Re: pgsql: Remove over-optimistic Assert. - Mailing list pgsql-committers

From Richard Guo
Subject Re: pgsql: Remove over-optimistic Assert.
Date
Msg-id CAMbWs4_dOrnBgOQ0WLU4axRDVae8+YMr895JeHEXmd755ZwH+w@mail.gmail.com
Whole thread Raw
In response to pgsql: Remove over-optimistic Assert.  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: pgsql: Remove over-optimistic Assert.  (Tom Lane <tgl@sss.pgh.pa.us>)
Re: pgsql: Remove over-optimistic Assert.  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-committers

On Thu, Feb 2, 2023 at 8:40 AM Tom Lane <tgl@sss.pgh.pa.us> wrote:
Remove over-optimistic Assert.

In commit 2489d76c4, I'd thought it'd be safe to assert that a
PlaceHolderVar appearing in a scan-level expression has empty
nullingrels.  However this is not so, as when we determine that a
join relation is certainly empty we'll put its targetlist into a
Result-with-constant-false-qual node, and nothing is done to adjust
the nullingrels of the Vars or PHVs therein.  (Arguably, a Result
used in this way isn't really a scan-level node, but it certainly
isn't an upper node either ...)
 
It seems this is the only case we can have PlaceHolderVar with non-empty
nullingrels at scan level.  So I wonder if we can manually adjust the
nullingrels of PHVs in this special case, and keep the assertion about
phnullingrels being NULL in fix_scan_expr.  I think that assertion is
asserting the right thing in most cases.  It's a pity to lose it.

Currently for the tlist of a childless Result, we special-case ROWID_VAR
Vars in set_plan_refs and thus keep assertions about varno != ROWID_VAR
in fix_scan_expr.  Do you think we can special-case PHVs at the same
place by setting its phnullingrels to NULL?  I'm imagining something
like attached.

Thanks
Richard
Attachment

pgsql-committers by date:

Previous
From: Robins Tharakan
Date:
Subject: Re: pgsql: Do assorted mop-up in the planner.
Next
From: Tom Lane
Date:
Subject: Re: pgsql: Do assorted mop-up in the planner.