Re: \du quite ugly in 8.4

From: Thom Brown <thombrown(at)gmail(dot)com>
To: Peter Eisentraut <peter_e(at)gmx(dot)net>
Cc: pgsql-hackers(at)postgresql(dot)org
Subject: Re: \du quite ugly in 8.4
Date: 2009-10-21 12:38:09
Message-ID: bddc86150910210538o544cda02u9f00dad1d74c2dd@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

2009/10/21 Peter Eisentraut <peter_e(at)gmx(dot)net>:
> Could someone clarify why this (from PG 8.4)
>
> # \du
>              List of roles
>   Role name   | Attributes  | Member of
> ---------------+-------------+------------
>  admin         | Create role | {}
>               : Create DB
>  postgres      | Superuser   | {}
>               : Create role
>               : Create DB
>  someotheruser |             | {}
>  someuser      |             | {}
>
> is an improvement over this (from PG 8.3)
>
> # \du
>                                 List of roles
>   Role name   | Superuser | Create role | Create DB | Connections |
> ---------------+-----------+-------------+-----------+-------------+------------
>  admin         | no        | yes         | yes       | no limit    | {}
>  postgres      | yes       | yes         | yes       | no limit    | {}
>  someotheruser | no        | no          | no        | no limit    | {}
>  someuser      | no        | no          | no        | no limit    | {}
> (4 rows)
>
> The way I see it, a perfectly clear, complete, and legible table has
> been turned into a denormalized, unreadable, and ugly pile of ASCII
> salad that moreover wastes valuable vertical screen space instead of
> using the abundant horizontal screen space.
>

I can see a possible advantage in that there is more room to list the
membership items (although I can't see that being useful most of the
time), and it also provides flexibility by providing an undefined list
of attributes compared to a defined list. For example, the new format
allows the display of "No inheritance", something not shown in the old
format.

However, if that is the limit of how this more flexible approach is
being taken advantage of, I'd tend to agree that the previous format
was fine.

Thom

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Tom Lane 2009-10-21 13:07:42 Re: Re: BUG #5065: pg_ctl start fails as administrator, with "could not locate matching postgres executable"
Previous Message Peter Eisentraut 2009-10-21 12:28:01 \du quite ugly in 8.4