Hi Dirk,
Thanks for your reply, at some point I did consider the possibility of
using Rcpp, but at that time I had already wrote a considerable amount
of regular C code.
Now I am only trying to extend that code a little, translating
everything in Rcpp would probably consume more time (although I agree
it might be easier to use).
I read the "Writing R Extensions" manual, I could only find how to
handle with R lists in C (section 5.9.6 Handling lists), but I could
not find how to <create> a list in C, and access its components, and
this is basically where I got stuck.
Best,
Adrian
On Tue, Nov 26, 2013 at 6:43 PM, Dirk Eddelbuettel <e...@debian.org> wrote:
>
> Hi Adrian,
>
> On 26 November 2013 at 18:12, Adrian Duşa wrote:
> | Dear R-devel,
> |
> | I am trying to do something similar to dynamic length lists in R, but
> | at C level.
> |
> | In R, that would be rather trivial:
> | - determine the length of the list
> | - create the list object
> | - store the values for each component
> | - access value components by using "[["
> |
> | At C level, for a single component where I need to store a vector of
> | length 5, I do:
> |
> | int *p_result;
> | SEXP my_list = PROTECT(allocVector(VECSXP, 1));
> | SEXP result = SET_VECTOR_ELT(my_list, 0, allocVector(INTSXP, 5));
> | p_result = INTEGER(result);
> |
> |
> | The number "1" (the length of "my_list") is dynamic, however.
> | Is there a web reference where I could do some further reading on this
> topic?
>
> If you are open to C++, there is a fair amount of documentation for Rcpp
> which some of us find easier.
>
> Turing-equivalence does of course hold, and all we do here can also be done
> at the C level given that we use the same 'SEXP foo(SEXP a, SEXP b, ...)'
> interface -- while hiding most of it.
>
> Here is a simple case where we define a creator function on the fly, compile,
> link and load it and then have it creates lists of length 2 and 4,
> respectively:
>
> R> library(Rcpp)
> R> cppFunction("List adrian(int n) { return List(n); } ")
> R> adrian(2)
> [[1]]
> NULL
>
> [[2]]
> NULL
>
> R> adrian(4)
> [[1]]
> NULL
>
> [[2]]
> NULL
>
> [[3]]
> NULL
>
> [[4]]
> NULL
>
> R>
>
> Hope this helps, Dirk
>
>
> --
> Dirk Eddelbuettel | e...@debian.org | http://dirk.eddelbuettel.com
--
Adrian Dusa
Romanian Social Data Archive
1, Schitu Magureanu Bd.
050025 Bucharest sector 5
Romania
Tel.:+40 21 3126618 \
+40 21 3120210 / int.101
Fax: +40 21 3158391
______________________________________________
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
