We Keep Coding

Sort excel by a column using shiftRows- Apache POI - XmlValueDisconnectedException

I have an XSSFWorkbook with n number of columns. And my requirement is to sort the entire sheet by the first column.
I referred to this link but did not get any information about sorting.
I have also tried the code from here but it gives exception at
sheet.shiftRows(row2.getRowNum(), row2.getRowNum(), -1);
I am using Apache POI 3.17.
Anyone has any suggestion or solution?

There seem to be a bug in POI when shifting columns, they say it was fixed in 3.9 but I used 3.17 and still have it:
Exception in thread "main" org.apache.xmlbeans.impl.values.XmlValueDisconnectedException
at org.apache.xmlbeans.impl.values.XmlObjectBase.check_orphaned(XmlObjectBase.java:1258)
at org.openxmlformats.schemas.spreadsheetml.x2006.main.impl.CTRowImpl.getR(Unknown Source)
at org.apache.poi.xssf.usermodel.XSSFRow.getRowNum(XSSFRow.java:394)
...
I assume it is the same you have. So I worked out an other way:
Sort your rows, then create a new workbook and copy rows in the correct order.
Then write this sorted workbook to the original file.
For simplicity, I assume all cell values are Strings. (if not, then modify accordingly)
private static final String FILE_NAME = "/home/userName/Workspace/fileToSort.xlsx";
public static void main(String[] args) {
Workbook originalWorkbook;
//create a workbook from your file
try(FileInputStream excelFile = new FileInputStream(new File(FILE_NAME))) {
originalWorkbook = new XSSFWorkbook(excelFile);
} catch (IOException e) {
throw new RuntimeException("Couldn't open file: " + FILE_NAME);
}
Sheet originalSheet = originalWorkbook.getSheetAt(0);
// Create a SortedMap<String, Row> where the key is the value of the first column
// This will automatically sort the rows
Map<String, Row> sortedRowsMap = new TreeMap<>();
// save headerRow
Row headerRow = originalSheet.getRow(0);
Iterator<Row> rowIterator = originalSheet.rowIterator();
// skip header row as we saved it already
rowIterator.next();
// sort the remaining rows
while(rowIterator.hasNext()) {
Row row = rowIterator.next();
sortedRowsMap.put(row.getCell(0).getStringCellValue(), row);
}
// Create a new workbook
try(Workbook sortedWorkbook = new XSSFWorkbook();
FileOutputStream out = new FileOutputStream(FILE_NAME)) {
Sheet sortedSheet = sortedWorkbook.createSheet(originalSheet.getSheetName());
// Copy all the sorted rows to the new workbook
// - header first
Row newRow = sortedSheet.createRow(0);
copyRowToRow(headerRow, newRow);
// then other rows, from row 1 up (not row 0)
int rowIndex = 1;
for(Row row : sortedRowsMap.values()) {
newRow = sortedSheet.createRow(rowIndex);
copyRowToRow(row, newRow);
rowIndex++;
}
// Write your new workbook to your file
sortedWorkbook.write(out);
} catch (Exception e) {
e.printStackTrace();
}
}
// Utility method to copy rows
private static void copyRowToRow(Row row, Row newRow) {
Iterator<Cell> cellIterator = row.cellIterator();
int cellIndex = 0;
while(cellIterator.hasNext()) {
Cell cell = cellIterator.next();
Cell newCell = newRow.createCell(cellIndex);
newCell.setCellValue(cell.getStringCellValue());
cellIndex++;
}
}
I tried it out on the following file
A B
---------------
Header1 Header2
a one
c three
d four
b two
and it sorts it this way:
A B
---------------
Header1 Header2
a one
b two
c three
d four

