Package org.apache.poi

Examples of org.apache.poi.POITextExtractor

86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
    InputStream stream = null;

    try {
      stream = rawDocument.getContentAsStream();
      POITextExtractor contentExtractor = ExtractorFactory.createExtractor(stream);
      setCleanedContent(contentExtractor.getText());
      POITextExtractor metadataExtractor = contentExtractor.getMetadataTextExtractor();

      Map<String, String> metaDataMap = createMetaDataMap(metadataExtractor.getText());
//      if (mLog.isDebugEnabled()) {
//      mLog.info("Found meta data ::" + metadataExtractor.getText()
//              + ":: in " + rawDocument.getUrl());
//      }
View Full Code Here
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
      is = new ByteArrayInputStream(str.getBytes());
    } else {
      throw new IllegalArgumentException("Parameter must be instance of byte[]");
    }
    String ret = null;
    POITextExtractor extractor;

    try {
      extractor = ExtractorFactory.createExtractor(is);

      ret = extractor.getText();

    } catch (IOException e) {
      throw new CRException(e);
    } catch (InvalidFormatException e) {
      throw new CRException(e);
View Full Code Here
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
      msgEmbMsg = getFileAndCheck(olTests, "attachment_msg_pdf.msg");
   }

   public void testFile() throws Exception {
      // Excel
      POITextExtractor xlsExtractor = ExtractorFactory.createExtractor(xls);
      assertNotNull("Had empty extractor for " + xls, xlsExtractor);
      assertTrue("Expected instanceof ExcelExtractor, but had: " + xlsExtractor.getClass(),
            xlsExtractor
            instanceof ExcelExtractor
      );
      assertTrue(
            xlsExtractor.getText().length() > 200
      );

      assertTrue(
            ExtractorFactory.createExtractor(xlsx)
            instanceof XSSFExcelExtractor
View Full Code Here
180
181
182
183
184
185
186
187
188
189
190
191
192
      new ExcelExtractor(simpleXLS);
   
    POITextExtractor[] extractors =
      new POITextExtractor[] { ooxmlExtractor, ole2Extractor };
    for (int i = 0; i < extractors.length; i++) {
      POITextExtractor extractor = extractors[i];
     
      String text = extractor.getText().replaceAll("[\r\t]", "");
      //System.out.println(text.length());
      //System.out.println(text);
      assertTrue(text.startsWith("First Sheet\nTest spreadsheet\n2nd row2nd row 2nd column\n"));
      Pattern pattern = Pattern.compile(".*13(\\.0+)?\\s+Sheet3.*", Pattern.DOTALL);
      Matcher m = pattern.matcher(text);
View Full Code Here
158
159
160
161
162
163
164
165
166
167
168
169
170
      new ExcelExtractor(HSSFTestDataSamples.openSampleWorkbook("SampleSS.xls"));
   
    POITextExtractor[] extractors =
      new POITextExtractor[] { ooxmlExtractor, ole2Extractor };
    for (int i = 0; i < extractors.length; i++) {
      POITextExtractor extractor = extractors[i];
     
      String text = extractor.getText().replaceAll("[\r\t]", "");
      assertTrue(text.startsWith("First Sheet\nTest spreadsheet\n2nd row2nd row 2nd column\n"));
      Pattern pattern = Pattern.compile(".*13(\\.0+)?\\s+Sheet3.*", Pattern.DOTALL);
      Matcher m = pattern.matcher(text);
      assertTrue(m.matches());     
    }
View Full Code Here
132
133
134
135
136
137
138
139
140
141
142
143
144
      new ExcelExtractor(HSSFTestDataSamples.openSampleWorkbook("SampleSS.xls"));
   
    POITextExtractor[] extractors =
      new POITextExtractor[] { ooxmlExtractor, ole2Extractor };
    for (int i = 0; i < extractors.length; i++) {
      POITextExtractor extractor = extractors[i];
     
      String text = extractor.getText().replaceAll("[\r\t]", "");
      assertTrue(text.startsWith("First Sheet\nTest spreadsheet\n2nd row2nd row 2nd column\n"));
      Pattern pattern = Pattern.compile(".*13(\\.0+)?\\s+Sheet3.*", Pattern.DOTALL);
      Matcher m = pattern.matcher(text);
      assertTrue(m.matches());     
    }
View Full Code Here
153
154
155
156
157
158
159
160
161
162
163
164
165
      new ExcelExtractor(HSSFTestDataSamples.openSampleWorkbook("SampleSS.xls"));
   
    POITextExtractor[] extractors =
      new POITextExtractor[] { ooxmlExtractor, ole2Extractor };
    for (int i = 0; i < extractors.length; i++) {
      POITextExtractor extractor = extractors[i];
     
      String text = extractor.getText().replaceAll("[\r\t]", "");
      assertTrue(text.startsWith("First Sheet\nTest spreadsheet\n2nd row2nd row 2nd column\n"));
      Pattern pattern = Pattern.compile(".*13(\\.0+)?\\s+Sheet3.*", Pattern.DOTALL);
      Matcher m = pattern.matcher(text);
      assertTrue(m.matches());     
    }
View Full Code Here
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
         System.out.println(DIVIDER);
        
         File f = new File(args[i]);
         System.out.println(f);
        
         POITextExtractor extractor =
            ExtractorFactory.createExtractor(f);
         POITextExtractor metadataExtractor =
            extractor.getMetadataTextExtractor();
        
         System.out.println("   " + DIVIDER);
         System.out.println(metadataExtractor.getText());
         System.out.println("   " + DIVIDER);
         System.out.println(extractor.getText());
         System.out.println(DIVIDER);
      }
   }
View Full Code Here
127
128
129
130
131
132
133
134
135
136
137
138
139
      new ExcelExtractor(HSSFTestDataSamples.openSampleWorkbook("SampleSS.xls"));
   
    POITextExtractor[] extractors =
      new POITextExtractor[] { ooxmlExtractor, ole2Extractor };
    for (int i = 0; i < extractors.length; i++) {
      POITextExtractor extractor = extractors[i];
     
      String text = extractor.getText().replaceAll("[\r\t]", "");
      assertTrue(text.startsWith("First Sheet\nTest spreadsheet\n2nd row2nd row 2nd column\n"));
      Pattern pattern = Pattern.compile(".*13(\\.0+)?\\s+Sheet3.*", Pattern.DOTALL);
      Matcher m = pattern.matcher(text);
      assertTrue(m.matches());     
    }
View Full Code Here
132
133
134
135
136
137
138
139
140
141
142
143
144
      new ExcelExtractor(HSSFTestDataSamples.openSampleWorkbook("SampleSS.xls"));
   
    POITextExtractor[] extractors =
      new POITextExtractor[] { ooxmlExtractor, ole2Extractor };
    for (int i = 0; i < extractors.length; i++) {
      POITextExtractor extractor = extractors[i];
     
      String text = extractor.getText().replaceAll("[\r\t]", "");
      assertTrue(text.startsWith("First Sheet\nTest spreadsheet\n2nd row2nd row 2nd column\n"));
      Pattern pattern = Pattern.compile(".*13(\\.0+)?\\s+Sheet3.*", Pattern.DOTALL);
      Matcher m = pattern.matcher(text);
      assertTrue(m.matches());     
    }
View Full Code Here
TOP

Related Classes of org.apache.poi.POITextExtractor

  • com.gentics.cr.lucene.indexer.transformer.POIContentTransformer
  • net.sf.regain.crawler.preparator.PoiMsOfficePreparator
  • org.apache.poi.extractor.CommandLineTextExtractor
  • org.apache.poi.extractor.TestExtractorFactory
  • org.apache.poi.hssf.extractor.TestHXFExcelExtractor
  • org.apache.poi.xssf.extractor.TestXSSFEventBasedExcelExtractor
  • org.apache.poi.xssf.extractor.TestXSSFExcelExtractor
  • org.dspace.app.mediafilter.PowerPointFilter
    • 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.