Thread: SELECT INTO vs. OIDs

Up to now, if you created a table with
SELECT ... INTO foo FROM ...

then the new table "foo" would have OIDs.

As of CVS tip I have changed this to create a table without OIDs.
I'd have preferred not to make such a change at the last minute,
but the hack we had in place was quite broken.  (InitPlan() was
trying to back-patch a decision to include OID header space into
an already-created plan tree.  This did not work in any but the
simplest cases.)

If anyone is really annoyed about this, we could probably find a
solution; but I'm not inclined to expend effort on it unless there's
someone out there who's seriously unhappy.  Comments?
        regards, tom lane

On Mon, 2 Sep 2002, Tom Lane wrote:

> As of CVS tip I have changed this to create a table without OIDs.
> ...
> If anyone is really annoyed about this, we could probably find a
> solution; but I'm not inclined to expend effort on it unless there's
> someone out there who's seriously unhappy.  Comments?

I actually prefer the new behaviour to the old.

cjs
-- 
Curt Sampson  <cjs@cynic.net>   +81 90 7737 2974   http://www.netbsd.org   Don't you know, in this new Dark Age, we're
alllight.  --XTC