Using Beanie ODM

Summary

Update time! I’m working on this tasks to get things started:

GitHub: Migrate models from homegrown PyMongo to Beanie

I want to move from Flask to FastAPI, and on a few past projects, I’ve enjoyed FastAPI’s native integration with Pydantic models. Since I use MongoDB on the backend, I’ve also used Beanie ODM for persisting my models to my DB.

As I move forward with this project I definitely want to have Administrator CRUD capabilities for user management.

As of now, I’ve rolled my own ODM and while that will work for what it has been used for, as I move to a more robust system, using something async like Beanie will definitely help.

Captains Log

2024-01-12

I needed to re-acquaint myself with Beanie ODM, so I spent the better part of the morning reading through an old project of mine, and the Beanie Docs (to see if anything has changed).

Problem 1: It seems that I forgot how Beanie structures linked documents, and since my homegrown solution won’t map directly to the same document style, I am going to need to migrate my ‘old’ data.

This might end up taking a bit of time to figure out all the bits and pieces of my old DB Schema that need to migrate and work through it.

OK, I’m done for the day – tomorrow will be writing migration scripts and testing them out on my development db.

2024-01-13

I have 6 models that need to be migrated, and today is the first day of the NFL Playoffs for the 2023 season so… I’m not sure how much I’ll get done. I’ll start with the ‘TGFPPick’ model.

Tomorrow’s post will elaborate on what it takes to do a migration from a manual ODM to a proper Beanie ODM (Pydantic) model.