From: Philippe Verdy (verdy_p@wanadoo.fr)
Date: Sat Aug 14 2004 - 02:44:03 CDT
From: "Michael Everson" <everson@evertype.com>
> No. There may have been an ambiguity there which you grasped at, but
> its roots were not deep, and the text will be altered. ZWNJ and ZWJ
> are intended for cursive scripts like Arabic and Brahmic (...)
Hic! Wasn't ZWNJ also defined to control ligatures in non-cursive scripts
like <f,ZWNJ,i> to instruct the renderer to NOT make a ligature between f
and i?
Are you saying now that sequences like <f,ZWNJ,i> and <f,ZWNJ,f> are non
standard, without meaning, because it breaks *now* with the new change in
Unicode rule D17 that changed the Unicode character model ?
If so, what is the proposed ignorable control to specify to a renderer that
it should NOT join <f,i> or <f,f> with a ligature?
Same thing for <f,ZWJ,i> to instruct the renderer that it SHOULD better join
the two grapheme clusters into a ligature...
What did I miss there?
This archive was generated by hypermail 2.1.5 : Sat Aug 14 2004 - 02:45:59 CDT