On 2022-01-26 3:13 am, cbur...@healthwise.org wrote:
Holger,
I am able to create a second property shape, ex:shape2, that uses the
values from my ex:shape1 in a very simple, non-transformative way and
successfully returns the values of ex:shape1 for ex:shape2:
ex:Class-shape2
a sh:PropertyShape ;
sh:path ex:shape2 ;
sh:class ex:Class
sh:values [
sh:path ex:shape1 ;
] ;
.
But the moment I try to introduce any more complexity, my sh:values
statement stops returning any results. For example, I tried:
sh:values [
sh:path ( ex:shape1 skos:broader ) ;
]
Right, complex path expressions don't work in that case. This would have
required a much more complex implementation that would hook into the
SPARQL-like path execution. Doable in principle, just not implemented yet.
The alternative to the above is to use nested sh:path expressions, try
something like
sh:values [
sh:path skos:broader ;
sh:nodes [
sh:path ex:shape1 ;
]
]
This is based on the optional second argument (sh:nodes) that
https://w3c.github.io/shacl/shacl-af/#node-expressions-path takes and
would first get all values of ex:shape1 and then get the broader values
of those. I believe this corresponds to your chained path expression.
Hope this works, didn't test... :)
Holger
and got back no values for ex:shape2. (All of the values returned for
ex:shape1 have broader concepts, although they are defined in a
different EDG asset collection.)
What am I doing wrong?
thanks,
Carl
On Thursday, January 20, 2022 at 5:01:18 PM UTC-8 Holger Knublauch wrote:
On 2022-01-21 10:54 am, cbur...@healthwise.org wrote:
Thanks, Holger. My need in this case isn't to query inferred data
using SPARQL or GraphQL, but rather to reuse the inferred data as
an input for another EDG property shape with another sh:values
rule in it. I think I'm hearing that that's not possible.
That scenario IS possible. You just need to use a path node
expression, see
https://w3c.github.io/shacl/shacl-af/#node-expressions-path
So if ex:prop1 is inferred then you can reference it in the
sh:values rule of another property using [ sh:path ex:prop1 ] and
it will compute that inference on the fly. This assumes that the
receiving property using a sh:values rule using SHACL node
expressions, not SPARQL.
Holger
Can you point me to where I can read more about the
materialization option and decide whether there's a way to make
that work for this use case? Thanks.
On Tuesday, January 18, 2022 at 4:03:32 PM UTC-8 Holger Knublauch
wrote:
On 2022-01-19 9:44 am, cbur...@healthwise.org wrote:
I have defined a property shape and am successfully
inferring its objects using a sh:values rule.
Can you remind me whether it is possible to use this
property and its inferred data in the sh:values rule for a
different property shape? If so, is there a special way that
it needs to be done -- like, does the data I inferred in the
first property shape need to be materialized in some way
before I can make use of it? --
In general, the RDF graphs that TopBraid operates on do not
automatically "see" the extra inferred triples, so they are
not computed as RDF triples on the fly. However, our higher
level query languages GraphQL and JavaScript/ADS do see them
whenever they are requested. The user interface goes through
GraphQL and will therefore display these values as if they
were materialized.
Now, within SHACL itself it depends on what you want to do.
If you have SHACL-AF inference rules including other
sh:values rules you can safely use path expressions
https://w3c.github.io/shacl/shacl-af/#node-expressions-path
and the values will be computed on the fly. For SHACL
constraints there is no such mechanism unless you are
expressing the constraints in SPARQL. There you can use the
magic property
(?focusNode ?predicate) tosh:values ?result
to query the inferred values.
The other option, of course, is to materialize the
inferences, e.g. using the Inferences panel or the Transform tab.
Does this help?
Holger
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-user...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/topbraid-users/d1fc11f5-f4a3-4d23-84b7-4a72666e6c3fn%40googlegroups.com
<https://groups.google.com/d/msgid/topbraid-users/d1fc11f5-f4a3-4d23-84b7-4a72666e6c3fn%40googlegroups.com?utm_medium=email&utm_source=footer>.
--
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-user...@googlegroups.com.
To view this discussion on the web visit
https://groups.google.com/d/msgid/topbraid-users/3d423d68-4f62-4180-a44a-3143e6a1128fn%40googlegroups.com
<https://groups.google.com/d/msgid/topbraid-users/3d423d68-4f62-4180-a44a-3143e6a1128fn%40googlegroups.com?utm_medium=email&utm_source=footer>.
--
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/6cd4990a-f08d-4461-b0bf-a15468f8d948n%40googlegroups.com
<https://groups.google.com/d/msgid/topbraid-users/6cd4990a-f08d-4461-b0bf-a15468f8d948n%40googlegroups.com?utm_medium=email&utm_source=footer>.
--
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/38344e6d-c2b1-bf0f-e5df-3f77435cdedc%40topquadrant.com.
