From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Pedro Gimeno <pgsql-002(at)personal(dot)formauri(dot)es> |
Cc: | pgsql-bugs(at)postgresql(dot)org |
Subject: | Re: BUG #4120: ERROR: cache lookup failed for function 0 |
Date: | 2008-04-21 18:39:27 |
Message-ID: | 608.1208803167@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-bugs |
Pedro Gimeno <pgsql-002(at)personal(dot)formauri(dot)es> writes:
> Tom Lane wrote:
> Here's a minimal test case:
> pruebas=# CREATE FUNCTION vc_gt(varchar,varchar) RETURNS boolean
> IMMUTABLE STRICT LANGUAGE SQL AS 'SELECT CAST($1 AS TEXT) > CAST($2 AS
> TEXT);';
> CREATE FUNCTION
> pruebas=# CREATE OPERATOR > (
> PROCEDURE = vc_gt,
> LEFTARG = character varying,
> RIGHTARG = character varying,
> COMMUTATOR = <,
> RESTRICT = scalargtsel,
> JOIN = scalargtjoinsel
> );
> CREATE OPERATOR
> pruebas=# SELECT ''::varchar < ''::varchar;
> ERROR: cache lookup failed for function 0
Ah. You have actually created a "shell" operator varchar < varchar
by having written the commutator link. The error message could be
more helpful, perhaps, but I can't get too excited about it considering
that it's a consequence of a bogus operator definition. Either provide
the < operator or drop the COMMUTATOR property.
> This was part of my workaround for bug #4083. I only needed MAX so I
> only wrote the > operator. BTW, has #4083 been dismissed?
I don't think we'll do anything about it, no.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2008-04-21 20:55:15 | Re: BUG #4113: server closed the connection unexpectedly |
Previous Message | Pedro Gimeno | 2008-04-21 18:17:00 | Re: BUG #4120: ERROR: cache lookup failed for function 0 |