[R-SIG-Mac] R as OS X Framework

[Simon Urbanek]

On Nov 29, 2003, at 7:32 PM, Jan de Leeuw wrote:

> This is a major step in the right direction. I know that Duncan has at 
> least
> the beginning of an R bridge to the Cocoa toolkits, which would be the

Does he? I was toying with a Cocoa package for R (Cocoa windows/widgets 
in R; quite interesting how to tell R package installation system to 
handle Obj-C sources ...). I thought about using parts of Rserve to put 
together something like JRclient but for Obj-C, plus a class that would 
encapsulate R services like the parser and the evaluator. Duncan could 
probably use his OH stuff for that - it might work. I'd be interested 
to lear more about his progress.

> second major step. It will make it possible to write stand-alone R 
> based
> applications with a Cocoa GUI, of which R.app will just be a major 
> example,
> but of which Rglm.app or Rcluster.app could be other examples (with
> problem specific menus and graphics). And these could all be developed
> nicely by having a Xcode plugin for R.

Definitely - I was thinking along these lines - with proper Obj-C 
classes encapsulating R functionality now exposed by R.framework an 
application could be "written" (though it's not the proper word any 
more ;)) with the Cocoa controller layer just by using drag&drop. That 
would be awesome...

>  These steps emphasize the strengths
> of OS X (as the quartz device already does) and do not just give 
> another
> boring shell around the command line interpreter. Getting away from 
> Carbon
> in this way is important, and you don't have to learn Objective C to
> program in Cocoa.

I guess that Quartz in Carbon is ok for now, because it actually 
doesn't use almost anything from Carbon except for the window creation 
routines ;). Everything else (especially widgets & co) could be Cocoa.

Simon

---
Simon Urbanek
Department of computer oriented statistics and data analysis
Universitätsstr. 14
86135 Augsburg
Germany

Tel: +49-821-598-2236
Fax: +49-821-598-2280

Simon.Urbanek at Math.Uni-Augsburg.de
http://simon.urbanek.info