> Did every one of your abstractions need to be represented by
> procedures?
No, very few. That's why I could get away with not using them. But, if
in the future the implementation changes, I will be back to the old
situation of scanning through the code looking for all references to
the data.
>
> The point of my message <[log in to unmask]>
> was that you don't need to write the accessors until you decide that
> the abstractions can't be represented as components, i.e., not until
> you need to represent them with the equivalent of C# get/put functions/
> updaters. Until then, you don't need the one-liners that bulk up the
> code, increasing development and maintenance costs, and slowing things
> down. To achieve this, uniform syntax of reference is crucial.
> Updaters
> on their own aren't useful unless they have the same syntax of
> reference
> as would the abstractions they implement if they were represented as
> data
> objects.
I fully agree. I like the idea of these updaters. Unfortunately they
are not available in fortran, and I have to make do with hand written
accessors for all data, if I want to garner the benefits of indirection
between data and data access.
Maybe updaters are something for F2013 ;-)
Drew
========================================
Dr. Drew McCormack (Kmr. R153)
Afd. Theoretische Chemie
Faculteit Exacte Wetenschappen
Vrije Universiteit Amsterdam
De Boelelaan 1083
1081 HV Amsterdam
The Netherlands
Email [log in to unmask]
Telephone +31 20 44 47623
Mobile +31 6 483 21307
Fax +31 20 44 47629
|