Trait Scorer

Source

pub trait Scorer {
    type World;
    type Expect;

    // Required method
    fn score(
        &self,
        expect: &Self::Expect,
        artifacts: &RunArtifacts,
        world: &Self::World,
    ) -> (String, bool);
}

Expand description

Scores one expectation against the world AND artifacts a case produced. The host implements this for a custom predicate/world; for the built-in assertions use BuiltinScorer.

score returns BOTH a human-readable label and the pass/fail in one call (the existing AetherCore scorer kept these as a paired score()/label(); folding them avoids re-deriving the label and guarantees the label always matches the verdict). The label is shown in the report’s per-predicate diagnostics, so make it identify WHICH predicate it is (e.g. "SolidPlaced(>= 4)").

Required Associated Types§

Source

type World

The world type to score against — must be the same world the paired Harness::World produces (the runner enforces Scorer<World = Harness::World>).

Source

type Expect

The host’s predicate type — matches the element type of EvalCase::expect.

Required Methods§

Source

fn score( &self, expect: &Self::Expect, artifacts: &RunArtifacts, world: &Self::World, ) -> (String, bool)

Score one expect predicate against the post-run world and the run’s artifacts.

Returns (label, passed): a human-readable label for per-predicate diagnostics, and whether the predicate held. Must be side-effect-free with respect to world (it takes a shared reference); the runner may call it for several predicates over the same world. A scorer that only inspects the world can ignore artifacts (and vice versa).