转载自:http://developer.51cto.com/art/201302/380622.htm

方法一:OleDb: 用这种方法读取Excel速度还是非常的快的,但这种方式读取数据的时候不太灵活,不过可以在 DataTable 中对数据进行一些删减修改。

优点:读取方式简单、读取速度快

缺点:除了读取过程不太灵活之外,这种读取方式还有个弊端就是,当Excel数据量很大时。会非常占用内存,当内存不够时会抛出内存溢出的异常。

不过一般情况下还是非常不错的。

DataTable GetDataFromExcelByConn(bool hasTitle = false)
{
OpenFileDialog openFile = new OpenFileDialog();
openFile.Filter = "Excel(*.xlsx)|*.xlsx|Excel(*.xls)|*.xls";
openFile.InitialDirectory = Environment.GetFolderPath(Environment.SpecialFolder.Desktop);
openFile.Multiselect = false;
if (openFile.ShowDialog() == DialogResult.Cancel) return null;
var filePath = openFile.FileName;
string fileType = System.IO.Path.GetExtension(filePath);
if (string.IsNullOrEmpty(fileType)) return null; using (DataSet ds = new DataSet())
{
string strCon = string.Format("Provider=Microsoft.Jet.OLEDB.{0}.0;" +
"Extended Properties=\"Excel {1}.0;HDR={2};IMEX=1;\";" +
"data source={3};",
(fileType == ".xls" ? : ), (fileType == ".xls" ? : ), (hasTitle ? "Yes" : "NO"), filePath);
string strCom = " SELECT * FROM [Sheet1$]";
using (OleDbConnection myConn = new OleDbConnection(strCon))
using (OleDbDataAdapter myCommand = new OleDbDataAdapter(strCom, myConn))
{
myConn.Open();
myCommand.Fill(ds);
}
if (ds == null || ds.Tables.Count <= ) return null;
return ds.Tables[];
}
}

方法二:Com组件的方式读取Excel

这种方式需要先引用 Microsoft.Office.Interop.Excel 。首选说下这种方式的优缺点

优点:可以非常灵活的读取Excel中的数据

缺点:如果是Web站点部署在IIS上时,还需要服务器机子已安装了Excel,有时候还需要为配置IIS权限。最重要的一点因为是基于单元格方式读取的,所以数据很慢(曾做过试验,直接读取千行、200多列的文件,直接读取耗时15分钟。即使采用多线程分段读取来提高CPU的利用率也需要8分钟。PS:CPU I3)

需要读取大文件的的童鞋们慎重。。。

DataTable GetDataFromExcelByCom(bool hasTitle = false)
{
OpenFileDialog openFile = new OpenFileDialog();
openFile.Filter = "Excel(*.xlsx)|*.xlsx|Excel(*.xls)|*.xls";
openFile.InitialDirectory = Environment.GetFolderPath(Environment.SpecialFolder.Desktop);
openFile.Multiselect = false;
if (openFile.ShowDialog() == DialogResult.Cancel) return null;
var excelFilePath = openFile.FileName; Excel.Application app = new Excel.Application();
Excel.Sheets sheets;
object oMissiong = System.Reflection.Missing.Value;
Excel.Workbook workbook = null;
DataTable dt = new DataTable(); try
{
if (app == null) return null;
workbook = app.Workbooks.Open(excelFilePath, oMissiong, oMissiong, oMissiong, oMissiong, oMissiong,
oMissiong, oMissiong, oMissiong, oMissiong, oMissiong, oMissiong, oMissiong, oMissiong, oMissiong);
sheets = workbook.Worksheets; //将数据读入到DataTable中
Excel.Worksheet worksheet = (Excel.Worksheet)sheets.get_Item();//读取第一张表
if (worksheet == null) return null; int iRowCount = worksheet.UsedRange.Rows.Count;
int iColCount = worksheet.UsedRange.Columns.Count;
//生成列头
for (int i = ; i < iColCount; i++)
{
var name = "column" + i;
if (hasTitle)
{
var txt = ((Excel.Range)worksheet.Cells[, i + ]).Text.ToString();
if (!string.IsNullOrWhiteSpace(txt)) name = txt;
}
while (dt.Columns.Contains(name)) name = name + "_1";//重复行名称会报错。
dt.Columns.Add(new DataColumn(name, typeof(string)));
}
//生成行数据
Excel.Range range;
int rowIdx = hasTitle ? : ;
for (int iRow = rowIdx; iRow <= iRowCount; iRow++)
{
DataRow dr = dt.NewRow();
for (int iCol = ; iCol <= iColCount; iCol++)
{
range = (Excel.Range)worksheet.Cells[iRow, iCol];
dr[iCol - ] = (range.Value2 == null) ? "" : range.Text.ToString();
}
dt.Rows.Add(dr);
}
return dt;
}
catch { return null; }
finally
{
workbook.Close(false, oMissiong, oMissiong);
System.Runtime.InteropServices.Marshal.ReleaseComObject(workbook);
workbook = null;
app.Workbooks.Close();
app.Quit();
System.Runtime.InteropServices.Marshal.ReleaseComObject(app);
app = null;
}
}

