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

Re: Attributes and namespaces

• From: David Megginson <david@m...>
• To: xml-dev@i...
• Date: 27 Sep 1999 16:18:51 -0400

"Shawn Silverman" <shawn@a...> writes:

> I'm confused about how namespaces should be applied to attributes.  Here
> are two questions:

> 1)
> 
> <?xml version="1.0"?>
>   <!-- all elements here are explicitly in the HTML namespace -->
>   <html:html xmlns:html='http://www.w3.org/TR/REC-html40'>
>     <html:head><html:title>Frobnostication</html:title></html:head>
>     <html:body><html:p>Moved to 
>       <html:a href='http://frob.com'>here.</html:a></html:p></html:body>
>   </html:html>
> 
> To which namespace does the "href" attribute in the "html:a" tag
> belong?  To the same one as its parent?  "html:"?

In the end, the Namespaces spec ended up making

  <html:a html:href="foo.xml">..</html:a>

and

  <html:a href="foo.xml">..</html:a>

distinguishable.  That doesn't mean that applications are *forced* to
distinguish the two href's, but that they can do so if they want (they
can also treat the two as equivalent if they wish).  There's an
unfortunately convoluted appendix in the Namespaces spec on this
topic, but it may not be helpful.

Personally, I objected fairly strongly (and verbosely) to this
decision at the time, because I thought that it dealt with an
extremely esoteric case at the cost of significant potential for
confusion and added API complexity.  So far, it seems that both RDF
and XHTML treat the two cases as equivalent (though neither says so
explicitly) and I have yet to see an XML+Namespaces application that
actually cares about the distinction, but there's still time for me to 
be proven wrong.

It's always hard to guess when you're writing specs ahead of the
implementors, and many people whose opinions I respect disagreed with
me.

> 2)
> 
> <?xml version="1.0"?>
>   <doc xmlns="mydefault">
>     <mytag myatt="blah"/>
>   </doc>
> 
> To which namespace does the "myatt" attribute belong?  The spec says
> that the default namespace doesn't apply, however from question 1, we
> can infer that attributes take on the namespace of their parents.  I'm
> assuming, though, that "myatt" belongs to no namespace.  What shall I
> assume here?

This is exactly the same situation as above -- it doesn't matter
whether you're using explicit prefixes or a default Namespace.


All the best,


David

-- 
David Megginson                 david@m...
           http://www.megginson.com/

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@i...
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo@i... the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo@i... the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa@i...)

• References:
  • Attributes and namespaces
    • From: "Shawn Silverman" <shawn@a...>

• Prev by Date: Attributes and namespaces
• Next by Date: Re: Attributes and namespaces
• Previous by thread: Attributes and namespaces
• Next by thread: Re: Attributes and namespaces
• Index(es):
  • Date
  • Thread