mirror of
https://github.com/NousResearch/hermes-agent.git
synced 2026-07-30 19:09:28 +00:00
Second, deeper pass over tools/gateway/hermes_cli plus first pass over the trees wave 1 missed (acp, acp_adapter, skills, computer_use, docker, dashboard, conformance, monitoring, secret_sources, hermes_state, providers). Same rubric as wave 1 (AGENTS.md test policy); security, alternation/caching invariants, issue-number regressions, and E2E kept. Real test-quality fixes found and rooted out along the way: - tests/tools/test_command_guards.py made real auxiliary-LLM HTTPS calls (DEFAULT_CONFIG smart-approval leaked in) — pinned approval mode=manual via autouse fixture: 17.4s → 0.4s. - test_model_switch_custom_providers.py / test_user_providers_model_switch.py silently probed live provider catalogs (~2s/test) — stubbed cached_provider_model_ids/provider_model_ids/fetch_api_models. - test_telegram_noise_filter.py: 15-platform copy-paste matrix over shared gateway.run logic → 3 representative platforms (55s → 3.9s). - test_gateway_shutdown.py: stop()'s 5s interrupt-deadline loop spun on MagicMock agents — interrupt.side_effect now clears _running_agents (22s → 1.0s). - test_gateway_inactivity_timeout.py poll-harness timings shrunk 3-5x (24s → 1.1s); test_mcp_stability.py backoff/SIGTERM-grace sleeps patched (15.4s → 2.5s); test_async_delegation.py negative-drain wait 5s → 0.5s. - test_telegram_init_deadline.py: loop-block margin restored to 1.0s with rationale comment — the watchdog-dump assertion needs the loop blocked well past deadline+grace under parallel load (flaked once in the 40-worker verification run at a 0.2s margin). Verification: full hermetic suite via scripts/run_tests.sh — 2,438 files, 21,718 tests passed, 0 failed, 293.9s wall. Suite totals vs original baseline: 46,820 → 19,757 test functions (−57.8%), wall 583.5s → 293.9s (−50%), subprocess CPU 13,564s → 11,623s.
207 lines
6.8 KiB
Python
207 lines
6.8 KiB
Python
"""Regression tests for active-session TEXT follow-up queueing.
|
|
|
|
When the agent is actively running, rapid text follow-ups should survive as
|
|
one next-turn pending message instead of clobbering each other. In
|
|
``busy_text_mode=queue`` those active follow-ups first pass through a short
|
|
debounce so bursty multi-message thoughts are merged before the active drain
|
|
hands off the next turn.
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
import asyncio
|
|
import sys
|
|
import types
|
|
from unittest.mock import AsyncMock, MagicMock, patch
|
|
|
|
import pytest
|
|
|
|
# Minimal telegram stub so importing gateway.platforms.base does not pull
|
|
# in the real python-telegram-bot dependency.
|
|
_tg = sys.modules.get("telegram") or types.ModuleType("telegram")
|
|
_tg.constants = sys.modules.get("telegram.constants") or types.ModuleType("telegram.constants")
|
|
_ct = MagicMock()
|
|
_ct.PRIVATE = "private"
|
|
_ct.GROUP = "group"
|
|
_ct.SUPERGROUP = "supergroup"
|
|
_tg.constants.ChatType = _ct
|
|
sys.modules.setdefault("telegram", _tg)
|
|
sys.modules.setdefault("telegram.constants", _tg.constants)
|
|
sys.modules.setdefault("telegram.ext", types.ModuleType("telegram.ext"))
|
|
|
|
from gateway.config import Platform, PlatformConfig
|
|
from gateway.platforms.base import (
|
|
BasePlatformAdapter,
|
|
MessageEvent,
|
|
MessageType,
|
|
SendResult,
|
|
)
|
|
from gateway.session import SessionSource, build_session_key
|
|
|
|
|
|
def _make_event(
|
|
text: str,
|
|
chat_id: str = "12345",
|
|
*,
|
|
chat_type: str = "dm",
|
|
user_id: str = "u1",
|
|
user_name: str | None = None,
|
|
thread_id: str | None = None,
|
|
) -> MessageEvent:
|
|
source = SessionSource(
|
|
platform=Platform.TELEGRAM,
|
|
chat_id=chat_id,
|
|
chat_type=chat_type,
|
|
user_id=user_id,
|
|
user_name=user_name,
|
|
thread_id=thread_id,
|
|
)
|
|
return MessageEvent(
|
|
text=text,
|
|
message_type=MessageType.TEXT,
|
|
source=source,
|
|
message_id=f"msg-{text[:8]}",
|
|
)
|
|
|
|
|
|
class _DummyAdapter(BasePlatformAdapter): # type: ignore[misc]
|
|
async def connect(self, *, is_reconnect: bool = False):
|
|
pass
|
|
|
|
async def disconnect(self):
|
|
pass
|
|
|
|
async def get_chat_info(self, chat_id):
|
|
return None
|
|
|
|
async def send(self, *args, **kwargs):
|
|
return SendResult(success=True, message_id="x")
|
|
|
|
|
|
def _make_initialized_adapter() -> BasePlatformAdapter:
|
|
return _DummyAdapter(PlatformConfig(enabled=True, token="***"), Platform.TELEGRAM)
|
|
|
|
|
|
def _make_adapter() -> BasePlatformAdapter:
|
|
"""Build a BasePlatformAdapter without running its heavy __init__."""
