Re: [HACKERS] A small type extension example for the contrib dire ctory (fwd)

From: Zeugswetter Andreas DBT <Andreas(dot)Zeugswetter(at)telecom(dot)at>
To: "'pgsql-hackers(at)hub(dot)org'" <pgsql-hackers(at)hub(dot)org>
Subject: Re: [HACKERS] A small type extension example for the contrib dire ctory (fwd)
Date: 1998-01-26 13:15:58
Message-ID: 219F68D65015D011A8E000006F8590C6010A51A9@sdexcsrv1.sd.spardat.at
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

I wrote:
>>> In the second query, the first two rows have been grouped, but
shouldn't
>>> they not be since b is NULL? I thought that NULL != NULL?
>
>Note that:
>NULL <> NULL is false
>NULL = NULL is false

This was wrong, I digged up some more docu and found:
<something> op NULL is unknown note that there exists a
third boolean state (name it ?)
`Unknown values occur when part of an expression that uses an arithmetic
operator is null.`
If a whole where expression evaluates to unknown the row is not chosen.
not: t -> f, f -> t, ? -> ?
and: ? and t -> ?, ? and f -> f, ? and ? -> ?
or: t or ? -> t, f or ? -> ?, ? or ? -> ?

Order by: `Null values are ordered as less than values that are not
null.`
but
NULL {>|<|<=|>=} value is unknown

Andreas

Browse pgsql-hackers by date

  From Date Subject
Next Message Bruce Momjian 1998-01-26 15:08:07 Re: [HACKERS] Re: [QUESTIONS] How is PostgreSQL doing?
Previous Message Zeugswetter Andreas DBT 1998-01-26 11:51:54 Re: [HACKERS] Group By, NULL values and inconsistent behaviour.