Re: tramp (2.2.3-24.1); Using tramp to run ess-remote session. Submitting more than one line of code at a time, causes emacs to wait indefinitely.

[Michael Albinus]

Tobias Muhlhofer <address@hidden> writes:

Hi Tobias,

> I use tramp to run an ess-remote session. To do this, I use M-x shell,
> log into the server and start R. Then I do M-x ess-remote. When I try to
> submit a region containing more than one line of code to the process (by
> highlighting the region and going C-c M-r) the spinning "Emacs is
> waiting" cursor appears and will not go away. If I hit C-g, I see that
> only the first line of code has been submitted.

Well, I neither use ess nor R, so it is hard to reproduce for me. I've
tried it like this:

Emacs 24.3.50.1
Tramp 2.2.7-pre
ess 5.14 (the one bundled in Ubuntu 12.04)

R 2.15.1 (on the remote host, the one bundled in Ubuntu 12.10)

I have written a local test file 123.R, derived from an R tutorial I've
found somewhere:

--8<---------------cut here---------------start------------->8---
fruit <- c(5, 10, 1, 20)
names(fruit) <- c("orange", "banana", "apple", "peach")
lunch <- fruit[c("apple","orange")]
print (fruit)
print (lunch)
--8<---------------cut here---------------end--------------->8---

I have opened a remote shell via "M-x shell". Started "R". Configured
via "M-x ess-remote". If I mark the whole buffer of 123.R, and I apply
"C-c M-r", I get in the shell buffer

--8<---------------cut here---------------start------------->8---
> fruit <- c(5, 10, 1, 20)
> names(fruit) <- c("orange", "banana", "apple", "peach")
> lunch <- fruit[c("apple","orange")]
> print (fruit)
orange banana  apple  peach 
     5     10      1     20 
> print (lunch)
 apple orange 
     1      5 
> 
--8<---------------cut here---------------end--------------->8---

That looks pretty good to me.

I do not believe, that the R version matters. But it could be, that
there are changes in either Emacs, Tramp, or ess, which bring the
problem to you. Don't know. Maybe you play with a setup similar to the
one I have used?

Looking around, there was a similar problem reported at
<http://blog.nguyenvq.com/2010/07/11/using-r-ess-remote-with-screen-in-emacs/>:

  I would also like to note (for archival reasons) that ess-remote does
  not work with ansi-term and multi-term because of the
  inferior-ess-mode command, which stems from comint-mode and
  inferior-ess-send-input. If you remove this command in ess-remote, you
  don’t get an error but u can only send one line of code at a time from
  the R file.

This report was not related to Tramp.

Btw, I have tried the same experiment with eshell, and there I got an
error. This looks like a Tramp problem; will try to fix it.

Best regards, Michael.