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 trigger_error( 'Use of ' . __METHOD__
. ' is deprecated', E_USER_NOTICE
);
175 return User
::getDefaultOptions();
178 function getFallbackLanguageCode() {
179 return self
::getFallbackFor( $this->mCode
);
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 = MWNamespace
::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 # Calculate Hebrew year
835 $hebrewYear = $year +
3760;
837 # Month number when September = 1, August = 12
846 # Calculate day of year from 1 September
848 for( $i = 1; $i < $month; $i++
) {
852 # Check if the year is leap
853 if( $year %
400 == 0 ||
( $year %
4 == 0 && $year %
100 > 0 ) ) {
856 } elseif( $i == 8 ||
$i == 10 ||
$i == 1 ||
$i == 3 ) {
863 # Calculate the start of the Hebrew year
864 $start = self
::hebrewYearStart( $hebrewYear );
866 # Calculate next year's start
867 if( $dayOfYear <= $start ) {
868 # Day is before the start of the year - it is the previous year
874 # Add days since previous year's 1 September
876 if( ( $year %
400 == 0 ) ||
( $year %
100 != 0 && $year %
4 == 0 ) ) {
880 # Start of the new (previous) year
881 $start = self
::hebrewYearStart( $hebrewYear );
884 $nextStart = self
::hebrewYearStart( $hebrewYear +
1 );
887 # Calculate Hebrew day of year
888 $hebrewDayOfYear = $dayOfYear - $start;
890 # Difference between year's days
891 $diff = $nextStart - $start;
892 # Add 12 (or 13 for leap years) days to ignore the difference between
893 # Hebrew and Gregorian year (353 at least vs. 365/6) - now the
894 # difference is only about the year type
895 if( ( $year %
400 == 0 ) ||
( $year %
100 != 0 && $year %
4 == 0 ) ) {
901 # Check the year pattern, and is leap year
902 # 0 means an incomplete year, 1 means a regular year, 2 means a complete year
903 # This is mod 30, to work on both leap years (which add 30 days of Adar I)
905 $yearPattern = $diff %
30;
907 $isLeap = $diff >= 30;
909 # Calculate day in the month from number of day in the Hebrew year
910 # Don't check Adar - if the day is not in Adar, we will stop before;
911 # if it is in Adar, we will use it to check if it is Adar I or Adar II
912 $hebrewDay = $hebrewDayOfYear;
915 while( $hebrewMonth <= 12 ) {
916 # Calculate days in this month
917 if( $isLeap && $hebrewMonth == 6 ) {
918 # Adar in a leap year
920 # Leap year - has Adar I, with 30 days, and Adar II, with 29 days
922 if( $hebrewDay <= $days ) {
926 # Subtract the days of Adar I
930 if( $hebrewDay <= $days ) {
936 } elseif( $hebrewMonth == 2 && $yearPattern == 2 ) {
937 # Cheshvan in a complete year (otherwise as the rule below)
939 } elseif( $hebrewMonth == 3 && $yearPattern == 0 ) {
940 # Kislev in an incomplete year (otherwise as the rule below)
943 # Odd months have 30 days, even have 29
944 $days = 30 - ( $hebrewMonth - 1 ) %
2;
946 if( $hebrewDay <= $days ) {
947 # In the current month
950 # Subtract the days of the current month
952 # Try in the next month
957 return array( $hebrewYear, $hebrewMonth, $hebrewDay, $days );
961 * This calculates the Hebrew year start, as days since 1 September.
962 * Based on Carl Friedrich Gauss algorithm for finding Easter date.
963 * Used for Hebrew date.
965 private static function hebrewYearStart( $year ) {
966 $a = intval( ( 12 * ( $year - 1 ) +
17 ) %
19 );
967 $b = intval( ( $year - 1 ) %
4 );
968 $m = 32.044093161144 +
1.5542417966212 * $a +
$b / 4.0 - 0.0031777940220923 * ( $year - 1 );
978 $c = intval( ( $Mar +
3 * ( $year - 1 ) +
5 * $b +
5 ) %
7);
979 if( $c == 0 && $a > 11 && $m >= 0.89772376543210 ) {
981 } else if( $c == 1 && $a > 6 && $m >= 0.63287037037037 ) {
983 } else if( $c == 2 ||
$c == 4 ||
$c == 6 ) {
987 $Mar +
= intval( ( $year - 3761 ) / 100 ) - intval( ( $year - 3761 ) / 400 ) - 24;
992 * Algorithm to convert Gregorian dates to Thai solar dates.
