Re: Can this be done with a trigger?

From: <>
Date: Tue, 5 Aug 2008 12:48:28 -0700 (PDT)
Message-ID: <>

On Aug 5, 12:16 pm, Aya the Vampire Slayer <> wrote:
> wa:
> <snip>
> >The SOURCE is not needed in this table. But, this would fire the
> >trigger where SOURCE is needed. The SOURCE will be derived within the
> >PL/SQL code. I do not think this is possible. I may have to let the
> >trigger fire, then go back and update the row.....
> Create a View on Table1 selecting * from Table1 and also adding a null
> column called SOURCE at the end.
> create or replace view vw_tbl1 as
> select t.*,
> null SOURCE
> from table1 t
> ;
> Change your PL/SQL procedure to insert into the view instead of directly
> into the table, passing whatever it is that the PL/SQL is able to figure
> out for the SOURCE column as part of the INSERT parameters (since you
> are now inserting into the view where SOURCE exists as a column).
> Then, create your trigger on the view and have the trigger insert all of
> the information into Table2, including the SOURCE column, after
> inserting into table1.
> If you are worried about people inserting directly into the table, then
> simply deny access to the table and only allow access to the view. (here
> is me hoping you aren't letting users log in directly to the schema that
> contains all of these objects...)
> --
> "Care must be exorcised when handring Opiticar System as it is apts to
> be sticked by dusts and hand-fat." --Japanese Translators
> "Keep your fingers off the lens." --Elton Byington, English Translator

I must say, I do like your approach, that is, if if it did not yield this result:

Tue Aug 5 14:41:58 2008
Errors in file /u01/NI00/admin/udump/ni00_ora_13221.trc: ORA-07445: exception encountered: core dump [_intel_fast_memcpy.J() +30] [SIGSEGV] [Address not mapped to object] [0x000000000] [] []

More investigation is needed. Hope I do not have to log an SR with Oracle.... Received on Tue Aug 05 2008 - 14:48:28 CDT

Original text of this message