如何高效的將excel導入到oracle?和前兩天的SqlBulkCopy 導入到sqlserver對應,oracle也有自身的方法,只是稍微復雜些.
那就是使用oracle的sql*loader功能，而sqlldr只支持類似csv格式的數據，所以要自己把excel轉換一下。
實現步驟：
用com組件讀取excel-保存為csv格式-處理最後一個字段為null的情況和表頭-根據excel結構建表-生成sqlldr的控制文件-用sqlldr命令導入數據
這個性能雖然沒有sql的bcp快，但還是相當可觀的，在我機器上1萬多數據不到4秒，而且導入過程代碼比較簡單，也同樣沒有循環拼接sql插入那麼難以維護。
這裡也提個問題:處理csv文件的表頭和最後一個字段為null的情況是否可以優化？除了我代碼中的例子，我實在想不出其他辦法。
using System;
using System.Data;
using System.Text;
using System.Windows.Forms;
using Microsoft.Office.Interop.Excel;
using System.Data.OleDb;
//引用-com-microsoft excel objects 11.0
namespace WindowsApplication5
{
public partial class Form1 : Form
{
public Form1()
{
InitializeComponent();
}
/// <SUMMARY>
/// excel導入到oracle
/// </SUMMARY>
/// <PARAM name="excelFile">文件名</PARAM>
/// <PARAM name="sheetName">sheet名</PARAM>
/// <PARAM name="sqlplusString">oracle命令sqlplus連接串</PARAM>
public void TransferData(string excelFile, string sheetName, string sqlplusString)
{
string strTempDir = System.IO.Path.GetDirectoryName(excelFile);
string strFileName = System.IO.Path.GetFileNameWithoutExtension(excelFile);
string strCsvPath = strTempDir +"\\"+strFileName + ".csv";
string strCtlPath = strTempDir + "\\" + strFileName + ".Ctl";
string strSqlPath = strTempDir + "\\" + strFileName + ".Sql";
if (System.IO.File.Exists(strCsvPath))
System.IO.File.Delete(strCsvPath);
//獲取excel對象
Microsoft.Office.Interop.Excel.Application ObjExcel = new Microsoft.Office.Interop.Excel.Application();
Microsoft.Office.Interop.Excel.Workbook ObjWorkBook;
Microsoft.Office.Interop.Excel.Worksheet ObjWorkSheet = null;
ObjWorkBook = ObjExcel.Workbooks.Open(excelFile, Type.Missing, Type.Missing, Type.Missing, Type.Missing, Type.Missing, Type.Missing, Type.Missing, Type.Missing, Type.Missing, Type.Missing, Type.Missing, Type.Missing, Type.Missing, Type.Missing);
foreach (Microsoft.Office.Interop.Excel.Worksheet sheet in ObjWorkBook.Sheets)
{
if (sheet.Name.ToLower() == sheetName.ToLower())
{
ObjWorkSheet = sheet;
break;
}
}
if (ObjWorkSheet == null) throw new Exception(string.Format("{0} not found!!", sheetName));
//保存為csv臨時文件
ObjWorkSheet.SaveAs(strCsvPath, Microsoft.Office.Interop.Excel.XlFileFormat.xlCSV, Type.Missing, Type.Missing, false, false, false, Type.Missing, Type.Missing, false);
ObjWorkBook.Close(false, Type.Missing, Type.Missing);
ObjExcel.Quit();
//讀取csv文件，需要將表頭去掉，並且將最後一列為null的字段處理為顯示的null，否則oracle不會識別，這個步驟有沒有好的替換方法?
System.IO.StreamReader reader = new System.IO.StreamReader(strCsvPath,Encoding.GetEncoding("gb2312"));
string strAll = reader.ReadToEnd();
reader.Close();
string strData = strAll.Substring(strAll.IndexOf("\r\n") + 2).Replace(",\r\n",",Null");
byte[] bytes = System.Text.Encoding.Default.GetBytes(strData);
System.IO.Stream ms = System.IO.File.Create(strCsvPath);
ms.Write(bytes, 0, bytes.Length);
ms.Close();
//獲取excel表結構
string strConn = "Provider=Microsoft.Jet.OLEDB.4.0;" + "Data Source=" + excelFile + ";" + "Extended Properties=Excel 8.0;";
OleDbConnection conn = new OleDbConnection(strConn);
conn.Open();
System.Data.DataTable table = conn.GetOleDbSchemaTable(System.Data.OleDb.OleDbSchemaGuid.Columns,
new object[] { null, null, sheetName+"$", null });
//生成sqlldr用到的控制文件,文件結構參考sql*loader功能，本示例已逗號分隔csv，數據帶逗號的用引號括起來。
string strControl = "load data\r\ninfile '{0}' \r\nappend into table {1}\r\n"+
"FIELDS TERMINATED BY ',' OPTIONALLY ENCLOSED BY '\"'\r\n(";
strControl = string.Format(strControl, strCsvPath,sheetName);
foreach (System.Data.DataRow drowColumns in table.Select("1=1", "Ordinal_Position"))
{
strControl += drowColumns["Column_Name"].ToString() + ",";
}
strControl = strControl.Substring(0, strControl.Length - 1) + ")";
bytes=System.Text.Encoding.Default.GetBytes(strControl);
ms= System.IO.File.Create(strCtlPath);
ms.Write(bytes, 0, bytes.Length);
ms.Close();
//生成初始化oracle表結構的文件
string strSql = @"drop table {0};
create table {0}
(";
strSql = string.Format(strSql, sheetName);
foreach (System.Data.DataRow drowColumns in table.Select("1=1", "Ordinal_Position"))
{
strSql += drowColumns["Column_Name"].ToString() + " varchar2(255),";
}
strSql = strSql.Substring(0, strSql.Length - 1) + ");\r\nexit;";
bytes = System.Text.Encoding.Default.GetBytes(strSql);
ms = System.IO.File.Create(strSqlPath);
ms.Write(bytes, 0, bytes.Length);
ms.Close();
//運行sqlplus，初始化表
System.Diagnostics.Process p = new System.Diagnostics.Process();
p.StartInfo = new System.Diagnostics.ProcessStartInfo();
p.StartInfo.FileName = "sqlplus";
p.StartInfo.Arguments = string.Format("{0} @{1}", sqlplusString, strSqlPath);
p.StartInfo.WindowStyle = System.Diagnostics.ProcessWindowStyle.Hidden;
p.StartInfo.UseShellExecute = false;
p.StartInfo.CreateNoWindow = true;
p.Start();
p.WaitForExit();
//運行sqlldr，導入數據
p = new System.Diagnostics.Process();
p.StartInfo = new System.Diagnostics.ProcessStartInfo();
p.StartInfo.FileName = "sqlldr";
p.StartInfo.Arguments = string.Format("{0} {1}", sqlplusString, strCtlPath);
p.StartInfo.WindowStyle = System.Diagnostics.ProcessWindowStyle.Hidden;
p.StartInfo.RedirectStandardOutput = true;
p.StartInfo.UseShellExecute = false;
p.StartInfo.CreateNoWindow = true;
p.Start();
System.IO.StreamReader r = p.StandardOutput;//截取輸出流
string line = r.ReadLine();//每次讀取一行
textBox3.Text += line + "\r\n";
while (!r.EndOfStream)
{
line = r.ReadLine();
textBox3.Text += line + "\r\n";
textBox3.Update();
}
p.WaitForExit();
//可以自行解決掉臨時文件csv，ctl和sql，代碼略去
}
private void button1_Click(object sender, EventArgs e)
{
TransferData(@"D:\test.xls", "Sheet1", "username/password@servicename");
}
}
}
