MSSQL教程之SQL 以日期静态更新保护的数据,一周排程...

其体积小、速度快、总体拥有成本低，尤其是开放源码这一特点，许多中小型网站为了降低网站总体拥有成本而选择了MySQL作为网站数据库。静态|计划|数据
以保护日期的分歧而改动显现内容.现在天是21号.保护了七天的数据即21--28.当在22号翻开保护界面的时分,保护内容显现为22--29号的数据和日期.
创立保护表:..

-----------------创立数据纪录表-------------------------

ifexists(select*fromdbo.sysobjectswhereid=object_id(N[dbo].[Pdl_WeekProcedure])andOBJECTPROPERTY(id,NIsUserTable)=1)

droptablePdl_WeekProcedure

GO

ifnotexists(select*fromdbo.sysobjectswhereid=object_id(N[dbo].[Pdl_WeekProcedure])andOBJECTPROPERTY(id,NIsUserTable)=1)

BEGIN

CreateTablePdl_WeekProcedure(

[Id]intIdentity(1,1)NotNull,

[DateId]datetimeNotNull,

[Partnum]varchar(20)Null,

[Revision]varchar(5)NullDEFAULT(0),

[Todate]datetimeNull,

[Qnty1]intNullDefault(0),

[Qnty2]intNullDefault(0),

[Qnty3]intNullDefault(0),

[Qnty4]intNullDefault(0),

[Qnty5]intNullDefault(0),

[Qnty6]intNullDefault(0),

[Qnty7]intNullDefault(0)--,

--[WeekDate]datetime

CONSTRAINT[PK_WeekProcedure]PRIMARYKEYCLUSTERED

(

[Id]

)ON[PRIMARY]

)ON[PRIMARY]



End

-----------------创立日期表--------------------

ifexists(select*fromdbo.sysobjectswhereid=object_id(N[dbo].[Pdl_WeekProcedureDate])andOBJECTPROPERTY(id,NIsUserTable)=1)

droptablePdl_WeekProcedureDate

GO

ifnotexists(select*fromdbo.sysobjectswhereid=object_id(N[dbo].[Pdl_WeekProcedureDate])andOBJECTPROPERTY(id,NIsUserTable)=1)

BEGIN

CreateTablePdl_WeekProcedureDate(

[Id]intIdentity(1,1)NotNull,

[DateId]datetimeNotNull,

[Date1]datetimeNull,

[Date2]datetimeNull,

[Date3]datetimeNull,

[Date4]datetimeNull,

[Date5]datetimeNull,

[Date6]datetimeNull,

[Date7]datetimeNull

CONSTRAINT[PK_WeekProcedureDate]PRIMARYKEYCLUSTERED

(

[Id]

)ON[PRIMARY]

)ON[PRIMARY]

End

-----------创立历一周交货排程史表----------------------------------

ifexists(select*fromdbo.sysobjectswhereid=object_id(N[dbo].[Pdl_WeekProcHistory])andOBJECTPROPERTY(id,NIsUserTable)=1)

droptablePdl_WeekProcHistory

GO

ifnotexists(select*fromdbo.sysobjectswhereid=object_id(N[dbo].[Pdl_WeekProcHistory])andOBJECTPROPERTY(id,NIsUserTable)=1)

BEGIN

CreateTablePdl_WeekProcHistory(

[Id]intIdentity(1,1)NotNull,

[DateId]datetimeNotNull,

[Partnum]varchar(20)Null,

[Revision]varchar(5)NullDEFAULT(0),

[Qnty1]intNullDefault(0),

[Qnty2]intNullDefault(0),

[Qnty3]intNullDefault(0),

[Qnty4]intNullDefault(0),

[Qnty5]intNullDefault(0),

[Qnty6]intNullDefault(0),

[Qnty7]intNullDefault(0),

[Date1]datetimeNull,

[Date2]datetimeNull,

[Date3]datetimeNull,

[Date4]datetimeNull,

[Date5]datetimeNull,

[Date6]datetimeNull,

[Date7]datetimeNull,

[IOTime]datetimeNotNullDefault(Getdate()),

[OutPutDate]datetimeNull

CONSTRAINT[PK_WeekProcHistory]PRIMARYKEYCLUSTERED

(

[Id]

)ON[PRIMARY]

)ON[PRIMARY]

