For what it's worth, I suspect this level of checking is only done
for *new* (or returned-from-archiving) packages. I agree that having
an option to have it checked by R CMD check would be valuable.
To repeat a question I've asked before: is there a canonical list
somewhere of the environment variables that control R CMD check's
behaviour, e.g. _R_CHECK_DEPENDS_ONLY_. R_CHECK_ENVIRON, etc. - I
found one in https://cran.r-project.org/doc/manuals/r-devel/R-ints.html#Tools
, is that complete?
On Fri, Jan 16, 2026 at 1:49?PM Josiah Parry <josiah.parry at gmail.com>
wrote:>
> I received a (valid) CRAN rejection which said
>
> "Please add \value to .Rd files regarding exported methods and explain
>
> the functions results in the documentation."
>
>
> My Rd file contained the following which is in the same spirit as the
> \value but misplaced.
>
> \details{
> > \code{encode()} returns a \link[blob:blob]{blob::blob} object which is
a
> > list of raw vectors.
> > If an error is encountered in \code{decode()} the result will be an
NA.
> > }
>
>
> This seems to be a new requirement from CRAN and I've failed this on
new
> packages a handful of times (sorry, team!). It would be great to have this
> for all packages, frankly.
>
> Are there any plans to have R CMD check catch this?
>
> I think it could save everyone time and energy.
>
> Thanks!
>
> - Josiah
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> R-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel