From: | Stephan Szabo <sszabo(at)megazone23(dot)bigpanda(dot)com> |
---|---|
To: | "M(dot) I(dot)" <google(dot)clp(at)alma(dot)ch> |
Cc: | Stephan Szabo <sszabo(at)megazone23(dot)bigpanda(dot)com>, <pgsql-general(at)postgresql(dot)org> |
Subject: | Re: Inheritance: delete parent deletes children |
Date: | 2002-10-04 16:13:12 |
Message-ID: | 20021004090851.B37537-100000@megazone23.bigpanda.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
On Fri, 4 Oct 2002, M. I. wrote:
> At 17.37 04.10.02, Stephan Szabo wrote:
>
> >On 2 Oct 2002, M. I. wrote:
> >
> > > But I just realized that things actually go into that "logging" table,
> > > not into the child tables. I wonder how the database knows which row
> > > in logging is related to which row in a child table.
> >
> >Well, selects, inserts, updates, deletes by default go through the
> >inheritance tree if you do them on a parent, so select * from
> >logging will show rows from the child tables. You can use ONLY
> >to turn off that behavior (select * from ONLY logging) should
> >show rows only in that particular table and not the children.
>
> Oh, I see. Yes, it seems to make sense.
>
> But the Delete is scary. It looks like I could inadvertently empty the
> whole database with a DELETE FROM logging; if I forget that it
> happens to have (many) children.
>
> The day I die, I certainly hope my children will survive me :-) ...
>
> Is there something I can do with rules, to protect me from this
> "exterminator" attitude of the database?
Actually there is the SQL_INHERITANCE configuration option which
changes the default behavior (see docs).
From | Date | Subject | |
---|---|---|---|
Next Message | Roland Roberts | 2002-10-04 16:24:09 | Re: [GENERAL] Fast Deletion For Large Tables |
Previous Message | scott.marlowe | 2002-10-04 16:08:54 | Re: [SQL] arrays |