[Rd] 1.2 R CMD check

Prof Brian D Ripley ripley@stats.ox.ac.uk
Fri, 8 Sep 2000 22:29:52 +0100 (BST)

On Fri, 8 Sep 2000, Paul Gilbert wrote:

> I've taken a quick look at the 1.2 snapshot to see what is coming. Is it my
> imagination or is it quite a bit faster?

On some machines and tasks, yes. It depends critically on the performance
of malloc we found, to the extent that on Windows we had to find another
malloc to avoid a massive slowdown.

> I noticed that "R CMD check" now seems to do much more: install, check examples,
> check dvi (&pdf?).  I think it is great to have this kind of comprehensive check
> of a package, especially once everything in the package is suppose to be
> working.

Most of the functionality of `build' has been moved to `check'.

> But when I am still in development mode I would also like to be able to run the
> pieces separately so that development is faster, and I don't see how to do
> that.  I can see how to set   --no-latex, but is there a way to avoid re-doing
> install?  (BTW, the usage for check says "Finally, the package is installed into
> the log directory" but I have the impression that the install is the first thing
> that is done.) Or should I consider "R CMD check" as the thing to do first and
> then somehow move the libraries installed in the *.Rcheck log directories into
> $RHOME/library so that everything that depends on them gets the new version? Or
> should I be somehow be setting things so that $RHOME/library is the  log
> directory?

It's not finished yet, you will notice from the perl code.  But yes, it
does do its own install as one of the things it is checking is that a
virgin install gets the results correct.  I just do an install into my
library directory of the version I intend to distribute after check and
build.  Nothing I have seen (not even nlme) takes anything like as long
to install as to check.

Brian D. Ripley,                  ripley@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 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch