Intentional, or bug?

From: Kovacs Baldvin <kb136(at)hszk(dot)bme(dot)hu>
To: <pgsql-sql(at)postgresql(dot)org>
Subject: Intentional, or bug?
Date: 2001-09-16 14:14:27
Message-ID: Pine.GSO.4.33.0109161601050.10242-100000@ural2
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-sql

Hi!

I probably found an inconsistence of the working of the =
operator. Please read ahead, if you're interested.

I have two tables, let't call them a and b. They are:

a)
x | y
-----------------------------
1 | 100
2 | NULL

b)
y | z
-----------------------------
1 | 'hallelujah'
NULL | 'hmm...'

Now let's issue this query:

select b.z from a,b where a.y=b.y;

The answer is: 'hallelujah', but it omits 'hmm...'.

Now let's issue the following:

select NULL = NULL;

The answer is: true!!!

My question is why does it omit the rows with NULL at the
first case?

However, I have a real life problem, when not all data is given in
a column, so I need a query to include the NULL=NULL rows. I can
workaround it as defining a new operator, ~=, with the
meaning "both null, or =". I did it, it works, but very slow.

So, is it an intentinal way of functioning, or it is bug somewhere?

Thanks,
Baldvin

Responses

Browse pgsql-sql by date

  From Date Subject
Next Message Kovacs Baldvin 2001-09-16 15:00:14 Correction for the previous letter:
Previous Message Tom Lane 2001-09-16 04:34:15 Re: optimizing queries and indexes...