[ESS] on "Check inferior-ess-prompt or ess-loop-timeout"

8rino-Luca Pantani ottorino-luca.pantani at unifi.it
Fri Oct 29 19:48:22 CEST 2004

Hi all,
I'm still fighthing with Xemacs on my laptop.

The conditions were as follows:
OS       win98
computer IBM Thinkpad 1161
ess      5.2.1
R        1.7.1
Xemacs   21.4.13
init.el from Dr. Fox (28 Dec 2003, Ver. 0.5.4, I will send it if necessary)

and I always get the error message
<Timeout waiting for prompt. Check inferior-ess-prompt or ess-loop-timeout.>

I've read some old posts on the matter and
Rich Heiberger underlined that:
>The most likely reason for running into an ess-loop-timeout problem is
>that the cursor is in the wrong place.  The fix is to put it in the right
>place.  The workaround is often:
>1. enter C-g to get control back.
>2. move the cursor to the end of the buffer and enter <Enter>

>The long term fix is to figure out why there was a cursor placement
>For that we need more detail.  Detail consists of two parts.
>a. run M-x ess-submit-bug report
>b. write an excessively detailed description of what you were doing when
>the window froze.

So I started playing with the section

>(setq win1 (selected-window))  ; win1 is on top
>(setq win2 (split-window)) ; win2 is on the bottom
>(select-window win2)
>(R) ; start R in win2 -- assumes that the R bin directory is on the path
>(if dedicated-R-window (set-window-dedicated-p win2 t)) ; window win2
dedicated to R process
>(select-window win1) ; back to win1
>(R-mode) ; put *scratch* buffer in R mode
>(balance-windows) ; equalize size of windows

of the init.el file, using the -q option, C-x C-e and M-x eval.region
trying to get the cursor in the right place, but without success.
Sometimes I was able to get things going, but I was not able to reproduce
the conditions.

Eventually I was frustrated a bit and decide to change ess and R
ess 5.2.1 	to      5.2.2
R   1.7.1 	to      1.8.1

and adding the line
(load "C:/Programmi/XEmacs/ess-5.2.2/lisp/ess-site")
init .el file
changing the path to R and modifying the path in the autoexec.

Using the -q option, M-x eval.region in the init.el up to line
<;;  and the R process at the bottom>
and submitting each line after,
I noticed that R starts in the upper window (win1) instead that in the lower
one (win2)

Playing with cursor position and line submission
I finally came to some changes in init.el,
eliminating the two lines as follows

;(select-window win1) ; back to win1
;(R-mode) ; put *scratch* buffer in R mode

Everything seems to go well, I can get the Dr Fox's icons and all,
except that I get only a R window. To correct this I therefore move to
scratch buffer,
send ls() and then C-x-o (or 0 do not have the manual handy)
and start my R session with the scratch buffer above and R buffer under.

Unfortunately I do not have the two items under the pull down menu
"just quit" and "quit saving R workspace".
To quit neatly I have to delete all buffers and finally exit Xemacs.

As you imagine is not practical to work like that so
what should I change in the init.el,
in order to have things going straightier?

Ottorino-Luca Pantani, Università di Firenze

More information about the ESS-help mailing list