处理来自多行的日期

Posted 2023-03-24

【中文标题】处理来自多行的日期

【英文标题】：Dealing with Dates from multiple rows

【发布时间】：2019-04-09 07:47:28

【问题描述】：

我需要为具有相同 ID 值的每一行生成一个结束日期，作为下一条记录的 date-1，如果该 ID 没有下一行，则结束日期应该为空。

例如，对于 ID 513，第一行是 2008-01-01 和 2010-04-16，第二行是 2010-04-17、2011-04-25。

我不确定如何在没有光标的情况下实现这一目标。

Create Table #Temp
(
    ID int,
    Amount money,
    StartDate datetime
)

insert into #Temp
(
    ID,
    Amount,
    StartDate
)
select 513,240.00,'2008-01-01 00:00:00' union all
select 513,240.00,'2010-04-17 00:00:00' union all
select 513,265.00,'2011-04-26 00:00:00' union all
select 513,275.00,'2012-04-17 00:00:00' union all
select 513,285.00,'2013-04-22 00:00:00' union all
select 513,325.00,'2015-06-15 00:00:00' union all
select 513,335.00,'2017-06-15 00:00:00' union all
select 514,280.00,'2001-01-22 00:00:00' union all
select 514,280.00,'2010-06-09 00:00:00' union all
select 515,240.00,'2019-01-01 00:00:00' union all
select 515,240.00,'2010-04-17 00:00:00' union all
select 515,265.00,'2011-04-26 00:00:00' union all
select 515,275.00,'2012-04-17 00:00:00' union all
select 515,285.00,'2013-04-22 00:00:00' union all
select 515,325.00,'2015-06-15 00:00:00' union all
select 515,335.00,'2017-06-15 00:00:00'

select * from #Temp

drop table #Temp

【参考方案1】：

您可以使用LEAD() 和DATEADD() 来实现您的结果：

SELECT *, DATEADD(DAY, -1, LEAD (StartDate, 1) OVER (PARTITION BY ID ORDER BY StartDate)) AS ENDDATE
FROM #Temp

Demo on db<>fiddle

【参考方案2】：

您可以使用铅，它可以按照您的要求工作，

select *, lead(StartDate-1) over(partition by id order by id,startdate) as EndDate 
from #Temp

【参考方案3】：

对于 SQL Server 2005+：

   WITH CTE AS(
        SELECT ROW_NUMBER() OVER(PARTITION BY ID ORDER BY StartDate) as row_num,
        ID, Amount, StartDate FROM #Temp 
    )
    SELECT
    prev.ID,
    prev.Amount,
    prev.StartDate,
    DATEADD(DAY, -1, ISNULL(prev.StartDate,1)) AS EndDate
    FROM CTE prev
    LEFT JOIN CTE nex ON nex.rownum = CTE.rownum + 1

【参考方案4】：

select t1.* ,  dateadd(day,-1,t2.dt) enddate
from  (select row_number()over(partition by ID order by ID) srno,  ID,  StartDate dt from #Temp) t1 
left join (select row_number()over(partition by ID order by ID) srno,   ID,  StartDate dt from #Temp)t2 on  t1.srno =(t2.srno-1)and  t1.ID = t2.ID