Date: Tue, 2 May 2000 11:23:12 +0200
From: Michael Meskes <meskes@postgresql.org>
To: Matthew Denny <mdenny@CS.Berkeley.EDU>
Cc: pgsql-interfaces@postgresql.org
Subject: Re: ECPG and WHENEVER statements
Message-ID: <20000502112312.A3097@fam-meskes.de>
Mail-Followup-To: Matthew Denny <mdenny@CS.Berkeley.EDU>,
	pgsql-interfaces@postgresql.org
References: <14592.44665.629100.20848@caracas.CS.Berkeley.EDU>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.1.9i
In-Reply-To: <14592.44665.629100.20848@caracas.CS.Berkeley.EDU>;
	from mdenny@CS.Berkeley.EDU on Fri, Apr 21, 2000 at 12:56:29PM
	-0700

[I'm still behind with my mail, so please take my spologies if this already
has been answererd.]

On Fri, Apr 21, 2000 at 12:56:29PM -0700, Matthew Denny wrote:
> In my ECPG code, I have a EXEC SQL WHENEVER NOT FOUND DO BREAK
> statement in order to end a cursor fetch loop when the cursor runs out of
> input tuples.  Unfortunately, it appears ECPG puts such a conditional
> break statement after every SQL EXEC in the file and then throws an

No, only after statements that may result in NOT FOUND. For instance it is
not set after a cursor close statement.

> error if these  subsequent statements are not in a switch or loop
> statement (which most of my INSERT, UPDATE, and DELETE calls are not).
> Thus, I currently  have to wrap all my INSERT, UPDATE, and DELETE
> calls in dummy switch statements.  Is there a more elegant way
> (e.g. via another EXEC SQL WHENEVER statement) to toggle WHENEVER NOT
> FOUND DO BREAK at different points in the source file?  A more elegnt
> solution would be greatly appreciated.

How about WHENEVER NOT FOUND BREAK before the loop and WHENEVER NOT FOUND
CONTINUE after it?

Michael
-- 
Michael Meskes                         | Go SF 49ers!
Th.-Heuss-Str. 61, D-41812 Erkelenz    | Go Rhein Fire!
Tel.: (+49) 2431/72651                 | Use Debian GNU/Linux!
Email: Michael@Fam-Meskes.De           | Use PostgreSQL!