- From: Peter F. Patel-Schneider <pfpschneider@gmail.com>
- Date: Wed, 22 Feb 2017 18:39:18 -0800
- To: "public-rdf-shapes@w3.org" <public-rdf-shapes@w3.org>
Last year I had produced two implementations of SHACL, one of a variant of SHACL Core and one of a variant of SHACL-SPARQL. Both implementations used the strategy of converting each shape with targets into a single SPARQL SELECT query, evaluating this query on the data graph, and using the resultant solution set as top-level validation results. I was planning on converting at least one of these implementations to the final version of SHACL Core. However, I will not be doing so with the current definition of SHACL. My implementation strategy does not produce validation results for any non-top-level constraint violations. It thus does not conform to the current definition of SHACL, which requires that all constraint violations MUST produce validation results. The validation reports that my implementation would create could conform to the rest of the current definition of SHACL, and could easily conform to most of the rest of the current definition of SHACL. Peter F. Patel-Schneider Nuance Communications
Received on Thursday, 23 February 2017 02:39:57 UTC