Re: Constant definitions (was: Visual RPG vs. Other Visual ....

Barbara,

is this an 'undocumented feature' or a bug?  intended, or screwup.

i'd like an example of where this would be beneficial, not just confusing.

If I ran across existing code like that below, without knowing what i know
now, i would assume that datastructure 'ds' was 22 bytes long, with a 10 in
positions 11,12... and maybe would have spent hours trying to figure out
why it wasn't true.

wouldn't anyone?

rick

----original message-------
Jon Paris wrote:
>
>  >>  Do you think it was retained for compatibility with RPG/400?
>
> I don't think it was retained.  In fact I know that it was deliberately
> _not_ supported.  We had to make a point of highlighting it as a
difference
> in all the early training material.
>
> As I noted - it doesn't work on V5R2 so the fact that it works on V4R5
may
> be an aberration.
>
> Either way there is a fix needed somewhere.  To the compiler, the
> documentation, or both.  I'm going to wait and see what Barbara has to
say.
>

Jon, it works on V5R2 for me.
     D ds              ds
     D  sub1                         10a   inz('sub1')
     D  sub1len        c                   10
     D  sub2                         10a   inz('sub2')
     C     sub1          dsply
     C     sub1len       dsply
     C     sub2          dsply
     C                   return

Looks like it's been supported but not documented since the beginning of
RPG IV.