Skip to content

Conversation

@dwreeves
Copy link
Collaborator

ready? 🫡

@dwreeves dwreeves requested a review from ewels September 15, 2025 01:26
@ewels
Copy link
Owner

ewels commented Sep 15, 2025

One last thing.. I was thinking about your integration tests, it's a really nice idea. I checked ClickPy stats to see who the most-trafficked consumers are of the library: https://clickpy.clickhouse.com/dashboard/rich-click

It looks like ScyPy is the biggest, since their v1.9.0 release. Can see it being pulled in here. Though... wait... it was removed again 😆 scipy/scipy#22887 ok, so not such a big deal. (Maybe could offer to help out with spin - see scientific-python/spin#23)

Others are apache-airflow (I guess you tried this already?) and beartype, then it starts to drop off a bit.

The only other project that I have that uses rich-click extensively is nf-core/tools.

Anyway, just a side thought! Not sure that this needs to be a blocker, I was thinking it was worth testing SciPy but if they're not using rich-click then I think we're probably fine to release.

Copy link
Owner

@ewels ewels left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Let's go!

@dwreeves
Copy link
Collaborator Author

dwreeves commented Sep 15, 2025

@ewels I've looked into scipy and Airflow. Both the scipy and airflow CLIs are for dev tools, so not end-user facing for 99.9% of people. I have not tested scipy; for Airflow when they upgraded to 3.0 my entire env broke and haven't fixed it so didn't test 😓 but in both cases it would impact very few people.

I haven't tested scipy because it is tricky to test because I'd have to set up the whole repo (I've been using uv run --with . --with [lib] [command] to do all the integration tests, which makes testing easy.)

I've not actually tested beartype, so I'll double check that.

I am pretty confident our integration testing has hit enough things; I even specifically chose libraries that do RichGroup subclasses.

@dwreeves
Copy link
Collaborator Author

Oh right, it's coming back to me. https://github.com/beartype/beartype/blob/4e978101e77f613efe77b32558324b2dd75e902b/beartype_test/a90_func/a50_external/test_click.py#L27 I did look at beartype. They don't actually use rich-click. They just have a unit-test where they assert that their @beartype decorator works with click commands.

@dwreeves
Copy link
Collaborator Author

@ewels There is nothing more that needs to be done on our end.

There is one bug that was just recently uncovered in esptool, a big library that uses rich-click.

... Unfortunately, this is technically a bug on their end, not ours, and there just isn't anything we can do about it. I am attempting to assist by providing a PR. espressif/esptool#1121

I think what we should do is merge this, but don't click the release button just yet. We should give the esptool maintainers a few hours to do a patch release that implements a fix. The maintainer of the repo is online right now and should get to it soon.

@dwreeves dwreeves merged commit e13d8f1 into ewels:main Sep 15, 2025
54 checks passed
@dwreeves
Copy link
Collaborator Author

Almost ready, but:

Please hold off on full release until esptool maintainers get their fix in or at least have enough time to respond.

😄

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

Successfully merging this pull request may close these issues.

2 participants