Implementing Data Product API functionality using [a python framework] deployed in Cloud Platform

[tom-webber]

Idea Description - Re-implement some/all of the functionality from the existing DaaP (Data as a Product) using a python framework such as FastAPI, LiteStar, Sanic, Django RF, deployed in Cloud Platform. As with the Lambda Powertools framework, this could be limited to the CRUD metadata/schema endpoints as a proof of concept.

Why Should We Do This - We'll get all of the functionality of the existing API, plus all of the integration features of using an API framework. As with the Lambda Powertools proposal, the hope is that this will reduce the size/complexity of the codebase, and help us to move faster in beta

Definition of Done - A fully or partially featured API available at a cloud-platform domain

What skills do we need? - API Dev, Cloud Platform deployment, API design experience

Working repo

[julialawrence]

Really cool idea! I was wondering if it's worth removing API Gateway from the equation!

[MatMoore]

Even if we don't eliminate API gateway, we could configure it as a proxy with a greedy path pattern so that we don't have to repeat all the routing logic there.

Also, we would grant IAM permissions at the granularity of the API service, rather than having to faff around it with every time we add endpoints.

[michaeljcollinsuk]

If you are using FastAPI think there will be some crossover with #2588 around setup, deployment etc so I'd like to be involved in that capacity at least

[MatMoore]

I'm interested on collaborating on this too (if there's time I can maybe split time between this one and the powertools one to compare/contrast)

[mitchdawson1982]

Interested on collaborating on this one as well as the Michael's Aws role microservice