Question

在SQL Server 2014中，我使用Northwind作为示例数据库，并使用Orders表。

选择第一行：

select top 1 orderID, CustomerID,EmployeeID,OrderDate,RequiredDate 
from orders

输出：

orderID CustomerID  EmployeeID  OrderDate   RequiredDate
10248   VINET   5   1996-07-04 00:00:00.000 1996-08-01 00:00:00.000

选择最后一行：

select top 1 orderID, CustomerID,EmployeeID,OrderDate,RequiredDate 
from orders order by orderID desc

输出：

orderID CustomerID  EmployeeID  OrderDate   RequiredDate
11077   RATTC   1   1998-05-06 00:00:00.000 1998-06-03 00:00:00.000

让我们使用Union ALL：

组合第一行和最后一行

select top 1 orderID, CustomerID,EmployeeID,OrderDate,RequiredDate 
from orders
union all
select top 1 orderID, CustomerID,EmployeeID,OrderDate,RequiredDate 
from orders order by orderID desc

输出：

orderID CustomerID  EmployeeID  OrderDate   RequiredDate
10248   VINET   5   1996-07-04 00:00:00.000 1996-08-01 00:00:00.000
10248   VINET   5   1996-07-04 00:00:00.000 1996-08-01 00:00:00.000

那么为什么最后的结果没有按预期显示数据。

我知道要获得第一行和最后一行，请使用下一行： -

select top 1 orderID, CustomerID,EmployeeID,OrderDate,RequiredDate 
from orders
union all
Select * from
(
select top 1 orderID, CustomerID,EmployeeID,OrderDate,RequiredDate 
from orders order by orderID desc ) a

输出： -

orderID CustomerID  EmployeeID  OrderDate   RequiredDate
10248   VINET   5   1996-07-04 00:00:00.000 1996-08-01 00:00:00.000
11077   RATTC   1   1998-05-06 00:00:00.000 1998-06-03 00:00:00.000

但我仍然感到困惑，为什么我应该第二次选择派生表！

Answer 1

这个

select top 1 orderID, CustomerID,EmployeeID,OrderDate,RequiredDate 
from orders
union all
select top 1 orderID, CustomerID,EmployeeID,OrderDate,RequiredDate 
from orders order by orderID desc

实际上如下：

(select top 1 orderID, CustomerID,EmployeeID,OrderDate,RequiredDate 
from orders
union all
select top 1 orderID, CustomerID,EmployeeID,OrderDate,RequiredDate 
from orders) order by orderID desc

因为除非它包含在子查询中，否则将在union的最后一个时间计算order by。

你已经知道了解决方法，所以我没有提到它。

使用Top with Union all

1 个答案: