Re: [RFC] Add :invisible face attribute

[Eli Zaretskii]

> From: Michal Nazarewicz <mina86@mina86.com>
> Date: Wed, 18 Dec 2024 17:08:12 +0100
> 
> Introduce :invisible face attribute which makes foreground to be the
> same as background rendering the text invisible; or when :invert-video
> is also in effect, background is the same as foreground.
> 
> Use it in Org mode for org-hide face eliminting the need for
> org-find-invisible-foreground function.  This also simplifies
> auto-dim-other-buffers NonGNU ELPA package removing the need to
> configure a separate hide face for org-hide remap.
> 
> To observe the atribute in action, set ‘org-hide-leading-stars’ option,
> open NEWS file and enable org-mode.  The initial stars in section
> headings are rendered using the new attribute rather than Org mode
> needing to explicitly match foregroun to background.

Thanks.

A new face attribute adds quite a bit of complexity, so we should
really believe this is a good idea.  Given that this can already be
achieved without a new attribute, I'm not sure.  Is this really a
frequent need/situation?

But let's see what others think about this.

> @@ -2911,6 +2925,13 @@ merge_face_ref (struct window *w,
>                 else
>                   err = true;
>               }
> +           else if (EQ (keyword, QCinvisible))
> +             {
> +               if (EQ (value, Qt) || NILP (value))
> +                 to[LFACE_INVISIBLE_INDEX] = value;
> +               else
> +                 err = true;
> +             }

Does this mean that the result of merging two faces with different
values for :invisible will depend on the order of the merge?  That is,
does the nil value override the t value?  Or am I missing something?

Btw, I think "invisible" is not the best name for this, because that's
not really what will happen on display.  It's more like "illegible" or
something to that effect.