alktunnels/tests/local_halves.rs

//! Tests for the `local` feature: real-socket halves through the real
//! establisher path (`local_dial` injected as the `DialFn`), TCP/UDP/
//! unix round-trips against real local echo servers, the OQ-TN-13
//! truncation fail-loud probe, and the 1400-byte MTU datagram.
//!
//! The forward topology (`wire_forward`) carries the full path: the
//! consumer opens (`TunnelSession::open`), the producer's establisher
//! dials via `local_dial`, the pump handler pumps — real sockets at
//! both ends.

#![cfg(feature = "local")]

mod harness;

use std::sync::Arc;

use alktunnels::params::{Substrate, TunnelParams};
use alktunnels::producer::ResourceRegistry;
use alktunnels::TunnelSession;

use harness::wire_forward;

/// A TCP echo server on an ephemeral port.
async fn tcp_echo_server() -> std::net::SocketAddr {
    let listener = tokio::net::TcpListener::bind(("127.0.0.1", 0))
        .await
        .expect("bind echo");
    let addr = listener.local_addr().expect("addr");
    tokio::spawn(async move {
        loop {
            let Ok((sock, _)) = listener.accept().await else {
                return;
            };
            tokio::spawn(async move {
                let (mut r, mut w) = tokio::io::split(sock);
                let _ = tokio::io::copy(&mut r, &mut w).await;
            });
        }
    });
    addr
}

/// A UDP echo server on an ephemeral port (one task, echo loop).
async fn udp_echo_server() -> std::net::SocketAddr {
    let sock = tokio::net::UdpSocket::bind(("127.0.0.1", 0))
        .await
        .expect("bind udp echo");
    let addr = sock.local_addr().expect("addr");
    tokio::spawn(async move {
        let mut buf = vec![0u8; 65535];
        loop {
            let Ok((n, peer)) = sock.recv_from(&mut buf).await else {
                return;
            };
            if sock.send_to(&buf[..n], peer).await.is_err() {
                return;
            }
        }
    });
    addr
}

#[tokio::test]
async fn tcp_dial_round_trips_through_the_real_establisher_path() {
    let addr = tcp_echo_server().await;
    let registry = ResourceRegistry::new();
    registry
        .register("echo", Substrate::Tcp, &addr.to_string())
        .await;
    let topo = wire_forward(registry, alktunnels::local::local_dial()).await;

    let mut session = TunnelSession::open(&topo.consumer, params("echo", Substrate::Tcp))
        .await
        .expect("tcp open");
    use tokio::io::{AsyncReadExt, AsyncWriteExt};
    let (read, write) = session.stream_halves().expect("halves");
    write.write_all(b"tcp-local").await.expect("write");
    write.flush().await.expect("flush");
    let mut buf = vec![0u8; 9];
    tokio::time::timeout(std::time::Duration::from_secs(5), read.read_exact(&mut buf))
        .await
        .expect("round trip")
        .expect("read");
    assert_eq!(&buf, b"tcp-local");
    session.close().await;
}

#[tokio::test]
async fn udp_datagram_round_trip_through_the_framed_adapter() {
    // The FRAMED adapter end-to-end (ADR-003): session sends the codec
    // frame; the pump copies raw bytes; the dialed UdpHalf strips the
    // frame on recv and delivers ONE datagram to the target; the echo
    // comes back through the same path.
    let addr = udp_echo_server().await;
    let registry = ResourceRegistry::new();
    registry
        .register("dns", Substrate::Udp, &addr.to_string())
        .await;
    let topo = wire_forward(registry, alktunnels::local::local_dial()).await;

    let mut session = TunnelSession::open(&topo.consumer, params("dns", Substrate::Udp))
        .await
        .expect("udp open");
    session.send_datagram(b"query").await.expect("send");
    let dg = tokio::time::timeout(std::time::Duration::from_secs(5), session.recv_datagram())
        .await
        .expect("recv timed out")
        .expect("io ok")
        .expect("datagram");
    assert_eq!(&dg[..], b"query");
    session.close().await;
}

#[tokio::test]
async fn udp_large_datagram_through_bounded_buffers() {
    // The POC's sizing test: a 1400-byte datagram (max ethernet MTU
    // payload) round-trips through the codec + the bounded-buffer path.
    let addr = udp_echo_server().await;
    let registry = ResourceRegistry::new();
    registry
        .register("big", Substrate::Udp, &addr.to_string())
        .await;
    let topo = wire_forward(registry, alktunnels::local::local_dial()).await;

    let mut session = TunnelSession::open(&topo.consumer, params("big", Substrate::Udp))
        .await
        .expect("udp open");
    let payload: Vec<u8> = (0..1400u32).map(|i| (i % 7) as u8).collect();
    session.send_datagram(&payload).await.expect("send");
    let got = tokio::time::timeout(std::time::Duration::from_secs(5), session.recv_datagram())
        .await
        .expect("recv timed out")
        .expect("io ok")
        .expect("datagram");
    assert_eq!(got.len(), payload.len());
    assert_eq!(&got[..], &payload[..]);
    session.close().await;
}

#[tokio::test]
async fn unix_dial_round_trips() {
    // OQ-TN-14: unix ships with local v1 (same halves shape as TCP).
    let dir = std::env::temp_dir().join(format!("alktunnels-test-{}", std::process::id()));
    let _ = std::fs::create_dir_all(&dir);
    let path = dir.join("echo.sock");
    let _ = std::fs::remove_file(&path);

    let listener = tokio::net::UnixListener::bind(&path).expect("bind unix echo");
    tokio::spawn(async move {
        loop {
            let Ok((sock, _)) = listener.accept().await else {
                return;
            };
            tokio::spawn(async move {
                let (mut r, mut w) = tokio::io::split(sock);
                let _ = tokio::io::copy(&mut r, &mut w).await;
            });
        }
    });

    let registry = ResourceRegistry::new();
    registry
        .register("uds", Substrate::Unix, &path.to_string_lossy())
        .await;
    let topo = wire_forward(registry, alktunnels::local::local_dial()).await;

    let mut session = TunnelSession::open(&topo.consumer, params("uds", Substrate::Unix))
        .await
        .expect("unix open");
    use tokio::io::{AsyncReadExt, AsyncWriteExt};
    let (read, write) = session.stream_halves().expect("halves");
    write.write_all(b"unix-local").await.expect("write");
    write.flush().await.expect("flush");
    let mut buf = vec![0u8; 10];
    tokio::time::timeout(std::time::Duration::from_secs(5), read.read_exact(&mut buf))
        .await
        .expect("round trip")
        .expect("read");
    assert_eq!(&buf, b"unix-local");
    session.close().await;
    let _ = std::fs::remove_file(&path);
}

#[tokio::test]
async fn truncation_fails_loud_oq_tn_13() {
    // The fail-loud posture (OQ-TN-13) at the FRAMED adapter: the
    // socket-facing recv uses the full 65535-byte scratch (try_recv
    // truncates silently at the buffer size), so a well-formed
    // >caller-buffer datagram is received whole and STAGED — never
    // silently truncated into a shorter frame. The staging surface is
    // pinned by `oversized_frame_stages_across_reads` (below); this
    // test pins the fail-loud path directly at the unit level: a
    // scratch smaller than the datagram WOULD have corrupted the
    // framing (the U-3 re-validation's warning), so the scratch must
    // be full-size. A datagram at the IP/UDP max (65507) is within
    // the u16 frame bound (65535) — it must arrive intact.
    let sock_a = tokio::net::UdpSocket::bind(("127.0.0.1", 0))
        .await
        .expect("bind a");
    let sock_b = tokio::net::UdpSocket::bind(("127.0.0.1", 0))
        .await
        .expect("bind b");
    sock_a
        .connect(sock_b.local_addr().expect("b addr"))
        .await
        .expect("connect a");
    sock_b
        .connect(sock_a.local_addr().expect("a addr"))
        .await
        .expect("connect b");

    // The producer-side adapter (what the establisher dials): sock_a
    // wrapped as halves. A 65507-byte datagram (the UDP theoretical
    // max — a well-formed wire's largest possible frame) from the
    // wire side.
    let (read, _write) = alktunnels::local::UdpHalf::split(sock_a);
    let mut read = read;
    let payload: Vec<u8> = (0..65507u32).map(|i| (i % 251) as u8).collect();
    sock_b.send(&payload).await.expect("send max-size");

    // The pump reads with the codec's chunk size (16 KiB) — far
    // smaller than the datagram: the adapter must STAGE, never
    // truncate. Drain through small reads and reassemble the frame.
    use tokio::io::AsyncReadExt;
    let mut got = Vec::new();
    let mut chunk = vec![0u8; 16 * 1024];
    while got.len() < payload.len() + 2 {
        let n = tokio::time::timeout(std::time::Duration::from_secs(5), read.read(&mut chunk))
            .await
            .expect("read timed out (datagram truncated?)")
            .expect("read ok");
        assert!(n > 0, "zero-length read before the frame completed");
        got.extend_from_slice(&chunk[..n]);
    }
    assert_eq!(&got[..2], &[0xff, 0xe3], "declared length = 65507");
    assert_eq!(
        &got[2..],
        &payload[..],
        "full payload intact — no truncation"
    );
}

/// The >caller-buffer (but within-frame-bound) datagram: the framed
/// adapter STAGES the frame and hands it out across reads — the pump's
/// far-side de-framer reassembles the true length (no silent
/// truncation, no shorter-datagram misread). The reassembly is pinned
/// by the end-to-end gates below; here the staging surface is pinned
/// directly: successive reads yield the full frame.
#[tokio::test]
async fn oversized_frame_stages_across_reads() {
    let sock_a = tokio::net::UdpSocket::bind(("127.0.0.1", 0))
        .await
        .expect("bind a");
    let sock_b = tokio::net::UdpSocket::bind(("127.0.0.1", 0))
        .await
        .expect("bind b");
    sock_a
        .connect(sock_b.local_addr().expect("b addr"))
        .await
        .expect("connect a");
    sock_b
        .connect(sock_a.local_addr().expect("a addr"))
        .await
        .expect("connect b");

    let (read, _write) = alktunnels::local::UdpHalf::split(sock_a);
    let mut read = read;
    let payload: Vec<u8> = (0..20000u32).map(|i| (i % 251) as u8).collect();
    sock_b.send(&payload).await.expect("send 20k");

    // Read through a small caller buffer (16 KiB — smaller than the
    // ~20002-byte frame): the first read fills it completely, the
    // second hands out the staged remainder.
    use tokio::io::AsyncReadExt;
    let mut got = Vec::new();
    let mut chunk = vec![0u8; 16 * 1024];
    let n1 = tokio::time::timeout(std::time::Duration::from_secs(5), read.read(&mut chunk))
        .await
        .expect("first read")
        .expect("read ok");
    got.extend_from_slice(&chunk[..n1]);
    let n2 = tokio::time::timeout(std::time::Duration::from_secs(5), read.read(&mut chunk))
        .await
        .expect("second read")
        .expect("read ok");
    got.extend_from_slice(&chunk[..n2]);

    assert_eq!(got.len(), payload.len() + 2, "frame length read out");
    assert_eq!(&got[..2], &[len_hi(payload.len()), len_lo(payload.len())]);
    assert_eq!(&got[2..], &payload[..], "payload bytes intact");
}

fn len_hi(n: usize) -> u8 {
    (n >> 8) as u8
}
fn len_lo(n: usize) -> u8 {
    (n & 0xff) as u8
}

