W3C home > Mailing lists > Public > public-rdf-dawg@w3.org > July to September 2006

Re: ISSUE: Malformed literals and non-lexical literals

From: Seaborne, Andy <andy.seaborne@hp.com>
Date: Thu, 17 Aug 2006 12:39:34 +0100
Message-ID: <44E45576.7040109@hp.com>
To: Bijan Parsia <bparsia@cs.man.ac.uk>
CC: Pat Hayes <phayes@ihmc.us>, RDF Data Access Working Group <public-rdf-dawg@w3.org>



Bijan Parsia wrote:
> Thanks Andy, that was interesting.
> 
> On Aug 17, 2006, at 10:20 AM, Seaborne, Andy wrote:
> 
>> Output from ARQ (using the standalone expression evaluator because  
>> it explains what is going on; ASK queries mask this):
>>
>> Reformatted for email.
>>
>> ==== Test 1
>>
>> arq.qexpr "'-5'^^xsd:positiveInteger < 5" ==> error
>>
>> 10:06:46 WARN  NodeValue
>>     Datatype format exception: "-5"^^xsd:positiveInteger
>> Exception: Can't compare "-5"^^<http://www.w3.org/2001/ 
>> XMLSchema#positiveInteger> and 5
> 
> You say that the warnings do not affect the results.

Warnings are output as logging messages.  Ignore then for the value of the 
expression.  ARQ just outputs some logging stuff for caes where the user may 
find it useful.  As ARQ (and Jena and Joseki) uses Jakarta Commons Logging 
(and log4j usually) the user/app can pick and choose logging messages anyway.

 > But this is
> wrong behavior, yes?
> 
> """Specifically, FILTERs eliminate any solutions that, when  
> substituted into the expression, result in either an effective  
> boolean value of false or produce an error. """

This is showing the result of an expression evaluation, not the full filter 
evaluation.  That's why I didn't use an ASK query.

	Andy

> 
> So, given data
> 	:b :p "-5"^^xsd:positiveInteger.
> 	:b :q "5"^^xsd:positiveInteger.
> 
> and query
> 	SELECT ?x {?x :p ?L1. ?x :q ?L2. FILTER ?L1 < ?L2}
> 
> ARQ returns
> 	?x/:b
> 
> instead of no answers?
> 
>> ==== Test 2
>>
>> arq.qexpr "isLiteral('-5'^^xsd:positiveInteger)" ==> true
>>
>> 10:06:46 WARN  NodeValue
>>     Datatype format exception: "-5"^^xsd:positiveInteger
>> true
> 
> But the isLiteral evaluation is correct.
> 
>> The warnings are execution logging warnings and do not affect the  
>> results.
>> It is trying to consider things as one of the understood types/ 
>> values, and falling back to a plain RDF literal (but not it's value).
> 
> I don't know exactly what that means for the example above.
> 
> Cheers,
> Bijan.
> 
Received on Thursday, 17 August 2006 11:41:15 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:15:27 GMT