diff options
Diffstat (limited to 'languages/Language.php')
-rw-r--r-- | languages/Language.php | 1031 |
1 files changed, 687 insertions, 344 deletions
diff --git a/languages/Language.php b/languages/Language.php index 3416fb27..fee5aec3 100644 --- a/languages/Language.php +++ b/languages/Language.php @@ -1,30 +1,34 @@ <?php /** - * @defgroup Language Language + * Internationalisation code * * @file * @ingroup Language */ -if( !defined( 'MEDIAWIKI' ) ) { +/** + * @defgroup Language Language + */ + +if ( !defined( 'MEDIAWIKI' ) ) { echo "This file is part of MediaWiki, it is not a valid entry point.\n"; exit( 1 ); } # Read language names global $wgLanguageNames; -require_once( dirname(__FILE__) . '/Names.php' ) ; +require_once( dirname( __FILE__ ) . '/Names.php' ); global $wgInputEncoding, $wgOutputEncoding; /** * These are always UTF-8, they exist only for backwards compatibility */ -$wgInputEncoding = "UTF-8"; -$wgOutputEncoding = "UTF-8"; +$wgInputEncoding = 'UTF-8'; +$wgOutputEncoding = 'UTF-8'; -if( function_exists( 'mb_strtoupper' ) ) { - mb_internal_encoding('UTF-8'); +if ( function_exists( 'mb_strtoupper' ) ) { + mb_internal_encoding( 'UTF-8' ); } /** @@ -34,20 +38,22 @@ if( function_exists( 'mb_strtoupper' ) ) { */ class FakeConverter { var $mLang; - function FakeConverter( $langobj ) { $this->mLang = $langobj; } - function autoConvertToAllVariants( $text ) { return $text; } + function __construct( $langobj ) { $this->mLang = $langobj; } + function autoConvertToAllVariants( $text ) { return array( $this->mLang->getCode() => $text ); } function convert( $t ) { return $t; } function convertTitle( $t ) { return $t->getPrefixedText(); } function getVariants() { return array( $this->mLang->getCode() ); } function getPreferredVariant() { return $this->mLang->getCode(); } + function getDefaultVariant() { return $this->mLang->getCode(); } + function getURLVariant() { return ''; } function getConvRuleTitle() { return false; } - function findVariantLink(&$l, &$n, $ignoreOtherCond = false) {} - function getExtraHashOptions() {return '';} - function getParsedTitle() {return '';} - function markNoConversion($text, $noParse=false) {return $text;} - function convertCategoryKey( $key ) {return $key; } - function convertLinkToAllVariants($text){ return array( $this->mLang->getCode() => $text); } - function armourMath($text){ return $text; } + function findVariantLink( &$l, &$n, $ignoreOtherCond = false ) { } + function getExtraHashOptions() { return ''; } + function getParsedTitle() { return ''; } + function markNoConversion( $text, $noParse = false ) { return $text; } + function convertCategoryKey( $key ) { return $key; } + function convertLinkToAllVariants( $text ) { return $this->autoConvertToAllVariants( $text ); } + function armourMath( $text ) { return $text; } } /** @@ -129,7 +135,7 @@ class Language { */ static function factory( $code ) { if ( !isset( self::$mLangObjCache[$code] ) ) { - if( count( self::$mLangObjCache ) > 10 ) { + if ( count( self::$mLangObjCache ) > 10 ) { // Don't keep a billion objects around, that's stupid. self::$mLangObjCache = array(); } @@ -158,10 +164,10 @@ class Language { $class = 'Language' . str_replace( '-', '_', ucfirst( $code ) ); // Preload base classes to work around APC/PHP5 bug if ( file_exists( "$IP/languages/classes/$class.deps.php" ) ) { - include_once("$IP/languages/classes/$class.deps.php"); + include_once( "$IP/languages/classes/$class.deps.php" ); } if ( file_exists( "$IP/languages/classes/$class.php" ) ) { - include_once("$IP/languages/classes/$class.php"); + include_once( "$IP/languages/classes/$class.php" ); } } @@ -169,7 +175,7 @@ class Language { throw new MWException( "Language fallback loop detected when creating class $class\n" ); } - if( ! class_exists( $class ) ) { + if ( !class_exists( $class ) ) { $fallback = Language::getFallbackFor( $code ); ++$recursionLevel; $lang = Language::newFromCode( $fallback ); @@ -202,7 +208,7 @@ class Language { } function __construct() { - $this->mConverter = new FakeConverter($this); + $this->mConverter = new FakeConverter( $this ); // Set the code to the name of the descendant if ( get_class( $this ) == 'Language' ) { $this->mCode = 'en'; @@ -225,7 +231,7 @@ class Language { * Hook which will be called if this is the content language. * Descendants can use this to register hook functions or modify globals */ - function initContLang() {} + function initContLang() { } /** * @deprecated Use User::getDefaultOptions() @@ -257,12 +263,12 @@ class Language { */ function getNamespaces() { if ( is_null( $this->namespaceNames ) ) { - global $wgExtraNamespaces, $wgMetaNamespace, $wgMetaNamespaceTalk; + global $wgMetaNamespace, $wgMetaNamespaceTalk, $wgExtraNamespaces; $this->namespaceNames = self::$dataCache->getItem( $this->mCode, 'namespaceNames' ); - if ( $wgExtraNamespaces ) { - $this->namespaceNames = $wgExtraNamespaces + $this->namespaceNames; - } + $validNamespaces = MWNamespace::getCanonicalNamespaces(); + + $this->namespaceNames = $wgExtraNamespaces + $this->namespaceNames + $validNamespaces; $this->namespaceNames[NS_PROJECT] = $wgMetaNamespace; if ( $wgMetaNamespaceTalk ) { @@ -272,6 +278,13 @@ class Language { $this->namespaceNames[NS_PROJECT_TALK] = $this->fixVariableInNamespace( $talk ); } + + # Sometimes a language will be localised but not actually exist on this wiki. + foreach( $this->namespaceNames as $key => $text ) { + if ( !isset( $validNamespaces[$key] ) ) { + unset( $this->namespaceNames[$key] ); + } + } # The above mixing may leave namespaces out of canonical order. # Re-order by namespace ID number... @@ -290,8 +303,8 @@ class Language { */ function getFormattedNamespaces() { $ns = $this->getNamespaces(); - foreach($ns as $k => $v) { - $ns[$k] = strtr($v, '_', ' '); + foreach ( $ns as $k => $v ) { + $ns[$k] = strtr( $v, '_', ' ' ); } return $ns; } @@ -320,7 +333,7 @@ class Language { */ function getFormattedNsText( $index ) { $ns = $this->getNsText( $index ); - return strtr($ns, '_', ' '); + return strtr( $ns, '_', ' ' ); } /** @@ -332,7 +345,7 @@ class Language { * @return mixed An integer if $text is a valid value otherwise false */ function getLocalNsIndex( $text ) { - $lctext = $this->lc($text); + $lctext = $this->lc( $text ); $ids = $this->getNamespaceIds(); return isset( $ids[$lctext] ) ? $ids[$lctext] : false; } @@ -365,14 +378,14 @@ class Language { # class-specific fixup hasn't been done. $this->mNamespaceIds = array(); foreach ( $this->getNamespaces() as $index => $name ) { - $this->mNamespaceIds[$this->lc($name)] = $index; + $this->mNamespaceIds[$this->lc( $name )] = $index; } foreach ( $this->getNamespaceAliases() as $name => $index ) { - $this->mNamespaceIds[$this->lc($name)] = $index; + $this->mNamespaceIds[$this->lc( $name )] = $index; } if ( $wgNamespaceAliases ) { foreach ( $wgNamespaceAliases as $name => $index ) { - $this->mNamespaceIds[$this->lc($name)] = $index; + $this->mNamespaceIds[$this->lc( $name )] = $index; } } } @@ -388,7 +401,7 @@ class Language { * @return mixed An integer if $text is a valid value otherwise false */ function getNsIndex( $text ) { - $lctext = $this->lc($text); + $lctext = $this->lc( $text ); if ( ( $ns = MWNamespace::getCanonicalIndex( $lctext ) ) !== null ) { return $ns; } @@ -480,7 +493,7 @@ class Language { global $IP; $names = array(); $dir = opendir( "$IP/languages/messages" ); - while( false !== ( $file = readdir( $dir ) ) ) { + while ( false !