Timeline
Aug 1, 2021:
- 11:59 PM Changeset in webkit [280526] by
-
- 3 edits in trunk/LayoutTests
[WinCairo] Unreviewed test gardening
- platform/wincairo-wk1/TestExpectations:
- platform/wincairo/TestExpectations:
- 9:56 PM Changeset in webkit [280525] by
-
- 6 edits in trunk/Source/WebCore
Migrate GlyphOverflow to LayoutUnits instead of ints
https://bugs.webkit.org/show_bug.cgi?id=133040
Patch by Rob Buis <rbuis@igalia.com> on 2021-08-01
Reviewed by Darin Adler.
Migrate GlyphOverflow to LayoutUnits instead of ints. This
is another step in the direction of subpixel layout/rendering.
- platform/graphics/FontCascade.h:
- rendering/LegacyInlineFlowBox.cpp:
(WebCore::LegacyInlineFlowBox::addTextBoxVisualOverflow):
- rendering/LegacyRootInlineBox.cpp:
(WebCore::LegacyRootInlineBox::ascentAndDescentForBox const):
- rendering/RenderText.cpp:
(WebCore::RenderText::computePreferredLogicalWidths):
- style/InlineTextBoxStyle.cpp:
(WebCore::extendIntToFloat):
- 7:46 PM Changeset in webkit [280524] by
-
- 2 edits in trunk/Tools
[Win][DumpRenderTree] Clean up IndexedDB files between tests
https://bugs.webkit.org/show_bug.cgi?id=228692
Reviewed by Ryosuke Niwa.
storage/indexeddb/getdatabases.html was failing because it counted
up the remaining databases of previous tests.
Mac DRT removes IndexedDB after the testing (Bug 192796, r240358).
Win DRT also should do it.
- DumpRenderTree/win/DumpRenderTree.cpp:
(runTest): Call TestRunner::clearAllDatabases after the testing.
- 9:02 AM Changeset in webkit [280523] by
-
- 12 edits in trunk/Source
Introduce an experimental feature to toggle WebRTC socket proxying
https://bugs.webkit.org/show_bug.cgi?id=228681
Reviewed by Eric Carlson.
Source/WebKit:
Decide whether to disable proxying based on experimental feature in WebProcess.
Send the disable relay value from WebProcess to NetworkProcess for TCP sockets.
Manually tested.
- NetworkProcess/webrtc/NetworkRTCProvider.cpp:
(WebKit::NetworkRTCProvider::createClientTCPSocket):
- NetworkProcess/webrtc/NetworkRTCProvider.h:
- NetworkProcess/webrtc/NetworkRTCProvider.messages.in:
- NetworkProcess/webrtc/NetworkRTCTCPSocketCocoa.h:
- NetworkProcess/webrtc/NetworkRTCTCPSocketCocoa.mm:
(WebKit::NetworkRTCTCPSocketCocoa::createClientTCPSocket):
(WebKit::NetworkRTCTCPSocketCocoa::NetworkRTCTCPSocketCocoa):
- WebProcess/Network/webrtc/LibWebRTCProvider.cpp:
(WebKit::RTCSocketFactory::CreateClientTcpSocket):
(WebKit::LibWebRTCProvider::createSocketFactory):
- WebProcess/Network/webrtc/LibWebRTCSocketFactory.cpp:
(WebKit::LibWebRTCSocketFactory::createClientTcpSocket):
- WebProcess/Network/webrtc/LibWebRTCSocketFactory.h:
Source/WTF:
- Scripts/Preferences/WebPreferencesExperimental.yaml:
- 7:10 AM Changeset in webkit [280522] by
-
- 2 edits in trunk/Tools
Add myself (Alexander Mikhaylenko) to contributors.json as a committer
Unreviewed.
- Scripts/webkitpy/common/config/contributors.json:
Jul 31, 2021:
- 3:48 PM Changeset in webkit [280521] by
-
- 2 edits in trunk/Source/WebCore
ListAttributeTargetObserver is needlessly created even when there is no list attribute
https://bugs.webkit.org/show_bug.cgi?id=228541
Reviewed by Yusuke Suzuki.
No new tests because no behavior change.
This is possibly a small improvement (~0.5%) on Speedometer subtests that do innerHTML, but not statistically significant.
Ported from a Chromium change by Adithya Srinivasan
- html/HTMLInputElement.cpp:
(WebCore::HTMLInputElement::resetListAttributeTargetObserver): check if there's a list attribute before making an observer.
