The basic idea is not to fail a CI check if the PR branch is not the cause.
Currently,
(1) "Build & Test / test-long" sometimes fails.
(2) It occurs frequently that one of "Build & Test using Conda(Meson) /
Conda" checks fails.
(1) is unfortunate, but this is our main engine checking the PR
I am struggling to understand how test scripts work for sage packages, such
as those developed for the admcycles package, available at
https://gitlab.com/modulispaces/admcycles/-/tree/master/tests?ref_type=heads
Could anyone recommend a primer on test scripts for sagemath packages, why
they
