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.
127 lines
4.2 KiB
Python
127 lines
4.2 KiB
Python
"""Tests for session handoff (CLI to gateway platform).
|
|
|
|
The handoff state machine lives on the ``sessions`` table:
|
|
|
|
None → "pending" → "running" → ("completed" | "failed")
|
|
|
|
CLI side calls ``request_handoff`` and poll-waits on ``get_handoff_state``.
|
|
Gateway side iterates ``list_pending_handoffs``, calls ``claim_handoff`` to
|
|
flip pending → running, and finishes with ``complete_handoff`` or
|
|
``fail_handoff``.
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
import time
|
|
|
|
import pytest
|
|
|
|
from hermes_state import SessionDB
|
|
|
|
|
|
class TestHandoffStateDB:
|
|
"""Test the handoff schema + helper methods on SessionDB."""
|
|
|
|
@pytest.fixture
|
|
def db(self, tmp_path, monkeypatch):
|
|
home = tmp_path / ".hermes"
|
|
home.mkdir()
|
|
monkeypatch.setenv("HERMES_HOME", str(home))
|
|
return SessionDB(db_path=home / "state.db")
|
|
|
|
def _make_session(self, db, session_id, source="cli", title=None):
|
|
"""Insert a session row directly for testing."""
|
|
def _do(conn):
|
|
conn.execute(
|
|
"INSERT OR IGNORE INTO sessions (id, source, title, started_at) "
|
|
"VALUES (?, ?, ?, ?)",
|
|
(session_id, source, title, time.time()),
|
|
)
|
|
db._execute_write(_do)
|
|
|
|
|
|
|
|
|
|
|
|
def test_list_pending_handoffs_excludes_running_and_terminal(self, db):
|
|
a, b, c, d = "sess-a", "sess-b", "sess-c", "sess-d"
|
|
for sid in (a, b, c, d):
|
|
self._make_session(db, sid)
|
|
|
|
db.request_handoff(a, "telegram")
|
|
db.request_handoff(b, "discord")
|
|
db.request_handoff(c, "telegram")
|
|
db.claim_handoff(c) # c is now running, not pending
|
|
db.request_handoff(d, "slack")
|
|
db.claim_handoff(d)
|
|
db.complete_handoff(d) # d is terminal
|
|
|
|
pending = db.list_pending_handoffs()
|
|
ids = [r["id"] for r in pending]
|
|
assert set(ids) == {a, b}
|
|
|
|
|
|
def test_complete_handoff_clears_error(self, db):
|
|
sid = "sess-complete"
|
|
self._make_session(db, sid)
|
|
db.request_handoff(sid, "telegram")
|
|
db.claim_handoff(sid)
|
|
db.fail_handoff(sid, "transient")
|
|
# User retries; mock the watcher path
|
|
db.request_handoff(sid, "telegram")
|
|
db.claim_handoff(sid)
|
|
db.complete_handoff(sid)
|
|
|
|
state = db.get_handoff_state(sid)
|
|
assert state["state"] == "completed"
|
|
assert state["error"] is None
|
|
|
|
|
|
|
|
|
|
def test_full_pending_to_completed_flow(self, db):
|
|
"""End-to-end sequence the CLI + gateway watcher follow."""
|
|
sid = "sess-flow"
|
|
self._make_session(db, sid, title="my session")
|
|
db.append_message(sid, "user", "Hello")
|
|
db.append_message(sid, "assistant", "Hi there!")
|
|
|
|
# CLI: request handoff
|
|
assert db.request_handoff(sid, "telegram") is True
|
|
assert db.get_handoff_state(sid)["state"] == "pending"
|
|
|
|
# Gateway watcher: discover + claim
|
|
pending = db.list_pending_handoffs()
|
|
assert len(pending) == 1
|
|
assert pending[0]["id"] == sid
|
|
assert db.claim_handoff(sid) is True
|
|
assert db.get_handoff_state(sid)["state"] == "running"
|
|
|
|
# Gateway uses get_messages to load the transcript (real flow uses
|
|
# session_store.switch_session which reads the same table).
|
|
messages = db.get_messages(sid)
|
|
assert [m["role"] for m in messages] == ["user", "assistant"]
|
|
|
|
# Gateway: mark completed
|
|
db.complete_handoff(sid)
|
|
assert db.get_handoff_state(sid)["state"] == "completed"
|
|
assert db.list_pending_handoffs() == []
|
|
|
|
|
|
class TestHandoffCommandRegistration:
|
|
"""Slash-command surface checks."""
|
|
|
|
def test_command_registered(self):
|
|
from hermes_cli.commands import resolve_command
|
|
cmd = resolve_command("handoff")
|
|
assert cmd is not None
|
|
assert cmd.name == "handoff"
|
|
assert cmd.category == "Session"
|
|
|
|
def test_command_is_cli_only(self):
|
|
"""`/handoff` is initiated from the CLI; gateway shouldn't expose it."""
|
|
from hermes_cli.commands import resolve_command, GATEWAY_KNOWN_COMMANDS
|
|
cmd = resolve_command("handoff")
|
|
assert cmd is not None
|
|
assert cmd.cli_only is True
|
|
assert "handoff" not in GATEWAY_KNOWN_COMMANDS
|