Open Xml - File Needs To Be Repaired To Open - openxml

Question

I am having an issue where the file generation process works as expected, but when I open the excel file it says that it is corrupt and needs to be repaired. When the repair is complete, the file opens and all of the data is there.
The error message that I am receiving is as follows:
Removed Records: Cell information from /xl/worksheets/sheet1.xml part
My code is as follows:
using (var workbookDocument = SpreadsheetDocument.Create(staging, DocumentFormat.OpenXml.SpreadsheetDocumentType.Workbook))
{
var count = query.Count();
var worksheetNumber = 1;
var worksheetCapacity = Convert.ToInt32(100000);
var worksheetCount = Convert.ToInt32(Math.Ceiling(Convert.ToDouble(count) / worksheetCapacity));
var workbookPart = workbookDocument.AddWorkbookPart();
var worksheetInfo = new List<WorksheetData>();
OpenXmlWriter worksheetWriter;
while (worksheetNumber <= worksheetCount)
{
var worksheetLine = 1;
var worksheetName = sheet + "_" + Convert.ToString(worksheetNumber);
var worksheetPart = workbookDocument.WorkbookPart.AddNewPart<WorksheetPart>
var worksheetId = workbookDocument.WorkbookPart.GetIdOfPart(worksheetPart);
var worksheetKey = Convert.ToUInt32(worksheetNumber);
var worksheetAttributes = new List<OpenXmlAttribute>();
worksheetAttributes.Add(new OpenXmlAttribute("r", null, worksheetLine.ToString()));
worksheetInfo.Add(new WorksheetData() { Id = worksheetId, Key = worksheetKey, Name = worksheetName });
worksheetWriter = OpenXmlWriter.Create(worksheetPart);
worksheetWriter.WriteStartElement(new Worksheet());
worksheetWriter.WriteStartElement(new SheetData());
worksheetWriter.WriteStartElement(new Row(), worksheetAttributes);
for (var i = 0; i < headers.Count; i++)
{
var worksheetCell = new DocumentFormat.OpenXml.Spreadsheet.Cell();
var worksheetValue = new DocumentFormat.OpenXml.Spreadsheet.CellValue(headers[i]);
worksheetAttributes.Clear();
worksheetAttributes.Add(new OpenXmlAttribute("t", null, "str"));
worksheetAttributes.Add(new OpenXmlAttribute("r", null, GetColumnReference(worksheetLine, i)));
worksheetWriter.WriteStartElement(worksheetCell, worksheetAttributes);
worksheetWriter.WriteElement(worksheetValue);
worksheetWriter.WriteEndElement();
}
worksheetWriter.WriteEndElement();
worksheetLine++;
var skip = ((worksheetNumber - 1) * worksheetCapacity);
var results = query.SelectProperties(columns).Skip(skip).Take(worksheetCapacity).ToList();
for (var j = 0; j < results.Count; j++)
{
worksheetAttributes.Clear();
worksheetAttributes.Add(new OpenXmlAttribute("r", null, worksheetLine.ToString()));
worksheetWriter.WriteStartElement(new Row());
for (var k = 0; k < columns.Count(); k++)
{
var column = columns[k].Split((".").ToCharArray()).Last();
var value = results[j].GetType().GetField(column).GetValue(results[j]);
var type = value?.GetType().Name;
var text = ExportFormatter.Format(type, value);
worksheetAttributes.Clear();
worksheetAttributes.Add(new OpenXmlAttribute("t", null, "str"));
worksheetAttributes.Add(new OpenXmlAttribute("r", null, GetColumnReference(worksheetLine, j)));
worksheetWriter.WriteStartElement(new Cell());
worksheetWriter.WriteElement(new CellValue(text));
worksheetWriter.WriteEndElement();
}
worksheetWriter.WriteEndElement();
worksheetLine++;
}
worksheetWriter.WriteEndElement();
worksheetWriter.WriteEndElement();
worksheetWriter.Close();
worksheetNumber++;
}
worksheetWriter = OpenXmlWriter.Create(workbookDocument.WorkbookPart);
worksheetWriter.WriteStartElement(new Workbook());
worksheetWriter.WriteStartElement(new Sheets());
for (var i = 0; i < worksheetInfo.Count; i++)
{
worksheetWriter.WriteElement(new Sheet()
{
Name = worksheetInfo[i].Name,
SheetId = worksheetInfo[i].Key,
Id = worksheetInfo[i].Id
});
}
worksheetWriter.WriteEndElement();
worksheetWriter.WriteEndElement();
worksheetWriter.Close();
}
I use the below class to track the worksheet information:
private class WorksheetData
{
public String Id { get; set; }
public UInt32 Key { get; set; }
public String Name { get; set; }
}
Can anyone identify why this is happening?
