Demo 与实际应用开发之间的距离

本文以一则示例说明一般技术人员演示demo与实际应用开发之间的差距。

      假设情况如下：一个项目当中需要用linq to sql 来作为数据访问层，如果未用过Linq to SQL 的话，一般大家会以DEMO示例 说明一张表的增、删、改、查的情况。

第一步  建好数据库表，假设建表脚本为

代码

CREATE   TABLE   [ dbo ] . [ NetPay_Order ] (
     [ SequenceId ]   [ int ]   IDENTITY ( 1 , 1 )  NOT   NULL ,
     [ MerId ]   [ nvarchar ] ( 50 )  NOT   NULL ,
     [ MerDate ]   [ nvarchar ] ( 50 )  NOT   NULL ,
     [ MerSeqId ]   [ nvarchar ] ( 50 )  NULL ,
     [ CardNo ]   [ nvarchar ] ( 50 )  NOT   NULL ,
     [ UsrName ]   [ nvarchar ] ( 50 )  NOT   NULL ,
     [ OpenBank ]   [ nvarchar ] ( 50 )  NOT   NULL ,
     [ Prov ]   [ nvarchar ] ( 50 )  NOT   NULL ,
     [ City ]   [ nvarchar ] ( 50 )  NOT   NULL ,
     [ TransAmt ]   [ decimal ] ( 8 ,  2 )  NOT   NULL ,
     [ Purpose ]   [ nvarchar ] ( 100 )  NULL ,
     [ Flag ]   [ nvarchar ] ( 5 )  NULL ,
     [ SubBank ]   [ nvarchar ] ( 50 )  NULL ,
     [ Version ]   [ nvarchar ] ( 10 )  NULL ,
     [ Submited ]   [ bit ]   NULL ,
     [ SubmitedResult ]   [ nvarchar ] ( 50 )  NULL ,
     [ SubmitedTime ]   [ datetime ]   NULL ,
     [ responsed ]   [ bit ]   NULL ,
     [ responseCode ]   [ nvarchar ] ( 50 )  NULL ,
     [ responseTime ]   [ datetime ]   NULL ,
     [ oracontrasted ]   [ bit ]   NULL ,
     [ oracontrastedTime ]   [ datetime ]   NULL ,
     [ oracontrastedResult ]   [ nvarchar ] ( 50 )  NULL ,
     [ EticketNo ]   [ nvarchar ] ( 50 )  NULL ,
     [ State ]   [ nvarchar ] ( 50 )  NULL ,
  CONSTRAINT   [ PK_NetPay_Order ]   PRIMARY   KEY   CLUSTERED  
(
     [ SequenceId ]   ASC
) WITH  (PAD_INDEX   =   OFF , STATISTICS_NORECOMPUTE   =   OFF , IGNORE_DUP_KEY  =   OFF , ALLOW_ROW_LOCKS   =   ON , ALLOW_PAGE_LOCKS   =   ON )  ON   [ PRIMARY ]
)  ON   [ PRIMARY ]

其中SequenceId是自增序列

 

第二步   打开Visual Studio.NET 2010,新建控制台项目TestLinqtoSQL ，然后依次点击Tool->Connect to Database 输入刚刚建好的数据库相关信息，在IDE 左侧看到数据库相关信息

第三步  新建Linq to SQL Classes ，右击项目依次点击add ->add item->linq to sql classes 并命名为DataAccess，在第二步当中看到的数据库信息找到第一步建好的表，直接拖到新建的linq to SQL Classes 上来。

第四步  前期工作完成，现在我们可以开始写代码了，我们要做的新增一条数据，我们会这样写

代码

 DataAccessDataContext _context  =   new  DataAccessDataContext();
            NetPay_Order recorder  =   new  NetPay_Order();
            recorder.MerId  =   " 0001 " ;
            recorder.MerDate  =  System.DateTime.Now.ToString( " yyyyMMdd " );
            recorder.MerSeqId  =   null ;
            recorder.CardNo  =   " 0002 " ;
            recorder.UsrName  =   " 赵一村 " ;
            recorder.City  =   " 苏州 " ;
            recorder.Prov  =   " 江苏 " ;
            recorder.Purpose  =   " 转账 " ;
            recorder.EticketNo  =  System.Guid.NewGuid().ToString();
            recorder.OpenBank  =   " 招商银行 " ;
            recorder.Flag  =   " 00 " ; 
            recorder.Version  =   " 20100906 " ;
            recorder.Submited  =   false ;  
            recorder.responsed  =   false ;
            recorder.oracontrasted  =   false ;
            _context.NetPay_Orders.InsertOnSubmit(recorder);
            _context.SubmitChanges();

 这样子，新增一笔数据就搞定了，用linq to SQL新增数据就这么简单。

                   

