RE: Enveloped signatures and XPath

Kent,

I agree with you that we should not require a serialize() function. 

I also suggest that we change the semantics of the XPath transform to "not",
so that the result of the XPath evaluation is those nodes that are pruned
from the document.

The current definition for serialize() makes the transform quite hard, as we
have to iterate through the result node-set, walk down each node recursively
and check for each node if it exists in the result node-set. Only if a node
does *not* exist in the result node-set is it removed.

The XSL WG has also suggested this change to "not" semantics 
(http://lists.w3.org/Archives/Public/w3c-ietf-xmldsig/2000JanMar/0190.html)


Petteri


> -----Original Message-----
> From: TAMURA Kent [mailto:kent@trl.ibm.co.jp]
> Sent: Tuesday, March 28, 2000 7:16 AM
> To: w3c-ietf-xmldsig@w3.org; Jonathan Marsh
> Subject: Re: Enveloped signatures and XPath
> 
> 
> 
> > This is true but doesn't address my point.  I think it's 
> great that a user
> > can return a nodeset and it'll get serialized 
> automatically.  In fact, it's
> > so great I can't see why anyone would ever need to bother 
> with calling
> > serialize() explicitly.
> 
> The same as Jonathan, I think the explicit serialize() function
> is not needed.  Of course, the automatical serialization for
> XPath result is needed.
> 
> -- 
> TAMURA Kent @ Tokyo Research Laboratory, IBM
>

Received on Tuesday, 28 March 2000 08:48:46 UTC