Re: [HACKERS] Subselects and NOTs

Поиск
Список
Период
Сортировка
От Vadim B. Mikheev
Тема Re: [HACKERS] Subselects and NOTs
Дата
Msg-id 34EAE46F.281C6CA7@sable.krasnoyarsk.su
обсуждение исходный текст
Ответ на Re: [HACKERS] Subselects and NOTs  (Bruce Momjian <maillist@candle.pha.pa.us>)
Список pgsql-hackers
Bruce Momjian wrote:
>
> >
> > Is 'not X in' the same as 'X not in' ? Currently we have:
>
> I am not sure about this, but I believe 'not X in subselect' is
> evaluated as 'not (x in subselect)' and not as 'X not in subselect'.  Am
> I missing something?

Even 'not (x in subselect)' doesn't help in Oracle! This works just
as 'x not in subselect'.

Ok, if we don't know what standard say then could you test this
in Informix, etc. Don't forget put NULL into table in subselect
to get 'no rows selected' for 'x not in subselect'...

>
> Also they show:
>
> select *
> from test
> where x <> (select y
>             from test2)
>
> When test2 returns no rows, the query returns no rows because the
> subquery returns a single row of NULL values.

This is exactly how it works in postgres and Oracle.

Vadim

В списке pgsql-hackers по дате отправления:

Предыдущее
От: "Vadim B. Mikheev"
Дата:
Сообщение: Re: [HACKERS] results of regression tests: NetBSD/i386 v1.3
Следующее
От: "Thomas G. Lockhart"
Дата:
Сообщение: Lost a function overloading capability?