Hare/yoi
1
1
Fork 0
Code Actions Packages Releases Activity

refactor: require annotated session log history

Browse Source
This commit is contained in:
Hare
2026-08-30 12:18:44 +09:00
parent 4ec56fe41e
commit 8493472983
28 changed files with 872 additions and 716 deletions
Download Patch File Download Diff File Expand all files Collapse all files
crates/session-store/src/history.rs
+10 -45
View File
@@ -1,9 +1,8 @@
//! Serializable history entries with restore-authoritative logical identity and origin.
use base64::{Engine as _, engine::general_purpose::URL_SAFE_NO_PAD};
use serde::{Deserialize, Serialize};
use crate::{LoggedItem, SessionId};
use crate::LoggedItem;
/// Stable logical identity of one model-visible history entry.
///
@@ -143,12 +142,15 @@ mod tests {
#[test]
fn annotated_segment_start_is_restore_visible_without_projecting_metadata() {
let session_id = uuid::Uuid::now_v7();
let history_entry = legacy_logged_history(LoggedItem::Message {
role: LoggedRole::Assistant,
content: vec![crate::LoggedContentPart::Text {
text: "answer".into(),
}],
});
let history_entry = LoggedHistoryEntry {
item: LoggedItem::Message {
role: LoggedRole::Assistant,
content: vec![crate::LoggedContentPart::Text {
text: "answer".into(),
}],
},
metadata: LoggedSessionHistoryMetadata::legacy_unknown(),
};
let state = crate::collect_state(&[crate::LogEntry::AnnotatedSegmentStart {
ts: 1,
session_id,
@@ -161,40 +163,3 @@ mod tests {
assert_eq!(state.history[0].as_text(), Some("answer"));
}
}
/// Legacy Session Logs did not persist annotations. Decode helpers explicitly
/// create `LegacyUnknown`; they never infer Human/System authority from role or
/// plaintext.
pub fn legacy_logged_history(item: LoggedItem) -> LoggedHistoryEntry {
LoggedHistoryEntry {
item,
metadata: LoggedSessionHistoryMetadata::legacy_unknown(),
}
}
pub fn legacy_segment_history(
session_id: SessionId,
items: impl IntoIterator<Item = LoggedItem>,
) -> Vec<LoggedHistoryEntry> {
items
.into_iter()
.enumerate()
.map(|(index, item)| LoggedHistoryEntry {
item,
metadata: LoggedSessionHistoryMetadata {
// Legacy logs have no persisted entry id. Derive one solely from
// durable segment content rather than minting a new random value
// on every restore/read. The explicit LegacyUnknown origin keeps
// this compatibility identity from becoming trust authority.
entry_id: {
let mut identity = Vec::with_capacity(24);
identity.extend_from_slice(session_id.as_bytes());
identity.extend_from_slice(&(index as u64).to_be_bytes());
LoggedSessionHistoryEntryId(format!("l-{}", URL_SAFE_NO_PAD.encode(identity)))
},
origin: LoggedSessionHistoryOrigin::LegacyUnknown,
derivation: None,
},
})
.collect()
}
crates/session-store/src/legacy_session_log.rs
+188
View File
@@ -0,0 +1,188 @@
//! Versioned decoder for Session schemas that predate canonical annotated history.
//!
//! These types are intentionally private to `session-store`. Current writers,
//! replay, and public projections use [`crate::LogEntry`] exclusively; only the
//! Worker Session schema migration is allowed to deserialize these shapes.
use agen::llm_client::types::RequestConfig;
use protocol::Segment;
use serde::Deserialize;
use crate::{
LogEntry, LoggedHistoryEntry, LoggedItem, LoggedSessionHistoryEntryId,
LoggedSessionHistoryMetadata, LoggedSessionHistoryOrigin, LoggedSystemHistoryEntry, SegmentId,
SegmentOrigin, SessionExtension, SessionId, SystemItem,
};
#[derive(Debug, Deserialize)]
#[serde(tag = "kind", rename_all = "snake_case")]
enum LegacyHistoryLogEntry {
SegmentStart {
ts: u64,
session_id: SessionId,
system_prompt: Option<String>,
config: RequestConfig,
history: Vec<LoggedItem>,
#[serde(default)]
forked_from: Option<SegmentOrigin>,
#[serde(default)]
compacted_from: Option<SegmentOrigin>,
},
UserInput {
ts: u64,
segments: Vec<Segment>,
#[serde(default)]
extensions: Vec<SessionExtension>,
},
AssistantItem {
ts: u64,
item: LoggedItem,
},
ToolResult {
ts: u64,
item: LoggedItem,
},
SystemItem {
ts: u64,
item: SystemItem,
},
}
/// Schema-v1 decoder. Non-history records already had their current shape, so
/// they pass through `LogEntry`; legacy history records are converted below.
#[derive(Debug, Deserialize)]
#[serde(untagged)]
enum LegacySessionLogEntryV1 {
History(LegacyHistoryLogEntry),
Current(LogEntry),
}
/// Schema v2 retained the v1 history shapes while adding non-history records.
/// Keep a distinct type so supported source versions remain explicit rather
/// than turning migration compatibility into the current `LogEntry` contract.
