mirror of
https://github.com/pezkuwichain/pezkuwi-subxt.git
synced 2026-05-30 11:41:02 +00:00
f8af277006
* compute ingress and routing in polkadot runtime * extract parent candidates from block when beginning consensus * fetch incoming messages when validating * fix consensus tests * parachain wasm execution uses messages * update parachain tests to check if messages are executed * abstract out network service to make room for network tests * skeleton for incoming data fetch * collate ingress from consensus-gossip * keep track of validated candidates in the shared-table * add some shared_table tests for new behavior * broadcast egress messages on gossip * test compute_ingress * move network tests to module folder * dummy network for consensus-network tests * make consensus network generic over executor * test egress broadcast and ingress fetch * fix test compilation * address some grumbles * address grumbles and fix parachain shuffle * remove broadcast parameter from consensus network trait
618 lines
18 KiB
Rust
618 lines
18 KiB
Rust
// Copyright 2017 Parity Technologies (UK) Ltd.
|
|
// This file is part of Polkadot.
|
|
|
|
// Polkadot is free software: you can redistribute it and/or modify
|
|
// it under the terms of the GNU General Public License as published by
|
|
// the Free Software Foundation, either version 3 of the License, or
|
|
// (at your option) any later version.
|
|
|
|
// Polkadot is distributed in the hope that it will be useful,
|
|
// but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
// GNU General Public License for more details.
|
|
|
|
// You should have received a copy of the GNU General Public License
|
|
// along with Polkadot. If not, see <http://www.gnu.org/licenses/>.
|
|
|
|
//! Statement routing and consensus table router implementation.
|
|
//!
|
|
//! During the consensus process, validators exchange statements on validity and availability
|
|
//! of parachain candidates.
|
|
//! The `Router` in this file hooks into the underlying network to fulfill
|
|
//! the `TableRouter` trait from `polkadot-consensus`, which is expected to call into a shared statement table
|
|
//! and dispatch evaluation work as necessary when new statements come in.
|
|
|
|
use sr_primitives::traits::{ProvideRuntimeApi, BlakeTwo256, Hash as HashT};
|
|
use polkadot_consensus::{
|
|
SharedTable, TableRouter, SignedStatement, GenericStatement, ParachainWork, Incoming,
|
|
Validated, Outgoing,
|
|
};
|
|
use polkadot_primitives::{Block, Hash, SessionKey};
|
|
use polkadot_primitives::parachain::{
|
|
BlockData, Extrinsic, CandidateReceipt, ParachainHost, Id as ParaId, Message
|
|
};
|
|
|
|
use codec::{Encode, Decode};
|
|
use futures::{future, prelude::*};
|
|
use futures::sync::oneshot::{self, Receiver};
|
|
use parking_lot::Mutex;
|
|
|
|
use std::collections::{hash_map::{Entry, HashMap}, HashSet};
|
|
use std::{io, mem};
|
|
use std::sync::Arc;
|
|
|
|
use consensus::{NetworkService, Knowledge, Executor};
|
|
|
|
type IngressPair = (ParaId, Vec<Message>);
|
|
type IngressPairRef<'a> = (ParaId, &'a [Message]);
|
|
|
|
fn attestation_topic(parent_hash: Hash) -> Hash {
|
|
let mut v = parent_hash.as_ref().to_vec();
|
|
v.extend(b"attestations");
|
|
|
|
BlakeTwo256::hash(&v[..])
|
|
}
|
|
|
|
fn incoming_message_topic(parent_hash: Hash, parachain: ParaId) -> Hash {
|
|
let mut v = parent_hash.as_ref().to_vec();
|
|
parachain.using_encoded(|s| v.extend(s));
|
|
v.extend(b"incoming");
|
|
|
|
BlakeTwo256::hash(&v[..])
|
|
}
|
|
|
|
/// Receiver for block data.
|
|
pub struct BlockDataReceiver {
|
|
outer: Receiver<Receiver<BlockData>>,
|
|
inner: Option<Receiver<BlockData>>
|
|
}
|
|
|
|
impl Future for BlockDataReceiver {
|
|
type Item = BlockData;
|
|
type Error = io::Error;
|
|
|
|
fn poll(&mut self) -> Poll<BlockData, io::Error> {
|
|
let map_err = |_| io::Error::new(
|
|
io::ErrorKind::Other,
|
|
"Sending end of channel hung up",
|
|
);
|
|
|
|
if let Some(ref mut inner) = self.inner {
|
|
return inner.poll().map_err(map_err);
|
|
}
|
|
match self.outer.poll().map_err(map_err)? {
|
|
Async::Ready(mut inner) => {
|
|
let poll_result = inner.poll();
|
|
self.inner = Some(inner);
|
|
poll_result.map_err(map_err)
|
|
}
|
|
Async::NotReady => Ok(Async::NotReady),
|
|
}
|
|
}
|
|
}
|
|
/// receiver for incoming data.
