From 4067bce3d0edaef93d89e00ac70ebd7297b7d4aa Mon Sep 17 00:00:00 2001 From: Vlada Dusek Date: Wed, 18 Nov 2026 07:46:21 +0100 Subject: [PATCH 1/2] test: drop `networkidle` wait in `block_requests` test --- tests/integration/crawlers/_playwright/test_utils.py | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/tests/integration/crawlers/_playwright/test_utils.py b/tests/integration/crawlers/_playwright/test_utils.py index c674401567..9d303770a6 100644 --- a/tests/integration/crawlers/_playwright/test_utils.py +++ b/tests/integration/crawlers/_playwright/test_utils.py @@ -114,11 +114,13 @@ async def test_block_requests_default(server_url: URL) -> None: target_url = str(server_url / 'resource_loading_page') + # `goto` returns on the `load` event, which the browser fires only after the script and image requests have + # finished or failed, so the tracked sets are complete once it returns. + # Default behavior, nothing is blocked page = await browser.new_page() requested_no_block, blocked_no_block = track_requests(page) await page.goto(target_url) - await page.wait_for_load_state('networkidle') await page.close() # With blocking @@ -126,7 +128,6 @@ async def test_block_requests_default(server_url: URL) -> None: requested_blocked, blocked_blocked = track_requests(page) await block_requests(page) await page.goto(target_url) - await page.wait_for_load_state('networkidle') await page.close() await browser.close() From 06dacb9d9058e984f57e75425c217ddb3b9dff43 Mon Sep 17 00:00:00 2001 From: Vlada Dusek Date: Wed, 18 Nov 2026 08:03:16 +0100 Subject: [PATCH 2/2] test: track blocked requests in the `block_requests` sibling tests --- .../crawlers/_playwright/test_utils.py | 26 ++++++++----------- 1 file changed, 11 insertions(+), 15 deletions(-) diff --git a/tests/integration/crawlers/_playwright/test_utils.py b/tests/integration/crawlers/_playwright/test_utils.py index 9d303770a6..edc289307a 100644 --- a/tests/integration/crawlers/_playwright/test_utils.py +++ b/tests/integration/crawlers/_playwright/test_utils.py @@ -17,6 +17,9 @@ def track_requests(page: Page) -> tuple[set[str], set[str]]: Chromium fails a request blocked by `Network.setBlockedURLs` with the `inspector` reason. Any other failure, such as `net::ERR_NO_BUFFER_SPACE` on a loaded Windows runner, is a transport error and does not count as blocked. + + The sets are complete once `page.goto` returns: it waits for the `load` event, which the browser fires only after + the page's script and image requests have finished or failed. """ requested: set[str] = set() blocked: set[str] = set() @@ -114,9 +117,6 @@ async def test_block_requests_default(server_url: URL) -> None: target_url = str(server_url / 'resource_loading_page') - # `goto` returns on the `load` event, which the browser fires only after the script and image requests have - # finished or failed, so the tracked sets are complete once it returns. - # Default behavior, nothing is blocked page = await browser.new_page() requested_no_block, blocked_no_block = track_requests(page) @@ -149,18 +149,16 @@ async def test_block_requests_with_extra_patterns(server_url: URL) -> None: target_url = str(server_url / 'resource_loading_page') page = await browser.new_page() - loaded_urls_blocked = [] - - page.on('requestfinished', lambda req: loaded_urls_blocked.append(req.url.rsplit('/', 1)[-1])) + requested, blocked = track_requests(page) await block_requests(page, extra_url_patterns=['*.js']) await page.goto(target_url) - await page.wait_for_load_state('networkidle') await page.close() await browser.close() - # With blocking, only HTML should load - assert set(loaded_urls_blocked) == {'resource_loading_page'} + # The page requests both resources, and both the image and the script are blocked + assert requested == {'resource_loading_page', 'test.js', 'test.png'} + assert blocked == {'test.js', 'test.png'} async def test_block_requests_with_custom_patterns(server_url: URL) -> None: @@ -171,15 +169,13 @@ async def test_block_requests_with_custom_patterns(server_url: URL) -> None: target_url = str(server_url / 'resource_loading_page') page = await browser.new_page() - loaded_urls_blocked = [] - - page.on('requestfinished', lambda req: loaded_urls_blocked.append(req.url.rsplit('/', 1)[-1])) + requested, blocked = track_requests(page) await block_requests(page, url_patterns=['*.js']) await page.goto(target_url) - await page.wait_for_load_state('networkidle') await page.close() await browser.close() - # With blocking, only PNG should load - assert set(loaded_urls_blocked) == {'resource_loading_page', 'test.png'} + # The page requests both resources, and only the script is blocked + assert requested == {'resource_loading_page', 'test.js', 'test.png'} + assert blocked == {'test.js'}