Re: JDBC behaviour

Bill Moran <wmoran@potentialtech.com> writes:
> On Sat, 20 Feb 2016 16:29:09 +0000
> Vitalii Tymchyshyn <vit@tym.im> wrote:
>> Well, I suppose replacing simple copy with procedural per-row function
>> would give huge performance hit. Also what method do you propose to use in
>> the code? Savepoints?

> Not at all. PL/PGSQL's ON ERROR handling can manage this without needing
> savepoints.

Actually, plpgsql's exception blocks *are* savepoints under the hood.
The backend engine does not have any way of recovering from errors other
than a (sub)transaction abort, which means you can't do this without a
savepoint or equivalent.

            regards, tom lane