I cannot think of anything in this inferencing area (with SPARQL-like
expressiveness) that is fully standardized or widely available.
Holger
On 11/25/2020 12:56 PM, Steve Ray wrote:
Oh dear. I'm doing this work in support of a standards committee where
I'm the only one using TBC. Others are using either Jena, or raw text
editors and home-rolled reasoners. (!) (I may be persuading one member
to get TBC though).
Of the solutions you are suggesting, which would you recommend that is
the most universal? Does Jena support SPIN (i.e. magic properties?).
I'm assuming tosh isn't available outside TBC/EDG?
Steve
On Tue, Nov 24, 2020 at 6:43 PM Holger Knublauch
<hol...@topquadrant.com <mailto:hol...@topquadrant.com>> wrote:
On 11/25/2020 12:32 PM, Steve Ray wrote:
Wow, fast!
I was just trying exactly that. I also dropped the prefixes since
I wasn't using any. So now I have:
ssh:Class_1
rdf:type owl:Class ;
rdf:type sh:NodeShape ;
rdfs:subClassOf owl:Thing ;
sh:property [
sh:path ssh:Property_1 ;
sh:values [
sh:select """SELECT ?result
WHERE {
BIND (\"Hello\" AS ?result) .
}""" ;
] ;
] ;
.
So far so good.
But executing this query:
*SELECT**
*WHERE*{
?subject a ssh:Class_1 .
?subject ssh:Property_1 ?object .
}
...is coming up empty.
This is because SPARQL only queries the asserted graph. To query
the inferred values, either use GraphQL, Active Data Shapes or use
this magic property:
PREFIX teamwork: <http://topbraid.org/teamwork#>
<http://topbraid.org/teamwork#>
SELECT *
WHERE {
...
(?subject ssh:Property_1 <urn:x-evn-master:geo>) teamwork:values
?object .
}
This takes the master graph that is holding the data in EDG (here:
the Geography ontology). This is the fastest variation for
performance, but this one here is more general:
(?subject ssh:Property_1) tosh:values ?object .
This design has some implications, e.g. you normally cannot query
inferred values in the inverse direction, e.g. if ?object is
given. They are just computed on the fly, for the duration of the
query. This means that the system doesn't need to keep track of
inferences or invalidate them if the data changes - it's always up
to date.
Holger
(Once I get one example working, I'm off to the races!)
Steve
On Tue, Nov 24, 2020 at 6:25 PM Holger Knublauch
<hol...@topquadrant.com <mailto:hol...@topquadrant.com>> wrote:
On 11/25/2020 12:18 PM, Steve Ray wrote:
Sorry to keep using up bandwidth here.
No problem. Admittedly, the official spec at
https://w3c.github.io/shacl/shacl-af/#select
<https://w3c.github.io/shacl/shacl-af/#select> is not exactly
full of examples. A failure of the editor :)
I have pored over all the documentation I can find on the
proper use of sh:values with SPARQL, and clearly I'm not
getting something. Here is my trivial toy example that is
supposed to just infer a hard-coded value of "Hello" for a
property. Can you see what I'm doing wrong?
TBC keeps saying: Property ssh:Property_1: values=(SHACL
node expression of unknown type:
java.lang.IllegalArgumentException: Malformed SHACL node
expression)
I'm also attaching the full file in case you want to just
load it.
Just drop the sh:sparql node and instead do something like
sh:values [
sh:select ...
sh:prefixes ...
]
HTH
Holger
ssh:Class_1
a owl:Class ;
a sh:NodeShape ;
rdfs:subClassOf owl:Thing ;
sh:property [
sh:path ssh:Property_1 ;
sh:values [
sh:sparql [
sh:prefixes <http://example.org/sparqlshape
<http://example.org/sparqlshape>> ;
sh:select """SELECT ?result
WHERE {
BIND (\"Hello\"AS ?result) .
}""" ;
] ;
] ;
] ;
.
ssh:Instance_1
a ssh:Class_1 ;
.
ssh:Property_1
a rdf:Property ;
.
Steve
On Tue, Nov 24, 2020 at 2:57 PM Holger Knublauch
<hol...@topquadrant.com <mailto:hol...@topquadrant.com>> wrote:
On 11/25/2020 8:48 AM, Steve Ray wrote:
I agree with you. Is there no way to have "variable"
predicates in native SHACL? Or is my only option here
to use the embedded SPARQL?
Exactly. SHACL node expressions do not have a notion of
variables. It is intentionally limited to be easier, but
then doesn't offer all features that SPARQL does. There
is nothing wrong with using SPARQL node expressions as a
fallback.
Holger
Steve
On Tue, Nov 24, 2020 at 12:51 PM Irene Polikoff
<ir...@topquadrant.com <mailto:ir...@topquadrant.com>>
wrote:
There is no path in your example that could get
from ?this to ?prop. A path specifies predicates.
You do not have predicates to specify - you are
finding out your predicates in the first WHERE
statement. This is why you need to have 2
statements in the WHERE clause and can’t boil them
down to a path expression in a single statement.
On Nov 24, 2020, at 12:39 PM, Steve Ray
<st...@steveray.com <mailto:st...@steveray.com>>
wrote:
I'm trying to get comfortable with using
SHACL code and wean myself from embedded SPARQL.
However, I'm having some trouble mapping certain
common SPARQL patterns into the SHACL counterpart.
Specifically, here's an example. I'm trying to
find all the values of any property that is a
subPropertyOf a parent property (c223:hasProperty).
