From: | Hannu Krosing <hannu(at)tm(dot)ee> |
---|---|
To: | Chris Bitmead <chris(at)bitmead(dot)com> |
Cc: | Chris Bitmead <chrisb(at)nimrod(dot)itg(dot)telstra(dot)com(dot)au>, PostgreSQL Development <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: Thus spoke SQL3 (on OO) |
Date: | 2000-05-22 09:38:57 |
Message-ID: | 39290031.EFB912F7@tm.ee |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Chris Bitmead wrote:
>
> Hannu Krosing wrote:
>
> > it would be hard to define RI by just saying that some field references "an
> > OID",
> > often you want to be able do define something more specific.
> >
> > It would be too much for most users to require that all primary and foreign
> > keys must be of type OID.
>
> Since it would be object and relational, you could do either. But all
> pure object databases _always_ rely on oid to define relationships, and
> that is likely to be all an ODMG inteface would support.
Is the ODMG interface available on the net, or is the plan to do a Poet clone
?
> Unless we want to break new ground anyway.
We would need some syntax to distinguish between REFERENCES (primary key) and
REFERENCES (oid).
Of course we would also need fast lookups by oid and oid->object lookup
tables(s)/function(s) but that's another part of the story.
--------------
Hannu
From | Date | Subject | |
---|---|---|---|
Next Message | Kardos, Dr. Andreas | 2000-05-22 09:47:07 | Re: OO / fe-be protocol |
Previous Message | Alfred Perlstein | 2000-05-22 09:15:06 | Re: rolap |