Le 15/06/10 20:45, Dominick Samperi a écrit :
I have tried something like
Rcpp::NumericVector nm(Rcpp::Dimension(2,3,4));
I get this:
> require( Rcpp )
Le chargement a nécessité le package : Rcpp
> require( inline )
Le chargement a nécessité le package : inline
> fx <- cxxfunction( signature(), ' return NumericVector(
Dimension(2,3,4) ) ;', plugin = "Rcpp" )
>
> fx()
, , 1
[,1] [,2] [,3]
[1,] 0 0 0
[2,] 0 0 0
, , 2
[,1] [,2] [,3]
[1,] 0 0 0
[2,] 0 0 0
, , 3
[,1] [,2] [,3]
[1,] 0 0 0
[2,] 0 0 0
, , 4
[,1] [,2] [,3]
[1,] 0 0 0
[2,] 0 0 0
which looks about right. although not much useful without some more
support such as extract or set the values.
But there seems to be a problem due to non-public fields.
What do you mean ?
Am I misunderstanding the purpose of Dimension() here?
Perhaps. Perhaps not. You did not tell us what you understand.
On Tue, Jun 15, 2010 at 2:19 PM, Romain Francois
<[email protected] <mailto:[email protected]>> wrote:
Le 15/06/10 20:08, Douglas Bates a écrit :
I may be confusing Armadillo classes and Rcpp classes but I seem to
recall that there is an Rcpp class for arrays with more than 2
dimensions. However, I can't find anything that looks like that in
the declarations or in the unit tests. I can easily create such an
object by installing a dim attribute on a Vector type but I didn't
want to reinvent the wheel.
Is there such a class or am I misremembering?
We do not have invented that wheel. It should not be too hard to
generalize what we do with Rcpp::Matrix
You can ignore the member functions that have diag in their names.
although, we would probably also have to extend the
Vector::operator() for more dimensions, and probably also
Vector::offset.
I feel somewhat lazy here, but if you or someone else decides to
pick it up, I will have a look.
Romain
--
Romain Francois
Professional R Enthusiast
+33(0) 6 28 91 30 30
http://romainfrancois.blog.free.fr
|- http://bit.ly/98Uf7u : Rcpp 0.8.1
|- http://bit.ly/c6YnCi : graph gallery collage
`- http://bit.ly/bZ7ltC : inline 0.3.5
_______________________________________________
Rcpp-devel mailing list
[email protected]
https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/rcpp-devel
