On 6 June 2024 18:47:43 BST, Matthias Koeppe <matthiaskoe...@gmail.com> wrote:
>On Thursday, June 6, 2024 at 8:40:28 AM UTC-7 Dima Pasechnik wrote:
>
>the incident with pytest 8 was very mild.
>
>
>These days, describing something as "very mild" could just mean that it is 
>not regularly leading to fatal hardware damage.
>
>The "incident" was that our use of pytest was completely broken. The 
>failure also caused the CI Build&Test to fail for months (see 
>https://github.com/sagemath/sage/pull/37999). 


It's a good question why nobody added a version pin -  which would have fixed 
it instantly.
But it's a question about our development practices, not about validity of the 
standard pip package approach.


>But yes, pytest is currently optional, and it is only required for testing 
>a very small set of modules in the Sage library. 
>So it is easy to ignore for developers, just like when other optional 
>packages break. 
>
>The proposal is to make it standard, though, and for that the robust 
>machinery of our normal/wheel packages is needed.
>

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/B0D6E92D-CDA3-4067-ACEB-2F83F8A86B54%40gmail.com.

Reply via email to