⚠ Archived content — this site is no longer maintained.   Current WebKit documentation is at docs.webkit.org.

Timeline



Nov 3, 2018:

4:24 PM Changeset in webkit [237777] by Devin Rousso
  • 19 edits in trunk

Web Inspector: Canvas: capture changes to <canvas> that would affect the recorded context
​https://bugs.webkit.org/show_bug.cgi?id=190854

Reviewed by Matt Baker.

Source/WebCore:

Updated existing tests: inspector/canvas/recording-2d.html

inspector/canvas/recording-bitmaprenderer.html
inspector/canvas/recording-webgl.html

  • html/HTMLCanvasElement.idl:

Apply CallTracingCallback=recordCanvasAction to the width and height attributes so
that they are recorded through the same path as CanvasRenderingContext.

  • html/CanvasBase.h:
  • html/CanvasBase.cpp:

(WebCore::CanvasBase::callTracingActive const): Added.

  • bindings/js/CallTracer.h:
  • bindings/js/CallTracer.cpp:

(WebCore::CallTracer::recordCanvasAction):

Source/WebInspectorUI:

  • UserInterface/Models/RecordingAction.js:

(WI.RecordingAction):
(WI.RecordingAction.isFunctionForType):
(WI.RecordingAction.constantNameForParameter):
(WI.RecordingAction.prototype.get contextReplacer): Added.
(WI.RecordingAction.prototype.async.swizzle):
(WI.RecordingAction.prototype.apply):
Create a constant list of actions for each recording type that need to replace the context
with a different value before being applied (e.g. width should be applied to the
context's canvas instead of directly to the context).

  • UserInterface/Views/RecordingContentView.js:

(WI.RecordingContentView.prototype._generateContentCanvas2D.actionModifiesPath): Added.
(WI.RecordingContentView.prototype._generateContentCanvas2D):
(WI.RecordingContentView._actionModifiesPath): Deleted.
Generate the path context after the actions are applied to the preview context so that the
final width/height are known and can be used. This is needed because changing the
width/height causes the content to be erased.

  • UserInterface/Views/RecordingActionTreeElement.js:

(WI.RecordingActionTreeElement._generateDOM):
(WI.RecordingActionTreeElement._classNameForAction):

  • UserInterface/Views/RecordingActionTreeElement.css:

(.tree-outline:focus .item.action.selected:not(.invalid, .initial-state, .has-context-replacer) > .icon): Added.
(.item.action > .titles .context-replacer::after): Added.
(.item.action.has-context-replacer > .icon): Added.
(@media (prefers-dark-interface) .item.action:not(.invalid, .initial-state, .has-context-replacer) > .icon): Added.
(.tree-outline:focus .item.action.selected:not(.initial-state, .invalid) > .icon): Deleted.
(@media (prefers-dark-interface) .item.action:not(.initial-state) > .icon): Deleted.
(@media (prefers-dark-interface) .tree-outline:not(.hide-disclosure-buttons) .item.action:not(.initial-state, .parent) > .icon): Deleted.
Add the context replacer text to the beginning of the action's name if it exists.

  • UserInterface/Views/CanvasContentView.js:

(WI.CanvasContentView.prototype._refreshPixelSize):
(WI.CanvasContentView.prototype._updatePixelSize): Deleted.
Update preview image when the canvas' size changes.

LayoutTests:

  • inspector/canvas/recording-2d-expected.txt:
  • inspector/canvas/recording-2d.html:
  • inspector/canvas/recording-bitmaprenderer-expected.txt:
  • inspector/canvas/recording-bitmaprenderer.html:
  • inspector/canvas/recording-webgl-expected.txt:
  • inspector/canvas/recording-webgl.html:
11:45 AM Changeset in webkit [237776] by aestes@apple.com
  • 5 edits in trunk

[Payment Request] PaymentResponse.retry()'s errorFields should be optional
​https://bugs.webkit.org/show_bug.cgi?id=191212

Reviewed by Youenn Fablet.

Source/WebCore:

Per WebIDL, the errorFields argument to PaymentResponse.retry() should be optional. See
<​https://github.com/w3c/payment-request/issues/804> for details.

Added test case to http/tests/paymentrequest/payment-response-retry-method.https.html.

  • Modules/paymentrequest/PaymentResponse.idl:

LayoutTests:

  • http/tests/paymentrequest/payment-response-retry-method.https-expected.txt:
  • http/tests/paymentrequest/payment-response-retry-method.https.html:
10:57 AM Changeset in webkit [237775] by Alan Bujtas
  • 6 edits
    2 adds in trunk

[LFC][BFC] Resolve percentage height values.
​https://bugs.webkit.org/show_bug.cgi?id=191224

Reviewed by Antti Koivisto.

Source/WebCore:

10.5 Content height: the 'height' property
The percentage is calculated with respect to the height of the generated box's containing block.
If the height of the containing block is not specified explicitly (i.e., it depends on content height),
and this element is not absolutely positioned, the used height is calculated as if 'auto' was specified.

