Re: primary key display in psql

From: Robert Haas <robertmhaas(at)gmail(dot)com>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: Peter Eisentraut <peter_e(at)gmx(dot)net>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: primary key display in psql
Date: 2010-01-13 22:03:33
Message-ID: 603c8f071001131403u12d3fac8x5400c58879dc4a58@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Wed, Jan 13, 2010 at 4:47 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> Peter Eisentraut <peter_e(at)gmx(dot)net> writes:
>> I think we could easily improve that by having it look something like
>> this instead:
>
>>     Table "public.test2"
>>  Column |  Type   | Modifiers
>> --------+---------+-----------
>>  a      | integer | PK
>>  b      | integer | PK
>> Indexes:
>>     "test2_pkey" PRIMARY KEY, btree (a, b)
>
> Spelling out "primary key" would seem to be more in keeping with existing
> entries in that column, eg we have "not null" not "NN".
>
> I think this is a sensible proposal for a single-column PK, but am less
> sure that it makes sense for multi-col.  The modifiers column is
> intended to describe column constraints; which a multi-col PK is not,
> by definition.

Yeah, IIRC, MySQL shows PRI for each column of a multi-column primary
key, and I think it's horribly confusing. I wouldn't even be in favor
of doing this just for the single-column case, on the grounds that it
makes the single and multiple column cases asymmetrical. IMO, the \d
output has too many bells and whistles already; the last thing we
should do is add more.

...Robert

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Robert Haas 2010-01-13 22:05:06 Re: patch to implement ECPG side tracing / tracking ...
Previous Message Kevin Grittner 2010-01-13 21:54:41 Re: Serializable Isolation without blocking