#[derive(Debug, Deserialize)]
#[serde(untagged)]
enum LegacySessionLogEntryV2 {
History(LegacyHistoryLogEntry),
Current(LogEntry),
}
pub(crate) fn decode_entry(
schema_version: u32,
line: &str,
session_id: SessionId,
segment_id: SegmentId,
line_index: usize,
) -> Result<LogEntry, serde_json::Error> {
let entry = match schema_version {
1 => match serde_json::from_str::<LegacySessionLogEntryV1>(line)? {
LegacySessionLogEntryV1::History(entry) => Entry::History(entry),
LegacySessionLogEntryV1::Current(entry) => Entry::Current(entry),
},
2 => match serde_json::from_str::<LegacySessionLogEntryV2>(line)? {
LegacySessionLogEntryV2::History(entry) => Entry::History(entry),
LegacySessionLogEntryV2::Current(entry) => Entry::Current(entry),
},
_ => unreachable!("legacy decoder called for unsupported schema {schema_version}"),
};
Ok(match entry {
Entry::History(entry) => {
canonicalize_history_entry(session_id, segment_id, line_index, entry)
}
Entry::Current(entry) => entry,
})
}
enum Entry {
History(LegacyHistoryLogEntry),
Current(LogEntry),
}
fn legacy_metadata(
segment_id: SegmentId,
line_index: usize,
item_index: usize,
) -> LoggedSessionHistoryMetadata {
let mut identity = Vec::with_capacity(32);
identity.extend_from_slice(segment_id.as_bytes());
identity.extend_from_slice(&(line_index as u64).to_be_bytes());
identity.extend_from_slice(&(item_index as u64).to_be_bytes());
LoggedSessionHistoryMetadata {
entry_id: LoggedSessionHistoryEntryId(format!(
"l-{}",
base64::Engine::encode(&base64::engine::general_purpose::URL_SAFE_NO_PAD, identity)
)),
origin: LoggedSessionHistoryOrigin::LegacyUnknown,
derivation: None,
}
}
fn canonicalize_history_entry(
_session_id: SessionId,
segment_id: SegmentId,
line_index: usize,
entry: LegacyHistoryLogEntry,
) -> LogEntry {
match entry {
LegacyHistoryLogEntry::SegmentStart {
ts,
session_id,
system_prompt,
config,
history,
forked_from,
compacted_from,
} => LogEntry::AnnotatedSegmentStart {
ts,
session_id,
system_prompt,
config,
history: history
.into_iter()
.enumerate()
.map(|(item_index, item)| LoggedHistoryEntry {
item,
metadata: legacy_metadata(segment_id, line_index, item_index),
})
.collect(),
forked_from,
compacted_from,
},
LegacyHistoryLogEntry::UserInput {
ts,
segments,
extensions,
} => LogEntry::AnnotatedUserInput {
ts,
history: vec![LoggedHistoryEntry {
item: LoggedItem::from(agen::Item::user_message(Segment::flatten_to_text(
&segments,
))),
metadata: legacy_metadata(segment_id, line_index, 0),
}],
segments,
extensions,
},
LegacyHistoryLogEntry::AssistantItem { ts, item } => LogEntry::AnnotatedAssistantItem {
ts,
entry: LoggedHistoryEntry {
item,
metadata: legacy_metadata(segment_id, line_index, 0),
},
},
LegacyHistoryLogEntry::ToolResult { ts, item } => LogEntry::AnnotatedToolResult {
ts,
entry: LoggedHistoryEntry {
item,
metadata: legacy_metadata(segment_id, line_index, 0),
},
},
LegacyHistoryLogEntry::SystemItem { ts, item } => LogEntry::AnnotatedSystemItem {
ts,
entry: LoggedSystemHistoryEntry {
item,
metadata: legacy_metadata(segment_id, line_index, 0),
},
},
}
}
crates/session-store/src/lib.rs
+4 -3
View File
@@ -26,13 +26,14 @@
//! let (session_id, segment_id) = create_segment(&store, SegmentStartState {
//! system_prompt: None,
//! config: &config,
//! history: &[],
//! history: Vec::new(),
//! })?;
//! ```
pub mod event_trace;
pub mod fs_store;
pub mod history;
mod legacy_session_log;
pub mod logged_item;
pub mod public_snapshot;
pub mod segment;
@@ -49,11 +50,11 @@ pub use fs_store::FsStore;
pub use history::{
LoggedHistoryDerivation, LoggedHistoryEntry, LoggedSessionHistoryEntryId,
LoggedSessionHistoryMetadata, LoggedSessionHistoryOrigin, LoggedSystemHistoryEntry,
LoggedWorkerSubject, legacy_logged_history, legacy_segment_history,
LoggedWorkerSubject,
};
pub use logged_item::{LoggedContentPart, LoggedItem, LoggedRole, from_logged, to_logged};
pub use segment::{
SegmentStartState, append_entry, append_system_item, classify_history_item,
SegmentStartState, append_entry, append_system_item, classify_logged_history_entry,
create_compacted_segment, create_segment, create_segment_with_ids, ensure_head_or_fork, fork,
fork_at, restore, restore_by_segment, save_config_changed, save_delta, save_extension,
save_run_completed, save_run_errored, save_turn_end, save_usage, save_user_input,
crates/session-store/src/public_snapshot.rs
+17 -43
View File
@@ -12,13 +12,12 @@ use crate::{
LoggedSessionHistoryOrigin, SessionId, SystemItem,
};
/// Project a complete current-segment log. A valid segment always starts with
/// one of the two SegmentStart records; malformed partial input uses the nil
/// session only to keep the public failure projection deterministic.
/// Project a complete current-segment log. A valid segment starts with one
/// canonical annotated SegmentStart record; malformed partial input uses the
/// nil session only to keep the public failure projection deterministic.
pub fn project_current_session_snapshot(log: &[LogEntry]) -> SessionSnapshot {
let session_id = log.iter().find_map(|entry| match entry {
LogEntry::SegmentStart { session_id, .. }
| LogEntry::AnnotatedSegmentStart { session_id, .. } => Some(*session_id),
LogEntry::AnnotatedSegmentStart { session_id, .. } => Some(*session_id),
_ => None,
});
project_session_snapshot(session_id.unwrap_or_else(SessionId::nil), log)
@@ -32,20 +31,6 @@ pub fn project_session_snapshot(session_id: SessionId, log: &[LogEntry]) -> Sess
for (log_index, record) in log.iter().enumerate() {
match record {
LogEntry::SegmentStart {
ts,
session_id,
history,
..
} => {
session_key = *session_id;
entries.clear();
for (item_index, item) in history.iter().enumerate() {
if let Some(data) = project_item(item) {
entries.push(legacy_entry(&session_key, log_index, item_index, *ts, data));
}
}
}
LogEntry::AnnotatedSegmentStart {
ts,
session_id,
@@ -56,39 +41,18 @@ pub fn project_session_snapshot(session_id: SessionId, log: &[LogEntry]) -> Sess
entries.clear();
extend_history(&mut entries, history, None, *ts);
}
LogEntry::UserInput { ts, segments, .. } => entries.push(legacy_entry(
&session_key,
log_index,
0,
*ts,
SessionSnapshotEntryData::UserInput {
segments: segments.clone(),
},
)),
LogEntry::AnnotatedUserInput {
ts,
segments,
history,
..
} => extend_history(&mut entries, history, Some(segments), *ts),
LogEntry::AssistantItem { ts, item } | LogEntry::ToolResult { ts, item } => {
if let Some(data) = project_item(item) {
entries.push(legacy_entry(&session_key, log_index, 0, *ts, data));
}
}
LogEntry::AnnotatedAssistantItem { ts, entry }
| LogEntry::AnnotatedToolResult { ts, entry } => {
if let Some(data) = project_item(&entry.item) {
entries.push(history_entry(entry, *ts, data));
}
}
LogEntry::SystemItem { ts, item } => entries.push(system_entry(
item,
legacy_entry_id(&session_key, log_index, 0),
*ts,
SessionEntryProvenance::LegacyUnknown,
Vec::new(),
)),
LogEntry::AnnotatedSystemItem { ts, entry } => entries.push(system_entry(
&entry.item,
entry.metadata.entry_id.0.clone(),
@@ -332,9 +296,9 @@ mod tests {
use crate::{LoggedSessionHistoryEntryId, LoggedSessionHistoryMetadata, LoggedWorkerSubject};
#[test]
fn legacy_projection_is_stable_and_hides_reasoning_and_system_prompts() {
fn current_projection_is_stable_and_hides_reasoning_and_system_prompts() {
let session_id = crate::new_session_id();
let log = vec![LogEntry::SegmentStart {
let log = vec![LogEntry::AnnotatedSegmentStart {
ts: 1,
session_id,
system_prompt: None,
@@ -358,7 +322,17 @@ mod tests {
text: "visible".into(),
}],
},
],
]
.into_iter()
.map(|item| LoggedHistoryEntry {
item,
metadata: LoggedSessionHistoryMetadata {
entry_id: LoggedSessionHistoryEntryId::new(),
origin: LoggedSessionHistoryOrigin::LegacyUnknown,
derivation: None,
},
})
.collect(),
forked_from: None,
compacted_from: None,
}];
crates/session-store/src/segment.rs
+39 -45
View File
@@ -4,11 +4,9 @@
//! The caller (typically Worker) holds the Engine directly and calls these
//! functions after state-mutating operations.
use crate::logged_item::{LoggedItem, to_logged};
use crate::segment_log::{self, LogEntry, SegmentOrigin};
use crate::store::{Store, StoreError};
use crate::system_item::SystemItem;
use crate::{SegmentId, SessionId};
use crate::{LoggedHistoryEntry, LoggedSystemHistoryEntry, SegmentId, SessionId};
use agen::EngineResult;
use agen::llm_client::RequestConfig;
use agen::llm_client::types::Item;
@@ -18,7 +16,7 @@ use protocol::Segment;
pub struct SegmentStartState<'a> {
pub system_prompt: Option<&'a str>,
pub config: &'a RequestConfig,
pub history: &'a [Item],
pub history: Vec<LoggedHistoryEntry>,
}
/// Create a new session + initial segment, writing the initial
@@ -44,12 +42,12 @@ pub fn create_segment_with_ids(
segment_id: SegmentId,
state: SegmentStartState<'_>,
) -> Result<(), StoreError> {
let entry = LogEntry::SegmentStart {
let entry = LogEntry::AnnotatedSegmentStart {
ts: segment_log::now_millis(),
session_id,
system_prompt: state.system_prompt.map(String::from),
config: state.config.clone(),
history: to_logged(state.history),
history: state.history.to_vec(),
forked_from: None,
compacted_from: None,
};
@@ -70,12 +68,12 @@ pub fn create_compacted_segment(
source_turn_count: usize,
) -> Result<SegmentId, StoreError> {
let segment_id = crate::new_segment_id();
let entry = LogEntry::SegmentStart {
let entry = LogEntry::AnnotatedSegmentStart {
ts: segment_log::now_millis(),
session_id: source_session_id,
system_prompt: state.system_prompt.map(String::from),
config: state.config.clone(),
history: to_logged(state.history),
history: state.history.to_vec(),
forked_from: None,
compacted_from: Some(SegmentOrigin {
segment_id: source_segment_id,
@@ -154,12 +152,12 @@ pub fn ensure_head_or_fork(
}
let source_segment_id = *segment_id;
let fork_id = crate::new_segment_id();
let entry = LogEntry::SegmentStart {
let entry = LogEntry::AnnotatedSegmentStart {
ts: segment_log::now_millis(),
session_id,
system_prompt: state.system_prompt.map(String::from),
config: state.config.clone(),
history: to_logged(state.history),
history: state.history.to_vec(),
forked_from: Some(SegmentOrigin {
segment_id: source_segment_id,
at_turn_index,
@@ -183,8 +181,9 @@ pub fn save_user_input(
session_id: SessionId,
segment_id: SegmentId,
segments: Vec<Segment>,
history: Vec<LoggedHistoryEntry>,
) -> Result<(), StoreError> {
save_user_input_with_extensions(store, session_id, segment_id, segments, Vec::new())
save_user_input_with_extensions(store, session_id, segment_id, segments, history, Vec::new())
}
/// Atomically persist one typed user submission and Runtime-owned session
@@ -194,15 +193,17 @@ pub fn save_user_input_with_extensions(
session_id: SessionId,
segment_id: SegmentId,
segments: Vec<Segment>,
history: Vec<LoggedHistoryEntry>,
extensions: Vec<segment_log::SessionExtension>,
) -> Result<(), StoreError> {
append_entry(
store,
session_id,
segment_id,
LogEntry::UserInput {
LogEntry::AnnotatedUserInput {
ts: segment_log::now_millis(),
segments,
history,
extensions,
},
)
@@ -220,64 +221,57 @@ pub fn save_delta(
store: &impl Store,
session_id: SessionId,
segment_id: SegmentId,
new_items: &[Item],
new_items: &[LoggedHistoryEntry],
) -> Result<(), StoreError> {
if new_items.is_empty() {
return Ok(());
}
let ts = segment_log::now_millis();
for item in new_items {
for entry in new_items {
let item = Item::from(entry.item.clone());
if item.is_user_message() {
// Already persisted by save_user_input at submit time.
continue;
}
let entry = classify_history_item(item, ts);
let entry = classify_logged_history_entry(entry.clone(), ts);
append_entry(store, session_id, segment_id, entry)?;
}
Ok(())
}
/// Map one history item to its singular `LogEntry` form. Used by the
/// fallback `save_delta` path and the controller's worker-callback
/// classifier so write classification lives in one place.
pub fn classify_history_item(item: &Item, ts: u64) -> LogEntry {
/// Map one annotated history entry to its singular `LogEntry` form. Used by
/// the fallback `save_delta` path and the controller's worker-callback
/// classifier so write classification lives in one place without discarding
/// identity or provenance.
/// Map one already-annotated history entry to its singular canonical record
/// without changing its identity or provenance.
pub fn classify_logged_history_entry(entry: LoggedHistoryEntry, ts: u64) -> LogEntry {
let item = Item::from(entry.item.clone());
if item.is_tool_result() {
LogEntry::ToolResult {
ts,
item: LoggedItem::from(item),
}
} else if item.is_assistant_message() || item.is_tool_call() || item.is_reasoning() {
LogEntry::AssistantItem {
ts,
item: LoggedItem::from(item),
}
LogEntry::AnnotatedToolResult { ts, entry }
} else {
// Defensive: anything else (future Item kinds) routes through
// AssistantItem rather than getting silently dropped.
LogEntry::AssistantItem {
ts,
item: LoggedItem::from(item),
}
// Assistant messages, tool calls, reasoning, and future non-user
// items all use the assistant-side canonical record.
LogEntry::AnnotatedAssistantItem { ts, entry }
}
}
/// Append a single typed system item as `LogEntry::SystemItem`. Helper
/// for the Worker-side interceptor commit path; mirrors the per-item
/// commit shape used for assistant / tool result entries.
/// Append one typed system item and its history metadata as a canonical
/// `LogEntry::AnnotatedSystemItem`.
pub fn append_system_item(
store: &impl Store,
session_id: SessionId,
segment_id: SegmentId,
item: SystemItem,
entry: LoggedSystemHistoryEntry,
) -> Result<(), StoreError> {
append_entry(
store,
session_id,
segment_id,
LogEntry::SystemItem {
LogEntry::AnnotatedSystemItem {
ts: segment_log::now_millis(),
item,
entry,
},
)
}
@@ -430,12 +424,12 @@ pub fn fork(
) -> Result<(SessionId, SegmentId), StoreError> {
let session_id = crate::new_session_id();
let fork_id = crate::new_segment_id();
let entry = LogEntry::SegmentStart {
let entry = LogEntry::AnnotatedSegmentStart {
ts: segment_log::now_millis(),
session_id,
system_prompt: state.system_prompt.map(String::from),
config: state.config.clone(),
history: to_logged(state.history),
history: state.history.to_vec(),
forked_from: None,
compacted_from: None,
};
@@ -470,7 +464,7 @@ pub fn fork_at(
// segment), before any turn completes.
entries
.iter()
.position(|e| !matches!(e, LogEntry::SegmentStart { .. }))
.position(|e| !matches!(e, LogEntry::AnnotatedSegmentStart { .. }))
.unwrap_or(entries.len())
} else {
entries
@@ -482,12 +476,12 @@ pub fn fork_at(
let state = segment_log::collect_state(&entries[..cut]);
let fork_id = crate::new_segment_id();
let entry = LogEntry::SegmentStart {
let entry = LogEntry::AnnotatedSegmentStart {
ts: segment_log::now_millis(),
session_id: source_session_id,
system_prompt: state.system_prompt,
config: state.config,
history: to_logged(&state.history),
history: state.annotated_history,
forked_from: Some(SegmentOrigin {
segment_id: source_id,
at_turn_index,
crates/session-store/src/segment_log.rs
+86 -133
View File
@@ -16,7 +16,6 @@ use serde::{Deserialize, Serialize};
use crate::history::{LoggedHistoryEntry, LoggedSystemHistoryEntry};
use crate::logged_item::LoggedItem;
use crate::system_item::SystemItem;
/// A single segment log entry, serialized as one JSONL line.
