RE: Target for numeric operation too small...you're joking, right?

["Reeve Fritchman" <reeve@xxxxxxxxxx>]

No joke.

The math majors know this stuff cold, but here's the short form: when you're
working with two numbers, you can always figure out the maximum value, and
from that you can always figure out the (work/internal) field size to hold
that value.  If you're adding two numbers together, the largest field size
you need is (# of digits in the largest field + 1).  999+999=1,998, right?
You don't need a 5.0 variable unless you stick with odd length numerics.
999 * 999 = 998,001; add the # of digits in each field to get the maximum
size.

So the compiler /can/ figure out if a field's going to be too small.  So we
get a message in the compiler output, and 99.999% of us will ignore it
anyway.

Otherwise, let's consider the numeric data management options we already
have: EXRPOPTS, EXTBINIT, FIXNBR, FLTDIV, INTPREC, and TRUNCNBR.  We need
one more option, to make it an odd number of options.

-----Original Message-----
From: rpg400-l-admin@midrange.com [mailto:rpg400-l-admin@midrange.com]On
Behalf Of Leland, David
Sent: Thursday, December 06, 2001 1:16 PM
To: 'rpg400-l@midrange.com'
Subject: RE: Target for numeric operation too small...hello, Toronto?

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.
--
[ Picked text/plain from multipart/alternative ]
You're joking, right?

-----Original Message-----
From: Tyra_Brodsky@vfc.com [mailto:Tyra_Brodsky@vfc.com]
Sent: Thursday, December 06, 2001 1:14 PM
To: rpg400-l@midrange.com
Subject: RE: Target for numeric operation too small...hello, Toronto?



But wouldn't it be nice to know it at time of compile instead of time of
execution????

That way I could fix the problem BEFORE it halts on me.





                    Buck Calabro
                    <Buck.Calabro@com       To:     rpg400-l@midrange.com
                    msoft.net>              cc:
                    Sent by:                Subject:     RE: Target for
numeric operation too small...hello,
                    rpg400-l-admin@mi        Toronto?
                    drange.com


                    12/06/01 10:08 AM
                    Please respond to
                    rpg400-l






>>I frankly would have liked to have seen
>>a different option on the error
>>message; "allow truncation."
>
>How about: EVAL(T)

Hi Mark!
I liken numeric truncation to an invalid array index.  Rather than add an
operand extender to allow the invalid index (or set it to 1) I would prefer
to have a hard halt so that the operator is aware that we're about to lose
some precision.

After consulting with the programmer (me), it would be nice if the operator
could take a "T" option, allow truncation and continue.  If it's a first
test, I might like to see the full scale results because the truncation may
be only one of several issues I need to address.

But I _prefer_ the hard error.
  --buck
_______________________________________________
This is the RPG programming on the AS400 / iSeries (RPG400-L) mailing list
To post a message email: RPG400-L@midrange.com
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/cgi-bin/listinfo/rpg400-l
or email: RPG400-L-request@midrange.com
Before posting, please take a moment to review the archives
at http://archive.midrange.com/rpg400-l.





_______________________________________________
This is the RPG programming on the AS400 / iSeries (RPG400-L) mailing list
To post a message email: RPG400-L@midrange.com
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/cgi-bin/listinfo/rpg400-l
or email: RPG400-L-request@midrange.com
Before posting, please take a moment to review the archives
at http://archive.midrange.com/rpg400-l.
_______________________________________________
This is the RPG programming on the AS400 / iSeries (RPG400-L) mailing list
To post a message email: RPG400-L@midrange.com
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/cgi-bin/listinfo/rpg400-l
or email: RPG400-L-request@midrange.com
Before posting, please take a moment to review the archives
at http://archive.midrange.com/rpg400-l.