Files
ss-tools/backend/tests/test_datasets.py
busya 4205618ee6 chore: eliminate all deprecation warnings from tests and linter
Warnings fixed:
- datetime.utcnow() → datetime.now(UTC) across 48+ files (src/ + tests/)
- datetime.utcnow (callback ref) → lambda: datetime.now(UTC) in model fields (18 files)
- Pydantic class Config → model_config = ConfigDict(...) (16 files)
- Pydantic .dict() → .model_dump() (8 files)
- ConfigDict(allow_population_by_field_name=True) → validate_by_name=True
- SQLAlchemy declarative_base() import path updated
- FastAPI on_event → lifespan context manager (app.py)
- Import sorting (ruff I001) auto-fixed across all files
- Fixed broken re-export chains that ruff F401 cleanup broke:
  _validate_bcp47: service.py now imports from dictionary_validation directly
  job_to_response: _job_routes.py and test imports from service_utils directly
  fetch_datasource_metadata: restored re-export in service.py
- Added missing TranslateJobService import in _job_routes.py (was deleted by F401)
- Added ConfigDict(protected_namespaces=()) for DashboardDatasetItem schema field
- pytest.ini: replaced deprecated importmode with asyncio_mode

All 440 tests pass with zero deprecation warnings.
2026-05-26 19:18:28 +03:00

420 lines
16 KiB
Python

# #region TestDatasetsApi [C:2] [TYPE Module]
# @RELATION DEPENDS_ON -> [DatasetsApi]
# @SEMANTICS: tests, datasets, api, metrics, inline-edit
# @PURPOSE: Contract tests for dataset endpoints (list+stats, detail+metrics, inline-edit descriptions).
# @LAYER Tests
import pytest
from unittest.mock import AsyncMock, MagicMock, patch
from fastapi.testclient import TestClient
from src.app import app
from src.dependencies import (
get_config_manager,
get_current_user,
get_resource_service,
get_task_manager,
has_permission,
)
client = TestClient(app)
# #region mock_user [C:2] [TYPE Function]
# @PURPOSE: Create a mock user with Admin role and all permissions for testing.
def _make_mock_user():
admin_role = MagicMock()
admin_role.name = "Admin"
admin_role.permissions = []
mock_user = MagicMock()
mock_user.username = "test-user"
mock_user.roles = [admin_role]
return mock_user
# #region mock_deps [C:2] [TYPE Function]
# @PURPOSE: Provide dependency override fixture for dataset route tests.
# @TEST_FIXTURE: dataset_route_overrides -> INLINE_JSON
@pytest.fixture
def mock_deps():
config_manager = MagicMock()
task_manager = MagicMock()
resource_service = MagicMock()
env = MagicMock()
env.id = "ss-dev"
env.name = "SS Dev"
env.superset_url = "http://superset:8088"
env.superset_token = "mock-token"
config_manager.get_environments.return_value = [env]
task_manager.get_all_tasks.return_value = []
resource_service.get_datasets_with_status = AsyncMock(
return_value=[
{
"id": 1,
"table_name": "users",
"schema": "public",
"database": "postgres",
"mapped_fields": {"total": 10, "mapped": 7},
"last_task": {"task_id": "t1", "status": "SUCCESS"},
"linked_dashboard_count": 2,
},
{
"id": 2,
"table_name": "orders",
"schema": "public",
"database": "postgres",
"mapped_fields": {"total": 5, "mapped": 0},
"last_task": None,
"linked_dashboard_count": 0,
},
{
"id": 3,
"table_name": "products",
"schema": "public",
"database": "postgres",
"mapped_fields": None,
"last_task": None,
"linked_dashboard_count": 0,
},
]
)
mock_user = _make_mock_user()
app.dependency_overrides[get_config_manager] = lambda: config_manager
app.dependency_overrides[get_task_manager] = lambda: task_manager
app.dependency_overrides[get_resource_service] = lambda: resource_service
app.dependency_overrides[get_current_user] = lambda: mock_user
# Override specific permission dependencies used by datasets routes
app.dependency_overrides[has_permission("plugin:migration", "READ")] = lambda: mock_user
app.dependency_overrides[has_permission("plugin:migration", "WRITE")] = lambda: mock_user
yield
app.dependency_overrides.clear()