- 1:47 PM Changeset in webkit [280520] by
-
- 4 edits in trunk/Source
[JSC] Move JIT_UNICODE_EXPRESSIONS to ENABLE(YARR_JIT_UNICODE_EXPRESSIONS) in PlatformEnable.h
https://bugs.webkit.org/show_bug.cgi?id=228669
Reviewed by Alexey Shvayka.
Source/JavaScriptCore:
Move JIT_UNICODE_EXPRESSIONS to PlatformEnable.h to align to the other YarrJIT flags.
This makes it easy to read that this is enabled on ARM64 and X86_64.
- yarr/YarrJIT.cpp:
Source/WTF:
- wtf/PlatformEnable.h:
- 1:43 PM Changeset in webkit [280519] by
-
- 6 edits in trunk/Source/JavaScriptCore
[JSC] Clean up Yarr by using
static constexpr
https://bugs.webkit.org/show_bug.cgi?id=228678
Reviewed by Alexey Shvayka.
Modernize Yarr code by using static constexpr.
- assembler/AbstractMacroAssembler.h:
(JSC::AbstractMacroAssembler::TrustedImm32::TrustedImm32):
- yarr/YarrCanonicalizeUCS2.cpp:
- yarr/YarrCanonicalizeUCS2.js:
(set characters.hex.set string_appeared_here):
- yarr/YarrErrorCode.cpp:
(JSC::Yarr::errorMessage):
- yarr/YarrJIT.cpp:
- 10:37 AM Changeset in webkit [280518] by
-
- 11 edits in trunk/Source
Improve LocalSampleBufferDisplayLayer logging
https://bugs.webkit.org/show_bug.cgi?id=228582
Reviewed by Eric Carlson.
Source/WebCore:
Add logging to correlate a LocalSampleBufferDisplayLayer to its HTMLMediaElement and its MediaStreamTrack.
Introduce setLogIdentifer for that purpose.
No observable change of behavior.
- platform/graphics/avfoundation/SampleBufferDisplayLayer.h:
- platform/graphics/avfoundation/objc/LocalSampleBufferDisplayLayer.h:
- platform/graphics/avfoundation/objc/LocalSampleBufferDisplayLayer.mm:
(WebCore::LocalSampleBufferDisplayLayer::enqueueSample):
(WebCore::LocalSampleBufferDisplayLayer::onIrregularFrameRateNotification):
- platform/graphics/avfoundation/objc/MediaPlayerPrivateMediaStreamAVFObjC.mm:
(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::ensureLayers):
(WebCore::MediaPlayerPrivateMediaStreamAVFObjC::checkSelectedVideoTrack):
Source/WebKit:
Send through IPC the log identifier to the GPU display layer.
- GPUProcess/webrtc/RemoteSampleBufferDisplayLayer.cpp:
(WebKit::RemoteSampleBufferDisplayLayer::setLogIdentifier):
- GPUProcess/webrtc/RemoteSampleBufferDisplayLayer.h:
- GPUProcess/webrtc/RemoteSampleBufferDisplayLayer.messages.in:
- WebProcess/GPU/webrtc/SampleBufferDisplayLayer.cpp:
(WebKit::SampleBufferDisplayLayer::setLogIdentifier):
- WebProcess/GPU/webrtc/SampleBufferDisplayLayer.h:
- 9:53 AM Changeset in webkit [280517] by
-
- 3 edits in trunk/Source/WebCore
AVVideoCaptureSource should not start/stop the camera if trying to apply the same size/frame rate configuration
https://bugs.webkit.org/show_bug.cgi?id=228634
<rdar://81174160>
Reviewed by Eric Carlson.
In some configurations, pages want to set a frame rate of 30, but the actual frame rate set will be 29.9999.
In that case, our heuristic to not recalibrate the camera fails.
To prevent this, we add checks in AVVideoCaptureSource to validate whether the new presets are the same as the already applied ones.
Manually tested.
- platform/mediastream/mac/AVVideoCaptureSource.h:
- platform/mediastream/mac/AVVideoCaptureSource.mm:
(WebCore::AVVideoCaptureSource::setSessionSizeAndFrameRate):
- 9:52 AM Changeset in webkit [280516] by
-
- 2 edits in trunk/Source/WebKit
RemoteAudioMediaStreamTrackRendererInternalUnit should handle changes of requested sampleCount
https://bugs.webkit.org/show_bug.cgi?id=228631
<rdar://80690191>
Reviewed by Eric Carlson.
The requested number of samples might change in case the output device changes.
When changing device, we might have a generated buffer of 4096 while 1024 samples are requested.
For each 1024 sample chunk, we will request a buffer of 4096, which cannot be provided by WebProcess.
