My solution to Opply Technical Challenge.

Table of Contents:

• Tools
• Installation and Usage
• Authentication and Authorization
• Steps on How to Authenticate
• Guide on Endpoint Usage
• API Documentation
• Pagination
• Admin
• Project Limitations
• Deployment Guide

Tools

• Django
• DRF
• JWT Token Authentication
• Django Debug Toolkit
• API Documentation with Swagger UI and Redoc
• Model model for generating tests fixtures and data.

Installation and Usage

Use the package manager pip for installation.

• Clone this repo
• Set up and activate a Python virtual environment with python3 -m venv venv
• Navigate to the src directory where manage.py is located.
• Create a .env file in the src folder, add a secret key parameter e.g: SECRET_KEY=very_secure_randomly_generated_key
• Run pip install -r requirements.txt, you can also use the requirements files in the requirements folder within the core directory.
• Run python manage.py makemigrations
• Run python manage.py migrate
• Create a superuser: run python manage.py createsuperuser
• Run checks python manage.py check
• Run the tests python manage.py test
• Run the server python manage.py runserver
• You can populate the products through the admin panel for testing.

Authentication and Authorization

• The project utilizes JWT Authentication for its token authentication using the djangorestframework-simplejwt package.
• The JWT authentication provides two tokens
• An access token, which should have a shorter expiry time such as 5 minutes, but for testing this project, 1 day is set.
• A refresh token, which should have a longer expiry time, 15 days is set.

Steps on How to Authenticate

• Create a new user using the endpoint ${HOST}/api/customers/new. If you already have a registered user, you can skip this step.
• Make a post request to the login endpoint ${HOST}/api/customers/login using a valid username and password, you will get a refresh and access token.
• Use the access token to send authorized request to endpoints requiring authentication by setting the authorization header either in clients such as Postman, Insomnia or using the Swagger UI and Redoc provided. E.g Authorization: Bearer <access_token>

Guide on Endpoint Usage

There are currently 9 active endpoints. ${HOST} is the address of the local host or the server where it is hosted.

Endpoints	Authentication Required	Method(s)	Action
${HOST}/api/customers/new	False	POST	Create a new user as a customer
${HOST}/api/customers/login	False	POST	Generate an access and refresh JWT token for authentication and authorization
${HOST}/api/customers/refresh	True	POST	Generate a new access token using a refresh token.
${HOST}/api/products/	False	GET	List all available products in a paginated format.
${HOST}/api/products/{id}/	False	GET	Get single product using the id.
${HOST}/api/products/orders/	True	GET	Get a list of orders pertaining to a customer
${HOST}/api/products/orders/	True	POST	Create an order for a product
${HOST}/api/products/orders/{order_id}/	True	GET	Get a single order using the order_id
${HOST}/api/customers/order-history/	True	GET	Get the order history of an authenticated customer

API Documentation

This project has an API documentation with Swagger UI as well as Redoc.

• Access the swagger UI API Doc via ${HOST}/api/swagger/
• Access the Redoc API Doc via ${HOST}/api/redoc/

Pagination

• There is pagination for all list endpoints with a minimum of 10 objects per page. The pagination utilizes a page format.

Admin

• The admin can be accessed via: ${HOST}/admin. Ideally, you would want to populate just the Product and the Custom Users table. The data of every other table is self-generated when using the endpoints including the Order table. You can also create new users via the endpoint for new customers.

Project Limitations

• This project does not go in-depth in User Registration processes since it's not the primary scope of the project. It provides a basic registration procedure with auto activation set for every registered user. Hence, there is no change password, reset password, email activation etc.

• There could be possibility of catching more advance edge cases and validations.

Deploying to Heroku

Here's a short guide on deploying this project to Heroku

• Create a file named Procfile, in it, specify the type of server and the location of you wsgi file in your project using dot notation. E.g with gunicorn will be web: gunicorn core.wsgi.

• Create a "runtime.txt" file, and in it, enter the version of python you want this project to run on. This project was built using Python 3.10.1. E.g of an entry will be python-3.10.1.

• With an active Heroku account, create an app instance, you can deploy directly using Heroku cli or create a CI/CD pipeline with Heroku and Github, which will automatically deploy when you push to the Github repo.