Re: Entering emojis

[Eli Zaretskii]

> Date: Thu, 28 Oct 2021 11:20:37 +0000
> From: Gregory Heytings <gregory@heytings.org>
> cc: mattiase@acm.org, raman@google.com, schwab@linux-m68k.org, 
>     stefankangas@gmail.com, emacs-devel@gnu.org
> 
> > There are examples of hieroglyph formatting in section 11.4 of the 
> > Unicode Standard: they show the sequence of codepoints and the expected 
> > display.  Do these work correctly with that font?  I mean figures 11-2, 
> > 11-3, and 11-4, as well as tables 11-2 and 11-3.  Here, we are free from 
> > the need to be aegyptologists, since Someone™ did the work for us.
> 
> This is such a specialized area that I trust the work of an aegyptologist 
> way more than that of the Unicode consortium.  The Aegyptus font contains 
> ~10000 glyphs and ligatures, more precisely, ~7500 glyphs and ~2500 
> ligatures.

That font was released more than a year ago, and the Unicode
formatting characters for the hieroglyphs are relatively new, as you
see from the standard's description.  So it could be the fonts didn't
yet catch up.

In any case, I'd be very surprised to learn that the Unicode Standard
is so wrong.  They have specialists aboard as well.

> The method chosen in that font is to use ligatures instead of explicit 
> formatting characters, with an "escape" (zero-width non-joiner U+200C) to 
> display adjacent characters that would have been subject to a ligature 
> separately (one after the other).  Both LibreOffice and Harfbuzz 
> understand this, and render Egyptian texts correctly.

Like I said: the new way of formatting this script is not yet
supported widely enough.  There was a discussion on the HarfBuzz forum
about a year ago, from which I understood that these controls are not
yet supported by fonts.

> I just looked at section 11.4 of the Unicode Standard.  The first 
> character sequence in figure 11-2 is not a valid quadrat.

Why not?  And it isn't supposed to be a quadrat, AFAIU, anyway.

> The second one are two characters that would normally be placed one
> above each other, so to obtain what is displayed in the Unicode
> Standard it's necessary to separate them with a zero-width
> non-joiner.

AFAIU, U+13431 is the joiner to be used in that case.

And you didn't answer my question: does LibreOffice with the Aegyptus
font display those sequences correctly?