Re: definintion of signed integer?

[Charles Wilt <charles.wilt@xxxxxxxxx>]

I'm pretty sure you're being bit by eval's handling of intermediate results...

Since PositiveValue is an unsigned integer, and 21 is a constant that
could also be considered an unsigned integer, eval is treating the
results as an unsigned integer which obviously won't hold a negative
value.

Solution:
1) redefine PositiveValue as a signed integer
2) let eval know you're dealing with signed values
    a) NegativeValue = %int(PositiveValue) - 21;
    b) NegativeValue = PositiveValue +  (-21);

note that 2a) would blow up given a PositiveValue greater than 127.

I'd go with option 1) myself....

HTH,
Charles



On Mon, Sep 27, 2010 at 4:04 PM, Koester, Michael
<mkoester@xxxxxxxxxxxxx> wrote:
> I have a variable defined "3i 0", local to a subprocedure that I expect
> could be any whole number value from -128 to +127, per my understanding
> of a signed integer.  When my subprocedure puts what I expect to be a
> minus 18 into the field, I get the dreaded "MCH1216 -  too small to hold
> result" crash & burn.
>
> I'm sure I can fix it instantly by defining packed numerics, but I need
> to understand the rules.
>
> D PositiveValue            3u  0
> D NegativeValue            3i  0
>  /free
>  PositiveValue = 3;
>  NegativeValue = PositiveValue - 21;
>   [down in flames we go]
>  /end-free
>
> What could be so tough about subtracting 3 from 21 and expecting -18?
> Please set me straight here.  It's an SQLRPGLE program on a 6.1 box.
>
> Many thanks.
> Michael Koester
>
> --
> This is the RPG programming on the IBM i / System i (RPG400-L) mailing list
> To post a message email: RPG400-L@xxxxxxxxxxxx
> To subscribe, unsubscribe, or change list options,
> visit: http://lists.midrange.com/mailman/listinfo/rpg400-l
> or email: RPG400-L-request@xxxxxxxxxxxx
> Before posting, please take a moment to review the archives
> at http://archive.midrange.com/rpg400-l.