class PropagatingFormula:
SATISFIED: State = Formula.SATISFIED
UNRESOLVED: State = Formula.UNRESOLVED
UNSATISFIED: State = Formula.UNSATISFIED
def __init__(self: PropagatingFormula, file_object: TextIO) -> None:
self.formula = Formula(file_object)
self.decision_level: DecisionLevel = 0
self.propagate()
self.decision_history: List[Optional[Tuple[Variable, Value]]] = [None]
def _sat_variable_value_from_unit_clause(
self: PropagatingFormula,
clause: Clause) -> Tuple[Variable, Value]:
head_lit, _ = clause.get_head_tail_lit()
return abs(head_lit), 0 if head_lit < 0 else 1
def propagate(self: PropagatingFormula) -> None:
while self.formula.get_unit_clauses() and self.formula.UNRESOLVED:
clause = next(iter(self.formula.get_unit_clauses()))
variable, value = self._sat_variable_value_from_unit_clause(clause)
self.formula.assign(self.decision_level, variable, value, clause)
def add_clause(self: PropagatingFormula, clause: List[int]) -> None:
self.formula.add_clause(clause)
self.propagate()
def assign(self: PropagatingFormula, variable: Variable,
value: Value) -> None:
self.decision_level += 1
self.formula.assign(self.decision_level, variable, value, None)
self.decision_history.append((variable, value))
self.propagate()
def get_current_state(self: PropagatingFormula) -> State:
return self.formula.get_current_state()
def get_current_decision_level(self: PropagatingFormula) -> DecisionLevel:
return self.decision_level
def backtrack(self: PropagatingFormula, d: DecisionLevel) -> None:
self.decision_level = d
self.formula.backtrack(d)
while len(self.decision_history) > d:
self.decision_history.pop()
def get_partial_assignment(self: PropagatingFormula) -> Assignment:
return self.formula.get_partial_assignment()
def get_unit_clauses(self: PropagatingFormula) -> Set[Clause]:
return self.formula.get_unit_clauses()
def get_unsat_clauses(self: PropagatingFormula) -> Set[Clause]:
return self.formula.get_unsat_clauses()
def get_decision_level(self: PropagatingFormula) -> DecisionLevel:
return self.formula.get_decision_level()
