]> git.bitcoin.ninja Git - rust-lightning/commitdiff
Add subcrate that implements network socket handling with Tokio
authorMatt Corallo <git@bluematt.me>
Sat, 1 Jun 2019 11:02:31 +0000 (07:02 -0400)
committerMatt Corallo <git@bluematt.me>
Sat, 1 Jun 2019 11:48:33 +0000 (07:48 -0400)
This is still pretty raw and ist mostly just a straight move from
rust-lightning-bitcoinrpc, but first step is to get it there,
second is to actually make it actually safe to use.

.travis.yml
net-tokio/Cargo.toml [new file with mode: 0644]
net-tokio/src/lib.rs [new file with mode: 0644]

index efdff130b0df85d87b7f7b34e4964aa76a8db100..38ddd94d1fd09cc8018647c719889bf9e6a715e1 100644 (file)
@@ -14,3 +14,4 @@ script:
   - cargo build --verbose
   - cargo test --verbose
   - if [ "$(rustup show | grep default | grep 1.29.2)" != "" ]; then cd fuzz && cargo test --verbose && ./travis-fuzz.sh; fi
+  - if [ "$(rustup show | grep default | grep stable)" != "" ]; then cd net-tokio && cargo build --verbose; fi
diff --git a/net-tokio/Cargo.toml b/net-tokio/Cargo.toml
new file mode 100644 (file)
index 0000000..a6fc047
--- /dev/null
@@ -0,0 +1,19 @@
+[package]
+name = "lightning-net-tokio"
+version = "0.0.1"
+authors = ["Matt Corallo"]
+license = "Apache-2.0"
+description = """
+Implementation of the rust-lightning network stack using Tokio.
+For Rust-Lightning clients which wish to make direct connections to Lightning P2P nodes, this is a simple alternative to implementing the nerequired network stack, especially for those already using Tokio.
+"""
+
+[dependencies]
+bitcoin = "0.18"
+bitcoin_hashes = "0.3"
+lightning = { version = "0.0.8", path = "../" }
+secp256k1 = "0.12"
+tokio-codec = "0.1"
+futures = "0.1"
+tokio = "0.1"
+bytes = "0.4"
diff --git a/net-tokio/src/lib.rs b/net-tokio/src/lib.rs
new file mode 100644 (file)
index 0000000..54752fd
--- /dev/null
@@ -0,0 +1,270 @@
+extern crate bytes;
+extern crate tokio;
+extern crate tokio_codec;
+extern crate futures;
+extern crate lightning;
+extern crate secp256k1;
+
+use bytes::BufMut;
+
+use futures::future;
+use futures::future::Future;
+use futures::{AsyncSink, Stream, Sink};
+use futures::sync::mpsc;
+
+use secp256k1::key::PublicKey;
+
+use tokio::timer::Delay;
+use tokio::net::TcpStream;
+
+use lightning::ln::peer_handler;
+use lightning::ln::peer_handler::SocketDescriptor as LnSocketTrait;
+
+use std::mem;
+use std::net::SocketAddr;
+use std::sync::{Arc, Mutex};
+use std::sync::atomic::{AtomicU64, Ordering};
+use std::time::{Duration, Instant};
+use std::vec::Vec;
+use std::hash::Hash;
+
+static ID_COUNTER: AtomicU64 = AtomicU64::new(0);
+
+/// A connection to a remote peer. Can be constructed either as a remote connection using
+/// Connection::setup_outbound o
+pub struct Connection {
+       writer: Option<mpsc::Sender<bytes::Bytes>>,
+       event_notify: mpsc::Sender<()>,
+       pending_read: Vec<u8>,
+       read_blocker: Option<futures::sync::oneshot::Sender<Result<(), ()>>>,
+       read_paused: bool,
+       need_disconnect: bool,
+       id: u64,
+}
+impl Connection {
+       fn schedule_read(peer_manager: Arc<peer_handler::PeerManager<SocketDescriptor>>, us: Arc<Mutex<Self>>, reader: futures::stream::SplitStream<tokio_codec::Framed<TcpStream, tokio_codec::BytesCodec>>) {
+               let us_ref = us.clone();
+               let us_close_ref = us.clone();
+               let peer_manager_ref = peer_manager.clone();
+               tokio::spawn(reader.for_each(move |b| {
+                       let pending_read = b.to_vec();
+                       {
+                               let mut lock = us_ref.lock().unwrap();
+                               assert!(lock.pending_read.is_empty());
+                               if lock.read_paused {
+                                       lock.pending_read = pending_read;
+                                       let (sender, blocker) = futures::sync::oneshot::channel();
+                                       lock.read_blocker = Some(sender);
+                                       return future::Either::A(blocker.then(|_| { Ok(()) }));
+                               }
+                       }
+                       //TODO: There's a race where we don't meet the requirements of disconnect_socket if its
+                       //called right here, after we release the us_ref lock in the scope above, but before we
+                       //call read_event!
