Re: [HACKERS] Subselects are in CVS...

ocie@paracel.com

From: ocie@paracel.com
To: vadim@sable.krasnoyarsk.su (Vadim B. Mikheev)
Cc: hackers@postgreSQL.org
Date: 1998-02-13T18:14:34Z
Lists: pgsql-hackers
Impressive, just one comment:

> 5. I need in advice: if subquery introduced with NOT IN doesn't return
>    any tuples then qualification is failed, yes ?

I think this is backwards.  Consider the following statements:

"1 not in (1,2)" FALSE
"1 not in (2)" TRUE
"1 not in ()" ?

I would tend to think that the ? should be TRUE.  I.E. every value
qualifies as not in an empty list of tuples.  The second item was
true, so why should removing the 2 from the list make the statement
false?

Ocie Mitchell