PYTHON-1330 Decode JSON binary type 0 to Python 3 bytes.

bson/json_util.py

@@ -412,6 +412,8 @@ def _get_binary(doc, json_options):
         return uuid.UUID(bytes=data)
     if subtype == UUID_SUBTYPE:
         return uuid.UUID(bytes=data)
+    if PY3 and subtype == 0:
+        return data
     return Binary(data, subtype)
 
 

doc/changelog.rst

@@ -34,6 +34,9 @@ Changes and Deprecations:
   MongoDB >= 3.0. Instead, it returns an empty
   :class:`~pymongo.command_cursor.CommandCursor` to make the behavior
   consistent across all MongoDB versions.
+- In Python 3, :meth:`~bson.json_util.loads` now automatically decodes JSON
+  binary with a subtype of 0 into :class:`bytes` instead of
+  :class:`~bson.binary.Binary`. See the :doc:`/python3` for more details.
 
 Changes in Version 3.4
 ----------------------

doc/python3.rst

@@ -14,7 +14,7 @@ Are there any PyMongo behavior changes with Python 3?
 Only one intentional change. Instances of :class:`bytes`
 are encoded as BSON type 5 (Binary data) with subtype 0.
 In Python 3 they are decoded back to :class:`bytes`. In
-Python 2 they will be decoded to :class:`~bson.binary.Binary`
+Python 2 they are decoded to :class:`~bson.binary.Binary`
 with subtype 0.
 
 For example, let's insert a :class:`bytes` instance using Python 3 then
@@ -42,6 +42,30 @@ to :class:`~bson.binary.Binary`::
   {u'binary': Binary('this is a byte string', 0), u'_id': ObjectId('4f9086b1fba5222021000000')}
 
 
+There is a similar change in behavior in parsing JSON binary with subtype 0.
+In Python 3 they are decoded into :class:`bytes`. In Python 2 they are
+decoded to :class:`~bson.binary.Binary` with subtype 0.
+
+For example, let's decode a JSON binary subtype 0 using Python 3. Notice the
+byte string is decoded to :class:`bytes`::
+
+  Python 3.6.1 (v3.6.1:69c0db5050, Mar 21 2017, 01:21:04)
+  [GCC 4.2.1 (Apple Inc. build 5666) (dot 3)] on darwin
+  Type "help", "copyright", "credits" or "license" for more information.
+  >>> from bson.json_util import loads
+  >>> loads('{"b": {"$binary": "dGhpcyBpcyBhIGJ5dGUgc3RyaW5n", "$type": "00"}}')
+  {'b': b'this is a byte string'}
+
+Now decode the same JSON in Python 2 . Notice the byte string is decoded
+to :class:`~bson.binary.Binary`::
+
+  Python 2.7.10 (default, Feb  7 2017, 00:08:15)
+  [GCC 4.2.1 Compatible Apple LLVM 8.0.0 (clang-800.0.34)] on darwin
+  Type "help", "copyright", "credits" or "license" for more information.
+  >>> from bson.json_util import loads
+  >>> loads('{"b": {"$binary": "dGhpcyBpcyBhIGJ5dGUgc3RyaW5n", "$type": "00"}}')
+  {u'b': Binary('this is a byte string', 0)}
+
 Why can't I share pickled ObjectIds between some versions of Python 2 and 3?
 ----------------------------------------------------------------------------
 

test/test_json_util.py

@@ -276,7 +276,10 @@ class TestJsonUtil(unittest.TestCase):
             strict_uuid=True, uuid_representation=CSHARP_LEGACY))
 
     def test_binary(self):
-        bin_type_dict = {"bin": Binary(b"\x00\x01\x02\x03\x04")}
+        if PY3:
+            bin_type_dict = {"bin": b"\x00\x01\x02\x03\x04"}
+        else:
+            bin_type_dict = {"bin": Binary(b"\x00\x01\x02\x03\x04")}
         md5_type_dict = {
             "md5": Binary(b' n7\x18\xaf\t/\xd1\xd1/\x80\xca\xe7q\xcc\xac',
                 MD5_SUBTYPE)}
@@ -286,6 +289,14 @@ class TestJsonUtil(unittest.TestCase):
         self.round_trip(md5_type_dict)
         self.round_trip(custom_type_dict)
 
+        # Binary with subtype 0 is decoded into bytes in Python 3.
+        bin = json_util.loads(
+            '{"bin": {"$binary": "AAECAwQ=", "$type": "00"}}')['bin']
+        if PY3:
+            self.assertEqual(type(bin), bytes)
+        else:
+            self.assertEqual(type(bin), Binary)
+
         # PYTHON-443 ensure old type formats are supported
         json_bin_dump = json_util.dumps(bin_type_dict)
         self.assertTrue('"$type": "00"' in json_bin_dump)
@@ -370,7 +381,7 @@ class TestJsonUtilRoundtrip(IntegrationTest):
             {'foo': [1, 2]},
             {'bar': {'hello': 'world'}},
             {'code': Code("function x() { return 1; }")},
-            {'bin': Binary(b"\x00\x01\x02\x03\x04")},
+            {'bin': Binary(b"\x00\x01\x02\x03\x04", USER_DEFINED_SUBTYPE)},
             {'dbref': {'_ref': DBRef('simple',
                                ObjectId('509b8db456c02c5ab7e63c34'))}}
         ]