+                       match peer_manager.read_event(&mut SocketDescriptor::new(us_ref.clone(), peer_manager.clone()), pending_read) {
+                               Ok(pause_read) => {
+                                       if pause_read {
+                                               let mut lock = us_ref.lock().unwrap();
+                                               lock.read_paused = true;
+                                       }
+                               },
+                               Err(e) => {
+                                       us_ref.lock().unwrap().need_disconnect = false;
+                                       return future::Either::B(future::result(Err(std::io::Error::new(std::io::ErrorKind::InvalidData, e))));
+                               }
+                       }
+
+                       if let Err(e) = us_ref.lock().unwrap().event_notify.try_send(()) {
+                               // Ignore full errors as we just need them to poll after this point, so if the user
+                               // hasn't received the last send yet, it doesn't matter.
+                               assert!(e.is_full());
+                       }
+
+                       future::Either::B(future::result(Ok(())))
+               }).then(move |_| {
+                       if us_close_ref.lock().unwrap().need_disconnect {
+                               peer_manager_ref.disconnect_event(&SocketDescriptor::new(us_close_ref, peer_manager_ref.clone()));
+                               println!("Peer disconnected!");
+                       } else {
+                               println!("We disconnected peer!");
+                       }
+                       Ok(())
+               }));
+       }
+
+       fn new(event_notify: mpsc::Sender<()>, stream: TcpStream) -> (futures::stream::SplitStream<tokio_codec::Framed<TcpStream, tokio_codec::BytesCodec>>, Arc<Mutex<Self>>) {
+               let (writer, reader) = tokio_codec::Framed::new(stream, tokio_codec::BytesCodec::new()).split();
+               let (send_sink, send_stream) = mpsc::channel(3);
+               tokio::spawn(writer.send_all(send_stream.map_err(|_| -> std::io::Error {
+                       unreachable!();
+               })).then(|_| {
+                       future::result(Ok(()))
+               }));
+               let us = Arc::new(Mutex::new(Self { writer: Some(send_sink), event_notify, pending_read: Vec::new(), read_blocker: None, read_paused: false, need_disconnect: true, id: ID_COUNTER.fetch_add(1, Ordering::AcqRel) }));
+
+               (reader, us)
+       }
+
+       /// Process incoming messages and feed outgoing messages on the provided socket generated by
+       /// accepting an incoming connection (by scheduling futures with tokio::spawn).
+       ///
+       /// You should poll the Receive end of event_notify and call get_and_clear_pending_events() on
+       /// ChannelManager and ChannelMonitor objects.
+       pub fn setup_inbound(peer_manager: Arc<peer_handler::PeerManager<SocketDescriptor>>, event_notify: mpsc::Sender<()>, stream: TcpStream) {
+               let (reader, us) = Self::new(event_notify, stream);
+
+               if let Ok(_) = peer_manager.new_inbound_connection(SocketDescriptor::new(us.clone(), peer_manager.clone())) {
+                       Self::schedule_read(peer_manager, us, reader);
+               }
+       }
+
+       /// Process incoming messages and feed outgoing messages on the provided socket generated by
+       /// making an outbound connection which is expected to be accepted by a peer with the given
+       /// public key (by scheduling futures with tokio::spawn).
+       ///
+       /// You should poll the Receive end of event_notify and call get_and_clear_pending_events() on
+       /// ChannelManager and ChannelMonitor objects.
+       pub fn setup_outbound(peer_manager: Arc<peer_handler::PeerManager<SocketDescriptor>>, event_notify: mpsc::Sender<()>, their_node_id: PublicKey, stream: TcpStream) {
+               let (reader, us) = Self::new(event_notify, stream);
+
+               if let Ok(initial_send) = peer_manager.new_outbound_connection(their_node_id, SocketDescriptor::new(us.clone(), peer_manager.clone())) {
+                       if SocketDescriptor::new(us.clone(), peer_manager.clone()).send_data(&initial_send, 0, true) == initial_send.len() {
+                               Self::schedule_read(peer_manager, us, reader);
+                       } else {
+                               println!("Failed to write first full message to socket!");
+                       }
+               }
+       }
+
+       /// Process incoming messages and feed outgoing messages on a new connection made to the given
+       /// socket address which is expected to be accepted by a peer with the given public key (by
+       /// scheduling futures with tokio::spawn).
+       ///
+       /// You should poll the Receive end of event_notify and call get_and_clear_pending_events() on
+       /// ChannelManager and ChannelMonitor objects.
