FlaskEmp

Flask CRUD app using REST API and MongoDB

Table of Contents

1. Setting up
2. Research
   • Choosing Pydantic
   • To build application for scalability
     • References
     • Take aways
3. Not within the scope of this project

Prerequisites

• Docker (with Docker Compose)
• Postman (for testing the API endpoints)

Setting up

1. Clone the repository
2. Install dependencies

pip install -r requirements.txt

3. Build and run containers

docker-compose up --build

4. The Flask app should now be running on http://localhost:5000 and accessible from Postman or a browser.
5. Go to http://localhost:5000/api/login and send

{
    "username": "admin",
    "password": "admin"
}

6. Copy the token from the response and in Authorization tab send the Bearer Token with that token value as a get request to http://localhost:5000/api/employee be authenticated as admin and see all employees

7. GET http://localhost:5000/api/employee?empid=xyz to see employee with that empid

8. POST http://localhost:5000/api/employee to add employee with schema

    empid: integer
    name: string
    email: email string
    pword: str = Minimum 6 chars long string

9. PUT to http://localhost:5000/api/employee?empid=xyz to update employee with same schema as above

10. DELETE to http://localhost:5000/api/employee?empid=xyz to delete employee

11. POST to http://localhost:5000/api/logout to log out of the admin session

Research

In this I have built the API from scratch

But Flask-Smorest is a powerful extension for Flask that streamlines the development process for robust and well-structured RESTful APIs Can research further on this:
https://rest-apis-flask.teclado.com/docs/flask_smorest/why_flask_smorest/#:~:text=Of%20course%2C%20you%20can%20keep,still%20a%20totally%20viable%20option.
https://datascienceafrica.medium.com/level-up-your-flask-apis-with-flask-smorest-3dc00dd7bc19

Choosing Pydantic

Researched upon which data validation library is the best and by popular opinion it is pydantic because of its immense features and above all scalability as it is 10x faster than some of the other options like marshmallow or jsonschema

To build application for scalability

References(major) :

1. https://www.linkedin.com/pulse/10-tips-write-scalable-flask-applications-vijay-londhe-mxcoc/
2. https://medium.com/@joseleonsalgado/building-scalable-apis-with-flask-best-practices-for-software-engineers-c4305a687ed6

Take awaysS

1. Better structure your application. This means building reusable components and initializing extensions inside a separate function

2. Blueprints help to modularize your application Instead of one app, create separate blueprints for pieces of functionality such as authentication, user management, or APIs

3. The application factory pattern allows you to create multiple instances of your Flask application, each with its own configuration settings

4. Rate Limit API Requests: Use rate-limit solution to avoid abuse and handle spike in traffic. Use Libraries like Flask-Limiter
   https://flask-limiter.readthedocs.io/en/stable/
   Advanced rate limiting: Redis
   I have not used a storage but in production use we have to specify that
   

5. Scalability is also about resilience: Flask-Talisman to enforce security headers, and ensure that user inputs are validated to prevent injection attacks
   https://pypi.org/project/flask-talisman/
   

6. Set up logging with SENTRY or Flask-Logging. Use Prometheus and Grafana to provide observability to monitor performance and test bottlenecks [NOT IMPLEMENTED]

7. Deploy your Flask app in a Docker container for isolating environment dependencies

8. Use caching layers to minimize the stress on your Flask app and database. To cache responses for repeated requests, you can do it via extensions like Flask-Caching [NOT IMPLEMENTED]

9. Secure Your API with JWT https://flask-jwt-extended.readthedocs.io/en/stable/ https://flask-jwt-extended.readthedocs.io/en/stable/blocklist_and_token_revoking.html

Not within the scope of this project:

1. WSGI server: WSGI stands for Web Server Gateway Interface. It's a standard interface that allows web servers to communicate with Python web applications. WSGI is used to forward requests from a web server to a Python web application, and then pass the response back to the web server. Using Gunicorn for apps in production. It is a WSGI HTTP server for UNIX. It's designed to serve Python web applications by handling multiple requests concurrently. Gunicorn can efficiently manage multiple worker processes to handle large loads, making it a great choice for production deployments.

2. Asynchronous Task Execution with Celery