@ud and @ub updated

[Bruce Easton]

Kenneth Brody Wednesday, April 11, 2007 5:16 PM:
>
> Quoting Bruce Easton (Wed, 11 Apr 2007 16:58:32 -0400):
>
> > Dennis Malen wrote Wednesday, April 11, 2007 4:42 PM:
> [...]
> > > Can we conclude for everyone's benefit that in auto processing it
> > > is not a good idea to write to a real field but it is OK to write
> > > to a dummy field?????
> >
> > Yes, you can conclude that.  You are allowed to do it of course
> > in auto prc, but you also should understand what results to
> > expect when you do.
>
> Note that the only reason this is allowed by filePro is that the
> original versions of filePro didn't disallow it, and too many
> people wrote (broken) code that depends on this (broken) ability.
> Disallowing it would "break" the (already broken, but "working")
> code that depends on this.
>
> [...]
>

Ah yes, and who am I to argue, being the champion of backward
compatibility that I am :).  And I suppose this explains the difference
in the later-applied @entsel where real-fld writing is proteced.
(Other than that difference and leaving out the auto table's way of
communicating values globally to other tables, I always just considered
@entsel as a duplicate of the first of the three runnings of automatic
processing that occurs with a typical filepro update transaction.  And
that it is in a more convenient place for comparing dummy fields and
just code used elsewhere in input.  At least in terms of populating
dummy fields on the screen.)

Bruce

Bruce Easton
STN, Inc.