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.
362 lines
13 KiB
Python
362 lines
13 KiB
Python
"""Tests for Telegram inline keyboard approval buttons."""
|
|
|
|
import os
|
|
import sys
|
|
from pathlib import Path
|
|
from types import SimpleNamespace
|
|
from unittest.mock import AsyncMock, MagicMock, patch
|
|
|
|
import pytest
|
|
|
|
# ---------------------------------------------------------------------------
|
|
# Ensure the repo root is importable
|
|
# ---------------------------------------------------------------------------
|
|
_repo = str(Path(__file__).resolve().parents[2])
|
|
if _repo not in sys.path:
|
|
sys.path.insert(0, _repo)
|
|
|
|
|
|
# ---------------------------------------------------------------------------
|
|
# Minimal Telegram mock so TelegramAdapter can be imported
|
|
# ---------------------------------------------------------------------------
|
|
def _ensure_telegram_mock():
|
|
"""Wire up the minimal mocks required to import TelegramAdapter."""
|
|
if "telegram" in sys.modules and hasattr(sys.modules["telegram"], "__file__"):
|
|
return
|
|
|
|
mod = MagicMock()
|
|
mod.ext.ContextTypes.DEFAULT_TYPE = type(None)
|
|
mod.constants.ParseMode.MARKDOWN = "Markdown"
|
|
mod.constants.ParseMode.MARKDOWN_V2 = "MarkdownV2"
|
|
mod.constants.ParseMode.HTML = "HTML"
|
|
mod.constants.ChatType.PRIVATE = "private"
|
|
mod.constants.ChatType.GROUP = "group"
|
|
mod.constants.ChatType.SUPERGROUP = "supergroup"
|
|
mod.constants.ChatType.CHANNEL = "channel"
|
|
# Provide real exception classes so ``except (NetworkError, ...)`` in
|
|
# connect() doesn't blow up under xdist when this mock leaks.
|
|
mod.error.NetworkError = type("NetworkError", (OSError,), {})
|
|
mod.error.TimedOut = type("TimedOut", (OSError,), {})
|
|
mod.error.BadRequest = type("BadRequest", (Exception,), {})
|
|
|
|
for name in ("telegram", "telegram.ext", "telegram.constants", "telegram.request"):
|
|
sys.modules.setdefault(name, mod)
|
|
sys.modules.setdefault("telegram.error", mod.error)
|
|
|
|
|
|
_ensure_telegram_mock()
|
|
|
|
from plugins.platforms.telegram.adapter import TelegramAdapter
|
|
from gateway.config import Platform, PlatformConfig
|
|
|
|
|
|
def _make_adapter(extra=None):
|
|
"""Create a TelegramAdapter with mocked internals."""
|
|
config = PlatformConfig(enabled=True, token="test-token", extra=extra or {})
|
|
adapter = TelegramAdapter(config)
|
|
adapter._bot = AsyncMock()
|
|
adapter._app = MagicMock()
|
|
return adapter
|
|
|
|
|
|
class _AuthRunner:
|
|
"""Minimal runner shim for callback auth tests."""
|
|
|
|
def __init__(self, authorized: bool):
|
|
self.authorized = authorized
|
|
self.last_source = None
|
|
|
|
async def _handle_message(self, event):
|
|
return None
|
|
|
|
def _is_user_authorized(self, source):
|
|
self.last_source = source
|
|
return self.authorized
|
|
|
|
|
|
# ===========================================================================
|
|
# send_exec_approval — inline keyboard buttons
|
|
# ===========================================================================
|
|
|
|
class TestTelegramExecApproval:
|
|
"""Test the send_exec_approval method sends InlineKeyboard buttons."""
