make html_unquote('') consistently return u'' everywhere -- re.sub('', '', u'')

started acting differently as of 2.5.1

1 files changed, 5 insertions, 1 deletions

diff --git a/paste/util/quoting.py b/paste/util/quoting.py
index cd87f2d..bd3ae0c 100644
--- a/paste/util/quoting.py
+++ b/paste/util/quoting.py
@@ -50,13 +50,17 @@ def html_unquote(s, encoding=None):
     >>> html_unquote('<hey you>')
     u'<hey\xa0you>'
     >>> html_unquote('')
-    ''
+    u''
     >>> html_unquote('&blahblah;')
     u'&blahblah;'
     >>> html_unquote('\xe1\x80\xa9')
     u'\u1029'
     """
     if isinstance(s, str):
+        if s == '':
+            # workaround re.sub('', '', u'') returning '' < 2.5.1
+            # instead of u'' >= 2.5.1
+            return u''
         s = s.decode(encoding or default_encoding)
     return _unquote_re.sub(_entity_subber, s)