Re: any way for ORDER BY x to imply NULLS FIRST in 8.3?

From: Simon Riggs <simon(at)2ndquadrant(dot)com>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: rihad <rihad(at)mail(dot)ru>, pgsql-general(at)postgresql(dot)org
Subject: Re: any way for ORDER BY x to imply NULLS FIRST in 8.3?
Date: 2007-11-07 11:25:40
Message-ID: 1194434740.4251.34.camel@ebony.site
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On Tue, 2007-11-06 at 09:48 -0500, Tom Lane wrote:

> Essentially the same text appears in SQL2003. Any application that
> depends on one particular choice here is therefore broken, or at least
> has chosen to work with only about half of the DBMSes in the world.

If an application has already made that choice then we should allow them
the opportunity to work with PostgreSQL. The application may be at
fault, but PostgreSQL is the loser because of that decision.

The SQL Standard says that the default for this is defined by the
implementation; that doesn't bar us from changing the implementation if
we wish. We can do that without changing PostgreSQL's historic default.

Perhaps we can have a parameter?

default_null_sorting = 'last' # may alternatively be set to 'first'

(or another wording/meaning.)

That is what I thought you'd implemented, otherwise I would have
suggested this myself way back. This new parameter would be a small
change, but will make a major difference to application portability.

This seems like the key to unlocking your new functionality for most
people.

--
Simon Riggs
2ndQuadrant http://www.2ndQuadrant.com

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message rihad 2007-11-07 11:25:46 prepared statements suboptimal?
Previous Message Reg Me Please 2007-11-07 11:19:48 Re: Postgresql simple query performance question