trex/trex-core
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix strategy state persistence corrections

Browse Source
This commit is contained in:
Hermes Agent
2026-08-29 16:11:02 +09:00
parent 4699de61cc
commit 3854aabd0d
11 changed files with 190 additions and 85 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/adapters/persistence/test_migration_and_drift.py
+20
View File
@@ -28,6 +28,7 @@ _EXPECTED_TABLES = {
"broker_account_snapshot", # dcy-041 / GitHub #45
"ledger_skip", # GitHub #68 skipped-ledger durable audit
"reconcile_notification_state", # req-0007 / ac-0024
"slot_strategy_state", # req-0009 / ac-0027
}
@@ -59,6 +60,25 @@ def test_daily_slot_columns_match_expected_contract(pg_clean):
assert actual == set(EXPECTED_SLOT_FIELDS)
def test_strategy_state_schema_matches_instance_map_contract(pg_clean):
with psycopg.connect(pg_clean) as conn:
with conn.cursor() as cur:
cur.execute(
"SELECT column_name, data_type, udt_name "
"FROM information_schema.columns "
"WHERE table_schema = 'public' AND table_name = 'slot_strategy_state'"
)
columns = {row[0]: (row[1], row[2]) for row in cur.fetchall()}
cur.execute(
"SELECT a.attname FROM pg_index i "
"JOIN pg_attribute a ON a.attrelid = i.indrelid AND a.attnum = ANY(i.indkey) "
"WHERE i.indrelid = 'slot_strategy_state'::regclass AND i.indisprimary"
)
primary_key = {row[0] for row in cur.fetchall()}
assert columns == {"slot_id": ("uuid", "uuid"), "data": ("jsonb", "jsonb")}
assert primary_key == {"slot_id"}
def test_drift_detected_when_contract_extended(pg_clean):
"""Negative case: a contract field absent from the table is caught."""
actual = _columns(pg_clean, "daily_slot")
tests/adapters/persistence/test_postgresql_repository.py
+49 -3
View File
@@ -164,7 +164,7 @@ def test_strategy_state_store_pg(pg_clean):
repo = PostgreSQLSlotRepository(pg_clean)
store = PgStrategyStateStore(pg_clean)
svc = SlotService(repo, instance_key=INST, strategy_state_store=store)
svc = SlotService(repo, instance_key=INST)
repo.reconcile_universe(INST, {"pf-a": "005930"})
slot = repo.find_slot_by_portfolio(INST, "pf-a")
svc.transit_slot(slot.slot_id, SlotState.INACTIVE, SlotState.PENDING_BUY,
@@ -189,12 +189,58 @@ def test_strategy_state_store_pg(pg_clean):
))
assert all(write.ok for write in writes)
assert {"instance-c", "instance-d"} <= set(store.get_for_slots([slot.slot_id])[slot.slot_id])
assert store.get_for_slots([slot.slot_id])[slot.slot_id]["instance-a"].version == 1
# Separate connections race on the same entry/version: only one CAS wins.
with ThreadPoolExecutor(max_workers=2) as executor:
competing = list(executor.map(
lambda data: store.set(
slot.slot_id, "instance-a", expected_version=1, data=data, slot_active=True,
),
("winner-one", "winner-two"),
))
assert sum(result.ok for result in competing) == 1, competing
assert sum(result.reason == ss.REASON_VERSION_CONFLICT for result in competing) == 1
winner = next(data for data, result in zip(("winner-one", "winner-two"), competing) if result.ok)
assert store.get_for_slots([slot.slot_id])[slot.slot_id]["instance-a"].data == winner
assert store.get_for_slots([slot.slot_id])[slot.slot_id]["instance-a"].version == 2
# A fresh adapter sees the byte-identical persisted entry and can continue CAS.
before_restart = store.get_for_slots([slot.slot_id])[slot.slot_id]["instance-a"]
after_restart = PgStrategyStateStore(pg_clean).get_for_slots([slot.slot_id])[slot.slot_id]["instance-a"]
assert after_restart == before_restart
svc.transit_slot(slot.slot_id, SlotState.HOLDING, SlotState.PENDING_SELL,
reason="sell", actor="t")
svc.transit_slot(slot.slot_id, SlotState.PENDING_SELL, SlotState.INACTIVE,
reason="sold", actor="t")
assert store.get_for_slots([slot.slot_id])[slot.slot_id]["instance-a"].version == 1
assert store.get_for_slots([slot.slot_id])[slot.slot_id]["instance-a"].version == 2
reused = svc.transit_slot(slot.slot_id, SlotState.INACTIVE, SlotState.PENDING_BUY,
reason="buy", actor="t")
reason="buy", actor="t")
assert reused.occupancy_seq == 2
rehydrated = PgStrategyStateStore(pg_clean)
persisted = rehydrated.get_for_slots([slot.slot_id])[slot.slot_id]["instance-a"]
assert persisted == before_restart
assert rehydrated.set(
slot.slot_id, "instance-a", expected_version=2, data="continued", slot_active=True,
).version == 3
def test_strategy_state_write_cannot_follow_close(pg_clean):
"""The store's locked lifecycle read rejects a write after close wins."""
