[Freeciv-Dev] Re: Ruleset Development - improvements

[Ben Webb <ben@xxxxxxxxxxxxxxxxxxxxxx>]

On Tue, 4 Dec 2001, Mike Kaufman wrote:

> Obviously, someone is going to have to come up with a list of possible
> effects (impr. affecting certains geographical areas,  affecting units,
> affected by governments, whatever) and then implement just those. There's
> no way to get a perfectly generic solution for improvements. I can dream
> up arbitrarily complex examples of improvement effects, so it's got to
> stop somewhere.
> 
> I would say to start, we ought to have fields available to allow the
> current paycivs (and AC) work as required. At that point we might think
> about a couple more effects that allow some flexibility or
> extensibility but then _stop_ there. In our ruleset documentation, we
> say "this is the level of generality that we support". Hmm, but maybe
> this is already done?

        Yup, this is already done. Check the header of buildings.ruleset. 
The problem is supporting this ruleset information in code; my present 
implementation supports the ruleset completely (although not all of the 
effects are actually applied, or understood by the AI) but at a 
computational cost, and some odd things happen if you use certain ruleset 
features. It may become necessary to restrict the ruleset in the future to 
remove these "odd things" or we may want to consider them as "features".

        Ben
-- 
ben@xxxxxxxxxxxxxxxxxxxxxx           http://bellatrix.pcl.ox.ac.uk/~ben/
"The dog did nothing in the night-time." "That was the curious 
incident," remarked Sherlock Holmes.
        - 'Silver Blaze', Sir Arthur Conan Doyle