我使用的是 Visual Studio 2008，我需要使用 Open XML SDK 2.0 从 Excel 工作表创建一个DataTable。我需要用工作表第一行的 DataTable 列创建它，并用其余的值完成它。
有人有示例代码或链接可以帮助我做到这一点吗？
【问题讨论】：
标签： c# openxml openxml-sdk import-from-excel
我认为这应该符合您的要求。如果您有共享字符串，则另一个功能只是为了处理，我假设您在列标题中执行此操作。不确定这是否完美，但我希望它有所帮助。
static void Main(string[] args)
{
DataTable dt = new DataTable();
using (SpreadsheetDocument spreadSheetDocument = SpreadsheetDocument.Open(@"..\..\example.xlsx", false))
{
WorkbookPart workbookPart = spreadSheetDocument.WorkbookPart;
IEnumerable<Sheet> sheets = spreadSheetDocument.WorkbookPart.Workbook.GetFirstChild<Sheets>().Elements<Sheet>();
string relationshipId = sheets.First().Id.Value;
WorksheetPart worksheetPart = (WorksheetPart)spreadSheetDocument.WorkbookPart.GetPartById(relationshipId);
Worksheet workSheet = worksheetPart.Worksheet;
SheetData sheetData = workSheet.GetFirstChild<SheetData>();
IEnumerable<Row> rows = sheetData.Descendants<Row>();
foreach (Cell cell in rows.ElementAt(0))
{
dt.Columns.Add(GetCellValue(spreadSheetDocument, cell));
}
foreach (Row row in rows) //this will also include your header row...
{
DataRow tempRow = dt.NewRow();
for (int i = 0; i < row.Descendants<Cell>().Count(); i++)
{
tempRow[i] = GetCellValue(spreadSheetDocument, row.Descendants<Cell>().ElementAt(i-1));
}
dt.Rows.Add(tempRow);
}
}
dt.Rows.RemoveAt(0); //...so i'm taking it out here.
}
public static string GetCellValue(SpreadsheetDocument document, Cell cell)
{
SharedStringTablePart stringTablePart = document.WorkbookPart.SharedStringTablePart;
string value = cell.CellValue.InnerXml;
if (cell.DataType != null && cell.DataType.Value == CellValues.SharedString)
{
return stringTablePart.SharedStringTable.ChildElements[Int32.Parse(value)].InnerText;
}
else
{
return value;
}
}
【讨论】：
tempRow 中将 A1、D1 和 G1 彼此相邻。同样在某些情况下CellValue 为空，我在GetCellValue 的第二行中遇到异常。
您好，上面的代码工作正常，除了一个变化
替换下面的代码行
tempRow[i] = GetCellValue(spreadSheetDocument, row.Descendants<Cell>().ElementAt(i-1));
与
tempRow[i] = GetCellValue(spreadSheetDocument, row.Descendants<Cell>().ElementAt(i));
如果使用 (i-1) 会抛出异常：
specified argument was out of the range of valid values. parameter name index.
【讨论】：
此解决方案适用于没有空单元格的电子表格。
要处理空单元格，您需要替换此行：
tempRow[i] = GetCellValue(spreadSheetDocument, row.Descendants<Cell>().ElementAt(i-1));
类似这样的：
Cell cell = row.Descendants<Cell>().ElementAt(i);
int index = CellReferenceToIndex(cell);
tempRow[index] = GetCellValue(spreadSheetDocument, cell);
并添加此方法：
private static int CellReferenceToIndex(Cell cell)
{
int index = -1;
string reference = cell.CellReference.ToString().ToUpper();
foreach (char ch in reference)
{
if (Char.IsLetter(ch))
{
int value = (int)ch - (int)'A';
index = (index + 1) * 26 + value;
}
else
return index;
}
return index;
}
【讨论】：
(index == 0) ? value : ((index + 1) * 26) + value => 将从A 开始的多字符索引返回不正确的结果，例如AA、AB、AAC 等，因为A 值将转换为@ 987654330@ 代替 26 乘数
这是我的完整解决方案，其中还考虑了空单元格。
public static class ExcelHelper
{
//To get the value of the cell, even it's empty. Unable to use loop by index
private static string GetCellValue(WorkbookPart wbPart, List<Cell> theCells, string cellColumnReference)
{
Cell theCell = null;
string value = "";
foreach (Cell cell in theCells)
{
if (cell.CellReference.Value.StartsWith(cellColumnReference))
{
theCell = cell;
break;
}
}
if (theCell != null)
{
value = theCell.InnerText;
// If the cell represents an integer number, you are done.
