I am guessing, once the bad number is written somewhere, it gets
reentered again (documents, etc).

1. Create a new table within your current database design called Bad-Cust-Num.
2. Create a display / edit screen to populate the Bad-Cust-Num database.
3. Add a function to this screen to change 1 Cust-Num from the Bad
value to the Good value.  This transaction could run a while,
depending on how many records it has to update.
4. Modify your data entry program.  When they enter a Cust-Num,
Inquire to the Bad-Cust-Num database.  If found, utilize the correct
number instead.
5. Modify your inquiry program to query the Bad-Cust-Num database and
respond with the Good-Cust-Num.

Possibly, a better name would be the Old-Cust-Num database?  Just use
the same program in case you need to change a legitimate customer
number for some reason.  I. E. two customers who keep using the same
number when they call in.

On Tue, Nov 20, 2012 at 2:41 PM, Joel C. Ewing <[email protected]> wrote:
> On 11/20/2012 10:57 AM, Ron Thomas wrote:
>>
>> Hello.
>>
>> We  have a input file with the below format
>>
>> Date          Bad Custno    Good Cust No
>> 01/01/2012     989898            12345
>> 02/02/2012     12346567         9898790
>> 03/04/2012     676767            767689
>>
>> This file is a sequentail file  wil get appended on a daly basis , i.e
>> they will be inputting a bad customer number and a good customer number. The
>> customer numbers are used in around 20 tables and because of some typo
>> errors or any other thing the bad customer numbers got entered in to the
>> tables.
>>
>> We need to correct the same on a daily or weekly process. could some one
>> throws some light on what should be the best process we should consider
>> here. The customer numbers are primary keys in many of the tables.
>>
>> Thanks in Advance
>>
>> Regards
>> Ron T
>>
>>
> What is the logic that detects that the customer number is "bad"? The best
> solution of course would be to apply that logic at original data entry
> process, detect that the value is "bad" and force whoever is doing the
> original data entry to supply a valid number before it is saved in any
> tables - better yet include  cross checks in the process to also minimize
> the possibility that the number is valid but belongs to a different
> customer.
>
> As others have pointed out, a simple daily substitution list won't do much
> good in fixing mis-entered data in your tables if the "bad" value is valid,
> but for a different customer than the one intended. If the customer whose
> customer number was misappropriated also has transactions on the same day,
> how is a simple daily substitution list going to distinguish between valid
> and invalid entries for that customer number?
>
> --
> Joel C. Ewing,    Bentonville, AR       [email protected]
>
> ----------------------------------------------------------------------
> For IBM-MAIN subscribe / signoff / archive access instructions,
> send email to [email protected] with the message: INFO IBM-MAIN



-- 
Mike A Schwab, Springfield IL USA
Where do Forest Rangers go to get away from it all?

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to