Re: ISO vs Unicode UTF-8 (was RE: UTF-8 signature in web and email)

From: Michael \(michka\) Kaplan (michka@trigeminal.com)
Date: Wed May 30 2001 - 19:51:08 EDT


Simon,

Would you care to answer (officially) why exactly Oracle needs for anything
to be done here? Per the spec, it is not illegal for a process to interpret
5/6-byte supplementary characters; it is only illegal to emit them. It seems
that Oracle and everyone else is well covered with the existing standard and
spec:

1) Any time Oracle insists on emiting the incorrect form, a process can
accept it and read it. Thus everything will work.

2) Any time Oracle is sent legal and proper UTF-8, it will (per your words
below) read it correctly and once again things will work.

3) Any time someone is sent the bad UTF-8 then Oracle will of course be able
to read it properly.

4) Any time Oracle emits legal UTF-8, it will of course work anywhere else.

It seems that nothing needs to be done for things to work. The entire
"requirement" that the proposal talks about is simply asking (insisting?)
that the UTC bless the needs of a specific implementation that go way beyond
the scope of what Unicode of any of the transformation formats ever promised
to do.

MichKa

Michael Kaplan
Trigeminal Software, Inc.
http://www.trigeminal.com/



This archive was generated by hypermail 2.1.2 : Fri Jul 06 2001 - 00:18:17 EDT