提取每月时段数据

Question

I'm dealing with sales data that looks like this

Date	Time	Net Amount	Store No	Item Category Code
2021-03-21 00:00:00.000	1754-01-01 08:01:14.627	100	001	FOOD
2021-01-31 00:00:00.000	1754-01-01 15:42:21.670	120	002	FOOD
2021-04-24 00:00:00.000	1754-01-01 17:35:30.070	160	002	FOOD
2021-03-14 00:00:00.000	1754-01-01 13:08:02.073	150	003	FOOD
2021-02-27 00:00:00.000	1754-01-01 18:33:53.847	95	001	FOOD
2021-02-20 00:00:00.000	1754-01-01 21:30:24.007	180	002	DRINK
2021-04-09 00:00:00.000	1754-01-01 07:36:33.930	110	003	FOOD
2021-01-29 00:00:00.000	1754-01-01 08:02:49.703	180	002	FOOD
2021-01-24 00:00:00.000	1754-01-01 11:01:00.953	110	003	FOOD

What I need is to pull out monthly data by daypart that would result to this

	Jan 2021	Feb 2021	Mar 2021	Apr 2021
4am to 10:59am	180	0	100	110
11am to 1:59pm	110	0	150	0
2pm to 3:59am	120	95	0	160

Here's the query I'm currently using:

with FoodSales as
(
    select distinct
           concat(year(s.[Date]),'',datename(month,s.[Date])) as TransMonth
           ,sum( s.[Net Amount]*-1) as Sales

    from [dbo].[vw_Sales Entry] as s

    left join [dbo].[vw_Discount Entry] as d
          on s.[Receipt No_]=d.[Receipt No_] 


    where s.[Date] >= '2021-01-01 04:00:00' and s.[Date] <= '2021-04-30 10:59:59'
          and s.[Item Category Code]='FOOD'
          and (d.[Discount Name] is null or d.[Discount Name]='')
         

    group by 
           concat(year(s.[Date]),'',datename(month,s.[Date]))

)

    select * 
    from FoodSales
    pivot( sum(Sales)
    for TransMonth in ([2021January]
                       ,[2021February]
                       ,[2021March]
                       ,[2021April]

      )
      ) as p


;

I've also tried using

[Date] between '2021-01-01 04:00:00' and '2021-04-30 10:59:59'

but when both are tried with another interval, the sales amount just comes out the same.

Answer 1

I would suggest to you to store time as part of datetime column itself. If you want to keep time separately, store them in the time datatype. I have put sample below considering both date and time being stored in the date column.

You can use DATEPART to demarcate the time ranges and then use pivot to generate result set as given below:

declare @TABLE TABLE(dateval datetime, Netamount int, storeno char(3), itemcategorycode char(4))

insert into @table
values
('2021-03-21 04:00:00.000', 100,'001','FOOD'), 
('2021-04-24 10:00:00.000', 160,'002','FOOD')   


select * from
(
SELECT case when datepart(hh,dateval) >= 4 and datepart(hh,dateval) < 11 then '4am to 10:59am' 
            when datepart(hh,dateval) >= 11 and datepart(hh,dateval) < 14 then '11am to 1:59pm' 
            when datepart(hh,dateval) >= 14 and datepart(hh,dateval) < 16 then '2pm to 3:59am' end as timerange,
            
            format(dateval,'MMM yy') AS Month_Year,netamount FROM @table 
            ) as t
            pivot (sum(NetAmount) for month_year in ([Mar 21],[Apr 21])) as pvt

timerange	Mar 21	Apr 21
4am to 10:59am	100	160