Date: Sat, 10 Jun 2017 23:25:25 -0700
From: Noah Misch <noah@leadboat.com>
To: kgrittn@gmail.com, thomas.munro@enterprisedb.com
Cc: pgsql-hackers@postgresql.org
Subject: RTE_NAMEDTUPLESTORE, enrtuples and comments
Message-ID: <20170611062525.GA1628882@rfd.leadboat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.5.24 (2015-08-30)
Precedence: bulk
Sender: pgsql-hackers-owner@postgresql.org

While completing my annual src/backend/nodes/*funcs.c audit, I noticed defects
in commit 18ce3a4 changes to RangeTblEntry:

1. Field relid is under a comment saying it is valid for RTE_RELATION only.
   Fields coltypes, coltypmods and colcollations are under a comment saying
   they are valid for RTE_VALUES and RTE_CTE only.  But _outRangeTblEntry()
   treats all of the above as valid for RTE_NAMEDTUPLESTORE.  Which is right?

2. New fields enrname and enrtuples are set only for RTE_NAMEDTUPLESTORE, yet
   they're under the comment for RTE_VALUES and RTE_CTE.  This pair needs its
   own comment.

3. Each of _{copy,equal,out,read}RangeTblEntry() silently ignores enrtuples.
   _equalRangeTblEntry() ignores enrname, too.  In each case, the function
   should either use the field or have a comment to note that skipping the
   field is intentional.  Which should it be?

This fourth point is not necessarily a defect: I wonder if RangeTblEntry is
the right place for enrtuples.  It's a concept regularly seen in planner data
structures but not otherwise seen at parse tree level.

nm


-- 
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers