Re: feature/icomplete-vertical

[Eli Zaretskii]

> Date: Sun, 20 Sep 2020 17:09:46 +0000
> From: Gregory Heytings <ghe@sdf.org>
> cc: spacibba@aol.com, emacs-devel@gnu.org, casouri@gmail.com,
>         joaotavora@gmail.com
> 
> >> trying to calculate the height of the completion candidate list, which 
> >> amount (as Stefan wrote) to redoing what redisplay does.
> >
> > I don't agree with Stefan, if this interpretation of what he said is 
> > correct.  We have window-text-pixel-size to measure the size of text on 
> > display without redoing what redisplay does.
> >
> 
> Do you mean that every function which wants to display something in the 
> minibuffer should use window-text-pixel-size to create a text that fits in 
> the minibuffer without hiding the prompt?

I didn't say that every function needs to do it, or anything to that
effect.  I was responding to the claim that measuring the size of text
on display needs to redo what redisplay does.

> And that this is better than only expecting such functions to approximate 
> the available space, and to rely on the fact that if the text is too large 
> the display engine will hide the end of the text instead of the prompt?

I didn't say that, either, not in such a general form, anyway.

I explained the state of affairs for the issue at hand in the bug
report; if something there is unclear, I will happily answer specific
questions regarding the issue we are facing.