Re: reference to variable fields in IBM copybooks

[Tom Liotta <qsrvbas@xxxxxxxxxxxx>]

Jon Paris wrote:

> On 9-Jun-09, at 5:43 PM, Scott Lindstrom wrote:
>
> You guys frighten the heck out of me!  If IBM changed the copy member 
> don't you think there might have been a reason? Like the data is now 
> returned differently or ...
>
> Exactly what copy member is it you are using Scott?

Jon:

I don't have access to older /COPY members; but, IIRC, the 
significant change was simply that IBM commented out the more or 
less useless field definitions following other fields that held the 
number of elements. Particularly with a varying number of array 
elements, it was meaningless to provide a live definition for a 
single PIC X(10) field when it was really an array element. (In 
fact, maybe they only supplied a PIC X(0001)?)

Now, if there was a way they could have generated something like 
[OCCURS i TO j DEPENDING ON n], where i might be zero, then the 
definition might have been useful.

But that's pure speculation over hazy memory.

Tom Liotta