原文的方法二还提供了多线程处理数据的代码,一并复制到此(此处出现了一个SheetOptions的类型,无法考证其来源,如果知晓,请留言,谢谢。):

/// <summary>
/// 使用COM,多线程读取Excel(1 主线程、4 副线程)
/// </summary>
/// <param name="excelFilePath">路径</param>
/// <returns>DataTabel</returns>
public System.Data.DataTable ThreadReadExcel(string excelFilePath)
{
Excel.Application app = new Excel.Application();
Excel.Sheets sheets = null;
Excel.Workbook workbook = null;
object oMissiong = System.Reflection.Missing.Value;
System.Data.DataTable dt = new System.Data.DataTable(); try
{
if (app == null)
{
return null;
} workbook = app.Workbooks.Open(excelFilePath, oMissiong, oMissiong, oMissiong, oMissiong, oMissiong, oMissiong,
oMissiong, oMissiong, oMissiong, oMissiong, oMissiong, oMissiong, oMissiong, oMissiong); //将数据读入到DataTable中——Start
sheets = workbook.Worksheets;
Excel.Worksheet worksheet = (Excel.Worksheet)sheets.get_Item();//读取第一张表
if (worksheet == null)
return null; string cellContent;
int iRowCount = worksheet.UsedRange.Rows.Count;
int iColCount = worksheet.UsedRange.Columns.Count;
Excel.Range range; //负责列头Start
DataColumn dc;
int ColumnID = ;
range = (Excel.Range)worksheet.Cells[, ];
//while (range.Text.ToString().Trim() != "")
while (iColCount >= ColumnID)
{
dc = new DataColumn();
dc.DataType = System.Type.GetType("System.String"); string strNewColumnName = range.Text.ToString().Trim();
if (strNewColumnName.Length == ) strNewColumnName = "_1";
//判断列名是否重复
for (int i = ; i < ColumnID; i++)
{
if (dt.Columns[i - ].ColumnName == strNewColumnName)
strNewColumnName = strNewColumnName + "_1";
} dc.ColumnName = strNewColumnName;
dt.Columns.Add(dc); range = (Excel.Range)worksheet.Cells[, ++ColumnID];
}
//End //数据大于500条,使用多进程进行读取数据
if (iRowCount - > )
{
//开始多线程读取数据
//新建线程
int b2 = (iRowCount - ) / ;
DataTable dt1 = new DataTable("dt1");
dt1 = dt.Clone();
SheetOptions sheet1thread = new SheetOptions(worksheet, iColCount, , b2 + , dt1);
Thread othread1 = new Thread(new ThreadStart(sheet1thread.SheetToDataTable));
othread1.Start(); //阻塞 1 毫秒,保证第一个读取 dt1
Thread.Sleep(); DataTable dt2 = new DataTable("dt2");
dt2 = dt.Clone();
SheetOptions sheet2thread = new SheetOptions(worksheet, iColCount, b2 + , b2 * + , dt2);
Thread othread2 = new Thread(new ThreadStart(sheet2thread.SheetToDataTable));
