Question

我有这个名为history的表（日期在DD-MM-YYYY中）：

====================
| Buy        | Qty |
====================
| 01-01-2012 |  1  |
| 01-01-2012 |  1  |
| 01-02-2012 |  1  |
| 01-03-2012 |  1  |
| 01-05-2012 |  1  |
| 01-07-2012 |  1  |
| 01-12-2012 |  1  |
====================

注意：在第4,6,8,9,10,11个月没有购买。

如果我跑：

SELECT MONTH(buy) AS day, YEAR(buy) as year, SUM(qty)
FROM history
GROUP BY MONTH(buy),YEAR(buy)

我得到了这个结果：

======================
| Month | Year | Qty |
======================
| 01    | 2012 |  2  |
| 02    | 2012 |  1  |
| 03    | 2012 |  1  |
| 05    | 2012 |  1  |
| 07    | 2012 |  1  |
| 12    | 2012 |  1  |
======================

我想要显示4,6,8,9,10,18个月，但Qty为0（零），如下所示：

======================
| Month | Year | Qty |
======================
| 01    | 2012 |  2  |
| 02    | 2012 |  1  |
| 03    | 2012 |  1  |
| 04    | 2012 |  0  |
| 05    | 2012 |  1  |
| 06    | 2012 |  0  |
| 07    | 2012 |  1  |
| 08    | 2012 |  0  |
| 09    | 2012 |  0  |
| 10    | 2012 |  0  |
| 11    | 2012 |  0  |
| 12    | 2012 |  1  |
======================

我该怎么做？

Answer 1

您可以使用整数查找表。

我在这种情况下使用的是通过CTE生成号码的功能/程序（以下示例）

其余的是使用结果数字作为月份创建日期 - 然后加入上面的结果。

declare @min int 
declare @max int

set @min = 1
set @max = 12
;
with numbers(n) as
(
    select @min as n 
    union all 
    select n + 1
    from numbers
    where n + 1 <= @max 
)
select row_number() over (order by n) [row], n
from numbers option(maxrecursion 0);

Answer 2

试试这个：

Declare @Sample table 
(Buy datetime ,Qty int)

Insert into @Sample values
( '01-01-2012' ,1),
('01-01-2012',1 ),
('01-02-2012',1 ),
('01-03-2012',1 ),
('01-05-2012',1 ),
('01-07-2012',1 ),
('01-12-2012',1 )

;with cte as 
(
  select top 12 row_number() over(order by t1.number) as N
  from   master..spt_values t1 
   cross join master..spt_values t2
 )
select t.N as month,
isnull(datepart(year,y.buy),'2012') as Year,
sum(isnull(y.qty,0)) as Quantity
from cte t
left join @Sample y
on month(convert(varchar(20),buy,103)) = t.N
group by y.buy,t.N

创建一个Month表来存储从1到12的值。而不是master..spt_values，你也可以使用sys.all_objects

  select row_number() over (order by object_id) as months
  from sys.all_objects

或使用递归cte生成月份表

;with cte(N) as 
(
Select 1 
union all
Select 1+N from cte where N<12
)
Select * from cte

然后使用Left join将month表中的值与表进行比较，并使用isnull函数来处理空值。

Answer 3

下面的代码可以帮助您，您也可以按年过滤数据

SELECT MonthYearTable.Month , MonthYearTable.Year, ISNULL(SUM(qty),0)
FROM history
Right JOIN (SELECT Month, Year FROM (SELECT 1 AS Month 
                            Union SELECT 2 
                            Union SELECT 3 
                            Union SELECT 4 
                            Union SELECT 5 
                            Union SELECT 6 
                            Union SELECT 7 
                            Union SELECT 8 
                            Union SELECT 9 
                            Union SELECT 10 
                            Union SELECT 11 
                            Union SELECT 12) AS MonthTable 
            INNER JOIN (SELECT DISTINCT YEAR(buy) AS year FROM history) YearTable ON 1=1
            ) AS MonthYearTable ON Month(history.buy) = MonthYearTable.Month AND Year(history.buy) = MonthYearTable.Year
GROUP BY MonthYearTable.Month, MonthYearTable.Year
ORDER BY MonthYearTable.Year, MonthYearTable.Month

Answer 4

试试这个：

CREATE TABLE history(Buy datetime,Qty int) 

INSERT INTO history
VALUES('01-01-2012',1),
 ('01-01-2012',1),  
 ('01-02-2012',1),  
 ('01-03-2012',1),  
 ('01-05-2012',1),  
 ('01-07-2012',1),  
 ('01-12-2012',1)  


Declare @mindate datetime
Declare @maxdate datetime
select @mindate=MIN(convert(datetime,buy,103)),@maxdate=MAX(convert(datetime,buy,103)) from history 


select datepart(month,a.dt) as month,datepart(year,a.dt) as year,isnull(SUM(qty),0) as Quantity  
from 
(select DATEADD(mm,number,convert(varchar(10),@mindate,103)) as dt from master..spt_values 
where type='p' and DATEADD(mm,number,convert(varchar(10),@mindate,103)) <= convert(varchar(10),@maxdate,103)
) a left join history h
on DATEPART(month,a.dt) = DATEPART(month,convert(varchar(10),h.Buy ,103))
group by a.dt 
order by a.dt

SQL查询组按月和年份

4 个答案: