On May 23, 2012, at 10:35 AM, Robert Haas wrote:
>> "Exclusion constraints generalize uniqueness constraints by allowing
>> arbitrary comparison operators, not just equality....This is useful
>> for time periods and other ranges, as well as arrays."
>
> I objected to it at the time, but lost the argument.
It does rather seem to be a mistake to say that they can be used with arrays when they cannot. Or can they in some other way I haven’t noticed?
David