X-Git-Url: http://git.cyclocoop.org/?a=blobdiff_plain;f=languages%2Fclasses%2FLanguageKk.php;h=cf90e96e6e419a77fc8022c0ac585c5afcd6063e;hb=70ab71ec3a18652b9fc5fd84162b8ec230ce4848;hp=71ca081df4f9772f3fdcf8c309a7409e473a1cbf;hpb=2f12a58d93a4783acb97956c945d0f28875e42c4;p=lhc%2Fweb%2Fwiklou.git diff --git a/languages/classes/LanguageKk.php b/languages/classes/LanguageKk.php index 71ca081df4..cf90e96e6e 100644 --- a/languages/classes/LanguageKk.php +++ b/languages/classes/LanguageKk.php @@ -2,211 +2,456 @@ /** Kazakh (Қазақша) * converter routines * - * @package MediaWiki - * @subpackage Language + * @addtogroup Language */ require_once( dirname(__FILE__).'/../LanguageConverter.php' ); -require_once( dirname(__FILE__).'/LanguageKk_kz.php' ); +require_once( dirname(__FILE__).'/LanguageKk_cyrl.php' ); + +define( 'KK_C_UC', 'АӘБВГҒДЕЁЖЗИЙКҚЛМНҢОӨПРСТУҰҮФХҺЦЧШЩЪЫІЬЭЮЯ' ); # Kazakh Cyrillic uppercase +define( 'KK_C_LC', 'аәбвгғдеёжзийкқлмнңоөпрстуұүфхһцчшщъыіьэюя' ); # Kazakh Cyrillic lowercase +define( 'KK_L_UC', 'AÄBCÇDEÉFGĞHIİÏJKLMNÑOÖPQRSŞTUÜVWXYÝZ' ); # Kazakh Latin uppercase +define( 'KK_L_LC', 'aäbcçdeéfgğhıiïjklmnñoöpqrsştuüvwxyýz' ); # Kazakh Latin lowercase +//define( 'KK_A', 'ٴابپتجحدرزسشعفقكلمنڭەوۇۋۆىيچھ' ); # Kazakh Arabic +define( 'H_HAMZA', 'Ù´' ); # U+0674 ARABIC LETTER HIGH HAMZA +//define( 'ZWNJ', '‌' ); # U+200C ZERO WIDTH NON-JOINER class KkConverter extends LanguageConverter { - var $mLatinToCyrillic = array( - 'YA' => 'Я', 'Ya' => 'Я', 'ya' => 'я', 'YE' => 'Е', 'Ye' => 'У', 'ye' => 'е', - 'YO' => 'Ё', 'Yo' => 'Ё', 'yo' => 'ё', 'YU' => 'Ю', 'Yu' => 'Ю', 'yu' => 'ю', - 'YW' => 'Ю', 'Yw' => 'Ю', 'yw' => 'ю', - - 'bʺ' => 'бъ', 'dʺ' => 'дъ', 'fʺ' => 'фъ', 'gʺ' => 'гъ', 'kʺ' => 'къ', 'lʺ' => 'лъ', - 'mʺ' => 'мъ', 'nʺ' => 'нъ', 'pʺ' => 'пъ', 'rʺ' => 'ръ', 'sʺ' => 'съ', 'tʺ' => 'тъ', - 'vʺ' => 'въ', 'zʺ' => 'зъ', - /* 'jʺ' => 'жъ', 'cʺ' => 'цъ', 'çʺ' => 'чъ', 'şʺ' => 'шъ', */ - - 'ŞÇʹ'=> 'ЩЬ', 'Şçʹ'=> 'Щь', 'Bʺ' => 'БЪ', 'Dʺ' => 'ДЪ', 'Fʺ' => 'ФЪ', 'Gʺ' => 'ГЪ', 'Kʺ' => 'КЪ', 'Lʺ' => 'ЛЪ', - 'Mʺ' => 'МЪ', 'Nʺ' => 'НЪ', 'Pʺ' => 'ПЪ', 'Rʺ' => 'РЪ', 'Sʺ' => 'СЪ', 'Tʺ' => 'ТЪ', - 'Vʺ' => 'ВЪ', 'Zʺ' => 'ЗЪ', - /* 'Jʺ' => 'ЖЪ', 'Cʺ' => 'ЦЪ', 'Çʺ' => 'ЧЪ', 'Şʺ' => 'ШЪ', */ - - 'şçʹ'=> 'щь', 'bʹ' => 'бь', 'dʹ' => 'дь', 'fʹ' => 'фь', 'gʹ' => 'гь', 'kʹ' => 'кь', 'lʹ' => 'ль', - 'mʹ' => 'мь', 'nʹ' => 'нь', 'pʹ' => 'пь', 'rʹ' => 'рь', 'sʹ' => 'сь', 'tʹ' => 'ть', - 'vʹ' => 'вь', 'zʹ' => 'зь', 'jʹ' => 'жь', 'cʹ' => 'ць', 'çʹ' => 'чь', 'şʹ' => 'шь', - - 'Bʹ' => 'БЬ', 'Dʹ' => 'ДЬ', 'Fʹ' => 'ФЬ', 'Gʹ' => 'ГЬ', 'Kʹ' => 'КЬ', 'Lʹ' => 'ЛЬ', - 'Mʹ' => 'МЬ', 'Nʹ' => 'НЬ', 'Pʹ' => 'ПЬ', 'Rʹ' => 'РЬ', 'Sʹ' => 'СЬ', 'Tʹ' => 'ТЬ', - 'Vʹ' => 'ВЬ', 'Zʹ' => 'ЗЬ', 'Jʹ' => 'ЖЬ', 'Cʹ' => 'ЦЬ', 'Çʹ' => 'ЧЬ', 'Şʹ' => 'ШЬ', - - 'ŞÇ' => 'Щ', 'Şç' => 'Щ', 'şç' => 'щ', - - 'a' => 'а', 'ä' => 'ә', 'b' => 