Re: Program failure analysis

On Fri, Oct 13, 2017 at 10:15 AM, Buck Calabro <kc2hiz@xxxxxxxxx> wrote:

It's somewhat surprising that more RPG programmers haven't chimed in
with their pattern for logging at the moment of failure. That thousands
of programmers recreate errors in the dev environment is powerful
evidence that I need to rethink what I'm doing.

Personally, I do not believe the evidence is as powerful as you seem to.

First, specifically regarding ON-EXIT, this is very new feature in
RPG. I think it is reasonable to expect that even folks who try to
stay fairly current are still working out how to best incorporate it
(if at all!) into their programming practices. In other words, you are
near the front of the pack here. There aren't a lot of folks to
follow.

Second, just because "thousands of programmers" do something, doesn't
mean they are doing the best thing. And related to that: even if the
thousands of programmers are doing something good and right, this
doesn't rule out other alternatives that may *also* be good and right.
(All the more so if those alternatives are new and untested.)

I think that this thread has taught me that
ON-EXIT really really is for sub-procedures; blocks that do one thing.
At least I'll be less likely to misapply it next time :-)

That's not one of the takeaways I got from this thread.

I still think ON-EXIT is for procedures, period. In other words,
whatever you want to stick in a procedure, that's what ON-EXIT is
*for*.

Now, it may be the case that, as a rule of thumb, it is desirable to
have smaller procedures. Thus *if* it is desirable for ON-EXIT to have
smaller scope, that's merely secondary to the desirability of smaller
procedures.

John Y.