All internal encoding of output now uses the 'xmlcharrefreplace' error handler. Also added a note to the docs. Anyone doing their own encoding of output should be as well.

1 files changed, 1 insertions, 1 deletions

diff --git a/tests/__init__.py b/tests/__init__.py
index 8fbc5d9..b274b1e 100644
--- a/tests/__init__.py
+++ b/tests/__init__.py
@@ -51,7 +51,7 @@ def get_args(file, config):
 
 def normalize(text):
     """ Normalize whitespace for a string of html using tidy. """
-    return str(tidy.parseString(text.encode('utf-8'), 
+    return str(tidy.parseString(text.encode('utf-8', errors='xmlcharrefreplace'), 
                                     drop_empty_paras=0,
                                     fix_backslash=0,
                                     fix_bad_comments=0,