To prevent this case, we keep track of a requested sample count overshoot so that next time the render callback is called,
we might not request to generate another chunk.
- GPUProcess/webrtc/RemoteAudioMediaStreamTrackRendererInternalUnitManager.cpp:
(WebKit::RemoteAudioMediaStreamTrackRendererInternalUnitManager::Unit::start):
(WebKit::RemoteAudioMediaStreamTrackRendererInternalUnitManager::Unit::render):
- 8:43 AM Changeset in webkit [280515] by
-
- 4 edits2 adds in trunk
REGRESSION(r279564): [GTK] Mouse wheel scrolling scrolls too little (take two)
https://bugs.webkit.org/show_bug.cgi?id=228572
Reviewed by Adrian Perez de Castro.
Source/WebCore:
Test: platform/gtk/fast/scrolling/scroll-wheel-repeated-events.html
When scrolling by a delta with ScrollAnimationSmooth, base the target point
on the current scroll target instead of on the current position. We also
pass the point to updatePerAxisData instead of the delta in order to make
it more explicit what the desired behavior is for this method.
- platform/ScrollAnimationSmooth.h: Update argument name.
- platform/ScrollAnimationSmooth.cpp:
(WebCore::ScrollAnimationSmooth::scroll): Base the delta on the target position and
pass that as the argument to updatePerAxisData. When scrolling to a point, simply
pass that to updatePerAxisData.
(WebCore::ScrollAnimationSmooth::updatePerAxisData): Accept a point instead of a
delta, which makes it more obvious how to use this function and allows it to
be used the same way by both version of ScrollAnimationSmooth::scroll.
LayoutTests:
- platform/gtk/fast/scrolling/scroll-wheel-repeated-events-expected.txt: Added.
- platform/gtk/fast/scrolling/scroll-wheel-repeated-events.html: Added.
- 2:43 AM Changeset in webkit [280514] by
-
- 4773 edits322 adds in trunk/JSTests
[JSC] Upgrade test262
https://bugs.webkit.org/show_bug.cgi?id=228627
Reviewed by Ross Kirsling.
- test262/config.yaml:
- test262/expectations.yaml:
- 1:29 AM Changeset in webkit [280513] by
-
- 2 edits in trunk/Source/JavaScriptCore
Unreviewed. Remove the build warning below since r280493.
warning: variable ‘block’ set but not used [-Wunused-but-set-variable]
- b3/B3CanonicalizePrePostIncrements.cpp:
(JSC::B3::canonicalizePrePostIncrements):
- 1:06 AM Changeset in webkit [280512] by
-
- 1 edit9 moves in trunk/LayoutTests
Move DOM-unrelated tests out of fast/regex/dom
https://bugs.webkit.org/show_bug.cgi?id=228661
Reviewed by Saam Barati.
By moving to fast/regex/, we can run them from JSC tests too.
- fast/regex/lastIndex-expected.txt: Renamed from LayoutTests/fast/regex/dom/lastIndex-expected.txt.
- fast/regex/lastIndex.html: Renamed from LayoutTests/fast/regex/dom/lastIndex.html.
- fast/regex/non-pattern-characters-expected.txt: Renamed from LayoutTests/fast/regex/dom/non-pattern-characters-expected.txt.
- fast/regex/non-pattern-characters.html: Renamed from LayoutTests/fast/regex/dom/non-pattern-characters.html.
- fast/regex/script-tests/lastIndex.js: Renamed from LayoutTests/fast/regex/dom/script-tests/lastIndex.js.
- fast/regex/script-tests/non-pattern-characters.js: Renamed from LayoutTests/fast/regex/dom/script-tests/non-pattern-characters.js.
- fast/regex/script-tests/unicodeCaseInsensitive.js: Renamed from LayoutTests/fast/regex/dom/script-tests/unicodeCaseInsensitive.js.
- fast/regex/unicodeCaseInsensitive-expected.txt: Renamed from LayoutTests/fast/regex/dom/unicodeCaseInsensitive-expected.txt.
- fast/regex/unicodeCaseInsensitive.html: Renamed from LayoutTests/fast/regex/dom/unicodeCaseInsensitive.html.
- 12:52 AM Changeset in webkit [280511] by
-
- 7 edits1 delete in trunk
Unreviewed, reverting r280488.
https://bugs.webkit.org/show_bug.cgi?id=228677
media/modern-media-controls/ tests are failing or timed-out
Reverted changeset:
"[Cocoa] Remove support for AVAssetImageGenerator"
https://bugs.webkit.org/show_bug.cgi?id=228560
https://commits.webkit.org/r280488