/// A RECORDING UDP "target" (review 001's verification-gate probe
/// shape, made permanent): it records every datagram it receives
/// (byte-for-byte, to the shared log) and replies with canned,
/// configurable datagrams — including EMPTY and RAW-UNFRAMED ones the
/// echo server can't produce.
struct RecordingTarget {
    addr: std::net::SocketAddr,
    received: Arc<tokio::sync::Mutex<Vec<Vec<u8>>>>,
}

async fn recording_udp_target(replies: Vec<Vec<u8>>) -> RecordingTarget {
    let sock = tokio::net::UdpSocket::bind(("127.0.0.1", 0))
        .await
        .expect("bind recording target");
    let addr = sock.local_addr().expect("addr");
    let received = Arc::new(tokio::sync::Mutex::new(Vec::new()));
    let log = Arc::clone(&received);
    tokio::spawn(async move {
        let mut buf = vec![0u8; 65535];
        // Reply cyclically (raw, as a real target would) to each
        // datagram that arrives, then keep recording.
        let mut next = 0usize;
        loop {
            let Ok((n, peer)) = sock.recv_from(&mut buf).await else {
                return;
            };
            log.lock().await.push(buf[..n].to_vec());
            if !replies.is_empty() {
                let reply = &replies[next % replies.len()];
                next += 1;
                if sock.send_to(reply, peer).await.is_err() {
                    return;
                }
            }
        }
    });
    RecordingTarget { addr, received }
}

fn params(resource: &str, substrate: Substrate) -> TunnelParams {
    TunnelParams {
        resource: resource.to_string(),
        substrate,
    }
}

/// GATE 1a (review 001): the target receives PAYLOAD BYTES ONLY — no
/// `[len: u16 BE]` codec prefix (the U-1 defect's primary face).
#[tokio::test]
async fn gate_target_receives_payload_bytes_only() {
    let target = recording_udp_target(vec![]).await;
    let registry = ResourceRegistry::new();
    registry
        .register("rec", Substrate::Udp, &target.addr.to_string())
        .await;
    let topo = wire_forward(registry, alktunnels::local::local_dial()).await;

    let mut session = TunnelSession::open(&topo.consumer, params("rec", Substrate::Udp))
        .await
        .expect("udp open");
    session.send_datagram(b"query").await.expect("send");
    // Give the datagram time to cross the pump.
    for _ in 0..50 {
        if !target.received.lock().await.is_empty() {
            break;
        }
        tokio::time::sleep(std::time::Duration::from_millis(20)).await;
    }
    let received = target.received.lock().await;
    assert_eq!(received.len(), 1, "exactly one datagram at the target");
    assert_eq!(
        &received[0][..],
        b"query",
        "the target sees the payload, not the codec frame (no 00 05 prefix)"
    );
    session.close().await;
}

/// GATE 1b: an EMPTY datagram round-trips over the REAL socket path —
/// `recv_datagram` yields `Some(b"")` and the channel stays alive
/// (the F-2 fix end-to-end; the raw adapter collapsed it to EOF).
#[tokio::test]
async fn gate_empty_datagram_round_trips_over_the_real_socket_path() {
    // The target replies cyclically: empty datagram first (the
    // round-trip under test), then `still-alive` (the liveness probe).
    let target = recording_udp_target(vec![vec![], b"still-alive".to_vec()]).await;
    let registry = ResourceRegistry::new();
    registry
        .register("empty", Substrate::Udp, &target.addr.to_string())
        .await;
    let topo = wire_forward(registry, alktunnels::local::local_dial()).await;

    let mut session = TunnelSession::open(&topo.consumer, params("empty", Substrate::Udp))
        .await
        .expect("udp open");
    session.send_datagram(b"").await.expect("send empty");
    let dg = tokio::time::timeout(std::time::Duration::from_secs(5), session.recv_datagram())
        .await
        .expect("recv timed out (empty datagram collapsed to EOF?)")
        .expect("io ok")
        .expect("empty datagram is Some, NOT None (EOF)");
    assert!(dg.is_empty(), "the empty datagram round-trips");

    // The channel is still alive: a second exchange works (the
    // canned reply cycles — the target sends b"still-alive" back).
    session.send_datagram(b"second").await.expect("send 2");
    let dg2 = tokio::time::timeout(std::time::Duration::from_secs(5), session.recv_datagram())
        .await
        .expect("recv 2 timed out")
        .expect("io ok 2")
        .expect("datagram 2");
    assert_eq!(&dg2[..], b"still-alive");
    session.close().await;
}

/// GATE 2: a >8 KiB datagram round-trips as ONE datagram at the
/// target (the 8 KiB copy-buffer split is gone — the frame reassembles
/// at the adapter before crossing the pump).
#[tokio::test]
async fn gate_large_datagram_arrives_as_one_datagram() {
    let target = recording_udp_target(vec![]).await;
    let registry = ResourceRegistry::new();
    registry
        .register("big", Substrate::Udp, &target.addr.to_string())
        .await;
    let topo = wire_forward(registry, alktunnels::local::local_dial()).await;

    let mut session = TunnelSession::open(&topo.consumer, params("big", Substrate::Udp))
        .await
        .expect("udp open");
    // 9000 bytes — the review's probe size that split as 8192 + 810.
    let payload: Vec<u8> = (0..9000u32).map(|i| (i % 251) as u8).collect();
    session.send_datagram(&payload).await.expect("send 9000");
    for _ in 0..100 {
        if !target.received.lock().await.is_empty() {
            break;
        }
        tokio::time::sleep(std::time::Duration::from_millis(20)).await;
    }
    let received = target.received.lock().await;
    assert_eq!(received.len(), 1, "ONE datagram at the target, not two");
    assert_eq!(
        received[0].len(),
        9000,
        "full payload in one datagram (no 8192 split)"
    );
    assert_eq!(&received[0][..], &payload[..], "payload intact");
    session.close().await;
}

