Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: Which one is better? Oracel 9i or DB2 7.2??

Re: Which one is better? Oracel 9i or DB2 7.2??

From: Mike Ault <mikerault_at_earthlink.net>
Date: 6 May 2002 08:58:54 -0700
Message-ID: <37fab3ab.0205060758.385e04b9@posting.google.com>


pobox002_at_bebub.com (Martin Burbridge) wrote in message news:<45a06b65.0205050819.641bd5ef_at_posting.google.com>...
> "The Nomad" <nobody_at_nowhere.com> wrote in message news:<5MSA8.396689$K52.64110834_at_typhoon.tampabay.rr.com>...
> >
> > I don't think I stated the issue clearly. You cannot update other rows in
> > the table being updated. My issue was as follows:
> >
> > I have parent-child relationships in a table. So, updating one row results
> > in the need to update maybe 10 other rows. You can't do this in Oracle 8i
> > because you can't modify a mutating table. During the update process, I have
> > to cache up the changes into arrays. Then, in the after part of the trigger,
> > I process all the child updates. It is VERY painful.
> >
> > Marc
>
> Oh I'm sorry, I thought I misunderstood, the answer did seem too
> obvious.
>
> I would say though that if you are designing an application that
> requires this type of logic, I would put it in packaged stored
> procedures, where you don't have these kinds of problems, and have
> those called to issue the updates. I mostly use triggers for low level
> stuff, primary keys, time stamps etc.
>
> I understand this isn't always possible when you need to co-exist in
> an environment with lots of different processes and applications
> inserting and updating by issuing straight SQL to the database. But
> then you're in a bit of a mess to begin with, so I'm not really
> surprised that things could get difficult.
>
> Anyway it sounds like you found the solution despite the challenges,
> good luck.
>
> Martin

Is there some over riding reason why you aren't following third-normal form in this design? Is it a required BOM structure? Sounds like a classic IUD anomaly from a non-normal design artifact.

Mike Received on Mon May 06 2002 - 10:58:54 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US