​https://www.w3.org/TR/CSS22/visudet.html#propdef-height

Test: fast/block/basic/height-percentage-simple.html

  • layout/FormattingContextGeometry.cpp:

(WebCore::Layout::isHeightAuto):
(WebCore::Layout::computedHeightValue):
(WebCore::Layout::contentHeightForFormattingContextRoot):
(WebCore::Layout::FormattingContext::Geometry::computedMaxHeight):
(WebCore::Layout::FormattingContext::Geometry::computedMinHeight):
(WebCore::Layout::FormattingContext::Geometry::outOfFlowNonReplacedVerticalGeometry):
(WebCore::Layout::FormattingContext::Geometry::complicatedCases):
(WebCore::Layout::FormattingContext::Geometry::inlineReplacedHeightAndMargin):
(WebCore::Layout::FormattingContext::Geometry::inlineReplacedWidthAndMargin):

Tools:

  • LayoutReloaded/misc/LFC-passing-tests.txt:

LayoutTests:

  • fast/block/basic/height-percentage-simple-expected.txt: Added.
  • fast/block/basic/height-percentage-simple.html: Added.
  • platform/ios/TestExpectations:
8:22 AM Changeset in webkit [237774] by achristensen@apple.com
  • 9 edits
    2 adds in trunk

Mac production builds should sign the network process xpc service with entitlements
​https://bugs.webkit.org/show_bug.cgi?id=191209
<rdar://problem/45377609>

Reviewed by Geoff Garen.

Source/WebKit:

  • Configurations/Network-OSX.entitlements: Copied from Configurations/Network-OSX-sandbox.entitlements.
  • Configurations/NetworkService.xcconfig:

Set CODE_SIGN_ENTITLEMENTS_COCOA_TOUCH_NO like we do for the web process,
except with the corresponoding network process entitlements file.

  • UIProcess/API/Cocoa/WKProcessPool.mm:

(-[WKProcessPool _networkProcessHasEntitlementForTesting:]):

  • UIProcess/API/Cocoa/WKProcessPoolPrivate.h:
  • UIProcess/Cocoa/WebProcessPoolCocoa.mm:

(WebKit::WebProcessPool::networkProcessHasEntitlementForTesting):

  • UIProcess/WebProcessPool.h:
  • WebKit.xcodeproj/project.pbxproj:

Add infrastructure to test that the network process actually has an entitlement.

Tools:

  • TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
  • TestWebKitAPI/Tests/WebKitCocoa/NetworkProcess.mm: Added.

(TEST):

7:57 AM Changeset in webkit [237773] by youenn@apple.com
  • 88 edits
    2 copies
    4 moves
    43 adds
    3 deletes in trunk/LayoutTests

Refresh WPT webrtc tests
​https://bugs.webkit.org/show_bug.cgi?id=191133

Reviewed by Eric Carlson.

LayoutTests/imported/w3c:

  • web-platform-tests/webrtc/: Refreshed.
  • web-platform-tests/webrtc/w3c-import.log:

LayoutTests:

7:29 AM Changeset in webkit [237772] by Philippe Normand
  • 2 edits in trunk/Source/WebKit

[WPE] Add the WebKitVersion.h header
​https://bugs.webkit.org/show_bug.cgi?id=191015
<rdar://problem/45647211>

Reviewed by Michael Catanzaro.

  • PlatformWPE.cmake: Install the generated header as well.
7:04 AM Changeset in webkit [237771] by Michael Catanzaro
  • 2 edits in trunk/WebDriverTests

Unreviewed, fix WebDriver expectations syntax after previous commit

6:30 AM Changeset in webkit [237770] by eric.carlson@apple.com
  • 9 edits
    3 adds in trunk

[MediaStream] enumerateDevices should not expose devices that are not available to getUserMedia
​https://bugs.webkit.org/show_bug.cgi?id=191177
<rdar://problem/45747873>

Reviewed by Jer Noble.

Source/WebCore:

Test: http/tests/media/media-stream/enumerate-devices-iframe-allow-attribute.html

  • Modules/mediastream/MediaDevicesRequest.cpp:

(WebCore::MediaDevicesRequest::start): Only expose devices that are available to gUM.

  • Modules/mediastream/UserMediaController.cpp:

(WebCore::isSecure): Moved from UserMediaRequest.cpp.
(WebCore::isAllowedToUse): Ditto.
(WebCore::UserMediaController::canCallGetUserMedia): Modified from UserMediaRequest.cpp.
(WebCore::UserMediaController::logGetUserMediaDenial): Log reason for denial.

  • Modules/mediastream/UserMediaController.h:
  • Modules/mediastream/UserMediaRequest.cpp:

(WebCore::UserMediaRequest::start): Use UserMediaController::canCallGetUserMedia.
(WebCore::isSecure): Deleted.
(WebCore::isAllowedToUse): Deleted.
(WebCore::canCallGetUserMedia): Deleted.

LayoutTests:

  • http/tests/media/media-stream/enumerate-devices-iframe-allow-attribute-expected.txt: Added.
  • http/tests/media/media-stream/enumerate-devices-iframe-allow-attribute.html: Added.
  • http/tests/media/media-stream/resources/enumerate-devices-iframe.html: Added.
  • http/tests/ssl/media-stream/get-user-media-different-host-expected.txt: Rebased for updated logging.
  • http/tests/ssl/media-stream/get-user-media-nested-expected.txt: Ditto.

Nov 2, 2018:

11:25 PM Changeset in webkit [237769] by achristensen@apple.com
  • 8 edits
    1 delete in trunk

Unreviewed, rolling out r237761.

Makes most API tests timeout and WK2 tests crash

Reverted changeset:

"Mac production builds should sign the network process xpc
service with entitlements"
​https://bugs.webkit.org/show_bug.cgi?id=191209
​https://trac.webkit.org/changeset/237761

10:53 PM Changeset in webkit [237768] by Michael Catanzaro
  • 2 edits in trunk/WebDriverTests

Unreviewed WebDriver expectations gardening

There are still a few more flaky tests, but this should improve things significantly.

10:27 PM Changeset in webkit [237767] by Michael Catanzaro
  • 2 edits in trunk/Tools

Unreviewed API test gardening

  • TestWebKitAPI/glib/TestExpectations.json:
9:01 PM Changeset in webkit [237766] by Justin Michaud
  • 30 edits
    5 copies
    3 moves
    15 adds
    2 deletes in trunk

Add new global object and preliminary Worklets support for CSS painting api
​https://bugs.webkit.org/show_bug.cgi?id=190979

Reviewed by Chris Dumez.

