Struct ShardHandle 

pub struct ShardHandle { /* private fields */ }

A cloneable handle for sending commands to a shard task.

Wraps the mpsc sender so callers don’t need to manage oneshot channels directly.

Implementations

impl ShardHandle

pub async fn send( &self, request: ShardRequest, ) -> Result<ShardResponse, ShardError>

Sends a request and waits for the response.

Returns ShardError::Unavailable if the shard task has stopped.

pub async fn dispatch( &self, request: ShardRequest, ) -> Result<Receiver<ShardResponse>, ShardError>

Sends a request and returns the reply channel without waiting for the response. Used by Engine::broadcast to fan out to all shards before collecting results, and by Engine::dispatch_to_shard for the dispatch-collect pipeline.

pub async fn dispatch_reusable( &self, request: ShardRequest, reply: Sender<ShardResponse>, ) -> Result<(), ShardError>

Sends a request using a caller-owned reply channel.

Unlike dispatch(), this doesn’t allocate a oneshot per call. The caller reuses the same mpsc::Sender across commands, saving a heap allocation per P=1 round-trip.

pub async fn dispatch_batch( &self, requests: Vec<ShardRequest>, ) -> Result<Vec<Receiver<ShardResponse>>, ShardError>

Sends a batch of requests as a single channel message.

Returns one receiver per request, in the same order. For a single request, falls through to dispatch() to avoid the batch overhead.

This is the key optimization for pipelining: N commands targeting the same shard consume 1 channel slot instead of N.

Trait Implementations

impl Clone for ShardHandle

fn clone(&self) -> ShardHandle

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ShardHandle

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

Formats the value using the given formatter.