用Jena工具包实现数据源的融合

All My Eye 博客上最近有几篇帖子在写如何实现几个RDF数据源的联合查询、以及实现遗留的RDB数据库和RDF数据源的融合。前者基于Jena实现，后者基于Jena + SquirrelRDF实现。

用Jena实现几个RDF数据源联合查询的关键在于基于named graph的RDF复杂查询，以all my eye上面这个例子中的Articles和Authors本体以及Berners-Lee本人主页上的CARD.rdf为例，由于card.rdf中有几个URI是../..等的形式，Jena报了些小错，权且把这3个文件下载下来、作为文件读取了，代码如下：

import com.hp.hpl.jena.rdf.model.* ;
import com.hp.hpl.jena.query.*;

import java.util.* ;
import java.io.* ;

。。。。

1     String queryString =
2                     "PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>" +
3                     "PREFIX dc: <http://purl.org/dc/elements/1.1/>" +
4                     "PREFIX foaf: <http://xmlns.com/foaf/0.1/>" +
5                     "SELECT ?ms_articleid ?ms_title ?ms_authorid ?join_familyname ?ext_authorid  ?ext_workplace ?ext_phone" +
6                     "WHERE {" +
7                         "?ms_articleid dc:identifier <http://metastore.ingenta.com/content/stc/00493155/v46n1/s9>." +
8                         "?ms_articleid dc:title ?ms_title ." +
9                         "?ms_articleid foaf:maker ?ms_authorid ." +
10                         "GRAPH <http://card.rdf> {"+
11                         "?ext_authorid foaf:family_name ?join_familyname ." +
12                         "?ext_authorid foaf:workplaceHomepage ?ext_workplace ." +
13                         "?ext_authorid foaf:phone ?ext_phone " +
14                         "}"+
15                         "GRAPH <http://authors.rdf> {" +
16                         "?ms_authorid foaf:family_name ?join_familyname" +
17                         "}" +
18                         "FILTER (?ext_authorid != ?ms_authorid)" +
19                     "}";
20
21             Query query = QueryFactory.create(queryString);
22
23             // data source
24             DataSource ds = DatasetFactory.create();
25
30             Model articlemodel = ModelFactory.createDefaultModel();
32             articlemodel.read(new FileInputStream(path + "articles.rdf"),null) ;
33
34
35             Model authormodel = ModelFactory.createDefaultModel();
36             authormodel.read(new FileInputStream(path + "authors.rdf"),null) ;
37
38
39             Model cardmodel = ModelFactory.createDefaultModel();
40             cardmodel.read(new FileInputStream(path + "card.rdf"),null) ;
41
42             ds.setDefaultModel(articlemodel);
43             ds.addNamedModel("http://card.rdf",cardmodel);
44             ds.addNamedModel("http://authors.rdf",authormodel);
45
46             QueryExecution qe = QueryExecutionFactory.create(query,ds);
47
48             ResultSet rs = qe.execSelect();
49             ResultSetFormatter.out(System.out, rs, query);
50
51             String ss = ResultSetFormatter.asXMLString(rs);
52             System.out.println(ss);
53
54             qe.close();

按照上面的代码，得到一个XML文件：

返回结果

当然也可以直接从URL从文件中读到RDF图中：

URL url = new URL("http://xx/yy.rdf");
InputStream stream = url.openStream();
xxxxmodel.read(stream,null);

all my eye上另外一个例子利用了SquirrelRDF，它实现了从RDB到RDF格式的自动转换，从而可以将遗留的数据格式统一起来。而D2RQ则是另外一个类似的软件包，而且据说它除了能将RDF、RDB统一起来，还实现了多个不同格式数据源查询的优化。