In article <50E0094A.1010500@gmx.at>, martin rudalics writes: > I obviously have no idea how this is supposed to work :-( > We could replace the condition > (when (and rmail-summary-buffer (buffer-name rmail-summary-buffer) > rmail-redisplay-summary) > in `rmail-maybe-display-summary' either by > (when (and rmail-summary-buffer (buffer-name rmail-summary-buffer) > (or rmail-redisplay-summary > (get-buffer-window rmail-summary-buffer 0))) > or by > (when (and rmail-summary-buffer (buffer-name rmail-summary-buffer) > (or rmail-redisplay-summary rmail-summary-window-size)) > Both should work if the summary window was specially created for the > summary. They won't work when the window got merely reused - but in > that case resizing the window doesn't strike me as a good idea in the > first place. Can you try them? I've just updated my working tree of Emacs from trunk and found that you have already committed a new patch. So, I tried it and confirmed that it worked well. Thank you!! > If this doesn't help, can you please tell me when and why you want to > resize the summary window and why you want to do that with > `rmail-redisplay-summary' set to nil? I can obviously restore the old > logic but would like to understand it first. My understanding of rmail-redisplay-summary is that when it is non-nil, rmail-summary-buffer is forced to be shown. For instance, when rmail-summary-buffer is buried and I change the current message in RMAIL buffer by typing p, rmail-summary-buffer pops up. As I don't want that behavior, I haven't customized rmail-redisplay-summary to t. What I want is that rmail-summary-buffer is shown in some window of the specified height when I type h in RMAIL buffer. --- Kenichi Handa handa@gnu.org