[R] Failing to install {rggobi} on win-7 R 2.12.0

Prof Brian Ripley ripley at stats.ox.ac.uk
Wed Jan 26 08:17:58 CET 2011


On Tue, 25 Jan 2011, Tom La Bone wrote:

> I recall that my problem on Windows was related to having a number of stray
> versions of GTK+ installed. I went back and deleted all versions and
> reinstalled the latest GTK+ and that seemed to fix things. However, when I
> went to do any work of substance ggobi locked up and became unresponsive.
> Never did get it working right on Windows. Had much more luck with R/ggobi
> on Ubuntu 10.10.

I've just been setting rggobi up for our classroom.  It seems that on 
Windows we now need to use Rgui in SDI mode to run rggobi without 
lookups.  (That was not the case last year, so it might be due to the 
change in GTK+ version or it might be due to the change from XP to x64 
Windows 7 on those machines.)

The rggobi binary on CRAN extras is statically linked against 
everything except GTK+, but the www.ggobi.org ggobi DLL needs both 
GTK+ DLLs and libxml2.dll (which needs iconv.dll and zlib1.dll). 
Late last year there was a problem in that GTK+ and libxml2.dll needed 
different zlib1.dll's, but AFAICS this is now resolved by using 
ftp://ftp.zlatkovic.com/libxml/libxml2-2.7.7.win32.zip and 
ftp://ftp.zlatkovic.com/libxml/iconv-1.9.2.win32.zip.  (Unpack those 
and drop the DLLs into somewhere on your path, e.g. the GTK+ bin 
directory.)

We've had a lot of trouble over zlib1.dll: those prepared from zLib 
1.2.3 and 1.2.5 are incompatible.  The whole point of the '1' in the 
name is to change the name in that case!  I suspect very few of those 
benefitting from Windows binary packages have any idea how much work 
goes into circumventing such issues.


-- 
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



More information about the R-help mailing list