Re: Value type constraints

On Friday, March 27, 2015, Richard Cyganiak <richard@cyganiak.de> wrote:

> Jose,
>
> I suggest adding this as a proposed requirement.
>
> The xsd:string datatype doesn’t support language tags though. In “old”
> RDF, language-tagged literals were special and didn’t have a datatype at
> all. In RDF 1.1, they have the datatype rdf:langString [1]. So you’d have
> to say:
>
>     sh:property [
>         sh:predicate rdfs:label;
>         sh:valueDatatype rdf:langString;
>         sh:languageTag "es";
>     ].
>
> Or maybe the use of sh:languageTag could imply that value for
> sh:valueDatatype. Then you could just say:
>
>     sh:property [
>         sh:predicate rdfs:label;
>         sh:languageTag "es";
>     ].
>
> This would work, as rdf:langString is the only datatype that supports
> language tags anyway [2].
>
> +1 for that simplified variant.  I was going to suggest it also...

I think there is already a requirement about language tags: R6.4






> Best,
> Richard
>
>

>
> [1] http://www.w3.org/TR/rdf11-concepts/#section-Datatypes
> [2] http://www.w3.org/TR/rdf11-concepts/#section-Graph-Literal
>
>
> > On 27 Mar 2015, at 17:02, Jose Emilio Labra Gayo <jelabra@gmail.com
> <javascript:;>> wrote:
> >
> >
> > sh:literalType instead of sh:datatype - since it is about literal values
> >
> > For me, both are ok.
> >
> > I would also like to suggest a way to constraint the language tag. For
> example, I would like to express concepts that have rdfs:label.
> > ex:shape a sh:Shape
> >    sh:property [
> >      sh:predicate rdfs:label ;
> >      sh:valueDatatype xsd:string ;
> >
> >      sh:languageTag "es"
> >    ] .
> >
> > I have no preference for some specific name of that property, but I
> think there should be some way to constraint the language tag of a string
> literal.
> >
> > Best regards, Jose Labra
> >
> >
> > On 3/27/15, 7:30 AM, "Richard Cyganiak" <richard@cyganiak.de
> <javascript:;>> wrote:
> >
> > >Or maybe even:
> > >
> > >    sh:type - for constraining the value¹s rdf:type
> > >    sh:datatype - for constraining the value¹s literal datatype
> >
> >
> >
> >
> >
> >
> > --
> > -- Jose Labra
> >
>
>

-- 
-- Jose Labra

Received on Friday, 27 March 2015 19:35:24 UTC