mirror of
https://github.com/bytedance/deer-flow.git
synced 2026-10-27 13:40:37 +00:00
6783073a39
* fix(image-search): screen DuckDuckGo result URLs * fix(image-search): unify provider result URL screening * docs(image-search): scope URL guidance to avoid inherited size limits --------- Co-authored-by: huyengiang101086-spec <270684046+huyengiang101086-spec@users.noreply.github.com> Co-authored-by: Willem Jiang <willem.jiang@gmail.com>
46 lines
1.5 KiB
Python
46 lines
1.5 KiB
Python
"""The image providers enforce the same string-only result URL policy."""
|
|
|
|
import importlib
|
|
from unittest.mock import patch
|
|
|
|
import pytest
|
|
|
|
|
|
@pytest.fixture(params=["url_safety", "brave.tools", "serper.tools"])
|
|
def screen(request):
|
|
module = importlib.import_module(f"deerflow.community.{request.param}")
|
|
return module.safe_public_result_url if request.param == "url_safety" else module._safe_public_url
|
|
|
|
|
|
@pytest.mark.parametrize(
|
|
"url",
|
|
[
|
|
"http://metadata.google.internal/image",
|
|
"http://[64:ff9b::7f00:1]/image",
|
|
"http://[::ffff:127.0.0.1]/image",
|
|
"http://[2002:7f00:1::]/image",
|
|
"http://[::10.0.0.1]/image",
|
|
"http://224.0.0.1/image",
|
|
"http://3758096385/image",
|
|
"http://[ff02::1]/image",
|
|
"http://[2002:e000:1::]/image",
|
|
"http://240.0.0.1/image",
|
|
"http://127.0.0.1/image",
|
|
"http://0x7f000001/image",
|
|
"http://0177.0.0.1/image",
|
|
"http://localhost./image",
|
|
"file:///etc/passwd",
|
|
],
|
|
)
|
|
def test_image_providers_reject_non_public_results(screen, url):
|
|
assert screen(url) == ""
|
|
|
|
|
|
@pytest.mark.parametrize(
|
|
"url",
|
|
["https://images.example/image", "http://8.8.8.8/image", "http://134744072/image", "http://[2606:4700:4700::1111]/image"],
|
|
)
|
|
def test_image_providers_keep_public_results_without_dns(screen, url):
|
|
with patch("socket.getaddrinfo", side_effect=AssertionError("Result screening must not perform DNS")):
|
|
assert screen(url) == url
|