RE: Easy way to define a DS char subfield the same size as somenumeric field?

Scott,

Nothing in you email is new to me, but thank you for the time and effort
you put into your reply.

Apparently I did a poor job of getting across what I wanted to do.

Consider the following example:

     D myNumFld        s             10p0  

     D myDS            DS                  qualified
     D                                9a   inz('Value is:')
     D  myCharFld                    10a   

     C                   eval      myDs.myCharFld =
%edtcde(myNumFld:'X')


What I'd like to be able to do is the following:

     D myNumFld        s             10p0  

     D myDS            DS                  qualified
     D                                9a   inz('Value is:')
     D  myCharFld
like(%size(%edtcde(myNumFld:'X'))

     C                   eval      myDs.myCharFld =
%edtcde(myNumFld:'X')


I'd agree that EVAL and BIFS are the better solution.  But I'm still
wondering if there's any way to do what I'm asking.

Thanks,

Charles Wilt
--
iSeries Systems Administrator / Developer
Mitsubishi Electric Automotive America
ph: 513-573-4343
fax: 513-398-1121
  

> -----Original Message-----
> From: rpg400-l-bounces@xxxxxxxxxxxx 
> [mailto:rpg400-l-bounces@xxxxxxxxxxxx] On Behalf Of Scott Klement
> Sent: Wednesday, April 19, 2006 12:47 PM
> To: RPG programming on the AS400 / iSeries
> Subject: Re: Easy way to define a DS char subfield the same 
> size as somenumeric field?
> 
> > I've got a standalone field defined as packed(3,0).
> > Is there an easy way to define a DS subfiled as char(3)?
> 
> That would be CHAR(2), actually.  Packed fields are always digits+1/2 
> bytes long.  In this case, 2 bytes.
> 
> 
> > In other words, given any specific numeric field, how can a define a
> > corresponding character field large enough to hold the numeric data
> > without having to hardcode the subfield size?
> 
>       D char            ds
>       D   packed                       3P 0
> 
> or
> 
>       D                 ds
>       D   char
>       D   packed                       3P 0 overlay(char)
> 
> In either case, the character field is automatically created 
> as 2 bytes 
> long, and you don't have to hard code the size.
> 
> > The DS is used to send out an email message.
> 
> It would NOT make sense to do this in an E-mail message.  
> E-mail messages 
> are text-only, so you'd want to convert the number to 
> human-readable form.
> 
> For example, if you had the packed value -123, this would be 
> stored in 
> memory in a packed field as x'123D'.  That means that your 2 
> character 
> field would be made up of 2 unprintable control characters.  
> You wouldn't 
> want to put that in an e-mail message!
> 
> So I'm thinking to myself, "maybe he meant zoned decimal instead of 
> packed?".  OKay, in that case you'd define it this way:
> 
>       D                 ds
>       D   char
>       D   zoned                        3S 0 overlay(char)
> 
> That'd be closer.  For example, if the zoned field had a 
> value of 19, it'd 
> be stored in memory as x'F0F1F9'.  If you tried to view it as 
> text, it'd 
> be 019 which at least would be printable!  However, it does have that 
> annoying leading zero.   Now consider what happens if you make it a 
> negative number, such as -123.  the bytes are set to x'F1F2D3', which 
> would appear on the screen as 12L, since x'D3' is the EBCDIC letter L.
> 
> Truly, you do NOT want to use a data structure for this.  A 
> data structure 
> just overlays two fields so that they have the same hex byte 
> values.  In 
> reality, you want to convert from the numeric field to a 
> human-readable, 
> nicely formatted, character field.  Not just copy the bytes 
> from one to 
> the other!
> 
> > If there isn't an easy way to do what I want, I'll just get 
> rid of the 
> > DS altogether and use eval + some bifs to build the message.
> 
> That's a MUCH MUCH MUCH better idea.  EVAL and BIFs will make 
> this much 
> easier to write, and using %char() or %editc() you can format 
> the number 
> any way you want.