Re: [HACKERS] alter table

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Vince Vielhaber <vev(at)michvhf(dot)com>
Cc: hackers(at)postgreSQL(dot)org
Subject: Re: [HACKERS] alter table
Date: 1999-07-06 21:12:55
Message-ID: 24681.931295575@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Vince Vielhaber <vev(at)michvhf(dot)com> writes:
> Can a serial type be added to an existing table using ALTER TABLE? If so
> why doesn't it work on hub? If not, I guess that explains why it doesn't
> work, so I'm open to suggestions on how to add it easily...

I'm not sure about this, but I seem to recall that ALTER TABLE ADD
COLUMN doesn't know anything about adding constraints or defaults.
And, of course, the default clause for a serial column is what *really*
makes it go. Given this lack, the fact that ALTER TABLE is also
unprepared to create the underlying sequence object for a SERIAL column
is the least of your worries ;-)

There was some discussion a few months ago about redesigning the support
for SERIAL columns to make them less of an add-on kluge and more of a
real integrated feature. I'd be inclined to think that that should
happen before we try to teach ALTER TABLE about serial columns ---
otherwise it'll just be another kluge in need of replacement. Does
anyone recall what happened with that discussion thread?

regards, tom lane

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Vince Vielhaber 1999-07-06 21:21:49 Re: [HACKERS] alter table
Previous Message Camil Coaja 1999-07-06 19:59:29 Trigger functions written in C never get called