I knew what you posted in the previous email and I did not ask to you
post it. Why did you complain that you were tied of it?
The straight forward answer to my question in the title is that
2013-01-05-r-function-from-c++.cpp should not be directly put to src/
direction in an R package. The function in that file has to be wrapped
in a function like the following to be callable in R. Isn't this
answer very simple and straightforward to my question? I don't see why
you spent so much time without directly addressing the question. Is
there anything not clear in the way that I stated my question?
SEXP myfun(SEXP f, SEXP x){
...
}
On Sat, May 25, 2013 at 6:31 PM, Dirk Eddelbuettel <e...@debian.org> wrote:
>
> On 25 May 2013 at 18:05, Peng Yu wrote:
> | > You. Are. Still. Doing. It. Wrong.
> |
> | I know that I did was wrong. Otherwise it should have worked. What's
> | the point of reiterating it and emphasizing it? Something more
> | constructive is more helpful.
>
> You still don't know what functions to call when. Here are two files. Copy
> and paste from between the dashed lines.
>
> --file 1: pengyu.cpp ------------------------------------------------------
>
> #include <Rcpp.h>
>
> using namespace Rcpp;
>
> // [[Rcpp::export]]
> NumericVector callFunction(NumericVector x, Function f) {
> NumericVector res = f(x);
> return res;
> }
> -----------------------------------------------------------------------------
>
> You will note that this is the _same code as in the Rcpp Gallery post_.
>
> Use it together with file 2:
>
> --file 1: pengyu.R ------------------------------------------------------
> library(Rcpp)
>
> set.seed(42)
> x <- rnorm(1e5)
> print(fivenum(x))
>
> sourceCpp("pengyu.cpp")
> print(callFunction(x, fivenum)
>
> cat("Done\n")
> -----------------------------------------------------------------------------
>
> Use it as follows:
>
> -----------------------------------------------------------------------------
> edd@max:/tmp$ Rscript pengyu.R
> [1] -4.04327635 -0.68238450 -0.00206637 0.67332471 4.32809127
> [1] -4.04327635 -0.68238450 -0.00206637 0.67332471 4.32809127
> Done
> edd@max:/tmp$
> -----------------------------------------------------------------------------
>
> You could have run this straight from the Rcpp Gallery as well. Here I use
> the URL for the raw file, download to a given filename and then call
> sourceCpp() on that file:
>
> -----------------------------------------------------------------------------
> edd@max:/tmp$ wget
> https://github.com/jjallaire/rcpp-gallery/raw/gh-pages/src/2013-01-05-r-function-from-c%2B%2B.cpp
> -OgalleryPost.cpp -q
> edd@max:/tmp$ Rscript -e 'library(Rcpp); sourceCpp("galleryPost.cpp")'
>
> R> set.seed(42)
>
> R> x <- rnorm(1e+05)
>
> R> fivenum(x)
> [1] -4.04327635 -0.68238450 -0.00206637 0.67332471 4.32809127
>
> R> callFunction(x, fivenum)
> [1] -4.04327635 -0.68238450 -0.00206637 0.67332471 4.32809127
> edd@max:/tmp$
> -----------------------------------------------------------------------------
>
> I have not hidden any secrets from you. You just managed to completely
> misunderstand a few things, and then decided that you should just
> semi-randomly guess a few new commands.
>
> Dirk, really tired of this
>
> --
> Dirk Eddelbuettel | e...@debian.org | http://dirk.eddelbuettel.com
--
Regards,
Peng
_______________________________________________
Rcpp-devel mailing list
Rcpp-devel@lists.r-forge.r-project.org
https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/rcpp-devel
