Re: Inspecting generated PlannerInfo - Mailing list pgsql-novice

From Tom Lane
Subject Re: Inspecting generated PlannerInfo
Date
Msg-id 18571.1457019773@sss.pgh.pa.us
Whole thread Raw
In response to Inspecting generated PlannerInfo  (Fábio Moreira <fabio@dias.moreira.nom.br>)
List pgsql-novice
=?UTF-8?Q?F=C3=A1bio_Moreira?= <fabio@dias.moreira.nom.br> writes:
> From what I could find, PostgreSQL’s tests check for output correctness but
> not if the query planner is actually applying specific optimizations to the
> input query; is that correct?

There are quite a few regression tests that look at EXPLAIN output to
verify that particular optimizations happen (or not) as intended.
There's no user-facing mechanism other than that.  For development
purposes, you could consider inserting pprint() calls on Paths at
critical points, or just invoking pprint manually from gdb --- the latter
is what I usually do.

            regards, tom lane


pgsql-novice by date:

Previous
From: Fábio Moreira
Date:
Subject: Inspecting generated PlannerInfo
Next
From: "Miguel Beltran R."
Date:
Subject: Re: Divide table raw into chunks