mirror of
https://github.com/tiennm99/DocsGPT.git
synced 2026-10-05 02:13:24 +00:00
Alex
574f96341e
The backend import package is now docsgpt, the name it will carry on PyPI; application was far too generic to install into anyone's site-packages. git mv plus a mechanical rewrite of every import, dotted string and path reference: 734 Python files, the compose files, Dockerfile, workflows, docs, setup scripts, devcontainer, k8s manifests, vscode config, pytest and coverage config, .gitignore. Behaviour is unchanged. Kept for one release: - A top-level application package whose meta-path finder resolves application.x.y to the already-imported docsgpt.x.y object, so old imports and entry points (celery -A application.app.celery, uvicorn application.asgi:asgi_app) keep working with a FutureWarning. - Celery registers every application.* task name as an alias of its docsgpt.* task on start-up, so messages queued by the previous release still run. The redbeat key prefix moves to redbeat:docsgpt:v2: so schedule entries the previous release wrote are left unread instead of firing twice. The backend image builds from the repository root (docker build -f docsgpt/Dockerfile .) so it can ship the alias package; a root .dockerignore allow-lists docsgpt/ and application/ and keeps caches, local data, .env files, the sample index files and the Dockerfile out. Compose and the image workflows point at the new context.
201 lines
7.1 KiB
Python
201 lines
7.1 KiB
Python
"""Tests for the GraphRAG enable route in docsgpt/api/user/sources/routes.py.
|
|
|
|
``graphrag_available`` and ``extract_graph.delay`` are mocked so no live
|
|
vector store, LLM, or model calls run; the ``sources`` row is real so the
|
|
authz lookup and config write read back.
|
|
"""
|
|
|
|
import uuid
|
|
from contextlib import contextmanager
|
|
from unittest.mock import patch
|
|
|
|
import pytest
|
|
from flask import Flask
|
|
|
|
from docsgpt.storage.db.repositories.sources import SourcesRepository
|
|
from docsgpt.storage.db.source_config import SourceConfig
|
|
|
|
|
|
@pytest.fixture
|
|
def app():
|
|
return Flask(__name__)
|
|
|
|
|
|
@contextmanager
|
|
def _patch_db(conn):
|
|
@contextmanager
|
|
def _yield():
|
|
yield conn
|
|
|
|
with patch(
|
|
"docsgpt.api.user.sources.routes.db_session", _yield
|
|
), patch(
|
|
"docsgpt.api.user.sources.routes.db_readonly", _yield
|
|
):
|
|
yield
|
|
|
|
|
|
def _grant_team_access(pg_conn, owner, member, source_id, access_level):
|
|
from docsgpt.storage.db.repositories.team_members import (
|
|
TeamMembersRepository,
|
|
)
|
|
from docsgpt.storage.db.repositories.team_resource_grants import (
|
|
TeamResourceGrantsRepository,
|
|
)
|
|
from docsgpt.storage.db.repositories.teams import TeamsRepository
|
|
|
|
team = TeamsRepository(pg_conn).create(
|
|
"Acme", f"acme-{uuid.uuid4().hex[:8]}", owner
|
|
)
|
|
TeamMembersRepository(pg_conn).add_member(
|
|
team["id"], member, role="team_member"
|
|
)
|
|
TeamResourceGrantsRepository(pg_conn).grant(
|
|
team["id"], "source", source_id, owner_id=owner, granted_by=owner,
|
|
access_level=access_level,
|
|
)
|
|
|
|
|
|
@pytest.mark.unit
|
|
class TestEnableSourceGraphRAG:
|
|
def test_returns_401_unauthenticated(self, app):
|
|
from docsgpt.api.user.sources.routes import EnableSourceGraphRAG
|
|
|
|
with app.test_request_context(
|
|
"/api/sources/x/graphrag/enable", method="POST"
|
|
):
|
|
from flask import request
|
|
request.decoded_token = None
|
|
response = EnableSourceGraphRAG().post("x")
|
|
assert response.status_code == 401
|
|
|
|
def test_unavailable_returns_400(self, app, pg_conn):
|
|
from docsgpt.api.user.sources.routes import EnableSourceGraphRAG
|
|
|
|
user = "u-graph-unavail"
|
|
src = SourcesRepository(pg_conn).create(
|
|
"files", user_id=user, type="file",
|
|
directory_structure={"a.md": {"type": "text/markdown"}},
|
|
)
|
|
sid = str(src["id"])
|
|
|
|
with _patch_db(pg_conn), patch(
|
|
"docsgpt.api.user.sources.routes.graphrag_available",
|
|
return_value=False,
|
|
), patch(
|
|
"docsgpt.api.user.sources.routes.extract_graph.delay"
|
|
) as mock_extract, app.test_request_context(
|
|
