[ESS] How do I customise the frame split when I run a line of R from ESS?

Paul Johnson pauljohn32 at gmail.com
Mon Sep 3 23:29:21 CEST 2012

On Sat, Sep 1, 2012 at 3:13 AM, Chris Evans <chrishold at psyctc.org> wrote:
> I am running Emacs with ESS both installed from the Ubuntu 12.04 defaults.
> Help says Emacs is:
> "GNU Emacs 23.3.1 (i686-pc-linux-gnu, GTK+ Version 2.24.10)
>  of 2012-03-25 on roseapple, modified by Debian"
> ESS is ESS version 12.04-4
> The default behaviour when I run a line of code from the icon in ESS seems
> to be a vertical split.

I run Debian and I've never seen a vertical split, so there must be
something unusual in your start up files. Given the way laptop
displays are going (too wide, too short IMHO), I can see some benefit
in vertical splits. Maybe you should count your blessings.

For my part, I never want Emacs splitting windows or obscuring buffers
i currently have open, so I want Emacs to pop up a new frame for *R*.
Info on how I do that is in this intro to Emacs/Ess lecture I did last


I noticed that emacs 24.1 has changed terminology on split windows. At
least on MS, in stead of split windows, the File menu says "open
window below" or such. That was interesting.

If you start emacs, and then tell it to split horizontally (I mean,
"open window below"), then click in that new window, and then you hit
the blue R, doesn't *R* go into the split below?

I'm absolutely sure that, about 5 years ago, I had my .emacs set set
to do that automatically.   As I recall, the hard part was getting
Emacs (or ESS?) to do what I wanted, not what it wanted. My work habit
would be to open *.R, then hit the blue R button, hoping that

1) the window would automatically split, and
2) the R session would always open in THE OTHER window.

As others noted in this thread, ESS pursues the opposite approach, of
starting *R* in the current window, so that if you have an R file
open, *R* covers it over. I hate that completely. You have to go
through a dance of splitting, choosing the other one, and then
starting R.  But I bet if you try, you can script it  up.

Paul E. Johnson
Professor, Political Science    Assoc. Director
1541 Lilac Lane, Room 504     Center for Research Methods
University of Kansas               University of Kansas
http://pj.freefaculty.org            http://quant.ku.edu

More information about the ESS-help mailing list