W3C home > Mailing lists > Public > public-rdf-dawg-comments@w3.org > June 2006

Blank nodes in queries and variables

From: Jan Wielemaker <wielemak@science.uva.nl>
Date: Mon, 12 Jun 2006 17:06:08 +0200
To: public-rdf-dawg-comments@w3.org
Message-Id: <200606121706.08772.wielemak@science.uva.nl>


After fixing the remaining obvious bugs in the test-cases my Prolog parser
now has 100% coverage on the syntax.  Time for semantics ...

Just a brief question for my understanding. Is the following true? If
not, please explain why ...

	Variables (?name) and blank nodes (_:name) in queries are
	closely related.  The name-spaces for variables and blank nodes
	are not related (i.e. ?a is not related to _:a).  Blank nodes
	do not appear in the result-set of SELECT *.  Where a variable
	matches any resource or literal, blank nodes only match resources.

Notably the latter is unclear as -if I recall well-, the SPARQL document
anticipates on allowing literals as subject in the future.  Comments?

I suggest to add a paragraph to the SPARQL definition explaining the
relation between blank nodes and variables.

	Cheers --- Jan

P.s.	Ask if you want the diff between the original (XML) testcases
	and the fixed.
Received on Monday, 12 June 2006 15:06:29 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:52:07 UTC