994 * Link: http://en.wikipedia.org/wiki/Thai_solar_calendar
996 * @param string $ts 14-character timestamp
997 * @return array converted year, month, day
999 private static function tsToThai( $ts ) {
1000 $gy = substr( $ts, 0, 4 );
1001 $gm = substr( $ts, 4, 2 );
1002 $gd = substr( $ts, 6, 2 );
1004 # Add 543 years to the Gregorian calendar
1005 # Months and days are identical
1006 $gy_thai = $gy +
543;
1008 return array( $gy_thai, $gm, $gd );
1013 * Roman number formatting up to 3000
1015 static function romanNumeral( $num ) {
1016 static $table = array(
1017 array( '', 'I', 'II', 'III', 'IV', 'V', 'VI', 'VII', 'VIII', 'IX', 'X' ),
1018 array( '', 'X', 'XX', 'XXX', 'XL', 'L', 'LX', 'LXX', 'LXXX', 'XC', 'C' ),
1019 array( '', 'C', 'CC', 'CCC', 'CD', 'D', 'DC', 'DCC', 'DCCC', 'CM', 'M' ),
1020 array( '', 'M', 'MM', 'MMM' )
1023 $num = intval( $num );
1024 if ( $num > 3000 ||
$num <= 0 ) {
1029 for ( $pow10 = 1000, $i = 3; $i >= 0; $pow10 /= 10, $i-- ) {
1030 if ( $num >= $pow10 ) {
1031 $s .= $table[$i][floor($num / $pow10)];
1033 $num = $num %
$pow10;
1039 * Hebrew Gematria number formatting up to 9999
1041 static function hebrewNumeral( $num ) {
1042 static $table = array(
1043 array( '', 'א', 'ב', 'ג', 'ד', 'ה', 'ו', 'ז', 'ח', 'ט', 'י' ),
1044 array( '', 'י', 'כ', 'ל', 'מ', 'נ', 'ס', 'ע', 'פ', 'צ', 'ק' ),
1045 array( '', 'ק', 'ר', 'ש', 'ת', 'תק', 'תר', 'תש', 'תת', 'תתק', 'תתר' ),
1046 array( '', 'א', 'ב', 'ג', 'ד', 'ה', 'ו', 'ז', 'ח', 'ט', 'י' )
1049 $num = intval( $num );
1050 if ( $num > 9999 ||
$num <= 0 ) {
1055 for ( $pow10 = 1000, $i = 3; $i >= 0; $pow10 /= 10, $i-- ) {
1056 if ( $num >= $pow10 ) {
1057 if ( $num == 15 ||
$num == 16 ) {
1058 $s .= $table[0][9] . $table[0][$num - 9];
1061 $s .= $table[$i][intval( ( $num / $pow10 ) )];
1062 if( $pow10 == 1000 ) {
1067 $num = $num %
$pow10;
1069 if( strlen( $s ) == 2 ) {
1072 $str = substr( $s, 0, strlen( $s ) - 2 ) . '"';
1073 $str .= substr( $s, strlen( $s ) - 2, 2 );
1075 $start = substr( $str, 0, strlen( $str ) - 2 );
1076 $end = substr( $str, strlen( $str ) - 2 );
1079 $str = $start . 'ך';
1082 $str = $start . 'ם';
1085 $str = $start . 'ן';
1088 $str = $start . 'ף';
1091 $str = $start . 'ץ';
1098 * This is meant to be used by time(), date(), and timeanddate() to get
1099 * the date preference they're supposed to use, it should be used in
1103 * function timeanddate([...], $format = true) {
1104 * $datePreference = $this->dateFormat($format);
1109 * @param mixed $usePrefs: if true, the user's preference is used
1110 * if false, the site/language default is used
1111 * if int/string, assumed to be a format.
1114 function dateFormat( $usePrefs = true ) {
1117 if( is_bool( $usePrefs ) ) {
1119 $datePreference = $wgUser->getDatePreference();
1121 $options = User
::getDefaultOptions();
1122 $datePreference = (string)$options['date'];
1125 $datePreference = (string)$usePrefs;
1129 if( $datePreference == '' ) {
1133 return $datePreference;
1138 * @param mixed $ts the time format which needs to be turned into a
1139 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
1140 * @param bool $adj whether to adjust the time output according to the
1141 * user configured offset ($timecorrection)
1142 * @param mixed $format true to use user's date format preference
1143 * @param string $timecorrection the time offset as returned by
1144 * validateTimeZone() in Special:Preferences
1147 function date( $ts, $adj = false, $format = true, $timecorrection = false ) {
1150 $ts = $this->userAdjust( $ts, $timecorrection );
1153 $pref = $this->dateFormat( $format );
1154 if( $pref == 'default' ||
!isset( $this->dateFormats
["$pref date"] ) ) {
1155 $pref = $this->defaultDateFormat
;
1157 return $this->sprintfDate( $this->dateFormats
["$pref date"], $ts );
1162 * @param mixed $ts the time format which needs to be turned into a
1163 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
1164 * @param bool $adj whether to adjust the time output according to the
1165 * user configured offset ($timecorrection)
1166 * @param mixed $format true to use user's date format preference
1167 * @param string $timecorrection the time offset as returned by
1168 * validateTimeZone() in Special:Preferences
1171 function time( $ts, $adj = false, $format = true, $timecorrection = false ) {
1174 $ts = $this->userAdjust( $ts, $timecorrection );
1177 $pref = $this->dateFormat( $format );
1178 if( $pref == 'default' ||
!isset( $this->dateFormats
["$pref time"] ) ) {
1179 $pref = $this->defaultDateFormat
;
1181 return $this->sprintfDate( $this->dateFormats
["$pref time"], $ts );
1186 * @param mixed $ts the time format which needs to be turned into a
1187 * date('YmdHis') format with wfTimestamp(TS_MW,$ts)
1188 * @param bool $adj whether to adjust the time output according to the
1189 * user configured offset ($timecorrection)
1191 * @param mixed $format what format to return, if it's false output the
1192 * default one (default true)
1193 * @param string $timecorrection the time offset as returned by
1194 * validateTimeZone() in Special:Preferences
1197 function timeanddate( $ts, $adj = false, $format = true, $timecorrection = false) {
1200 $ts = wfTimestamp( TS_MW
, $ts );
1203 $ts = $this->userAdjust( $ts, $timecorrection );
1206 $pref = $this->dateFormat( $format );
1207 if( $pref == 'default' ||
!isset( $this->dateFormats
["$pref both"] ) ) {
1208 $pref = $this->defaultDateFormat
;
1211 return $this->sprintfDate( $this->dateFormats
["$pref both"], $ts );
1214 function getMessage( $key ) {
1216 return isset( $this->messages
[$key] ) ?
