mirror of
https://github.com/NousResearch/hermes-agent.git
synced 2026-07-31 19:16:29 +00:00
39975613b1
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.
91 lines
3.5 KiB
Python
91 lines
3.5 KiB
Python
"""Regression tests for #15165 — gateway session shutdown must pass the
|
|
agent's conversation transcript to ``shutdown_memory_provider`` so memory
|
|
providers' ``on_session_end`` hooks see the real messages instead of an
|
|
empty list.
|
|
|
|
Before the fix, ``_cleanup_agent_resources`` called
|
|
``agent.shutdown_memory_provider()`` with no arguments, which in turn
|
|
invoked ``on_session_end([])`` on every memory provider. Providers with
|
|
an empty-guard (Holographic, Hindsight, etc.) exited early and never
|
|
persisted the session's facts, so the next gateway start-up surfaced no
|
|
memories from the prior conversation.
|
|
|
|
The fix reads ``agent._session_messages`` (set on ``AIAgent.__init__``
|
|
and refreshed every turn via ``_persist_session``) and forwards it to
|
|
``shutdown_memory_provider``. Test stubs built via ``object.__new__``
|
|
or plain ``MagicMock()`` still exercise the legacy no-arg path, so the
|
|
change is backward-compatible with existing suites.
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
import sys
|
|
import types
|
|
from unittest.mock import MagicMock
|
|
|
|
import pytest
|
|
|
|
|
|
@pytest.fixture(autouse=True)
|
|
def _mock_dotenv(monkeypatch):
|
|
"""gateway.run imports dotenv at module load; stub so tests run bare."""
|
|
fake = types.ModuleType("dotenv")
|
|
fake.load_dotenv = lambda *a, **kw: None
|
|
monkeypatch.setitem(sys.modules, "dotenv", fake)
|
|
|
|
|
|
def _make_runner():
|
|
from gateway.run import GatewayRunner
|
|
|
|
runner = object.__new__(GatewayRunner)
|
|
return runner
|
|
|
|
|
|
# A lightweight stand-in for AIAgent so ``isinstance(..., list)`` correctly
|
|
# discriminates between "attribute set to a list" and "attribute absent /
|
|
# MagicMock auto-synthesised". Using MagicMock directly for the agent
|
|
# would also work for the populated case, but attribute access on a
|
|
# MagicMock always yields a child MagicMock — we want a real Python
|
|
# object we can shape per-test.
|
|
class _FakeAgent:
|
|
def __init__(self, session_messages=None, has_shutdown=True):
|
|
if session_messages is not None:
|
|
self._session_messages = session_messages
|
|
if has_shutdown:
|
|
self.shutdown_memory_provider = MagicMock()
|
|
self.close = MagicMock()
|
|
|
|
|
|
class TestCleanupAgentResourcesPassesMessages:
|
|
"""_cleanup_agent_resources forwards the agent's session messages."""
|
|
|
|
def test_populated_messages_forwarded(self):
|
|
"""Real-world path: an agent that ran a turn has a populated
|
|
``_session_messages`` list and the cleanup call forwards it."""
|
|
runner = _make_runner()
|
|
transcript = [
|
|
{"role": "user", "content": "remember my dog is named Biscuit"},
|
|
{"role": "assistant", "content": "Got it — Biscuit."},
|
|
]
|
|
agent = _FakeAgent(session_messages=transcript)
|
|
|
|
runner._cleanup_agent_resources(agent)
|
|
|
|
# The fix must call shutdown_memory_provider with the exact list
|
|
# identity — providers iterate it to extract facts.
|
|
agent.shutdown_memory_provider.assert_called_once_with(transcript)
|
|
|
|
def test_empty_list_still_forwarded(self):
|
|
"""An agent that initialised but ran no turns has an empty list
|
|
on ``_session_messages``. Forwarding it (rather than falling
|
|
through to the no-arg path) makes the absence of content
|
|
explicit to providers and matches the pre-fix observable
|
|
behaviour (``on_session_end([])``)."""
|
|
runner = _make_runner()
|
|
agent = _FakeAgent(session_messages=[])
|
|
|
|
runner._cleanup_agent_resources(agent)
|
|
|
|
agent.shutdown_memory_provider.assert_called_once_with([])
|
|
|
|
|