+       pub fn connect_outbound(peer_manager: Arc<peer_handler::PeerManager<SocketDescriptor>>, event_notify: mpsc::Sender<()>, their_node_id: PublicKey, addr: SocketAddr) {
+               let connect_timeout = Delay::new(Instant::now() + Duration::from_secs(10)).then(|_| {
+                       future::err(std::io::Error::new(std::io::ErrorKind::TimedOut, "timeout reached"))
+               });
+               tokio::spawn(TcpStream::connect(&addr).select(connect_timeout)
+                       .and_then(move |stream| {
+                               Connection::setup_outbound(peer_manager, event_notify, their_node_id, stream.0);
+                               future::ok(())
+                       }).or_else(|_| {
+                               //TODO: return errors somehow
+                               future::ok(())
+                       }));
+       }
+}
+
+#[derive(Clone)]
+pub struct SocketDescriptor {
+       conn: Arc<Mutex<Connection>>,
+       id: u64,
+       peer_manager: Arc<peer_handler::PeerManager<SocketDescriptor>>,
+}
+impl SocketDescriptor {
+       fn new(conn: Arc<Mutex<Connection>>, peer_manager: Arc<peer_handler::PeerManager<SocketDescriptor>>) -> Self {
+               let id = conn.lock().unwrap().id;
+               Self { conn, id, peer_manager }
+       }
+}
+impl peer_handler::SocketDescriptor for SocketDescriptor {
+       fn send_data(&mut self, data: &Vec<u8>, write_offset: usize, resume_read: bool) -> usize {
+               macro_rules! schedule_read {
+                       ($us_ref: expr) => {
+                               tokio::spawn(future::lazy(move || -> Result<(), ()> {
+                                       let mut read_data = Vec::new();
+                                       {
+                                               let mut us = $us_ref.conn.lock().unwrap();
+                                               mem::swap(&mut read_data, &mut us.pending_read);
+                                       }
+                                       if !read_data.is_empty() {
+                                               let mut us_clone = $us_ref.clone();
+                                               match $us_ref.peer_manager.read_event(&mut us_clone, read_data) {
+                                                       Ok(pause_read) => {
+                                                               if pause_read { return Ok(()); }
+                                                       },
+                                                       Err(_) => {
+                                                               //TODO: Not actually sure how to do this
+                                                               return Ok(());
+                                                       }
+                                               }
+                                       }
+                                       let mut us = $us_ref.conn.lock().unwrap();
+                                       if let Some(sender) = us.read_blocker.take() {
+                                               sender.send(Ok(())).unwrap();
+                                       }
+                                       us.read_paused = false;
+                                       if let Err(e) = us.event_notify.try_send(()) {
+                                               // Ignore full errors as we just need them to poll after this point, so if the user
+                                               // hasn't received the last send yet, it doesn't matter.
+                                               assert!(e.is_full());
+                                       }
+                                       Ok(())
+                               }));
+                       }
+               }
+
+               let mut us = self.conn.lock().unwrap();
+               if resume_read {
+                       let us_ref = self.clone();
+                       schedule_read!(us_ref);
+               }
+               if data.len() == write_offset { return 0; }
+               if us.writer.is_none() {
+                       us.read_paused = true;
+                       return 0;
+               }
+
+               let mut bytes = bytes::BytesMut::with_capacity(data.len() - write_offset);
+               bytes.put(&data[write_offset..]);
+               let write_res = us.writer.as_mut().unwrap().start_send(bytes.freeze());
+               match write_res {
+                       Ok(res) => {
+                               match res {
+                                       AsyncSink::Ready => {
+                                               data.len() - write_offset
+                                       },
+                                       AsyncSink::NotReady(_) => {
+                                               us.read_paused = true;
+                                               let us_ref = self.clone();
+                                               tokio::spawn(us.writer.take().unwrap().flush().then(move |writer_res| -> Result<(), ()> {
+                                                       if let Ok(writer) = writer_res {
+                                                               {
+                                                                       let mut us = us_ref.conn.lock().unwrap();
+                                                                       us.writer = Some(writer);
+                                                               }
+                                                               schedule_read!(us_ref);
+                                                       } // we'll fire the disconnect event on the socket reader end
+                                                       Ok(())
+                                               }));
+                                               0
+                                       }
+                               }
+                       },
+                       Err(_) => {
+                               // We'll fire the disconnected event on the socket reader end
+                               0
+                       },
+               }
+       }
+
+       fn disconnect_socket(&mut self) {
+               let mut us = self.conn.lock().unwrap();
+               us.need_disconnect = true;
+               us.read_paused = true;
+       }
+}
+impl Eq for SocketDescriptor {}
+impl PartialEq for SocketDescriptor {
+       fn eq(&self, o: &Self) -> bool {
+               self.id == o.id
+       }
+}
+impl Hash for SocketDescriptor {
+       fn hash<H: std::hash::Hasher>(&self, state: &mut H) {
+               self.id.hash(state);
+       }
+}
+