Re: Assert failure of the cross-check for nullingrels

Richard Guo <guofenglinux(at)gmail(dot)com> writes:
> On Thu, May 18, 2023 at 3:34 AM Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>> After some poking at it I hit on what seems like a really simple
>> solution: we should be checking syn_righthand not min_righthand
>> to see whether a Var should be considered nullable by a given OJ.

> I thought about this solution before but proved it was not right in
> https://www.postgresql.org/message-id/CAMbWs48fObJJ%3DYVb4ip8tnwxwixUNKUThfnA1eGfPzJxJRRgZQ%40mail.gmail.com
> I checked the query shown there and it still fails with v3 patch.

Bleah. The other solution I'd been poking at involved adding an
extra check for clone clauses, as attached (note this requires
8a2523ff3). This survives your example, but I wonder if it might
reject all the clones in some cases. It seems a bit expensive
too, although as I said before, I don't think the clone cases get
traversed all that often.

Perhaps another answer could be to compare against syn_righthand
for clone clauses and min_righthand for non-clones? That seems
mighty unprincipled though.

regards, tom lane