Date: Mon, 6 Mar 2006 11:30:42 -0700
From: Michael Fuhr <mike@fuhr.org>
To: Emil Rachovsky <zsevgymko@yahoo.com>
Cc: "pgsql-general@postgresql.org" <pgsql-general@postgresql.org>,
	"pgsql-sql@postgresql.org" <pgsql-sql@postgresql.org>
Subject: Re: Syntax error in Execute statement
Message-ID: <20060306183042.GA14001@winnie.fuhr.org>
References: <20060306152158.81224.qmail@web54312.mail.yahoo.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20060306152158.81224.qmail@web54312.mail.yahoo.com>

On Mon, Mar 06, 2006 at 07:21:58AM -0800, Emil Rachovsky wrote:
> I created a function "coalescec" which behaves the
> same as coalesce, but uses an integer and a string.

Is there a reason you can't use the standard COALESCE and cast the
integer value to text/varchar?

> Now I'm getting an error in the folowing statements : 
> 
> query := 'select "UID" from S_Users_To_Connection
> where ConnID = ' || coalescec(conn_id,'null');
> execute query into nUID;
> 
> ERROR:  syntax error at or near "$2" at character 20
> QUERY:  SELECT   $1  into  $2 

What version of PostgreSQL are you running?  EXECUTE INTO is new
in 8.1; earlier versions would print an error like the above.

> I have put UID in quotes because it seems to be a
> keyword. 

How did you determine that?  UID isn't shown in the "SQL Key Words"
appendix of the documentation and I don't see it in the backend or
PL/pgSQL grammars.  Using it unquoted works here:

test=> CREATE TABLE foo (uid integer);
CREATE TABLE
test=> INSERT INTO foo (uid) VALUES (1);
INSERT 0 1
test=> SELECT uid FROM foo;
 uid 
-----
   1
(1 row)

I'd guess that you created the column with an uppercase quoted
identifier.  See the documentation regarding case folding and
quoted identifiers:

http://www.postgresql.org/docs/8.1/interactive/sql-syntax.html#SQL-SYNTAX-IDENTIFIERS

-- 
Michael Fuhr