Re: Procedure Binding in CL

[Scott Klement <midrange-l@xxxxxxxxxxxxxxxx>]

Hi Dave,

Thanks for the explanation, I now see that the CALL command lacks *OMIT, 
that makes sense that you'd want to add that to the CALL command.

I don't think adding EXTPGM makes much sense.  Yes, it eliminates static 
binding, but it does so by making it a program call (rather than a 
procedure call), so it makes much more sense to use CALL ("call a 
program") in that case rather than CALLPRC ("call a procedure").

Since adding *OMIT to CALL solves the issue, then I'd say it's a much 
cleaner solution.

FWIW, it's possible to fake it out so that CALL does the equivalent of 
*OMIT by passing a parameter that's based on pointer, and making sure 
the pointer is set to *NULL.  That is all that *OMIT does, under the covers.

-SK

On 3/20/18 9:39 AM, dlclark@xxxxxxxxxxxxxxxx wrote:
> The benefit is that CALLPRC supports *OMIT while CALL does not.
>          Currently, yes.  But that is what the addition of the EXTPGM
> parameter is all about.  It would remove the compile-time binding
> requirement just like the RPG/ILE compiler does when the EXTPGM keyword is
> specified in a prototype.