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_sends_inline_keyboard(self):
|
|
adapter = _make_adapter()
|
|
mock_msg = MagicMock()
|
|
mock_msg.message_id = 42
|
|
adapter._bot.send_message = AsyncMock(return_value=mock_msg)
|
|
|
|
result = await adapter.send_exec_approval(
|
|
chat_id="12345",
|
|
command="rm -rf /important",
|
|
session_key="agent:main:telegram:group:12345:99",
|
|
description="dangerous deletion",
|
|
)
|
|
|
|
assert result.success is True
|
|
assert result.message_id == "42"
|
|
|
|
adapter._bot.send_message.assert_called_once()
|
|
kwargs = adapter._bot.send_message.call_args[1]
|
|
assert kwargs["chat_id"] == 12345
|
|
assert "rm -rf /important" in kwargs["text"]
|
|
assert "dangerous deletion" in kwargs["text"]
|
|
assert kwargs["reply_markup"] is not None # InlineKeyboardMarkup
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_non_smart_allow_permanent_false_keeps_session(self, monkeypatch):
|
|
adapter = _make_adapter()
|
|
adapter._bot.send_message = AsyncMock(return_value=SimpleNamespace(message_id=42))
|
|
buttons = []
|
|
monkeypatch.setattr(
|
|
"plugins.platforms.telegram.adapter.InlineKeyboardButton",
|
|
lambda text, callback_data: buttons.append(text) or text,
|
|
)
|
|
monkeypatch.setattr(
|
|
"plugins.platforms.telegram.adapter.InlineKeyboardMarkup", lambda rows: rows
|
|
)
|
|
|
|
await adapter.send_exec_approval(
|
|
chat_id="12345", command="curl example.test", session_key="s",
|
|
allow_permanent=False,
|
|
)
|
|
|
|
assert buttons == ["✅ Allow Once", "✅ Session", "❌ Deny"]
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_full_approval_keyboard_is_two_by_two(self, monkeypatch):
|
|
"""Regression: d48bf743f flattened all buttons into one row (4x1)."""
|
|
adapter = _make_adapter()
|
|
adapter._bot.send_message = AsyncMock(return_value=SimpleNamespace(message_id=42))
|
|
captured_rows = []
|
|
monkeypatch.setattr(
|
|
"plugins.platforms.telegram.adapter.InlineKeyboardButton",
|
|
lambda text, callback_data: text,
|
|
)
|
|
monkeypatch.setattr(
|
|
"plugins.platforms.telegram.adapter.InlineKeyboardMarkup",
|
|
lambda rows: captured_rows.extend(rows) or rows,
|
|
)
|
|
|
|
await adapter.send_exec_approval(
|
|
chat_id="12345", command="curl example.test", session_key="s",
|
|
)
|
|
|
|
assert captured_rows == [
|
|
["✅ Allow Once", "✅ Session"],
|
|
["✅ Always", "❌ Deny"],
|
|
]
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_smart_deny_two_buttons_share_one_row(self, monkeypatch):
|
|
"""smart_deny yields 2 buttons — they pair into a single readable row."""
|
|
adapter = _make_adapter()
|
|
adapter._bot.send_message = AsyncMock(return_value=SimpleNamespace(message_id=42))
|
|
captured_rows = []
|
|
monkeypatch.setattr(
|
|
"plugins.platforms.telegram.adapter.InlineKeyboardButton",
|
|
lambda text, callback_data: text,
|
|
)
|
|
monkeypatch.setattr(
|
|
"plugins.platforms.telegram.adapter.InlineKeyboardMarkup",
|
|
lambda rows: captured_rows.extend(rows) or rows,
|
|
)
|
|
|
|
await adapter.send_exec_approval(
|
|
chat_id="12345", command="curl example.test", session_key="s",
|
|
allow_permanent=False, smart_denied=True,
|
|
)
|
|
|
|
assert captured_rows == [
|
|
["✅ Allow Once", "❌ Deny"],
|
|
]
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_send_update_prompt_escapes_dynamic_prompt(self):
|
|
adapter = _make_adapter()
|
|
sent = {}
|
|
|
|
async def mock_send_message(**kwargs):
|
|
sent.update(kwargs)
|
|
return SimpleNamespace(message_id=55)
|
|
|
|
adapter._bot.send_message = AsyncMock(side_effect=mock_send_message)
|
|
|
|
result = await adapter.send_update_prompt(
|
|
chat_id="12345",
|
|
prompt="Fix [issue]_1 and verify *markdown*",
|
|
default="alpha_beta",
|
|
metadata={"thread_id": "999"},
|
|
)
|
|
|
|
assert result.success is True
|
|
assert "MARKDOWN_V2" in repr(sent["parse_mode"])
|
|
assert "Fix \\[issue\\]\\_1" in sent["text"]
|
|
assert "alpha\\_beta" in sent["text"]
|
|
|
|
# _handle_callback_query — approval button clicks
|
|
# ===========================================================================
|
|
|
|
class TestTelegramApprovalCallback:
|
|
"""Test the approval callback handling in _handle_callback_query."""
