- From: Peter F. Patel-Schneider <pfpschneider@gmail.com>
- Date: Tue, 28 Jul 2015 07:42:22 -0700
- To: Holger Knublauch <holger@topquadrant.com>, RDF Data Shapes Working Group <public-data-shapes-wg@w3.org>
The current editor's draft uses a particular mechanism for the extension langauge to invoke named expressions, but the particular mechanism there does not need to be used to invoke named shapes, independent of the syntax used for this purpose. As far as I can tell, there are at least two potenntial connections between the extension language and the core language: 1/ Invoking named shapes from the extension language. 2/ Embedding core constructs (e.g., conjunctions) in expressions written in the extension language. A clearer statement of ISSUE-26 would be to just say: Can the extension language invoke named shapes. peter On 07/27/2015 04:39 PM, Holger Knublauch wrote: > ISSUE-26 [1] is about whether extensions such as SPARQL can invoke the > high-level language. I believe this needs to be answered individually for each > extension language. For SPARQL, my proposal to resolve this is as currently > drafted, by relying on the sh:hasShape function: > > http://w3c.github.io/data-shapes/shacl-ref/#hasShape > > I would be OK with a solution that makes this function *optional* if we find > work-arounds to all places where it is currently used. But I believe this > capability is generally useful and greatly extends what users can achieve with > SHACL. > > Holger > > [1] http://www.w3.org/2014/data-shapes/track/issues/26 >
Received on Tuesday, 28 July 2015 14:43:05 UTC