Re: [HACKERS] transition table behavior with inheritance appearsbroken (was: Declarative partitioning - another take)

On Mon, May 1, 2017 at 10:01 AM, Robert Haas <robertmhaas@gmail.com> wrote:

> It seems pretty clear to me that this is busted.

I don't think you actually tested anything that is dependent on any
of my patches there.

> Adding this as an open item.  Kevin?

It will take some time to establish what legacy behavior is and how
the new transition tables are impacted.  My first reaction is that a
trigger on the parent should fire for any related action on a child
(unless maybe the trigger is defined with an ONLY keyword???) using
the TupleDesc of the parent.  Note that the SQL spec mandates that
even in a AFTER EACH ROW trigger the transition tables must
represent all rows affected by the STATEMENT.  I think that this
should be independent of triggers fired at the row level.  I think
the rules should be similar for updateable views.

This will take some time to investigate, discuss and produce a
patch.  I think best case is Friday.

--
Kevin Grittner
VMware vCenter Server
https://www.vmware.com/