End








更新保护表日期数据:
/*

ProcedureName:一周交货排程数据更新

AuthorName:lyf

Date:01/13/2005

*/



AlterProcedurePdl_WeekProcDateUpdateAs



Declare@numint,@num1int,@num2int,@Countint

Declare@Strvarchar(4000)

Declare@DateIddatetime,@Date1datetime,@Date2datetime,@Date3datetime,@Date4datetime,

@Date5datetime,@Date6datetime,@Date7datetime



Select@DateId=Convert(datetime,Convert(varchar,Getdate(),101))

Select@Date1=Convert(datetime,Convert(varchar,Getdate(),101))

Select@Date2=@Date1+1,@Date3=@Date1+2,@Date4=@Date1+3,

@Date5=@Date1+4,@Date6=@Date1+5,@Date7=@Date1+6



---------------修改更新为当天日期后向后推7天--------

IFNotExists(Select*FromPdl_WeekProcedureDate

WhereConvert(datetime,Convert(varchar,DateId,101))=Convert(datetime,Convert(varchar,@DateId,101)))

InsertIntoPdl_WeekProcedureDate(DateId,Date1,Date2,Date3,Date4,Date5,Date6,Date7)

Values(@DateId,@Date1,@Date2,@Date3,@Date4,@Date5,@Date6,@Date7)



----------将表Pdl_WeekProcedure里的数据换算成当天最新数据"对应程序保护接口"—

Select@num=Isnull(DatedIFF(day,(SelectMin(DateId)FromPdl_WeekProcedure),

(SelectMax(DateId)FromPdl_WeekProcedureDate)),0)



IF@num>0

Begin

Set@Count=1

Set@num1=@num

Set@num2=@num

Set@num=7-@num

Set@Str=



While@Count<=@num

Begin

Select@Str=@Str+Qnty+Convert(varchar,@Count)+=Qnty+Convert(varchar,@num1+1)+,

Set@Count=@Count+1

Set@num1=@num1+1

End



While@num2>0

Begin

Select@Str=@Str+Qnty+Convert(varchar,(7-@num2+1))+=0,

Set@num2=@num2-1

End

Select@str=Left(@str,Len(@str)-1)



--------------拔出一周交货排程保护汗青------------

InsertIntoPdl_WeekProcHistory(DateId,Partnum,Revision,Qnty1,Qnty2,Qnty3,Qnty4,Qnty5,Qnty6,Qnty7,

Date1,Date2,Date3,Date4,Date5,Date6,Date7,IOTime,OutPutDate)

Selectt1.DateId,t1.Partnum,t1.Revision,t1.Qnty1,t1.Qnty2,t1.Qnty3,t1.Qnty4,t1.Qnty5,t1.Qnty6,t1.Qnty7,

t2.Date1,t2.Date2,t2.Date3,t2.Date4,t2.Date5,t2.Date6,t2.Date7,Getdate(),t1.Todate

FromPdl_WeekProceduret1(nolock),Pdl_WeekProcedureDatet2(nolock)

Wheret1.Dateid=t2.DateId



------------------------更新Pdl_WeekProcedure数据----------------------------

Exec(UpdatePdl_WeekProcedureSet+@Str)



UpdatePdl_WeekProcedure

SetDateId=(SelectMax(DateId)FromPdl_WeekProcedureDate)



-------------------------更新当天出货工夫字段--------------------------

UpdatePdl_WeekProcedureSetTodate=Null

WhereConvert(datetime,Convert(varchar,Todate,101))Convert(datetime,Convert(varchar,Getdate(),101))

End



--deletePdl_WeekProcedureDatewheredateid=2005-01-1500:00:00.000

--UpdatePdl_WeekProceduresetdateid=convert(datetime,convert(varchar,getdate(),101))-1


天生最初数据汇合:

/*

ProcedureName:一周交货排程报表

AuthorName:lyf

Date:01/18/2005

*/

--execPdl_WeekDeliveryProcedure,,,,,



AlterProcedurePdl_WeekDeliveryProcedure

@InBPartnumvarchar(20)=,--DH024-030,--,MH075-013

@InEPartnumvarchar(20)=,--DH024-030,--,MH075-013

@InBRevisionvarchar(5)=,

@InERevisionvarchar(5)=,

@BDatevarchar(100),--datetime,--=01/18/2005,

@EDatevarchar(100),--datetime--=01/21/2005

@Flageint

As



Declare@BDateTimedatetime,@EDateTimedatetime



--Set@BDate=isnull(@BDate,getdate())

--Set@EDate=isnull(@EDate,getdate()+6)



-------------挪用更新保护记录表数据存储历程-------------------

ExecPdl_WeekProcDateUpdate



Select@BDateTime=convert(datetime,Convert(varchar,Getdate(),101))

Select@EDateTime=convert(datetime,Convert(varchar,@BDateTime+7,101))



/*取周出货形态数据表布局*/

Selecttop0PartNum,Revision,Qnty=0,WeekDate=DateId,Com=A9--=Convert(varchar(5),WeekDate,101)

into#WeekConfig

FromPdl_WeekProcedure(nolock)



/*在成品数据*/

selectt1.Partnum,t1.Revision,

Qnty=sum(caset1.POP

when0

thenceiling(1*t1.QNTY/(power(ISNULL(t3.LPiece,1),caset1.POPwhen0then0else1end)*

power(ISNULL(t3.LLPiece,1),caset1.POPwhen2then1else0end)))

when1

thenceiling(1*t1.QNTY*power(ISNULL(t3.LPiece,1),caset1.POPwhen0then1else0end)/

power(ISNULL(t3.LLPiece,1),caset1.POPwhen2then1else0end))

when2

then(1*t1.QNTY*power(ISNULL(t3.LPiece,1),caset1.POPwhen0then1else0end)*

power(ISNULL(t3.LLPiece,1),caset1.POPwhen2then0else1end))

end),DelDate=min(t4.ExpStkDate),UPP=convert(varchar,t3.LPiece)+*+convert(varchar,t3.LLPiece),

t5.Id,t5.name

Into#tmpStknwip

fromStknwipt1(nolock),

LotInfot2(nolock),

ProdBasict3(nolock),

OrderDetailt4(nolock),

PDL_ProcGroupt5(nolock),

PDL_ProcGroupDetailt6(nolock)

wheret1.LotNum=t2.LotNum

andt1.Layer=t2.Layer

andt2.IsInSC1

andt1.Qnty>0

andt1.ProcCodenotin(990,000)

andt1.Partnum=t3.Partnum

andt1.Revision=t3.Revision

--andt1.Layer=t3.Layer

andt2.PoNum*=t4.PoNum

andt2.Poitem*=t4.SerialNum

andt1.ProcCode=t6.ProcCode

andt5.Id=t6.Id

and((t1.PartNum>=@InBPartnumor@InBPartnum=)

and(t1.PartNum<=@InEPartnumor@InEPartnum=))

and((t1.Revision=@InBRevisionor@InBRevision=)

and(t1.Revision=@InERevisionor@InERevision=))

GroupByt1.Partnum,t1.Revision,

t3.LPiece,t3.LLPiece,t5.Id,t5.name

OrderByt5.Id,t1.partnum



/*wip完全性天生*/

Declare@CheckPartNumvarchar(20),@CheckRevisionvarchar(5)

Declare@WipProcCodevarchar(20),@WipIdvarchar(5)



SelectTop1@CheckPartNum=Partnum,@CheckRevision=RevisionFrom#tmpStknwip



DECLARECheckWipINSENSITIVECURSOR

FOR

SelectProcName=RTrim(LTrim(t1.Name)),Id=RTrim(LTrim(t1.Id))

FromPDL_ProcGroupt1(nolock),PDL_ProcGroupDetailt2(nolock)

Wheret1.Id=t2.Id

GroupByt1.Id,t1.Name

OPENCheckWip

BEGIN

FETCHNEXTFROMCheckWipINTO@WipProcCode,@WipId

WHILE@@FETCH_STATUS=0

BEGIN

IFNotExists(Select*From#tmpStknwipWhereName=@WipProcCodeandId=@WipId)

InsertInto#tmpStknwip(Partnum,Revision,Qnty,Name,Id)

Select@CheckPartNum,@CheckRevision,0,@WipProcCode,@WipId



FETCHNEXTFROMCheckWipINTO@WipProcCode,@WipId

END

END

CLOSECheckWip

DEALLOCATECheckWip



/*算计在成品数目*/

SelectPartnum,Revision,Qnty=sum(Qnty),

DelDate,UPP,ProcCode=Name,Id

Into#Stknwip

From#tmpStknwip

GroupByPartnum,Revision,DelDate,UPP,Id,Name

OrderByID



/*判别完全性*/

IF((NotExists(Select*from#WeekConfig))and(NotExists(Select*from#Stknwip)))

