Skip to content

armgit / dify

Public Imported from GitHub
0

Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype to production without rebuilding the stack.

test: use real datasource crawl and stream event values (#43413)

Asuka Minato2372ab58e8d4
diff --git a/api/tests/unit_tests/core/datasource/test_stream_node_events.py b/api/tests/unit_tests/core/datasource/test_stream_node_events.py
index 58e18ca00e4..1b3005af1ac 100644
--- a/api/tests/unit_tests/core/datasource/test_stream_node_events.py
+++ b/api/tests/unit_tests/core/datasource/test_stream_node_events.py
@@ -7,6 +7,7 @@ from pytest_mock import MockerFixture
from core.datasource.datasource_manager import DatasourceManager
from core.datasource.entities.datasource_entities import DatasourceMessage
from graphon.node_events import StreamCompletedEvent
+from graphon.runtime import VariablePool
def _gen_var_stream() -> Generator[DatasourceMessage]:
@@ -35,10 +36,11 @@ def test_stream_node_events_accumulates_variables(mocker: MockerFixture) -> None
tenant_id="t",
parameters_for_log={},
datasource_info={"user_id": "u"},
- variable_pool=mocker.Mock(),
- datasource_param=type("P", (), {"workspace_id": "w", "page_id": "pg", "type": "t"})(),
+ variable_pool=VariablePool(),
+ datasource_param=None,
online_drive_request=None,
credentials={},
)
)
assert isinstance(events[-1], StreamCompletedEvent)
+ assert events[-1].node_run_result.outputs["a"] == "xy"
diff --git a/api/tests/unit_tests/core/datasource/test_website_crawl.py b/api/tests/unit_tests/core/datasource/test_website_crawl.py
index d589acb747b..fc2ff7b1e22 100644
--- a/api/tests/unit_tests/core/datasource/test_website_crawl.py
+++ b/api/tests/unit_tests/core/datasource/test_website_crawl.py
@@ -14,6 +14,7 @@ and ensure proper handling of crawl options, status checking, and data retrieval
from unittest.mock import Mock, create_autospec, patch
+import httpx
import pytest
from pytest_mock import MockerFixture
@@ -568,8 +569,7 @@ class TestWebsiteService:
"api_key": "test_key",
}
- mock_response = Mock()
- mock_response.json.return_value = {"code": 200, "data": {"taskId": "task-789"}}
+ mock_response = httpx.Response(200, json={"code": 200, "data": {"taskId": "task-789"}})
mock_httpx_post = mocker.patch(
"services.data_source.website_service._adaptive_http_client.post",
return_value=mock_response,
@@ -913,8 +913,7 @@ class TestErrorHandling:
# Simulate API error for non-existent job
from core.rag.extractor.watercrawl.exceptions import WaterCrawlBadRequestError
- mock_response = Mock()
- mock_response.status_code = 404
+ mock_response = httpx.Response(404, json={"message": "Crawl job not found", "errors": {}})
mock_instance.get_crawl_request.side_effect = WaterCrawlBadRequestError(mock_response)
provider = WaterCrawlProvider(api_key="test_key")
@@ -1356,8 +1355,7 @@ class TestProviderSpecificFeatures:
"api_key": "test_key",
}
- mock_response = Mock()
- mock_response.json.return_value = {
+ response_data = {
"code": 200,
"data": {
"title": "Single Page Title",
@@ -1365,6 +1363,7 @@ class TestProviderSpecificFeatures:
"url": "https://example.com/page",
},
}
+ mock_response = httpx.Response(200, json=response_data)
mocker.patch("services.data_source.website_service._jina_http_client.get", return_value=mock_response)
from services.data_source.website_service import WebsiteCrawlApiRequest