Files
hermes-agent/tests/gateway/test_stop_thread_sibling.py
kshitijk4poor 0fb56906fc fix(gateway): a /stop in a thread stops every run of that thread
The thread-sibling tier short-circuited the chat-scope tier, so when a per-user
thread sibling was live the handler interrupted ONLY that run and replied
"Stopped" while a same-thread run under a differently shaped key — the
channel-keyed #286 shape this fallback exists for — kept going. The chat tier is
a superset of the sibling tier (a sibling needs the caller's own thread slot,
which satisfies the chat predicate), so it is the set acted on; the reason still
resolves to `stop_command_thread_sibling` when the chat tier found nothing beyond
siblings, so hook consumers keep that label.

Both tiers also share ONE scan now (they each called `_same_chat_runs`, building
the running-agent snapshot twice), and the matcher is a single text pass:
`_same_chat_key_slots` takes the caller's key prefix instead of re-deriving the
namespace per key, and the whole-slot rule lives once in `_strip_slot` (the tail
check uses it too). Concrete annotations on the three methods (`List[...]` was
missing from the typing import), and the prefix construction reads as a named
namespace rather than a nested f-string.

WhatsApp DMs: `build_session_key` canonicalises the DM chat id, so the fallback
matched nothing there — it now canonicalises the same way, which makes the
chat-scope fallback reach a run keyed from any JID/LID alias.

Docs: `sessions.md` described the tiers sequentially and over-claimed the
widening's reach — an in-thread stop reaches that thread plus a thread-less
room-wide run, never another thread or a peer's per-sender top-level run.
`hooks.md` pointed at `gateway/run.py` for `_interrupt_and_clear_session`, which
lives in `gateway/run_agent_cache.py`.

Guards: the partial-stop case (per-user sibling + same-thread channel run, both
must be interrupted), the lone-sibling reason label, and the WhatsApp canonical
id. Each is mutation-checked individually: restoring the short-circuit, forcing
the reason to chat_scope, or dropping the canonicalisation each fails exactly
its own test. 21 passed in the two /stop test files.
2026-09-18 15:32:17 +05:30

