Re: Comments on ORDER BY

Richard Cyganiak wrote:
> Comments on 
> http://www.w3.org/2001/sw/DataAccess/rq23/
> regarding the ORDER BY clause:
> 
> 
>>From the grammar:
> 
> | [16] OrderExpression ::= FunctionCall | Var
> | [59] FunctionCall ::= URI '(' ArgList ')'
> 
> This doesn't allow expressions like "?a + ?b" in the ORDER BY clause. Is
> this intentional?

The working group has decided to allow expressions in ORDER BY clauses.

To avoid ambiguities, expressions must be enclosed in ().

	Andy

> 
> If not, this sentence from section 10.1 also needs updating:
> 
> | An ordering condition can be a variable or a function call.
> 
> 
> 
>>From section 10.1:
> 
> | When ordering a solution sequence involves an expression, it
> | is possible that the ordering conditions do no give a
> | completely determined ordering for the sequence. In this case
> | the ordering of solutions that are not distinguished, is not
> | determined.
> 
> What does the first part of the sentence refer to? AFAICT, ordering a
> solution sequence always involves an expression. There's also some
> redundancy with this later paragraph:
> 
> | If the ordering criteria do not specify the order of values,
> | then the ordering in the solution sequence is undefined.
> | However, an implementation must consistently impose the same
> | order so that applying LIMIT/OFFSET will not miss any solutions.
> 
> Also, s/ no / not / in the first paragraph.
> 
> 
> 
> Section 10.1 defines an order for different types of RDF terms, starting
> with "no value assigned to the variable":
> 
> | 1. (Lowest) no value assigned to the variable in this solution.
> 
> I have two issues with this: There might be no "the variable", e.g.
> "ORDER BY my:func(?var1, ?var2)". And it should be more explicit if the
> sentence applies only to unbound variables, or also to solutions that
> generate type errors, e.g. "ORDER BY xsd:integer(?x)" where ?x is not
> bound to an appropriate literal. 
> 
> 
>

Received on Tuesday, 7 June 2005 17:48:02 UTC