| From: | Robert Haas <robertmhaas(at)gmail(dot)com> |
|---|---|
| To: | Simon Riggs <simon(at)2ndquadrant(dot)com> |
| Cc: | Csaba Nagy <ncslists(at)googlemail(dot)com>, Josh Berkus <josh(at)agliodbs(dot)com>, pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: ALTER TABLE ... REPLACE WITH |
| Date: | 2010-12-15 16:21:00 |
| Message-ID: | AANLkTim2qC6BC+oPjXrptc2ecZFt9TETTQN=XVffcPfm@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Wed, Dec 15, 2010 at 10:50 AM, Simon Riggs <simon(at)2ndquadrant(dot)com> wrote:
> On Wed, 2010-12-15 at 12:17 +0100, Csaba Nagy wrote:
>
>> But one problem would be when the replaced table is the _parent_ for a
>> foreign key relationship. I don't think you can have that constraint
>> pre-verified on the replacement table and simply replacing the content
>> could leave the child relations with orphans.
>
> Good point.
>
> The only sensible way to handle this is by putting the FK checks into
> check pending state (as discussed on a different thread).
>
> We would probably need to disallow FKs with DELETE or UPDATE CASCADE
> since it would be difficult to execute those.
I'm still wondering if TRUNCATE CONCURRENTLY would be a more elegant solution.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Florian Pflug | 2010-12-15 16:24:00 | Re: CommitFest wrap-up |
| Previous Message | Andrew Dunstan | 2010-12-15 16:19:53 | Re: Complier warnings on mingw gcc 4.5.0 |