Re: Cascade update

From: Thomas Kyte <tkyte_at_us.oracle.com>
Date: 1997/05/01
Message-ID: <3368e4e8.1126139_at_newshost>#1/1

On Tue, 29 Apr 1997 21:42:57 -0300, Roberto Bruno <rbruno_at_centroin.com.br> wrote:

>Jeremy Rickard wrote:
>
> In article <335f3321.360066_at_news.pacific.net.sg>, ECatz <ee-
> ket.tang_at_dbsp.stems.com> writes
> >I know that Oracle allows cascade delete, but what about cascade
> >update ?
>
> Don't know. Some databases do, some don't.
>
> Personally, I can't see a need for this feature, so long as you
> choose
> sensible (stable) primary keys.
> --
> Jeremy Rickard
>
> It can't. And if you have dealt with long history, you'll wish it
>could. Time shows that the hardest thing is to get a key that's really
>stable.
>
>Right now, the only way I know of doing it is temporarily disabling the
>key, updating on both tables and then enabling it again. Of course, that
>involves locking both tables during this operation. I hope you are not
>needing this for any high-availability distributed system.
>
>If anyone knows that I'm wrong, please let me know.
>

There exists a package on http://govt.us.oracle.com/ under downloadable utilities (see the quick picks on the home page) that implement cascade update while declaritive RI (primary/foreign keys) are enabled. It does it with triggers and avoid all mutating table issues.

also, in a future release of oracle, you will be able to defer the validiation of constraints until commit time, allowing you to more flexibly cascade an update.

Thomas Kyte
tkyte_at_us.oracle.com
Oracle Government
Bethesda MD

http://govt.us.oracle.com/ -- downloadable utilities

---

Opinions are mine and do not necessarily reflect those of Oracle Corporation Received on Thu May 01 1997 - 00:00:00 CEST