Re: Move 'ABC' to Numeric? -- RPG400-L

Has anyone else tried this line of code?
 
I just tried it out.
 
  C    move    'ABC'             NUM3             3 0
 
It compiled. I ran it in debug, and low and behold, the value of NUM3 = 123.
 
Since it actually works I'd guess someone at sometime was dealing with a
problem long since resolved. Simple encryption, like ROT13? or perhaps
showing off? Or trying out some knowledge from an article or seminar? In any
event, it does appear to actually work.
 
  
---------------------------------------------------------
Booth Martin http://www.MartinVT.com
Booth@xxxxxxxxxxxx
---------------------------------------------------------
 
-------Original Message-------
 
From: RPG programming on the AS400 / iSeries
Date: Monday, July 28, 2003 9:17:51 AM
To: RPG programming on the AS400 / iSeries
Subject: Re: Move 'ABC' to Numeric?
 
Steve,
 
I don't think Tom was saying it was CASE, or any other tool dependent. I
think he was saying that no sober person would put that line in their
program on their own. Therefore if a vendor see's a program in another
vendor's package with that line then it is pretty obvious that the other
package stole the logic from you.
 
Rob Berendt
--
"They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety."
Benjamin Franklin
 
 
 
 
 
"Steve Landess" <steve_landess@xxxxxxxxxxx>
Sent by: rpg400-l-bounces@xxxxxxxxxxxx
07/26/2003 08:39 AM
Please respond to RPG programming on the AS400 / iSeries
 
To: "RPG programming on the AS400 / iSeries"
<rpg400-l@xxxxxxxxxxxx>
cc:
Fax to:
Subject: Re: Move 'ABC' to Numeric?
 
 
>>Bill wrote:
>> 2. Move 'ABC' to Numeric? (Bill Hopkins)
>>Anyone know why you would see this in code.
>> MOVE 'ABC' NUM 8P0
>>More code follows; then
>>NUM IFEQ 123
>>More code
>> ENDIF
>>
>>Found in bridge from JDE Sales Orders to PKMS
 
>
> Tom wrote:
> Does anything happen if it's not equal? Or is there an implication that
something somewhere else won't >happen if the IF-test fails because the
enclosed statements didn't run?
>
> I can _imagine_ someone from years gone by using something like this as
a
part of code that was
> generated so that under different circumstance a different source line
would have been auto-gen'd
>into the code to result in a value other than 123, perhaps because the
referenced numeric field might
> sometimes be signed and otherwise unsigned.
>
> Or perhaps the 'ABC' constant would be used as a marker in the compiled
object in order
>to recognize a reverse-engineered kind of rip-off? A few of these
sprinkled
around weren't
>unheard of and I believe still are used. It might just be a remnant that
never got cleaned up.
 
Tom -
 
IMO, This was not code generated by the JDE CASE tool. I have been
working
with JDE software since 1987, and I have never seen anything like this in
any JDE program. Some human bozo coded it. I have seen many modified JDE
programs that had stupid modifications like this.
 
Once, when I was working at a client in the Dallas area, a JDE _employee_
came in to fix a problem with the invoice print program. Later, when the
program was still working incorrectly, I looked at the mods he made.
Instead
of fixing the real problem, he had coded a GOTO around a block of code so
that the program would not bomb.
 
This speaks to the need of having a systematic way of QA'ing programs. In
college I was taught the structured methodology and using peer reviews,
the
"ego-less" approach to reviewing code. Too often the mavericks slip
things
in the code that should be caught _before_ the program gets into
production.
 
Steve.