From: John Cowan (jcowan@reutershealth.com)
Date: Thu Jul 31 2003 - 16:56:13 EDT
Ted Hopp scripsit:
> So would this new right-holam character be a combining character?
Just so.
> If so, its
> use should be highly restricted, similar to what is done with shin dot and
> sin dot. Applying a right-holam character to anything other than a bare vav
> should be considered an error (no other combining marks should be allowed,
> including HEBREW ACCENT characters).
Unicode allows any combining character to be attached to any base character
whatsoever. However, putting a dagesh into a DEVANAGARI KA, or placing a
circumflex over an ARABIC MEEM, is pretty certain to cause bad rendering, and
may screw up other text processes such as syllabication.
> Would FB4B continue to decompose into 05D5 05B9?
Yes. Normalization stability requires it.
> It seems to me that either I'm misinterpreting things, or most people in
> this discussion would prefer a new combining character to a new base
> character. If this is so, I'd appreciate an explanation of why, because I
> don't understand it.
Assertions of the form "Mark X is only used with base form Y" have proven to
be false too often in the past.
-- John Cowan <jcowan@reutershealth.com> http://www.ccil.org/~cowan http://www.reutershealth.com Charles li reis, nostre emperesdre magnes, Set anz totz pleinz ad ested in Espagnes.
This archive was generated by hypermail 2.1.5 : Thu Jul 31 2003 - 17:51:16 EDT