// For dates, this code returns the serialized value that represents the date. The code handles strings and
// Booleans individually. For shared strings, the code looks up the corresponding value in the shared string table. For Booleans, the code converts the value into the words TRUE or FALSE.
if (theCell.DataType != null)
{
switch (theCell.DataType.Value)
{
case CellValues.SharedString:
// For shared strings, look up the value in the shared strings table.
var stringTable = wbPart.GetPartsOfType<SharedStringTablePart>().FirstOrDefault();
// If the shared string table is missing, something is wrong. Return the index that is in the cell. Otherwise, look up the correct text in the table.
if (stringTable != null)
{
value = stringTable.SharedStringTable.ElementAt(int.Parse(value)).InnerText;
}
break;
case CellValues.Boolean:
switch (value)
{
case "0":
value = "FALSE";
break;
default:
value = "TRUE";
break;
}
break;
}
}
}
return value;
}
private static string GetCellValue(WorkbookPart wbPart, List<Cell> theCells, int index)
{
return GetCellValue(wbPart, theCells, GetExcelColumnName(index));
}
private static string GetExcelColumnName(int columnNumber)
{
int dividend = columnNumber;
string columnName = String.Empty;
int modulo;
while (dividend > 0)
{
modulo = (dividend - 1) % 26;
columnName = Convert.ToChar(65 + modulo).ToString() + columnName;
dividend = (int)((dividend - modulo) / 26);
}
return columnName;
}
//Only xlsx files
public static DataTable GetDataTableFromExcelFile(string filePath, string sheetName = "")
{
DataTable dt = new DataTable();
try
{
using (SpreadsheetDocument document = SpreadsheetDocument.Open(filePath, false))
{
WorkbookPart wbPart = document.WorkbookPart;
IEnumerable<Sheet> sheets = document.WorkbookPart.Workbook.GetFirstChild<Sheets>().Elements<Sheet>();
string sheetId = sheetName != "" ? sheets.Where(q => q.Name == sheetName).First().Id.Value : sheets.First().Id.Value;
WorksheetPart wsPart = (WorksheetPart)wbPart.GetPartById(sheetId);
SheetData sheetdata = wsPart.Worksheet.Elements<SheetData>().FirstOrDefault();
int totalHeaderCount = sheetdata.Descendants<Row>().ElementAt(0).Descendants<Cell>().Count();
//Get the header
for (int i = 1; i <= totalHeaderCount; i++)
{
dt.Columns.Add(GetCellValue(wbPart, sheetdata.Descendants<Row>().ElementAt(0).Elements<Cell>().ToList(), i));
}
foreach (Row r in sheetdata.Descendants<Row>())
{
if (r.RowIndex > 1)
{
DataRow tempRow = dt.NewRow();
//Always get from the header count, because the index of the row changes where empty cell is not counted
for (int i = 1; i <= totalHeaderCount; i++)
{
tempRow[i - 1] = GetCellValue(wbPart, r.Elements<Cell>().ToList(), i);
}
dt.Rows.Add(tempRow);
}
}
}
}
catch (Exception ex)
{
}
return dt;
}
}
【讨论】：
Public Shared Function ExcelToDataTable(filename As String) As DataTable
Try
Dim dt As New DataTable()
Using doc As SpreadsheetDocument = SpreadsheetDocument.Open(filename, False)
Dim workbookPart As WorkbookPart = doc.WorkbookPart
Dim sheets As IEnumerable(Of Sheet) = doc.WorkbookPart.Workbook.GetFirstChild(Of Sheets)().Elements(Of Sheet)()
Dim relationshipId As String = sheets.First().Id.Value
Dim worksheetPart As WorksheetPart = DirectCast(doc.WorkbookPart.GetPartById(relationshipId), WorksheetPart)
Dim workSheet As Worksheet = worksheetPart.Worksheet
Dim sheetData As SheetData = workSheet.GetFirstChild(Of SheetData)()
Dim rows As IEnumerable(Of Row) = sheetData.Descendants(Of Row)()
For Each cell As Cell In rows.ElementAt(0)
dt.Columns.Add(GetCellValue(doc, cell))
Next
For Each row As Row In rows
'this will also include your header row...
