[R-SIG-Mac] Error: package 'RODBC' was built for universal-apple-darwin8.6.1

[Randall C Johnson [Contr.]]

Hello,
I did install R from the CRAN universal binary, but had installed RODBC from
source. Perhaps there were other problems with my installation as well...?
Anyhow, when I deleted and re-installed R again from the binary and started
out with the RODBC binary everything worked just fine.

Thanks,
Randy


On 7/6/06 9:48 AM, "Simon Urbanek" <simon.urbanek at r-project.org> wrote:

> Randy,
> 
> On Jul 5, 2006, at 11:43 AM, Randall C. Johnson [Contr] wrote:
> 
>> I recently installed RODBC (binary) on my G4 PowerBook, and I am
>> not able to load the library (I thought I had it working at one
>> point). I also tried installing from source, but I have other
>> issues there... Any ideas? I did upgrade to 10.4.7 this morning,
>> could that have something to do with it?
>> 
>>> library(RODBC) # binary installation
>> Error: package 'RODBC' was built for universal-apple-darwin8.6.1
>> 
> 
> It seems that you have compiled R from sources as non-universal
> binary, so you're pretty much on your own, because that is not
> compatible with the CRAN binary of R.
> 
> 
>>> library(RODBC) # source installation
>> Error in dyn.load(x, as.logical(local), as.logical(now)) :
>>     unable to load shared library
>> '/Library/Frameworks/R.framework/Versions/2.3/Resources/library/
>> RODBC/libs/R ODBC.so':
>> 
>> dlopen(/Library/Frameworks/R.framework/Versions/2.3/Resources/
>> library/RODBC/libs/RODBC.so, 6): Symbol not found: _libintl_dgettext
> 
> Pretty much the same answer - try re-compiling R and RODBC (also
> don't forget to delete the previously installed binary package!).
> According to the log you're using fink which supplies its own libintl
> so my guess would be that it's one of the problems. It's not uncommon
> that things break in fink after OS updates, but that's just a wild
> guess, so you may want to update your fink as well (or don't use it
> for R - sometimes it causes more harm than good due to compiler issues).
> 
> If you want to avoid having to compile everything yourself, my actual
> recommendation would be to use the CRAN binary to avoid problems like
> that, so you can use binary packages.
> 
> Cheers,
> Simon
> 
> _______________________________________________
> R-SIG-Mac mailing list
> R-SIG-Mac at stat.math.ethz.ch
> https://stat.ethz.ch/mailman/listinfo/r-sig-mac
> 

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Randall C Johnson
Bioinformatics Analyst
SAIC-Frederick, Inc (Contractor)
Laboratory of Genomic Diversity
NCI-Frederick, P.O. Box B
Bldg 560, Rm 11-85
Frederick, MD 21702
Phone: (301) 846-1304
Fax: (301) 846-1686
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~