[Rcpp-devel] About calling C/C++ functions in R

Hi,

I totally second Dirk in the advice about reading the appropriate
documents. I would add the book from John Chambers "Programming with
Data" as it has very clear chapters about R and C(++).

Once you digested these documents, you can read the Rcpp-modules
vignette from Rcpp, which might help you.

Also, I just commited some code in Rcpp (which will be available in the
next version of Rcpp : 0.8.3) that adds the Rcpp::InternalFunction
class, that you can use like this:

// grab the global environment
Environment g = Environment::global_env() ;

// assign the "hello" R variable to the internal function
g["hello"] = InternalFunction( &hello ) ;



Here is a complete example you can run from the R prompt:

require( Rcpp)
require( inline)

inc <- '

const char* hello( std::string who ){
std::string result( "hello " ) ;
result += who ;
return result.c_str() ;
}

'
code <- '
Environment g = Environment::global_env() ;
g["hello"] = InternalFunction( &hello ) ;
return R_NilValue ;
'
fx <- cxxfunction( signature(), code, inc, plugin = "Rcpp" )
f <- fx() # force loading the dynamic library

hello( "world" )

I have not tried it yet from RInside.

Romain

Le 16/06/10 02:13, xiagao1982 a ?crit :

Dear friends,
I am a newcomer of Rcpp and RInside. I installed them in my system and
successfully build the following example:
#include <RInside.h> // for the embedded R via RInside
Rcpp::NumericMatrix createMatrix(const int n) {
Rcpp::NumericMatrix M(n,n);
for (int i=0; i<n; i++) {
for (int j=0; j<n; j++) {
M(i,j) = i*10+j;
}
}
return(M);
}
 >
int main(int argc, char *argv[]) {
const int mdim = 4; // let the matrices be 4 by 4
SEXP ans;
RInside R(argc, argv); // create an embedded R instance
Rcpp::NumericMatrix M = createMatrix(mdim); // create and fill a sample
data Matrix
R["M"] = M; // assign C++ matrix M to R's 'M' var
std::string evalstr = "\
cat('Running ls()\n'); print(ls()); \
cat('Showing M\n'); print(M); \
cat('Showing colSums()\n'); Z <- colSums(M); print(Z); \
Z"; // returns Z
ans = R.parseEval(evalstr); // eval the init string -- Z is now in ans
Rcpp::NumericVector v(ans); // convert SEXP ans to a vector of doubles
for (int i=0; i< v.size(); i++) { // show the result
std::cout << "In C++ element " << i << " is " << v[i] << std::endl;
}
exit(0);
}
Now I add a function in the C++ code:
const char* hello( std::string who ){
std::string result( "hello " ) ;
result += who ;
return result.c_str() ;
}
And I try to call this function in R:
std::string txt = "ret = hello('Friends'); print(ret);";
R.parseEvalQ(txt); // eval string quietly, no result
It fails to do that.
Could you please tell me how to achieve that? Thanks very much!
Gao Xia
------------------------------------------------------------------------
xiagao1982
2010-06-16