helper functions for Diaspora cert mangling

[friendica.git] / mod / parse_url.php

diff --git a/mod/parse_url.php b/mod/parse_url.php
index acfe624cb233dede87cc86c2f304007ca35d2a65..46c6b46e97998e3db59b542d11b3b15b937c8fe4 100644 (file)
--- a/mod/parse_url.php
+++ b/mod/parse_url.php
@@ -1,31 +1,64 @@
 <?php
 
 require_once('library/HTML5/Parser.php');
+require_once('library/HTMLPurifier.auto.php');
 
 function parse_url_content(&$a) {
 
-       $url = trim($_GET['url']);
+       logger('parse_url: ' . $_GET['url']);
+
+       $url = trim(hex2bin($_GET['url']));
+
+       logger('parse_url: ' . $url);
 
        $text = null;
 
-       $template = "<a href=\"%s\" >%s</a>%s";
+       $template = "<br /><a href=\"%s\" >%s</a>%s<br />";
+
+
+       $arr = array('url' => $url, 'text' => '');
+
+       call_hooks('parse_link', $arr);
+
+       if(strlen($arr['text'])) {
+               echo $arr['text'];
+               killme();
+       }
 
-       if($url) 
+       if($url) {
                $s = fetch_url($url);
-       else {
+       } else {
                echo '';
                killme();
        }
-       
+
+       logger('parse_url: data: ' . $s, LOGGER_DATA);
+
        if(! $s) {
                echo sprintf($template,$url,$url,'');
                killme();
        }
 
-       $dom = HTML5_Parser::parse($s);
+       if(strpos($s,'<title>')) {
+               $title = substr($s,strpos($s,'<title>')+7,64);
+               if(strpos($title,'<') !== false)
+                       $title = strip_tags(substr($title,0,strpos($title,'<')));
+       }
+
+       $config = HTMLPurifier_Config::createDefault();
+       $config->set('Cache.DefinitionImpl', null);
 
-       if(! $dom)
-               return $ret;
+       $purifier = new HTMLPurifier($config);
+       $s = $purifier->purify($s);
+
+//     logger('parse_url: purified: ' . $s, LOGGER_DATA);
+
+       $dom = @HTML5_Parser::parse($s);
+
+       if(! $dom) {
+               echo sprintf($template,$url,$url,'');
+               killme();
+       }
 
        $items = $dom->getElementsByTagName('title');
 
@@ -36,24 +69,31 @@ function parse_url_content(&$a) {
                }
        }
 
-
        $divs = $dom->getElementsByTagName('div');
        if($divs) {
                foreach($divs as $div) {
                        $class = $div->getAttribute('class');
-                       if($class && stristr($class,'article')) {
+                       if($class && (stristr($class,'article') || stristr($class,'content'))) {
                                $items = $div->getElementsByTagName('p');
                                if($items) {
                                        foreach($items as $item) {
                                                $text = $item->textContent;
+                                               if(stristr($text,'<script')) {
+                                                       $text = '';
+                                                       continue;
+                                               }
                                                $text = strip_tags($text);
-                                               if(strlen($text) < 100)
+                                               if(strlen($text) < 100) {
+                                                       $text = '';
                                                        continue;
+                                               }
                                                $text = substr($text,0,250) . '...' ;
                                                break;
                                        }
                                }
                        }
+                       if($text)
+                               break;
                }
        }
 
@@ -62,9 +102,13 @@ function parse_url_content(&$a) {
                if($items) {
                        foreach($items as $item) {
                                $text = $item->textContent;
+                               if(stristr($text,'<script'))
+                                       continue;
                                $text = strip_tags($text);
-                               if(strlen($text) < 100)
+                               if(strlen($text) < 100) {
+                                       $text = '';
                                        continue;
+                               }
                                $text = substr($text,0,250) . '...' ;
                                break;
                        }
@@ -72,9 +116,9 @@ function parse_url_content(&$a) {
        }
 
        if(strlen($text)) {
-               $text = '<br />' . $text;
+               $text = '<br /><br />' . $text;
        }
 
-       echo sprintf($template,$url,$title,$text);
+       echo sprintf($template,$url,($title) ? $title : $url,$text);
        killme();
-}
\ No newline at end of file
+}