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: question about plsql exceptions

Re: question about plsql exceptions

From: Frank van Bortel <frank.van.bortel_at_gmail.com>
Date: Sun, 26 Feb 2006 11:58:35 +0100
Message-ID: <dts0t1$ku4$4@news3.zwoll1.ov.home.nl>


Mladen Gogala wrote:
> On Sat, 25 Feb 2006 15:11:12 -0800, Jim Kennedy wrote:
>

>> I find it very useful.

>
> Actually, I don't. When I'm loading something that is not supposed to be
> in the database (vast majority of my data loads), I'm always doing that
> with errors=0 option, so that the load is terminated when the 1st error
> is encountered and I'm informed of any problems. Of course, I always load
> into a staging schema first, check the untrusted data and process them. I
> have to confess, I didn't need MERGE statement more then twice in my whole
> career. That doesn't diminish what you said to the OP, which I completely
> agree with, I only don't see MERGE as being tremendously useful.
>

A merge is more effective that a pl/sql construct with an exception handler, is all I claimed.
And yes - only when you *know* your data being merged is correct.
Maybe you ought to look into it for the final stage of your data loads ;)
-- 
Regards,
Frank van Bortel

Top-posting is one way to shut me up...
Received on Sun Feb 26 2006 - 04:58:35 CST

Original text of this message

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