///
@@ -50,28 +49,7 @@ impl SessionExtension {
#[derive(Debug, Clone, Serialize, Deserialize)]
#[serde(tag = "kind", rename_all = "snake_case")]
pub enum LogEntry {
/// Segment start. Always the first entry in a segment log.
/// For forked segments, `history` contains the seed state from the parent.
SegmentStart {
ts: u64,
/// Session this segment belongs to. Compaction / fork inherits
/// the source segment's session_id; only fresh "new conversation"
/// segments mint a new session_id.
session_id: crate::SessionId,
system_prompt: Option<String>,
config: RequestConfig,
history: Vec<LoggedItem>,
/// Origin: forked from a sibling segment at a specific turn boundary.
/// The referenced segment is guaranteed to share `session_id`.
#[serde(default, skip_serializing_if = "Option::is_none")]
forked_from: Option<SegmentOrigin>,
/// Origin: compacted from a sibling segment at a specific turn boundary.
/// The referenced segment is guaranteed to share `session_id`.
#[serde(default, skip_serializing_if = "Option::is_none")]
compacted_from: Option<SegmentOrigin>,
},
/// Schema-v2 segment seed. Retained entries keep their stable logical
/// Canonical segment seed. Retained entries keep their stable logical
/// identity and origin across fork/compaction/restore.
AnnotatedSegmentStart {
ts: u64,
@@ -105,22 +83,7 @@ pub enum LogEntry {
/// restore conservatively instead of re-running a dangling tool call.
Invoke { ts: u64, trigger: InvokeKind },
/// User input accepted at submit time. Carries the original typed
/// `Vec<Segment>` so clients can re-render typed atoms (paste chips,
/// file refs) on segment restore.
/// Replay flattens these into a `Item::user_message` for the worker
/// history; the worker layer never sees segments directly.
UserInput {
ts: u64,
segments: Vec<Segment>,
/// Typed durable state committed atomically with this input record.
/// Runtime-owned Flow invocation uses this to avoid a Backend-instance
/// commit that can get ahead of Worker history.
#[serde(default, skip_serializing_if = "Vec::is_empty")]
extensions: Vec<SessionExtension>,
},
/// Schema-v2 user submission with its exact model-visible entries. Typed
/// Canonical user submission with its exact model-visible entries. Typed
/// Flow instructions and caller-attributed input remain separate entries.
AnnotatedUserInput {
ts: u64,
@@ -130,35 +93,19 @@ pub enum LogEntry {
history: Vec<LoggedHistoryEntry>,
},
/// Schema-v2 model output and metadata committed as one journal record.
/// Canonical model output and metadata committed as one journal record.
AnnotatedAssistantItem { ts: u64, entry: LoggedHistoryEntry },
/// One assistant-side item appended to history — assistant message,
/// reasoning, or tool call. Singular: one entry per history item so
/// the wire-side `Event::*` lane and on-disk LogEntry stay 1:1.
AssistantItem { ts: u64, item: LoggedItem },
/// Schema-v2 tool output and metadata committed as one journal record.
/// Canonical tool output and metadata committed as one journal record.
AnnotatedToolResult { ts: u64, entry: LoggedHistoryEntry },
/// One tool-execution result appended to history.
ToolResult { ts: u64, item: LoggedItem },
/// Schema-v2 typed system event and model-visible metadata committed
/// Canonical typed system event and model-visible metadata committed
/// together.
AnnotatedSystemItem {
ts: u64,
entry: LoggedSystemHistoryEntry,
},
/// One typed agent-injected system item: notification, child-Worker
/// lifecycle event, `@<path>` / `/<slug>` resolution payload. Each
/// `SystemItem` carries kind metadata that the LLM
/// itself never sees (the LLM gets `Item::system_message` with the
/// item's denormalised `body`), but live clients and replay paths
/// dispatch on `kind` for typed rendering.
SystemItem { ts: u64, item: SystemItem },
/// Turn boundary. Records the turn count after increment.
TurnEnd { ts: u64, turn_count: usize },
@@ -260,6 +207,10 @@ pub struct RestoredState {
pub system_prompt: Option<String>,
pub config: RequestConfig,
pub history: Vec<Item>,
/// Canonical persisted history with stable identity and provenance. This is
/// the authority for rewrites, forks, and annotated restore; `history` is
/// retained as the model-facing item projection.
pub annotated_history: Vec<LoggedHistoryEntry>,
pub turn_count: usize,
/// AgentTurns consumed by the active paused/yielded logical run.
pub active_run_turn_count: Option<usize>,
@@ -276,7 +227,7 @@ pub struct RestoredState {
/// session-store は domain を不透明扱いし、各ドメインが自前で fold する。
pub extensions: Vec<(String, serde_json::Value)>,
/// User submissions in original typed form, in submit order.
/// One entry per `LogEntry::UserInput`; the K-th entry corresponds to
/// One entry per `LogEntry::AnnotatedUserInput`; the K-th entry corresponds to
/// the K-th `Item::user_message` derived during replay (modulo
/// pre-compaction history seeded via `SegmentStart.history`, whose
/// original segments are not preserved). Used by clients to re-render
@@ -291,6 +242,7 @@ pub fn collect_state(entries: &[LogEntry]) -> RestoredState {
system_prompt: None,
config: RequestConfig::default(),
history: Vec::new(),
annotated_history: Vec::new(),
turn_count: 0,
active_run_turn_count: None,
last_run_interrupted: false,
@@ -304,18 +256,6 @@ pub fn collect_state(entries: &[LogEntry]) -> RestoredState {
state.entries_count += 1;
match entry {
LogEntry::SegmentStart {
session_id,
system_prompt,
config,
history,
..
} => {
state.session_id = Some(*session_id);
state.system_prompt = system_prompt.clone();
state.config = config.clone();
state.history = history.iter().cloned().map(Item::from).collect();
}
LogEntry::AnnotatedSegmentStart {
session_id,
system_prompt,
@@ -326,6 +266,7 @@ pub fn collect_state(entries: &[LogEntry]) -> RestoredState {
state.session_id = Some(*session_id);
state.system_prompt = system_prompt.clone();
state.config = config.clone();
state.annotated_history = history.clone();
state.history = history
.iter()
.cloned()
@@ -338,26 +279,13 @@ pub fn collect_state(entries: &[LogEntry]) -> RestoredState {
state.last_run_interrupted = true;
state.active_run_turn_count = Some(0);
}
LogEntry::UserInput {
segments,
extensions,
..
} => {
let text = Segment::flatten_to_text(segments);
state.history.push(Item::user_message(text));
state.user_segments.push(segments.clone());
state.extensions.extend(
extensions
.iter()
.map(|extension| (extension.domain.clone(), extension.payload.clone())),
);
}
LogEntry::AnnotatedUserInput {
segments,
extensions,
history,
..
