Principle: Read the source file of a site, then use regular expressions to analyze its source code, and get all the links
))$url = "http://www.csdn.net/expert/";//Set url
$site=substr($url,0,strpos($url,"/",8)); //Site
$base=substr($url,0,strrpos($url,"/")+1);//The directory where the file is located
$fp = fopen($url, "r " );//Open url
while(!feof($fp))$contents.=fread($fp,1024);//
$pattern="|href=["]? ([^ "]+)[" ]|U";
preg_match_all($pattern,$contents, $regArr, PREG_SET_ORDER);//Match all href=
for($i=0;$i< count($regArr);$i++){//Traverse all matches
if(!eregi("://",$regArr[$i][1]))//Whether it is a relative path, that is, whether it is still There are://
if(substr($regArr[$i][1],0,1)=="/")//Whether it is the root directory of the site
echo "link".($i +1).":".$site.$regArr[$i][1]."
";//Root directory
else
echo "link".($i+ 1).":".$base.$regArr[$i][1]."
";//Current directory
else
echo "link".($i+1 ).":".$regArr[$i][1]."
";//Relative path
}
fclose($fp);
?>
http://www.bkjia.com/PHPjc/508875.html