Numpy 2.0.0 problems

[pc494]

This has filtered down from upstream (see pyxem/pyxem#1099). quaternion on which we depend is currently (and may not for some time be) compatible with numpy 2.0.0.

It's possible we could drag and drop a sister package in place (see discussion moble/quaternion#229) but I'm acutely aware this project is really short of dev resourcing.

The alternative is that we make ourselves numpy 1.x for now.

[CSSFrancis]

@pc494 I'm going to try to make the switch this week when I have the time! (Which I should)

[hakonanes]

As I mentioned in #507:

The dependency on numpy-quaternion is NOT important. We only use it for quaternion-quaternion and quaternion-vector multiplication. Our "old" implementation used Numba, and as far as I can remember, it was only twice as slow.
If numpy-quaternion becomes a hassle, I'd suggest we make it an optional dependency: use it when available, but fall back to Numba if it's not. What do you think?

[pc494]

Sorry @hakonanes I had completely forgotten the #507 discussion when I opened this.

[hakonanes]

Nono, you opened this long before my post, I just reposted here to say that we're on it (:

[hakonanes]

orix 0.13.1 is now available from PyPI, with numpy-quaternion as an optional dependency.

But whaddayaknow, latest diffpy.structure is not compatible with NumPy 2.0: diffpy/diffpy.structure#68.

So orix is still not compatible with NumPy 2.0.

[hakonanes]

Reports from the diffpy.structure folks is that NumPy 2.0 compatibility will be addressed in a week's time.

[pc494]

patch to 0.13.2 with a minimum version on diffpy.structure?

[hakonanes]

After they release a new diffpy.structure release compatible with NumPy 2.0? Yes, that sounds like a plan.

[hakonanes]

With #525 and an official release of diffpy.structure v3.2.2 (currently an RC out https://pypi.org/project/diffpy.structure/3.2.2rc0/), orix is compatible with NumPy 2.0 (without numpy-quaternion).