Hi Vinayak,
I completely agree with Neil. The way I think about this is that it
is namespace bindings are a parse time thing. After that the ns attrs have
no effect at all. The only reason we are adding them in is so the document
can be output and then read back in and be logically equivalent.
Gareth
On Tue, 11 Mar 2003 [EMAIL PROTECTED] wrote:
> Hi Vinayak,
>
> I think you might be attaching over-much importance to the serialized form
> of the document.
>
> Nowadays--where everything XML'ish tends to get cast in terms of the XML
> Information Set--I think the DOM spec writers don't think of the DOM as
> conveying information about particular serialized forms, but rather they
> think in terms of modeling an abstract collection of information--an XML
> serialization of which is another alternative. Looked at in this way, the
> original child2 element doesn't inherit anything from its parent: it has
> its own namespace and namespace prefix property which the DOM models.
> When you rename the child1 node in this way, abstractly what you're doing
> it changing its namespace infoset property.
>
> The DOM3 serialization API's are meant to provide as lossless a bridge as
> possible between the DOM rendering of the infoset and its serialized form.
> So it makes all the sense in the world not to change anything about the
> child2 node, while simultaneously avoiding declaring redundant namespace
> prefixes everywhere.
>
> Hope that makes some sense. :)
>
> Cheers,
> Neil
> Neil Graham
> XML Parser Development
> IBM Toronto Lab
> Phone: 905-413-3519, T/L 969-3519
> E-mail: [EMAIL PROTECTED]
>
--
Gareth Reakes, Head of Product Development +44-1865-203192
DecisionSoft Limited http://www.decisionsoft.com
XML Development and Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
