Puzzler of the day - 5.0 Windows Setup

[Kenneth Brody]

On 5/3/2011 4:49 PM, Nancy Palmquist wrote:

[... Computers A and C work.  Computer B doesn't properly select records...]

> When it is broken, here is what I get.  Output prints two lines for each
> record (doubling the statement balance) and it starts the output at
> billing account 05575 instead of 05502 (all data before 05575 is gone).
[...]

I was going to suggest the possibility that computer B was pointed to a 
different data set, until you wrote:

> Now if I have Computer B, run from 05502 to 05574 - it works correctly.
> If I have Computer B, run from 05576 to 05683 - it works correctly.
> If I have Computer B, run just 05575 - it works correctly.
>
> I do not use lookup (dash) in the programming so I am not roaming
> around.  I do use the PRINT command but that would not cause the index
> to include the duplicate records since that happens on the generating
> side not sort/select side.
>
> To my experience, I can say SELECT - 20 times in a scan table and I
> still only get one selection.

Correct.

> But just to make sure, I did check and I
> only issue the SELECT command once, without condition, so it is not
> repeating that.
>
> BTW - No associated fields are used in this file.  Eliminate that
> possibility.
[...]

How about having your sort/select processing log each record number 
processed, along with "selected"/"not selected", and see what you get?  How 
does computer B compare to the ones that work?

-- 
Kenneth Brody