[R-pkg-devel] EHRtemporalVariability package check warning, cannot fix

[Ayala Hernandez, Rafael]

Hi Carlos,

Your package seems to use BiocStyle to output the vignette. I have encountered a similar error with my NutrienTrackeR package. The issue was triggered by the latest update to the knitr package, which caused knitting of vignettes in BiocStyle output to fail.

The issue was resolved by the Bioconductor team by pushing a new version of BiocStyle. This fixed all packages dependent on BiocStyle on Bioconductor, and seems to also have fixed the build reports for NutrienTrackeR for all linux and MacOS builds.

However, it has not fixed the error for the windows builds (https://cran.r-project.org/web/checks/check_results_NutrienTrackeR.html) . The error seems very similar to what you are getting, also being triggered by the first line of code in the vignette. If the issue is the same as what I was having, it will be triggered no matter how many chunks you remove, since it is the underlying combination of updated knitr and possibly not latest BiocStyle which triggers it.

Perhaps winbuilder and the machine building CRAN packages under windows are still not using the latest BiocStyle version for some reason?

Best wishes,

Rafa
________________________________
De: R-package-devel <r-package-devel-bounces using r-project.org> en nombre de Carlos Sáez <carsaesi using ibime.upv.es>
Enviado: lunes, 5 de octubre de 2020 18:43
Para: r-package-devel using r-project.org <r-package-devel using r-project.org>
Asunto: [R-pkg-devel] EHRtemporalVariability package check warning, cannot fix


*******************
This email originates from outside Imperial. Do not click on links and attachments unless you recognise the sender.
If you trust the sender, add them to your safe senders list https://spam.ic.ac.uk/SpamConsole/Senders.aspx to disable email stamping for this address.
*******************
Hi,



Some weeks ago we received the CRAN notification to update our
EHRtemporalVariability package in light of some Warnings found on the
vignette compilation.



We managed to solve most by including the �webshot� package in the Suggests
list in DESCRIPTION, since the Vignette is rendered in .html including
dynamic Plotly plots.



However, we cannot manage to solve the last one, and thus we cannot update
the package in CRAN. Local checks show no Warnings (only a Note due to
having 5.1MB file). Even updated R to the latest ver 4.0.2 and R tools to
the latest 4, and all checks ok. But submitting on winbuilder the Warning is
still there.



The logs are here:

 <https://win-builder.r-project.org/9IvFwBBcgDlI>
https://win-builder.r-project.org/9IvFwBBcgDlI



This is the error:



        Error(s) in re-building vignettes:

--- re-building 'EHRtemporalVariability.Rmd' using rmarkdown

Quitting from lines 35-36 (EHRtemporalVariability.Rmd)

Error: processing vignette 'EHRtemporalVariability.Rmd' failed with
diagnostics:

argument is of length zero

--- failed re-building 'EHRtemporalVariability.Rmd'



The package is exactly that found in
https://github.com/hms-dbmi/EHRtemporalVariability



Those lines 35-36 are the first chunk in the .Rmd vignette file:

```{r setup, include=FALSE}

knitr::opts_chunk$set(echo = TRUE, warning=FALSE)

```

Surprisingly, if removing this chunk and resubmitting, the same �argument is
of length zero� error is present but in the following (now first) chunk of
code.



Any help would be really appreciated. We would kindly provide any additional
info that could help.

Best regards!

Carlos


        [[alternative HTML version deleted]]


	[[alternative HTML version deleted]]