Message-ID: <4D6D4EC2.70306@gmail.com>
Date: Tue, 01 Mar 2011 12:53:38 -0700
From: Rob Sargent <robjsargent@gmail.com>
User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US;
	rv:1.9.1.16) Gecko/20101125 SUSE/3.0.11 Lightning/1.0b1
	Thunderbird/3.0.11
MIME-Version: 1.0
To: pgsql-sql@postgresql.org
References: <AANLkTinSR8Wuago7HdATetEiXQLU8fjbnm2tX5n1RGHG@mail.gmail.com>
In-Reply-To: <AANLkTinSR8Wuago7HdATetEiXQLU8fjbnm2tX5n1RGHG@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Subject: Re: distinguishing identical columns after joins


On 03/01/2011 12:47 PM, S G wrote:
> This question is particularly geared towards self-joins, but can apply
> to any join where the tables involved have any identical column names.
>  Aside from explicit column references, is there any way to pull all
> columns (*) from each table in a join and quickly append/prepend some
> identifier to distinguish them from each other?  For example, table t1
> contains columns named col1 and col2:
> 
> SELECT
>    *
> FROM
>    t1 AS a
> INNER JOIN
>    t1 AS b
> ON
>    a.col1 = b.col1
> 
> would yield a result set with column names: col1, col2, col1, col2.
> I'm looking for something that would automatically rename the columns
> like: a_col1, a_col2, b_col1, b_col2.  Does such functionality exist?
> It's not such a big deal in this example, but it can be quite tedious
> to explicitly reference and rename every single column for such joins
> when the tables involved have a very large number of columns.
> 
> I would beg for the same functionality when expanding compound
> datatypes.  For example, a compound datatype cd1 exists with fields
> named f1 and f2:
> 
> SELECT
>    ((value1, value2)::cd1).* AS a
> 
> normally produces a result set with column names: f1, f2.  I'm looking
> for something that would produce column names: a_f1, a_f2.
> 
> Thanks!
> sg
> 
select a.col1 as a_col1 etc doesn't do it for you?