Begin

InsertInto#Stknwip(Partnum,Revision,Qnty,ProcCode,Id)

Selectnotdata,no,0,ProcCode=Name,Id

FromPDL_ProcGroup



InsertInto#WeekConfig(PartNum,Revision,Qnty,WeekDate,Com)

Selectnotdata,no,0,@BDateTime,A9

End

Else

Begin

IFNotExists(Select*from#Stknwip)

InsertInto#Stknwip(Partnum,Revision,Qnty,ProcCode,Id)

Selectnotdata,no,0,ProcCode=Name,Id

FromPDL_ProcGroup



IFNotExists(Select*From#WeekConfig)

InsertInto#WeekConfig(PartNum,Revision,Qnty,WeekDate,Com)

SelectTop1t1.Partnum,t1.Revision,Qnty=0,@BDateTime,A9

FromStknwipt1(nolock),LotInfot2(nolock)

Wheret1.LotNum=t2.LotNum

andt1.Layer=t2.Layer

andt2.IsInSC1

andt1.Qnty>0

andt1.ProcCodenotin(990,000)

and((t1.PartNum>=@InBPartnumor@InBPartnum=)

and(t1.PartNum<=@InEPartnumor@InEPartnum=))

and((t1.Revision=@InBRevisionor@InBRevision=)

and(t1.Revision=@InERevisionor@InERevision=))

End



/*天生出货形态工夫格局和完全性天生*/

Declare@Partnumvarchar(20),@Revisionvarchar(5),@Countint,@DateCountint

Declare@WeekDatedatetime,@Comvarchar(200)

Select@Com=A



DECLAREAddDateListINSENSITIVECURSOR

FOR

SelectPartnum,RevisionFrom#WeekConfig

GroupbyPartnum,Revision

OPENAddDateList

BEGIN

FETCHNEXTFROMAddDateListINTO@Partnum,@Revision

WHILE@@FETCH_STATUS=0

BEGIN

Select@Count=Count(WeekDate)From#WeekConfig

WherePartnum=@PartnumandRevision=@RevisionandWeekDate>=@BDateTimeandWeekDate<=@EDateTime



Select@DateCount=0



IF@Count<7

Begin



While@DateCount<7

Begin



Select@WeekDate=convert(datetime,Convert(varchar,(@BDateTime+@DateCount),101))



IFNotExists(Select*From#WeekConfig

WherePartnum=@PartnumandRevision=@Revision

andconvert(datetime,Convert(varchar,WeekDate,101))=@WeekDate

)

Begin

InsertInto#WeekConfig(partnum,revision,qnty,weekdate,Com)

Values(@Partnum,@Revision,0,convert(datetime,Convert(varchar,(@BDateTime+@DateCount),101)),

@Com+Convert(varchar,@DateCount))

Select@DateCount=@DateCount+1

End

Else

Begin

Update#WeekConfigSetCom=@Com+Convert(varchar,@DateCount)

WherePartnum=@PartnumandRevision=@Revision

andconvert(datetime,Convert(varchar,WeekDate,101))=@WeekDate

Select@DateCount=@DateCount+1

End

End

End

FETCHNEXTFROMAddDateListINTO@Partnum,@Revision

END

END

CLOSEAddDateList

DEALLOCATEAddDateList



/*天生交织数据集*/

InsertInto#Stknwip(Partnum,Revision,Qnty,ProcCode,Id)

SelectPartNum,Revision,Qnty,Com,1A

From#WeekConfig

WhereWeekDate>=@BDateTimeandWeekDate<=@EDateTime



Declare@StkPartnumvarchar(20),@StkRevisionvarchar(5),@StkDelDatedatetime,@StkUPPvarchar(10)

