[Rd] Windows build not running on r-forge

S Ellison S.Ellison at lgc.co.uk
Mon Mar 28 19:07:35 CEST 2011


Thanks for the advice; blindingly simple. I was foxed by the fact that
the R-forge linux 'builds' ran successfully with warning but not error.

Regret that I couldn't provide R version detail, but the problem was
the build at the r-forge end and not my own installed version of R; I
could only tell you what the R-forge log told me. 


Steve Ellison


>>> Prof Brian Ripley <ripley at stats.ox.ac.uk> 28/03/2011 05:32 >>>
On Mon, 28 Mar 2011, S Ellison wrote:

> Please forgive any mis-post, and do feel free to point me to a more
> appropriate list if this isn't properly R-dev.
>
> I have a package on R-forge that shows correct linux and other *nix
> builds, but no windows build. The log for the patched version shows
the
> error below, which appears to be due to a lack of /src files, a
problem
> that does not halt the *nix builds.
>
> The package contains no compiled code (src is intentionally empty).
>
>
> Log as follows:
> * installing to library 'R:/lib/R/CRAN/2.12'
> * installing *source* package 'metRology' ...
> ** libs
>
> *** arch - i386
> Error in file.copy(Sys.glob("src/*"), ss, recursive = TRUE) :
>  no files to copy from
> * removing 'R:/lib/R/CRAN/2.12/metRology'
> Run time: 1.27 seconds.
>
> Advice would be welcome on what I can do about it...?

Don't have an empty 'src' directory: that is not a valid source 
package.

And please do tell us the version of R as per the posting guide.  I am

guessing 2.12.2 here.

>
> Steve Ellison
>
>
> *******************************************************************
> This email and any attachments are confidential. Any\ ...{{dropped:24}}



More information about the R-devel mailing list