NAKANO Yoshihisa <nakano(dot)yosihisa(at)jp(dot)fujitsu(dot)com> writes:
> Tom Lane wrote:
>> After looking at the test case a bit, I have an alternate approach:
>> constraints on the column will have DEPENDENCY_AUTO type, while
>> constraints using the column will have DEPENDENCY_NORMAL type.
> O.K. It seems nicer than my solution.
> Please find the patch attached. I fixed the patch to decide the order
> of the list by deptype of the pg_depend record.
Patch applied with some minor comment improvements. Thanks!
regards, tom lane