Re: XPath 2.0 filter enchancement

On Wednesday 07 August 2002 04:02 pm, Aleksey Sanin wrote:
> I think that we can
> simply extend the XPath 2.0
> filter spec to allow this. The only change we need to do is to allow one
> more
> child <dsig-xpath:XPointer> of the "dsig-xpath:XPathType" type in
> the <dsigTransform> element.

Hi Aleksy,

Independent of the technical merit of the proposal, the spec is already 
rather mature/baked. The Candidate Rec period formally ends today and I'm 
trying to schedule the Proposed Rec review -- though vacation schedules are 
tricky. So, for any new functionality, I'd want to hear *very* strong 
consensus that this was essential; if so, does the change necessitate 
recycling at Last Call, a new CR, a new namespace/identifier? And is the 
delay worth it? (We're in that stage where even reasonable and modest 
proposals have a very high bar to leap! <smile/>)

On the technical issue, in your example I'm not sure why:
  <dsig-xpath:XPointer 
    Filter="subtract"> xpointer(id('foo'))</dsig-xpath:XPath>
is all that superior to:
  <dsig-xpath:XPath 
    Filter="subtract">id('foo')</dsig-xpath:XPath>
There might be some more exotic expressions where it's a little easier, but 
even so, I'm comfortable with a REQUIRED reliance upon a normative 
reference to the XPath Recommendation; I'm less comfortable with a reliance 
on the old (abandoned) XPointer Candidate REC, or the new Working Drafts. 
(In fact, this is inappropriate going back to the maturity issue.)


-- 
* I will be on holiday the week of August 12th; I will try to return any 
calls or emails received the following week. 

Joseph Reagle Jr.                 http://www.w3.org/People/Reagle/
W3C Policy Analyst                mailto:reagle@w3.org
IETF/W3C XML-Signature Co-Chair   http://www.w3.org/Signature/
W3C XML Encryption Chair          http://www.w3.org/Encryption/2001/

Received on Thursday, 8 August 2002 12:13:34 UTC