Re: AW: %Addr() on Const parm @v4r5

["M. Lazarus" <mlazarus@xxxxxxxx>]

Stu,

 I had compiled the same program to both 5.2 and 4.5.  It worked 
fine on 5.2 and failed on 4.5.  Using the CEETSTA works great on 4.5.

 -mark

At 7/13/10 06:19 PM, you wrote:
> All,
>
> For what it's worth, I always do this for my optional parms declared as *
> const* and it works like a charm (v5r4 and v5r3 prior to that):
>
> if %parms > *zero and %addr( parm ) <> *null;
> ...
> endif;
>
> You cannot assign the %addr() to a variable, nor can you base anything on
> that %addr() (unless you fool the compiler), but you should be able to use
> it as a temporary value in an expression.
>
> Stu
>
>
> On Tue, Jul 13, 2010 at 15:13, M. Lazarus <mlazarus@xxxxxxxx> wrote:
>
> > Yes, it ran fine @ 5.2.
> >
> >
> > At 7/13/10 03:49 PM, you wrote:
> > >OK, I just checked V5R2 using this URL:
> > >http://publib.boulder.ibm.com/iseries/v5r2/ic2924/books/c0925083545.htm
> > >
> > >Nothing in there about this. The closest there is (which is also in the
> > V5R3
> > >manual) is this:
> > >
> > >"If the variable is specified as a PARM of the *ENTRY PLIST, %ADDR returns
> > >the address passed to the program by the caller."
> > >
> > >It compiled, but did it run?
> > >
> > >On Tue, Jul 13, 2010 at 12:45 PM, Rory Hewitt <rory.hewitt@xxxxxxxxx>
> > wrote:
> > >
> > > > Could be :) I won't even check to the V5R2 manual online, because it
> > uses a
> > > > Java applet which takes an *age* to load.
> > > >
> > > >
> > > > On Tue, Jul 13, 2010 at 12:32 PM, M. Lazarus <mlazarus@xxxxxxxx>
> > wrote:
> > > >
> > > >> Rory,
> > > >>
> > > >>  That's interesting, because I was able to compile it (in free
> > > >> format) at v5r2. Could it have been an undocumented feature then?
> > > >>
> > > >>  -mark