Albe Laurenz [2009-08-14 07:34]:
> Radoslaw Zielinski wrote:
[...]
>> The "orphans" count should be 0, obviously.
> Just to make sure that there is really an inconsistency:
> Could you pg_dump both tables and try to load them into
> another database? If that works without errors, we must have
> m
Radoslaw Zielinski wrote:
> radek=# \d kandydaci
> Table "public.kandydaci"
> Column | Type | Modifiers
> ---+--+---
>id_rekordu| bigint | not null
>id_osoby | integer | n
Grzegorz Jaśkiewicz [2009-08-13 14:23]:
> On Thu, Aug 13, 2009 at 12:36 PM, Radoslaw Zielinski
> wrote:
[...]
>> "kandydaci_fk_id_rekordu" FOREIGN KEY (id_rekordu) REFERENCES
>> rekordy(id) ON DELETE CASCADE
[...]
> since you do LEFT JOIN, indeed you can get r.id to be null.
There is a for
On Thu, Aug 13, 2009 at 12:36 PM, Radoslaw Zielinski wrote:
> Hello,
>
> I have reported this yesterday via WWW as bug 4979, but I can't see it
> in the -bugs archive. Has it been lost or are the bug reports being
> moderated...?
>
> Anyway. Pg 8.4.0 from yum.postgresql.org running on 64bit RHEL
Hello,
I have reported this yesterday via WWW as bug 4979, but I can't see it
in the -bugs archive. Has it been lost or are the bug reports being
moderated...?
Anyway. Pg 8.4.0 from yum.postgresql.org running on 64bit RHEL 5.3.
radek=# \d kandydaci
Table "public.kandydaci"
