Re: [PATCHES] Inherited Constraints

Поиск
Список
Период
Сортировка
От Hannu Krosing
Тема Re: [PATCHES] Inherited Constraints
Дата
Msg-id 1141822424.4009.5.camel@localhost.localdomain
обсуждение исходный текст
Ответ на Re: [PATCHES] Inherited Constraints  (Bruce Momjian <pgman@candle.pha.pa.us>)
Ответы Re: [PATCHES] Inherited Constraints  (Stephan Szabo <sszabo@megazone.bigpanda.com>)
Список pgsql-hackers
Ühel kenal päeval, E, 2006-03-06 kell 17:25, kirjutas Bruce Momjian:
> Hannu Krosing wrote:
> > ?hel kenal p?eval, E, 2006-03-06 kell 12:12, kirjutas Bruce Momjian:
> > > Added to TODO:
> > > 
> > >       o Prevent parent tables from altering or dropping constraints
> > >         like CHECK that are inherited by child tables
> > > 
> > >         Dropping constraints should only be possible with CASCADE.
> > > 
> > > and we already have this in TODO:
> > > 
> > >         o %Prevent child tables from altering or dropping constraints
> > >           like CHECK that were inherited from the parent table
> > > 
> > > so I think we now have all the failure cases documented.
> > 
> > If you want to be consistent, then ALTER TABLE ONLY ADD CONSTRAINT  .. 
> > should also be forbidden, so you can't create non-inherited constraints
> 
> I don't have a problem with creating ONLY constraints on parents and
> children.  We just don't want them to be removed/modified if they are
> shared.

Well, when you delete a constraint from child, the constraint becomes an
"ONLY" constraint on parent. If you allow ONLY constraints on parents,
then why disallow dropping them from childs ?

IIRC the original complaint about being able to drop constraints from
children was that inherited tables not being bound by constraints on
parents was unexpected/broken.

I.E when you have 

CREATE TABLE T(i int check (i>0));

then you would be really surprised by getting -1 out from that table.

---------------
Hannu




В списке pgsql-hackers по дате отправления:

Предыдущее
От: Markus Schaber
Дата:
Сообщение: Re: [SQL] Interval subtracting
Следующее
От: Simon Riggs
Дата:
Сообщение: Re: Merge algorithms for large numbers of "tapes"