fix(chunk-context): address PR #767 review — extract bbox helper, fix page_number overwrite
Resolves both 🟡 important issues from the latest review: 1. `page_number` was unconditionally overwritten in `viz_routes.py:696` even when Qdrant's payload lacked the field, clobbering the value resolved from `chunk_context.page_number`. The new helper returns each field independently and both call sites only overwrite via `is not None` guards, matching the existing logic in `visualization.py`. 2. The ~60-line `if chunk_index is not None: ... else: ...` Qdrant scroll block was duplicated between `api/visualization.py` and `auth/viz_routes.py`. Extracted into `get_chunk_bbox_and_page_from_qdrant` in `search/context.py` alongside the existing private `_get_chunk_*_from_qdrant` helpers; both routes now share ~12 lines of caller code. New unit tests at `tests/unit/test_chunk_bbox_helper.py` cover the indexed and offset paths, the `(bbox, None)` regression case, and graceful degradation on Qdrant strict-mode 400 (which also closes nit #4). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
Chris Coutinho
co-authored by
Claude Opus 4.7
Claude Opus 4.7
parent
c780f96d2b
commit
7ef8760d27
@@ -18,7 +18,6 @@ from pathlib import Path
|
||||
import anyio
|
||||
import numpy as np
|
||||
from jinja2 import Environment, FileSystemLoader
|
||||
from qdrant_client.models import FieldCondition, Filter, MatchValue
|
||||
from starlette.authentication import requires
|
||||
from starlette.requests import Request
|
||||
from starlette.responses import HTMLResponse, JSONResponse
|
||||
@@ -34,13 +33,15 @@ from nextcloud_mcp_server.search import (
|
||||
BM25HybridSearchAlgorithm,
|
||||
SemanticSearchAlgorithm,
|
||||
)
|
||||
from nextcloud_mcp_server.search.context import get_chunk_with_context
|
||||
from nextcloud_mcp_server.search.context import (
|
||||
get_chunk_bbox_and_page_from_qdrant,
|
||||
get_chunk_with_context,
|
||||
)
|
||||
from nextcloud_mcp_server.vector.oauth_sync import (
|
||||
NotProvisionedError,
|
||||
get_user_client_basic_auth,
|
||||
)
|
||||
from nextcloud_mcp_server.vector.pca import PCA
|
||||
from nextcloud_mcp_server.vector.placeholder import get_placeholder_filter
|
||||
from nextcloud_mcp_server.vector.qdrant_client import get_qdrant_client
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
@@ -625,82 +626,22 @@ async def chunk_context_endpoint(request: Request) -> JSONResponse:
|
||||
|
||||
# For PDF files, also fetch the chunk bbox from Qdrant so the client
|
||||
# can overlay a highlight on top of a render-on-demand page image
|
||||
# (Deck #76).
|
||||
# (Deck #76). Qdrant's page_number is trusted over the
|
||||
# context-expansion fallback when present.
|
||||
chunk_bbox = None
|
||||
page_number = chunk_context.page_number
|
||||
if doc_type == "file":
|
||||
try:
|
||||
settings = get_settings()
|
||||
qdrant_client = await get_qdrant_client()
|
||||
|
||||
# Prefer chunk_index for the chunk-bbox lookup (always indexed);
|
||||
# fall back to (chunk_start_offset, chunk_end_offset) when not provided.
|
||||
if chunk_index is not None:
|
||||
points_response = await qdrant_client.scroll(
|
||||
collection_name=settings.get_collection_name(),
|
||||
scroll_filter=Filter(
|
||||
must=[
|
||||
get_placeholder_filter(),
|
||||
FieldCondition(
|
||||
key="doc_id", match=MatchValue(value=doc_id_int)
|
||||
),
|
||||
FieldCondition(
|
||||
key="user_id", match=MatchValue(value=user_id)
|
||||
),
|
||||
FieldCondition(
|
||||
key="chunk_index",
|
||||
match=MatchValue(value=chunk_index),
|
||||
),
|
||||
]
|
||||
),
|
||||
limit=1,
|
||||
with_vectors=False,
|
||||
with_payload=["chunk_bbox", "page_number"],
|
||||
)
|
||||
else:
|
||||
# Legacy fallback for clients that don't send chunk_index
|
||||
# (pre-cbcoutinho/astrolabe#75). chunk_start/end_offset
|
||||
# aren't indexed in Qdrant Cloud strict mode, so this
|
||||
# call may fail with HTTP 400 there; the outer except
|
||||
# logs a warning and the response degrades gracefully
|
||||
# (no chunk_bbox).
|
||||
points_response = await qdrant_client.scroll(
|
||||
collection_name=settings.get_collection_name(),
|
||||
scroll_filter=Filter(
|
||||
must=[
|
||||
get_placeholder_filter(),
|
||||
FieldCondition(
|
||||
key="doc_id", match=MatchValue(value=doc_id_int)
|
||||
),
|
||||
FieldCondition(
|
||||
key="user_id", match=MatchValue(value=user_id)
|
||||
),
|
||||
FieldCondition(
|
||||
key="chunk_start_offset",
|
||||
match=MatchValue(value=start),
|
||||
),
|
||||
FieldCondition(
|
||||
key="chunk_end_offset",
|
||||
match=MatchValue(value=end),
|
||||
),
|
||||
]
|
||||
),
|
||||
limit=1,
|
||||
with_vectors=False,
|
||||
with_payload=["chunk_bbox", "page_number"],
|
||||
)
|
||||
|
||||
points = points_response[0]
|
||||
if points and points[0].payload:
|
||||
chunk_bbox = points[0].payload.get("chunk_bbox")
|
||||
page_number = points[0].payload.get("page_number")
|
||||
if chunk_bbox:
|
||||
logger.info(
|
||||
f"Found chunk bbox: page={page_number}, "
|
||||
f"rects={len(chunk_bbox)}"
|
||||
)
|
||||
except Exception as e:
|
||||
logger.warning(f"Failed to fetch chunk bbox: {e}")
|
||||
qdrant_bbox, qdrant_page = await get_chunk_bbox_and_page_from_qdrant(
|
||||
user_id=user_id,
|
||||
doc_id=doc_id_int,
|
||||
chunk_index=chunk_index,
|
||||
chunk_start=start,
|
||||
chunk_end=end,
|
||||
)
|
||||
if qdrant_bbox is not None:
|
||||
chunk_bbox = qdrant_bbox
|
||||
if qdrant_page is not None:
|
||||
page_number = qdrant_page
|
||||
|
||||
# Return response compatible with frontend expectations
|
||||
response_data: dict = {
|
||||
|
||||
Reference in New Issue
Block a user