[PATCH] pgbench: fix incorrect long option name in error message

From: Semih Doğan <semih702do(at)gmail(dot)com>
To: pgsql-hackers(at)lists(dot)postgresql(dot)org
Subject: [PATCH] pgbench: fix incorrect long option name in error message
Date: 2026-07-15 18:30:24
Message-ID: CALOtZ7tuWisV=v0cUY_q6PLHJ-fOiQ7ZN476JwmM0PyV0t5i7Q@mail.gmail.com
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hi hackers,

While experimenting with pgbench I passed a non-numeric value to -c
and got:

pgbench -c abc
pgbench: error: invalid value "abc" for option -c/--clients

The message refers to --clients, but no such option exists -- the
actual long option is --client (singular), as shown in --help and in
the documentation. Following the message's suggestion fails:

pgbench --clients=10
pgbench: illegal option -- clients

All the sibling option_parse_int() calls in pgbench.c spell the
option name correctly (-j/--jobs, -T/--time, ...); this one looks
like a simple typo. I verified it is present on master and in
released versions (18.4).

The attached patch fixes the message and updates the TAP test in
002_pgbench_no_server.pl that was asserting the wrong spelling.

Regards,
Semih Doğan

Attachment Content-Type Size
0001-Fix-incorrect-long-option-name-in-pgbench-error-mess.patch application/octet-stream 1.5 KB

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Jeff Davis 2026-07-15 18:36:56 Re: REASSIGN OWNED vs. relisshared dep on !relisshared
Previous Message Dmitry Fomin 2026-07-15 18:29:21 Re: [PATCH v1 0/7] Wait event timing and tracing instrumentation