|
|
#[derive(Clone)]
|
|
pub struct IncomingReceiver {
|
|
inner: future::Shared<Receiver<Incoming>>
|
|
}
|
|
|
|
impl Future for IncomingReceiver {
|
|
type Item = Incoming;
|
|
type Error = io::Error;
|
|
|
|
fn poll(&mut self) -> Poll<Incoming, io::Error> {
|
|
match self.inner.poll() {
|
|
Ok(Async::NotReady) => Ok(Async::NotReady),
|
|
Ok(Async::Ready(i)) => Ok(Async::Ready(Incoming::clone(&*i))),
|
|
Err(_) => Err(io::Error::new(
|
|
io::ErrorKind::Other,
|
|
"Sending end of channel hung up",
|
|
)),
|
|
}
|
|
}
|
|
}
|
|
|
|
/// Table routing implementation.
|
|
pub struct Router<P, E, N: NetworkService, T> {
|
|
table: Arc<SharedTable>,
|
|
network: Arc<N>,
|
|
api: Arc<P>,
|
|
exit: E,
|
|
task_executor: T,
|
|
parent_hash: Hash,
|
|
attestation_topic: Hash,
|
|
knowledge: Arc<Mutex<Knowledge>>,
|
|
fetch_incoming: Arc<Mutex<HashMap<ParaId, IncomingReceiver>>>,
|
|
deferred_statements: Arc<Mutex<DeferredStatements>>,
|
|
}
|
|
|
|
impl<P, E, N: NetworkService, T> Router<P, E, N, T> {
|
|
pub(crate) fn new(
|
|
table: Arc<SharedTable>,
|
|
network: Arc<N>,
|
|
api: Arc<P>,
|
|
task_executor: T,
|
|
parent_hash: Hash,
|
|
knowledge: Arc<Mutex<Knowledge>>,
|
|
exit: E,
|
|
) -> Self {
|
|
Router {
|
|
table,
|
|
network,
|
|
api,
|
|
task_executor,
|
|
parent_hash,
|
|
attestation_topic: attestation_topic(parent_hash),
|
|
knowledge,
|
|
fetch_incoming: Arc::new(Mutex::new(HashMap::new())),
|
|
deferred_statements: Arc::new(Mutex::new(DeferredStatements::new())),
|
|
exit,
|
|
}
|
|
}
|
|
|
|
/// Get the attestation topic for gossip.
|
|
pub(crate) fn gossip_topic(&self) -> Hash {
|
|
self.attestation_topic
|
|
}
|
|
}
|
|
|
|
impl<P, E: Clone, N: NetworkService, T: Clone> Clone for Router<P, E, N, T> {
|
|
fn clone(&self) -> Self {
|
|
Router {
|
|
table: self.table.clone(),
|
|
network: self.network.clone(),
|
|
api: self.api.clone(),
|
|
task_executor: self.task_executor.clone(),
|
|
parent_hash: self.parent_hash.clone(),
|
|
attestation_topic: self.attestation_topic.clone(),
|
|
deferred_statements: self.deferred_statements.clone(),
|
|
fetch_incoming: self.fetch_incoming.clone(),
|
|
knowledge: self.knowledge.clone(),
|
|
exit: self.exit.clone(),
|
|
}
|
|
}
|
|
}
|
|
|
|
impl<P: ProvideRuntimeApi + Send + Sync + 'static, E, N, T> Router<P, E, N, T> where
|
|
P::Api: ParachainHost<Block>,
|
|
N: NetworkService,
|
|
T: Clone + Executor + Send + 'static,
|
|
E: Future<Item=(),Error=()> + Clone + Send + 'static,
|
|
{
|
|
/// Import a statement whose signature has been checked already.
