bug#39799: 28.0.50; Most emoji sequences don’t render correctly

[Eli Zaretskii]

> From: Robert Pluim <rpluim@gmail.com>
> Cc: rgm@gnu.org,  39799@debbugs.gnu.org,  mfabian@redhat.com
> Date: Wed, 22 Sep 2021 10:59:00 +0200
> 
>     Eli> AFAICT, the composition machinery doesn't allow to have a rule with
>     Eli> non-zero PREV-CHARS where PREV-CHARS are supposed to match characters
>     Eli> that have their own rules in composition-function-table.  Such rules
>     Eli> are never considered, because once the rules for a character are
>     Eli> processed, we never reconsider what to do with that character.  I 
> will
>     Eli> ask Kenichi Handa whether my conclusion is indeed correct, but for
>     Eli> now, I think we should take the fire escape you mentioned earlier:
> 
> OK. Those are surprising semantics. If it does turn out to be like
> that, we should document it.

We could probably lift this restriction if we change the logic in
composition_compute_stop_pos: it currently returns as soon as it finds
the first character with a valid entry in composition-function-table,
instead of keeping looking until it finds the one whose rule's PATTERN
starts at the smallest character position.  But even if it turns out
there are no problems with such a change, I wouldn't do it for Emacs 28.