8797 sujets

Développement web côté serveur, CMS

TOUT est dans le titre?

Est-ce possible?

J'ai déjà tenté quelquechose mais en vain...

############################

<?php


function RecursiveRetrieveURL($p1_s_url_src, $p2_i_limit_recur)
{
static $a_urls, $i_count_recur;

$i_count_recur++;

if($i_count_recur == 1) $a_urls [] = '';

if($i_count_recur <= $p2_i_limit_recur)
{
if(!empty($p1_s_url_src))
{
if(!in_array($p1_s_url_src, $a_urls))
{

$m_content_url2 = fopen($p1_s_url_src, 'rb');

$s_content_url = stream_get_contents($m_content_url2);

fclose($m_content_url2);

if(is_string($s_content_url))
{

if(preg_match_all('#(http://)?www\.[a-z_-]+\.[a-z_-]{2,4}#i', $s_content_url, $a_matches))
{
foreach($a_matches [0] as $s_urls):


if(!preg_match('#http#i', $s_urls))
{

$s_formated_urls = preg_replace('#(www\.)([a-z_-]+\.[a-z_-]{2,4})#i', 'http://$1$2', $s_urls);

if(!in_array($s_formated_urls, $a_urls)) $a_urls [] = $s_formated_urls;

RecursiveRetrieveURL($s_formated_urls, $p2_i_limit_recur);


}else
{
if(!in_array($s_urls, $a_urls)) $a_urls [] = $s_urls;

RecursiveRetrieveURL($s_urls, $p2_i_limit_recur);
}

endforeach;

return $a_urls;

}// else string not matching with an url

}else var_dump($m_content_url); // OUTPUT nothing?

}//else the requested URL has already been parsed


}//else empty url

}// else beyond R limitation
}

var_dump(RecursiveRetrieveURL('http://google.fr', 100));
?>