Re: Query to find first missing integer in a field
From: Michael Austin <maustin_at_firstdbasource.com>
Date: Tue, 07 Apr 2009 15:40:10 -0500
Message-ID: <PSOCl.15777$W06.14182_at_flpi148.ffdc.sbc.com>
ddf wrote:
> On Apr 7, 2:51 pm, dean <deanbrow..._at_yahoo.com> wrote:
>
> This has 'DON'T DO THIS' written all over it in large, bold, red
> letters as it's a disaster in the making. I suppose you're looking to
> 'backfill' missing sequence numbers and that is a recipe for failure.
>
> Seriously rethink this before you're in too deep and end up with an
> application that won't run because it won't scale.
>
>
>
> David Fitzjarrell
Date: Tue, 07 Apr 2009 15:40:10 -0500
Message-ID: <PSOCl.15777$W06.14182_at_flpi148.ffdc.sbc.com>
ddf wrote:
> On Apr 7, 2:51 pm, dean <deanbrow..._at_yahoo.com> wrote:
>> Does anyone know a good way to find the first free integer number that >> does not exist in a table in a specified column? I can think of a >> couple of tedious ways of doing this but they are hardly elegant. >> Assume there could be any integer present up to 2 billion.
>
> This has 'DON'T DO THIS' written all over it in large, bold, red
> letters as it's a disaster in the making. I suppose you're looking to
> 'backfill' missing sequence numbers and that is a recipe for failure.
>
> Seriously rethink this before you're in too deep and end up with an
> application that won't run because it won't scale.
>
>
>
> David Fitzjarrell
not to mention any referential integrity that may get munged because it was designed to have missing numbers. IMO there are VERY few reasons to EVER have non-gap sequential numbers. Even in accounting you can VOID a check number... take the hint. Received on Tue Apr 07 2009 - 15:40:10 CDT