lphuberdeau / hyp

Partial JSON API implementation in Python

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Hyp Build Status

JSON-API responses in Python.

About

Hyp is a library implementing the must parts of the JSON-API response specification. This means that you can use Hyp to serialize your models into responses that contain links and linked compound documents. It works really good in combination with your micro web framework of choice, preferably Flask.

It has built in support for both Schematics and Marshmallow in the sense that you can use any of them for serializing your models (or primitives) into JSON that Hyp creates responses from. To add support for more data serialization libraries such as Colander should be trivial.

Depending on which serialization library that you would like to use make sure to add it to your app's requirements.

Tutorial

First let's define some serializers for your models:

from marshmallow import Schema, fields


class CommentSchema(Schema):
    id = fields.Integer()
    content = fields.String()


class PersonSchema(Schema):
    id = fields.Integer()
    name = fields.String()


class PostSchema(Schema):
    id = fields.Integer()
    title = fields.String()

We can then create our own responders using the hyp.Responders class:

from hyp.marshmallow import Responder


class CommentResponder(Responder):
    TYPE = 'comments'
    SERIALIZER = CommentSchema


class PersonResponder(Responder):
    TYPE = 'people'
    SERIALIZER = PersonSchema


class PostResponder(Responder):
    TYPE = 'posts'
    SERIALIZER = PostSchema
    LINKS = {
        'comments': {
            'responder': CommentResponder,
            'href': 'http://example.com/comments/{posts.comments}',
        },
        'author': {
            'responder': PersonResponder,
            'href': 'http://example.com/people/{posts.author}',
        },
    }

Finally we can use our responders for creating responses. These responses goes perfectly into any Flask application out there:

post = {
    'id': 1,
    'title': 'My post',
    'comments': [
        {'id': 1, 'content': 'A comment'},
        {'id': 2, 'content': 'Another comment'},
    ]
}

json = PostResponder.respond(post, linked={'comments': post['comments']})

Alternatively, the linked items can be automatically collected. The resulting structure will be the same. The LINKS definitions in the responders will be used to identify the linked items in the provided data recursively. If the same element is referenced multiple times (identified by the id only), the last entry will be used.

post = {
    'id': 1,
    'title': 'My post',
    'comments': [
        {'id': 1, 'content': 'A comment'},
        {'id': 2, 'content': 'Another comment'},
    ]
}

json = PostResponder.respond(post, collect=True)

The json variable will now contain some freshly squeezed JSON ready for sending back to the client:

{
    "posts": [
        {
            "id": 1,
            "title": "My title",
            "links": {
                "comments": [1, 2]
            }
        }
    ],
    "linked": {
        "comments": [
            {
                "id": 1,
                "content": "My comment"
            },
            {
                "id": 2,
                "content": "Another comment"
            }
        ]
    },
    "links": {
        "posts.comments": {
            "type": "comments",
            "href": "http://example.com/comments/{posts.comments}"
        }
    }
}

If you'd like to get have dict returned instead of json, for example if you want to use flask's jsonify, then you can use the build method instead:

post = {
    'id': 1,
    'title': 'My post',
    'comments': [
        {'id': 1, 'content': 'A comment'},
        {'id': 2, 'content': 'Another comment'},
    ]
}

response = PostResponder.build(post, linked={'comments': post['comments']})
json = flask.jsonify(response)

About

Partial JSON API implementation in Python

License:MIT License


Languages

Language:Python 99.2%Language:Makefile 0.8%