Package org.exoplatform.services.document

Examples of org.exoplatform.services.document.DocumentReader

74
75
76
77
78
79
80
81
82
83
84
85
   public void testPDFDocumentReaderServiceXMPMetadataTikasFile() throws Exception
   {
      InputStream is = TestPropertiesExtracting.class.getResourceAsStream("/tikaTestPDF.pdf");
      try
      {
         DocumentReader rdr = service.getDocumentReader("application/pdf");
         Properties testprops = rdr.getProperties(is);
         Properties etalon = new Properties();
         etalon.put(DCMetaData.TITLE, "Document title");
         etalon.put(DCMetaData.CREATOR, "Document author");
         evalProps(etalon, testprops);
      }
View Full Code Here
48
49
50
51
52
53
54
55
56
57
58
59
      InputStream is = TestHtmlDocumentReader.class.getResourceAsStream("/test.html");
      try
      {
         String mimeType = mimetypeResolver.getMimeType("test.html");

         DocumentReader dr = service.getDocumentReader(mimeType);
         String text = dr.getContentAsText(is);
         assertTrue(text.contains("This is the third maintenance release of the redesigned 2.0"));
      }
      finally
      {
         is.close();
View Full Code Here
64
65
66
67
68
69
70
71
72
73
74
   {

      InputStream is = TestHtmlDocumentReader.class.getResourceAsStream("/ch-core.html");
      String mimeType = mimetypeResolver.getMimeType("ch-core.html");

      DocumentReader dr = service.getDocumentReader(mimeType);
      String text = dr.getContentAsText(is);

      assertTrue((normalizeWhitespaces(text))
         .contains("The eXo Core is a set of common services that are used by eXo products and modules, it also can be used in the business logic. It's Authentication and Security, Organization, Database, Logging, JNDI, LDAP, Document reader and other services."));
   }
View Full Code Here
47
48
49
50
51
52
53
54
55
56
57
58
      InputStream is = TestHtmlDocumentReader.class.getResourceAsStream("/test.html");
      try
      {
         String mimeType = mimetypeResolver.getMimeType("test.html");

         DocumentReader dr = service.getDocumentReader(mimeType);
         String text = dr.getContentAsText(is);
         assertTrue(text.contains("This is the third maintenance release of the redesigned 2.0"));
      }
      finally
      {
         is.close();
View Full Code Here
62
63
64
65
66
67
68
69
70
71
72
73
   public void testXHTMLGetContentAsString() throws Exception
   {
      InputStream is = TestHtmlDocumentReader.class.getResourceAsStream("/testXHTML.html");
      try
      {
         DocumentReader dr = service.getDocumentReader("application/xhtml+xml");
         String text = dr.getContentAsText(is);
         assertTrue(text
            .contains("This document tests the ability of Apache Tika to extract content from an XHTML document."));
      }
      finally
      {
View Full Code Here
69
70
71
72
73
74
75
76
77
78
79
80
   public void testPDFDocumentReaderServiceXMPMetadata() throws Exception
   {
      InputStream is = TestPropertiesExtracting.class.getResourceAsStream("/MyTest.pdf");
      try
      {
         DocumentReader rdr = service.getDocumentReader("application/pdf");
         Properties testprops = rdr.getProperties(is);
         Properties etalon = new Properties();
         etalon.put(DCMetaData.TITLE, "Test de convertion de fichier tif");
         etalon.put(DCMetaData.CREATOR, "Christian Klaus");
         etalon.put(DCMetaData.DESCRIPTION, "20080901 TEST Christian Etat OK");
         //         Calendar c = ISO8601.parseEx("2008-09-01T08:01:10+00:00");
View Full Code Here
91
92
93
94
95
96
97
98
99
100
101
102
   {
      InputStream is = TestPropertiesExtracting.class.getResourceAsStream("/pfs_accapp.pdf");
      try
      {