DECLARECU_StknWipINSENSITIVECURSOR

FOR

SelectPartnum,RevisionFrom#Stknwip

GroupByPartnum,Revision

OPENCU_StknWip

BEGIN

FETCHNEXTFROMCU_StknWipINTO@StkPartnum,@StkRevision

WHILE@@FETCH_STATUS=0

BEGIN

select@StkDelDate=min(deldate),@StkUPP=min(upp)

from#Stknwip

WherePartnum=@StkPartnumandRevision=@StkRevision



Updatet1Sett1.DelDate=@StkDelDate,t1.UPP=@StkUPPFrom#Stknwipt1

Wheret1.Partnum=@StkPartnumandt1.Revision=@StkRevision

and(isnull(t1.DelDate,01/01/1900)=01/01/1900orisnull(t1.UPP,-)=-)



FETCHNEXTFROMCU_StknWipINTO@StkPartnum,@StkRevision

END

END

CLOSECU_StknWip

DEALLOCATECU_StknWip



Declare@svarchar(8000)

Set@s=

Select@s=@s+,[+rtrim(proccode)+]=max(caseproccodewhen+rtrim(proccode)+thenQntyelseend)

From#StknwipGroupByproccode,IdOrderById



ifexists(select*fromdbo.sysobjectswhereid=object_id(N[dbo].[X_tmpWeekProc])andOBJECTPROPERTY(id,NIsUserTable)=1)

droptable[dbo].[X_tmpWeekProc]



Exec(selectPartnum,Revision,DelDate=min(DelDate),UPP+@s+IntoX_tmpWeekProcfrom#StknwipgroupbyPartnum,Revision,UPP)



Select*Into#StknWipTotalFromX_tmpWeekProc

DropTableX_tmpWeekProc



Selectt2.PartNum,t2.Revision,PQnty=Sum(t2.PQnty)

Into#InWarehouse

FromPDL_PassBast1(nolock),

PDL_PassDtlt2(nolock),

LotInfot3(nolock)

Wheret1.PaperNo=t2.PaperNo

and(t1.PaperDate>=@BDateTime)

and(t1.PaperDate<=@EDateTime)

andt2.AftProc>=800

andt2.LotNum=t3.LotNum

andt1.finished=1

andt1.IsCancel=0

GroupByt2.PartNum,t2.Revision



------------------------------------增添仓位字段------------------------------

Exec(ALTERTABLE#StknWipTotaladd[InWarehouse]intnotNULLDEFAULT(0))



---------修正进库数----------

Updatet1Sett1.InWarehouse=isnull(t2.PQnty,0)From#StknWipTotalt1,#InWarehouset2

Wheret1.Partnum=t2.Partnum



------------修正保护数据,便利工夫限制局限-------------------------

Selecttop0Dateid,PartNum,Revision,TodayTime=DateId,Qnty=0,DateStr=Qnty10,Todate

Into#ToDate

FromPdl_WeekProcedure



Declare@DayNumint,@CheckDatedatetime,@StrIntvarchar(8000)

Declare@PartnumDatevarchar(20),@RevisionDatevarchar(5)



Set@CheckDate=Convert(datetime,Convert(varchar,getdate(),101))

Set@StrInt=



DECLARECu_PartDateINSENSITIVECURSOR

FOR

SelectDistinctPartNum,RevisionFromPdl_WeekProcedure

OPENCu_PartDate

BEGIN

FETCHNEXTFROMCu_PartDateINTO@PartnumDate,@RevisionDate

WHILE@@FETCH_STATUS=0

BEGIN

Set@DayNum=1

While@DayNum<=7

Begin

Set@StrInt=Convert(varchar,@Daynum)

Exec(InsertInto#ToDate(Dateid,PartNum,Revision,TodayTime,Qnty,DateStr,Todate)

Selectt2.Dateid,Partnum=+@PartnumDate+,Revision=+@RevisionDate+,

t1.[Date+@StrInt+],Qnty=sum(Isnull(t2.[Qnty+@StrInt+],0)),

DateStr=+Qnty+@StrInt+,t2.Todate

FromPdl_WeekProcedureDatet1(nolock),

Pdl_WeekProceduret2(nolock)

Wheret1.DateId=t2.DateIdandt1.DateId=+@CheckDate+

andt2.Partnum=+@PartnumDate+andt2.Revision=+@RevisionDate+

GroupByt1.[Date+@StrInt+],t2.Dateid,t2.Todate)

