On Thu, 2003-06-19 at 15:00, Tom Lane wrote:
> Rod Taylor <rbt@rbt.ca> writes:
> > Right now if the column exists in the child table, the add column is
> > rejected. I assume that will remain.
>
> Have you actually tried it?
I used different datatypes which, of course, was the wrong test.
When I update the values in t1 I need to ensure I only get data from t1
that has the new column NULL.
Constraints to be checked are still only the ones attached to the
altered table directly. A constraint on t1 only will not affect data on
t2.
--
Rod Taylor <rbt@rbt.ca>
PGP Key: http://www.rbt.ca/rbtpub.asc