Closed akabla closed 3 years ago
Interesting, TBH codecov is a bit mysterious to me in exactly how it arrives at % lines tested. For example, codecov and coveralls always gave slightly different numbers based on the same data. But I had a read of this which at least goes some way to explaining the difference between codecov/project
and codecov/patch
:
https://docs.codecov.io/docs/commit-status
I normally don't follow it closely but treat it as an approximate measure of repository health.
edit: saw your comment, nice work figuring out the exact change.
Surprising that codecov fails here. Why would this be different now?
I get it - this is because by setting the flags to bypass FW to true all the time, there is a bit of function that is never executed (definitions.jl lines 738 and 746). That's fine as this will be deleted soon enough.
Surprising that codecov fails here. Why would this be different now?
I get it - this is because by setting the flags to bypass FW to true all the time, there is a bit of function that is never executed (definitions.jl lines 738 and 746). That's fine as this will be deleted soon enough.
This link may work to flag the problem. https://app.codecov.io/gh/JuliaRheology/RHEOS.jl/compare/104/diff
I think this is quite impressive actually!
Surprising that codecov fails here. Why would this be different now?