== ( $file = readdir( $dir ) ) ) { $code = self::getCodeFromFileName( $file, 'Messages' ); if ( $code && isset( $allNames[$code] ) ) { $names[$code] = $allNames[$code]; @@ -509,39 +522,39 @@ class Language { } function getMonthName( $key ) { - return $this->getMessageFromDB( self::$mMonthMsgs[$key-1] ); + return $this->getMessageFromDB( self::$mMonthMsgs[$key - 1] ); } function getMonthNameGen( $key ) { - return $this->getMessageFromDB( self::$mMonthGenMsgs[$key-1] ); + return $this->getMessageFromDB( self::$mMonthGenMsgs[$key - 1] ); } function getMonthAbbreviation( $key ) { - return $this->getMessageFromDB( self::$mMonthAbbrevMsgs[$key-1] ); + return $this->getMessageFromDB( self::$mMonthAbbrevMsgs[$key - 1] ); } function getWeekdayName( $key ) { - return $this->getMessageFromDB( self::$mWeekdayMsgs[$key-1] ); + return $this->getMessageFromDB( self::$mWeekdayMsgs[$key - 1] ); } function getWeekdayAbbreviation( $key ) { - return $this->getMessageFromDB( self::$mWeekdayAbbrevMsgs[$key-1] ); + return $this->getMessageFromDB( self::$mWeekdayAbbrevMsgs[$key - 1] ); } function getIranianCalendarMonthName( $key ) { - return $this->getMessageFromDB( self::$mIranianCalendarMonthMsgs[$key-1] ); + return $this->getMessageFromDB( self::$mIranianCalendarMonthMsgs[$key - 1] ); } function getHebrewCalendarMonthName( $key ) { - return $this->getMessageFromDB( self::$mHebrewCalendarMonthMsgs[$key-1] ); + return $this->getMessageFromDB( self::$mHebrewCalendarMonthMsgs[$key - 1] ); } function getHebrewCalendarMonthNameGen( $key ) { - return $this->getMessageFromDB( self::$mHebrewCalendarMonthGenMsgs[$key-1] ); + return $this->getMessageFromDB( self::$mHebrewCalendarMonthGenMsgs[$key - 1] ); } function getHijriCalendarMonthName( $key ) { - return $this->getMessageFromDB( self::$mHijriCalendarMonthMsgs[$key-1] ); + return $this->getMessageFromDB( self::$mHijriCalendarMonthMsgs[$key - 1] ); } /** @@ -574,36 +587,42 @@ class Language { $minDiff = 0; if ( $data[0] == 'System' || $tz == '' ) { - # Global offset in minutes. - if( isset($wgLocalTZoffset) ) $minDiff = $wgLocalTZoffset; + # Global offset in minutes. + if ( isset( $wgLocalTZoffset ) ) { + $minDiff = $wgLocalTZoffset; + } } else if ( $data[0] == 'Offset' ) { $minDiff = intval( $data[1] ); } else { $data = explode( ':', $tz ); - if( count( $data ) == 2 ) { + if ( count( $data ) == 2 ) { $data[0] = intval( $data[0] ); $data[1] = intval( $data[1] ); $minDiff = abs( $data[0] ) * 60 + $data[1]; - if ( $data[0] < 0 ) $minDiff = -$minDiff; + if ( $data[0] < 0 ) { + $minDiff = -$minDiff; + } } else { $minDiff = intval( $data[0] ) * 60; } } # No difference ? Return time unchanged - if ( 0 == $minDiff ) return $ts; + if ( 0 == $minDiff ) { + return $ts; + } wfSuppressWarnings(); // E_STRICT system time bitching # Generate an adjusted date; take advantage of the fact that mktime # will normalize out-of-range values so we don't have to split $minDiff # into hours and minutes. $t = mktime( ( - (int)substr( $ts, 8, 2) ), # Hours + (int)substr( $ts, 8, 2 ) ), # Hours (int)substr( $ts, 10, 2 ) + $minDiff, # Minutes (int)substr( $ts, 12, 2 ), # Seconds (int)substr( $ts, 4, 2 ), # Month (int)substr( $ts, 6, 2 ), # Day - (int)substr( $ts, 0, 4 ) ); #Year + (int)substr( $ts, 0, 4 ) ); # Year $date = date( 'YmdHis', $t ); wfRestoreWarnings(); @@ -617,9 +636,8 @@ class Language { * escaping format. * * Supported format characters are dDjlNwzWFmMntLoYyaAgGhHiscrU. See the - * PHP manual for definitions. "o" format character is supported since - * PHP 5.1.0, previous versions return literal o. - * There are a number of extensions, which start with "x": + * PHP manual for definitions. There are a number of extensions, which + * start with "x": * * xn Do not translate digits of the next numeric format character * xN Toggle raw digit (xn) flag, stays set until explicitly unset @@ -669,7 +687,6 @@ class Language { * @param $ts String: 14-character timestamp * YYYYMMDDHHMMSS * 01234567890123 - * @todo emulation of "o" format character for PHP pre 5.1.0 * @todo handling of "o" format character for Iranian, Hebrew, Hijri & Thai? */ function sprintfDate( $format, $ts ) { @@ -730,51 +747,73 @@ class Language { $num = intval( substr( $ts, 6, 2 ) ); break; case 'xij': - if ( !$iranian ) $iranian = self::tsToIranian( $ts ); + if ( !$iranian ) { + $iranian = self::tsToIranian( $ts ); + } $num = $iranian[2]; break; case 'xmj': - if ( !$hijri ) $hijri = self::tsToHijri( $ts ); + if ( !$hijri ) { + $hijri = self::tsToHijri( $ts ); + } $num = $hijri[2]; break; case 'xjj': - if ( !$hebrew ) $hebrew = self::tsToHebrew( $ts ); + if ( !$hebrew ) { + $hebrew = self::tsToHebrew( $ts ); + } $num = $hebrew[2]; break; case 'l': - if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts ); + if ( !$unix ) { + $unix = wfTimestamp( TS_UNIX, $ts ); + } $s .= $this->getWeekdayName( gmdate( 'w', $unix ) + 1 ); break; case 'N': - if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts ); + if ( !$unix ) { + $unix = wfTimestamp( TS_UNIX, $ts ); + } $w = gmdate( 'w', $unix ); $num = $w ? $w : 7; break; case 'w': - if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts ); + if ( !$unix ) { + $unix = wfTimestamp( TS_UNIX, $ts ); + } $num = gmdate( 'w', $unix ); break; case 'z': - if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts ); + if ( !$unix ) { + $unix = wfTimestamp( TS_UNIX, $ts ); + } $num = gmdate( 'z', $unix ); break; case 'W': - if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts ); + if ( !$unix ) { + $unix = wfTimestamp( TS_UNIX, $ts ); + } $num = gmdate( 'W', $unix ); break; case 'F': $s .= $this->getMonthName( substr( $ts, 4, 2 ) ); break; case 'xiF': - if ( !$iranian ) $iranian = self::tsToIranian( $ts ); + if ( !$iranian ) { + $iranian = self::tsToIranian( $ts ); + } $s .= $this->getIranianCalendarMonthName( $iranian[1] ); break; case 'xmF': - if ( !$hijri ) $hijri = self::tsToHijri( $ts ); + if ( !$hijri ) { + $hijri = self::tsToHijri( $ts ); + } $s .= $this->getHijriCalendarMonthName( $hijri[1] ); break; case 'xjF': - if ( !$hebrew ) $hebrew = self::tsToHebrew( $ts ); + if ( !$hebrew ) { + $hebrew = self::tsToHebrew( $ts ); + } $s .= $this->getHebrewCalendarMonthName( $hebrew[1] ); break; case 'm': @@ -787,64 +826,84 @@ class Language { $num = intval( substr( $ts, 4, 2 ) ); break; case 'xin': - if ( !$iranian ) $iranian = self::tsToIranian( $ts ); + if ( !$iranian ) { + $iranian = self::tsToIranian( $ts ); + } $num = $iranian[1]; break; case 'xmn': - if ( !$hijri ) $hijri = self::tsToHijri ( $ts ); + if ( !$hijri ) { + $hijri = self::tsToHijri ( $ts ); + } $num = $hijri[1]; break; case 'xjn': - if ( !$hebrew ) $hebrew = self::tsToHebrew( $ts ); + if ( !$hebrew ) { + $hebrew = self::tsToHebrew( $ts ); + } $num = $hebrew[1]; break; case 't': - if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts ); + if ( !$unix ) { + $unix = wfTimestamp( TS_UNIX, $ts ); + } $num = gmdate( 't', $unix ); break; case 'xjt': - if ( !$hebrew ) $hebrew = self::tsToHebrew( $ts ); + if ( !$hebrew ) { + $hebrew = self::tsToHebrew( $ts ); + } $num = $hebrew[3]; break; case 'L': - if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts ); + if ( !$unix ) { + $unix = wfTimestamp( TS_UNIX, $ts ); + } $num = gmdate( 'L', $unix ); break; - # 'o' is supported since PHP 5.1.0 - # return literal if not supported - # TODO: emulation for pre 5.1.0 versions case 'o': - if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts ); - if ( version_compare(PHP_VERSION, '5.1.0') === 1 ) - $num = date( 'o', $unix ); - else - $s .= 'o'; + if ( !$unix ) { + $unix = wfTimestamp( TS_UNIX, $ts ); + } + $num = date( 'o', $unix ); break; case 'Y': $num = substr( $ts, 0, 4 ); break; case 'xiY': - if ( !