From 8a33bcbed46eccd2860136ec2c7332c9cfa639f7 Mon Sep 17 00:00:00 2001
From: Andres Rey <andreskrey@gmail.com>
Date: Thu, 15 Jun 2017 17:54:39 +0100
Subject: Safecheck for really bad HTML

---
 CHANGELOG.md       | 2 ++
 src/HTMLParser.php | 2 +-
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index e181081..685cfb3 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -3,6 +3,8 @@ All notable changes to this project will be documented in this file.
 
 ## Unreleased
 
+- Added a safecheck for really nasty HTML
+
 ## [v0.2.1](https://github.com/andreskrey/readability.php/releases/tag/v0.2.1)
 
 - Added `normalizeEntities` flag to convert UTF-8 characters to its HTML Entity equivalent. Fixes bugs on htmls with mixed encoding.
diff --git a/src/HTMLParser.php b/src/HTMLParser.php
index 6e77e3f..9aa9974 100644
--- a/src/HTMLParser.php
+++ b/src/HTMLParser.php
@@ -125,7 +125,7 @@ class HTMLParser
         $this->metadata['title'] = $this->getTitle();
 
         // Checking for minimum HTML to work with.
-        if (!($root = $this->dom->getElementsByTagName('body')->item(0))) {
+        if (!($root = $this->dom->getElementsByTagName('body')->item(0)) || !$root->firstChild) {
             return false;
         }
 
-- 
cgit v1.2.3