FS#242 Don't use UTF8 in search when possible

Backend

• 

  2005-04-07 andi

  The search is significantly slower whith UTF-8 support. A check should be made if the search string contains ASCII-7 only, if yes a non UTF8-aware method can be used.
• 

  2005-07-14 geofffilippi

  I would bump up the severity a little. My users complain because their searches don't come back. Not knowing much about UTF-8, it seems to me that for english-only sites, the admin should be able to bypass this functionality.
  
  I get the following message:
  
  Fatal error: Maximum execution time of 30 seconds exceeded in /usr/local/apache2/htdocs/dokuwiki/inc/utf8.php on line 377
  
  Looking at the code:
  
    374  function unicode_to_utf8( $str ) {
    375    $utf8 = '';
    376    foreach( $str as $unicode ) {
    377      if ( $unicode < 128 ) {
    378        $utf8.= chr( $unicode );
    379      } elseif ( $unicode < 2048 ) {
    380        $utf8.= chr( 192 +  ( ( $unicode - ( $unicode % 64 ) ) / 64 ) );
    381        $utf8.= chr( 128 + ( $unicode % 64 ) );
    382      } else {
    383        $utf8.= chr( 224 + ( ( $unicode - ( $unicode % 4096 ) ) / 4096 ) )
  ;
    384        $utf8.= chr( 128 + ( ( ( $unicode % 4096 ) - ( $unicode % 64 ) ) /
  64 ) );
    385        $utf8.= chr( 128 + ( $unicode % 64 ) );
    386      }
    387    }
    388    return $utf8;
    389  }
  
• 

  2005-09-12 andi

  obsoleted by new index based search