From 5cdc5d015d68f569bc7c0fbab3fe0cafcd699147 Mon Sep 17 00:00:00 2001
From: Emanuele Cesena <emanuele.cesena@gmail.com>
Date: Thu, 15 Jan 2015 15:35:08 -0800
Subject: [PATCH] Fix nodeToString() to return valid HTML

In some cases the output of nodeToString() is not valid HTML, e.g. for iframe nodes.

Solution found here: http://stackoverflow.com/questions/27020950/lxml-modify-tags-prevent

An example of bad behavior is an article with a youtube video inside. The video embed code is transformed into `<iframe ... />`, that is not valid, and thus the result in the browser is a page with no visible content after the video.
---
 newspaper/parsers.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/newspaper/parsers.py b/newspaper/parsers.py
index f81c7177..c40e09d6 100644
--- a/newspaper/parsers.py
+++ b/newspaper/parsers.py
@@ -85,7 +85,7 @@ def nodeToString(cls, node):
         """`decode` is needed at the end because `etree.tostring`
         returns a python bytestring
         """
-        return lxml.etree.tostring(node).decode()
+        return lxml.etree.tostring(node, method='html').decode()
 
     @classmethod
     def replaceTag(cls, node, tag):