Sum columns of different record count in joined tables

Question

I have a problem with a query, please help.
I have two tables:
create table rec_a (key_code varchar(20),
                TOT_AMT varchar(20),
                INV_AMT varchar(20))
create table rec_B (key_code varchar(20),
                INVOICE_AMT varchar(20),
               PAID_AMT varchar(20))
insert into REC_A values (123123,  1168182.16, 1168182.16);
insert into REC_B values (123123, 205699.04,205699.04);
insert into REC_B values (123123,130912.78,130912.78);
insert into REC_B values (123123, 81622.87,81622.87);
insert into REC_B values (123123, 438032.43,438032.43);
insert into REC_B values (123123, 159936.17,159936.17);
insert into REC_B values (123123, 151978.87,151978.87);
table 1
KEY_CODE     TOT_AMT     INV_AMT
123123     1168182.16     1168182.16
table 2
KEY_CODE     INVOICE_AMT     PAID_AMT
123123      205699.04     205699.04
123123      130912.78     130912.78
123123      81622.87     81622.87
123123      438032.43     438032.43
123123      159936.17     159936.17
123123      151978.87     151978.87
I wrote a query to sum up all the fields (tot_amt,inv_amt,invoice_Amt,paid_amt)
SELECT B.key_code,sum(A.invoice_amt),sum(a.paid_amt),SUM(b.tot_Amt),SUM(b.inv_amt) FROM rec_a B, rec_b A WHERE A.KEY_CODE=B.KEY_CODE AND B.KEY_CODE in ('123123')
group by B.key_codeI got this:
KEY_CODE     SUM(A.INVOICE_AMT)     SUM(A.PAID_AMT)     SUM(B.TOT_AMT)     SUM(B.INV_AMT)
123123      1168182.16     1168182.16     7009092.96     7009092.96
But I expected this
KEY_CODE     SUM(A.INVOICE_AMT)     SUM(A.PAID_AMT)     SUM(B.TOT_AMT)     SUM(B.INV_AMT)
123123      1168182.16     1168182.16     1168182.16     1168182.16
Can anyone help me here?
Thanks
Edited by: 1003064 on Apr 29, 2013 3:19 AM

Answer 1

Hi,
welcome to the forum.
Please read SQL and PL/SQL FAQ
Particularly post CREATE TABLE and INSERT statement, describe the logic and post your expected output.
When you put some code and/or output please enclose it between two lines starting with {noformat}{noformat}
i.e.:
{noformat}{noformat}
SELECT ...
{noformat}{noformat}
It's also considered a good practice to mark questions as answered when the answers are satisfying your question or provide additional details.
How does rec_a and rec_b tables relates to table1 and table2?
It seems that yours is a very simple error. When you join 2 tables all the common rows get repeated.
So without grouping your table will have something like this:table 1 table 2
KEY_CODE TOT_AMT INV_AMT KEY_CODE INVOICE_AMT PAID_AMT
123123 1168182.16 1168182.16 123123 205699.04 205699.04
123123 1168182.16 1168182.16 123123 130912.78 130912.78
123123 1168182.16 1168182.16 123123 81622.87 81622.87
123123 1168182.16 1168182.16 123123 438032.43 438032.43
123123 1168182.16 1168182.16 123123 159936.17 159936.17
123123 1168182.16 1168182.16 123123 151978.87 151978.87
Now sum up all the columns and you will understand why you get that result.
Regards.
Al
Regards.
Al
Edited by: Alberto Faenza on Apr 29, 2013 12:17 PM