# Conflicts: # gateway/config.py # gateway/config_loader.py # gateway/readiness.py # hermes_cli/managed_scope.py # hermes_cli/plugin_python_deps.py # hermes_cli/plugins_cmd.py # hermes_cli/update_cmd_maint.py # plugin-catalog/hindsight.yaml # plugins/plugin_loader.py # providers/__init__.py # scripts/run_tests.sh # tests/gateway/test_control_socket_windows_live.py # tests/gateway/test_gateway_streaming_nested_config.py # tests/hermes_cli/test_doctor.py # tests/hermes_cli/test_plan_reconciliation_windows_live.py # tests/hermes_cli/test_update_apply_shallow_count.py # tests/hermes_cli/test_update_concurrent_quarantine.py # tests/hermes_cli/test_update_shim_self_lock.py # tests/hermes_cli/test_verify_console_scripts.py # tests/tools/test_lazy_deps.py # tests/tui_gateway/test_subprocess_encoding.py # tools/lazy_deps.py
541 lines
20 KiB
Python
541 lines
20 KiB
Python
"""Tests for gateway proxy mode — forwarding messages to a remote API server."""
|
|
|
|
import types
|
|
from unittest.mock import AsyncMock, MagicMock, patch
|
|
|
|
import pytest
|
|
|
|
from gateway.config import Platform, StreamingConfig
|
|
from gateway.platforms.base import resolve_proxy_url
|
|
from gateway.run import GatewayRunner
|
|
from gateway.session import SessionSource
|
|
|
|
def _make_runner(proxy_url=None):
|
|
"""Create a minimal GatewayRunner for proxy tests."""
|
|
runner = object.__new__(GatewayRunner)
|
|
runner.adapters = {}
|
|
runner.config = MagicMock()
|
|
runner.config.streaming = StreamingConfig()
|
|
runner._running_agents = {}
|
|
runner._session_run_generation = {}
|
|
runner._session_model_overrides = {}
|
|
runner._agent_cache = {}
|
|
runner._agent_cache_lock = None
|
|
return runner
|
|
|
|
def _make_source(platform=Platform.MATRIX):
|
|
return SessionSource(
|
|
platform=platform,
|
|
chat_id="!room:server.org",
|
|
chat_name="Test Room",
|
|
chat_type="group",
|
|
user_id="@user:server.org",
|
|
user_name="testuser",
|
|
thread_id=None,
|
|
)
|
|
|
|
class _FakeSSEResponse:
|
|
"""Simulates an aiohttp response with SSE streaming."""
|
|
|
|
def __init__(self, status=200, sse_chunks=None, error_text=""):
|
|
self.status = status
|
|
self._sse_chunks = sse_chunks or []
|
|
self._error_text = error_text
|
|
self.content = self
|
|
|
|
async def text(self):
|
|
return self._error_text
|
|
|
|
async def iter_any(self):
|
|
for chunk in self._sse_chunks:
|
|
if isinstance(chunk, str):
|
|
chunk = chunk.encode("utf-8")
|
|
yield chunk
|
|
|
|
async def __aenter__(self):
|
|
return self
|
|
|
|
async def __aexit__(self, *args):
|
|
pass
|
|
|
|
class _FakeSession:
|
|
"""Simulates an aiohttp.ClientSession with captured request args."""
|
|
|
|
def __init__(self, response):
|
|
self._response = response
|
|
self.captured_url = None
|
|
self.captured_json = None
|
|
self.captured_headers = None
|
|
|
|
def post(self, url, json=None, headers=None, **kwargs):
|
|
self.captured_url = url
|
|
self.captured_json = json
|
|
self.captured_headers = headers
|
|
return self._response
|
|
|
|
async def __aenter__(self):
|
|
return self
|
|
|
|
async def __aexit__(self, *args):
|
|
pass
|
|
|
|
def _patch_aiohttp(session):
|
|
"""Install the optional aiohttp boundary without requiring the extra."""
