php写的文章采集URL补全函数(FormatUrl)

<?php
$surl="http://www.jb200.com/";
$gethtm = '<a href="/index.htm">首页</a><a href="Resolvent/index.htm">解决方案</a>';
echo formaturl($gethtm,$surl);
?>

<?php
function formaturl($l1,$l2){
if (preg_match_all("/(<img[^>]+src="([^"]+)"[^>]*>)|(<a[^>]+href="([^"]+)"[^>]*>)|(<img[^>]+src='([^']+)'[^>]*>)|(<a[^>]+href='([^']+)'[^>]*>)/i",$l1,$regs)){
foreach($regs[0] as $num => $url){
$l1 = str_replace($url,lIIIIl($url,$l2),$l1);
}
}
return $l1;
}
function lIIIIl($l1,$l2){
if(preg_match("/(.*)(href|src)=(.+?)( |/>|>).*/i",$l1,$regs)){$I2 = $regs[3];}
if(strlen($I2)>0){
$I1 = str_replace(chr(34),"",$I2);
$I1 = str_replace(chr(39),"",$I1);
}else{return $l1;}
$url_parsed = parse_url($l2);
$scheme = $url_parsed["scheme"];if($scheme!=""){$scheme = $scheme."://";}
$host = $url_parsed["host"];
$l3 = $scheme.$host;
if(strlen($l3)==0){return $l1;}
$path = dirname($url_parsed["path"]);if($path[0]==""){$path="";}
$pos = strpos($I1,"#");
if($pos>0) $I1 = substr($I1,0,$pos);
//判断类型
if(preg_match("/^(http|https|ftp):(//|\)(([w/+-~`@:%])+.)+([w/.=?+-~`@':!%#]|(&)|&)+/i",$I1)){return $l1; }//http开头的url类型要跳过
elseif($I1[0]=="/"){$I1 = $l3.$I1;}//绝对路径
elseif(substr($I1,0,3)=="../"){//相对路径
while(substr($I1,0,3)=="../"){
$I1 = substr($I1,strlen($I1)-(strlen($I1)-3),strlen($I1)-3);
if(strlen($path)>0){
$path = dirname($path);
}
}
$I1 = $l3.$path."/".$I1;
}
elseif(substr($I1,0,2)=="./"){
$I1 = $l3.$path.substr($I1,strlen($I1)-(strlen($I1)-1),strlen($I1)-1);
}
elseif(strtolower(substr($I1,0,7))=="mailto:"||strtolower(substr($I1,0,11))=="javascript:"){
return $l1;
}else{
$I1 = $l3.$path."/".$I1;
}
return str_replace($I2,""$I1"",$l1);
}
?>