fix(vector-sync): address PR review — dict guard, symmetric backstop, metrics

- purge route: 400 (not 500) on a valid-JSON non-object body
- scanner: backstop-purge admin-disabled note/news_item/deck_card points
  (their deletion-tracking lives inside the skipped scan_* fns), mirroring the
  files path; gated on a concrete allow-set so fail-open never deletes
- processor: record_ingest_dropped("admin_disabled") so consent-skipped index
  tasks are observable/alertable
- app.py: list /api/v1/vector-sync/purge in the endpoints log line
- capabilities: drop empty-string doc types; return frozenset throughout
- purge: document the count-before-delete approximation
- tests: non-object body -> 400, ProvisioningRequiredError -> 428, cache TTL
  expiry refetch, and the scanner consent backstop

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
Chris Coutinho
2026-06-16 00:54:50 +02:00
co-authored by Claude Opus 4.8
parent ef5b3f3873
commit 477fb02b0a
10 changed files with 258 additions and 43 deletions
+34 -20
View File
@@ -2,6 +2,8 @@
from __future__ import annotations
from unittest.mock import AsyncMock
import nextcloud_mcp_server.capabilities as cap
from nextcloud_mcp_server.capabilities import (
_parse_enabled_doc_types,
@@ -84,22 +86,19 @@ def test_is_doc_type_allowed_empty_set_blocks_all():
# ---------------------------------------------------------------------------
class _FakeClient:
def __init__(self, payload=None, raises: Exception | None = None):
self._payload = payload
self._raises = raises
self.calls = 0
async def capabilities(self):
self.calls += 1
if self._raises is not None:
raise self._raises
return self._payload
def _client(payload=None, raises: Exception | None = None) -> AsyncMock:
"""An object with an async ``capabilities()`` method (AsyncMock-backed)."""
m = AsyncMock()
if raises is not None:
m.capabilities.side_effect = raises
else:
m.capabilities.return_value = payload
return m
async def test_allowed_doc_types_parses_and_caches():
clear_cache()
client = _FakeClient(_payload(["note", "file"]))
client = _client(_payload(["note", "file"]))
first = await allowed_doc_types(client, "alice")
second = await allowed_doc_types(client, "alice")
@@ -107,38 +106,53 @@ async def test_allowed_doc_types_parses_and_caches():
assert first == frozenset({"note", "file"})
assert second == frozenset({"note", "file"})
# Second call served from the cache — only one OCS round-trip.
assert client.calls == 1
assert client.capabilities.await_count == 1
async def test_allowed_doc_types_missing_block_returns_none():
clear_cache()
client = _FakeClient({"ocs": {"data": {"capabilities": {}}}})
client = _client({"ocs": {"data": {"capabilities": {}}}})
assert await allowed_doc_types(client, "bob") is None
async def test_allowed_doc_types_fail_open_not_cached():
clear_cache()
client = _FakeClient(raises=RuntimeError("ocs down"))
client = _client(raises=RuntimeError("ocs down"))
assert await allowed_doc_types(client, "carol") is None
# Failures are not cached — the next call retries the OCS lookup.
assert await allowed_doc_types(client, "carol") is None
assert client.calls == 2
assert client.capabilities.await_count == 2
async def test_allowed_doc_types_cache_is_per_user():
clear_cache()
alice = _FakeClient(_payload(["note"]))
bob = _FakeClient(_payload(["file"]))
alice = _client(_payload(["note"]))
bob = _client(_payload(["file"]))
assert await allowed_doc_types(alice, "alice") == frozenset({"note"})
assert await allowed_doc_types(bob, "bob") == frozenset({"file"})
async def test_allowed_doc_types_refetches_after_ttl(monkeypatch):
clear_cache()
client = _client(_payload(["note"]))
# Drive the module clock so the second call lands past the TTL window.
clock = {"now": 1000.0}
monkeypatch.setattr(cap.time, "monotonic", lambda: clock["now"])
await allowed_doc_types(client, "erin")
clock["now"] += cap._CACHE_TTL_SECONDS + 1
await allowed_doc_types(client, "erin")
assert client.capabilities.await_count == 2
async def test_clear_cache_forces_refetch():
clear_cache()
client = _FakeClient(_payload(["note"]))
client = _client(_payload(["note"]))
await allowed_doc_types(client, "dave")
cap.clear_cache()
await allowed_doc_types(client, "dave")
assert client.calls == 2
assert client.capabilities.await_count == 2