Skip to main content

A library for converting between BSON and JSON.

Project description

View build status

See github for the latest source.


Shane Harvey <>


A fast BSON to MongoDB Extended JSON converter for Python that uses libbson.


python-bsonjs can be installed with pip:

$ python -m pip install python-bsonjs


>>> import bsonjs
>>> bson_bytes = bsonjs.loads('{"hello": "world"}')
>>> bson_bytes
>>> bsonjs.dumps(bson_bytes)
'{ "hello" : "world" }'

Using bsonjs with pymongo to insert a RawBSONDocument.

>>> import bsonjs
>>> from pymongo import MongoClient
>>> from bson.raw_bson import RawBSONDocument
>>> client = MongoClient("localhost", 27017, document_class=RawBSONDocument)
>>> db = client.test
>>> bson_bytes = bsonjs.loads('{"_id": 1, "x": 2}')
>>> bson_bytes
>>> result = db.test.insert_one(RawBSONDocument(bson_bytes))
>>> result.inserted_id  # NOTE: inserted_id is None
>>> result.acknowledged
>>> raw_doc = db.test.find_one({'x': 2})
>>> raw_doc.raw == bson_bytes
>>> bsonjs.dumps(raw_doc.raw)
'{ "_id" : 1, "x" : 2 }'


bsonjs is roughly 10-15x faster than PyMongo’s json_util at decoding BSON to JSON and encoding JSON to BSON. See

$ python
Timing: bsonjs.dumps(b)
10000 loops, best of 3: 0.110911846161
Timing: json_util.dumps(bson.BSON(b).decode())
10000 loops, best of 3: 1.46571397781
bsonjs is 13.22x faster than json_util

Timing: bsonjs.loads(j)
10000 loops, best of 3: 0.0628039836884
Timing: bson.BSON().encode(json_util.loads(j))
10000 loops, best of 3: 0.683200120926
bsonjs is 11.72x faster than json_util


Top Level Arrays

Because libbson does not distinguish between top level arrays and top level documents, neither does python-bsonjs. This means that if you give dumps or dump a top level array it will give you back a dictionary. Below are two examples of this behavior

>>> import bson
>>> from bson import json_util
>>> import bsonjs
>>> bson.decode(bsonjs.loads(json_util.dumps(["a", "b", "c"])))
{'0': 'a', '1': 'b', '2': 'c'}
>>> bson.decode(bsonjs.loads(json_util.dumps([])))

One potential solution to this problem is to wrap your list in a dictionary, like so

>>> list = ["a", "b", "c"]
>>> dict = {"data": list}
>>> wrapped = bson.decode(bsonjs.loads(json_util.dumps(dict)))
{'data': ['a', 'b', 'c']}
>>> wrapped["data"]
['a', 'b', 'c']

Installing From Source

python-bsonjs supports CPython 3.7+.


You must build python-bsonjs separately for each version of Python. On Windows this means you must use the same C compiler your Python version was built with.

  • Windows build requires Microsoft Visual Studio 2015


You can download the source using git:

$ git clone


Once you have the source properly downloaded, build and install the package:

$ pip install -v .


To run the test suite:

$ python -m pytest

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

python-bsonjs-0.4.0.tar.gz (181.9 kB view hashes)

Uploaded source

Built Distributions

python_bsonjs-0.4.0-cp37-abi3-win_amd64.whl (117.8 kB view hashes)

Uploaded cp37

python_bsonjs-0.4.0-cp37-abi3-win32.whl (97.7 kB view hashes)

Uploaded cp37

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page