From 5bd1fb4e36f19455c7c291eb0616236043b5a7d3 Mon Sep 17 00:00:00 2001
From: Marius Gedminas <marius@gedmin.as>
Date: Thu, 21 May 2020 19:01:33 +0300
Subject: [PATCH] Fix internal error on empty HTML files

When BeautifulSoup finds an empty file on disk, it sets
original_encoding to None.  It doesn't matter what encoding we pick for
empty files, so let's just pick one.

I don't know if there are any circumstances where BeautifulSoup might
set the encoding to None for a non-empty file.

Closes #392.
---
 linkcheck/checker/urlbase.py         | 6 ++++--
 tests/checker/data/empty.html        | 0
 tests/checker/data/empty.html.result | 5 +++++
 tests/checker/test_file.py           | 3 +++
 4 files changed, 12 insertions(+), 2 deletions(-)
 create mode 100644 tests/checker/data/empty.html
 create mode 100644 tests/checker/data/empty.html.result

diff --git a/linkcheck/checker/urlbase.py b/linkcheck/checker/urlbase.py
index 57f0d28a..f4081301 100644
--- a/linkcheck/checker/urlbase.py
+++ b/linkcheck/checker/urlbase.py
@@ -632,8 +632,10 @@ class UrlBase:
         if self.text is None:
             self.get_raw_content()
             self.soup = htmlsoup.make_soup(self.data)
-            self.text = self.data.decode(self.soup.original_encoding)
-            self.encoding = self.soup.original_encoding
+            # Sometimes soup.original_encoding is None!  Better mangled text
+            # than an internal crash, eh?
+            self.encoding = self.soup.original_encoding or 'ISO-8859-1'
+            self.text = self.data.decode(self.encoding)
         return self.text
 
     def read_content(self):
diff --git a/tests/checker/data/empty.html b/tests/checker/data/empty.html
new file mode 100644
index 00000000..e69de29b
diff --git a/tests/checker/data/empty.html.result b/tests/checker/data/empty.html.result
new file mode 100644
index 00000000..7df3f536
--- /dev/null
+++ b/tests/checker/data/empty.html.result
@@ -0,0 +1,5 @@
+url file://%(curdir)s/%(datadir)s/empty.html
+cache key file://%(curdir)s/%(datadir)s/empty.html
+real url file://%(curdir)s/%(datadir)s/empty.html
+name %(datadir)s/empty.html
+valid
diff --git a/tests/checker/test_file.py b/tests/checker/test_file.py
index ba9ebc15..e91c144d 100644
--- a/tests/checker/test_file.py
+++ b/tests/checker/test_file.py
@@ -68,6 +68,9 @@ class TestFile(LinkCheckTest):
     def test_php(self):
         self.file_test("file.php")
 
+    def test_empty(self):
+        self.file_test("empty.html")
+
     @need_word
     def test_word(self):
         confargs = dict(enabledplugins=["WordParser"])