Asynchronous Reminding Todo app with full async functionality using Telegram bots (Backend-side)

Technologies used:

• FastAPI for API

• PostgreSQL for database

• Async SQLAlchemy as the ORM coupled with Alembic for migrations

• Aiogram for using a previously created telegram bot (used for sending OTPs and reminding user's todos)

• Celery for task scheduling

  • Redis as Celery's results Backend
  • RabbitMQ as the message Broker between the App and Celery

---

Requirements

Telegram Bot

Using @BotFather in Telegram, first create a bot and go through configuring it (BotFather will guide you, making you an offer you can't refuse ;} ). You will be given a Token in the end. Set this token inside your .env file.

Environment Variables

You also need to create a single .env file and specify the below required environment variables:

PostgreSQL

• POSTGRES_USER
• POSTGRES_PASSWORD
• POSTGRES_DB

pgAdmin

• PGADMIN_DEFAULT_EMAIL
• PGADMIN_DEFAULT_PASSWORD

FastAPI

• DB_URL
• SECRET_KEY
• ALGORITHM
• TOKEN_EXPIRATION_IN_SECONDS

Celery

• BROKER_URL
• BACKEND_URL

Note: Put these all inside one file named .env

Note: If launching without Docker, you need to first load the environment variables, e.g. using dotenv module's load_dotenv() method

Note: Secret_key and Algorithm are used for login token handling

Docker

It's highly recommended to launch this using Docker.

Otherwise you are gonna have to install different applications, and use two seperate terminals for launching the app. (one for FastAPI and one for the Celery worker)

---

Finally (using Docker)

Use docker-compose up command to launch the application.

Otherwise (not using Docker)

After setting up PostgreSQL, Redis and RabbitMQ:

in one terminal

run uvicorn main:app

in another terminal

on Linux / MacOS, run: celery -A utils.celery_worker.app worker --loglevel=info

on Windows, run: celery -A utils.celery_worker.app worker --pool=solo --loglevel=info

---

Finally (for real)

Check out localhost:8000 for FastAPI, and its docs (Swagger) at localhost:8000/docs

Using Docker only access pgAdmin at localhost:5050