Re: How to convert character field w/ negative numeric data (255N)

Scott, Bob, Matt, and others,

Thank you for the explanations!  I had thought since MOVE is not supported 
in free-format, there would have been an alternative BIF to do what I 
wanted.  I like the overlay idea, too, so I may use that. 

The field is in a vendor supplied file and I do not know why they chose to 
define a numeric field as character and populate it using that format.  It 
isn't something that I am allowed to change.

Thanks for the information!

-Mike

rpg400-l-bounces@xxxxxxxxxxxx wrote on 03/09/2006 03:50:37 PM:


> The data in your field is not human-readable character data.  You could 
> not reasonably expect a user to see "100N" and know that it means -1005.
> 
> The %DEC() and %INT() BIFs are expecting human-readable character data. 
> For example, it could be '   -1234' or '     -85.65' and they'd have no 
> trouble interpreting it.  But '00000856N'they won't understand because 
> it's not human-readable character data.
> 
> What you have in your field is the contents of a zoned decimal field. 
I'm 
> curious as to why you have this loaded into a character field in your 
> program.  Why not read it into a zoned decimal field in the first place? 

> When you read the file, just code the field as zoned decimal.   (Either 
in 
> the external definition, or in the I-spec if it's program described.)
> 
> IF for some obscure reason you can't do that, you can either use the 
MOVE 
> op-code (as you've already discovered) or you can simply overlay the 
> character field with a zoned field in a data structure.  For example:
> 
>       D                 ds
>       D  myCharFld                     9A
>       D  myZonedFld                    9S 2 overlay(myCharFld)
> 
>        /free
> 
>            myCharFld = '00000856N';
>            // myZonedFld now contains -85.65.