Source/WebCore:

Test: fast/css-custom-paint/worklet.html

Add a new ScriptExecutionContext and global object to run worklets in. This is mostly copy+paste
from Workers, but without any of the threading. Worklet.addModule does not yet support loading scripts
or doing cross origin checking. There are quite a few parts of the ScriptExecutionContext api that are
left as ASSERT_NOT_REACHED().

  • CMakeLists.txt:
  • DerivedSources.make:
  • Sources.txt:
  • WebCore.xcodeproj/project.pbxproj:
  • bindings/js/JSDOMGlobalObject.cpp:

(WebCore::JSDOMGlobalObject::scriptExecutionContext const):

  • bindings/js/JSPaintWorkletGlobalScopeCustom.cpp: Renamed from Source/WebCore/bindings/js/JSCSSPaintWorkletGlobalScopeCustom.cpp.

(WebCore::JSPaintWorkletGlobalScope::visitAdditionalChildren):
(WebCore::JSPaintWorkletGlobalScope::registerPaint):

  • bindings/js/JSWorkletGlobalScopeBase.cpp: Added.

(WebCore::JSWorkletGlobalScopeBase::JSWorkletGlobalScopeBase):
(WebCore::JSWorkletGlobalScopeBase::finishCreation):
(WebCore::JSWorkletGlobalScopeBase::clearDOMGuardedObjects):
(WebCore::JSWorkletGlobalScopeBase::visitChildren):
(WebCore::JSWorkletGlobalScopeBase::destroy):
(WebCore::JSWorkletGlobalScopeBase::scriptExecutionContext const):
(WebCore::JSWorkletGlobalScopeBase::supportsRichSourceInfo):
(WebCore::JSWorkletGlobalScopeBase::shouldInterruptScript):
(WebCore::JSWorkletGlobalScopeBase::shouldInterruptScriptBeforeTimeout):
(WebCore::JSWorkletGlobalScopeBase::javaScriptRuntimeFlags):
(WebCore::toJS):
(WebCore::toJSWorkletGlobalScope):

  • bindings/js/JSWorkletGlobalScopeBase.h: Added.

(WebCore::JSWorkletGlobalScopeBase::wrapped const):
(WebCore::JSWorkletGlobalScopeBase::proxy const):
(WebCore::JSWorkletGlobalScopeBase::createStructure):
(WebCore::toJS):

  • bindings/js/ScriptState.cpp:

(WebCore::execStateFromWorkerGlobalScope):
(WebCore::execStateFromWorkletGlobalScope):

  • bindings/js/ScriptState.h:
  • bindings/js/WebCoreBuiltinNames.h:
  • bindings/scripts/CodeGeneratorJS.pm:

(IsDOMGlobalObject):
(ShouldUseGlobalObjectPrototype):
(GenerateHeader):
(GenerateRuntimeEnableConditionalStringForExposed):
(GenerateImplementation):

  • bindings/scripts/preprocess-idls.pl:
  • css/CSSPaintImageValue.cpp:

(WebCore::CSSPaintImageValue::image):

  • css/DOMCSSPaintWorklet.cpp:

(WebCore::DOMCSSPaintWorklet::ensurePaintWorklet):
(WebCore::DOMCSSPaintWorklet::ensurePaintWorkletGlobalScope): Deleted.

  • css/DOMCSSPaintWorklet.h:
  • css/DOMCSSPaintWorklet.idl:
  • dom/Document.cpp:

(WebCore::Document::ensureCSSPaintWorklet):
(WebCore::Document::setCSSPaintWorkletGlobalScope):
(WebCore::Document::ensureCSSPaintWorkletGlobalScope): Deleted.

  • dom/Document.h:

(WebCore::Document::getCSSPaintWorkletGlobalScope):

  • dom/EventTargetFactory.in:
  • dom/ScriptExecutionContext.cpp:

(WebCore::ScriptExecutionContext::vm):
(WebCore::ScriptExecutionContext::execState):

  • dom/ScriptExecutionContext.h:

(WebCore::ScriptExecutionContext::isWorkletGlobalScope const):

  • inspector/agents/worker/WorkerDebuggerAgent.cpp:

(WebCore::WorkerDebuggerAgent::injectedScriptForEval):

  • inspector/agents/worker/WorkerRuntimeAgent.cpp:

(WebCore::WorkerRuntimeAgent::injectedScriptForEval):

  • platform/graphics/CustomPaintImage.cpp:

(WebCore::CustomPaintImage::CustomPaintImage):
(WebCore::CustomPaintImage::doCustomPaint):

  • platform/graphics/CustomPaintImage.h:
  • worklets/PaintWorkletGlobalScope.cpp: Copied from Source/WebCore/css/CSSPaintWorkletGlobalScope.cpp.

(WebCore::PaintWorkletGlobalScope::create):
(WebCore::PaintWorkletGlobalScope::PaintWorkletGlobalScope):
(WebCore::PaintWorkletGlobalScope::devicePixelRatio):
(WebCore::PaintWorkletGlobalScope::addRegisteredPaint):

  • worklets/PaintWorkletGlobalScope.h: Renamed from Source/WebCore/css/CSSPaintWorkletGlobalScope.h.
  • worklets/PaintWorkletGlobalScope.idl: Copied from Source/WebCore/css/CSSPaintWorkletGlobalScope.idl.
  • worklets/Worklet.cpp: Copied from Source/WebCore/css/CSSPaintWorkletGlobalScope.cpp.