} => {
state.annotated_history.extend(history.iter().cloned());
state
.history
.extend(history.iter().cloned().map(|entry| Item::from(entry.item)));
@@ -370,20 +298,16 @@ pub fn collect_state(entries: &[LogEntry]) -> RestoredState {
}
LogEntry::AnnotatedAssistantItem { entry, .. }
| LogEntry::AnnotatedToolResult { entry, .. } => {
state.annotated_history.push(entry.clone());
state.history.push(Item::from(entry.item.clone()));
}
LogEntry::AnnotatedSystemItem { entry, .. } => {
state.annotated_history.push(LoggedHistoryEntry {
item: LoggedItem::from(entry.item.to_history_item()),
metadata: entry.metadata.clone(),
});
state.history.push(entry.item.to_history_item());
}
LogEntry::AssistantItem { item, .. } => {
state.history.push(Item::from(item.clone()));
}
LogEntry::ToolResult { item, .. } => {
state.history.push(Item::from(item.clone()));
}
LogEntry::SystemItem { item, .. } => {
state.history.push(item.to_history_item());
}
LogEntry::TurnEnd { turn_count, .. } => {
if let Some(active_turn_count) = &mut state.active_run_turn_count {
*active_turn_count += turn_count.saturating_sub(state.turn_count);
@@ -465,6 +389,20 @@ pub fn now_millis() -> u64 {
#[cfg(test)]
mod tests {
use super::*;
use crate::{
LoggedSessionHistoryEntryId, LoggedSessionHistoryMetadata, LoggedSessionHistoryOrigin,
};
fn annotated(item: Item) -> LoggedHistoryEntry {
LoggedHistoryEntry {
item: LoggedItem::from(item),
metadata: LoggedSessionHistoryMetadata {
entry_id: LoggedSessionHistoryEntryId::new(),
origin: LoggedSessionHistoryOrigin::LegacyUnknown,
derivation: None,
},
}
}
#[test]
fn replay_empty() {
@@ -476,12 +414,12 @@ mod tests {
#[test]
fn replay_segment_start_sets_initial_state() {
let state = collect_state(&[LogEntry::SegmentStart {
let state = collect_state(&[LogEntry::AnnotatedSegmentStart {
ts: 1000,
session_id: uuid::Uuid::nil(),
system_prompt: Some("You are helpful.".into()),
config: RequestConfig::default().with_max_tokens(1024),
history: vec![Item::user_message("seed").into()],
history: vec![annotated(Item::user_message("seed"))],
forked_from: None,
compacted_from: None,
}]);
@@ -494,7 +432,7 @@ mod tests {
#[test]
fn replay_full_turn() {
let state = collect_state(&[
LogEntry::SegmentStart {
LogEntry::AnnotatedSegmentStart {
ts: 1000,
session_id: uuid::Uuid::nil(),
system_prompt: None,
@@ -503,14 +441,15 @@ mod tests {
forked_from: None,
compacted_from: None,
},
LogEntry::UserInput {
LogEntry::AnnotatedUserInput {
ts: 2000,
extensions: vec![],
segments: vec![Segment::text("Hello")],
history: vec![annotated(Item::user_message("Hello"))],
},
LogEntry::AssistantItem {
LogEntry::AnnotatedAssistantItem {
ts: 3000,
item: Item::assistant_message("Hi!").into(),
entry: annotated(Item::assistant_message("Hi!")),
},
LogEntry::TurnEnd {
ts: 3100,
@@ -531,7 +470,7 @@ mod tests {
#[test]
fn replay_incomplete_invoke_is_interrupted() {
let state = collect_state(&[
LogEntry::SegmentStart {
LogEntry::AnnotatedSegmentStart {
ts: 1000,
session_id: uuid::Uuid::nil(),
system_prompt: None,
@@ -544,14 +483,15 @@ mod tests {
ts: 2000,
trigger: InvokeKind::UserSend,
},
LogEntry::UserInput {
LogEntry::AnnotatedUserInput {
ts: 2001,
extensions: vec![],
segments: vec![Segment::text("run a tool")],
history: vec![annotated(Item::user_message("run a tool"))],
},
LogEntry::AssistantItem {
LogEntry::AnnotatedAssistantItem {
ts: 3000,
item: Item::tool_call("call_1", "side_effect", "{}").into(),
entry: annotated(Item::tool_call("call_1", "side_effect", "{}")),
},
]);
@@ -561,7 +501,7 @@ mod tests {
#[test]
fn replay_with_tool_calls() {
let state = collect_state(&[
LogEntry::SegmentStart {
LogEntry::AnnotatedSegmentStart {
ts: 1000,
session_id: uuid::Uuid::nil(),
system_prompt: None,
@@ -570,22 +510,27 @@ mod tests {
forked_from: None,
compacted_from: None,
},
LogEntry::UserInput {
LogEntry::AnnotatedUserInput {
ts: 2000,
extensions: vec![],
segments: vec![Segment::text("Check weather")],
history: vec![annotated(Item::user_message("Check weather"))],
},
LogEntry::AssistantItem {
LogEntry::AnnotatedAssistantItem {
ts: 3000,
item: Item::tool_call("call_1", "get_weather", r#"{"city":"Tokyo"}"#).into(),
entry: annotated(Item::tool_call(
"call_1",
"get_weather",
r#"{"city":"Tokyo"}"#,
)),
},
LogEntry::ToolResult {
LogEntry::AnnotatedToolResult {
ts: 3500,
item: Item::tool_result("call_1", "Sunny, 25C").into(),
entry: annotated(Item::tool_result("call_1", "Sunny, 25C")),
},
LogEntry::AssistantItem {
LogEntry::AnnotatedAssistantItem {
ts: 4000,
item: Item::assistant_message("It's sunny in Tokyo!").into(),
entry: annotated(Item::assistant_message("It's sunny in Tokyo!")),
},
LogEntry::TurnEnd {
ts: 4100,
@@ -599,9 +544,9 @@ mod tests {
#[test]
fn replay_restores_durable_tool_image_detail() {
let entry = LogEntry::ToolResult {
let entry = LogEntry::AnnotatedToolResult {
ts: 3500,
item: Item::tool_result_item_with_attachments(
entry: annotated(Item::tool_result_item_with_attachments(
"call_image",
"attached",
None,
@@ -609,8 +554,7 @@ mod tests {
vec![agen::tool::Attachment::Image(
agen::tool::ImageAttachment::new("image/png", b"durable-image".to_vec()),
)],
)
.into(),
)),
};
let persisted = serde_json::to_string(&entry).unwrap();
