Coverage calculation accounting for concurrency#566
Conversation
Codecov Report
@@ Coverage Diff @@
## develop #566 +/- ##
==========================================
+ Coverage 70.44% 70.55% +0.1%
==========================================
Files 64 64
Lines 5518 5518
Branches 767 767
==========================================
+ Hits 3887 3893 +6
+ Misses 1425 1420 -5
+ Partials 206 205 -1
Continue to review full report at Codecov.
|
wtgee
left a comment
There was a problem hiding this comment.
LGTM but was almost expecting a huge jump in coverage. :) Will be nice to be more accurate. Will leave for @jamessynge to see if he has comments.
|
@wtgee It would make more difference if |
jamessynge
left a comment
There was a problem hiding this comment.
I don't have an opinion yet as to how well it works, but would like to try!
I am surprised that we don't see more of a jump, but I'll take a look once merged and I can try it on my laptop.
|
I'm looking forward to trying this out, and hopefully getting some improved coverage. |
While working on #547 I ran into problems with inaccurate test coverage reports because the existing
coverageand Travis CI configurations produced results which did not take into account concurrency, in particular scripts started withsubprocess. I was able to come up with some changes to the configurations that appear to resolve these problems, and should properly calculate test coverage involvingmultiprocessing,threadingandsubprocessconcurrency. At @wtgee's suggestion I have split these config changes out into a Pull Request of their own.