RE: Array over fields in a data structure

["Lim Hock-Chai" <Lim.Hock-Chai@xxxxxxxx>]

Just one more little problem.  This data structure is part of the *entry parm. 

The only solution I come up with so far is:
        D tFecDs      DS
        D   tFEC1                    LIKE(MYFEC1)        
        D   tFEC2              LIKE(MYFEC1)
        D   tFEC3              LIKE(MYFEC1)
        D   tFEC4              LIKE(MYFEC1)
         etc
        D   tFEC_ARR           LIKE(MYFEC1) overlay(fecDS)
        D                      DIM(whatever)

         * loop the file and load the array.
        C         . . . . 
       * create a sub-procedure that basically eval MYFEC1 = tFEC1, .....


Another method is to use QUSLFLD to get the field location and do substring by 
field name.  (A little bit of overkill). 



-----Original Message-----
From: rpg400-l-bounces@xxxxxxxxxxxx
[mailto:rpg400-l-bounces@xxxxxxxxxxxx]On Behalf Of Barbara Morris
Sent: Thursday, November 18, 2004 11:03 AM
To: rpg400-l@xxxxxxxxxxxx
Subject: Re: Array over fields in a data structure


Lim Hock-Chai wrote:
> 
> What if there is no pattern in between each field.  ( The actual data 
> structure has no pattern).
> If I change the structure to be as follow:
>       A          R TESTDSFMT
>       A            MYFEC1         3
>       A            MYOTHER1       9   0
>       A            MYFEC1OK       1
>       A            MYFEC2         3
>       A            MYOTHER2       5
>       A            MYOTHER3       7
>       A            MYFEC2OK       1
>       A            MYFEC3         3
>       A            MYFEC3OK       1
>       A            MYFEC4         3
...

To get the MYFEC fields to be contiguous in the program, then instead of
using an externally-described DS for this file, define a USROPN F spec
for the file, and then define a DS like this (with just the names, no
type information):

D fecDs       DS
D   MYFEC1
D   MYFEC2
D   MYFEC3
D   MYFEC4
 etc
D   MYFEC_ARR          LIKE(MYFEC1) overlay(fecDS)
D                      DIM(whatever)