Move tests to GitHub Actions #36
Conversation
| @@ -0,0 +1,15 @@ | |||
| name: Cancel | |||
There was a problem hiding this comment.
This workflow will cancel duplicate CI builds. For example, when rapidly pushing commits to a PR, if there are CI builds still running from an older commit they will be cancelled
| os: [windows-latest, ubuntu-latest, macos-latest] | ||
| python-version: [3.6, 3.7, 3.8, 3.9] |
There was a problem hiding this comment.
I went ahead and added a Python 3.9 build as well as windows and macOS builds
ci/environment-3.6.yml
Outdated
| # Dask 2021.3.0 introduced a regression which broke xhistogram. | ||
| # This regression should be fixed in the next Dask release and we | ||
| # should remove the dask version pinning below. | ||
| # see https://github.com/dask/dask/pull/7391 | ||
| # and https://github.com/xgcm/xhistogram/issues/27 | ||
| - dask!=2021.3.0 |
| # see https://github.com/dask/dask/pull/7391 | ||
| # and https://github.com/xgcm/xhistogram/issues/27 | ||
| - dask!=2021.3.0 | ||
| - numpy=1.16 |
There was a problem hiding this comment.
Instead of testing NumPy 1.16 and 1.17 on Python 3.6, the changes here test NumPy 1.16 on Python 3.6 and NumPy 1.17 on Python 3.7. My guess is this isn't a big deal, but let me know if we should changes what's here
Codecov Report
@@ Coverage Diff @@
## master #36 +/- ##
=======================================
Coverage 94.41% 94.41%
=======================================
Files 3 3
Lines 215 215
Branches 58 58
=======================================
Hits 203 203
Misses 8 8
Partials 4 4 Continue to review full report at Codecov.
|
|
This is amazing @jrbourbeau! You're like an open source superhero! 🦸 Is merging #34 what caused the conflicts? I'm happy to revert that, since this supersedes it. |
|
No worries, just needed to merge |
This PR takes a crack at moving the current CI builds that run on Travis CI to GitHub Actions
xref #32