Newly approved TCs - 122, 123, 126, 131, 134 and 140

[Manu Sporny]

The XHTML+RDFa test suite has been updated to include a number of test
cases that we've approved over the past 4 months (but I failed to mark
as approved until just now).

I vaguely recall us approving TCs 122, 123 and 126, but could not find a
record of us doing so. Those test cases pass at least 3 implementations
and there was nothing that jumped out at being invalid about those
tests. However, if somebody else could check them out and make sure I
didn't make a mistake, that would be great.

There is a clear record of TCs 131, 134 and 140 being approved.

We had said that TC 140 shouldn't generate the triple listed in the
SPARQL, but I can't remember why now (and it wasn't minuted). TC140 has
been published as a negative test, but Philip meant it to be a positive
test. Why should TC 140 not generate the triple listed in the SPARQL?

-- manu

--
Manu Sporny (skype: msporny, twitter: manusporny)
President/CEO - Digital Bazaar, Inc.
blog: Establishing an Open Digital Media Commerce Standard
http://blog.digitalbazaar.com/2009/09/28/a-digital-content-commerce-standard/

[Ivan Herman-2]

Manu Sporny wrote:

> The XHTML+RDFa test suite has been updated to include a number of test
> cases that we've approved over the past 4 months (but I failed to mark
> as approved until just now).
>
> I vaguely recall us approving TCs 122, 123 and 126, but could not find a
> record of us doing so. Those test cases pass at least 3 implementations
> and there was nothing that jumped out at being invalid about those
> tests. However, if somebody else could check them out and make sure I
> didn't make a mistake, that would be great.
>

TC 122: seems o.k. to me
TC 123: seems o.k. to me (and I distinctly remember the discussion on
that, but I did not find the right mail references)
TC 126: seems o.k. to me

Ivan

> There is a clear record of TCs 131, 134 and 140 being approved.
>
> We had said that TC 140 shouldn't generate the triple listed in the
> SPARQL, but I can't remember why now (and it wasn't minuted). TC140 has
> been published as a negative test, but Philip meant it to be a positive
> test. Why should TC 140 not generate the triple listed in the SPARQL?
>
> -- manu
>

--

Ivan Herman, W3C Semantic Web Activity Lead
Home: http://www.w3.org/People/Ivan/
mobile: [phone]
PGP Key: http://www.ivan-herman.net/pgpkey.html
FOAF: http://www.ivan-herman.net/foaf.rdf

[Philip Taylor-5]

Manu Sporny wrote:
> We had said that TC 140 shouldn't generate the triple listed in the
> SPARQL, but I can't remember why now (and it wasn't minuted). TC140 has
> been published as a negative test, but Philip meant it to be a positive
> test. Why should TC 140 not generate the triple listed in the SPARQL?

If it's agreed that no triple should be generated for this test case,
then the SPARQL test is inadequate for finding implementation bugs. In
particular, librdfa generates the triple:

<http://rdfa.digitalbazaar.com/test-suite/test-cases/xhtml1/0140.xhtml>
<http://rdfa.digitalbazaar.com/test-suite/test-cases/xhtml1/0140.xhtml_>
"Test"

which looks like a clear error, yet the test case does not detect it
since it's only testing for absence of one specific triple. Some other
implementations have different errors (treating it like a blank node).

I would suggest changing the query to say:

   # This is a negative test, no triples may be generated.
   ASK WHERE {
     ?s ?p ?o .
   }

--
Philip Taylor
pjt47@...