Re: Convert from free to fixed

Wow,
Scott really let it go this time! Scott for what its worth, I have always
appreciated your comments, thoughts and suggestions. I know that I have
from time to time shown my dislike for a few things or individuals on
here, but your comments are well spoken and hit the nail on the head!

We need more of this from you......

Douglas




From: Scott Klement <rpg400-l@xxxxxxxxxxxxxxxx>
To: RPG programming on the IBM i / System i <rpg400-l@xxxxxxxxxxxx>
Date: 09/28/2010 01:52 PM
Subject: Re: Convert from free to fixed
Sent by: rpg400-l-bounces@xxxxxxxxxxxx



Hello,

1) *PSSR won't catch file errors... unless you code INFSR(*PSSR), that
is. Even if you do, it only catches file errors for files that are
global to the module. Not for those local to the subprocedure.

2) *PSSR can call itself again when an error occurs. This makes it
clumsy.

While I certainly see your point about *PSSR being a "catch-all"
routine... it's not a perfect solution for one. On the other hand, if
you do this:

MONITOR;
exsr MyMainLogic; // or callp
ON-ERROR;
// do catch-all logic here
ENDMON;

You will catch all errors, without futzing around with INFSR or worrying
about local/global scope. And you don't have to worry about it being
called in a loop because an error occurs in the ON-ERROR.

Granted, there are ways to make *PSSR work... but MONITOR is cleaner
and easier... and it can also handle errors (as opposed to just being a
catch-all) which means if you have to learn only one technique, you can
do it all with MONITOR.

I would suggest at least trying MONITOR in a few places, get acquainted
with it, and see if you like it. If you've used it a bunch of times and
you really still prefer *PSSR, then stick with *PSSR. But I personally
find MONITOR much easier and less clumsy.


On 9/28/2010 10:55 AM, hockchai Lim wrote:

There is no doubt that one should use monitor on places where failure is
likely to occur and handle that specific error accordingly. But if I

want

to guarantee that my RPG program never goes into a MSGW status, I

usually

use *pssr. Why wrap the entire main procedure with an ugly monitor

opcode

when I can use a *pssr to provide me that guarantee.