next big code change at work just started. got the unit test written. now i'm on "step 2, draw the rest of the owl" stage
#unit tests
briefly had a moment where my unit test for parsing [NaN] (and comparing to Python's output) was failing ... and then I realized that, like. definitionally it would fail, because NaN != NaN
"oh"
I love mandatory flaky frontend tests on PRs. I love when they run for twenty minutes and then fail for reasons that have nothing to do with my code changes
what is your opinion on colocated unit tests? i will put my thoughts under the fold.
This page's posts are visible only to users who are logged in.