Horton

Re: Early WIP/PoC for inlining CTEs

Nico Williams <nico@cryptonector.com>

View thread
From: Nico Williams <nico@cryptonector.com>
To: David Fetter <david@fetter.org>
Cc: Andrew Gierth <andrew@tao11.riddles.org.uk>, pgsql-hackers@postgresql.org
Date: 2018-07-25T15:32:06Z
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 Wed, Jul 25, 2018 at 07:42:37AM +0200, David Fetter wrote:
> Please find attached a version rebased atop 167075be3ab1547e18 with
> what I believe are appropriate changes to regression test output.  The
> other changes to the regression tests output are somewhat puzzling, as
> they change the actual results of queries.  I've also attached both
> the "leftover" diff and the files to which it should be applied.

I think the SQL programmer needs some control over whether a CTE is:

 - a materialized view -- and therefore a barrier
 - a view (which can then be inlined by the optimizer)

It is possible to add a keyword for this purpose in the WITH syntax:

    WITH   VIEW (...) AS a_view
         , MATERIALIZED VIEW (...) AS a_barrier
    ...;

This would be a lot like creating TEMP views, but without the catalog
overhead.

(I wonder how hard it would be to partiion the OID namespace into
temp/persistent ranges so that temp schema elements need not be written
into the catalog.)

Nico
--