Device Readings Interview Project

Summary

This app runs a API Rails 7 with Ruby 3.3 all running in Docker. It is structured a using classic MVC rails conventions. To avoid using any disk space, I used ActiveModel for models instead of ActiveRecord. I stored objects in memory using Redis (with only in memory caching).

Assumptions

I assumed that the duplicates timestamps could come in the same request or over multiple requests. In the case where they came in the same request, the first occurrences of the timestamp are kept and all other occurrences are removed.

Improvements

Things I would have done given more time:

• Removed more parts of the boilerplate project that uses views. In hindsight, I should have found a boilerplate project that was for API specifically.
• Created more robust validation on Reading model. Right now, a bad timestamp or count field will through an unhandled error.
• Create more thorough specs. Could have used a Reading model spec, routing specs, more edge cases in existing specs.
• Fill out documentation with more explicit schemas and errors

Setup

Please ensure you are using Docker Compose V2. This project relies on the docker compose command, not the previous docker-compose standalone program.

https://docs.docker.com/compose/#compose-v2-and-the-new-docker-compose-command

Check your docker compose version with:

% docker compose version
Docker Compose version v2.10.2

Initial setup

cp .env.example .env
docker compose build
docker compose run --rm web bin/rails db:setup

Running the Rails app

docker compose up

Running tests

docker compose run --rm web bin/rspec

Documentation

can be found at docs/index.html

Credits

Based on boilerplate app https://github.com/ryanwi/rails7-on-docker