You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@jena.apache.org by Myungjin Lee <xm...@gmail.com> on 2018/07/11 05:27:22 UTC
Create NodeIterator from SPARQL Result
Hi.
I have a question. I am developing small application with Jena and ARQ.
Is it possible to create NodeIterator instance from SPARQL result?
In Jena and ARQ APIs, SPARQL query(qexec.execSelect() method) returns an
instance of ResultSet interface.
But I want to create a method that returns NodeIterator instance after
executing SPARQL query like below.
public NodeIterator getObjects(Resource s, Property p) {
String queryString = "SELECT ?o WHERE { <" + s.getURI() + "> <" +
p.getURI() + "> ?o . }";
Query query = QueryFactory.create(queryString);
QueryExecution qexec =
QueryExecutionFactory.sparqlService("SPARQL_ENDPOINT_ADDRESS", query);
ResultSet results = qexec.execSelect();
NodeIterator ni = ??? // to create NodeIterator instance from
ResultSet
return ni;
}
Is it possible?
Thank you for your help.
Best regards,
Myungjin Lee
Re: Create NodeIterator from SPARQL Result
Posted by Myung Jin Lee <xm...@gmail.com>.
I will check it.
Thanks for your help.
2018-07-11 20:43 GMT+09:00 Andy Seaborne <an...@apache.org>:
> On 11/07/18 10:00, Myungjin Lee wrote:
>
>> Thank you for your help.
>>
>> As you know, there are several triple stores such as Virtuoso,
>> AllegroGraph,
>> and so forth.
>> So I want to develop Jena base interfaces for different triple stores.
>> I know that some triple stores already support Jena, but some triple
>> stores
>> that are not famous only support SPARQL endpoint.
>>
>
> Have you looked at Jena's RDFConnection?
>
>
>
>> For these triple stores, I need some classes that implements Jena
>> interfaces.
>> It is one of classes and methods to return NodeIterator from SPARQL
>> result.
>>
>> I thought there are, maybe, a simple way to create NodeIterator from
>> ResultSet.
>> But your code is a good solution for me.
>>
>> Thank you.
>>
>>
>>
>> -----Original Message-----
>> From: Lorenz Buehmann <bu...@informatik.uni-leipzig.de>
>> Sent: Wednesday, July 11, 2018 3:25 PM
>> To: users@jena.apache.org
>> Subject: Re: Create NodeIterator from SPARQL Result
>>
>> A NodeIterator for what? A ResultSet wraps an iterator of QuerySolution
>> objects, each of which contains the mapping of a variable name to the
>> corresponding RDF resource (if bound). Thus, each QuerySolution can wrap a
>> set of nodes...
>>
>> Clearly, you could generate a wrapping iterator by yourself. Just
>> implement
>> the NodeIterator interface and the corresponding methods. The constructor
>> indeed should take the ResultSet and a variable name as arguments (if you
>> want all nodes for each QuerySolution, it'S also
>> possible) - the rest is straightforward Java coding.
>>
>>
>> Untested stub:
>>
>>
>> class ResultSetWrappingNodeIterator implements NodeIterator {
>>
>> final ResultSet rs;
>>
>> final String varName;
>>
>> ResultSetWrappingNodeIterator(ResultSet rs, String varName) {
>>
>> this.rs = rs;
>>
>> this.varName = varName;
>>
>> }
>>
>>
>> public RDFNode next() {
>>
>> return rs.next().get(varName).asNode();
>>
>> }
>>
>>
>> ...
>>
>> }
>>
>>
>>
>> On 11.07.2018 07:27, Myungjin Lee wrote:
>>
>>> Hi.
>>>
>>>
>>> I have a question. I am developing small application with Jena and ARQ.
>>>
>>>
>>> Is it possible to create NodeIterator instance from SPARQL result?
>>>
>>> In Jena and ARQ APIs, SPARQL query(qexec.execSelect() method) returns
>>> an instance of ResultSet interface.
>>>
>>>
>>> But I want to create a method that returns NodeIterator instance after
>>> executing SPARQL query like below.
>>>
>>>
>>> public NodeIterator getObjects(Resource s, Property p) {
>>>
>>> String queryString = "SELECT ?o WHERE { <" + s.getURI() +
>>> "> <" +
>>> p.getURI() + "> ?o . }";
>>>
>>> Query query = QueryFactory.create(queryString);
>>>
>>> QueryExecution qexec =
>>> QueryExecutionFactory.sparqlService("SPARQL_ENDPOINT_ADDRESS", query);
>>>
>>> ResultSet results = qexec.execSelect();
>>>
>>> NodeIterator ni = ??? // to create NodeIterator instance
>>> from ResultSet
>>>
>>> return ni;
>>>
>>> }
>>>
>>>
>>> Is it possible?
>>>
>>> Thank you for your help.
>>>
>>>
>>> Best regards,
>>>
>>> Myungjin Lee
>>>
>>>
>>>
>>
>>
>>
Re: Create NodeIterator from SPARQL Result
Posted by Andy Seaborne <an...@apache.org>.
On 11/07/18 10:00, Myungjin Lee wrote:
> Thank you for your help.
>
> As you know, there are several triple stores such as Virtuoso, AllegroGraph,
> and so forth.
> So I want to develop Jena base interfaces for different triple stores.
> I know that some triple stores already support Jena, but some triple stores
> that are not famous only support SPARQL endpoint.
Have you looked at Jena's RDFConnection?
>
> For these triple stores, I need some classes that implements Jena
> interfaces.
> It is one of classes and methods to return NodeIterator from SPARQL result.
>
> I thought there are, maybe, a simple way to create NodeIterator from
> ResultSet.
> But your code is a good solution for me.
>
> Thank you.
>
>
>
> -----Original Message-----
> From: Lorenz Buehmann <bu...@informatik.uni-leipzig.de>
> Sent: Wednesday, July 11, 2018 3:25 PM
> To: users@jena.apache.org
> Subject: Re: Create NodeIterator from SPARQL Result
>
> A NodeIterator for what? A ResultSet wraps an iterator of QuerySolution
> objects, each of which contains the mapping of a variable name to the
> corresponding RDF resource (if bound). Thus, each QuerySolution can wrap a
> set of nodes...
>
> Clearly, you could generate a wrapping iterator by yourself. Just implement
> the NodeIterator interface and the corresponding methods. The constructor
> indeed should take the ResultSet and a variable name as arguments (if you
> want all nodes for each QuerySolution, it'S also
> possible) - the rest is straightforward Java coding.
>
>
> Untested stub:
>
>
> class ResultSetWrappingNodeIterator implements NodeIterator {
>
> final ResultSet rs;
>
> final String varName;
>
> ResultSetWrappingNodeIterator(ResultSet rs, String varName) {
>
> this.rs = rs;
>
> this.varName = varName;
>
> }
>
>
> public RDFNode next() {
>
> return rs.next().get(varName).asNode();
>
> }
>
>
> ...
>
> }
>
>
>
> On 11.07.2018 07:27, Myungjin Lee wrote:
>> Hi.
>>
>>
>>
>> I have a question. I am developing small application with Jena and ARQ.
>>
>>
>>
>> Is it possible to create NodeIterator instance from SPARQL result?
>>
>> In Jena and ARQ APIs, SPARQL query(qexec.execSelect() method) returns
>> an instance of ResultSet interface.
>>
>>
>>
>> But I want to create a method that returns NodeIterator instance after
>> executing SPARQL query like below.
>>
>>
>>
>> public NodeIterator getObjects(Resource s, Property p) {
>>
>> String queryString = "SELECT ?o WHERE { <" + s.getURI() +
>> "> <" +
>> p.getURI() + "> ?o . }";
>>
>> Query query = QueryFactory.create(queryString);
>>
>> QueryExecution qexec =
>> QueryExecutionFactory.sparqlService("SPARQL_ENDPOINT_ADDRESS", query);
>>
>> ResultSet results = qexec.execSelect();
>>
>> NodeIterator ni = ??? // to create NodeIterator instance
>> from ResultSet
>>
>> return ni;
>>
>> }
>>
>>
>>
>> Is it possible?
>>
>> Thank you for your help.
>>
>>
>>
>> Best regards,
>>
>> Myungjin Lee
>>
>>
>
>
>
RE: Create NodeIterator from SPARQL Result
Posted by Myungjin Lee <xm...@gmail.com>.
Thank you for your help.
As you know, there are several triple stores such as Virtuoso, AllegroGraph,
and so forth.
So I want to develop Jena base interfaces for different triple stores.
I know that some triple stores already support Jena, but some triple stores
that are not famous only support SPARQL endpoint.
For these triple stores, I need some classes that implements Jena
interfaces.
It is one of classes and methods to return NodeIterator from SPARQL result.
I thought there are, maybe, a simple way to create NodeIterator from
ResultSet.
But your code is a good solution for me.
Thank you.
-----Original Message-----
From: Lorenz Buehmann <bu...@informatik.uni-leipzig.de>
Sent: Wednesday, July 11, 2018 3:25 PM
To: users@jena.apache.org
Subject: Re: Create NodeIterator from SPARQL Result
A NodeIterator for what? A ResultSet wraps an iterator of QuerySolution
objects, each of which contains the mapping of a variable name to the
corresponding RDF resource (if bound). Thus, each QuerySolution can wrap a
set of nodes...
Clearly, you could generate a wrapping iterator by yourself. Just implement
the NodeIterator interface and the corresponding methods. The constructor
indeed should take the ResultSet and a variable name as arguments (if you
want all nodes for each QuerySolution, it'S also
possible) - the rest is straightforward Java coding.
Untested stub:
class ResultSetWrappingNodeIterator implements NodeIterator {
final ResultSet rs;
final String varName;
ResultSetWrappingNodeIterator(ResultSet rs, String varName) {
this.rs = rs;
this.varName = varName;
}
public RDFNode next() {
return rs.next().get(varName).asNode();
}
...
}
On 11.07.2018 07:27, Myungjin Lee wrote:
> Hi.
>
>
>
> I have a question. I am developing small application with Jena and ARQ.
>
>
>
> Is it possible to create NodeIterator instance from SPARQL result?
>
> In Jena and ARQ APIs, SPARQL query(qexec.execSelect() method) returns
> an instance of ResultSet interface.
>
>
>
> But I want to create a method that returns NodeIterator instance after
> executing SPARQL query like below.
>
>
>
> public NodeIterator getObjects(Resource s, Property p) {
>
> String queryString = "SELECT ?o WHERE { <" + s.getURI() +
> "> <" +
> p.getURI() + "> ?o . }";
>
> Query query = QueryFactory.create(queryString);
>
> QueryExecution qexec =
> QueryExecutionFactory.sparqlService("SPARQL_ENDPOINT_ADDRESS", query);
>
> ResultSet results = qexec.execSelect();
>
> NodeIterator ni = ??? // to create NodeIterator instance
> from ResultSet
>
> return ni;
>
> }
>
>
>
> Is it possible?
>
> Thank you for your help.
>
>
>
> Best regards,
>
> Myungjin Lee
>
>
Re: Create NodeIterator from SPARQL Result
Posted by Lorenz Buehmann <bu...@informatik.uni-leipzig.de>.
A NodeIterator for what? A ResultSet wraps an iterator of QuerySolution
objects, each of which contains the mapping of a variable name to the
corresponding RDF resource (if bound). Thus, each QuerySolution can wrap
a set of nodes...
Clearly, you could generate a wrapping iterator by yourself. Just
implement the NodeIterator interface and the corresponding methods. The
constructor indeed should take the ResultSet and a variable name as
arguments (if you want all nodes for each QuerySolution, it'S also
possible) - the rest is straightforward Java coding.
Untested stub:
class ResultSetWrappingNodeIterator implements NodeIterator {
final ResultSet rs;
final String varName;
ResultSetWrappingNodeIterator(ResultSet rs, String varName) {
this.rs = rs;
this.varName = varName;
}
public RDFNode next() {
return rs.next().get(varName).asNode();
}
...
}
On 11.07.2018 07:27, Myungjin Lee wrote:
> Hi.
>
>
>
> I have a question. I am developing small application with Jena and ARQ.
>
>
>
> Is it possible to create NodeIterator instance from SPARQL result?
>
> In Jena and ARQ APIs, SPARQL query(qexec.execSelect() method) returns an
> instance of ResultSet interface.
>
>
>
> But I want to create a method that returns NodeIterator instance after
> executing SPARQL query like below.
>
>
>
> public NodeIterator getObjects(Resource s, Property p) {
>
> String queryString = "SELECT ?o WHERE { <" + s.getURI() + "> <" +
> p.getURI() + "> ?o . }";
>
> Query query = QueryFactory.create(queryString);
>
> QueryExecution qexec =
> QueryExecutionFactory.sparqlService("SPARQL_ENDPOINT_ADDRESS", query);
>
> ResultSet results = qexec.execSelect();
>
> NodeIterator ni = ??? // to create NodeIterator instance from
> ResultSet
>
> return ni;
>
> }
>
>
>
> Is it possible?
>
> Thank you for your help.
>
>
>
> Best regards,
>
> Myungjin Lee
>
>