Classical natural tests

[ssiccha]

No description provided.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 80.39%. Comparing base (2df1f6c) to head (68495a1).
⚠️ Report is 1 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #200      +/-   ##
==========================================
+ Coverage   80.08%   80.39%   +0.31%     
==========================================
  Files          44       44              
  Lines       18381    18382       +1     
==========================================
+ Hits        14720    14778      +58     
+ Misses       3661     3604      -57     

see 5 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[ssiccha]

Unfortunately on my machine, some of the smaller GLs fail. relatively regularly. I'll try to find out which ones that are. I have seen that there already are related issues, so I'll look there too.

[fingolfin]

We now have tst/working/quick/ClassicalNaturalNaming.tst. Might be better to extend that instead. Perhaps also get rid of tst/testclassicalnatural.g; for that, we maybe should...

• make the number of repetitions in TestNaming an argument??
• teach TestNaming to use randomize generating sets (but really, even without that we already find plenty bugs)
• add more input tests; inspired by tst/testclassicalnatural.g and this PR?

[fingolfin]

Removing sl3.tst seems sensible.

But I would not bother moving individual tests from veryslow to quick these days. If we want to have some of those tests run in "quick" mode, I'd suggest we instead move the test to tst/combined and use #@if comments to run parts only in (very)slow mode