|
|
module = types.SimpleNamespace(
|
|
ClientSession=MagicMock(return_value=session),
|
|
ClientTimeout=MagicMock(),
|
|
)
|
|
return patch.dict("sys.modules", {"aiohttp": module})
|
|
|
|
class TestGetProxyUrl:
|
|
"""Test _get_proxy_url() config resolution."""
|
|
|
|
def test_returns_none_when_not_configured(self, monkeypatch):
|
|
monkeypatch.delenv("GATEWAY_PROXY_URL", raising=False)
|
|
runner = _make_runner()
|
|
with patch("gateway.run._load_gateway_config", return_value={}):
|
|
assert runner._get_proxy_url() is None
|
|
|
|
def test_reads_from_config_yaml(self, monkeypatch):
|
|
monkeypatch.delenv("GATEWAY_PROXY_URL", raising=False)
|
|
runner = _make_runner()
|
|
cfg = {"gateway": {"proxy_url": "http://10.0.0.1:8642"}}
|
|
with patch("gateway.run._load_gateway_config", return_value=cfg):
|
|
assert runner._get_proxy_url() == "http://10.0.0.1:8642"
|
|
|
|
class _SelectiveScope(dict):
|
|
"""Bound scope that resolves GATEWAY_PROXY_URL but fails on the KEY read."""
|
|
def get(self, name, default=None):
|
|
if name == "GATEWAY_PROXY_URL":
|
|
return "http://proxy.local:8642"
|
|
if name == "GATEWAY_PROXY_KEY":
|
|
raise RuntimeError("resolver boom")
|
|
return dict.get(self, name, default)
|
|
|
|
|
|
class TestProxyKeyScopeFailure:
|
|
"""The proxy key read must propagate a bound-scope failure -- the ambient env
|
|
may hold another profile's credential (pre-fix: ``except Exception -> os.getenv``)."""
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_proxy_key_scope_failure_never_borrows_env(self, monkeypatch):
|
|
from agent import secret_scope as ss
|
|
|
|
monkeypatch.setenv("GATEWAY_PROXY_KEY", "foreign-key")
|
|
runner = _make_runner()
|
|
runner._run_still_current_fn = lambda *a, **k: True
|
|
|
|
ss.set_multiplex_active(True)
|
|
token = ss.set_secret_scope(_SelectiveScope())
|
|
try:
|
|
with _patch_aiohttp(MagicMock()):
|
|
with pytest.raises(RuntimeError, match="resolver boom"):
|
|
await runner._run_agent_via_proxy("hi", "ctx", [], _make_source(), "sess-1")
|
|
finally:
|
|
ss.reset_secret_scope(token)
|
|
ss.set_multiplex_active(False)
|
|
|
|
|
|
class TestResolveProxyUrl:
|
|
|
|
def test_no_proxy_bypasses_matching_host(self, monkeypatch):
|
|
for key in ("HTTPS_PROXY", "HTTP_PROXY", "ALL_PROXY",
|
|
"https_proxy", "http_proxy", "all_proxy", "NO_PROXY", "no_proxy"):
|
|
monkeypatch.delenv(key, raising=False)
|
|
monkeypatch.setenv("HTTPS_PROXY", "http://proxy.example:8080")
|
|
monkeypatch.setenv("NO_PROXY", "api.telegram.org")
|
|
|
|
assert resolve_proxy_url(target_hosts="api.telegram.org") is None
|
|
|
|
def test_no_proxy_bypasses_cidr_target(self, monkeypatch):
|
|
for key in ("HTTPS_PROXY", "HTTP_PROXY", "ALL_PROXY",
|
|
"https_proxy", "http_proxy", "all_proxy", "NO_PROXY", "no_proxy"):
|
|
monkeypatch.delenv(key, raising=False)
|
|
monkeypatch.setenv("HTTPS_PROXY", "http://proxy.example:8080")
|
|
monkeypatch.setenv("NO_PROXY", "149.154.160.0/20")
|
|
|
|
assert resolve_proxy_url(target_hosts=["149.154.167.220"]) is None
|
|
|
|
@pytest.mark.platforms("macos")
|
|
class TestMacosProxyProbeCache:
|
|
"""``scutil --proxy`` is a ~11 ms fork and resolve_proxy_url runs it on the SEND path —
|
|
per chunk of an outbound message and per media attachment."""
