This is quite interesting, thank you!
1) Yes - but combined with other workarounds, such as pushing the property
shapes up the hierarchy, or querying multiple subclasses in parallel,
provides options
2) Yes - it changes the structure with respect to a common interface/union
class, which we don't have anyway (!)
Davide
Out of curiosity: if there was a way for users to declare unions or
interfaces via a triple in the graphql: namespace, so that the
union/interface declaration could be
added to the GraphQL schema, would the GraphQL engine in EDG work out of
the box?
On Tuesday, November 28, 2023 at 1:07:33 PM UTC-6 Holger Knublauch wrote:
> One more thought: it seems like you want to dynamically cast from one type
> to another. You can theoretically do that through a virtual property that
> points at the instance itself but with a different declared sh:class. Add
> these triples to your ontology;
>
> hst:AAA
> sh:property hst:AAA-asBBB .
>
> hst:AAA-asBBB
> a sh:PropertyShape ;
> sh:path hst:asBBB ;
> dash:hidden true ;
> sh:class hst:BBB ;
> sh:maxCount 1 ;
> sh:name "as BBB" ;
> sh:values [
> sh:filterShape [
> sh:class hst:BBB ;
> ] ;
> sh:nodes sh:this ;
> ] ;
> .
>
> This computes an inferred value that has as values the focus node itself
> (sh:this), but declared as BBB via sh:class. It uses a filterShape to
> return null if the focus node is NOT an instance of BBB.
>
> You can then formulate queries such as
>
> {
> aaas {
> uri
> asBBB {
> propB
> }
> }
> }
>
> with an example output such as
>
> {
> "data": {
> "aaas": [
> {
> "uri": "
> http://ontology.mayo.edu/data-graphs/HierarchicalDataGraphTest#a1";,
> "asBBB": null
> },
> {
> "uri": "
> http://ontology.mayo.edu/data-graphs/HierarchicalDataGraphTest#d1";,
> "asBBB": null
> },
> {
> "uri": "
> http://ontology.mayo.edu/data-graphs/HierarchicalDataGraphTest#b1";,
> "asBBB": {
> "propB": "valueB"
> }
> },
> {
> "uri": "
> http://ontology.mayo.edu/data-graphs/HierarchicalDataGraphTest#c1";,
> "asBBB": {
> "propB": "valueBC"
> }
> }
> ]
> }
> }
>
> Note this approach already works but is a compromise:
>
> 1) It requires changes to the ontology for all cases where you want to
> typecast
> 2) It probably returns a different nesting in the result JSON (not sure)?
>
> Does this help?
>
> Holger
>
>
>
> On 28 Nov 2023, at 5:01 pm, Holger Knublauch <[email protected]>
> wrote:
>
>
>
> On 28 Nov 2023, at 3:08 pm, Davide Sottara <[email protected]> wrote:
>
> I have put together a simple ontology with a class hierarchy, a property
> on each class, and one individual per class - attached
> AAA
> DDD
> BBB
> CCC
>
>
> Thanks for the file, which I was able to play with. I also tried something
> like
>
> {
> aaas {
> uri
> ... on BBB {
> propA
> }
> }
> }
>
> but since the schema doesn't know the relationships between AAA and BBB
> this doesn't work. All types in the generated schemas are independent, no
> implements or such.
>
>
> I was looking for a behavior that replicates unions or interfaces:
> https://www.apollographql.com/docs/apollo-server/schema/unions-interfaces/
>
> In fact, have tried setting graphql:isInterface = true, but that does not
> seem to have any effect?
>
>
> graphql:isInterface is only used for round-tripping GraphQL Schema files
> from elsewhere, not those that are auto-generated.
>
> So I am afraid this syntax isn't working and you may need to revert to
> other technologies like SPARQL or ADS JavaScript to formulate such queries.
>
> Holger
>
>
>
> Thank you for any insight\
> Davide
>
> On Wednesday, November 22, 2023 at 1:44:58 AM UTC-6 Holger Knublauch wrote:
>
>> Hi Davide,
>>
>> thanks for the error message. I had to read up on GraphQL fragments as I
>> have never used them.
>> It seems that they would need to be declared using the fragment keyword
>> but I don't see that in your query.
>> Our GraphQL engine does not declare such fragments automatically.
>>
>> It would streamline this discussion if we had a complete, self-contained
>> example.
>>
>> Thanks
>> Holger
>>
>>
>> On 22 Nov 2023, at 3:59 am, Davide Sottara <[email protected]> wrote:
>>
>> Unfortunately I get an error
>> "Fragment cannot be spread here as objects of type (super) can never be
>> of type (sub)
>>
>> On Tuesday, November 21, 2023 at 2:55:51 PM UTC-6 Holger Knublauch wrote:
>>
>>> Have you tried it out? What happened?
>>>
>>> Holger
>>>
>>>
>>> On 21 Nov 2023, at 5:35 am, Davide Sottara <[email protected]> wrote:
>>>
>>> I have a class hierarchy, where the classes are also node shapes with
>>> property shapes
>>>
>>> class AA
>>> propA :String
>>>
>>> class BB subClassOf AA
>>> propB : String
>>>
>>> Is a query like the following supported?
>>>
>>> aaas {
>>> propA
>>> ... on bbb {
>>> propB
>>> }
>>> }
>>>
>>> Thank you
>>>
>>> --
>>> The topics of this mailing list include TopBraid EDG and related
>>> technologies such as SHACL.
>>> To post to this group, send email to [email protected]
>>> ---
>>> 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 [email protected].
>>> To view this discussion on the web visit
>>> https://groups.google.com/d/msgid/topbraid-users/efa87ebe-4872-42f7-aaa7-27fc8dcadbb1n%40googlegroups.com
>>>
>>> <https://groups.google.com/d/msgid/topbraid-users/efa87ebe-4872-42f7-aaa7-27fc8dcadbb1n%40googlegroups.com?utm_medium=email&utm_source=footer>
>>> .
>>>
>>>
>>>
>> --
>> The topics of this mailing list include TopBraid EDG and related
>> technologies such as SHACL.
>> To post to this group, send email to [email protected]
>> ---
>> 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 [email protected].
>>
>> To view this discussion on the web visit
>> https://groups.google.com/d/msgid/topbraid-users/746c10ff-c6ef-4fe3-9c4c-0ef1fa714fdcn%40googlegroups.com
>>
>> <https://groups.google.com/d/msgid/topbraid-users/746c10ff-c6ef-4fe3-9c4c-0ef1fa714fdcn%40googlegroups.com?utm_medium=email&utm_source=footer>
>> .
>>
>>
>>
> --
> The topics of this mailing list include TopBraid EDG and related
> technologies such as SHACL.
> To post to this group, send email to [email protected]
> ---
> 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 [email protected].
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/topbraid-users/eb1d3a18-c776-4924-b365-8f0de0e16cc0n%40googlegroups.com
>
> <https://groups.google.com/d/msgid/topbraid-users/eb1d3a18-c776-4924-b365-8f0de0e16cc0n%40googlegroups.com?utm_medium=email&utm_source=footer>
> .
> <HierarchicalDataGraphTest.trig>
>
>
>
>
--
The topics of this mailing list include TopBraid EDG and related technologies
such as SHACL.
To post to this group, send email to [email protected]
---
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 [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/topbraid-users/25252b80-746d-42cf-be0b-cf2c930fdd29n%40googlegroups.com.
