Date: Mon, 4 Nov 2013 13:15:34 -0800 (PST)
From: David Johnston <polobo@yahoo.com>
To: pgsql-sql@postgresql.org
Message-ID: <1383599734773-5776920.post@n5.nabble.com>
In-Reply-To: <1383599125836-5776914.post@n5.nabble.com>
References: <1383596556949-5776903.post@n5.nabble.com>
 <1383597496358-5776905.post@n5.nabble.com>
 <1383599125836-5776914.post@n5.nabble.com>
Subject: Re: json and aggregate
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Precedence: bulk
Sender: pgsql-sql-owner@postgresql.org

Diway wrote
> I could use something like this (not tested!)
> --
> select fk_item_header_id as id, json_agg(row(message_id, dt_created,
> lines, size, current_part, status, fk_item_cat_id)) as data from item
> group by fk_item_header_id;
> --

CREATE TYPE 

http://www.postgresql.org/docs/9.3/interactive/sql-createtype.html

Since you have a known set of columns you define a custom type that matches
the structure and then:

SELECT fk_item_header_id, json_agg(  ( message_id, dt_create, ...
)::custom_type  ) FROM ...;

Not tested as I do not have a JSON capable release available at the moment. 
But since the type information includes the column names the conversion to
JSON should keep those names intact.

David J.


--
View this message in context: http://postgresql.1045698.n5.nabble.com/json-and-aggregate-tp5776903p5776920.html
Sent from the PostgreSQL - sql mailing list archive at Nabble.com.


-- 
Sent via pgsql-sql mailing list (pgsql-sql@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-sql