/// GATE 3: an unframed target-initiated datagram resolves at the
/// session (the session codec decodes the framed bytes the adapter
/// produced — real-target traffic is not parked mid-frame).
#[tokio::test]
async fn gate_unframed_target_datagram_resolves_at_the_session() {
    // A raw "HELLO" — a real server's greeting the raw adapter parked
    // mid-frame (the probe's 2s timeout).
    let target = recording_udp_target(vec![b"HELLO".to_vec()]).await;
    let registry = ResourceRegistry::new();
    registry
        .register("greet", Substrate::Udp, &target.addr.to_string())
        .await;
    let topo = wire_forward(registry, alktunnels::local::local_dial()).await;

    let mut session = TunnelSession::open(&topo.consumer, params("greet", Substrate::Udp))
        .await
        .expect("udp open");
    session.send_datagram(b"ping").await.expect("send");
    let dg = tokio::time::timeout(std::time::Duration::from_secs(5), session.recv_datagram())
        .await
        .expect("recv timed out — unframed target traffic was parked/dropped")
        .expect("io ok")
        .expect("datagram");
    assert_eq!(&dg[..], b"HELLO");
    session.close().await;
}

/// U-2 (review 001): `connect_udp`'s bind is family-correct — a v6
/// target dials from a v6-bound socket (the hardcoded `0.0.0.0` bind
/// broke every v6-only target at connect). A real ::1 target
/// round-trips through the full path.
#[tokio::test]
async fn connect_udp_binds_the_matching_family_v6_targets_dial() {
    let sock = tokio::net::UdpSocket::bind(("::1", 0))
        .await
        .expect("bind v6 echo");
    let addr = sock.local_addr().expect("v6 addr");
    tokio::spawn(async move {
        let mut buf = vec![0u8; 65535];
        loop {
            let Ok((n, peer)) = sock.recv_from(&mut buf).await else {
                return;
            };
            if sock.send_to(&buf[..n], peer).await.is_err() {
                return;
            }
        }
    });

    let registry = ResourceRegistry::new();
    registry
        .register("v6", Substrate::Udp, &addr.to_string())
        .await;
    let topo = wire_forward(registry, alktunnels::local::local_dial()).await;

    let mut session = TunnelSession::open(&topo.consumer, params("v6", Substrate::Udp))
        .await
        .expect("v6 udp open (family-correct bind)");
    session.send_datagram(b"v6-query").await.expect("send v6");
    let dg = tokio::time::timeout(std::time::Duration::from_secs(5), session.recv_datagram())
        .await
        .expect("recv timed out")
        .expect("io ok")
        .expect("datagram");
    assert_eq!(&dg[..], b"v6-query");
    session.close().await;
}

#[tokio::test]
async fn dial_refusal_maps_to_dial_failed() {
    // Error mapping: a refused dial (nothing listening on the port) is
    // `dial_failed` on the wire (`channel:open_failed` reason).
    let registry = ResourceRegistry::new();
    registry
        .register("dead", Substrate::Tcp, "127.0.0.1:1")
        .await;
    let topo = wire_forward(registry, alktunnels::local::local_dial()).await;

    let err = match TunnelSession::open(&topo.consumer, params("dead", Substrate::Tcp)).await {
        Ok(_) => panic!("refused dial must fail"),
        Err(e) => e,
    };
    assert_eq!(err.open_ref().establishment_reason(), Some("dial_failed"));
}

#[tokio::test]
async fn listen_producer_over_a_real_tcp_listener() {
    // The listen shape + local halves compose: a real TcpListenerHalves
    // accept loop feeds an AcceptQueue; a listen-registered producer
    // pops accepted sockets; the consumer pumps the tunnel against the
    // accepted local socket.
    let queue = alktunnels::producer::AcceptQueue::new();
    let listener = alktunnels::local::bind_tcp("127.0.0.1:0")
        .await
        .expect("bind");
    let addr = listener.local_addr().expect("addr");

    let accept_loop = tokio::spawn({
        let queue = queue.clone();
        async move { listener.accept_loop(queue).await }
    });

    // A local client connects (the accepted connection's source). The
    // driving end comes from `try_clone` — the pump owns the accepted
    // handle; the driver pushes/reads the echo through the same real
    // socket.
    // std-level try_clone BEFORE converting to tokio streams: one fd
    // for the driver, one wrapped for the pump (two independent
    // handles to the same socket). O_NONBLOCK is shared across the
    // dup, so BOTH ends go nonblocking and the driver is driven
    // async too.
    let local_std = std::net::TcpStream::connect(addr).expect("std connect");
    let driver_std = local_std.try_clone().expect("try_clone driver");
    local_std.set_nonblocking(true).expect("nonblocking");
    let mut driver = tokio::net::TcpStream::from_std(driver_std).expect("tokio driver");

    let registry = ResourceRegistry::new();
    registry
        .register("listener", Substrate::Tcp, "assembly-owned")
        .await;
    let accept_fn: alktunnels::producer::AcceptFn = Arc::new(move || {
        let queue = queue.clone();
        Box::pin(async move {
            queue.pop().await.ok_or_else(|| {
                alktunnels::producer::TunnelEstablishError::DialFailed(
                    "listener closed".to_string(),
                )
            })
        })
    });
    let topo = harness::wire_listen(registry, accept_fn).await;

    let channel_id = alktunnels::open_reverse_channel(
        &topo.consumer_call,
        &params("listener", Substrate::Tcp),
        None,
    )
    .await
    .expect("listen open");

    let session = TunnelSession::adopt(
        &topo.consumer_manager,
        channel_id,
        Substrate::Tcp,
        alktunnels::TUNNEL_ALPN,
    )
    .await
    .expect("adopt");
    // The producer's pump (wrapper-managed) copies channel <-> the
    // accepted handle; the accepted handle IS the client socket — so
    // the session's borrowed halves drive it directly: the driver end
    // writes into the client socket, the pump carries it over the
    // channel, and the session halves read it (and back).
    let mut session = session;
    use tokio::io::{AsyncReadExt, AsyncWriteExt};
    driver.write_all(b"listen-local").await.expect("write");
    driver.flush().await.expect("flush");
    let (read, _write) = session.stream_halves().expect("halves");
    let mut buf = vec![0u8; 12];
    tokio::time::timeout(std::time::Duration::from_secs(5), read.read_exact(&mut buf))
        .await
        .expect("round trip")
        .expect("read");
    assert_eq!(&buf, b"listen-local");

    session.close().await;
    accept_loop.abort();
}

// =====================================================================
// UdpHalf write-side machinery (review 001's U-1 invariants, the
// never-blocks paths the end-to-end gates don't reach): WouldBlock
// stashing, the accepted-byte high-water backpressure, the flush
// drain, and the shutdown best-effort drain — driven with poll_fn
// over a shrunken-SO_SNDBUF socket (deterministic, no timing).
// =====================================================================