# #region test_get_datasets_returns_stats [C:2] [TYPE Function]
# @PURPOSE: Verify GET /api/datasets returns stats object with correct counts.
# @TEST_CONTRACT: datasets_list -> datasets with stats payload
# @TEST_SCENARIO: datasets_with_stats -> HTTP 200 returns datasets, stats, and pagination.
# @TEST_INVARIANT: stats_counts_correct -> VERIFIED_BY: [datasets_with_stats]
def test_get_datasets_returns_stats(mock_deps):
response = client.get(
"/api/datasets?env_id=ss-dev&page=1&page_size=10"
)
assert response.status_code == 200
data = response.json()
assert "datasets" in data
assert "stats" in data
assert data["stats"]["total"] == 3
assert data["stats"]["mapped_count"] == 1 # users: 7/10
assert data["stats"]["unmapped_count"] == 1 # orders: 0/5
assert data["stats"]["linked_count"] == 1 # users has 2 linked dashboards
assert data["total"] == 3
assert data["page"] == 1
assert len(data["datasets"]) == 3
# #region test_get_datasets_filter_unmapped [C:2] [TYPE Function]
# @PURPOSE: Verify server-side filtering by unmapped datasets.
# @TEST_SCENARIO: filter_unmapped -> Only returns datasets with mapped=0.
def test_get_datasets_filter_unmapped(mock_deps):
response = client.get(
"/api/datasets?env_id=ss-dev&filter=unmapped&page=1&page_size=10"
)
assert response.status_code == 200
data = response.json()
# stats are global (unfiltered)
assert data["stats"]["total"] == 3
# filtered list: only orders (mapped=0)
assert data["total"] == 1
assert len(data["datasets"]) == 1
assert data["datasets"][0]["table_name"] == "orders"
# #region test_get_datasets_filter_mapped [C:2] [TYPE Function]
# @PURPOSE: Verify server-side filtering by mapped datasets.
def test_get_datasets_filter_mapped(mock_deps):
response = client.get(
"/api/datasets?env_id=ss-dev&filter=mapped&page=1&page_size=10"
)
assert response.status_code == 200
data = response.json()
assert data["stats"]["total"] == 3
assert data["total"] == 1
assert data["datasets"][0]["table_name"] == "users"
# #region test_get_datasets_filter_linked [C:2] [TYPE Function]
# @PURPOSE: Verify server-side filtering by datasets linked to dashboards.
def test_get_datasets_filter_linked(mock_deps):
response = client.get(
"/api/datasets?env_id=ss-dev&filter=linked&page=1&page_size=10"
)
assert response.status_code == 200
data = response.json()
assert data["stats"]["total"] == 3
assert data["total"] == 1
assert data["datasets"][0]["table_name"] == "users"
# #region test_get_datasets_filter_all [C:2] [TYPE Function]
# @PURPOSE: Verify filter=all returns all datasets (same as no filter).
def test_get_datasets_filter_all(mock_deps):
response = client.get(
"/api/datasets?env_id=ss-dev&filter=all&page=1&page_size=10"
)
assert response.status_code == 200
data = response.json()
