Update hssf.extractor.ExcelExtractor to optionally output blank cells too

git-svn-id: https://svn.apache.org/repos/asf/poi/trunk@697589 13f79535-47bb-0310-9956-ffa450edef68

src/documentation/content/xdocs/changes.xml

@@ -37,6 +37,7 @@
 
 		<!-- Don't forget to update status.xml too! -->
         <release version="3.2-alpha1" date="2008-??-??">
+           <action dev="POI-DEVELOPERS" type="add">Update hssf.extractor.ExcelExtractor to optionally output blank cells too</action>
            <action dev="POI-DEVELOPERS" type="add">Include the sheet name in the output of examples.XLS2CSVmra</action>
            <action dev="POI-DEVELOPERS" type="fix">45784 - Support long chart titles in SeriesTextRecords</action>
            <action dev="POI-DEVELOPERS" type="fix">45777 - Throw an exception if HSSF Footer or Header is attemped to be set too long, rather than having it break during writing out</action>

src/documentation/content/xdocs/status.xml

@@ -34,6 +34,7 @@
 	<!-- Don't forget to update changes.xml too! -->
     <changes>
         <release version="3.2-alpha1" date="2008-??-??">
+           <action dev="POI-DEVELOPERS" type="add">Update hssf.extractor.ExcelExtractor to optionally output blank cells too</action>
            <action dev="POI-DEVELOPERS" type="add">Include the sheet name in the output of examples.XLS2CSVmra</action>
            <action dev="POI-DEVELOPERS" type="fix">45784 - Support long chart titles in SeriesTextRecords</action>
            <action dev="POI-DEVELOPERS" type="fix">45777 - Throw an exception if HSSF Footer or Header is attemped to be set too long, rather than having it break during writing out</action>

src/java/org/apache/poi/hssf/extractor/ExcelExtractor.java

@@ -44,6 +44,7 @@ public class ExcelExtractor extends POIOLE2TextExtractor {
 	private boolean includeSheetNames = true;
 	private boolean formulasNotResults = false;
 	private boolean includeCellComments = false;
+	private boolean includeBlankCells = false;
 	
 	public ExcelExtractor(HSSFWorkbook wb) {
 		super(wb);
@@ -73,6 +74,14 @@ public class ExcelExtractor extends POIOLE2TextExtractor {
     public void setIncludeCellComments(boolean includeCellComments) {
         this.includeCellComments = includeCellComments;
     }
+	/**
+	 * Should blank cells be output? Default is to only
+	 *  output cells that are present in the file and are
+	 *  non-blank.
+	 */
+	public void setIncludeBlankCells(boolean includeBlankCells) {
+		this.includeBlankCells = includeBlankCells;
+	}
 	
 	/**
 	 * Retreives the text contents of the file
@@ -80,6 +89,11 @@ public class ExcelExtractor extends POIOLE2TextExtractor {
 	public String getText() {
 		StringBuffer text = new StringBuffer();
 
+		// We don't care about the differnce between
+		//  null (missing) and blank cells
+		wb.setMissingCellPolicy(HSSFRow.RETURN_BLANK_AS_NULL);
+		
+		// Process each sheet in turn
 		for(int i=0;i<wb.getNumberOfSheets();i++) {
 			HSSFSheet sheet = wb.getSheetAt(i);
 			if(sheet == null) { continue; }
@@ -108,15 +122,19 @@ public class ExcelExtractor extends POIOLE2TextExtractor {
 				// Check each cell in turn
 				int firstCell = row.getFirstCellNum();
 				int lastCell = row.getLastCellNum();
+				if(includeBlankCells) {
+					firstCell = 0;
+				}
+				
 				for(int k=firstCell;k<lastCell;k++) {
 					HSSFCell cell = row.getCell(k);
-					if(cell == null) { continue; }
 					boolean outputContents = true;
 
+					if(cell == null) {
+						// Only output if requested
+						outputContents = includeBlankCells;
+					} else {
 						switch(cell.getCellType()) {
-						case HSSFCell.CELL_TYPE_BLANK:
-							outputContents = false;
-							break;
 							case HSSFCell.CELL_TYPE_STRING:
 								text.append(cell.getRichStringCellValue().getString());
 								break;
@@ -166,6 +184,7 @@ public class ExcelExtractor extends POIOLE2TextExtractor {
 						    String commentText = comment.getString().getString().replace('\n', ' ');
 						    text.append(" Comment by "+comment.getAuthor()+": "+commentText);
 						}
+					}
 					
 					// Output a tab if we're not on the last cell
 					if(outputContents && k < (lastCell-1)) {

src/testcases/org/apache/poi/hssf/extractor/TestExcelExtractor.java

@@ -187,6 +187,27 @@ public final class TestExcelExtractor extends TestCase {
 		);
 	}
 	
+	public void testWithBlank() throws Exception {
+		ExcelExtractor extractor = createExtractor("MissingBits.xls");
+		String def = extractor.getText();
+		extractor.setIncludeBlankCells(true);
+		String padded = extractor.getText();
+		
+		assertTrue(def.startsWith(
+				"Sheet1\n" +
+				"&[TAB]\t\n" +
+				"Hello\t\n" +
+				"11.0\t23.0\t\n"
+		));
+		
+		assertTrue(padded.startsWith(
+				"Sheet1\n" +
+				"&[TAB]\t\n" +
+				"Hello\t\t\t\t\t\t\t\t\t\t\t\n" +
+				"11.0\t\t\t23.0\t\t\t\t\t\t\t\t\n"
+		));
+	}
+	
 	
 	/**
 	 * Embded in a non-excel file