例如,我有以下数据:
select *
into t1
from (
select 'apple' as company, 'Tom' as people
union select 'apple' as company, 'Jessi' as people
union select 'fb' as company, 'Alex' as people
union select 'fb' as company, 'Nick' as people
) as a
select * from t1
数据如下:
company people
apple Jessi
apple Tom
fb Alex
fb Nick
我想知道每家公司是否都有汤姆 - 在每一行都有。
是否有类似的内容:
select *,
any(people = 'Tom') over (partition by company order by company) as find_tom
from t1
解决方法:
-- step 1: binary indicator has_tom
select *
,case when people = 'Tom' then 1 else 0 end
as has_tom
into t2
from t1
-- use max(has_tom) to see if tom exists for each partition
select *,
case when max(has_tom) over (partition by company order by company) = 1 then 'has tom' else 'no tom' end
as find_tom
from t2
这会产生我想要的结果:
company people has_tom find_tom
apple Jessi 0 has tom
apple Tom 1 has tom
fb Alex 0 no tom
fb Nick 0 no tom
答案 0 :(得分:2)
这有用吗?
Select t1.*,
CASE WHEN
MAX(CASE WHEN people='Tom' THEN 1 ELSE 0 END)
OVER(Partition by company Order by (select null)) =1 THEN 'has Tom'
ELSE 'no Tom' END
FROM t1
答案 1 :(得分:1)
我会像这样使用case
:
select t1.*,
(case when sum(case when people = 'tom' then 1 else 0 end) over (partition by company) > 0
then 'has tom' else 'no tom'
end) as find_tom
from t1;