From: | Kevin Hunter <hunteke(at)earlham(dot)edu> |
---|---|
To: | PostgresQL General <pgsql-general(at)postgresql(dot)org> |
Subject: | COALESCE function |
Date: | 2006-12-31 05:44:22 |
Message-ID: | 45974E36.1030108@earlham.edu |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Hello All,
Attempting to select two different column types with COALESCE returns
this error:
ERROR: COALESCE types smallint and character varying cannot be matched
Attempting the same thing with a CASE statement returns a similar error:
ERROR: CASE types smallint and character varying cannot be matched
I also checked Oracle's NVL command, and it throws a similar error.
Clearly, I'm not supposed to intermix two different column types into a
SELECT statement. This is because the engine needs to return a set
given the criteria, and it's difficult to do that with criteria that
/depends on the data/, yes?
Could someone explain a/the more formal reason why I can't do what I'm
trying to do?
The relevant part of my SELECT statement:
SELECT
...,
COALESCE(f.number, f.name),
...
FROM
...,
field AS f,
...
WHERE
...
;
f.number ∈ SMALLINT
f.name ∈ CHARACTER VARYING
Thank you in advance!
Kevin
P.S. If something gets lost in bit/encoding translation, ∈ = "Element Of"
From | Date | Subject | |
---|---|---|---|
Next Message | mike | 2006-12-31 06:10:29 | Re: COALESCE function |
Previous Message | David Fetter | 2006-12-31 01:36:32 | Re: Generic timestamp function for updates where field names vary |