BUG: Copy meshgrid after broadcasting #8617
Conversation
44ea15d to
4b2f725
Compare
Also, remove some unused variables
4b2f725 to
d932fcd
Compare
| @@ -179,7 +179,7 @@ def _broadcast_shape(*args): | |||
| supplied arrays against each other. | |||
| """ | |||
| if not args: | |||
| raise ValueError('must provide at least one argument') | |||
| return () | |||
| # use the old-iterator because np.nditer does not handle size 0 arrays | |||
| # consistently | |||
| b = np.broadcast(*args[:32]) | |||
There was a problem hiding this comment.
Is there any reason that np.broadcast() is forbidden? Seems to work just fine with the check for ndim >= 1 removed.
There was a problem hiding this comment.
@mhvk: So you'd be in favor of a separate PR that removes that restriction from the C code?
There was a problem hiding this comment.
I slightly misunderstood -- I agreed with your change here for sure. I'm less sure about messing with the C code, but this is mostly because quite a while ago, on earlier changes in stride_tricks, there was the following #5371 (comment)
2cc1694 to
d9b26f8
Compare
|
This look all OK to me; missing only is a possible addition to the release notes (this one is probably a bit more important to add at all than #8618, which arguably is not something that people have to be specifically alerted to). |
|
AFAICT, this regression was introduced after 1.12. So it doesn't need to make it into 1.12.1, nor does it need a release note entry? |
|
Yes, you're right! So, I think we're set then, and I'll use my newly acquired powers... |
|
@mhvk: I'm curious, what's the upgrade path for acquiring some of those powers? Being able to stick labels on my PRs would be handy |
|
I just was asked a few days ago to join the team, so now I can set labels and merge... I do wonder whether it would not make sense to let everyone at least set labels on their own PRs. |
That's not possible given the not very granular permissions available:
Usually people don't ask but just continue sending and reviewing good PRs until a core dev finally wakes up and approaches them. However, people are very welcome to ask or express the ambition to join the core team (either in private to a core dev or publicly). And since you did just ask: your high activity level hasn't gone unnoticed @eric-wieser! I'll bring it up with the rest of the team and get back to you. |
|
I tried to find a description of how we manage commit rights works in https://docs.scipy.org/doc/numpy-dev/dev/, either in the contributor or governance section, but it's not explained. Will have a look at fixing that. |
Also, remove some unused variables
Fixes #8561