LanguageJa.php 1.3 KB

1234567891011121314151617181920212223242526272829303132333435363738394041
  1. <?php
  2. /**
  3. * Japanese (日本語)
  4. *
  5. * @ingroup Language
  6. */
  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
  11. $s = $string;
  12. # Strip known punctuation ?
  13. #$s = preg_replace( '/\xe3\x80[\x80-\xbf]/', '', $s ); # U3000-303f
  14. # Space strings of like hiragana/katakana/kanji
  15. $hiragana = '(?:\xe3(?:\x81[\x80-\xbf]|\x82[\x80-\x9f]))'; # U3040-309f
  16. $katakana = '(?:\xe3(?:\x82[\xa0-\xbf]|\x83[\x80-\xbf]))'; # U30a0-30ff
  17. $kanji = '(?:\xe3[\x88-\xbf][\x80-\xbf]'
  18. . '|[\xe4-\xe8][\x80-\xbf]{2}'
  19. . '|\xe9[\x80-\xa5][\x80-\xbf]'
  20. . '|\xe9\xa6[\x80-\x99])';
  21. # U3200-9999 = \xe3\x88\x80-\xe9\xa6\x99
  22. $s = preg_replace( "/({$hiragana}+|{$katakana}+|{$kanji}+)/", ' $1 ', $s );
  23. # Double-width roman characters: ff00-ff5f ~= 0020-007f
  24. $s = preg_replace( '/\xef\xbc([\x80-\xbf])/e', 'chr((ord("$1") & 0x3f) + 0x20)', $s );
  25. $s = preg_replace( '/\xef\xbd([\x80-\x99])/e', 'chr((ord("$1") & 0x3f) + 0x60)', $s );
  26. # Do general case folding and UTF-8 armoring
  27. return parent::stripForSearch( $s );
  28. }
  29. # Italic is not appropriate for Japanese script
  30. # Unfortunately most browsers do not recognise this, and render <em> as italic
  31. function emphasize( $text ) {
  32. return $text;
  33. }
  34. }