- From: Pierre-Antoine CHAMPIN <champin@cpe.fr>
- Date: Thu, 19 Aug 1999 18:22:48 +0000
- To: ML RDF-comment <www-rdf-comments@w3.org>, Sergey Melnik <melnik@db.stanford.edu>
By the way, I found another strange "Literalisation" in SiRPAC,
in method processTypedNode, I made the following modifs :
// special case: should the typedNode have aboutEach attribute,
// the type predicate should distribute to pointed
// collection also -> create a child node to the typedNode
Enumeration eTargets = typedNode.targets ();
if (sAboutEach != null &&
eTargets.hasMoreElements()) {
Element newPredicate = new Element (RDFMS + "type",
new AttributeListImpl());
//-> CHANGED by PA
// Data newData = new Data (typedNode.name());
// newPredicate.addChild (newData);
newPredicate.resource (typedNode.name());
//->end CHANGED by PA
typedNode.addChild (newPredicate);
} else {
because puting a Data element into a 'type' one
resulted in a 'type' arc pointing to a Litteral
(which looked just the same but is invalid)
Pierre-Antoine
Received on Friday, 20 August 1999 03:55:26 UTC