[ESS-bugs] select-window (C-c M-j) doesn't give input focus to target (*R*) [Carbon GNU Emacs]

[Stephen Eglen]

David,
sorry for the delay in looking at this.

 > I understand that, in ESS-mode, C-c M-j (and friends) are supposed to  
 > select the *R* window and the buffer shown therein, so the input  
 > focus is in that window in the end.
 > ESS selects the appropriate frame (`select-frame') and window  
 > (`select-window') to do so in the function `ess-show-buffer'.
 > 
 > In current CVS versions of GNU Emacs (head, Carbon port) and ESS  
 > 5.2.8, this doesn't work. The *R* window is selected by `ess-show- 
 > buffer' (as I have verified, both visually and by evaluating  
 > (selected-window)), but the input focus (cursor is solid block, not  
 > frame) remains on the original window. As soon as keyboard input is  
 > made, the source window gets selected again (visibly).

 > To reproduce:
 > Arrange two frames: one showing just a buffer in R-mode, the other  
 > one showing the *R* process buffer. Enter some R code into the first  
 > buffer. Try to evaluate a line from within the first buffer by typing  
 > C-c M-j. This reliably produces the bug.

I cannot reproduce this on linux, using latest ESS sources and CVS
Emacs compiled from today's sources.


 > There has been a change to the behavior of display-buffer (doesn't  
 > select target frame any more) a few weeks ago, which should have  
 > brought the display-buffer in line with its documentation.

Can you tell me which revision you are referring to here?  A quick
browse of src/window.c change log showed this to be the newest change
to display_buffer:

revision 1.525
date: 2005-11-03 17:11:50 +0000;  author: monnier;  state: Exp;  lines: +2 -0
(Fdisplay_buffer): Fix last change to not use a dedicated window.

Is that what you are referring to?

 > I'm not sure if this is a problem with ESS or with (Carbon) Emacs -
 > please excuse the cross-posting.

Could you try again with ESS 5.2.12 and a recent CVS Emacs?


Thanks, Stephen