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_resume_typing_after_inline_approval(self):
|
|
"""Clicking an inline approval button must un-pause the chat's typing.
|
|
|
|
Regression for #27853: the text /approve path resumed typing, but the
|
|
ea: callback path did not, so the typing indicator stayed gone for the
|
|
rest of a long-running turn after a button click.
|
|
"""
|
|
adapter = _make_adapter()
|
|
adapter._approval_state[5] = "agent:main:telegram:group:12345:99"
|
|
adapter.pause_typing_for_chat("12345")
|
|
assert "12345" in adapter._typing_paused
|
|
|
|
query = AsyncMock()
|
|
query.data = "ea:once:5"
|
|
query.message = MagicMock()
|
|
query.message.chat_id = 12345
|
|
query.from_user = MagicMock()
|
|
query.from_user.first_name = "Norbert"
|
|
query.from_user.id = "12345"
|
|
query.answer = AsyncMock()
|
|
query.edit_message_text = AsyncMock()
|
|
|
|
update = MagicMock()
|
|
update.callback_query = query
|
|
context = MagicMock()
|
|
|
|
with patch.dict(os.environ, {"TELEGRAM_ALLOWED_USERS": "*"}, clear=False):
|
|
with patch("tools.approval.resolve_gateway_approval", return_value=1):
|
|
await adapter._handle_callback_query(update, context)
|
|
|
|
assert "12345" not in adapter._typing_paused
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_approval_callback_escapes_dynamic_user_name(self):
|
|
adapter = _make_adapter()
|
|
adapter._approval_state[3] = "agent:main:telegram:group:12345:99"
|
|
|
|
query = AsyncMock()
|
|
query.data = "ea:once:3"
|
|
query.message = MagicMock()
|
|
query.message.chat_id = 12345
|
|
query.from_user = MagicMock()
|
|
query.from_user.first_name = "Alice_Bob"
|
|
query.answer = AsyncMock()
|
|
query.edit_message_text = AsyncMock()
|
|
|
|
update = MagicMock()
|
|
update.callback_query = query
|
|
context = MagicMock()
|
|
query.from_user.id = "12345"
|
|
|
|
with patch.dict(os.environ, {"TELEGRAM_ALLOWED_USERS": "*"}, clear=False):
|
|
with patch("tools.approval.resolve_gateway_approval", return_value=1):
|
|
await adapter._handle_callback_query(update, context)
|
|
|
|
edit_kwargs = query.edit_message_text.call_args[1]
|
|
assert "MARKDOWN_V2" in repr(edit_kwargs["parse_mode"])
|
|
assert "Alice\\_Bob" in edit_kwargs["text"]
|
|
assert "Approved once" in edit_kwargs["text"]
|
|
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_update_prompt_callback_not_affected(self, tmp_path):
|
|
"""Ensure update prompt callbacks still work."""
