The code below is used to parse input tags in html form. If you have a script to crawl a website and it does have input form with many inputs , this will save you a lot of time.
function parse_html_input($html)
{
$inputs=array();
$regexp = "/<input(.*?)>/is";
if(preg_match_all ("$regexp", $html, $matches))
{
foreach($matches[0] as $match)
{
preg_match('/name=[\"\']{0,1}(\w+)[\"\']{0,1}/i',$match,$result);
if(!$result)
continue;
$value="";
$input_name = $result[1];
if($input_name=="")
continue;
if($result)
{
$match=str_replace(">"," >",$match);
if(strpos(strtolower($match),"value=\"")!==false)
preg_match('/value=[\"\']{0,1}[^"]*[\"\']{0,1}/i',$match,$result2);
else
preg_match('/value=[^\n\r"]+ /i',$match,$result2);
if($result2)
{
$value= $result2[0];
if(strtolower(substr($value,0,6))=="value=")
{
$value=substr($value,6,strlen($value)-6);
}
$value=str_replace("\"","",$value);
}
}
$inputs[$input_name]=$value;
}
}
$form_regexp = "<form.+>";
$form_action="";
if(preg_match_all ("/$form_regexp/i", $html, $matches))
{
foreach($matches[0] as $match)
{
if(strpos(strtolower($match),"action=\"")!==false)
{
preg_match('/action=[\"\']{0,1}[^\n\r"]*[\"\']{0,1}/i',$match,$result2);
$form_action=explode("\"",$result2[0])[1];
}
elseif(strpos(strtolower($match),"action=")!==false)
{
preg_match('/action=[^\n\r"]+ /i',$match,$result2);
$form_action=explode("=",$result2[0])[1];
}
}
}
if($form_action!="")
{
$inputs["form_action"]=$form_action;
}
return $inputs;
}