src/scratchpad/ooxml-src/org/apache/poi/hssf/extractor/HXFExcelExtractor.java

   1 /* ====================================================================
   2    Licensed to the Apache Software Foundation (ASF) under one or more
   3    contributor license agreements.  See the NOTICE file distributed with
   4    this work for additional information regarding copyright ownership.
   5    The ASF licenses this file to You under the Apache License, Version 2.0
   6    (the "License"); you may not use this file except in compliance with
   7    the License.  You may obtain a copy of the License at
   8
   9        http://www.apache.org/licenses/LICENSE-2.0
  10
  11    Unless required by applicable law or agreed to in writing, software
  12    distributed under the License is distributed on an "AS IS" BASIS,
  13    WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  14    See the License for the specific language governing permissions and
  15    limitations under the License.
  16 ==================================================================== */
  17 package org.apache.poi.hssf.extractor;
  18
  19 import java.io.File;
  20 import java.io.IOException;
  21
  22 import org.apache.poi.POIXMLTextExtractor;
  23 import org.apache.poi.hssf.HSSFXML;
  24 import org.apache.poi.hssf.usermodel.HSSFXMLCell;
  25 import org.apache.poi.hssf.usermodel.HSSFXMLWorkbook;
  26 import org.apache.poi.hxf.HXFDocument;
  27 import org.apache.xmlbeans.XmlException;
  28 import org.openxml4j.exceptions.OpenXML4JException;
  29 import org.openxml4j.opc.Package;
  30 import org.openxmlformats.schemas.spreadsheetml.x2006.main.CTCell;
  31 import org.openxmlformats.schemas.spreadsheetml.x2006.main.CTRow;
  32 import org.openxmlformats.schemas.spreadsheetml.x2006.main.CTSheet;
  33 import org.openxmlformats.schemas.spreadsheetml.x2006.main.CTWorksheet;
  34
  35 /**
  36  * Helper class to extract text from an OOXML Excel file
  37  */
  38 public class HXFExcelExtractor extends POIXMLTextExtractor {
  39         private HSSFXMLWorkbook workbook;
  40         private boolean includeSheetNames = true;
  41         private boolean formulasNotResults = false;
  42
  43         public HXFExcelExtractor(Package container) throws XmlException, OpenXML4JException, IOException {
  44                 this(new HSSFXMLWorkbook(
  45                                 new HSSFXML(container)
  46                 ));
  47         }
  48         public HXFExcelExtractor(HSSFXMLWorkbook workbook) {
  49                 super(workbook);
  50                 this.workbook = workbook;
  51         }
  52
  53         public static void main(String[] args) throws Exception {
  54                 if(args.length < 1) {
  55                         System.err.println("Use:");
  56                         System.err.println("  HXFExcelExtractor <filename.xlsx>");
  57                         System.exit(1);
  58                 }
  59                 POIXMLTextExtractor extractor =
  60                         new HXFExcelExtractor(HXFDocument.openPackage(
  61                                         new File(args[0])
  62                         ));
  63                 System.out.println(extractor.getText());
  64         }
  65
  66         /**
  67          * Should sheet names be included? Default is true
  68          */
  69         public void setIncludeSheetNames(boolean includeSheetNames) {
  70                 this.includeSheetNames = includeSheetNames;
  71         }
  72         /**
  73          * Should we return the formula itself, and not
  74          *  the result it produces? Default is false
  75          */
  76         public void setFormulasNotResults(boolean formulasNotResults) {
  77                 this.formulasNotResults = formulasNotResults;
  78         }
  79
  80         /**
  81          * Retreives the text contents of the file
  82          */
  83         public String getText() {
  84                 StringBuffer text = new StringBuffer();
  85
  86                 CTSheet[] sheetRefs =
  87                         workbook._getHSSFXML().getSheetReferences().getSheetArray();
  88                 for(int i=0; i<sheetRefs.length; i++) {
  89                         try {
  90                                 CTWorksheet sheet =
  91                                         workbook._getHSSFXML().getSheet(sheetRefs[i]);
  92                                 CTRow[] rows =
  93                                         sheet.getSheetData().getRowArray();
  94
  95                                 if(i > 0) {
  96                                         text.append("\n");
  97                                 }
  98                                 if(includeSheetNames) {
  99                                         text.append(sheetRefs[i].getName() + "\n");
 100                                 }
 101
 102                                 for(int j=0; j<rows.length; j++) {
 103                                         CTCell[] cells = rows[j].getCArray();
 104                                         for(int k=0; k<cells.length; k++) {
 105                                                 CTCell cell = cells[k];
 106                                                 if(k > 0) {
 107                                                         text.append("\t");
 108                                                 }
 109
 110                                                 boolean done = false;
 111
 112                                                 // Is it a formula one?
 113                                                 if(cell.getF() != null) {
 114                                                         if(formulasNotResults) {
 115                                                                 text.append(cell.getF().getStringValue());
 116                                                                 done = true;
 117                                                         }
 118                                                 }
 119                                                 if(!done) {
 120                                                         HSSFXMLCell uCell = new HSSFXMLCell(cell, workbook);
 121                                                         text.append(uCell.getStringValue());
 122                                                 }
 123                                         }
 124                                         text.append("\n");
 125                                 }
 126                         } catch(Exception e) {
 127                                 throw new RuntimeException(e);
 128                         }
 129                 }
 130
 131                 return text.toString();
 132         }
 133 }