Question

First of all, my table structure is something similar like this:

CREATE TABLE Testing(
       [ID]        NVARCHAR(50),
       [DATE]      DATETIME,
       [TOTAL]     INT,
       [ITEM]      NVARCHAR(50),
       [Warehouse] NVARCHAR(50)
)ON[PRIMARY]

I put some sample here:

[ID] [Date]      [Total] [Item] [Warehouse]
1    2011-04-04  400     A0001   B12
2    2011-05-04  500     A0001   B13
3    2011-04-30  400     A0001   B12
4    2011-04-25  400     A0001   B13
5    2011-06-05  600     A0001   B12
6    2011-03-02  300     A0001   B11
7    2011-05-28  500     A0001   B13   

I am trying to group by [Item] and [Warehouse] and [Date] by month as well
For example output:

The result should be like this

[Date]        [Total]   [Item]  [Warehouse]
March 2011     300       A0001   B11
April 2011     800       A0001   B12
June 2011      500       A0001   B12
April 2011     400       A0001   B13
May 2011       1000      A0001   B13

I tried the sql something like, that i parse in month and year part to made the selection

SELECT [Item],[Warehouse],SUM(Total) AS Total
FROM [Testing]
WHERE Datepart(month,[Date]) = 4 AND DATEPART(year,[Date]) = 2011
GROUP BY [Item],[Warehouse]

I get the expected result?Is there any way to do? i actually trying to came out a close balance for each month and year by distinct of warehouse and item?

Sound to me its need to loop through a prefix table..
Its that anyway to do so?

Thanks

Regards
Liangck

Answer 1

I think that formatting date on SQL Server side is (usually) a bed idea, but anyway you can try:

select datename(mm, dateadd(m, datediff(m, 0, [date]), 0)) + ' ' +
      select cast(datepart(yy, dateadd(m, datediff(m, 0, [date]), 0)) as varchar)
      [date] sum(total) as total, item,warehouse
from testing
group by item, warehouse, dateadd(m, datediff(m, 0, [date]), 0)