On Fri, 17 Feb 2006, "Tomás A. Rossi" wrote:
> "SELECT * FROM sso.organismos o LEFT OUTER JOIN motivos_bloqueos m ON(
> o.id_motivo_ultimo_bloqueo=m.id) WHERE no(o.baja_fisica) AND nombre ~* 'a'
> AND activo ORDER BY nombre"
>
> Your debug code printed:
> "[id] [nombre] [dominio] [activo] [fecha_ultima_baja] [bloqueado]
> [cant_bloqueos_heredados] [fecha_ultimo_bloqueo] [detalle_ultimo_bloqueo]
> [id_motivo_ultimo_bloqueo] [baja_fisica] [descripcion] [id] [motivo]
> [baja_fisica]"
>
> Seems that two columns has the same name (id) or something like that? How to
> workaround this?
> This didn't happen when using postgres 7.3 with driver 7.3.
>
Previous driver was retrieving the first id, but the newer one is
retrieving the second id. The behavior of duplicate names is undefined
and this change was part of a performance enhancement put in to speed up
searches by name. You need to alias one of the columns to a different
name.
SELECT o.id AS oid, m.id AS mid, ... FROM ...
Then you may safely say getInt("oid").
Kris Jurka
