Horton

Re: Early WIP/PoC for inlining CTEs

Nico Williams <nico@cryptonector.com>

View thread
From: Nico Williams <nico@cryptonector.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Andres Freund <andres@anarazel.de>, Jeremy Finzel <finzelj@gmail.com>, Andrew Gierth <andrew@tao11.riddles.org.uk>, pgsql-hackers@postgresql.org
Date: 2018-07-25T18:39:19Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Prevent inlining of multiply-referenced CTEs with outer recursive refs.

  2. Allow user control of CTE materialization, and change the default behavior.

  3. Split QTW_EXAMINE_RTES flag into QTW_EXAMINE_RTES_BEFORE/_AFTER.

  4. document when PREPARE uses generic plans

On Tue, Jul 24, 2018 at 07:57:49PM -0400, Tom Lane wrote:
> Andres Freund <andres@anarazel.de> writes:
> > On 2018-07-24 19:49:19 -0400, Tom Lane wrote:
> >> However, a singly-referenced SELECT CTE could reasonably be treated as
> >> equivalent to a sub-select-in-FROM, and then you would have the same
> >> mechanisms for preventing inlining as you do for those cases,
> >> e.g. OFFSET 0.  And sticking in OFFSET 0 would be backwards-compatible
> >> too: your code would still work the same in older releases, unlike if
> >> we invent new syntax for this.
> 
> > I still think this is just doubling down on prior mistakes.
> 
> Not following what you think a better alternative is?  I'd be the
> first to agree that OFFSET 0 is a hack, but people are used to it.
> 
> Assuming that we go for inline-by-default for at least some cases,
> there's a separate discussion to be had about whether it's worth
> making a planner-control GUC to force the old behavior.  I'm not
> very excited about that, but I bet some people will be.

It is widely known that CTEs in PG are optimizer barriers.

That actually is useful, and I do make use of that fact (though I'm not
proud of it).

My proposal is that PG add an extension for specifying that a CTE is to
be materialized (barrier) or not (then inlined).

Nico
--