1 | |
|
2 | |
|
3 | |
|
4 | |
|
5 | |
|
6 | |
|
7 | |
|
8 | |
|
9 | |
|
10 | |
|
11 | |
|
12 | |
|
13 | |
|
14 | |
|
15 | |
|
16 | |
|
17 | |
|
18 | |
package org.apache.any23.plugin.htmlscraper; |
19 | |
|
20 | |
import de.l3s.boilerpipe.BoilerpipeExtractor; |
21 | |
import de.l3s.boilerpipe.BoilerpipeProcessingException; |
22 | |
import de.l3s.boilerpipe.extractors.ArticleExtractor; |
23 | |
import de.l3s.boilerpipe.extractors.CanolaExtractor; |
24 | |
import de.l3s.boilerpipe.extractors.DefaultExtractor; |
25 | |
import de.l3s.boilerpipe.extractors.LargestContentExtractor; |
26 | |
import org.apache.any23.extractor.ExtractionContext; |
27 | |
import org.apache.any23.extractor.ExtractionException; |
28 | |
import org.apache.any23.extractor.ExtractionParameters; |
29 | |
import org.apache.any23.extractor.ExtractionResult; |
30 | |
import org.apache.any23.extractor.Extractor; |
31 | |
import org.apache.any23.extractor.ExtractorFactory; |
32 | |
import org.apache.any23.extractor.SimpleExtractorFactory; |
33 | |
import org.apache.any23.vocab.SINDICE; |
34 | |
import org.openrdf.model.URI; |
35 | |
import org.openrdf.model.impl.ValueFactoryImpl; |
36 | |
|
37 | |
import java.io.IOException; |
38 | |
import java.io.InputStream; |
39 | |
import java.io.InputStreamReader; |
40 | |
import java.util.ArrayList; |
41 | |
import java.util.Arrays; |
42 | |
import java.util.List; |
43 | |
|
44 | |
|
45 | |
|
46 | |
|
47 | |
|
48 | |
|
49 | |
|
50 | 0 | public class HTMLScraperExtractor implements Extractor.ContentExtractor { |
51 | |
|
52 | |
public final static String NAME = "html-scraper"; |
53 | |
|
54 | 0 | public final static URI PAGE_CONTENT_DE_PROPERTY = |
55 | |
ValueFactoryImpl.getInstance().createURI(SINDICE.NS + "pagecontent/de"); |
56 | 0 | public final static URI PAGE_CONTENT_AE_PROPERTY = |
57 | |
ValueFactoryImpl.getInstance().createURI(SINDICE.NS + "pagecontent/ae"); |
58 | 0 | public final static URI PAGE_CONTENT_LCE_PROPERTY = |
59 | |
ValueFactoryImpl.getInstance().createURI(SINDICE.NS + "pagecontent/lce"); |
60 | 0 | public final static URI PAGE_CONTENT_CE_PROPERTY = |
61 | |
ValueFactoryImpl.getInstance().createURI(SINDICE.NS + "pagecontent/ce"); |
62 | |
|
63 | 0 | protected final static ExtractorFactory<HTMLScraperExtractor> factory = |
64 | |
SimpleExtractorFactory.create( |
65 | |
NAME, |
66 | |
null, |
67 | |
Arrays.asList("text/html;q=0.02", "application/xhtml+xml;q=0.02"), |
68 | |
null, |
69 | |
HTMLScraperExtractor.class |
70 | |
); |
71 | |
|
72 | 0 | private final List<ExtractionRule> extractionRules = new ArrayList<ExtractionRule>(); |
73 | |
|
74 | 0 | public HTMLScraperExtractor() { |
75 | 0 | loadDefaultRules(); |
76 | 0 | } |
77 | |
|
78 | |
public void addTextExtractor(String name, URI property, BoilerpipeExtractor extractor) { |
79 | 0 | extractionRules.add( new ExtractionRule(name, property, extractor) ); |
80 | 0 | } |
81 | |
|
82 | |
public String[] getTextExtractors() { |
83 | 0 | final List<String> extractors = new ArrayList<String>(); |
84 | 0 | for(ExtractionRule er : extractionRules) { |
85 | 0 | extractors.add(er.name); |
86 | |
} |
87 | 0 | return extractors.toArray( new String[extractors.size()] ); |
88 | |
} |
89 | |
|
90 | |
public void run( |
91 | |
ExtractionParameters extractionParameters, |
92 | |
ExtractionContext extractionContext, |
93 | |
InputStream inputStream, |
94 | |
ExtractionResult extractionResult |
95 | |
) throws IOException, ExtractionException { |
96 | |
try { |
97 | 0 | final URI documentURI = extractionContext.getDocumentURI(); |
98 | 0 | for (ExtractionRule extractionRule : extractionRules) { |
99 | 0 | final String content = extractionRule.boilerpipeExtractor.getText(new InputStreamReader(inputStream)); |
100 | 0 | extractionResult.writeTriple( |
101 | |
documentURI, |
102 | |
extractionRule.property, |
103 | |
ValueFactoryImpl.getInstance().createLiteral(content) |
104 | |
); |
105 | 0 | } |
106 | 0 | } catch (BoilerpipeProcessingException bpe) { |
107 | 0 | throw new ExtractionException("Error while applying text processor " + ArticleExtractor.class, bpe); |
108 | 0 | } |
109 | 0 | } |
110 | |
|
111 | |
public ExtractorFactory getDescription() { |
112 | 0 | return factory; |
113 | |
} |
114 | |
|
115 | |
public void setStopAtFirstError(boolean b) { |
116 | |
|
117 | 0 | } |
118 | |
|
119 | |
private void loadDefaultRules() { |
120 | 0 | addTextExtractor("default-extractor" , PAGE_CONTENT_DE_PROPERTY , DefaultExtractor.getInstance()); |
121 | 0 | addTextExtractor("article-extractor" , PAGE_CONTENT_AE_PROPERTY , ArticleExtractor.getInstance()); |
122 | 0 | addTextExtractor("large-content-extractor", PAGE_CONTENT_LCE_PROPERTY, LargestContentExtractor.getInstance()); |
123 | 0 | addTextExtractor("canola-extractor" , PAGE_CONTENT_CE_PROPERTY , CanolaExtractor.getInstance()); |
124 | 0 | } |
125 | |
|
126 | |
|
127 | |
|
128 | |
|
129 | |
class ExtractionRule { |
130 | |
|
131 | |
public final String name; |
132 | |
public final URI property; |
133 | |
public final BoilerpipeExtractor boilerpipeExtractor; |
134 | |
|
135 | 0 | ExtractionRule(String name, URI property, BoilerpipeExtractor boilerpipeExtractor) { |
136 | 0 | if(name == null) { |
137 | 0 | throw new NullPointerException("name cannot be null."); |
138 | |
} |
139 | 0 | if(property == null) { |
140 | 0 | throw new NullPointerException("property cannot be null."); |
141 | |
} |
142 | 0 | if(boilerpipeExtractor == null) { |
143 | 0 | throw new NullPointerException("extractor cannot be null."); |
144 | |
} |
145 | 0 | this.name = name; |
146 | 0 | this.property = property; |
147 | 0 | this.boilerpipeExtractor = boilerpipeExtractor; |
148 | 0 | } |
149 | |
|
150 | |
} |
151 | |
} |