Re: Add --{no-,}bypassrls flags to createuser - Mailing list pgsql-hackers

From Kyotaro Horiguchi
Subject Re: Add --{no-,}bypassrls flags to createuser
Date
Msg-id 20220415.153341.594036939490942414.horikyota.ntt@gmail.com
Whole thread Raw
In response to Re: Add --{no-,}bypassrls flags to createuser  (Shinya Kato <Shinya11.Kato@oss.nttdata.com>)
Responses Re: Add --{no-,}bypassrls flags to createuser  (Robert Haas <robertmhaas@gmail.com>)
List pgsql-hackers
At Fri, 15 Apr 2022 14:55:48 +0900, Shinya Kato <Shinya11.Kato@oss.nttdata.com> wrote in 
> I understand. For backward compatibility, I left the ROLE clause
> option as it is and changed the IN ROLE clause option to --membership
> option.

Thanks!

-    printf(_("  -g, --role=ROLE           new role will be a member of this role\n"));
+    printf(_("  -g, --role=ROLE        new role will be a member of this role\n"));

This looks lik an unexpected change.  We shoudl preserve it, but *I*
think that we can add a synonym of the old --role for
understandability/memorability.  (By the way "-g" looks like coming
from "group", which looks somewhat strange..)

> printf(_("  -b, --belongs-to=ROLE     new role will be a member of this role\n"));

+    printf(_("  -m, --membership=ROLE     this role will be a member of new role\n"));

membership sounds somewhat obscure, it seems *to me* members is clearer

> printf(_("  -m, --member=ROLE        new role will be a member of this role\n"));

I'd like to hear others' opinions.

regards.

-- 
Kyotaro Horiguchi
NTT Open Source Software Center



pgsql-hackers by date:

Previous
From: Kyotaro Horiguchi
Date:
Subject: Re: Printing backtrace of postgres processes
Next
From: Noah Misch
Date:
Subject: Re: Skipping logical replication transactions on subscriber side