Kylie provides mappings between JSON data structures and Python objects.
Project description
Kylie provides mappings between JSON data structures and Python objects. It provides a reasonable amount of power with only a tiny bit of magic, and it has 100% code coverage.
Free software: BSD license
Documentation: https://kylie.readthedocs.org/
It supports name and type-mapping, and it’s not bound to JSON, although it is geared to serializing and de-serializing data to and from JSON-type data structures such as dicts, lists, strings, numbers, and bools.
Example
class SpanishInquisitionModel(Model): inquisition_id = Attribute('id') expected = Attribute(python_type=bool, serialized_type=int) >>> surprise = SpanishInquisitionModel(inquisition_id=1234, expected=False) >>> surprise.inquisition_id 1234 >>> surprise.serialize() {'id': 1234, expected=0}
Note that the attribute inquisition_id becomes the dict key "id", and expected is mapped to 0 instead of False.
We can now take this dict, dumps it to JSON, and somewhere else call the following on the json_data (which is a dict returned from loads):
>>> my_surprise = SpanishInquisitionModel().deserialize(json_data) >>> my_surprise.inquisition_id 1234 >>> my_surprise.expected False
Kylie also supports nested models, so you can embed other Model instances inside the data, and Kylie will manage serialization and deserialization of them for you. See the docs for details.
Features
Allows properties to have different names between Models and python dictionary keys.
Allows type conversion when serializing and de-serializing objects.
Automatic serialization/deserialization of attributes that are also Models.
Not bound to JSON in any way, and should also be useful for MessagePack
Non-Features
So what doesn’t Kylie do yet? Well, there are quite a few things, because it’s very new:
Can’t currently map to lists of nested objects (well, you can, but you have write the attribute mapper functions yourself.
Doesn’t have any mechanism for validation. I am planning to add this, once I decide the best way to do it. Ideas welcome!
Doesn’t have any built-in mechanism for choosing between different types to deserialize a dict to. Again, you can do this through customized type mapper functions. At the very least, I’ll document this soon. I plan to add the ability to automatically deserialize to a type based on, for example, a __type__ item in the dict.
No post-serialize or post-deserialize options, unless you do it yourself. This would allow wiring up of objects that are referred to by id and provided elsewhere in the serialized data-structure, for example.
So, lots to do! The amount of non-features currently outweighs the list of features, but I think Kylie is already useful.
Why is it called Kylie?
I named it especially for you. Back in the late 80’s (I’m old!) Kylie and Jason were today’s Kim and Kanye. This Kylie works well with JSON. Geddit?
History
0.1.0 (2015-04-12)
First release on PyPI.
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
Built Distribution
Hashes for kylie-0.1.1-py2.py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 0d61006235e2fe64bb04783aa6872c5735df1224a0b1dc1e4ff8047c1a5cb679 |
|
MD5 | dc332508672843cdb7dbaaa00bc9dd63 |
|
BLAKE2b-256 | c504e7020a54ceb6b0bbdd7676e38ff196b3d1ed0f95bf2770e11c1f17c18b0c |