mirror of
https://github.com/openai/codex.git
synced 2026-04-26 15:45:02 +00:00
## Problem The first user turn can pay websocket handshake latency even when a session has already started. We want to reduce that initial delay while preserving turn semantics and avoiding any prompt send during startup. Reviewer feedback also called out duplicated connect/setup paths and unnecessary preconnect state complexity. ## Mental model `ModelClient` owns session-scoped transport state. During session startup, it can opportunistically warm one websocket handshake slot. A turn-scoped `ModelClientSession` adopts that slot once if available, restores captured sticky turn-state, and otherwise opens a websocket through the same shared connect path. If startup preconnect is still in flight, first turn setup awaits that task and treats it as the first connection attempt for the turn. Preconnect is handshake-only. The first `response.create` is still sent only when a turn starts. ## Non-goals This change does not make preconnect required for correctness and does not change prompt/turn payload semantics. It also does not expand fallback behavior beyond clearing preconnect state when fallback activates. ## Tradeoffs The implementation prioritizes simpler ownership and shared connection code over header-match gating for reuse. The single-slot cache keeps lifecycle straightforward but only benefits the immediate next turn. Awaiting in-flight preconnect has the same app-level connect-timeout semantics as existing websocket connect behavior (no new timeout class introduced by this PR). ## Architecture `core/src/client.rs`: - Added session-level preconnect lifecycle state (`Idle` / `InFlight` / `Ready`) carrying one warmed websocket plus optional captured turn-state. - Added `pre_establish_connection()` startup warmup and `preconnect()` handshake-only setup. - Deduped auth/provider resolution into `current_client_setup()` and websocket handshake wiring into `connect_websocket()` / `build_websocket_headers()`. - Updated turn websocket path to adopt preconnect first, await in-flight preconnect when present, then create a new websocket only when needed. - Ensured fallback activation clears warmed preconnect state. - Added documentation for lifecycle, ownership, sticky-routing invariants, and timeout semantics. `core/src/codex.rs`: - Session startup invokes `model_client.pre_establish_connection(...)`. - Turn metadata resolution uses the shared timeout helper. `core/src/turn_metadata.rs`: - Centralized shared timeout helper used by both turn-time metadata resolution and startup preconnect metadata building. `core/tests/common/responses.rs` + websocket test suites: - Added deterministic handshake waiting helper (`wait_for_handshakes`) with bounded polling. - Added startup preconnect and in-flight preconnect reuse coverage. - Fallback expectations now assert exactly two websocket attempts in covered scenarios (startup preconnect + turn attempt before fallback sticks). ## Observability Preconnect remains best-effort and non-fatal. Existing websocket/fallback telemetry remains in place, and debug logs now make preconnect-await behavior and preconnect failures easier to reason about. ## Tests Validated with: 1. `just fmt` 2. `cargo test -p codex-core websocket_preconnect -- --nocapture` 3. `cargo test -p codex-core websocket_fallback -- --nocapture` 4. `cargo test -p codex-core websocket_first_turn_waits_for_inflight_preconnect -- --nocapture`
126 lines
4.5 KiB
Rust
126 lines
4.5 KiB
Rust
#![allow(clippy::expect_used, clippy::unwrap_used)]
|
|
|
|
use anyhow::Result;
|
|
use core_test_support::responses::WebSocketConnectionConfig;
|
|
use core_test_support::responses::ev_assistant_message;
|
|
use core_test_support::responses::ev_completed;
|
|
use core_test_support::responses::ev_done;
|
|
use core_test_support::responses::ev_reasoning_item;
|
|
use core_test_support::responses::ev_response_created;
|
|
use core_test_support::responses::ev_shell_command_call;
|
|
use core_test_support::responses::mount_response_sequence;
|
|
use core_test_support::responses::sse;
|
|
use core_test_support::responses::sse_response;
|
|
use core_test_support::responses::start_mock_server;
|
|
use core_test_support::responses::start_websocket_server_with_headers;
|
|
use core_test_support::skip_if_no_network;
|
|
use core_test_support::test_codex::test_codex;
|
|
use pretty_assertions::assert_eq;
|
|
|
|
const TURN_STATE_HEADER: &str = "x-codex-turn-state";
|
|
|
|
#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
|
|
async fn responses_turn_state_persists_within_turn_and_resets_after() -> Result<()> {
|
|
skip_if_no_network!(Ok(()));
|
|
|
|
let server = start_mock_server().await;
|
|
let call_id = "shell-turn-state";
|
|
|
|
let first_response = sse(vec![
|
|
ev_response_created("resp-1"),
|
|
ev_reasoning_item("rsn-1", &["thinking"], &[]),
|
|
ev_shell_command_call(call_id, "echo turn-state"),
|
|
ev_completed("resp-1"),
|
|
]);
|
|
let second_response = sse(vec![
|
|
ev_response_created("resp-2"),
|
|
ev_assistant_message("msg-1", "done"),
|
|
ev_completed("resp-2"),
|
|
]);
|
|
let third_response = sse(vec![
|
|
ev_response_created("resp-3"),
|
|
ev_assistant_message("msg-2", "done"),
|
|
ev_completed("resp-3"),
|
|
]);
|
|
|
|
// First response sets turn_state; follow-up request in the same turn should echo it.
