Re: WDSC vs SEU

I fully agree, understanding activation groups is the proper step. But
considering how many people are in attendance at conference sessions on
this topic, it's clearly not a quick thing to get one's hands around.
And while the scenario of unchanged defaults and running converted
programs is fine, there will always be the time when someone will change
a program to take advantage of subprocedures or some ILE function, and
may get into a bind (no pun intended) by not having a good grasp of
where various pieces are running.
I just wanted to point out that there can be a risk in doing a mass
change of CLP to CLLE without somehow addressing the override issue.

Bill Reed
Rock of Ages Corp.


-----Original Message-----
From: wdsci-l-bounces+breed=barre.rockofages.com@xxxxxxxxxxxx
[mailto:wdsci-l-bounces+breed=barre.rockofages.com@xxxxxxxxxxxx] On
Behalf Of Wilt, Charles
Sent: Thursday, October 25, 2007 9:10 AM
To: Websphere Development Studio Client for iSeries
Subject: Re: [WDSCI-L] WDSC vs SEU

Bill,

Assuming the defaults haven't been changed, if you convert RPG III to
RPG IV using CVTRPGSRC, convert
CLP to CLLE and continue to use CRTBNDxxx (PDM option 14-Compile).

Then both the CLLE and the RPGLE programs will be created with
DFTACTGRP(*YES) and will continue to
work like OPM programs.

These conversions are indeed trivial.

Things become not trivial, once you specify DFTACTGRP(*NO) in any of the
programs in the job stream.
Before doing so, you need to understand activation groups.

While OVRSCOPE(*JOB) may fix problems, it is IMHO not usually the
"correct" answer. OVRSCOPE(*JOB) is
sometimes needed, but more often it is used as a brute force fix by
those who don't understand
activation groups.

Charles