From: James Kass (thunder-bird@earthlink.net)
Date: Tue Feb 12 2008 - 05:08:39 CST
I wrote,
> 2. If the CGJ function is only to prevent canonical re-ordering, then there
> would never be a need for a rendering engine to use it as part of a font
> look-up.
> 3. T.U.S. 5.0 pages 540-541 does say that CGJ only blocks re-ordering.
Oh, but wait...
Quoting from T.U.S. 5.0 page 541,
"The CGJ has no visible glyph and no other format effect on neighboring characters
but simply blocks reodering of combining marks."
Quoting from T.U.S. 5.0 page 542,
"However, some older implementations may treat a sequence of grapheme clusters
linked by combining grapheme joiners as a single unit for the application of
combining enclosing marks."
So, the statement on page 541 isn't true.
And, if CGJs are used to actually join graphemes, as the character name implies
and as it says on page 542, the rendering engines would be submitting CGJs
as part of look-up strings and fonts would be including CGJ in GSUB look-ups.
Best regards,
James Kass
This archive was generated by hypermail 2.1.5 : Tue Feb 12 2008 - 05:14:46 CST