Re: BUG #6236: Query execution bug

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: "Boris Matkov" <borism(at)devart(dot)com>
Cc: pgsql-bugs(at)postgresql(dot)org
Subject: Re: BUG #6236: Query execution bug
Date: 2011-09-30 15:33:35
Message-ID: 7767.1317396815@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

"Boris Matkov" <borism(at)devart(dot)com> writes:
> INSERT INTO btest0 (id, name) VALUES ((SELECT DISTINCT id FROM btest0 WHERE
> name = $1 LIMIT 1), $1) - was executed successfully

> INSERT INTO btest0 (name,id) VALUES ($1, (SELECT DISTINCT id FROM btest0
> WHERE name = $1 LIMIT 1)) - I got error:
> ---------------------------
> inconsistent types deduced for parameter $1
> ---------------------------

> Script for the btest0 table creating:

> CREATE TABLE btest0
> (
> id integer NOT NULL,
> name character varying(50) NOT NULL,
> value double precision,
> CONSTRAINT pk_btest0 PRIMARY KEY (id)
> )

It may be annoying, but it's not a bug. The equality test uses the
"text = text" operator, so if that's seen first then the parameter gets
resolved as text. You'd be better off to declare the type of the
parameter instead of assuming that it will always be resolved the way
you want ... but if you can't be bothered to do that, using text instead
of varchar as the column type would avoid most of the cases where you'll
see something like this.

regards, tom lane

In response to

Browse pgsql-bugs by date

  From Date Subject
Next Message Tom Lane 2011-09-30 17:35:08 Re: BUG #6235: Delete fails with ON DELETE rule on inherited table
Previous Message Kevin Grittner 2011-09-30 14:40:55 Re: FW: PostGre compatible to RHEL 6.1