/// An UNCONNECTED UdpHalf write half: `try_send` fails
/// deterministically with `EDESTADDRREQ` ("destination address
/// required") — the deterministic probe for the write path's ERROR
/// arms on this host, where no UDP sender-blocking shape was
/// reachable (loopback/veth drop silently; see the host-notes on the
/// drain test).
fn unconnected_write_half() -> Box<dyn tokio::io::AsyncWrite + Send + Sync + Unpin> {
    let std_sock = std::net::UdpSocket::bind("127.0.0.1:0").expect("bind unconnected");
    std_sock.set_nonblocking(true).expect("nonblocking");
    let sock = tokio::net::UdpSocket::from_std(std_sock).expect("tokio socket");
    let (_read, write) = alktunnels::local::UdpHalf::split(sock);
    write
}

/// Drive one `poll_write` with the given buffer (one poll — the
/// pump's copy shape).
async fn poll_write_once<W: tokio::io::AsyncWrite + Unpin>(
    write: &mut W,
    buf: &[u8],
) -> std::io::Result<usize> {
    futures::future::poll_fn(|cx| tokio::io::AsyncWrite::poll_write(Pin::new(write), cx, buf)).await
}

/// Drive one `poll_flush` to completion (the copy's flush).
async fn poll_flush_once<W: tokio::io::AsyncWrite + Unpin>(write: &mut W) -> std::io::Result<()> {
    futures::future::poll_fn(|cx| tokio::io::AsyncWrite::poll_flush(Pin::new(write), cx)).await
}

/// Drive one `poll_shutdown`.
async fn poll_shutdown_once<W: tokio::io::AsyncWrite + Unpin>(
    write: &mut W,
) -> std::io::Result<()> {
    futures::future::poll_fn(|cx| tokio::io::AsyncWrite::poll_shutdown(Pin::new(write), cx)).await
}

use std::pin::Pin;

#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
async fn udp_half_write_queue_drains_on_flush_and_shutdown() {
    // The queue/flush machinery on a WORKING socket: poll_write
    // accepts frames (the de-framer's queue + accepted accounting),
    // flush drains them onto the socket (Ready(Ok)), and shutdown's
    // best-effort drain reports Ready(Ok) (UDP has no shutdown).
    // The peer reads (a real probe) — the datagrams land.
    //
    // Host-notes (probed 2026-09-09): (a) a fresh-socket first
    // `try_send` may report WouldBlock spuriously on this kernel
    // (one readiness+retry cycle clears it, ~20% of runs) — the
    // flush loop tolerates it, which also exercises the `sending`
    // stash + re-offer path whenever it fires. (b) The test runs
    // multi-thread: on a current-thread runtime the kernel's
    // socket-registration timing left datagrams undelivered at
    // bound-but-unread peers (tokio 1.53 probe result), so the
    // delivery assertion is only trustworthy multi-threaded.
    let peer = tokio::net::UdpSocket::bind(("127.0.0.1", 0))
        .await
        .expect("bind peer");
    let snd_sock = socket2::Socket::new(
        socket2::Domain::IPV4,
        socket2::Type::DGRAM,
        Some(socket2::Protocol::UDP),
    )
    .expect("raw sender socket");
    snd_sock.set_nonblocking(true).expect("nonblocking");
    snd_sock
        .bind(&socket2::SockAddr::from(
            "127.0.0.1:0"
                .parse::<std::net::SocketAddr>()
                .expect("parse"),
        ))
        .expect("bind sender");
    let snd_std: std::net::UdpSocket = snd_sock.into();
    let snd = tokio::net::UdpSocket::from_std(snd_std).expect("tokio sender");
    snd.connect(peer.local_addr().expect("peer addr"))
        .await
        .expect("connect sender");

    let (_read, mut write) = alktunnels::local::UdpHalf::split(snd);

    // Accept three frames (the de-framer queue holds them).
    for i in 0..3u8 {
        let frame = [0, 3, i, i, i];
        let res = poll_write_once(&mut write, &frame).await;
        assert!(
            matches!(res, Ok(5)),
            "poll_write accepted the frame: {res:?}"
        );
    }

    // Flush drains the queue onto the socket — bounded cycles: the
    // quirk's stash+Pending resolves on the next flush (poll_send_ready
    // is Ok), a saturated socket stays queued (not this host's case).
    let mut flushed = false;
    for _ in 0..8 {
        match tokio::time::timeout(
            std::time::Duration::from_secs(5),
            poll_flush_once(&mut write),
        )
        .await
        .expect("flush timed out")
        {
            Ok(()) => {
                flushed = true;
                break;
            }
            Err(e) if e.kind() == std::io::ErrorKind::WouldBlock => {}
            Err(e) => panic!("unexpected flush error: {e:?}"),
        }
    }
    assert!(flushed, "flush drained the queue within 8 cycles");

    // The peer saw all three datagrams (the accounting released).
    let mut buf = vec![0u8; 65535];
    let mut received = 0;
    for _ in 0..50 {
        match peer.try_recv(&mut buf) {
            Ok(_) => received += 1,
            Err(_) if received < 3 => {
                tokio::time::sleep(std::time::Duration::from_millis(20)).await;
            }
            Err(_) => break,
        }
    }
    assert_eq!(received, 3, "the peer saw the flushed datagrams");

