Re: psql: Add role's membership options to the \du+ command - Mailing list pgsql-hackers

From Tom Lane
Subject Re: psql: Add role's membership options to the \du+ command
Date
Msg-id 1525117.1680628353@sss.pgh.pa.us
Whole thread Raw
In response to Re: psql: Add role's membership options to the \du+ command  (Robert Haas <robertmhaas@gmail.com>)
Responses Re: psql: Add role's membership options to the \du+ command
List pgsql-hackers
Robert Haas <robertmhaas@gmail.com> writes:
> I'm not sure what the right thing to do is here. It's a problem to
> have new information in the catalogs that you can't view via
> \d<whatever>. But displaying that information as a string of
> characters that will be gibberish to anyone but an expert doesn't
> necessarily seem like it really solves the problem. However, if we
> spell out the words, then it gets bulky. But maybe bulky is better
> than incomprehensible.

The existing precedent in \du definitely leans towards "bulky":

regression=# \du
                                   List of roles
 Role name |                         Attributes                         | Member of
-----------+------------------------------------------------------------+-----------
 alice     | Cannot login                                               | {bob}
 bob       | Cannot login                                               | {}
 postgres  | Superuser, Create role, Create DB, Replication, Bypass RLS | {}

It seems pretty inconsistent to me to treat the role attributes this
way and then economize in the adjacent column.

Another advantage to spelling out the SQL keywords is that it removes
the question of whether we should translate them.

I wonder if, while we're here, we should apply the idea of
joining-with-newlines-not-commas to the attributes column too.
That's another source of inconsistency in the proposed display.

            regards, tom lane



pgsql-hackers by date:

Previous
From: Pavel Stehule
Date:
Subject: Re: possible proposal plpgsql GET DIAGNOSTICS oid = PG_ROUTINE_OID
Next
From: Pavel Stehule
Date:
Subject: Re: proposal: psql: show current user in prompt