(WebCore::Worklet::create):
(WebCore::Worklet::Worklet):
(WebCore::Worklet::addModule):

  • worklets/Worklet.h: Renamed from Source/WebCore/css/CSSPaintWorkletGlobalScope.cpp.
  • worklets/Worklet.idl: Copied from Source/WebCore/css/CSSPaintWorkletGlobalScope.idl.
  • worklets/WorkletConsoleClient.cpp: Copied from Source/WebCore/inspector/agents/worker/WorkerDebuggerAgent.cpp.

(WebCore::WorkletConsoleClient::WorkletConsoleClient):
(WebCore::WorkletConsoleClient::messageWithTypeAndLevel):
(WebCore::WorkletConsoleClient::count):
(WebCore::WorkletConsoleClient::time):
(WebCore::WorkletConsoleClient::timeEnd):
(WebCore::WorkletConsoleClient::profile):
(WebCore::WorkletConsoleClient::profileEnd):
(WebCore::WorkletConsoleClient::takeHeapSnapshot):
(WebCore::WorkletConsoleClient::timeStamp):
(WebCore::WorkletConsoleClient::record):
(WebCore::WorkletConsoleClient::recordEnd):

  • worklets/WorkletConsoleClient.h: Added.
  • worklets/WorkletGlobalScope.cpp: Added.

(WebCore::WorkletGlobalScope::WorkletGlobalScope):
(WebCore::WorkletGlobalScope::~WorkletGlobalScope):
(WebCore::WorkletGlobalScope::evaluate):
(WebCore::WorkletGlobalScope::isJSExecutionForbidden const):
(WebCore::WorkletGlobalScope::disableEval):
(WebCore::WorkletGlobalScope::disableWebAssembly):
(WebCore::WorkletGlobalScope::completeURL const):
(WebCore::WorkletGlobalScope::logExceptionToConsole):
(WebCore::WorkletGlobalScope::addConsoleMessage):
(WebCore::WorkletGlobalScope::addMessage):

  • worklets/WorkletGlobalScope.h: Added.

(WebCore::WorkletGlobalScope::isPaintWorkletGlobalScope const):
(WebCore::WorkletGlobalScope::identifier const):
(WebCore::WorkletGlobalScope::script):
(WebCore::WorkletGlobalScope::jsRuntimeFlags const):
(isType):

  • worklets/WorkletGlobalScope.idl: Renamed from Source/WebCore/css/CSSPaintWorkletGlobalScope.idl.
  • worklets/WorkletScriptController.cpp: Added.

(WebCore::WorkletScriptController::WorkletScriptController):
(WebCore::WorkletScriptController::~WorkletScriptController):
(WebCore::WorkletScriptController::forbidExecution):
(WebCore::WorkletScriptController::isExecutionForbidden const):
(WebCore::WorkletScriptController::disableEval):
(WebCore::WorkletScriptController::disableWebAssembly):
(WebCore::WorkletScriptController::initScriptWithSubclass):
(WebCore::WorkletScriptController::initScript):
(WebCore::WorkletScriptController::evaluate):
(WebCore::WorkletScriptController::setException):

  • worklets/WorkletScriptController.h: Added.

(WebCore::WorkletScriptController::workletGlobalScopeWrapper):
(WebCore::WorkletScriptController::vm):
(WebCore::WorkletScriptController::initScriptIfNeeded):

Tools:

  • Scripts/webkitpy/bindings/main.py:

(BindingsTests.generate_supplemental_dependency):
(BindingsTests.main):

LayoutTests:

Add a new test harness for worklet tests, since they do not have access to the dom.

  • fast/css-custom-paint/basic-expected.txt:
  • fast/css-custom-paint/basic.html:
  • fast/css-custom-paint/registerPaintBindings-expected.html: Added.
  • fast/css-custom-paint/registerPaintBindings-expected.txt: Removed.
  • fast/css-custom-paint/registerPaintBindings.html:
  • fast/css-custom-paint/resources/testharness-worklet.js: Added.

(format_value):

  • fast/css-custom-paint/resources/testharness.js: Added.

(async.importWorklet):

  • fast/css-custom-paint/worklet-expected.html: Added.
  • fast/css-custom-paint/worklet.html: Added.
7:36 PM Changeset in webkit [237765] by mmaxfield@apple.com
  • 27 edits in trunk/Source/WebCore

Clean up drawLineForDocumentMarker()
​https://bugs.webkit.org/show_bug.cgi?id=191215

Reviewed by Zalan Bujtas.

In preparation for ​https://bugs.webkit.org/show_bug.cgi?id=190764, I need to do a little bit of refactoring.
This patch has no behavior change; it just does the following:

  1. Renames drawLineForDocumentMarker() to drawDotsForDocumentMarker(), because 2 of the 3 implementations draw dots
  2. Moves our implementation back into GraphicsContext, because it's simpler and GraphicsContext is already platform-

specific.

  1. The signature used to accept a location and a width, but without a height, it's difficult to know what the bounding

box is. In particular, knowing the bounding box must be possible without a GraphicsContext. So, I've modified
the signature to accept a rectangle instead. The GraphicsContext draws only within this rectangle.

No new tests because there is no behavior change.

  • platform/graphics/GraphicsContext.h:
  • platform/graphics/GraphicsContextImpl.h:
  • platform/graphics/cairo/CairoOperations.cpp:

(WebCore::Cairo::drawDotsForDocumentMarker):
(WebCore::Cairo::drawLineForDocumentMarker): Deleted.

  • platform/graphics/cairo/CairoOperations.h:
  • platform/graphics/cairo/GraphicsContextCairo.cpp:

(WebCore::GraphicsContext::drawDotsForDocumentMarker):
(WebCore::GraphicsContext::drawLineForDocumentMarker): Deleted.

