On 14-12-2024 20:12, Miro Hrončok wrote:
On 14. 12. 24 11:35, Sandro wrote:
On 12-12-2024 21:07, Gwyn Ciesla via devel wrote:
I hadn't, but that's a good idea. Here you are.
https://bugzilla.redhat.com/show_bug.cgi?id=2332159
Thanks. I saw a handful of FTBFS/FTI bugs filed by a script[1].
On 14. 12. 24 11:35, Sandro wrote:
On 12-12-2024 21:07, Gwyn Ciesla via devel wrote:
I hadn't, but that's a good idea. Here you are.
https://bugzilla.redhat.com/show_bug.cgi?id=2332159
Thanks. I saw a handful of FTBFS/FTI bugs filed by a script[1]. Obviously
that's just the tip of the icebe
ckages
started to fail when NumPy was updated.
I think it would be helpful to have bugs filed for all packages
currently failing due to NumPy 2.x, either directly or indirectly. Some
will be trivial to fix others will require more work (updates,
bootstrapping, etc). Having the bugs filed and
; > Most FTBFS and FTI issues can be resolved by patching out numpy < 2 pins.
> > Python code using the C API will need to be rebuilt.
>
>
> Do you plan on creating a tracking bug? I think that may prove helpful.
>
> I have blocked a few packages waiting for NumPy 2.x a
rebuilt.
Do you plan on creating a tracking bug? I think that may prove helpful.
I have blocked a few packages waiting for NumPy 2.x against the update
bug [1] while the change proposal was awaiting approval. But I think a
tracking bug would be more suitable.
[1] https://bugzilla.redhat.com/show
In accordance with: https://fedoraproject.org/wiki/Changes/NumPy2
I'm building numpy 2.2.0 for Rawhide/f42 now.
Most FTBFS and FTI issues can be resolved by patching out numpy < 2 pins.
Python code using the C API will need to be rebuilt.
--
Gwyn Ciesla
she/her/hers
--