let restored_entry: LogEntry = serde_json::from_str(&persisted).unwrap();
@@ -630,7 +574,7 @@ mod tests {
#[test]
fn replay_config_changed() {
let state = collect_state(&[
LogEntry::SegmentStart {
LogEntry::AnnotatedSegmentStart {
ts: 1000,
session_id: uuid::Uuid::nil(),
system_prompt: None,
@@ -650,7 +594,7 @@ mod tests {
#[test]
fn replay_llm_usage_appends_to_usage_history() {
let state = collect_state(&[
LogEntry::SegmentStart {
LogEntry::AnnotatedSegmentStart {
ts: 1000,
session_id: uuid::Uuid::nil(),
system_prompt: None,
@@ -659,10 +603,11 @@ mod tests {
forked_from: None,
compacted_from: None,
},
LogEntry::UserInput {
LogEntry::AnnotatedUserInput {
ts: 2000,
extensions: vec![],
segments: vec![Segment::text("hi")],
history: vec![annotated(Item::user_message("hi"))],
},
LogEntry::LlmUsage {
ts: 2100,
@@ -672,9 +617,9 @@ mod tests {
cache_write_tokens: 0,
output_tokens: 10,
},
LogEntry::AssistantItem {
LogEntry::AnnotatedAssistantItem {
ts: 2200,
item: Item::assistant_message("yo").into(),
entry: annotated(Item::assistant_message("yo")),
},
LogEntry::LlmUsage {
ts: 3100,
@@ -698,7 +643,7 @@ mod tests {
#[test]
fn replay_without_llm_usage_keeps_usage_history_empty() {
let state = collect_state(&[
LogEntry::SegmentStart {
LogEntry::AnnotatedSegmentStart {
ts: 1000,
session_id: uuid::Uuid::nil(),
system_prompt: None,
@@ -707,10 +652,11 @@ mod tests {
forked_from: None,
compacted_from: None,
},
LogEntry::UserInput {
LogEntry::AnnotatedUserInput {
ts: 2000,
extensions: vec![],
segments: vec![Segment::text("hi")],
history: vec![annotated(Item::user_message("hi"))],
},
]);
assert!(state.usage_history.is_empty());
@@ -771,7 +717,7 @@ mod tests {
#[test]
fn replay_invoke_marker_only_mutates_interrupted_state() {
let state = collect_state(&[
LogEntry::SegmentStart {
LogEntry::AnnotatedSegmentStart {
ts: 0,
session_id: uuid::Uuid::nil(),
system_prompt: None,
@@ -784,10 +730,11 @@ mod tests {
ts: 100,
trigger: InvokeKind::UserSend,
},
LogEntry::UserInput {
LogEntry::AnnotatedUserInput {
ts: 101,
extensions: vec![],
segments: vec![Segment::text("hi")],
history: vec![annotated(Item::user_message("hi"))],
},
LogEntry::TurnEnd {
ts: 200,
@@ -806,7 +753,7 @@ mod tests {
#[test]
fn replay_paused_turn_abandoned_clears_interrupted_marker() {
let state = collect_state(&[
LogEntry::SegmentStart {
LogEntry::AnnotatedSegmentStart {
ts: 0,
session_id: uuid::Uuid::nil(),
system_prompt: None,
@@ -830,7 +777,7 @@ mod tests {
#[test]
fn replay_restores_active_run_budget_across_compaction_checkpoint() {
let state = collect_state(&[
LogEntry::SegmentStart {
LogEntry::AnnotatedSegmentStart {
ts: 0,
session_id: uuid::Uuid::nil(),
system_prompt: None,
@@ -861,7 +808,7 @@ mod tests {
}))
.expect("legacy run-completed entry");
let state = collect_state(&[
LogEntry::SegmentStart {
LogEntry::AnnotatedSegmentStart {
ts: 0,
session_id: uuid::Uuid::nil(),
system_prompt: None,
@@ -924,7 +871,7 @@ mod tests {
#[test]
fn replay_extension_collects_domain_payload_pairs() {
let state = collect_state(&[
LogEntry::SegmentStart {
LogEntry::AnnotatedSegmentStart {
ts: 1000,
session_id: uuid::Uuid::nil(),
system_prompt: None,
@@ -983,9 +930,12 @@ mod tests {
#[test]
fn user_input_extensions_restore_with_the_same_committed_input() {
let segments = vec![Segment::text("Flow instructions"), Segment::text("Ticket")];
let entry = LogEntry::UserInput {
let entry = LogEntry::AnnotatedUserInput {
ts: 9999,
segments: segments.clone(),
history: vec![annotated(Item::user_message(Segment::flatten_to_text(
&segments,
)))],
extensions: vec![SessionExtension::new(
"flow.runtime.v1",
serde_json::json!({ "state": "implement", "revision": 0 }),
@@ -1000,7 +950,7 @@ mod tests {
assert_eq!(state.extensions[0].1["state"], "implement");
}
/// Mixed segments survive a JSON round-trip through `LogEntry::UserInput`,
/// Mixed segments survive a JSON round-trip through `LogEntry::AnnotatedUserInput`,
/// and `collect_state` derives `Item::user_message` from the flattened
/// text while preserving the original segments separately. This covers
/// the segments → flatten → Item replay path from the ticket.
@@ -1020,16 +970,19 @@ mod tests {
path: "src/main.rs".into(),
},
];
let entry = LogEntry::UserInput {
let entry = LogEntry::AnnotatedUserInput {
ts: 4242,
extensions: vec![],
segments: segments.clone(),
history: vec![annotated(Item::user_message(Segment::flatten_to_text(
&segments,
)))],
};
// JSON round-trip preserves the variant byte-for-byte.
let json = serde_json::to_string(&entry).unwrap();
let parsed: LogEntry = serde_json::from_str(&json).unwrap();
let state = collect_state(&[
LogEntry::SegmentStart {
LogEntry::AnnotatedSegmentStart {
ts: 1,
session_id: uuid::Uuid::nil(),
system_prompt: None,
crates/session-store/src/system_item.rs
+1 -1
View File
@@ -8,7 +8,7 @@
//! `kind` instead of parsing text prefixes like `[Notification] …` or
//! `[File: …]`.
//!
//! Persisted as the payload of [`crate::LogEntry::SystemItem`] (one
//! Persisted as the payload of [`crate::LogEntry::AnnotatedSystemItem`] (one
//! entry per item), and broadcast live as the payload of
//! `Event::SystemItem` on the wire.
//!
