As I'm writing quite large xlsx files, I'm using OpenXmlReader and OpenXmlWriter as recommended on this page:
https://blogs.msdn.microsoft.com/brian_jones/2010/06/22/writing-large-excel-files-with-the-open-xml-sdk/
What I only do is change formulas inside existing cells and making sure that their value is discarded so that it is recalculated when Excel opens the file.
Here is the function that I'm using:
public void Save(Stream Input, Stream Output)
{
Input.Position = 0;
if (Input != Output)
Input.CopyTo(Output);
using (SpreadsheetDocument document = SpreadsheetDocument.Open(Output, true))
{
WorkbookPart wbPart = document.WorkbookPart;
// force recalculation as we change formulas
wbPart.Workbook.CalculationProperties.ForceFullCalculation = true;
wbPart.Workbook.CalculationProperties.FullCalculationOnLoad = true;
// store the worksheet parts in a separate list because the loop below
// adds and removes elements inside wbPart.WorksheetParts
List<WorksheetPart> originalWsParts = new List<WorksheetPart>();
foreach (WorksheetPart inputWsPart in wbPart.WorksheetParts)
originalWsParts.Add(inputWsPart);
// process all worksheets in the workbook
foreach (WorksheetPart inputWsPart in originalWsParts)
{
string origninalSheetId = wbPart.GetIdOfPart(inputWsPart);
WorksheetPart replacementWsPart = wbPart.AddNewPart<WorksheetPart>();
string replacementWsPartId = wbPart.GetIdOfPart(replacementWsPart);
OpenXmlReader reader = OpenXmlReader.Create(inputWsPart);
OpenXmlWriter writer = OpenXmlWriter.Create(replacementWsPart);
while (reader.Read())
{
logger.Debug(reader.ElementType.Name);
if (reader.ElementType == typeof(Cell) && reader.IsStartElement)
{
writer.WriteStartElement(reader);
// write the cell content, changing the formula and skipping the value
while (reader.Read() && !(reader.ElementType == typeof(Cell) && reader.IsEndElement))
{
if (reader.IsStartElement)
{
if (reader.ElementType == typeof(CellFormula))
{
CellFormula element = reader.LoadCurrentElement() as CellFormula;
element.Text = "SUM(1,2)";
element.CalculateCell = true;
writer.WriteElement(element);
}
else if (reader.ElementType != typeof(CellValue))
{
writer.WriteStartElement(reader);
string elementText = reader.GetText();
if (!String.IsNullOrEmpty(elementText))
writer.WriteString(elementText);
}
}
else if (reader.IsEndElement)
{
if (reader.ElementType != typeof(CellValue))
writer.WriteEndElement();
}
}
writer.WriteEndElement();
}
else
{
if (reader.IsStartElement)
{
writer.WriteStartElement(reader);
string elementText = reader.GetText();
if (!String.IsNullOrEmpty(elementText))
writer.WriteString(elementText);
}
else if (reader.IsEndElement)
{
writer.WriteEndElement();
}
}
}
reader.Close();
writer.Close();
Sheet sheet = wbPart.Workbook.Descendants<Sheet>()
.Where(s => s.Id.Value.Equals(origninalSheetId)).First();
sheet.Id.Value = replacementWsPartId;
wbPart.DeletePart(inputWsPart);
}
}
}
It works quite well on the simplest workbooks, but it creates unreadable files when there are drawings inside sheets in the file.
For instance, if I have a drawings on Sheet1, when Excel opens the saved file, it complains that the file has unreadable parts and shows me the drawings in the list of things it has deleted.
I unzipped the xlsx file and compared the sheetX.xml files, and apart from the added x: prefix, they are the same.
Obviously, I'm missing something but reading the various docs I could find, nothing came to me. I believe there is a reference to the original worksheet part that has not been updated but I don't see any drawings descendant in the workbook.
Any help is most welcome.
Update
I looked more closely at the files content and there are two folders missing inside the xl folder: charts and drawings
So clearly, I'm missing code that would add those into the final archive, but I can't (yet) figure out what code this is.
The above code creates a new WorksheetPart inside the loop, when the intention was always to clone the input part.
I thus went looking for a way to clone a worksheet, but there is no ClonePart() method on WorkbookPart.