         DocumentReader rdr = service.getDocumentReader("application/pdf");
         Properties testprops = rdr.getProperties(is);
         Properties etalon = new Properties();
         etalon.put(DCMetaData.TITLE, "Personal Account Opening Form VN");
         etalon.put(DCMetaData.CREATOR, "mr");
         etalon.put(DCMetaData.PUBLISHER, "Adobe LiveCycle Designer ES 8.2");
         evalProps(etalon, testprops, false);
View Full Code Here
110
111
112
113
114
115
116
117
118
119
120
121
   public void testPDFDocumentReaderServiceXMPUsecase1() throws Exception
   {
      InputStream is = TestPropertiesExtracting.class.getResourceAsStream("/Trait_union.06.Mai_2009.pdf");
      try
      {
         DocumentReader rdr = service.getDocumentReader("application/pdf");
         Properties testprops = rdr.getProperties(is);
         Properties etalon = new Properties();
         etalon.put(DCMetaData.TITLE, "journal interne mai 2009.qxp");
         etalon.put(DCMetaData.CREATOR, "presse");
         evalProps(etalon, testprops, false);
      }
View Full Code Here
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
                     Constants.JCR_DATA, 0), ItemType.PROPERTY));

               // index if have jcr:mimeType sibling for this binary property only
               try
               {
                  DocumentReader dreader =
                     extractor.getDocumentReader(new String(pmime.getValues().get(0).getAsByteArray()));

                  data = propData.getValues();

                  if (data == null)
                     log.warn("null value found at property " + prop.getQPath().getAsString());

                  // check the jcr:encoding property
                  PropertyData encProp =
                     (PropertyData)stateProvider.getItemData(node, new QPathEntry(Constants.JCR_ENCODING, 0),
                        ItemType.PROPERTY);

                  if (encProp != null)
                  {
                     // encoding parameter used
                     String encoding = new String(encProp.getValues().get(0).getAsByteArray());
                     for (ValueData pvd : data)
                     {
                        InputStream is = null;
                        try
                        {
                           is = pvd.getAsStream();
                           Reader reader = new StringReader(dreader.getContentAsText(is, encoding));
                           doc.add(createFulltextField(reader));

                        }
                        finally
                        {
                           try
                           {
                              is.close();
                           }
                           catch (Throwable e)
                           {
                           }
                        }
                     }
                  }
                  else
                  {
                     // no encoding parameter
                     for (ValueData pvd : data)
                     {
                        InputStream is = null;
                        try
                        {
                           is = pvd.getAsStream();
                           Reader reader = new StringReader(dreader.getContentAsText(is));
                           doc.add(createFulltextField(reader));
                        }
                        finally
                        {
                           try
View Full Code Here
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
                     Constants.JCR_DATA, 0)));

               // index if have jcr:mimeType sibling for this binary property only
               try
               {
                  DocumentReader dreader =
                     extractor.getDocumentReader(new String(pmime.getValues().get(0).getAsByteArray()));

                  data = propData.getValues();

                  if (data == null)
                     log.warn("null value found at property " + prop.getQPath().getAsString());

                  // check the jcr:encoding property
                  PropertyData encProp =
                     (PropertyData)stateProvider.getItemData(node, new QPathEntry(Constants.JCR_ENCODING, 0));

                  if (encProp != null)
                  {
                     // encoding parameter used
                     String encoding = new String(encProp.getValues().get(0).getAsByteArray());
                     for (ValueData pvd : data)
                     {
                        InputStream is = null;
                        try
                        {
                           is = pvd.getAsStream();
                           Reader reader = new StringReader(dreader.getContentAsText(is, encoding));
                           doc.add(createFulltextField(reader));

                        }
                        finally
                        {
                           try
                           {
                              is.close();
                           }
                           catch (Throwable e)
                           {
                           }
                        }
                     }
                  }
                  else
                  {
                     // no encoding parameter
                     for (ValueData pvd : data)
                     {
                        InputStream is = null;
                        try
                        {
                           is = pvd.getAsStream();
                           Reader reader = new StringReader(dreader.getContentAsText(is));
                           doc.add(createFulltextField(reader));
                        }
                        finally
                        {
                           try
View Full Code Here
TOP

Related Classes of org.exoplatform.services.document.DocumentReader

  • org.exoplatform.services.document.impl.DocumentReaderServiceImpl
  • org.exoplatform.services.document.impl.tika.TikaDocumentReaderServiceImpl
  • org.exoplatform.services.document.test.TestHtmlDocumentReader
  • org.exoplatform.services.document.test.TestPropertiesExtracting
  • org.exoplatform.services.document.tika.TestHtmlOnTikaDocumentReader
  • org.exoplatform.services.document.tika.TestPropertiesExtractionOnTika
  • org.exoplatform.services.jcr.impl.core.query.lucene.NodeIndexer
  • org.exoplatform.services.jcr.impl.core.query.TestExcelFileSearch
    • 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.