Python 3 FAQ

    PyMongo supports CPython 3.4+ and PyPy3.5+.

    Only one intentional change. Instances of are encoded as BSON type 5 (Binary data) with subtype 0. In Python 3 they are decoded back to bytes. In Python 2 they are decoded to with subtype 0.

    Now retrieve the same document in Python 2. Notice the byte string is decoded to Binary:

    2. [GCC 4.7.3] on linux2
    3. Type "help", "copyright", "credits" or "license" for more information.
    4. >>> import pymongo
    5. >>> c.test.bintest.find_one()
    6. {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 . In Python 2 they are decoded to Binary with subtype 0.

    Now decode the same JSON in Python 2 . Notice the byte string is decoded to :

    1. Python 2.7.10 (default, Feb  7 2017, 00:08:15)
    2. [GCC 4.2.1 Compatible Apple LLVM 8.0.0 (clang-800.0.34)] on darwin
    3. Type "help", "copyright", "credits" or "license" for more information.
    5. >>> loads('{"b": {"$binary": "dGhpcyBpcyBhIGJ5dGUgc3RyaW5n", "$type": "00"}}')
    6. {u'b': Binary('this is a byte string', 0)}

    Instances of ObjectId pickled using Python 2 can always be unpickled using Python 3.

    If you need to pickle ObjectIds using Python 3 and unpickle them using Python 2 you must use protocol <= 2:

    1. Python 3.6.5 (default, Jun 21 2018, 15:09:09)
    2. [GCC 7.3.0] on linux
    3. Type "help", "copyright", "credits" or "license" for more information.
    4. >>> from bson.objectid import ObjectId
    6. >>> oid
    7. ObjectId('4f96f20c430ee6bd06000000')
    8. >>> pickle.dumps(oid, protocol=2)
    9. b'\x80\x02cbson.objectid\nObjectId\nq\x00)\x81q\x01c_codecs\nencode\...'
    11. Python 2.7.15 (default, Jun 21 2018, 15:00:48)
    12. [GCC 7.3.0] on linux2
    13. Type "help", "copyright", "credits" or "license" for more information.
    14. >>> import pickle
    15. ObjectId('4f96f20c430ee6bd06000000')