fix: address PR #814 reviewer follow-ups

- gateway_client: guard token cache with a lazy anyio.Lock so concurrent
  embed calls share one M2M token request instead of racing
- status subscriber: distinguish idle fetch timeouts from real broker
  errors (log + 5s backoff) instead of swallowing all and spinning
- nats: warn when the bus URL uses unencrypted transport (non-tls://)
- collection_metadata: accept an optional shared httpx client, make TLS
  verify explicit, document the unauthenticated control-plane contract
- replace python -O-stripped asserts with explicit ValueError in the bus
  status builder and the api metadata source
- document why the nil-UUID sentinel point can't collide with content ids

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
Chris Coutinho
2026-05-31 19:42:09 +02:00
co-authored by Claude Opus 4.8
parent b5ed1e3b4d
commit 2d845cb70f
8 changed files with 209 additions and 37 deletions
@@ -116,3 +116,47 @@ async def test_token_provider_caches_and_refreshes(monkeypatch):
t3 = await tp.get_token()
assert t3 == "tok2"
assert calls["n"] == 2
async def test_token_provider_concurrent_callers_issue_single_request(monkeypatch):
"""Two concurrent get_token() calls must share one token request, not race."""
import anyio
calls = {"n": 0}
async def handler(request: httpx.Request) -> httpx.Response:
calls["n"] += 1
# Hold the "network" open so a second caller arrives mid-flight.
await anyio.sleep(0.05)
return httpx.Response(
200, json={"access_token": f"tok{calls['n']}", "expires_in": 3600}
)
transport = httpx.MockTransport(handler)
orig_async_client = httpx.AsyncClient
def _client(*args, **kwargs):
kwargs["transport"] = transport
return orig_async_client(*args, **kwargs)
monkeypatch.setattr(httpx, "AsyncClient", _client)
tp = GatewayTokenProvider(
token_url="https://idp.example/oauth2/token",
client_id="cid",
client_secret="sec",
)
results: list[str] = []
async def _fetch():
results.append(await tp.get_token())
async with anyio.create_task_group() as tg:
tg.start_soon(_fetch)
tg.start_soon(_fetch)
# The lock serialises the check-then-fetch cycle: only one HTTP request,
# and both callers observe the same cached token.
assert calls["n"] == 1
assert results == ["tok1", "tok1"]
@@ -70,6 +70,29 @@ async def test_api_source(mocker):
assert meta["embedding_identity"] == "amazon.titan-embed-text-v2:0"
async def test_api_source_uses_shared_http_client():
"""When a caller passes http_client, the read reuses it (no own client)."""
def handler(request: httpx.Request) -> httpx.Response:
assert request.url.path == "/v1/qdrant-collections/col/metadata"
return httpx.Response(200, json={"embedding_identity": "mistral-embed"})
shared = httpx.AsyncClient(transport=httpx.MockTransport(handler))
async with shared:
meta = await cm._read_from_api("https://cp", "col", client=shared)
assert meta["embedding_identity"] == "mistral-embed"
async def test_api_missing_url_falls_back_to_env(mocker):
"""COLLECTION_METADATA_SOURCE=api with no URL must not crash the query path."""
# Bypass Settings validation to simulate a -O / mutated-state edge case.
settings = Settings()
settings.collection_metadata_source = "api"
settings.collection_metadata_api_url = None
meta = await cm.read_collection_metadata(mocker.AsyncMock(), "col", settings)
assert meta == cm.env_default_metadata(settings)
async def test_upsert_sentinel_builds_point(mocker):
client = mocker.AsyncMock()
await cm.upsert_sentinel(
+19
View File
@@ -12,6 +12,7 @@ from nextcloud_mcp_server.vector.queue.nats import (
NatsTaskProducer,
_modified_at_rfc3339,
msg_id,
warn_if_insecure_nats_url,
)
from nextcloud_mcp_server.vector.queue.postgres import PostgresTaskProducer
from nextcloud_mcp_server.vector.scanner import DocumentTask
@@ -134,3 +135,21 @@ def test_transport_for(url, expected):
async def test_postgres_producer_is_a_seam():
with pytest.raises(NotImplementedError, match="documented seam"):
await PostgresTaskProducer.connect(object())
@pytest.mark.parametrize(
"url,should_warn",
[
("nats://nats:4222", True),
("ws://nats:8080", True),
("tls://nats:4222", False),
("wss://nats:8080", False),
],
)
def test_warn_if_insecure_nats_url(url, should_warn, caplog):
import logging
with caplog.at_level(logging.WARNING):
warn_if_insecure_nats_url(url)
warned = any("unencrypted transport" in r.getMessage() for r in caplog.records)
assert warned is should_warn