Re: [Bulk] Re: New Template Type

I just want something simple, and to be honest, even subprocedures do not guarantee parameter matches, pointers or no. You don't even have to have a prototype except when calling a procedure in a RPL or COBOL program. If you use different prototypes for each definition or call of a given procedure, you are on your own. But using a prototype, or "Procedure Template" can ensure that you are passing a procedure pointer, or a character string, or whatever data type it is that you say you want to pass. Using a signature mechanism will gain no additional security with significant additional risk. Consider the signatures on service programs. They are largely ignored now by using constant signatures. And ILE applications continue running with no problem. But start using generated signatures, and your binding language becomes exponentially more complex. And the same restrictions remain: you can't change the parameters of an existing procedure except to add *NoPass parameters to the end of the list, you can't rearrange the procedures in the service program, you can't remove a procedure from the service program, unless you want to recompile everything that uses the service program.

Mark Murphy
STAR BASE Consulting, Inc.
mmurphy@xxxxxxxxxxxxxxx


-----Mark S Waterbury <mark.s.waterbury@xxxxxxxxxxxxx> wrote: -----
To: "RPG programming on the IBM i (AS/400 and iSeries)" <rpg400-l@xxxxxxxxxxxx>
From: Mark S Waterbury <mark.s.waterbury@xxxxxxxxxxxxx>
Date: 03/25/2015 11:26AM
Subject: Re: [Bulk] Re: New Template Type

John, Charles and Mark:

The ideas put forth are all "good" ... but we can never achieve any real
"guarantees" when you pass a procedure pointer, at run-time, that it
will actually point to a procedure that actually "matches" the
definition used at "compile-time" ... at least, within the confines of
the current ILE architecture, such as it is.

...