Set@DayNum=@DayNum+1

End

FETCHNEXTFROMCu_PartDateINTO@PartnumDate,@RevisionDate

END

END

CLOSECu_PartDate

DEALLOCATECu_PartDate



---------------查询用户设置的工夫局限------------------------------

selectt1.*into#tmpTodateFrom#ToDatet1,

(SelectdistinctPartnumRevision=Partnum+Revision

From#ToDate

WhereQnty>0and((TodayTime>=@BDateor@BDate=)

and(TodayTime<=@EDateor@EDate=)))t2

Where(t1.Partnum+t1.Revision)=t2.PartnumRevision



------------------------------转换数据显现格局-----------------------------

Selecttop0Dateid,Partnum,Revision,Qnty1,Qnty2,Qnty3,Qnty4,Qnty5,Qnty6,Qnty7,Todate

Into#WeekProcedure

FromPdl_WeekProcedure



declare@strvarchar(8000)

set@str=

select@str=@str+,[+rtrim(DateStr)+]=max(caseTodayTimewhen+rtrim(TodayTime)+thenQntyelseend)

from#tmpTodategroupbyTodayTime,DateStrOrderbyDateStr



Select@str=Isnull(@str,0,0,0,0,0,0,0)





If(Exists(select*from#tmpTodate))and(@Flage=0)

Begin

InsertInto#WeekProcedure(Dateid,Partnum,Revision,Qnty1,Qnty2,Qnty3,Qnty4,Qnty5,Qnty6,Qnty7,Todate)

exec(SelectDateid,Partnum,Revision+@str+,TodateFrom#tmpTodateGroupByDateid,Partnum,Revision,Todate)



Updatet1Sett1.[A0]=t2.Qnty1,t1.[A1]=t2.Qnty2,t1.[A2]=t2.Qnty3,t1.DelDate=t2.Todate,

t1.[A3]=t2.Qnty4,t1.[A4]=t2.Qnty5,t1.[A5]=t2.Qnty6,t1.[A6]=t2.Qnty7

From#StknWipTotalt1,#WeekProceduret2(nolock)

WhereLtrim(Rtrim(t1.PartNum))=Ltrim(Rtrim(t2.PartNum))

andLtrim(Rtrim(t1.Revision))=Ltrim(Rtrim(t2.Revision))

andconvert(datetime,@BDateTime)=t2.DateId



Selectt1.*,BDateTime=convert(datetime,@BDateTime),EDateTime=convert(datetime,@EDateTime)

From#StknWipTotalt1,#WeekProceduret2

Wheret1.Partnum=t2.Partnumandt1.Revision=t2.Revision

End

Else

If@Flage=1

Begin

Updatet1SetA0=Qnty1,A1=Qnty2,A2=Qnty3,A3=Qnty4,A4=Qnty5,A5=Qnty6,A6=Qnty7,t1.DelDate=t2.Todate

From#StknWipTotalt1,Pdl_WeekProceduret2

WhereLtrim(Rtrim(t1.PartNum))=Ltrim(Rtrim(t2.PartNum))

andLtrim(Rtrim(t1.Revision))=Ltrim(Rtrim(t2.Revision))

andconvert(datetime,@BDateTime)=t2.DateId



Selectt1.*,BDateTime=convert(datetime,@BDateTime),EDateTime=convert(datetime,@EDateTime)

From#StknWipTotalt1

End






提供TCP/IP、ODBC和JDBC等多种数据库连接路径。

从项目平台的选择上讲，我们关心的，应该是一款产品能不能满足任务需求，而不是网上怎么说。

代替了原来VB式的错误判断。比Oracle高级不少。

分区表是个亮点！从分区表也能看出微软要做大作强SQLServer的信心。资料很多，这里不详细说。但是重点了解的是：现在的SQLServer2005的表，都是默认为分区表的。因为它要支持滑动窗口的这个特性。这种特性对历史数据和实时数据的处理是很有帮助的。

SQLServer的异构移植功能个人感觉最好了。（如果对比过SQLServer的链接服务器和Oracle的透明网关的朋友会发现SQLServer的sp_addlinkedserver(openquery)异构数据库系列比Oracle真是强太多了。）

无法深入到数据库系统层面去了解和探究

一直以来个人感觉SQLServer的优化器要比Oracle的聪明。SQL2005的更是比2k聪明了不少。（有次作试验发现有的语句在200万级时还比50万级的相同语句要快show_text的一些提示没有找到解释。一直在奇怪。）

所以你总能得到相应的升级版本，来满足你的需求。

是要和操作系统进行Socket通讯的场景。否则建议慎重！