RE: [OpenType] Further thoughts on ZWJ ligation

From: John Hudson (
Date: Fri Jul 19 2002 - 15:35:41 EDT

At 12:01 PM 19-07-02, Paul Nelson wrote:

>You have some incorrect assumptions about implementation. Not to go into
>details, but your statement that "ligature substituions will not happen
>when the control characters are not displayed" is not true.
>There are a number of ways to "hide" the control character glyph. The
>magic is WHEN to do that. If you do that after the ligature is formed
>you don't have a control character glyph to handle any longer.

Fair enough. This seems to be similar to what I understand to be Apple's

What is the current implementation? If, for example, I perform a test in
which I hijack a couple of Arabic codepoints (so I know that the <rlig>
feature will be applied during processing), to use in sequence with the ZWJ
character, will the ligature form correctly in an application like Wordpad?
Or are your comments in reference to what could be done by changing the
existing implementation?

John Hudson

Tiro Typeworks
Vancouver, BC

Language must belong to the Other -- to my linguistic community
as a whole -- before it can belong to me, so that the self comes to its
unique articulation in a medium which is always at some level
indifferent to it. - Terry Eagleton

This archive was generated by hypermail 2.1.2 : Fri Jul 19 2002 - 13:50:05 EDT