Package org.apache.tika.config

Examples of org.apache.tika.config.TikaConfig

64
65
66
67
68
69
70
71
72
73
74
75
        assertEquals(a, parsers.get(0));
        assertEquals(b, parsers.get(1));
    }

    public void testDefaultParser() throws Exception {
       TikaConfig config = TikaConfig.getDefaultConfig();

       CompositeParser parser = (CompositeParser) config.getParser();

       // Check it has the full registry
       assertEquals(config.getMediaTypeRegistry(), parser.getMediaTypeRegistry());
    }
View Full Code Here
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
       Parser bmpAliasParser = new DummyParser(
             new HashSet<MediaType>(Arrays.asList(bmpAlias)),
             bmpAliasMetadata, null
       );
      
       TikaConfig config = TikaConfig.getDefaultConfig();
       CompositeParser canonical = new CompositeParser(
             config.getMediaTypeRegistry(), bmpCanonicalParser
       );
       CompositeParser alias = new CompositeParser(
             config.getMediaTypeRegistry(), bmpAliasParser
       );
       CompositeParser both = new CompositeParser(
             config.getMediaTypeRegistry(), bmpCanonicalParser, bmpAliasParser
       );
      
       ContentHandler handler = new BodyContentHandler();
       Metadata metadata;
      
View Full Code Here
45
46
47
48
49
50
51
52
53
54
    private URL u;

    private static final File f = new File("/a/b/c/x.pdf");

    protected void setUp() throws Exception{
        TikaConfig config = TikaConfig.getDefaultConfig();
        repo = config.getMimeRepository();
        tika = new Tika(config);
        u = new URL("http://mydomain.com/x.pdf?x=y");
    }
View Full Code Here
41
42
43
44
45
46
47
48
49
50
51
public class OOXMLParserTest extends TestCase {
    private Parser parser;
  
    @Override
    protected void setUp() throws Exception {
       TikaConfig config = TikaConfig.getDefaultConfig();
       ContainerAwareDetector detector = new ContainerAwareDetector(
             config.getMimeRepository()
       );
       parser = new AutoDetectParser(detector);
    }
View Full Code Here
68
69
70
71
72
73
74
75
76
77
78
79
        assertEquals(b, parsers.get(1));
    }

    @Test
    public void testDefaultParser() throws Exception {
       TikaConfig config = TikaConfig.getDefaultConfig();

       CompositeParser parser = (CompositeParser) config.getParser();

       // Check it has the full registry
       assertEquals(config.getMediaTypeRegistry(), parser.getMediaTypeRegistry());
    }
View Full Code Here
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
       Parser bmpAliasParser = new DummyParser(
             new HashSet<MediaType>(Arrays.asList(bmpAlias)),
             bmpAliasMetadata, null
       );
      
       TikaConfig config = TikaConfig.getDefaultConfig();
       CompositeParser canonical = new CompositeParser(
             config.getMediaTypeRegistry(), bmpCanonicalParser
       );
       CompositeParser alias = new CompositeParser(
             config.getMediaTypeRegistry(), bmpAliasParser
       );
       CompositeParser both = new CompositeParser(
             config.getMediaTypeRegistry(), bmpCanonicalParser, bmpAliasParser
       );
      
       ContentHandler handler = new BodyContentHandler();
       Metadata metadata;
      
View Full Code Here
563
564
565
566
567
568
569
570
571
572
    }


    private void configure(String configFilePath) throws Exception {
        this.configFilePath = configFilePath;
        TikaConfig config = new TikaConfig(new File(configFilePath));
        parser = new AutoDetectParser(config);
        detector = config.getDetector();
        context.set(Parser.class, parser);
    }
View Full Code Here
156
157
158
159
160
161
162
163
164
        }
    }

    @Test
    public void testBundleParsers(BundleContext bc) throws Exception {
        TikaConfig tika = new TikaConfig();

        // TODO Implement as with Detectors
    }
View Full Code Here
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
     *
     * @param args ignored
     * @throws Exception if an error occurs
     */
    public static void main(String[] args) throws Exception {
        TikaConfig config = TikaConfig.getDefaultConfig();
        if (args.length > 0) {
            File configFile = new File(args[0]);
            config = new TikaConfig(configFile);
        }
        UIManager.setLookAndFeel(UIManager.getSystemLookAndFeelClassName());
        final TikaConfig finalConfig = config;
        SwingUtilities.invokeLater(new Runnable() {
            public void run() {
                new TikaGUI(new AutoDetectParser(finalConfig)).setVisible(true);
            }
        });
View Full Code Here
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
            Writer writer = new OutputStreamWriter(new FileOutputStream(configFile), encoding);
            ex.dump(TikaConfig.getDefaultConfig(), writer, encoding);
            writer.flush();
            writer.close();

            TikaConfig c = new TikaConfig(configFile);
            assertEquals(CompositeParser.class, c.getParser().getClass());
            assertEquals(CompositeDetector.class, c.getDetector().getClass());

            CompositeParser p = (CompositeParser) c.getParser();
            assertTrue("enough parsers?", p.getParsers().size() > 130);

            CompositeDetector d = (CompositeDetector) c.getDetector();
            assertTrue("enough detectors?", d.getDetectors().size() > 3);
            //just try to load it into autodetect to make sure no errors are thrown
            Parser auto = new AutoDetectParser(c);
        }
    }
View Full Code Here
TOP

Related Classes of org.apache.tika.config.TikaConfig

  • com.nidhinova.tika.server.TikaService
  • org.apache.any23.mime.TikaMIMETypeDetector
  • org.apache.clerezza.uima.metadatagenerator.mediatype.TikaTextExtractor
  • org.apache.jackrabbit.core.query.lucene.JackrabbitParser
  • org.apache.jackrabbit.core.query.lucene.SearchIndex
  • org.apache.nutch.parse.tika.TikaParser
  • org.apache.nutch.util.MimeUtil
  • org.apache.solr.handler.dataimport.TikaEntityProcessor
  • org.apache.solr.handler.extraction.ExtractingRequestHandler
  • org.apache.tika.bundle.BundleIT
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.