Horton

Re: TupleDescAttr bounds checks

Robert Haas <robertmhaas@gmail.com>

View thread View on postgr.es
From: Robert Haas <robertmhaas@gmail.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2026-03-24T10:45:55Z
Lists: pgsql-hackers 
On Sun, Mar 22, 2026 at 11:54 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
>  Anyway, you should get this fix pushed.

Done, and I'll plan to commit the other patches later today.

Also, if on any occasion you happen to feel that I'm not being
aggressive enough in committing something I've previously posted, feel
free to take matters into your own hands. I often wait a bit to see if
anybody will object to things or comment on them, and in this case
there were compounding factors like (1) the weekend and (2) being very
busy looking into other problems that test_plan_advice turned up.
Since this was such a simple fix and you'd +1'd it, I would have felt
comfortable putting it in right away, but I simply haven't had a
moment to spare until now, and I use that term loosely given that I do
not normally use the time between 6am and 7am for to commit patches.
Anyway, the point is: I'm virtually always happy when someone else
decides to commit one of my patches; it saves me a non-trivial amount
of time and I'm not offended.

Thanks,

-- 
Robert Haas
EDB: http://www.enterprisedb.com

Commits

  1. Avoid unsafe access to negative index in a TupleDesc.

  2. Disallow system columns in COPY FROM WHERE conditions.

  3. Add a test for creating an index on a whole-row expression.

  4. Bounds-check access to TupleDescAttr with an Assert.

  5. Prevent spurious "indexes on virtual generated columns are not supported".