[rkward-devel] browser in rkward

[Romain Francois]

Thomas Friedrichsmeier wrote:
> On Tuesday 26 May 2009, Romain Francois wrote:
> > It is probably better if you proceed first with the modifications you
> > have in mind.
>
>
> Ok, I'm done with that for now. There's more cruft to clean up, but 
> that's for another day. Also, on and off, I'm working on getting 
> rembedinternal.cpp to compile on win32. However, those changes should 
> be relatively easy to merge with whatever will be needed for your 
> debugger.

That was quick. I'll have a look now, don't expect me to be as quick 
though, because there is some learning involved ...

> > See also:
> > - http://r-forge.r-project.org/projects/rinside/
> > - http://r-forge.r-project.org/projects/rcpp/
>
>
> I took a quick look, now, and it's nice stuff, but the value for 
> RKWard is limited:
> - We only need a rather "narrow" interface to R. Our goal isn't to do 
> speedy calculations in C++, but to add a GUI on to of R. Hence, for 
> most things that we want to put into R-space, we can use plain R syntax.
> - The other way around it's more interesting, and I'll take a closer 
> look at the date/time classes one day. However, the place we really 
> need this is the data editor, and we're generally well advised to try 
> to use QStrings/QLists and such, there (+ we need to implement all the 
> range checks and goodies from rcpp, in the data model, anyway). 
> (core/rkvariable.cpp, our main data-storage for the data editor will 
> need to be revised a lot, as well, but probably towards using Qt-classes).
>
>
> So it seems like a good place to look at, when in doubt about how to 
> deal with some R structures. But not something that is of much 
> *direct* use to RKWard, I think.
>
>
> Regards
> Thomas
-- 
Romain Francois
Independent R Consultant
+33(0) 6 28 91 30 30
http://romainfrancois.blog.free.fr