You can also get the current prefix mapping from Fuseki if you do this
query:

CONSTRUCT WHERE { }

You can also update the prefix mapping with

LOAD <file:///my/prefixes.ttl>

Cheers,


On Mon, 2024-08-26 at 10:49 +0200, christophe heligon wrote:
> Dear Andy,
> thank you for this explanation.
> Christophe
> 
> ----- Mail original -----
> De: "Andy Seaborne" <[email protected]>
> À: "users" <[email protected]>
> Envoyé: Vendredi 23 Août 2024 12:46:56
> Objet: Re: jena vs jena-fuseki: different behaviour for namespace
> replacement by URI in query answer
> 
> On 22/08/2024 13:24, christophe heligon wrote:
> > Dear community,
> > 
> > Jena and Jena-fuseki 4.9/4.10/5.1 do not display the results
> > (namespace/ 
> > prefix vs uri) similarly in my hands and I wonder what I did wrong.
> 
> Nothing!
> 
> SPARQL SELECT results (any format) do not include prefixes and the
> 3rd 
> party component (the YASR part of YASGUI) using CodeMirror doesn't
> have 
> a way of getting them.
> 
> Prefixes are only surface syntax and not part of RDF conceptual data 
> model. Each dataset records prefixes.
> 
> We recently introduced a "prefixes API" so applications can get and 
> update the prefixes for a dataset.
> 
> https://jena.apache.org/documentation/fuseki2/prefixes-service
> 
> but of course that's a Jena feature, not a standard.
> 
> It would be good if YASR could be prefixes sensitive. It may be 
> something the Jena UI javascript can do as YASR is supposed to be 
> modular and flexible.
> 
>      Andy
> 
> > I use the applications via downloading from jena website or through
> > the 
> > stain/jena docker images.
> > I load the same data set in both context(tdb2.tdbloader --loc DB
> > $files 
> > or upload manually in fuseki GUI).
> > I send the same query to both either through ./bin/tdb2.tdbquery --
> > loc 
> > DB --file request.sq or through fuseki GUI.
> > The replies are different as fuseki will replace all prefixes by
> > URIs 
> > and jena will replace some only and I did not see why some and not 
> > others. See below.
> > 
> > Can someone explain this behaviour to me?
> > 
> > Best regards,
> > Christophe
> 

Reply via email to