From 4bcf65f77c4df7144ca07b7f7622395a354a5430 Mon Sep 17 00:00:00 2001
From: Jeremy Cline <jeremy@jcline.org>
Date: Oct 21 2016 17:49:24 +0000
Subject: Handle empty files in detect_encodings


If there's no data, chardet returns ``None``. This just returns ASCII
with 100% confidence if the file is empty.

Signed-off-by: Jeremy Cline <jeremy@jcline.org>

---

diff --git a/pagure/lib/encoding_utils.py b/pagure/lib/encoding_utils.py
index 7f6d0a8..23eb4ef 100644
--- a/pagure/lib/encoding_utils.py
+++ b/pagure/lib/encoding_utils.py
@@ -71,6 +71,10 @@ def detect_encodings(data):
     :return: A dictionary mapping possible encodings to confidence levels
     :rtype:  dict
     """
+    if not data:
+        # It's an empty string so we can safely say it's ascii
+        return {'ascii': 1.0}
+
     # We can't use ``chardet.detect`` because we want to dig in the internals
     # of the detector to bias the utf-8 result.
     detector = universaldetector.UniversalDetector()
diff --git a/tests/test_pagure_lib_encoding_utils.py b/tests/test_pagure_lib_encoding_utils.py
index 0060595..8c63d4a 100644
--- a/tests/test_pagure_lib_encoding_utils.py
+++ b/tests/test_pagure_lib_encoding_utils.py
@@ -29,6 +29,13 @@ class TestGuessEncoding(unittest.TestCase):
         self.assertEqual(result, 'utf-8')
         self.assertEqual(chardet_result['encoding'], 'ISO-8859-2')
 
+    def test_guess_encoding_no_data(self):
+        result = encoding_utils.guess_encoding(u''.encode('utf-8'))
+        self.assertEqual(result, 'ascii')
+
+
+class TestDecode(unittest.TestCase):
+
     def test_decode(self):
         data = u'Šabata'
         self.assertEqual(data, encoding_utils.decode(data.encode('utf-8')))