'б', 'c' => 'ц', 'ç' => 'ч', 'd' => 'д', 'e' => 'е', - 'é' => 'э', 'f' => 'ф', 'g' => 'г', 'ğ' => 'ғ', 'h' => 'Ò»', 'i' => 'і', 'ı' => 'ы', - 'ï' => 'и', 'j' => 'ж', 'k' => 'к', 'l' => 'л', 'm' => 'м', 'n' => 'н', 'ñ' => 'Ò£', - 'o' => 'о', 'ö' => 'Ó©', 'p' => 'п', 'q' => 'қ', 'r' => 'р', 's' => 'с', 'ş' => 'ш', - 't' => 'т', 'u' => 'Ò±', 'ü' => 'Ò¯', 'v' => 'в', 'w' => 'у', 'x' => 'х', 'ý' => 'й', - 'z' => 'з', - - 'A' => 'А', 'Ä' => 'Ә', 'B' => 'Б', 'C' => 'Ц', 'Ç' => 'Ч', 'D' => 'Д', 'E' => 'Е', - 'É' => 'Э', 'F' => 'Ф', 'G' => 'Г', 'Ğ' => 'Ғ', 'H' => 'Òº', 'Ä°' => 'І', 'I' => 'Ы', - 'Ï' => 'И', 'J' => 'Ж', 'K' => 'К', 'L' => 'Л', 'M' => 'М', 'N' => 'Н', 'Ñ' => 'Ò¢', - 'O' => 'О', 'Ö' => 'Ó¨', 'P' => 'П', 'Q' => 'Қ', 'R' => 'Р', 'S' => 'С', 'Ş' => 'Ш', - 'T' => 'Т', 'U' => 'Ò°', 'Ü' => 'Ò®', 'V' => 'В', 'W' => 'У', 'Ý' => 'Й', 'X' => 'Ð¥', - 'Z' => 'З' - ); - - var $mCyrillicToLatin = array( - 'а' => 'a', 'ә' => 'ä', 'б' => 'b', 'в' => 'v', 'г' => 'g', 'ғ' => 'ğ', - 'д' => 'd', 'е' => 'e', 'ё' => 'yo', 'ж' => 'j', 'з' => 'z', 'и' => 'ï', - 'й' => 'ý', 'к' => 'k', 'қ' => 'q', 'л' => 'l', 'м' => 'm', 'н' => 'n', - 'Ò£' => 'ñ', 'о' => 'o', 'Ó©' => 'ö', 'п' => 'p', 'р' => 'r', 'с' => 's', - 'т' => 't', 'у' => 'w', 'Ò±' => 'u', 'Ò¯' => 'ü', 'ф' => 'f', 'х' => 'x', - 'Ò»' => 'h', 'ц' => 'c', 'ч' => 'ç', 'ш' => 'ş', 'щ' => 'şç', 'ъ' => 'ʺ', - 'ы' => 'ı', 'ь' => 'ʹ', 'і' => 'i', 'э' => 'é', 'ю' => 'yw', 'я' => 'ya', - - 'А' => 'A', 'Ә' => 'Ä', 'Б' => 'B', 'В' => 'V', 'Г' => 'G', 'Ғ' => 'Ğ', - 'Д' => 'D', 'Е' => 'E', 'Ё' => 'Yo', 'Ж' => 'J', 'З' => 'Z', 'И' => 'Ï', - 'Й' => 'Ý', 'К' => 'K', 'Қ' => 'Q', 'Л' => 'L', 'М' => 'M', 'Н' => 'N', - 'Ò¢' => 'Ñ', 'О' => 'O', 'Ó¨' => 'Ö', 'П' => 'P', 'Р' => 'R', 'С' => 'S', - 'Т' => 'T', 'У' => 'W', 'Ò°' => 'U', 'Ò®' => 'Ü', 'Ф' => 'F', 'Ð¥' => 'X', - 'Òº' => 'H', 'Ц' => 'C', 'Ч' => 'Ç', 'Ш' => 'Ş', 'Щ' => 'Şç', 'Ъ' => 'ʺ', - 'Ы' => 'I', 'Ь' => 'ʹ', 'І' => 'Ä°', 'Э' => 'É', 'Ю' => 'Yw', 'Я' => 'Ya' - ); - - var $mCyrillicToArabic = array( - 'ла' => 'لا', 'ЛА' => 'لا', 'Ла' => 'لا', - - 'а' => 'ا', 'ә' => 'ٴا', 'б' => 'ب', 'в' => 'ۆ', 'г' => 'Ú¯', 'ғ' => 'ع', - 'д' => 'د', 'е' => 'ە', 'ё' => 'يو', 'ж' => 'ج', 'з' => 'ز', 'и' => 'ي', - 'й' => 'ي', 'к' => 'ك', 'қ' => 'ق', 'л' => 'ل', 'м' => 'م', 'н' => 'ن', - 'Ò£' => 'Ú­', 'о' => 'و', 'Ó©' => 'ٴو', 'п' => 'Ù¾', 'р' => 'ر', 'с' => 'س', - 'т' => 'ت', 'у' => 'ۋ', 'Ò±' => 'ۇ', 'Ò¯' => 'ٴۇ', 'ф' => 'ف', 'х' => 'Ø­', - 'Ò»' => 'Ú¾', 'ц' => 'تس', 'ч' => 'چ', 'ш' => 'Ø´', 'щ' => 'شش', 'ъ' => 'ي', - 'ы' => 'ى', 'ь' => 'ي', 'і' => 'ٴى', 'э' => 'ە', 'ю' => 'يۋ', 'я' => 'يا', - - 'А' => 'ا', 'Ә' => 'ٴا', 'Б' => 'ب', 'В' => 'ۆ', 'Г' => 'Ú¯', 'Ғ' => 'ع', - 'Д' => 'د', 'Е' => 'ە', 'Ё' => 'يو', 'Ж' => 'ج', 'З' => 'ز', 'И' => 'ي', - 'Й' => 'ي', 'К' => 'ك', 'Қ' => 'ق', 'Л' => 'ل', 'М' => 'م', 'Н' => 'ن', - 