我正在尝试透视sql表中的数据,以便每个记录ID仅获得一行。但是,我正在运行的代码不断返回空值作为最终结果。
我的桌子如下:
"Sequence" "University Issuing Body"**
"1" "College1"
"1" "College2"
"2" "College1"
"2" "College1"
"2" "College2"
我需要表格如下所示:
"Sequence" "University Issuing Body1" "University Issuing Body2" "University Issuing Body3"
"1" "College1" "College2" "NULL"
"2" "College1" "College1" "College2"
我正在运行的代码如下:
select
Sequence]
,[University Issuing Body_1]
,[University Issuing Body_2]
,[University Issuing Body_3]
from (
select
[Sequence]
,[University Issuing Body]
,'University Issuing Body' + cast(row_number() over(partition by [Sequence] order by [University Issuing Body]) as varchar(12)) as hdg
from [AB_DCU_IP_2018].[dbo].[PR_Q_Joined]
) d
pivot(
max([University Issuing Body])
for hdg in ([University Issuing Body_1], [University Issuing Body_2], [University Issuing Body_3])
)pvt
但是,当我运行上面的代码时,我所有的值都显示为空,并且得到以下结果:
"Sequence" "University Issuing Body_1" "University Issuing Body_2" "University Issuing Body_3"
"1" "NULL" "NULL" "NULL"
"2" "NULL" "NULL" "NULL"
任何帮助将不胜感激,因为我一直在在线寻找解决方案,而且似乎找不到我要去哪里。
谢谢。
答案 0 :(得分:1)
只需使用条件聚合:
select Sequence,
max(case when seqnum = 1 then [University Issuing Body] end) as University_Issuing_Body_1,
max(case when seqnum = 2 then [University Issuing Body] end) as University_Issuing_Body_2,
max(case when seqnum = 3 then [University Issuing Body] end) as University_Issuing_Body_3
from (select pqj.*
row_number() over (partition by [Sequence] order by [University Issuing Body]) as seqnum
from [AB_DCU_IP_2018].[dbo].[PR_Q_Joined] pqj
) d
group by [University Issuing Body];
如果您确实要使用pivot
,则无需定义hdg
:
select [Sequence],
[1] as [University Issuing Body_1],
[2] as [University Issuing Body_2],
[3] as [University Issuing Body_3]
from (select [Sequence], [University Issuing ,Body]
row_number() over (partition by [Sequence] order by [University Issuing Body]) as seqnum
from [AB_DCU_IP_2018].[dbo].[PR_Q_Joined]
) d
pivot (max([University Issuing Body])
for seqnum in (1, 2, 3)
)pvt
答案 1 :(得分:1)
您正在使用hdg构造中间表,格式为University Issuing Body<n>
。但是,随后您将提取hdg格式为University Issuing Body_<n>
的值。毫不奇怪,这与任何内容都不匹配-因此您会得到NULL。
只需在您的hdg构造中加一个下划线:
select
[Sequence]
,[University Issuing Body_1]
,[University Issuing Body_2]
,[University Issuing Body_3]
from (
select
[Sequence]
,[University Issuing Body]
,'University Issuing Body_' + cast(row_number() over(partition by [Sequence] order by [University Issuing Body]) as varchar(12)) as hdg
from [AB_DCU_IP_2018].[dbo].[PR_Q_Joined]
) d
pivot(
max([University Issuing Body])
for hdg in ([University Issuing Body_1], [University Issuing Body_2], [University Issuing Body_3])
) pvt