Skip to content

Update Python Frameworks: Remove Unmaintained Packages and Add Litestar Support #9899

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
pythonhubdev opened this issue May 15, 2025 · 3 comments

Comments

@pythonhubdev
Copy link
Contributor

pythonhubdev commented May 15, 2025

Hey everyone,

I’ve been keeping an eye on the FrameworkBenchmarks project for a while now, and I’ve got to say, it’s such a cool tool for anyone trying to figure out which web frameworks perform best. That said, I’ve noticed a couple of things about the Python section that could use some attention to keep it as useful and accurate as possible.

First off, I think it’d be worth taking a look at the Python frameworks currently in the benchmarks and weeding out any that aren’t being maintained anymore, for example: Japronto. There might be some in there that have been abandoned or replaced by newer options, and if they’re still hanging around, it could throw off the results for folks trying to pick a framework for their next project.

Maybe we could check the last commit dates on their GitHub repos or see if there’s any official word on their status? If it helps, I’d be happy to dig into the current list and flag any that haven’t seen updates in, say, the last year or so.

Second, I wanted to bring up something specific I noticed—Starlite, which I think was benchmarked a while back, has been deprecated for quite some time now. It’s been replaced by Litestar, which is the actively maintained version. From what I’ve seen, like in this Reddit post where the maintainers talked about the rename (they called it LiteStar there, but it’s definitely Litestar now), and the PyPI page showing recent updates, it’s clear Litestar is where the action’s at.

There’s even an issue in the Strawberry GraphQL repo where they’re dropping Starlite support in favor of Litestar, which pretty much seals the deal on that transition.
Ref web: [https://www.reddit.com/r/Python/comments/126owko/starlite_to_become_litestar/]
Ref web: [https://pypi.org/project/litestar/] Ref web: [https://github.com/strawberry-graphql/strawberry/issues/3539]

Since I didn’t spot a Starlite folder in the recent repository—and honestly, it wouldn’t make sense to keep it there anyway—I think adding Litestar would be a great move. It’s production-ready, picking up steam in the Python community, and including it would give a clearer picture of how it stacks up against other frameworks. Plus, I’m kind of excited to see how it performs myself!

So, here’s what I’m suggesting:

Do a quick review of all the Python frameworks in the benchmarks and ditch any that aren’t maintained anymore.
Add support for Litestar to replace the outdated Starlite stuff.

If there’s anything I can do to pitch in—like tracking down unmaintained packages or helping get Litestar set up in the benchmarks—just let me know. I’d love to see this project stay as awesome as it is!

Thanks for taking a look at this, and keep up the killer work!

Best,
Vetrichelvan

@joanhey
Copy link
Contributor

joanhey commented May 18, 2025

I think that is better create a PR to fix it, than create a issue.
This benchmark is open source to any language and framework !!

You are welcome !!

@pythonhubdev
Copy link
Contributor Author

Hi @joanhey thanks for letting me know. Here is a #9904 to remove un-maintained packages.

@pythonhubdev
Copy link
Contributor Author

Hi @joanhey PR #9905 added support for litestar but this is a DraftPR. I need to check with various responses as of now I have ported FastAPI related tests to litestar based but planning to add granian server test as well.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants