that all makes sense! Looking forward to your PR!

The more I hear about and try uv it looks like it will kick Poetry's ass :) You might want to hold off on putting more effort into a poetry based build process right now...

Ah man yeah uv looks great - it's really become more feature complete since I last looked at it ~6 months ago or so

I'm hoping most of the build stuff we have is re-usable - it can use the pyproject.toml as it is for the most part, I suppose it's going to be the funkier stuff like the auto-scm versioning that might need some thinking (unless they have this now too :D)

I was actually working on a branch to clean up the pipeline which I'll put up now so you can at least have a look at it, maybe it's a reasonable interim step (which would make swapping for uv easier ultimately)

^ WIP branch added above, hey at least the tests are green.

I think betting against the Astral guys long term is probably going to be a bad idea :D I might start playing with uv again, even the docs look more polished

I know they also have rye which uses uv but seems to be more all-encompassing, not sure if this will follow the same path 🤔

I've been eyeing up uv for some internal projects just to try it out, I think from the POV of Locust builds, the main issues would be:

• Dynamic versioning isn't currently supported
• Hooks aren't supported (original reason for switching to poetry - pre-building of the UI) (now provided via plugin) as far as I can tell

It's obviously early days, and they will hopefully support this sort of tinkering in the future - but in the meantime for switching we'd need to work a bit differently with respect to the versioning (storing this in source control or strictly using tags) and hooks (multi-step build process for the caller)

For now, I'm calling this done 🥳

Yes, the tagged build worked great as well! Thanks 🙏