othread2.Start(); DataTable dt3 = new DataTable("dt3");
dt3 = dt.Clone();
SheetOptions sheet3thread = new SheetOptions(worksheet, iColCount, b2 * + , b2 * + , dt3);
Thread othread3 = new Thread(new ThreadStart(sheet3thread.SheetToDataTable));
othread3.Start(); DataTable dt4 = new DataTable("dt4");
dt4 = dt.Clone();
SheetOptions sheet4thread = new SheetOptions(worksheet, iColCount, b2 * + , b2 * + , dt4);
Thread othread4 = new Thread(new ThreadStart(sheet4thread.SheetToDataTable));
othread4.Start(); //主线程读取剩余数据
for (int iRow = b2 * + ; iRow <= iRowCount; iRow++)
{
DataRow dr = dt.NewRow();
for (int iCol = ; iCol <= iColCount; iCol++)
{
range = (Excel.Range)worksheet.Cells[iRow, iCol];
cellContent = (range.Value2 == null) ? "" : range.Text.ToString();
dr[iCol - ] = cellContent;
}
dt.Rows.Add(dr);
} othread1.Join();
othread2.Join();
othread3.Join();
othread4.Join(); //将多个线程读取出来的数据追加至 dt1 后面
foreach (DataRow dr in dt.Rows)
dt1.Rows.Add(dr.ItemArray);
dt.Clear();
dt.Dispose(); foreach (DataRow dr in dt2.Rows)
dt1.Rows.Add(dr.ItemArray);
dt2.Clear();
dt2.Dispose(); foreach (DataRow dr in dt3.Rows)
dt1.Rows.Add(dr.ItemArray);
dt3.Clear();
dt3.Dispose(); foreach (DataRow dr in dt4.Rows)
dt1.Rows.Add(dr.ItemArray);
dt4.Clear();
dt4.Dispose(); return dt1;
}
else
{
for (int iRow = ; iRow <= iRowCount; iRow++)
{
DataRow dr = dt.NewRow();
for (int iCol = ; iCol <= iColCount; iCol++)
{
range = (Excel.Range)worksheet.Cells[iRow, iCol];
cellContent = (range.Value2 == null) ? "" : range.Text.ToString();
dr[iCol - ] = cellContent;
}
dt.Rows.Add(dr);
}
}
//将数据读入到DataTable中——End
return dt;
}
catch
{
return null;
}
finally
{
workbook.Close(false, oMissiong, oMissiong);
System.Runtime.InteropServices.Marshal.ReleaseComObject(workbook);
System.Runtime.InteropServices.Marshal.ReleaseComObject(sheets);
workbook = null;
app.Workbooks.Close();
app.Quit();
System.Runtime.InteropServices.Marshal.ReleaseComObject(app);
app = null;
GC.Collect();
GC.WaitForPendingFinalizers();
}
}

原文还提供了第三种方法,感兴趣的可以关心一下:

方法三:NPOI方式读取Excel,NPOI是一组开源的组件,类似Java的 POI。包括:NPOI、NPOI.HPSF、NPOI.HSSF、NPOI.HSSF.UserModel、NPOI.POIFS、NPOI.Util,下载的时候别只下一个噢

优点:读取Excel速度较快,读取方式操作灵活性

缺点:只支持03的Excel,xlsx的无法读取。由于这点,使用这种方式的人不多啊,没理由要求客户使用03版Excel吧,再说03版Excel对于行数还有限制,只支持65536行。

