RE: Cleaning up garbage in packed numeric fields??

["David C. Shea" <dshea@xxxxxxxxxxxx>]

I checked the archives and found an interesting approach...

RPGLE will fix bad data based on an H spec.  But - you can only fix
either packed fields or zoned fields... can't do both.

I put this little rpg program together.  Fortunately, I haven't tripped
over a file with both packed and zoned that are bad.  Any way around
this?  It appears that I can't have two h specs for the same key word,
and I can't specify both *inputpacked and *zoned on the one fixnbr h
spec.  Or, is my syntax wrong?

h* fixnbr(*inputpacked)                   
h fixnbr(*zoned)                          
Ffilename  up   e             disk        
c                   update    fmtname 


 
********************************************************
Go... FASTER! Without an Upgrade!
ARCTOOLS/400(tm) and ReorgWizard(tm)
http://www.arctools.com
DCSoftware, Inc.
Ph: (508) 435-8243
Fax: (508) 435-4498
********************************************************

-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx
[mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of David C. Shea
Sent: Friday, August 12, 2005 8:56 AM
To: Midrange Systems Technical Discussion
Subject: Cleaning up garbage in packed numeric fields??

I have a data file with some packed numeric fields.  On some records,
instead of 12345F, there is 1b73x1 or some other gibberish that is
obviously not valid numeric data.  It's not null, it's garbage.

A few bad apples are ruining the whole bunch.

I'm trying to CPYTOIMPF from this DDS file to a delimited stream file.
Because of this bad data, I get some records approaching the bad record,
but nothing afterwards.

I've tried CPYF to see if that would initialize the bad stuff, but no
good.

Any ideas how I can clean up these records - maybe reinitialize these
fields to zeroes?