Closed jranke closed 2 years ago
A timeout would kill the whole check, not just the tests. This seems like a crash during the test cases. Use a different reporter to see where the crash happens:
test_check(..., reporter = "summary")
OK, thanks. I believe this means I need to run the tests locally. The tests run fine on Linux (local, travis, r-hub) and Winbuilder with R-release, so it seems I need to install R-devel on Windows to see what happens with R-devel on Winbuilder and r-hub...
No, you don't need to run it locally, just modify that line in testthat.R
as I suggested and then at least you will see where it crashes.
Ah, now I see what you mean! Thanks!
Well, I followed your advice, and now I get
✔ checking for unstated dependencies in 'tests'
─ checking tests (850ms)
E Running 'testthat.R' (4m 22s)
Running the tests in 'tests/testthat.R' failed.
Complete output:
> library(testthat)
> library(mkin)
Loading required package: parallel
>
> test_check("mkin", reporter = "summary")
✔ checking for unstated dependencies in vignettes
✔ checking package vignettes in 'inst/doc'
─ checking running R code from vignettes
'FOCUS_D.rmd' using 'UTF-8'... OK
'FOCUS_L.rmd' using 'UTF-8'... OK
'mkin.rmd' using 'UTF-8'... OK
'twa.rmd' using 'UTF-8'... OK
NONE
✔ checking re-building of vignette outputs (18.3s)
✔ checking PDF version of manual (16.9s)
─ Done with R CMD check
─ Cleaning up files and user
I guess it fails in the setup script. I will check using r-devel on Windows.
You need to look at at testthat.Rout
file in the artifacts, that's where testthat's output is. Or testthat.Rout.fail
.
I only get a testthat.Rout.fail
containing
R Under development (unstable) (2022-02-21 r81789 ucrt) -- "Unsuffered Consequences"
Copyright (C) 2022 The R Foundation for Statistical Computing
Platform: x86_64-w64-mingw32/x64 (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.
> library(testthat)
> library(mkin)
Loading required package: parallel
>
> test_check("mkin", reporter = "summary")
and nothing more... It seems testthat does not get to the stage where it reports something.
Yeah, then it does seem like it crashes in the setup file.
Thanks for your help anyways! I learned a lot :)
rhub is cool
Unfortunately this is hard to debug without access to a Windows machine, but at least you can try R-devel on win-builder as well, to see if this might be specific to R-hub, or not.
R-devel on winbuilder also just fails without any more info
R-devel on Windows is up and running over here now, I have a virtual Windows machine that I can use for this kind of stuff.
OK, I need to get Rtools 4.2, how often do you update Rtools on your windows machines?
No often, but rtools version very-very rarely matters. Especially if you don't have any compiled code.
I do rely on compiled code, but it does seem to be something else. Still investigating...
In some cases nlmixr (one of the packages that mkin uses as a backend) crashes the R instance on current R-devel on Windows... Therefore my test setup script never completes and I only get a testthat.Rout.fail
file.
I have no idea why this fails, I don't see any clue in the build log
https://builder.r-hub.io/status/mkin_1.1.0.tar.gz-7a0d47e9722249fc98fc81de87b9acdd
or the artifacts
https://artifacts.r-hub.io/mkin_1.1.0.tar.gz-7a0d47e9722249fc98fc81de87b9acdd/mkin.Rcheck/tests/testthat.Rout.fail
Is there a timeout?