Package org.apache.poi.hssf.extractor

Examples of org.apache.poi.hssf.extractor.ExcelExtractor

116
117
118
119
120
121
122
123
124
125
126
    //  out from
    for(Iterator entries = poifsDir.getEntries(); entries.hasNext(); ) {
      Entry entry = (Entry)entries.next();
     
      if(entry.getName().equals("Workbook")) {
        return new ExcelExtractor(poifsDir, fs);
      }
      if(entry.getName().equals("WordDocument")) {
        return new WordExtractor(poifsDir, fs);
      }
      if(entry.getName().equals("PowerPoint Document")) {
View Full Code Here
125
126
127
128
129
130
131
132
133
134
135
    //  out from
    for(Iterator entries = poifsDir.getEntries(); entries.hasNext(); ) {
      Entry entry = (Entry)entries.next();
     
      if(entry.getName().equals("Workbook")) {
        return new ExcelExtractor(poifsDir, fs);
      }
      if(entry.getName().equals("WordDocument")) {
        return new WordExtractor(poifsDir, fs);
      }
      if(entry.getName().equals("PowerPoint Document")) {
View Full Code Here
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
          assertEquals("Top Right Cell", s.getRow(0).getCell(255).getStringCellValue());
          assertEquals("Bottom Left Cell", s.getRow(65535).getCell(0).getStringCellValue());
          assertEquals("Bottom Right Cell", s.getRow(65535).getCell(255).getStringCellValue());
         
          // Extract and check
          ExcelExtractor ex = new ExcelExtractor(wb);
          String text = ex.getText();
          assertTrue(text.contains("Top Left Cell"));
          assertTrue(text.contains("Top Right Cell"));
          assertTrue(text.contains("Bottom Left Cell"));
          assertTrue(text.contains("Bottom Right Cell"));
       }
View Full Code Here
18
19
20
21
22
23
24
25
26
27
28
29
 
  public IndexDocument getIndexedDocument(File2Index fileData)
      throws SolrException {
    try {
      POIFSFileSystem fs = new POIFSFileSystem(new ByteArrayInputStream(fileData.data));
      ExcelExtractor extractor = new ExcelExtractor(fs);
      String excelText = extractor.getText();

      return new IndexDocument(fileData.path, excelText, null);
    } catch (IOException e) {
      String msg = "Failed to write to the index";
      log.error(msg, e);
View Full Code Here
251
252
253
254
255
256
257
258
259
260
261
262
                resourceURL = new URL(url);
                is = resourceURL.openStream();
            }

            POIFSFileSystem fs = new POIFSFileSystem(is);
            ExcelExtractor extractor = new ExcelExtractor(fs);
            String excelText = extractor.getText();

            Document document = new Document();
            document.add(new Field("id", id, Field.Store.YES, Field.Index.TOKENIZED));
            document.add(
                    new Field("content", excelText, Field.Store.NO, Field.Index.TOKENIZED));
View Full Code Here
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
          assertEquals("Top Right Cell", s.getRow(0).getCell(255).getStringCellValue());
          assertEquals("Bottom Left Cell", s.getRow(65535).getCell(0).getStringCellValue());
          assertEquals("Bottom Right Cell", s.getRow(65535).getCell(255).getStringCellValue());
         
          // Extract and check
          ExcelExtractor ex = new ExcelExtractor(wb);
          String text = ex.getText();
          assertTrue(text.contains("Top Left Cell"));
          assertTrue(text.contains("Top Right Cell"));
          assertTrue(text.contains("Bottom Left Cell"));
          assertTrue(text.contains("Bottom Right Cell"));
       }
View Full Code Here
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
          assertEquals("Top Right Cell", s.getRow(0).getCell(255).getStringCellValue());
          assertEquals("Bottom Left Cell", s.getRow(65535).getCell(0).getStringCellValue());
          assertEquals("Bottom Right Cell", s.getRow(65535).getCell(255).getStringCellValue());
         
          // Extract and check
          ExcelExtractor ex = new ExcelExtractor(wb);
          String text = ex.getText();
          assertTrue(text.contains("Top Left Cell"));
          assertTrue(text.contains("Top Right Cell"));
          assertTrue(text.contains("Bottom Left Cell"));
          assertTrue(text.contains("Bottom Right Cell"));
          ex.close();
       }
    }
View Full Code Here
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
   */
  public void testComparedToOLE2() throws Exception {
    // A fairly simple file - ooxml
     XSSFEventBasedExcelExtractor ooxmlExtractor = getExtractor("SampleSS.xlsx");

    ExcelExtractor ole2Extractor =
      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());     
    }
   
    ole2Extractor.close();
    ooxmlExtractor.close();
  }
View Full Code Here
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
   */
  public void testComparedToOLE2() throws IOException {
    // A fairly simple file - ooxml
    XSSFExcelExtractor ooxmlExtractor = getExtractor("SampleSS.xlsx");

    ExcelExtractor ole2Extractor =
      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());     
    }

    ole2Extractor.close();
    ooxmlExtractor.close();
  }
View Full Code Here
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
          assertEquals("Top Right Cell", s.getRow(0).getCell(255).getStringCellValue());
          assertEquals("Bottom Left Cell", s.getRow(65535).getCell(0).getStringCellValue());
          assertEquals("Bottom Right Cell", s.getRow(65535).getCell(255).getStringCellValue());
         
          // Extract and check
          ExcelExtractor ex = new ExcelExtractor(wb);
          String text = ex.getText();
          assertTrue(text.contains("Top Left Cell"));
          assertTrue(text.contains("Top Right Cell"));
          assertTrue(text.contains("Bottom Left Cell"));
          assertTrue(text.contains("Bottom Right Cell"));
          ex.close();
       }
    }
View Full Code Here
TOP

Related Classes of org.apache.poi.hssf.extractor.ExcelExtractor

  • org.apache.jackrabbit.extractor.MsExcelTextExtractor
  • org.apache.poi.extractor.ExtractorFactory
  • org.apache.poi.hpsf.extractor.TestHPSFPropertiesExtractor
  • org.apache.poi.hssf.usermodel.HSSFCell
  • org.apache.poi.hssf.usermodel.HSSFCellStyle
  • org.apache.poi.hssf.usermodel.HSSFComment
  • org.apache.poi.hssf.usermodel.HSSFDataFormatter
  • org.apache.poi.hssf.usermodel.HSSFRichTextString
  • org.apache.poi.hssf.usermodel.HSSFRow
  • org.apache.poi.hssf.usermodel.HSSFSheet
    • 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.