Re: shapes-ISSUE-194 (valueStem): stems in value sets

Both cases can be covered with existing features (sh:or and sh:not) and 
I am against adding redundant special handling just for stem. The same 
argument could be applied to almost every other constraint type, e.g. 
sh:datatype. Unless I am missing something, I propose closing without 
action.

Holger


On 8/11/2016 23:39, RDF Data Shapes Working Group Issue Tracker wrote:
> shapes-ISSUE-194 (valueStem): stems in value sets
>
> http://www.w3.org/2014/data-shapes/track/issues/194
>
> Raised by: Eric Prud'hommeaux
> On product:
>
> The current SHACL definition for stem (see 4.5.4 sh:stem <http://w3c.github.io/data-shapes/shacl/#StemConstraintComponent>) treats them as a Parameter while ShEx uses them in value sets (see 4.4.6 Values Constraint <https://shexspec.github.io/spec/#values>). The SHACL definition use doesn't offer much value over an XSD pattern. It's pretty common for clinical value sets to require any value with starting with one of a number of base IRIs (e.g. terminologies like LOINC, SNOMED, CPT, ...).
>
> Another diff is that ShEx value sets have exclusions which can in turn be stems (see 10 Value Sets <http://shex.io/primer/#h-value-sets>)
>
>
>

Received on Friday, 11 November 2016 00:27:37 UTC