fix(viz_routes): address PR #767 review — param parity + always-on page_number

- Replace bare int() casts for start/end/context_chars in chunk_context_endpoint
  with _parse_int_param, matching visualization.py bounds (0–10M for offsets,
  0–10K for context_chars), and add the missing end > start guard.
- Initialize page_number from chunk_context.page_number so non-file doc_types
  surface it; include page_number, chunk_index, and total_chunks unconditionally
  in the response. Only highlighted_page_image stays gated on its own truthiness.
- Add a chunk_index forwarding regression test that asserts the new kwargs reach
  get_chunk_with_context and appear in the response payload.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
Chris Coutinho
2026-05-08 19:14:43 +02:00
co-authored by Claude Opus 4.7
parent a33a365a69
commit 53e6dba5a2
2 changed files with 67 additions and 6 deletions
+15 -6
View File
@@ -538,7 +538,6 @@ async def chunk_context_endpoint(request: Request) -> JSONResponse:
end_str = request.query_params.get("end")
chunk_index_str = request.query_params.get("chunk_index")
total_chunks_str = request.query_params.get("total_chunks")
context_chars = int(request.query_params.get("context", "500"))
# Validate required parameters
if not all([doc_type, doc_id, start_str, end_str]):
@@ -556,8 +555,17 @@ async def chunk_context_endpoint(request: Request) -> JSONResponse:
assert start_str is not None
assert end_str is not None
start = int(start_str)
end = int(end_str)
context_chars = _parse_int_param(
request.query_params.get("context"),
500,
0,
10000,
"context_chars",
)
start = _parse_int_param(start_str, 0, 0, 10000000, "start")
end = _parse_int_param(end_str, 0, 0, 10000000, "end")
if end <= start:
raise ValueError("end must be greater than start")
chunk_index: int | None = None
if chunk_index_str is not None:
chunk_index = _parse_int_param(
@@ -617,7 +625,7 @@ async def chunk_context_endpoint(request: Request) -> JSONResponse:
# For PDF files, also fetch the highlighted page image from Qdrant
highlighted_page_image = None
page_number = None
page_number = chunk_context.page_number
if doc_type == "file":
try:
settings = get_settings()
@@ -697,12 +705,13 @@ async def chunk_context_endpoint(request: Request) -> JSONResponse:
"after_context": chunk_context.after_context,
"has_more_before": chunk_context.has_before_truncation,
"has_more_after": chunk_context.has_after_truncation,
"page_number": page_number,
"chunk_index": chunk_context.chunk_index,
"total_chunks": chunk_context.total_chunks,
}
# Add image data if available
if highlighted_page_image:
response_data["highlighted_page_image"] = highlighted_page_image
response_data["page_number"] = page_number
return JSONResponse(response_data)
@@ -259,6 +259,58 @@ class TestChunkContextCredentialPath:
assert "failed to fetch chunk context" in data["error"].lower()
class TestChunkContextParameterForwarding:
"""Verify new chunk_index / total_chunks query params reach the lookup.
Regression guard for PR #767: the whole point of the fix is that callers
pass chunk_index, and it must arrive at get_chunk_with_context as the
primary Qdrant lookup key.
"""
def test_chunk_index_and_total_chunks_forwarded(self):
mock_nc_client = _make_mock_nc_client()
mock_ctx = _make_mock_chunk_context()
mock_ctx.chunk_index = 7
mock_ctx.total_chunks = 10
with (
patch(
"nextcloud_mcp_server.api.visualization.validate_token_and_get_user",
new_callable=AsyncMock,
return_value=("testuser", True),
),
patch(
"nextcloud_mcp_server.api.visualization.get_user_client_basic_auth",
new_callable=AsyncMock,
return_value=mock_nc_client,
),
patch(
"nextcloud_mcp_server.api.visualization.get_chunk_with_context",
new_callable=AsyncMock,
return_value=mock_ctx,
) as mock_get_chunk,
):
app = create_test_app()
client = TestClient(app)
response = client.get(
"/api/v1/chunk-context?doc_type=note&doc_id=42"
"&start=0&end=10&chunk_index=7&total_chunks=10",
headers={"Authorization": "Bearer test-token"},
)
assert response.status_code == 200
kwargs = mock_get_chunk.await_args.kwargs
assert kwargs["chunk_index"] == 7
assert kwargs["total_chunks"] == 10
data = response.json()
assert data["chunk_index"] == 7
assert data["total_chunks"] == 10
# page_number must be present even when None (frontend may scroll
# by it for non-file doc types)
assert "page_number" in data
class TestChunkContextConfigErrors:
"""Tests for configuration failure paths."""