On Fri, 20 Jul 2007 03:43:50 +0200, Jonas Sicking <[EMAIL PROTECTED]> wrote:
In the current draft, the algorithm for matching access items says that
both the origin label and item label should be converted using the
ToASCII algorithm. However the pattern for the label part of items does
not allow any other characters than alphanumeric and '-'.
Should we use other patterns for the subdomain part to allow
internationalized characters? Or should we change the algorithm to not
apply ToASCII to the item label?
I agree that STD13 (RFC1034) is a bit vague but subdomain (and label) do
allow other characters. They are both defined as a serie of octets. The
more restricted BNF at the end of the specification seems to be
non-normative.
--
Anne van Kesteren
<http://annevankesteren.nl/>
<http://www.opera.com/>
