feat(browser): migrate to CloakBrowser for browser emulation and streamline dependency management

- Replace Playwright-based browser emulation with CloakBrowser as default - Update config to support CloakBrowser options and humanization presets - Refactor browser helper to use CloakBrowser context and remove cf_clearance dependency - Update Dockerfile, entrypoint, and update scripts to install CloakBrowser runtime - Ensure CloakBrowser kernel is pre-installed during local setup and dependency updates - Add tests for CloakBrowser integration and legacy compatibility
2026-06-08 17:20:22 +08:00 · 2026-05-16 20:51:38 +08:00
parent 9069dccb2a
commit 0ee9fec1d2
10 changed files with 293 additions and 140 deletions
--- a/tests/test_browser_helper.py
+++ b/tests/test_browser_helper.py
@@ -0,0 +1,95 @@
+from __future__ import annotations
+
+import unittest
+from unittest.mock import patch
+
+from app.helper.browser import PlaywrightHelper
+
+
+class _FakePage:
+    def __init__(self) -> None:
+        self.headers = None
+        self.loaded_url = None
+        self.closed = False
+
+    def set_extra_http_headers(self, headers: dict[str, str]) -> None:
+        self.headers = headers
+
+    def goto(self, url: str) -> None:
+        self.loaded_url = url
+
+    def wait_for_load_state(self, _state: str, timeout: int) -> None:
+        self.timeout = timeout
+
+    def content(self) -> str:
+        return "<html>ok</html>"
+
+    def close(self) -> None:
+        self.closed = True
+
+
+class _FakeContext:
+    def __init__(self, page: _FakePage) -> None:
+        self.page = page
+        self.closed = False
+
+    def new_page(self) -> _FakePage:
+        return self.page
+
+    def close(self) -> None:
+        self.closed = True
+
+
+class BrowserHelperTests(unittest.TestCase):
+    def _assert_get_page_source_uses_cloakbrowser(self, emulation: str) -> None:
+        page = _FakePage()
+        context = _FakeContext(page)
+
+        with patch("app.helper.browser.settings.BROWSER_EMULATION", emulation), \
+                patch.object(
+                    PlaywrightHelper,
+                    "_PlaywrightHelper__launch_cloakbrowser_context",
+                    return_value=context,
+                ) as launch_context:
+            source = PlaywrightHelper().get_page_source(
+                url="https://example.com",
+                cookies="uid=1",
+                ua="UA",
+                timeout=3,
+            )
+
+        self.assertEqual(source, "<html>ok</html>")
+        launch_context.assert_called_once_with(
+            headless=False,
+            user_agent="UA",
+            proxies=None,
+        )
+        self.assertEqual(page.headers, {"cookie": "uid=1"})
+        self.assertEqual(page.loaded_url, "https://example.com")
+        self.assertTrue(page.closed)
+        self.assertTrue(context.closed)
+
+    def test_default_emulation_uses_cloakbrowser_context(self):
+        self._assert_get_page_source_uses_cloakbrowser("cloakbrowser")
+
+    def test_legacy_playwright_emulation_uses_cloakbrowser_context(self):
+        self._assert_get_page_source_uses_cloakbrowser("Playwright")
+
+    def test_legacy_browser_type_constructor_is_accepted(self):
+        page = _FakePage()
+        context = _FakeContext(page)
+
+        with patch.object(
+            PlaywrightHelper,
+            "_PlaywrightHelper__launch_cloakbrowser_context",
+            return_value=context,
+        ):
+            source = PlaywrightHelper(browser_type="firefox").get_page_source(
+                url="https://example.com"
+            )
+
+        self.assertEqual(source, "<html>ok</html>")
+
+
+if __name__ == "__main__":
+    unittest.main()