assert data["total"] == 3
# #region test_get_dataset_detail_returns_metrics [C:2] [TYPE Function]
# @PURPOSE: Verify GET /api/datasets/{id} returns metrics and metric_count.
# @TEST_SCENARIO: detail_with_metrics -> Response includes metrics list and metric_count.
def test_get_dataset_detail_returns_metrics(mock_deps):
with patch("src.api.routes.datasets.SupersetClient") as MockClient:
mock_client = MagicMock()
MockClient.return_value = mock_client
mock_client.get_dataset_detail.return_value = {
"id": 1,
"table_name": "users",
"schema": "public",
"database": "postgres",
"description": "User table",
"columns": [
{"id": 10, "name": "id", "type": "INT", "is_dttm": False, "is_active": True, "description": "PK"},
{"id": 11, "name": "name", "type": "STRING", "is_dttm": False, "is_active": True, "description": ""},
],
"column_count": 2,
"metrics": [
{"id": 100, "metric_name": "count", "expression": "COUNT(*)", "verbose_name": "Row Count", "description": "Total rows", "metric_type": "count"},
],
"metric_count": 1,
"sql": "SELECT * FROM users",
"linked_dashboards": [{"id": 5, "title": "Users Dashboard", "slug": "users-dash"}],
"linked_dashboard_count": 1,
"is_sqllab_view": False,
"created_on": "2024-01-01T00:00:00",
"changed_on": "2024-06-01T00:00:00",
}
response = client.get(
"/api/datasets/1?env_id=ss-dev"
)
assert response.status_code == 200
data = response.json()
assert "metrics" in data
assert "metric_count" in data
assert data["metric_count"] == 1
assert len(data["metrics"]) == 1
assert data["metrics"][0]["metric_name"] == "count"
assert data["metrics"][0]["expression"] == "COUNT(*)"
# #region test_update_column_description [C:2] [TYPE Function]
# @PURPOSE: Verify PUT /api/datasets/{id}/columns/{col_id}/description saves description.
# @TEST_SCENARIO: save_column_desc -> Saves and returns updated description.
def test_update_column_description(mock_deps):
with patch("src.api.routes.datasets.SupersetClient") as MockClient:
mock_client = MagicMock()
MockClient.return_value = mock_client
mock_client.get_dataset.return_value = {
"id": 1,
"result": {
"id": 1,
"columns": [
{"id": 10, "column_name": "id", "type": "INT", "description": "old desc"},
{"id": 11, "column_name": "name", "type": "STRING", "description": ""},
],
"metrics": [],
},
}
mock_client.update_dataset.return_value = {"id": 1}
response = client.put(
"/api/datasets/1/columns/10/description?env_id=ss-dev",
json={"description": "new description"},
)
assert response.status_code == 200
data = response.json()
assert data["success"] is True
assert data["description"] == "new description"
# Verify update_dataset was called with correct column modified
update_call = mock_client.update_dataset.call_args
assert update_call[0][0] == 1 # dataset_id
# Check override_columns=false was passed
assert update_call[1]["override_columns"] is False
# #region test_update_metric_description [C:2] [TYPE Function]
# @PURPOSE: Verify PUT /api/datasets/{id}/metrics/{metric_id}/description saves correctly.
# @TEST_SCENARIO: save_metric_desc -> Saves and returns updated description.
def test_update_metric_description(mock_deps):
with patch("src.api.routes.datasets.SupersetClient") as MockClient:
mock_client = MagicMock()
MockClient.return_value = mock_client
mock_client.get_dataset.return_value = {
"id": 1,
"result": {
"id": 1,
"columns": [],
"metrics": [
{"id": 100, "metric_name": "count", "expression": "COUNT(*)", "description": ""},
],
},
}
mock_client.update_dataset.return_value = {"id": 1}
response = client.put(
"/api/datasets/1/metrics/100/description?env_id=ss-dev",
json={"description": "Total row count"},
)
assert response.status_code == 200
data = response.json()
assert data["success"] is True
assert data["description"] == "Total row count"
# #region test_update_column_description_not_found [C:2] [TYPE Function]
# @PURPOSE: Verify 404 when column not found.
def test_update_column_description_not_found(mock_deps):
with patch("src.api.routes.datasets.SupersetClient") as MockClient:
mock_client = MagicMock()
MockClient.return_value = mock_client
mock_client.get_dataset.return_value = {
"id": 1,
"result": {
"id": 1,
"columns": [{"id": 10, "column_name": "id"}],
"metrics": [],
},
}
response = client.put(
"/api/datasets/1/columns/999/description?env_id=ss-dev",
json={"description": "test"},
)
assert response.status_code == 404
# #region test_update_column_description_validation [C:2] [TYPE Function]
# @PURPOSE: Verify 400 when description exceeds max length.
def test_update_column_description_validation(mock_deps):
response = client.put(
"/api/datasets/1/columns/10/description?env_id=ss-dev",
json={"description": "x" * 2001},
)