$this->messages
[$key] : null;
1219 function getAllMessages() {
1221 return $this->messages
;
1224 function iconv( $in, $out, $string ) {
1225 # For most languages, this is a wrapper for iconv
1226 return iconv( $in, $out . '//IGNORE', $string );
1229 // callback functions for uc(), lc(), ucwords(), ucwordbreaks()
1230 function ucwordbreaksCallbackAscii($matches){
1231 return $this->ucfirst($matches[1]);
1234 function ucwordbreaksCallbackMB($matches){
1235 return mb_strtoupper($matches[0]);
1238 function ucCallback($matches){
1239 list( $wikiUpperChars ) = self
::getCaseMaps();
1240 return strtr( $matches[1], $wikiUpperChars );
1243 function lcCallback($matches){
1244 list( , $wikiLowerChars ) = self
::getCaseMaps();
1245 return strtr( $matches[1], $wikiLowerChars );
1248 function ucwordsCallbackMB($matches){
1249 return mb_strtoupper($matches[0]);
1252 function ucwordsCallbackWiki($matches){
1253 list( $wikiUpperChars ) = self
::getCaseMaps();
1254 return strtr( $matches[0], $wikiUpperChars );
1257 function ucfirst( $str ) {
1258 if ( empty($str) ) return $str;
1259 if ( ord($str[0]) < 128 ) return ucfirst($str);
1260 else return self
::uc($str,true); // fall back to more complex logic in case of multibyte strings
1263 function uc( $str, $first = false ) {
1264 if ( function_exists( 'mb_strtoupper' ) ) {
1266 if ( self
::isMultibyte( $str ) ) {
1267 return mb_strtoupper( mb_substr( $str, 0, 1 ) ) . mb_substr( $str, 1 );
1269 return ucfirst( $str );
1272 return self
::isMultibyte( $str ) ?
mb_strtoupper( $str ) : strtoupper( $str );
1275 if ( self
::isMultibyte( $str ) ) {
1276 list( $wikiUpperChars ) = $this->getCaseMaps();
1277 $x = $first ?
'^' : '';
1278 return preg_replace_callback(
1279 "/$x([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/",
1280 array($this,"ucCallback"),
1284 return $first ?
ucfirst( $str ) : strtoupper( $str );
1289 function lcfirst( $str ) {
1290 if ( empty($str) ) return $str;
1291 if ( is_string( $str ) && ord($str[0]) < 128 ) {
1292 // editing string in place = cool
1293 $str[0]=strtolower($str[0]);
1296 else return self
::lc( $str, true );
1299 function lc( $str, $first = false ) {
1300 if ( function_exists( 'mb_strtolower' ) )
1302 if ( self
::isMultibyte( $str ) )
1303 return mb_strtolower( mb_substr( $str, 0, 1 ) ) . mb_substr( $str, 1 );
1305 return strtolower( substr( $str, 0, 1 ) ) . substr( $str, 1 );
1307 return self
::isMultibyte( $str ) ?
mb_strtolower( $str ) : strtolower( $str );
1309 if ( self
::isMultibyte( $str ) ) {
1310 list( , $wikiLowerChars ) = self
::getCaseMaps();
1311 $x = $first ?
'^' : '';
1312 return preg_replace_callback(
1313 "/$x([A-Z]|[\\xc0-\\xff][\\x80-\\xbf]*)/",
1314 array($this,"lcCallback"),
1318 return $first ?
strtolower( substr( $str, 0, 1 ) ) . substr( $str, 1 ) : strtolower( $str );
1321 function isMultibyte( $str ) {
1322 return (bool)preg_match( '/[\x80-\xff]/', $str );
1325 function ucwords($str) {
1326 if ( self
::isMultibyte( $str ) ) {
1327 $str = self
::lc($str);
1329 // regexp to find first letter in each word (i.e. after each space)
1330 $replaceRegexp = "/^([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)| ([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/";
1332 // function to use to capitalize a single char
1333 if ( function_exists( 'mb_strtoupper' ) )
1334 return preg_replace_callback(
1336 array($this,"ucwordsCallbackMB"),
1340 return preg_replace_callback(
1342 array($this,"ucwordsCallbackWiki"),
1347 return ucwords( strtolower( $str ) );
1350 # capitalize words at word breaks
1351 function ucwordbreaks($str){
1352 if (self
::isMultibyte( $str ) ) {
1353 $str = self
::lc($str);
1355 // since \b doesn't work for UTF-8, we explicitely define word break chars
1356 $breaks= "[ \-\(\)\}\{\.,\?!]";
1358 // find first letter after word break
1359 $replaceRegexp = "/^([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)|$breaks([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/";
1361 if ( function_exists( 'mb_strtoupper' ) )
1362 return preg_replace_callback(
1364 array($this,"ucwordbreaksCallbackMB"),
1368 return preg_replace_callback(
1370 array($this,"ucwordsCallbackWiki"),
1375 return preg_replace_callback(
1376 '/\b([\w\x80-\xff]+)\b/',
1377 array($this,"ucwordbreaksCallbackAscii"),
1382 * Return a case-folded representation of $s
1384 * This is a representation such that caseFold($s1)==caseFold($s2) if $s1
1385 * and $s2 are the same except for the case of their characters. It is not
1386 * necessary for the value returned to make sense when displayed.
1388 * Do *not* perform any other normalisation in this function. If a caller
1389 * uses this function when it should be using a more general normalisation
1390 * function, then fix the caller.
1392 function caseFold( $s ) {
1393 return $this->uc( $s );
1396 function checkTitleEncoding( $s ) {
1397 if( is_array( $s ) ) {
1398 wfDebugDieBacktrace( 'Given array to checkTitleEncoding.' );
1400 # Check for non-UTF-8 URLs
1401 $ishigh = preg_match( '/[\x80-\xff]/', $s);
1402 if(!$ishigh) return $s;
1404 $isutf8 = preg_match( '/^([\x00-\x7f]|[\xc0-\xdf][\x80-\xbf]|' .
1405 '[\xe0-\xef][\x80-\xbf]{2}|[\xf0-\xf7][\x80-\xbf]{3})+$/', $s );
1406 if( $isutf8 ) return $s;
1408 return $this->iconv( $this->fallback8bitEncoding(), "utf-8", $s );
1411 function fallback8bitEncoding() {
1413 return $this->fallback8bitEncoding
;
1417 * Some languages have special punctuation to strip out
1418 * or characters which need to be converted for MySQL's
1419 * indexing to grok it correctly. Make such changes here.
