starmen1265
e152b8ea05
- Add StealthConfig Pydantic model with all playwright-stealth options - Add apply_stealth_sync/apply_stealth_async functions for browser context - Add get_default_stealth_config convenience function - Add 9 unit tests covering config defaults, stealth application, and disabled mode - Update conftest.py with stealth_config and mock_browser_context fixtures Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
120 lines
3.1 KiB
Python
120 lines
3.1 KiB
Python
"""Shared fixtures for Zhihu Crawler tests"""
|
|
|
|
import pytest
|
|
from unittest.mock import MagicMock, AsyncMock
|
|
|
|
|
|
@pytest.fixture
|
|
def mock_page():
|
|
"""Mock Playwright page object.
|
|
|
|
Provides a mock page with route method for API interception testing.
|
|
"""
|
|
page = MagicMock()
|
|
page.route = MagicMock()
|
|
page.goto = MagicMock()
|
|
page.wait_for_selector = MagicMock()
|
|
page.query_selector = MagicMock()
|
|
page.close = MagicMock()
|
|
return page
|
|
|
|
|
|
@pytest.fixture
|
|
def mock_browser():
|
|
"""Mock PlaywrightBrowser instance.
|
|
|
|
Provides a mock browser with page attribute and lifecycle methods.
|
|
"""
|
|
browser = MagicMock()
|
|
browser.page = MagicMock()
|
|
browser.start = MagicMock()
|
|
browser.close = MagicMock()
|
|
browser.goto = MagicMock()
|
|
return browser
|
|
|
|
|
|
@pytest.fixture
|
|
def sample_answer_api_response():
|
|
"""Sample answer API response data.
|
|
|
|
Mimics the structure of /api/v4/answers/{id} response.
|
|
"""
|
|
return {
|
|
"id": 123456789,
|
|
"type": "answer",
|
|
"question": {
|
|
"id": 987654321,
|
|
"title": "What is the best way to learn programming?",
|
|
"type": "question"
|
|
},
|
|
"content": "<p>This is a sample answer content in HTML format.</p>",
|
|
"excerpt": "This is a sample answer content...",
|
|
"author": {
|
|
"id": "user123",
|
|
"name": "John Doe",
|
|
"url_token": "john-doe-123",
|
|
"headline": "Software Engineer"
|
|
},
|
|
"voteup_count": 1000,
|
|
"comment_count": 50,
|
|
"created_time": 1700000000,
|
|
"updated_time": 1700100000
|
|
}
|
|
|
|
|
|
@pytest.fixture
|
|
def sample_article_api_response():
|
|
"""Sample article API response data.
|
|
|
|
Mimics the structure of zhuanlan article API response.
|
|
"""
|
|
return {
|
|
"id": 11223344,
|
|
"title": "How to Build a Web Scraper",
|
|
"content": "<p>This article explains how to build a web scraper...</p>",
|
|
"excerpt": "This article explains how to build a web scraper...",
|
|
"author": {
|
|
"id": "author456",
|
|
"name": "Jane Smith",
|
|
"url_token": "jane-smith-456",
|
|
"headline": "Tech Writer"
|
|
},
|
|
"voteup_count": 500,
|
|
"comment_count": 25,
|
|
"created_time": 1699900000,
|
|
"updated_time": 1700000000
|
|
}
|
|
|
|
|
|
@pytest.fixture
|
|
def sample_answer_url():
|
|
"""Sample Zhihu answer URL."""
|
|
return "https://www.zhihu.com/question/987654321/answer/123456789"
|
|
|
|
|
|
@pytest.fixture
|
|
def sample_article_url():
|
|
"""Sample Zhihu zhuanlan article URL."""
|
|
return "https://zhuanlan.zhihu.com/p/11223344"
|
|
|
|
|
|
@pytest.fixture
|
|
def sample_invalid_url():
|
|
"""Sample invalid URL for testing error handling."""
|
|
return "https://www.example.com/some-page"
|
|
|
|
|
|
@pytest.fixture
|
|
def stealth_config():
|
|
"""Default StealthConfig for testing."""
|
|
from kb_workflow.crawler.stealth import StealthConfig
|
|
return StealthConfig()
|
|
|
|
|
|
@pytest.fixture
|
|
def mock_browser_context():
|
|
"""Mock Playwright BrowserContext for stealth testing."""
|
|
context = MagicMock()
|
|
context.add_init_script = MagicMock()
|
|
return context
|