Automatic update from web-platform-tests
[Azure Pipelines] tweak job and display names
Non-lowercase macOS in the job names looks out of place now, so use _mac
to match _win.
Parenthesis in display names aren't great because in the GitHub Checks
API integration, the check name is "$pipelineName ($jobName)", leading
to check names like "Azure Pipelines (tools/ unittests: macOS (Python 3.6))".
(The pipeline name is indeed "Azure Pipelines".)
--
wpt-commits: b1c034d31dccb568749e042938db02dd1be2b72c
wpt-pr: 23550
Automatic update from web-platform-tests
Mark idlharness.html test as timeout=long
This is symptomatic of crbug.com/1047818, but to stop the pain for
sheriffs lets mark it timeout=long for now.
Bug: 1080606
Change-Id: I3dc21ea2afca8aff08dcf09e146bb41c196aa1b3
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2197288
Auto-Submit: Stephen McGruer <smcgruer@chromium.org>
Commit-Queue: Stephen McGruer <smcgruer@chromium.org>
Commit-Queue: David Bokan <bokan@chromium.org>
Reviewed-by: David Bokan <bokan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#767979}
--
wpt-commits: 53419f622f307b330fcbe4c49aaed9a8c7630348
wpt-pr: 23553
Automatic update from web-platform-tests
Introduce test infrastructure for QuicTransport (#22844)
This change introduces test infrastructure for QuicTransport.
See also: https://github.com/web-platform-tests/rfcs/blob/master/rfcs/quic.md
tools/quic contains the test server and files needed by the server such as
certificate files (TODO: we will switch to the same certificate used by
wptserve once aioquic 0.8.8 is released).
tools/quic/quic_transport_server.py is based on
https://github.com/aiortc/aioquic/blob/master/examples/http3_server.py
webtransport/quic contains a test example and a sample custom handler.
This change doesn't contain a means to run the QuicTransport server
automatically.
Tracking issue: #19114
--
wpt-commits: 512cf2462e561c4657e83ca2f6dc2f9fb3ca133e
wpt-pr: 22844
Automatic update from web-platform-tests
Remove the `[SecureContext]` restriction from Trusted Types.
While it's reasonable to exclude new APIs from non-secure contexts, the
ancestry requirements allow attackers to disable restricted APIs from
embedded contexts. This is usually excellent, as it means that data
won't leak from secure to non-secure contexts. For security features,
on the other hand, this gives the attacker some advantage with regard
to embedded contexts' mitigtions.
This is unfortunate, and this patch removes the restriction to ensure
that embedded contexts can continue to mitigate the effect of XSS
attack by reverting https://crrev.com/c/2093214 and
https://crrev.com/c/2098076.
Bug: 1059554
Change-Id: Ib948437310509f1d29cacff1e6c74ab7cbc30d11
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2195965
Commit-Queue: Mike West <mkwst@chromium.org>
Reviewed-by: Daniel Vogelheim <vogelheim@chromium.org>
Cr-Commit-Position: refs/heads/master@{#767894}
--
wpt-commits: 094353fedf808caf83e82b4959b5edf02e1be92e
wpt-pr: 23525
Automatic update from web-platform-tests
[Azure Pipelines] run wpt_integration_{macOS,win}_{py36,py38} entirely using py3 (#23536)
py2 was needed to update the manifest only, but that now works with py3.
Incidentally fixes https://github.com/web-platform-tests/wpt/issues/23441,
where the problem appears to have been some mixing of py2 and py3.
This also renames job names from _py3 to _py36.
--
wpt-commits: 01e8699486be3c1e3111476310cba4ae70271d86
wpt-pr: 23536
Automatic update from web-platform-tests
device orientation: Properly reject promise in waitForEvent().
Follow-up to commit 4b04dea7f0ef ("device orientation: Try to remove
flakiness from tests").
When assertEventEquals() fails in waitForEvent(), the exception should be
propagated through a promise rejection, otherwise the promise will neither
resolve nor reject, and the test will time out.
The source of the flakiness still needs to be determined, but the flaky
tests should now fail fast rather than time out.
Bug: 1078298
Change-Id: I990f2c7982084938bfa36d76d1b0f18ec458e38e
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2196584
Commit-Queue: Raphael Kubo da Costa <raphael.kubo.da.costa@intel.com>
Commit-Queue: Reilly Grant <reillyg@chromium.org>
Auto-Submit: Raphael Kubo da Costa <raphael.kubo.da.costa@intel.com>
Reviewed-by: Reilly Grant <reillyg@chromium.org>
Cr-Commit-Position: refs/heads/master@{#767861}
--
wpt-commits: 030299ef4d6a5ea18d45e77f92de75a15a769108
wpt-pr: 23540
Automatic update from web-platform-tests
Pass logger explicitly in to Executors This means that we are able to report errors which happen during setup
--
wpt-commits: 48b773a266795216e28ac6b8e55f300bb6ad2541
wpt-pr: 23477
Automatic update from web-platform-tests
Add MathML for automatic scriptlevel attribute (#23543)
* Add MathML for automatic scriptlevel attribute
This also checks dynamic/case-insensitiveness of "accent" and "accentunder"
https://github.com/mathml-refresh/mathml/issues/57
--
wpt-commits: 14ec365a5d98c7de9c5debbfd0bd1c75a9be3c41
wpt-pr: 23543
Automatic update from web-platform-tests
Add state to check if Resize Observer Loop Limit error was dispatched
Currently, if the resize observer loop limit error is dispatched and
the error callback affects layout causing lifecycle state to be <
kPrePaintClean, an infinite loop may occur causing unexpected behaviour.
