Null cell equivalent to blank cell consideration improvised diff

Author: abhisheksarkar30

.gitignore

@@ -78,4 +78,6 @@ local.properties
 
 # Custom
 /target/
-*.xls*
+*.xls*
+.idea
+*.iml

apps/excel-diff-checker.jar

@@ -116,7 +116,7 @@ private static List<SheetProcessorTask> processAllSheets(XSSFWorkbook resultWork
 		List<SheetProcessorTask> tasks = new ArrayList<>();
 
 		Consumer<Sheet> consumer = sheet1 -> {
-			XSSFSheet sheet2 = (XSSFSheet) workbook2.getSheet(sheet1.getSheetName());
+			XSSFSheet sheet2 = workbook2.getSheet(sheet1.getSheetName());
 
 			if(sheet2 == null) {
 				System.out.println(String.format("Sheet[%s] doesn't exist in workbook[%s]", sheet1.getSheetName(), FILE_NAME2));
@@ -135,13 +135,13 @@ private static List<SheetProcessorTask> processSpecifiedSheets(XSSFWorkbook resu
 		List<String> sheets = Arrays.asList(specifiedSheets.split(","));
 
 		Consumer<String> consumer = sheetName -> {
-			XSSFSheet sheet1 = (XSSFSheet) resultWorkbook.getSheet(sheetName);
-			XSSFSheet sheet2 = (XSSFSheet) workbook2.getSheet(sheetName);
+			XSSFSheet sheet1 = resultWorkbook.getSheet(sheetName);
+			XSSFSheet sheet2 = workbook2.getSheet(sheetName);
 
 			if(sheet1 == null || sheet2 == null ) {
 				System.out.println(String.format("Sheet[%s] doesn't exist in both workbooks", sheetName));
 			} else
-				tasks.add(new SheetProcessorTask((XSSFSheet) sheet1, sheet2, remarksOnly));
+				tasks.add(new SheetProcessorTask(sheet1, sheet2, remarksOnly));
 		};
 
 		sheets.forEach(consumer);

src/main/java/edu/abhi/poi/excel/ExcelDiffChecker.java

@@ -39,8 +39,8 @@ public CallableValue call() {
 
 	private void processAllRows() throws Exception {
 		for (int rowIndex = 0; rowIndex <= sheet1.getLastRowNum(); rowIndex++) {
-			XSSFRow row1 = (XSSFRow) sheet1.getRow(rowIndex);
-			XSSFRow row2 = (XSSFRow) sheet2.getRow(rowIndex);
+			XSSFRow row1 = sheet1.getRow(rowIndex);
+			XSSFRow row2 = sheet2.getRow(rowIndex);
 
 			if (row1 == null || row2 == null) {
 				if (!(row1 == null && row2 == null)) {
@@ -59,28 +59,30 @@ private void processAllColumns(XSSFRow row1, XSSFRow row2) throws Exception {
 		boolean isRow1Blank = true, isRow2Blank = true;
 
 		for (int columnIndex = 0; columnIndex <= row1.getLastCellNum(); columnIndex++) {
-			XSSFCell cell1 = (XSSFCell) row1.getCell(columnIndex);
-			XSSFCell cell2 = (XSSFCell) row2.getCell(columnIndex);
+			XSSFCell cell1 = row1.getCell(columnIndex);
+			XSSFCell cell2 = row2.getCell(columnIndex);
+			String cell1Value = Utility.getCellValue(cell1);
+			String cell2Value = Utility.getCellValue(cell2);
 
-			if (Utility.hasNoContent(cell1)) {
-				if (Utility.hasContent(cell2)) {
+			if (cell1Value.isEmpty()) {
+				if (!cell2Value.isEmpty()) {
 					isRow2Blank = false;
 					crt.setDiffFlag(true);
 					Utility.processDiffForColumn(cell1 == null ? row1.createCell(columnIndex) : cell1, remarksOnly,
-							Utility.getCellValue(cell2), rowRemarks);
+							cell2Value, rowRemarks);
 				}
-			} else if (Utility.hasNoContent(cell2)) {
-				if (Utility.hasContent(cell1)) {
+			} else if (cell2Value.isEmpty()) {
+				if (!cell1Value.isEmpty()) {
 					isRow1Blank = false;
 					crt.setDiffFlag(true);
-					Utility.processDiffForColumn(cell1, remarksOnly, cell2 == null? null : Utility.getCellValue(cell2), rowRemarks);
+					Utility.processDiffForColumn(cell1, remarksOnly, cell2Value, rowRemarks);
 				}
 			} else {
 				isRow1Blank = isRow2Blank = false;
 
