Re: sqlsmith: ERROR: XX000: bogus varno: 2

Robert Haas <robertmhaas(at)gmail(dot)com> writes:
> On Sun, Dec 19, 2021 at 4:17 PM Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>> I don't see anything particularly surprising here. pg_get_expr is only
>> able to cope with expression trees over a single relation, but ON UPDATE
>> rules can refer to both OLD and NEW relations. Maybe we could make the
>> error message more friendly, but there's not much else to be done,
>> I think.

> +1 for making the error message more friendly.

The problem is that the spot where it's thrown doesn't have a lot of
context. We can fix that by having pg_get_expr itself check for
out-of-spec Vars before starting the recursion, which adds a bit of
overhead but I don't think we're terribly concerned about that.

I figured this would be just a quick hack in ruleutils.c, but was
dismayed to find the regression tests falling over, because some
bozo neglected to teach nodeFuncs.c about partition expressions.
It might be a good idea to back-patch that part, before we find
some other place that fails.

regards, tom lane