Struct ddo::core::implementation::heuristics::NaturalOrder

pub struct NaturalOrder;

This is the default variable branching heuristic This strategy branches on the variables in their ''natural'' order. That is, it will first pick Variable(0) then Variable(1), Variable(2), etc... until there are no variables left to branch on.

Example

let mut variables = VarSet::all(3);
assert_eq!(Some(Variable(0)), NaturalOrder.next_var(&variables, current_layer, next_layer));

variables.remove(Variable(0)); // move on to the next layer
assert_eq!(Some(Variable(1)), NaturalOrder.next_var(&variables, current_layer, next_layer));

variables.remove(Variable(1)); // move on to the next layer
assert_eq!(Some(Variable(2)), NaturalOrder.next_var(&variables, current_layer, next_layer));

variables.remove(Variable(2)); // move on to the last layer, no more var to branch on
assert_eq!(None, NaturalOrder.next_var(&variables, current_layer, next_layer));

Note:

Even though any variable heuristic may access the current and next layers of the mdd being developed, the natural ordering heuristic does not use that access.

Trait Implementations

impl Clone for NaturalOrder

fn clone(&self) -> NaturalOrder

Returns a copy of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for NaturalOrder

fn fmt(&self, f: &mut Formatter) -> Result

Formats the value using the given formatter.

impl Default for NaturalOrder

fn default() -> NaturalOrder

Returns the "default value" for a type.

impl<T> VariableHeuristic<T> for NaturalOrder

fn next_var<'a>(
    &self,
    free_vars: &'a VarSet,
    _c: Layer<'a, T>,
    _n: Layer<'a, T>
) -> Option<Variable>

Returns the best variable to branch on from the set of free_vars or None in case no branching is useful (free_vars is empty, no decision left to make, etc...).