'Ò¢' => 'Ú­', 'О' => 'و', 'Ó¨' => 'ٴو', 'П' => 'Ù¾', 'Р' => 'ر', 'С' => 'س', - 'Т' => 'ت', 'У' => 'ۋ', 'Ò°' => 'ۇ', 'Ò®' => 'ٴۇ', 'Ф' => 'ف', 'Ð¥' => 'Ø­', - 'Òº' => 'Ú¾', 'Ц' => 'تس', 'Ч' => 'چ', 'Ш' => 'Ø´', 'Щ' => 'شش', 'Ъ' => 'ي', - 'Ы' => 'ى', 'Ь' => 'ي', 'І' => 'ٴى', 'Э' => 'ە', 'Ю' => 'يۋ', 'Я' => 'يا', - ); - - function loadDefaultTables() { + + function __construct($langobj, $maincode, + $variants=array(), + $variantfallbacks=array(), + $markup=array(), + $flags = array()) { + parent::__construct( $langobj, $maincode, + $variants, $variantfallbacks, $markup, $flags ); + + // No point delaying this since they're in code. + // Waiting until loadDefaultTables() means they never get loaded + // when the tables themselves are loaded from cache. + $this->loadRegs(); + } + + function loadDefaultTables() { + // require( dirname(__FILE__)."/../../includes/KkConversion.php" ); + // Placeholder for future implementing. Remove variables declarations + // after generating KkConversion.php + $kk2Cyrl = array(); + $kk2Latn = array(); + $kk2Arab = array(); + $kk2KZ = array(); + $kk2TR = array(); + $kk2CN = array(); + $this->mTables = array( - 'kk-kz' => new ReplacementArray( $this->mLatinToCyrillic ), - 'kk-tr' => new ReplacementArray( $this->mCyrillicToLatin ), - 'kk-cn' => new ReplacementArray( $this->mCyrillicToArabic ), - 'kk' => new ReplacementArray() + 'kk-cyrl' => new ReplacementArray( $kk2Cyrl ), + 'kk-latn' => new ReplacementArray( $kk2Latn ), + 'kk-arab' => new ReplacementArray( $kk2Arab ), + 'kk-kz' => new ReplacementArray( array_merge($kk2Cyrl, $kk2KZ) ), + 'kk-tr' => new ReplacementArray( array_merge($kk2Latn, $kk2TR) ), + 'kk-cn' => new ReplacementArray( array_merge($kk2Arab, $kk2CN) ), + 'kk' => new ReplacementArray() ); - } - - /* - * Override function from LanguageConvertor - */ - function getPreferredVariant(){ - global $wgUser, $wgRequest, $wgTitle;; - // Additional checks: - // - There should be no conversion for Talk pages - if($wgTitle!=NULL && $wgTitle->isTalkPage()){ - return $this->mMainLanguageCode; - } - return parent::getPreferredVariant(); - } - - /* - * A function wrapper, if there is no selected variant, - * leave the link names as they were - */ - function findVariantLink( &$link, &$nt ) { - $oldlink=$link; - parent::findVariantLink($link,$nt); - if($this->getPreferredVariant()==$this->mMainLanguageCode) - $link=$oldlink; - } - - /* - * We want our external link captions to be converted in variants, - * so we return the original text instead -{$text}-, except for URLs - */ - function markNoConversion($text) { - if(preg_match("/^https?:\/\/|ftp:\/\/|irc:\/\//",$text)) - return parent::markNoConversion($text); - return $text; - } - - /* - * An ugly function wrapper for parsing Image titles - * (to prevent image name conversion) - */ - function