RE: blank or 0 in database field

There are shops with "legacy" data that might have something like this...

Say the file name in question is XFiles, and the field name is XFieldNum.

****  Method One: 

Insert a MONITOR before using the field, and an ENDMON just after, but before 
the ENDMON use ON-ERROR to check for an error on the field.


****  Method Two: 

a.Determine the start-position of the culprit field in the record format.
b.Define an externally described data structure in the D-specs.
c.Define another character subfield of length 9 for the externally described 
data structure, and use OVERLAY(FieldPosn) to overlay it to the culprit field 
to test.
d.In the RPG code, test against the overlaid character subfield. 

D XFilesDS    E  DS   Extname( XFiles )
D   xFieldCh           Overlay( XFieldNum )


****  Method Three: 

a.Define a pointer and initialize it to the field address, and then a define a 
character field of 9 long based on that pointer:

D XFieldNumPointer   S     *   inz(%addr( XFieldNum ) 
D XFieldChar         S     *   based( XFieldNumPointer )

===> In your code you can use xFieldCH in the first case, or XFieldChar in the 
second, to test for blanks before using the numeric field. 

--Alan



-----Original Message-----
From: rpg400-l-bounces@xxxxxxxxxxxx
[mailto:rpg400-l-bounces@xxxxxxxxxxxx]On Behalf Of Jay Vaughn
Sent: Wednesday, October 18, 2006 4:44 PM
To: rpg400-l@xxxxxxxxxxxx
Subject: blank or 0 in database field


We have a PF with a field defined 9s0...

A non-as/400 application moves data to that field and according to the 
programmer, it is either 0 or blank... 

Is there a way to determine in my rpg pgm whether he moved a 0 or blank to 
that field?