RE: Indicator enhancements

[Buck Calabro <buck.calabro@xxxxxxxxxxxxxxxxx>]

Gary Guthrie wrote:

>Ah, but it does have general usefulness. It's just that 
>the silly example I used doesn't cause the usefulness 
>to jump out at you.
>
>Consider that
>
>- -1 is a dead customer
>- -2 is a customer who never pays
>- -3 is an industrial spy
>
>Consider further that the procedure is likely housed 
>in a service program. Some applications might need 
>to know specific errors (hey, is this a dead customer 
>or a deadbeat customer!), others don't really care
>about the specific error. Rather they just care that 
>there's some error.
>
>So, the usefulness of *NEGATIVE and *POSITIVE figurative 
>constants is in the ability to code a layer that is independent 
>of the details. This would be clearer ...

-snip-

With all due respect, which is clearer and more independent of the details?

if GetCustomerStatus(CustomerID) = *Negative
...
endif


if CustomerError(CustomerID)
...
endif


callp GetCustomerStatusAndErrorState(CustomerID: Status: ErrorState)
if ErrorState = WARNING
...
endif


Why overload GetCustomerStatus with the implied job of reporting an "error"
as well as a specific status?  If you break out the error reporting as a
separate parameter or function, you get to have other facets of the customer
that can cause an error, and not limit yourself to negative status codes.
Once you've made the leap from "negative means error," the next logical step
is "how severe is the error?"  It gets ugly fast if you start tying the
severity to a range of negative numbers.  I suspect that the tradition
carries on in C because... it's traditional! <grin>

Buck Calabro
+---
| This is the RPG/400 Mailing List!
| To submit a new message, send your mail to RPG400-L@midrange.com.
| To subscribe to this list send email to RPG400-L-SUB@midrange.com.
| To unsubscribe from this list send email to RPG400-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---