Closed DrChainsaw closed 3 years ago
Merging #33 into master will increase coverage by
0.04%
. The diff coverage is87.50%
.
@@ Coverage Diff @@
## master #33 +/- ##
==========================================
+ Coverage 84.17% 84.22% +0.04%
==========================================
Files 6 6
Lines 316 317 +1
==========================================
+ Hits 266 267 +1
Misses 50 50
Impacted Files | Coverage Δ | |
---|---|---|
src/parallelism.jl | 80.76% <87.50%> (+0.76%) |
:arrow_up: |
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 3a26a5c...74baeae. Read the comment docs.
Looks good to me!
Use case seems to work out at my end as well. Go ahead and merge when ready.
Ugh, just found an issue with mapexec. Just a sec and I'll patch it
I think I'm good now :)
Last commit might be poking a little bit too much into Dagger internals just to test that a context is never accidentally created with Context(). Feel free to exclude it or suggest a better way to test that a fallback never happened.
Travis build failed due to what seems like some random download error when fetching the registry. It seems like I can't restart it though
Looks great! :rocket: Thanks a lot!
Fixes #32
I don’t love the mocking-ish testcase, but I couldn’t think of another way to test that it is propagated correctly.
I chose to only make the exported compute functions (
exec
andcompute
) have non-context versions. Let me know if you prefer it for any other method as well.I will run some more tests for my use case just to make sure and I’ll post right back here once I have done so.