|
|
pub(crate) fn import_statement(&self, statement: SignedStatement) {
|
|
trace!(target: "p_net", "importing consensus statement {:?}", statement.statement);
|
|
|
|
// defer any statements for which we haven't imported the candidate yet
|
|
let c_hash = {
|
|
let candidate_data = match statement.statement {
|
|
GenericStatement::Candidate(ref c) => Some(c.hash()),
|
|
GenericStatement::Valid(ref hash)
|
|
| GenericStatement::Invalid(ref hash)
|
|
=> self.table.with_candidate(hash, |c| c.map(|_| *hash)),
|
|
};
|
|
match candidate_data {
|
|
Some(x) => x,
|
|
None => {
|
|
self.deferred_statements.lock().push(statement);
|
|
return;
|
|
}
|
|
}
|
|
};
|
|
|
|
// import all statements pending on this candidate
|
|
let (mut statements, _traces) = if let GenericStatement::Candidate(_) = statement.statement {
|
|
self.deferred_statements.lock().get_deferred(&c_hash)
|
|
} else {
|
|
(Vec::new(), Vec::new())
|
|
};
|
|
|
|
// prepend the candidate statement.
|
|
debug!(target: "consensus", "Importing statements about candidate {:?}", c_hash);
|
|
statements.insert(0, statement);
|
|
let producers: Vec<_> = self.table.import_remote_statements(
|
|
self,
|
|
statements.iter().cloned(),
|
|
);
|
|
// dispatch future work as necessary.
|
|
for (producer, statement) in producers.into_iter().zip(statements) {
|
|
self.knowledge.lock().note_statement(statement.sender, &statement.statement);
|
|
|
|
if let Some(work) = producer.map(|p| self.create_work(c_hash, p)) {
|
|
trace!(target: "consensus", "driving statement work to completion");
|
|
let work = work.select2(self.exit.clone()).then(|_| Ok(()));
|
|
self.task_executor.spawn(work);
|
|
}
|
|
}
|
|
}
|
|
|
|
/// Broadcast outgoing messages to peers.
|
|
pub(crate) fn broadcast_egress(&self, outgoing: Outgoing) {
|
|
use slice_group_by::LinearGroupBy;
|
|
|
|
let mut group_messages = Vec::new();
|
|
for egress in outgoing {
|
|
let source = egress.from;
|
|
let messages = egress.messages.outgoing_messages;
|
|
|
|
let groups = LinearGroupBy::new(&messages, |a, b| a.target == b.target);
|
|
for group in groups {
|
|
let target = match group.get(0) {
|
|
Some(msg) => msg.target,
|
|
None => continue, // skip empty.
|
|
};
|
|
|
|
group_messages.clear(); // reuse allocation from previous iterations.
|
|
group_messages.extend(group.iter().map(|msg| msg.data.clone()).map(Message));
|
|
|
|
debug!(target: "consensus", "Circulating messages from {:?} to {:?} at {}",
|
|
source, target, self.parent_hash);
|
|
|
|
// this is the ingress from source to target, with given messages.
|
|
let target_incoming = incoming_message_topic(self.parent_hash, target);
|
|
let ingress_for: IngressPairRef = (source, &group_messages[..]);
|
|
|
|
self.network.gossip_message(target_incoming, ingress_for.encode());
|
|
}
|
|
}
|
|
}
|
|
|
|
fn create_work<D>(&self, candidate_hash: Hash, producer: ParachainWork<D>)
|
|
-> impl Future<Item=(),Error=()> + Send + 'static
|
|
where
|
|
D: Future<Item=(BlockData, Incoming),Error=io::Error> + Send + 'static,
|
|
{
|
|
let table = self.table.clone();
|
|
let network = self.network.clone();
|
|
let knowledge = self.knowledge.clone();
|
|
let attestation_topic = self.attestation_topic.clone();
|
|
|
|
producer.prime(self.api.clone())
|
|
.map(move |validated| {
|
|
// store the data before broadcasting statements, so other peers can fetch.
|
|
knowledge.lock().note_candidate(
|
|
candidate_hash,
|
|
Some(validated.block_data().clone()),
|
|
validated.extrinsic().cloned(),
|
|
);
|
|
|
|
// propagate the statement.
|
|
// consider something more targeted than gossip in the future.
