Peter Eisentraut <peter.eisentraut@2ndquadrant.com> writes:
> On 06/09/2018 18:25, Tom Lane wrote:
>> Also, in further service of providing insulation between SPI nesting
>> levels, I thought it'd be a good idea for SPI_connect() to reset the
>> globals to zero/NULL after saving them. This ensures that a nesting
>> level can't accidentally be affected by the state of an outer level.
>> It's barely possible that there's somebody out there who's *intentionally*
>> accessing state from a calling SPI level, but that seems like it'd be
>> pretty dangerous programming practice. Still, maybe there's an argument
>> for omitting that hunk in released branches.
> These look like good changes. I'm not sure about backpatching them, but
> as you say the chances that someone wants to do this intentionally are low.
Yeah. I'd put a higher probability on the idea that this will fix
somebody's latent bug in code that's never been tested inside an
outer level of SPI call. It would, for example, work to rely on
SPI_tuptable being initially NULL, as long as you hadn't tried it
inside an outer call.
regards, tom lane