7 class LanguageJa
extends Language
{
8 function stripForSearch( $string ) {
9 # MySQL fulltext index doesn't grok utf-8, so we
10 # need to fold cases and convert to hex
13 # Strip known punctuation ?
14 #$s = preg_replace( '/\xe3\x80[\x80-\xbf]/', '', $s ); # U3000-303f
16 # Space strings of like hiragana/katakana/kanji
17 $hiragana = '(?:\xe3(?:\x81[\x80-\xbf]|\x82[\x80-\x9f]))'; # U3040-309f
18 $katakana = '(?:\xe3(?:\x82[\xa0-\xbf]|\x83[\x80-\xbf]))'; # U30a0-30ff
19 $kanji = '(?:\xe3[\x88-\xbf][\x80-\xbf]'
20 . '|[\xe4-\xe8][\x80-\xbf]{2}'
21 . '|\xe9[\x80-\xa5][\x80-\xbf]'
22 . '|\xe9\xa6[\x80-\x99])';
23 # U3200-9999 = \xe3\x88\x80-\xe9\xa6\x99
24 $s = preg_replace( "/({$hiragana}+|{$katakana}+|{$kanji}+)/", ' $1 ', $s );
26 # Double-width roman characters: ff00-ff5f ~= 0020-007f
27 $s = preg_replace( '/\xef\xbc([\x80-\xbf])/e', 'chr((ord("$1") & 0x3f) + 0x20)', $s );
28 $s = preg_replace( '/\xef\xbd([\x80-\x99])/e', 'chr((ord("$1") & 0x3f) + 0x60)', $s );
30 # Do general case folding and UTF-8 armoring
31 return parent
::stripForSearch( $s );
34 # Italic is not appropriate for Japanese script
35 # Unfortunately most browsers do not recognise this, and render <em> as italic
36 function emphasize( $text ) {