host
d7f0f1be96
Docker tasks do not inherit the worker process environment, they only get extra_env (-e). SEARXNG_URL in the worker unit EnvironmentFile therefore never reached codex -> MCP inside the container, so the auto chain skipped SearXNG and only DuckDuckGo was tried (live e2e: 'all search providers failed: duckduckgo (DNS)'). _docker_task_context now merges any set SEARXNG_URL / PERPLEXITY_API_KEY / EXA_API_KEY / TAVILY_API_KEY / BRAVE_API_KEY from the worker environment into extra_env (values never logged; paid keys follow the same extra_env path as gateway keys). Tests: 2 new docker-env tests; websearch + tenant isolation = 25 passing; full cloud suite 1038 passing (4 skipped).
166 lines
6.7 KiB
Python
166 lines
6.7 KiB
Python
"""Worker wiring for web search MCP: tenant seeding + egress filter.
|
|
|
|
No Docker, no network: _tenant_dirs runs against temp dirs, and the
|
|
egress check only reads the repo filter file.
|
|
"""
|
|
|
|
import os
|
|
import sys
|
|
import tempfile
|
|
import unittest
|
|
from pathlib import Path
|
|
|
|
try:
|
|
from cloud.server import init_db
|
|
from cloud.worker import Worker
|
|
except ImportError:
|
|
sys.path.insert(0, str(Path(__file__).resolve().parents[2]))
|
|
from cloud.server import init_db
|
|
from cloud.worker import Worker
|
|
|
|
REPO_ROOT = Path(__file__).resolve().parents[2]
|
|
|
|
|
|
class WorkerWebsearchTestCase(unittest.TestCase):
|
|
def setUp(self):
|
|
self.tmp = tempfile.TemporaryDirectory()
|
|
self.db_path = os.path.join(self.tmp.name, "control.db")
|
|
init_db(self.db_path)
|
|
self.template_home = os.path.join(self.tmp.name, "tpl-home")
|
|
os.makedirs(self.template_home)
|
|
open(os.path.join(self.template_home, "config.toml"),
|
|
"w").write('[model_provider.xcmapi]\n'
|
|
'base_url="https://xcmapi.org/v1"\n')
|
|
self.template_repo = os.path.join(self.tmp.name, "tpl-repo")
|
|
os.makedirs(os.path.join(self.template_repo, ".git"))
|
|
open(os.path.join(self.template_repo, "sample.py"),
|
|
"w").write("def add(a, b):\n return a + b\n")
|
|
|
|
def tearDown(self):
|
|
self.tmp.cleanup()
|
|
|
|
def _worker(self, **kwargs):
|
|
kwargs.setdefault("env", {"XCM_API_KEY": "test-placeholder"})
|
|
kwargs.setdefault("tenants_root",
|
|
os.path.join(self.tmp.name, "tenants"))
|
|
kwargs.setdefault("tenant_template_home", self.template_home)
|
|
kwargs.setdefault("tenant_template_repo", self.template_repo)
|
|
kwargs.setdefault("tenant_isolation", True)
|
|
return Worker(self.db_path, "/fake/host.py", **kwargs)
|
|
|
|
def test_flag_off_is_old_behaviour(self):
|
|
worker = self._worker()
|
|
self.assertFalse(worker.websearch_mcp)
|
|
dirs = worker._tenant_dirs("projA")
|
|
config = open(os.path.join(dirs["codex_home"],
|
|
"config.toml")).read()
|
|
self.assertNotIn("[mcp_servers.websearch]", config)
|
|
self.assertFalse(os.path.exists(
|
|
os.path.join(dirs["codex_home"], "websearch_mcp.py")))
|
|
|
|
def test_flag_on_seeds_container_path_and_scripts(self):
|
|
worker = self._worker(websearch_mcp=True)
|
|
dirs = worker._tenant_dirs("projA")
|
|
home = dirs["codex_home"]
|
|
config = open(os.path.join(home, "config.toml")).read()
|
|
self.assertIn("[mcp_servers.websearch]", config)
|
|
# Container path, not the host temp path: this config is
|
|
# read inside the task container, where the tenant home is
|
|
# mounted at /var/lib/codex-app-server/codex-home.
|
|
self.assertIn("/var/lib/codex-app-server/codex-home/"
|
|
"websearch_mcp.py", config)
|
|
self.assertNotIn(self.tmp.name, config)
|
|
self.assertTrue(os.path.exists(
|
|
os.path.join(home, "websearch.py")))
|
|
self.assertTrue(os.path.exists(
|
|
os.path.join(home, "websearch_mcp.py")))
