"RDFa Processing in detail" gets some details wrong?

In section http://www.w3.org/TR/rdfa-syntax/#s_rdfaindetail

I see:

[[
The value of base may change the initial value of [current subject]:

<html>
  <head>
    <base href="http://www.example.org/jo/blog" />
    <title>Jo's Friends and Family Blog</title>
    <link rel="foaf:primaryTopic" href="#bbq" />
    <meta property="dc:creator" content="Jo" />
  </head>
  <body>
    ...
  </body>
</html>
A parser should now generate the following triples, regardless of the
URL from which the XHTML document is served:

<http://www.example.org/jo/blog> foaf:primaryTopic <#bbq> .
<http://www.example.org/jo/blog> dc:creator "Jo" .
]]

but when I made that example into a test case for the RDFa parser
I'm developing, it failed, and I don't think the problem is
in my code.

That <#bbq> object should be <http://www.example.org/jo/blog#bbq>, no?


Then under 6.1.1.2. Using @about, the last 2 of the 5
"should generate" triples look wrong:

<#bbq> cal:summary "one last summer barbecue" .
<#bbq> cal:dtastart "2007-09-16T16:00:00-05:00"^^xsd:dateTime .

dtastart is an obvious typo.

But "one last summer barbecue" suggests there's some sort of
whitespace stripping in RDFa parsing, but step 9 in
section 5 makes it pretty clear that there is no whitespace stripping
in this case:

"The actual literal is either the value of @content (if present) or a
string created by concatenating the text content of each of the
descendant elements of the [current element] in document order."


These examples do help with understanding the spec quite a bit,
but these details are a huge distraction.

I strongly suggest (a) adding these examples to the test suite
and (b) establishing an automated check that they stay in sync
with the test suite.

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

Received on Wednesday, 3 February 2010 22:33:19 UTC