[R-SIG-Mac] Re: aqua/x11/tcl/tk

[stefano iacus]

Well this is not exact.
My binary built of R has an Aqua GUI but uses X11/TclTk. The libs are 
distributed in the R.framework.

This means that, after starting the X window server, you have access to 
X11, for example the X11 graphic device.

you can also load the tcltk package in R (with the AQUA GUI or from the 
command line) and have also the Tk widgets. Of course you don't have  
Aqua Tk widgets but you can use everything smoothly.

If you build R against Aqua TclTk libs, then Jan's trick for using Aqua 
Tcl/Tk is working but, if I remember well, it only works correctly 
running R from the shell, i.e. you cannot run Aqua TclTk from R GUI 
without problems with the event loop (unless you start the tkStartGUI, 
which replaces the R GUI with the tk widgets, but I found this way of 
using the GUI a bit more than strange).

stefano

On Apr 27, 2004, at 5:10 PM, James Wettenhall wrote:

> Hi Jean,
>
> I'm not a Mac person, just someone with an interest in Tcl/Tk,
> so you'll probably get some more informed answers soon, but...
>
> My understanding is that Aqua Tcl/Tk and X11 Tcl/Tk are not
> compatible and if you combine them, you can get strange errors.
> It looks like you have built R with an Aqua version
> of Tcl/Tk (in /Library/Frameworks/... or wherever), so you may
> run into trouble if you run R from X11 or if use addTclPath()
> with Tcl/Tk directories built with X11/Fink.  You can have a
> look at your Tcl Path from within R, by typing:
> library(tcltk)
> tclvalue("auto_path")
>
> As mentioned, the /Library/Frameworks/... Tcl/Tk directories
> will be Aqua, and from memory, Fink puts things in /sw/ by
> default...
>
> OK, now I'll let someone more knowledgeable answer your question
> and I'll try to convince my I.T. guys to let me play around with
> installing R and Tcl/Tk on a Mac OSX machine some time to see
> what I can learn... ;-)
>
> Regards,
> James
>
> _______________________________________________
> R-SIG-Mac mailing list
> R-SIG-Mac at stat.math.ethz.ch
> https://www.stat.math.ethz.ch/mailman/listinfo/r-sig-mac
>