A complicated query , group by issue (I guess)

Question

Hi , on a bad structured database 10g running on Windows XP , I have two tables , one for the clients paid money and the other is for clients invoices , I need to calculate the sum of invoices of one client - which identified by his FirstNAme and LastName minus the sum of his paid money beside oney that already paid when making the invoice for example x buy an item , he made an invoice ith 100$ , he paid immediately 10$ nad then later - this will be on the other table - he paid 40$ , so my final result should be 50$ ,all in respect to the same year, here is what I did
SELECT NVL(SUM(NORMAL_CLIENTS.INV_VALUE), 0) - (NVL(SUM(NORMAL_CLIENTS.PAID), 0) + NVL(SUM(NORMAL_CLIENTS_RECEIPTS.REC_VALUE), 0))
AS EXPR1, NORMAL_CLIENTS.NAME, NORMAL_CLIENTS.SURNAME
FROM NORMAL_CLIENTS INNER JOIN
NORMAL_CLIENTS_RECEIPTS ON NORMAL_CLIENTS.NAME = NORMAL_CLIENTS_RECEIPTS.NAME AND
NORMAL_CLIENTS.SURNAME = NORMAL_CLIENTS_RECEIPTS.SURNAME
WHERE (TO_CHAR(NORMAL_CLIENTS.INV_DATE, 'YYYY') =
(SELECT DATE3
FROM TEMP)) AND (TO_CHAR(NORMAL_CLIENTS_RECEIPTS.REC_DATE, 'YYYY') =
(SELECT DATE3
FROM TEMP TEMP_1))
GROUP BY NORMAL_CLIENTS.NAME, NORMAL_CLIENTS.SURNAME
so this worked for me in one condition , the client has to have records on both table , otherwise it returns null
any help will be much appreciated

Answer 1

Hi,
So that we can help you, can you provide us with some DDLs to create the tables, and some INSERTs to populate them with representative data?
You don't need to give the full structure of your tables, just relevant columns for the problem.
Also, to preserve formatting and readability, please use the tag to enclose code snippets.
Thanks.