Python 3 - properly handle __str__ vs __unicode__ .

Also test for it, which increases test coverage to 100%.

Author: joncotton

embedly/models.py

@@ -1,11 +1,8 @@
-from __future__ import unicode_literals
-
-try:  # pragma: no cover
-    from collections import UserDict as IterableUserDict  # pragma: no cover
-except ImportError:  # Python 2  # pragma: no cover
-    from UserDict import IterableUserDict  # pragma: no cover
+from __future__ import absolute_import, unicode_literals
+from .py3_utils import python_2_unicode_compatible, IterableUserDict
 
 
+@python_2_unicode_compatible
 class Url(IterableUserDict, object):
     """
     A dictionary with two additional attributes for the method and url.
@@ -19,7 +16,4 @@ def __init__(self, data=None, method=None, original_url=None, **kwargs):
         self.original_url = original_url
 
     def __str__(self):
-        return self.__unicode__().encode("utf-8")
-
-    def __unicode__(self):
         return '<%s %s>' % (self.method.title(), self.original_url or "")

embedly/py3_utils.py

@@ -0,0 +1,28 @@
+import sys
+
+# 2to3 doesn't handle the UserDict relocation
+# put the import logic here for cleaner usage
+try:
+    from collections import UserDict as IterableUserDict
+except ImportError:  # Python 2
+    from UserDict import IterableUserDict
+
+
+def python_2_unicode_compatible(klass):
+    """
+    A decorator that defines __unicode__ and __str__ methods under Python 2.
+    Under Python 3 it does nothing.
+
+    From django.utils.encoding.py in 1.4.2+, minus the dependency on Six.
+
+    To support Python 2 and 3 with a single code base, define a __str__ method
+    returning text and apply this decorator to the class.
+    """
+    if sys.version_info[0] == 2:
+        if '__str__' not in klass.__dict__:
+            raise ValueError("@python_2_unicode_compatible cannot be applied "
+                             "to %s because it doesn't define __str__()." %
+                             klass.__name__)
+        klass.__unicode__ = klass.__str__
+        klass.__str__ = lambda self: self.__unicode__().encode('utf-8')
+    return klass

embedly/tests.py

@@ -1,5 +1,6 @@
 from __future__ import unicode_literals
 import re
+import sys
 import json
 
 try:  # pragma: no cover
@@ -78,6 +79,18 @@ def test_model_data_can_serialize(self):
         unserialzed = json.loads(json.dumps(obj.data))
         self.assertDictEqual(obj.data, unserialzed)
 
+    def test_str_representation(self):
+        unistr = 'I\xf1t\xebrn\xe2ti\xf4n\xe0liz\xe6tion'
+        url = "http://test.com"
+        obj = Url(method=unistr, original_url=url)
+
+        if sys.version_info[0] == 2:
+            self.assertTrue(unistr.encode('utf-8') in str(obj))
+            self.assertTrue(url.encode('utf-8') in str(obj))
+        else:
+            self.assertTrue(unistr in str(obj))
+            self.assertTrue(url in str(obj))
+
 
 class EmbedlyTestCase(unittest.TestCase):
     def setUp(self):