[Bioc-devel] 3 unit tests skipped for CytoPipeline package on Bioc devel

[Martin Morgan]

I looked at the first reported skip test-gating.R:42:5 and see https://code.bioconductor.org/browse/CytoPipeline/blob/devel/tests/testthat/test-gating.R#L42. I googled for the help page of `vdiffr::expect_doppelganger` https://vdiffr.r-lib.org/reference/expect_doppelganger.html and saw the `can` argument — the documentation indicates that the test is disabled on ‘CRAN’, but actually testthat relies on an environment variable ’NOT_CRAN’ to be defined for the test to run https://testthat.r-lib.org/reference/skip.html . `devtools::test()` defines this variable, but R CMD check (and hence the Biocondcutor build system) does not, so the test is skipped. 

Once the test is skipped on line 42, then all the remaining lines of that `testthat()` expression, including several other calls to `expect_doppelganger()`, are also skipped but silently (the code is never evaluated).

You could add `cran = TRUE` as an argument if you disagree with the arguments on `?expect_doppelganger()` as to why this might be a bad idea.

Martin

> On Oct 17, 2024, at 8:57 AM, Philippe Hauchamps via Bioc-devel <bioc-devel using r-project.org> wrote:
> 
> Hi dear members of Bioc-devel community,
> 
> Recently I noticed a 'strange' log in the BUILD/CHECK reports of CytoPipeline, one of the package I am currently maintaining, on Bioc-devel:
> https://bioconductor.org/checkResults/devel/bioc-LATEST/CytoPipeline/teran2-checksrc.html
> 
>    Skipped tests (3)                                                            
> • On CRAN (3): 'test-gating.R:42:5', 'test-ggplots.R:185:5',
>  'test-ggplots.R:310:5'
> 
> [ FAIL 0 | WARN 0 | SKIP 3 | PASS 248 ]
> 
> It seems three of my unit tests (using testthat) are skipped, while I can't remember having deactivated any of my tests using testthat.
> 
> These 3 tests have something in common, they all use the vdiffr package. However, many other unit tests do also use vdiffr (vdiffr::expect_doppelganger()).
> 
> Would you have an clue about what might be happening ?
> 
> Thanks a lot,
> 
> Best wishes,
> 
> Philippe Hauchamps.
> 
> 
> 	[[alternative HTML version deleted]]
> 
> _______________________________________________
> Bioc-devel using r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel


	[[alternative HTML version deleted]]