Re: remove_useless_joins vs. bug #19560

Richard Guo <guofenglinux(at)gmail(dot)com> writes:
> ... The exception is
> flatten_join_alias_vars, which descends into every PHV and asserts on
> SubPlans. I changed it to leave PHVs of upper levels alone; such a
> PHV cannot contain join aliases of the level being flattened anyway.

BTW, this comment reminds me strongly of an earlier draft I had,
which eliminated the assertion failure by just not letting
ChangeVarNodes recurse into outer-level PHVs, on the grounds that
we can ignore their contents until they get replaced by Params.
That didn't seem terribly principled on its own, so I also made
eval_const_expressions not do anything to outer-level PHVs, but
then I started to wonder just how far the implications should extend.
When I realized that a localized fix in extract_lateral_references()
seemed possible I gave up on that approach. But maybe it's the way
to go. It has the attractive property of removing useless work.
Attached is as far as I got with the idea before switching to the
other way.

regards, tom lane