Re: Re: "Oracle's ROWNUM"

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Barry Lind <barry(at)xythos(dot)com>
Cc: pgsql-general(at)postgresql(dot)org
Subject: Re: Re: "Oracle's ROWNUM"
Date: 2001-07-30 18:12:12
Message-ID: 21527.996516732@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

Barry Lind <barry(at)xythos(dot)com> writes:
> ... For example rownum = 2 will return
> no rows because the first row returned by the query has by definition a
> rownum of 1, but the where predicate prevents this row from being
> returned, thus it can never get to a rownum value of 2 to satisfy the
> where predicate.

So in other words, a construct accessible in the WHERE clause is defined
in terms of what happens far downstream of WHERE. This cannot possibly
have sane behavior. I won't even ask about join queries...

AFAICT, LIMIT/OFFSET do the same job in a much more logical fashion.
Let's stick with those, and not try to copy the more brain-dead aspects
of Oracle.

regards, tom lane

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Peter Choe 2001-07-30 20:22:24 database information
Previous Message Stephan Szabo 2001-07-30 17:05:39 Re: RULE vs TRIGGER