f"/api/sources/{sid}/graphrag/enable", method="POST"
|
|
):
|
|
from flask import request
|
|
request.decoded_token = {"sub": user}
|
|
response = EnableSourceGraphRAG().post(sid)
|
|
|
|
assert response.status_code == 400
|
|
mock_extract.assert_not_called()
|
|
got = SourcesRepository(pg_conn).get_any(sid, user)
|
|
assert SourceConfig.parse(got.get("config")).kind == "classic"
|
|
|
|
def test_owner_sets_config_and_enqueues(self, app, pg_conn):
|
|
from docsgpt.api.user.sources.routes import EnableSourceGraphRAG
|
|
|
|
user = "u-graph-owner"
|
|
src = SourcesRepository(pg_conn).create(
|
|
"files", user_id=user, type="file",
|
|
directory_structure={"a.md": {"type": "text/markdown"}},
|
|
)
|
|
sid = str(src["id"])
|
|
|
|
fake_task = type("T", (), {"id": "task-g"})()
|
|
with _patch_db(pg_conn), patch(
|
|
"docsgpt.api.user.sources.routes.graphrag_available",
|
|
return_value=True,
|
|
), patch(
|
|
"docsgpt.worker._reset_graph_for_source",
|
|
) as mock_reset, patch(
|
|
"docsgpt.api.user.sources.routes.extract_graph.delay",
|
|
return_value=fake_task,
|
|
) as mock_extract, app.test_request_context(
|
|
f"/api/sources/{sid}/graphrag/enable", method="POST"
|
|
):
|
|
from flask import request
|
|
request.decoded_token = {"sub": user}
|
|
response = EnableSourceGraphRAG().post(sid)
|
|
|
|
assert response.status_code == 200
|
|
assert response.json["success"] is True
|
|
assert response.json["task_id"] == "task-g"
|
|
|
|
got = SourcesRepository(pg_conn).get_any(sid, user)
|
|
cfg = SourceConfig.parse(got.get("config"))
|
|
assert cfg.kind == "graphrag"
|
|
assert cfg.retrieval.retriever == "graphrag"
|
|
|
|
# Each enable wipes any prior graph so it rebuilds from scratch.
|
|
mock_reset.assert_called_once_with(sid)
|
|
mock_extract.assert_called_once()
|
|
assert mock_extract.call_args.args[0] == sid
|
|
assert mock_extract.call_args.args[1] == user
|
|
# The key varies with the fresh ``updated_at`` the config write bumped,
|
|
# so each enable produces a new key that re-runs the worker.
|
|
key = mock_extract.call_args.kwargs["idempotency_key"]
|
|
assert key.startswith(f"extract-graph:{sid}:")
|
|
assert key != f"extract-graph:{sid}:"
|
|
|
|
def test_viewer_rejected_403(self, app, pg_conn):
|
|
from docsgpt.api.user.sources.routes import EnableSourceGraphRAG
|
|
|
|
owner = "alice-graph"
|
|
viewer = "bob-graph-viewer"
|
|
src = SourcesRepository(pg_conn).create(
|
|
"files", user_id=owner, type="file",
|
|
directory_structure={"a.md": {"type": "text/markdown"}},
|
|
)
|
|
sid = str(src["id"])
|
|
_grant_team_access(pg_conn, owner, viewer, sid, "viewer")
|
|
|
|
with _patch_db(pg_conn), patch(
|
|
"docsgpt.api.user.sources.routes.graphrag_available",
|
|
return_value=True,
|
|
), patch(
|
|
"docsgpt.api.user.sources.routes.extract_graph.delay"
|
|
) as mock_extract, app.test_request_context(
|
|
f"/api/sources/{sid}/graphrag/enable", method="POST"
|
|
):
|
|
from flask import request
|
|
request.decoded_token = {"sub": viewer}
|
|
response = EnableSourceGraphRAG().post(sid)
|
|
|
|
assert response.status_code == 403
|
|
mock_extract.assert_not_called()
|
|
got = SourcesRepository(pg_conn).get_any(sid, owner)
|
|
assert SourceConfig.parse(got.get("config")).kind == "classic"
|
|
|
|
|
|
@pytest.mark.unit
|
|
class TestConfigPatchCannotSetGraphrag:
|
|
"""The config PATCH endpoint must not flip kind to graphrag (D28)."""
|
|
|
|
def test_patch_kind_graphrag_rejected_400(self, app, pg_conn):
|
|
from docsgpt.api.user.sources.routes import SourceConfigResource
|
|
|
|
user = "u-patch-graph"
|
|
src = SourcesRepository(pg_conn).create(
|
|
"files", user_id=user, type="file",
|
|
directory_structure={"a.md": {"type": "text/markdown"}},
|
|
)
|
|
sid = str(src["id"])
|
|
|
|
with _patch_db(pg_conn), app.test_request_context(
|
|
f"/api/sources/{sid}/config",
|
|
method="PATCH",
|
|
json={"kind": "graphrag"},
|
|
):
|
|
from flask import request
|
|
request.decoded_token = {"sub": user}
|
|
response = SourceConfigResource().patch(sid)
|
|
|
|
assert response.status_code == 400
|
|
got = SourcesRepository(pg_conn).get_any(sid, user)
|
|
assert SourceConfig.parse(got.get("config")).kind == "classic"
|