Canvas test failures

[gap-packages]

Canvas test failures

[jamesjer]

This also affects the output of GetLinks(graph); in graph.tst.

[mcmartins]

That's true, the ID are uuids. The weird thing is that locally and on the github actions the tests were passing, e.g.: https://github.com/gap-packages/francy/actions/runs/4237207803/jobs/7363113640

I've added gap> RANDOM_SEED(1); on all tests in the develop branch (which I believe should do the trick).

[jamesjer]

That's interesting. I see that if I run the tests multiple times, I get the same ID in each run. But if I completely remove the buildroot and spin up a new one, I get a different ID, which also persists across multiple tests. Does that mean that the random number generator is given a predictable seed? If so, then RANDOM_SEED(1) should indeed do the trick.

[mcmartins]

Yes, that should do the trick if the tests are ran on the same "session".
Though, for the system packages release tests run twice which then fails, see comment on #92
I've removed the offending test cases.