|
|
|
|
SCUTIL_OUT = "<dictionary> {\n HTTPEnable : 1\n HTTPProxy : 10.0.0.1\n HTTPPort : 3128\n}"
|
|
|
|
@pytest.fixture(autouse=True)
|
|
def _isolate(self):
|
|
import gateway.platforms.base as base
|
|
base.reset_macos_proxy_cache()
|
|
yield
|
|
base.reset_macos_proxy_cache()
|
|
|
|
def _count_forks(self, monkeypatch):
|
|
import gateway.platforms.base as base
|
|
calls = []
|
|
|
|
def fake(*a, **kw):
|
|
calls.append(a)
|
|
return self.SCUTIL_OUT
|
|
monkeypatch.setattr(base.subprocess, "check_output", fake)
|
|
return base, calls
|
|
|
|
def test_repeated_probes_fork_scutil_once(self, monkeypatch):
|
|
base, calls = self._count_forks(monkeypatch)
|
|
results = [base._detect_macos_system_proxy() for _ in range(10)]
|
|
assert len(calls) == 1, f"expected 1 scutil fork for 10 probes, got {len(calls)}"
|
|
assert results == ["http://10.0.0.1:3128"] * 10
|
|
|
|
def test_expired_ttl_re_reads(self, monkeypatch):
|
|
base, calls = self._count_forks(monkeypatch)
|
|
clock = {"t": 1000.0}
|
|
monkeypatch.setattr(base.time, "monotonic", lambda: clock["t"])
|
|
base._detect_macos_system_proxy()
|
|
clock["t"] += base._MACOS_PROXY_TTL_SECONDS + 1
|
|
base._detect_macos_system_proxy()
|
|
assert len(calls) == 2
|
|
|
|
|
|
class TestRunAgentProxyDispatch:
|
|
"""Test that _run_agent() delegates to proxy when configured."""
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_run_agent_delegates_to_proxy(self, monkeypatch):
|
|
monkeypatch.setenv("GATEWAY_PROXY_URL", "http://host:8642")
|
|
runner = _make_runner()
|
|
source = _make_source()
|
|
|
|
expected_result = {
|
|
"final_response": "Hello from remote!",
|
|
"messages": [
|
|
{"role": "user", "content": "hi"},
|
|
{"role": "assistant", "content": "Hello from remote!"},
|
|
],
|
|
"api_calls": 1,
|
|
"tools": [],
|
|
}
|
|
|
|
runner._run_agent_via_proxy = AsyncMock(return_value=expected_result)
|
|
|
|
result = await runner._run_agent(
|
|
message="hi",
|
|
context_prompt="",
|
|
history=[],
|
|
source=source,
|
|
session_id="test-session-123",
|
|
session_key="test-key",
|
|
run_generation=7,
|
|
)
|
|
|
|
assert result["final_response"] == "Hello from remote!"
|
|
runner._run_agent_via_proxy.assert_called_once()
|
|
assert runner._run_agent_via_proxy.call_args.kwargs["run_generation"] == 7
|
|
|
|
class TestRunAgentViaProxy:
|
|
"""Test the actual proxy HTTP forwarding logic."""
