Re: Odd CALLPRC behavior when using RPG indicator with RTNVAL

On Thu, Mar 12, 2009 at 4:57 PM, David Gibbs <david@xxxxxxxxxxxx> wrote:

Charles Wilt wrote:

While it may be a pain, David has had plenty of time to stop using the
original work around once IBM provided a correct way to handle the
issue.

"If it ain't broke, don't fix it".

What about when it is broke and just has a band-aid on it? ;)

Should it be noted as being band-aided, with an intent to actually fix
it when possible?

Then again, it depends on how strongly IBM noted that the band-aid was band-aid.

This is a technique that's worked for many years ... and there was no expectation that it would stop working ... it's hard to justify a change for a strictly 'technical' reason.

And, to be honest, I wasn't aware of the new technique ... even though it was made available 4 releases ago.  Add to that, we always target a release a number of revs below the current version.  So our coding techniques have to lag also.

And I don't honestly recall the workaround <grin>. I started on the
box at v3r6, but mostly maintaining RPG III. Don't recall when I
started with RPG IV and ILE, but honestly, I to this day I don't have
a lot of CLLE --> RPGLE interaction let alone lots where I need the
ExtProc(*CL) keyword.

Charles Wilt