from trex_core.adapters.persistence.postgresql import PgStrategyStateStore
repo = PostgreSQLSlotRepository(pg_clean)
svc = SlotService(repo, instance_key=INST)
store = PgStrategyStateStore(pg_clean)
repo.reconcile_universe(INST, {"pf-close": "005930"})
slot = repo.find_slot_by_portfolio(INST, "pf-close")
svc.transit_slot(slot.slot_id, SlotState.INACTIVE, SlotState.PENDING_BUY, reason="buy", actor="t")
svc.transit_slot(slot.slot_id, SlotState.PENDING_BUY, SlotState.HOLDING, reason="fill", actor="t", qty=1)
assert store.set(slot.slot_id, "instance", expected_version=0, data="before", slot_active=True).ok
svc.transit_slot(slot.slot_id, SlotState.HOLDING, SlotState.PENDING_SELL, reason="sell", actor="t")
svc.transit_slot(slot.slot_id, SlotState.PENDING_SELL, SlotState.INACTIVE, reason="sold", actor="t")
result = store.set(slot.slot_id, "instance", expected_version=1, data="after", slot_active=True)
assert not result.ok and result.reason == "slot_inactive"
assert store.get_for_slots([slot.slot_id])[slot.slot_id]["instance"].data == "before"
tests/application/test_slots_handler_quantity_mismatch.py
+27
View File
@@ -332,6 +332,33 @@ def test_ac7_warn_store_failure_fail_closed():
assert "excluded_slots" in resp.payload
def test_ac23_exclusion_precedes_broker_and_strategy_state_readers():
repo = InMemorySlotRepository()
svc = SlotService(repo)
warn_store = InMemoryReconcileWarningStore()
slot = _make_holding_slot(repo, svc, portfolio_id="pf-A", symbol=SYM_A, qty=10)
warn_store.set_warned(str(slot.slot_id), portfolio_id="pf-A", case_code="C")
class ExplodingReader:
def holdings_for_slots_rpc(self):
raise AssertionError("excluded slot reached broker reader")
class ExplodingStateStore:
def get_for_slots(self, _slot_ids):
raise AssertionError("excluded slot reached state reader")
handler = make_slots_rpc_handler(
repo,
instance_key=INSTANCE,
warn_store=warn_store,
broker_quantity_reader=ExplodingReader(),
strategy_state_store=ExplodingStateStore(),
)
response = handler(_req(filter="holding")).payload
assert response["slots"] == []
assert response["excluded_slots"] == {"reconcile_quantity_mismatch": 1}
# ---------------------------------------------------------------------------
# AC-8: intent rejected if C-case warning set after last slots RPC (race)
# ---------------------------------------------------------------------------
tests/application/test_slots_handler_strategy_state.py
+11
View File
@@ -58,3 +58,14 @@ def test_operator_without_or_with_id_receives_full_map_without_watchdog_update()
state = _call(handler, payload)["slots"][0]["strategy_state"]
assert set(state) == {"instance-a", "instance-b"}
assert watchdog.last_seen(str(slot.slot_id)) is None
@pytest.mark.parametrize("spy", ["false", "true", 1, [], {}])
def test_non_boolean_spy_is_rejected_without_visibility_or_watchdog_update(spy):
slot, _, watchdog, handler = _wire()
payload = _call(handler, {"spy": spy})
assert payload == {
"slots": [], "available_slots": None, "degraded": False,
"reason": "bad_request:spy", "message": "spy must be a boolean",
}
assert watchdog.last_seen(str(slot.slot_id)) is None
tests/conftest.py
+1
View File
@@ -27,6 +27,7 @@ _PG_TABLES = (
"daily_symbol_entries",
"daily_slot",
"reconcile_notification_state",
"slot_strategy_state",
)
tests/domain/slot/test_occupancy_seq_and_close_delete.py
+1 -1
View File
@@ -8,7 +8,7 @@ from trex_core.domain.slot.in_memory_repository import InMemorySlotRepository
def test_close_preserves_state_and_reopen_continues_version():
repo, store = InMemorySlotRepository(), InMemoryStrategyStateStore()
service = SlotService(repo, strategy_state_store=store)
service = SlotService(repo)
repo.reconcile_universe("default", {"pf": "SYM"})
slot = repo.list_slots("default")[0]
service.transit_slot(slot.slot_id, SlotState.INACTIVE, SlotState.PENDING_BUY, reason="buy", actor="t")
trex_core/adapters/persistence/postgresql.py
+34 -34
View File
@@ -716,53 +716,53 @@ class PgStrategyStateStore:
data: str,
slot_active: bool,
) -> SetResult:
if not slot_active:
return SetResult(ok=False, reason="slot_inactive")
if len(data.encode("utf-8")) > _STATE_CAP_BYTES:
return SetResult(ok=False, reason="state_too_large")
now = datetime.now(tz=timezone.utc)
with self._psycopg.connect(self._dsn, autocommit=False) as conn:
try:
with conn.cursor() as cur:
