Re: Early WIP/PoC for inlining CTEs

On Wed, Jul 25, 2018 at 04:18:42PM +0100, Andrew Gierth wrote:
> >>>>> "David" == David Fetter <david(at)fetter(dot)org> writes:
>
> David> Please find attached a version rebased atop 167075be3ab1547e18
> David> with what I believe are appropriate changes to regression test
> David> output. The other changes to the regression tests output are
> David> somewhat puzzling, as they change the actual results of queries.
>
> Both of those changes are the result of volatile CTEs being inlined more
> than once (in one case, as part of an explicit test to ensure that CTEs
> are being materialized and not multiply evaluated).
>
> If you look for the XXX comment in the patch, it should be easy to add a
> check that skips inlining if cterefcount > 1 and
> contains_volatile_functions is true.

Thanks for the broad hints!

Please find attached the next version, which passes 'make check'.

Best,
David.
--
David Fetter <david(at)fetter(dot)org> http://fetter.org/
Phone: +1 415 235 3778

Remember to vote!
Consider donating to Postgres: http://www.postgresql.org/about/donate