SQL Query SUM issue - sql

Question

I am writing a SQL query where i want to list down and calculate some data every thing looks to work fine but the only problem I am having is when i SUM values in a given column the final result is a overall total but i want this total only to be for a month of October.
The scenario is as followed i need to calculate a total amount of deposits made by costomers over the month of October with other information that works well its only the sum issue.
SQL:
SELECT customers.id AS 'Customer ID',
customers.firstname AS
'Customer First Name',
customers.lastname AS 'Customer Last Name'
,
users.firstname AS
'Employee First Name',
users.lastname AS 'Employee Last Name'
,
positions.currency AS 'Currency',
Sum(customer_deposits.amount) AS 'Total Deposits',
Sum(positions.status) AS 'Total Bets',
Sum(customer_total_statistics.totalbonusdeposits) AS 'Total Bonuses',
positions.date AS 'Date'
FROM customers
LEFT JOIN customer_deposits
ON customers.id = customer_deposits.customerid
LEFT JOIN users
ON customers.employeeinchargeid = users.id
LEFT JOIN positions
ON customers.id = positions.customerid
LEFT JOIN customer_total_statistics
ON customers.id = customer_total_statistics.customerid
WHERE customer_deposits.status = 'approved'
AND positions.status = 'won'
AND positions.date BETWEEN '2013-10-01' AND '2013-11-01'
AND customers.isdemo = '0'
GROUP BY customers.id
ORDER BY customers.id

Answer 1

Well, you don't filter your customer_deposits at all, so the sum over customer_deposits.amount will of course give you the sum of all the deposits. You'll have to add another where to filter customer_deposits by date as well, not just positions.

Answer 2

It looks like you have multiple 1-n relationships so your joins are creating cartesian products. So if you had a customer with 2 deposits and 2 positions:
Positions customer_deposits
CustomerID | Amount CustomerID | Amount
1 | 10 1 | 30
1 | 20 1 | 40
When you do this join:
SELECT c.ID, cd.Amount AS DepositAmount, p.Amount AS PositionAmount
FROM Customers c
INNER JOIN customer_deposits cd
ON c.ID = cd.CustomerID
INNER JOIN Positions p
ON c.ID = p.CustomerID
You end up with 4 rows with all combinations of deposit amount and position amount:
ID DepositAmount PositionAmount
1 10 30
1 10 40
1 20 30
1 20 40
So if you sum this up you end up with incorrect sums because rows are duplicated. You need to move your SUMs to subqueries:
SELECT customers.id AS 'Customer ID',
customers.firstname AS 'Customer First Name',
customers.lastname AS 'Customer Last Name',
users.firstname AS 'Employee First Name',
users.lastname AS 'Employee Last Name',
positions.currency AS 'Currency',
cd.amount AS 'Total Deposits',
P.status AS 'Total Bets',
cs.totalbonusdeposits AS 'Total Bonuses',
positions.date AS 'Date'
FROM customers
INNER JOIN
( SELECT CustomerID, SUM(Amount) AS Amount
FROM customer_deposits
WHERE customer_deposits.status = 'approved'
GROUP BY CustomerID
) cd
ON customers.id = cd.customerid
LEFT JOIN users
ON customers.employeeinchargeid = users.id
INNER JOIN
( SELECT CustomerID, Date, SUM(Status) AS Status
FROM positions
WHERE positions.status = 'won'
AND positions.date BETWEEN '2013-10-01' AND '2013-11-01'
GROUP BY CustomerID, Date
) P
ON customers.id = positions.customerid
LEFT JOIN
( SELECT CustomerID, SUM(totalbonusdeposits) AS totalbonusdeposits
FROM customer_total_statistics
GROUP BY CustomerID
) cs
ON customers.id = customer_total_statistics.customerid
WHERE customers.isdemo = '0';
Note, you have used fields in the where clause for left joined tables which effectively turns them into an inner join, e.g.
positions.status = 'won'
If there is no match in positions, the status will be NULL, and NULL = 'won' evaluates to NULL (not true) so it will exclude all rows where there is no match in position. As such I have changed your LEFT JOIN's to INNER.