                    可是这只能算是Demo 不足以实际就应用

遇到的问题1    批量操作与事务性，这里只是新增一笔数据，实际开发当中经常会有批量新增的事情，经常修改，代码会变成如下样子

   static   void  Main( string [] args)
        {
            DataAccessDataContext _context  =   new  DataAccessDataContext();
             foreach  (NetPay_Order recorder  in  getOrders())
            {           
                _context.NetPay_Orders.InsertOnSubmit(recorder);          
            } 
            _context.SubmitChanges();
        }
         private   static  NetPay_Order[] getOrders()
        {
            NetPay_Order[] recorders  =   new  NetPay_Order[ 2 ];
            NetPay_Order recorder1,recorder2;
            recorder1  =   new  NetPay_Order();
            recorder1.MerId  =   " 1001 " ;
            recorder1.MerDate  =  System.DateTime.Now.ToString( " yyyyMMdd " );
            recorder1.MerSeqId  =   null ;
            recorder1.CardNo  =   " 0002 " ;
            recorder1.UsrName  =   " 赵一村 " ;
            recorder1.City  =   " 苏州 " ;
            recorder1.Prov  =   " 江苏 " ;
            recorder1.Purpose  =   " 转账 " ;
            recorder1.EticketNo  =  System.Guid.NewGuid().ToString();
            recorder1.OpenBank  =   " 招商银行 " ;
            recorder1.Flag  =   " 00 " ;
            recorder1.Version  =   " 20100906 " ;
            recorder1.Submited  =   false ;
            recorder1.responsed  =   false ;
            recorder1.oracontrasted  =   false ;
            recorders[ 0 ]  =  recorder1;

            recorder2  =   new  NetPay_Order();
           //   recorder2.MerId = "1002";
            recorder2.MerDate  =  System.DateTime.Now.ToString( " yyyyMMdd " );
            recorder2.MerSeqId  =   null ;
            recorder2.CardNo  =   " 0002 " ;
            recorder2.UsrName  =   " 赵一村 " ;
            recorder2.City  =   " 苏州 " ;
            recorder2.Prov  =   " 江苏 " ;
            recorder2.Purpose  =   " 转账 " ;
            recorder2.EticketNo  =  System.Guid.NewGuid().ToString();
            recorder2.OpenBank  =   " 招商银行 " ;
            recorder2.Flag  =   " 00 " ;
            recorder2.Version  =   " 20100906 " ;
            recorder2.Submited  =   false ;
            recorder2.responsed  =   false ;
            recorder2.oracontrasted  =   false ;
            recorders[ 1 ]  =  recorder2;
             return  recorders;

        }

       其中DataAccessDataContext _context = new DataAccessDataContext();

            foreach (NetPay_Order recorder in getOrders())

            {          

                _context.NetPay_Orders.InsertOnSubmit(recorder);         

            }

            _context.SubmitChanges();

       这几段代码中畎认情况linq to SQL 会为你提供事务功能

 

遇到的问题2   假如MerSeqId字段是根据MerDate+ SequenceId 两字字段组合起来作为流水号呢，我们知道程序热行到_context.NetPay_Orders.InsertOnSubmit(recorder)这一句时并没有新增数据库，而是要_context.SubmitChanges()以后才可以更新数据库，可谓有真正更新数据库的时候SequenceId 字段才会真正有值，此时照平常想法，代码会变成如下样子

DataAccessDataContext _context = new DataAccessDataContext();

            foreach (NetPay_Order recorder in getOrders())

            {          

                _context.NetPay_Orders.InsertOnSubmit(recorder);

                _context.SubmitChanges();

                recorder.MerSeqId = recorder.MerDate.ToString() + recorder.SequenceId.ToString();

                _context.SubmitChanges();

            }

这样子_context.SubmitChanges()以后再更新MerSeqId字段值然后再SubmitChanges一次，到这里，好像问题已经解决了。

 

                                可是这只能算是DEMO,还不足以实际应用

 

遇到的问题3    像上面的解决方案失去了事务性了，如果对事务没有要求的话，这样子问题也不大，而如果这段保存操作是作为WCF作务客户端调用呢，问题就有了，客户端提交一批数据上来，服务要保证其事务性，为此我们又要对程序改造成如下样子

 DataAccessDataContext _context  =   new  DataAccessDataContext();
             if  (_context.Connection  !=   null )
                _context.Connection.Open();
            DbTransaction tran  =  _context.Connection.BeginTransaction();
             try
            {
                 foreach  (NetPay_Order recorder  in  getOrders())
                {
                    _context.NetPay_Orders.InsertOnSubmit(recorder);
                    _context.SubmitChanges();
                    recorder.MerSeqId  =  recorder.MerDate.ToString()  +  recorder.SequenceId.ToString();
                    _context.SubmitChanges();
                } 
            }
             catch  (Exception ex)
            {
                tran.Rollback();
            }
            

 

到这里已经解决了问题2，3了

 

 

 

转载于:https://www.cnblogs.com/zycblog/archive/2010/09/06/1819325.html