FIX: Fix unit example so that we can unpin numpy<2.1 #29616
Conversation
Closes matplotlib#28780. The underlying problem is that operations on numpy scalars try to eagerly convert the other operand to an array. As a result `scalar = np .float64 (2); scalar * radians` would result in a numpy scalar. But we don't want that. Instead we enforce `radians.__rmul__(scalar)` by giving the unit a higher `__array_priority__`. See also https://github .com/numpy/numpy/issues/17650. I haven't found any specific change notes on this in numpy 2.1. Interestingly, the full story is even more complex. Also for numpy<2.1 `radians.__rmul__(scalar)` is not called, but there seems another mechanism through __array__ and __array_warp__ catching back in so that the result is again a TaggedValue. But I have not fully investigated why it worked previously. In fact, we want the solution here with going through __rmul__, and that works for all numpy versions. `
|
Owee, I'm MrMeeseeks, Look at me. There seem to be a conflict, please backport manually. Here are approximate instructions:
And apply the correct labels and milestones. Congratulations — you did some good work! Hopefully your backport PR will be tested by the continuous integration and merged soon! Remember to remove the If these instructions are inaccurate, feel free to suggest an improvement. |
|
Owee, I'm MrMeeseeks, Look at me. There seem to be a conflict, please backport manually. Here are approximate instructions:
And apply the correct labels and milestones. Congratulations — you did some good work! Hopefully your backport PR will be tested by the continuous integration and merged soon! Remember to remove the If these instructions are inaccurate, feel free to suggest an improvement. |
|
Don't see any point of backporting to the doc branch if 3.10.1 is imminent. |
…3.10.x Backport PR #29616 on branch v3.10.x (FIX: Fix unit example so that we can unpin numpy<2.1)
|
It's not urgent but somehow these changes should propagate to the doc branch. How does it work? Is doc reset to 3.10.x when that is released or is 3.10.x merged into doc? |
|
It's merged together before a release, and v3.10.1-doc starts fresh after v3.10.1 |
Closes #28780.
The underlying problem is that operations on numpy scalars try to eagerly convert the other operand to an array. As a result
scalar = np .float64(2); scalar * radianswould result in a numpy scalar. But we don't want that.Instead we enforce
radians.__rmul__(scalar)by giving the unit a higher__array_priority__. See also numpy/numpy#17650.I haven't found any specific change notes on this in numpy 2.1. Interestingly, the full story is even more complex. Also for numpy<2.1
radians.__rmul__(scalar)is not called, but there seems another mechanism through__array__and__array_warp__catching back in so that the result is again a TaggedValue. But I have not fully investigated why it worked previously. In fact, we want the solution here with going through__rmul__, and that works for all numpy versions.