使用Jena解析RDF递归 [英] Parsing RDF recursive with Jena

查看：78
本文介绍了使用Jena解析RDF递归的处理方法，对大家解决问题具有一定的参考价值，需要的朋友们下面随着小编来一起学习吧！

问题描述

我正在尝试使用Apache Jena解析RDF文档递归.它由像这样的数据集组成:

I'm trying to parse a RDF document recursive using Apache Jena. It consists out of datasets like this: 

<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，我可以得到每条语句，它直接位于dcat:Dataset(

I'm so far to get every statement, which is directly beneath dcat:Dataset (Iterate over specific resource in RDF file with Jena), but I want to find every triple in every level. My output should look like this: 

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
...

我用递归函数进行了尝试，该函数遍历语句，当语句不是文字时，它将对象跟随到下一个节点.像这样:

I've tried it with a recursive function, which iterates over the statements and when a statement is not a literal it follows the object to the next node. Like this:

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;

}

但是以某种方式，我用这种方法一无所获. 非常感谢您的每一次见解.

But somehow I'm getting nowhere with this method. Thank you very much for every insight.

推荐答案

根据您链接到的先前的问题，我完成了您的数据，以便我们可以使用一些工作数据.这是完整的数据:

Based on the earlier question that you linked to, I completed your data so that we have some working data to use. Here is the completed data:

<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>

听起来您只是想对每个类型的元素进行深度优先搜索 dcat:Dataset.这很容易做到.我们只需选择dcat:Dataset类型的每个元素，然后从该RDFNode开始深度优先搜索.

It sounds like you are just trying to do a depth first search on each element of type dcat:Dataset. That's easy enough to do. We just select each element of type dcat:Dataset and then start a depth first search from that 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() + "]=> " );
                }
            }
        }
    }
}

这将产生输出:

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

比您描述的输出要漂亮，但似乎正是您想要的.

which is less pretty than the output you described, but seems to be what you want.

该问题使用了每个语句直接在dcat:Dataset下面"的表示法，我认为值得指出的是，以防万一，RDF是基于图的表示形式.确实可以使用RDF/XML序列化来提供一些易于理解的结构化XML，但是没有任何要求XML表示具有这种结构.要了解这种差异，请注意，以下RDF/XML表示与该答案前面发布的图相同的图.

The question used the notation "every statement, which is directly beneath dcat:Dataset," and I think that it is worth pointing out, just in case there is any confusion, that RDF is a graph-based representation. It is true that the RDF/XML serialization can be used to provide some nicely structured XML that is human readable, but there is nothing that requires that that XML representation has that sort of structure. To see this difference, note that the following RDF/XML represents the same graph as the one posted earlier in this answer.

<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>

RDF图完全相同，即使 XML结构非常不同.我只是在此强调一个事实，即使将RDF 作为图形而不是作为分层XML来使用，也很重要，即使特定的序列化可能表明我们可以使用后者.

The RDF graph is exactly the same, even though the XML structure is very different. I only bring this up to highlight the fact that it really is important to work with RDF as a graph, not as hierarchical XML, even if a particular serialization might suggest that we could work with the latter.

这篇关于使用Jena解析RDF递归的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！

查看全文
相关文章
其他开发最新文章
热门教程
热门工具
登录 关闭
扫码关注1秒登录
发送“验证码”获取 | 15天全站免登陆