bug#57669: 29.0.50; C-n, C-p off under long lines

[Eli Zaretskii <eliz@gnu.org>]

> From: dick <dick.r.chiang@gmail.com>
> Cc: 57669@debbugs.gnu.org
> Date: Sat, 10 Sep 2022 08:07:24 -0400
> 
> EZ> So I think this part of the code is equivalent to what we do, in
> EZ> back_to_previous_visible_line_start
> 
> I'm glad you noticed!  It means you're actually reading the code!

You cannot say anything of an essence without ad-hominem, can you?

> One difference is I gate line scanning (as opposed to character
> scanning) with behaved_p, whereas GNU crosses its fingers and hopes its
> POS_LIMIT estimate is far enough.  So oddly enough, for all my cavailier
> assumptions, at least this part of the code [xdisp.c 9537:9577] is
> *more* conservative than GNU.

But since behaved_p is actually inaccurate at best, that doesn't
really matter, does it?  It just trades one class of cases where the
shortcut doesn't work for another.  You simply haven't yet met the
situations where those problems happen, most probably because you
tested only simple cases, like all-ASCII files which use a single
fixed-pitch font.  The code on master went through much more testing,
by more users, during several years.