[Bioc-devel] packages modifying 'library' during R CMD check

[Sean Davis]

On Wed, Nov 29, 2017 at 9:43 AM, Obenchain, Valerie <
Valerie.Obenchain at roswellpark.org> wrote:

> Hi developers,
>
> Several Bioconductor packages have been flagged as modifying 'library'
> during R CMD check. Thanks to Tomas Kalibera for identifying these.
>
> Behavior falls roughly into 3 categories:
>
> 1) installing/updating packages:
>
> Some packages call biocLite(), install.packages(), update.packages() or
> the devtools equivalent in code accessible from examples or vignettes. This
> modifies packages in .libPaths() which is a problem for reproducibility.
> The user is expecting a snapshot of particular package(s) and this behavior
> changes versions out from under them.
>
> 2) writing to disk instead of tempfile():
>

For some cases where longer-than-R-session storage is needed, there may be
a use for the rappdirs package:
https://CRAN.R-project.org/package=rappdirs
<https://cran.r-project.org/package=rappdirs>

Sean


>
> Packages that write result or intermediate files to disk are in danger of
> overwriting the user's carefully created local files. In some cases, the
> setwd() is modified such that the files are written out in the library when
> the package is installed, e.g.,
>
>    setwd(system.file("extra", package="DiffBind"))
>
> 3) RUnit summary file:
>
> RUnit has an option to output results to "unitTests/reportSummary.txt"
> which is written to the installation tree of a package. Because the output
> summarizes test results, which one presumably wants to see after the R
> process has ended, using tempdir() or tempfile() may not be appropriate.
> Possibly the file should be written to the .Rcheck directory. We aren't
> sure how to best handle this situation yet - Tomas may have some guidance
> eventually.
>
>
> Tomas ran CRAN+BIOC check with read-only "library" directory to prevent
> installation and updates of packages during tests. The packages below
> failed their tests as they tried to write to "library". If you are a
> maintainer of one of these please visit this link for details on the exact
> failure:
>
>   https://github.com/kalibera/rlibwrite
>
> agilp
> alpine
> annmap
> BadRegionFinder
> BBCAnalyzer
> BeadDataPackR
> CoverageView
> crlmm
> DiffBind
> geneRxCluster
> GeneticsPed
> LOLA
> mammaPrintData
> metagenomeSeq
> MinimumDistance
> mQTL.NMR
> multiClust
> oligo
> oligoClasses
> PAA
> plasFIA
> proFIA
> RCAS
> Rdisop
> RImmPort
> SIMAT
> SNPchip
> SRGnet
> triform
> Uniquorn
> VanillaICE
> xcms
> XDE
> xmapbridge
>
> Thanks.
> Valerie
>
>
> This email message may contain legally privileged and/or confidential
> information.  If you are not the intended recipient(s), or the employee or
> agent responsible for the delivery of this message to the intended
> recipient(s), you are hereby notified that any disclosure, copying,
> distribution, or use of this email message is prohibited.  If you have
> received this message in error, please notify the sender immediately by
> e-mail and delete this email message from your computer. Thank you.
>         [[alternative HTML version deleted]]
>
> _______________________________________________
> Bioc-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel
>



-- 
Sean Davis, MD, PhD
Center for Cancer Research
National Cancer Institute
National Institutes of Health
Bethesda, MD 20892
https://seandavi.github.io/
https://twitter.com/seandavis12

	[[alternative HTML version deleted]]