[Harbour] Extended field types in RDDs

[Phil Barnett]

On Saturday 01 September 2007, Przemyslaw Czerpak wrote:
> If you have any other propositions then I'd like to hear about them.
>
> Please think about it. We have to chose sth and we cannot leave
> it as is because each RDD developer will have to chse sth himself
> creating incompatible solution strongly oriented to databse
> supported by RDD (f.e. current ADS RDDs). So it's real problem.

It's difficult to retain the existing header structure and realize all of 
these things. We could extend the header to have two sections for each field 
instead of one, but we'd have to teach the code how to understand it.

Are you proposing we always follow the existing header structure?

It would be trivial to modify the header structure to accomodate these 
changes, but that would lead to new and non backward compatible RDD's.. I'm 
not sure that is a problem, though.

-- 
Phil Barnett
AI4OF
SKCC #600