NewlineJSON 0.3

Example

Calling newlinejson.open() returns a loaded instance of newlinejson.Stream(), which generally acts like a file-like object. See help(newlinejson.Stream) for more information.

import newlinejson as nlj

with nlj.open('sample-data/dictionaries.json') as src, \
        with nlj.open('out.json', 'w') as dst:
    for line in src:
        dst.write(line)

with open('out.json') as f:
    print(f.read()))
{'field2': 'l1f2', 'field3': 'l1f3', 'field1': 'l1f1'}
{'field2': 'l2f2', 'field3': 'l2f3', 'field1': 'l2f1'}
{'field2': 'l3f2', 'field3': 'l3f3', 'field1': 'l3f1'}
{'field2': 'l4f2', 'field3': 'l4f3', 'field1': 'l4f1'}
{'field2': 'l5f2', 'field3': 'l5f3', 'field1': 'l5f1'}

Can’t I do everything this module does with one function?

Pretty much - this is the simplest newline delimited JSON API:

import json

def reader(stream):
    for line in stream:
        yield json.loads(line)

with open('sample-data/lists.json') as src, open('outfile.json', 'w') as dst:
    for line in reader(src):
        dst.write(json.dumps(line))

But it doesn’t handle failures and every time it needs to be used it has to be re-written, which means it needs to be packaged, which means it needs unittests, may as well be a little more Pythonic, and now we’re back to this module. Add in transparent compression (in the next version) and its much easier to just import newlinejson and know that it will work rather than mess with a custom solution every time.

Why is this better than MsgPack, Protobuf, or any other packed-binary format?

It probably isn’t. If you’re looking for a module to incorporate into a high capacity data pipeline or bandwidth limited environment you definitely want a packed-binary format. If you’re working with a small amount of local data to produce a one-off product, proofing a workflow, or want to provide additional I/O capabilities to a commandline application reading/writing from/to stdin/stdout, this module is pretty easy to work with.

The goal of this module is to fill a gap in the Python ecosystem in an easy to use and intuitive manner, not to provide highly optimized I/O. If Python’s built-in JSON library isn’t fast enough but newline delimited JSON is the right answer to your problem, one of many faster JSON libraries can be used globally with newlinejson.core.JSON = module or by setting json_lib=module as a keyword argument in open(), load(), etc.

Installing

Via pip:

$ pip install NewlineJSON

From master:

$ git clone https://github.com/geowurster/NewlineJSON.git
$ cd NewlineJSON
$ python setup.py install

Developing

Install:

$ pip install virtualenv
$ git clone https://github.com/geowurster/NewlineJSON
$ cd NewlineJSON
$ virtualenv venv
$ source venv/bin/activate
$ pip install -r requirements-dev.txt -e .
$ nosetests --with-coverage
$ pep8 --max-line-length=95 newlinejson