AdSpeed.org | Giving it back: Simple & fast spider detection

Simple & fast spider detection

Developers can use the browscap.ini to detect if the User-Agent is a search engine or data collection spider or not. However, that file is rather large and will be expensive on resources. If you need a faster, simpler (albeit less accurate) method, this class can be useful.

<?php
/** a quick/less resource intensive browser detection
* @author Son Nguyen
* @since 1/29/2007
* @package Framework
* @subpackage Misc
*/
class CBrowserDetect {
 private $mUA;
 /** constructor */
 function __construct($pUA=NULL) {
  $this->setUA($pUA);
 }

/** comment */
 function setUA($pUA=NULL) {
  if (is_null($pUA) && isset($_SERVER['HTTP_USER_AGENT'])) {
   $this->mUA = $_SERVER['HTTP_USER_AGENT'];
  } else {
   $this->mUA = $pUA;
  } // fi
 }

/** if this is a spider/crawler */
 function isSpider($pUA=NULL) {
  $this->setUA($pUA);
  $vBots = array( 
   'AdsBot-Google',
   'alexa',
   'appie',
   'Ask Jeeves',
   'Baiduspider',
   'crawler',
   'FAST',
   'Firefly',
   'froogle',
   'girafabot',
   'Googlebot',
   'ia_archiver',
   'InfoSeek',
   'inktomi',
   'Java/',
   'looksmart',
   'msnbot',
   'NationalDirectory',
   'rabaz',
   'Scooter',
   'Slurp',
   'Spade',
   'TECNOSEEK',
   'Teoma',
   'URL_Spider_SQL',
   'WebBug',
   'WebCapture',
   'WebFindBot',
   'Wget/',
   'ZyBorg',
   'libwww-perl/',
   'httpunit/',
   'WebZIP/',
  );
  foreach ($vBots AS $vPat) {
   if (stripos($this->mUA,$vPat)!==FALSE) {
    return TRUE;
   } // fi
  } // rof
  return FALSE;
 }
}
?>

Labels: code

0 comments

AdSpeed.org | Giving it back

Simple & fast spider detection

0 Comments:

PHP5 Classes

Recent Submissions