Perhaps and extra ending tag or ones that missing?

Answer 1

Finally got this to work, there were a few issues.
The cell references A1 A2 A3, etc... were not correct in the code.
The row number were not beign tracked corretly.
The attributes being applied to the cell elements were not correct because they wern't being cleared prior to writing.
The usage of the CallValue was not functioning as expected. Not sure exactly why, but when the Cell Value is used it doesn't open properly in excel. Simply using the cell and setting the DataType and CellValue properties did work. Note - The underlying xml looked exactly the same between the two, but only the second worked.
The final code from this is as follows:
public static ExportInfo Export<T>(this IQueryable<T> query, String temp, String path, List<ExportField> fields)
{
var worker = new ExportWorker();
return worker.Export<T>(query, temp, path, fields);
}
public static class ExportFormatter
{
public static String Format(String type, Object value)
{
if (value == null)
{
return "";
}
else
{
var text = "";
switch (type)
{
case "Decimal":
var decimalValue = (Decimal)value;
text = decimal.Round(decimalValue, 2, MidpointRounding.AwayFromZero).ToString();
break;
case "DateTimeOffset":
var dateTimeOffset = (DateTimeOffset)value;
text = dateTimeOffset.ToUniversalTime().ToString("MM/dd/yyyy");
break;
case "DateTime":
var dateTime = (DateTime)value;
text = dateTime.ToUniversalTime().ToString("MM/dd/yyyy");
break;
default:
text = Convert.ToString(value);
break;
}
return text;
}
public class ExportWorker
{
String record;
String staging;
String destination;
Thread thread;
Timer timer;
public ExportInfo Export<T>(IQueryable<T> query, String temp, String path, List<ExportField> fields)
{
var selections = from a in fields group a by new { a.Field } into g select new { g.Key.Field, Header = g.Max(x => x.Header) };
var headers = (from a in selections select a.Header).ToList();
var columns = (from a in selections select a.Field).Distinct().ToList();
var entity = query.ElementType.ToString();
var array = entity.Split((".").ToCharArray());
var sheet = array[array.Length - 1];
var key = Guid.NewGuid().ToString().Replace("-", "_");
var name = key + ".xlsx";
var log = key + ".txt";
var timeout = 60 * 60000;
staging = temp + name;
destination = path + name;
record = path + log;
thread = new Thread(
new ThreadStart(() =>
{
try
{
using (var workbookDocument = SpreadsheetDocument.Create(staging, DocumentFormat.OpenXml.SpreadsheetDocumentType.Workbook))
{
var count = query.Count();
var worksheetNumber = 1;
var worksheetCapacity = Convert.ToInt32(100000);
var worksheetCount = Convert.ToInt32(Math.Ceiling(Convert.ToDouble(count) / worksheetCapacity));
var workbookPart = workbookDocument.AddWorkbookPart();
var worksheetInfo = new List<WorksheetData>();
OpenXmlWriter worksheetWriter;
while (worksheetNumber <= worksheetCount)
{
var worksheetLine = 1;
var worksheetThrottle = 0;
var worksheetName = sheet + "_" + Convert.ToString(worksheetNumber);
var worksheetPart = workbookDocument.WorkbookPart.AddNewPart<WorksheetPart>();
var worksheetId = workbookDocument.WorkbookPart.GetIdOfPart(worksheetPart);
var worksheetKey = Convert.ToUInt32(worksheetNumber);
var worksheetAttributes = new List<OpenXmlAttribute>();
worksheetAttributes.Add(new OpenXmlAttribute("r", null, worksheetLine.ToString()));
