Loads a MiniZinc from ModelFile. The problem is set up using a mapping to a concrete ECLiPSe solver, as specified in the SolverOrOptions argument. Neither search nor output are done.
To pass parameters into the model, a ParMap can be given, consisting of a list of FznId=ECLiPSeGroundTerm correspondences. Here, FznId is an atom (the FlatZinc parameter identifier within the model), and ECLiPSeGroundTerm is the corresponding ECLiPSe constant. Alternatively, an instance file can be specified.
To access the ECLiPSe variables corresponding to the model's variables, VarMap can be given, consisting of a list of FznId=ECLiPSeTerm correspondences. Here, FznId is an atom (the FlatZinc variable identifier within the model), and ECLiPSeTerm is the corresponding ECLiPSe constant, variable or array.
The mzn_load/5 predicate returns a FlatZinc solver state which can be used to lookup further information about the model (fzn_var_lookup/3, fzn_obj_lookup/2), to perform the standard search (fzn_search/1), or to perform the model's output actions (fzn_output/1).
?- mzn_load("queens", fzn_ic, [n=8], [q=Q], FznState).
Q = [](_2492{1..8}, _2512{1..8}, _2532{1..8}, _2552{1..8}, ...]
FznState = state(...)
There are 84 delayed goals.
Yes (0.02s cpu)
?- mzn_load("queens", fzn_ic, [n=8], [q=Q], FznState),
ic:labeling(Q).
Q = [](1, 5, 8, 6, 3, 7, 2, 4)
FznState = state(...)
Yes (0.03s cpu, solution 1, maybe more)
?- mzn_load("queens", fzn_ic, [n=8], [q=Q], FznState),
ic:labeling(Q),
fzn_output(FznState).
% output from fzn_output:
q = [1,5,8,6,3,7,2,4];
% Total time 0.030s cpu (0.020 setup)
% output from ECLiPSe toplevel:
Q = [](1, 5, 8, 6, 3, 7, 2, 4)
FznState = state(...)
Yes (0.03s cpu, solution 1, maybe more)
</PRE>