autoConvert($text, $toVariant=false) { - global $wgTitle; - if($wgTitle->getNameSpace()==NS_IMAGE){ - $imagename = $wgTitle->getNsText(); - if(preg_match("/^$imagename:/",$text)) return $text; - } - return parent::autoConvert($text,$toVariant); - } + } - /** - * It translates text into variant, specials: - * - ommiting roman numbers - */ - function translate($text, $toVariant){ - $breaks = '[^\w\x80-\xff]'; + function postLoadTables() { + $this->mTables['kk-kz']->merge( $this->mTables['kk-cyrl'] ); + $this->mTables['kk-tr']->merge( $this->mTables['kk-latn'] ); + $this->mTables['kk-cn']->merge( $this->mTables['kk-arab'] ); + } - // regexp for roman numbers - $roman = 'M{0,4}(CM|CD|D?C{0,3})(XC|XL|L?X{0,3})(IX|IV|V?I{0,3})'; + function loadRegs() { + + $this->mCyrl2Latn = array( + ## Punctuation + '/№/u' => 'No.', + ## Е after vowels + '/([АӘЕЁИОӨҰҮЭЮЯЪЬ])Е/u' => '$1YE', + '/([АӘЕЁИОӨҰҮЭЮЯЪЬ])е/ui' => '$1ye', + ## leading ЁЮЯЩ + '/^Ё(['.KK_C_UC.']|$)/u' => 'YO$1', '/^Ё(['.KK_C_LC.']|$)/u' => 'Yo$1', + '/^Ю(['.KK_C_UC.']|$)/u' => 'YU$1', '/^Ю(['.KK_C_LC.']|$)/u' => 'Yu$1', + '/^Я(['.KK_C_UC.']|$)/u' => 'YA$1', '/^Я(['.KK_C_LC.']|$)/u' => 'Ya$1', + '/^Щ(['.KK_C_UC.']|$)/u' => 'ŞÇ$1', '/^Щ(['.KK_C_LC.']|$)/u' => 'Şç$1', + ## other ЁЮЯ + '/Ё/u' => 'YO', '/ё/u' => 'yo', + '/Ю/u' => 'YU', '/ю/u' => 'yu', + '/Я/u' => 'YA', '/я/u' => 'ya', + '/Щ/u' => 'ŞÇ', '/щ/u' => 'şç', + ## soft and hard signs + '/[ъЪ]/u' => 'ʺ', '/[ьЬ]/u' => 'ʹ', + ## other characters + '/А/u' => 'A', '/а/u' => 'a', '/Ә/u' => 'Ä', '/ә/u' => 'ä', + '/Б/u' => 'B', '/б/u' => 'b', '/В/u' => 'V', '/в/u' => 'v', + '/Г/u' => 'G', '/г/u' => 'g', '/Ғ/u' => 'Ğ', '/ғ/u' => 'ğ', + '/Д/u' => 'D', '/д/u' => 'd', '/Е/u' => 'E', '/е/u' => 'e', + '/Ж/u' => 'J', '/ж/u' => 'j', '/З/u' => 'Z', '/з/u' => 'z', + '/И/u' => 'Ï', '/и/u' => 'ï', '/Й/u' => 'Ý', '/й/u' => 'ý', + '/К/u' => 'K', '/к/u' => 'k', '/Қ/u' => 'Q', '/қ/u' => 'q', + '/Л/u' => 'L', '/л/u' => 'l', '/М/u' => 'M', '/м/u' => 'm', + '/Н/u' => 'N', '/н/u' => 'n', '/Ò¢/u' => 'Ñ', '/Ò£/u' => 'ñ', + '/О/u' => 'O', '/о/u' => 'o', '/Ó¨/u' => 'Ö', '/Ó©/u' => 'ö', + '/П/u' => 'P', '/п/u' => 'p', '/Р/u' => 'R', '/р/u' => 'r', + '/С/u' => 'S', '/с/u' => 's', '/Т/u' => 'T', '/т/u' => 't', + '/У/u' => 'W', '/у/u' => 'w', '/Ò°/u' => 'U', '/Ò±/u' => 'u', + '/Ò®/u' => 'Ü', '/Ò¯/u' => 'ü', '/Ф/u' => 'F', '/ф/u' => 'f', + '/Ð¥/u' => 'X', '/х/u' => 'x', '/Òº/u' => 'H', '/Ò»/u' => 'h', + '/Ц/u' => 'C', '/ц/u' => 'c', '/Ч/u' => 'Ç', '/ч/u' => 'ç', + '/Ш/u' => 'Ş', '/ш/u' => 'ş', '/Ы/u' => 'I', '/ы/u' => 'ı', + '/І/u' => 'Ä°', '/і/u' => 'i', '/Э/u' => 'É', '/э/u' => 'é', + ); - $reg = '/^'.$roman.'$|^'.$roman.$breaks.'|'.$breaks.$roman.'$|'.$breaks.$roman.$breaks.'/'; + $this->mLatn2Cyrl = array( + ## Punctuation + '/#|No\./' => '№', + ## Şç + '/ŞÇʹ/u'=> 'ЩЬ', '/Şçʹ/u'=> 'Щь', '/Şçʹ/u'=> 'Щь', + '/Ş[Çç]/u' => 'Щ', '/şç/u' => 'щ', + ## soft and hard signs + '/(['.KK_L_UC.'])