bug#23871: 25.1.50; Undo unexpectedly leads to blank buffer

[Eli Zaretskii <eliz@gnu.org>]

> From: Markus Triska <triska@metalevel.at>
> Cc: 23871@debbugs.gnu.org
> Date: Thu, 30 Jun 2016 20:00:31 +0200
> 
> Eli Zaretskii <eliz@gnu.org> writes:
> 
> 
> > I suspect this is a redisplay problem, not an undo problem.  How do
> > you see that point is in the wrong position?  What does "C-x =" say
> > immediately after the undo?
> 
> I see from two facts that it is in the wrong position:
> 
> 1) I need to press C-p to go to the position *before* the undo. From
>    this I see that point after undo is at least one line off.
>    
> 2) C-x = immediately after undo confirms this, saying:
> 
>    point=323 of 322 (EOB) column=0
> 
>    While point before undo is point 281 of 322.

I'm not sure this is not the expected behavior.  This means the blank
buffer you see is just the last empty line with EOB.  The Emacs
display engine should not normally allow that, it should show the last
line before EOB.

> > Can you try coming up with a recipe that doesn't involve SWI-Prolog?
> 
> Please see #21722. It shows that point position after undo is different
> from what it was before, just as in this case.

When I repeat the recipe there, both with Emacs 25.0.95 and 25.1.50,
point is positioned at the end of the line, both after "C-M-x C-/" and
"C-h f C-g C-M-x C-/".  So I see consistent behavior.  Perhaps undo
gurus could comment whether this is also the correct behavior.

> The font and even window size definitely have an impact on this
> issue. Please try a different window size, and I hope you can then
> reproduce the issue, or at least #21722.

Unfortunately, I can't reproduce any of them.