[GRASS-dev] Strange behaviour of some v.* in system() in R

[Markus Neteler]

On Mon, Aug 21, 2006 at 09:15:50PM +0200, Roger Bivand wrote:
> On Mon, 21 Aug 2006, Markus Neteler wrote:
> > On Mon, Aug 21, 2006 at 06:36:30PM +0200, Roger Bivand wrote:
> > > On Mon, 21 Aug 2006, Markus Neteler wrote:
> > > > Dear Roger,
> > > > 
> > > > did you run
> > > > make distclean
> > > > ? There were parser updates which require a clean recompile.
> > > 
> > > Dear Markus,
> > > 
> > > I made a full build from source tarball of the 6.2 pre-beta announced 
> > > yesterday, so there was nothing there before that I'm aware of (installed 
> > > to a new directory, so I hope it isn't picking up dynammically loaded 
> > > *.so from stale versions??). 
> > 
> > It should now - you could just launch the system() command with
> > ldd
> > added to verify.
> 
> Dear Markus,
> 
> Thank you, system("ldd /home/rsb/.../grass-6.2.cvs/bin/v.out.ogr")
> 
> had:
> 
>         libgrass_gis.so => /usr/local/lib/libgrass_gis.so (0xb7b76000)
>         libgrass_datetime.so => /usr/local/lib/libgrass_datetime.so (0xb7b6e000)
> 
> left over from a much earlier version, not the current ones. That seems to 
> have been the problem. With them removed, the spgrass6 functions appear to 
> work again.

Glad to hear that, Roger!
I have taken liberty to CC the list to reduce confusion.

Best,
Markus