RE: Updated c14n Spec

I need to read the attributes of a element anyway, and I also need to sort
them using the attribute name and the namespace uri as sort keys. This is
the minimum requirement in all cases.

What I would like to avoid is walking the parent elements of a node looking
for inherited namespace declarations. 

Petteri


> -----Original Message-----
> From: David Blondeau [mailto:blondeau@intalio.com]
> Sent: Monday, June 12, 2000 9:38 PM
> To: Petteri Stenius; 'John Boyer'; Joseph Reagle
> Cc: XML DSig
> Subject: Re: Updated c14n Spec
> 
> 
> Hi Petteri,
> 
> > The current XPath canonicalization is starting to look quite nice.
> I agree except the optional XPath expression that in my 
> opinion prevent
> non-XPath implementation.
> 
> > My main motivation is performance: the namespace axis 
> approach requires
> one
> > to walk the parent nodes of a element when canonicalizing 
> and keep track
> of
> > inherited namespaces etc., the old C14N spec is very simple 
> to implement
> in
> > one single pass.
> [move this part]
> > The idea is that an element always declares only those 
> namespaces that
> > appear on the element itself and on the element's attributes.
> [end moved part]
> If your main motivation is performance, I am not sure your 
> suggestion is
> good since you will need to parse all attribute values to know if a
> namespace prefix is used in them -> difficult and time consumig!
> The namespace axis is the best approach in my opinion.
> 
> >Other motivations are that the original C14N approach (even without
> >namespace rewriting) is simpler to implement and that many 
> XML processors
> do
> >not implement the namespace axis functionality.
> IMO, the namespace axis functionality is really simple to 
> implement using
> SAX, maybe more complex in DOM, but the attribute values 
> parsing is really
> not a good option.
> 
> David
>

Received on Wednesday, 14 June 2000 02:49:10 UTC