W3C Basic XML Schema Patterns for Databinding
- Dear soapbuilders,
You may be interested to take a look at the work of the
W3C XML Schema Patterns for Databinding WG:
which has published a Last Call Working Draft of the
"Basic XML Schema Patterns for Databinding 1.0" specification:
The Last Call review period closed on Jan 12 2007.
We look forward to receiving any comments you may have -
instructions on filing comments can be found in the Status section.
To accompany the spec, we are building a collection of
an XSLT 2.0 stylesheet to report on the patterns detected in a
given schema or WSDL document:
which we plan to provide as a W3C service, RSN.
Note, many schema constructs remain "unknown" we're working
on that using our collection of example Schema documents
wrapped up inside WSDL 1.1, 2.0 along with instance documents
wrapped up as POX, SOAP 1.1 and 1.2 messages:
But most importantly we're in the process of developing a test suite
around the examples. You might like to consider trying your
with the testsuite and help by contributing test cases and test logs
which we can use
to publicise and improve the "state of the art" of databinding
I'm at XML 2006 next week, so consider attending the presentation:
and feel free to collar me directly if you'd like to learn more about
efforts and how you may contribute and participate.
on behalf of the W3C XML Schema Patterns for Databinding WG
- Hi Steve
> > The Last Call review period closed on Jan 12 2007.
> Paul, when exactly was this email posted :)
> > http://www.w3.org/2002/ws/databinding/testsuite/This one goes further - the spec is the test suite - it's
> I celebrate any standard that is accompanied by a test suite.I just
> have one little process question.
mostly just a bunch of XPath "patterns" bundle in to a XSLT
detector and a load of examples we wrapup in SOAP and WSDL and
fire at toolkits.
> Will the specification be considered 'final' before or after the testNo. For Basic, if a "state of the art" toolkit doesn't work with a
> suite is written?
it's "at risk" so likely to be removed before we move to
For "Advanced" we've yet to decide on our criteria: the test suite
will be there, but what happens if a pattern widely used in schemata
in the wild is ignored by all databinding implementations?
So the risk is we don't hear from implementers at all: maybe their
kit is defunct,
and not "state of the art", maybe it's top-secret, or maybe the owners
believe they support the whole of schema and don't need to bother
.. you decide :-)
// hope this doesn't take 36 hours to hit the list ..