[R-pkg-devel] Warning when the --as-cran flag is used, not otherwise.

David Winsemius dw|n@em|u@ @end|ng |rom comc@@t@net
Wed Jan 17 08:19:54 CET 2024


Speaking as a potential user of such a package, I wonder if that 
information could be put in a help page for the entire package? I've 
seen several packages that do such a "global" overview. I realize there 
might be a Suggests: entry in the DESCRIPTION file but not everyone 
knows how to find it and even those who do might first try for a help 
page if it was in the Index.

On 1/16/24 22:39, Rolf Turner wrote:
> On Tue, 16 Jan 2024 23:53:15 +0100
> Sebastian Meyer <seb.meyer using fau.de> wrote:
>
>> There is a note on your help page that says
>>
>> These data are **not** immediately available in the `eglhmm` package.
>>
>> which seems to be in line with the check warning.
> OK.  Thanks.  Now that you have pointed this out, it's all too obvious.
> The user is invited to install the monoCyteSim package (from a specified
> github repository).  After the user has done so, then the data sets
> bivarSim and ccSim are available in the usual way.
>
> When I ran R CMD check without the --as-cran flag, there was no problem
> because I had already installed the monoCyteSim package on my local
> computer.  *With* that flag, R CMD check carries on as if it were
> looking at the eghlmm package in the context of running on a CRAN
> server.  This server would *not* have the monoCyteSim package
> installed, whence it would not be able to find the data sets bivarSim
> and ccCM, whence the warning.  Makes perfect sense now.
>
> It turns out I can make the warning go away by removing the lines
>
> \usage{
>      bivarSim
>      ccSim
> }
>
> They serve no useful purpose and are not required.
>
> Thanks again.
>
> cheers,
>
> Rolf Turner
>



More information about the R-package-devel mailing list