> > According to the SAX documentation namespace attributes should have no
> > namespace URI, and no local name:
> >     ...
> > This information is based on Namespaces in XML REC.
> > 
> > However, the XML Information Set W3C REC specifies that namespace
> > attributes should be bound to "http://www.w3.org/2000/xmlns/"; :
> >     ...
> > In my opinion, SAX definition for namespace attributes is incorrect and
> > needs a fix.
> > Any comments?

When you believe that sort of thing, you should file a bug or RFE against
SAX using the bug database accessible through the URL below.

> My guess is that SAX2 was defined before InfoSet came out.  I hope you
> don't mind if I cc [EMAIL PROTECTED] (http://www.saxproject.org).

The SAX2 spec does indeed conform to the Namespace spec, rather than
to the (new) Infoset spec or (less new) DOM L2 spec.  The namespace folk
were pretty adamant that "xmlns" isn't a namespace at all.  I went around
with them on that issue more than once.  Though I can't say I'm glad to see
the W3C change its mind on that issue well _after_ the Namespace REC
was finalized!  (*)

Re what to do ... changing the default behavior of a SAX2 parser isn't
something that could be done while retaining backwards compatibility.

One could define a feature flag (another "SAX2 Extensions 1.1 flag)
to kick in this behavior as a non-default mode, but I'm not sure that'd
be very beneficial -- apps couldn't rely on it, they'd have to code the
workaround in any case, and given that, what's the benefit of adding
another confusion of codepaths that needs testing/updating?

- Dave

(*) The first of the three DOM L2 Candidate Recommendations,
    dated 10-Dec-1999, conformed to the 14-Jan-1999 Namespace
    REC ... but the second of three (!) DOM L2 CRs, 7-Mar-2000,
    changed this.  SAX2 finalized two months after that, DOM L2
    in November 2000, and Infoset just a few months ago.  Just to
    track the publicly visible milestones of this change.



---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to