Struct ProvenanceAnnotation 

pub struct ProvenanceAnnotation {
    pub rule_name: String,
    pub clause_index: usize,
    pub support: Vec<ProofTerm>,
    pub along_values: HashMap<String, Value>,
    pub iteration: usize,
    pub fact_row: FactRow,
    pub proof_probability: Option<f64>,
    pub neural_calls: Vec<NeuralProvenance>,
}

Provenance annotation for a derived fact (Green et al. 2007, Definition 3.1).

Captures a single derivation witness: the rule and clause that produced the fact, plus the support set of proof terms that contributed to it.

Fields

rule_name: String

Name of the rule that derived this fact.

clause_index: usize

Index of the clause within the rule that produced this fact.

support: Vec<ProofTerm>

Proof terms: IS-ref input facts (populated when IS-ref tracking is available).

along_values: HashMap<String, Value>

ALONG column values captured at derivation time.

iteration: usize

Fixpoint iteration number when the fact was first derived.

fact_row: FactRow

Full fact row stored for Mode A filtering/display.

proof_probability: Option<f64>

Probability of this specific proof path (populated by top-k filtering).

neural_calls: Vec<NeuralProvenance>

Phase C B1–B3: per neural-model invocation that contributed to this fact’s derivation. Populated by LocyModelInvokeExec when it runs a classifier; consumed by Mode A EXPLAIN to surface model_name + raw + calibrated

• optional confidence band per call.

Trait Implementations

impl Clone for ProvenanceAnnotation

fn clone(&self) -> ProvenanceAnnotation

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ProvenanceAnnotation

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

Formats the value using the given formatter.