From: Raymond Mercier (RaymondM@compuserve.com)
Date: Mon May 23 2005 - 02:37:03 CDT
Gregg Reynolds writes
> Ok; but that doesn't answer the question. Font encoding is not the same
> as "character" encoding, so I'm still not really sure what text you're
> looking at.
The ad hoc font here, as for most people, is created by pasting into the
cells with lower ACSII hex values the various Hebrew glyphs. So if you
select that font in Word you get Hebrew glyphs when you just type a, b, c,
... etc. I made the font using Fontographer, with ANSI character set and
'don't care' font family. That's what one did in those days. I have done
that for Greek and Arabic as well. The conversion of Greek to Unicode is
easy, but Arabic also raises the issue of ordering.
If the resulting document is saved as text (not unicode .txt) and opened in
Notepad you will get Hebrew if the font is changed to the one I am talking
about (which of course I named Hebrew.ttf). Since most of my docs are only
partly Hebrew, and the rest English, of course it looks all wrong in
Notepad, but in Word different fonts can be mixed. Also in Word the global
replacement macro can be written to select only the Hebrew font.
The real problem, as I see it, is to cancel the ordering somehow. Is the
bidi algorithm in Uniscribe ? I am not sure how it works.
I am hopeful about the SIL Converter.
Raymond
This archive was generated by hypermail 2.1.5 : Mon May 23 2005 - 02:37:49 CDT