On Sat, Mar 13, 2010 at 7:09 AM, Romain Francois <[email protected]> wrote: > > Le 13/03/10 13:18, Douglas Bates a écrit : >> >> On Sat, Mar 13, 2010 at 5:45 AM, Douglas Bates<[email protected]> >> wrote: >>> >>> In a recent versions of R-devel calling Rscript from within R or in, >>> say, R CMD INSTALL produces a warning. >>> >>>> system("Rscript -e 'Rcpp:::LdFlags()'") >>> >>> WARNING: ignoring environment value of R_HOME >>> -L/home/bates/R/i686-pc-linux-gnu-library/2.11/Rcpp/lib -lRcpp >>> -Wl,-rpath,/home/bates/R/i686-pc-linux-gnu-library/2.11/Rcpp/lib> >>> >>> This interferes with the standard way of determining the Makevars file >>> for packages that use Rcpp. >>> >>> I'll check if there is a way to suppress that. >> >> For the time being I would suggest piping the output through >> sed /^WARNING/d > > That's odd. I just "svn co" R-devel in fedora and I cannot reproduce this. > > some other potential options are: > --vanilla > 2> /dev/null
You're right that it is not a matter of the version of R, as I initially thought it was. I'm not sure what I am doing to cause this - it may be related to running R within ESS in emacs. In any case, the sed pipe I suggested is benign if there is no warning. _______________________________________________ Rcpp-devel mailing list [email protected] https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/rcpp-devel
