From: | tgl(at)postgresql(dot)org (Tom Lane) |
---|---|
To: | pgsql-committers(at)postgresql(dot)org |
Subject: | pgsql: Fix equivclass.c's not-quite-right strategy for handling X=X |
Date: | 2009-09-29 01:21:02 |
Message-ID: | 20090929012102.A25DA753FB7@cvs.postgresql.org |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-committers |
Log Message:
-----------
Fix equivclass.c's not-quite-right strategy for handling X=X clauses.
The original coding correctly noted that these aren't just redundancies
(they're effectively X IS NOT NULL, assuming = is strict). However, they
got treated that way if X happened to be in a single-member EquivalenceClass
already, which could happen if there was an ORDER BY X clause, for instance.
The simplest and most reliable solution seems to be to not try to process
such clauses through the EquivalenceClass machinery; just throw them back
for traditional processing. The amount of work that'd be needed to be
smarter than that seems out of proportion to the benefit.
Per bug #5084 from Bernt Marius Johnsen, and analysis by Andrew Gierth.
Tags:
----
REL8_3_STABLE
Modified Files:
--------------
pgsql/src/backend/optimizer:
README (r1.41 -> r1.41.2.1)
(http://anoncvs.postgresql.org/cvsweb.cgi/pgsql/src/backend/optimizer/README?r1=1.41&r2=1.41.2.1)
pgsql/src/backend/optimizer/path:
equivclass.c (r1.9.2.2 -> r1.9.2.3)
(http://anoncvs.postgresql.org/cvsweb.cgi/pgsql/src/backend/optimizer/path/equivclass.c?r1=1.9.2.2&r2=1.9.2.3)
pgsql/src/test/regress/expected:
select.out (r1.18 -> r1.18.2.1)
(http://anoncvs.postgresql.org/cvsweb.cgi/pgsql/src/test/regress/expected/select.out?r1=1.18&r2=1.18.2.1)
pgsql/src/test/regress/sql:
select.sql (r1.14 -> r1.14.2.1)
(http://anoncvs.postgresql.org/cvsweb.cgi/pgsql/src/test/regress/sql/select.sql?r1=1.14&r2=1.14.2.1)
From | Date | Subject | |
---|---|---|---|
Next Message | User Eulerto | 2009-09-29 03:38:34 | pgspy - pgspy: Imported Sources |
Previous Message | Tom Lane | 2009-09-29 01:20:55 | pgsql: Fix equivclass.c's not-quite-right strategy for handling X=X |