[GRASS-stats] rgdal for R 2.7?

Hamish hamish_b at yahoo.com
Wed Mar 10 09:09:03 EST 2010


Hamish:
> > I am trying to get spgrass6 running on Debian/stable which ships
> > with R 2.7.1, and clean up the GRASS/R wiki page while I'm at it:
> >   http://grass.osgeo.org/wiki/R
..
> > I get as far as installing the rgdal package, which complains
..
> > The package page shows that it wants R >= 2.10.0
..
> > I see a bunch of old rgdal versions in the archive; according to the
> > spgrass6 dependency listing anything newer than 0.6-7 (Feb-2009) should
> > be ok.  http://cran.r-project.org/src/contrib/Archive/rgdal/

Roger:
> Check the dates for a version current when 2.7.1 was current and maybe
> get  more recent. Do the same with sp if it complains - the most
> recent one needs >= 2.9.1

No, I just did install.packages("sp") and it did not complain.
library("sp") works ok. shrug.


> Hope this helps,

Yes, thanks. Actually I only had to go back one release to -24. 
Very easy:

 wget http://cran.r-project.org/src/contrib/Archive/rgdal/rgdal_0.6-24.tar.gz
 R CMD INSTALL -l /usr/local/lib/R/site-library rgdal_0.6-24.tar.gz

done!


> PS. Debian policy is very odd, R is three stable releases
> on from there,

note that is only Debian/*stable*. If you want modern stuff debian/testing
and unstable both have R 2.10.  Debian/stable is for people who don't
want to deal with bugs, everything is tested for a long long time before
it is allowed in. Desktop users may groan, but people maintaining rock-
solid low-maintenance servers couldn't be happier. In a general sense
Debian's QA policies are so strict that even the unstable branch is more
stable than most other distros I have used. And then you have Debian/
experimental if you really want bleeding edge... it's just a matter of
choosing your pain threshold. And of course there is backports.org and
Ubuntu to play with if the above don't fit your needs.

(fwiw GRASS in Debian/stable is @ 6.2.3 vs. 6.4.svn40k in /testing and
/unstable)


cheers,
Hamish



      


More information about the grass-stats mailing list