1424 function stripForSearch( $string ) {
1426 if ( $wgDBtype != 'mysql' ) {
1430 # MySQL fulltext index doesn't grok utf-8, so we
1431 # need to fold cases and convert to hex
1433 wfProfileIn( __METHOD__
);
1434 if( function_exists( 'mb_strtolower' ) ) {
1435 $out = preg_replace(
1436 "/([\\xc0-\\xff][\\x80-\\xbf]*)/e",
1437 "'U8' . bin2hex( \"$1\" )",
1438 mb_strtolower( $string ) );
1440 list( , $wikiLowerChars ) = self
::getCaseMaps();
1441 $out = preg_replace(
1442 "/([\\xc0-\\xff][\\x80-\\xbf]*)/e",
1443 "'U8' . bin2hex( strtr( \"\$1\", \$wikiLowerChars ) )",
1446 wfProfileOut( __METHOD__
);
1450 function convertForSearchResult( $termsArray ) {
1451 # some languages, e.g. Chinese, need to do a conversion
1452 # in order for search results to be displayed correctly
1457 * Get the first character of a string.
1462 function firstChar( $s ) {
1464 preg_match( '/^([\x00-\x7f]|[\xc0-\xdf][\x80-\xbf]|' .
1465 '[\xe0-\xef][\x80-\xbf]{2}|[\xf0-\xf7][\x80-\xbf]{3})/', $s, $matches);
1467 return isset( $matches[1] ) ?
$matches[1] : "";
1470 function initEncoding() {
1471 # Some languages may have an alternate char encoding option
1472 # (Esperanto X-coding, Japanese furigana conversion, etc)
1473 # If this language is used as the primary content language,
1474 # an override to the defaults can be set here on startup.
1477 function recodeForEdit( $s ) {
1478 # For some languages we'll want to explicitly specify
1479 # which characters make it into the edit box raw
1480 # or are converted in some way or another.
1481 # Note that if wgOutputEncoding is different from
1482 # wgInputEncoding, this text will be further converted
1483 # to wgOutputEncoding.
1484 global $wgEditEncoding;
1485 if( $wgEditEncoding == '' or
1486 $wgEditEncoding == 'UTF-8' ) {
1489 return $this->iconv( 'UTF-8', $wgEditEncoding, $s );
1493 function recodeInput( $s ) {
1494 # Take the previous into account.
1495 global $wgEditEncoding;
1496 if($wgEditEncoding != "") {
1497 $enc = $wgEditEncoding;
1501 if( $enc == 'UTF-8' ) {
1504 return $this->iconv( $enc, 'UTF-8', $s );
1509 * For right-to-left language support
1519 * A hidden direction mark (LRM or RLM), depending on the language direction
1523 function getDirMark() {
1524 return $this->isRTL() ?
"\xE2\x80\x8F" : "\xE2\x80\x8E";
1528 * An arrow, depending on the language direction
1532 function getArrow() {
1533 return $this->isRTL() ?
'←' : '→';
1537 * To allow "foo[[bar]]" to extend the link over the whole word "foobar"
1541 function linkPrefixExtension() {
1543 return $this->linkPrefixExtension
;
1546 function &getMagicWords() {
1548 return $this->magicWords
;
1551 # Fill a MagicWord object with data from here
1552 function getMagic( &$mw ) {
1553 if ( !$this->mMagicHookDone
) {
1554 $this->mMagicHookDone
= true;
1555 wfRunHooks( 'LanguageGetMagic', array( &$this->mMagicExtensions
, $this->getCode() ) );
1557 if ( isset( $this->mMagicExtensions
[$mw->mId
] ) ) {
1558 $rawEntry = $this->mMagicExtensions
[$mw->mId
];
1560 $magicWords =& $this->getMagicWords();
1561 if ( isset( $magicWords[$mw->mId
] ) ) {
1562 $rawEntry = $magicWords[$mw->mId
];
1564 # Fall back to English if local list is incomplete
1565 $magicWords =& Language
::getMagicWords();
1566 $rawEntry = $magicWords[$mw->mId
];
1570 if( !is_array( $rawEntry ) ) {
1571 error_log( "\"$rawEntry\" is not a valid magic thingie for \"$mw->mId\"" );
1573 $mw->mCaseSensitive
= $rawEntry[0];
1574 $mw->mSynonyms
= array_slice( $rawEntry, 1 );
1579 * Add magic words to the extension array
1581 function addMagicWordsByLang( $newWords ) {
1582 $code = $this->getCode();
1583 $fallbackChain = array();
1584 while ( $code && !in_array( $code, $fallbackChain ) ) {
1585 $fallbackChain[] = $code;
1586 $code = self
::getFallbackFor( $code );
1588 if ( !in_array( 'en', $fallbackChain ) ) {
1589 $fallbackChain[] = 'en';
1591 $fallbackChain = array_reverse( $fallbackChain );
1592 foreach ( $fallbackChain as $code ) {
1593 if ( isset( $newWords[$code] ) ) {
1594 $this->mMagicExtensions
= $newWords[$code] +
$this->mMagicExtensions
;
1600 * Get special page names, as an associative array
1601 * case folded alias => real name
1603 function getSpecialPageAliases() {
1605 if ( !isset( $this->mExtendedSpecialPageAliases
) ) {
1606 $this->mExtendedSpecialPageAliases
= $this->specialPageAliases
;
1607 wfRunHooks( 'LanguageGetSpecialPageAliases',
1608 array( &$this->mExtendedSpecialPageAliases
, $this->getCode() ) );
1610 return $this->mExtendedSpecialPageAliases
;
1614 * Italic is unsuitable for some languages
1618 * @param string $text The text to be emphasized.
1621 function emphasize( $text ) {
1622 return "<em>$text</em>";
1626 * Normally we output all numbers in plain en_US style, that is
1627 * 293,291.235 for twohundredninetythreethousand-twohundredninetyone
1628 * point twohundredthirtyfive. However this is not sutable for all
1629 * languages, some such as Pakaran want ੨੯੩,੨੯੫.੨੩੫ and others such as
1630 * Icelandic just want to use commas instead of dots, and dots instead
1631 * of commas like "293.291,235".
1633 * An example of this function being called:
1635 * wfMsg( 'message', $wgLang->formatNum( $num ) )
1638 * See LanguageGu.php for the Gujarati implementation and
1639 * LanguageIs.php for the , => . and . => , implementation.
1641 * @todo check if it's viable to use localeconv() for the decimal
1644 * @param mixed $number the string to be formatted, should be an integer or
1645 * a floating point number.
1646 * @param bool $nocommafy Set to true for special numbers like dates
1649 function formatNum( $number, $nocommafy = false ) {
1650 global $wgTranslateNumerals;
1652 $number = $this->commafy($number);
1653 $s = $this->separatorTransformTable();
1654 if (!is_null($s)) { $number = strtr($number, $s); }
1657 if ($wgTranslateNumerals) {
1658 $s = $this->digitTransformTable();
1659 if (!is_null($s)) { $number = strtr($number, $s); }
1665 function parseFormattedNumber( $number ) {
1666 $s = $this->digitTransformTable();
1667 if (!is_null($s)) { $number = strtr($number, array_flip($s)); }
1669 $s = $this->separatorTransformTable();
1670 if (!is_null($s)) { $number = strtr($number, array_flip($s)); }
1672 $number = strtr( $number, array (',' => '') );
1677 * Adds commas to a given number
1682 function commafy($_) {
1683 return strrev((string)preg_replace('/(\d{3})(?=\d)(?!\d*\.)/','$1,',strrev($_)));
1686 function digitTransformTable() {
1688 return $this->digitTransformTable
;
1691 function separatorTransformTable() {
1693 return $this->separatorTransformTable
;
1698 * For the credit list in includes/Credits.php (action=credits)
1703 function listToText( $l ) {
1706 for ($i = $m; $i >= 0; $i--) {
1709 } else if ($i == $m - 1) {
1710 $s = $l[$i] . ' ' . $this->getMessageFromDB( 'and' ) . ' ' . $s;
1712 $s = $l[$i] . ', ' . $s;
1719 * Truncate a string to a specified length in bytes, appending an optional
1720 * string (e.g. for ellipses)
1722 * The database offers limited byte lengths for some columns in the database;
1723 * multi-byte character sets mean we need to ensure that only whole characters
1724 * are included, otherwise broken characters can be passed to the user
1726 * If $length is negative, the string will be truncated from the beginning
1728 * @param string $string String to truncate
1729 * @param int $length Maximum length (excluding ellipses)
1730 * @param string $ellipses String to append to the truncated text
1733 function truncate( $string, $length, $ellipsis = "" ) {
1734 if( $length == 0 ) {
1737 if ( strlen( $string ) <= abs( $length ) ) {
1741 $string = substr( $string, 0, $length );
1742 $char = ord( $string[strlen( $string ) - 1] );
1744 if ($char >= 0xc0) {
1745 # We got the first byte only of a multibyte char; remove it.
1746 $string = substr( $string, 0, -1 );
1747 } elseif( $char >= 0x80 &&
1748 preg_match( '/^(.*)(?:[\xe0-\xef][\x80-\xbf]|' .
1749 '[\xf0-\xf7][\x80-\xbf]{1,2})$/', $string, $m ) ) {
1750 # We chopped in the middle of a character; remove it
1753 return $string . $ellipsis;
1755 $string = substr( $string, $length );
1756 $char = ord( $string[0] );
1757 if( $char >= 0x80 && $char < 0xc0 ) {
1758 # We chopped in the middle of a character; remove the whole thing
1759 $string = preg_replace( '/^[\x80-\xbf]+/', '', $string );
1761 return $ellipsis . $string;
1766 * Grammatical transformations, needed for inflected languages
1767 * Invoked by putting {{grammar:case|word}} in a message
1769 * @param string $word
1770 * @param string $case
1773 function convertGrammar( $word, $case ) {
1774 global $wgGrammarForms;
1775 if ( isset($wgGrammarForms[$this->getCode()][$case][$word]) ) {
1776 return $wgGrammarForms[$this->getCode()][$case][$word];
1782 * Plural form transformations, needed for some languages.
1783 * For example, there are 3 form of plural in Russian and Polish,
1784 * depending on "count mod 10". See [[w:Plural]]
1785 * For English it is pretty simple.
1787 * Invoked by putting {{plural:count|wordform1|wordform2}}
1788 * or {{plural:count|wordform1|wordform2|wordform3}}
1790 * Example: {{plural:{{NUMBEROFARTICLES}}|article|articles}}
1792 * @param integer $count Non-localized number
1793 * @param array $forms Different plural forms
1794 * @return string Correct form of plural for $count in this language
1796 function convertPlural( $count, $forms ) {
1797 if ( !count($forms) ) { return ''; }
1798 $forms = $this->preConvertPlural( $forms, 2 );
1800 return ( abs($count) == 1 ) ?
$forms[0] : $forms[1];
1804 * Checks that convertPlural was given an array and pads it to requested
1805 * amound of forms by copying the last one.
1807 * @param integer $count How many forms should there be at least
1808 * @param array $forms Array of forms given to convertPlural
1809 * @return array Padded array of forms or an exception if not an array
1811 protected function preConvertPlural( /* Array */ $forms, $count ) {
1812 while ( count($forms) < $count ) {
1813 $forms[] = $forms[count($forms)-1];
1819 * For translaing of expiry times
1820 * @param string The validated block time in English
1821 * @return Somehow translated block time
1822 * @see LanguageFi.php for example implementation
1824 function translateBlockExpiry( $str ) {
1826 $scBlockExpiryOptions = $this->getMessageFromDB( 'ipboptions' );
1828 if ( $scBlockExpiryOptions == '-') {
1832 foreach (explode(',', $scBlockExpiryOptions) as $option) {
1833 if ( strpos($option, ":") === false )
1835 list($show, $value) = explode(":", $option);
1836 if ( strcmp ( $str, $value) == 0 ) {
1837 return htmlspecialchars( trim( $show ) );
1845 * languages like Chinese need to be segmented in order for the diff
1848 * @param string $text
1851 function segmentForDiff( $text ) {
1856 * and unsegment to show the result
1858 * @param string $text
1861 function unsegmentForDiff( $text ) {
1865 # convert text to different variants of a language.
1866 function convert( $text, $isTitle = false) {
1867 return $this->mConverter
->convert($text, $isTitle);
1870 # Convert text from within Parser
1871 function parserConvert( $text, &$parser ) {
1872 return $this->mConverter
->parserConvert( $text, $parser );
1875 # Check if this is a language with variants
1876 function hasVariants(){
1877 return sizeof($this->getVariants())>1;
1880 # Put custom tags (e.g. -{ }-) around math to prevent conversion
1881 function armourMath($text){
1882 return $this->mConverter
->armourMath($text);
1887 * Perform output conversion on a string, and encode for safe HTML output.
1888 * @param string $text
1889 * @param bool $isTitle -- wtf?
1891 * @todo this should get integrated somewhere sane
1893 function convertHtml( $text, $isTitle = false ) {
1894 return htmlspecialchars( $this->convert( $text, $isTitle ) );
1897 function convertCategoryKey( $key ) {
1898 return $this->mConverter
->convertCategoryKey( $key );
1902 * get the list of variants supported by this langauge
1903 * see sample implementation in LanguageZh.php
1905 * @return array an array of language codes
1907 function getVariants() {
1908 return $this->mConverter
->getVariants();
1912 function getPreferredVariant( $fromUser = true ) {
1913 return $this->mConverter
->getPreferredVariant( $fromUser );
1917 * if a language supports multiple variants, it is
1918 * possible that non-existing link in one variant
1919 * actually exists in another variant. this function
1920 * tries to find it. See e.g. LanguageZh.php
1922 * @param string $link the name of the link
1923 * @param mixed $nt the title object of the link
1924 * @return null the input parameters may be modified upon return
1926 function findVariantLink( &$link, &$nt ) {
1927 $this->mConverter
->findVariantLink($link, $nt);
1931 * If a language supports multiple variants, converts text
1932 * into an array of all possible variants of the text:
1933 * 'variant' => text in that variant
1936 function convertLinkToAllVariants($text){
1937 return $this->mConverter
->convertLinkToAllVariants($text);
1942 * returns language specific options used by User::getPageRenderHash()
1943 * for example, the preferred language variant
1948 function getExtraHashOptions() {
1949 return $this->mConverter
->getExtraHashOptions();
1953 * for languages that support multiple variants, the title of an
1954 * article may be displayed differently in different variants. this
1955 * function returns the apporiate title defined in the body of the article.
1959 function getParsedTitle() {
1960 return $this->mConverter
->getParsedTitle();
1964 * Enclose a string with the "no conversion" tag. This is used by
1965 * various functions in the Parser
1967 * @param string $text text to be tagged for no conversion
1968 * @return string the tagged text
1970 function markNoConversion( $text, $noParse=false ) {
1971 return $this->mConverter
->markNoConversion( $text, $noParse );
1975 * A regular expression to match legal word-trailing characters
1976 * which should be merged onto a link of the form [[foo]]bar.
1981 function linkTrail() {
1983 return $this->linkTrail
;
1986 function getLangObj() {
1991 * Get the RFC 3066 code for this language object
1993 function getCode() {
1994 return $this->mCode
;
1997 function setCode( $code ) {
1998 $this->mCode
= $code;
2001 static function getFileName( $prefix = 'Language', $code, $suffix = '.php' ) {
2002 return $prefix . str_replace( '-', '_', ucfirst( $code ) ) . $suffix;
2005 static function getMessagesFileName( $code ) {
2007 return self
::getFileName( "$IP/languages/messages/Messages", $code, '.php' );
2010 static function getClassFileName( $code ) {
2012 return self
::getFileName( "$IP/languages/classes/Language", $code, '.php' );
2015 static function getLocalisationArray( $code, $disableCache = false ) {
2016 self
::loadLocalisation( $code, $disableCache );
2017 return self
::$mLocalisationCache[$code];
2021 * Load localisation data for a given code into the static cache
2023 * @return array Dependencies, map of filenames to mtimes
2025 static function loadLocalisation( $code, $disableCache = false ) {
2026 static $recursionGuard = array();
2027 global $wgMemc, $wgCheckSerialized;
2030 throw new MWException( "Invalid language code requested" );
2033 if ( !$disableCache ) {
2034 # Try the per-process cache
2035 if ( isset( self
::$mLocalisationCache[$code] ) ) {
2036 return self
::$mLocalisationCache[$code]['deps'];
2039 wfProfileIn( __METHOD__
);
2041 # Try the serialized directory
2042 $cache = wfGetPrecompiledData( self
::getFileName( "Messages", $code, '.ser' ) );
2044 if ( $wgCheckSerialized && self
::isLocalisationOutOfDate( $cache ) ) {
2046 wfDebug( "Language::loadLocalisation(): precompiled data file for $code is out of date\n" );
2048 self
::$mLocalisationCache[$code] = $cache;
2049 wfDebug( "Language::loadLocalisation(): got localisation for $code from precompiled data file\n" );
2050 wfProfileOut( __METHOD__
);
2051 return self
::$mLocalisationCache[$code]['deps'];
2055 # Try the global cache
2056 $memcKey = wfMemcKey('localisation', $code );
2057 $fbMemcKey = wfMemcKey('fallback', $cache['fallback'] );
2058 $cache = $wgMemc->get( $memcKey );
2060 if ( self
::isLocalisationOutOfDate( $cache ) ) {
2061 $wgMemc->delete( $memcKey );
2062 $wgMemc->delete( $fbMemcKey );
2064 wfDebug( "Language::loadLocalisation(): localisation cache for $code had expired\n" );
2066 self
::$mLocalisationCache[$code] = $cache;
2067 wfDebug( "Language::loadLocalisation(): got localisation for $code from cache\n" );
