How to AutoFit excel column using OpenXML Package

Question

This code to generate Excel spreadsheet Using openxml package. Please anyone tell how to auto fit its column width.

OpenXmlPackage.SpreadsheetDocument spreadsheetDocument = OpenXmlPackage.SpreadsheetDocument.Create(downloadFilePath, OpenXml.SpreadsheetDocumentType.Workbook);
// Add a WorkbookPart to the document.
OpenXmlPackage.WorkbookPart workbookpart = spreadsheetDocument.AddWorkbookPart();
// Add a WorksheetPart to the WorkbookPart.
workbookpart.Workbook = new OpenXmlSpreadsheet.Workbook();
int numDates = datesObject.Length;

// Add Sheets to the Workbook.
OpenXmlSpreadsheet.Sheets sheets = new OpenXmlSpreadsheet.Sheets();
OpenXml.UInt32Value sheetId = 1;

OpenXmlPackage.WorksheetPart firstWorksheetPart = workbookpart.AddNewPart<OpenXmlPackage.WorksheetPart>();
firstWorksheetPart.Worksheet = new OpenXmlSpreadsheet.Worksheet(new OpenXmlSpreadsheet.SheetData());
// Append a new worksheet and associate it with the workbook.
OpenXmlSpreadsheet.Sheet firstSheet = new OpenXmlSpreadsheet.Sheet() { Id = spreadsheetDocument.WorkbookPart.GetIdOfPart(firstWorksheetPart), SheetId = sheetId, Name = "Summary" };
sheets.Append(firstSheet);
sheetId++;

 OpenXmlSpreadsheet.SheetData firstSheetData = firstWorksheetPart.Worksheet.GetFirstChild<OpenXmlSpreadsheet.SheetData>();

 DataTable summaryTable = new DataTable();
 summaryTable.Clear();
 summaryTable.Columns.Add("name");
 summaryTable.Columns.Add("value");

 DataRow _summaryInfo = summaryTable.NewRow();
 _summaryInfo["name"] = "Clinic Name";
 _summaryInfo["value"] = userInfo[0];
 summaryTable.Rows.Add(_summaryInfo);



 int firstRowCount = summaryTable.Rows.Count;

 for (int rowNumber = 1; rowNumber <= firstRowCount; rowNumber++)
 {
      DataRow dataRow = summaryTable.Rows[rowNumber - 1];
      OpenXmlSpreadsheet.Row contentRow = ExcelHandler.createContentRow(dataRow, rowNumber);
      firstSheetData.AppendChild(contentRow);
 }

 firstWorksheetPart.Worksheet.Save();

Solution

The auto-fit logic is something which is implemented by Microsoft Excel, and is not a part of the OpenXML spreadsheet format. Auto-fit involves measuring the width (or height) of the value in each cell and finding the maximum value.

In order to implement auto-fit in your own code, you will have to manually measure the text; you can use TextRenderer.MeasureText or Graphics.MeasureString with appropriate format flags (disable prefix characters). This will give you a size in pixels, which you will need to convert to Excel's convoluted column width units. The formula for this is:

width = Truncate([{Number of Characters} * {Maximum Digit Width} + {5 pixel padding}]/{Maximum Digit Width}*256)/256

Taken from this article: Column Class (DocumentFormat.OpenXml.Spreadsheet)

(Maximum Digit Width can be determined by measuring the width of the '0' character using the workbook's default font - told you it was convoluted!)

Once you have obtained the cell widths using this formula, you can find the maximum value and apply it to the Column.Width property.

There are subtle differences in the way Microsoft Excel renders text (compared to how GDI/GDI+ does it), so this method is not 100% accurate - but it is sufficient for most purposes and you can always add some extra padding to ensure a proper fit.

Comments

[Emaborsa]

What does the Truncate exactly do and why is it needed?

[Bradley Smith]

@Emaborsa Truncate discards the decimal portion of the number, effectively rounding it down to the nearest integer. This is needed because pixels are expressed as whole numbers.

[Emaborsa]

OK, but the Column.Width property takes a double (I think in points) and not an int for the pixel...

[Bradley Smith]

Sorry, my last comment was wrong; the Column.Width property is measured in Excel's column width units (not pixels or points). So the Truncate function is used purely because Microsoft decided to discard the decimal portion of the result before dividing by 256 - maybe this is for precision reasons? Anyway, it does not make a huge difference to the final figure.

[Lucy82]

@Prakash Gupta take a look at my answer. Posted code is in DOM method, but I created a solution with SAX approach too and results are outstanding - no memory drain (writing directly from DataReader) and much faster than Interop library. Only down-side is Autofit - you ave to read same data twice.