Re: [OK?] Re: some questions on ORDER BY[closed]

Quoting Lee Feigenbaum <feigenbl@us.ibm.com>:

> Hello Faisal,
> 
> I received the following blank reply from you this morning; just wanted to 
> point that out in case the lack of content was an error. We'd appreciate a 
> reply on this issue to the public-dawg-comments list whenever you might 
> have a chance.
> 
> thanks,
> Lee
> 
> Faisal.Alkhateeb@inrialpes.fr wrote on 03/06/2006 03:42:25 AM:
> 
> > Quoting Lee Feigenbaum <feigenbl@us.ibm.com>:
> > 
> > > 
> > > Hello Faisal,
> > > 
> > > Faisal.Alkhateeb@inrialpes.fr wrote on 03/02/2006 05:29:44 AM:
> > > 
> > > > Hello,
> > > > 
> > > >   I have some questions regarding ORDER BY. If we have the following 
> 
> > > one:
> > > > 
> > > >               ORDER BY (lang (_:b))
> > > > How we can evaluate this and then ordering the set of solutions?
> > > > 
> > > > Also if we have:
> > > > 
> > > >          ORDER BY (<http://example.org>)
> > > > How we can order the results?
> > > 
> > > The SPARQL Query Language specification does not define ordering on 
> > > bnodes; and while it does define ordering on IRIs, ordering by a 
> constant 
> > > IRI will, of course, not produce a total ordering. For both of these 
> > > cases, then, the following text from section 10.1.3 of the 
> specification 
> > > applies ( http://www.w3.org/2001/sw/DataAccess/rq23/#modOrderBy ):
> > > 
> > > """
> > > If the ordering criteria do not specify the order of values, then the 
> > > ordering in the solution sequence is undefined.
> > > """
> > > 
> > > > What are the semantics of Order Conditions when they don not 
> > > containvariables 
> > > > like the above examples?
> > > 
> > > The semantics of ORDER BY are not altered by the presence or lack of 
> > > variables (though many instances of ORDER BY without variables will 
> (as 
> > > above) not result in a total ordering of the solutions). The text from 
> the 
> > > above cited section of the Last Call document reads:
> > > 
> > > """
> > > The "<" operator (see the Operator Mapping) defines the relative order 
> of 
> > > pairs of numerics, xsd:dateTimes and xsd:strings.
> > > 
> > > IRIs are ordered by comparing the character strings making up each IRI 
> 
> > > using the "<" operator.
> > > 
> > > SPARQL also defines a fixed, arbitrary order between some kinds of RDF 
> 
> > > terms that would not otherwise be ordered. This arbitrary order is 
> > > necessary to provide slicing of query solutions by use of LIMIT and 
> > > OFFSET.
> > > 
> > >    1. (Lowest) no value assigned to the variable or expression in this 
> 
> > > solution.
> > >    2. Blank nodes
> > >    3. IRIs
> > >    4. RDF literals
> > >    5. A plain literal is lower than an RDF literal with type 
> xsd:string of 
> > >  the same lexical form.
> > > 
> > > If the ordering criteria do not specify the order of values, then the 
> > > ordering in the solution sequence is undefined.
> > > 
> > > Ordering a sequence of solutions always results in a sequence with the 
> 
> > > same number of solutions in it, even if the ordering criteria does not 
> 
> > > differentiate between two solutions.
> > > """
> > > 
> > > Please let us know if you find this response satisfactory.
> > > 
> > > thanks,
> > > Lee
> > > 
> > > 
> > > 
> > > 
> > 
> > 
> 
> 
>

Received on Monday, 6 March 2006 13:47:02 UTC