使用 Jena 解析 RDF 递归

我正在尝试使用 Apache Jena 递归解析 RDF 文档。它由如下数据集组成：

<dcat:dataset>
    <dcat:Dataset rdf:about="http://url/" > 
        <dct:description xml:lang="ca">Description</dct:description>
        <dct:license rdf:resource="http://creativecommons.org/licenses/by/3.0/"/>
        <dcat:keyword xml:lang="ca">Keyword1</dcat:keyword>
        <dcat:distribution>
            <dcat:Download>
                <dcat:accessURL>http:/url/</dcat:accessURL>
                <dct:format>
                    <dct:IMT>
                        <rdf:value>application/pdf</rdf:value>
                        <rdfs:label>pdf</rdfs:label>
                    </dct:IMT>
                </dct:format>
                <dct:modified rdf:datatype="http://www.w3.or/2001/XMLSchema#date">2012-11-09T16:23:22</dct:modified>
           </dcat:Download>
        </dcat:distribution>
        <dct:publisher>
           <foaf:Organization>
              <dct:title xml:lang="en">Company</dct:title>
              <foaf:homepage rdf:resource="http://url/"/>
           </foaf:Organization>
        </dct:publisher>
    </dcat:Dataset>
</dcat:dataset>

到目前为止，我已经得到了位于 dcat:Dataset 正下方的每一条语句（使用 Jena 迭代 RDF 文件中的特定资源 https://stackoverflow.com/questions/16939159/iterate-over-specific-resource-in-rdf-file-with-jena），但我想找到每个级别中的每个三元组。我的输出应该如下所示：

description: Description
license: http://creativecommons.org/licenses/by/3.0/
keyword: Keyword1
distribution -> Download -> accessurl: http:/url/
distribution -> Download -> format -> IMT -> value: application/pdf
distribution -> Download -> format -> IMT -> label: pdf
...

我已经用递归函数尝试过，该函数迭代语句，当语句不是文字时，它会跟随对象到达下一个节点。像这样：

private String recursiveQuery(Statement stmt) {
    Resource subject = stmt.getSubject();
    Property predicate = stmt.getPredicate();
    RDFNode object = stmt.getObject();

    if(object.isLiteral()) {
        out.println("LIT: " + predicate.getLocalName());
        return object.toString();

    } else {
        out.println(predicate.getLocalName());
        Resource r = stmt.getResource();
        StmtIterator stmts = r.listProperties();
        while (stmts.hasNext()) {
            Statement s = stmts.next();
            out.println(s.getPredicate().getLocalName());
            return recursiveQuery(s);
        }
    }
    return null;

}

但不知怎的，我用这种方法毫无进展。 非常感谢您的每一个见解。

根据您链接到的先前问题，我完成了您的数据，以便我们有一些工作数据可供使用。这是完成的数据：

<rdf:RDF
    xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
    xmlns:dcat="http://www.w3.org/ns/dcat#"
    xmlns:skos="http://www.w3.org/2004/02/skos/core#"
    xmlns:foaf="http://xmlns.com/foaf/0.1/"
    xmlns:owl="http://www.w3.org/2002/07/owl#"
    xmlns:dct="http://purl.org/dc/terms/"
    xmlns:dctypes="http://purl.org/dc/dcmitype/"
    xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#">
  <dcat:Catalog rdf:about="http://uri/">
    <dcat:dataset>
    <dcat:Dataset rdf:about="http://url/" > 
        <dct:description xml:lang="ca">Description</dct:description>
        <dct:license rdf:resource="http://creativecommons.org/licenses/by/3.0/"/>
        <dcat:keyword xml:lang="ca">Keyword1</dcat:keyword>
        <dcat:distribution>
            <dcat:Download>
                <dcat:accessURL>http:/url/</dcat:accessURL>
                <dct:format>
                    <dct:IMT>
                        <rdf:value>application/pdf</rdf:value>
                        <rdfs:label>pdf</rdfs:label>
                    </dct:IMT>
                </dct:format>
                <dct:modified rdf:datatype="http://www.w3.or/2001/XMLSchema#date">2012-11-09T16:23:22</dct:modified>
           </dcat:Download>
        </dcat:distribution>
        <dct:publisher>
           <foaf:Organization>
              <dct:title xml:lang="en">Company</dct:title>
              <foaf:homepage rdf:resource="http://url/"/>
           </foaf:Organization>
        </dct:publisher>
    </dcat:Dataset>
    </dcat:dataset>
  </dcat:Catalog>
 </rdf:RDF>

