Re: RPG-III to -IV conversion rehash worries - Part II

On 12/19/05, Buck <kc2hiz@xxxxxxxxx> wrote:
>
> In EBCDIC, all the letters have a valid numeric character portion.  MOVE
> strips off the character portion of the character, and uses the numeric
> portion to convert to numeric.  That's what MOVE is supposed to do.  The
> last character of the 'from ' field is checked for a 'D' character
> portion.  If it is a 'D', the resulting number is treated as a negative,
> otherwise it is positive.


<snip>

Interesting discussion, but I think I've gotten a little OT here.  Bottom
line is that I want the converted RPG-IV application to work exactly the
same as the RPG-III app it was converted from via CVTRPGSRC.  If the RPG-IV
app "allows" the same "aberations" (?) as the RPG-III app, then that's OK
for the initial conversion.  Actually, there are now two RPG-III programs
that I am converting, and they are 9,100 and 14,000 lines long.  As far as
the chance for problems to pop up (Murphy's Law notwithstanding), it's
comforting to know these apps have been in production for almost 20 years
now.

Anywhere else where the standards are either weak or non-enforced, I'd
probably be a little more cautious.  But I've been impressed by the
consistency of following standards here.  It helps that we've got a
top-notch QA person here.  Also, when a default value for a parameter in the
compile command is changed for a program, it is consistently documented.  I
have not seen any instance of IGNDECERR(*YES) documented anywhere.

As a sanity check, I might just whip up a quickie proggy and find all of the
MOVEx statements in the source and check to see if any are moving alpha to
numeric.  Maybe another to check data structure overlays.

Thanks for all the suggestions!
- Dan