2068 wfProfileOut( __METHOD__
);
2069 return $cache['deps'];
2073 wfProfileIn( __METHOD__
);
2076 # Default fallback, may be overridden when the messages file is included
2077 if ( $code != 'en' ) {
2083 # Load the primary localisation from the source file
2084 $filename = self
::getMessagesFileName( $code );
2085 if ( !file_exists( $filename ) ) {
2086 wfDebug( "Language::loadLocalisation(): no localisation file for $code, using implicit fallback to en\n" );
2090 $deps = array( $filename => filemtime( $filename ) );
2091 require( $filename );
2092 $cache = compact( self
::$mLocalisationKeys );
2093 wfDebug( "Language::loadLocalisation(): got localisation for $code from source\n" );
2096 if ( !empty( $fallback ) ) {
2097 # Load the fallback localisation, with a circular reference guard
2098 if ( isset( $recursionGuard[$code] ) ) {
2099 throw new MWException( "Error: Circular fallback reference in language code $code" );
2101 $recursionGuard[$code] = true;
2102 $newDeps = self
::loadLocalisation( $fallback, $disableCache );
2103 unset( $recursionGuard[$code] );
2105 $secondary = self
::$mLocalisationCache[$fallback];
2106 $deps = array_merge( $deps, $newDeps );
2108 # Merge the fallback localisation with the current localisation
2109 foreach ( self
::$mLocalisationKeys as $key ) {
2110 if ( isset( $cache[$key] ) ) {
2111 if ( isset( $secondary[$key] ) ) {
2112 if ( in_array( $key, self
::$mMergeableMapKeys ) ) {
2113 $cache[$key] = $cache[$key] +
$secondary[$key];
2114 } elseif ( in_array( $key, self
::$mMergeableListKeys ) ) {
2115 $cache[$key] = array_merge( $secondary[$key], $cache[$key] );
2116 } elseif ( in_array( $key, self
::$mMergeableAliasListKeys ) ) {
2117 $cache[$key] = array_merge_recursive( $cache[$key], $secondary[$key] );
2121 $cache[$key] = $secondary[$key];
2125 # Merge bookstore lists if requested
2126 if ( !empty( $cache['bookstoreList']['inherit'] ) ) {
2127 $cache['bookstoreList'] = array_merge( $cache['bookstoreList'], $secondary['bookstoreList'] );
2129 if ( isset( $cache['bookstoreList']['inherit'] ) ) {
2130 unset( $cache['bookstoreList']['inherit'] );
2134 # Add dependencies to the cache entry
2135 $cache['deps'] = $deps;
2137 # Replace spaces with underscores in namespace names
2138 $cache['namespaceNames'] = str_replace( ' ', '_', $cache['namespaceNames'] );
2140 # And do the same for specialpage aliases. $page is an array.
2141 foreach ( $cache['specialPageAliases'] as &$page ) {
2142 $page = str_replace( ' ', '_', $page );
2144 # Decouple the reference to prevent accidental damage
2147 # Save to both caches
2148 self
::$mLocalisationCache[$code] = $cache;
2149 if ( !$disableCache ) {
2150 $wgMemc->set( $memcKey, $cache );
2151 $wgMemc->set( $fbMemcKey, (string) $cache['fallback'] );
2154 wfProfileOut( __METHOD__
);
2159 * Test if a given localisation cache is out of date with respect to the
2160 * source Messages files. This is done automatically for the global cache
2161 * in $wgMemc, but is only done on certain occasions for the serialized
2164 * @param $cache mixed Either a language code or a cache array
2166 static function isLocalisationOutOfDate( $cache ) {
2167 if ( !is_array( $cache ) ) {
2168 self
::loadLocalisation( $cache );
2169 $cache = self
::$mLocalisationCache[$cache];
2172 foreach ( $cache['deps'] as $file => $mtime ) {
2173 if ( !file_exists( $file ) ||
filemtime( $file ) > $mtime ) {
2182 * Get the fallback for a given language
2184 static function getFallbackFor( $code ) {
2185 if ( $code === 'en' ) return false;
2188 $memcKey = wfMemcKey( 'fallback', $code );
2189 $fbcode = $wgMemc->get( $memcKey );
2191 if ( $fbcode !== false ) {
2192 wfDebug( __METHOD__
. ": got fallback for $code from memc: '$fbcode'\n" );
2193 if ( $fbcode === '' ) $fbcode = false;
2197 self
::loadLocalisation( $code );
2198 $fbcode = self
::$mLocalisationCache[$code]['fallback'];
2199 $wgMemc->set( $memcKey, (string) $fbcode );
2204 * Get all messages for a given language
2206 static function getMessagesFor( $code ) {
2207 self
::loadLocalisation( $code );
2208 return self
::$mLocalisationCache[$code]['messages'];
2212 * Get a message for a given language
2214 static function getMessageFor( $key, $code ) {
2215 self
::loadLocalisation( $code );
2216 return isset( self
::$mLocalisationCache[$code]['messages'][$key] ) ? self
::$mLocalisationCache[$code]['messages'][$key] : null;
2220 * Load localisation data for this object
2223 if ( !$this->mLoaded
) {
2224 self
::loadLocalisation( $this->getCode() );
2225 $cache =& self
::$mLocalisationCache[$this->getCode()];
2226 foreach ( self
::$mLocalisationKeys as $key ) {
2227 $this->$key = $cache[$key];
2229 $this->mLoaded
= true;
2231 $this->fixUpSettings();
2236 * Do any necessary post-cache-load settings adjustment
2238 function fixUpSettings() {
2239 global $wgExtraNamespaces, $wgMetaNamespace, $wgMetaNamespaceTalk,
2240 $wgNamespaceAliases, $wgAmericanDates;
2241 wfProfileIn( __METHOD__
);
2242 if ( $wgExtraNamespaces ) {
2243 $this->namespaceNames
= $wgExtraNamespaces +
$this->namespaceNames
;
2246 $this->namespaceNames
[NS_PROJECT
] = $wgMetaNamespace;
2247 if ( $wgMetaNamespaceTalk ) {
2248 $this->namespaceNames
[NS_PROJECT_TALK
] = $wgMetaNamespaceTalk;
2250 $talk = $this->namespaceNames
[NS_PROJECT_TALK
];
2251 $talk = str_replace( '$1', $wgMetaNamespace, $talk );
2253 # Allow grammar transformations
2254 # Allowing full message-style parsing would make simple requests
2255 # such as action=raw much more expensive than they need to be.