# Lock the lifecycle row before changing state. A close UPDATE
# and this write therefore have one serial order: a writer that
# follows a close observes INACTIVE and cannot recreate state.
cur.execute(
"INSERT INTO slot_strategy_state (slot_id, data) "
"SELECT %s, jsonb_build_object(%s, jsonb_build_object("
"'version', 1, 'data', %s, 'updated_at', %s)) WHERE %s = 0 "
"ON CONFLICT (slot_id) DO UPDATE SET data = jsonb_set("
"slot_strategy_state.data, ARRAY[%s], jsonb_build_object("
"'version', %s + 1, 'data', %s, 'updated_at', %s), true) "
"WHERE COALESCE((slot_strategy_state.data -> %s ->> 'version')::bigint, 0) = %s "
"RETURNING data -> %s ->> 'version'",
(
slot_id,
strategy_instance_id,
data,
now,
expected_version,
strategy_instance_id,
expected_version,
data,
now,
strategy_instance_id,
expected_version,
strategy_instance_id,
),
"SELECT status FROM daily_slot WHERE slot_id = %s::uuid FOR UPDATE",
(slot_id,),
)
row = cur.fetchone()
if row is None:
cur.execute(
"SELECT COALESCE((data -> %s ->> 'version')::bigint, 0) "
"FROM slot_strategy_state WHERE slot_id = %s",
(strategy_instance_id, slot_id),
)
conflict = cur.fetchone()
slot = cur.fetchone()
if slot is None or slot[0] not in {"pending_buy", "holding", "pending_sell"}:
return SetResult(ok=False, reason="slot_inactive")
cur.execute(
"SELECT data FROM slot_strategy_state WHERE slot_id = %s::uuid",
(slot_id,),
)
state_row = cur.fetchone()
state_map = state_row[0] if state_row is not None else {}
entry = state_map.get(strategy_instance_id)
version = int(entry["version"]) if entry is not None else 0
if expected_version != version:
return SetResult(
ok=False,
reason="version_conflict",
version=int(conflict[0]) if conflict is not None else 0,
version=version,
)
entry_json = "jsonb_build_object('version', %s::bigint, " \
"'data', %s::text, 'updated_at', %s::timestamptz)"
if state_row is None:
cur.execute(
"INSERT INTO slot_strategy_state (slot_id, data) VALUES ("
"%s::uuid, jsonb_build_object(%s::text, " + entry_json + "))",
(slot_id, strategy_instance_id, version + 1, data, now),
)
else:
cur.execute(
"UPDATE slot_strategy_state SET data = jsonb_set(data, "
"ARRAY[%s::text]::text[], " + entry_json + ", true) "
"WHERE slot_id = %s::uuid",
(strategy_instance_id, version + 1, data, now, slot_id),
)
conn.commit()
return SetResult(ok=True, version=int(row[0]))
return SetResult(ok=True, version=version + 1)
except BaseException:
conn.rollback()
raise
trex_core/adapters/persistence/sql/027_replace_slot_strategy_state_with_instance_map.sql
+1 -1
View File
@@ -12,7 +12,7 @@ BEGIN
DROP TABLE slot_strategy_state;
END IF;
END $$;
CREATE TABLE slot_strategy_state (
CREATE TABLE IF NOT EXISTS slot_strategy_state (
slot_id UUID PRIMARY KEY,
data JSONB NOT NULL DEFAULT '{}'::jsonb
);
trex_core/application/messaging_handlers.py
+43 -39
View File
@@ -223,7 +223,16 @@ def make_slots_rpc_handler(
def handler(req: RpcRequest) -> RpcResponse:
body = req.payload or {}
filter_val = str(body.get("filter") or "all").lower()
spy = bool(body.get("spy", False))
spy_value = body.get("spy", False)
if not isinstance(spy_value, bool):
return RpcResponse(payload={
"slots": [],
"available_slots": None,
"degraded": False,
"reason": "bad_request:spy",
"message": "spy must be a boolean",
})
spy = spy_value
strategy_instance_id = body.get("strategy_instance_id")
if not spy and (not isinstance(strategy_instance_id, str) or not strategy_instance_id):
return RpcResponse(payload={
@@ -268,10 +277,42 @@ def make_slots_rpc_handler(
for s in slots
]
