Skip to content

[Bioc-devel] nightly build errors while checking tests

2 messages · Kelly Street, Hervé Pagès

Original
Kelly Street
# 
Hello,

I've recently noticed that the nightly builds for the slingshot package
(both release and devel versions) are failing on merida1, but I can't
figure out why. It has something to do with the tests, but I am unable to
reproduce the error. Here is the build report for the release version:
http://bioconductor.org/checkResults/release/bioc-LATEST/slingshot/merida1-checksrc.html

R CMD check provides this output:

* checking tests ...
  Running ?testthat.R?
 ERROR
Running the tests in ?tests/testthat.R? failed.
Complete output:
  > library(testthat)
  > library(slingshot)
  Loading required package: princurve
  >
  > test_check("slingshot")

And the slingshot.Rcheck/tests/testthat.Rout.fail output is similar:

R version 4.0.5 (2021-03-31) -- "Shake and Throw"
Copyright (C) 2021 The R Foundation for Statistical Computing
Platform: x86_64-apple-darwin17.0 (64-bit)

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.
Loading required package: princurve
There's not a lot of information to go on, so has anybody else seen this
kind of behavior? I went through the list of packages with errors in the
nightly builds and didn't see anything exactly like this, although
the paxtoolsr package was failing in a similar way (seemingly due to a Java
issue?). Any help would be greatly appreciated.

Thanks,
Kelly Street
Hervé Pagès
# 
Hi Kelly,

It looks like the most recent versions of rgl are causing problems (the 
package was updated twice in less than a week on CRAN):

     https://cran.r-project.org/package=rgl

Here is what I did:

- I went on merida1 (our Mac builder for the release builds) and was 
able to reproduce the error we see on the build report for slingshot 
(http://bioconductor.org/checkResults/release/bioc-LATEST/slingshot/merida1-checksrc.html) 
using the latest rgl (0.106.8).

- However after downgrading rgl to version 0.105.22 (archived on CRAN 
here https://cran.r-project.org/src/contrib/Archive/rgl/), I was able to 
run 'R CMD check' on slingshot without problem.

You have a unit test in slingshot ("3D plotting functions work as 
expected") that uses rgl. This test fails with the latest rgl and when 
run in the context of 'R CMD check'. Note however that it doesn't fail 
when I try to run the tests interactively by doing 
test_check("slingshot") at the command line.

Unfortunately the slingshot.Rcheck/tests/testthat.Rout.fail file 
produced by 'R CMD check' does not provide any details about the failure 
so debugging this to understand exactly what's going on would require 
more time and effort than what I'm able to offer at the moment.

Anyways, an easy workaround is to simply skip the problematic test on 
Mac for now. I think you can do this by adding something like this at 
the beginning of the test:

   if (Sys.info()[["sysname"]] == "Darwin")
     skip('latest rgl is causing problem on Mac')

Hope this helps,

H.
On 4/28/21 12:24 PM, Kelly Street wrote: