Re: BUG #6064: != NULL, <> NULL do not work [sec=UNCLASSIFIED]

=20
>> The real problem here then is that the documentation showing
>> the boolean comparison operators does not mention this
=20
> This page does, at length:

But not in the table, in fact it doesn't even mention the IS NULL, IS NOT N=
ULL operators at all.
=20
>http://www.postgresql.org/docs/9.0/interactive/functions-comparison.html
=20
>What page are you looking at?
I was looking at that page but didn't read the detail because I thought the=
 information was in the=20
table and the detail was just textual examples. In so much of the documenta=
tion, the rest of the page
just writes out in long hand what can be inferred from the syntax. People d=
o not read online like a book,
they scan. If the information is not upfront it just won't be seen.
=20
>> And indeed the parser does not generate warnings either.
=20
>It would be totally wrong for the parser to generate warnings about
>correct behavior.

The behaviour of the generated code may well be correct and indeed I agree =
that it is but from=20
everything you and the detailed documentation have said column !=3D NULL is=
 at least deprecated=20
and is highly likely to indicate a programming error. It is totally normal =
for a parser to warn=20
against archaic or dangerous constructs. The idea being to avoid subtle run=
time bugs that are
hard to track down. This happens in Ada, Java and even C and many other lan=
guages.

Regards,
Michael
=20

IMPORTANT: This email remains the property of the Department of Defence and=
 is subject to the jurisdiction of section 70 of the Crimes Act 1914. If yo=
u have received this email in error, you are requested to contact the sende=
r and delete the email.