crates/session-store/src/worker_session_store.rs
+184 -184
View File
@@ -12,11 +12,7 @@
use crate::event_trace::TraceEntry;
use crate::segment_log::LogEntry;
use crate::store::{Store, StoreError};
use crate::{
LoggedHistoryEntry, LoggedItem, LoggedSessionHistoryEntryId, LoggedSessionHistoryMetadata,
LoggedSessionHistoryOrigin, LoggedSystemHistoryEntry, SegmentId, SessionId,
};
use base64::{Engine as _, engine::general_purpose::URL_SAFE_NO_PAD};
use crate::{SegmentId, SessionId};
use serde::{Deserialize, Serialize};
use std::fs::{self, File, OpenOptions};
use std::io::{Read, Seek, SeekFrom, Write};
@@ -56,7 +52,11 @@ impl WorkerSessionStore {
validate_canonical_segment_logs(&root)?;
}
PREVIOUS_SESSION_SCHEMA_VERSION | LEGACY_SESSION_SCHEMA_VERSION => {
migrate_segment_logs_to_v3(&root, manifest.session_id)?;
migrate_segment_logs_to_v3(
&root,
manifest.session_id,
manifest.schema_version,
)?;
manifest.schema_version = SESSION_SCHEMA_VERSION;
atomic_write_json(&root.join(SESSION_FILE), &manifest)?;
}
@@ -151,13 +151,7 @@ impl WorkerSessionStore {
.join(format!("{segment_id}.trace.jsonl"))
}
fn append_log_entry(
&self,
path: &Path,
session_id: SessionId,
segment_id: SegmentId,
entry: &LogEntry,
) -> Result<(), StoreError> {
fn append_log_entry(&self, path: &Path, entry: &LogEntry) -> Result<(), StoreError> {
let _guard = self
.append_lock
.lock()
@@ -172,9 +166,8 @@ impl WorkerSessionStore {
file.seek(SeekFrom::Start(0))?;
let mut existing = Vec::new();
file.read_to_end(&mut existing)?;
let line_index = parse_jsonl::<LogEntry>(&existing)?.len();
let entry = canonicalize_log_entry(session_id, segment_id, line_index, entry.clone());
let line = serde_json::to_string(&entry)?;
parse_jsonl::<LogEntry>(&existing)?;
let line = serde_json::to_string(entry)?;
let mut record = Vec::with_capacity(line.len() + 1);
record.extend_from_slice(line.as_bytes());
record.push(b'\n');
@@ -232,7 +225,7 @@ impl Store for WorkerSessionStore {
entry: &LogEntry,
) -> Result<(), StoreError> {
self.ensure_session(session_id, true)?;
self.append_log_entry(&self.log_path(segment_id), session_id, segment_id, entry)
self.append_log_entry(&self.log_path(segment_id), entry)
}
fn read_all(
@@ -285,9 +278,8 @@ impl Store for WorkerSessionStore {
) -> Result<(), StoreError> {
self.ensure_session(session_id, true)?;
let mut content = Vec::new();
for (line_index, entry) in entries.iter().enumerate() {
let entry = canonicalize_log_entry(session_id, segment_id, line_index, entry.clone());
serde_json::to_writer(&mut content, &entry)?;
for entry in entries {
serde_json::to_writer(&mut content, entry)?;
content.push(b'\n');
}
atomic_write_bytes(&self.log_path(segment_id), &content)?;
@@ -380,7 +372,11 @@ fn segment_log_paths(root: &Path) -> Result<Vec<(SegmentId, PathBuf)>, StoreErro
Ok(paths)
}
fn migrate_segment_logs_to_v3(root: &Path, session_id: SessionId) -> Result<(), StoreError> {
fn migrate_segment_logs_to_v3(
root: &Path,
session_id: SessionId,
source_schema_version: u32,
) -> Result<(), StoreError> {
struct MigrationPlan {
path: PathBuf,
source: Vec<u8>,
@@ -392,21 +388,14 @@ fn migrate_segment_logs_to_v3(root: &Path, session_id: SessionId) -> Result<(),
let mut plans = Vec::new();
for (segment_id, path) in segment_log_paths(root)? {
let source = fs::read(&path)?;
let entries: Vec<LogEntry> = parse_jsonl(&source).map_err(|error| StoreError::Corrupt {
line: 0,
message: format!(
"cannot migrate Worker Session log {}: {error}",
path.display()
),
})?;
let canonical = entries
.into_iter()
.enumerate()
.map(|(line_index, entry)| {
canonicalize_log_entry(session_id, segment_id, line_index, entry)
})
.collect::<Vec<_>>();
validate_canonical_entries(&path, &canonical)?;
let canonical = parse_legacy_jsonl(source_schema_version, session_id, segment_id, &source)
.map_err(|error| StoreError::Corrupt {
line: 0,
message: format!(
"cannot migrate Worker Session log {}: {error}",
path.display()
),
})?;
let mut output = Vec::new();
for entry in canonical {
serde_json::to_writer(&mut output, &entry)?;
@@ -442,120 +431,33 @@ fn migrate_segment_logs_to_v3(root: &Path, session_id: SessionId) -> Result<(),
fn validate_canonical_segment_logs(root: &Path) -> Result<(), StoreError> {
for (_, path) in segment_log_paths(root)? {
let entries: Vec<LogEntry> = parse_jsonl(&fs::read(&path)?)?;
validate_canonical_entries(&path, &entries)?;
let _: Vec<LogEntry> = parse_jsonl(&fs::read(&path)?)?;
}
Ok(())
}
fn validate_canonical_entries(path: &Path, entries: &[LogEntry]) -> Result<(), StoreError> {
for (line_index, entry) in entries.iter().enumerate() {
if matches!(
entry,
LogEntry::SegmentStart { .. }
| LogEntry::UserInput { .. }
| LogEntry::AssistantItem { .. }
| LogEntry::ToolResult { .. }
| LogEntry::SystemItem { .. }
) {
return Err(StoreError::Corrupt {
line: line_index + 1,
message: format!(
"Worker Session schema v3 contains legacy history record in {}",
path.display()
),
});
}
}
Ok(())
}
fn legacy_metadata(
_session_id: SessionId,
segment_id: SegmentId,
line_index: usize,
item_index: usize,
) -> LoggedSessionHistoryMetadata {
let mut identity = Vec::with_capacity(32);
identity.extend_from_slice(segment_id.as_bytes());
identity.extend_from_slice(&(line_index as u64).to_be_bytes());
identity.extend_from_slice(&(item_index as u64).to_be_bytes());
LoggedSessionHistoryMetadata {
entry_id: LoggedSessionHistoryEntryId(format!("l-{}", URL_SAFE_NO_PAD.encode(identity))),
origin: LoggedSessionHistoryOrigin::LegacyUnknown,
derivation: None,
}
}
fn canonicalize_log_entry(
fn parse_legacy_jsonl(
schema_version: u32,
session_id: SessionId,
segment_id: SegmentId,
line_index: usize,
entry: LogEntry,
) -> LogEntry {
match entry {
LogEntry::SegmentStart {
ts,
session_id,
system_prompt,
config,
history,
forked_from,
compacted_from,
} => LogEntry::AnnotatedSegmentStart {
ts,
session_id,
system_prompt,
config,
history: history
.into_iter()
.enumerate()
.map(|(item_index, item)| LoggedHistoryEntry {
item,
metadata: legacy_metadata(session_id, segment_id, line_index, item_index),
})
.collect(),
forked_from,
compacted_from,
},
LogEntry::UserInput {
ts,
segments,
extensions,
} => LogEntry::AnnotatedUserInput {
ts,
history: vec![LoggedHistoryEntry {
item: LoggedItem::from(agen::Item::user_message(
protocol::Segment::flatten_to_text(&segments),
)),
metadata: legacy_metadata(session_id, segment_id, line_index, 0),
}],
segments,
extensions,
},
LogEntry::AssistantItem { ts, item } => LogEntry::AnnotatedAssistantItem {
ts,
entry: LoggedHistoryEntry {
item,
metadata: legacy_metadata(session_id, segment_id, line_index, 0),
},
},
LogEntry::ToolResult { ts, item } => LogEntry::AnnotatedToolResult {
ts,
