优化此 C# 代码的其他方法

本文关键字:其他 方法 代码 优化 | 更新日期: 2023-09-27 18:34:29

我创建了 mdb 记录到 sql 记录的转换,对我来说它工作正常。但是当记录大得多时,速度会很慢。我认为还有其他方法可以优化记录的解析和提取?

using System;
using System.Collections;
using System.Collections.Generic;
using System.IO;
using System.Text;
using System.Data;
using System.Data.OleDb;    
namespace conMDBToSQL
{
    class Program
    {
        static void Main()
        {
            // string sFileName = "C:''sqlrecords.mdb";
            string sFileName = "C:''sample.mdb";
            OleDbConnection _con = new OleDbConnection( @"Provider=Microsoft.Jet.OLEDB.4.0;Data Source=" + sFileName + ";Mode=Share Exclusive;User Id=admin;Password=;" );
            try
            {
                _con.Open();
                string sTableName = "users";
                string strSQL = String.Empty;
                strSQL = "SELECT * FROM " + sTableName + " ORDER BY 1";
                OleDbCommand _cmd = new OleDbCommand( strSQL, _con );
                _cmd.CommandType = CommandType.Text;
                OleDbDataReader _dr = _cmd.ExecuteReader( CommandBehavior.SchemaOnly );
                ArrayList _Arr = new ArrayList();

                DataTable dt = _dr.GetSchemaTable();
                foreach ( DataRow drow in dt.Rows )
                {
                    // Console.Write( drow[0].ToString() + "'n" );
                    _Arr.Add( drow[0].ToString() );
                }
                _dr.Close();
                _cmd.Dispose();

                strSQL = "SELECT * FROM " + sTableName + " ORDER BY 1";
                _cmd = new OleDbCommand(strSQL, _con);
                _cmd.CommandType = CommandType.Text;
                _dr = _cmd.ExecuteReader();
                string s_fields = String.Empty;
                string s_values = String.Empty;
                int arr_count = _Arr.Count - 1;
                while (_dr.Read()) 
                {
                    for(int i_a = 0; i_a <= arr_count;i_a++)
                    {
                        if (i_a <= arr_count)
                        {
                            s_fields += _Arr[i_a].ToString() + ",";
                            s_values += "'" + _dr[i_a].ToString() + "',";
                        }
                        else
                        {
                            s_fields += _Arr[i_a].ToString();
                            s_values += "'" + _dr[i_a].ToString() + "'";
                        }
                    }
                    // Build, query
                    strSQL = String.Empty;
                    strSQL = "INSERT INTO " + sTableName + "("+ s_fields +") VALUES ("+ s_values +")";
                    // Write, text file
                    StreamWriter sw = new StreamWriter(sFileName, true, Encoding.UTF8);
                    sw.Write(strSQL);
                    sw.Close();
                    sw = null;
                }

                Console.WriteLine("'n");
                Console.WriteLine("Finished..");
                Console.ReadKey();
            }
            catch ( OleDbException olex )
            {
                throw olex;
            }
            finally 
            {
                _con.Close();
            }
        }
    }
}

谢谢

优化此 C# 代码的其他方法

SqlBulkCopy 类帮助了我一次。链接在这里

我看到两个主要问题。首先,您应该在循环开始时打开StreamReader,并在完成后关闭它。喜欢这个:

using (StreamWriter sw = new StreamWriter(sFilename, false, Encoding.UTF8)
{
    while (dr.Read())
    {
        // convert here
        // Each line is written to the StreamWriter
    }
}

using结构将确保StreamWriter得到妥善关闭和处置。

第二个问题是将内容附加到字符串。不要附加字符串,而是使用 StringBuilder

StringBuilder s_fields = new StringBuilder();
StringBuilder s_values = new StringBuilder();
int arr_count = _Arr.Count - 1;
while (_dr.Read()) 
{
    for(int i_a = 0; i_a <= arr_count;i_a++)
    {
        if (i_a <= arr_count)
        {
            s_fields.AppendFormat("{0},", _Arr[i_a].ToString());
            s_values.AppendFormat("'{0}',", _dr[i_a].ToString());
        }
        else
        {
            s_fields.Append(_Arr[i_a].ToString());
            s_values.AppendFormat("'{0}'", _dr[i_a].ToString());
        }
    }
    strSQL = String.Format("INSERT INTO {0} ({1}) VALUES ({2})",
        sTableName, s_fields.ToString(), s_values.ToString());
    sw.WriteLine(strSQL);
}

这应该会大大提高您的表现。

我认为你的主要罪魁祸首是new StreamWriter(sFileName, true, Encoding.UTF8)行代码。对于每个数据行,您正在打开文件,添加单个命令,然后关闭文件。那将非常低效。

尝试使用 StringBuilder 而不是写入流,然后在读取循环之外,将使用StringBuilder生成的字符串仅写入文件一次。

我想你会发现这大大加快了速度。