array('timeout' => $timeout, // Timeout : time until we stop waiting for the response. 'header'=> "User-Agent: ".$useragent, // spoot Mozilla Firefox 'follow_location' => true ))); // only download page lesser than 4MB $data = @file_get_contents($url, false, $context, -1, 4000000); // We download at most 4 MB from source. // echo "

http_response_header : ".print_r($http_response_header);

        if(isset($http_response_header) and isset($http_response_header[0])) {
            $httpcodeOK = isset($http_response_header) and isset($http_response_header[0]) and ((strpos($http_response_header[0], '200 OK') !== FALSE) or (strpos($http_response_header[0], '301 Moved Permanently') !== FALSE));
        }
    }

    // if response is not empty and response is OK
    if (isset($data) and isset($httpcodeOK) and $httpcodeOK ) {

        // take charset of page and get it
        preg_match('##Usi', $data, $meta);

        // if meta tag is found
        if (!empty($meta[0])) {
            // retrieve encoding in $enc
            preg_match('#charset="?(.*)"#si', $meta[0], $enc);

            // if charset is found set it otherwise, set it to utf-8
            $html_charset = (!empty($enc[1])) ? strtolower($enc[1]) : 'utf-8';

        } else {
            $html_charset = 'utf-8';
            $enc[1] = '';
        }

        // replace charset of url to charset of page
        $data = str_replace('charset='.$enc[1], 'charset='.$html_charset, $data);

        return $data;
    }
    else {
        return FALSE;
    }
}

/**
 * Préparation de l'URL avec récupération du contenu avant insertion en base
 */
function prepare_url($url)
{
    $parametres = array();
    $url    = html_entity_decode(trim($url));

    // We remove the annoying parameters added by FeedBurner and GoogleFeedProxy (?utm_source=...)
    // from shaarli, by sebsauvage
    $i=strpos($url,'&utm_source='); if ($i!==false) $url=substr($url,0,$i);
    $i=strpos($url,'?utm_source='); if ($i!==false) $url=substr($url,0,$i);
    $i=strpos($url,'#xtor=RSS-'); if ($i!==false) $url=substr($url,0,$i);

    $title  = $url;
    if (!preg_match('!^https?://!i', $url))
        $url = 'http://' . $url;

    $html = Encoding::toUTF8(get_external_file($url,15));
    if (isset($html) and strlen($html) > 0)
    {
        $r = new Readability($html, $url);
        if($r->init())
        {
            $title = $r->articleTitle->innerHTML;
        }
    }

    $parametres['title']    = $title;
    $parametres['content']  = $r->articleContent->innerHTML;

    return $parametres;
}