|
|
adapter = object.__new__(_DummyAdapter)
|
|
adapter.config = PlatformConfig(enabled=True, token="***")
|
|
adapter.platform = Platform.TELEGRAM
|
|
adapter._message_handler = AsyncMock(return_value=None)
|
|
adapter._busy_session_handler = None
|
|
adapter._active_sessions = {}
|
|
adapter._pending_messages = {}
|
|
adapter._session_tasks = {}
|
|
adapter._background_tasks = set()
|
|
adapter._post_delivery_callbacks = {}
|
|
adapter._expected_cancelled_tasks = set()
|
|
adapter._fatal_error_code = None
|
|
adapter._fatal_error_message = None
|
|
adapter._fatal_error_retryable = True
|
|
adapter._fatal_error_handler = None
|
|
adapter._running = True
|
|
adapter._busy_text_mode = "queue"
|
|
adapter._busy_text_debounce_seconds = 0.1
|
|
adapter._busy_text_hard_cap_seconds = 1.0
|
|
adapter._text_debounce = {}
|
|
adapter._auto_tts_default = False
|
|
adapter._auto_tts_enabled_chats = set()
|
|
adapter._auto_tts_disabled_chats = set()
|
|
adapter._typing_paused = set()
|
|
return adapter
|
|
|
|
|
|
def _debounced_event(adapter: BasePlatformAdapter, session_key: str) -> MessageEvent:
|
|
return adapter._text_debounce[session_key].event
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_rapid_text_followups_accumulate_instead_of_replacing():
|
|
"""Rapid TEXT follow-ups must all survive in the pending event."""
|
|
adapter = _make_adapter()
|
|
adapter._busy_text_mode = "" # direct-merge behavior, no debounce
|
|
first = _make_event("part one")
|
|
session_key = build_session_key(first.source)
|
|
adapter._active_sessions[session_key] = asyncio.Event()
|
|
|
|
await adapter.handle_message(_make_event("part two"))
|
|
await adapter.handle_message(_make_event("part three"))
|
|
|
|
pending = adapter._pending_messages[session_key]
|
|
assert pending.text == "part two\npart three"
|
|
assert not adapter._active_sessions[session_key].is_set()
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_debounce_resets_timer_on_new_arrival():
|
|
adapter = _make_adapter()
|
|
adapter._busy_text_debounce_seconds = 0.1
|
|
|
|
first = _make_event("one")
|
|
session_key = build_session_key(first.source)
|
|
adapter._active_sessions[session_key] = asyncio.Event()
|
|
|
|
await adapter.handle_message(first)
|
|
task1 = adapter._text_debounce[session_key].task
|
|
assert task1 is not None
|
|
assert not task1.done()
|
|
|
|
await adapter.handle_message(_make_event("two"))
|
|
task2 = adapter._text_debounce[session_key].task
|
|
assert task2 is not None
|
|
assert task2 is not task1
|
|
await asyncio.sleep(0)
|
|
assert task1.cancelled() or task1.done()
|
|
assert adapter._text_debounce[session_key].task is task2
|
|
|
|
await adapter.handle_message(_make_event("three"))
|
|
task3 = adapter._text_debounce[session_key].task
|
|
assert task3 is not None
|
|
assert task3 is not task2
|
|
|
|
await asyncio.sleep(0.2)
|
|
assert session_key not in adapter._text_debounce
|
|
assert adapter._pending_messages[session_key].text == "one\ntwo\nthree"
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_control_and_clarify_messages_bypass_text_debounce():
|
|
adapter = _make_adapter()
|
|
started: list[str] = []
|
|
|
|
def _fake_start(event, session_key, *, interrupt_event=None):
|
|
started.append(event.text)
|
|
return True
|
|
|
|
adapter._start_session_processing = _fake_start # type: ignore[method-assign]
|
|
|
|
await adapter.handle_message(_make_event("/status"))
|
|
assert started == ["/status"]
|
|
assert adapter._text_debounce == {}
|
|
|
|
answer = _make_event("clarify answer")
|
|
session_key = build_session_key(answer.source)
|
|
adapter._active_sessions[session_key] = asyncio.Event()
|
|
adapter._message_handler = AsyncMock(return_value=None)
|
|
|
|
with patch("tools.clarify_gateway.get_pending_for_session", return_value=object()):
|
|
await adapter.handle_message(answer)
|
|
|
|
adapter._message_handler.assert_awaited_once_with(answer)
|
|
assert session_key not in adapter._text_debounce
|
|
assert session_key not in adapter._pending_messages
|
|
|
|
|
|
def test_adapter_defaults_to_interrupt_mode(monkeypatch):
|
|
monkeypatch.delenv("HERMES_GATEWAY_BUSY_TEXT_MODE", raising=False)
|
|
adapter = _make_initialized_adapter()
|
|
assert adapter._busy_text_mode == "interrupt"
|
|
assert not adapter._is_queue_text_debounce_candidate(_make_event("hello"))
|
|
|
|
|
|
def test_command_messages_bypass_debounce_even_in_queue_mode():
|
|
adapter = _make_adapter()
|
|
assert not adapter._is_queue_text_debounce_candidate(_make_event(""))
|
|
assert not adapter._is_queue_text_debounce_candidate(_make_event("/stop"))
|
|
|
|
|