Re: [eclipse-clp-users] A complex Problem

No, the variables of the compiled clauses are not the point ... but the 
variables of the solutions must stay the same.
Except the d_next call, this is already working as u can see.

Kish Shen schrieb:
> Christian Wirth wrote:
>
>>
>> if i only call getunboundmove(Player,Move,Term,State), the solution is:
>>
>> Player = Player
>> Move = move(north)
>> Term = [d_role(Player, _219), 'd_++'(_265, _340, _219), d_cell(_264, 
>> _340, _219)] - Read as: role Player exists  | _340 is _264 + 1 | cell 
>> x=_263,y=_340 exists (the last var is State=_219, but these facts are 
>> state independent)
>> State = [location(_73, _264, _265), \+ blockednorth(_264, _265)] - 
>> Player _73 is in position _264/_265 (Here _73 should be Player, but 
>> thats not importent) | the north of position _264/_265 is not blocked
>>
>> and for effect, i want to get:
>>
>> Effects=[location(Player,_264,_340)]
>>
>> It is essential, that the variable numbers are staying correct, 
>> because i need to map the effects to the relevant State/Term conditions
>
> I haven't been able to understand your code in detail, but am I 
> correct in thinking that you are compiling predicates such as d_role 
> and d_cell etc. in your example, which share variables (_219 in your 
> case), and you expect the compiled version of these predicates will 
> still share the variables?
>
> If so, this is a rather fundamental misunderstanding of variables in 
> Prolog. Variables exists during the runtime only, and variables in 
> your source are not the same as the runtime variables.
>
> If you really want to do meta-programming using compiled/asserted 
> predicates with some notion about shared variables across the 
> predicates, you will probably need to have some ground representation 
> of variables (such as an id number) in your predicates, which you 
> convert to runtime variables.
>
> Cheers,
>
> Kish
>