List domains that contain a string - ontology

Question

first off all I´m new to Jena. I´ve created an ontology and i´ve 4 classes, procedure, buyer, supplier and Zip Code.And i´ve the following properties:
Procedure hasBuyer Buyers
Procedure hasSupplier Supplier
Supplier hasZipCode Zip Code
Buyer hasZipCode Zip Code
What i want to know it´s the best approach in Jena to return all domains that contain a string "3333".
For example:
Procedure 1 have buyer 1 with zip code 333, supplier 1 with zip code 333 and supplier2 with zip code 334.
Procedure 2 have buyer 2 with zip code 331, supplier 2 with zip code 334 and supplier2 with zip code 335.
Procedure 3 have buyer 3 with zip code 333, supplier 1 with zip code 333 and supplier3 with zip code 335.
The results must be:
Procedures - Procedure1 and Procedure3
Buyer - Buyer1 and Buyer 3
Supplier - supplier1
NG

Answer 1

First of all, you can't use "supplier 2 with zip code 334", and then "supplier 2 with zip code 335", because it's the same individual, and you'll see "supplier 2 with zip code 334 and zip code 334" in application both times.
There is some variants of realisation.
With plain Jena API:
Model model; // your model
Resource supplierClass = model.getResource(YOUR_NS + "Supplier");
Resource buyerClass = model.getResource(YOUR_NS + "Buyer");
Resource procClass = model.getResource(YOUR_NS + "Procedure");
Property zipCodeProp = model.getProperty(YOUR_NS + "zipCode");
Property hasBuyerProp = model.getProperty(YOUR_NS + "hasBuyer");
Property hasSupplierProp = model.getProperty(YOUR_NS + "hasSupplier");
StmtIterator iter =
model.listStatements(new SimpleSelector(null, zipCodeProp, "333"));
while (iter.hasNext()) {
Resource subject = iter.next().getSubject();
if (!subject.hasProperty(RDF.type))
continue;
Resource subjectClass = subject.getPropertyResourceValue(RDF.type);
SimpleSelector sel;
if (subjectClass.equals(supplierClass))
sel = new SimpleSelector(null, hasSupplierProp, subject);
else if (subjectClass.equals(buyerClass))
sel = new SimpleSelector(null, hasBuyerProp, subject);
else
continue;
StmtIterator innerIter = model.listStatements(sel);
while (innerIter.hasNext()) {
Resource proc = innerIter.next().getSubject();
if (!proc.hasProperty(RDF.type) ||
!proc.getPropertyResourceValue(RDF.type).equals(procClass))
continue;
// now you can retrieve linked entities from this procedure
}
}
And SPARQL query:
PREFIX yourns: <YOUR_NS>
SELECT DISTINCT ?proc
{
?proc a yourns:Procedure;
yourns:hasBuyer ?buyer;
yourns:hasSupplier ?supplier.
?supplier zipCode ?supplierZip.
?buyer zipCode ?buyerZip.
FILTER (?supplierZip = '333' || ?buyerZip = '333')
}
with further use of ARQ:
Query query = QueryFactory.create(queryString);
QueryExecution qe = QueryExecutionFactory.create(query, model);
ResultSet results = qe.execSelect();
while (results.hasNext()) {
QuerySolution qs = results.next();
Resource proc = qs.getResource("proc");
// now again you can retrieve linked entities
}