languages/classes/LanguageJa.php

   1 <?php
   2 /**
   3  * Japanese (日本語)
   4  *
   5  * @package MediaWiki
   6  * @subpackage Language
   7  */
   8 class LanguageJa extends Language {
   9         function stripForSearch( $string ) {
  10                 # MySQL fulltext index doesn't grok utf-8, so we
  11                 # need to fold cases and convert to hex
  12                 $s = $string;
  13
  14                 # Strip known punctuation ?
  15                 #$s = preg_replace( '/\xe3\x80[\x80-\xbf]/', '', $s ); # U3000-303f
  16
  17                 # Space strings of like hiragana/katakana/kanji
  18                 $hiragana = '(?:\xe3(?:\x81[\x80-\xbf]|\x82[\x80-\x9f]))'; # U3040-309f
  19                 $katakana = '(?:\xe3(?:\x82[\xa0-\xbf]|\x83[\x80-\xbf]))'; # U30a0-30ff
  20                 $kanji = '(?:\xe3[\x88-\xbf][\x80-\xbf]'
  21                         . '|[\xe4-\xe8][\x80-\xbf]{2}'
  22                         . '|\xe9[\x80-\xa5][\x80-\xbf]'
  23                         . '|\xe9\xa6[\x80-\x99])';
  24                         # U3200-9999 = \xe3\x88\x80-\xe9\xa6\x99
  25                 $s = preg_replace( "/({$hiragana}+|{$katakana}+|{$kanji}+)/", ' $1 ', $s );
  26
  27                 # Double-width roman characters: ff00-ff5f ~= 0020-007f
  28                 $s = preg_replace( '/\xef\xbc([\x80-\xbf])/e', 'chr((ord("$1") & 0x3f) + 0x20)', $s );
  29                 $s = preg_replace( '/\xef\xbd([\x80-\x99])/e', 'chr((ord("$1") & 0x3f) + 0x60)', $s );
  30
  31                 # Do general case folding and UTF-8 armoring
  32                 return parent::stripForSearch( $s );
  33         }
  34
  35         # Italic is not appropriate for Japanese script
  36         # Unfortunately most browsers do not recognise this, and render <em> as italic
  37         function emphasize( $text ) {
  38                 return $text;
  39         }
  40 }
  41
  42 ?>