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_builds_correct_request(self, monkeypatch):
|
|
monkeypatch.setenv("GATEWAY_PROXY_URL", "http://host:8642")
|
|
monkeypatch.setenv("GATEWAY_PROXY_KEY", "test-key-123")
|
|
runner = _make_runner()
|
|
source = _make_source()
|
|
|
|
resp = _FakeSSEResponse(
|
|
status=200,
|
|
sse_chunks=[
|
|
'data: {"choices":[{"delta":{"content":"Hello"}}]}\n\n'
|
|
'data: {"choices":[{"delta":{"content":" world"}}]}\n\n'
|
|
"data: [DONE]\n\n"
|
|
],
|
|
)
|
|
session = _FakeSession(resp)
|
|
|
|
with patch("gateway.run._load_gateway_config", return_value={}):
|
|
with _patch_aiohttp(session):
|
|
with patch("aiohttp.ClientTimeout"):
|
|
result = await runner._run_agent_via_proxy(
|
|
message="How are you?",
|
|
context_prompt="You are helpful.",
|
|
history=[
|
|
{"role": "user", "content": "Hello"},
|
|
{"role": "assistant", "content": "Hi there!"},
|
|
],
|
|
source=source,
|
|
session_id="session-abc",
|
|
)
|
|
|
|
# Verify request URL
|
|
assert session.captured_url == "http://host:8642/v1/chat/completions"
|
|
|
|
# Verify auth header
|
|
assert session.captured_headers["Authorization"] == "Bearer test-key-123"
|
|
|
|
# Verify session ID header
|
|
assert session.captured_headers["X-Hermes-Session-Id"] == "session-abc"
|
|
|
|
# Verify messages include system, history, and current message
|
|
messages = session.captured_json["messages"]
|
|
assert messages[0] == {"role": "system", "content": "You are helpful."}
|
|
assert messages[1] == {"role": "user", "content": "Hello"}
|
|
assert messages[2] == {"role": "assistant", "content": "Hi there!"}
|
|
assert messages[3] == {"role": "user", "content": "How are you?"}
|
|
|
|
# Verify streaming is requested
|
|
assert session.captured_json["stream"] is True
|
|
|
|
# Verify response was assembled
|
|
assert result["final_response"] == "Hello world"
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_handles_connection_error(self, monkeypatch):
|
|
monkeypatch.setenv("GATEWAY_PROXY_URL", "http://unreachable:8642")
|
|
monkeypatch.delenv("GATEWAY_PROXY_KEY", raising=False)
|
|
runner = _make_runner()
|
|
source = _make_source()
|
|
|
|
class _ErrorSession:
|
|
def post(self, *args, **kwargs):
|
|
raise ConnectionError("Connection refused")
|
|
|
|
async def __aenter__(self):
|
|
return self
|
|
|
|
async def __aexit__(self, *args):
|
|
pass
|
|
|
|
with patch("gateway.run._load_gateway_config", return_value={}):
|
|
with _patch_aiohttp(_ErrorSession()):
|
|
with patch("aiohttp.ClientTimeout"):
|
|
result = await runner._run_agent_via_proxy(
|
|
message="hi",
|
|
context_prompt="",
|
|
history=[],
|
|
source=source,
|
|
session_id="test",
|
|
)
|
|
|
|
assert "Connection refused" in result["final_response"]
|
|
assert result["api_calls"] == 0
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_no_system_message_when_context_empty(self, monkeypatch):
|
|
monkeypatch.setenv("GATEWAY_PROXY_URL", "http://host:8642")
|
|
monkeypatch.delenv("GATEWAY_PROXY_KEY", raising=False)
|
|
runner = _make_runner()
|
|
source = _make_source()
|
|
|
|
resp = _FakeSSEResponse(
|
|
status=200,
|
|
sse_chunks=[b'data: {"choices":[{"delta":{"content":"ok"}}]}\n\ndata: [DONE]\n\n'],
|
|
)
|
|
session = _FakeSession(resp)
|
|
|
|
with patch("gateway.run._load_gateway_config", return_value={}):
|
|
with _patch_aiohttp(session):
|
|
with patch("aiohttp.ClientTimeout"):
|
|
await runner._run_agent_via_proxy(
|
|
message="hello",
|
|
context_prompt="",
|
|
history=[],
|
|
source=source,
|
|
session_id="test",
|
|
)
|
|
|
|
# No system message should appear when context_prompt is empty
|
|
messages = session.captured_json["messages"]
|
|
assert len(messages) == 1
|
|
assert messages[0]["role"] == "user"
|
|
assert messages[0]["content"] == "hello"
|
|
|
|
class TestStreamingResilience:
|
|
"""Tests for SSE streaming robustness — hang avoidance and malformed-chunk tolerance."""