    // Shutdown: the best-effort drain (the queue is already empty) →
    // Ready(Ok).
    let res = poll_shutdown_once(&mut write).await;
    assert!(res.is_ok(), "shutdown is Ready(Ok): {res:?}");
    drop(peer);
}

#[tokio::test]
async fn udp_half_write_error_maps_loud_on_unconnected_socket() {
    // The ERROR arms: an unconnected socket's `try_send` fails
    // deterministically (`EDESTADDRREQ`) — poll_write's drain maps
    // the socket error loud (the `pump_queue_to_socket` Err arm),
    // and flush re-raises it (the `poll_flush` error surface). The
    // high-water backpressure and the WouldBlock stash remain
    // host-dependent (see connected_udp_pair's note) — covered by the
    // leftover-coverage task.
    let mut write = unconnected_write_half();

    let frame = [0, 3, b'x', b'y', b'z'];
    let res = poll_write_once(&mut write, &frame).await;
    match res {
        // The opportunistic drain swallows the error on the accept
        // path (`let _ =`); the queue still holds the datagram.
        Ok(_) => {}
        Err(e) => panic!("poll_write must ACCEPT (the drain error is swallowed): {e:?}"),
    }

    // Flush surfaces the socket error loud (the drain's Err arm).
    let res = tokio::time::timeout(
        std::time::Duration::from_secs(5),
        poll_flush_once(&mut write),
    )
    .await
    .expect("flush timed out");
    assert!(res.is_err(), "flush must surface the unconnected error");
}

#[tokio::test]
async fn udp_half_poll_shutdown_drains_best_effort_and_is_idempotent() {
    // UDP has no shutdown: poll_shutdown drains the queue best-effort
    // and reports Ready(Ok) — the drop-is-close posture. On a
    // saturated socket the drain is best-effort too (the error is
    // swallowed — shutdown never fails the pump).
    let mut write = unconnected_write_half();
    use tokio::io::AsyncWriteExt;
    write
        .write_all(&[0, 3, b'x', b'y', b'z'])
        .await
        .expect("accept");
    let res = poll_shutdown_once(&mut write).await;
    assert!(
        res.is_ok(),
        "shutdown is Ready(Ok) even when the drain errors: {res:?}"
    );
}

#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
async fn udp_half_flush_drains_repeatedly_until_the_queue_is_empty() {
    // The flush loop's pump-again shape (tokio's copy polls flush
    // after every window drain): each flush drains that window's
    // queued datagrams; five write+flush cycles → five datagrams at
    // the peer (the fresh-socket quirk tolerated per cycle — see the
    // host-notes on the drain test; multi-thread runtime for the same
    // delivery-timing reason).
    let peer = tokio::net::UdpSocket::bind(("127.0.0.1", 0))
        .await
        .expect("bind peer");
    let snd_sock = socket2::Socket::new(
        socket2::Domain::IPV4,
        socket2::Type::DGRAM,
        Some(socket2::Protocol::UDP),
    )
    .expect("raw sender socket");
    snd_sock.set_nonblocking(true).expect("nonblocking");
    snd_sock
        .bind(&socket2::SockAddr::from(
            "127.0.0.1:0"
                .parse::<std::net::SocketAddr>()
                .expect("parse"),
        ))
        .expect("bind sender");
    let snd_std: std::net::UdpSocket = snd_sock.into();
    let snd = tokio::net::UdpSocket::from_std(snd_std).expect("tokio sender");
    snd.connect(peer.local_addr().expect("peer addr"))
        .await
        .expect("connect sender");

    let (_read, mut write) = alktunnels::local::UdpHalf::split(snd);

    for i in 0..5u8 {
        let frame = [0, 1, i];
        poll_write_once(&mut write, &frame).await.expect("accept");
        // Each cycle flushes (bounded retries — the fresh-socket
        // quirk's stash clears on the second readiness cycle).
        let mut flushed = false;
        for _ in 0..8 {
            match poll_flush_once(&mut write).await {
                Ok(()) => {
                    flushed = true;
                    break;
                }
                Err(e) if e.kind() == std::io::ErrorKind::WouldBlock => {}
                Err(e) => panic!("unexpected flush {i} error: {e:?}"),
            }
        }
        assert!(flushed, "flush cycle {i} drained");
    }

    let mut buf = vec![0u8; 65535];
    let mut received = 0;
    for _ in 0..50 {
        match peer.try_recv(&mut buf) {
            Ok(_) => received += 1,
            Err(_) if received < 5 => {
                tokio::time::sleep(std::time::Duration::from_millis(20)).await;
            }
            Err(_) => break,
        }
    }
    assert_eq!(received, 5, "five flush cycles, five datagrams");
    drop(peer);
}

/// The resolve arm (`connect_udp`'s no-addresses mapping): a target
/// whose name cannot be resolved (an RFC 6761 `.invalid` name —
/// guaranteed NXDOMAIN, resolved instantly and without network I/O)
/// maps to `dial_failed` (the resolve step's error posture, ADR-004's
/// `DialFn` contract). Rootless, deterministic.
#[tokio::test]
async fn connect_udp_unresolvable_target_maps_to_dial_failed() {
    let err = match alktunnels::local::connect_udp("definitely-not-a-real-host-xyz.invalid:1").await
    {
        Ok(_) => panic!("an unresolvable target must fail"),
        Err(e) => e,
    };
    assert!(
        matches!(
            err,
            alktunnels::producer::TunnelEstablishError::DialFailed(_)
        ),
        "resolve failure maps to dial_failed: {err:?}"
    );
}