|
|
let responses = vec![
|
|
sse_response(first_response).insert_header(TURN_STATE_HEADER, "ts-1"),
|
|
sse_response(second_response),
|
|
sse_response(third_response),
|
|
];
|
|
let request_log = mount_response_sequence(&server, responses).await;
|
|
|
|
let test = test_codex().build(&server).await?;
|
|
test.submit_turn("run a shell command").await?;
|
|
test.submit_turn("second turn").await?;
|
|
|
|
let requests = request_log.requests();
|
|
assert_eq!(requests.len(), 3);
|
|
// Initial turn request has no header; follow-up has it; next turn clears it.
|
|
assert_eq!(requests[0].header(TURN_STATE_HEADER), None);
|
|
assert_eq!(
|
|
requests[1].header(TURN_STATE_HEADER),
|
|
Some("ts-1".to_string())
|
|
);
|
|
assert_eq!(requests[2].header(TURN_STATE_HEADER), None);
|
|
|
|
Ok(())
|
|
}
|
|
|
|
#[tokio::test(flavor = "multi_thread", worker_threads = 2)]
|
|
async fn websocket_turn_state_persists_within_turn_and_resets_after() -> Result<()> {
|
|
skip_if_no_network!(Ok(()));
|
|
|
|
let call_id = "ws-shell-turn-state";
|
|
// First connection delivers turn_state; second (same turn) must send it; third (new turn) must not.
|
|
let server = start_websocket_server_with_headers(vec![
|
|
WebSocketConnectionConfig {
|
|
requests: vec![vec![
|
|
ev_response_created("resp-1"),
|
|
ev_reasoning_item("rsn-1", &["thinking"], &[]),
|
|
ev_shell_command_call(call_id, "echo websocket"),
|
|
ev_done(),
|
|
]],
|
|
response_headers: vec![(TURN_STATE_HEADER.to_string(), "ts-1".to_string())],
|
|
accept_delay: None,
|
|
},
|
|
WebSocketConnectionConfig {
|
|
requests: vec![vec![
|
|
ev_response_created("resp-2"),
|
|
ev_assistant_message("msg-1", "done"),
|
|
ev_completed("resp-2"),
|
|
]],
|
|
response_headers: Vec::new(),
|
|
accept_delay: None,
|
|
},
|
|
WebSocketConnectionConfig {
|
|
requests: vec![vec![
|
|
ev_response_created("resp-3"),
|
|
ev_assistant_message("msg-2", "done"),
|
|
ev_completed("resp-3"),
|
|
]],
|
|
response_headers: Vec::new(),
|
|
accept_delay: None,
|
|
},
|
|
])
|
|
.await;
|
|
|
|
let mut builder = test_codex();
|
|
let test = builder.build_with_websocket_server(&server).await?;
|
|
test.submit_turn("run the echo command").await?;
|
|
test.submit_turn("second turn").await?;
|
|
|
|
let handshakes = server.handshakes();
|
|
assert_eq!(handshakes.len(), 3);
|
|
assert_eq!(handshakes[0].header(TURN_STATE_HEADER), None);
|
|
assert_eq!(
|
|
handshakes[1].header(TURN_STATE_HEADER),
|
|
Some("ts-1".to_string())
|
|
);
|
|
assert_eq!(handshakes[2].header(TURN_STATE_HEADER), None);
|
|
|
|
server.shutdown().await;
|
|
Ok(())
|
|
}
|