The following feedback is a result of parsing the test cases with RDFLib's rdf/xml parser. --eikeon
The following version of the test cases was used: http://www.w3.org/2001/08/rdf-test/ (CVS 1.156).
Along with RDF/XML Syntax Specification (Revised) W3C Working Draft 6 November 2002 yet to be published at: http://www.w3.org/TR/2002/WD-rdf-syntax-grammar-20021108/
The comments are currently only on test cases that are indicated as approved in the Manifest.
The 6 positive test cases that follow are the same ones I pointed out before. And at least 4 of the 6 seem to be incorrect? Am I not running against the latest test cases?
Of the 126 approved positive test cases 120 are passing with no issues along the way. The remaining 6 test cases follow:
The expected output contains the following triple:
<http://www.w3.org/2000/10/rdf-tests/rdfcore/rdf-containers-syntax-vs-schema/test004.rdf#e4> <http://foo/bar> "foobar".where RDFLib is getting:
_:a6 <http://foo/bar> "foobar".Best I can tell, the triple should be the result of an empty property element. In which case, the subject comes from the r in the otherwise section and since there is no rdf:resource attribute, r should be a BNode.
This test case is not working for the same reason as test004 above.
Is rdf:aboutEach allowed as an attribute here? This test case fails since the parser reports the invalid property attribute of http://www.w3.org/1999/02/22-rdf-syntax-ns#aboutEach.
It appears that the xml:base in the expected output does not match the actually xml:base as it is missing an "xmlbase/".
Python's urlparse.urljoin is returning:
#fooas the result of joining:
mailto:Jeremy_Carroll@hp.com and #fooHave yet to dive into RFC 2396 to see which is correct.
Have yet to deal with any charmod issues.
Of the 38 approved negative test cases 26 are passing with no issues along the way. Three are due to RDFLib not yet supporting the chmod stuff. The other 9 are a result of the rdf-ns-prefix-confusion tests no longer being valid due to a change in the RDF/XML syntax.
As now stated in section 6.1.4 Attribute Event: "support for a limited set of non-namespaced names is REQUIRED". Which invalidates this test case as well as error0002 -- error0009.
Have yet to deal with any charmod issues.
Have yet to deal with any charmod issues.
Have yet to deal with any charmod issues.