$iranian ) $iranian = self::tsToIranian( $ts ); + if ( !$iranian ) { + $iranian = self::tsToIranian( $ts ); + } $num = $iranian[0]; break; case 'xmY': - if ( !$hijri ) $hijri = self::tsToHijri( $ts ); + if ( !$hijri ) { + $hijri = self::tsToHijri( $ts ); + } $num = $hijri[0]; break; case 'xjY': - if ( !$hebrew ) $hebrew = self::tsToHebrew( $ts ); + if ( !$hebrew ) { + $hebrew = self::tsToHebrew( $ts ); + } $num = $hebrew[0]; break; case 'xkY': - if ( !$thai ) $thai = self::tsToYear( $ts, 'thai' ); + if ( !$thai ) { + $thai = self::tsToYear( $ts, 'thai' ); + } $num = $thai[0]; break; case 'xoY': - if ( !$minguo ) $minguo = self::tsToYear( $ts, 'minguo' ); + if ( !$minguo ) { + $minguo = self::tsToYear( $ts, 'minguo' ); + } $num = $minguo[0]; break; case 'xtY': - if ( !$tenno ) $tenno = self::tsToYear( $ts, 'tenno' ); + if ( !$tenno ) { + $tenno = self::tsToYear( $ts, 'tenno' ); + } $num = $tenno[0]; break; case 'y': @@ -877,15 +936,21 @@ class Language { $num = substr( $ts, 12, 2 ); break; case 'c': - if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts ); + if ( !$unix ) { + $unix = wfTimestamp( TS_UNIX, $ts ); + } $s .= gmdate( 'c', $unix ); break; case 'r': - if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts ); + if ( !$unix ) { + $unix = wfTimestamp( TS_UNIX, $ts ); + } $s .= gmdate( 'r', $unix ); break; case 'U': - if ( !$unix ) $unix = wfTimestamp( TS_UNIX, $ts ); + if ( !$unix ) { + $unix = wfTimestamp( TS_UNIX, $ts ); + } $num = $unix; break; case '\\': @@ -922,13 +987,12 @@ class Language { } elseif ( $roman ) { $s .= self::romanNumeral( $num ); $roman = false; - } elseif( $hebrewNum ) { + } elseif ( $hebrewNum ) { $s .= self::hebrewNumeral( $num ); $hebrewNum = false; } else { $s .= $this->formatNum( $num, true ); } - $num = false; } } return $s; @@ -950,19 +1014,19 @@ class Language { $gd = substr( $ts, 6, 2 ) -1; # Days passed from the beginning (including leap years) - $gDayNo = 365*$gy - + floor(($gy+3) / 4) - - floor(($gy+99) / 100) - + floor(($gy+399) / 400); + $gDayNo = 365 * $gy + + floor( ( $gy + 3 ) / 4 ) + - floor( ( $gy + 99 ) / 100 ) + + floor( ( $gy + 399 ) / 400 ); // Add days of the past months of this year - for( $i = 0; $i < $gm; $i++ ) { + for ( $i = 0; $i < $gm; $i++ ) { $gDayNo += self::$GREG_DAYS[$i]; } // Leap years - if ( $gm > 1 && (($gy%4===0 && $gy%100!==0 || ($gy%400==0)))) { + if ( $gm > 1 && ( ( $gy % 4 === 0 && $gy % 100 !== 0 || ( $gy % 400 == 0 ) ) ) ) { $gDayNo++; } @@ -971,26 +1035,27 @@ class Language { $jDayNo = $gDayNo - 79; - $jNp = floor($jDayNo / 12053); + $jNp = floor( $jDayNo / 12053 ); $jDayNo %= 12053; - $jy = 979 + 33*$jNp + 4*floor($jDayNo/1461); + $jy = 979 + 33 * $jNp + 4 * floor( $jDayNo / 1461 ); $jDayNo %= 1461; if ( $jDayNo >= 366 ) { - $jy += floor(($jDayNo-1)/365); - $jDayNo = floor(($jDayNo-1)%365); + $jy += floor( ( $jDayNo - 1 ) / 365 ); + $jDayNo = floor( ( $jDayNo - 1 ) % 365 ); } for ( $i = 0; $i < 11 && $jDayNo >= self::$IRANIAN_DAYS[$i]; $i++ ) { $jDayNo -= self::$IRANIAN_DAYS[$i]; } - $jm= $i+1; - $jd= $jDayNo+1; + $jm = $i + 1; + $jd = $jDayNo + 1; - return array($jy, $jm, $jd); + return array( $jy, $jm, $jd ); } + /** * Converting Gregorian dates to Hijri dates. * @@ -998,39 +1063,40 @@ class Language { * * @link http://phpnuke.org/modules.php?name=News&file=article&sid=8234&mode=thread&order=0&thold=0 */ - private static function tsToHijri ( $ts ) { + private static function tsToHijri( $ts ) { $year = substr( $ts, 0, 4 ); $month = substr( $ts, 4, 2 ); $day = substr( $ts, 6, 2 ); $zyr = $year; - $zd=$day; - $zm=$month; - $zy=$zyr; - - - - if (($zy>1582)||(($zy==1582)&&($zm>10))||(($zy==1582)&&($zm==10)&&($zd>14))) - { - - - $zjd=(int)((1461*($zy + 4800 + (int)( ($zm-14) /12) ))/4) + (int)((367*($zm-2-12*((int)(($zm-14)/12))))/12)-(int)((3*(int)(( ($zy+4900+(int)(($zm-14)/12))/100)))/4)+$zd-32075; - } - else - { - $zjd = 367*$zy-(int)((7*($zy+5001+(int)(($zm-9)/7)))/4)+(int)((275*$zm)/9)+$zd+1729777; - } + $zd = $day; + $zm = $month; + $zy = $zyr; + + if ( + ( $zy > 1582 ) || ( ( $zy == 1582 ) && ( $zm > 10 ) ) || + ( ( $zy == 1582 ) && ( $zm == 10 ) && ( $zd > 14 ) ) + ) + { + $zjd = (int)( ( 1461 * ( $zy + 4800 + (int)( ( $zm - 14 ) / 12 ) ) ) / 4 ) + + (int)( ( 367 * ( $zm - 2 - 12 * ( (int)( ( $zm - 14 ) / 12 ) ) ) ) / 12 ) - + (int)( ( 3 * (int)( ( ( $zy + 4900 + (int)( ( $zm - 14 ) / 12 ) ) / 100 ) ) ) / 4 ) + + $zd - 32075; + } else { + $zjd = 367 * $zy - (int)( ( 7 * ( $zy + 5001 + (int)( ( $zm - 9 ) / 7 ) ) ) / 4 ) + + (int)( ( 275 * $zm ) / 9 ) + $zd + 1729777; + } - $zl=$zjd-1948440+10632; - $zn=(int)(($zl-1)/10631); - $zl=$zl-10631*$zn+354; - $zj=((int)((10985-$zl)/5316))*((int)((50*$zl)/17719))+((int)($zl/5670))*((int)((43*$zl)/15238)); - $zl=$zl-((int)((30-$zj)/15))*((int)((17719*$zj)/50))-((int)($zj/16))*((int)((15238*$zj)/43))+29; - $zm=(int)((24*$zl)/709); - $zd=$zl-(int)((709*$zm)/24); - $zy=30*$zn+$zj-30; + $zl = $zjd -1948440 + 10632; + $zn = (int)( ( $zl - 1 ) / 10631 ); + $zl = $zl - 10631 * $zn + 354; + $zj = ( (int)( ( 10985 - $zl ) / 5316 ) ) * ( (int)( ( 50 * $zl ) / 17719 ) ) + ( (int)( $zl / 5670 ) ) * ( (int)( ( 43 * $zl ) / 15238 ) ); + $zl = $zl - ( (int)( ( 30 - $zj ) / 15 ) ) * ( (int)( ( 17719 * $zj ) / 50 ) ) - ( (int)( $zj / 16 ) ) * ( (int)( ( 15238 * $zj ) / 43 ) ) + 29; + $zm = (int)( ( 24 * $zl ) / 709 ); + $zd = $zl - (int)( ( 709 * $zm ) / 24 ); + $zy = 30 * $zn + $zj - 30; - return array ($zy, $zm, $zd); + return array( $zy, $zm, $zd ); } /** @@ -1055,7 +1121,7 @@ class Language { # Month number when September = 1, August = 12 $month += 4; - if( $month > 12 ) { + if ( $month > 12 ) { # Next year $month -= 12; $year++; @@ -1064,15 +1130,15 @@ class Language { # Calculate day of year from 1 September $dayOfYear = $day; - for( $i = 1; $i < $month; $i++ ) { - if( $i == 6 ) { + for ( $i = 1; $i < $month; $i++ ) { + if ( $i == 6 ) { # February $dayOfYear += 28; # Check if the year is leap - if( $year % 400 == 0 || ( $year % 4 == 0 && $year % 100 > 0 ) ) { + if ( $year % 400 == 0 || ( $year % 4 == 0 && $year % 100 > 0 ) ) { $dayOfYear++; } - } elseif( $i == 8 || $i == 10 || $i == 1 || $i == 3 ) { + } elseif ( $i == 8 || $i == 10 || $i == 1 || $i == 3 ) { $dayOfYear += 30; } else { $dayOfYear += 31; @@ -1083,7 +1149,7 @@ class Language { $start = self::hebrewYearStart( $hebrewYear ); # Calculate next year's start - if( $dayOfYear <= $start ) { + if ( $dayOfYear <= $start ) { # Day is before the start of the year - it is the previous year # Next year's start $nextStart = $start; @@ -1092,7 +1158,7 @@ class Language { $hebrewYear--; # Add days since previous year's 1 September $dayOfYear += 365; - if( ( $year % 400 == 0 ) || ( $year % 100 != 0 && $year % 4 == 0 ) ) { + if ( ( $year % 400 == 0 ) || ( $year % 100 != 0 && $year % 4 == 0 ) ) { # Leap year $dayOfYear++; } @@ -1111,7 +1177,7 @@ class Language { # Add 12 (or 13 for leap years) days to ignore the difference between # Hebrew and Gregorian year (353 at least vs. 365/6) - now the # difference is only about the year type - if( ( $year % 400 == 0 ) || ( $year % 100 != 0 && $year % 4 == 0 ) ) { + if ( ( $year % 400 == 0 ) || ( $year % 100 != 0 && $year % 4 == 0 ) ) { $diff += 13; } else { $diff += 12; @@ -1131,14 +1197,14 @@ class Language { $hebrewDay = $hebrewDayOfYear; $hebrewMonth = 1; $days = 0; - while( $hebrewMonth <= 12 ) { + while ( $hebrewMonth <= 12 ) { # Calculate days in this month - if( $isLeap && $hebrewMonth == 6 ) { + if ( $isLeap && $hebrewMonth == 6 ) { # Adar in a leap year - if( $isLeap ) { + if ( $isLeap ) { # Leap year - has