X-Git-Url: https://scripts.mit.edu/gitweb/autoinstallsdev/mediawiki.git/blobdiff_plain/19e297c21b10b1b8a3acad5e73fc71dcb35db44a..6932310fd58ebef145fa01eb76edf7150284d8ea:/languages/classes/LanguageKk.php diff --git a/languages/classes/LanguageKk.php b/languages/classes/LanguageKk.php index e3caacc8..1a1438c6 100644 --- a/languages/classes/LanguageKk.php +++ b/languages/classes/LanguageKk.php @@ -1,7 +1,25 @@ mTables = array( + $kk2Cyrl = []; + $kk2Latn = []; + $kk2Arab = []; + $kk2KZ = []; + $kk2TR = []; + $kk2CN = []; + + $this->mTables = [ '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() - ); + '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() + ]; } function postLoadTables() { @@ -60,8 +86,7 @@ class KkConverter extends LanguageConverter { } function loadRegs() { - - $this->mCyrl2Latn = array( + $this->mCyrl2Latn = [ # # Punctuation '/№/u' => 'No.', # # Е after vowels @@ -98,13 +123,13 @@ class KkConverter extends LanguageConverter { '/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', - ); + ]; - $this->mLatn2Cyrl = array( + $this->mLatn2Cyrl = [ # # Punctuation '/#|No\./' => '№', # # Şç - '/ŞÇʹ/u' => 'ЩЬ', '/Şçʹ/u' => 'Щь', '/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', @@ -137,15 +162,15 @@ class KkConverter extends LanguageConverter { '/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', - ); + ]; - $this->mCyLa2Arab = array( + $this->mCyLa2Arab = [ # # Punctuation -> Arabic '/#|№|No\./u' => '؀', # ؀ '/\,/' => '،', # ، - '/;/' => '؛', # ؛ + '/;/' => '؛', # ؛ '/\?/' => '؟', # ؟ - '/%/' => 'Ùª', # ٪ + '/%/' => 'Ùª', # ٪ '/\*/' => 'Ù­', # ٭ # # Digits -> Arabic '/0/' => 'Û°', # ۰ @@ -187,112 +212,100 @@ class KkConverter extends LanguageConverter { '/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',*/ - ); - } - - /* 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 $carray; + ]; } - /* + /** * A function wrapper: * - if there is no selected variant, leave the link * names as they were * - do not try to find variants for usernames + * + * @param string &$link + * @param Title &$nt + * @param bool $ignoreOtherCond */ function findVariantLink( &$link, &$nt, $ignoreOtherCond = false ) { // check for user namespace if ( is_object( $nt ) ) { $ns = $nt->getNamespace(); - if ( $ns == NS_USER || $ns == NS_USER_TALK ) + if ( $ns == NS_USER || $ns == NS_USER_TALK ) { return; + } } $oldlink = $link; parent::findVariantLink( $link, $nt, $ignoreOtherCond ); - if ( $this->getPreferredVariant() == $this->mMainLanguageCode ) + 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_FILE ) { - $imagename = $wgTitle->getNsText(); - if ( preg_match( "/^$imagename:/", $text ) ) return $text; } - return parent::autoConvert( $text, $toVariant ); } /** * It translates text into variant + * + * @param string $text + * @param string $toVariant + * + * @return string */ function translate( $text, $toVariant ) { - global $wgLanguageCode; $text = parent::translate( $text, $toVariant ); - switch( $toVariant ) { + switch ( $toVariant ) { case 'kk-cyrl': case 'kk-kz': $letters = KK_L_UC . KK_L_LC . 'ʺʹ#0123456789'; - $wgLanguageCode = 'kk'; break; case 'kk-latn': case 'kk-tr': $letters = KK_C_UC . KK_C_LC . '№0123456789'; - $wgLanguageCode = 'kk-Latn'; break; case 'kk-arab': case 'kk-cn': $letters = KK_C_UC . KK_C_LC . /*KK_L_UC.KK_L_LC.'ʺʹ'.*/',;\?%\*№0123456789'; - $wgLanguageCode = 'kk-Arab'; break; default: - $wgLanguageCode = 'kk'; return $text; } // disable conversion variables like $1, $2... $varsfix = '\$[0-9]'; - $matches = preg_split( '/' . $varsfix . '[^' . $letters . ']+/u', $text, -1, PREG_SPLIT_OFFSET_CAPTURE ); + $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 .= substr( $text, $mstart, $m[1] - $mstart ); $ret .= $this->regsConverter( $m[0], $toVariant ); $mstart = $m[1] + strlen( $m[0] ); } + return $ret; } + /** + * @param string $text + * @param string $toVariant + * @return mixed|string + */ function regsConverter( $text, $toVariant ) { if ( $text == '' ) { return $text; } - switch( $toVariant ) { + 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'*/; + $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; @@ -302,7 +315,9 @@ class KkConverter extends LanguageConverter { // 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] ) ) { + 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 .= $m[0]; @@ -334,20 +349,13 @@ class KkConverter extends LanguageConverter { } } - /* - * We want our external link captions to be converted in variants, - * so we return the original text instead -{$text}-, except for URLs + /** + * @param string $key + * @return string */ - 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' ); } - } /** @@ -357,72 +365,63 @@ class KkConverter extends LanguageConverter { * @ingroup Language */ class LanguageKk extends LanguageKk_cyrl { - function __construct() { - global $wgHooks; parent::__construct(); - $variants = array( 'kk', 'kk-cyrl', 'kk-latn', 'kk-arab', 'kk-kz', 'kk-tr', 'kk-cn' ); - $variantfallbacks = array( - 'kk' => 'kk-cyrl', + $variants = [ 'kk', 'kk-cyrl', 'kk-latn', 'kk-arab', 'kk-kz', 'kk-tr', 'kk-cn' ]; + $variantfallbacks = [ + 'kk' => 'kk-cyrl', 'kk-cyrl' => 'kk', 'kk-latn' => 'kk', 'kk-arab' => 'kk', - 'kk-kz' => 'kk-cyrl', - 'kk-tr' => 'kk-latn', - 'kk-cn' => 'kk-arab' - ); + '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(); - } - } - - /* * It fixes issue with ucfirst for transforming 'i' to 'Ä°' * + * @param string $string + * + * @return string */ - 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 ); + public function ucfirst( $string ) { + if ( substr( $string, 0, 1 ) === 'i' ) { + $variant = $this->getPreferredVariant(); + if ( $variant == 'kk-latn' || $variant == 'kk-tr' ) { + return 'Ä°' . substr( $string, 1 ); + } } - return $string; + return parent::ucfirst( $string ); } - /* + /** * It fixes issue with lcfirst for transforming 'I' to 'ı' * + * @param string $string + * + * @return string */ - 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 ); + function lcfirst( $string ) { + if ( substr( $string, 0, 1 ) === 'I' ) { + $variant = $this->getPreferredVariant(); + if ( $variant == 'kk-latn' || $variant == 'kk-tr' ) { + return 'ı' . substr( $string, 1 ); + } } - return $string; + return parent::lcfirst( $string ); } + /** + * @param string $word + * @param string $case + * @return string + */ function convertGrammar( $word, $case ) { - wfProfileIn( __METHOD__ ); - $variant = $this->getPreferredVariant(); switch ( $variant ) { case 'kk-arab': @@ -440,7 +439,6 @@ class LanguageKk extends LanguageKk_cyrl { $word = parent::convertGrammarKk_cyrl( $word, $case ); } - wfProfileOut( __METHOD__ ); return $word; } }