RE: ARGH!!! Pointers

["Joep Beckeringh" <joep@xxxxxxxxxxxxxxxxxxx>]

Mark,

Are these MODS (what is the plural of MODS?) by any chance defined in
the subprocedures?  If so, the memory occupied by these MODS is freed
when the subprocedures end.  This would explain why the second procedure
gives you the same pointer: it just reused the same memory, which was
just released by teh first procedure.  If this is the case, you could
define the MODS as STATIC.  This prevents the releasing of the memory,
so you can use it through the pointer.

Joep Beckeringh


> -----Oorspronkelijk bericht-----
> Van: rpg400-l-admin@midrange.com
> [mailto:rpg400-l-admin@midrange.com] Namens MWalter@hanoverwire.com
> Verzonden: donderdag 30 mei 2002 22:25
> Aan: rpg400-l@midrange.com
> Onderwerp: ARGH!!! Pointers
>
>
> Ok,
>
> I thought I had a handle on pointers. I have a service program. In the
> service program I have a module that contains two subprocedures. Each
> returns a pointer to a MODS. The program calls the first
> subproc and gets
> the pointer. Now I have a valid mods with which to work.
> Next, I write a
> subfile record based on the data from the first occurrence of
> that MODS.
> Now I call the second subproc and the pointer address value
> returned is the
> same as the previous one overwriting my data. At least this
> is what the
> interactive debugger tells me. (which negates my previous post about
> debugging based structures). From this message, can anyone
> tell me what's
> going on. If need be, I'll post the code.
>
>
> Thanks,
>
> Mark
>
>
> Mark Walter
> Sr. Programmer/Analyst
> Hanover Wire Cloth a div of CCX, Inc.
> mwalter@hanoverwire.com
> http://www.hanoverwire.com
> 717.637.3795 Ext.3040