  • platform/graphics/cairo/GraphicsContextImplCairo.cpp:

(WebCore::GraphicsContextImplCairo::drawDotsForDocumentMarker):
(WebCore::GraphicsContextImplCairo::drawLineForDocumentMarker): Deleted.

  • platform/graphics/cairo/GraphicsContextImplCairo.h:
  • platform/graphics/cocoa/FontCascadeCocoa.mm:
  • platform/graphics/cocoa/GraphicsContextCocoa.mm:

(WebCore::colorForMarkerLineStyle):
(WebCore::GraphicsContext::drawDotsForDocumentMarker):
(WebCore::GraphicsContext::drawLineForDocumentMarker): Deleted.

  • platform/graphics/displaylists/DisplayListItems.cpp:

(WebCore::DisplayList::Item::sizeInBytes):
(WebCore::DisplayList::DrawDotsForDocumentMarker::apply const):
(WebCore::DisplayList::DrawDotsForDocumentMarker::localBounds const):
(WebCore::DisplayList::operator<<):
(WebCore::DisplayList::DrawLineForDocumentMarker::apply const): Deleted.
(WebCore::DisplayList::DrawLineForDocumentMarker::localBounds const): Deleted.

  • platform/graphics/displaylists/DisplayListItems.h:

(WebCore::DisplayList::DrawDotsForDocumentMarker::create):
(WebCore::DisplayList::DrawDotsForDocumentMarker::rect const):
(WebCore::DisplayList::DrawDotsForDocumentMarker::DrawDotsForDocumentMarker):
(WebCore::DisplayList::DrawLineForDocumentMarker::create): Deleted.
(WebCore::DisplayList::DrawLineForDocumentMarker::point const): Deleted.
(WebCore::DisplayList::DrawLineForDocumentMarker::width const): Deleted.
(WebCore::DisplayList::DrawLineForDocumentMarker::DrawLineForDocumentMarker): Deleted.

  • platform/graphics/displaylists/DisplayListRecorder.cpp:

(WebCore::DisplayList::Recorder::drawDotsForDocumentMarker):
(WebCore::DisplayList::Recorder::drawLineForDocumentMarker): Deleted.

  • platform/graphics/displaylists/DisplayListRecorder.h:
  • platform/graphics/nicosia/cairo/NicosiaCairoOperationRecorder.cpp:

(Nicosia::CairoOperationRecorder::drawDotsForDocumentMarker):
(Nicosia::CairoOperationRecorder::drawLineForDocumentMarker): Deleted.

  • platform/graphics/nicosia/cairo/NicosiaCairoOperationRecorder.h:
  • platform/graphics/win/GraphicsContextCGWin.cpp:

(WebCore::GraphicsContext::drawDotsForDocumentMarker):
(WebCore::GraphicsContext::drawLineForDocumentMarker): Deleted.

  • platform/graphics/win/GraphicsContextDirect2D.cpp:

(WebCore::GraphicsContext::drawDotsForDocumentMarker):
(WebCore::GraphicsContext::drawLineForDocumentMarker): Deleted.

  • rendering/InlineTextBox.cpp:

(WebCore::InlineTextBox::paintPlatformDocumentMarker):

  • rendering/RenderTheme.cpp:

(WebCore::RenderTheme::drawLineForDocumentMarker): Deleted.

  • rendering/RenderTheme.h:
  • rendering/RenderThemeCocoa.h:
  • rendering/RenderThemeCocoa.mm:

(WebCore::RenderThemeCocoa::drawLineForDocumentMarker): Deleted.

  • rendering/RenderThemeIOS.h:
  • rendering/RenderThemeIOS.mm:

(WebCore::RenderThemeIOS::colorForMarkerLineStyle): Deleted.

  • rendering/RenderThemeMac.h:
  • rendering/RenderThemeMac.mm:

(WebCore::RenderThemeMac::colorForMarkerLineStyle): Deleted.

6:52 PM Changeset in webkit [237764] by keith_miller@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

Assert JSValues can fit into a pointer when API casting
​https://bugs.webkit.org/show_bug.cgi?id=191220

Reviewed by Michael Saboff.

  • API/APICast.h:

(toJS):
(toJSForGC):
(toRef):

6:27 PM Changeset in webkit [237763] by msaboff@apple.com
  • 7 edits
    1 add in trunk

Rolling in r237753 with unreviewed build fix.

Fixed issues with DECLARE_THROW_SCOPE placement.

5:49 PM Changeset in webkit [237762] by Ryan Haddad
  • 2 edits in trunk/Source/WebKit

Unreviewed, attempt to fix the build with recent SDKs.

  • Shared/EntryPointUtilities/mac/XPCService/XPCServiceEntryPoint.mm:

(WebKit::XPCServiceExit):

5:33 PM Changeset in webkit [237761] by achristensen@apple.com
  • 8 edits
    1 add in trunk

Mac production builds should sign the network process xpc service with entitlements
​https://bugs.webkit.org/show_bug.cgi?id=191209
<rdar://problem/45377609>

Reviewed by Geoff Garen.

Source/WebKit:

  • Configurations/NetworkService.xcconfig:

Set CODE_SIGN_ENTITLEMENTS_COCOA_TOUCH_NO like we do for the web process,
except with the corresponoding network process entitlements file.

  • UIProcess/API/Cocoa/WKProcessPool.mm:

(-[WKProcessPool _networkProcessHasEntitlementForTesting:]):

  • UIProcess/API/Cocoa/WKProcessPoolPrivate.h:
  • UIProcess/Cocoa/WebProcessPoolCocoa.mm:

(WebKit::WebProcessPool::networkProcessHasEntitlementForTesting):

  • UIProcess/WebProcessPool.h:

Add infrastructure to test that the network process actually has an entitlement.