|
|
let signed = table.import_validated(validated);
|
|
network.gossip_message(attestation_topic, signed.encode());
|
|
})
|
|
.map_err(|e| debug!(target: "p_net", "Failed to produce statements: {:?}", e))
|
|
}
|
|
|
|
}
|
|
|
|
impl<P: ProvideRuntimeApi, E, N, T> Router<P, E, N, T> where
|
|
P::Api: ParachainHost<Block>,
|
|
N: NetworkService,
|
|
T: Executor,
|
|
E: Future<Item=(),Error=()> + Clone + Send + 'static,
|
|
{
|
|
fn do_fetch_incoming(&self, parachain: ParaId) -> IncomingReceiver {
|
|
use polkadot_primitives::BlockId;
|
|
let (tx, rx) = {
|
|
let mut fetching = self.fetch_incoming.lock();
|
|
match fetching.entry(parachain) {
|
|
Entry::Occupied(entry) => return entry.get().clone(),
|
|
Entry::Vacant(entry) => {
|
|
// has not been requested yet.
|
|
let (tx, rx) = oneshot::channel();
|
|
let rx = IncomingReceiver { inner: rx.shared() };
|
|
entry.insert(rx.clone());
|
|
|
|
(tx, rx)
|
|
}
|
|
}
|
|
};
|
|
|
|
let parent_hash = self.parent_hash;
|
|
let topic = incoming_message_topic(parent_hash, parachain);
|
|
let gossip_messages = self.network.gossip_messages_for(topic)
|
|
.map_err(|()| panic!("unbounded receivers do not throw errors; qed"))
|
|
.filter_map(|msg| IngressPair::decode(&mut msg.as_slice()));
|
|
|
|
let canon_roots = self.api.runtime_api().ingress(&BlockId::hash(parent_hash), parachain)
|
|
.map_err(|e| format!("Cannot fetch ingress for parachain {:?} at {:?}: {:?}",
|
|
parachain, parent_hash, e)
|
|
);
|
|
|
|
let work = canon_roots.into_future()
|
|
.and_then(move |ingress_roots| match ingress_roots {
|
|
None => Err(format!("No parachain {:?} registered at {}", parachain, parent_hash)),
|
|
Some(roots) => Ok(roots.into_iter().collect())
|
|
})
|
|
.and_then(move |ingress_roots| ComputeIngress {
|
|
inner: gossip_messages,
|
|
ingress_roots,
|
|
incoming: Vec::new(),
|
|
})
|
|
.map(move |incoming| if let Some(i) = incoming { let _ = tx.send(i); })
|
|
.select2(self.exit.clone())
|
|
.then(|_| Ok(()));
|
|
|
|
self.task_executor.spawn(work);
|
|
|
|
rx
|
|
}
|
|
}
|
|
|
|
impl<P: ProvideRuntimeApi + Send, E, N, T> TableRouter for Router<P, E, N, T> where
|
|
P::Api: ParachainHost<Block>,
|
|
N: NetworkService,
|
|
T: Clone + Executor + Send + 'static,
|
|
E: Future<Item=(),Error=()> + Clone + Send + 'static,
|
|
{
|
|
type Error = io::Error;
|
|
type FetchCandidate = BlockDataReceiver;
|
|
type FetchIncoming = IncomingReceiver;
|
|
|
|
fn local_candidate(&self, receipt: CandidateReceipt, block_data: BlockData, extrinsic: Extrinsic) {
|
|
// produce a signed statement
|
|
let hash = receipt.hash();
|
|
let validated = Validated::collated_local(receipt, block_data.clone(), extrinsic.clone());
|
|
let statement = self.table.import_validated(validated);
|
|
|
|
// give to network to make available.
|
|
self.knowledge.lock().note_candidate(hash, Some(block_data), Some(extrinsic));
|
|
self.network.gossip_message(self.attestation_topic, statement.encode());
|
|
}
|
|
|
|
fn fetch_block_data(&self, candidate: &CandidateReceipt) -> BlockDataReceiver {
|
|
let parent_hash = self.parent_hash.clone();
|
|
let candidate = candidate.clone();
|
|
let (tx, rx) = ::futures::sync::oneshot::channel();
|
|
self.network.with_spec(move |spec, ctx| {
|
|
let inner_rx = spec.fetch_block_data(ctx, &candidate, parent_hash);
|
|
let _ = tx.send(inner_rx);
|
|
});
|
|
BlockDataReceiver { outer: rx, inner: None }
|
|
}
|
|
|
|
fn fetch_incoming(&self, parachain: ParaId) -> Self::FetchIncoming {
|
|
self.do_fetch_incoming(parachain)
|
|
}
|
|
}
|
|
|
|
impl<P, E, N: NetworkService, T> Drop for Router<P, E, N, T> {
|
|
fn drop(&mut self) {
|
|
let parent_hash = self.parent_hash.clone();
|
|
self.network.with_spec(move |spec, _| spec.remove_consensus(&parent_hash));
|
|
self.network.drop_gossip(self.attestation_topic);
|
|
|
|
{
|
|
let mut incoming_fetched = self.fetch_incoming.lock();
|
|
for (para_id, _) in incoming_fetched.drain() {
|
|
self.network.drop_gossip(incoming_message_topic(
|
|
self.parent_hash,
|
|
para_id,
|
|
));
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
// A unique trace for valid statements issued by a validator.
