X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=mod%2Fparse_url.php;h=46c6b46e97998e3db59b542d11b3b15b937c8fe4;hb=cfc7f8308f5e1756d1ea7b2fadd317821442d16c;hp=33381a0d271be0363535bf2a6e29aa5ad892ccc6;hpb=36b66dccb60a7d65724da542f9d7a2d6722fc6aa;p=friendica.git

diff --git a/mod/parse_url.php b/mod/parse_url.php
index 33381a0d27..46c6b46e97 100644
--- a/mod/parse_url.php
+++ b/mod/parse_url.php
@@ -1,32 +1,124 @@
 <?php
 
 require_once('library/HTML5/Parser.php');
+require_once('library/HTMLPurifier.auto.php');
 
 function parse_url_content(&$a) {
-	$url = trim($_GET['url']);
 
-	$template = "<a href=\"%s\" >%s</a>";
+	logger('parse_url: ' . $_GET['url']);
 
-	if($url) 
+	$url = trim(hex2bin($_GET['url']));
+
+	logger('parse_url: ' . $url);
+
+	$text = null;
+
+	$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) {
 		$s = fetch_url($url);
-	
+	} else {
+		echo '';
+		killme();
+	}
+
+	logger('parse_url: data: ' . $s, LOGGER_DATA);
+
 	if(! $s) {
-		echo sprintf($template,$url,$url);
+		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);
+
+	$purifier = new HTMLPurifier($config);
+	$s = $purifier->purify($s);
+
+//	logger('parse_url: purified: ' . $s, LOGGER_DATA);
 
-	if(! $dom)
-		return $ret;
+	$dom = @HTML5_Parser::parse($s);
+
+	if(! $dom) {
+		echo sprintf($template,$url,$url,'');
+		killme();
+	}
 
 	$items = $dom->getElementsByTagName('title');
-	
-	foreach($items as $item) {
-		$title = $item->textContent;
-		break;
+
+	if($items) {
+		foreach($items as $item) {
+			$title = trim($item->textContent);
+			break;
+		}
+	}
+
+	$divs = $dom->getElementsByTagName('div');
+	if($divs) {
+		foreach($divs as $div) {
+			$class = $div->getAttribute('class');
+			if($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) {
+							$text = '';
+							continue;
+						}
+						$text = substr($text,0,250) . '...' ;
+						break;
+					}
+				}
+			}
+			if($text)
+				break;
+		}
+	}
+
+	if(! $text) {
+		$items = $dom->getElementsByTagName('p');
+		if($items) {
+			foreach($items as $item) {
+				$text = $item->textContent;
+				if(stristr($text,'<script'))
+					continue;
+				$text = strip_tags($text);
+				if(strlen($text) < 100) {
+					$text = '';
+					continue;
+				}
+				$text = substr($text,0,250) . '...' ;
+				break;
+			}
+		}
+	}
+
+	if(strlen($text)) {
+		$text = '<br /><br />' . $text;
 	}
 
-	echo sprintf($template,$url,$title);
+	echo sprintf($template,$url,($title) ? $title : $url,$text);
 	killme();
-}
\ No newline at end of file
+}