RE: Imperfect solutions

From: Stephan Szabo <sszabo(at)megazone23(dot)bigpanda(dot)com>
To: Christopher Kings-Lynne <chriskl(at)familyhealth(dot)com(dot)au>
Cc: Hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: RE: Imperfect solutions
Date: 2001-06-06 01:58:46
Message-ID: Pine.BSF.4.21.0106051855210.12756-100000@megazone23.bigpanda.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers


On Wed, 6 Jun 2001, Christopher Kings-Lynne wrote:

> > > Those two points are already mentioned - I have another 90%
> > patch ready to
> > > go that will add that functionality as well...
> >
> > As a question, are you doing anything to handle dropping referenced unique
> > constraints or are we just waiting on that until a referencing system
> > is built?
>
> By that do you mean: what happens when you drop a primary key that is
> referenced by a foreign key?
>
> My answer: Forgot about that ;) I'll see what I can do but anytime
> investigation of foreign keys is required it's a real pain. Foreign keys
> are kinda next on my list for work, so I might look at it then if it's too
> difficult right now. (I've got a query that can find all foreign keys on a
> relation, and what they relate to, that I'm going to add to psql).

I wouldn't worry all that much about it since you could still break it
with drop index, but I wanted to know if you'd done anything with it
and if so how general it was.

How'd you do the splitting of the arguments to get the columns referenced?
That was the biggest problem I was having, trying to get the bytea split
up. (Well, without writing a function to do it for me)

> My other questions then are:
>
> Does anything else (other than fk's) ever reference a primary key?
> What can reference a unique key?

Foreign keys are the only one I know of, but they can reference either.

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Christopher Kings-Lynne 2001-06-06 02:09:55 RE: Imperfect solutions
Previous Message Christopher Kings-Lynne 2001-06-06 01:58:05 RE: Feature request : Remove identifier length constraints