|
|
#[derive(Hash, PartialEq, Eq, Clone, Debug)]
|
|
enum StatementTrace {
|
|
Valid(SessionKey, Hash),
|
|
Invalid(SessionKey, Hash),
|
|
}
|
|
|
|
// helper for deferring statements whose associated candidate is unknown.
|
|
struct DeferredStatements {
|
|
deferred: HashMap<Hash, Vec<SignedStatement>>,
|
|
known_traces: HashSet<StatementTrace>,
|
|
}
|
|
|
|
impl DeferredStatements {
|
|
fn new() -> Self {
|
|
DeferredStatements {
|
|
deferred: HashMap::new(),
|
|
known_traces: HashSet::new(),
|
|
}
|
|
}
|
|
|
|
fn push(&mut self, statement: SignedStatement) {
|
|
let (hash, trace) = match statement.statement {
|
|
GenericStatement::Candidate(_) => return,
|
|
GenericStatement::Valid(hash) => (hash, StatementTrace::Valid(statement.sender, hash)),
|
|
GenericStatement::Invalid(hash) => (hash, StatementTrace::Invalid(statement.sender, hash)),
|
|
};
|
|
|
|
if self.known_traces.insert(trace) {
|
|
self.deferred.entry(hash).or_insert_with(Vec::new).push(statement);
|
|
}
|
|
}
|
|
|
|
fn get_deferred(&mut self, hash: &Hash) -> (Vec<SignedStatement>, Vec<StatementTrace>) {
|
|
match self.deferred.remove(hash) {
|
|
None => (Vec::new(), Vec::new()),
|
|
Some(deferred) => {
|
|
let mut traces = Vec::new();
|
|
for statement in deferred.iter() {
|
|
let trace = match statement.statement {
|
|
GenericStatement::Candidate(_) => continue,
|
|
GenericStatement::Valid(hash) => StatementTrace::Valid(statement.sender, hash),
|
|
GenericStatement::Invalid(hash) => StatementTrace::Invalid(statement.sender, hash),
|
|
};
|
|
|
|
self.known_traces.remove(&trace);
|
|
traces.push(trace);
|
|
}
|
|
|
|
(deferred, traces)
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
// computes ingress from incoming stream of messages.
|
|
// returns `None` if the stream concludes too early.
|
|
#[must_use = "futures do nothing unless polled"]
|
|
struct ComputeIngress<S> {
|
|
ingress_roots: HashMap<ParaId, Hash>,
|
|
incoming: Vec<IngressPair>,
|
|
inner: S,
|
|
}
|
|
|
|
impl<S> Future for ComputeIngress<S> where S: Stream<Item=IngressPair> {
|
|
type Item = Option<Incoming>;
|
|
type Error = S::Error;
|
|
|
|
fn poll(&mut self) -> Poll<Option<Incoming>, Self::Error> {
|
|
loop {
|
|
if self.ingress_roots.is_empty() {
|
|
return Ok(Async::Ready(
|
|
Some(mem::replace(&mut self.incoming, Vec::new()))
|
|
))
|
|
}
|
|
|
|
let (para_id, messages) = match try_ready!(self.inner.poll()) {
|
|
None => return Ok(Async::Ready(None)),
|
|
Some(next) => next,
|
|
};
|
|
|
|
match self.ingress_roots.entry(para_id) {
|
|
Entry::Vacant(_) => continue,
|
|
Entry::Occupied(occupied) => {
|
|
let canon_root = occupied.get().clone();
|
|
let messages = messages.iter().map(|m| &m.0[..]);
|
|
if ::polkadot_consensus::message_queue_root(messages) != canon_root {
|
|
continue;
|
|
}
|
|
|
|
occupied.remove();
|
|
}
|
|
}
|
|
|
|
let pos = self.incoming.binary_search_by_key(
|
|
¶_id,
|
|
|&(id, _)| id,
|
|
)
|
|
.err()
|
|
.expect("incoming starts empty and only inserted when \
|
|
para_id not inserted before; qed");
|
|
|
|
self.incoming.insert(pos, (para_id, messages));
|
|
}
|
|
}
|
|
}
|
|
|
|
#[cfg(test)]
|
|
mod tests {
|
|
use super::*;
|
|
use substrate_primitives::H512;
|
|
use futures::stream;
|
|
|
|
#[test]
|
|
fn deferred_statements_works() {
|
|
let mut deferred = DeferredStatements::new();
|
|
let hash = [1; 32].into();
|
|
let sig = H512::from([2; 64]).into();
|
|
let sender = [255; 32].into();
|
|
|
|
let statement = SignedStatement {
|
|
statement: GenericStatement::Valid(hash),
|
|
sender,
|
|
signature: sig,
|
|
};
|
|
|
|
// pre-push.