Fortunately for me, someone else already had this issue and Brian Jones did a part on his blog about this:
https://blogs.msdn.microsoft.com/brian_jones/2009/02/19/how-to-copy-a-worksheet-within-a-workbook/
So, I changed the start of the using statement to this:
using (SpreadsheetDocument document = SpreadsheetDocument.Open(Output, true), tmpDocument = SpreadsheetDocument.Create(new MemoryStream(), document.DocumentType))
{
WorkbookPart wbPart = document.WorkbookPart;
WorkbookPart tmpWbPart = tmpDocument.AddWorkbookPart();
Then instead of calling AddNewPart<WorksheetPart>, I now have the following code:
// can't directly clone, so add to the temporary workbook part and then back
// into the working workbook part
WorksheetPart tmpWsPart = tmpWbPart.AddPart(inputWsPart);
WorksheetPart replacementWsPart = wbPart.AddPart(tmpWsPart);
tmpWbPart.DeletePart(tmpWsPart);
tmpWsPart = null;
With those changes, I now have my Drawing part in the worksheet, the associated folders in the xlsx file.
As a result, Excel no longer complains when opening the file and the graphs are all there and updated.
Related
I am using Office.Interop.Excel to read data from Excel using C# ASP.Net & Dotnet 6.
I can read the Data and everything seems to be working fine.
But I have a challenge here.
The excel which I am reading data from would be updated every second.
But I am seeing an error while trying to open it and update random data.
The error says that the file is locked for editing.
Please have a look at the code below:
public double GetGoldPrice()
{
string filename = #"D:\Test.xlsx";
int row = 1;
int column = 1;
Application excelApplication = new Application();
Workbook excelWorkBook = excelApplication.Workbooks.Open(filename);
string workbookName = excelWorkBook.Name;
int worksheetcount = excelWorkBook.Worksheets.Count;
if (worksheetcount > 0)
{
Worksheet worksheet = (Worksheet)excelWorkBook.Worksheets[1];
string firstworksheetname = worksheet.Name;
var data = ((Microsoft.Office.Interop.Excel.Range) worksheet.Cells[row, column]).Value;
excelApplication.Quit();
return data;
}
else
{
Console.WriteLine("No worksheets available");
excelApplication.Quit();
return 0;
}
}
My end goal is to get live data from Excel whenever I fire the function.
The Excel would be open and can be editing any time.
Please help!
You said your file is xlsx so you would be better not using Interop but Open XML SDK 2.5. Then you can open the file in read only mode:
using (SpreadsheetDocument spreadsheetDocument =
SpreadsheetDocument.Open(fileName, false))
{
// Code removed here.
}
Check here to get familiar with Open XML SDK
I am trying to embed object into .xlsx document and copy sheets with embedded objects.
1. Copying sheets
This looks like straight forward issue. I have created method to copy the sheets:
static void CopySheetInsideWorkbook(string filename, string sheetName, string clonedSheetName)
{
using (SpreadsheetDocument spreadsheetDocument = SpreadsheetDocument.Open(filename, true))
{
WorkbookPart workbookPart = spreadsheetDocument.WorkbookPart;
WorksheetPart sourceSheetPart = GetWorksheetPartByName(spreadsheetDocument, sheetName);
SpreadsheetDocument tempSheet =
SpreadsheetDocument.Create(new MemoryStream(), spreadsheetDocument.DocumentType);
WorkbookPart tempWorkbookPart = tempSheet.AddWorkbookPart();
WorksheetPart tempWorksheetPart = tempWorkbookPart.AddPart<WorksheetPart>(sourceSheetPart);
WorksheetPart clonedSheet = workbookPart.AddPart<WorksheetPart>(tempWorksheetPart);
Sheets sheets = workbookPart.Workbook.GetFirstChild<Sheets>();
Sheet copiedSheet = new Sheet
{
Name = clonedSheetName,
Id = workbookPart.GetIdOfPart(clonedSheet),
SheetId = (uint) sheets.ChildElements.Count + 1
};
sheets.Append(copiedSheet);
workbookPart.Workbook.Save();
}
}
The ouput is as expected but the embedded files are copied as "Picture" rather than "Object". I unzipped .xlsx file and all looks legit ie. similar to the sheet I copied. Yet still the file cannot be opened on the copied sheet. All images, strings are displayed in correct way.
2. Embedding the object
What I understand I need to do is:
Convert object into oleObject - this will be separate fun.
Add DrawingsPart - It looks like it's read-only and I can only add ImagePart.
