[Rd] packages checks on OS X with 2.12 devel
Prof Brian Ripley
ripley at stats.ox.ac.uk
Tue Aug 10 18:27:12 CEST 2010
On Tue, 10 Aug 2010, Max Kuhn wrote:
> I'm checking packages under the devel version on OS X
>
>> sessionInfo()
> R version 2.12.0 Under development (unstable) (2010-08-08 r52687)
> Platform: x86_64-apple-darwin9.8.0/x86_64 (64-bit)
>
> I install packages via install.packages with type = "source" (I stick
> to arch=x86_64). I've had no issues with installs.
What is very strange is that you seem to have built R with two archs:
why if you don't want to use i386? If you build R with just one arch,
it will only install/check ... that one arch. So this is shooting
yourself in the foot!
> When I check packages via
>
> R --arch=x86_64 CMD check package
>
> I see:
>
> * using platform: x86_64-apple-darwin9.8.0 (64-bit)
> …
> * loading checks for arch ‘i386’
> …
> * loading checks for arch ‘x86_64’
> …
> ** running examples for arch ‘i386’
>
> The check then fails since the packages are not installed on i386. I tried
>
> R --arch=x86_64 --no-multiarch CMD check package
>
> but that still runs checks on x86_64 and i386. I didn't find anything
> else in the "what's new" for the latest devel version.
--no-multiarch is an option to check, not to R.
> I'd prefer not to install the full crane and bioconductor set in both
> architectures. Any suggestions (especially for the documentation that
> I'm sure exists but I can't find).
Rdev CMD check --help and the R-exts manual both cover this.
--
Brian D. Ripley, ripley at stats.ox.ac.uk
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UK Fax: +44 1865 272595
More information about the R-devel
mailing list