If Arabic non-punctuation characters are being displayed using their Unicode
code points, we’re doing it wrong. They need contextual shaping, not to
mention bidi. Pango seems capable of this, as does Win32’s Uniscribe and
Carbon on OS X, and I’ve been looking into what’s necessary for the work in
what dead time I have. Nothing concrete yet, though.
Basically, I don’t see any reason to keep that approach around for Arabic in
any form.
Ar an triú lá déag de mí Lúnasa, scríobh Stephen J. Turnbull:
QUERY
This really should be commented out, not simply deleted. We
eventually need to find a way to handle non-built-in charsets, and a
note here that this table doesn't work on non-built-in charsets
currently would be a useful reminder.
Aidan Kehoe writes:
> changeset: 4494:ea66743ff20155e3a4e479120d7bd68bfe12ac9f
> user: Aidan Kehoe <kehoea(a)parhasard.net>
> date: Sun Aug 10 12:03:10 2008 +0200
> files: src/ChangeLog src/objects-xlike-inc.c
> description:
> Fix the XFT build, thank you Robert Delius Royar.
>
> 2008-08-10 Aidan Kehoe <kehoea(a)parhasard.net>
>
> * objects-xlike-inc.c (charset_table): Remove the entry for
> Vcharset_arabic_iso8859_7, thank you Robert Delius Royar.
--
¿Dónde estará ahora mi sobrino Yoghurtu Nghé, que tuvo que huir
precipitadamente de la aldea por culpa de la escasez de rinocerontes?
_______________________________________________
XEmacs-Patches mailing list
XEmacs-Patches(a)xemacs.org
http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-patches