Get all links by reading the source files of a site

Read the source file of a certain site, then use regular expressions to analyze its source code and get all the links.

/**********qiushuiwuhen(2002-5-20)***********/
if(empty($url))$url = "http://www.csdn.net/expert/";//Set url
$site =substr($url,0,strpos($url,"/",8));//Site
$base=substr($url,0,strrpos($url,"/") 1);//File Directory


$fp = fopen($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;$iif(!eregi("://",$regArr[$i][1]))//Whether It is a relative path, that is, whether there is ://
if(substr($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);
?>