fix(chunk-context): address PR #767 round-2 review — gate, parity, doc
Latest reviewer comment flagged five items on top of the original PR. This commit addresses every one: 🟡 1. Skip the offset-based Qdrant fallback for `doc_type=file` when `chunk_index` is supplied. Qdrant Cloud's strict mode rejects unindexed filter fields with HTTP 400, which `_get_chunk_from_qdrant` catches and logs at `logger.error` — masking real Qdrant problems in monitoring. Notes/cards keep the offset fallback (cheap, useful for legacy data). 🟡 2. Add a `logger.warning` and clarifying inline comment in the doc-text fallback path when `chunk_index` is None — surfaces the pre-existing "0/N misreport" so callers can detect it. Type-nullability propagation is deferred to a follow-up (out of scope for this hotfix). 🟢 3. Simplify `if chunk_text and doc_id_int is not None:` → `if chunk_text:` with an inner `assert doc_id_int is not None` for `ty` narrowing. The outer second clause was dead. 🟢 4. Add `doc_type` `FieldCondition` to the offset-based image lookup in both `visualization.py` and `viz_routes.py` for parity with the `chunk_index` branches. 🟢 5. Inline the `chunk_filter` local in `visualization.py` directly into the `must=[]` list (matches `viz_routes.py` style). Adds `tests/unit/test_chunk_context_offset_gate.py` with three regression tests covering the gate matrix: (file, with-index → skip offset), (note, with-index → still tries offset), (file, no-index → still tries offset). Lives at top-level rather than `tests/unit/search/` to side-step a pre-existing circular-init issue in `nextcloud_mcp_server.search`. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
co-authored by
Claude Opus 4.7
parent
8457c427a5
commit
51c1d42ea3
@@ -0,0 +1,137 @@
|
||||
"""Unit tests for `nextcloud_mcp_server.search.context.get_chunk_with_context`.
|
||||
|
||||
Focused on the chunk-lookup gate that decides whether to fall back from the
|
||||
indexed `chunk_index` path to the unindexed `(chunk_start, chunk_end)` path.
|
||||
The behaviour matters because Qdrant Cloud's strict mode rejects filters on
|
||||
unindexed fields with HTTP 400 — a fall-through there surfaces a misleading
|
||||
`logger.error` even when the caller's request would correctly resolve as a
|
||||
404 via the file fast-fail.
|
||||
"""
|
||||
|
||||
from unittest.mock import AsyncMock, MagicMock, patch
|
||||
|
||||
import pytest
|
||||
|
||||
# Import via the auth surface first to side-step a known circular-init issue
|
||||
# in `nextcloud_mcp_server.search.__init__` when `search` is imported as the
|
||||
# first entry point (also affects pre-existing tests under tests/unit/search/).
|
||||
import nextcloud_mcp_server.auth.viz_routes # noqa: F401 (init-order fixup)
|
||||
from nextcloud_mcp_server.search import context as context_module
|
||||
from nextcloud_mcp_server.search.context import get_chunk_with_context
|
||||
|
||||
pytestmark = pytest.mark.unit
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
def mock_nc_client() -> MagicMock:
|
||||
return MagicMock()
|
||||
|
||||
|
||||
class TestOffsetFallbackGate:
|
||||
"""When chunk_index is provided AND doc_type=='file', the offset fallback
|
||||
must be skipped — see PR #767 review (🟡 spurious Qdrant error log).
|
||||
"""
|
||||
|
||||
async def test_file_with_chunk_index_skips_offset_fallback_on_miss(
|
||||
self, mock_nc_client
|
||||
):
|
||||
with (
|
||||
patch.object(
|
||||
context_module,
|
||||
"_get_chunk_by_index_from_qdrant",
|
||||
new_callable=AsyncMock,
|
||||
return_value=None,
|
||||
) as mock_indexed,
|
||||
patch.object(
|
||||
context_module,
|
||||
"_get_chunk_from_qdrant",
|
||||
new_callable=AsyncMock,
|
||||
return_value="should-not-be-returned",
|
||||
) as mock_offset,
|
||||
):
|
||||
result = await get_chunk_with_context(
|
||||
nc_client=mock_nc_client,
|
||||
user_id="alice",
|
||||
doc_id=12345,
|
||||
doc_type="file",
|
||||
chunk_start=0,
|
||||
chunk_end=100,
|
||||
chunk_index=3,
|
||||
total_chunks=20,
|
||||
)
|
||||
|
||||
assert result is None, "file fast-fail must return None on Qdrant miss"
|
||||
mock_indexed.assert_awaited_once()
|
||||
mock_offset.assert_not_awaited()
|
||||
|
||||
async def test_note_with_chunk_index_still_uses_offset_fallback(
|
||||
self, mock_nc_client
|
||||
):
|
||||
"""Notes/deck cards keep the offset fallback (cheap, useful for legacy
|
||||
data): the gate is file-specific.
|
||||
"""
|
||||
with (
|
||||
patch.object(
|
||||
context_module,
|
||||
"_get_chunk_by_index_from_qdrant",
|
||||
new_callable=AsyncMock,
|
||||
return_value=None,
|
||||
) as mock_indexed,
|
||||
patch.object(
|
||||
context_module,
|
||||
"_get_chunk_from_qdrant",
|
||||
new_callable=AsyncMock,
|
||||
return_value=None,
|
||||
) as mock_offset,
|
||||
patch.object(
|
||||
context_module,
|
||||
"_fetch_document_text",
|
||||
new_callable=AsyncMock,
|
||||
return_value=None,
|
||||
),
|
||||
):
|
||||
await get_chunk_with_context(
|
||||
nc_client=mock_nc_client,
|
||||
user_id="alice",
|
||||
doc_id=42,
|
||||
doc_type="note",
|
||||
chunk_start=0,
|
||||
chunk_end=10,
|
||||
chunk_index=2,
|
||||
total_chunks=5,
|
||||
)
|
||||
|
||||
mock_indexed.assert_awaited_once()
|
||||
mock_offset.assert_awaited_once()
|
||||
|
||||
async def test_file_without_chunk_index_uses_offset_fallback(self, mock_nc_client):
|
||||
"""Files with no chunk_index supplied still use the offset path —
|
||||
the gate only kicks in once the indexed lookup has been attempted.
|
||||
"""
|
||||
with (
|
||||
patch.object(
|
||||
context_module,
|
||||
"_get_chunk_by_index_from_qdrant",
|
||||
new_callable=AsyncMock,
|
||||
return_value=None,
|
||||
) as mock_indexed,
|
||||
patch.object(
|
||||
context_module,
|
||||
"_get_chunk_from_qdrant",
|
||||
new_callable=AsyncMock,
|
||||
return_value=None,
|
||||
) as mock_offset,
|
||||
):
|
||||
await get_chunk_with_context(
|
||||
nc_client=mock_nc_client,
|
||||
user_id="alice",
|
||||
doc_id=12345,
|
||||
doc_type="file",
|
||||
chunk_start=0,
|
||||
chunk_end=100,
|
||||
chunk_index=None,
|
||||
total_chunks=20,
|
||||
)
|
||||
|
||||
mock_indexed.assert_not_awaited()
|
||||
mock_offset.assert_awaited_once()
|
||||
Reference in New Issue
Block a user