Re: Storing derived and derivable data

Jay Dee wrote:

>
> I took a look; I didn't see anything among the automations,
> chains, or cascading actions that looked like it would be
> a good fit. The description of automations, which seem to
> be column and foreign-key (a new feature) oriented, didn't
> seem like good fits. The phrase
> [i]t is easy enough to provide any number of ways to
> add rows to TABLE X on the event of changes to TABLE Y
> seemed a bit of a teaser. Perhaps what you're looking for
> might be, in Andromeda terms, a "row automation?"
>

You are right about that sentence, it is too philosophical in a document that should be a review and light reference. I took it out. I'm afraid the current sentence is still to philosophical, but perhaps a few more iterations will do it.

In any event, yes, the idea is to define a new method of row automation not currently provided.

The concept in my previous post, of populating a child table automatically on INSERT to parent, has appeal to me because it rounds out the operations Andromeda provides around foreign keys. If this operation were put in, you would be able to create parents, create children, copy down and aggregate up through a foreign key. It becomes the 4-lane highway of automation.

-- 
Kenneth Downs
Secure Data Software, Inc.
(Ken)nneth_at_(Sec)ure(Dat)a(.com)