Re: Potential bug in the logic of rmail-select-summary

["Göktuğ Kayaalp" <self@gkayaalp.com>]

On 2021-01-19 16:50 +02, Eli Zaretskii <eliz@gnu.org> wrote:
> In what version of Emacs do you see that?

Build of 36d33776c21b3765b8a611f09ae7d86417abee8a, within last few days.

> FWIW, I cannot reproduce this, neither in Emacs 27 nor in Emacs 28.
> Does this happen in "emacs -Q"?  (I use Rmail all the time, and I
> always have both the mbox file and its summary displayed, so if this
> would be happening to me, I'd sure have noticed.)

Indeed, I should’ve known better and tested with -Q.  I’m sorry about
that.

So apparently the problem was caused by some interaction with the
following bit of config:

 display-buffer-alist
 '(("\\*Quail Completions" . (display-buffer-in-side-window))
   ("\\*.*Completions\\*" . (display-buffer-in-side-window . ((side . bottom))))
   ("\\*Help\\*" . (display-buffer-reuse-window))
   ("Checkdoc" . (display-buffer-pop-up-window))
   ("Calendar" . (display-buffer-in-side-window . ((side . bottom))))
   ("help\\[R\\]" . (display-buffer-pop-up-window))
   (".*" . (display-buffer-same-window)))

When I commented out that last line it worked.  FWIW, I added the
following advice which seems to help:

(define-advice rmail-show-message
    (:around (fn &rest args) inhibit-display-buffer-alist)
  (let ((display-buffer-alist nil))
    (apply fn args)))

Is this an expected interaction or a bug?  AFAIU rmail-pop-to-buffer
could use some form of display-buffer-* to avoid going through
display-buffer-alist.

-- 
İ. Göktuğ Kayaalp / @cadadr / <https://www.gkayaalp.com/>
pgp:   024C 30DD 597D 142B 49AC 40EB 465C D949 B101 2427