W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > June 2002

rdf:Description & test cases

From: Jeremy Carroll <jjc@hplb.hpl.hp.com>
Date: Fri, 7 Jun 2002 08:39:05 +0200
To: <w3c-rdfcore-wg@w3.org>
Message-ID: <MABBLGKMPIJFCKFGDBEPAENCCAAA.jjc@hplb.hpl.hp.com>

Issues:
+ legal uses of rdf:Description
+ obsoleting rdf-ns-prefix-confusion/test0007



Good news:

I now have a (not yet public) version of ARP that passes nearly all the test
cases from the April publication. This is driven from the Manifest file

Bad news:

the errors ...

The ones it doesn't attempt are:

(Manifest is unhelpful, tests themselves are probably OK)
nested-bag-IDs/test010a
nested-bag-IDs/test011a
nested-bag-IDs/test012a
nested-bag-IDs/test010b
nested-bag-IDs/test011b
nested-bag-IDs/test012b

Should be obsoleted:

rdf-ns-prefix-confusion/test0007

Plain wrong (IMO):
rdfms-rdf-names-use/test-013
rdfms-rdf-names-use/test-026

i.e. test-013 legal
  <rdf:RDF
 xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
  >
    <rdf:Description
        rdf:about="http://example.org/node1">
      <rdf:Description
        rdf:resource="http://example.org/node2" />
     </rdf:Description>
  </rdf:RDF>

test-026 legal

<rdf:Description
    rdf:about="http://example.org/node1"
    rdf:Description="string" />

These are about rdf:Description being used as a property name.
I see rdf:Description as a syntactic word like rdf:ID or rdf:about, and it
is not and never can be the predicate of a triple. (Not withstanding my
proposal about the RDF Graph!).

Also I did not find the minute approving these test cases; the link from the
Test Cases WD takes us to the action to create them.

http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001Nov/0651.html

[[[

   ACTION 2001-11-30#4 Dave Beckett: Create test cases covering reserved
use of rdf names in this proposal.

]]]

(Of course I wouldn't raise a procedural question if I hadn't discovered
that I disagreed with the outcome!)

Other Manifest errors (Apr 29) include:
 missing test:warning on

unrecognised-xml-attributes/test001
unrecognised-xml-attributes/test002

and possibly on some of the
rdfms-rdf-names-use/test-0XX


ARP is also happy with all the PENDING test cases in the April 29 zip.

I should get ARP running off the web site today, currently I am testing
against the zip; it will fail the charmod-uri tests still since I haven't
sorted that yet.

Jeremy
Received on Friday, 7 June 2002 02:33:12 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:49:10 EDT