sql聚合函数优化

时间:2017-07-21 19:32:37

标签: sql

我正试图找到一种优化查询的方法,我认为这需要花费大量时间。这是我的SQL查询:

select
ISNULL((select sum(s.durationp) from Schedule s where AccountID = @AccountID and ClientID = @clientidvalue and status=2 and (@StaffID = 0 OR (@StaffID <> 0 AND s.StaffID = @StaffID)) and s.Date= @sunday), 0),
ISNULL((select sum(s.durationp) from Schedule s where AccountID = @AccountID and ClientID = @clientidvalue and status=2 and (@StaffID = 0 OR (@StaffID <> 0 AND s.StaffID = @StaffID)) and s.Date= @monday), 0),
ISNULL((select sum(s.durationp) from Schedule s where AccountID = @AccountID and ClientID = @clientidvalue and status=2 and  (@StaffID = 0 OR (@StaffID <> 0 AND s.StaffID = @StaffID)) and s.Date= @tuesday), 0),
ISNULL((select sum(s.durationp) from Schedule s where AccountID = @AccountID and ClientID = @clientidvalue and status=2 and (@StaffID = 0 OR (@StaffID <> 0 AND s.StaffID = @StaffID)) and s.Date= @wednesday), 0),
ISNULL((select sum(s.durationp) from Schedule s where AccountID = @AccountID and ClientID = @clientidvalue and status=2 and (@StaffID = 0 OR (@StaffID <> 0 AND s.StaffID = @StaffID)) and s.Date= @thursday), 0),
ISNULL((select sum(s.durationp) from Schedule s where AccountID = @AccountID and ClientID = @clientidvalue and status=2 and (@StaffID = 0 OR (@StaffID <> 0 AND s.StaffID = @StaffID)) and s.Date= @friday), 0),
ISNULL((select sum(s.durationp) from Schedule s where AccountID = @AccountsqlID and ClientID = @clientidvalue and status=2 and (@StaffID = 0 OR (@StaffID <> 0 AND s.StaffID = @StaffID)) and s.Date= @saturday), 0).

我认为查询的每一列都在执行相同的逻辑,除了将日期更改为最后一个约束。有没有办法可以先计算整个逻辑,比如将结果刷新到临时表中,然后用date作为where子句查询它。

1 个答案:

答案 0 :(得分:0)

使用条件聚合:

select sum(case when s.Date = @sunday then s.durationp else 0 end) as sunday,
       sum(case when s.Date = @monday then s.durationp else 0 end) as monday,
       . . .
from Schedule s 
where AccountID = @AccountID and ClientID = @clientidvalue and
      status = 2 and
      (@StaffID = 0 OR (@StaffID <> 0 AND s.StaffID = @StaffID));

注意:假设至少有一行符合NULL条件,我遗漏了WHERE转换。