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.microformats2;
19  
20  import org.apache.any23.extractor.ExtractionResult;
21  import org.apache.any23.extractor.ExtractorDescription;
22  import org.apache.any23.extractor.TagSoupExtractionResult;
23  import org.apache.any23.vocab.VCard;
24  import org.eclipse.rdf4j.model.BNode;
25  import org.eclipse.rdf4j.model.vocabulary.RDF;
26  import org.w3c.dom.Node;
27  import org.apache.any23.extractor.html.EntityBasedMicroformatExtractor;
28  import org.apache.any23.extractor.html.HTMLDocument;
29  
30  import java.util.ArrayList;
31  
32  /**
33   * Extractor for the <a href="http://microformats.org/wiki/h-geo">h-geo</a> microformat.
34   *
35   * @author Nisala Nirmana
36   */
37  public class HGeoExtractor extends EntityBasedMicroformatExtractor {
38  
39      private static final VCard vVCARD = VCard.getInstance();
40  
41      private static final String[] geoFields = { "latitude", "longitude", "altitude" };
42  
43      @Override
44      public ExtractorDescription getDescription() {
45          return HGeoExtractorFactory.getDescriptionInstance();
46      }
47  
48      protected String getBaseClassName() {
49          return Microformats2Prefixes.CLASS_PREFIX + "geo";
50      }
51  
52      @Override
53      protected void resetExtractor() {
54          // Empty.
55      }
56  
57      protected boolean extractEntity(Node node, ExtractionResult out) {
58          if (null == node)
59              return false;
60          final HTMLDocumentl/HTMLDocument.html#HTMLDocument">HTMLDocument document = new HTMLDocument(node);
61          BNode geo = getBlankNodeFor(node);
62          out.writeTriple(geo, RDF.TYPE, vVCARD.Location);
63          final String extractorName = getDescription().getExtractorName();
64          ArrayList<HTMLDocument.TextField> geoNodes = new ArrayList<HTMLDocument.TextField>();
65          for (String field : geoFields) {
66              geoNodes.add(document.getSingularTextField(Microformats2Prefixes.PROPERTY_PREFIX + field));
67          }
68          if (geoNodes.get(0).source() == null) {
69              String[] composed = document.getSingularUrlField(Microformats2Prefixes.CLASS_PREFIX + "geo").value()
70                      .split(";");
71              for (int counter = 0; counter < composed.length; counter++) {
72                  conditionallyAddStringProperty(
73                          document.getSingularUrlField(Microformats2Prefixes.CLASS_PREFIX + "geo").source(), geo,
74                          vVCARD.getProperty(geoFields[counter]), composed[counter]);
75              }
76          } else {
77              for (int counter = 0; counter < geoNodes.size(); counter++) {
78                  conditionallyAddStringProperty(geoNodes.get(counter).source(), geo,
79                          vVCARD.getProperty(geoFields[counter]), geoNodes.get(counter).value());
80              }
81          }
82          final TagSoupExtractionResult../org/apache/any23/extractor/TagSoupExtractionResult.html#TagSoupExtractionResult">TagSoupExtractionResult tser = (TagSoupExtractionResult) getCurrentExtractionResult();
83          tser.addResourceRoot(document.getPathToLocalRoot(), geo, this.getClass());
84          return true;
85      }
86  
87  }