On Fri, Mar 23, 2018 at 6:55 AM, Simon Riggs <simon@2ndquadrant.com> wrote:
> Peter, if you have the code and you consider it important that this
> subfeature is in PostgreSQL, why not post the code so we can commit
> it?
Fair enough. Attached patch shows what I'm on about. This should be
applied on top of 0001_merge_v23e_onconflict_work.patch +
0002_merge_v23e_main.patch. I'm not expecting an authorship credit for
posting this patch.
One thing that the test output shows that is interesting is that there
is never a "SubPlan 1" or "InitPlan 1" in EXPLAIN output -- it seems
to always start at "SubPlan 2". This probably has nothing to do with
CTEs in particular. I didn't notice this before now, although there
were no existing tests of EXPLAIN in the patch that show subplans or
initplans.
Is this somehow related to the issue of using two RTEs for the target
relation? That's certainly why we always see unaliased target table
"m" with the alias "m_1" in EXPLAIN output, so I would not be
surprised if it caused another EXPLAIN issue.
--
Peter Geoghegan