mirror of
https://github.com/NousResearch/hermes-agent.git
synced 2026-07-31 19:16:29 +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.
143 lines
5 KiB
Python
143 lines
5 KiB
Python
"""Regression tests for task/session cwd propagation in terminal_tool."""
|
|
|
|
import json
|
|
from types import SimpleNamespace
|
|
|
|
import tools.terminal_tool as terminal_tool
|
|
|
|
|
|
def _minimal_terminal_config(cwd="/default"):
|
|
return {
|
|
"env_type": "local",
|
|
"cwd": cwd,
|
|
"timeout": 60,
|
|
"lifetime_seconds": 3600,
|
|
}
|
|
|
|
|
|
def test_foreground_command_uses_registered_task_cwd_for_existing_environment(monkeypatch):
|
|
"""ACP can update task cwd after the local env exists; foreground must honor it."""
|
|
calls = []
|
|
|
|
class FakeEnv:
|
|
env = {}
|
|
|
|
def execute(self, command, **kwargs):
|
|
calls.append((command, kwargs))
|
|
return {"output": "ok", "returncode": 0}
|
|
|
|
task_id = "acp-session-1"
|
|
monkeypatch.setattr(terminal_tool, "_active_environments", {task_id: FakeEnv()})
|
|
monkeypatch.setattr(terminal_tool, "_last_activity", {})
|
|
monkeypatch.setattr(terminal_tool, "_task_env_overrides", {task_id: {"cwd": "/workspace/acp"}})
|
|
monkeypatch.setattr(terminal_tool, "_get_env_config", lambda: _minimal_terminal_config())
|
|
monkeypatch.setattr(
|
|
terminal_tool,
|
|
"_check_all_guards",
|
|
lambda command, env_type, **kwargs: {"approved": True},
|
|
)
|
|
|
|
result = json.loads(terminal_tool.terminal_tool(command="pwd", task_id=task_id))
|
|
|
|
assert result["exit_code"] == 0
|
|
assert calls == [("pwd", {"timeout": 60, "cwd": "/workspace/acp", "bounded_capture": True})]
|
|
|
|
|
|
def test_explicit_workdir_still_wins_over_registered_task_cwd(monkeypatch):
|
|
calls = []
|
|
|
|
class FakeEnv:
|
|
env = {}
|
|
|
|
def execute(self, command, **kwargs):
|
|
calls.append(kwargs)
|
|
return {"output": "ok", "returncode": 0}
|
|
|
|
task_id = "acp-session-1"
|
|
monkeypatch.setattr(terminal_tool, "_active_environments", {task_id: FakeEnv()})
|
|
monkeypatch.setattr(terminal_tool, "_last_activity", {})
|
|
monkeypatch.setattr(terminal_tool, "_task_env_overrides", {task_id: {"cwd": "/workspace/acp"}})
|
|
monkeypatch.setattr(terminal_tool, "_get_env_config", lambda: _minimal_terminal_config())
|
|
monkeypatch.setattr(
|
|
terminal_tool,
|
|
"_check_all_guards",
|
|
lambda command, env_type, **kwargs: {"approved": True},
|
|
)
|
|
|
|
result = json.loads(
|
|
terminal_tool.terminal_tool(
|
|
command="pwd",
|
|
task_id=task_id,
|
|
workdir="/explicit/workdir",
|
|
)
|
|
)
|
|
|
|
assert result["exit_code"] == 0
|
|
assert calls == [{"timeout": 60, "cwd": "/explicit/workdir", "bounded_capture": True}]
|
|
|
|
|
|
def test_background_command_prefers_recorded_session_cwd_over_init_time_cwd(monkeypatch):
|
|
"""Background process launches must also use the recorded session cwd."""
|
|
|
|
class FakeEnv:
|
|
env = {}
|
|
cwd = "/workspace/live"
|
|
|
|
class FakeRegistry:
|
|
def __init__(self):
|
|
self.calls = []
|
|
self.pending_watchers = []
|
|
|
|
def spawn_local(self, **kwargs):
|
|
self.calls.append(kwargs)
|
|
return SimpleNamespace(id="proc_test", pid=1234)
|
|
|
|
import tools.process_registry as process_registry_mod
|
|
|
|
registry = FakeRegistry()
|
|
task_id = "session-live-cwd-bg"
|
|
monkeypatch.setattr(terminal_tool, "_active_environments", {task_id: FakeEnv()})
|
|
monkeypatch.setattr(terminal_tool, "_last_activity", {})
|
|
monkeypatch.setattr(terminal_tool, "_session_cwd", {})
|
|
monkeypatch.setattr(terminal_tool, "_task_env_overrides", {task_id: {"cwd": "/workspace/init"}})
|
|
monkeypatch.setattr(terminal_tool, "_get_env_config", lambda: _minimal_terminal_config(cwd="/workspace/init"))
|
|
monkeypatch.setattr(terminal_tool, "_start_cleanup_thread", lambda: None)
|
|
monkeypatch.setattr(terminal_tool, "_resolve_container_task_id", lambda value: value or "default")
|
|
monkeypatch.setattr(
|
|
terminal_tool,
|
|
"_check_all_guards",
|
|
lambda command, env_type, **kwargs: {"approved": True},
|
|
)
|
|
monkeypatch.setattr(process_registry_mod, "process_registry", registry)
|
|
terminal_tool.record_session_cwd(task_id, "/workspace/live")
|
|
|
|
result = json.loads(
|
|
terminal_tool.terminal_tool(
|
|
command="sleep 1",
|
|
task_id=task_id,
|
|
background=True,
|
|
)
|
|
)
|
|
|
|
assert result["exit_code"] == 0
|
|
# session_key falls back to the raw task_id when no gateway contextvar is set
|
|
# (it doesn't propagate to tool-worker threads), so process.kill / stop can
|
|
# still find and terminate this background process.
|
|
assert registry.calls == [{
|
|
"command": "sleep 1",
|
|
"cwd": "/workspace/live",
|
|
"task_id": task_id,
|
|
"session_key": task_id,
|
|
"env_vars": {},
|
|
"use_pty": False,
|
|
}]
|
|
|
|
|
|
def test_safe_getcwd_falls_back_to_home_when_no_terminal_cwd(monkeypatch):
|
|
def _boom():
|
|
raise FileNotFoundError()
|
|
|
|
monkeypatch.setattr(terminal_tool.os, "getcwd", _boom)
|
|
monkeypatch.delenv("TERMINAL_CWD", raising=False)
|
|
monkeypatch.setattr(terminal_tool.os.path, "expanduser", lambda p: "/home/me")
|
|
assert terminal_tool._safe_getcwd() == "/home/me"
|