(听他们的开发人员说会在2012年底推出新版,支持xlsx的读取。但一直很忙没时间去关注这个事情,有兴趣的同学可以瞧瞧去)

using System;
using System.Data;
using System.IO;
using System.Web;
using NPOI;
using NPOI.HPSF;
using NPOI.HSSF;
using NPOI.HSSF.UserModel;
using NPOI.POIFS;
using NPOI.Util;
using System.Text;
using System.Configuration; public class NPOIHelper
{
private static int ExcelMaxRow = Convert.ToInt32(ConfigurationManager.AppSettings["ExcelMaxRow"]);
/// <summary>
/// 由DataSet导出Excel
/// </summary>
/// <param name="sourceTable">要导出数据的DataTable</param>
/// <param name="sheetName">工作表名称</param>
/// <returns>Excel工作表</returns>
private static Stream ExportDataSetToExcel(DataSet sourceDs)
{
HSSFWorkbook workbook = new HSSFWorkbook();
MemoryStream ms = new MemoryStream(); for (int i = ; i < sourceDs.Tables.Count; i++)
{
HSSFSheet sheet = (HSSFSheet)workbook.CreateSheet(sourceDs.Tables[i].TableName);
HSSFRow headerRow = (HSSFRow)sheet.CreateRow();
// handling header.
foreach (DataColumn column in sourceDs.Tables[i].Columns)
headerRow.CreateCell(column.Ordinal).SetCellValue(column.ColumnName);
// handling value.
int rowIndex = ;
foreach (DataRow row in sourceDs.Tables[i].Rows)
{
HSSFRow dataRow = (HSSFRow)sheet.CreateRow(rowIndex);
foreach (DataColumn column in sourceDs.Tables[i].Columns)
{
dataRow.CreateCell(column.Ordinal).SetCellValue(row[column].ToString());
}
rowIndex++;
}
}
workbook.Write(ms);
ms.Flush();
ms.Position = ;
workbook = null;
return ms;
}
/// <summary>
/// 由DataSet导出Excel
/// </summary>
/// <param name="sourceTable">要导出数据的DataTable</param>
/// <param name="fileName">指定Excel工作表名称</param>
/// <returns>Excel工作表</returns>
public static void ExportDataSetToExcel(DataSet sourceDs, string fileName)
{
//检查是否有Table数量超过65325
for (int t = ; t < sourceDs.Tables.Count; t++)
{
if (sourceDs.Tables[t].Rows.Count > ExcelMaxRow)
{
DataSet ds = GetdtGroup(sourceDs.Tables[t].Copy());
sourceDs.Tables.RemoveAt(t);
//将得到的ds插入 sourceDs中
for (int g = ; g < ds.Tables.Count; g++)
{
DataTable dt = ds.Tables[g].Copy();
sourceDs.Tables.Add(dt);
}
t--;
}
} MemoryStream ms = ExportDataSetToExcel(sourceDs) as MemoryStream;
HttpContext.Current.Response.AppendHeader("Content-Disposition", "attachment;filename=" + fileName);
HttpContext.Current.Response.BinaryWrite(ms.ToArray());
HttpContext.Current.ApplicationInstance.CompleteRequest();
//HttpContext.Current.Response.End();
ms.Close();
ms = null;
}
/// <summary>
/// 由DataTable导出Excel
/// </summary>
/// <param name="sourceTable">要导出数据的DataTable</param>
/// <returns>Excel工作表</returns>
private static Stream ExportDataTableToExcel(DataTable sourceTable)
{
HSSFWorkbook workbook = new HSSFWorkbook();
MemoryStream ms = new MemoryStream();
HSSFSheet sheet = (HSSFSheet)workbook.CreateSheet(sourceTable.TableName);
HSSFRow headerRow = (HSSFRow)sheet.CreateRow();
// handling header.
foreach (DataColumn column in sourceTable.Columns)
headerRow.CreateCell(column.Ordinal).SetCellValue(column.ColumnName);
// handling value.
int rowIndex = ;