Adar I, with 30 days, and Adar II, with 29 days $days = 30; - if( $hebrewDay <= $days ) { + if ( $hebrewDay <= $days ) { # Day in Adar I $hebrewMonth = 13; } else { @@ -1146,23 +1212,23 @@ class Language { $hebrewDay -= $days; # Try Adar II $days = 29; - if( $hebrewDay <= $days ) { + if ( $hebrewDay <= $days ) { # Day in Adar II $hebrewMonth = 14; } } } - } elseif( $hebrewMonth == 2 && $yearPattern == 2 ) { + } elseif ( $hebrewMonth == 2 && $yearPattern == 2 ) { # Cheshvan in a complete year (otherwise as the rule below) $days = 30; - } elseif( $hebrewMonth == 3 && $yearPattern == 0 ) { + } elseif ( $hebrewMonth == 3 && $yearPattern == 0 ) { # Kislev in an incomplete year (otherwise as the rule below) $days = 29; } else { # Odd months have 30 days, even have 29 $days = 30 - ( $hebrewMonth - 1 ) % 2; } - if( $hebrewDay <= $days ) { + if ( $hebrewDay <= $days ) { # In the current month break; } else { @@ -1185,21 +1251,21 @@ class Language { $a = intval( ( 12 * ( $year - 1 ) + 17 ) % 19 ); $b = intval( ( $year - 1 ) % 4 ); $m = 32.044093161144 + 1.5542417966212 * $a + $b / 4.0 - 0.0031777940220923 * ( $year - 1 ); - if( $m < 0 ) { + if ( $m < 0 ) { $m--; } $Mar = intval( $m ); - if( $m < 0 ) { + if ( $m < 0 ) { $m++; } $m -= $Mar; - $c = intval( ( $Mar + 3 * ( $year - 1 ) + 5 * $b + 5 ) % 7); - if( $c == 0 && $a > 11 && $m >= 0.89772376543210 ) { + $c = intval( ( $Mar + 3 * ( $year - 1 ) + 5 * $b + 5 ) % 7 ); + if ( $c == 0 && $a > 11 && $m >= 0.89772376543210 ) { $Mar++; - } else if( $c == 1 && $a > 6 && $m >= 0.63287037037037 ) { + } else if ( $c == 1 && $a > 6 && $m >= 0.63287037037037 ) { $Mar += 2; - } else if( $c == 2 || $c == 4 || $c == 6 ) { + } else if ( $c == 2 || $c == 4 || $c == 6 ) { $Mar++; } @@ -1215,57 +1281,74 @@ class Language { * http://en.wikipedia.org/wiki/Minguo_calendar * http://en.wikipedia.org/wiki/Japanese_era_name * - * @param $ts String: 14-character timestamp, calender name - * @return array converted year, month, day + * @param $ts String: 14-character timestamp + * @param $cName String: calender name + * @return Array: converted year, month, day */ private static function tsToYear( $ts, $cName ) { $gy = substr( $ts, 0, 4 ); $gm = substr( $ts, 4, 2 ); $gd = substr( $ts, 6, 2 ); - if (!strcmp($cName,'thai')) { + if ( !strcmp( $cName, 'thai' ) ) { # Thai solar dates # Add 543 years to the Gregorian calendar # Months and days are identical $gy_offset = $gy + 543; - } else if ((!strcmp($cName,'minguo')) || !strcmp($cName,'juche')) { + } else if ( ( !strcmp( $cName, 'minguo' ) ) || !strcmp( $cName, 'juche' ) ) { # Minguo dates # Deduct 1911 years from the Gregorian calendar # Months and days are identical $gy_offset = $gy - 1911; - } else if (!strcmp($cName,'tenno')) { + } else if ( !strcmp( $cName, 'tenno' ) ) { # Nengō dates up to Meiji period # Deduct years from the Gregorian calendar # depending on the nengo periods # Months and days are identical - if (($gy < 1912) || (($gy == 1912) && ($gm < 7)) || (($gy == 1912) && ($gm == 7) && ($gd < 31))) { + if ( ( $gy < 1912 ) || ( ( $gy == 1912 ) && ( $gm < 7 ) ) || ( ( $gy == 1912 ) && ( $gm == 7 ) && ( $gd < 31 ) ) ) { # Meiji period $gy_gannen = $gy - 1868 + 1; $gy_offset = $gy_gannen; - if ($gy_gannen == 1) + if ( $gy_gannen == 1 ) { $gy_offset = '元'; - $gy_offset = '明治'.$gy_offset; - } else if ((($gy == 1912) && ($gm == 7) && ($gd == 31)) || (($gy == 1912) && ($gm >= 8)) || (($gy > 1912) && ($gy < 1926)) || (($gy == 1926) && ($gm < 12)) || (($gy == 1926) && ($gm == 12) && ($gd < 26))) { + } + $gy_offset = '明治' . $gy_offset; + } else if ( + ( ( $gy == 1912 ) && ( $gm == 7 ) && ( $gd == 31 ) ) || + ( ( $gy == 1912 ) && ( $gm >= 8 ) ) || + ( ( $gy > 1912 ) && ( $gy < 1926 ) ) || + ( ( $gy == 1926 ) && ( $gm < 12 ) ) || + ( ( $gy == 1926 ) && ( $gm == 12 ) && ( $gd < 26 ) ) + ) + { # Taishō period $gy_gannen = $gy - 1912 + 1; $gy_offset = $gy_gannen; - if ($gy_gannen == 1) + if ( $gy_gannen == 1 ) { $gy_offset = '元'; - $gy_offset = '大正'.$gy_offset; - } else if ((($gy == 1926) && ($gm == 12) && ($gd >= 26)) || (($gy > 1926) && ($gy < 1989)) || (($gy == 1989) && ($gm == 1) && ($gd < 8))) { + } + $gy_offset = '大正' . $gy_offset; + } else if ( + ( ( $gy == 1926 ) && ( $gm == 12 ) && ( $gd >= 26 ) ) || + ( ( $gy > 1926 ) && ( $gy < 1989 ) ) || + ( ( $gy == 1989 ) && ( $gm == 1 ) && ( $gd < 8 ) ) + ) + { # Shōwa period $gy_gannen = $gy - 1926 + 1; $gy_offset = $gy_gannen; - if ($gy_gannen == 1) + if ( $gy_gannen == 1 ) { $gy_offset = '元'; - $gy_offset = '昭和'.$gy_offset; + } + $gy_offset = '昭和' . $gy_offset; } else { # Heisei period $gy_gannen = $gy - 1989 + 1; $gy_offset = $gy_gannen; - if ($gy_gannen == 1) + if ( $gy_gannen == 1 ) { $gy_offset = '元'; - $gy_offset = '平成'.$gy_offset; + } + $gy_offset = '平成' . $gy_offset; } } else { $gy_offset = $gy; @@ -1293,7 +1376,7 @@ class Language { $s = ''; for ( $pow10 = 1000, $i = 3; $i >= 0; $pow10 /= 10, $i-- ) { if ( $num >= $pow10 ) { - $s .= $table[$i][floor($num / $pow10)]; + $s .= $table[$i][floor( $num / $pow10 )]; } $num = $num % $pow10; } @@ -1324,14 +1407,14 @@ class Language { $num = 0; } else { $s .= $table[$i][intval( ( $num / $pow10 ) )]; - if( $pow10 == 1000 ) { + if ( $pow10 == 1000 ) { $s .= "'"; } } } $num = $num % $pow10; } - if( strlen( $s ) == 2 ) { + if ( strlen( $s ) == 2 ) { $str = $s . "'"; } else { $str = substr( $s, 0, strlen( $s ) - 2 ) . '"'; @@ -1379,19 +1462,18 @@ class Language { function dateFormat( $usePrefs = true ) { global $wgUser; - if( is_bool( $usePrefs ) ) { - if( $usePrefs ) { + if ( is_bool( $usePrefs ) ) { + if ( $usePrefs ) { $datePreference = $wgUser->getDatePreference(); } else { - $options = User::getDefaultOptions(); - $datePreference = (string)$options['date']; + $datePreference = (string)User::getDefaultOption( 'date' ); } } else { $datePreference = (string)$usePrefs; } // return int - if( $datePreference == '' ) { + if ( $datePreference == '' ) { return 'default'; } @@ -1431,6 +1513,7 @@ class Language { * @return string */ function date( $ts, $adj = false, $format = true, $timecorrection = false ) { + $ts = wfTimestamp( TS_MW, $ts ); if ( $adj ) { $ts = $this->userAdjust( $ts, $timecorrection ); } @@ -1449,6 +1532,7 @@ class Language { * @return string */ function time( $ts, $adj = false, $format = true, $timecorrection = false ) { + $ts = wfTimestamp( TS_MW, $ts ); if ( $adj ) { $ts = $this->userAdjust( $ts, $timecorrection ); } @@ -1467,7 +1551,7 @@ class Language { * validateTimeZone() in Special:Preferences * @return string */ - function timeanddate( $ts, $adj = false, $format = true, $timecorrection = false) { + function timeanddate( $ts, $adj = false, $format = true, $timecorrection = false ) { $ts = wfTimestamp( TS_MW, $ts ); if ( $adj ) { $ts = $this->userAdjust( $ts, $timecorrection ); @@ -1477,11 +1561,21 @@ class Language { } function getMessage( $key ) { - return self::$dataCache->getSubitem( $this->mCode, 'messages', $key ); + // Don't change getPreferredVariant() to getCode() / mCode, because: + + // 1. Some language like Chinese has multiple variant languages. Only + // getPreferredVariant() (in LanguageConverter) could return a + // sub-language which would be more suitable for the user. + // 2. To languages without multiple variants, getPreferredVariant() + // (in FakeConverter) functions exactly same as getCode() / mCode, + // it won't break anything. + + // The same below. + return self::$dataCache->getSubitem( $this->getPreferredVariant(), 'messages', $key ); } function getAllMessages() { - return self::$dataCache->getItem( $this->mCode, 'messages' ); + return