View Javadoc
1   /*
2    * Licensed to the Apache Software Foundation (ASF) under one or more
3    * contributor license agreements.  See the NOTICE file distributed with
4    * this work for additional information regarding copyright ownership.
5    * The ASF licenses this file to You under the Apache License, Version 2.0
6    * (the "License"); you may not use this file except in compliance with
7    * the License.  You may obtain a copy of the License at
8    *
9    *  http://www.apache.org/licenses/LICENSE-2.0
10   *
11   * Unless required by applicable law or agreed to in writing, software
12   * distributed under the License is distributed on an "AS IS" BASIS,
13   * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14   * See the License for the specific language governing permissions and
15   * limitations under the License.
16   */
17  
18  package org.apache.any23.extractor.html;
19  
20  import org.apache.any23.extractor.ExtractionContext;
21  import org.apache.any23.extractor.ExtractionException;
22  import org.apache.any23.extractor.ExtractionParameters;
23  import org.apache.any23.extractor.ExtractionResult;
24  import org.apache.any23.extractor.ExtractorDescription;
25  import org.apache.any23.rdf.Any23ValueFactoryWrapper;
26  import org.apache.any23.extractor.Extractor.TagSoupDOMExtractor;
27  import org.eclipse.rdf4j.model.BNode;
28  import org.eclipse.rdf4j.model.IRI;
29  import org.eclipse.rdf4j.model.ValueFactory;
30  import org.eclipse.rdf4j.model.impl.SimpleValueFactory;
31  import org.w3c.dom.Document;
32  
33  import java.io.IOException;
34  
35  /**
36   * Extractor for "ICBM coordinates" provided as META headers in the head of an HTML page.
37   *
38   * @author Gabriele Renzi
39   * @author Richard Cyganiak (richard@cyganiak.de)
40   */
41  public class ICBMExtractor implements TagSoupDOMExtractor {
42  
43      @Override
44      public void run(ExtractionParameters extractionParameters, ExtractionContext extractionContext, Document in,
45              ExtractionResult out) throws IOException, ExtractionException {
46  
47          // ICBM is the preferred method, if two values are available it is meaningless to read both
48          String props = DomUtils.find(in, "//META[@name=\"ICBM\" or @name=\"geo.position\"]/@content");
49          if ("".equals(props))
50              return;
51  
52          String[] coords = props.split("[;,]");
53          float lat, lon;
54          try {
55              lat = Float.parseFloat(coords[0]);
56              lon = Float.parseFloat(coords[1]);
57          } catch (NumberFormatException nfe) {
58              return;
59          }
60  
61          final ValueFactory factory = new Any23ValueFactoryWrapper(SimpleValueFactory.getInstance(), out);
62          BNode point = factory.createBNode();
63          out.writeTriple(extractionContext.getDocumentIRI(), expand("dcterms:related"), point);
64          out.writeTriple(point, expand("rdf:type"), expand("geo:Point"));
65          out.writeTriple(point, expand("geo:lat"), factory.createLiteral(Float.toString(lat)));
66          out.writeTriple(point, expand("geo:long"), factory.createLiteral(Float.toString(lon)));
67      }
68  
69      private IRI expand(String curie) {
70          return getDescription().getPrefixes().expand(curie);
71      }
72  
73      @Override
74      public ExtractorDescription getDescription() {
75          return ICBMExtractorFactory.getDescriptionInstance();
76      }
77  
78  }