我的数据库看起来像这样
DailyData
rID int
Stock varchar
rDate date
Shares int
price float
我要做的是获取两个日期的数据。
示例数据
rID stock rDate Shares price
11 Stock1 21/03/2016 15 1.22
12 Stock2 21/03/2016 22 2.23
13 Stock3 21/03/2016 17 3.32
14 Stock4 21/03/2016 10 4.24
15 Stock1 22/03/2016 15 1.25
16 Stock2 22/03/2016 20 2.27
17 Stock3 22/03/2016 17 3.32
18 Stock1 23/03/2016 15 1.28
19 Stock2 23/03/2016 20 2.20
20 Stock3 23/03/2016 17 3.32
21 Stock4 23/03/2016 10 4.24
预期输出
Stock Shares-21 Shares-20
Stock1 15 15
Stock2 22 20
Stock3 17 17
Stock4 10 0
我对SQL Server CE数据库的查询:
Select
DD1.Stock, sum(DD1.Shares) as Shares-21, sum(DD2.shares) as Shares-20
from
DailyData DD1, DailyData DD2
where
DD1.rDate = '21/03/2016' and DD2.rDate = '20/03/2016'
and DD1.Stock = DD2.Stock
group by
DD1.Stock
我得到7行数据而不是4行。
请帮助查询。
*******************************新修改************** ******* 我按照建议遵循,但似乎没有奏效。这是一个真正的sql脚本。
Select P.pName,DD.Stock,
sum(case DD.rDate when '03/21/2016' then DD.Shares else 0 end) as Shares21,
sum(case DD.rDate when '03/20/2016' then DD.Shares else 0 end) as Shares20
from dailyData DD, Portfolios P
where DD.rDate = '03/21/2016' or DD.rDate = '03/20/2016'
and DD.pID = P.pID
and DD.pID=1
group by P.pName,DD.stock
order by P.pName,DD.Stock
现在对于pID = 1,20-Mar和21-Mar有23条记录
运行此查询后,它返回的方式超过23.我只期待23条记录。
答案 0 :(得分:3)
实际上它不需要使用JOIN。您可以像这样使用CASE WHEN语句
SELECT stock,
Sum(CASE rdate
WHEN '21/03/2016' THEN shares
ELSE 0
END) AS Shares21,
Sum(CASE rdate
WHEN '20/03/2016' THEN shares
ELSE 0
END) AS Shares20
FROM dailydata
WHERE rdate = '21/03/2016'
OR rdate = '20/03/2016'
GROUP BY stock
答案 1 :(得分:0)
您可以将PIVOT与DATEPART一起使用
SELECT Stock, [21] AS [Shares-21], [20] AS [Shares-20] FROM
(
SELECT s.Stock, Datepart(d,s.rDate) rDate , s.Shares FROM DailyData s
) src
PIVOT
(
MAX(Shares) for rDate in ([21], [20])
) pvt