Re: psql: Add role's membership options to the \du+ command

From: Pavel Luzanov <p(dot)luzanov(at)postgrespro(dot)ru>
To: Kyotaro Horiguchi <horikyota(dot)ntt(at)gmail(dot)com>
Cc: david(dot)g(dot)johnston(at)gmail(dot)com, tgl(at)sss(dot)pgh(dot)pa(dot)us, robertmhaas(at)gmail(dot)com, david(dot)zhang(at)highgo(dot)ca, pgsql-hackers(at)lists(dot)postgresql(dot)org
Subject: Re: psql: Add role's membership options to the \du+ command
Date: 2023-04-15 13:16:26
Message-ID: e90e6775-63a0-6400-1a46-5e4a11124d50@postgrespro.ru
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On 14.04.2023 10:28, Kyotaro Horiguchi wrote:
> As David-G appears to express concern in upthread, I don't think a
> direct Japanese translation from "rolename from rolename" works well,
> as the "from" needs accompanying verb. I, as a Japanese speaker, I
> prefer a more non-sentence-like notation, similar to David's
> suggestion but with slight differences:
>
> "pg_read_all_stats (grantor: postgres, inherit, set)"

In this form, it confuses me that 'postgres' and 'inherit, set' look
like a common list.

> Come to think of this, I recalled a past discussion in which we
> concluded that translating punctuation marks appearing between a
> variable number of items within list expressions should be avoided...
>
> Thus, I'd like to propose to use an ACL-like notation, which doesn't
> need translation.
>
> ..| Mamber of |
> ..| pg_read_server_files=ais/horiguti,pg_execute_server_program=/postgres |

It's very tempting to do so. But I don't like this approach. Showing
membership options as an ACL-like column will be confusing.
Even in your example, my first reaction is that
pg_execute_server_program is available to public.
(As for the general patterns, we can also consider combining three
options into one column, like pg_class.reloptions.)

So, yet another way to discuss:

pg_read_all_stats(inherit,set/horiguti)
pg_execute_server_program(empty/postgres)

One more point. Grants without any option probably should be prohibited
as useless. But this is for a new thread.

--
Pavel Luzanov
Postgres Professional: https://postgrespro.com

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Dmitry Dolgov 2023-04-15 14:40:57 Re: [RFC] Add jit deform_counter
Previous Message Mikael Kjellström 2023-04-15 05:48:52 Re: Direct I/O