From: | Jan Wieck <JanWieck(at)Yahoo(dot)com> |
---|---|
To: | Boris Folgmann <boris(at)folgmann(dot)de> |
Cc: | pgsql-bugs(at)postgresql(dot)org |
Subject: | Re: BUG #6774: FOR IN SELECT LOOP ignores ORDER BY |
Date: | 2012-07-30 16:21:38 |
Message-ID: | 5016B492.5030807@Yahoo.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-bugs |
On 7/30/2012 5:56 AM, Boris Folgmann wrote:
> Hi,
>
> hubert depesz lubaczewski schrieb/wrote:
>> generally - order by datname is understood as "order by *variable
>> datname*". - which is null.
>
> It's clear that it's a shadowing problem. But it's not a "FOR IN EXECUTE"
> where a variable makes sense. I mean why is a "ORDER BY variable" valid in
> "FOR IN"? In SQL you can't use ORDER BY NULL. If the order by clause is not
> correct the whole statement should raise an error.
Note that PL/pgSQL replaces all local variables inside a query with
$-parameters for the prepared SPI plan. The parser rejects ordering by
non-integer constants, but it does not reject ordering by $-parameters
or constant expressions. (maybe it should). You can for example
SELECT * FROM something ORDER BY 'foo'||'bar';
Jan
--
Anyone who trades liberty for security deserves neither
liberty nor security. -- Benjamin Franklin
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2012-07-30 16:36:13 | Re: BUG #6774: FOR IN SELECT LOOP ignores ORDER BY |
Previous Message | Bruce Momjian | 2012-07-30 16:14:16 | Re: BUG #6785: Memory Leak in plpgsql |