From: | David Pratt <fairwinds(at)eastlink(dot)ca> |
---|---|
To: | Jason Minion <jason(dot)minion(at)sigler(dot)com> |
Cc: | pgsql-admin(at)postgresql(dot)org |
Subject: | Re: Question of Table Design and Foreign Keys |
Date: | 2005-05-31 16:33:37 |
Message-ID: | BD59D321-D1F1-11D9-AEE7-000A27B3B070@eastlink.ca |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-admin |
Hi Jason, many thanks Jason for your reply.
On Tuesday, May 31, 2005, at 11:25 AM, Jason Minion wrote:
> I think the way you want to look at it is: if a record exists in
> sample_attributesjoin, then it needs to reference both samples
> and attributes. You don't need to worry about samples or attributes
> being "off" because they don't have "enough" references, instead
> you want to ensure that if a record exists (whether inserted,
> updated, or references are deleted) in sample_attributesjoin
> that the sample_id and attribute_id are always valid.
>
Yes, this is what I was thinking but just needed to ensure this was the
right approach since I have a number of tables and similar
circumstances. So if there is no attributes for a sample record when I
join the tables and query, I will just have nothing returned for these
records and no trouble with integrity, am I correct? When I up update
or delete records that I have added to attributesjoin, the referential
integrity of attributes and samples table is maintained using the
constraints, have I got this right?
Regards,
David
From | Date | Subject | |
---|---|---|---|
Next Message | Marc G. Fournier | 2005-05-31 17:45:56 | Major flood of mail to lists ... |
Previous Message | Myli | 2005-05-31 15:44:46 | new antidote discovered |