Package org.apache.tika.parser

Examples of org.apache.tika.parser.Parser

87
88
89
90
91
92
93
94
95
96
97
    Map<MediaType,Parser> parsers = parser.getParsers();
    parsers.put(MediaType.APPLICATION_XML, new HtmlParser());
    parser.setParsers(parsers);

    parser.setFallback(new Parser() {
      public Set<MediaType> getSupportedTypes(ParseContext parseContext) {
        return parser.getSupportedTypes(parseContext);
      }

      public void parse(InputStream inputStream, ContentHandler contentHandler, Metadata metadata, ParseContext parseContext) {
View Full Code Here
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
    /**
     * Test that with only ID3v1 tags, we get some information out  
     */
    @Test
    public void testMp3ParsingID3v1() throws Exception {
        Parser parser = new AutoDetectParser(); // Should auto-detect!
        ContentHandler handler = new BodyContentHandler();
        Metadata metadata = new Metadata();

        InputStream stream = Mp3ParserTest.class.getResourceAsStream(
                "/test-documents/testMP3id3v1.mp3");
        try {
            parser.parse(stream, handler, metadata, new ParseContext());
        } finally {
            stream.close();
        }

        assertEquals("audio/mpeg", metadata.get(Metadata.CONTENT_TYPE));
View Full Code Here
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
     * Test that with only ID3v2 tags, we get the full
     *  set of information out.
     */
    @Test
    public void testMp3ParsingID3v2() throws Exception {
        Parser parser = new AutoDetectParser(); // Should auto-detect!
        ContentHandler handler = new BodyContentHandler();
        Metadata metadata = new Metadata();

        InputStream stream = Mp3ParserTest.class.getResourceAsStream(
                "/test-documents/testMP3id3v2.mp3");
        try {
            parser.parse(stream, handler, metadata, new ParseContext());
        } finally {
            stream.close();
        }

        // Check core properties
View Full Code Here
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
     * Test that with both id3v2 and id3v1, we prefer the
     *  details from id3v2
     */
    @Test
    public void testMp3ParsingID3v1v2() throws Exception {
        Parser parser = new AutoDetectParser(); // Should auto-detect!
        ContentHandler handler = new BodyContentHandler();
        Metadata metadata = new Metadata();

        InputStream stream = Mp3ParserTest.class.getResourceAsStream(
                "/test-documents/testMP3id3v1_v2.mp3");
        try {
            parser.parse(stream, handler, metadata, new ParseContext());
        } finally {
            stream.close();
        }

        assertEquals("audio/mpeg", metadata.get(Metadata.CONTENT_TYPE));
View Full Code Here
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
     * Test that with only ID3v2 tags, of version 2.4, we get the full
     *  set of information out.
     */
    @Test
    public void testMp3ParsingID3v24() throws Exception {
        Parser parser = new AutoDetectParser(); // Should auto-detect!
        ContentHandler handler = new BodyContentHandler();
        Metadata metadata = new Metadata();

        InputStream stream = Mp3ParserTest.class.getResourceAsStream(
                "/test-documents/testMP3id3v24.mp3");
        try {
            parser.parse(stream, handler, metadata, new ParseContext());
        } finally {
            stream.close();
        }

        assertEquals("audio/mpeg", metadata.get(Metadata.CONTENT_TYPE));
View Full Code Here
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
     * Tests that a file with characters not in the ISO 8859-1
     *  range is correctly handled
     */
    @Test
    public void testMp3ParsingID3i18n() throws Exception {
       Parser parser = new AutoDetectParser(); // Should auto-detect!
       ContentHandler handler = new BodyContentHandler();
       Metadata metadata = new Metadata();

       InputStream stream = Mp3ParserTest.class.getResourceAsStream(
               "/test-documents/testMP3i18n.mp3");
       try {
           parser.parse(stream, handler, metadata, new ParseContext());
       } finally {
           stream.close();
       }

       assertEquals("audio/mpeg", metadata.get(Metadata.CONTENT_TYPE));
View Full Code Here
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
     * Tests that a file with both lyrics and
     *  ID3v2 tags gets both extracted correctly
     */
    @Test
    public void testMp3ParsingLyrics() throws Exception {
        Parser parser = new AutoDetectParser(); // Should auto-detect!
        ContentHandler handler = new BodyContentHandler();
        Metadata metadata = new Metadata();

        // Note - our test file has a lyrics tag, but lacks any
        //  lyrics in the tags, so we can't test that bit
        // TODO Find a better sample file
       
        InputStream stream = Mp3ParserTest.class.getResourceAsStream(
                "/test-documents/testMP3lyrics.mp3");
        try {
            parser.parse(stream, handler, metadata, new ParseContext());
        } finally {
            stream.close();
        }

        assertEquals("audio/mpeg", metadata.get(Metadata.CONTENT_TYPE));
View Full Code Here
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
     * This test will check for the complicated set of ID3v2.4
     *  tags.
     */
    @Test
    public void testTIKA424() throws Exception {
       Parser parser = new AutoDetectParser(); // Should auto-detect!
       ContentHandler handler = new BodyContentHandler();
       Metadata metadata = new Metadata();

       InputStream stream = Mp3ParserTest.class.getResourceAsStream(
               "/test-documents/test2.mp3");
       if(stream == null) {
          // You haven't downloaded the file
          // Skip the test
          return;
       }
      
       try {
           parser.parse(stream, handler, metadata, new ParseContext());
       } finally {
           stream.close();
       }

       assertEquals("audio/mpeg", metadata.get(Metadata.CONTENT_TYPE));
View Full Code Here
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
     * In this case, it is a file with JPEG data in the ID3, which
     *  is trunacted before the end of the JPEG bit of the ID3 frame.
     */
    @Test
    public void testTIKA474() throws Exception {
       Parser parser = new AutoDetectParser(); // Should auto-detect!
       ContentHandler handler = new BodyContentHandler();
       Metadata metadata = new Metadata();

       InputStream stream = Mp3ParserTest.class.getResourceAsStream(
               "/test-documents/testMP3truncated.mp3");
      
      
       try {
           parser.parse(stream, handler, metadata, new ParseContext());
       } finally {
           stream.close();
       }

       // Check we could get the headers from the start
View Full Code Here
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
    }

    // TIKA-1024
    @Test
    public void testNakedUTF16BOM() throws Exception {
       Parser parser = new AutoDetectParser(); // Should auto-detect!
       ContentHandler handler = new BodyContentHandler();
       Metadata metadata = new Metadata();

       InputStream stream = Mp3ParserTest.class.getResourceAsStream(
               "/test-documents/testNakedUTF16BOM.mp3");
      
       try {
           parser.parse(stream, handler, metadata, new ParseContext());
       } finally {
           stream.close();
       }
       assertEquals("audio/mpeg", metadata.get(Metadata.CONTENT_TYPE));
       assertEquals("", metadata.get(XMPDM.GENRE));
View Full Code Here
TOP

Related Classes of org.apache.tika.parser.Parser

  • bixo.parser.TikaCallableTest
  • com.baasbox.controllers.File
  • com.cloudera.cdk.morphline.solrcell.SolrCellBuilder$SolrCell
  • com.digitalpebble.behemoth.tika.TikaProcessor
  • com.findwise.hydra.stage.tika.SimpleFetchingTikaStageTest
  • com.ikanow.infinit.e.harvest.extraction.document.file.FileHarvester
  • com.ikanow.infinit.e.harvest.extraction.text.legacy.TextExtractorTika
  • com.nidhinova.tika.server.TikaService
  • com.tamingtext.tika.TikaTest
  • ipcress.Extraction
    • 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.