Re: Sets of related test cases

On Thu, 2007-02-15 at 09:32 -0500, Clark, John wrote:
> I put in some work on action 2 from the 2007-02-07 meeting[0] (which was
> actually an action for both Chime and myself).  The test suite only
> required minor modifications; I should have attached both the patch and
> one new file - noxinclude1.rdf - to this email.

The patch modifies testlist3.rdf, but that file is generated from
testlist3.html. See section "Manifest Format"
  http://www.w3.org/2001/sw/grddl-wg/td/testlist1#testlistdoc
and the transformation
  http://www.w3.org/2001/sw/grddl-wg/td/aboutTests

I might be able to back-port it, but I'd have
to actually think to do that, and I'm just in admin mode right now...


>   I also made minor
> modifications to Chime's GRDDL.py client.  I should have also attached
> that entire program to this email, but please be aware that it is in an
> experimental state until he has a chance to review it.

You did attach the whole thing. And it seems your work doesn't
include the DEBUG factoring Chime did, so I applied your
patch against the 6d7d38577f3e version, resulting in
  http://homer.w3.org:8000/?cs=97f71f8cc60e

We now have two heads that should be merged; this one and
the one resulting from Chime's DEBUG factoring work.
  http://homer.w3.org:8000/?cs=8a5bc100babf


> As part of this work, I've been thinking about the fact that cases like
> this can have multiple GRDDL results.

That looks like a pretty reasonable idea, though I'm not done
thinking it thru.

Please think about how to push it back into the HTML source of the
test manifests, updating the aboutTests transformation while you're
at it.



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

Received on Friday, 16 February 2007 02:56:36 UTC