On Tue, 21 Nov 2006 09:55:01 +0100, Alexey Proskuryakov
<[EMAIL PROTECTED]> wrote:
I'm fine with always serializing to UTF-8 if that's what browsers do.
One less character encoding issue for authors to deal with. :-)
There's one difference between WebKit and Firefox in this regard,
however. WebKit serializes the document without any XML declaration,
so encoding as
UTF-8 is only natural. However, Firefox includes a declaration, with an
encoding that doesn't match the actual one - which looks like a bug to
me.
Yeah, sounds like it. My suggestion would be to require that the XML
Declaration be dropped. Or, if present, is set to UTF-8 (which is the
encoding anyway).
Perhaps, the requrement could be that the document should be encoded to
be valid - i.e. if an XML declaration is present, it should match
the actual encoding?
That's what the specification currently says. :-) What do you want?
--
Anne van Kesteren
<http://annevankesteren.nl/>
<http://www.opera.com/>
