Re: Odd CALLPRC behavior when using RPG indicator with RTNVAL

[Jon Paris <Jon.Paris@xxxxxxxxxxxxxx>]

On 13-Mar-09, at 11:19 AM, midrange-l-request@xxxxxxxxxxxx wrote:

> Interesting point Scott, but I assume IBM had reasons for it.  Does
> anybody (Jon, Barbara) know why?

Rochester decided that CLLE would handle indicators (i.e. a single  
char) in the same way that C did.  It made sense because after all  
everybody knows that there are so many more C programmers than RPG and  
that they don't understand anything about handling parameters or other  
complicated stuff - so it was important that their job should be as  
easy as possible even if it made it harder for those few RPG and COBOL  
folks out there.

Yeah - right.

The reasoning was beyond my comprehension.  If I recall correctly, the  
fact that it was being done this way was not that obvious from the  
documentation that we reviewed and so it didn't raise a flag until it  
was too late to turn the ship.

Jon Paris

www.Partner400.com
www.SystemiDeveloper.com