Joe Conway <mail@joeconway.com> writes:
> But BuildDescForRelation leaves initializes the tupdesc like this:
> desc = CreateTemplateTupleDesc(natts, UNDEFOID);
> The UNDEFOID later causes an assertion failure in heap_formtuple when
> you try to use the tupdesc to build a tuple.
So far, I haven't seen any reason to think that that three-way has-OID
stuff accomplishes anything but causing trouble ... but I've applied
this patch for the moment. I hope to get around to reviewing the
HeapTupleHeader hacks later in the weekend.
regards, tom lane