• From: Sean McGrath <sean.mcgrath@p...>
• To: Robert Koberg <rob@k...>
• Date: Sat, 11 Jul 2009 10:46:01 -0500

Robert Koberg wrote:
> ...
> Maybe I am missing the utility of this because of too simple an 
> example, but why would you do this?
Robert,

I suspect my example was too simple. I'm just using an element from the 
html namespace as
an example to illustrate the masking. I start most of my processing 
pipelines by masking all
the namespace decls and masking all prefixes.

I do it so that all the weirdness about namespace inheritance, bleed, 
race conditions etc. all gets
neutralized. I can then proceed to re-arrange/search over the instance 
without worrying about
what namespaces are in scope and what ones are inherited and what ones 
have nested declarations
and what prefixes have multiple different bindings and what prefixes 
have multiple identical bindings
and and and.....

I don't want parsers and other XML tools "helping me" by doing a bunch 
of namespace processing
before I get to work with the document.

I do a lot of XML processing work where I make changes in-situ : i.e. 
author/edit-style operations.

I also do a lot of XML processing where I need tight control over the 
input and output syntaxes
so that I can detect exactly what has changed.

Both of these are greatly simplified by neutralizing the namespaces in 
my opinion.

Sean

• References:
  • Re: RE: Namespace use cases
    • From: Sean McGrath <sean.mcgrath@p...>
  • Re: RE: Namespace use cases
    • From: Robert Koberg <rob@k...>