foreach (DataRow row in sourceTable.Rows)
{
HSSFRow dataRow = (HSSFRow)sheet.CreateRow(rowIndex);
foreach (DataColumn column in sourceTable.Columns)
{
dataRow.CreateCell(column.Ordinal).SetCellValue(row[column].ToString());
}
rowIndex++;
}
workbook.Write(ms);
ms.Flush();
ms.Position = ;
sheet = null;
headerRow = null;
workbook = null;
return ms;
}
/// <summary>
/// 由DataTable导出Excel
/// </summary>
/// <param name="sourceTable">要导出数据的DataTable</param>
/// <param name="fileName">指定Excel工作表名称</param>
/// <returns>Excel工作表</returns>
public static void ExportDataTableToExcel(DataTable sourceTable, string fileName)
{
//如数据超过65325则分成多个Table导出
if (sourceTable.Rows.Count > ExcelMaxRow)
{
DataSet ds = GetdtGroup(sourceTable);
//导出DataSet
ExportDataSetToExcel(ds, fileName);
}
else
{
MemoryStream ms = ExportDataTableToExcel(sourceTable) as MemoryStream;
HttpContext.Current.Response.AppendHeader("Content-Disposition", "attachment;filename=" + fileName);
HttpContext.Current.Response.BinaryWrite(ms.ToArray());
HttpContext.Current.ApplicationInstance.CompleteRequest();
//HttpContext.Current.Response.End();
ms.Close();
ms = null;
}
} /// <summary>
/// 传入行数超过65325的Table,返回DataSet
/// </summary>
/// <param name="dt"></param>
/// <returns></returns>
public static DataSet GetdtGroup(DataTable dt)
{
string tablename = dt.TableName; DataSet ds = new DataSet();
ds.Tables.Add(dt); double n = dt.Rows.Count / Convert.ToDouble(ExcelMaxRow); //创建表
for (int i = ; i < n; i++)
{
DataTable dtAdd = dt.Clone();
dtAdd.TableName = tablename + "_" + i.ToString();
ds.Tables.Add(dtAdd);
} //分解数据
for (int i = ; i < ds.Tables.Count; i++)
{
//新表行数达到最大 或 基表数量不足
while (ds.Tables[i].Rows.Count != ExcelMaxRow && ds.Tables[].Rows.Count != ExcelMaxRow)
{
ds.Tables[i].Rows.Add(ds.Tables[].Rows[ExcelMaxRow].ItemArray);
ds.Tables[].Rows.RemoveAt(ExcelMaxRow); }
} return ds;
} /// <summary>
/// 由DataTable导出Excel
/// </summary>
/// <param name="sourceTable">要导出数据的DataTable</param>
/// <param name="fileName">指定Excel工作表名称</param>
/// <returns>Excel工作表</returns>
public static void ExportDataTableToExcelModel(DataTable sourceTable, string modelpath, string modelName, string fileName, string sheetName)
{
int rowIndex = ;//从第二行开始,因为前两行是模板里面的内容
int colIndex = ;
FileStream file = new FileStream(modelpath + modelName + ".xls", FileMode.Open, FileAccess.Read);//读入excel模板
HSSFWorkbook hssfworkbook = new HSSFWorkbook(file);
HSSFSheet sheet1 = (HSSFSheet)hssfworkbook.GetSheet("Sheet1");
sheet1.GetRow().GetCell().SetCellValue("excelTitle"); //设置表头
foreach (DataRow row in sourceTable.Rows)
{ //双循环写入sourceTable中的数据
rowIndex++;
colIndex = ;
HSSFRow xlsrow = (HSSFRow)sheet1.CreateRow(rowIndex);
foreach (DataColumn col in sourceTable.Columns)
{
xlsrow.CreateCell(colIndex).SetCellValue(row[col.ColumnName].ToString());
colIndex++;
}
}
sheet1.ForceFormulaRecalculation = true;
FileStream fileS = new FileStream(modelpath + fileName + ".xls", FileMode.Create);//保存
hssfworkbook.Write(fileS);
fileS.Close();
file.Close();
}
}