assert response.status_code == 400
# #region test_get_datasets_superset_503 [C:2] [TYPE Function]
# @PURPOSE: Verify GET /api/datasets returns 503 when Superset/ResourceService throws.
# @TEST_SCENARIO: upstream_superset_failure -> 503 with error detail.
# @TEST_INVARIANT: upstream_failure_503 -> VERIFIED_BY: [superset_failure]
def test_get_datasets_superset_503(mock_deps):
# Override resource_service to throw during get_datasets_with_status
from src.dependencies import get_resource_service
failing_resource_service = MagicMock()
failing_resource_service.get_datasets_with_status = AsyncMock(
side_effect=Exception("Superset connection refused")
)
app.dependency_overrides[get_resource_service] = lambda: failing_resource_service
try:
response = client.get(
"/api/datasets?env_id=ss-dev&page=1&page_size=10"
)
assert response.status_code == 503
assert "Superset" in response.text or "Failed" in response.text
finally:
app.dependency_overrides[get_resource_service] = None
# #region test_update_column_description_superset_502 [C:2] [TYPE Function]
# @PURPOSE: Verify PUT column description returns 502 when SupersetClient.update_dataset throws.
# @TEST_SCENARIO: superset_put_failure -> 502 with error detail.
def test_update_column_description_superset_502(mock_deps):
with patch("src.api.routes.datasets.SupersetClient") as MockClient:
mock_client = MagicMock()
MockClient.return_value = mock_client
# GET succeeds
mock_client.get_dataset.return_value = {
"id": 1,
"result": {
"id": 1,
"columns": [{"id": 10, "column_name": "id", "type": "INT", "description": "old desc"}],
"metrics": [],
},
}
# PUT fails
mock_client.update_dataset.side_effect = Exception("Superset write failure")
response = client.put(
"/api/datasets/1/columns/10/description?env_id=ss-dev",
json={"description": "new description"},
)
assert response.status_code == 502
data = response.json()
assert "Superset" in data.get("detail", "")
# #region test_update_column_description_strips_html [C:2] [TYPE Function]
# @PURPOSE: Verify HTML tags are stripped from description before saving.
# @TEST_SCENARIO: html_in_description -> HTML tags stripped, clean text saved.
def test_update_column_description_strips_html(mock_deps):
with patch("src.api.routes.datasets.SupersetClient") as MockClient:
mock_client = MagicMock()
MockClient.return_value = mock_client
mock_client.get_dataset.return_value = {
"id": 1,
"result": {
"id": 1,
"columns": [{"id": 10, "column_name": "id", "type": "INT", "description": ""}],
"metrics": [],
},
}
mock_client.update_dataset.return_value = {"id": 1}
response = client.put(
"/api/datasets/1/columns/10/description?env_id=ss-dev",
json={"description": "Hello <b>World</b> with <script>alert('xss')</script>"},
)
assert response.status_code == 200
data = response.json()
# HTML stripped
assert data["description"] == "Hello World with alert('xss')"
assert "<b>" not in data["description"]
assert "<script>" not in data["description"]
# Verify the stripped description was sent to Superset
update_call = mock_client.update_dataset.call_args
updated_columns = update_call[0][1]["columns"]
assert updated_columns[0]["description"] == "Hello World with alert('xss')"
# #region test_dataset_item_has_metric_count [C:2] [TYPE Function]
# @PURPOSE: Verify DatasetItem in list response includes metric_count field.
def test_dataset_item_has_metric_count(mock_deps):
response = client.get(
"/api/datasets?env_id=ss-dev&page=1&page_size=10"
)
assert response.status_code == 200
data = response.json()
for ds in data["datasets"]:
assert "metric_count" in ds
assert isinstance(ds["metric_count"], int)