// =====================================================================
// The sender-block machinery (the coverage-weak-spots task's option 1,
// CLOSED — the recipe was found): a veth fixture with the peer
// address UNASSIGNED and the neighbor table flushed puts the
// connected UDP sender into a PERMANENT WouldBlock/ENOSPC state
// (the unresolved neighbor never resolves; the blocked write path's
// skb stays charged to the socket's wmem with a shrunken SO_SNDBUF).
// Assigning the peer address + binding the peer resolves ARP; the
// queued/stashed datagrams drain and DELIVER. Validated on this host
// (kernel 6.8.0-110-generic, tokio 1.53.1, socket2 0.6.5) — 3/3
// deterministic probe runs, 2026-09-09.
//
// Root-gated (`ip link add`): auto-skipped without root. Never
// touches the default route or any real interface.
// =====================================================================

/// The veth fixture (created and torn down per run): `alktx/alktx-p`
/// with 10.240.0.1/24 (the sender side) and 10.240.0.9 (the peer
/// side, UNASSIGNED during the block phase — its absence is what
/// pins the sender). The peer address is assigned mid-test (the
/// resolve step). A dead-simple guard skips the test when `ip` is
/// unavailable or the command fails (non-root hosts).
struct VethFixture {
    peer_addr_assigned: bool,
}

/// Run a privileged `ip` command: bare `ip` first (uid-0 test runs),
/// `sudo ip` second (the uid-1000 dev runs). `None` = unavailable.
async fn ip_cmd(args: &[&str]) -> Option<bool> {
    let mut spawned = false;
    for prog in ["ip", "sudo"] {
        let mut argv = Vec::with_capacity(args.len() + 1);
        if prog == "sudo" {
            argv.push("ip");
        }
        argv.extend_from_slice(args);
        match tokio::process::Command::new(prog)
            .args(&argv)
            .output()
            .await
        {
            Ok(o) => {
                spawned = true;
                if o.status.success() {
                    return Some(true);
                }
            }
            Err(_) => continue,
        }
    }
    spawned.then_some(false)
}

impl VethFixture {
    async fn try_create() -> Option<Self> {
        // Skip loud-free: the coverage task documents the host boundary.
        let created = ip_cmd(&[
            "link", "add", "alktx", "type", "veth", "peer", "name", "alktx-p",
        ])
        .await
        .unwrap_or(false);
        if !created {
            eprintln!("skipping: veth fixture unavailable (non-root host)");
            return None;
        }
        let _ = ip_cmd(&["addr", "add", "10.240.0.1/24", "dev", "alktx"]).await;
        // A jumbo MTU (65536): a 60 KB datagram crosses as ONE frame —
        // no IP fragmentation, so no fragment-reassembly races at the
        // peer (fragment-ID collisions under the drain burst merged
        // datagrams on an MTU-1500 veth during the probe runs).
        let _ = ip_cmd(&["link", "set", "alktx", "mtu", "65535"]).await;
        let _ = ip_cmd(&["link", "set", "alktx", "up"]).await;
        let _ = ip_cmd(&["link", "set", "alktx-p", "up"]).await;
        Some(Self {
            peer_addr_assigned: false,
        })
    }

    fn assign_peer(&mut self) -> bool {
        if self.peer_addr_assigned {
            return true;
        }
        let ok = std::process::Command::new("ip")
            .args(["addr", "add", "10.240.0.9/24", "dev", "alktx-p"])
            .output()
            .ok()
            .map(|o| o.status.success())
            .unwrap_or(false)
            || std::process::Command::new("sudo")
                .args(["ip", "addr", "add", "10.240.0.9/24", "dev", "alktx-p"])
                .output()
                .ok()
                .map(|o| o.status.success())
                .unwrap_or(false);
        self.peer_addr_assigned = ok;
        ok
    }
}

impl Drop for VethFixture {
    fn drop(&mut self) {
        let _ = std::process::Command::new("ip")
            .args(["link", "del", "alktx"])
            .output();
        let _ = std::process::Command::new("sudo")
            .args(["ip", "link", "del", "alktx"])
            .output();
    }
}

/// A sender socket blocked on the unresolved neighbor: connect to the
/// UNASSIGNED peer address, flush the neighbor table, shrunken
/// SO_SNDBUF. `try_send` on a 60 KB datagram (truesize beyond the
/// buffer) then reports WouldBlock/ENOSPC persistently — the state
/// the write-side machinery is for. Returns the tokio socket ready
/// for `UdpHalf::split`.
async fn blocked_sender_socket(peer_port: u16) -> Option<tokio::net::UdpSocket> {
    // The BLOCK-STATE PROBE runs on a THROWAWAY socket: an oversized
    // probe datagram's partially-built skb chain stays corked in
    // sk_write_queue on EAGAIN (the kernel flushes it on the next
    // successful send) — poisoning the socket with a stale partial
    // datagram. The test socket below is created fresh afterwards.
    let probe = socket2::Socket::new(
        socket2::Domain::IPV4,
        socket2::Type::DGRAM,
        Some(socket2::Protocol::UDP),
    )
    .ok()?;
    probe.set_nonblocking(true).ok()?;
    probe.set_send_buffer_size(32768).ok()?;
    probe
        .bind(&"10.240.0.1:0".parse::<std::net::SocketAddr>().ok()?.into())
        .ok()?;
    let probe_std: std::net::UdpSocket = probe.into();
    let probe_sock = tokio::net::UdpSocket::from_std(probe_std).ok()?;
    probe_sock.connect(("10.240.0.9", peer_port)).await.ok()?;

    // Confirm the block actually holds (the host invariant): with a
    // 64 KiB SO_SNDBUF, the FIRST 60 KB datagram queues on the
    // unresolved neighbor (Ok — its skb is charged to wmem); every
    // SUBSEQUENT send EAGAINs (the charge exceeds the buffer) until
    // the neighbor resolves. That persistent EAGAIN/ENOBUFS pair IS
    // the block state. A resolved neighbor would send Ok throughout.
    let big = vec![0u8; 60000];
    let mut blocked = 0;
    for _ in 0..3 {
        match probe_sock.try_send(&big) {
            Ok(_) => {} // the first datagram queues; expected
            Err(e) if e.kind() == std::io::ErrorKind::WouldBlock => blocked += 1,
            Err(e) if e.raw_os_error() == Some(105) => blocked += 1,
            Err(_) => return None,
        }
    }
    if blocked < 1 {
        return None;
    }
    drop(probe_sock);

