SQL 计算出现在不同列中的不同值

Question

我想计算不同的值并报告它们在 SQL 中的出现。问题是我需要计算不同列上的出现次数，然后为每一列报告每个值的出现次数。所以，假设我从一个名为 Result :

的 table 中获得了这样的数据

Id	OwnerId	PilotId	CoachId
A	1	2	3
B	2	2	3
C	3	1	3

那么我的结果应该是：

Id	Owner	Pilot	Coach
1	1	1	0
2	1	2	0
3	1	0	3

我尝试将 COUNT 函数与联合一起使用以单独获取每个结果并且它有效。但我想在同一行 per Id 上得到结果。谁能帮我解决这个问题？

Answer 1

我认为您想将列逆透视为行，然后计算每个值的出现次数。你可以这样做：

select v.id,
    sum(case when v.col = 'owner' then 1 else 0 end) owner,
    sum(case when v.col = 'pilot' then 1 else 0 end) pilot,
    sum(case when v.col = 'coach' then 1 else 0 end) coach
from mytable t
cross apply (values (t.ownerid, 'owner'), (t.pilotid, 'pilot'), (t.coachid, 'coach')) as v(id, col)
group by v.id

Demo on DB Fiddle:

id | owner | pilot | coach
-: | ----: | ----: | ----:
 1 |     1 |     1 |     0
 2 |     1 |     2 |     0
 3 |     1 |     0 |     3

Answer 2

同样使用基本的 SQL 命令。确保在每个 DBMS 上工作：

SELECT 
  ids.id, 
  SUM(CASE WHEN ro.ownerid = ids.id THEN 1 ELSE 0 END) AS Owner,
  SUM(CASE WHEN po.pilotid = ids.id THEN 1 ELSE 0 END) AS Pilot,
  SUM(CASE WHEN co.coachid = ids.id THEN 1 ELSE 0 END) AS Coach
FROM (
  SELECT ownerID AS id FROM result UNION
  SELECT pilotID FROM result UNION
  SELECT coachID FROM result  
) ids
  LEFT JOIN result ro ON ids.id = ro.ownerid
  LEFT JOIN result po ON ids.id = po.pilotid
  LEFT JOIN result co ON ids.id = co.coachid
GROUP BY ids.id

这里SQLFiddler

SQL 计算出现在不同列中的不同值

SQL count distinct values appearing in different columns

sql

sql-server

count

pivot

unpivot