~95% on SimpleQA (e.g. Qwen3.6-27B on a 3090). Supports all local and cloud LLMs (llama.cpp, Ollama, Google, ...). 10+ search engines - arXiv, PubMed, your private documents. Everything Local & Encrypted.
Python academiaanthropicarxivbravedeep-researchencryptionhome-automationhomeserverlocallocal-deep-researchlocal-llmmistralollamaopenaipubmedresearchresearch-toolretrieval-augmented-generationsearxngself-hosted
"""
Fixtures for content_fetcher tests.
"""
import pytest
from unittest.mock import MagicMock, patch
@pytest.fixture
def mock_html_response():
"""Mock HTML response for testing."""
return """
<!DOCTYPE html>
<html>
<head>
<title>Test Article Title</title>
<meta name="description" content="This is a test article description.">
<meta name="author" content="Test Author">
</head>
<body>
<nav>Navigation menu to be removed</nav>
<article>
<h1>Main Article Heading</h1>
<p>This is the first paragraph of the article content. It contains important information about the topic being discussed.</p>
<p>This is the second paragraph with more details about the subject matter.</p>
<h2>Subheading</h2>
<p>Additional content under the subheading explaining more concepts.</p>
</article>
<footer>Footer content to be removed</footer>
</body>
</html>
"""
@pytest.fixture
def mock_arxiv_html():
"""Mock arXiv abstract page."""
return """
<!DOCTYPE html>
<html>
<head>
<title>[2301.12345] Attention Is All You Need</title>
</head>
<body>
<div id="abs">
<h1>Attention Is All You Need</h1>
<div class="authors">Vaswani et al.</div>
<blockquote class="abstract">
The dominant sequence transduction models are based on complex recurrent or
convolutional neural networks. We propose a new simple network architecture,
the Transformer, based solely on attention mechanisms.
</blockquote>
</div>
</body>
</html>
"""
@pytest.fixture
def mock_session():
"""Mock requests session."""
mock = MagicMock()
mock.headers = {}
return mock
@pytest.fixture
def mock_successful_response(mock_html_response):
"""Mock successful HTTP response."""
response = MagicMock()
response.status_code = 200
response.text = mock_html_response
response.content = mock_html_response.encode("utf-8")
response.headers = {"content-type": "text/html; charset=utf-8"}
return response
@pytest.fixture
def mock_pdf_response():
"""Mock PDF response (minimal valid PDF)."""
# Minimal PDF content
pdf_content = b"%PDF-1.4\n1 0 obj\n<<>>\nendobj\ntrailer\n<<>>\n%%EOF"
response = MagicMock()
response.status_code = 200
response.content = pdf_content
response.headers = {"content-type": "application/pdf"}
return response
@pytest.fixture
def mock_rate_tracker():
"""Mock rate tracker."""
with patch(
"local_deep_research.research_library.downloaders.html.AdaptiveRateLimitTracker"
) as mock:
tracker = MagicMock()
tracker.apply_rate_limit.return_value = 0
mock.return_value = tracker
yield tracker