听起来你只是想做一个深度优先搜索 http://en.wikipedia.org/wiki/Depth_first_search在类型的每个元素上dcat:Dataset。这很容易做到。我们只需选择 type 的每个元素dcat:Dataset然后从中开始深度优先搜索RDFNode.

import java.util.HashSet;
import java.util.Set;

import com.hp.hpl.jena.rdf.model.Model;
import com.hp.hpl.jena.rdf.model.ModelFactory;
import com.hp.hpl.jena.rdf.model.RDFNode;
import com.hp.hpl.jena.rdf.model.Statement;
import com.hp.hpl.jena.rdf.model.StmtIterator;
import com.hp.hpl.jena.vocabulary.RDF;


public class DFSinRDFwithJena {
    public static void main(String[] args) {
        Model model = ModelFactory.createDefaultModel();
        model.read( "rdfdfs.rdf" );

        StmtIterator stmts = model.listStatements( null, RDF.type, model.getResource( "http://www.w3.org/ns/dcat#" + "Dataset" ));
        while ( stmts.hasNext() ) {
            rdfDFS( stmts.next().getSubject(), new HashSet<RDFNode>(), "" );
        }
        model.write( System.out, "N3" );
    }

    public static void rdfDFS( RDFNode node, Set<RDFNode> visited, String prefix ) {
        if ( visited.contains( node )) {
            return;
        }
        else {
            visited.add( node );
            System.out.println( prefix + node );
            if ( node.isResource() ) {
                StmtIterator stmts = node.asResource().listProperties();
                while ( stmts.hasNext() ) {
                    Statement stmt = stmts.next();
                    rdfDFS( stmt.getObject(), visited, prefix + node + " =[" + stmt.getPredicate() + "]=> " );
                }
            }
        }
    }
}

这会产生输出：

http://url/
http://url/ =[http://purl.org/dc/terms/publisher]=> -f6d9b42:13f2e8dc5fb:-7ffd
http://url/ =[http://purl.org/dc/terms/publisher]=> -f6d9b42:13f2e8dc5fb:-7ffd =[http://purl.org/dc/terms/title]=> Company@en
http://url/ =[http://purl.org/dc/terms/publisher]=> -f6d9b42:13f2e8dc5fb:-7ffd =[http://www.w3.org/1999/02/22-rdf-syntax-ns#type]=> http://xmlns.com/foaf/0.1/Organization
http://url/ =[http://www.w3.org/ns/dcat#distribution]=> -f6d9b42:13f2e8dc5fb:-7fff
http://url/ =[http://www.w3.org/ns/dcat#distribution]=> -f6d9b42:13f2e8dc5fb:-7fff =[http://purl.org/dc/terms/modified]=> 2012-11-09T16:23:22^^http://www.w3.or/2001/XMLSchema#date
http://url/ =[http://www.w3.org/ns/dcat#distribution]=> -f6d9b42:13f2e8dc5fb:-7fff =[http://purl.org/dc/terms/format]=> -f6d9b42:13f2e8dc5fb:-7ffe
http://url/ =[http://www.w3.org/ns/dcat#distribution]=> -f6d9b42:13f2e8dc5fb:-7fff =[http://purl.org/dc/terms/format]=> -f6d9b42:13f2e8dc5fb:-7ffe =[http://www.w3.org/2000/01/rdf-schema#label]=> pdf
http://url/ =[http://www.w3.org/ns/dcat#distribution]=> -f6d9b42:13f2e8dc5fb:-7fff =[http://purl.org/dc/terms/format]=> -f6d9b42:13f2e8dc5fb:-7ffe =[http://www.w3.org/1999/02/22-rdf-syntax-ns#value]=> application/pdf
http://url/ =[http://www.w3.org/ns/dcat#distribution]=> -f6d9b42:13f2e8dc5fb:-7fff =[http://purl.org/dc/terms/format]=> -f6d9b42:13f2e8dc5fb:-7ffe =[http://www.w3.org/1999/02/22-rdf-syntax-ns#type]=> http://purl.org/dc/terms/IMT
http://url/ =[http://www.w3.org/ns/dcat#distribution]=> -f6d9b42:13f2e8dc5fb:-7fff =[http://www.w3.org/ns/dcat#accessURL]=> http:/url/
http://url/ =[http://www.w3.org/ns/dcat#distribution]=> -f6d9b42:13f2e8dc5fb:-7fff =[http://www.w3.org/1999/02/22-rdf-syntax-ns#type]=> http://www.w3.org/ns/dcat#Download
http://url/ =[http://www.w3.org/ns/dcat#keyword]=> Keyword1@ca
http://url/ =[http://purl.org/dc/terms/license]=> http://creativecommons.org/licenses/by/3.0/
http://url/ =[http://purl.org/dc/terms/description]=> Description@ca
http://url/ =[http://www.w3.org/1999/02/22-rdf-syntax-ns#type]=> http://www.w3.org/ns/dcat#Dataset

