bug#20140: 24.4; M17n shaper output rejected

[Eli Zaretskii <eliz@gnu.org>]

> Date: Tue, 15 Feb 2022 21:06:05 +0000
> From: Richard Wordingham <richard.wordingham@ntlworld.com>
> Cc: larsi@gnus.org, 20140@debbugs.gnu.org
> 
> > > Not off the top of my head, but compare لحج with the presentation
> > > form ‎ﳊ U+FCCA ARABIC LIGATURE LAM WITH HAH INITIAL FORM for the
> > > first two letters.  The lam part is a vertical line in the middle
> > > of the glyph; the 'hah' part forms the lower part of the glyph.  
> > 
> > They look identical here (using the default Courier New font).  With
> > what font did you think they will look wrong?
> 
> In the Courier New font in Windows 10 of 2017 (+ automatic updates),
> U+FCCA looks like the image in the Unicode code chart, and bears little
> resemblance to the righthand two thirds of <U+0644, U+062D, U+062C>.
> In keeping with its Latin part, the sequence of three characters looks
> as one would expect from a typewriter when one enters text letter by
> letter.

It sounds like Courier New in Windows 10 was "improved" by removing
the capability of ligating those 2 characters.  On Windows XP, their
standard Courier New shows the first 2 characters ligate into a single
glyph, which looks just like U+FCCA, but on Windows 10 they don't
ligate.  I don't know why is that; perhaps Arabic typesetting experts
decided these should not ligate?

> I must admit I'm having trouble laying my hand on a font which
> does these ligatures.

Try the Arabic Typesetting font, there I see on Windows 10 that the
first 2 characters look like U+FCCA.

IOW, this is a font issue, not an Emacs or HarfBuzz issue.