How to create a great local Python development environment with Docker, and a TODO application.

Work in progress

1. Why Docker?
2. Dockerize an app
3. Immediate file changes (volumes)
4. Use IDE in Docker
5. Docker Compose
6. Add more services (Redis)
7. Debug Python code inside a container

1. Why Docker?

Python versions, more than just a virtual env, ...

2. Dockerize an app

FROM python:3.10-slim
WORKDIR /code
COPY ./requirements.txt ./
RUN pip install --no-cache-dir --upgrade -r requirements.txt
COPY ./src ./src
CMD ["uvicorn", "src.main:app", "--host", "0.0.0.0", "--port", "80", "--reload"]

Use Dockerfile and the code in scr directory

(Use slim or alpine for smaller versions)

docker build -t fastapi-image .
docker run --name fastapi-container -p 80:80 fastapi-image
docker run -d --name fastapi-container -p 80:80 fastapi-image

3. Immediate file changes (volumes)

 docker stop fastapi-container
 docker ps -a
 docker rm fastapi-container

 docker run -d --name fastapi-container -p 80:80 -v $(pwd):/code fastapi-image

4. Use the IDE to jump into a python venv

Running this project in a python venv and use pip to install our deps.

python -m venv venv
. venv/bin/activate
pip install -r requirements.txt

We can then run this locally like before by using this console command.

uvicorn src.main:app --reload

Or just copy the same command from part 2/3 to be more explicit in port selection.

After this we can use the command:

deactivate

to jump out of our venv and back to our host machine.

5. Docker Compose

services:
  app:
    build: .
    container_name: python-server
    command: uvicorn src.main:app --host 0.0.0.0 --port 80 --reload
    ports:
      - 80:80
      - 5678:5678
    volumes:
      - .:/code

Use docker-compose.yml

docker-compose up
docker-compose down

6. Add more services (Redis)

services:
  app:
    ...
    depends_on:
      - redis

  redis:
    image: redis:alpine

7. Debug Python code inside a container

Add this in the code

import debugpy

debugpy.listen(("0.0.0.0", 5678))

# print("Waiting for client to attach...")
# debugpy.wait_for_client()

And the port in yml:

services:
  app:
    ...
    ports:
      - 80:80
      - 5678:5678

Attach to running container

8. Docs are generated from FastAPI

We just need to go to /docs from our url to find all our endpoints automatically have been ported to Swagger.

Try a Python version easily with Docker

 docker pull python:3.11-slim
 docker run -d  -i --name python_dev python:3.11-slim
 docker exec -it python_dev /bin/sh

Further Resources:

• https://towardsdatascience.com/debugging-for-dockerized-ml-applications-in-python-2f7dec30573d
• https://github.com/Wyntuition/try-python-flask-redis-docker-compose
• https://docs.docker.com/compose/gettingstarted/
• https://www.docker.com/blog/containerized-python-development-part-1/
• https://www.uvicorn.org/
• https://fastapi.tiangolo.com/

Other commands for cleaning up

docker rm container_name
docker image rm image_name
docker system prune
docker images prune

Check folder size:

du -sh *