Re: Eglot "inlay hints" landed

[Eli Zaretskii <eliz@gnu.org>]

> From: João Távora <joaotavora@gmail.com>
> Date: Thu, 23 Feb 2023 19:26:10 +0000
> Cc: Stefan Monnier <monnier@iro.umontreal.ca>, dalal.chinmay.0101@gmail.com, 
> 	emacs-devel@gnu.org, dimitri@belopopsky.com, luangruo@yahoo.com
> 
> > I actually gave you a recipe for demonstrating the problems I have in
> > mind: scroll the window under pixel-scroll-precision-mode.  AFAIK, we
> > don't call window-scroll-functions in that case.
> 
> I don't have a pixel-scroll-enabled Emacs to test.  Can you scroll
> large portions of the window like that and w-s-functions will never
> get called?  I'd say that's a bug we should fix.
> 
> > Another situation where we don't call window-scroll-functions is when
> > the user types into the buffer.
> 
> That is handled by Eglot's after-change-functions already so it
> isn't a problem.
> 
> > Yet another situation is when you type "C-x 1" to delete all the other
> > windows on the frame, leaving the current window that now shows more
> > stuff than before.
> 
> Reproduced.  Easy enough to fix with window-configuration-change-functions.
> I pushed a fix.

My point is that using jit-lock machinery, you will never miss an
update, because redisplay is very good at knowing when something needs
to be redrawn.  It has to.  By contrast, all those hooks are less
reliable, and also make Emacs sluggish, because the hooks are many
times triggered when there's nothing to do wrt display.