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.

<br /> <?php<br /> /** a quick/less resource intensive browser detection<br /> * @author Son Nguyen<br /> * @since 1/29/2007<br /> * @package Framework<br /> * @subpackage Misc<br /> */<br /> class CBrowserDetect {<br /> private $mUA;<br /> /** constructor */<br /> function __construct($pUA=NULL) {<br /> $this->setUA($pUA);<br /> }</p> <p> /** comment */<br /> function setUA($pUA=NULL) {<br /> if (is_null($pUA) && isset($_SERVER['HTTP_USER_AGENT'])) {<br /> $this->mUA = $_SERVER['HTTP_USER_AGENT'];<br /> } else {<br /> $this->mUA = $pUA;<br /> } // fi<br /> }</p> <p> /** if this is a spider/crawler */<br /> function isSpider($pUA=NULL) {<br /> $this->setUA($pUA);<br /> $vBots = array(<br /> ‘AdsBot-Google’,<br /> ‘alexa’,<br /> ‘appie’,<br /> ‘Ask Jeeves’,<br /> ‘Baiduspider’,<br /> ‘crawler’,<br /> ‘FAST’,<br /> ‘Firefly’,<br /> ‘froogle’,<br /> ‘girafabot’,<br /> ‘Googlebot’,<br /> ‘ia_archiver’,<br /> ‘InfoSeek’,<br /> ‘inktomi’,<br /> ‘Java/’,<br /> ‘looksmart’,<br /> ‘msnbot’,<br /> ‘NationalDirectory’,<br /> ‘rabaz’,<br /> ‘Scooter’,<br /> ‘Slurp’,<br /> ‘Spade’,<br /> ‘TECNOSEEK’,<br /> ‘Teoma’,<br /> ‘URL_Spider_SQL’,<br /> ‘WebBug’,<br /> ‘WebCapture’,<br /> ‘WebFindBot’,<br /> ‘Wget/’,<br /> ‘ZyBorg’,<br /> ‘libwww-perl/’,<br /> ‘httpunit/’,<br /> ‘WebZIP/’,<br /> );<br /> foreach ($vBots AS $vPat) {<br /> if (stripos($this->mUA,$vPat)!==FALSE) {<br /> return TRUE;<br /> } // fi<br /> } // rof<br /> return FALSE;<br /> }<br /> }<br /> ?><br />