If you are fine with another package doing the legwork for you, calling an R
function from C++ is very easy:
R> library(Rcpp)
R> cppFunction('NumericVector fun(NumericMatrix X, NumericVector y, Function s)
{ return s(X, y); }')
R> set.seed(42); solve(matrix(rnorm(9),3,3), rep(1,3))
[1] -0.778649 1.553893 0.717221
R> set.seed(42); fun(matrix(rnorm(9),3,3), rep(1,3), solve)
[1] -0.778649 1.553893 0.717221
R>
So the C++ function 'fun' we created using Rcpp, and which just calls the
supplied function on the first two arguments, returns us the same answer from
C++ as we get when we call solve(X, y) in R.
Dirk
--
Dirk Eddelbuettel | e...@debian.org | http://dirk.eddelbuettel.com
______________________________________________
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