Tools:

  • TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
  • TestWebKitAPI/Tests/WebKitCocoa/NetworkProcess.mm: Added.

(TEST):

5:06 PM Changeset in webkit [237760] by Chris Dumez
  • 3 edits in trunk/LayoutTests

REGRESSION (r237699): Layout Test http/tests/security/anchor-download-block-crossorigin.html is a flaky failure
​https://bugs.webkit.org/show_bug.cgi?id=191199

Reviewed by Dewei Zhu.

Navigate the subframe instead of the main frame, to make sure that the console message is consistently logged.

  • http/tests/security/anchor-download-block-crossorigin-expected.txt:
  • http/tests/security/anchor-download-block-crossorigin.html:
4:33 PM Changeset in webkit [237759] by Kocsen Chung
  • 24 edits in tags/Safari-607.1.12.1

Apply patch. rdar://problem/45749094
​https://bugs.webkit.org/show_bug.cgi?id=191201

4:20 PM Changeset in webkit [237758] by Jon Davis
  • 2 edits in trunk/Websites/webkit.org

Follow-up fix for the webkit.org blog index page
​https://bugs.webkit.org/show_bug.cgi?id=191216

Reviewed by Timothy Hatcher.

  • wp-content/themes/webkit/index.php:
3:47 PM Changeset in webkit [237757] by Ryan Haddad
  • 7 edits
    1 delete in trunk

Unreviewed, rolling out r237753.

Introduced JSC test failures

Reverted changeset:

"Running out of stack space not properly handled in
RegExp::compile() and its callers"
​https://bugs.webkit.org/show_bug.cgi?id=191206
​https://trac.webkit.org/changeset/237753

3:46 PM Changeset in webkit [237756] by pvollan@apple.com
  • 2 edits in trunk/LayoutTests

[Win] Layout Test fast/block/basic/inline-content-with-floating-image.html is failing
​https://bugs.webkit.org/show_bug.cgi?id=191194

Unreviewed test gardening.

  • platform/win/TestExpectations:
3:35 PM Changeset in webkit [237755] by Lucas Forschler
  • 2 edits in trunk/Tools

Add Windows 10 archive bots
​https://bugs.webkit.org/show_bug.cgi?id=190964

This Change updates the Apple Windows Buildbot configuration.

  • Add Windows 10 specific builders and testers for release and debug.
  • Refactor Windows -> Windows 7.
  • Remove Windows 7 Debug.

Reviewed by Aakash Jain.

  • BuildSlaveSupport/build.webkit.org-config/config.json:
3:07 PM Changeset in webkit [237754] by ajuma@chromium.org
  • 3 edits in trunk/Source/WebCore

requestAnimationFrame causes bad location of position:fixed inside overflow:auto and iframe
​https://bugs.webkit.org/show_bug.cgi?id=176243

Reviewed by Simon Fraser.

When a new layer tree is committed to the UIProcess, the positions of layers for fixed
or sticky nodes in the newly-committed tree can be stale, because of scrolling that has
happened in the UIProcess since the tree was updated in the WebProcess. To handle this,
RemoteLayerTreeDrawingAreaProxy::commitLayerTree updates the positions of these layers
by calling RemoteScrollingCoordinatorProxy::viewportChangedViaDelegatedScrolling, which
leads to a recursive traversal of the ScrollingTree to update each such layer. However,
since ScrollingTreeFrameScrollingNodeIOS didn't implement updateLayersAfterAncestorChange,
this traversal never descended into scrolling nodes within an iframe, so the layers for
these nodes were left with stale positions.

Implement ScrollingTreeFrameScrollingNodeIOS::updateLayersAfterAncestorChange so that
fixed and sticky layers within an iframe do get their positions updated when a new layer
tree is committed.

  • page/scrolling/ios/ScrollingTreeFrameScrollingNodeIOS.h:
  • page/scrolling/ios/ScrollingTreeFrameScrollingNodeIOS.mm:

(WebCore::ScrollingTreeFrameScrollingNodeIOS::updateLayersAfterAncestorChange):

3:05 PM Changeset in webkit [237753] by msaboff@apple.com
  • 7 edits
    1 add in trunk

Running out of stack space not properly handled in RegExp::compile() and its callers
​https://bugs.webkit.org/show_bug.cgi?id=191206

Reviewed by Filip Pizlo.

JSTests:

New regression test.

  • stress/regexp-compile-oom.js: Added.

(recurseAndTest):

Source/JavaScriptCore:

Eliminated two RELEASE_ASSERT_NOT_REACHED() for errors returned by Yarr parsing code. Bubbled those errors
up to where they are turned into the appropriate exceptions in matchInline(). If the errors are not due
to syntax, we reset the RegExp state in case the parsing is tried with a smaller stack.

  • runtime/RegExp.cpp:

(JSC::RegExp::compile):
(JSC::RegExp::compileMatchOnly):

  • runtime/RegExp.h:
  • runtime/RegExpInlines.h:

(JSC::RegExp::compileIfNecessary):
(JSC::RegExp::matchInline):
(JSC::RegExp::compileIfNecessaryMatchOnly):

  • runtime/RegExpObjectInlines.h:

(JSC::RegExpObject::execInline):

  • yarr/YarrErrorCode.h:

(JSC::Yarr::hasHardError):

2:43 PM Changeset in webkit [237752] by Jon Davis
  • 9 edits in trunk/Websites/webkit.org

Improved the SVG icons to use SVG symbols
​https://bugs.webkit.org/show_bug.cgi?id=191205

Reviewed by Timothy Hatcher.

