RE: Last record keeps repeating

["Haas, Matt" <Matt.Haas@xxxxxxxxxxx>]

The only problem with sqlstt = '00000' is that a warning message (like
data conversion happening on a FETCH) will cause your loop to terminate
when you may not want it to. Scott's example also has an issue. The
issue is that if there is an I/O problem (like if you're connected to a
remote machine) or a data issue that makes the query invalid, it will
stay in the loop and reprocess the last read record.

What I've found to be best is to check the first two characters (which
are the error class) of SQLSTT to see if they are '00' (success) or '01'
(warning). This, of course, begs the question: When do you know if it's
okay to skip 01's? That answer requires that you take a look at the
SQLSTT documentation
(http://publib.boulder.ibm.com/iseries/v5r1/ic2924/index.htm?info/rzala/
rzalastc.html -- I know this is for V5R1 but SQLSTT values are a
standard and they don't change very often so this should be perfect for
pretty much anyone on the list) and make an informed decision.

Matt

-----Original Message-----
From: rpg400-l-bounces@xxxxxxxxxxxx
[mailto:rpg400-l-bounces@xxxxxxxxxxxx] On Behalf Of Carel Teijgeler
Sent: Friday, September 16, 2005 12:32 PM
To: rpg400-l@xxxxxxxxxxxx
Subject: Re: Last record keeps repeating

It could be changed into 

>>  if sqlstt = '00000';

Regards,
Carel teijgeler
*********** REPLY SEPARATOR  ***********

On 16-9-05 at 11:15 Scott Klement wrote:

>>  dow record_loop = 'Y';
>> c+ fetch next from maincursor into :main_ds
>>  if sqlstt <> '02000';
>>     do whatever you do here
>> else;
>>   // Leave loop
>> endif;
>> enddo;
>
>Won't this cause an endless loop if any error aside from EOF occurs?