Docker is required to run the solution. In order to run the server you need to build the image and run the container.
make build
make run
Building the image may take a moment as it will download the face_recognition Python library which weight is significant.
Once the server is running the app should be accessible on port 8282 on host.
To connect to the websocket use the following URL:
ws://localhost:8282/faces
You can also take a look at the API documentation at http://localhost:8282/docs
To run tests:
make test
The choices that I considered were Django and FastAPI. The advantages of Django that I could see:
- Django's file systems classes and easier file management
Advantages of FastAPI
- Less boilerplate code
- Everything is async and doesn't require a dedicated tool for background tasks / queues.
- Easy to implement websockets
Given that, I chose FastAPI.
Tools such as poetry / pmd / pipenv when used together with Docker almost always require a multi-stage build to do it correctly.
I decided for a basic pip and requirements.txt that works and doesn't require a lengthy Dockerfile configuration.
A few integration tests were added to test the HTTP API.