This change makes it far easier to add new SVG icons to the
icons.svg sprite and provides better positioning control
over the icons within the blog post tile frame.

  • wp-content/themes/webkit/front-page.php:
  • wp-content/themes/webkit/functions.php:
  • wp-content/themes/webkit/header.php:
  • wp-content/themes/webkit/images/icons.svg:
  • wp-content/themes/webkit/index.php:
  • wp-content/themes/webkit/loop.php:
  • wp-content/themes/webkit/style.css:

(@supports ( -webkit-backdrop-filter: blur(10px) )):
(footer):
(#icons):
(.tile .background-image):
(.tile .background-image svg):
(.tile.category-web-inspector svg):
(.tile.category-performance svg):
(.tile.category-javascript svg):
(.tile.category-css svg):
(.tile.category-standards svg):
(.tile.category-contributing svg):
(.tile.category-storage svg):
(.tile.category-layout svg):
(.tile.category-safari-technology-preview svg):
(.tile.category-accessibility svg):
(.tile.category-security svg):
(.tile.category-privacy svg):
(header .menu-item-has-children .label-toggle::after):
(@media only screen and (max-width: 920px)):
(.main-menu.label-toggle):
(header .menu):
(@media only screen and (max-width: 690px)):
(@media only screen and (max-width: 415px)):
(@media only screen and (max-height: 415px)):
(.tile.category-web-inspector .background-image): Deleted.
(.tile.category-performance .background-image): Deleted.
(.tile.category-javascript .background-image): Deleted.
(.tile.category-css .background-image): Deleted.
(.tile.category-standards .background-image): Deleted.
(.tile.category-contributing .background-image): Deleted.
(.tile.category-storage .background-image): Deleted.
(.tile.category-layout .background-image): Deleted.
(.tile.category-safari-technology-preview .background-image): Deleted.
(.tile.category-accessibility .background-image): Deleted.
(.tile.tag-console .background-image,): Deleted.
(.tile.tag-console .background-image): Deleted.
(.tile.tag-debugger .background-image): Deleted.
(.tile.tag-shortcuts .background-image): Deleted.
(.tile.tag-timeline .background-image): Deleted.

  • wp-content/themes/webkit/widgets/post.php:
1:38 PM Changeset in webkit [237751] by bshafiei@apple.com
  • 2 edits in tags/Safari-607.1.12.1/Source/JavaScriptCore

Cherry-pick r237745. rdar://problem/45765887

API should use wrapper object if address is 32-bit
​https://bugs.webkit.org/show_bug.cgi?id=191203

Reviewed by Filip Pizlo.

  • API/APICast.h: (toJS): (toJSForGC): (toRef):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@237745 268f45cc-cd09-0410-ab3c-d52691b4dbfc

1:33 PM Changeset in webkit [237750] by bshafiei@apple.com
  • 125 edits
    4 adds
    29 deletes in tags/Safari-607.1.12.1

Revert r237547. rdar://problem/45767917

1:33 PM Changeset in webkit [237749] by bshafiei@apple.com
  • 3 edits
    1 delete in tags/Safari-607.1.12.1

Revert r237553. rdar://problem/45768563

1:33 PM Changeset in webkit [237748] by bshafiei@apple.com
  • 2 edits in tags/Safari-607.1.12.1/Source/JavaScriptCore

Revert r237576. rdar://problem/45767917

1:33 PM Changeset in webkit [237747] by bshafiei@apple.com
  • 6 edits in tags/Safari-607.1.12.1/Source/JavaScriptCore

Revert r237627. rdar://problem/45771226

12:39 PM Changeset in webkit [237746] by Matt Baker
  • 4 edits in trunk/Source/WebInspectorUI

Web Inspector: support multiple selection/deletion of cookie records
​https://bugs.webkit.org/show_bug.cgi?id=66381
<rdar://problem/19281525>

Reviewed by Devin Rousso.

  • Localizations/en.lproj/localizedStrings.js:
  • UserInterface/Views/CookieStorageContentView.js:

(WI.CookieStorageContentView):
(WI.CookieStorageContentView.prototype.get scrollableElements):
(WI.CookieStorageContentView.prototype.tableNumberOfRows):
(WI.CookieStorageContentView.prototype.tableSortChanged):
(WI.CookieStorageContentView.prototype.tableCellContextMenuClicked):
(WI.CookieStorageContentView.prototype.tableDidRemoveRows):
(WI.CookieStorageContentView.prototype.tablePopulateCell):
(WI.CookieStorageContentView.prototype.initialLayout):
(WI.CookieStorageContentView.prototype._generateSortComparator):
(WI.CookieStorageContentView.prototype._refreshButtonClicked):
(WI.CookieStorageContentView.prototype._reloadCookies):
(WI.CookieStorageContentView.prototype._updateSort):
(WI.CookieStorageContentView.prototype._handleTableKeyDown):
(WI.CookieStorageContentView.prototype.update): Deleted.
(WI.CookieStorageContentView.prototype._rebuildTable): Deleted.
(WI.CookieStorageContentView.prototype._sortDataGrid.localeCompare): Deleted.
(WI.CookieStorageContentView.prototype._sortDataGrid.numberCompare): Deleted.
(WI.CookieStorageContentView.prototype._sortDataGrid.expiresCompare): Deleted.
(WI.CookieStorageContentView.prototype._sortDataGrid): Deleted.
(WI.CookieStorageContentView.prototype._deleteCallback): Deleted.
Replace DataGrid with Table. The content view serves as the table
delegate and data source, and handles delete and backspace key events
to allow deleting the selected cookies. Cookies may also be deleted from
the table context menu and a new button in the navigation bar.

