Installation and Intro to DRF

Introduction to DRF

DRF, the Django Rest Framework, is a library built on top of Django that enables developers to build web APIs. It comes with a lot of functionalities including authentication packages for both OAuth1 and OAuth2, serializers, pagination, validators, parsers, and lots more. It also allows us to define and structure the URLs. It is used by famous organizations like Mozilla and Red Hat.

How to Install Django REST Framework?

DRF can be installed using the python package manager pip running the command

Windows

Linux

After successful installation, we need to include 'rest_framework' in the settings.py file of our project as shown

RESTful Structure

REST is a Representational State Transfer built to interact with the web through HTTP protocols and methods. The RESTful structure is comprised of a client, a server, and a response. The HTTP request is sent by the client and an HTTP response is received in return by the server.

REST APIs consist of mainly four types of methods, i.e GET, PUT, POST, and DELETE.

Models

As Django follows Model-View-Template architecture, models act as a mediator between the database and the views, that is the views where writing backend scripts we can fetch data from the database and can also update it.

For the instance, let's create a model for a book store to save the name of the book along with its author and page count of the book in the models.py file

Register this model in the admin.py file so that we can access it through the admin dashboard

Now run the migrations command so that this model reflects in the database

Create a super user running the command

After login, you will find a similar UI as shown in the figure

Click on Book models and add some data accordingly then click save.

Now so far we have successfully created a model and added some data.

Now we need to serialize this model.

Serialization

Serializers are used to serialize the data i.e to convert objects, and models into data types that can be stored and transmitted into a form that is understandable by JavaScript and other front-end frameworks.

Here, we will use the Model Serializers which generate fields based on the models and can be accessed by simply implementing the .create() and .update() commands.

To implement this, we need to create a file named serializers.py and write

In this code above, at first, we import serializers from rest_framework. Now we need to import the model which we want to serialize.

Now create a class inheriting the ModelsSerializer from the serializers. Define a Meta class with specific models and the field you want to serialize. In the metaclass of serializers, we define which models and its field are required to be serialized.

As our serializers are ready, we can import them to the views.py file to write the backend logic to access this.

In the views.py file, we need to make some imports first including viewsets from rest_framework and the respective models and serializers

The class BookList inherits the properties of ModelViewSet which is a tool that helps you to define functions that match common actions like CREATE, RETRIEVE, UPDATE, and DELETE.

URLs

Now we need to create URLs to access the APIs we have created.

To implement this, we will use the concept of routers. The router can be defined as the URLs that represent their respective controllers and the view logic written.

In the code written above, we first imported routers from the rest_framework and BookList from the views file. We have used the .register() method which requires two arguments, one is a prefix and the other is the ViewSet, every prefix represents a view set mentioned. Here r'books' is the prefix, which can also be defined as the URL path or the endpoints, and BookList is the model view set we have created.

Now the API URL is ready and we can get all the data from the BookModel by routing the URL GET /books/

Pagination

The concept of pagination allows us to control the number of objects we want to view per page. For this, we need to include the following code in the settings.py file

For now, it is set to 5 which means that it will allow viewing 5 objects of the model per page.

Testing Our API

Now as we have created our Django rest framework application, and created and serialized the models we can test our endpoints.

First run the server, by running the command

and route to http://localhost:8000/ or to the specific development server port.

Now, as mentioned above books is one of the routes registered. So now we can route to /books i.e http://localhost:8000/books/. This will return you to the page displayed below in the image with a UI provided by the Django Rest Framework itself.

Now as you can see, we have retrieved the data from the models using the GET HTTP method.