Spam Reporting Script 5/23

<?php
$sites = array('[redacted]');
shuffle($sites);
$api_key = '[redacted]';

function PostToHost($data) {
   $fp = fsockopen("www.stopforumspam.com",80);
   fputs($fp, "POST /add.php HTTP/1.1\n" );
   fputs($fp, "Host: www.stopforumspam.com\n" );
   fputs($fp, "Content-type: application/x-www-form-urlencoded\n" );
   fputs($fp, "Content-length: ".strlen($data)."\n" );
   fputs($fp, "Connection: close\n\n" );
   fputs($fp, $data);
   fclose($fp);
}

function curl_get($url, $refuseblank = false) {
    $ch = curl_init ();
    curl_setopt ( $ch, CURLOPT_URL, $url);
    curl_setopt ( $ch, CURLOPT_FOLLOWLOCATION, 1 );
    curl_setopt ( $ch, CURLOPT_RETURNTRANSFER, 1 );
    curl_setopt ( $ch, CURLOPT_ENCODING, "" );
    curl_setopt ( $ch, CURLOPT_USERAGENT, "Mozilla/5.0 (Windows; U; Windows NT 6.0; en-US; rv:1.9.2) Gecko/20100115 Firefox/3.6 (.NET CLR 3.5.30729)" );
    if (defined('SERVER_LOGIN')) { //my test server requires a login
        curl_setopt($ch, CURLOPT_USERPWD, SERVER_LOGIN);
        curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, false);
    }
    $out = '';
    if ($refuseblank) {
        while ($out == '') {
            $out = curl_exec ($ch);
        }
    } else {
        $out = curl_exec($ch);
    }
    curl_close($ch);
    return $out;
}

$spamurls = array();
$i = 0;
$spamcounts = array();
foreach ($sites as $url) {
    $spamcounts[$url] = array();
    $sitecount = 0;
    $continue = true;
    echo 'Trying site ' . $url . "\n";
    //get number of registered users
    $site = curl_get($url);
    while ($continue) {
        $data = curl_get($url . '/spam.php?salt=' . time());
        if (!strstr($data, '<?xml')) {
            echo 'Site could not be harvested. Check that it is working properly.' . "\n";
            $continue = false;
            break;
        }
        $xml = new SimpleXMLElement($data);
        if (!isset($xml->user)) {
            $continue = false;
            break;
        }
        foreach ($xml->user as $user) {
            $username = (string)$user->username;
            $email = (string)$user->email;
            $ip = (string)$user->ip;
            $evidence = (string)$user->evidence;

            $resp = posttohost('username=' . rawurlencode($username) . '&ip_addr=' . rawurlencode($ip) . '&evidence=' . rawurlencode($evidence) . '&email=' . rawurlencode($email) . '&api_key=' . $api_key);
            $i++;
            $sitecount++;
            if ($sitecount % 10 == 0) {
                echo ' -> ' . $sitecount . ' entries...' . "\n";
            }

            //any URLs?
            $lines = explode("\n", $evidence);
            foreach ($lines as $line) {
                if (strpos($line, 'http') === 0) {
                    $spamurls[] = $line;
                }
                if (strpos($line, 'Registered: ') === 0) {
                    preg_match('%(\d+) (.*?) (\d+)%', $line, $matches);
                    $timestamp = $matches[0];
                    if (isset($spamcounts[$url][$timestamp])) {
                        $spamcounts[$url][$timestamp]++;
                    } else {
                        $spamcounts[$url][$timestamp] = 1;
                    }
                }
            }
        }
    }
    if (isset($spamcounts[$url])) {
        $data = file_get_contents('stats.txt');
        foreach ($spamcounts[$url] as $timestamp => $count) {
            $data .= $url . ',' . $timestamp . ',' . $count . "\n";
        }
        file_put_contents('stats.txt', $data);
    }
    echo ' -> ' . $sitecount . ' entries for that site' . "\n";
}
echo $i . ' entries processed.' . "\n";

//record any new spam urls
$existingspamurls = explode("\n", file_get_contents('spamurls.txt'));
$spamurls = array_unique(array_merge($existingspamurls, $spamurls));
file_put_contents('spamurls.txt', implode("\n", $spamurls));
unset($spamurls, $existingspamurls);