Re: Invalid XHTML Re: Another test suggesting change in the spec

On Mon, 2007-04-23 at 17:23 +0100, Jeremy Carroll wrote:
> 
> See
> http://www.w3.org/2001/sw/grddl-wg/td/html-and-grddl-xform-attr
> for a test case.
>
> input
> 
> <?xml version="1.0" encoding="UTF-8"?>
> <html xmlns="http://www.w3.org/1999/xhtml"
>     xmlns:g="http://www.w3.org/2003/g/data-view#"
>     g:transformation="two" >
>    <head profile="http://www.w3.org/2003/g/data-view">
>      <title>HTML Doc with grddl:transformation Attribute</title>
>    </head>
> 
>    <body>
>      <h1>HTML Doc with grddl:transformation Attribute</h1>
> 
> 
>      <p>GRDDL results for this entry come both from this inline
>        <a rel="GRDDL
>        transformation" href="one">one</a>,
>        and the transformation on the root element.</p>
> 
>    </body>
> </html>
> 
> The current spec does not license transformation "one", since the 
> document is not valid, by virtue of the g:transformation and xmlns:g 
> attributes on the root element.
> 
> Some choices are:
> a) change the spec along the lines suggested

The suggestions so far haven't affected the normative
rules/assertions, so they don't support this test either.

The rule currently starts:

  Given an XHTML family document[XHTML] with XPath root node N,

I would love it if the XHTML specs provided a definition
of the set of documents we're interested in, but I can't
find it.

I suppose we could derive a specification from our code.
In your message of Mon, 23 Apr 2007 17:13:06 +0100, you wrote...

> But currently our implementations do not in general check for DTD 
> validity, yet formally they should since the rel="transformation"
thing 
> is only defined for DTD valid docs.

What _do_ the implementations check or depend on?
MIME type, XML-wf-ness, and root element namespace?

If so, I'd specify something like this...

  If an information resource has a text/html representation
  whose body is an XML document whose root element
  bears the local name 'html' and the
  namespace name 'http://www.w3.org/1999/xhtml', then ...

> b) reject the test and remain silent
> c) modify the test result to show only two
> d) modify the test to have two different answers
>       two
>     and
>       one + two
> 
> I would oppose (c) and abstain from (b) and (d), while I support (a).
> I would be happy if the test was also informative.
> 
> Jeremy

-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/
D3C2 887B 0F92 6005 C541  0875 0F91 96DE 6E52 C29E

Received on Monday, 23 April 2007 17:45:40 UTC