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

[Robert Pluim <rpluim@gmail.com>]

>>>>> On Wed, 04 Mar 2020 18:04:40 +0200, Eli Zaretskii <eliz@gnu.org> said:

    >> From: Robert Pluim <rpluim@gmail.com>
    >> Cc: monnier@iro.umontreal.ca,  emacs-devel@gnu.org
    >> Date: Wed, 04 Mar 2020 10:25:45 +0100
    >> 
    >> One issue this has thrown up is that the hypothetical 'emoji' script
    >> completely subsumes U+1F000..U+1F02F (mahjong-tile) and
    >> U+1F030..U+1F09F (domino-tile), so those scripts would no longer
    >> exist.

    Eli> That shouldn't happen.  The 'emoji' script should include the
    Eli> following blocks:

    Eli>    1F300 .. 1F5FF (Miscellaneous Symbols and Pictographs)
    Eli>    1F600 .. 1F64F (Emoticons)
    Eli>    1F680 .. 1F6FF (Transport and Map Symbols)
    Eli>    1F900 .. 1F9FF (Supplemental Symbols and Pictographs)

    Eli> I understand that there are scattered Emoji elsewhere, but even if we
    Eli> make them individually belong to Emoji, we don't need to do that for
    Eli> mahjong-tile and domino-tile blocks.

My mistake. Those two ranges have the 'Extended_Pictographic' property
in emoji-data.txt, not 'Emoji', so I should just ignore those (and for
consistency not set the individual codepoints within mahjong-tile and
domino-tile as emoji). My nice clean awk script is growing warts :-)

    >> Your concern here is what Emacs should do when one of the characters in this
    >> range is followed by U+FE0E or U+FE0F?

    Eli> The latter.  The former doesn't affect the visual appearance of those
    Eli> base characters, so it's basically a no-op.

If we put the specific codepoints in the 2XXX and 3XXX blocks that
have 'Emoji' property in the emoji script, they'll display by default
with a nice font, and will be composed if they're followed by
FE0F. Would that be such a bad thing?

Robert