- From: Paul Nelson <paulnel@microsoft.com>
- Date: Wed, 10 Oct 2001 15:35:31 -0700
- To: "John Hudson" <tiro@tiro.com>, "Unicode List" <unicode@unicode.org>
- Cc: <www-international@w3.org>
Why would you do something like this? These are control characters and
should not be played with. The shaping engine should set the shapes of
the charaters on either side correctly. Since glyphs for control
characters are not normally displayed there is no reason for this
lookup. Actually, your lookup stuff will not effect the text at all
because the shaping is determined before you ever start glyphing.
Unicode operations with characters have priority over glyphing.
Paul
-----Original Message-----
From: John Hudson [mailto:tiro@tiro.com]
Sent: Wednesday, October 10, 2001 3:07 PM
To: Unicode List
Cc: www-international@w3.org; Paul Nelson
Subject: ZWJ+ZWNJ+ZWJ (Was: ZWJ and Turkish)
At 14:23 10/10/2001, Roozbeh Pournader wrote on the Unicode list:
>Right, but just make sure that ZWJ+ZWNJ+ZWJ (the worst thing in
>Unicode, as I call it) also works on your Arabic fonts...
Presuming that the result here is that the final ZWJ should override the
ZWNJ, the easiest way to handle this would be to put a pre-emptive
lookup
in the OT Character Composition/Decomposition <ccmp> feature:
ZWJ + ZWNJ -> ZWNJ
ZWNJ + ZWJ -> ZWJ
I'm not sure if the current version of Uniscribe would make use of this
lookup. I've cc'd this message to Paul Nelson in the MS Typography
group,
who should be able to comment on this.
John Hudson
Tiro Typeworks www.tiro.com
Vancouver, BC tiro@tiro.com
Type is something that you can pick up and hold in your hand.
- Harry Carter
Received on Wednesday, 10 October 2001 18:46:36 UTC