I'd like to run the following scenario by you to confirm whether my
understanding of what I can and cannot do with TBC and SHACL is correct.

Suppose I have the following pattern:

[image: image.png]

I want to use a SHACL PropertyShape with sh:values to give me an inferred
"shortcut" relation from Class A to Class D. However, in doing so, I want
to filter on values of Prop B and Prop C, so it is not just a simple matter
of a 3-step sh:path statement.
I read your documentation at https://www.topquadrant.com/graphql/values.html
with interest on this topic, and have concluded that while a single
PropertyShape cannot achieve this, I can nest two PropertyShapes:
the first gives me a filtered "shortcut" relation from Class A to Class C
(filtering on Prop B), which I can then use in defining a second filtered
"shortcut" that filters on Prop C and gets me to Class D. This conclusion
is based on the fact that one cannot nest sh:values statements inside a
single PropertyShape definition - is this true? (If it IS possible to nest
sh:values statements, does the inner sh:values statement play the role of
sh:nodes for the outer filterShape?)

Corollary questions:
Must Class A be the targetClass of both of the above PropertyShapes?
Must I run the SHACL reasoner beforehand for the inferred relations to be
usable?

I also tried to achieve the same behavior with an embedded SPARQL query
instead of native SHACL statements, which allowed me to create the inferred
shortcut relation with a single SPARQL query. However, as you document at the
same link above <https://www.topquadrant.com/graphql/values.html>, under
the section "Use of Inferred Values using SPARQL", I can only use the
shortcut property if:
a) I use the       (?a ?shortcut-property) tosh:values ?d          syntax,
or
b) I run the SHACL reasoner before making a query.

Is this correct?

Finally, you also solicit feedback on whether the integration with SPARQL
should work more directly, without the use of the tosh:values magic
property, to which I will add a strong "YES PLEASE!". I say this because my
colleagues are looking for use-cases where the inferred shortcut property
is indistinguishable from a regular explicit property when querying.



Steve

-- 
You received this message because you are subscribed to the Google Groups 
"TopBraid Suite Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to topbraid-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/topbraid-users/CAGUep8515RB8ZCxqx2YBou_dp%2BxyFKE5ZkRqB0Ho5jC-prixCQ%40mail.gmail.com.

Reply via email to