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_done_marker_stops_reading_trailing_chunks(self, monkeypatch):
|
|
"""After `[DONE]`, no further SSE chunks must be processed.
|
|
|
|
A buggy upstream that holds the connection open and streams more
|
|
chunks after `[DONE]` should not leak those chunks into the
|
|
response. Regression test for the inner `break` that only exited
|
|
the line-parse loop, leaving the outer chunk loop to keep reading
|
|
until sock_read timeout.
|
|
"""
|
|
monkeypatch.setenv("GATEWAY_PROXY_URL", "http://host:8642")
|
|
monkeypatch.delenv("GATEWAY_PROXY_KEY", raising=False)
|
|
runner = _make_runner()
|
|
source = _make_source()
|
|
|
|
# Content → [DONE] → MORE content. The trailing chunk must be
|
|
# dropped. With the pre-fix code it would be appended to
|
|
# full_response, since `break` only exited the inner loop.
|
|
resp = _FakeSSEResponse(
|
|
status=200,
|
|
sse_chunks=[
|
|
'data: {"choices":[{"delta":{"content":"Hello"}}]}\n',
|
|
'data: [DONE]\n',
|
|
'data: {"choices":[{"delta":{"content":" IGNORED"}}]}\n',
|
|
],
|
|
)
|
|
session = _FakeSession(resp)
|
|
|
|
with patch("gateway.run._load_gateway_config", return_value={}):
|
|
with _patch_aiohttp(session):
|
|
with patch("aiohttp.ClientTimeout"):
|
|
result = await runner._run_agent_via_proxy(
|
|
message="hi",
|
|
context_prompt="",
|
|
history=[],
|
|
source=source,
|
|
session_id="test",
|
|
)
|
|
|
|
assert result["final_response"] == "Hello"
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_residual_buffer_flushed_after_eof(self, monkeypatch):
|
|
"""A final SSE frame without a trailing newline must not be dropped.
|
|
|
|
The line loop only consumes complete lines; if the upstream's last
|
|
frame lacks the newline, its content sat in ``buffer`` at EOF and
|
|
was silently discarded (pi#8997's bug class). The residual buffer
|
|
is now flushed after the read loop.
|
|
"""
|
|
monkeypatch.setenv("GATEWAY_PROXY_URL", "http://host:8642")
|
|
monkeypatch.delenv("GATEWAY_PROXY_KEY", raising=False)
|
|
runner = _make_runner()
|
|
source = _make_source()
|
|
|
|
resp = _FakeSSEResponse(
|
|
status=200,
|
|
sse_chunks=[
|
|
'data: {"choices":[{"delta":{"content":"Hello"}}]}\n',
|
|
'data: {"choices":[{"delta":{"content":" world"}}]}', # no newline, then EOF
|
|
],
|
|
)
|
|
session = _FakeSession(resp)
|
|
|
|
with patch("gateway.run._load_gateway_config", return_value={}):
|
|
with _patch_aiohttp(session):
|
|
with patch("aiohttp.ClientTimeout"):
|
|
result = await runner._run_agent_via_proxy(
|
|
message="hi",
|
|
context_prompt="",
|
|
history=[],
|
|
source=source,
|
|
session_id="test",
|
|
)
|
|
|
|
assert result["final_response"] == "Hello world"
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_eof_without_done_and_no_content_is_an_error(self, monkeypatch):
|
|
"""Clean EOF with no [DONE] and no content must surface an error, not
|
|
an empty 'response'. With content, the partial text is kept (and the
|
|
truncation is logged) rather than thrown away."""