Apache POI writes only one record

I am writing some data into excel file using Apache POI but for some reason the file shows only the last record (1 record only). I have list of POLJO that I am passing. I am also iterating through the cells but all I get is just one record.
Method to write in excel
public void writeToExcel(List<NYProgramTO> to){
try {
Workbook workBook = new HSSFWorkbook();
CreationHelper helper = workBook.getCreationHelper();
Sheet sheet = workBook.createSheet("NY_PPA_P3_Sheet");
Row headerRow = sheet.createRow(0);
headerRow.createCell(0).setCellValue("First Name");
headerRow.createCell(1).setCellValue("Last Name");
headerRow.createCell(2).setCellValue("Policy Number");
headerRow.createCell(3).setCellValue("Zip Code");
headerRow.createCell(4).setCellValue("Date of Birth");
if(to != null){
int size = to.size();
for(int i = 0; i < size; i++){
NYProgramTO nyP= to.get(i);
Row row = sheet.createRow(1);
row.createCell(0).setCellValue(nyP.getFirstName());
row.createCell(1).setCellValue(nyP.getLastName());
row.createCell(2).setCellValue(nyP.getPolicyNumber());
row.createCell(3).setCellValue(nyP.getZipCode());
row.createCell(4).setCellValue(nyP.getDateOfBirth());
}
}
FileOutputStream stream = new FileOutputStream("NY_PPA_P3.xlsx");
workBook.write(stream);
stream.close();
System.out.println("NY_PPA_P3.xlsx created successfully.");
} catch (Exception ex) {
ex.printStackTrace();
}
}

If by "only one record" you mean that only one row is appearing, this is probably easily fixable by making sure that you increment the Row that is being created before writing the Cells.
Try changing:
Row row = sheet.createRow(1);
to:
Row row = sheet.createRow(i+1);

How to retrieve some specific rows and columns from an excel sheet?

I am reading an xlsx file using java (Apache POI).
I have created a Document class (having all excel column heading as variables)
i have to read each row in the excel and map to the Document class by creating a collection of Document class.
The problem I am facing is that I have to start reading from row 2 and from column 7 to column 35 and map the corresponding values to the document class.
Unable to to figure out exactly how the code should be ?
I have written the following lines of code.
List sheetData = new ArrayList();
InputStream excelFile = new BufferedInputStream(new FileInputStream("D:\\Excel file\\data.xlsx"));
Workbook workBook = new XSSFWorkbook(excelFile); // Creates Workbook
XSSFSheet sheet = (XSSFSheet) workBook.getSheet("Daily");
DataFormatter formatter = new DataFormatter();
for (int i = 7; i <= 35; i++) {
XSSFRow row = sheet.getRow(i);
Cell cell = row.getCell(i);
String val = formatter.formatCellValue(cell);
sheetData.add(val);
}

Assuming I've understood your question correctly, I believe you want to process every row which exists from row 2 onwards to the end of the file, and for each of those rows consider the cells in columns 7 through 35. I believe you also might need to process those values, but you haven't said how, so for this example I'll just stuff them in a list of strings and hope for the best...
This is based on the Apache POI documentation for iterating over rows and cells
File excelFile = new File("D:\\Excel file\\data.xlsx");
Workbook workBook = WorkbookFactory.create(excelFile);
Sheet sheet = workBook.getSheet("Daily");
DataFormatter formatter = new DataFormatter();
// Start from the 2nd row, processing all to the end
// Note - Rows and Columns in Apache POI are 0-based not 1-based
for (int rn=1; rn<=sheet.getLastRowNum(); rn++) {
Row row = sheet.getRow(rn);
if (row == null) {
// Whole row is empty. Handle as required here
continue;
}
List<String> values = new ArrayList<String>();
for (int cn=6; cn<35; cn++) {
Cell cell = row.getCell(cn);
String val = null;
if (cell != null) { val = formatter.formatCellValue(cell); }
if (val == null || val.isEmpty()) {
// Cell is empty. Handle as required here
}
// Save the value to list. Save to an object instead if required
values.append(val);
}
}
workBook.close();
Depending on your business requirements, put in logic for handling blank rows and cells. Then, do whatever you need to do with the values you find, again as per your business requirements!

You could iterate with an Iterator in the document, but there is also an function "getRow() and getCell()"
Workbook workbook = new XSSFWorkbook(excelFile);
// defines the standard pointer in document in the first Sheet
XSSFSheet data = this.workbook.getSheetAt(0);
// you could iterate the document with an iterator
Iterator<Cell> iterator = this.data.iterator();
// x/y pointer at the document
Row row = data.getRow(y);
Cell pointingCell = row.getCell(x);
String pointingString = pointingCell.getStringCellValue();

Read and write data to excel using POI - It deletes my original data from excel after write new data

I am first reading data from excel and then write in same excel. It works fine first time. Just after data written , It deletes my original data which was there from very beginning.
code is given below :
public static void main (String args[]) throws Exception {
//CODE TO REMOVE UNNECESSARY WARNING
System.setProperty("org.apache.commons.logging.Log", "org.apache.commons.logging.impl.Jdk14Logger");
//CALL FIREFOX DRIVER TO OPEN IT
WebDriver driver = new FirefoxDriver();
driver.get("https://www.google.co.in/?gfe_rd=cr&ei=VQiAVOeCFavM8gf59IHACg&gws_rd=ssl#q=software+testing");
java.util.List<WebElement> links = driver.findElements(By.tagName("h3"));
int sizecount = links.size();
System.out.println(sizecount);
FileInputStream input = new FileInputStream("D:\\sel.xls");
int count=0;
HSSFWorkbook wb = new HSSFWorkbook(input);
HSSFSheet sh = wb.getSheet("sheet1");
HSSFRow row = sh.getRow(count);
String data = row.getCell(0).toString();
System.out.println(data);
FileOutputStream webdata = new FileOutputStream ("D:\\sel.xls");
int inc = 1;
for(int i=1;i<=links.size()-1;i++)
{
HSSFRow row1 = sh.createRow(count);
row1.createCell(inc).setCellValue(links.get(i).getText());
count++;
}
wb.write(webdata);

if you are updating it, you should also update the count variable with the last line of the excel
int inc = 1;
count = sh.getPhysicalNumberOfRows();
and do the rest...

you are facing the issue because the index of excel sheets are same when you are writing.
First you should check wheater the cell is empty or not then you should write.
Or you can first do getrows() or getcolumn() it will give you the size of the rows and columns.
once you have the size you can write your data after the that row size and column.

Apache POI. Copying sheets

I'm using apache poi to create an excel document. To create new sheet in workbook I write next code:
Workbook wb = new HSSFWorkbook();
Sheet sh = wb.createSheet();
this code create and add sheet to workbook. But I want to create sheet formerly and then add it to workbook. Smth like this:
Sheet sh = new HSSFSheet();
wb.addSheet(sh);
I need such thing, because I want to copy data from one sheet of one workbook to another sheet of another workbook(Workbook interface has method Sheet cloneSheet(int)). But Workbook interface doesn't have method like addSheet(Sheet sh).
Also HSSFWorkbook is final class so I can't extend it to implement add method
How can I do this?

You can't just take a Sheet object from one Workbook, and add it to a different Workbook.
What you'll need to do is to open the old workbook and the new workbooks at the same time, and create the sheet in the new workbook. Next, clone all the styles you used in the old sheet onto the new one (HSSFCellStyle has a method for cloning a style from one workbook to another). Finally, iterate over all the cells and copy them over.

You should use RangeCopier.
XSSFWorkbook workbookFrom = new XSSFWorkbook(new File("/path/to/workbookFrom.xlsx"));
XSSFSheet sheetFrom = workbookFrom.getSheetAt(0);
XSSFWorkbook workbookTo = new XSSFWorkbook(new File("/path/to/workbookTo.xlsx"));
XSSFSheet sheetTo = workbookTo.createSheet("sheet1");
workbookTo.setSheetOrder("sheet1", 0);
XSSFRangeCopier xssfRangeCopier = new XSSFRangeCopier(sheetFrom, sheetTo);
int lastRow = sheetFrom.getLastRowNum();
int lastCol = 0;
for (int i = 0; i < lastRow; i++) {
Row row = sheetFrom.getRow(i);
if (row != null) {
if (row.getLastCellNum() > lastCol) {
lastCol = row.getLastCellNum();
}
sheetTo.setDefaultRowHeight(sheetFrom.getDefaultRowHeight());
}
}
for (int j = 0; j < lastCol; j++) {
sheetTo.setColumnWidth(j, sheetFrom.getColumnWidth(j));
}
CellRangeAddress cellAddresses = new CellRangeAddress(0, lastRow, 0, lastCol);
xssfRangeCopier.copyRange(cellAddresses, cellAddresses, true, true);
workbookTo.write(new FileOutputStream(new File("/path/to/worksheetTo.xlsx")));

