> One of the more interesting things in the 3.2 revision is a
> few "mainly
> private use" characters set aside in the Arabic region. We
> can safely use
> these as initial/final/medial/isol forms of characters that
> don't have all
> their forms encoded as glyphs already.
Well, the code points you refer to are now (or soon, to
be precise) "not-a-character"s, and have the same status
as FFFE, FFFF, 1FFFE, etc. They are really inteded for
"process internal" things, which may vary from application
to application, but include such things as table cell
boundary, footnote anchor, etc. Just like FFFF in UTF-16
is suitable as EOF. They are not intended for Arabic
glyph indices...
http://www.unicode.org/charts/draftunicode32/U32-FB50.pdf.
Kind regards
/kent k
-
Linux-UTF8: i18n of Linux on all levels
Archive: http://mail.nl.linux.org/lists/
