test case details by 26 Oct telcon

Last tuesday we struggled to make progress on the SOURCE
issue in part because I couldn't figure out exactly
how tests work.

Let's please get the testing details worked out
during, if not before, next week's teleconference.

http://www.w3.org/2001/sw/DataAccess/tests/

I hope to put the question on the first few tests...
http://www.w3.org/2001/sw/DataAccess/tests/#dawg-triple-pattern-001
thru
http://www.w3.org/2001/sw/DataAccess/tests/#dawg-triple-pattern-003

The organization is kinda inconvenient for me;
the description of how the tests actually work is in
 http://www.w3.org/2001/sw/DataAccess/tests/README.html
and even there it's not quite clear to me exactly what
the definition of correct results is. Hmm...

"A test passes if the graph from the action is the same as the graph
named in the result. "Same as" is are the two graphs bNode isomorphic:
same shape, same labels."

I guess I need to study that a bit more.

I'm particularly interested in whether the result
{solutionX, solutionX} is considered a passing or
failing result when the expected result is {solutionX}.


I had hoped to do more homework before sending this out;
sorry I don't have more details.

I only got as far as studying the definitions again.
http://lists.w3.org/Archives/Public/public-rdf-dawg/2004OctDec/0151.html

Anybody who finds time to work on this, please connect
to irc://irc.w3.org:6665/dawg while you're at it, if
that's not too much hassle.

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