POI version < v4.0
Okay I tried to do what Gagravarr said above. This solution works for me. This code will work if the sheets don't have tables, etc. If the sheets contain simple text (String, boolean, int etc), formulas, this solution will work.
Workbook oldWB = new XSSFWorkbook(new FileInputStream("C:\\input.xlsx"));
Workbook newWB = new XSSFWorkbook();
CellStyle newStyle = newWB.createCellStyle(); // Need this to copy over styles from old sheet to new sheet. Next step will be processed below
Row row;
Cell cell;
for (int i = 0; i < oldWB.getNumberOfSheets(); i++) {
XSSFSheet sheetFromOldWB = (XSSFSheet) oldWB.getSheetAt(i);
XSSFSheet sheetForNewWB = (XSSFSheet) newWB.createSheet(sheetFromOldWB.getSheetName());
for (int rowIndex = 0; rowIndex < sheetFromOldWB.getPhysicalNumberOfRows(); rowIndex++) {
row = sheetForNewWB.createRow(rowIndex); //create row in this new sheet
for (int colIndex = 0; colIndex < sheetFromOldWB.getRow(rowIndex).getPhysicalNumberOfCells(); colIndex++) {
cell = row.createCell(colIndex); //create cell in this row of this new sheet
Cell c = sheetFromOldWB.getRow(rowIndex).getCell(colIndex, Row.CREATE_NULL_AS_BLANK ); //get cell from old/original WB's sheet and when cell is null, return it as blank cells. And Blank cell will be returned as Blank cells. That will not change.
if (c.getCellType() == Cell.CELL_TYPE_BLANK){
System.out.println("This is BLANK " + ((XSSFCell) c).getReference());
}
else { //Below is where all the copying is happening. First It copies the styles of each cell and then it copies the content.
CellStyle origStyle = c.getCellStyle();
newStyle.cloneStyleFrom(origStyle);
cell.setCellStyle(newStyle);
switch (c.getCellTypeEnum()) {
case STRING:
cell.setCellValue(c.getRichStringCellValue().getString());
break;
case NUMERIC:
if (DateUtil.isCellDateFormatted(cell)) {
cell.setCellValue(c.getDateCellValue());
} else {
cell.setCellValue(c.getNumericCellValue());
}
break;
case BOOLEAN:
cell.setCellValue(c.getBooleanCellValue());
break;
case FORMULA:
cell.setCellValue(c.getCellFormula());
break;
case BLANK:
cell.setCellValue("who");
break;
default:
System.out.println();
}
}
}
}
}
//Write over to the new file
FileOutputStream fileOut = new FileOutputStream("C:\\output.xlsx");
newWB.write(fileOut);
oldWB.close();
newWB.close();
fileOut.close();
If your requirement is to copy full sheets without leaving or adding anything. I think The process of elimination works better and faster then the above code. And you don't have to worry about losing formulas, drawings, tables, styles, fonts, etc.
XSSFWorkbook wb = new XSSFWorkbook("C:\\abc.xlsx");
for (int i = wb.getNumberOfSheets() - 1; i >= 0; i--) {
if (!wb.getSheetName(i).contentEquals("January")) //This is a place holder. You will insert your logic here to get the sheets that you want.
wb.removeSheetAt(i); //Just remove the sheets that don't match your criteria in the if statement above
}
FileOutputStream out = new FileOutputStream(new File("C:\\xyz.xlsx"));
wb.write(out);
out.close();
POI version >= v4.0
As of version 4.0, Cell.CELL_TYPE_BLANK and Row.CREATE_NULL_AS_BLANK don't exist (they deprecated). Use CellType.* and Row.MissingCellPolicy.* instead.