RE: %scan BIF error or not?

["Bob Cozzi" <cozzi@xxxxxxxxx>]

Brad,
This is the expected/observed behavior on V4R5. When I wrote CGILIB I ran
into similar issues with %SCAN and now use other, faster techniques that
avoid %SCAN except in very simple situations.
On V5Rx that changed things in order to implement MONITOR/On-Error. I have
no evidence that %SCAN changed, but you are right, I see similar results on
V5.2 as well.
-Bob 

-----Original Message-----
From: rpg400-l-bounces@xxxxxxxxxxxx [mailto:rpg400-l-bounces@xxxxxxxxxxxx]
On Behalf Of Brad Stone
Sent: Tuesday, January 25, 2005 1:30 PM
To: RPG programming on the AS400 / iSeries
Subject: %scan BIF error or not?

I've encountered an error on a V4R5 machine with the %scan
bif that doesn't appear to happen on V5R2 and functions as
I would assume it would.

Here's the example:
eval string = 'thisisatest'
eval j = %scan('test':string:10)

Notice that the start position plus (10) the length of the
lookup string (4) is longer than the string I am scanning
(11) (this is where the problem seems to lie)

Now, I get the following error on V4R5:
Length or start position is out of range for the string
operation.

In V5R2 I simply get 0 for j and things work fine which is
how I assumed it would work.  

I have made sure the starting position isn't more than the
length of the string being scanned... it seems when the
starting position plus the length of the lookup string is
longer than the string to scan it throws this error.

Is this correct behavior, or is the behavior from V5R2
correct?  If V5R2 is correct, is there a PTF to fix V4R5?
-- 
This is the RPG programming on the AS400 / iSeries (RPG400-L) mailing list
To post a message email: RPG400-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/rpg400-l
or email: RPG400-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/rpg400-l.