self::$dataCache->getItem( $this->getPreferredVariant(), 'messages' ); } function iconv( $in, $out, $string ) { @@ -1499,63 +1593,68 @@ class Language { } // callback functions for uc(), lc(), ucwords(), ucwordbreaks() - function ucwordbreaksCallbackAscii($matches){ - return $this->ucfirst($matches[1]); + function ucwordbreaksCallbackAscii( $matches ) { + return $this->ucfirst( $matches[1] ); } - function ucwordbreaksCallbackMB($matches){ - return mb_strtoupper($matches[0]); + function ucwordbreaksCallbackMB( $matches ) { + return mb_strtoupper( $matches[0] ); } - function ucCallback($matches){ + function ucCallback( $matches ) { list( $wikiUpperChars ) = self::getCaseMaps(); return strtr( $matches[1], $wikiUpperChars ); } - function lcCallback($matches){ + function lcCallback( $matches ) { list( , $wikiLowerChars ) = self::getCaseMaps(); return strtr( $matches[1], $wikiLowerChars ); } - function ucwordsCallbackMB($matches){ - return mb_strtoupper($matches[0]); + function ucwordsCallbackMB( $matches ) { + return mb_strtoupper( $matches[0] ); } - function ucwordsCallbackWiki($matches){ + function ucwordsCallbackWiki( $matches ) { list( $wikiUpperChars ) = self::getCaseMaps(); return strtr( $matches[0], $wikiUpperChars ); } + /** + * Make a string's first character uppercase + */ function ucfirst( $str ) { $o = ord( $str ); - if ( $o < 96 ) { + if ( $o < 96 ) { // if already uppercase... return $str; } elseif ( $o < 128 ) { - return ucfirst($str); + return ucfirst( $str ); // use PHP's ucfirst() } else { // fall back to more complex logic in case of multibyte strings - return self::uc($str,true); + return $this->uc( $str, true ); } } + /** + * Convert a string to uppercase + */ function uc( $str, $first = false ) { if ( function_exists( 'mb_strtoupper' ) ) { if ( $first ) { - if ( self::isMultibyte( $str ) ) { + if ( $this->isMultibyte( $str ) ) { return mb_strtoupper( mb_substr( $str, 0, 1 ) ) . mb_substr( $str, 1 ); } else { return ucfirst( $str ); } } else { - return self::isMultibyte( $str ) ? mb_strtoupper( $str ) : strtoupper( $str ); + return $this->isMultibyte( $str ) ? mb_strtoupper( $str ) : strtoupper( $str ); } } else { - if ( self::isMultibyte( $str ) ) { - list( $wikiUpperChars ) = $this->getCaseMaps(); + if ( $this->isMultibyte( $str ) ) { $x = $first ? '^' : ''; return preg_replace_callback( "/$x([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/", - array($this,"ucCallback"), + array( $this, 'ucCallback' ), $str ); } else { @@ -1569,7 +1668,7 @@ class Language { if ( !$o ) { return strval( $str ); } elseif ( $o >= 128 ) { - return self::lc( $str, true ); + return $this->lc( $str, true ); } elseif ( $o > 96 ) { return $str; } else { @@ -1579,85 +1678,91 @@ class Language { } function lc( $str, $first = false ) { - if ( function_exists( 'mb_strtolower' ) ) - if ( $first ) - if ( self::isMultibyte( $str ) ) + if ( function_exists( 'mb_strtolower' ) ) { + if ( $first ) { + if ( $this->isMultibyte( $str ) ) { return mb_strtolower( mb_substr( $str, 0, 1 ) ) . mb_substr( $str, 1 ); - else + } else { return strtolower( substr( $str, 0, 1 ) ) . substr( $str, 1 ); - else - return self::isMultibyte( $str ) ? mb_strtolower( $str ) : strtolower( $str ); - else - if ( self::isMultibyte( $str ) ) { - list( , $wikiLowerChars ) = self::getCaseMaps(); + } + } else { + return $this->isMultibyte( $str ) ? mb_strtolower( $str ) : strtolower( $str ); + } + } else { + if ( $this->isMultibyte( $str ) ) { $x = $first ? '^' : ''; return preg_replace_callback( "/$x([A-Z]|[\\xc0-\\xff][\\x80-\\xbf]*)/", - array($this,"lcCallback"), + array( $this, 'lcCallback' ), $str ); - } else + } else { return $first ? strtolower( substr( $str, 0, 1 ) ) . substr( $str, 1 ) : strtolower( $str ); + } + } } function isMultibyte( $str ) { return (bool)preg_match( '/[\x80-\xff]/', $str ); } - function ucwords($str) { - if ( self::isMultibyte( $str ) ) { - $str = self::lc($str); + function ucwords( $str ) { + if ( $this->isMultibyte( $str ) ) { + $str = $this->lc( $str ); // regexp to find first letter in each word (i.e. after each space) $replaceRegexp = "/^([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)| ([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/"; // function to use to capitalize a single char - if ( function_exists( 'mb_strtoupper' ) ) + if ( function_exists( 'mb_strtoupper' ) ) { return preg_replace_callback( $replaceRegexp, - array($this,"ucwordsCallbackMB"), + array( $this, 'ucwordsCallbackMB' ), $str ); - else + } else { return preg_replace_callback( $replaceRegexp, - array($this,"ucwordsCallbackWiki"), + array( $this, 'ucwordsCallbackWiki' ), $str ); - } - else + } + } else { return ucwords( strtolower( $str ) ); + } } - # capitalize words at word breaks - function ucwordbreaks($str){ - if (self::isMultibyte( $str ) ) { - $str = self::lc($str); + # capitalize words at word breaks + function ucwordbreaks( $str ) { + if ( $this->isMultibyte( $str ) ) { + $str = $this->lc( $str ); // since \b doesn't work for UTF-8, we explicitely define word break chars - $breaks= "[ \-\(\)\}\{\.,\?!]"; + $breaks = "[ \-\(\)\}\{\.,\?!]"; // find first letter after word break $replaceRegexp = "/^([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)|$breaks([a-z]|[\\xc0-\\xff][\\x80-\\xbf]*)/"; - if ( function_exists( 'mb_strtoupper' ) ) + if ( function_exists( 'mb_strtoupper' ) ) { return preg_replace_callback( $replaceRegexp, - array($this,"ucwordbreaksCallbackMB"), + array( $this, 'ucwordbreaksCallbackMB' ), $str ); - else + } else { return preg_replace_callback( $replaceRegexp, - array($this,"ucwordsCallbackWiki"), + array( $this, 'ucwordsCallbackWiki' ), $str ); - } - else + } + } else { return preg_replace_callback( - '/\b([\w\x80-\xff]+)\b/', - array($this,"ucwordbreaksCallbackAscii"), - $str ); + '/\b([\w\x80-\xff]+)\b/', + array( $this, 'ucwordbreaksCallbackAscii' ), + $str + ); + } } /** @@ -1676,18 +1781,22 @@ class Language { } function checkTitleEncoding( $s ) { - if( is_array( $s ) ) { + if ( is_array( $s ) ) { wfDebugDieBacktrace( 'Given array to checkTitleEncoding.' ); } # Check for non-UTF-8 URLs - $ishigh = preg_match( '/[\x80-\xff]/', $s); - if(!$ishigh) return $s; + $ishigh = preg_match( '/[\x80-\xff]/', $s ); + if ( !$ishigh ) { + return $s; + } $isutf8 = preg_match( '/^([\x00-\x7f]|[\xc0-\xdf][\x80-\xbf]|' . '[\xe0-\xef][\x80-\xbf]{2}|[\xf0-\xf7][\x80-\xbf]{3})+$/', $s ); - if( $isutf8 ) return $s; + if ( $isutf8 ) { + return $s; + } - return $this->iconv( $this->fallback8bitEncoding(), "utf-8", $s ); + return $this->iconv( $this->fallback8bitEncoding(), 'utf-8', $s ); } function fallback8bitEncoding() { @@ -1703,15 +1812,15 @@ class Language { function hasWordBreaks() { return true; } - + /** * Some languages such as Chinese require word segmentation, * Specify such segmentation when overridden in derived class. - * + * * @param $string String * @return String */ - function wordSegmentation( $string ) { + function segmentByWord( $string ) { return $string; } @@ -1723,7 +1832,7 @@ class Language { * @return String */ function normalizeForSearch( $string ) { - return $string; + return self::convertDoubleWidth( $string ); } /** @@ -1731,8 +1840,17 @@ class Language { * range: ff00-ff5f ~= 0020-007f */ protected static function convertDoubleWidth( $string ) { - $string = preg_replace( '/\xef\xbc([\x80-\xbf])/e', 'chr((ord("$1") & 0x3f) + 0x20)', $string ); - $string = preg_replace( '/\xef\xbd([\x80-\x99])/e', 'chr((ord("$1") & 0x3f) + 0x60)', $string ); + static $full = null; + static $half = null; + + if ( $full === null ) { + $fullWidth = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz"; + $halfWidth = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz"; + $full = str_split( $fullWidth, 3 ); + $half = str_split( $halfWidth ); + } + + $string = str_replace( $full, $half, $string ); return $string; } @@ -1756,8 +1874,12 @@ class Language { */ function firstChar( $s ) { $matches = array(); - preg_match( '/^([\x00-\x7f]|[\xc0-\xdf][\x80-\xbf]|' . - '[\xe0-\xef][\x80-\xbf]{2}|[\xf0-\xf7][\x80-\xbf]{3})/', $s, $matches); + preg_match( + '/^([\x00-\x7f]|[\xc0-\xdf][\x80-\xbf]|' . + '[\xe0-\xef][\x80-\xbf]{2}|[\xf0-\xf7][\x80-\xbf]{3})/', + $s, + $matches + ); if ( isset( $matches[1] ) ) { if ( strlen( $matches[1] ) != 3 ) { @@ -1766,7 +1888,7 @@ class Language { // Break down Hangul syllables to grab the first jamo $code = utf8ToCodepoint( $matches[1] ); - if ( $code < 0xac00 || 0xd7a4 <= $code) { + if ( $code < 0xac00 || 0xd7a4 <= $code ) { return $matches[1]; } elseif ( $code < 0xb098 ) { return "\xe3\x84\xb1"; @@ -1798,7 +1920,7 @@ class Language { return "\xe3\x85\x8e"; } } else { - return ""; + return ''; } } @@ -1817,8 +1939,7 @@ class Language { # wgInputEncoding, this text will be further converted # to wgOutputEncoding. global $wgEditEncoding; - if( $wgEditEncoding == '' or - $wgEditEncoding == 'UTF-8' ) { + if ( $wgEditEncoding == '' || $wgEditEncoding == 'UTF-8' ) { return $s; } else { return $this->iconv( 'UTF-8', $wgEditEncoding, $s ); @@ -1828,12 +1949,12 @@ class Language { function recodeInput( $s ) { # Take the previous into account. global $wgEditEncoding; - if($wgEditEncoding != "") { + if ( $wgEditEncoding != '' ) { $enc = $wgEditEncoding; } else { $enc = 'UTF-8'; } - if( $enc == 'UTF-8' ) { + if ( $enc == 'UTF-8' ) { return $s; } else { return $this->iconv( $enc, 'UTF-8', $s ); @@ -1848,7 +1969,14 @@ class Language { * This is language-specific for performance reasons only. */ function normalize( $s ) { - return UtfNormal::cleanUp( $s ); + global $wgAllUnicodeFixes; + $s = UtfNormal::cleanUp( $s ); + if ( $wgAllUnicodeFixes ) { + $s = $this->transformUsingPairFile( 'normalize-ar.ser', $s ); + $s = $this->transformUsingPairFile( 'normalize-ml.ser', $s ); + } + + return $s; } /** @@ -1863,7 +1991,7 @@ class Language { if ( !isset( $this->transformData[$file] ) ) { $data = wfGetPrecompiledData( $file ); if ( $data === false ) { - throw new MWException( __METHOD__.": The transformation file $file is missing" ); + throw new MWException( __METHOD__ . ": The transformation file $file is missing" ); } $this->transformData[$file] = new ReplacementArray( $data ); } @@ -1965,7 +2093,7 @@ class Language { } } - if( !is_array( $rawEntry ) ) { + if ( !is_array( $rawEntry ) ) { error_log( "\"$rawEntry\" is not a valid magic thingie for \"$mw->mId\"" ); } else { $mw->mCaseSensitive = $rawEntry[0]; @@ -2047,15 +2175,19 @@ class Language { */ function formatNum( $number, $nocommafy = false ) { global $wgTranslateNumerals; - if (!$nocommafy) { - $number = $this->commafy($number); + if ( !$nocommafy ) { + $number = $this->commafy( $number ); $s = $this->separatorTransformTable(); - if ($s) { $number = strtr($number, $s); } + if ( $s ) { + $number = strtr( $number, $s ); + } } - if ($wgTranslateNumerals) { + if ( $wgTranslateNumerals ) { $s = $this->digitTransformTable(); - if ($s) { $number = strtr($number, $s); } + if ( $s ) { + $number = strtr( $number, $s ); + } } return $number; @@ -2063,12 +2195,16 @@ class Language { function parseFormattedNumber( $number ) { $s = $this->digitTransformTable(); - if ($s) { $number = strtr($number, array_flip($s)); } + if ( $s ) { + $number = strtr( $number, array_flip( $s ) ); + } $s = $this->separatorTransformTable(); - if ($s) { $number = strtr($number, array_flip($s)); } + if ( $s ) { + $number = strtr( $number, array_flip( $s ) ); + } - $number = strtr( $number, array (',' => '') ); + $number = strtr( $number, array( ',' => '' ) ); return $number; } @@ -2078,8 +2214,8 @@ class Language { * @param $_ mixed * @return string */ - function commafy($_) { - return strrev((string)preg_replace('/(\d{3})(?=\d)(?!\d*\.)/','$1,',strrev($_))); + function commafy( $_ ) { + return strrev( (string)preg_replace( '/(\d{3})(?=\d)(?!\d*\.)/', '$1,', strrev( $_ ) ) ); } function digitTransformTable() { @@ -2090,7 +2226,6 @@ class Language { return self::$dataCache->getItem( $this->mCode, 'separatorTransformTable' ); } - /** * Take a list of strings and build a locale-friendly comma-separated * list, using the local comma-separator message. @@ -2102,14 +2237,13 @@ class Language { function listToText( $l ) { $s = ''; $m = count( $l ) - 1; - if( $m == 1 ) { + if ( $m == 1 ) { return $l[0] . $this->getMessageFromDB( 'and' ) . $this->getMessageFromDB( 'word-separator' ) . $l[1]; - } - else { + } else { for ( $i = $m; $i >= 0; $i-- ) { if ( $i == $m ) { $s = $l[$i]; - } else if( $i == $m - 1 ) { + } else if ( $i == $m - 1 ) { $s = $l[$i] . $this->getMessageFromDB( 'and' ) . $this->getMessageFromDB( 'word-separator' ) . $s; } else { $s = $l[$i] . $this->getMessageFromDB( 'comma-separator' ) . $s; @@ -2128,7 +2262,11 @@ class Language { function commaList( $list ) { return implode( $list, - wfMsgExt( 'comma-separator', array( 'parsemag', 'escapenoentities', 'language' => $this ) ) ); + wfMsgExt( + 'comma-separator', + array( 'parsemag', 'escapenoentities', 'language' => $this ) + ) + ); } /** @@ -2140,7 +2278,11 @@ class Language { function semicolonList( $list ) { return implode( $list, - wfMsgExt( 'semicolon-separator', array( 'parsemag', 'escapenoentities', 'language' => $this ) ) ); + wfMsgExt( + 'semicolon-separator', + array( 'parsemag', 'escapenoentities', 'language' => $this ) + ) + ); } /** @@ -2151,7 +2293,11 @@ class Language { function pipeList( $list ) { return implode( $list, - wfMsgExt( 'pipe-separator', array( 'escapenoentities', 'language' => $this ) ) ); + wfMsgExt( + 'pipe-separator', + array( 'escapenoentities', 'language' => $this ) + ) + ); } /** @@ -2171,42 +2317,26 @@ class Language { */ function truncate( $string, $length, $ellipsis = '...' ) { # Use the localized ellipsis character - if( $ellipsis == '...' ) { + if ( $ellipsis == '...' ) { $ellipsis = wfMsgExt( 'ellipsis', array( 'escapenoentities', 'language' => $this ) ); } - - if( $length == 0 ) { + # Check if there is no need to truncate + if ( $length == 0 ) { return $ellipsis; - } - if ( strlen( $string ) <= abs( $length ) ) { + } elseif ( strlen( $string ) <= abs( $length ) ) { return $string; } $stringOriginal = $string; - if( $length > 0 ) { - $string = substr( $string, 0, $length ); - $char = ord( $string[strlen( $string ) - 1] ); - $m = array(); - if ($char >= 0xc0) { - # We got the first byte only of a multibyte char; remove it. - $string = substr( $string, 0, -1 ); - } elseif( $char >= 0x80 && - preg_match( '/^(.*)(?:[\xe0-\xef][\x80-\xbf]|' . - '[\xf0-\xf7][\x80-\xbf]{1,2})$/', $string, $m ) ) { - # We chopped in the middle of a character; remove it - $string = $m[1]; - } + if ( $length > 0 ) { + $string = substr( $string, 0, $length ); // xyz... + $string = $this->removeBadCharLast( $string ); $string = $string . $ellipsis; - } else { - $string = substr( $string, $length ); - $char = ord( $string[0] ); - if( $char >= 0x80 && $char < 0xc0 ) { - # We chopped in the middle of a character; remove the whole thing - $string = preg_replace( '/^[\x80-\xbf]+/', '', $string ); - } + $string = substr( $string, $length ); // ...xyz + $string = $this->removeBadCharFirst( $string ); $string = $ellipsis . $string; } - # Do not truncate if the ellipsis actually make the string longer. Bug 22181 + # Do not truncate if the ellipsis makes the string longer/equal (bug 22181) if ( strlen( $string ) < strlen( $stringOriginal ) ) { return $string; } else { @@ -2215,6 +2345,188 @@ class Language { } /** + * Remove bytes that represent an incomplete Unicode character + * at the end of string (e.g. bytes of the char are missing) + * + * @param $string String + * @return string + */ + protected function removeBadCharLast( $string ) { + $char = ord( $string[strlen( $string ) - 1] ); + $m = array(); + if ( $char >= 0xc0 ) { + # We got the first byte only of a multibyte char; remove it. + $string = substr( $string, 0, -1 ); + } elseif ( $char >= 0x80 && + preg_match( '/^(.