extracting indivdiual bibo:Articles from RDF document

Question

I have a RDF/XML document with this format:

<?xml version="1.0" encoding="UTF-8"?>
<rdf:RDF xmlns:ags="http://purl.org/agmes/1.1/" xmlns:foaf="http://xmlns.com/foaf/0.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#" xmlns:bibo="http://purl.org/ontology/bibo/" xmlns:dct="http://purl.org/dc/terms/">
  <bibo:Article rdf:about="http://xxxxx/NO8500391">
    <dct:identifier>NO8500391</dct:identifier>
    ...
  </bibo:Article>
  <bibo:Article rdf:about="http://xxxxx/NO8500523">
    ...
  </bibo:Article>
  <bibo:Article rdf:about="http://xxxxx/NO8500496">
  ...
  </bibo:Article>
</rdf:RDF>

As you can see, in a single RDF/XML file, there are many bibo:Articles, could be thousands. What I want is to extract each article and convert it to RDF/JSON (I know how to write a model) using Apache Jena, so I can have a separate document for each article, and later import them all to a index like CouchDB or Elasticsearch to perform searches.

How can I extract each article in the model (Jena)? The dirty way that I was thinking is to process the file as XML and extract each bibo:Article element.

Answer 1

Firstly can I ask for some clarfication on your question? I think what you are asking is to split each bibo:Article entry into its own document right?

As an aside note that this is not the same as splitting each first level node because RDF/XML is not a canonical serialization i.e. the same RDF may be serialized by multiple different RDF/XML documents and there is no guarantee that they will always be the first level nodes.

Now to try and answer your question, there are two main ways to achieve your aim.

Using SPARQL Queries

Firstly issue a SELECT query to retrieve all articles instances, then for each result issue a DESCRIBE query on the article URI which will give you a new Jena Model containing only information about that URI.

Note that you can change exactly how DESCRIBE queries by creating a custom DescribeHandler if you wish but that may be overkill.

You can then serialize the results of each DESCRIBE query to a new document.

Using the Model API

Use the listStatements() method (the overload that takes search criteria) to first find the articles, then similar to the first solution issue further calls for each discovered article URI to find statements about it. These can be added to a new model and serialized out as desired.