Closed k-doering-NOAA closed 5 months ago
These all sound like great things to set up! Please send me a meeting invite for a time that works for you.
@Schiano-NOAA it looks like it is renv that is causing the call-r-cmd-check action to not run all the way through. I tried taking out everything related to renv on this branch and the workflow ran through.
I think what is happening is it sets up the dependencies according to renv, but then doesn't use renv when running check()
(because everything related to renv is in r build ignore), so there is a mismatch. Updating the lock file and ensuring it matches the DESCRIPTION might remove this mismatch, though?
I don't know much about renv, but I wonder if it is useful for a package that I think we will want to work with the most recent versions of packages? Philosophically, I'm not sure. There are a few helpful vignettes available: one for using renv with CI and one for developing packages.
I think there are ways to run gha and keep renv, though, if it feels helpful to you!
In the meantime, running devtools::check()
in your local clone of ASAR should probably show some errors with package checking! I was able to run it in a codespace successfully.
Interesting. I will check this out and see if I can get the error to go away. Thanks for looking more into this.
Issue: dependencies were updated in DESCRIPTION file to reflect those in the renv lock file. The issue persisted.
Solution: renv was disabled and associated files were deleted. Since the package was formally created, the renv environment isn't necessary since dependencies are called and loaded from the DESCRIPTION/NAMESPACE. Other errors found here was also fixed in subsequent edits.
Sounds good - We can definitely troubleshoot using renv in the future if needed!
@Schiano-NOAA would you like github actions to:
If any of these sound useful, perhaps we can schedule a meeting to walk through setting them up with ghactions4r? I think we could set these up in an hour meeting.