FLASK APP

Overview

This is a Flask Application that has API Endpoints to perform CRUD operations on User data that is stored locally.

Features

• User management (Create, Read, Update, Delete) # refer the User Model
• User search using partial names and cities with pages
• Get detailed summary of the users
• OpenAPI 3.0 Documentation for the end points
• Logging for debugging
• Functional Tests checking the APIs for code 200
• Pre-commit hooks and code quality tools (ruff, black)
• Self Contained dependencies with poetry
• Containerised for easy deployment.
• JWT Authentication using username and password.
• Rate-Limiters for end points.

Prerequisites

Make sure you have the following installed:

• Docker

Setup

1. Clone the Repository

git clone <url>
cd Flask-Assignment

2. Build the Docker Image

docker build -t <container_name> .

3. Start the Docker Container

docker container run -d -p 5000:<port> <container_name>

Note: If you don't want to detach the cli of the container, do not use -d. Use any port that you want to run the flask app on.

Check if Container has stared: (Only in detached mode/different terminal session)

docker container ls

Note down the first 4 characters of the container, as we will use it to access it. It is used as <container_id>

The flask server has started and is accessible at http://localhost:<port> The API Documentation can be found at http://localhost:<port>/apidocs

4. Stopping the Docker Container:

docker stop <container_id>

API Documentation

Once running, access API documentation at:

http://localhost:5000/apidocs

Testing

When you are inside a docker's CLI, testing can be done using

cd App
pytest tests.py -v

Schema of the User Table:

erDiagram
    USERS {
        INT ID PK "Primary Key, Auto Increment"
        VARCHAR(50) FirstName "Not Null"
        VARCHAR(50) LastName "Not Null"
        VARCHAR(100) CompanyName
        INT Age
        VARCHAR(50) City
        VARCHAR(50) State
        VARCHAR(20) Zip
        VARCHAR(100) Email "Unique, Not Null"
        VARCHAR(100) Web
    }

Loading

Logging

Application logs has been created in app.log inside the docker container.

Authentication

To use make changes to the database, you first need to login, this can be done by posting to http://127.0.0.1/login Using the args:

• uid : "admin"
• pass : "1243"

Access Rate Limit

Access rate limit has been setup on the GET end points. Default rate: 10 per hour

Pre-Commit Hooks

Pre-Commit Hooks have been enabled along with black and ruff, which runs automatically when comitting.

Afterwords

How long did it take?

This project took me three nights to complete, as shown in the commit history.

What was the most challenging?

Learning most of the tools used for the first time was the major hurdle. It was more time consuming than I had thought. Learning these and then applying them took most of my time, which would be reflected by the slow commit rates, as I was also looking at the documentations and resources as I coded.

What was unclear?

Most of what I had to do had been clearly instructed, I found no ambiguity.

Any unexpected challenges?

The hardest part was documenting the API using OpenAPI, which took most of my last day. I had not expected a documentation to take so much time.

Is the difficulty appropriate?

Yes, the difficulty is appropriate. However given longer time, the final app and the tests could have been more thorough.

Why the chosen tools?

Most of the tools chosen were for convenience and clear documentation to learn from.

Any assumptions or decisions made?

Testing is only done for 200 response codes, as a very basic implementation.