Package org.apache.tika.detect

Examples of org.apache.tika.detect.Detector

220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
     * We don't currently support the .xlsb file format
     *  (an OOXML container with binary blobs), but we
     *  shouldn't break on these files either (TIKA-826) 
     */
    public void testExcelXLSB() throws Exception {
       Detector detector = new DefaultDetector();
       AutoDetectParser parser = new AutoDetectParser();
      
       InputStream input = ExcelParserTest.class.getResourceAsStream(
             "/test-documents/testEXCEL.xlsb");
       Metadata m = new Metadata();
       m.add(Metadata.RESOURCE_NAME_KEY, "excel.xlsb");
      
       // Should be detected correctly
       MediaType type = null;
       try {
          type = detector.detect(input, m);
          assertEquals("application/vnd.ms-excel.sheet.binary.macroenabled.12", type.toString());
       } finally {
          input.close();
       }
      
View Full Code Here
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
    }

    if (mediaType !=null) {
      metadata.add(org.apache.tika.metadata.HttpHeaders.CONTENT_TYPE, mediaType.toString());

      final Detector detector = parser.getDetector();

      parser.setDetector(new Detector() {
        public MediaType detect(InputStream inputStream, Metadata metadata) throws IOException {
          String ct = metadata.get(org.apache.tika.metadata.HttpHeaders.CONTENT_TYPE);

          if (ct!=null) {
            return MediaType.parse(ct);
          } else {
            return detector.detect(inputStream, metadata);
          }
        }
      });
    }
  }
View Full Code Here
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
    }

    if (mediaType !=null) {
      metadata.add(org.apache.tika.metadata.HttpHeaders.CONTENT_TYPE, mediaType.toString());

      final Detector detector = parser.getDetector();

      parser.setDetector(new Detector() {
        public MediaType detect(InputStream inputStream, Metadata metadata) throws IOException {
          String ct = metadata.get(org.apache.tika.metadata.HttpHeaders.CONTENT_TYPE);

          if (ct!=null) {
            return MediaType.parse(ct);
          } else {
            return detector.detect(inputStream, metadata);
          }
        }
      });
    }
  }
View Full Code Here
103
104
105
106
107
108
109
110
111
112
113
114
        String data = "<!DOCTYPE html>\n<html><body><p>test <span>content</span></p></body></html>";
        InputStream stream = new ByteArrayInputStream(data.getBytes("UTF-8"));
        Writer writer = new StringWriter();
        ContentHandler contentHandler = new BodyContentHandler(writer);
        Metadata metadata = new Metadata();
        Detector contentTypeDetector = new DefaultDetector(classLoader);
        MediaType type = contentTypeDetector.detect(stream, metadata);
        assertEquals(type.toString(), "text/html");
        metadata.add(Metadata.CONTENT_TYPE, type.toString());
        ParseContext parseCtx = new ParseContext();
        parser.parse(stream, contentHandler, metadata, parseCtx);
        writer.flush();
View Full Code Here
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
     *  (an OOXML container with binary blobs), but we
     *  shouldn't break on these files either (TIKA-826) 
     */
    @Test
    public void testExcelXLSB() throws Exception {
       Detector detector = new DefaultDetector();
       AutoDetectParser parser = new AutoDetectParser();
      
       InputStream input = ExcelParserTest.class.getResourceAsStream(
             "/test-documents/testEXCEL.xlsb");
       Metadata m = new Metadata();
       m.add(Metadata.RESOURCE_NAME_KEY, "excel.xlsb");
      
       // Should be detected correctly
       MediaType type = null;
       try {
          type = detector.detect(input, m);
          assertEquals("application/vnd.ms-excel.sheet.binary.macroenabled.12", type.toString());
       } finally {
          input.close();
       }
      
View Full Code Here
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
     * We don't currently support the old Excel 95 .xls file format,
     *  but we shouldn't break on these files either (TIKA-976) 
     */
    @Test
    public void testExcel95() throws Exception {
       Detector detector = new DefaultDetector();
       AutoDetectParser parser = new AutoDetectParser();
      
       InputStream input = ExcelParserTest.class.getResourceAsStream(
             "/test-documents/testEXCEL_95.xls");
       Metadata m = new Metadata();
       m.add(Metadata.RESOURCE_NAME_KEY, "excel_95.xls");
      
       // Should be detected correctly
       MediaType type = null;
       try {
          type = detector.detect(input, m);
          assertEquals("application/vnd.ms-excel", type.toString());
       } finally {
          input.close();
       }
      
View Full Code Here
188
189
190
191
192
193
194
195
196
197
198
       ForkParser parser = new ForkParser(
             ForkParserIntegrationTest.class.getClassLoader(),
             tika.getParser());
      
       ParseContext context = new ParseContext();
       context.set(Detector.class, new Detector() {
          public MediaType detect(InputStream input, Metadata metadata) {
             return MediaType.OCTET_STREAM;
          }
       });
View Full Code Here
115
116
117
118
119
120
121
122
123
124
125
                "for example: <mimeTypeRepository resource=\"/org/apache/tika/mime/tika-mimetypes.xml\"/>");
        rootElement.appendChild(mimeComment);
    }

    private void addDetectors(Element rootElement, Document doc, TikaConfig config) throws Exception {
        Detector detector = config.getDetector();
        Element detectorsElement = doc.createElement("detectors");

        if (detector instanceof DefaultDetector) {
            List<Detector> children = ((DefaultDetector)detector).getDetectors();
            for (Detector d : children) {
View Full Code Here
248
249
250
251
252
253
254
255
256
257
258
259
260
   
    private String getExtension(TikaInputStream is, Metadata metadata) {
        String cType = metadata.get(Metadata.CONTENT_TYPE);
        TikaConfig config = getConfig();
        if (cType == null) {
            Detector detector = config.getDetector();
            try {
                MediaType mediaType = detector.detect(is, metadata);
                MimeTypes types = config.getMimeRepository();
                MimeType mime = types.forName(mediaType.toString());
                metadata.set(Metadata.CONTENT_TYPE, mediaType.getSubtype());
                return mime.getExtension();
            } catch (IOException e) {
View Full Code Here
39
40
41
42
43
44
45
46
47
48
49
50
51
52
   private static final String file = "testWINMAIL.dat";
  
   @Test
   public void testBasics() throws Exception {
      TikaInputStream stream = getTestFile(file);
      Detector detector = new DefaultDetector();
      try {
         assertEquals(
                 MediaType.application("vnd.ms-tnef"),
                 detector.detect(stream, new Metadata()));
     } finally {
         stream.close();
     }
   }
View Full Code Here
TOP

Related Classes of org.apache.tika.detect.Detector

  • com.nidhinova.tika.server.TikaService
  • org.apache.tika.bundle.BundleIT
  • org.apache.tika.config.TikaConfig
  • org.apache.tika.example.DumpTikaConfigExample
  • org.apache.tika.parser.fork.ForkParserIntegrationTest
  • org.apache.tika.parser.microsoft.ExcelParserTest
  • org.apache.tika.parser.microsoft.TNEFParserTest
  • org.apache.tika.parser.rtf.RTFEmbObjHandler
  • org.apache.tika.server.TikaResource
  • org.jasig.portal.io.xml.JaxbPortalDataHandlerService
    • 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.