The SPARQL should (I think) be something like this:
c223:PropertiesShape
rdf:type sh:PropertyShape ;
sh:path c223:hasProperty ;
sh:name "PropertiesShape" ;
sh:values [
sh:sparql [
sh:prefixes
<http://www.w3.org/2000/01/rdf-schema
<http://www.w3.org/2000/01/rdf-schema>> ;
sh:select """SELECT DISTINCT ?prop
WHERE {
?property rdfs:subPropertyOf* c223:hasProperty .
$this ?property ?prop .
}""" ;
] ;
] ;
.
I know the following is wrong, but not sure what
sh: calls I should be using:
c223:PropertiesShape
rdf:type sh:PropertyShape ;
sh:path c223:hasProperty ;
sh:name "PropertiesShape" ;
sh:values [
sh:distinct [
sh:nodes [
sh:path (
[
sh:zeroOrMorePath rdfs:subPropertyOf ;
]
c223:hasProperty
) ;
] ;
] ;
] ;
.
This gives me a "malformed SHACL expression", so
there are definitely problems.
My real question is, how does one duplicate this
pattern in SHACL?:
WHERE {
?property rdfs:subPropertyOf*
<parentProperty> .
$this ?property ?result .
}
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
<mailto:topbraid-users+unsubscr...@googlegroups.com>.
To view this discussion on the web visit
https://groups.google.com/d/msgid/topbraid-users/CAGUep87nmSVo5eziHPh6ExbPD1%3DD%3DWrfTNOwNtK0PBww21%3DpgA%40mail.gmail.com
<https://groups.google.com/d/msgid/topbraid-users/CAGUep87nmSVo5eziHPh6ExbPD1%3DD%3DWrfTNOwNtK0PBww21%3DpgA%40mail.gmail.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
<mailto:topbraid-users+unsubscr...@googlegroups.com>.
To view this discussion on the web visit
https://groups.google.com/d/msgid/topbraid-users/11EA64E3-07D1-4DAF-A609-21C84A739254%40topquadrant.com
<https://groups.google.com/d/msgid/topbraid-users/11EA64E3-07D1-4DAF-A609-21C84A739254%40topquadrant.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
<mailto:topbraid-users+unsubscr...@googlegroups.com>.
To view this discussion on the web visit
https://groups.google.com/d/msgid/topbraid-users/CAGUep87Hq%3DPv7rCT0f3aQPU%2BzYo27dA1eodYyGe9ZOoh4D9opA%40mail.gmail.com
<https://groups.google.com/d/msgid/topbraid-users/CAGUep87Hq%3DPv7rCT0f3aQPU%2BzYo27dA1eodYyGe9ZOoh4D9opA%40mail.gmail.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
<mailto:topbraid-users+unsubscr...@googlegroups.com>.
To view this discussion on the web visit
https://groups.google.com/d/msgid/topbraid-users/41fa8018-8314-3e9b-b854-7fbc1e70f87a%40topquadrant.com
<https://groups.google.com/d/msgid/topbraid-users/41fa8018-8314-3e9b-b854-7fbc1e70f87a%40topquadrant.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
<mailto:topbraid-users+unsubscr...@googlegroups.com>.
To view this discussion on the web visit
https://groups.google.com/d/msgid/topbraid-users/CAGUep87xZen2zPqno3u%2BVjXKHBCHcx-b8VJeX4cHTnrzP9eE1g%40mail.gmail.com
<https://groups.google.com/d/msgid/topbraid-users/CAGUep87xZen2zPqno3u%2BVjXKHBCHcx-b8VJeX4cHTnrzP9eE1g%40mail.gmail.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
<mailto:topbraid-users+unsubscr...@googlegroups.com>.
To view this discussion on the web visit
https://groups.google.com/d/msgid/topbraid-users/6b37a258-8507-22ce-e8e4-384b893c2dd9%40topquadrant.com
<https://groups.google.com/d/msgid/topbraid-users/6b37a258-8507-22ce-e8e4-384b893c2dd9%40topquadrant.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
<mailto:topbraid-users+unsubscr...@googlegroups.com>.
To view this discussion on the web visit
https://groups.google.com/d/msgid/topbraid-users/CAGUep86jmegagE%3D9W84epw_%3DofMCb6ktcCZa3cVJS7Sq0M27eQ%40mail.gmail.com
<https://groups.google.com/d/msgid/topbraid-users/CAGUep86jmegagE%3D9W84epw_%3DofMCb6ktcCZa3cVJS7Sq0M27eQ%40mail.gmail.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
<mailto:topbraid-users+unsubscr...@googlegroups.com>.
To view this discussion on the web visit
https://groups.google.com/d/msgid/topbraid-users/21a18751-29ba-e10c-4c8e-103ff8aa94df%40topquadrant.com
<https://groups.google.com/d/msgid/topbraid-users/21a18751-29ba-e10c-4c8e-103ff8aa94df%40topquadrant.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
<mailto:topbraid-users+unsubscr...@googlegroups.com>.
To view this discussion on the web visit
https://groups.google.com/d/msgid/topbraid-users/CAGUep8435H%3DwTMOHKm1exe-vKMe47FvGuJ3BvDLaMWst_260Qg%40mail.gmail.com
<https://groups.google.com/d/msgid/topbraid-users/CAGUep8435H%3DwTMOHKm1exe-vKMe47FvGuJ3BvDLaMWst_260Qg%40mail.gmail.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/d35076e7-a0fe-ab6c-9388-6c3c31ad4bff%40topquadrant.com.
