Defer persistence of rollout file (#11028)

- Defer rollout persistence for fresh threads (`InitialHistory::New`):
keep rollout events in memory and only materialize rollout file + state
DB row on first `EventMsg::UserMessage`.
- Keep precomputed rollout path available before materialization.
- Change `thread/start` to build thread response from live config
snapshot and optional precomputed path.
- Improve pre-materialization behavior in app-server/TUI: clearer
invalid-request errors for file-backed ops and a friendlier `/fork` “not
ready yet” UX.
- Update tests to match deferred semantics across
start/read/archive/unarchive/fork/resume/review flows.
- Improved resilience of user_shell test, which should be unrelated to
this change but must be affected by timing changes

For Reviewers:
* The primary change is in recorder.rs
* Most of the other changes were to fix up broken assumptions in
existing tests

Testing:
* Manually tested CLI
* Exercised app server paths by manually running IDE Extension with
rebuilt CLI binary
* Only user-visible change is that `/fork` in TUI generates visible
error if used prior to first turn
This commit is contained in:
Eric Traut
2026-02-07 23:05:03 -08:00
committed by GitHub
parent 6d08298f4e
commit b3de6c7f2b
19 changed files with 983 additions and 195 deletions

View File

@@ -1,12 +1,17 @@
use anyhow::Result;
use app_test_support::McpProcess;
use app_test_support::create_mock_responses_server_repeating_assistant;
use app_test_support::to_response;
use codex_app_server_protocol::JSONRPCError;
use codex_app_server_protocol::JSONRPCResponse;
use codex_app_server_protocol::RequestId;
use codex_app_server_protocol::ThreadArchiveParams;
use codex_app_server_protocol::ThreadArchiveResponse;
use codex_app_server_protocol::ThreadStartParams;
use codex_app_server_protocol::ThreadStartResponse;
use codex_app_server_protocol::TurnStartParams;
use codex_app_server_protocol::TurnStartResponse;
use codex_app_server_protocol::UserInput;
use codex_core::ARCHIVED_SESSIONS_SUBDIR;
use codex_core::find_thread_path_by_id_str;
use std::path::Path;
@@ -16,9 +21,10 @@ use tokio::time::timeout;
const DEFAULT_READ_TIMEOUT: std::time::Duration = std::time::Duration::from_secs(10);
#[tokio::test]
async fn thread_archive_moves_rollout_into_archived_directory() -> Result<()> {
async fn thread_archive_requires_materialized_rollout() -> Result<()> {
let server = create_mock_responses_server_repeating_assistant("Done").await;
let codex_home = TempDir::new()?;
create_config_toml(codex_home.path())?;
create_config_toml(codex_home.path(), &server.uri())?;
let mut mcp = McpProcess::new(codex_home.path()).await?;
timeout(DEFAULT_READ_TIMEOUT, mcp.initialize()).await??;
@@ -38,17 +44,73 @@ async fn thread_archive_moves_rollout_into_archived_directory() -> Result<()> {
let ThreadStartResponse { thread, .. } = to_response::<ThreadStartResponse>(start_resp)?;
assert!(!thread.id.is_empty());
// Locate the rollout path recorded for this thread id.
let rollout_path = find_thread_path_by_id_str(codex_home.path(), &thread.id)
.await?
.expect("expected rollout path for thread id to exist");
let rollout_path = thread.path.clone().expect("thread path");
assert!(
!rollout_path.exists(),
"fresh thread rollout should not exist yet at {}",
rollout_path.display()
);
assert!(
find_thread_path_by_id_str(codex_home.path(), &thread.id)
.await?
.is_none(),
"thread id should not be discoverable before rollout materialization"
);
// Archive should fail before the rollout is materialized.
let archive_id = mcp
.send_thread_archive_request(ThreadArchiveParams {
thread_id: thread.id.clone(),
})
.await?;
let archive_err: JSONRPCError = timeout(
DEFAULT_READ_TIMEOUT,
mcp.read_stream_until_error_message(RequestId::Integer(archive_id)),
)
.await??;
assert!(
archive_err
.error
.message
.contains("no rollout found for thread id"),
"unexpected archive error: {}",
archive_err.error.message
);
// Materialize rollout via a real user turn and confirm archive succeeds.
let turn_start_id = mcp
.send_turn_start_request(TurnStartParams {
thread_id: thread.id.clone(),
input: vec![UserInput::Text {
text: "materialize".to_string(),
text_elements: Vec::new(),
}],
..Default::default()
})
.await?;
let turn_start_response: JSONRPCResponse = timeout(
DEFAULT_READ_TIMEOUT,
mcp.read_stream_until_response_message(RequestId::Integer(turn_start_id)),
)
.await??;
let _: TurnStartResponse = to_response::<TurnStartResponse>(turn_start_response)?;
timeout(
DEFAULT_READ_TIMEOUT,
mcp.read_stream_until_notification_message("turn/completed"),
)
.await??;
assert!(
rollout_path.exists(),
"expected {} to exist",
"expected rollout path {} to exist after first user message",
rollout_path.display()
);
// Archive the thread.
let discovered_path = find_thread_path_by_id_str(codex_home.path(), &thread.id)
.await?
.expect("expected rollout path for thread id to exist after materialization");
assert_paths_match_on_disk(&discovered_path, &rollout_path)?;
let archive_id = mcp
.send_thread_archive_request(ThreadArchiveParams {
thread_id: thread.id.clone(),
@@ -80,14 +142,32 @@ async fn thread_archive_moves_rollout_into_archived_directory() -> Result<()> {
Ok(())
}
fn create_config_toml(codex_home: &Path) -> std::io::Result<()> {
fn create_config_toml(codex_home: &Path, server_uri: &str) -> std::io::Result<()> {
let config_toml = codex_home.join("config.toml");
std::fs::write(config_toml, config_contents())
std::fs::write(config_toml, config_contents(server_uri))
}
fn config_contents() -> &'static str {
r#"model = "mock-model"
fn config_contents(server_uri: &str) -> String {
format!(
r#"model = "mock-model"
approval_policy = "never"
sandbox_mode = "read-only"
model_provider = "mock_provider"
[model_providers.mock_provider]
name = "Mock provider for test"
base_url = "{server_uri}/v1"
wire_api = "responses"
request_max_retries = 0
stream_max_retries = 0
"#
)
}
fn assert_paths_match_on_disk(actual: &Path, expected: &Path) -> std::io::Result<()> {
let actual = actual.canonicalize()?;
let expected = expected.canonicalize()?;
assert_eq!(actual, expected);
Ok(())
}