wip

Author: albu-diku

mig/shared/base.py

@@ -490,12 +490,28 @@ def verify_local_url(configuration, req_url):
     return False
 
 
+def is_bytes_type(thetype):
+    """Return boolean indicating if val is a unicode string. We avoid
+    `isinstance(val, unicode)`
+    and the like since it breaks when combined with python-future and futurize.
+    """
+    return (thetype == bytes)
+
+
 def is_unicode(val):
     """Return boolean indicating if val is a unicode string. We avoid
     `isinstance(val, unicode)`
     and the like since it breaks when combined with python-future and futurize.
     """
-    return (type(u"") == type(val))
+    return is_unicode_type(type(val))
+
+
+def is_unicode_type(thetype):
+    """Return boolean indicating if val is a unicode string. We avoid
+    `isinstance(val, unicode)`
+    and the like since it breaks when combined with python-future and futurize.
+    """
+    return (thetype == type(u""))
 
 
 def force_utf8(val, highlight=''):
@@ -510,14 +526,14 @@ def force_utf8(val, highlight=''):
     return codecs.encode("%s%s%s" % (highlight, val, highlight), 'utf8')
 
 
-def _walk_and_convert_recursive(input_obj, highlight='', _as_bytes=False, _force_primitive=None, _force_recursive=None):
+def _walk_and_convert_recursive(input_obj, highlight='', _is_primitive=None, _force_primitive=None, _force_recursive=None):
     thetype = type(input_obj)
     if issubclass(thetype, dict):
         return {_force_recursive(i, highlight): _force_recursive(j, highlight) for (i, j) in
                 input_obj.items()}
     elif issubclass(thetype, (list, tuple)):
         return thetype((_force_recursive(i, highlight) for i in input_obj))
-    elif is_unicode(input_obj):
+    elif not _is_primitive(thetype):
         return _force_primitive(input_obj, highlight)
     else:
         return input_obj
@@ -528,7 +544,7 @@ def force_utf8_rec(input_obj, highlight=''):
     dictionaries with nested unicode strings to a pure utf8 version. Actual
     changes are marked out with the highlight string if given.
     """
-    return _walk_and_convert_recursive(input_obj, highlight, _force_primitive=force_utf8, _force_recursive=force_utf8_rec)
+    return _walk_and_convert_recursive(input_obj, highlight, _is_primitive=is_bytes_type, _force_primitive=force_utf8, _force_recursive=force_utf8_rec)
 
 
 def force_unicode(val, highlight=''):
@@ -548,7 +564,7 @@ def force_unicode_rec(input_obj, highlight=''):
     dictionaries with nested utf8 strings to a pure unicode version. Actual
     changes are marked out with the highlight string if given.
     """
-    return _walk_and_convert_recursive(input_obj, highlight, _force_primitive=force_unicode, _force_recursive=force_unicode_rec)
+    return _walk_and_convert_recursive(input_obj, highlight, _is_primitive=is_unicode_type, _force_primitive=force_unicode, _force_recursive=force_unicode_rec)
 
 
 def _force_default_coding(input_obj, kind, highlight=''):

tests/test_mig_shared_base.py

@@ -58,35 +58,22 @@ def test_encode_within_a_dict(self):
     def test_encode_within_a_list(self):
         output = force_default_fs_coding_rec(['foo', 'bar', 'baz'])
 
-        truth = [force_default_str_coding('foo'),
-                 force_default_str_coding('bar'),
-                 force_default_str_coding('baz')]
+        self.assertEqual(output, [b'foo', b'bar', b'baz'])
 
-        self.assertEqual(output, truth)
-
-    def test_encode_within_a_tuple(self):
+    def test_encode_within_a_tuple_string(self):
         output = force_default_fs_coding_rec(('foo', 'bar', 'baz'))
-        truth = (force_default_str_coding('foo'),
-                 force_default_str_coding('bar'),
-                 force_default_str_coding('baz'))
 
-        self.assertEqual(output, truth)
+        self.assertEqual(output, (b'foo', b'bar', b'baz'))
 
     def test_encode_within_a_tuple_bytes(self):
-        output = force_default_fs_coding_rec(('foo', 'bar', 'baz'))
-        truth = (force_default_str_coding('foo'),
-                 force_default_str_coding('bar'),
-                 force_default_str_coding('baz'))
+        output = force_default_fs_coding_rec((b'foo', b'bar', b'baz'))
 
-        self.assertEqual(output, truth)
+        self.assertEqual(output, (b'foo', b'bar', b'baz'))
 
     def test_encode_within_a_tuple_unicode(self):
-        output = force_default_fs_coding_rec(('foo', 'bar', 'baz'))
-        truth = (force_default_str_coding('foo'),
-                 force_default_str_coding('bar'),
-                 force_default_str_coding('baz'))
+        output = force_default_fs_coding_rec((u'foo', u'bar', u'baz'))
 
-        self.assertEqual(output, truth)
+        self.assertEqual(output, (b'foo', b'bar', b'baz'))
 
 
 class MigSharedBase__force_utf8(MigTestCase):
@@ -95,29 +82,36 @@ class MigSharedBase__force_utf8(MigTestCase):
     def test_encode_string(self):
         output = force_utf8('foobar')
 
-        self.assertEqual(output, truth)
+        self.assertEqual(output, b'foobar')
 
     def test_encode_bytes(self):
         output = force_utf8(b'foobar')
 
-        self.assertEqual(output, truth)
+        self.assertEqual(output, b'foobar')
 
     def test_encode_unicode(self):
         output = force_utf8(u'foobar')
 
-        self.assertEqual(output, truth)
+        self.assertEqual(output, b'foobar')
 
 
 class MigSharedBase__force_unicode(MigTestCase):
     """Unit tests of mig.shared.base force_unicode()"""
 
-    def test_encode_a_string(self):
-        # input_vals = (b'foobar', u'foobar')
-        # truth = u'foobar'
-        # for val in input_vals:
-        #     output = force_unicode(val)
-        #     self.assertEqual(output, truth)
-        pass
+    def test_encode_string(self):
+        output = force_unicode('foobar')
+
+        self.assertEqual(output, u'foobar')
+
+    def test_encode_bytes(self):
+        output = force_unicode(b'foobar')
+
+        self.assertEqual(output, u'foobar')
+
+    def test_encode_unicode(self):
+        output = force_unicode(u'foobar')
+
+        self.assertEqual(output, u'foobar')
 
 if __name__ == '__main__':
     testmain(failfast=True)