>The procecures are created from tabular data. -- case statements based on logic that must be evaluated sequentially -- Since the rules are based on many different variables, the determination of a result is not as simple as a single or even multi key lookup. I created an fxp file that produced a huge preformance improvement. Now it isseven times faster than a looping structure that sequentially evaluates sets of rules.

I went exactly the same route as you did for a rule engine trying to establish the probability of doubles in adress management. The rule are different / differently weighted depending on the nature of the adresses and the intended use of the database and a singular prg is generated at program start depending on the settings in a config file.

As long as you stay in the prg format you have IMHO more benefits:
- the worker function/method can be a function, which is called faster.
- you can define/write out help functions
- I personally find #define handling easier to follow if written explicitely (which I use here as well)
- can be easily checked with any ASCII-editor while the program is running
- you still can create object based routines.

HTH

thomas