From bf5344ce04b8f37b658f481a09f1742f84c182a3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Michael=20T=C3=A4nzer?= <neo@nhng.de>
Date: Wed, 14 Aug 2013 22:45:32 +0200
Subject: bug 1190: RSS may contain HTML markup => decode it and then encode it
 again
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Michael Tänzer <neo@nhng.de>
---
 pages/index/0.php | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

(limited to 'pages/index/0.php')

diff --git a/pages/index/0.php b/pages/index/0.php
index a2c2e5a..db46673 100644
--- a/pages/index/0.php
+++ b/pages/index/0.php
@@ -53,7 +53,13 @@
 
 		$query = "./description";
 		$nodeList = $xpath->query($query, $item);
-		$description = recode_string("UTF8..html" , $nodeList->item(0)->nodeValue);
+		$description = $nodeList->item(0)->nodeValue;
+		// The description may contain HTML entities => convert them
+		$description = html_entity_decode($description, ENT_COMPAT | ENT_HTML401, 'UTF-8');
+		// Description may contain HTML markup and unicode characters => encode them
+		// If we didn't decode and then encode again, (i.e. take the content
+		// as it is in the RSS feed) we might inject harmful markup
+		$description = recode_string("UTF8..html", $description);
 
 		printf("<h3> %s </h3>\n", $title);
 		printf("<p> %s </p>\n", $description);
-- 
cgit v1.2.1