从同一个表中按不同条件对多列求和，然后按日期分组

Question

I tried to SUM values of multiple columns (CASH+CARD+CHEQUE+REST) and group them to display totals for each day.

Actual table:

Date	CASH	CARD	CHEQUE	REST	TYPE
2022-06-02	150.00	200.00	0.00	12.00	STORE1
2022-06-02	150.00	240.00	56.00	67.00	STORE2
2022-06-02	45.00	459.00	150.00	0.00	STORE3
2022-06-02	45.00	400.00	150.00	34.00	TRAVEL1
2022-06-03	87.00	59.00	150.00	400.00	STORE1
2022-06-03	45.00	790.00	450.00	104.00	STORE2
2022-06-03	70.00	30.00	0.00	241.00	STORE3
2022-06-03	30.00	120.00	11.00	72.00	TRAVEL1

I want it to make it as new table as:

Date	TOTAL_SALE	STORE_TOTAL	TRAVEL_TOTAL
2022-06-02	2158.00	1529.00	629.00
2022-06-03	2659.00	2426.00	233.00

I tried with UNION but that puts the result one below other and not like this in the same row, I also tried the option below but am getting this error

#1111 invalid use of group function

Code I use that produces an error:

SELECT DATE, (SUM(CASH) + SUM(CARD) + SUM(CHEQUE) + SUM(REST)) AS TOTAL_SALE,

SUM(case when TYPE LIKE 'STORE%' then (SUM(CASH) + SUM(CARD) + SUM(CHEQUE) + SUM(REST))
else 0 end) as STORE_TOTAL,
SUM(case when TYPE LIKE 'TRAVEL%' then (SUM(CASH) + SUM(CARD) + SUM(CHEQUE) + SUM(REST))
else 0 end) as TRAVEL_TOTAL

FROM tbl_Payment where DATE BETWEEN '2022-06-02' AND '2022-06-03'
GROUP BY DATE ASC

Answer 1

First, group by does not take an order. No asc .

The problem is putting sum inside a sum . Whatever is inside sum will be summed per row , there's no need to sum it again.

Similarly, you don't have to sum each column and then add the sums. Add the columns then sum. This is a bit more succinct.

SELECT
  "DATE",
  SUM(CASH + CARD + CHEQUE + REST) AS TOTAL_SALE,
  SUM(
    case when TYPE LIKE 'STORE%' then
      CASH + CARD + CHEQUE + REST
    else
      0
    end
  ) as STORE_TOTAL,
  SUM(
    case when TYPE LIKE 'TRAVEL%' then
      CASH + CARD + CHEQUE + REST
    else
      0
    end
  ) as TRAVEL_TOTAL
FROM tbl_Payment
where "DATE" BETWEEN '2022-06-02' AND '2022-06-03'
GROUP BY "DATE"

Note: date is a SQL keyword. It can be confused with the type date . Avoid using it as a column name. Use the at and on conventions for naming timestamp and date columns. For example, paid_on .