languages/classes/LanguageZh.php

   1 <?php
   2 /**
   3  * Chinese specific code.
   4  *
   5  * This program is free software; you can redistribute it and/or modify
   6  * it under the terms of the GNU General Public License as published by
   7  * the Free Software Foundation; either version 2 of the License, or
   8  * (at your option) any later version.
   9  *
  10  * This program is distributed in the hope that it will be useful,
  11  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  12  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
  13  * GNU General Public License for more details.
  14  *
  15  * You should have received a copy of the GNU General Public License along
  16  * with this program; if not, write to the Free Software Foundation, Inc.,
  17  * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
  18  * http://www.gnu.org/copyleft/gpl.html
  19  *
  20  * @file
  21  * @ingroup Language
  22  */
  23
  24 /**
  25  * @ingroup Language
  26  */
  27 class ZhConverter extends LanguageConverter {
  28         /**
  29          * @param Language $langobj
  30          * @param string $maincode
  31          * @param array $variants
  32          * @param array $variantfallbacks
  33          * @param array $flags
  34          * @param array $manualLevel
  35          */
  36         function __construct( $langobj, $maincode,
  37                                                                 $variants = [],
  38                                                                 $variantfallbacks = [],
  39                                                                 $flags = [],
  40                                                                 $manualLevel = [] ) {
  41                 $this->mDescCodeSep = '：';
  42                 $this->mDescVarSep = '；';
  43                 parent::__construct( $langobj, $maincode,
  44                                                                         $variants,
  45                                                                         $variantfallbacks,
  46                                                                         $flags,
  47                                                                         $manualLevel );
  48                 $names = [
  49                         'zh' => '原文',
  50                         'zh-hans' => '简体',
  51                         'zh-hant' => '繁體',
  52                         'zh-cn' => '大陆',
  53                         'zh-tw' => '台灣',
  54                         'zh-hk' => '香港',
  55                         'zh-mo' => '澳門',
  56                         'zh-sg' => '新加坡',
  57                         'zh-my' => '大马',
  58                 ];
  59                 $this->mVariantNames = array_merge( $this->mVariantNames, $names );
  60         }
  61
  62         function loadDefaultTables() {
  63                 $this->mTables = [
  64                         'zh-hans' => new ReplacementArray( MediaWiki\Languages\Data\ZhConversion::$zh2Hans ),
  65                         'zh-hant' => new ReplacementArray( MediaWiki\Languages\Data\ZhConversion::$zh2Hant ),
  66                         'zh-cn' => new ReplacementArray( MediaWiki\Languages\Data\ZhConversion::$zh2CN ),
  67                         'zh-hk' => new ReplacementArray( MediaWiki\Languages\Data\ZhConversion::$zh2HK ),
  68                         'zh-mo' => new ReplacementArray( MediaWiki\Languages\Data\ZhConversion::$zh2HK ),
  69                         'zh-my' => new ReplacementArray( MediaWiki\Languages\Data\ZhConversion::$zh2CN ),
  70                         'zh-sg' => new ReplacementArray( MediaWiki\Languages\Data\ZhConversion::$zh2CN ),
  71                         'zh-tw' => new ReplacementArray( MediaWiki\Languages\Data\ZhConversion::$zh2TW ),
  72                         'zh' => new ReplacementArray
  73                 ];
  74         }
  75
  76         function postLoadTables() {
  77                 $this->mTables['zh-cn']->setArray(
  78                         $this->mTables['zh-cn']->getArray() + $this->mTables['zh-hans']->getArray()
  79                 );
  80                 $this->mTables['zh-hk']->setArray(
  81                         $this->mTables['zh-hk']->getArray() + $this->mTables['zh-hant']->getArray()
  82                 );
  83                 $this->mTables['zh-mo']->setArray(
  84                         $this->mTables['zh-mo']->getArray() + $this->mTables['zh-hant']->getArray()
  85                 );
  86                 $this->mTables['zh-my']->setArray(
  87                         $this->mTables['zh-my']->getArray() + $this->mTables['zh-hans']->getArray()
  88                 );
  89                 $this->mTables['zh-sg']->setArray(
  90                         $this->mTables['zh-sg']->getArray() + $this->mTables['zh-hans']->getArray()
  91                 );
  92                 $this->mTables['zh-tw']->setArray(
  93                         $this->mTables['zh-tw']->getArray() + $this->mTables['zh-hant']->getArray()
  94                 );
  95         }
  96
  97         /**
  98          * @param string $key
  99          * @return string
 100          */
 101         function convertCategoryKey( $key ) {
 102                 return $this->autoConvert( $key, 'zh' );
 103         }
 104 }
 105
 106 /**
 107  * class that handles both Traditional and Simplified Chinese
 108  * right now it only distinguish zh_hans, zh_hant, zh_cn, zh_tw, zh_sg and zh_hk.
 109  *
 110  * @ingroup Language
 111  */
 112 class LanguageZh extends LanguageZh_hans {
 113         function __construct() {
 114                 parent::__construct();
 115
 116                 $variants = [
 117                         'zh',
 118                         'zh-hans',
 119                         'zh-hant',
 120                         'zh-cn',
 121                         'zh-hk',
 122                         'zh-mo',
 123                         'zh-my',
 124                         'zh-sg',
 125                         'zh-tw'
 126                 ];
 127
 128                 $variantfallbacks = [
 129                         'zh' => [ 'zh-hans', 'zh-hant', 'zh-cn', 'zh-tw', 'zh-hk', 'zh-sg', 'zh-mo', 'zh-my' ],
 130                         'zh-hans' => [ 'zh-cn', 'zh-sg', 'zh-my' ],
 131                         'zh-hant' => [ 'zh-tw', 'zh-hk', 'zh-mo' ],
 132                         'zh-cn' => [ 'zh-hans', 'zh-sg', 'zh-my' ],
 133                         'zh-sg' => [ 'zh-hans', 'zh-cn', 'zh-my' ],
 134                         'zh-my' => [ 'zh-hans', 'zh-sg', 'zh-cn' ],
 135                         'zh-tw' => [ 'zh-hant', 'zh-hk', 'zh-mo' ],
 136                         'zh-hk' => [ 'zh-hant', 'zh-mo', 'zh-tw' ],
 137                         'zh-mo' => [ 'zh-hant', 'zh-hk', 'zh-tw' ],
 138                 ];
 139                 $ml = [
 140                         'zh' => 'disable',
 141                         'zh-hans' => 'unidirectional',
 142                         'zh-hant' => 'unidirectional',
 143                 ];
 144
 145                 $this->mConverter = new ZhConverter( $this, 'zh',
 146                                                                 $variants, $variantfallbacks,
 147                                                                 [],
 148                                                                 $ml );
 149         }
 150
 151         /**
 152          * this should give much better diff info
 153          *
 154          * @param string $text
 155          * @return string
 156          */
 157         function segmentForDiff( $text ) {
 158                 return preg_replace( '/[\xc0-\xff][\x80-\xbf]*/', ' $0', $text );
 159         }
 160
 161         /**
 162          * @param string $text
 163          * @return string
 164          */
 165         function unsegmentForDiff( $text ) {
 166                 return preg_replace( '/ ([\xc0-\xff][\x80-\xbf]*)/', '$1', $text );
 167         }
 168
 169         /**
 170          * auto convert to zh-hans and normalize special characters.
 171          *
 172          * @param string $string
 173          * @param string $autoVariant Defaults to 'zh-hans'
 174          * @return string
 175          */
 176         function normalizeForSearch( $string, $autoVariant = 'zh-hans' ) {
 177                 // always convert to zh-hans before indexing. it should be
 178                 // better to use zh-hans for search, since conversion from
 179                 // Traditional to Simplified is less ambiguous than the
 180                 // other way around
 181                 $s = $this->mConverter->autoConvert( $string, $autoVariant );
 182                 // LanguageZh_hans::normalizeForSearch
 183                 $s = parent::normalizeForSearch( $s );
 184                 return $s;
 185         }
 186
 187         /**
 188          * @param array $termsArray
 189          * @return array
 190          */
 191         function convertForSearchResult( $termsArray ) {
 192                 $terms = implode( '|', $termsArray );
 193                 $terms = self::convertDoubleWidth( $terms );
 194                 $terms = implode( '|', $this->mConverter->autoConvertToAllVariants( $terms ) );
 195                 $ret = array_unique( explode( '|', $terms ) );
 196                 return $ret;
 197         }
 198 }