2256 # This will hopefully cover most cases.
2257 $talk = preg_replace_callback( '/{{grammar:(.*?)\|(.*?)}}/i',
2258 array( &$this, 'replaceGrammarInNamespace' ), $talk );
2259 $talk = str_replace( ' ', '_', $talk );
2260 $this->namespaceNames
[NS_PROJECT_TALK
] = $talk;
2263 # The above mixing may leave namespaces out of canonical order.
2264 # Re-order by namespace ID number...
2265 ksort( $this->namespaceNames
);
2267 # Put namespace names and aliases into a hashtable.
2268 # If this is too slow, then we should arrange it so that it is done
2269 # before caching. The catch is that at pre-cache time, the above
2270 # class-specific fixup hasn't been done.
2271 $this->mNamespaceIds
= array();
2272 foreach ( $this->namespaceNames
as $index => $name ) {
2273 $this->mNamespaceIds
[$this->lc($name)] = $index;
2275 if ( $this->namespaceAliases
) {
2276 foreach ( $this->namespaceAliases
as $name => $index ) {
2277 $this->mNamespaceIds
[$this->lc($name)] = $index;
2280 if ( $wgNamespaceAliases ) {
2281 foreach ( $wgNamespaceAliases as $name => $index ) {
2282 $this->mNamespaceIds
[$this->lc($name)] = $index;
2286 if ( $this->defaultDateFormat
== 'dmy or mdy' ) {
2287 $this->defaultDateFormat
= $wgAmericanDates ?
'mdy' : 'dmy';
2289 wfProfileOut( __METHOD__
);
2292 function replaceGrammarInNamespace( $m ) {
2293 return $this->convertGrammar( trim( $m[2] ), trim( $m[1] ) );
2296 static function getCaseMaps() {
2297 static $wikiUpperChars, $wikiLowerChars;
2298 if ( isset( $wikiUpperChars ) ) {
2299 return array( $wikiUpperChars, $wikiLowerChars );
2302 wfProfileIn( __METHOD__
);
2303 $arr = wfGetPrecompiledData( 'Utf8Case.ser' );
2304 if ( $arr === false ) {
2305 throw new MWException(
2306 "Utf8Case.ser is missing, please run \"make\" in the serialized directory\n" );
2309 wfProfileOut( __METHOD__
);
2310 return array( $wikiUpperChars, $wikiLowerChars );
2313 function formatTimePeriod( $seconds ) {
2314 if ( $seconds < 10 ) {
2315 return $this->formatNum( sprintf( "%.1f", $seconds ) ) . wfMsg( 'seconds-abbrev' );
2316 } elseif ( $seconds < 60 ) {
2317 return $this->formatNum( round( $seconds ) ) . wfMsg( 'seconds-abbrev' );
2318 } elseif ( $seconds < 3600 ) {
2319 return $this->formatNum( floor( $seconds / 60 ) ) . wfMsg( 'minutes-abbrev' ) .
2320 $this->formatNum( round( fmod( $seconds, 60 ) ) ) . wfMsg( 'seconds-abbrev' );
2322 $hours = floor( $seconds / 3600 );
2323 $minutes = floor( ( $seconds - $hours * 3600 ) / 60 );
2324 $secondsPart = round( $seconds - $hours * 3600 - $minutes * 60 );
2325 return $this->formatNum( $hours ) . wfMsg( 'hours-abbrev' ) .
2326 $this->formatNum( $minutes ) . wfMsg( 'minutes-abbrev' ) .
2327 $this->formatNum( $secondsPart ) . wfMsg( 'seconds-abbrev' );
2331 function formatBitrate( $bps ) {
2332 $units = array( 'bps', 'kbps', 'Mbps', 'Gbps' );
2334 return $this->formatNum( $bps ) . $units[0];
2336 $unitIndex = floor( log10( $bps ) / 3 );
2337 $mantissa = $bps / pow( 1000, $unitIndex );
2338 if ( $mantissa < 10 ) {
2339 $mantissa = round( $mantissa, 1 );
2341 $mantissa = round( $mantissa );
2343 return $this->formatNum( $mantissa ) . $units[$unitIndex];
2347 * Format a size in bytes for output, using an appropriate
2348 * unit (B, KB, MB or GB) according to the magnitude in question
2350 * @param $size Size to format
2351 * @return string Plain text (not HTML)
2353 function formatSize( $size ) {
2354 // For small sizes no decimal places necessary
2356 if( $size > 1024 ) {
2357 $size = $size / 1024;
2358 if( $size > 1024 ) {
2359 $size = $size / 1024;
2360 // For MB and bigger two decimal places are smarter
2362 if( $size > 1024 ) {
2363 $size = $size / 1024;
2364 $msg = 'size-gigabytes';
2366 $msg = 'size-megabytes';
2369 $msg = 'size-kilobytes';
2372 $msg = 'size-bytes';
2374 $size = round( $size, $round );
2375 $text = $this->getMessageFromDB( $msg );
2376 return str_replace( '$1', $this->formatNum( $size ), $text );