worksheetInfo.Add(new WorksheetData() { Id = worksheetId, Key = worksheetKey, Name = worksheetName });
worksheetWriter = OpenXmlWriter.Create(worksheetPart);
worksheetWriter.WriteStartElement(new Worksheet());
worksheetWriter.WriteStartElement(new SheetData());
worksheetWriter.WriteStartElement(new Row(), worksheetAttributes);
for (var i = 0; i < headers.Count; i++)
{
var worksheetCell = new Cell();
worksheetCell.DataType = CellValues.String;
worksheetCell.CellValue = new CellValue(headers[i]);
worksheetWriter.WriteElement(worksheetCell);
worksheetAttributes.Clear();
}
worksheetWriter.WriteEndElement();
worksheetLine++;
var skip = ((worksheetNumber - 1) * worksheetCapacity);
var results = query.SelectProperties(columns).Skip(skip).Take(worksheetCapacity).ToList();
for (var j = 0; j < results.Count; j++)
{
if (worksheetThrottle >= 5) { worksheetThrottle = 0; System.Threading.Thread.Sleep(1); }
worksheetAttributes.Clear();
worksheetAttributes.Add(new OpenXmlAttribute("r", null, worksheetLine.ToString()));
worksheetWriter.WriteStartElement(new Row(), worksheetAttributes);
for (var k = 0; k < columns.Count(); k++)
{
var column = columns[k].Split((".").ToCharArray()).Last();
var value = results[j].GetType().GetField(column).GetValue(results[j]);
var type = value?.GetType().Name;
var text = (String)ExportFormatter.Format(type, value);
var worksheetCell = new Cell();
worksheetCell.DataType = CellValues.String;
worksheetCell.CellValue = new CellValue(text);
worksheetWriter.WriteElement(worksheetCell);
worksheetAttributes.Clear();
}
worksheetWriter.WriteEndElement();
worksheetLine++;
worksheetThrottle++;
}
worksheetWriter.WriteEndElement();
worksheetWriter.WriteEndElement();
worksheetWriter.Close();
worksheetNumber++;
}
worksheetWriter = OpenXmlWriter.Create(workbookDocument.WorkbookPart);
worksheetWriter.WriteStartElement(new Workbook());
worksheetWriter.WriteStartElement(new Sheets());
for (var i = 0; i < worksheetInfo.Count; i++)
{
worksheetWriter.WriteElement(new Sheet()
{
Name = worksheetInfo[i].Name,
SheetId = worksheetInfo[i].Key,
Id = worksheetInfo[i].Id
});
}
worksheetWriter.WriteEndElement();
worksheetWriter.WriteEndElement();
worksheetWriter.Close();
}
var logsfile = File.CreateText(record);
var datafile = (new DirectoryInfo(temp)).GetFiles().FirstOrDefault(a => a.Name == name);
datafile.MoveTo(destination);
logsfile.WriteLine("succeeded");
logsfile.Close();
}
catch (Exception ex)
{
try { File.Delete(staging); } catch (Exception) { }
var logsfile = File.CreateText(record);
logsfile.WriteLine("failed");
logsfile.WriteLine(ex.ToString());
logsfile.Close();
}
})
);
thread.Start();
timer = new Timer(Expire, null, timeout, Timeout.Infinite);
return new ExportInfo() { File = destination, Log = record };
}
void Expire(object state)
{
try { File.Delete(staging); } catch (Exception) { }
var logsfile = File.CreateText(record);
logsfile.WriteLine("timeout");
logsfile.Close();
thread.Abort();
}
private class WorksheetData
{
public String Id { get; set; }
public UInt32 Key { get; set; }
public String Name { get; set; }
}
}
After making those adjustments, the export works beautifully.
Also, open xml solved a lot of problems that I was having with memory management.
Using the above approach allowed me to export 3 files, each with 1.5 million rows (40 columns) in about 10 minutes.
During the export process, CPU utilization never exceeded 35% and it never used more than 1/10 of a gig of memory. Bravo...