Andras Deak

Ha :P Like I said, most are nitpicks or tangential. As always, feel free to ignore/dismiss any or all of them. The potentially more substantial points are probably easy to...

I'm very happy if we have both @akaszynski. Most of my concern comes from @banesullivan's comment from earlier: > **With these changes, should we completely change PyVista's Plotting API to...

> > With these changes, should we completely change PyVista's Plotting API to strictly use pipelining? > > The current implementation is already hard to follow, and I'm hoping that...

All good then, thanks for clarifying @banesullivan.

Almost done, thanks for bearing with me @banesullivan. One comment posted just now, and https://github.com/pyvista/pyvista/pull/3318#discussion_r1067534921 still seems open. Neither are blockers.

~~I've also seen some doc build errors but couldn't investigate from mobile yet.~~ Aah, you fixed it in https://github.com/pyvista/pyvista/pull/3318/commits/c374454a0f74963d6d60c124e714e29142d1f7a3.

The (recently appeared) array casting errors are mostly easy to handle, I've fixed almost all locally. The garbage collection issue, on the other hand, is very very hairy. I'm still...

So [we might be innocent](https://github.com/numpy/numpy/issues/26545), or some extra steps are needed now to subclass `ndarray` properly.

NumPy bug confirmed and fix PR already open upstream. We might want to use nightly wheels once it's fixed to be able to use CI.

Actually, no, we'd need nightly builds of the 2.0.0 release branch... `main` is 2.1.0 dev now.