On 20/09/2025 18:23, Graham Inggs wrote:
Interesting that the release notes for pynormaliz [1] state:
The source code of v2.22 is identical to v2.21
I will re-assign this bug to pynormaliz and remove pynormaliz from
testing, in the hopes of unblocking normaliz, macaulay2 and polymake,
and finishing off the flint transition.
If there is a bug, it would be in normaliz, not pynormaliz. I don't think
reassigning to pynormaliz makes sense in this case. That's only hiding the
possible ABI break.
Emilio
