Re: Multiple occurrence data structure anomaly

Hi Trevor -

On Fri, 5 Jul 2013 13:34:46 -0400, "Briggs, Trevor (TBriggs2)"
<TBriggs2@xxxxxxxxxxx> wrote:

Sorry, you are correct. Too many words starting with "pro.." :-)

However, I have just re-checked the old program under debug, and at the
point it calls the second program, the occurrence is set to 50. And,
after the called program returns it is still set to 50. The next thing
the program does is set the occurrence to 1 and look for a value - and
there it is!

Barbara Morris's said "When you pass a MODS as a prototyped parameter,
it passes the current occurrence. (As you've seen.)"

By implication, I take it that an unprototyped program call always
passes the address of the first occurrence.

I have personally experienced a prototyped call (in my case a true
procedure call not a program call) passing the current occurrence and
added code to set the occurrence to 1 before making the call.

I think I have unprototyped program calls that pass MODS, but never
really paid any attention because I wanted the address of the first
occurrence and that's what I got.

I very seldom use prototyped program calls (mostly just for system
APIs) so I never ran into MODS issue with that.

Ken
Opinions expressed are my own and do not necessarily represent the views
of my employer or anyone in their right mind.