Closed neroist closed 3 months ago
this make sense to me, thanks. the only pause of reflection I have is: the additional verbosity might be useful if a test is failing? maybe the answer is no, at the moment I am not entirely sure about it.
Yes I'm also wondering what it looks like when you get a CI failure now. Maybe you could temporarily change a test to fail and we'll see what it looks like? 😄
this make sense to me, thanks. the only pause of reflection I have is: the additional verbosity might be useful if a test is failing? maybe the answer is no, at the moment I am not entirely sure about it.
I see, fine point! In that case switch("define", "nimibNoLog")
may be removed. I don't see a point in keeping the warnings.
Odd, tests fail on my end...
rather remove the warnings than suppress
Were you able to understand CI failures? Not sure I understand the last comment and why closing but you likely now best about it, just curious.
Were you able to understand CI failures? Not sure I understand the last comment and why closing but you likely now best about it, just curious.
Not too much. I think that the tests are better without -d:nimibQuiet
(the logs would be very helpful for debug purposes).
I closed the PR because I'd rather remove the source of the warnings (e.g. unused imports & spacing issues) than just suppress them during compilation. Already done aswell.
The commits explain the changes made -- the purpose for this PR is to make
nimble test
less noisy with less Nim compiler & Nimib printing to stdout. This PR also depends on #242 withswitch("define", "nimibNoLog")
, but that line can be easily removed if needed.getNimibVersion()
is changed as I wanted it to use theargs
parameter similar to the other invocations ofexecProcess()
.