Struct ReasoningLedger 

pub struct ReasoningLedger { /* private fields */ }

Validated append-only reasoning participant content.

Implementations

impl ReasoningLedger

pub fn records(&self) -> &[ReasoningRecord]

Borrow validated records in canonical order.

use graphforge_knowledge::ReasoningLedger;
let mut ledger = ReasoningLedger::default();
ledger.records.clear();

pub fn into_records(self) -> Vec<ReasoningRecord>

Consume this ledger and return its validated records in canonical order.

pub fn new(records: Vec<ReasoningRecord>) -> Result<Self, KnowledgeError>

Validate, sort, and construct one complete participant.

pub fn merge(&self, staged: &Self) -> Result<Self, KnowledgeError>

Merge staged append-only records with idempotent exact replay.

Both inputs are validated ledgers and the row vector is private, so existing records cannot be mutated after validation. The only new whole-ledger condition is the combined row limit. A staged row that refers to a base-only predecessor is rejected while constructing the staged ledger because its predecessor is missing there. If the staged ledger includes that base row to satisfy its own chain, merge accepts it only as exact replay. Base rows cannot point to newly appended rows, since the base ledger was already validated without those identities. Therefore a cycle spanning base and staged rows is unreachable, while any identity reuse is exact replay or a conflict.

pub fn append(&self, record: ReasoningRecord) -> Result<Self, KnowledgeError>

Append one record to this validated ledger without revalidating stored rows.

The incoming record is validated because its fields are public. The existing ledger is immutable and already validated, so only combined identity, predecessor, assertion, cycle, and row-limit conditions are checked against its trusted records.

pub fn record_fingerprint( &self, reasoning_uuid: Uuid, ) -> Result<[u8; 32], KnowledgeError>

Canonical fingerprint over the exact immutable record.

pub fn current_for(&self, assertion_uuid: Uuid) -> Option<&ReasoningRecord>

Resolve the current leaf without mutating or hiding branch history.

pub fn batch(&self) -> Result<RecordBatch, KnowledgeError>

Build the authoritative Arrow batch.

pub fn from_batches(batches: &[RecordBatch]) -> Result<Self, KnowledgeError>

Decode authoritative batches and revalidate the complete ledger.

Trait Implementations

impl Clone for ReasoningLedger

fn clone(&self) -> Self

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ReasoningLedger

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

Formats the value using the given formatter.

impl Default for ReasoningLedger

fn default() -> Self

Returns the “default value” for a type.

impl Eq for ReasoningLedger

impl PartialEq for ReasoningLedger

fn eq(&self, other: &Self) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl StructuralPartialEq for ReasoningLedger