Codecov Report

Merging #4846 into master will increase coverage by 0.54%.
The diff coverage is 66.66%.

@@            Coverage Diff             @@
##           master    #4846      +/-   ##
==========================================
+ Coverage   94.28%   94.82%   +0.54%     
==========================================
  Files         113      113              
  Lines       25160    25161       +1     
  Branches     2497     2497              
==========================================
+ Hits        23723    23860     +137     
+ Misses       1100      971     -129     
+ Partials      337      330       -7

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update df8869c...3d0ef12. Read the comment docs.

After 1st commit:

increase coverage by 0.99%.

Now without windows:

will increase coverage by 0.31%.

Now:

will increase coverage by 0.46%.

Need to investigate/compare coverage again.

btw: so apparently tests are failing with PYTEST_ADDOPTS=-vv on Travis.. :/
(https://travis-ci.org/pytest-dev/pytest/builds/499355099?utm_source=github_status&utm_medium=notification)

btw: so apparently tests are failing with PYTEST_ADDOPTS=-vv on Travis.. :/

😞

codecov/patch link still times out, with only 5 uploads now.. https://codecov.io/gh/pytest-dev/pytest/commit/11b531be0a38e3c9ce9e43e52fbc3c56fb72b6a5/build
So the base might also be relevant here.. :/

@nicoddemus
But we can see that only ~5 build jobs are necessary to have ~current coverage.

The diff status is pretty much expected, since I've added comments where coverage regressed due to #4839.

codecov/patch link still times out, with only 5 uploads now.. codecov.io/gh/pytest-dev/pytest/commit/11b531be0a38e3c9ce9e43e52fbc3c56fb72b6a5/build
So the base might also be relevant here.. :/

Not sure how to proceed then. Can we reach out to them perhaps? Or even drop the current base and start fresh, if that's the issue?

I'm slacking with them since September last year about this at least.. :/

I will compare this manually again, clean it up, and then we should merge it.

I will bring back the jobs as before, but only ~5 then with coverage reporting.

I also plan to e.g. replace py37 with py37-pexpect (to get one less job) etc.

I'm slacking with them since September last year about this at least.. :/

Bummer. They have no idea why our coverage report view times out?

Yeah, not really apparently.. the suggested it might be flags, or combinations in general.
I've also seen it with other projects though (awesome/awesomeWM), where using less flags appeared to help.. there are less jobs though.

@nicoddemus
FWIW I could invite you on the slack conversation they've created with me, but it would not really help I guess.

Yeah, I agree.

Feel free to merge this then when you clean it up and deem it good enough.

Nearly there - will hopefully finish it tomorrow or on the weekend then.

With master being merged here this also now does not time out anymore on codecov.io.

So in summary flags were to blame then?

So in summary flags were to blame then?

Looks like it, yes.

Good, I'm sure they would like to know that. 👍

It is something they suspect already. But it is not the only factor apparently.

Continued / restarted in #4865.