-				if (!Utility.getCellValue(cell1).equals(Utility.getCellValue(cell2))) {
+				if (!cell1Value.equals(cell2Value)) {
 					crt.setDiffFlag(true);
-					Utility.processDiffForColumn(cell1, remarksOnly, Utility.getCellValue(cell2), rowRemarks);
+					Utility.processDiffForColumn(cell1, remarksOnly, cell2Value, rowRemarks);
 				}
 			}
 		}
@@ -97,24 +99,24 @@ else if(isRow1Blank && !isRow2Blank)
 
 	public void processNullRow(XSSFSheet sheet1, int rowIndex, XSSFRow row2) throws Exception {
 		XSSFRow row1 = sheet1.getRow(rowIndex);
-		StringBuilder rowRemarks = new StringBuilder();
+//		StringBuilder rowRemarks = new StringBuilder();
 
 		if (row1 == null) {
 			if (row2.getPhysicalNumberOfCells() != 0) {
 				row1 = sheet1.createRow(rowIndex);
 				crt.setDiffFlag(true);
-				for (int columnIndex = 0; columnIndex <= row2.getLastCellNum(); columnIndex++) {
-					Utility.processDiffForColumn(row1.createCell(0), remarksOnly,
-							Utility.getCellValue(row2.getCell(columnIndex)), rowRemarks);
-				}
+//				for (int columnIndex = 0; columnIndex <= row2.getLastCellNum(); columnIndex++) {
+//					Utility.processDiffForColumn(row1.createCell(0), remarksOnly,
+//							Utility.getCellValue(row2.getCell(columnIndex)), rowRemarks);
+//				}
 				crt.getDiffContainer().append(String.format("\nAdded Row[%s] in Sheet[%s]",
 						(row1.getRowNum() + 1), sheet1.getSheetName()));
 			}
 		} else {
 			if (row1.getPhysicalNumberOfCells() != 0) {
 				crt.setDiffFlag(true);
-				XSSFCell cell1 = row1.getCell(0);
-				Utility.processDiffForColumn(cell1 == null ? row1.createCell(0) : cell1, remarksOnly, "Null row", rowRemarks);
+//				XSSFCell cell1 = row1.getCell(0);
+//				Utility.processDiffForColumn(cell1 == null ? row1.createCell(0) : cell1, remarksOnly, "Null row", rowRemarks);
 				crt.getDiffContainer().append(String.format("\nRemoved Row[%s] of Sheet[%s]",
 						(row1.getRowNum() + 1), sheet1.getSheetName()));
 			}

src/main/java/edu/abhi/poi/excel/SheetProcessorTask.java

@@ -20,14 +20,6 @@
  */
 public class Utility {
 
-	public static boolean hasNoContent(XSSFCell cell) {
-		return cell == null || cell.getRawValue() == null || cell.getRawValue().equals("");
-	}
-
-	public static boolean hasContent(XSSFCell cell) {
-		return cell != null && cell.getRawValue() != null && !cell.getRawValue().equals("");
-	}
-
 	@SuppressWarnings("rawtypes")
 	public static void processDiffForColumn(XSSFCell cell1, boolean remarksOnly, String note, StringBuilder sb) throws Exception {
 
@@ -57,7 +49,7 @@ public static void processDiffForColumn(XSSFCell cell1, boolean remarksOnly, Str
 			Comment comment = drawing.createCellComment(anchor);
 
 			//set the comment text and author
-			comment.setString(factory.createRichTextString("Found " + note));
+			comment.setString(factory.createRichTextString("Found: " + note));
 			comment.setAuthor("SYSTEM");
 
 			cell1.setCellComment(comment);
@@ -67,27 +59,23 @@ public static void processDiffForColumn(XSSFCell cell1, boolean remarksOnly, Str
 	public static String getCellValue(XSSFCell cell) throws Exception {
 		String content = "";
 
+		if(cell == null) return content;
+
 		CellType cellType = cell.getCellType();
 
 		if(cellType == CellType.FORMULA)
 			cellType = cell.getCachedFormulaResultType();
 
 		switch(cellType) {
-		case BLANK:	content += null;
-		break;
-		case BOOLEAN: content += cell.getBooleanCellValue();
-		break;
-		case ERROR: content += cell.getErrorCellString();
-		break;
-		case STRING: content += cell.getRichStringCellValue();
-		break;
-		case NUMERIC: content += DateUtil.isCellDateFormatted(cell) ? cell.getDateCellValue() : 
-			cell.getNumericCellValue();
-		break;
-		case _NONE:	content += null;
-		break;
-		default: throw new Exception(String.format("Unexpected Cell[%s] Type[%s] of Sheet[%s]", cell.getReference(), 
-				cell.getCellType(), cell.getSheet().getSheetName()));
+		case BLANK:
+		case _NONE: break;
+		case BOOLEAN: content += cell.getBooleanCellValue(); break;
+		case ERROR: content += trimToEmpty(cell.getErrorCellString()); break;
+		case STRING: content += trimToEmpty(cell.getRichStringCellValue().getString()); break;
+		case NUMERIC: content += DateUtil.isCellDateFormatted(cell) ? cell.getDateCellValue() :
+				!content.equals(cell.getRawValue())? cell.getNumericCellValue() : content; break;
+		default: throw new Exception(String.format("Unexpected Cell[%s] Type[%s] of Sheet[%s]", cell.getReference(),
+			cell.getCellType(), cell.getSheet().getSheetName()));
 		}
 		return content;
 	}
@@ -99,4 +87,8 @@ public static boolean deleteIfExists(File file) {
 		return false;		
 	}
 
+	private static String trimToEmpty(String value) {
+		return value == null? "" : value.trim();
+	}
+
 }