fixup! Generate tokenizer tests from testdata files

Author: gsnedders

html5lib/tests/tokenizer.py

@@ -1,5 +1,6 @@
 from __future__ import absolute_import, division, unicode_literals
 
+import codecs
 import json
 import warnings
 import re
@@ -185,7 +186,7 @@ def capitalize(s):
 
 class TokenizerFile(pytest.File):
     def collect(self):
-        with open(str(self.fspath), "rb") as fp:
+        with codecs.open(str(self.fspath), "r", encoding="utf-8") as fp:
             tests = json.load(fp)
         if 'tests' in tests:
             for i, test in enumerate(tests['tests']):