Re: emacs-27 60c84ad: ; * etc/TODO: Fix last change.

[Eli Zaretskii <eliz@gnu.org>]

> From: Robert Pluim <rpluim@gmail.com>
> Cc: monnier@iro.umontreal.ca,  emacs-devel@gnu.org
> Date: Tue, 03 Mar 2020 13:51:14 +0100
> 
> >>>>> On Mon, 02 Mar 2020 21:55:32 +0100, Robert Pluim <rpluim@gmail.com> said:
>     Robert> One thing we could do with emoji-data.txt is extract which codepoints
>     Robert> are emoji and use that to set up the fontsets appropriately.
> 
> That looks easy enough. Of course the fontsets end up not being used
> for certain codepoints because of use-default-font-for-symbols
> defaulting to t.
> 
> What should be done here? Create another char table analagous to
> char-script-table, add all the emojis from emoji-data.txt to it, and
> check whether the char is in that table in fontset.c:face_for_char?

I think it would be simpler and cleaner to define a new script,
'emoji', and update admin/unidata/blocks.awk to produce it for the
relevant blocks (including Variation Selectors, btw).  Then no change
is needed in face_for_char, since that is only in effect for the
'symbol' script.  As a nice bonus, this will also allow a much easier
customization of the fontsets, so that fonts for symbols can be
different from fonts for emoji.

There's an issue with the codepoints in the U+2XXX and U+3XXX blocks:
some of them can have Emoji glyph variations, but putting them into
the 'emoji' script is probably not a good idea, as long as Emacs
selects a font based on a single character.  I guess we will have to
leave them in 'symbol' for now.

The file admin/notes/unicode tells what has to be done when a new
script is added to Emacs.