To fix this, state is added to the resize observer controller in order
to achieve the following flow:
- Loop limit error is reached
- Error callback causes lifecycle state to be < kPrePaintClean
- Lifecycle steps including resize observer steps are run again to move
state to kPrePaintClean
- Error is not dispatched a second time
- Proceed to the paint step
Bug: 1076205
Change-Id: I904392d1dc73caf874f3d7ecf572a74422f0ab11
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2189352
Reviewed-by: Philip Rogers <pdr@chromium.org>
Reviewed-by: Daniel Libby <dlibby@microsoft.com>
Commit-Queue: Sahir Vellani <sahir.vellani@microsoft.com>
Cr-Commit-Position: refs/heads/master@{#767841}
--
wpt-commits: 166ced72a2f9487dfa477ab79e6616b4fc890fbf
wpt-pr: 23514
Automatic update from web-platform-tests
[manifest] Add "quic" property to tests
--
[wptrunner] Add a switch to include QUIC tests
--enable-quic, set to false by default.
This flag also controls whether to assign a port to QUIC in the server
config, which will then be passed to `wpt serve` to determine whether to
start the QUIC server (to be implemented).
--
Document QUIC-related flags and metadata
--
wpt-commits: af9ed5fb0295c7e3f108bde6a757d2d3ce4f17a3, 041acf40dfc13df66f5db09248d00c3e102ba3c2, d5d85c26602ec949dd1ca1dcb52cda1ec7b724b5
wpt-pr: 23468
Automatic update from web-platform-tests
Handle k-rate AudioParam inputs for AudioWorkletNode
Support k-rate input to AudioParams for an AudioWorkletNode.
Previously, inputs to a k-rate AudioParam were ignored, which is
incorrect. The input must be summed with any timeline value.
Manually ran the new test case with Chrome stable and the tests fail
as expected (output is constantly 0 which is wrong).
Bug: 1015760
Test: k-rate-audioworklet-connections.https.html
Change-Id: I5dbba311fb84b9a162e1b2ac4fc9ef902ff36eb6
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2194189
Commit-Queue: Raymond Toy <rtoy@chromium.org>
Reviewed-by: Hongchan Choi <hongchan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#767820}
--
wpt-commits: f0799f95887c1d0401fc2abee68f0e5cf8c20463
wpt-pr: 23518
Automatic update from web-platform-tests
Update some scriptlevel tests to match the new syntax (part 2). (#23538)
math-script-level-003: remove the reference left in previous commit.
math-script-level-auto-and-math-style-*: use font-size: scriptlevel(auto) and
update some assert doc.
https://github.com/mathml-refresh/mathml/issues/174
--
wpt-commits: 907d95d79df0f6d403b97ac9e494292eea7f0e9e
wpt-pr: 23538
Automatic update from web-platform-tests
Update flake8 from 3.7.9 to 3.8.1 (#23523)
Co-authored-by: Robert Ma <robertma@chromium.org>
--
wpt-commits: 942150b8b8f95bcecd5b5cecb8f8cd7ef0683703
wpt-pr: 23523
Automatic update from web-platform-tests
Add a test to check the video detector helper function.
This also checks that the video luma settles to a reasonable value
for a large range of values, so could be considered a codec test.
WPT bug: https://github.com/web-platform-tests/wpt/issues/23340
Bug: none
Change-Id: I6ce0537b90903d823945e2fe3c5389fa4c132608
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/2172971
Commit-Queue: Harald Alvestrand <hta@chromium.org>
Reviewed-by: Philip Jägenstedt <foolip@chromium.org>
Cr-Commit-Position: refs/heads/master@{#767270}
--
wpt-commits: cdd4b43f2f89422f36d81546df70eb09f6aa8ed0
wpt-pr: 23341
With current gecko, there are cases that compositor window is used even when it is not necessary. For example, fallback from DirectComposite with WebRender and fallback from double buffering with Compositor. With some gpu drivers, the fallbacks with compositor window causes rendering problem.
Differential Revision: https://phabricator.services.mozilla.com/D76470
Forgot to add test to the first commit.
Moved bug reference link to seperate link tag in scroll-target-margin-004.html and changed comment on modified css line.
Differential Revision: https://phabricator.services.mozilla.com/D76685
IsInNativeAnonymousSubtree and IsRootOfNativeAnonymousSubtree are
exposed in nsINode, so we don't need to make sure to call them on
nsIContent objects.
Depends on D76681
Differential Revision: https://phabricator.services.mozilla.com/D76682
In favor of the NativeAnonymous versions which they forward to.
Done automatically with:
rg -l 'IsInAnonymousSubtree' | xargs sed -i 's/IsInAnonymousSubtree/IsInNativeAnonymousSubtree/g'
And removing the function definitions afterwards.
Differential Revision: https://phabricator.services.mozilla.com/D76681