bai_du 采集代码（已过期）

<?php
$url = "http://www.baidu.com/s?wd=site:www.xxxxxx.com+inurl:hot&tn=baidulaonian&pn=";
$i=0;
while($i<64){
$url1 = $url.($i*10);
$html = gethtml($url1);
$string = htmlspecialchars ($html);
if(strpos($string,"http://www.faxingzhan.com")==false){
break;
}
//获取列表
preg_match_all('/[a-zA-z]+:\/\/www.xxxxxx.com\/[^\s]*/',$string,$data);
$result = $data[0];
if(is_array($result)){
foreach($result as $item){
echo $item .'<br />';
}
}else{
continue;
}
$i++;
}

function gethtml($url){
$header = array (
'User-Agent: Mozilla/5.0 (Windows NT 5.1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/33.0.1750.146 Safari/537.36'
);
$ch = curl_init ();
curl_setopt ( $ch, CURLOPT_URL, $url );
curl_setopt ( $ch, CURLOPT_HTTPHEADER, $header );
curl_setopt ( $ch, CURLOPT_RETURNTRANSFER, 1 );
// 执行
$content = curl_exec ( $ch );
if ($content == FALSE) {
echo "error:" . curl_error ( $ch );
}
// 关闭
curl_close ( $ch );

//输出结果
return $content;
}    
?>

正则写的还有些缺陷 适合百度老年版（目前已经失效，仅供参考）