快速将Excel导入DataTable

Question

我试图将Excel文件读入Data.DataTable列表,虽然使用我当前的方法可能需要很长时间.我特意按工作表逐个单元地去工作表,它往往需要很长时间.有更快的方法吗？这是我的代码:

    List<DataTable> List = new List<DataTable>();

    // Counting sheets
    for (int count = 1; count < WB.Worksheets.Count; ++count)
    {
        // Create a new DataTable for every Worksheet
        DATA.DataTable DT = new DataTable();

        WS = (EXCEL.Worksheet)WB.Worksheets.get_Item(count);

        textBox1.Text = count.ToString();

        // Get range of the worksheet
        Range = WS.UsedRange;


        // Create new Column in DataTable
        for (cCnt = 1; cCnt <= Range.Columns.Count; cCnt++)
        {
            textBox3.Text = cCnt.ToString();


                Column = new DataColumn();
                Column.DataType = System.Type.GetType("System.String");
                Column.ColumnName = cCnt.ToString();
                DT.Columns.Add(Column);

            // Create row for Data Table
            for (rCnt = 0; rCnt <= Range.Rows.Count; rCnt++)
            {
                textBox2.Text = rCnt.ToString();

                try
                {
                    cellVal = (string)(Range.Cells[rCnt, cCnt] as EXCEL.Range).Value2;
                }
                catch (Microsoft.CSharp.RuntimeBinder.RuntimeBinderException)
                {
                    ConvertVal = (double)(Range.Cells[rCnt, cCnt] as EXCEL.Range).Value2;
                    cellVal = ConvertVal.ToString();
                }

                // Add to the DataTable
                if (cCnt == 1)
                {

                    Row = DT.NewRow();
                    Row[cCnt.ToString()] = cellVal;
                    DT.Rows.Add(Row);
                }
                else
                {

                    Row = DT.Rows[rCnt];
                    Row[cCnt.ToString()] = cellVal;

                }
            }
        }
        // Add DT to the list. Then go to the next sheet in the Excel Workbook
        List.Add(DT);
    }

Answer 1

Caling .Value2是一项昂贵的操作,因为它是一个COM-interop调用.我会将整个范围读入一个数组,然后遍历数组:

object[,] data = Range.Value2;

// Create new Column in DataTable
for (int cCnt = 1; cCnt <= Range.Columns.Count; cCnt++)
{
    textBox3.Text = cCnt.ToString();

    var Column = new DataColumn();
    Column.DataType = System.Type.GetType("System.String");
    Column.ColumnName = cCnt.ToString();
    DT.Columns.Add(Column);

    // Create row for Data Table
    for (int rCnt = 1; rCnt <= Range.Rows.Count; rCnt++)
    {
        textBox2.Text = rCnt.ToString();

        string CellVal = String.Empty;
        try
        {
            cellVal = (string)(data[rCnt, cCnt]);
        }
        catch (Microsoft.CSharp.RuntimeBinder.RuntimeBinderException)
        {
            ConvertVal = (double)(data[rCnt, cCnt]);
            cellVal = ConvertVal.ToString();
        }

        DataRow Row;

        // Add to the DataTable
        if (cCnt == 1)
        {

            Row = DT.NewRow();
            Row[cCnt.ToString()] = cellVal;
            DT.Rows.Add(Row);
        }
        else
        {

            Row = DT.Rows[rCnt + 1];
            Row[cCnt.ToString()] = cellVal;

        }
    }
} 

Answer 2

如果其他人正在使用EPPlus。这个实现很幼稚，但是有一些评论引起了人们的注意。如果要GetWorkbookAsDataSet()在顶部再增加一种方法，它将满足OP的要求。

    /// <summary>
    /// Assumption: Worksheet is in table format with no weird padding or blank column headers.
    /// 
    /// Assertion: Duplicate column names will be aliased by appending a sequence number (eg. Column, Column1, Column2)
    /// </summary>
    /// <param name="worksheet"></param>
    /// <returns></returns>
    public static DataTable GetWorksheetAsDataTable(ExcelWorksheet worksheet)
    {
        var dt = new DataTable(worksheet.Name);
        dt.Columns.AddRange(GetDataColumns(worksheet).ToArray());
        var headerOffset = 1; //have to skip header row
        var width = dt.Columns.Count;
        var depth = GetTableDepth(worksheet, headerOffset);
        for (var i = 1; i <= depth; i++)
        {
            var row = dt.NewRow();
            for (var j = 1; j <= width; j++)
            {
                var currentValue = worksheet.Cells[i + headerOffset, j].Value;

                //have to decrement b/c excel is 1 based and datatable is 0 based.
                row[j - 1] = currentValue == null ? null : currentValue.ToString();
            }

            dt.Rows.Add(row);
        }

        return dt;
    }

    /// <summary>
    /// Assumption: There are no null or empty cells in the first column
    /// </summary>
    /// <param name="worksheet"></param>
    /// <returns></returns>
    private static int GetTableDepth(ExcelWorksheet worksheet, int headerOffset)
    {
        var i = 1;
        var j = 1;
        var cellValue = worksheet.Cells[i + headerOffset, j].Value;
        while (cellValue != null)
        {
            i++;
            cellValue = worksheet.Cells[i + headerOffset, j].Value;
        }

        return i - 1; //subtract one because we're going from rownumber (1 based) to depth (0 based)
    }

    private static IEnumerable<DataColumn> GetDataColumns(ExcelWorksheet worksheet)
    {
        return GatherColumnNames(worksheet).Select(x => new DataColumn(x));
    }

    private static IEnumerable<string> GatherColumnNames(ExcelWorksheet worksheet)
    {
        var columns = new List<string>();

        var i = 1;
        var j = 1;
        var columnName = worksheet.Cells[i, j].Value;
        while (columnName != null)
        {
            columns.Add(GetUniqueColumnName(columns, columnName.ToString()));
            j++;
            columnName = worksheet.Cells[i, j].Value;
        }

        return columns;
    }

    private static string GetUniqueColumnName(IEnumerable<string> columnNames, string columnName)
    {
        var colName = columnName;
        var i = 1;
        while (columnNames.Contains(colName))
        {
            colName = columnName + i.ToString();
            i++;
        }

        return colName;
    }