ʺ(['.KK_L_UC.'])/u' => '$1Ъ$2', + '/ʺ(['.KK_L_LC.'])/u' => 'ъ$1', + '/(['.KK_L_UC.'])ʹ(['.KK_L_UC.'])/u' => '$1Ь$2', + '/ʹ(['.KK_L_LC.'])/u' => 'ь$1', + '/ʺ/u' => 'ъ', + '/ʹ/u' => 'ь', + ## Ye Yo Yu Ya. + '/Y[Ee]/u' => 'Е', '/ye/u' => 'е', + '/Y[Oo]/u' => 'Ё', '/yo/u' => 'ё', + '/Y[UWuw]/u' => 'Ю', '/y[uw]/u' => 'ю', + '/Y[Aa]/u' => 'Я', '/ya/u' => 'я', + ## other characters + '/A/u' => 'А', '/a/u' => 'а', '/Ä/u' => 'Ә', '/ä/u' => 'ә', + '/B/u' => 'Б', '/b/u' => 'б', '/C/u' => 'Ц', '/c/u' => 'ц', + '/Ç/u' => 'Ч', '/ç/u' => 'ч', '/D/u' => 'Д', '/d/u' => 'д', + '/E/u' => 'Е', '/e/u' => 'е', '/É/u' => 'Э', '/é/u' => 'э', + '/F/u' => 'Ф', '/f/u' => 'ф', '/G/u' => 'Г', '/g/u' => 'г', + '/Ğ/u' => 'Ғ', '/ğ/u' => 'ғ', '/H/u' => 'Òº', '/h/u' => 'Ò»', + '/I/u' => 'Ы', '/ı/u' => 'ы', '/Ä°/u' => 'І', '/i/u' => 'і', + '/Ï/u' => 'И', '/ï/u' => 'и', '/J/u' => 'Ж', '/j/u' => 'ж', + '/K/u' => 'К', '/k/u' => 'к', '/L/u' => 'Л', '/l/u' => 'л', + '/M/u' => 'М', '/m/u' => 'м', '/N/u' => 'Н', '/n/u' => 'н', + '/Ñ/u' => 'Ò¢', '/ñ/u' => 'Ò£', '/O/u' => 'О', '/o/u' => 'о', + '/Ö/u' => 'Ó¨', '/ö/u' => 'Ó©', '/P/u' => 'П', '/p/u' => 'п', + '/Q/u' => 'Қ', '/q/u' => 'қ', '/R/u' => 'Р', '/r/u' => 'р', + '/S/u' => 'С', '/s/u' => 'с', '/Ş/u' => 'Ш', '/ş/u' => 'ш', + '/T/u' => 'Т', '/t/u' => 'т', '/U/u' => 'Ò°', '/u/u' => 'Ò±', + '/Ü/u' => 'Ò®', '/ü/u' => 'Ò¯', '/V/u' => 'В', '/v/u' => 'в', + '/W/u' => 'У', '/w/u' => 'у', '/Ý/u' => 'Й', '/ý/u' => 'й', + '/X/u' => 'Ð¥', '/x/u' => 'х', '/Z/u' => 'З', '/z/u' => 'з', + ); - $matches = preg_split($reg, $text, -1, PREG_SPLIT_OFFSET_CAPTURE); + $this->mCyLa2Arab = array( + ## Punctuation -> Arabic + '/#|№|No\./u' => '؀', # ؀ + '/\,/' => '،', # ، + '/;/' => '؛', # ؛ + '/\?/' => '؟', # ؟ + '/%/' => 'Ùª', # ٪ + '/\*/' => 'Ù­', # ٭ + ## Digits -> Arabic + '/0/' => 'Û°', # ۰ + '/1/' => 'Û±', # ۱ + '/2/' => 'Û²', # ۲ + '/3/' => 'Û³', # ۳ + '/4/' => 'Û´', # ۴ + '/5/' => 'Ûµ', # ۵ + '/6/' => 'Û¶', # ۶ + '/7/' => 'Û·', # ۷ + '/8/' => 'Û¸', # ۸ + '/9/' => 'Û¹', # ۹ + ## Cyrillic -> Arabic + '/Аллаһ/ui' => 'ï·²', + '/([АӘЕЁИОӨҰҮЭЮЯЪЬ])е/ui' => '$1يە', + '/[еэ]/ui' => 'ە', '/[ъь]/ui' => '', + '/[аә]/ui' => 'ا', '/[оө]/ui' => 'و', '/[Ò±Ò¯]/ui' => 'ۇ', '/[ыі]/ui' => 'ى', + '/[и]/ui' => 'ىي', '/ё/ui' => 'يو', '/ю/ui' => 'يۋ', '/я/ui' => 'يا', '/[й]/ui' => 'ي', + '/ц/ui' => 'تس', '/щ/ui' => 'شش', + '/Ò»/ui' => 'Ø­', '/ч/ui' => 'تش', + #'/Ò»/ui' => 'Ú¾', '/ч/ui' => 'چ', + '/б/ui' => 'ب', '/в/ui' => 'ۆ', '/г/ui' => 'Ú¯', '/ғ/ui' => 'ع', + '/д/ui' => 'د', '/ж/ui' => 'ج', '/з/ui' => 'ز', '/к/ui' => 'ك', + '/қ/ui' => 'ق', '/л/ui' => 'ل', '/м/ui' => 'م', '/н/ui' => 'ن', + '/Ò£/ui' => 'Ú­', '/п/ui' => 'Ù¾', '/р/ui' => 'ر', '/с/ui' => 'س', + '/т/ui' => 'ت', '/у/ui' => 'ۋ', '/ф/ui' => 'ف', '/х/ui' => 'Ø­', + '/ш/ui' => 