我得到一个 ERROR subquery must return only one column。我该如何解决这个问题？

Question

select count(distinct "WorkContracts"."SubstituteTeacher") as "Count", "Accounts"."FirstName", "Accounts"."LastName", "Candidates"."Confirmed", "Candidates"."FinalStatus"
from dbo."Accounts"
left join dbo."WorkContracts" on "Accounts"."ID" = "WorkContracts"."SubstituteTeacher" 
full outer join dbo."SubstituteTeacherPeriods" on "Accounts"."ID" = "SubstituteTeacherPeriods"."CreatedBy" 
full outer join dbo."Candidates" ON "Accounts"."ID" = "Candidates"."ModifiedBy" 
where "WorkContracts"."WorkContractType" != 3 
and "SubstituteTeacherPeriods"."StartTime" > TO_TIMESTAMP('2018-09-18 00:00:00', 'yyyy-MM-dd HH24:mi:ss') 
and "SubstituteTeacherPeriods"."EndTime" < TO_TIMESTAMP('2018-09-19 00:00:00', 'yyyy-MM-dd HH24:mi:ss') 
and "SubstituteTeacherPeriods"."Discriminator" = 'AvailabilityPeriod'
and "Candidates"."FinalStatus" = (select max("FinalStatus") from dbo."Candidates")
and "Candidates"."Confirmed"= (select max("Confirmed")"Confirmed", "SubstituteTeacher" from dbo."Candidates" group by "SubstituteTeacher")
group by "Accounts"."FirstName", "Accounts"."LastName", "Candidates"."FinalStatus", "Candidates"."Confirmed"

Answer 1

这似乎是导致错误的行：

and "Candidates"."Confirmed" = (select max("Confirmed")"Confirmed", "SubstituteTeacher"
    from dbo."Candidates" group by "SubstituteTeacher")

首先，将单个标量值与具有两个值的记录进行比较没有意义。但是，比这更糟糕的是，RHS 上的子查询可能返回不止一条记录。尝试更改为：

AND "Candidates"."Confirmed" IN (SELECT MAX("Confirmed") FROM dbo."Candidates"
                                 GROUP BY "SubstituteTeacher")

这是您在修复此问题后重构的整个查询：

SELECT
    COUNT(DISTINCT wc.SubstituteTeacher) AS Count,
    a.FirstName,
    a.LastName,
    c.Confirmed,
    c.FinalStatus
FROM dbo.Accounts a
LEFT JOIN dbo.WorkContracts wc
    ON a.ID = wc.SubstituteTeacher
FULL OUTER JOIN dbo.SubstituteTeacherPeriods s
    ON a.ID = s.CreatedBy
FULL OUTER JOIN dbo.Candidates c
    ON a.ID = c.ModifiedBy
WHERE
    wc.WorkContractType != 3 AND
    s.StartTime >= '2018-09-18 00:00:00' AND s.EndTime < '2018-09-19 00:00:00' AND
    s.Discriminator = 'AvailabilityPeriod' AND
    c.FinalStatus = (SELECT MAX(FinalStatus) FROM dbo.Candidates) AND
    c.Confirmed IN (SELECT MAX(Confirmed) FROM dbo.Candidates
                             GROUP BY SubstituteTeacher)
GROUP BY
    a.FirstName,
    a.LastName,
    c.FinalStatus,
    c.Confirmed;

请注意，通过引入 table 别名并在各处删除不必要的双引号，查询变得更具可读性。

如果您想要使用 Confirmed 列的每位教师的最新记录，那么我们可以使用 ROW_NUMBER:

WITH cte AS (
    SELECT
        COUNT(DISTINCT wc.SubstituteTeacher) AS Count,
        a.FirstName,
        a.LastName,
        c.Confirmed,
        c.FinalStatus,
        ROW_NUMBER() OVER (PARTITION BY a.FirstName, a.LastName ORDER BY Confirmed DESC) rn
    FROM dbo.Accounts a
    ...         -- the rest of the above query
)

SELECT
    Count,
    FirstName,
    LastName,
    Confirmed,
    FinalStatus
FROM cte
WHERE rn = 1;

我得到一个 ERROR subquery must return only one column。我该如何解决这个问题？

I get an ERROR subquery must return only one column. How do I solve this?

sql

subquery

nested-function