- From: Daniel Veillard <veillard@redhat.com>
- Date: Mon, 7 Feb 2005 06:29:57 -0500
- To: Anne van Kesteren <fora@annevankesteren.nl>
- Cc: public-xml-id@w3.org
On Mon, Feb 07, 2005 at 12:03:25PM +0100, Anne van Kesteren wrote:
>
> Although I am too late for last call comments I was wondering if the WG
> could clarify which ID attribute should get presedence when multiple are
> applied to a single element.
Could you explain what you mean by "presedence" there ?
> That is invalid per the XML specification but it does not make the
> document ill-formed so UAs which do not use a validating parser need a
> way to handle this.
"handle this" in what sense ? Given a node what is its ID (assuming
it has one)? In general the problem is given a string what is the element
carrying an ID attribute with that value. That is what is needed for
pointing, and is IMHO orthogonal to "presedence" or "preference".
> The multiple ID question has been brought up before on this mailing
> list[1][2].
yes but not for "presedence". That seems unrelated to me.
> [1] seems to be ignored. At least, I could not find any response to the
> e-mail in question.
>
> Also, as mentioned before, this causes problems for UAs[3].
Not for pointing, as #foo has the semantic #xpointer(id('foo'))
which is inherited from XPath and is the first element in document order
containg an ID attribute of value 'foo'
I don't see where in the discussion of that bug a "presedence" or a
"preference" is needed when multiple ID are defined on a node, it is a validity
error.
> [1]<http://lists.w3.org/Archives/Public/public-xml-id/2004Apr/0012.html>
> [2]<http://lists.w3.org/Archives/Public/public-xml-id/2005Jan/0066.html>
> [3]<https://bugzilla.mozilla.org/show_bug.cgi?id=258238>
Daniel
--
Daniel Veillard | Red Hat Desktop team http://redhat.com/
veillard@redhat.com | libxml GNOME XML XSLT toolkit http://xmlsoft.org/
http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/
Received on Monday, 7 February 2005 11:30:00 UTC