'Ø´', + ## Latin -> Arabic // commented for now... + /*'/Allah/ui' => 'ï·²', + '/[eé]/ui' => 'ە', '/[yý]/ui' => 'ي', '/[ʺʹ]/ui' => '', + '/[aä]/ui' => 'ا', '/[oö]/ui' => 'و', '/[uü]/ui' => 'ۇ', + '/[ï]/ui' => 'ىي', '/[ıIiÄ°]/u' => 'ى', + '/c/ui' => 'تس', + '/ç/ui' => 'تش', '/h/ui' => 'Ø­', + #'/ç/ui' => 'چ', '/h/ui' => 'Ú¾', + '/b/ui' => 'ب','/d/ui' => 'د', + '/f/ui' => 'ف', '/g/ui' => 'Ú¯', '/ğ/ui' => 'ع', + '/j/ui' => 'ج', '/k/ui' => 'ك', '/l/ui' => 'ل', '/m/ui' => 'م', + '/n/ui' => 'ن', '/ñ/ui' => 'Ú­', '/p/ui' => 'Ù¾', '/q/ui' => 'ق', + '/r/ui' => 'ر', '/s/ui' => 'س', '/ş/ui' => 'Ø´', '/t/ui' => 'ت', + '/v/ui' => 'ۆ', '/w/ui' => 'ۋ', '/x/ui' => 'Ø­', '/z/ui' => 'ز',*/ + ); - $m = array_shift($matches); - $ret = strtr($m[0], $this->mTables[$toVariant]); - $mstart = $m[1]+strlen($m[0]); - foreach($matches as $m) { - $ret .= substr($text, $mstart, $m[1]-$mstart); - $ret .= parent::translate($m[0], $toVariant); - $mstart = $m[1] + strlen($m[0]); + } + + /* rules should be defined as -{ekavian | iyekavian-} -or- + -{code:text | code:text | ...}- + update: delete all rule parsing because it's not used + currently, and just produces a couple of bugs + */ + function parseManualRule($rule, $flags=array()) { + if(in_array('T',$flags)){ + return parent::parseManualRule($rule, $flags); } + // otherwise ignore all formatting + foreach($this->mVariants as $v) { + $carray[$v] = $rule; + } + + return $carray; + } + + // Do not convert content on talk pages + function parserConvert( $text, &$parser ){ + if(is_object($parser->getTitle() ) && $parser->getTitle()->isTalkPage()) + $this->mDoContentConvert=false; + else + $this->mDoContentConvert=true; + + return parent::parserConvert($text, $parser ); + } + + /* + * A function wrapper: + * - if there is no selected variant, leave the link + * names as they were + * - do not try to find variants for usernames + */ + function findVariantLink( &$link, &$nt ) { + // check for user namespace + if(is_object($nt)){ + $ns = $nt->getNamespace(); + if($ns==NS_USER || $ns==NS_USER_TALK) + return; + } + + $oldlink=$link; + parent::findVariantLink($link,$nt); + if($this->getPreferredVariant()==$this->mMainLanguageCode) + $link=$oldlink; + } + + /* + * An ugly function wrapper for parsing Image titles + * (to prevent image name conversion) + */ + function autoConvert($text, $toVariant=false) { + global $wgTitle; + if(is_object($wgTitle) && $wgTitle->getNameSpace()==NS_IMAGE){ + $imagename = $wgTitle->getNsText(); + if(preg_match("/^$imagename:/",$text)) return $text; + } + return parent::autoConvert($text,$toVariant); + } + + /** + * It translates text into variant + */ + function translate( $text, $toVariant ){ + global $wgContLanguageCode; + $text = parent::translate( $text, $toVariant ); + + $letters = ''; + switch( $toVariant ) { + case 'kk-cyrl': + case 'kk-kz': + $letters = KK_L_UC . KK_L_LC . 'ʺʹ#0123456789'; + $wgContLanguageCode = 'kk'; + break; + case 'kk-latn': + case 'kk-tr': + $letters = KK_C_UC . KK_C_LC . '№0123456789'; + $wgContLanguageCode = 'kk-Latn'; + break; + case 'kk-arab': + case 'kk-cn': + $letters = KK_C_UC.KK_C_LC./*KK_L_UC.KK_L_LC.'ʺʹ'.*/',;\?%\*№0123456789'; + $wgContLanguageCode = 'kk-Arab'; + break; + default: + $wgContLanguageCode = 'kk'; + return $text; + } + // disable conversion variables like $1, $2... + $varsfix = '\$[0-9]'; + + $matches = preg_split( '/' . $varsfix . '[^' . $letters . ']+/u', $text, -1, PREG_SPLIT_OFFSET_CAPTURE); + $mstart = 0; + $ret = ''; + foreach( $matches as $m ) { + $ret .= substr( $text, $mstart, $m[1]-$mstart ); + $ret .= $this->regsConverter( $m[0], $toVariant ); + $mstart = $m[1] + strlen($m[0]); + } return $ret; } + function regsConverter( $text, $toVariant ) { + if ($text == '') return $text; + + $pat = array(); + $rep = array(); + switch( $toVariant ) { + case 'kk-arab': + case 'kk-cn': + $letters = KK_C_LC.KK_C_UC/*.KK_L_LC.KK_L_UC*/; + $front = 'әөүіӘӨҮІ'/*.'äöüiÄÖÜİ'*/; + $excludes = 'еэгғкқЕЭГҒКҚ'/*.'eégğkqEÉGĞKQ'*/; + // split text to words + $matches = preg_split( '/[\b\s\-\.:]+/', $text, -1, PREG_SPLIT_OFFSET_CAPTURE); + $mstart = 0; + $ret = ''; + foreach( $matches as $m ) { + $ret .= substr( $text, $mstart, $m[1] - $mstart ); + // is matched the word to front vowels? + // exclude a words matched to е, э, г, к, к, қ, + // them should be without hamza + if ( preg_match('/['.$front.']/u', $m[0]) && !preg_match('/['.$excludes.']/u', $m[0]) ) { + $ret .= preg_replace('/['.$letters.']+/u', H_HAMZA.'$0', $m[0]); + } else { + $ret .= $m[0]; + } + $mstart = $m[1] + strlen($m[0]); + } + $text =& $ret; + foreach( $this->mCyLa2Arab as $pat => $rep ) { + $text = preg_replace( $pat, $rep, $text ); + } + return $text; + break; + case 'kk-latn': + case 'kk-tr': + foreach( $this->mCyrl2Latn as $pat => $rep ) { + $text = preg_replace( $pat, $rep, $text ); + } + return $text; + break; + case 'kk-cyrl': + case 'kk-kz': + foreach( $this->mLatn2Cyrl as $pat => $rep ) { + $text = preg_replace( $pat, $rep, $text ); + } + return $text; + break; + default: + return $text; + } + } + + /* + * We want our external link captions to be converted in variants, + * so we return the original text instead -{$text}-, except for URLs + */ + function markNoConversion( $text, $noParse=false ) { + if( $noParse || preg_match( "/^https?:\/\/|ftp:\/\/|irc:\/\//", $text ) ) + return parent::markNoConversion( $text ); + return $text; + } + + function convertCategoryKey( $key ) { + return $this->autoConvert( $key, 'kk' ); + } + } -class LanguageKk extends LanguageKk_kz { +/* class that handles Cyrillic, Latin and Arabic scripts for Kazakh + right now it only distinguish