OpenRAG is a comprehensive, single package Retrieval-Augmented Generation platform built on Langflow, Docling, and Opensearch.
Python
from types import SimpleNamespace
from unittest.mock import AsyncMock, MagicMock
import pytest
from fastapi import HTTPException
from api.files import upload_histogram
from services.file_service_v2 import FileServiceV2, _distinct_files_agg
USER = SimpleNamespace(user_id="u1", jwt_token="jwt")
def _call(file_service, **overrides):
params = {
"start": "2024-05-01T00:00:00.000000Z",
"end": "2024-06-20T23:59:59.000000Z",
"tz": "America/New_York",
"connector_type": None,
"mimetype": None,
"owner": None,
"data_sources": None,
"file_service": file_service,
"user": USER,
}
params.update(overrides)
return upload_histogram(**params)
@pytest.mark.asyncio
async def test_passes_window_timezone_and_filters_to_service():
service = MagicMock()
service.upload_histogram = AsyncMock(return_value={"total": 2, "days": {"2024-06-01": 2}})
response = await _call(service, owner=["u1"])
assert response.status_code == 200
kwargs = service.upload_histogram.await_args.kwargs
assert kwargs["time_zone"] == "America/New_York"
assert kwargs["start"] == "2024-05-01T00:00:00.000000Z"
assert kwargs["owner"] == ["u1"]
@pytest.mark.asyncio
@pytest.mark.parametrize(
"overrides",
[
{"start": "20240501"},
{"end": "not-a-date"},
{"start": "2024-07-01T00:00:00Z"},
{"tz": "Mars/Olympus_Mons"},
],
)
async def test_rejects_bad_input(overrides):
service = MagicMock()
service.upload_histogram = AsyncMock()
with pytest.raises(HTTPException) as exc:
await _call(service, **overrides)
assert exc.value.status_code == 422
service.upload_histogram.assert_not_awaited()
@pytest.mark.asyncio
async def test_returns_500_when_service_fails():
service = MagicMock()
service.upload_histogram = AsyncMock(side_effect=RuntimeError("down"))
response = await _call(service)
assert response.status_code == 500
@pytest.mark.asyncio
async def test_service_buckets_by_local_day_and_keeps_all_time_total():
client = MagicMock()
client.search = AsyncMock(
return_value={
"aggregations": {
"total": {"value": 7},
"window": {
"days": {
"buckets": [
{"key_as_string": "2024-06-01", "files": {"value": 2}},
{"key_as_string": "2024-06-03", "files": {"value": 5}},
]
}
},
}
}
)
session_manager = MagicMock()
session_manager.get_user_opensearch_client.return_value = client
result = await FileServiceV2(session_manager).upload_histogram(
user_id="u1",
start="2024-05-01T00:00:00.000000Z",
end="2024-06-20T23:59:59.000000Z",
time_zone="Asia/Kolkata",
connector_type=["local"],
)
assert result == {"total": 7, "days": {"2024-06-01": 2, "2024-06-03": 5}}
body = client.search.await_args.kwargs["body"]
# The date window narrows only the histogram, never the all-time total.
assert "range" not in str(body["query"])
assert {"term": {"connector_type": "local"}} in body["query"]["bool"]["filter"]
window = body["aggs"]["window"]
assert window["filter"] == {
"range": {
"indexed_time": {
"gte": "2024-05-01T00:00:00.000000Z",
"lte": "2024-06-20T23:59:59.000000Z",
}
}
}
histogram = window["aggs"]["days"]["date_histogram"]
assert histogram["time_zone"] == "Asia/Kolkata"
assert histogram["calendar_interval"] == "day"
# A file is a (filename, document_id) pair, matching the file list's rows.
for agg in (body["aggs"]["total"], window["aggs"]["days"]["aggs"]["files"]):
script = agg["cardinality"]["script"]["source"]
assert "doc['filename']" in script
assert "doc['document_id']" in script
# Painless rejects "\\n" in a literal; the separator must come from params.
assert "\\" not in script
assert "params.separator" in script
assert agg["cardinality"]["script"]["params"] == {"separator": "\n"}
assert agg["cardinality"]["precision_threshold"] == 3000
def test_chunk_count_terms_agg_groups_by_the_same_file_identity_as_the_count():
body = FileServiceV2(MagicMock())._build_terms_aggregation_for_chunk_count(
query={"bool": {}}, page_size=25, offset=0, sort_order="desc"
)
terms = body["aggs"]["files"]["terms"]
assert "field" not in terms
assert terms["script"] == _distinct_files_agg()["cardinality"]["script"]
assert terms["script"]["params"] == {"separator": "\n"}
def test_script_grouped_bucket_falls_back_to_the_filename_part_of_its_key():
files = FileServiceV2(MagicMock())._buckets_to_files(
[
{
"key": "report.pdf\nabc123",
"file_metadata": {"hits": {"hits": [{"_source": {"document_id": "abc123"}}]}},
}
]
)
assert files[0]["filename"] == "report.pdf"