[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: namespace handling models

• To: bryan <bry@i...>
• Subject: Re: namespace handling models
• From: Bob Foster <bob@o...>
• Date: Mon, 05 Jan 2004 07:14:35 -0600
• Cc: xml-dev@l...
• In-reply-to: <007501c3d380$22de8970$2001a8c0@bryans>
• References: <007501c3d380$22de8970$2001a8c0@bryans>
• User-agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.5) Gecko/20031007

bryan wrote:
>>transform: elements and children in the unknown namespace are 
>>transformed into known elements and text. Ex: Henry Thompson's normal
>>forms.
> Thanks for the mention of normal forms, hadn't seen it before, now I
> have some reading for next weekend. However looking at what I googled up
> 
> http://www.ltg.ed.ac.uk/~ht/normalForms.html
> 
>  I couldn't see right off how one manages transform of unknown namespace
> elements to known elements and text:

Here is an entirely trivial example:

<x:foo xmlns:x="unknown" a="1" x:a="2">
   <y:bar xmlns:y="other"/>
</x:foo>

Becomes:

<element name="foo" ns="unknown" xmlns="known">
   <attribute name="a" value="1"/>
   <attribute name="a" ns="unknown" value="2"/>
   <text>
   </text>
   <element name="bar" ns="other"/>
   <text>
</text>
</element>

(I think I got this from a post by Simon St. Laurent.) It's a kind of 
transparent tunneling.

> If I have <x:code xmlns:x="urn:whatever:namespace">90952</x:code>
> somewhere in the xml coming into my application, and I have absolutely
> no knowledge of the namespace and the element of local name code in it,
> how does normal Forms or any technology help me to securely transform
> that to <zip>90952</zip>. This is what I think you're saying it does,
> and this is what I find strange and mystifying that it could do,
> although I am willing to be astounded if it be so.

It's not about meaning, it's about markup. You don't have to lose 
information just because you don't understand it.

Bob

>>In the spirit of your list, this doesn't include strategies by which 
>>unknown namespaces become known, such as James Clark's NXML locate and 
>>David Tolpin's Regular Associations, recently mentioned on this list.
> 
> 
> Thanks for the info, still not sure if I understand the Regular
> Associations post yet. Yeah, how an unknown namespace becomes known
> isn't what I'm interested in right now. What I'm thinking about
> specifically is models for how unknown namespaces are treated that
> remain unknown all the way through an application, mainly because we
> have an application in which I suppose that is going to be a big
> problem. 

• Follow-Ups:
  • RE: namespace handling models
    • From: "bryan" <bry@i...>

• References:
  • RE: namespace handling models
    • From: "bryan" <bry@i...>

• Prev by Date: active-uri was RE: Globbing versus Regular Expressions (was: RegularAssociations)
• Next by Date: RE: namespace handling models
• Previous by thread: RE: namespace handling models
• Next by thread: RE: namespace handling models
• Index(es):
  • Date
  • Thread