kk_cyrl, kk_latn, kk_arab and kk_kz, kk_tr, kk_cn. +*/ +class LanguageKk extends LanguageKk_cyrl { - function __construct() { - global $wgHooks; + function __construct() { + global $wgHooks; parent::__construct(); - $variants = array('kk', 'kk-kz', 'kk-tr', 'kk-cn'); - $variantfallbacks = array( - 'kk' => 'kk-kz', - 'kk-kz' => 'kk-kz', - 'kk-tr' => 'kk-tr', - 'kk-cn' => 'kk-cn' - ); + $variants = array( 'kk', 'kk-cyrl', 'kk-latn', 'kk-arab', 'kk-kz', 'kk-tr', 'kk-cn' ); + $variantfallbacks = array( + 'kk' => 'kk-kz', + 'kk-cyrl' => 'kk', + 'kk-latn' => 'kk', + 'kk-arab' => 'kk', + 'kk-kz' => 'kk-cyrl', + 'kk-tr' => 'kk-latn', + 'kk-cn' => 'kk-arab' + ); + + $this->mConverter = new KkConverter( $this, 'kk', $variants, $variantfallbacks ); + + $wgHooks['ArticleSaveComplete'][] = $this->mConverter; + } + + /** + * Work around for right-to-left direction support in kk-arab and kk-cn + * + * @return bool + */ + function isRTL() { + $variant = $this->getPreferredVariant(); + if ( $variant == 'kk-arab' || $variant == 'kk-cn' ) { + return true; + } else { + return parent::isRTL(); + } + } - $this->mConverter = new KkConverter( $this, 'kk', $variants, $variantfallbacks ); - $wgHooks['ArticleSaveComplete'][] = $this->mConverter; - } + /* + * It fixes issue with ucfirst for transforming 'i' to 'Ä°' + * + */ + function ucfirst ( $string ) { + $variant = $this->getPreferredVariant(); + if ( ($variant == 'kk-latn' || $variant == 'kk-tr') && $string[0] == 'i' ) { + $string = 'Ä°' . substr( $string, 1 ); + } else { + $string = parent::ucfirst( $string ); + } + return $string; + } + + /* + * It fixes issue with lcfirst for transforming 'I' to 'ı' + * + */ + function lcfirst ( $string ) { + $variant = $this->getPreferredVariant(); + if ( ($variant == 'kk-latn' || $variant == 'kk-tr') && $string[0] == 'I' ) { + $string = 'ı' . substr( $string, 1 ); + } else { + $string = parent::lcfirst( $string ); + } + return $string; + } function convertGrammar( $word, $case ) { - $fname="LanguageKk::convertGrammar"; - wfProfileIn( $fname ); - - //always convert to kk-kz before convertGrammar - $w1 = $word; - $word = $this->mConverter->autoConvert($word, 'kk-kz'); - $w2 = $word; - $word = parent::convertGrammar( $word, $case ); - //restore encoding - if( $w1 != $w2 ) { - $word = $this->mConverter->translate($word, 'kk-tr'); - } - wfProfileOut( $fname ); + wfProfileIn( __METHOD__ ); + + $variant = $this->getPreferredVariant(); + switch ( $variant ) { + case 'kk-arab': + case 'kk-cn': + $word = parent::convertGrammarKk_arab( $word, $case ); + break; + case 'kk-latn': + case 'kk-tr': + $word = parent::convertGrammarKk_latn( $word, $case ); + break; + case 'kk-cyrl': + case 'kk-kz': + case 'kk': + default: + $word = parent::convertGrammarKk_cyrl( $word, $case ); + } + + wfProfileOut( __METHOD__ ); return $word; } } - -?>