Lars Grüter

(If you prefer a different place to discuss this, I'm happy to post there!) Hey, thanks everyone for working on this in #87! I've just tested this with the current...

> [...] let's first detail out the usage issues a bit more. [...] can you explain in more detail how you use a debugger and what exactly is not working?...

Would `.mesonpy` contain symlinks to the source tree? That could indeed make synchronization unnecessary. I'm not sure if this is what you mean and how it will effect the debugging...

Just a note: on [v0.13.0rc0](https://pypi.org/project/meson-python/0.13.0rc0/) I replaced the `site-packages` folder in `build-install` with a symlink to my repository root which contains the `skimage` package. This workaround effectively reproduces the old...

Sorry, @mkcor I don't want to come of as devaluing your approach. I'm happy with your solution as long as it gets all edge cases. :) There are a lot...

No idea about the new [error](https://github.com/scikit-image/scikit-image/actions/runs/8819647825/job/24213215782?pr=7322#step:5:208) yet: ``` ../meson.build:1:0: ERROR: Compiler /usr/bin/clang cannot compile programs. ``` But it seems unrelated since this PR doesn't touch any part of the infrastructure..

Considering that there's already been put good work into the [gist](https://gist.github.com/jwindhager/71fa5b149e85d61f83c9613e57d5b3f4) provided by @jwindhager, I'd be happy to review a PR on this. @jwindhager would you be interested and able...

It's tricky to figure out the correct behavior for the edges of the inclusive percentile interval `[p0, p1]`. I think it's still not correct.

@mkcor, see the WIP at https://github.com/scikit-image/scikit-image/pull/7256.

+1 of paying close attention to how many parameters we want to make reachable via position. Not a fan of hard-coding a number. I'd rather recommend to use positional args...