Kerem Kat <keremkat@gmail.com> writes:
> On Sat, Sep 24, 2011 at 19:51, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> Why? CORRESPONDING at a given set-operation level doesn't affect either
>> sub-query, so I don't see why you'd need a different representation for
>> the sub-queries.
> In the planner to construct a subquery out of SetOperationStmt or
> RangeTblRef, a new RangeTblRef is needed.
> To create a RangeTableRef, parser state is needed and planner assumes
> root->parse->rtable be not modified
> after generating simple_rte_array.
Actually, after looking at the code again, I don't think you need any of
that, since there's already a SubqueryScan node being inserted into the
plan. You just need to improve generate_setop_tlist so that it can deal
with cases where the mapping from subplan targetlist to the setop output
columns isn't one-to-one.
regards, tom lane