*)(?:[\xe0-\xef][\x80-\xbf]|' . + '[\xf0-\xf7][\x80-\xbf]{1,2})$/', $string, $m ) ) + { + # We chopped in the middle of a character; remove it + $string = $m[1]; + } + return $string; + } + + /** + * Remove bytes that represent an incomplete Unicode character + * at the start of string (e.g. bytes of the char are missing) + * + * @param $string String + * @return string + */ + protected function removeBadCharFirst( $string ) { + $char = ord( $string[0] ); + if ( $char >= 0x80 && $char < 0xc0 ) { + # We chopped in the middle of a character; remove the whole thing + $string = preg_replace( '/^[\x80-\xbf]+/', '', $string ); + } + return $string; + } + + /* + * Truncate a string of valid HTML to a specified length in bytes, + * appending an optional string (e.g. for ellipses), and return valid HTML + * + * This is only intended for styled/linked text, such as HTML with + * tags like <span> and <a>, were the tags are self-contained (valid HTML) + * + * Note: tries to fix broken HTML with MWTidy + * + * @param string $text HTML string to truncate + * @param int $length (zero/positive) Maximum length (excluding ellipses) + * @param string $ellipsis String to append to the truncated text + * @returns string + */ + function truncateHtml( $text, $length, $ellipsis = '...' ) { + # Use the localized ellipsis character + if ( $ellipsis == '...' ) { + $ellipsis = wfMsgExt( 'ellipsis', array( 'escapenoentities', 'language' => $this ) ); + } + # Check if there is no need to truncate + if ( $length <= 0 ) { + return $ellipsis; // no text shown, nothing to format + } elseif ( strlen( $text ) <= $length ) { + return $text; // string short enough even *with* HTML + } + $text = MWTidy::tidy( $text ); // fix tags + $displayLen = 0; // innerHTML legth so far + $testingEllipsis = false; // checking if ellipses will make string longer/equal? + $tagType = 0; // 0-open, 1-close + $bracketState = 0; // 1-tag start, 2-tag name, 0-neither + $entityState = 0; // 0-not entity, 1-entity + $tag = $ret = ''; + $openTags = array(); // open tag stack + $textLen = strlen( $text ); + for ( $pos = 0; $pos < $textLen; ++$pos ) { + $ch = $text[$pos]; + $lastCh = $pos ? $text[$pos - 1] : ''; + $ret .= $ch; // add to result string + if ( $ch == '<' ) { + $this->truncate_endBracket( $tag, $tagType, $lastCh, $openTags ); // for bad HTML + $entityState = 0; // for bad HTML + $bracketState = 1; // tag started (checking for backslash) + } elseif ( $ch == '>' ) { + $this->truncate_endBracket( $tag, $tagType, $lastCh, $openTags ); + $entityState = 0; // for bad HTML + $bracketState = 0; // out of brackets + } elseif ( $bracketState == 1 ) { + if ( $ch == '/' ) { + $tagType = 1; // close tag (e.g. "</span>") + } else { + $tagType = 0; // open tag (e.g. "<span>") + $tag .= $ch; + } + $bracketState = 2; // building tag name + } elseif ( $bracketState == 2 ) { + if ( $ch != ' ' ) { + $tag .= $ch; + } else { + // Name found (e.g. "<a href=..."), add on tag attributes... + $pos += $this->truncate_skip( $ret, $text, "<>", $pos + 1 ); + } + } elseif ( $bracketState == 0 ) { + if ( $entityState ) { + if ( $ch == ';' ) { + $entityState = 0; + $displayLen++; // entity is one displayed char + } + } else { + if ( $ch == '&' ) { + $entityState = 1; // entity found, (e.g. " ") + } else { + $displayLen++; // this char is displayed + // Add on the other display text after this... + $skipped = $this->truncate_skip( + $ret, $text, "<>&", $pos + 1, $length - $displayLen ); + $displayLen += $skipped; + $pos += $skipped; + } + } + } + # Consider truncation once the display length has reached the maximim. + # Double-check that we're not in the middle of a bracket/entity... + if ( $displayLen >= $length && $bracketState == 0 && $entityState == 0 ) { + if ( !$testingEllipsis ) { + $testingEllipsis = true; + # Save where we are; we will truncate here unless + # the ellipsis actually makes the string longer. + $pOpenTags = $openTags; // save state + $pRet = $ret; // save state + } elseif ( $displayLen > ( $length + strlen( $ellipsis ) ) ) { + # Ellipsis won't make string longer/equal, the truncation point was OK. + $openTags = $pOpenTags; // reload state + $ret = $this->removeBadCharLast( $pRet ); // reload state, multi-byte char fix + $ret .= $ellipsis; // add ellipsis + break; + } + } + } + if ( $displayLen == 0 ) { + return ''; // no text shown, nothing to format + } + // Close the last tag if left unclosed by bad HTML + $this->truncate_endBracket( $tag, $text[$textLen - 1], $tagType, $openTags ); + while ( count( $openTags ) > 0 ) { + $ret .= '</' . array_pop( $openTags ) . '>'; // close open tags + } + return $ret; + } + + // truncateHtml() helper function + // like strcspn() but adds the skipped chars to $ret + private function truncate_skip( &$ret, $text, $search, $start, $len = -1 ) { + $skipCount = 0; + if ( $start < strlen( $text ) ) { + $skipCount = strcspn( $text, $search, $start, $len ); + $ret .= substr( $text, $start, $skipCount ); + } + return $skipCount; + } + + /* + * truncateHtml() helper function + * (a) push or pop $tag from $openTags as needed + * (b) clear $tag value + * @param String &$tag Current HTML tag name we are looking at + * @param int $tagType (0-open tag, 1-close tag) + * @param char $lastCh Character before the '>' that ended this tag + * @param array &$openTags Open tag stack (not accounting for $tag) + */ + private function truncate_endBracket( &$tag, $tagType, $lastCh, &$openTags ) { + $tag = ltrim( $tag ); + if ( $tag != '' ) { + if ( $tagType == 0 && $lastCh != '/' ) { + $openTags[] = $tag; // tag opened (didn't close itself) + } else if ( $tagType == 1 ) { + if ( $openTags && $tag == $openTags[count( $openTags ) - 1] ) { + array_pop( $openTags ); // tag closed + } + } + $tag = ''; + } + } + + /** * Grammatical transformations, needed for inflected languages * Invoked by putting {{grammar:case|word}} in a message * @@ -2224,7 +2536,7 @@ class Language { */ function convertGrammar( $word, $case ) { global $wgGrammarForms; - if ( isset($wgGrammarForms[$this->getCode()][$case][$word]) ) { + if ( isset( $wgGrammarForms[$this->getCode()][$case][$word] ) ) { return $wgGrammarForms[$this->getCode()][$case][$word]; } return $word; @@ -2239,11 +2551,17 @@ class Language { * These details may be overriden per language. */ function gender( $gender, $forms ) { - if ( !count($forms) ) { return ''; } + if ( !count( $forms ) ) { + return ''; + } $forms = $this->preConvertPlural( $forms, 2 ); - if ( $gender === 'male' ) return $forms[0]; - if ( $gender === 'female' ) return $forms[1]; - return isset($forms[2]) ? $forms[2] : $forms[0]; + if ( $gender === 'male' ) { + return $forms[0]; + } + if ( $gender === 'female' ) { + return $forms[1]; + } + return isset( $forms[2] ) ? $forms[2] : $forms[0]; } /** @@ -2262,7 +2580,9 @@ class Language { * @return string Correct form