|
|
# The copied MCP script is the real one (not a stub).
|
|
script = open(os.path.join(home, "websearch_mcp.py")).read()
|
|
self.assertIn("web_search", script)
|
|
self.assertIn("web_fetch", script)
|
|
|
|
def test_old_tenant_merges_on_next_task(self):
|
|
worker = self._worker()
|
|
dirs = worker._tenant_dirs("projA")
|
|
config_path = os.path.join(dirs["codex_home"], "config.toml")
|
|
self.assertNotIn("[mcp_servers.websearch]",
|
|
open(config_path).read())
|
|
# Flag flips on later (deploy enables it): next task's
|
|
# _tenant_dirs must merge, not require a new project.
|
|
worker2 = self._worker(websearch_mcp=True)
|
|
worker2._tenant_dirs("projA")
|
|
config = open(config_path).read()
|
|
self.assertIn("[mcp_servers.websearch]", config)
|
|
self.assertIn("xcmapi.org", config)
|
|
self.assertEqual(config.count("[mcp_servers.websearch]"), 1)
|
|
# Third call stays idempotent.
|
|
worker2._tenant_dirs("projA")
|
|
config = open(config_path).read()
|
|
self.assertEqual(config.count("[mcp_servers.websearch]"), 1)
|
|
|
|
|
|
class DockerEnvTestCase(unittest.TestCase):
|
|
def setUp(self):
|
|
self.tmp = tempfile.TemporaryDirectory()
|
|
self.db_path = os.path.join(self.tmp.name, "control.db")
|
|
init_db(self.db_path)
|
|
|
|
def tearDown(self):
|
|
self.tmp.cleanup()
|
|
|
|
def _docker_context(self):
|
|
worker = Worker(self.db_path, "/fake/host.py",
|
|
env={"XCM_API_KEY": "test-placeholder"},
|
|
tenant_isolation=False, egress_proxy_port=None)
|
|
return worker._docker_task_context({"project_id": "projA"})
|
|
|
|
def test_searxng_url_passes_into_container_env(self):
|
|
from unittest import mock
|
|
with mock.patch.dict(os.environ,
|
|
{"SEARXNG_URL": "http://searx.diante.me"}):
|
|
ok, _, _, env, err = self._docker_context()
|
|
self.assertTrue(ok)
|
|
self.assertIsNone(err)
|
|
self.assertEqual(env.get("SEARXNG_URL"), "http://searx.diante.me")
|
|
|
|
def test_no_search_env_means_no_extra_env(self):
|
|
from unittest import mock
|
|
with mock.patch.dict(os.environ, {}, clear=False):
|
|
for k in ("SEARXNG_URL", "PERPLEXITY_API_KEY", "EXA_API_KEY",
|
|
"TAVILY_API_KEY", "BRAVE_API_KEY"):
|
|
os.environ.pop(k, None)
|
|
ok, _, _, env, err = self._docker_context()
|
|
self.assertTrue(ok)
|
|
self.assertIsNone(err)
|
|
self.assertIsNone(env)
|
|
|
|
|
|
class EgressFilterTestCase(unittest.TestCase):
|
|
def test_search_domains_present_in_basic_form(self):
|
|
lines = open(REPO_ROOT / "deploy" / "egress" /
|
|
"tinyproxy-filter").read().splitlines()
|
|
entries = [ln for ln in lines
|
|
if ln and not ln.startswith("#")]
|
|
# No extended-regex syntax in any entry: + ? | and bare
|
|
# parens are literal in BASIC and would break the allowlist.
|
|
for entry in entries:
|
|
self.assertNotIn("+", entry)
|
|
self.assertNotIn("?", entry)
|
|
self.assertNotIn("|", entry)
|
|
self.assertNotIn("(", entry.replace("\\(", ""))
|
|
self.assertNotIn(")", entry.replace("\\)", ""))
|
|
expected = [
|
|
r"^\(.*\.\)*perplexity\.ai$",
|
|
r"^\(.*\.\)*exa\.ai$",
|
|
r"^\(.*\.\)*tavily\.com$",
|
|
r"^\(.*\.\)*search\.brave\.com$",
|
|
r"^\(.*\.\)*duckduckgo\.com$",
|
|
r"^searx\.diante\.me$",
|
|
]
|
|
for want in expected:
|
|
self.assertIn(want, entries, "missing filter entry: %s" % want)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
unittest.main()
|