[R] repeated - R package - Now rmutil Compilation Errors

[Roger Bivand]

On Thu, 11 Aug 2005 rab45+ at pitt.edu wrote:

> In my quest to install the repeated package, I have also to install the
> "rmutil" package. (BTW, I'm running Fedora Core 4 and R 2.1.1.) But I now
> get several compilation errors for rmutil and I'm afraid to try to fix
> them:
> 
> [root at ophth-bilora R]# R CMD INSTALL rmutil
> * Installing *source* package 'rmutil' ...
> ** libs
> gcc -I/usr/lib/R/include  -I/usr/local/include   -fPIC  -O2 -g -pipe
> -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -m32 -march=i386 -mtune=pentium4
> -fasynchronous-unwind-tables -c cutil.c -o cutil.o
> gcc -I/usr/lib/R/include  -I/usr/local/include   -fPIC  -O2 -g -pipe
> -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -m32 -march=i386 -mtune=pentium4
> -fasynchronous-unwind-tables -c dist.c -o dist.o
> gfortran   -fPIC  -O2 -g -pipe -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -m32
> -march=i386 -mtune=pentium4 -fasynchronous-unwind-tables -c gettvc.f -o
> gettvc.o
>  In file gettvc.f:55
> 
>             tvcov2(n,i)=-1e301
>                              1
> Error: Real constant overflows its kind at (1)
>  In file gettvc.f:79
> 
>                recx=1e301
>                         1
> Error: Real constant overflows its kind at (1)
>  In file gettvc.f:84
> 
>                reck=1e301
>                         1
> Error: Real constant overflows its kind at (1)
>  In file gettvc.f:86
> 
>             if((recx.ge.1e300).and.(reck.ge.1e300)) ldone=.true.
>                             1
> Error: Real constant overflows its kind at (1)
> make: *** [gettvc.o] Error 1
> ERROR: compilation failed for package 'rmutil'
> 
> I downloaded both repeated and rmutil (for R > 2.0) from:
> 
> http://popgen.unimaas.nl/~jlindsey/rcode.html
> 

Please note that you are using gfortran rather than g77 - they are not the
same, and very possibly differ in how they react to code. The 1e301
constants do overflow the maximum real value (of the order of 3e+38), but
1d301 is less than the maximum double value of the order of 1.7e+308. So
the variable declarations in the code do not seem to match the constants
being used, which probably ought to have been written as "d" rather than
"e" anyway. It doesn't look as though you'll be able to compile this code
with your chosen compiler, could you rather try g77? There seems to be
useful advice on the Fedora lists.

> 
> Rick B.
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
> 

-- 
Roger Bivand
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Helleveien 30, N-5045 Bergen,
Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43
e-mail: Roger.Bivand at nhh.no