entry: LoggedHistoryEntry {
item,
metadata: legacy_metadata(session_id, segment_id, line_index, 0),
},
},
LogEntry::SystemItem { ts, item } => LogEntry::AnnotatedSystemItem {
ts,
entry: LoggedSystemHistoryEntry {
item,
metadata: legacy_metadata(session_id, segment_id, line_index, 0),
},
},
canonical => canonical,
}
bytes: &[u8],
) -> Result<Vec<LogEntry>, serde_json::Error> {
let text = std::str::from_utf8(bytes).map_err(|error| {
serde_json::Error::io(std::io::Error::new(std::io::ErrorKind::InvalidData, error))
})?;
text.lines()
.enumerate()
.filter(|(_, line)| !line.trim().is_empty())
.map(|(line_index, line)| {
crate::legacy_session_log::decode_entry(
schema_version,
line,
session_id,
segment_id,
line_index,
)
})
.collect()
}
fn atomic_write_json<T: Serialize>(path: &Path, value: &T) -> Result<(), StoreError> {
@@ -659,7 +561,21 @@ fn truncate_uncommitted_tail(file: &mut File) -> std::io::Result<u64> {
#[cfg(test)]
mod tests {
use super::*;
use crate::{Store, new_segment_id, new_session_id};
use crate::{
LoggedHistoryEntry, LoggedItem, LoggedSessionHistoryEntryId, LoggedSessionHistoryMetadata,
LoggedSessionHistoryOrigin, Store, new_segment_id, new_session_id,
};
fn annotated(item: agen::Item) -> LoggedHistoryEntry {
LoggedHistoryEntry {
item: LoggedItem::from(item),
metadata: LoggedSessionHistoryMetadata {
entry_id: LoggedSessionHistoryEntryId::new(),
origin: LoggedSessionHistoryOrigin::LegacyUnknown,
derivation: None,
},
}
}
#[test]
fn canonical_layout_and_single_session_invariant() {
@@ -748,26 +664,27 @@ mod tests {
)
.unwrap();
let source = vec![
LogEntry::SegmentStart {
ts: 1,
session_id,
system_prompt: None,
config: agen::llm_client::RequestConfig::default(),
history: vec![LoggedItem::from(agen::Item::assistant_message("prior"))],
forked_from: None,
compacted_from: None,
},
LogEntry::UserInput {
ts: 2,
segments: vec![protocol::Segment::Text {
content: "hello".into(),
}],
extensions: Vec::new(),
},
LogEntry::AssistantItem {
ts: 3,
item: LoggedItem::from(agen::Item::assistant_message("reply")),
},
serde_json::json!({
"kind": "segment_start",
"ts": 1,
"session_id": session_id,
"system_prompt": null,
"config": agen::llm_client::RequestConfig::default(),
"history": [LoggedItem::from(agen::Item::assistant_message("prior"))],
"forked_from": null,
"compacted_from": null
}),
serde_json::json!({
"kind": "user_input",
"ts": 2,
"segments": [{ "kind": "text", "content": "hello" }],
"extensions": []
}),
serde_json::json!({
"kind": "assistant_item",
"ts": 3,
"item": LoggedItem::from(agen::Item::assistant_message("reply"))
}),
];
let path = root
.path()
@@ -829,15 +746,16 @@ mod tests {
.path()
.join(SEGMENTS_DIR)
.join(format!("{valid_segment}.jsonl"));
let valid_entry = LogEntry::SegmentStart {
ts: 1,
session_id,
system_prompt: None,
config: agen::llm_client::RequestConfig::default(),
history: vec![LoggedItem::from(agen::Item::assistant_message("prior"))],
forked_from: None,
compacted_from: None,
};
let valid_entry = serde_json::json!({
"kind": "segment_start",
"ts": 1,
"session_id": session_id,
"system_prompt": null,
"config": agen::llm_client::RequestConfig::default(),
"history": [LoggedItem::from(agen::Item::assistant_message("prior"))],
"forked_from": null,
"compacted_from": null
});
let mut valid_bytes = serde_json::to_vec(&valid_entry).unwrap();
valid_bytes.push(b'\n');
fs::write(&valid_path, &valid_bytes).unwrap();
@@ -862,7 +780,7 @@ mod tests {
}
#[test]
fn schema_v3_rejects_legacy_records_and_new_writes_are_canonical() {
fn current_jsonl_requires_annotations_across_append_rewrite_and_reopen() {
let root = tempfile::tempdir().unwrap();
let session_id = new_session_id();
let segment_id = new_segment_id();
@@ -871,12 +789,12 @@ mod tests {
.create_segment(
session_id,
segment_id,
&[LogEntry::SegmentStart {
&[LogEntry::AnnotatedSegmentStart {
ts: 1,
session_id,
system_prompt: None,
config: agen::llm_client::RequestConfig::default(),
history: Vec::new(),
history: vec![annotated(agen::Item::user_message("seed"))],
forked_from: None,
compacted_from: None,
}],
@@ -886,11 +804,94 @@ mod tests {
.append(
session_id,
segment_id,
&LogEntry::UserInput {
&LogEntry::AnnotatedAssistantItem {
ts: 2,
entry: annotated(agen::Item::assistant_message("reply")),
},
)
.unwrap();
let before_rewrite = store.read_all(session_id, segment_id).unwrap();
store
.create_segment(session_id, segment_id, &before_rewrite)
.unwrap();
drop(store);
let reopened = WorkerSessionStore::new(root.path()).unwrap();
let restored = reopened.read_all(session_id, segment_id).unwrap();
assert_eq!(
serde_json::to_value(&restored).unwrap(),
serde_json::to_value(&before_rewrite).unwrap()
);
for entry in &restored {
match entry {
LogEntry::AnnotatedSegmentStart { history, .. } => assert!(history.iter().all(
|entry| !entry.metadata.entry_id.0.is_empty()
&& matches!(
entry.metadata.origin,
LoggedSessionHistoryOrigin::LegacyUnknown
)
)),
LogEntry::AnnotatedAssistantItem { entry, .. } => {
assert!(!entry.metadata.entry_id.0.is_empty());
assert!(matches!(
entry.metadata.origin,
LoggedSessionHistoryOrigin::LegacyUnknown
));
}
_ => {}
}
}
let log = fs::read_to_string(reopened.log_path(segment_id)).unwrap();
for line in log.lines() {
let value: serde_json::Value = serde_json::from_str(line).unwrap();
let kind = value["kind"].as_str().unwrap();
assert!(
!matches!(
kind,
"segment_start"
| "user_input"
| "assistant_item"
| "tool_result"
| "system_item"
),
"current-schema JSONL contains legacy history record: {kind}"
);
}
}
#[test]
fn schema_v3_rejects_legacy_records_and_new_writes_are_canonical() {
let root = tempfile::tempdir().unwrap();
let session_id = new_session_id();
let segment_id = new_segment_id();
let store = WorkerSessionStore::new(root.path()).unwrap();
store
.create_segment(
session_id,
segment_id,
&[LogEntry::AnnotatedSegmentStart {
ts: 1,
session_id,
system_prompt: None,
config: agen::llm_client::RequestConfig::default(),
history: vec![annotated(agen::Item::assistant_message("seed"))],
forked_from: None,
compacted_from: None,
}],
)
.unwrap();
store
.append(
session_id,
segment_id,
&LogEntry::AnnotatedUserInput {
ts: 2,
segments: vec![protocol::Segment::Text {
content: "new".into(),
}],
history: vec![annotated(agen::Item::user_message("new"))],
extensions: Vec::new(),
},
)
@@ -907,12 +908,11 @@ mod tests {
let mut file = OpenOptions::new().append(true).open(path).unwrap();
serde_json::to_writer(
&mut file,
&LogEntry::SystemItem {
ts: 3,
item: crate::SystemItem::LegacyIgnored {
slug: "legacy".into(),
},
},
&serde_json::json!({
"kind": "system_item",
"ts": 3,
"item": { "kind": "legacy_ignored", "slug": "legacy" }
}),
)
.unwrap();
file.write_all(b"\n").unwrap();