[Rd] "R CMD check" leaving object files.

[Prof Brian Ripley]

On Mon, 7 Apr 2008, Barry Rowlingson wrote:

> If I run "R CMD check" on my package source directory I get no warnings
> (okay, at the moment I get a few, but lets imagine for the sake of
> argument I don't). Then if I run it again I get a warning about my
> source package having object files in it:
>
> * checking if this is a source package ... WARNING
> Subdirectory 'foo/src' contains object files.
>
>  Well, I didn't put them there!

You asked R CMD check to call R CMD INSTALL which did.

>  Can't find anyone else really griping about this on the R search
> stuff. So am I doing something wrong?
>
>  Is there a standard way I can do a 'clean' in the source subdirectory
> before the check?
>
>  Thanks in advance for any ideas, I'm really just putting off sorting
> out the code/documentation warnings...

I always do

R CMD build mypkg
R CMD check mypkg_version.tar.gz

That way I check a clean tarball, the one I might distribute.

What you are seeing is the aftermath of R CMD INSTALL.  That has arguments 
--clean and --pre-clean to deal with this, but 'clean' does not use them.
(My guess is that the faster install by not doing so is thought 
beneficial.)

>
> Barry
>
> ______________________________________________
> R-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel
>

-- 
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