Embed Object
Connect both drawing and embedded object part toghether and allocate to some range in spreadsheet.
static void EmbedFileXlsx(string path, string embeddedFilePath, string placeholderImagePath)
{
using (SpreadsheetDocument spreadsheetDocument = SpreadsheetDocument.Open(path, true))
{
WorksheetPart sourceSheetPart = GetWorksheetPartByName(spreadsheetDocument, "Test");
var imagePart = sourceSheetPart.AddImagePart(ImagePartType.Emf, "rId1");
imagePart.FeedData(File.Open(placeholderImagePath, FileMode.Open));
var embeddedObject =
sourceSheetPart.AddEmbeddedObjectPart(#"application/vnd.openxmlformats-officedocument.oleObject");
embeddedObject.FeedData(File.Open(embeddedFilePath, FileMode.Open));
spreadsheetDocument.Save();
}
}
This code just adds embedded objects into the file but does not create any type of relationship between them. This means that file is not visible on the spreadsheet.
I tried copying sheets using ClosedXML as well but unfortunately this is not supported nor the embedding.
I also managed to understand how I can copy sheet into new document with all embedded objects using .xml files inside spreadsheet but I do not think this would be much productive and I would like to achieve this using all the methods inside OpenXML. It looks everything is there but something is amiss.
I am no expert in this, but could this help you on your way?
spreadsheetDocument.CreateRelationshipToPart(SOME ID);
I'm trying to read a fairly simple Excel spreadheet (saved as an xlsx) using the OpenXML nuget package.
I'm able to locate the specific Sheet I'm interested in, but calling sheet.GetFirstChild<SheetData>() on it, in order to access the cell values, always returns null. I've looked at several examples online, and they all seem to agree that this is the correct way to access the data.
The xlsx file I'm trying to read can be downloaded here.
Here is the code I'm using to read the file. The line var data = sheet.GetFirstChild<SheetData>(); is where the null is returned.
void Main()
{
using (var s = System.IO.File.OpenRead("c:\\ImportTemplate.xlsx"))
using (var document = SpreadsheetDocument.Open(s, false))
{
foreach (var worksheetPart in document.WorkbookPart.WorksheetParts)
{
Sheet sheet = GetSheetFromWorkSheet(document.WorkbookPart, worksheetPart);
var name = sheet.Name.Value;
if (name == "ImportData")
{
var data = sheet.GetFirstChild<SheetData>(); // <-- data is set to null
foreach (var row in data.Descendants<Row>()){
// [...]
}
}
}
}
}
public static Sheet GetSheetFromWorkSheet(WorkbookPart workbookPart, WorksheetPart worksheetPart)
{
string relationshipId = workbookPart.GetIdOfPart(worksheetPart);
IEnumerable<Sheet> sheets = workbookPart.Workbook.Sheets.Elements<Sheet>();
return sheets.FirstOrDefault(s => s.Id.HasValue && s.Id.Value == relationshipId);
}
The above code as an easily runnable LINQPad file can be downloaded from here.
What am I doing wrong?
using the example here How to Copy a Worksheet within a Workbook
I have successfully been able to clone/copy sheets in my excel file, however when I open the excel the 2nd sheet is the active(visible) sheet. I haven't been able to locate a property that could do thins.....Is there any way to specify what sheet is active?
I've tried to force it by opening and editing the first sheet in the file thinking it was the last edited sheet that was active but that didn't work either.
any help would be great. TIA
update: looking at the workbook.xml created when renaming the .xlsx to .zip I came accross the 'activeTab' property. made a quick change to my code and seems to work just fine
public void SetFirstSheetInFocus(String xlsxFile)
{
using (SpreadsheetDocument spreadSheet = SpreadsheetDocument.Open(xlsxFile, true))
{
//Get a reference to access the main Workbook part, which contains all references
WorkbookPart _workbookPart = spreadSheet.WorkbookPart;
if (_workbookPart != null)
{
WorkbookView _workbookView = spreadSheet.WorkbookPart.Workbook.BookViews.ChildElements.First<WorkbookView>();
if (_workbookView != null)
{
_workbookView.ActiveTab = 0; // 0 for first or whatever tab you want to use
}
// Save the workbook.