# AC-0023 exclusion must precede all optional data readers so excluded
# slots cannot expose state or make an otherwise valid response fail.
excluded_count = 0
if warn_store is not None:
try:
c_warned_ids = warn_store.list_c_case_warned()
except Exception: # noqa: BLE001
_log.exception("warn_store.list_c_case_warned() failed — fail-closed")
return RpcResponse(payload={
"slots": [],
"available_slots": None,
"degraded": True,
"reason": "warn_store_unavailable",
"message": (
"reconcile warning store unavailable; "
"returning no slots (fail-closed)"
),
"excluded_slots": {"reconcile_quantity_mismatch": 0},
})
if c_warned_ids:
included = [
(slot, row)
for slot, row in zip(slots, slot_rows, strict=True)
if not (
slot.status.value == "holding"
and str(slot.slot_id) in c_warned_ids
)
]
excluded_count = len(slots) - len(included)
slots = [slot for slot, _ in included]
slot_rows = [row for _, row in included]
broker_qty_by_symbol: dict[str, int] = {}
broker_qty_available = False
broker_qty_captured_at: Optional[datetime] = None
if broker_quantity_reader is not None:
if broker_quantity_reader is not None and slots:
try:
holdings, broker_qty_captured_at = broker_quantity_reader.holdings_for_slots_rpc()
broker_qty_by_symbol = {
@@ -361,43 +402,6 @@ def make_slots_rpc_handler(
if strategy_filter is not None:
slot_rows = [r for r in slot_rows if r.get("strategy_name") == strategy_filter]
# AC-0023: exclude holding slots with active case_code='C' reconcile warning.
# Fail-closed: if warn_store exists but throws, return degraded=True / no slots.
excluded_count = 0
if warn_store is not None:
try:
c_warned_ids = warn_store.list_c_case_warned()
except Exception: # noqa: BLE001
_log.exception("warn_store.list_c_case_warned() failed — fail-closed")
return RpcResponse(payload={
"slots": [],
"available_slots": available_slots,
"degraded": True,
"reason": "warn_store_unavailable",
"message": (
"reconcile warning store unavailable; "
"returning no slots (fail-closed)"
),
"excluded_slots": {"reconcile_quantity_mismatch": 0},
})
if c_warned_ids:
# Rebuild slot lookup from the original `slots` list (before strategy
# filter reduced slot_rows). slot_rows[i] corresponds to slots[i]
# only before filtering; use slot_id stored in each row.
slot_by_id = {str(s.slot_id): s for s in slots}
filtered_rows: list[dict] = []
for row in slot_rows:
slot_obj = slot_by_id.get(str(row.get("slot_id", "")))
if (
slot_obj is not None
and str(slot_obj.slot_id) in c_warned_ids
and slot_obj.status.value == "holding"
):
excluded_count += 1
else:
filtered_rows.append(row)
slot_rows = filtered_rows
base_payload: dict = {
"slots": slot_rows,
"available_slots": available_slots,
trex_core/domain/slot/service.py
-3
View File
@@ -44,15 +44,12 @@ class SlotService:
self,
repository: SlotRepository,
instance_key: str = "default",
*,
strategy_state_store: Any = None,
) -> None:
self._repo = repository
# dcy-020/REQ-057: advisory lock is scoped per instance so a shared DB
# does not serialize transitions across instances (NFR-1).
self._instance_key = instance_key
self._lock_id = slot_advisory_lock_id(instance_key)
self._strategy_state_store = strategy_state_store
@property
def instance_key(self) -> str:
trex_core/runtime/daemon.py
+3 -4
View File
@@ -182,13 +182,12 @@ class Daemon:
# Persistence + services
self.slot_repo = PostgreSQLSlotRepository(config.dsn)
self.intent_store = PostgreSQLIntentStore(config.dsn)
# dcy-028 / REQ-079·081 (BAT-042): opaque strategy-state store. Injected
# into the slot service so close (→INACTIVE) deletes the slot's state.
# dcy-028 / REQ-079·081 (BAT-042): opaque strategy-state store persists
# independently of slot occupancy and is shared by both RPC handlers.
self.strategy_state_store = PgStrategyStateStore(config.dsn)
self.slot_service = SlotService(
self.slot_repo,
instance_key=config.instance_key,
strategy_state_store=self.strategy_state_store,
)
self.outbox = PostgreSQLLedgerOutbox(config.dsn)
@@ -458,7 +457,7 @@ class Daemon:
topology.slots_queue(self.config.instance_key), handler, self.stop)
def _strategy_state_set_loop(self) -> None:
# dcy-028 / REQ-080: opaque strategy-state write RPC (occupancy_seq+version CAS).
# dcy-028 / REQ-080: opaque strategy-state write RPC (entry-version CAS).
transport = _build_transport(self.config)
self._transports.append(transport)
try: