Re: Question about unique constraint

Just wanted to send a thank you on behalf of my colleague. He is using the information provided to complete this work.

on Behalf of Vadim
Michael

On Wed, Jun 29, 2016 at 4:23 AM, Jonathan Lewis <jonathan_at_jlcomp.demon.co.uk
> wrote:

>
> It's true that if you create a unique or primary key constraint in the
> novalidate state and there is no pre-existing index Oracle will create
> non-unique index, but if a unique index already exists then Oracle will
> happily use it. Unless you particularly want a non-unique index supporting
> the primary/unique key (which would be the case if you wanted it to be
> deferrable) it's far better to base it on a unique index because there are
> a number of small but significant differences in workload depending on your
> choice.
>
> Regards
> Jonathan Lewis
> http://jonathanlewis.wordpress.com
> _at_jloracle
>
> ________________________________________
> From: oracle-l-bounce_at_freelists.org [oracle-l-bounce_at_freelists.org] on
> behalf of Tefft, Michael J [Michael.J.Tefft_at_snapon.com]
> Sent: 29 June 2016 10:59
> To: Powell, Mark; napacunningham_at_gmail.com; oracle-l_at_freelists org
> Subject: RE: Question about unique constraint
>
> >> I believe you will need to create your primary-key constraint using a
> non-unique index <<
>
> > Why?
>
> Because he did not want the constraint to be validated at the time it was
> created.
>
> Mike Tefft
> --
> http://www.freelists.org/webpage/oracle-l
>
> --
> http://www.freelists.org/webpage/oracle-l
>
>
>

-- 
Michael Cunningham

--
http://www.freelists.org/webpage/oracle-l