_workbookPart.Workbook.Save();
}
}
}
If the name of your sheet is in the variable
sheetName
you can set the sheet with that name active like this:
using (var spreadsheetDoc = SpreadsheetDocument.Open(emptyHIPTemplatePath, true /* isEditable */, new OpenSettings { AutoSave = false }))
{
var workbookPart = spreadsheetDoc.WorkbookPart;
var workBook = spreadsheetDoc.WorkbookPart.Workbook;
var sheet = workBook.Descendants<Sheet>().FirstOrDefault(s => s.Name == sheetName);
var sheetIndex = workBook.Descendants<Sheet>().ToList().IndexOf(sheet);
var workBookView = workBook.Descendants<WorkbookView>().First();
workBookView.ActiveTab = Convert.ToUInt32(sheetIndex);
...
workBook.Save();
}
From Vincent Tan's book:
The SheetId property doesn't determine the order. The order of
appending the Sheet classes to the Sheets class, does.
When you add a sheet, it gets the next index, but a single sheet does not have an index. OpenXML gives it an index when you are done adding sheets. Again, from Vincent Tan's book:
Let's say you have 3 worksheets named Sheet1, Sheet2 and Sheet3.
However, when you appended the corresponding Sheet classes, you did it
as Sheet2, Sheet3 and Sheet1, in that order.
I am trying to convert a xls or xlsx file with multiple sheets into one CSV file using c# and the interop library. I am only getting the one sheet in the CSV file. I know I can specify the sheet to save as or change the active sheet to save that one but I am looking for a solution to append all the sheets to the same CSV file that will work with both xls and xlsx files. I am automating this and don't care what is in the excel document just want to pull the string values out and append it to the csv file. Here is the code I am using:
Microsoft.Office.Interop.Excel.Application app = new Microsoft.Office.Interop.Excel.Application();
app.Visible = false;
app.DisplayAlerts = false;
Workbook wkb = app.Workbooks.Open(fullFilePath);
wkb.SaveAs(newFileName, XlFileFormat.xlCSVWindows);
Is this even possible?
I'm just getting started tackling a similar situation, but I believe this may address your needs:
http://www.codeproject.com/Articles/246772/Convert-xlsx-xls-to-csv
This uses the ExcelDataReader api that you can get from NuGet
http://exceldatareader.codeplex.com/
Like Tim was saying, you're going to have to make sure and possibly validate that the columns and structure are the same between sheets. You may also have to eat the header rows on all the sheets after the first one. I'll post an update and some code samples once I've finished.
Update [7/15/2013]. Here's my finished code. Not very fancy, but it gets the job done. All of the sheets are tables in the DataSet, so you just loop through the tables adding onto your destination. I'm outputting to a MongoDB, but I'm guessing you can swap that out for a StreamWriter for your CSV file rather easily.
private static void ImportValueSetAttributeFile(string filePath)
{
FileStream stream = File.Open(filePath, FileMode.Open, FileAccess.Read);
// Reading from a OpenXml Excel file (2007 format; *.xlsx)
IExcelDataReader excelReader = ExcelReaderFactory.CreateOpenXmlReader(stream);
// DataSet - The result of each spreadsheet will be created in the result.Tables
DataSet result = excelReader.AsDataSet();
// Free resources (IExcelDataReader is IDisposable)
excelReader.Close();
var connectionString = ConfigurationManager.ConnectionStrings[0].ConnectionString;
var database = ConfigurationManager.AppSettings["database"];
var mongoAccess = new MongoDataAccess(connectionString, database);
var cdm = new BaseDataManager();
int ind = 0;
for (int i = 0; i < result.Tables.Count; i++)
{
int row_no = 1;
while (row_no < result.Tables[ind].Rows.Count) // ind is the index of table
// (sheet name) which you want to convert to csv
{
var currRow = result.Tables[ind].Rows[row_no];
var valueSetAttribute = new ValueSetAttribute()
{
CmsId = currRow[0].ToString(),
NqfNumber = currRow[1].ToString(),
ValueSetName = currRow[2].ToString(),
ValueSetOid = currRow[3].ToString(),
Definition = currRow[4].ToString(),
QdmCategory = currRow[5].ToString(),
Expansion = currRow[6].ToString(),
Code = currRow[7].ToString(),
Description = currRow[8].ToString(),
CodeSystem = currRow[9].ToString(),
CodeSystemOid = currRow[10].ToString(),
CodeSystemVersion = currRow[11].ToString()
};
cdm.AddRecords<ValueSetAttribute>(valueSetAttribute, "ValueSetAttributes");
row_no++;
}
ind++;
}
}