RE: Are subroutines codependent?

Nathan, I don't really see how your exsr example is any clearer than:

dow not %eof();
process-record();
read record;
enddo;

One nice advantage of subprocedures, to me, is that local variables are
set to their initial values on each invocation, replacing all those
clears and evals we used to need in subroutines.

Trevor Briggs
Analyst/Programmer
Lincare, Inc.
(727) 431-1246
TBriggs2@xxxxxxxxxxx

-----Original Message-----
From: RPG400-L [mailto:rpg400-l-bounces@xxxxxxxxxxxx] On Behalf Of
Nathan Andelin
Sent: Tuesday, July 08, 2014 2:31 PM
To: RPG programming on the IBM i (AS/400 and iSeries)
Subject: Re: Are subroutines codependent?

I think there's some risk of becoming pedantic about procedures vs.
subroutines. You can become overly concerned, even ostentatiously
concerned
about replacing subroutines with procedures. I see evidence of that
attitude. I have written thousands of procedures and hundreds of service
programs. But I still use subroutines to enhance readability of code.
One
common example is to use a subroutine to separate record processing from
a
record read loop.

read record;

dow not %eof();
exsr process-record;
read record;
enddo;

I'm also not comfortable with the term "global" vs. "local" to define
variable scope. You might distinguish rather between variables scoped to
a
procedure, a module, or even data exports and imports which are shared
between programs and service programs within an activation group. Be
more
precise about describing the scope of variables.

Use subroutines to block code for readability. Use procedures and
service
programs when you need a routine to return a value, and when the routine
will be called from multiple programs or service programs.