POS Application - Simplify SQL Mutli-Queries(MySQL) - mysql

Question

Hei guy I'm working on a POS app with MySQL. Here is my situation:
Table "purchased_item"
| id | name | check_id | real_price |
| 1 | iPhone5 | 0001 | 399 |
| 2 | iPhone4 | 0001 | 199 |
| 3 | iPhone5s | 0002 | 599 |
| 4 | iPhone5c | 0003 | 399 |
| 5 | iMac 21" | 0003 | 999 |
| 6 | iPod Touch | 0003 | 99 |
| 7 | iPhone5 | 0004 | 399 |
| 8 | iPhone3G | 0004 | 99 |
| 9 | iPhone6 | 0005 | 899 |
| 10 | iPhone3Gs | 0005 | 101 |
And I want to know how many checks's total are larger than or qual(>=) 1000, so what I'm doing now is to do several times of query. In this example, I do 5 times and sum it manually by the host program.
Later the data grow, the queries become slow because there're tons of checks everyday. So I change to record it to another table.
Table "checks"
| id | total | sales |
| 0001 | 598 | A |
| 0002 | 599 | A |
| 0003 | 1497 | B |
| 0004 | 498 | B |
| 0005 | 1000 | A |
But another problem occur in the later time: When I need to adjust the real_price in "purchased_item" table, I also need to maintain the "total" column in "checks" table. It sounds doesn't a big matter but I'd like to find a better way to solve it.
Solved:
SELECT * FROM purchased_item
GROUP BY check_id
HAVING sum(real_price) >= 1000
And the result will be:
| id | name | check_id | real_price |
| 4 | iPhone5c | 0003 | 399 |
| 9 | iPhone6 | 0005 | 899 |
Further question: If I want to count the total price for checks, how can I do it?
I found it:
SELECT check_id,sum(real_price) FROM purchased_item
GROUP BY check_id
HAVING sum(real_price) >= 1000

Answer 1

Try it this way
SELECT i.id, i.name, i.check_id, i.real_price
FROM
(
SELECT MIN(id) id
FROM purchased_item
GROUP BY check_id
HAVING SUM(real_price) >= 1000
) q JOIN purchased_item i
ON q.id = i.id
ORDER BY q.id DESC
Sample output:
| ID | NAME | CHECK_ID | REAL_PRICE |
|----|----------|----------|------------|
| 9 | iPhone6 | 5 | 899 |
| 4 | iPhone5c | 3 | 399 |
...I want to count how many checks's total are over 1000
For that you can just do this
SELECT COUNT(*) total
FROM
(
SELECT check_id
FROM purchased_item
GROUP BY check_id
HAVING SUM(real_price) >= 1000
) q;
Sample output:
| TOTAL |
|-------|
| 2 |
Here is SQLFiddle demo
To update total in checks after adjusting real_price in purchased_item
UPDATE checks c JOIN
(
SELECT check_id, SUM(real_price) total
FROM purchased_item
WHERE check_id IN(5) -- whatever check(s)'s total you want to recalculate
GROUP BY check_id
) p
ON c.id = p.check_id
SET c.total = p.total;
Here is SQLFiddle demo