableCache ) {
2124 $wgMemc->set( $memcKey, $cache );
2127 wfProfileOut( __METHOD__
);
2132 * Test if a given localisation cache is out of date with respect to the
2133 * source Messages files. This is done automatically for the global cache
2134 * in $wgMemc, but is only done on certain occasions for the serialized
2137 * @param $cache mixed Either a language code or a cache array
2139 static function isLocalisationOutOfDate( $cache ) {
2140 if ( !is_array( $cache ) ) {
2141 self
::loadLocalisation( $cache );
2142 $cache = self
::$mLocalisationCache[$cache];
2145 foreach ( $cache['deps'] as $file => $mtime ) {
2146 if ( !file_exists( $file ) ||
filemtime( $file ) > $mtime ) {
2155 * Get the fallback for a given language
2157 static function getFallbackFor( $code ) {
2158 self
::loadLocalisation( $code );
2159 return self
::$mLocalisationCache[$code]['fallback'];
2163 * Get all messages for a given language
2165 static function getMessagesFor( $code ) {
2166 self
::loadLocalisation( $code );
2167 return self
::$mLocalisationCache[$code]['messages'];
2171 * Get a message for a given language
2173 static function getMessageFor( $key, $code ) {
2174 self
::loadLocalisation( $code );
2175 return isset( self
::$mLocalisationCache[$code]['messages'][$key] ) ? self
::$mLocalisationCache[$code]['messages'][$key] : null;
2179 * Load localisation data for this object
2182 if ( !$this->mLoaded
) {
2183 self
::loadLocalisation( $this->getCode() );
2184 $cache =& self
::$mLocalisationCache[$this->getCode()];
2185 foreach ( self
::$mLocalisationKeys as $key ) {
2186 $this->$key = $cache[$key];
2188 $this->mLoaded
= true;
2190 $this->fixUpSettings();
2195 * Do any necessary post-cache-load settings adjustment
2197 function fixUpSettings() {
2198 global $wgExtraNamespaces, $wgMetaNamespace, $wgMetaNamespaceTalk,
2199 $wgNamespaceAliases, $wgAmericanDates;
2200 wfProfileIn( __METHOD__
);
2201 if ( $wgExtraNamespaces ) {
2202 $this->namespaceNames
= $wgExtraNamespaces +
$this->namespaceNames
;
2205 $this->namespaceNames
[NS_PROJECT
] = $wgMetaNamespace;
2206 if ( $wgMetaNamespaceTalk ) {
2207 $this->namespaceNames
[NS_PROJECT_TALK
] = $wgMetaNamespaceTalk;
2209 $talk = $this->namespaceNames
[NS_PROJECT_TALK
];
2210 $talk = str_replace( '$1', $wgMetaNamespace, $talk );
2212 # Allow grammar transformations
2213 # Allowing full message-style parsing would make simple requests
2214 # such as action=raw much more expensive than they need to be.
2215 # This will hopefully cover most cases.
2216 $talk = preg_replace_callback( '/{{grammar:(.*?)\|(.*?)}}/i',
2217 array( &$this, 'replaceGrammarInNamespace' ), $talk );
2218 $talk = str_replace( ' ', '_', $talk );
2219 $this->namespaceNames
[NS_PROJECT_TALK
] = $talk;
2222 # The above mixing may leave namespaces out of canonical order.
2223 # Re-order by namespace ID number...
2224 ksort( $this->namespaceNames
);
2226 # Put namespace names and aliases into a hashtable.
2227 # If this is too slow, then we should arrange it so that it is done
2228 # before caching. The catch is that at pre-cache time, the above
2229 # class-specific fixup hasn't been done.
2230 $this->mNamespaceIds
= array();
2231 foreach ( $this->namespaceNames
as $index => $name ) {
2232 $this->mNamespaceIds
[$this->lc($name)] = $index;
2234 if ( $this->namespaceAliases
) {
2235 foreach ( $this->namespaceAliases
as $name => $index ) {
2236 $this->mNamespaceIds
[$this->lc($name)] = $index;
2239 if ( $wgNamespaceAliases ) {
2240 foreach ( $wgNamespaceAliases as $name => $index ) {
2241 $this->mNamespaceIds
[$this->lc($name)] = $index;
2245 if ( $this->defaultDateFormat
== 'dmy or mdy' ) {
2246 $this->defaultDateFormat
= $wgAmericanDates ?
'mdy' : 'dmy';
2248 wfProfileOut( __METHOD__
);
2251 function replaceGrammarInNamespace( $m ) {
2252 return $this->convertGrammar( trim( $m[2] ), trim( $m[1] ) );
2255 static function getCaseMaps() {
2256 static $wikiUpperChars, $wikiLowerChars;
2257 if ( isset( $wikiUpperChars ) ) {
2258 return array( $wikiUpperChars, $wikiLowerChars );
2261 wfProfileIn( __METHOD__
);
2262 $arr = wfGetPrecompiledData( 'Utf8Case.ser' );
2263 if ( $arr === false ) {
2264 throw new MWException(
2265 "Utf8Case.ser is missing, please run \"make\" in the serialized directory\n" );
2268 wfProfileOut( __METHOD__
);
2269 return array( $wikiUpperChars, $wikiLowerChars );
2272 function formatTimePeriod( $seconds ) {
2273 if ( $seconds < 10 ) {
2274 return $this->formatNum( sprintf( "%.1f", $seconds ) ) . wfMsg( 'seconds-abbrev' );
2275 } elseif ( $seconds < 60 ) {
2276 return $this->formatNum( round( $seconds ) ) . wfMsg( 'seconds-abbrev' );
2277 } elseif ( $seconds < 3600 ) {
2278 return $this->formatNum( floor( $seconds / 60 ) ) . wfMsg( 'minutes-abbrev' ) .
2279 $this->formatNum( round( fmod( $seconds, 60 ) ) ) . wfMsg( 'seconds-abbrev' );
2281 $hours = floor( $seconds / 3600 );
2282 $minutes = floor( ( $seconds - $hours * 3600 ) / 60 );
2283 $secondsPart = round( $seconds - $hours * 3600 - $minutes * 60 );
2284 return $this->formatNum( $hours ) . wfMsg( 'hours-abbrev' ) .
2285 $this->formatNum( $minutes ) . wfMsg( 'minutes-abbrev' ) .
2286 $this->formatNum( $secondsPart ) . wfMsg( 'seconds-abbrev' );
2290 function formatBitrate( $bps ) {
2291 $units = array( 'bps', 'kbps', 'Mbps', 'Gbps' );
2293 return $this->formatNum( $bps ) . $units[0];
2295 $unitIndex = floor( log10( $bps ) / 3 );
2296 $mantissa = $bps / pow( 1000, $unitIndex );
2297 if ( $mantissa < 10 ) {
2298 $mantissa = round( $mantissa, 1 );
2300 $mantissa = round( $mantissa );
2302 return $this->formatNum( $mantissa ) . $units[$unitIndex];
2306 * Format a size in bytes for output, using an appropriate
2307 * unit (B, KB, MB or GB) according to the magnitude in question
2309 * @param $size Size to format
2310 * @return string Plain text (not HTML)
2312 function formatSize( $size ) {
2313 // For small sizes no decimal places necessary
2315 if( $size > 1024 ) {
2316 $size = $size / 1024;
2317 if( $size > 1024 ) {
2318 $size = $size / 1024;
2319 // For MB and bigger two decimal places are smarter
2321 if( $size > 1024 ) {
2322 $size = $size / 1024;
2323 $msg = 'size-gigabytes';
2325 $msg = 'size-megabytes';
2328 $msg = 'size-kilobytes';
2331 $msg = 'size-bytes';
2333 $size = round( $size, $round );
2334 $text = $this->getMessageFromDB( $msg );
2335 return str_replace( '$1', $this->formatNum( $size ), $text );