SQL Query Joins (Create view of combined sum from different table) - mysql

Question

Hi I need a help regarding this problem. I want to create a view where it displays the summary of purchased and sold items.
Theses are my tables:
**tblfruit**
ID Name
1 Apple
2 Orange
3 mango
**tblpurchaseditems**
ID fruit_id qty amount
1 1 3 75
2 1 2 50
3 2 1 10
4 3 3 30
**tblsolditems**
ID fruit_id qty amount
1 1 2 150
2 1 2 350
3 2 1 50
4 3 2 230
5 3 1 120
I want the output :
**tblsummary**
ID fruit_id totalqtypurchased totalamountpurchased totalqtysold totalamountsold
1 1 5 125 4 500
2 2 1 10 1 50
3 3 3 30 3 350

Answer 1

So there are a couple of ways of doing this, either online or using a join.
Inline:
Select name, (select sum(qty) from tableb) as totalqty
From tablea
Join:
Select a.name, sum(b.qty) as totalqty
From tablea a
Join tableb b
On b.itemid = a.itemid
Group by a.name
For multiple tables things can get trickier using joins due to duplication so inline queries can work better here, for smaller queries.

Answer 2

select a.fruit_id,sum(b.qty),sum(b.amount),sum(c.qty),sum(c.amount)
from tblfruit a,tblpurchaseditems b,tblsolditems c
where a.fruit_id=b.fruit_id and a.fruit_id=c.fruit_id
group by a.fruit_id

Answer 3

From your sample data and expected result
you can try to UNION ALL to combine tblpurchaseditems and tblpurchaseditems table and make grp to spite two result set. then use condition aggregate function to get your expect result.
CREATE TABLE tblpurchaseditems(
ID INT,
fruit_id INT,
qty INT,
amount INT
);
INSERT INTO tblpurchaseditems VALUES (1,1,3,75);
INSERT INTO tblpurchaseditems VALUES (2,1,2,50);
INSERT INTO tblpurchaseditems VALUES (3,2,1,10);
INSERT INTO tblpurchaseditems VALUES (4,3,3,30);
CREATE TABLE tblsolditems(
ID INT,
fruit_id INT,
qty INT,
amount INT
);
INSERT INTO tblsolditems VALUES (1,1,2,150);
INSERT INTO tblsolditems VALUES (2,1,2,350);
INSERT INTO tblsolditems VALUES (3,2,1,50);
INSERT INTO tblsolditems VALUES (4,3,2,230);
INSERT INTO tblsolditems VALUES (5,3,1,120);
Query 1:
SELECT (#RN:=#RN+1) ID,
fruit_id,
SUM(CASE WHEN grp = 1 THEN qty END) totalqtypurchased ,
SUM(CASE WHEN grp = 1 THEN amount END) totalamountpurchased ,
SUM(CASE WHEN grp = 2 THEN qty END) totalqtysold,
SUM(CASE WHEN grp = 2 THEN amount END) totalamountsold
FROM (
select 1 grp,fruit_id,qty,amount from tblpurchaseditems
UNION ALL
SELECT 2,fruit_id,qty,amount FROM tblpurchaseditems
)t1 CROSS JOIN (SELECT #RN:=0) v
GROUP BY fruit_id
Results:
| ID | fruit_id | totalqtypurchased | totalamountpurchased | totalqtysold | totalamountsold |
|----|----------|-------------------|----------------------|--------------|-----------------|
| 1 | 1 | 5 | 125 | 5 | 125 |
| 2 | 2 | 1 | 10 | 1 | 10 |
| 3 | 3 | 3 | 30 | 3 | 30 |
NOTE
I would use a column itemGroup in table, which can represent which type.
1 mean purchaseditems
2 mean solditems
That would not need to use UNION ALL to combine two table, more make sense.
So in the table schema will look like.
CREATE TABLE tblitems(
ID INT,
fruit_id INT,
qty INT,
amount INT,
itemGroup INT
);
INSERT INTO tblitems VALUES (1,1,3,75,1);
INSERT INTO tblitems VALUES (2,1,2,50,1);
INSERT INTO tblitems VALUES (3,2,1,10,1);
INSERT INTO tblitems VALUES (4,3,3,30,1);
INSERT INTO tblitems VALUES (1,1,2,150,2);
INSERT INTO tblitems VALUES (2,1,2,350,2);
INSERT INTO tblitems VALUES (3,2,1,50 ,2);
INSERT INTO tblitems VALUES (4,3,2,230,2);
INSERT INTO tblitems VALUES (5,3,1,120,2);
Query 1:
SELECT (#RN:=#RN+1) ID,
fruit_id,
SUM(CASE WHEN itemGroup = 1 THEN qty END) totalqtypurchased ,
SUM(CASE WHEN itemGroup = 1 THEN amount END) totalamountpurchased ,
SUM(CASE WHEN itemGroup = 2 THEN qty END) totalqtysold,
SUM(CASE WHEN itemGroup = 2 THEN amount END) totalamountsold
FROM tblitems t1 CROSS JOIN (SELECT #RN:=0) v
GROUP BY fruit_id
Results:
| ID | fruit_id | totalqtypurchased | totalamountpurchased | totalqtysold | totalamountsold |
|----|----------|-------------------|----------------------|--------------|-----------------|
| 1 | 1 | 5 | 125 | 4 | 500 |
| 2 | 2 | 1 | 10 | 1 | 50 |
| 3 | 3 | 3 | 30 | 3 | 350 |