00001 <?php 00002 00008 class LanguageJa extends Language { 00009 function stripForSearch( $string ) { 00010 # MySQL fulltext index doesn't grok utf-8, so we 00011 # need to fold cases and convert to hex 00012 $s = $string; 00013 00014 # Strip known punctuation ? 00015 #$s = preg_replace( '/\xe3\x80[\x80-\xbf]/', '', $s ); # U3000-303f 00016 00017 # Space strings of like hiragana/katakana/kanji 00018 $hiragana = '(?:\xe3(?:\x81[\x80-\xbf]|\x82[\x80-\x9f]))'; # U3040-309f 00019 $katakana = '(?:\xe3(?:\x82[\xa0-\xbf]|\x83[\x80-\xbf]))'; # U30a0-30ff 00020 $kanji = '(?:\xe3[\x88-\xbf][\x80-\xbf]' 00021 . '|[\xe4-\xe8][\x80-\xbf]{2}' 00022 . '|\xe9[\x80-\xa5][\x80-\xbf]' 00023 . '|\xe9\xa6[\x80-\x99])'; 00024 # U3200-9999 = \xe3\x88\x80-\xe9\xa6\x99 00025 $s = preg_replace( "/({$hiragana}+|{$katakana}+|{$kanji}+)/", ' $1 ', $s ); 00026 00027 # Double-width roman characters: ff00-ff5f ~= 0020-007f 00028 $s = preg_replace( '/\xef\xbc([\x80-\xbf])/e', 'chr((ord("$1") & 0x3f) + 0x20)', $s ); 00029 $s = preg_replace( '/\xef\xbd([\x80-\x99])/e', 'chr((ord("$1") & 0x3f) + 0x60)', $s ); 00030 00031 # Do general case folding and UTF-8 armoring 00032 return parent::stripForSearch( $s ); 00033 } 00034 00035 # Italic is not appropriate for Japanese script 00036 # Unfortunately most browsers do not recognise this, and render <em> as italic 00037 function emphasize( $text ) { 00038 return $text; 00039 } 00040 }