Re: shapes-ISSUE-44 (Graph dependencies): How to express dependencies between graphs [SHACL Spec]

Arthur,

you seem to assume that a graph only contains a single root resource? 
That would just be a special case of named graphs. My use cases are

- if someone opens a file in an editor, the system needs to be able to 
quickly figure out which other graphs need to be included, to build a 
union graph that can be queried by the editing tool

- same for constraint checking - if someone validates a whole graph, we 
need to know which shapes to collect, where the required SHACL templates 
are defined etc. Following all references in linked data style is often 
not feasible and not the best approach (although this is another option 
that we could support too).

Holger



On 4/17/15 6:50 AM, Arthur Ryman wrote:
> On Thu, Apr 16, 2015 at 4:34 PM, Holger Knublauch
> <holger@topquadrant.com> wrote:
>> oslc:instanceShape looks like sh:nodeShape [1] to me. As far as I understand
>> the OSLC spec, sh:include/sh:library would be closer to oslc:resourceShape.
> Holger,
>
> Your writeup of ISSUE-44 says:
> "1) sh:include - points from a graph (e.g. of instances) to other
> graphs (e.g. of class definitions)"
>
> If you are pointing from an instance graph to the shape that describes
> it, then in OSLC you use oslc:instanceShape. This is like on XML
> document pointing to an XML Schema. Is that what sh:include is for? If
> not, I think your writeup is misleading.
>
> -- Arthur

Received on Thursday, 16 April 2015 21:12:43 UTC