Files
deer-flow/tests/unit/tools/test_crawl.py
Willem Jiang d30c4d00d3 fix: convert crawl_tool dict return to JSON string for type consistency (#636)
Keep fixing #631
This pull request updates the crawl_tool function to return its results as a JSON string instead of a dictionary, and adjusts the unit tests accordingly to handle the new return type. The changes ensure consistent serialization of output and proper validation in tests.
2025-10-21 10:00:33 +08:00

113 lines
3.4 KiB
Python

import json
from unittest.mock import Mock, patch
from src.tools.crawl import crawl_tool
class TestCrawlTool:
@patch("src.tools.crawl.Crawler")
def test_crawl_tool_success(self, mock_crawler_class):
# Arrange
mock_crawler = Mock()
mock_article = Mock()
mock_article.to_markdown.return_value = (
"# Test Article\nThis is test content." * 100
)
mock_crawler.crawl.return_value = mock_article
mock_crawler_class.return_value = mock_crawler
url = "https://example.com"
# Act
result = crawl_tool(url)
# Assert
assert isinstance(result, str)
result_dict = json.loads(result)
assert result_dict["url"] == url
assert "crawled_content" in result_dict
assert len(result_dict["crawled_content"]) <= 1000
mock_crawler_class.assert_called_once()
mock_crawler.crawl.assert_called_once_with(url)
mock_article.to_markdown.assert_called_once()
@patch("src.tools.crawl.Crawler")
def test_crawl_tool_short_content(self, mock_crawler_class):
# Arrange
mock_crawler = Mock()
mock_article = Mock()
short_content = "Short content"
mock_article.to_markdown.return_value = short_content
mock_crawler.crawl.return_value = mock_article
mock_crawler_class.return_value = mock_crawler
url = "https://example.com"
# Act
result = crawl_tool(url)
# Assert
result_dict = json.loads(result)
assert result_dict["crawled_content"] == short_content
@patch("src.tools.crawl.Crawler")
@patch("src.tools.crawl.logger")
def test_crawl_tool_crawler_exception(self, mock_logger, mock_crawler_class):
# Arrange
mock_crawler = Mock()
mock_crawler.crawl.side_effect = Exception("Network error")
mock_crawler_class.return_value = mock_crawler
url = "https://example.com"
# Act
result = crawl_tool(url)
# Assert
assert isinstance(result, str)
assert "Failed to crawl" in result
assert "Network error" in result
mock_logger.error.assert_called_once()
@patch("src.tools.crawl.Crawler")
@patch("src.tools.crawl.logger")
def test_crawl_tool_crawler_instantiation_exception(
self, mock_logger, mock_crawler_class
):
# Arrange
mock_crawler_class.side_effect = Exception("Crawler init error")
url = "https://example.com"
# Act
result = crawl_tool(url)
# Assert
assert isinstance(result, str)
assert "Failed to crawl" in result
assert "Crawler init error" in result
mock_logger.error.assert_called_once()
@patch("src.tools.crawl.Crawler")
@patch("src.tools.crawl.logger")
def test_crawl_tool_markdown_conversion_exception(
self, mock_logger, mock_crawler_class
):
# Arrange
mock_crawler = Mock()
mock_article = Mock()
mock_article.to_markdown.side_effect = Exception("Markdown conversion error")
mock_crawler.crawl.return_value = mock_article
mock_crawler_class.return_value = mock_crawler
url = "https://example.com"
# Act
result = crawl_tool(url)
# Assert
assert isinstance(result, str)
assert "Failed to crawl" in result
assert "Markdown conversion error" in result
mock_logger.error.assert_called_once()