The following issue has been SUBMITTED. ====================================================================== http://bugs.librdf.org/mantis/view.php?id=508 ====================================================================== Reported By: normang Assigned To: ====================================================================== Project: Raptor RDF Syntax Library Issue ID: 508 Category: api Reproducibility: always Severity: major Priority: normal Status: new Syntax Name: Turtle ====================================================================== Date Submitted: 2012-06-10 21:12 Last Modified: 2012-06-10 21:12 ====================================================================== Summary: Raptor objects to possibly valid Turtle syntax Description: I'm not sure whose bug this is.
Protégé can save files in Turtle format. The way it saves the triples which assert mutual disjointness between classes is in a syntax which Raptor objects to. See the reduction below. Raptor is, I think, objecting to the '.' on line 9. Looking at the Turtle grammar at <http://www.w3.org/TR/turtle/#sec-grammar>, this file is indeed syntactically invalid: the [...] is not an instance of 'triples' in production [2]. Perhaps therefore the bug is Protégé's (and I'm about to mail the relevant feedback list and point them to this bugreport). However another possibility is that the error is in the grammar in the TR (and so in Raptor?). In section 2.1, we find the text: "A generated blank node may also be made with [] which is useful to provide the subject of RDF triples for each pair from the predicateObjectList or the root of the collection." This is a rather clumsily-written sentence, so that one reading of it suggests that [...] is valid, and so seems to suggest that the syntax in the reduction below (and thus the syntax emitted by Protégé) is acceptable. If so, it would appear that production [2] should be 'directive "." | triples "." | blankNodePropertyList'. In fact, I think this sentence is simply noting that the sequence "[]" may denote a blank node, referring to production [94s]. I do have the impression that I've seen this construction (namely blankNodePropertyList at the top level) somewhere, but can't now put my finger on it. It may be a bit of Turtle 'folk grammar'. More likely, it's a bit of Notation 3 (see <http://www.w3.org/TeamSubmission/n3/#bnodes>, under "Square bracket blank node syntax") which has been carried over into Protégé. So: I think on balance that this is Protégé's error, but I'm reporting it here because (a) it looks as if, despite protestations, Raptor can actually parse it (so the error might be accidental); (b) if this is a Turtle grammar problem, librdf.org will have more standing to report it than I will; and (c) Protégé doesn't have a public bugparade! Steps to Reproduce: % cat /tmp/try.ttl @prefix owl: <http://www.w3.org/2002/07/owl#> . @prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> . @prefix foo: <urn:example#>. [ rdf:type owl:AllDisjointClasses ; owl:members ( foo:Class1 foo:Class2 ) ] . % rapper -iturtle -ordfxml-abbrev /tmp/try.ttl rapper: Parsing URI file:///tmp/try.ttl with parser turtle rapper: Serializing with serializer rdfxml-abbrev rapper: Error - URI file:///tmp/try.ttl:9 - syntax error <?xml version="1.0" encoding="utf-8"?> <rdf:RDF xmlns:foo="urn:example#" xmlns:owl="http://www.w3.org/2002/07/owl#"; xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#";> <owl:AllDisjointClasses> <owl:members> <rdf:Description> <rdf:first rdf:resource="urn:example#Class1"/> <rdf:rest> <rdf:Description> <rdf:first rdf:resource="urn:example#Class2"/> <rdf:rest rdf:resource="http://www.w3.org/1999/02/22-rdf-syntax-ns#nil"/> </rdf:Description> </rdf:rest> </rdf:Description> </owl:members> </owl:AllDisjointClasses> </rdf:RDF> rapper: Parsing returned 6 triples % rapper --version 2.0.7 ====================================================================== Issue History Date Modified Username Field Change ====================================================================== 2012-06-10 21:12 normang New Issue ====================================================================== _______________________________________________ redland-dev mailing list [email protected] http://lists.librdf.org/mailman/listinfo/redland-dev
