Question

I’m currently programming away in emacs. I have a function defined in my .emacs that saves all my work and executes the interpreter to run my work in the currently open shell buffer. Typically I’ll be editing in one or more frames and have the shell open in a separate frame. The problem I have is when I run my save and execute function, everything is saved but the shell buffer then gets displayed in the frame I’m currently editing. So I have two frames showing the shell buffer and I can’t see the source code I was just editing. Quite often when I’m programming, I’d immediately like to compare the output of the code with the code I’ve just written. It’s a bit annoying to have to switch back to my code buffer and then go to the end of the other shell buffer to look at the output while referencing the just written code.

(defun execute-script ()
  "Switch to shell buffer and re-execute the last command."
  (interactive)
  (save-some-buffers)
  (switch-to-buffer "*shell*")
  (end-of-buffer)
  (comint-previous-input 0)
  (comint-send-input))

As you can see my function is rather primitive at the moment, just re-executing the most recently used command in the shell.

I know that Emacs has the functionality to switch to a buffer in a different frame, as the ido buffer switcher code does this. Does anybody have any pointers as to what I need to replace my call to switch-to-buffer with to get the desired effect?

Regards Giles.

Answer 1

Exchange

(switch-to-buffer "*shell*")

for

(switch-to-buffer-other-window "*shell*")

This will enforce a two window layout with your previous buffer on the left side and the shell buffer on the right side.
Edit: If you’re using a multiple frame layout and not a multiple window layout you can use

(switch-to-buffer-other-frame "*shell*")