Updated SPARQL Query Results XML Format draft

I've made the following changes as discussed in the last telcon at
http://www.w3.org/2001/sw/DataAccess/rf1/
(CVS Revision: 1.30 $ of $Date: 2005/07/13 13:51:42 $ or later)

1. Removed xsi:type from datatyped literals as W3C XML Schemas could not
record this as an optional attribute on an element with literal content.
Optional attributes require a ComplexType however xsi:type is only
allowed on a SimpleType.

2. Moved boolean (boolean results) to be an alternate for results
(variable binding results) to allow earlier detection of the results
format after feedback from AndyS in
http://lists.w3.org/Archives/Public/public-rdf-dawg/2005AprJun/0416.html

3. Added link to allow linking to metadata about the query results after
a comment by Ian Harrison in
http://lists.w3.org/Archives/Public/public-rdf-dawg-comments/2005Jun/0028.html

4. Switched XML namespace to http://www.w3.org/2005/06/sparqlResults to
match W3C policy for XML namespace names

5. Added publication note to switch result2.xsd to final location


However, I've also noticed a couple of items in Red Ink that still need
thinking about:

1. How/if to record duplicates in results. (Section 2.3.3)

When ORDER BY is given, the result format may record index="1",
index="2" on the <result> element.  (Side issue - "may" or "should" do
this?)

However when there are duplicates should it generate indexes 1, 2, 2, 3
where items #2 and #3 are duplicates?  (A query with ORDER BY but no
SELECT DISTINCT).


2. XML Schemas and normativeness (Section 4)

I'm happy with the RELAX NG to be normative but my confidence is lower
with the WXS being right or useful. AndyS told me it worked for working
with WSDL.  Just go with both and not worry?

Dave

Received on Wednesday, 13 July 2005 13:59:54 UTC