Jan Wieck <JanWieck(at)yahoo(dot)com> writes:
>> that plpgsql's handling of command-counter changes is broken, but it
>> does not follow that sprinkling the code with SetQuerySnapshot is wise.
> Why do you blame PL/pgSQL for that? I don't see a single
> reference to the command counter from the PL/pgSQL sources.
> All it does is using SPI. So does "using SPI" by itself count
> as "boken"?
Sorry: SPI is broken, not plpgsql. Does that make you feel better?
regards, tom lane