Dim tempRow As DataRow = dt.NewRow()
For i As Integer = 0 To row.Descendants(Of Cell)().Count() - 1
tempRow(i) = GetCellValue(doc, row.Descendants(Of Cell)().ElementAt(i))
Next
dt.Rows.Add(tempRow)
Next
End Using
dt.Rows.RemoveAt(0)
Return dt
Catch ex As Exception
Throw ex
End Try
End Function
Public Shared Function GetCellValue(document As SpreadsheetDocument, cell As Cell) As String
Try
If IsNothing(cell.CellValue) Then
Return ""
End If
Dim value As String = cell.CellValue.InnerXml
If cell.DataType IsNot Nothing AndAlso cell.DataType.Value = CellValues.SharedString Then
Dim stringTablePart As SharedStringTablePart = document.WorkbookPart.SharedStringTablePart
Return stringTablePart.SharedStringTable.ChildElements(Int32.Parse(value)).InnerText
Else
Return value
End If
Catch ex As Exception
Return ""
End Try
End Function
【讨论】：
首先将 ExcelUtility.cs 添加到您的项目中：
ExcelUtility.cs
using System.Data;
using System.Linq;
using DocumentFormat.OpenXml.Packaging;
using DocumentFormat.OpenXml.Spreadsheet;
namespace Core_Excel.Utilities
{
static class ExcelUtility
{
public static DataTable Read(string path)
{
var dt = new DataTable();
using (var ssDoc = SpreadsheetDocument.Open(path, false))
{
var sheets = ssDoc.WorkbookPart.Workbook.GetFirstChild<Sheets>().Elements<Sheet>();
var relationshipId = sheets.First().Id.Value;
var worksheetPart = (WorksheetPart) ssDoc.WorkbookPart.GetPartById(relationshipId);
var workSheet = worksheetPart.Worksheet;
var sheetData = workSheet.GetFirstChild<SheetData>();
var rows = sheetData.Descendants<Row>().ToList();
foreach (var row in rows) //this will also include your header row...
{
var tempRow = dt.NewRow();
var colCount = row.Descendants<Cell>().Count();
foreach (var cell in row.Descendants<Cell>())
{
var index = GetIndex(cell.CellReference);
// Add Columns
for (var i = dt.Columns.Count; i <= index; i++)
dt.Columns.Add();
tempRow[index] = GetCellValue(ssDoc, cell);
}
dt.Rows.Add(tempRow);
}
}
var m = dt.Rows[0][9];
return dt;
}
private static string GetCellValue(SpreadsheetDocument document, Cell cell)
{
var stringTablePart = document.WorkbookPart.SharedStringTablePart;
var value = cell.CellValue.InnerXml;
if (cell.DataType != null && cell.DataType.Value == CellValues.SharedString)
return stringTablePart.SharedStringTable.ChildElements[int.Parse(value)].InnerText;
return value;
}
public static int GetIndex(string name)
{
if (string.IsNullOrWhiteSpace(name))
return -1;
int index = 0;
foreach (var ch in name)
{
if (char.IsLetter(ch))
{
int value = ch - 'A' + 1;
index = value + index * 26;
}
else
break;
}
return index - 1;
}
}
}
用法：
var path = "D:\\Documents\\test.xlsx";
var dt = ExcelUtility.Read(path);
那就尽情享受吧！
【讨论】：
我知道自从这个帖子开始以来已经很久了。但是，上述解决方案都没有真正适合我。空单元格问题和其他问题。
我在 GitHub 上使用“MIT”许可证找到了一个非常好的解决方案： https://github.com/ExcelDataReader/ExcelDataReader 这对我来说适用于 C# 和 VBnet 应用程序。 来自 VBNET 的示例调用（c# 的示例代码在 GitHub 上）：
Using stream As FileStream = New FileStream(DataPath & "\" & fName.Name, FileMode.Open, FileAccess.Read, FileShare.ReadWrite)
Using reader As IExcelDataReader = ExcelReaderFactory.CreateReader(stream)
ds = reader.AsDataSet(New ExcelDataSetConfiguration() With {
.UseColumnDataType = False,
.ConfigureDataTable = Function(tableReader) New ExcelDataTableConfiguration() With {
.UseHeaderRow = True
}
})
End Using
End Using
结果是一个数据集，工作簿中的每个工作表都有一个表。
我真的很喜欢自己编译用 C# 制作的 dll，而不是使用现成的 dll。这样我就可以控制向客户交付的内容。
【讨论】：
如果 rows 值为 null 或为空，则获取值错误。
所有列都填充了数据，如果它正在工作的话。但也许所有行都不是
【讨论】：