这不如您描述的输出漂亮，但似乎是您想要的。

关于 RDF 作为图形表示的注意事项

该问题使用了“每个陈述，直接位于dcat:Dataset”，我认为值得指出的是，以防出现任何混淆，RDF 是一种基于图的表示。确实，RDF/XML 序列化可用于提供一些人类可读的结构良好的 XML，但并不要求 XML 表示具有这种结构。要查看此差异，请注意以下 RDF/XML 表示the same图表如本答案之前发布的图表。

<rdf:RDF
    xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
    xmlns:dcat="http://www.w3.org/ns/dcat#"
    xmlns:skos="http://www.w3.org/2004/02/skos/core#"
    xmlns:foaf="http://xmlns.com/foaf/0.1/"
    xmlns:owl="http://www.w3.org/2002/07/owl#"
    xmlns:dct="http://purl.org/dc/terms/"
    xmlns:dctypes="http://purl.org/dc/dcmitype/"
    xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#" > 
  <rdf:Description rdf:nodeID="A0">
    <dct:modified rdf:datatype="http://www.w3.or/2001/XMLSchema#date">2012-11-09T16:23:22</dct:modified>
    <dct:format rdf:nodeID="A1"/>
    <dcat:accessURL>http:/url/</dcat:accessURL>
    <rdf:type rdf:resource="http://www.w3.org/ns/dcat#Download"/>
  </rdf:Description>
  <rdf:Description rdf:about="http://uri/">
    <dcat:dataset rdf:resource="http://url/"/>
    <rdf:type rdf:resource="http://www.w3.org/ns/dcat#Catalog"/>
  </rdf:Description>
  <rdf:Description rdf:about="http://url/">
    <dct:publisher rdf:nodeID="A2"/>
    <dcat:distribution rdf:nodeID="A0"/>
    <dcat:keyword xml:lang="ca">Keyword1</dcat:keyword>
    <dct:license rdf:resource="http://creativecommons.org/licenses/by/3.0/"/>
    <dct:description xml:lang="ca">Description</dct:description>
    <rdf:type rdf:resource="http://www.w3.org/ns/dcat#Dataset"/>
  </rdf:Description>
  <rdf:Description rdf:nodeID="A2">
    <foaf:homepage rdf:resource="http://url/"/>
    <dct:title xml:lang="en">Company</dct:title>
    <rdf:type rdf:resource="http://xmlns.com/foaf/0.1/Organization"/>
  </rdf:Description>
  <rdf:Description rdf:nodeID="A1">
    <rdfs:label>pdf</rdfs:label>
    <rdf:value>application/pdf</rdf:value>
    <rdf:type rdf:resource="http://purl.org/dc/terms/IMT"/>
  </rdf:Description>
</rdf:RDF>

The RDF图是完全相同的，尽管XML结构是非常不同的。我提出这一点只是为了强调使用 RDF 确实很重要这一事实作为图表，而不是分层 XML，即使特定的序列化可能表明我们可以使用后者。