django-rest-framework-google-json-style-api 1.2.4

Make Google Json Style and Django Rest Framework play nice together.

Django REST Framework Google JSON Style API

Format specification

• https://google.github.io/styleguide/jsoncstyleguide.xml

Installation

At the command line:

$ pip install django-rest-framework-google-json-style-api

Add the render and parser to your django settings file.

# ...
REST_FRAMEWORK = {
    'PAGE_SIZE': 10,
    'DEFAULT_PAGINATION_CLASS':
        'rest_framework_google_json_style_api.pagination.GoogleJsonStylePageNumberPagination',
    'DEFAULT_RENDERER_CLASSES': (
        'rest_framework_google_json_style_api.renderers.JSONRenderer',
        # Any other renders
    ),
    'DEFAULT_PARSER_CLASSES': (
        'rest_framework_google_json_style_api.parsers.JSONParser',
        # Any other parsers
    ),
}
# ...

Goals

By default, Django REST Framework will produce a response like:

{
    "id": 1,
    "username": "scott",
    "full_name": "Scott Chang"
}

and

[
    {
        "id": 1,
        "username": "scott",
        "full_name": "Scott Chang"
    },
    {
        "id": 2,
        "username": "pocheng",
        "full_name": "Pocheng Huang"
    }
]

Google JSON Style Guide shows a response to look like the following:

{
    "data": {
        "id": 1,
        "username": "scott",
        "fullName": "Scott Chang"
    }
}

and

{
    "data": {
        "items":[
            {
                "id": 1,
                "username": "scott",
                "fullName": "Scott Chang"
            },
            {
                "id": 2,
                "username": "pocheng",
                "fullName": "Pocheng Huang"
            }
        ]
    }
}

Google JSON Style Guide uses camel case for field names. So, by default the package uses camel case. If you want to use underscores, you must specify it in your django settings file.

GOOGLE_JSON_STYLE_API = {
    # ...
    'CAMELIZE': False
    # ...
}

Pagination

{
    "method": "list",
    "params": {
        "page": "1",
        "pageSize": "2"
    },
    "data": {
        "currentItemCount": 2,
        "itemsPerPage": 2,
        "totalItems": 200,
        "pageIndex": 1,
        "totalPages": 100,
        "nextLink": "http://example.com/api/v1/?page=2&page_size=2",
        "previousLink": null,
        "items": [
            {
                "id": 1,
                "username": "scott",
                "fullName": "Scott Chang"
            },
            {
                "id": 2,
                "username": "pocheng",
                "fullName": "Pocheng Huang"
            }
        ]
    }
}

Attach meta to data object

Example

Let's add the count of items in the data object.

class AuthorViewSet(viewsets.ModelViewSet):
    queryset = Author.objects.all()
    # ...

    def list(self, request, *args, **kwargs):
        response = super(AuthorViewSet, self).list(request, *args, **kwargs)
        response.data = {
            'meta': {
                # Add meta data in here
                'num_items': self.queryset.count(),
            },
            # Keep original data in results
            'results': response.data
        }

Response

{
    "data": {
        "numItems": 2,
        "items":[
            {
                "id": 1,
                "username": "scott",
                "fullName": "Scott Chang"
            },
            {
                "id": 2,
                "username": "pocheng",
                "fullName": "Pocheng Huang"
            }
        ]
    }
}

Underscoreize Options

There are two conventions of snake case.

Case 1 (default)

v2Counter -> v_2_counter
fooBar2 -> foo_bar_2

Case 2

v2Counter -> v2_counter
fooBar2 -> foo_bar2

By default, the package uses the first case. To use the second case, specify it in your django settings file. The setting only works when you use camel case(default).

GOOGLE_JSON_STYLE_API = {
    # ...
    'JSON_UNDERSCOREIZE': {
        'no_underscore_before_number': True,
    },
    # ...
}