Closed milanmlft closed 1 year ago
I haven't been able to reproduce this also not in my most recent runs where the Bioc version is correct. So I will close this issue for now, as it might just have been a one-off event.
The problem with the inconsistent Bioc versions is still relevant though, but should be fixed in #7.
Thanks a lot for the two issues.
I think the {callr} problem was fixed by #5 and I made that available on a few hours before you created this issue.
Good spot on the Bioconductor version. Hopefully your patch will fix the problem, and I'll try to put a test in my workflow to alert me if it happens again. I'll merge that today.
Hi @grimbough,
Thanks a lot for the great actions, they really simplify my older GHA workflows a lot!
I'm currently running into an error on Bioc-devel though, during the
BiocCheck
step with the messageSee for example https://github.com/milanmlft/msqrob2/actions/runs/3204812867/jobs/5236552594#step:7:76
I've been trying to reproduce this error both locally and using my fork of your biocActionsExamples but so far without much success...
However, I want to point out that in the biocActionsExamples workflow, the Bioconductor version used during
BiocCheck
is not correct for theBioc-devel
instances, e.g. in your latest run forBioc devel (macOS-11)
:https://github.com/grimbough/biocActionsExamples/actions/runs/3205056371/jobs/5237096594#step:7:79
the bioc version indicated is actually 3.15 instead of 3.16... which is weird because in the
Set up R and Bioconductor
step, the version is correct. This might be a reason that the error is not showing up there. I thought that this might be due to some incorrect caching of the package dependencies, but I'm still exploring.