[php-blog-cvs] serendipity/bundled-libs/Services SURBL.php,NONE,1.1.2.1

[Sebastian N. <no...@us...>]

Update of /cvsroot/php-blog/serendipity/bundled-libs/Services
In directory sc8-pr-cvs1.sourceforge.net:/tmp/cvs-serv5805/bundled-libs/Services

Added Files:
      Tag: branch-smarty
	SURBL.php 
Log Message:
Added SURBL support to serendipity_event_spamblock

--- NEW FILE: SURBL.php ---
<?php
/* vim: set expandtab tabstop=4 shiftwidth=4 softtabstop=4: */
// +----------------------------------------------------------------------+
// | PEAR::Services_SURBL                                                 |
// +----------------------------------------------------------------------+
// | Copyright (c) 2004 Sebastian Nohn <seb...@no...>               |
// +----------------------------------------------------------------------+
// | This source file is subject to version 3.0 of the PHP license,       |
// | that is bundled with this package in the file LICENSE, and is        |
// | available through the world-wide-web at the following url:           |
// | http://www.php.net/license/3_0.txt.                                  |
// | If you did not receive a copy of the PHP license and are unable to   |
// | obtain it through the world-wide-web, please send a note to          |
// | li...@ph... so we can mail you a copy immediately.               |
// +----------------------------------------------------------------------+
// | Authors: Sebastian Nohn <seb...@no...>                         |
// +----------------------------------------------------------------------+
//
// $Id: SURBL.php,v 1.1.2.1 2004/11/11 13:31:43 nohn Exp $

/**
 * PEAR::Services_SURBL
 *
 * This class acts as interface to the SURBL - Spam URI Realtime Blocklists.
 *
 * Services_SURBL looks up an supplied URI if it's listed in a
 * Spam URI Realtime Blocklists.
 *
 * @author  Sebastian Nohn <seb...@no...>
 * @package Services_SURBL
 * @license http://www.php.net/license/3_0.txt
 * @version 0.3
 */
require_once "Net/CheckIP.php";
require_once "Cache.php";

class Services_SURBL {

    /**     
     * @var    string
     * @access protected
     */
    var $blacklist = "multi.surbl.org";

    /**
     * @var    string
     * @access protected
     */
    var $doubleCcTldFile = "http://spamcheck.freeapp.net/two-level-tlds";

    /**
     * @var    array
     * @access private
     */
    var $twoLevelCcTld = array();

    /**
     * Check if the last two parts of the FQDN are whitelisted
     *
     * @param  $fqdn
     * @access protected
     */
    function isDoubleCcTld($fqdn) {
        // 30 Day should be way enough
        $options = array('lifeTime' => 2);
        $id = md5($this->doubleCcTldFile);

        $cache = new Cache("file", $options);
        if ($data = $cache->get($id)) {
            // Cache hit
        } else {
            // Cache miss
            $fp = fopen ($this->doubleCcTldFile, "r");
            while (!feof($fp)) {
                $buffer = fgets($fp, 4096);
                $data[] = trim($buffer);
            } // while
            $data = array_flip($data);
            fclose ($fp);
            $cache->save($id, $data);
        } // if 
        if (array_key_exists($fqdn, $data)) {
            return true;
        } else {
            return false;
        } // if
    } // function

    /**
     * Get Hostname to ask for
     *
     * Performs the following steps:
     *
     * (1) Extract the hostname from the given URI
     * (2) Check if the "hostname" is an ip
     * (3a) IS_IP Reverse the IP (1.2.3.4 -> 4.3.2.1)
     * (3b) IS_FQDN Check if is in "CC-2-level-TLD"
     * (3b1) IS_IN_2LEVEL: we want the last three names
     * (3b2) IS_NOT_2LEVEL: we want the last two names
     * (4) return the FQDN to query
     *
     * @param  string $uri
     * @access protected
     */
    function getNameForCheck($uri) {
        $host       = "";
        // (1) Extract the hostname from the given URI
        $parsed_uri = parse_url($uri);
        $host       = $parsed_uri["host"];
        // (2) Check if the "hostname" is an ip
        if (Net_CheckIP::check_ip($host)) {
            // (3a) IS_IP Reverse the IP (1.2.3.4 -> 4.3.2.1)
            $host = implode(".", array_reverse(explode(".", $host)));
        } else {
            $host_elements = explode(".", $host);
            while (count($host_elements) > 3) {
                array_shift($host_elements);
            } // while
            $host_3_elements = implode(".", $host_elements);

            $host_elements = explode(".", $host);
            while (count($host_elements) > 2) {
                array_shift($host_elements);
            } // while
            $host_2_elements = implode(".", $host_elements);

            // (3b) IS_FQDN Check if is in "CC-2-level-TLD"
            if ($this->isDoubleCcTld($host_2_elements)) {
                // (3b1) IS_IN_2LEVEL: we want the last three names
                $host = $host_3_elements;
            } else {
                // (3b2) IS_NOT_2LEVEL: we want the last two names
                $host = $host_2_elements;
            } // if
        } // if
        // (4) return the FQDN to query
        $host      .= ".".$this->blacklist;
        return $host;
    } // function

    /**
     * Set the blacklist to a desired blacklist.
     * 
     * Default is multi.surbl.org.
     *
     * It is suggested to use one of these:
     *
     * sc.surbl.org - SpamCop message-body URI domains
     * ws.surbl.org - sa-blacklist domains as a SURBL
     * ob.surbl.org - OutBlaze spamvertised sites
     * ab.surbl.org - AbuseButler spamvertised sites
     * multi.surbl.org - Combined SURBL list
     * 
     * @param  string $blacklist
     * @access public
     */
    function setBlacklist($blacklist = "multi.surbl.org") {
        $this->blacklist = $blacklist;
    }

    /**
     * Get the blacklist.
     *
     * @access public
     */
    function getBlacklist() {
        return $this->blacklist;
    }

    /**
     * Checks if the supplied URL is listen in an SURBL
     *
     * @param  string $host
     * @access public
     */
    function isSpam($host) {
        $result = gethostbyname($this->getNameForCheck($host));
        if ($result != $this->getNameForCheck($host)) {
            return true;
        } else {
            return false;
        }
    } // function

} // class
?>