<?php
-require_once( dirname(__FILE__).'/../LanguageConverter.php' );
-require_once( dirname(__FILE__).'/LanguageZh_hans.php' );
+require_once( dirname( __FILE__ ) . '/../LanguageConverter.php' );
+require_once( dirname( __FILE__ ) . '/LanguageZh_hans.php' );
/**
* @ingroup Language
*/
class ZhConverter extends LanguageConverter {
- function __construct($langobj, $maincode,
- $variants=array(),
- $variantfallbacks=array(),
- $markup=array(),
+ /**
+ * @param $langobj Language
+ * @param $maincode string
+ * @param $variants array
+ * @param $variantfallbacks array
+ * @param $flags array
+ * @param $manualLevel array
+ */
+ function __construct( $langobj, $maincode,
+ $variants = array(),
+ $variantfallbacks = array(),
$flags = array(),
$manualLevel = array() ) {
$this->mDescCodeSep = ':';
$this->mDescVarSep = ';';
- parent::__construct($langobj, $maincode,
+ parent::__construct( $langobj, $maincode,
$variants,
$variantfallbacks,
- $markup,
$flags,
- $manualLevel);
+ $manualLevel );
$names = array(
'zh' => '原文',
'zh-hans' => '简体',
'zh-sg' => '新加坡',
'zh-my' => '大马',
);
- $this->mLanguageNames = array_merge($this->mLanguageNames,$names);
+ $this->mVariantNames = array_merge( $this->mVariantNames, $names );
}
function loadDefaultTables() {
- require( dirname(__FILE__)."/../../includes/ZhConversion.php" );
+ require( dirname( __FILE__ ) . "/../../includes/ZhConversion.php" );
$this->mTables = array(
'zh-hans' => new ReplacementArray( $zh2Hans ),
'zh-hant' => new ReplacementArray( $zh2Hant ),
- 'zh-cn' => new ReplacementArray( array_merge($zh2Hans, $zh2CN) ),
- 'zh-hk' => new ReplacementArray( array_merge($zh2Hant, $zh2HK) ),
- 'zh-mo' => new ReplacementArray( array_merge($zh2Hant, $zh2HK) ),
- 'zh-my' => new ReplacementArray( array_merge($zh2Hans, $zh2SG) ),
- 'zh-sg' => new ReplacementArray( array_merge($zh2Hans, $zh2SG) ),
- 'zh-tw' => new ReplacementArray( array_merge($zh2Hant, $zh2TW) ),
+ 'zh-cn' => new ReplacementArray( array_merge( $zh2Hans, $zh2CN ) ),
+ 'zh-hk' => new ReplacementArray( array_merge( $zh2Hant, $zh2HK ) ),
+ 'zh-mo' => new ReplacementArray( array_merge( $zh2Hant, $zh2HK ) ),
+ 'zh-my' => new ReplacementArray( array_merge( $zh2Hans, $zh2SG ) ),
+ 'zh-sg' => new ReplacementArray( array_merge( $zh2Hans, $zh2SG ) ),
+ 'zh-tw' => new ReplacementArray( array_merge( $zh2Hant, $zh2TW ) ),
'zh' => new ReplacementArray
);
}
$this->mTables['zh-tw']->merge( $this->mTables['zh-hant'] );
}
- /* there shouldn't be any latin text in Chinese conversion, so no need
- to mark anything.
- $noParse is there for compatibility with LanguageConvert::markNoConversion
+ /**
+ * there shouldn't be any latin text in Chinese conversion, so no need
+ * to mark anything.
+ * $noParse is there for compatibility with LanguageConvert::markNoConversion
+ *
+ * @param $text string
+ * @param $noParse bool
+ *
+ * @return string
*/
- function markNoConversion($text, $noParse = false) {
+ function markNoConversion( $text, $noParse = false ) {
return $text;
}
+ /**
+ * @param $key string
+ * @return String
+ */
function convertCategoryKey( $key ) {
return $this->autoConvert( $key, 'zh' );
}
global $wgHooks;
parent::__construct();
- $variants = array('zh','zh-hans','zh-hant','zh-cn','zh-hk','zh-mo','zh-my','zh-sg','zh-tw');
+ $variants = array( 'zh', 'zh-hans', 'zh-hant', 'zh-cn', 'zh-hk', 'zh-mo', 'zh-my', 'zh-sg', 'zh-tw' );
+
$variantfallbacks = array(
- 'zh' => array('zh-hans','zh-hant','zh-cn','zh-tw','zh-hk','zh-sg','zh-mo','zh-my'),
- 'zh-hans' => array('zh-cn','zh-sg','zh-my'),
- 'zh-hant' => array('zh-tw','zh-hk','zh-mo'),
- 'zh-cn' => array('zh-hans','zh-sg','zh-my'),
- 'zh-sg' => array('zh-hans','zh-cn','zh-my'),
- 'zh-my' => array('zh-hans','zh-sg','zh-cn'),
- 'zh-tw' => array('zh-hant','zh-hk','zh-mo'),
- 'zh-hk' => array('zh-hant','zh-mo','zh-tw'),
- 'zh-mo' => array('zh-hant','zh-hk','zh-tw'),
+ 'zh' => array( 'zh-hans', 'zh-hant', 'zh-cn', 'zh-tw', 'zh-hk', 'zh-sg', 'zh-mo', 'zh-my' ),
+ 'zh-hans' => array( 'zh-cn', 'zh-sg', 'zh-my' ),
+ 'zh-hant' => array( 'zh-tw', 'zh-hk', 'zh-mo' ),
+ 'zh-cn' => array( 'zh-hans', 'zh-sg', 'zh-my' ),
+ 'zh-sg' => array( 'zh-hans', 'zh-cn', 'zh-my' ),
+ 'zh-my' => array( 'zh-hans', 'zh-sg', 'zh-cn' ),
+ 'zh-tw' => array( 'zh-hant', 'zh-hk', 'zh-mo' ),
+ 'zh-hk' => array( 'zh-hant', 'zh-mo', 'zh-tw' ),
+ 'zh-mo' => array( 'zh-hant', 'zh-hk', 'zh-tw' ),
);
- $ml=array(
+ $ml = array(
'zh' => 'disable',
'zh-hans' => 'unidirectional',
'zh-hant' => 'unidirectional',
$this->mConverter = new ZhConverter( $this, 'zh',
$variants, $variantfallbacks,
- array(),array(),
- $ml);
+ array(),
+ $ml );
$wgHooks['ArticleSaveComplete'][] = $this->mConverter;
}
- # this should give much better diff info
+ /**
+ * this should give much better diff info
+ *
+ * @param $text string
+ * @return string
+ */
function segmentForDiff( $text ) {
return preg_replace(
"/([\\xc0-\\xff][\\x80-\\xbf]*)/e",
- "' ' .\"$1\"", $text);
+ "' ' .\"$1\"", $text );
}
+ /**
+ * @param $text string
+ * @return string
+ */
function unsegmentForDiff( $text ) {
return preg_replace(
"/ ([\\xc0-\\xff][\\x80-\\xbf]*)/e",
- "\"$1\"", $text);
+ "\"$1\"", $text );
}
- // word segmentation
- function stripForSearch( $string ) {
+ /**
+ * auto convert to zh-hans and normalize special characters.
+ *
+ * @param $string String
+ * @param $autoVariant String, default to 'zh-hans'
+ * @return String
+ */
+ function normalizeForSearch( $string, $autoVariant = 'zh-hans' ) {
wfProfileIn( __METHOD__ );
- // eventually this should be a word segmentation
- // for now just treat each character as a word
- // @fixme only do this for Han characters...
- $t = preg_replace(
- "/([\\xc0-\\xff][\\x80-\\xbf]*)/",
- " $1", $string);
-
- //always convert to zh-hans before indexing. it should be
- //better to use zh-hans for search, since conversion from
- //Traditional to Simplified is less ambiguous than the
- //other way around
-
- $t = $this->mConverter->autoConvert($t, 'zh-hans');
- $t = parent::stripForSearch( $t );
+ // always convert to zh-hans before indexing. it should be
+ // better to use zh-hans for search, since conversion from
+ // Traditional to Simplified is less ambiguous than the
+ // other way around
+ $s = $this->mConverter->autoConvert( $string, $autoVariant );
+ // LanguageZh_hans::normalizeForSearch
+ $s = parent::normalizeForSearch( $s );
wfProfileOut( __METHOD__ );
- return $t;
+ return $s;
}
+ /**
+ * @param $termsArray array
+ * @return array
+ */
function convertForSearchResult( $termsArray ) {
$terms = implode( '|', $termsArray );
+ $terms = self::convertDoubleWidth( $terms );
$terms = implode( '|', $this->mConverter->autoConvertToAllVariants( $terms ) );
- $ret = array_unique( explode('|', $terms) );
+ $ret = array_unique( explode( '|', $terms ) );
return $ret;
}
}