jonathansick
commented
2 years ago Thanks for reaching out! We do use this package everyday at Rubin Observatory, so it's not going anywhere. That said, we've tended to implement features and fix issues as we run into them so that's why the development of kafkit probably looks very on-and-off again.
I feel super guilty for not merging and making a new release for #12 so I'm going to deal with it in the next couple of weeks.
A few things our team has in mind for future development is:
- Use tox-docker to run integration tests against a real schema registry (this will help us merge things like #12 more confidently)
- Add httpx support, alongside the existing aiohttp support since we've started to use FastAPI a lot
- Look into integrating dataclasses and/or pydantic (https://github.com/marcosschroh/dataclasses-avroschema) to define schemas and get typed data structures
So although we're probably not going to add new maintainers from outside the observatory, if you want to talk about new features and PRs, that'd be really great. And in return, I'll do my best to integrate those in a much more timely manner.
I also notice that https://marcosschroh.github.io/python-schema-registry-client/ would be a great alternative to kafkit at this point too (I'm not sure it existed when we started Kafkit, but I think we'll be sticking with kafkit).
This is a very promising and appealing library which I'd love to make use of in my projects at work, however I notice maintainership has dropped off in the last year or so. Is this library currently maintained? Are their plans to continue maintainership?
I'm happy to help move issues along and publish updates.