misternumbertwo
Registered User.
- Local time
- , 07:26
- Joined
- Dec 29, 2008
- Messages
- 11
Hi everyone,
I'm currently working on a database that handles invoices. In a certain month, around 25,000 invoices are processed. Some of them are repeating but with different amounts that I need to get the total.
It shouldn't be a problem right? I just need to do a summation and group it by invoice numbers. However, here's my problem.
191992 - $20
191992A - $62
191992B - $32
191992C - $12
191992D - $55
The query should produce the sum of the invoices above as well. The suffix letter should be disregarded.
Can anyone figure out how i can do this? If i remove every record's last character, the first record in my example would not be counted. I also can't do a len() function because the invoice numbers vary in length.
I'm currently working on a database that handles invoices. In a certain month, around 25,000 invoices are processed. Some of them are repeating but with different amounts that I need to get the total.
It shouldn't be a problem right? I just need to do a summation and group it by invoice numbers. However, here's my problem.
191992 - $20
191992A - $62
191992B - $32
191992C - $12
191992D - $55
The query should produce the sum of the invoices above as well. The suffix letter should be disregarded.
Can anyone figure out how i can do this? If i remove every record's last character, the first record in my example would not be counted. I also can't do a len() function because the invoice numbers vary in length.