root/trackback/requirements/de.easy-coding.wcf.trackback/files/lib/util/XMLRPCServer.class.php @ 805

<?php
require_once(WBB_DIR.'lib/page/PublicSEORewriter.class.php');
require_once(WCF_DIR.'lib/util/StringUtil.class.php');
require_once(WBB_DIR.'lib/data/post/Post.class.php');
require_once(WBB_DIR.'lib/util/IXR.class.php');
require_once(WBB_DIR.'lib/util/TrackbackUtil.class.php');
require_once(WBB_DIR.'lib/data/thread/Thread.class.php');

// Some browser-embedded clients send cookies. We don't want them.
$_COOKIE = array();

// A bug in PHP < 5.2.2 makes $HTTP_RAW_POST_DATA not set by default,
// but we can do it ourself.
if(!isset($HTTP_RAW_POST_DATA)) {
        $HTTP_RAW_POST_DATA = file_get_contents('php://input');
}

# fix for mozBlog and other cases where '<?xml' isn't on the very first line
if(isset($HTTP_RAW_POST_DATA)) {
        $HTTP_RAW_POST_DATA = trim($HTTP_RAW_POST_DATA);
}


/**
 * xmlrpc util
 *
 * @author      Torben Brodt
 * @package     de.easy-coding.wcf.trackback
 * @license     GNU General Public License <http://opensource.org/licenses/gpl-3.0.html>
 */
class XMLRPCServer extends IXR_Server {
        protected $rewriter;

        /**
         * constructor
         */
        public function __construct() {
                $this->methods = array(
                        'pingback.ping' => 'this:pingback_ping'
                );
                
                $this->rewriter = new PublicSEORewriter();

                parent::IXR_Server($this->methods);
        }
        
        /**
         * escapes strings
         * @param arr
         */
        protected function escape(&$arr) {
                foreach ($arr as $k => $v ) {
                        $arr[$k] = escapeString($v);
                }
        }

        /**
         * gets a pingback and registers it 
         * @param args array(linkedfrom, linkedto)
         */
        public function pingback_ping($args) {
                $this->escape($args);

                $pagelinkedfrom = StringUtil::decodeHTML($args[0]);
                $pagelinkedto   = StringUtil::decodeHTML($args[1]);

                $title = '';

                $error_code = -1;

                // Check if the page linked to is in our site
                $pos1 = strpos($pagelinkedto, str_replace(array('http://www.','http://','https://www.','https://'), '', PAGE_URL));
                if(!$pos1) {
                        return new IXR_Error(0, 'Is there no link to us?');
                }

                // let's find which post is linked to
                $threadID = $this->rewriter->thread2threadID($pagelinkedto);
                if($threadID === NULL) {
                        return new IXR_Error(33, 'The specified target URL cannot be resolved as a thread url.');
                }
                
                $thread = new Thread($threadID);

                // Check if post exists
                if($thread->firstPostID) {
                        $post = new Post($thread->firstPostID);
                } else {
                        return new IXR_Error(33, 'The specified target URL cannot be used as a target. It doesn\'t exist.');
                }

                // Check if pings are on
                if(!$post->hasTrackback) {
                        return new IXR_Error(33, 'The specified target URL cannot be used as a target. It is not a pingback-enabled resource.');
                }

                // check for duplicates
                $sql = "SELECT  COUNT(*) AS c
                        FROM    wbb".WBB_N."_trackback 
                        WHERE   postID = ".$post->postID."
                        AND     url = '".escapeString($pagelinkedfrom)."' ";
                        
                $row = WBBCore::getDB()->getFirstRow($sql);
                if(intval($row['c']) > 0) {
                        return new IXR_Error(48, 'The pingback has already been registered.');
                }

                // very stupid, but gives time to the 'from' server to publish !
                sleep(1);

                // Let's check the remote site
                extract(parse_url($pagelinkedfrom), EXTR_SKIP);

                $path  = ( !isset($path) ) ? '/'          : $path;
                $path .= ( isset($query) ) ? '?' . $query : '';
                $port  = ( isset($port)  ) ? $port        : 80;

                // Try to connect to the server at $host
                $fp = @fsockopen($host, $port, $errno, $errstr, 2);
                if(!$fp) {
                        return new IXR_Error(16, 'The source URL does not exist.');
                }
                
                $request = "GET $path HTTP/1.1\r\nHost: $host\r\nUser-Agent: ".TrackbackUtil::$agent." \r\n\r\n";
                fputs($fp, $request);
                $linea = '';
                while (!feof($fp)) {
                        $linea .= fgets($fp, 4096);
                }

                // Work around bug in strip_tags():
                $linea = str_replace('<!DOC', '<DOC', $linea);
                $linea = preg_replace( '/[\s\r\n\t]+/', ' ', $linea ); // normalize spaces
                $linea = preg_replace( "/ <(h1|h2|h3|h4|h5|h6|p|th|td|li|dt|dd|pre|caption|input|textarea|button|body)[^>]*>/", "\n\n", $linea );

                preg_match('|<title>([^<]*?)</title>|is', $linea, $matchtitle);
                $title = $matchtitle[1];
                if(empty($title)) {
                        return new IXR_Error(32, 'We cannot find a title on that page.');
                }

                $linea = strip_tags($linea, '<a>'); // just keep the tag we need
                $p = explode("\n\n", $linea);

                $preg_target = preg_quote($pagelinkedto);

                foreach($p as $para) {
                        if(strpos($para, $pagelinkedto) !== false) { // it exists, but is it a link?
                                preg_match("|<a[^>]+?".$preg_target."[^>]*>([^>]+?)</a>|", $para, $context);

                                // If the URL isn't in a link context, keep looking
                                if(empty($context)) {
                                        continue;
                                }

                                // We're going to use this fake tag to mark the context in a bit
                                // the marker is needed in case the link text appears more than once in the paragraph
                                $excerpt = preg_replace('|\</?wpcontext\>|', '', $para);

                                // prevent really long link text
                                if(strlen($context[1]) > 100) {
                                        $context[1] = substr($context[1], 0, 100).'...';
                                }

                                $marker = '<wpcontext>'.$context[1].'</wpcontext>';    // set up our marker
                                $excerpt= str_replace($context[0], $marker, $excerpt); // swap out the link for our marker
                                $excerpt = strip_tags($excerpt, '<wpcontext>');        // strip all tags but our context marker
                                $excerpt = trim($excerpt);
                                $preg_marker = preg_quote($marker);
                                $excerpt = preg_replace("|.*?\s(.{0,100}$preg_marker.{0,100})\s.*|s", '$1', $excerpt);
                                $excerpt = strip_tags($excerpt); // YES, again, to remove the marker wrapper
                                break;
                        }
                }

                if(empty($context)) { 
                        // Link to target not found
                        return new IXR_Error(17, 'The source URL does not contain a link to the target URL, and so cannot be used as a source.');
                }

                // save incoming
                TrackbackUtil::save($postID, $pagelinkedto, $title, $excerpt, $pagelinkedfrom, $title);

                return sprintf('Pingback from %1$s to %2$s registered. Keep the web talking! :-)', $pagelinkedfrom, $pagelinkedto);
        }
}
?>