Question

我想在不使用临时表的情况下优化代码，这可能吗？我尝试使用JOINS，但是我不知道如何在符合条件的同一个表中使用JOIN：

select 'Quest1' q1, 
        date1 as date_q1,
        date1 as date_q2
into #Temp
from table1
where id in (select min(id) 
                from table1 
                where date1 = '2019-01-01'
                group by sy_id, date1) 
and sy_id is not null;

update #Temp
set date_q2 =  table1.date1
from table1 
where table1.cal_id = 7 
and #Temporal.sy_id = table1.sy_id
select q1, DATEDIFF(d, date_q1, date_q2) as av 
from #Temp
union all
select 'Quest2' q1, DATEDIFF(d, date_ref, date1) as av
from table1 
where id in (select min(id) 
                from table1 
                where date1 = '2019-01-01' 
                group by sy_id, date1)

编辑，解决。

Select q1, avg(Diferencia) as av from (select 'Quest1' q1, datediff(d, date1, (select top 1 d.date1 from table1 d where d.cal_id = 7 and d.sy_id = sy_id)) av from table1 where id in (select min(id) from table1 where date1 >= '2019-01-01' group by sy_id, date1) and sy_id is not null union all select 'Quest2' q1, datediff(d, date_ref, date1) av from table1 where id in (select min(id) from table1 where date1 >= '2019-01-01' group by sy_id) group by q1

Answer 1

您应该检查CTE：https://docs.microsoft.com/en-us/sql/t-sql/queries/with-common-table-expression-transact-sql?view=sql-server-2017

更具体地说： Select first row in each GROUP BY group?

使用此查询按组选择顶部行，然后即可加入。

或者，您可以使用子查询：

SELECT
    date_diff(
        some_date,
#subquery - min date or whatever
        SELECT MIN (...) FROM "x" AS "x2" WHERE "x2"."id"="x1"."id" 
    )
FROM "x" AS "x1"

在没有临时表的情况下优化我的SQL代码

1 个答案: