Re: Complex (for me) SQL question

[Dan <dan27649@xxxxxxxxx>]

In our case, the first occurrence of a duplicate for a part number made it
possible for all subsequent duplicates for that part number to occur.
Primarily because someone embedded SQL that did a SELECT ... INTO
datastructure.  When more than one record was in the result set, the query
failed.  They tested SQLCOD >= 0 and <> 100 to discern that query found a
record.  However, when more than one record is returned, SQLCOD = -811, and
RECORD_FOUND was set to false, causing a subsequent INSERT.  So, finding
when the first duplicate occurred allowed us to define a date range across
all affected part numbers and we were able to eliminate several potential
root causes.

- Dan

On Tue, Jan 19, 2016 at 7:25 PM, Michael Schutte <mschutte369@xxxxxxxxx>
wrote:

> If you were looking to see when duplicates occurred, I would have
> personally preferred...
> select item
> from file
> group by item
> having count(1) > 1
>
> So if this query was running periodically, I would send off an email as
> soon as at least one record was read from this.