54badce2d8
* Rename wordSegmentation() to segmentByWord(). * Consolidate search index locking and iteration to Maintenance.php * Add maintenance/updateDoubleWidthSearch.php to take care of new format for normalized double-width roman characters. * Add error checking to updateSearchIndex.php for creating $posFile. * Add note to UPGRADE about running updateDoubleWidthSearch.php.
32 lines
684 B
PHP
32 lines
684 B
PHP
<?php
|
|
/**
|
|
* @ingroup Language
|
|
*/
|
|
class LanguageYue extends Language {
|
|
function hasWordBreaks() {
|
|
return false;
|
|
}
|
|
|
|
/**
|
|
* Eventually this should be a word segmentation;
|
|
* for now just treat each character as a word.
|
|
* @todo Fixme: only do this for Han characters...
|
|
*/
|
|
function segmentByWord( $string ) {
|
|
$reg = "/([\\xc0-\\xff][\\x80-\\xbf]*)/";
|
|
$s = self::insertSpace( $string, $reg );
|
|
return $s;
|
|
}
|
|
|
|
function normalizeForSearch( $string ) {
|
|
wfProfileIn( __METHOD__ );
|
|
|
|
// Double-width roman characters
|
|
$s = self::convertDoubleWidth( $string );
|
|
$s = trim( $s );
|
|
$s = parent::normalizeForSearch( $s );
|
|
|
|
wfProfileOut( __METHOD__ );
|
|
return $s;
|
|
}
|
|
}
|