Generated Void Description display 0 triples on populated graph in GraphDB

[aibba19]

Hello,

I found this repository while searching for methods to extract metadata from my graph stored in GraphDB (currently on my local PC), and it seems very helpful for my use case.

I attempted to run the following bash commands as described in the repository:

java -jar /content/void-generator-0.6-uber.jar -r http://myIP:7200/repositories/Barcelona2 \
   -p http://myIP:7200/repositories/Barcelona2 \
   --void-file void-barcelona2.ttl \
   --iri-of-void 'http://myIP:7200/repositories/Barcelona2/' \
   -g http://myIP:7200/repositories/Barcelona2 

This resulted in the following output file:

@prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .
@prefix void: <http://rdfs.org/ns/void#> .
@prefix : <http://www.w3.org/ns/sparql-service-description#> .
@prefix void_ext: <http://ldf.fi/void-ext#> .
@prefix formats: <http://www.w3.org/ns/formats/> .
@prefix pav: <http://purl.org/pav/> .
@prefix xsd: <http://www.w3.org/2001/XMLSchema#> .
@prefix dcterms: <http://purl.org/dc/terms/> .

<http://myIP:7200/repositories/Barcelona2> a :Service, :NamedGraph;
  :defaultDataset <http://myIP:7200/repositories/Barcelona2#sparql-default-dataset>;
  :endpoint <http://myIP:7200/repositories/Barcelona2>;
  :resultFormat formats:SPARQL_Results_CSV, formats:SPARQL_Results_JSON, formats:N-Triples,
    formats:RDF_XML, formats:SPARQL_Results_TSV, formats:Turtle, formats:SPARQL_Results_XML;
  :supportedLanguage :SPARQL11Query;
  :feature :UnionDefaultGraph, :BasicFederatedQuery;
  :availableGraphs _:cb2e5e19164a411cb17ed8dde9d32bf911;
  :name <http://myIP:7200/repositories/Barcelona2>;
  :graph <http://myIP:7200/repositories/Barcelona2/void#_graph_Barcelona2!e1011618> .

<http://myIP:7200/repositories/Barcelona2#sparql-default-dataset> a :Dataset;
  :defaultGraph <http://myIP:7200/repositories/Barcelona2/void#sparql-default-graph>;
  :namedGraph <http://myIP:7200/repositories/Barcelona2> .

<http://myIP:7200/repositories/Barcelona2/void#sparql-default-graph> a :Graph;
  void:distinctObjects "15706"^^xsd:long;
  void_ext:distinctLiterals "9559"^^xsd:long;
  void_ext:distinctIRIReferenceObjects "6147"^^xsd:long;
  void:distinctSubjects "6132"^^xsd:long;
  void_ext:distinctIRIReferenceSubjects "6132"^^xsd:long .

_:cb2e5e19164a411cb17ed8dde9d32bf911 :namedGraph <http://myIP:7200/repositories/Barcelona2> .

<http://myIP:7200/repositories/Barcelona2/void#_graph_Barcelona2!e1011618>
  a :Graph;
 ** void:triples "0"^^xsd:long ** .

Since the repository is populated with many triples, I expected to see a different output instead of void:triples "0".

What am I doing wrong here? Any help or advice would be greatly appreciated.

[galgonek]

Hello,

How many triplets do you have in graph http://myIP:7200/repositories/Barcelona2? What result will the following SPARQL query return?

SELECT (COUNT(*) AS ?COUNT) WHERE {
  GRAPH <http://myIP:7200/repositories/Barcelona2> {
    ?S ?P ?O
  }
}

[aibba19]

The result of the query is 0.
While searching for a solution, I found that this issue might be because I am using the default graph in my GraphDB repository.

Below, you can see a screenshot of the graph overview from my GraphDB UI.

Should I specify the graph name in any of the terminal command options? If yes, in which commands and how?

[aibba19]

After several attempts, I solved the problem with the following steps:

I created a named graph (e.g., http://barcelona_with_geometry.com) in my GraphDB.
I modified the command with the following options :
!java -jar /content/void-generator-0.6-uber.jar -r http://151.56.105.209:7200/repositories/Barcelona2 \ -p http://151.56.105.209:7200/repositories/Barcelona2 \ --void-file void-barcelona2.ttl \ --iri-of-void 'http://151.56.105.209:7200/repositories/Barcelona2/' \ -g http://barcelona_with_geometry.com

The result is a file containing some information about my graph. However, I still have a question: Do I need to have the ontologies used to define the classes and properties in my graph also stored in GraphDB? My graph uses three different ontologies in addition to the OWL ontology.

[galgonek]

Should I specify the graph name in any of the terminal command options? If yes, in which commands and how?

The -g parameter is optional. If you do not use it, statistics will be calculated for the default graph and all named graphs.

Do I need to have the ontologies used to define the classes and properties in my graph also stored in GraphDB?

As far as I know, the current version of the VoID generator does not need the ontologies to be stored in the database.