Microsoft OLEDB 错误外部表不是预期的格式

Question

我编写了一个可以使用 Microsoft 数据访问引擎读取 excel 2007 文件的代码，下面的代码 sn-p 适用于大多数文件，但它恰好适用于大多数 excel 文件，即 .xlsx ,.xls 但在 objConn.Open() 失败时； excel格式有问题的excel文件请参考下图

打开 ole Db Connection 会失败，说明错误 External table is not in the expected format。这个导入过程的另一个问题是

            OleDbCommand objCmdSelect = new OleDbCommand("SELECT * FROM "+ SelectionSheet, objConn);

无法阅读以空格开头的工作表，我们将不胜感激任何有关解决此问题的帮助。

       public DataTable ReadExcel(string Path, ArrayList IgnoreString, ArrayList IgnoreColumn)
        {
            DataTable dtReturn = new DataTable();
            DataTable dtPrintable = new DataTable();
            DataTable dtTemp = new DataTable();
            try
            {
                string sConnectionString = "Provider=Microsoft.ACE.OLEDB.12.0;" +
                                        "Data Source=" + Path + ";" + "Extended Properties=\"Excel 12.0 Xml;HDR=YES;IMEX=1;\"";

                OleDbConnection objConn = new OleDbConnection(sConnectionString);




                objConn.Open();
                DataTable dtSheetnames = objConn.GetOleDbSchemaTable(OleDbSchemaGuid.Tables, null);

                DataTable dtTesting = objConn.GetOleDbSchemaTable(OleDbSchemaGuid.DbInfoLiterals, new object[] {});
                DataTable dtTesting2 = objConn.GetOleDbSchemaTable(OleDbSchemaGuid.Tables_Info, new object[] { });


                string SelectionSheet = dtSheetnames.Rows[0][2].ToString();

                if (SelectionSheet.Contains("'") )
                {
                    SelectionSheet = SelectionSheet.Remove(0, 1);
                    SelectionSheet = "[" + SelectionSheet;
                    SelectionSheet = SelectionSheet.Remove(SelectionSheet.Length - 1, 1);
                    // -- Mod by zeemz on 23 dec
                    //    string PrintArea = SelectionSheet + "Print_Area]";

                    SelectionSheet = SelectionSheet + "]";
                }
                else
                {
                    SelectionSheet = "["+ SelectionSheet + "]";
                }


                //OleDbCommandBuilder objCmdBuilder = new OleDbCommandBuilder(


                OleDbCommand objCmdSelect = new OleDbCommand("SELECT * FROM "+ SelectionSheet, objConn);




                OleDbDataAdapter objAdapter = new OleDbDataAdapter();
                DataSet objDataSet = new DataSet();

                DataSet PrintAreads = new DataSet();

                objAdapter.SelectCommand = objCmdSelect;
                objAdapter.Fill(objDataSet);

                // -- Mod by zeemz on 23 dec
                //objCmdSelect.CommandText = "SELECT * FROM " + PrintArea;
                //objAdapter.Fill(PrintAreads);




                objConn.Close();

                dtReturn = objDataSet.Tables[0].Copy();
               // dtPrintable = PrintAreads.Tables[0].Copy();

                // -- Mod by zeemz on 23 dec
                //if (dtPrintable.Columns.Count != dtReturn.Columns.Count)
                //{
                //    int TotalPrintable = dtPrintable.Columns.Count;
                //    int TotalComing = dtReturn.Columns.Count;
                //    int StartRemovingPos = TotalComing - TotalPrintable;

                //    for (int i = TotalPrintable; dtPrintable.Columns.Count != dtReturn.Columns.Count; i++)
                //    {

                //        dtReturn.Columns.RemoveAt(i);
                //        i = i - 1 ;
                //    }


                //}


                int iCount = 0;
                while (iCount <= dtReturn.Rows.Count - 1)
                {
                    if (isRowEmpty(dtReturn.Rows[iCount]))
                    {
                        dtReturn.Rows.RemoveAt(iCount);
                    }
                    else
                    {
                        iCount += 1;
                    }
                }


                //now applying the filters

                //column ignore
                for (int i = IgnoreColumn.Count - 1; i >= 0; i--)
                {
                    dtReturn.Columns.RemoveAt((int)IgnoreColumn[i]);
                }

                //string ignore
                for (int i = IgnoreString.Count - 1; i >= 0; i--)
                {
                    for (int j = dtReturn.Rows.Count - 1; j >= 0; i--)
                    {
                        foreach (DataColumn dCol in dtReturn.Columns)
                        {
                            if (dtReturn.Rows[j][dCol.ColumnName].ToString().ToLower().Contains(IgnoreString[i].ToString().ToLower()))
                            {
                                dtReturn.Rows.RemoveAt(j);
                                break;
                            }
                        }
                    }
                }


                /* Hack to get rid of DateTime Columns */
                // added by zeemz
                dtTemp = dtReturn.Clone();
                dtTemp.Clear();
                foreach (DataColumn tempColumn in dtTemp.Columns)
                {
                //    if (tempColumn.DataType == typeof(DateTime))
//                    {
                        tempColumn.DataType = typeof(String);
  //                  }
                }
                foreach (DataRow tempRow in dtReturn.Rows)
                {
                    DataRow insRow = dtTemp.NewRow();
                    foreach (DataColumn tempColumn in dtReturn.Columns)
                    {

                        if (tempColumn.DataType == typeof(DateTime))
                        {
                            if (!String.IsNullOrEmpty(tempRow[tempColumn.ColumnName.ToString()].ToString()))
                            {
                                insRow[tempColumn.ColumnName.ToString()] = Convert.ToDateTime(tempRow[tempColumn.ColumnName.ToString()].ToString()).ToString("yyyyMMddhhmmss");
                            }
                            else
                            {
                                insRow[tempColumn.ColumnName.ToString()] = "";
                            }
                        }
                        else
                        {

                            insRow[tempColumn.ColumnName.ToString()] = tempRow[tempColumn.ColumnName.ToString()].ToString();
                        }

                    }
                    dtTemp.Rows.Add(insRow);
                }



            }
            catch (Exception ex)
            {
                throw ex;
            }

            return dtTemp;
        }

Answer 1

我碰巧发现当您手动或以编程方式修改 xlsx 文件时存在问题文件和Excel它自己可以修复损坏的文件按预期显示文件。