Closed jbcrail closed 7 years ago
@jbcrail do you think travis should fail if flake8 fails ?
Also, Thanks!
Yeah, nice. A flake8 PR == <3 :)
@quasiben Yes, I'm partial to making Travis fail on any flake8 errors/warnings.
When resolving flake8 issues, I found an undefined method call (copy_
) in commit b17ff99. The flake8 error uncovered an issue that would've arisen eventually (the method isn't invoked at this time).
@jbcrail are there other files you want to fix ? should we agree on some linting styles (column limits) and ignore versioneer.py ? I'm +1 on failing travis is linting fails, do you want to add that as well ? Happy to do all these thing but thought I'd ask since you are well into fixing these errors :)
@quasiben I'm finished updating the files now that flake8 passes.
I just noticed that you're using YAPF and certainly want to stick with the project's agreed-upon styling. I had originally set flake8 to use 119 (Django default), but I can update flake8 to use YAPF's defined column width of 120. I've already updated Travis to fail on flake8 errors/warnings. I also can update Travis to run flake8 only on the dask_ec2
directory.
After making these changes, should I run YAPF to format the code?
@jbcrail, apologies for letting this drag on. I think we should merge this and remove YAPF. Thank you so much for this contribution!
merging
Coverage decreased (-0.07%) to 73.981% when pulling c1f53c4d940ed04d117897131cb7c10a6ff80b44 on jbcrail:run-flake8 into 073a9ec924e9054b269e01bcac08766a113c3e4f on dask:master.