Tom Lane wrote:
> Chris Campbell <chris@bignerdranch.com> writes:
>
>>True, as long as there's a hook to do the inlining/rewriting before
>>the query's planned. I guess we can see function calls at the parse
>>stage, check to see if they're SQL functions or not, grab the prosrc,
>>do the substitution, then re-parse?
>
>
> pull_up_subqueries in prepjointree.c would be the appropriate place
> I think: if it's an RTE_FUNCTION RTE, look to see if function is SQL
> and has the other needed properties, if so replace it by an RTE_SUBQUERY
> RTE with the correct subquery, then recurse to try to flatten the
> subquery. (Note: I'm in the middle of hacking that code to flatten
> UNION subqueries, so you might want to wait till I commit before
> starting on a patch ;-))
If we are talking about inserting the function definition into the query as a
subquery and then letting the parser treat it as a subquery, then I see no
reason to use either the existing function or view subsystems. It sounds more
like we are discussing a macro language.
CREATE MACRO foo(bar,baz) AS $$ select a from b where b > bar and b < baz $$;
Then when you query
SELECT * FROM foo(1,7) AS f WHERE f % 7 = 3
you get a macro expansion as such:
SELECT * FROM (a from b where b > bar and b < baz) AS f WHERE f % 7 = 3
Then whatever optimizations the query planner can manage against a subquery will
work for macros as well.
Thoughts?