|
|
{
|
|
let (signed, traces) = deferred.get_deferred(&hash);
|
|
assert!(signed.is_empty());
|
|
assert!(traces.is_empty());
|
|
}
|
|
|
|
deferred.push(statement.clone());
|
|
deferred.push(statement.clone());
|
|
|
|
// draining: second push should have been ignored.
|
|
{
|
|
let (signed, traces) = deferred.get_deferred(&hash);
|
|
assert_eq!(signed.len(), 1);
|
|
|
|
assert_eq!(traces.len(), 1);
|
|
assert_eq!(signed[0].clone(), statement);
|
|
assert_eq!(traces[0].clone(), StatementTrace::Valid(sender, hash));
|
|
}
|
|
|
|
// after draining
|
|
{
|
|
let (signed, traces) = deferred.get_deferred(&hash);
|
|
assert!(signed.is_empty());
|
|
assert!(traces.is_empty());
|
|
}
|
|
}
|
|
|
|
#[test]
|
|
fn compute_ingress_works() {
|
|
let actual_messages = [
|
|
(
|
|
ParaId::from(1),
|
|
vec![Message(vec![1, 3, 5, 6]), Message(vec![4, 4, 4, 4])],
|
|
),
|
|
(
|
|
ParaId::from(2),
|
|
vec![
|
|
Message(vec![1, 3, 7, 9, 1, 2, 3, 4, 5, 6]),
|
|
Message(b"hello world".to_vec()),
|
|
],
|
|
),
|
|
(
|
|
ParaId::from(5),
|
|
vec![Message(vec![1, 2, 3, 4, 5]), Message(vec![6, 9, 6, 9])],
|
|
),
|
|
];
|
|
|
|
let roots: HashMap<_, _> = actual_messages.iter()
|
|
.map(|&(para_id, ref messages)| (
|
|
para_id,
|
|
::polkadot_consensus::message_queue_root(messages.iter().map(|m| &m.0)),
|
|
))
|
|
.collect();
|
|
|
|
let inputs = [
|
|
(
|
|
ParaId::from(1), // wrong message.
|
|
vec![Message(vec![1, 1, 2, 2]), Message(vec![3, 3, 4, 4])],
|
|
),
|
|
(
|
|
ParaId::from(1),
|
|
vec![Message(vec![1, 3, 5, 6]), Message(vec![4, 4, 4, 4])],
|
|
),
|
|
(
|
|
ParaId::from(1), // duplicate
|
|
vec![Message(vec![1, 3, 5, 6]), Message(vec![4, 4, 4, 4])],
|
|
),
|
|
|
|
(
|
|
ParaId::from(5), // out of order
|
|
vec![Message(vec![1, 2, 3, 4, 5]), Message(vec![6, 9, 6, 9])],
|
|
),
|
|
(
|
|
ParaId::from(1234), // un-routed parachain.
|
|
vec![Message(vec![9, 9, 9, 9])],
|
|
),
|
|
(
|
|
ParaId::from(2),
|
|
vec![
|
|
Message(vec![1, 3, 7, 9, 1, 2, 3, 4, 5, 6]),
|
|
Message(b"hello world".to_vec()),
|
|
],
|
|
),
|
|
];
|
|
let ingress = ComputeIngress {
|
|
ingress_roots: roots,
|
|
incoming: Vec::new(),
|
|
inner: stream::iter_ok::<_, ()>(inputs.iter().cloned()),
|
|
};
|
|
|
|
assert_eq!(ingress.wait().unwrap().unwrap(), actual_messages);
|
|
}
|
|
}
|