171 lines
6.0 KiB
Python

"""Regression tests: /stop can interrupt a sibling participant's run in a
per-user thread.
When ``thread_sessions_per_user=True``, each participant in a thread gets an
isolated session key (``...:{thread_id}:{user_id}``). A run another user
started lives under a different key, so the caller's own ``/stop`` used to find
nothing and reply "no active task to stop". Authorized users should be able to
stop any run in the same thread.
"""
import pytest
from gateway.run import GatewayRunner, _AGENT_PENDING_SENTINEL, _INTERRUPT_REASON_STOP
from gateway.session import SessionSource, build_session_key
from gateway.platforms.base import Platform
from gateway.platforms.event import MessageEvent, MessageType
class _FakeAgent:
pass
def _thread_source(uid, thread_id="thr1", chat_id="chan1"):
return SessionSource(
platform=Platform.DISCORD,
chat_type="forum",
chat_id=chat_id,
thread_id=thread_id,
user_id=uid,
)
def _per_user_key(uid, thread_id="thr1", chat_id="chan1"):
return build_session_key(
_thread_source(uid, thread_id, chat_id),
thread_sessions_per_user=True,
)
# ---------------------------------------------------------------------------
# _sibling_thread_run_keys
# ---------------------------------------------------------------------------
def test_sibling_returns_empty_for_non_thread_source():
# The sibling TIER is thread-scoped: a non-thread group/channel source has no thread slot to
# match. The room-wide fallback for that shape is the chat tier's job — see
# tests/gateway/test_stop_chat_scope.py::test_stop_reaches_peer_run_in_per_sender_group.
runner = object.__new__(GatewayRunner)
nonthread = SessionSource(
platform=Platform.DISCORD, chat_type="group", chat_id="chan1", user_id="userA"
)
grp_b = build_session_key(
SessionSource(
platform=Platform.DISCORD, chat_type="group", chat_id="chan1", user_id="userB"
)
)
runner._running_agents = {grp_b: _FakeAgent()}
own_key = "agent:main:discord:group:chan1:userA"
assert runner._sibling_thread_run_keys(
nonthread, runner._same_chat_runs(nonthread, own_key)
) == []
def test_sibling_matches_named_profile_runs():
# A named-profile stop must match another participant's run under the SAME profile, taken from
# the caller's own key (the session store's answer)...
runner = object.__new__(GatewayRunner)
source = _thread_source("userA")
source.profile = "work"
key_b = build_session_key(
_thread_source("userB"), thread_sessions_per_user=True, profile="work"
)
runner._running_agents = {key_b: _FakeAgent()}
own_key = "agent:work:discord:forum:chan1:thr1:userA"
assert runner._sibling_thread_run_keys(source, runner._same_chat_runs(source, own_key)) == [key_b]
def test_sibling_does_not_cross_profiles():
# ...and never reaches a DIFFERENT profile's run in the same chat/thread.
runner = object.__new__(GatewayRunner)
source = _thread_source("userA")
source.profile = "work"
main_key = build_session_key(_thread_source("userB"), thread_sessions_per_user=True)
runner._running_agents = {main_key: _FakeAgent()}
own_key = "agent:work:discord:forum:chan1:thr1:userA"
assert runner._sibling_thread_run_keys(source, runner._same_chat_runs(source, own_key)) == []
# ---------------------------------------------------------------------------
# _handle_stop_command fallback path
# ---------------------------------------------------------------------------
class _StoreEntry:
def __init__(self, session_key):
self.session_key = session_key
class _FakeStore:
def __init__(self, session_key):
self._key = session_key
def get_or_create_session(self, source):
return _StoreEntry(self._key)
@pytest.mark.asyncio
async def test_stop_does_not_interrupt_sibling_when_unauthorized(monkeypatch):
runner = object.__new__(GatewayRunner)
key_a = _per_user_key("userA")
key_b = _per_user_key("userB")
runner._running_agents = {key_b: _FakeAgent()}
runner.session_store = _FakeStore(key_a)
interrupted = []
async def _fake_interrupt(session_key, source, *, interrupt_reason, invalidation_reason):
interrupted.append(session_key)
runner._interrupt_and_clear_session = _fake_interrupt
runner._is_user_authorized = lambda source: False
event = MessageEvent(
text="/stop", message_type=MessageType.TEXT, source=_thread_source("userA")
)
result = await runner._handle_stop_command(event)
assert interrupted == []
assert "no active" in str(getattr(result, "text", result)).lower()
# ---------------------------------------------------------------------------
# /stop with no active agent still clears a stuck platform status (#32295)
# ---------------------------------------------------------------------------
class _FakeStatusAdapter:
def __init__(self):
self.cleared = []
async def _stop_typing_with_metadata(self, chat_id, metadata=None):
self.cleared.append((chat_id, metadata))
@pytest.mark.asyncio
async def test_stop_no_active_agent_survives_status_clear_failure():
"""A failing adapter clear must not break the /stop reply."""
runner = object.__new__(GatewayRunner)
runner._running_agents = {}
key = _per_user_key("userA")
runner.session_store = _FakeStore(key)
runner._is_user_authorized = lambda source: True
class _BoomAdapter:
async def _stop_typing_with_metadata(self, chat_id, metadata=None):
raise RuntimeError("boom")
runner.adapters = {Platform.DISCORD: _BoomAdapter()}
runner._thread_metadata_for_source = (
lambda source, reply_to_message_id=None: None
)
runner._reply_anchor_for_event = lambda event: None
event = MessageEvent(
text="/stop", message_type=MessageType.TEXT, source=_thread_source("userA")
)
result = await runner._handle_stop_command(event)
assert "no active" in str(getattr(result, "text", result)).lower()