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 require_once( __DIR__ . '/../LanguageConverter.php' );
  25 require_once( __DIR__ . '/LanguageZh_hans.php' );
  26
  27 /**
  28  * @ingroup Language
  29  */
  30 class ZhConverter extends LanguageConverter {
  31
  32         /**
  33          * @param $langobj Language
  34          * @param $maincode string
  35          * @param $variants array
  36          * @param $variantfallbacks array
  37          * @param $flags array
  38          * @param $manualLevel array
  39          */
  40         function __construct( $langobj, $maincode,
  41                                                                 $variants = array(),
  42                                                                 $variantfallbacks = array(),
  43                                                                 $flags = array(),
  44                                                                 $manualLevel = array() ) {
  45                 $this->mDescCodeSep = '：';
  46                 $this->mDescVarSep = '；';
  47                 parent::__construct( $langobj, $maincode,
  48                                                                         $variants,
  49                                                                         $variantfallbacks,
  50                                                                         $flags,
  51                                                                         $manualLevel );
  52                 $names = array(
  53                         'zh'      => '原文',
  54                         'zh-hans' => '简体',
  55                         'zh-hant' => '繁體',
  56                         'zh-cn'   => '大陆',
  57                         'zh-tw'   => '台灣',
  58                         'zh-hk'   => '香港',
  59                         'zh-mo'   => '澳門',
  60                         'zh-sg'   => '新加坡',
  61                         'zh-my'   => '大马',
  62                 );
  63                 $this->mVariantNames = array_merge( $this->mVariantNames, $names );
  64         }
  65
  66         function loadDefaultTables() {
  67                 require( __DIR__ . "/../../includes/ZhConversion.php" );
  68                 $this->mTables = array(
  69                         'zh-hans' => new ReplacementArray( $zh2Hans ),
  70                         'zh-hant' => new ReplacementArray( $zh2Hant ),
  71                         'zh-cn'   => new ReplacementArray( array_merge( $zh2Hans, $zh2CN ) ),
  72                         'zh-hk'   => new ReplacementArray( array_merge( $zh2Hant, $zh2HK ) ),
  73                         'zh-mo'   => new ReplacementArray( array_merge( $zh2Hant, $zh2HK ) ),
  74                         'zh-my'   => new ReplacementArray( array_merge( $zh2Hans, $zh2SG ) ),
  75                         'zh-sg'   => new ReplacementArray( array_merge( $zh2Hans, $zh2SG ) ),
  76                         'zh-tw'   => new ReplacementArray( array_merge( $zh2Hant, $zh2TW ) ),
  77                         'zh'      => new ReplacementArray
  78                 );
  79         }
  80
  81         function postLoadTables() {
  82                 $this->mTables['zh-cn']->merge( $this->mTables['zh-hans'] );
  83                 $this->mTables['zh-hk']->merge( $this->mTables['zh-hant'] );
  84                 $this->mTables['zh-mo']->merge( $this->mTables['zh-hant'] );
  85                 $this->mTables['zh-my']->merge( $this->mTables['zh-hans'] );
  86                 $this->mTables['zh-sg']->merge( $this->mTables['zh-hans'] );
  87                 $this->mTables['zh-tw']->merge( $this->mTables['zh-hant'] );
  88         }
  89
  90         /**
  91          * there shouldn't be any latin text in Chinese conversion, so no need
  92          * to mark anything.
  93          * $noParse is there for compatibility with LanguageConvert::markNoConversion
  94          *
  95          * @param $text string
  96          * @param $noParse bool
  97          *
  98          * @return string
  99          */
 100         function markNoConversion( $text, $noParse = false ) {
 101                 return $text;
 102         }
 103
 104         /**
 105          * @param $key string
 106          * @return String
 107          */
 108         function convertCategoryKey( $key ) {
 109                 return $this->autoConvert( $key, 'zh' );
 110         }
 111 }
 112
 113 /**
 114  * class that handles both Traditional and Simplified Chinese
 115  * right now it only distinguish zh_hans, zh_hant, zh_cn, zh_tw, zh_sg and zh_hk.
 116  *
 117  * @ingroup Language
 118  */
 119 class LanguageZh extends LanguageZh_hans {
 120
 121         function __construct() {
 122                 global $wgHooks;
 123                 parent::__construct();
 124
 125                 $variants = array( 'zh', 'zh-hans', 'zh-hant', 'zh-cn', 'zh-hk', 'zh-mo', 'zh-my', 'zh-sg', 'zh-tw' );
 126
 127                 $variantfallbacks = array(
 128                         'zh'      => array( 'zh-hans', 'zh-hant', 'zh-cn', 'zh-tw', 'zh-hk', 'zh-sg', 'zh-mo', 'zh-my' ),
 129                         'zh-hans' => array( 'zh-cn', 'zh-sg', 'zh-my' ),
 130                         'zh-hant' => array( 'zh-tw', 'zh-hk', 'zh-mo' ),
 131                         'zh-cn'   => array( 'zh-hans', 'zh-sg', 'zh-my' ),
 132                         'zh-sg'   => array( 'zh-hans', 'zh-cn', 'zh-my' ),
 133                         'zh-my'   => array( 'zh-hans', 'zh-sg', 'zh-cn' ),
 134                         'zh-tw'   => array( 'zh-hant', 'zh-hk', 'zh-mo' ),
 135                         'zh-hk'   => array( 'zh-hant', 'zh-mo', 'zh-tw' ),
 136                         'zh-mo'   => array( 'zh-hant', 'zh-hk', 'zh-tw' ),
 137                 );
 138                 $ml = array(
 139                         'zh'      => 'disable',
 140                         'zh-hans' => 'unidirectional',
 141                         'zh-hant' => 'unidirectional',
 142                 );
 143
 144                 $this->mConverter = new ZhConverter( $this, 'zh',
 145                                                                 $variants, $variantfallbacks,
 146                                                                 array(),
 147                                                                 $ml );
 148
 149                 $wgHooks['PageContentSaveComplete'][] = $this->mConverter;
 150         }
 151
 152         /**
 153          * this should give much better diff info
 154          *
 155          * @param $text string
 156          * @return string
 157          */
 158         function segmentForDiff( $text ) {
 159                 return preg_replace(
 160                         "/([\\xc0-\\xff][\\x80-\\xbf]*)/e",
 161                         "' ' .\"$1\"", $text );
 162         }
 163
 164         /**
 165          * @param $text string
 166          * @return string
 167          */
 168         function unsegmentForDiff( $text ) {
 169                 return preg_replace(
 170                         "/ ([\\xc0-\\xff][\\x80-\\xbf]*)/e",
 171                         "\"$1\"", $text );
 172         }
 173
 174         /**
 175          * auto convert to zh-hans and normalize special characters.
 176          *
 177          * @param $string String
 178          * @param $autoVariant String, default to 'zh-hans'
 179          * @return String
 180          */
 181         function normalizeForSearch( $string, $autoVariant = 'zh-hans' ) {
 182                 wfProfileIn( __METHOD__ );
 183
 184                 // always convert to zh-hans before indexing. it should be
 185                 // better to use zh-hans for search, since conversion from
 186                 // Traditional to Simplified is less ambiguous than the
 187                 // other way around
 188                 $s = $this->mConverter->autoConvert( $string, $autoVariant );
 189                 // LanguageZh_hans::normalizeForSearch
 190                 $s = parent::normalizeForSearch( $s );
 191                 wfProfileOut( __METHOD__ );
 192                 return $s;
 193
 194         }
 195
 196         /**
 197          * @param $termsArray array
 198          * @return array
 199          */
 200         function convertForSearchResult( $termsArray ) {
 201                 $terms = implode( '|', $termsArray );
 202                 $terms = self::convertDoubleWidth( $terms );
 203                 $terms = implode( '|', $this->mConverter->autoConvertToAllVariants( $terms ) );
 204                 $ret = array_unique( explode( '|', $terms ) );
 205                 return $ret;
 206         }
 207 }
 208