c# 读取excel数据的两种方法(转)的更多相关文章

  1. c# 读取excel数据的两种方法

    转载自:http://developer.51cto.com/art/201302/380622.htm, 方法一:OleDb: 用这种方法读取Excel速度还是非常的快的,但这种方式读取数据的时候不 ...

  2. 【转】c# 读取excel数据的两种方法

    源地址:http://www.cnblogs.com/icyJ/p/ReadExcel.html

  3. 读取Excel文件的两种方法

    第一种方法:传统方法,采用OleDB读取EXCEL文件, 优点:写法简单,缺点:服务器必须安有此组件才能用,不推荐使用 private DataSet GetConnect_DataSet2(stri ...

  4. 读取Excel文件的两种方法比较 以及用NPOI写入Excel

    1. 采用NPOI方式,只需引用NPOI.dll,但目前最高只能到2.4.0版. 缺点:只支持.xls,不支持.xlsx格式.github上的2.4.1版支持.xlsx,但总提示缺ICSharpCod ...

  5. SQLServer 批量插入数据的两种方法

    SQLServer 批量插入数据的两种方法-发布:dxy 字体:[增加 减小] 类型:转载 在SQL Server 中插入一条数据使用Insert语句,但是如果想要批量插入一堆数据的话,循环使用Ins ...

  6. Java构造和解析Json数据的两种方法详解二

    在www.json.org上公布了很多JAVA下的json构造和解析工具,其中org.json和json-lib比较简单,两者使用上差不多但还是有些区别.下面接着介绍用org.json构造和解析Jso ...

  7. php获取数组中重复数据的两种方法

    分享下php获取数组中重复数据的两种方法. 1,利用php提供的函数,array_unique和array_diff_assoc来实现 <?php function FetchRepeatMem ...

  8. C++读取字符串数据的两种方式

    C++读取字符串数据的两种方式 对于同样的样例输入: ladder came tape soon leader acme RIDE lone Dreis peat ScAlE orb eye Ride ...

  9. Java构造和解析Json数据的两种方法详解二——org.json

    转自:http://www.cnblogs.com/lanxuezaipiao/archive/2013/05/24/3096437.html 在www.json.org上公布了很多JAVA下的jso ...

随机推荐

  1. HDU5723 Abandoned country (最小生成树+深搜回溯法)

    Description An abandoned country has n(n≤100000) villages which are numbered from 1 to n. Since aban ...

  2. JS实现添加至购物车功能

    效果图展示: 当将书拖拽至购物车一览时: 首先将页面的基本结构写出来: <!DOCTYPE html> <html lang="en"> <head& ...

  3. VS2013+MFC串口控件的简单上位机

    因为做东西,正好用到这里.所以就上传了文件分享一下. 利用VS带的MFC库,用起来还是比较方便的.空间的程序构架都是自动生成的,具体的程序自己加进去就行. 里面有整个的工程 还带有一个生成的EXE文件 ...

  4. F数圈圈

    Description 幼儿园的小朋友对数字其实不是很感兴趣,他们更感兴趣的是形状,现在给你一个数字,小朋友都会数出其中一共有多少圆圈圈 Input 一个数字n长度不超过19位 Output 输出其中 ...

  5. 最短路和次短路问题,dijkstra算法

    /*  *题目大意:  *在一个有向图中,求从s到t两个点之间的最短路和比最短路长1的次短路的条数之和;  *  *算法思想:  *用A*求第K短路,目测会超时,直接在dijkstra算法上求次短路; ...

  6. 35. Search Insert Position【leetcode】

    Given a sorted array and a target value, return the index if the target is found. If not, return the ...

  7. 在App Store中无法更新xcode 8.3.2时 以及更新新系统10.12

    今天更新Xcode 遇到三个坑坑1 要想更新Xcode最新版本 需要先把电脑系统更新到10.12 坑2 电脑系统更新到10.12 需要更新Mac Sierra 就可以了坑3 在App Store中下载 ...

  8. 前端到后台ThinkPHP开发整站(1)

    1.前言: 我个人从来没有写过博客文章,作为一个程序员没有自己的博客算是一个合格的程序员,所以我地想想也要经营起一个的博客,做一个小项目,写这博客算就做这个项目的一个项目笔记吧!现在自学着ThinkP ...

  9. .NET并行处理和并发1-Threads and Theading

    线程是操作系统分配处理器时间的基本单元,并且进程中可以有多个线程同时执行代码. 每个线程都维护异常处理程序.调度优先级和一组系统用于在调度该线程前保存线程上下文的结构. 线程上下文包括为使线程在线程的 ...

  10. 表空间移动(transporting tablespaces)

    --表空间移动(transporting tablespaces) --------------------------------------2014/01/15   1. 表空间传输步骤简介.   ...