|
|
adapter = _make_adapter()
|
|
|
|
query = AsyncMock()
|
|
query.data = "update_prompt:y"
|
|
query.message = MagicMock()
|
|
query.message.chat_id = 12345
|
|
query.from_user = MagicMock()
|
|
query.from_user.id = 123
|
|
query.answer = AsyncMock()
|
|
query.edit_message_text = AsyncMock()
|
|
|
|
update = MagicMock()
|
|
update.callback_query = query
|
|
context = MagicMock()
|
|
|
|
with patch("tools.approval.resolve_gateway_approval") as mock_resolve:
|
|
with patch("hermes_constants.get_hermes_home", return_value=tmp_path):
|
|
# Allow the caller — the new fail-closed allowlist gate
|
|
# (#24457) rejects empty TELEGRAM_ALLOWED_USERS, but this
|
|
# test isn't exercising that gate; it's verifying the
|
|
# update_prompt callback still writes the response.
|
|
with patch.dict(os.environ, {"TELEGRAM_ALLOWED_USERS": "*"}):
|
|
await adapter._handle_callback_query(update, context)
|
|
|
|
# Should NOT have triggered approval resolution
|
|
mock_resolve.assert_not_called()
|
|
assert (tmp_path / ".update_response").read_text() == "y"
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_update_prompt_callback_rejects_unauthorized_user(self, tmp_path):
|
|
"""Update prompt buttons should honor TELEGRAM_ALLOWED_USERS."""
|
|
adapter = _make_adapter()
|
|
|
|
query = AsyncMock()
|
|
query.data = "update_prompt:y"
|
|
query.message = MagicMock()
|
|
query.message.chat_id = 12345
|
|
query.from_user = MagicMock()
|
|
query.from_user.id = 222
|
|
query.answer = AsyncMock()
|
|
query.edit_message_text = AsyncMock()
|
|
|
|
update = MagicMock()
|
|
update.callback_query = query
|
|
context = MagicMock()
|
|
|
|
with patch("hermes_constants.get_hermes_home", return_value=tmp_path):
|
|
with patch.dict(os.environ, {"TELEGRAM_ALLOWED_USERS": "111"}):
|
|
await adapter._handle_callback_query(update, context)
|
|
|
|
query.answer.assert_called_once()
|
|
assert "not authorized" in query.answer.call_args[1]["text"].lower()
|
|
query.edit_message_text.assert_not_called()
|
|
assert not (tmp_path / ".update_response").exists()
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_update_prompt_callback_rejects_user_blocked_by_global_allowlist(self, tmp_path):
|
|
adapter = _make_adapter()
|
|
runner = _AuthRunner(authorized=False)
|
|
adapter._message_handler = runner._handle_message
|
|
|
|
query = AsyncMock()
|
|
query.data = "update_prompt:y"
|
|
query.message = MagicMock()
|
|
query.message.chat_id = 12345
|
|
query.message.chat.type = "private"
|
|
query.from_user = MagicMock()
|
|
query.from_user.id = 222
|
|
query.from_user.first_name = "Mallory"
|
|
query.answer = AsyncMock()
|
|
query.edit_message_text = AsyncMock()
|
|
|
|
update = MagicMock()
|
|
update.callback_query = query
|
|
context = MagicMock()
|
|
|
|
with patch("hermes_constants.get_hermes_home", return_value=tmp_path):
|
|
with patch.dict(os.environ, {"TELEGRAM_ALLOWED_USERS": ""}):
|
|
await adapter._handle_callback_query(update, context)
|
|
|
|
query.answer.assert_called_once()
|
|
assert "not authorized" in query.answer.call_args[1]["text"].lower()
|
|
query.edit_message_text.assert_not_called()
|
|
assert not (tmp_path / ".update_response").exists()
|
|
assert runner.last_source is not None
|
|
assert runner.last_source.platform == Platform.TELEGRAM
|
|
assert runner.last_source.user_id == "222"
|
|
|