Robert Haas <robertmhaas@gmail.com> writes:
> Just having an option to preload frozen tuples dodges all of these
> issues by throwing our hands up in the air, but it does have the
> advantage of being more general. Even if we do that I'm not sure it
> would be a bad thing to try to solve this issue in a somewhat more
> principled way, but it would surely reduce the urgency.
Yeah. ISTM the whole point of TRUNCATE is "I don't care about
serializability for this operation, give me efficiency instead".
So I see nothing wrong with a (non-default) option for COPY that
similarly trades away some semantic guarantees for efficiency's sake.
There are an awful lot of bulk-load scenarios where people will gladly
take that trade, and are not very interested in halfway points either.
regards, tom lane