diff --git a/src/endpoints.py b/src/endpoints.py index c7d0afd..49a9b29 100644 --- a/src/endpoints.py +++ b/src/endpoints.py @@ -7,7 +7,6 @@ from typing import Annotated from fastapi import APIRouter, Depends, HTTPException from fastapi.responses import RedirectResponse -from playwright.async_api import Page from playwright.async_api import TimeoutError as PlaywrightTimeoutError from playwright_captcha import CaptchaType @@ -52,23 +51,6 @@ async def health_check(sb: BrowserDep): return HealthcheckResponse(user_agent=health_check_request.solution.user_agent) -async def _wait_for_networkidle(page: Page, timer: TimeoutTimer) -> None: - """ - Wait for the network to go idle; a timeout is non-fatal. - - Some sites keep background connections open (analytics beacons, - websockets, ...), so ``networkidle`` may never settle even though the page - is fully usable once ``domcontentloaded`` has fired. Log and continue - instead of failing the request. - """ - try: - await page.wait_for_load_state("networkidle", timeout=timer.remaining() * 1000) - except PlaywrightTimeoutError: - logger.info( - "networkidle timed out after domcontentloaded; continuing with loaded page" - ) - - @router.post("/v1") async def read_item(request: LinkRequest, dep: BrowserDep) -> LinkResponse: """Handle POST requests.""" @@ -111,7 +93,18 @@ async def read_item(request: LinkRequest, dep: BrowserDep) -> LinkResponse: status = HTTPStatus.OK logger.debug("Challenge solved successfully.") else: - await _wait_for_networkidle(dep.page, timer) + # Best-effort: some sites keep background connections open + # (analytics beacons, websockets, ...), so ``networkidle`` may + # never settle. The page is fully usable once ``domcontentloaded`` + # has fired, so log and continue instead of failing the request. + try: + await dep.page.wait_for_load_state( + "networkidle", timeout=timer.remaining() * 1000 + ) + except PlaywrightTimeoutError: + logger.info( + "networkidle timed out after domcontentloaded; continuing with loaded page" + ) except (TimeoutError, PlaywrightTimeoutError) as e: logger.error("Timed out while loading the page or solving the challenge") raise HTTPException( diff --git a/tests/main_test.py b/tests/main_test.py index ac9c74b..df512db 100644 --- a/tests/main_test.py +++ b/tests/main_test.py @@ -1,6 +1,6 @@ from http import HTTPStatus from json import JSONDecodeError -from types import SimpleNamespace +from unittest.mock import AsyncMock, MagicMock import httpx import pytest @@ -106,55 +106,28 @@ def test_max_timeout_normalization(payload: dict, expected: int): assert request.max_timeout == expected -class FakeLoadStateError(PlaywrightTimeoutError): - """Raised by FakePage.wait_for_load_state for configured failing states.""" +def fake_dep(*, fail_states: set[str] | None = None) -> BrowserDepClass: + """Build a browser dependency triple backed by mocks.""" + page = AsyncMock() + page.url = "https://example.test/login" + page.goto.return_value = MagicMock( + status=HTTPStatus.OK, headers={"content-type": "text/html"} + ) + page.title.return_value = "Login" + page.evaluate.return_value = "UnitTestBrowser/1.0" + page.content.return_value = "