Struct BestScoreFeasibleTermination 

pub struct BestScoreFeasibleTermination<S, F>
where
    S: PlanningSolution,
    F: Fn(&S::Score) -> bool + Send + Sync,
{ /* private fields */ }

Terminates when the best score becomes feasible.

A score is considered feasible when it meets a feasibility check defined by a user-provided function. For HardSoftScore, this typically means hard score >= 0 (no hard constraint violations).

Zero-Erasure Design

The feasibility check function F is stored as a concrete generic type parameter, eliminating virtual dispatch overhead when checking termination.

Implementations

impl<S, F> BestScoreFeasibleTermination<S, F>

where S: PlanningSolution, F: Fn(&S::Score) -> bool + Send + Sync,

pub fn new(feasibility_check: F) -> Self

Creates a termination with a custom feasibility check.

impl<S: PlanningSolution> BestScoreFeasibleTermination<S, fn(&S::Score) -> bool>

pub fn score_at_least_zero() -> Self

Creates a termination that checks if score >= zero.

This is the typical feasibility check for most score types.

Trait Implementations

impl<S, F> Debug for BestScoreFeasibleTermination<S, F>

where S: PlanningSolution, F: Fn(&S::Score) -> bool + Send + Sync,

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

Formats the value using the given formatter.

impl<S, F> Termination<S> for BestScoreFeasibleTermination<S, F>

where S: PlanningSolution, F: Fn(&S::Score) -> bool + Send + Sync,

fn is_terminated(&self, solver_scope: &SolverScope<S>) -> bool

Returns true if solving should terminate.