Conversation
|
I can't see anything in here that would cause those loopy failures. Unrelated? Any ideas? :\ |
|
Yes, totally unrelated. Loopy CI got broken by inducer/islpy#150 and then broken a bit more by inducer/pymbolic#186. It'll get fixed up by inducer/loopy#937. As long as it's just the same three tests that are broken, I'm not worried. |
|
This should pass once pyopencl 2025.2 (which I've just released) is on conda-forge. Optimistically tomorrow. |
Thanks! |
|
Woo 🎉 loopy still seems unhappy though, with the same 3 errors as far as I can tell? From my reading of it seems like there's a change in behavior.. that -17 became a -16. Is that ok? EDIT: Just noticed the loopy PR isn't merged yet! Nevermind :D |
Yes, there is.
Good question! This had me concerned, too, so I looked at these changes quite carefully. It turns out it's equivalent. |
This enables pyright and fixed a lot of errors that it complained about (mainly around override and class variables).