JSON Object Strict Typing

Summary

Provides json unmarshalling to strict types for Python. This ensures you always get safe dictionaries for usage. This allows you to always do the below without worrying about KeyErrors.

schema = JSONObject(...)
json_parsed = schema.loads(json_unparsed)
json_parsed['my_key']['my_next_key']

Currently supports basic Python primitives: str, list, dict, float, int, bool

Coming soon: Python DateTime support

Usage

First, start by declaring your schema of objects you expect to get. For example, if I expect to get a user representation,

user_schema = JSONObject({
    username: str,
    id: int,
    name: str,
    active: bool,
    permissions: [
        {
            'resource': 'string',
            'action': 'string',
        }
    ]
})

# Make the relevant API call...
req = requests.get('...')
data = user_schema.loads(req.json()) # The loads signature is indifferent between json.loads(s) and s itself.

# You can now do things like
for perm in data['permissions']:
    print(perm['resource'])
# safely without worrying about KeyErrors

Example

from json_obj import JSONObject

>>> schema = {
  'a': str,
  'b': int,
  'c': float,
  'd': boolean,
}
>>> loader = JSONObject(schema)
>>> loader.loads({
  'a': 34,
  'b': True,
})

{
  'a': '34',
  'b': 1,
  'c': 0.0,
  'd': False,
}