Alvaro Herrera <alvherre at alvh.no-ip.org> writes:
> Christopher Browne wrote:
>> Alvaro Herrera <alvherre at alvh.no-ip.org> writes:
>> > Christopher Browne wrote:
>> >
>> >> That's not "line 487" as I see it...  The full function call was...
>> >> 
>> >> log_status = DatumGetInt32(SPI_getbinval(SPI_tuptable->vals[0],
>> >> SPI_tuptable->tupdesc, 1, NULL));
>> >
>> > I think the problem may be that you're no longer able to pass a NULL
>> > as fourth argument here.
>> 
>> There are several calls to SPI_getbinval(), just one of which fails in
>> this manner.
>
> Maybe that's the only one getting called?  Maybe the others do not
> resolve into exactly the same underlying crashing macro?

Perhaps I missaid that...

There are several calls to SPI_getbinval() - only one of them is passing
in NULL, and that's the one that was failing.

A set of tests against Slony-I 1.2 and 2.0 against PostgreSQL HEAD + 8.4
+ 8.3 all passed, so it's pretty clear that this characterizes the
problem & solution.  I committed this against 1.2, 2.0, and HEAD.

Thanks for your help!
-- 
let name="cbbrowne" and tld="ca.afilias.info" in name ^ "@" ^ tld;;
Christopher Browne
"Bother,"  said Pooh,  "Eeyore, ready  two photon  torpedoes  and lock
phasers on the Heffalump, Piglet, meet me in transporter room three"