Its not possible to update or insert data via SPARQL endpoint

[BjBe82]

Just tried to insert data in a named graph using the SPARQL endpoint:

PREFIX dtsc: <tag:esa.int,2025:dtsc:managment:vocabulary#>
INSERT DATA {
	  _:newProject a dtsc:Project ;
				   dtsc:name "name" ;
				   dtsc:description "description" .
	
}

This fails with the following error:

---

Checked this with Virtuoso where the same Query works as expected.

-> ->

[kenwenzel]

The frontend does currently not support SPARQL Update but the HTTP endpoint should do so.

[BjBe82]

This was the initial issue I had in Bruno/Java code

[kenwenzel]

The HTTP parameter is called update for SPARQL Update queries. Please check the following documentation:
https://www.w3.org/TR/2013/REC-sparql11-protocol-20130321/#update-operation

[BjBe82]

Thanks for the hint, this works.

[kenwenzel]

I'll plan to rework the frontend with a better editor. This will then also work using the UI.

[kenwenzel]

@BjBe82 Maybe you also just want to use the GrahphStore API to directly upload TTL or JSON-LD files.

[BjBe82]

@kenwenzel I already use the GrahphStore API to upload complete graphs. The problem is if i want to update small subsets of the graphs like name changes, relationship added/removed.

Maybe as a side question, are you intending to adapt the "model" (should be "using-named-graph-uri" based on the spec) parameter to the SPARQL spec so APIs like this: https://rdf4j.org/documentation/programming/repository/ can be used out of the box?

[kenwenzel]

@BjBe82 You mean renaming model to graph?
The RDF4J API can do more than SPARQL+GraphStore, it supports transactions, for example.

[BjBe82]

Just checked if i can use SPARQLRepository as mentioned here: https://rdf4j.org/documentation/programming/setup/ => "Accessing a SPARQL endpoint" but the Request will strictly follow the SPARQL spec as you mentioned. The Query String Parameters are called
"using-named-graph-uri" for update or "named-graph-uri" for query instead of "model". Which means SPARQLRepository will send a "named-graph-uri" parameter with specified Dataset URI.

But this is getting a little bit off-topic, my initial issue is solved, and the issue can be closed. Thanks for the support.

[kenwenzel]

@BjBe82 Could you open a specific issue for supporting the SparqlRepository? I think it's a good idea.

[kenwenzel]

@BjBe82 You should be able to use the SparqlRepository if you encode the model directly in the endpoint URL Afterwards you treat all triples as If they we're contained in the default (unnamed) graph.