of plural for $count in this language */ function convertPlural( $count, $forms ) { - if ( !count($forms) ) { return ''; } + if ( !count( $forms ) ) { + return ''; + } $forms = $this->preConvertPlural( $forms, 2 ); return ( $count == 1 ) ? $forms[0] : $forms[1]; @@ -2277,31 +2597,31 @@ class Language { * @return array Padded array of forms or an exception if not an array */ protected function preConvertPlural( /* Array */ $forms, $count ) { - while ( count($forms) < $count ) { - $forms[] = $forms[count($forms)-1]; + while ( count( $forms ) < $count ) { + $forms[] = $forms[count( $forms ) - 1]; } return $forms; } /** - * For translaing of expiry times + * For translating of expiry times * @param $str String: the validated block time in English * @return Somehow translated block time * @see LanguageFi.php for example implementation */ function translateBlockExpiry( $str ) { - $scBlockExpiryOptions = $this->getMessageFromDB( 'ipboptions' ); - if ( $scBlockExpiryOptions == '-') { + if ( $scBlockExpiryOptions == '-' ) { return $str; } - foreach (explode(',', $scBlockExpiryOptions) as $option) { - if ( strpos($option, ":") === false ) + foreach ( explode( ',', $scBlockExpiryOptions ) as $option ) { + if ( strpos( $option, ':' ) === false ) { continue; - list($show, $value) = explode(":", $option); - if ( strcmp ( $str, $value) == 0 ) { + } + list( $show, $value ) = explode( ':', $option ); + if ( strcmp( $str, $value ) == 0 ) { return htmlspecialchars( trim( $show ) ); } } @@ -2331,8 +2651,8 @@ class Language { } # convert text to all supported variants - function autoConvertToAllVariants($text) { - return $this->mConverter->autoConvertToAllVariants($text); + function autoConvertToAllVariants( $text ) { + return $this->mConverter->autoConvertToAllVariants( $text ); } # convert text to different variants of a language. @@ -2346,16 +2666,15 @@ class Language { } # Check if this is a language with variants - function hasVariants(){ - return sizeof($this->getVariants())>1; + function hasVariants() { + return sizeof( $this->getVariants() ) > 1; } # Put custom tags (e.g. -{ }-) around math to prevent conversion - function armourMath($text){ - return $this->mConverter->armourMath($text); + function armourMath( $text ) { + return $this->mConverter->armourMath( $text ); } - /** * Perform output conversion on a string, and encode for safe HTML output. * @param $text String text to be converted @@ -2372,7 +2691,7 @@ class Language { } /** - * get the list of variants supported by this langauge + * Get the list of variants supported by this langauge * see sample implementation in LanguageZh.php * * @return array an array of language codes @@ -2381,20 +2700,27 @@ class Language { return $this->mConverter->getVariants(); } - - function getPreferredVariant( $fromUser = true, $fromHeader = false ) { - return $this->mConverter->getPreferredVariant( $fromUser, $fromHeader ); + function getPreferredVariant() { + return $this->mConverter->getPreferredVariant(); + } + + function getDefaultVariant() { + return $this->mConverter->getDefaultVariant(); + } + + function getURLVariant() { + return $this->mConverter->getURLVariant(); } /** - * if a language supports multiple variants, it is + * If a language supports multiple variants, it is * possible that non-existing link in one variant * actually exists in another variant. this function * tries to find it. See e.g. LanguageZh.php * * @param $link String: the name of the link * @param $nt Mixed: the title object of the link - * @param boolean $ignoreOtherCond: to disable other conditions when + * @param $ignoreOtherCond Boolean: to disable other conditions when * we need to transclude a template or update a category's link * @return null the input parameters may be modified upon return */ @@ -2406,12 +2732,13 @@ class Language { * If a language supports multiple variants, converts text * into an array of all possible variants of the text: * 'variant' => text in that variant + * + * @deprecated Use autoConvertToAllVariants() */ - function convertLinkToAllVariants($text){ - return $this->mConverter->convertLinkToAllVariants($text); + function convertLinkToAllVariants( $text ) { + return $this->mConverter->convertLinkToAllVariants( $text ); } - /** * returns language specific options used by User::getPageRenderHash() * for example, the preferred language variant @@ -2423,7 +2750,7 @@ class Language { } /** - * for languages that support multiple variants, the title of an + * For languages that support multiple variants, the title of an * article may be displayed differently in different variants. this * function returns the apporiate title defined in the body of the article. * @@ -2441,7 +2768,7 @@ class Language { * @param $noParse * @return string the tagged text */ - function markNoConversion( $text, $noParse=false ) { + function markNoConversion( $text, $noParse = false ) { return $this->mConverter->markNoConversion( $text, $noParse ); } @@ -2543,7 +2870,9 @@ class Language { } function fixVariableInNamespace( $talk ) { - if ( strpos( $talk, '$1' ) === false ) return $talk; + if ( strpos( $talk, '$1' ) === false ) { + return $talk; + } global $wgMetaNamespace; $talk = str_replace( '$1', $wgMetaNamespace, $talk ); @@ -2579,20 +2908,34 @@ class Language { } function formatTimePeriod( $seconds ) { - if ( $seconds < 10 ) { - return $this->formatNum( sprintf( "%.1f", $seconds ) ) . wfMsg( 'seconds-abbrev' ); - } elseif ( $seconds < 60 ) { - return $this->formatNum( round( $seconds ) ) . wfMsg( 'seconds-abbrev' ); - } elseif ( $seconds < 3600 ) { - return $this->formatNum( floor( $seconds / 60 ) ) . wfMsg( 'minutes-abbrev' ) . - $this->formatNum( round( fmod( $seconds, 60 ) ) ) . wfMsg( 'seconds-abbrev' ); + if ( round( $seconds * 10 ) < 100 ) { + return $this->formatNum( sprintf( "%.1f", round( $seconds * 10 ) / 10 ) ) . $this->getMessageFromDB( 'seconds-abbrev' ); + } elseif ( round( $seconds ) < 60 ) { + return $this->formatNum( round( $seconds ) ) . $this->getMessageFromDB( 'seconds-abbrev' ); + } elseif ( round( $seconds ) < 3600 ) { + $minutes = floor( $seconds / 60 ); + $secondsPart = round( fmod( $seconds, 60 ) ); + if ( $secondsPart == 60 ) { + $secondsPart = 0; + $minutes++; + } + return $this->formatNum( $minutes ) . $this->getMessageFromDB( 'minutes-abbrev' ) . ' ' . + $this->formatNum( $secondsPart ) . $this->getMessageFromDB( 'seconds-abbrev' ); } else { $hours = floor( $seconds / 3600 ); $minutes = floor( ( $seconds - $hours * 3600 ) / 60 ); $secondsPart = round( $seconds - $hours * 3600 - $minutes * 60 ); - return $this->formatNum( $hours ) . wfMsg( 'hours-abbrev' ) . - $this->formatNum( $minutes ) . wfMsg( 'minutes-abbrev' ) . - $this->formatNum( $secondsPart ) . wfMsg( 'seconds-abbrev' ); + if ( $secondsPart == 60 ) { + $secondsPart = 0; + $minutes++; + } + if ( $minutes == 60 ) { + $minutes = 0; + $hours++; + } + return $this->formatNum( $hours ) . $this->getMessageFromDB( 'hours-abbrev' ) . ' ' . + $this->formatNum( $minutes ) . $this->getMessageFromDB( 'minutes-abbrev' ) . ' ' . + $this->formatNum( $secondsPart ) . $this->getMessageFromDB( 'seconds-abbrev' ); } } @@ -2621,13 +2964,13 @@ class Language { function formatSize( $size ) { // For small sizes no decimal places necessary $round = 0; - if( $size > 1024 ) { + if ( $size > 1024 ) { $size = $size / 1024; - if( $size > 1024 ) { + if ( $size > 1024 ) { $size = $size / 1024; // For MB and bigger two decimal places are smarter $round = 2; - if( $size > 1024 ) { + if ( $size > 1024 ) { $size = $size / 1024; $msg = 'size-gigabytes'; } else { |