|
|
monkeypatch.setenv("GATEWAY_PROXY_URL", "http://host:8642")
|
|
monkeypatch.delenv("GATEWAY_PROXY_KEY", raising=False)
|
|
runner = _make_runner()
|
|
source = _make_source()
|
|
|
|
resp = _FakeSSEResponse(status=200, sse_chunks=[])
|
|
session = _FakeSession(resp)
|
|
|
|
with patch("gateway.run._load_gateway_config", return_value={}):
|
|
with _patch_aiohttp(session):
|
|
with patch("aiohttp.ClientTimeout"):
|
|
result = await runner._run_agent_via_proxy(
|
|
message="hi",
|
|
context_prompt="",
|
|
history=[],
|
|
source=source,
|
|
session_id="test",
|
|
)
|
|
|
|
assert result["final_response"]
|
|
assert result["api_calls"] == 0
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_client_timeout_sets_sock_connect(self, monkeypatch):
|
|
"""ClientTimeout must bound the TCP connect phase.
|
|
|
|
Without an explicit ``sock_connect``, an unreachable proxy host
|
|
hangs for the OS default (minutes) before failing. The fix sets
|
|
a short connect cap so the gateway surfaces the error quickly.
|
|
"""
|
|
monkeypatch.setenv("GATEWAY_PROXY_URL", "http://host:8642")
|
|
monkeypatch.delenv("GATEWAY_PROXY_KEY", raising=False)
|
|
runner = _make_runner()
|
|
source = _make_source()
|
|
|
|
resp = _FakeSSEResponse(status=200, sse_chunks=['data: [DONE]\n'])
|
|
session = _FakeSession(resp)
|
|
|
|
captured = {}
|
|
|
|
def _capture_timeout(**kwargs):
|
|
captured.update(kwargs)
|
|
return MagicMock()
|
|
|
|
with patch("gateway.run._load_gateway_config", return_value={}):
|
|
with _patch_aiohttp(session):
|
|
with patch("aiohttp.ClientTimeout", side_effect=_capture_timeout):
|
|
await runner._run_agent_via_proxy(
|
|
message="hi",
|
|
context_prompt="",
|
|
history=[],
|
|
source=source,
|
|
session_id="test",
|
|
)
|
|
|
|
assert "sock_connect" in captured, (
|
|
"ClientTimeout should set sock_connect to bound TCP connect"
|
|
)
|
|
assert 0 < captured["sock_connect"] <= 60, (
|
|
f"sock_connect should be a short, reasonable cap — got {captured['sock_connect']}"
|
|
)
|
|
|
|
@pytest.mark.asyncio
|
|
async def test_malformed_chunk_is_skipped_not_fatal(self, monkeypatch):
|
|
"""One bad SSE chunk must not abort the whole stream.
|
|
|
|
Pre-fix: `choices[0].get(...)` raised ``AttributeError`` when
|
|
``choices[0]`` was ``None``, escaping the narrow
|
|
``except json.JSONDecodeError`` and bubbling to the outer
|
|
``except Exception`` which returned whatever partial response
|
|
was accumulated. All later chunks were lost.
|
|
|
|
Post-fix: type guards + broader exception handling skip the bad
|
|
chunk and keep parsing.
|
|
"""
|
|
monkeypatch.setenv("GATEWAY_PROXY_URL", "http://host:8642")
|
|
monkeypatch.delenv("GATEWAY_PROXY_KEY", raising=False)
|
|
runner = _make_runner()
|
|
source = _make_source()
|
|
|
|
resp = _FakeSSEResponse(
|
|
status=200,
|
|
sse_chunks=[
|
|
'data: {"choices":[{"delta":{"content":"Hello"}}]}\n',
|
|
'data: {"choices":[null]}\n',
|
|
'data: {"choices":"wrong-type"}\n',
|
|
'data: {"choices":[{"delta":"wrong-type"}]}\n',
|
|
'data: {"choices":[{"delta":{"content":" world"}}]}\n',
|
|
'data: [DONE]\n',
|
|
],
|
|
)
|
|
session = _FakeSession(resp)
|
|
|
|
with patch("gateway.run._load_gateway_config", return_value={}):
|
|
with _patch_aiohttp(session):
|
|
with patch("aiohttp.ClientTimeout"):
|
|
result = await runner._run_agent_via_proxy(
|
|
message="hi",
|
|
context_prompt="",
|
|
history=[],
|
|
source=source,
|
|
session_id="test",
|
|
)
|
|
|
|
assert result["final_response"] == "Hello world"
|