Re: Break referential integrity. - Mailing list pgsql-sql

From Jan Wieck
Subject Re: Break referential integrity.
Date
Msg-id 3F0C67BA.2050209@Yahoo.com
Whole thread Raw
In response to Re: Break referential integrity.  (Stephan Szabo <sszabo@megazone23.bigpanda.com>)
List pgsql-sql
Stephan Szabo wrote:
> On Wed, 2 Jul 2003, Rudi Starcevic wrote:
> 
>> Hi,
>>
>> I know that if you have a trigger and function then drop/replace the
>> function the trigger needs
>> to be drop/replaced too so that it can see the new function.
>>
>> Is it the same for Ref. Integ. on table's too ?
>>
>> If table B's foreign key references table A and you drop/replace table A
>> then the reference from table B to table A is broken and needs to be
>> recreated ?
> 
> In recent versions, you should not be drop table A without specifying
> cascade which will drop the constraint for you (and thus you'll need to
> recreate it).  In older versions, I'm not 100% sure what'd happen, but you
> probably should drop and recreate it for good measure.

Older versions spit out a NOTICE and dropped the constraint, so they 
kinda defaulted to CASCADE.


Jan

-- 
#======================================================================#
# It's easier to get forgiveness for being wrong than for being right. #
# Let's break this rule - forgive me.                                  #
#================================================== JanWieck@Yahoo.com #



pgsql-sql by date:

Previous
From: Rod Taylor
Date:
Subject: Re: max length of sql select statement ?
Next
From: Michael A Nachbaur
Date:
Subject: Home-brewed table syncronization