    // The test socket: fresh (no corked state), same posture.
    let sock = socket2::Socket::new(
        socket2::Domain::IPV4,
        socket2::Type::DGRAM,
        Some(socket2::Protocol::UDP),
    )
    .ok()?;
    sock.set_nonblocking(true).ok()?;
    sock.set_send_buffer_size(32768).ok()?;
    sock.bind(&"10.240.0.1:0".parse::<std::net::SocketAddr>().ok()?.into())
        .ok()?;
    let snd_std: std::net::UdpSocket = sock.into();
    let snd = tokio::net::UdpSocket::from_std(snd_std).ok()?;
    snd.connect(("10.240.0.9", peer_port)).await.ok()?;
    Some(snd)
}

/// THE sender-block machinery, end to end (option 1 — one test, the
/// coherent behavior): the WouldBlock stash (`poll_write`'s
/// opportunistic drain stashes the mid-send datagram and registers
/// the send waker — `pump_queue_to_socket`'s Pending arms), the
/// accepted-byte high-water backpressure (`poll_write` Pends when the
/// queue stays over [`alktunnels::local::WRITE_QUEUE_HIGH_WATER`]),
/// the flush re-pump (readiness → drain), and the post-resolve
/// delivery (the queued datagrams reach the peer).
#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
async fn udp_half_write_side_blocks_stashes_and_drains_on_resolve() {
    let Some(mut veth) = VethFixture::try_create().await else {
        eprintln!("skipping: veth fixture unavailable (non-root host)");
        return;
    };

    // The peer port is EPHEMERAL (stale datagrams from earlier runs
    // of this test family target any fixed port; an ephemeral port
    // sidesteps the leftovers).
    let probe = tokio::net::UdpSocket::bind(("10.240.0.1", 0))
        .await
        .expect("port probe bind");
    let peer_port = probe.local_addr().expect("port probe addr").port();
    drop(probe);
    let snd = blocked_sender_socket(peer_port)
        .await
        .expect("blocked sender state");
    let (_read, mut write) = alktunnels::local::UdpHalf::split(snd);

    // Phase A: poll_write ACCEPTS everything (the accept-all
    // contract) while the socket blocks; the queued bytes cross the
    // high-water line and the next poll_write Pends.
    // Drive single-shot poll_write polls with a noop waker (the
    // pump's copy shape): each returns Ok (accepted) or Pending
    // (the high-water). The neighbor's INCOMPLETE→FAILED cycle
    // drops queued skbs periodically (charge released → sends
    // succeed again), so the drive loops until a Pend is observed.
    let mut body = vec![0xABu8; 60002];
    body[0] = 0; // len = 60000 BE
    body[1] = 0xea;
    body[2] = 0x60;

    let noop_waker = futures::task::noop_waker();
    let mut cx = std::task::Context::from_waker(&noop_waker);

    let mut pending_observed = false;
    for _ in 0..60 {
        let fut = std::future::poll_fn(|cx| {
            tokio::io::AsyncWrite::poll_write(Pin::new(&mut write), cx, &body)
        });
        let mut fut = std::pin::pin!(fut);
        match std::future::Future::poll(fut.as_mut(), &mut cx) {
            std::task::Poll::Ready(Ok(60002)) => {} // accepted
            std::task::Poll::Ready(Ok(n)) => panic!("poll_write accepted {n}?"),
            std::task::Poll::Ready(Err(e)) => panic!("poll_write error: {e:?}"),
            std::task::Poll::Pending => {
                pending_observed = true;
                break;
            }
        }
    }
    assert!(
        pending_observed,
        "the high-water Pends once the queue rides over the line"
    );

    // Flush: the blocked socket → the stash + readiness-Pending arms
    // (bounded retries; the resolve below unblocks the LAST one).
    let mut pending_flushes = 0usize;
    for _ in 0..10 {
        match tokio::time::timeout(
            std::time::Duration::from_millis(300),
            poll_flush_once(&mut write),
        )
        .await
        {
            Err(_elapsed) => {
                pending_flushes += 1;
            } // readiness stayed Pending (blocked socket)
            Ok(Ok(())) => break, // drained (only possible after resolve)
            Ok(Err(e)) => panic!("flush error: {e:?}"),
        }
    }
    assert!(pending_flushes > 0, "the blocked socket Pends the flush");

    // Phase B: resolve — assign the peer address (the neighbor
    // completes), bind the peer, wait for ARP; the queued datagrams
    // drain and DELIVER.
    assert!(veth.assign_peer(), "assign the peer address");
    let peer = tokio::net::UdpSocket::bind(("10.240.0.9", peer_port))
        .await
        .expect("bind peer (resolve)");

    let mut flushed = false;
    for _ in 0..20 {
        match tokio::time::timeout(
            std::time::Duration::from_millis(300),
            poll_flush_once(&mut write),
        )
        .await
        {
            Ok(Ok(())) => {
                flushed = true;
                break;
            }
            Ok(Err(e)) => panic!("post-resolve flush error: {e:?}"),
            Err(_) => {}
        }
    }
    assert!(flushed, "the queued datagrams drained after the resolve");

    // The peer received the queued datagrams (delivery across the
    // blocked→resolved window — the stash/queue never lost them; at
    // least the stashed tail must arrive — the periodic
    // INCOMPLETE→FAILED drops may consume some queued ones, which
    // is kernel-side loss, not adapter loss).
    tokio::time::sleep(std::time::Duration::from_millis(1500)).await;
    let mut got = 0usize;
    let mut buf = vec![0u8; 65535];
    while peer.try_recv(&mut buf).is_ok() {
        got += 1;
    }
    assert!(
        got >= 1,
        "the stashed/queued datagrams delivered after resolve"
    );

    drop(peer);
}