W3C home > Mailing lists > Public > public-ws-semann@w3.org > May 2007

RE: Test suite SAWSDL update

From: Jonathan Marsh <jonathan@wso2.com>
Date: Sun, 27 May 2007 08:06:30 -0700
To: "'Jacek Kopecky'" <jacek.kopecky@deri.org>
Cc: <public-ws-semann@w3.org>
Message-ID: <016501c7a070$9c101730$d4304590$@com>

OK, thanks for the analysis.  I copied all the testcases again (and updated
to the new namespace) and fixed some of the test metadata so all the
testcases would run.  I checked a couple and found a QName reference bug in
my implementation, and so got different results for many of the testcases 5
and above.

I didn't do a manual check of _all_ the results to ensure they are correct
to my understanding of the spec, perhaps you could take a look at each of
the test-suite/results/Baseline/SAWSDL*.canonical.wsdlcm files and see if
the results are correct.

It is a known limitation that wsdl-xslt doesn't remove duplicate uris during
a merge.  That's difficult and so I skipped it until we have a second
implementation to test ;-).

Likewise the test suite has a limitation in that the canonical format
doesn't sort the list of uris.  Again not vital until we have a second
implementation.

Jonathan Marsh - http://www.wso2.com - http://auburnmarshes.spaces.live.com
 

> -----Original Message-----
> From: Jacek Kopecky [mailto:jacek.kopecky@deri.org]
> Sent: Friday, May 25, 2007 6:53 AM
> To: Jonathan Marsh
> Cc: public-ws-semann@w3.org
> Subject: RE: Test suite SAWSDL update
> 
> Hi Jonathan, the answer is yes.
> 
> The behavior should show up on a number of the test cases:
> 05-simpletype-annotation.wsdl propagates mRef from simple type
> 06-complextype-annotation.wsdl propagates mRef from complex type
> 06-multiple-complextype-annotation.wsdl checks set merging on
> propagation
> 10-type-lifting.wsdl propagates liftingSM from simple type
> 12-* propagate loweringSM from complex type
> 13-* have propagation overriding
> 
> Note that 06-... are updated because previously only an embedded
> element
> declaration referenced the global type definition with a sawsdl
> annotation, so no propagation would show up in the WSDL components.
> 
> If you generate the results for these test cases, I'll be happy to
> eyeball them for correctness. 8-)
> 
> If you wish, for some of the cases I could also add test files that are
> equivalent after propagation, i.e. as though the propagation was done
> manually.
> 
> Jacek
> 
> On Tue, 2007-05-22 at 10:53 -0700, Jonathan Marsh wrote:
> > So wsdl-xslt should add all three annotations to the Element
> Declaration
> > Component that appear on the Type Definition Component that it
> references
> > (if it does reference one)?
> >
> > Hoping the answer is yes I added this to wsdl-xslt, and did some
> temporary
> > test case tweaking to do a smoke test, but none of the test cases
> currently
> > in the WSDL test suite test this functionality.  Would you be willing
> to
> > write up some test cases to verify that my implementation is correct?
> >
> > Jonathan Marsh - http://www.wso2.com -
> http://auburnmarshes.spaces.live.com
> >
Received on Sunday, 27 May 2007 15:06:39 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 19:58:46 UTC