  • UserInterface/Views/Table.js:

(WI.Table.prototype.isRowSelected):
(WI.Table.prototype.selectRow):
(WI.Table.prototype.deselectRow):
(WI.Table.prototype.removeRow):
(WI.Table.prototype._getOrCreateRow):
(WI.Table.prototype._handleMouseDown):
(WI.Table.prototype._removeRows):
(WI.Table.prototype._isRowSelected): Deleted.
Make isSelectedRow public. It is more convenient and efficient than
calling Table.prototype.selectedRows.includes(rowIndex).

12:03 PM Changeset in webkit [237745] by keith_miller@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

API should use wrapper object if address is 32-bit
​https://bugs.webkit.org/show_bug.cgi?id=191203

Reviewed by Filip Pizlo.

  • API/APICast.h:

(toJS):
(toJSForGC):
(toRef):

11:31 AM Changeset in webkit [237744] by Alan Coon
  • 11 edits
    1 delete in tags/Safari-607.1.12.1/Source

Cherry-pick r237721. rdar://problem/45722391

Replace CommonRandom SPI with API
​https://bugs.webkit.org/show_bug.cgi?id=191178
<rdar://problem/45722391>

Reviewed by Brent Fulgham.

Source/bmalloc:

  • bmalloc/CryptoRandom.cpp: (bmalloc::ARC4RandomNumberGenerator::stir):

Source/WebCore:

The API is available since macOS 10.10 and iOS 10, and therefore it is safe to replace
all SPI usages with API.

No change of behaviors.

  • crypto/CommonCryptoUtilities.h:
  • crypto/mac/CryptoKeyMac.cpp: (WebCore::CryptoKey::randomData):
  • crypto/mac/SerializedCryptoKeyWrapMac.mm: (WebCore::createAndStoreMasterKey): (WebCore::wrapSerializedCryptoKey):
  • page/Crypto.cpp: (WebCore::Crypto::getRandomValues):

Source/WTF:

  • WTF.xcodeproj/project.pbxproj:
  • wtf/CMakeLists.txt:
  • wtf/RandomDevice.cpp: (WTF::RandomDevice::cryptographicallyRandomValues):
  • wtf/spi/darwin/CommonCryptoSPI.h: Removed.

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@237721 268f45cc-cd09-0410-ab3c-d52691b4dbfc

11:10 AM Changeset in webkit [237743] by Wenson Hsieh
  • 7 edits
    2 adds in trunk

[iOS] Changing view scale sometimes does not zoom the page to the new initial scale, when the page is at initial scale
​https://bugs.webkit.org/show_bug.cgi?id=191180
<rdar://problem/45744786>

Reviewed by Simon Fraser.

Source/WebCore:

When computing the minimum scale in ViewportConfiguration::minimumScale, if our content width or height is
shorter than the view width or height, then we recompute the minimum scale such that the content dimensions will
fill the bounds of the view by setting the minimum scale to the view width or height divided by the content
width or height.

Suppose the minimum scale is equal to some value s; additionally, let w_c denote the content width and w_v
denote the view width (as integers). If w_v / s is not an integral value, the computed content width w_c may
be rounded, such that w_v / w_c is not precisely equal to s. In the case that w_v / w_c is ever so
slightly larger than s, we will end up overriding the minimum scale s with w_v / w_c.

As a result, specifying a viewport with a decimal minimum-scale will sometimes cause the computed minimum
scale of the viewport (and platform view) to be very slightly different from the minimum scale. The new layout
test below exercises this scenario, specifying a viewport with minimum and initial scales of 0.94 that results
in ViewportConfiguration::minimumScale returning 0.94158.

With the WebPage::setViewportConfigurationViewLayoutSize check added in r237127, this means setting
_viewScale: when the page is at initial scale sometimes doesn't zoom the page to the new initial scale when it
should, since the page scale factor and the initial scale are different enough such that
areEssentiallyEqualAsFloat returns false.

This patch addresses these issues by snapping to the minimum scale if the computed scale that fits content
dimensions to view dimensions results in a minimum scale that is close enough to the configuration's minimum
scale, such that the difference can be attributed to rounding error when computing content or view dimensions.

Test: fast/viewport/ios/viewport-minimum-and-initial-scale.html

  • page/ViewportConfiguration.cpp:

(WebCore::ViewportConfiguration::minimumScale const):

LayoutTests:

Add a layout test, and make some adjustments to UIHelper.

  • fast/viewport/ios/constant-width-viewport-after-changing-view-scale.html:
  • fast/viewport/ios/device-width-viewport-after-changing-view-scale.html:
  • fast/viewport/ios/viewport-minimum-and-initial-scale-expected.txt: Added.
  • fast/viewport/ios/viewport-minimum-and-initial-scale.html: Added.

Add a new layout test that contains a viewport meta tag with minimum and initial scales set to 0.94, and checks
that the resulting minimum and initial scales are 0.94 instead of 0.94158.

  • fast/viewport/watchos/viewport-disable-extra-zoom-adaptations.html:
  • resources/ui-helper.js:

Make UIHelper.zoomScale return a number rather than a string, and adjust a few call sites.

(window.UIHelper.zoomScale):
(window.UIHelper.minimumZoomScale):
(window.UIHelper):

10:45 AM Changeset in webkit [237742] by Philippe Normand
  • 2 edits in trunk/Source/WebCore

[GTK][WPE] Unreviewed, another --no-video --no-web-audio build fix following r237677

  • inspector/agents/InspectorDOMAgent.cpp:

(WebCore::InspectorDOMAgent::addEventListenersToNode):

10:42 AM Changeset in webkit [237741] by Philippe Normand
  • 3 edits in trunk/Tools

[GTK][WPE] Enable GST_DEBUG_NO_COLOR environment variable pass-through

Rubber-stamped by Michael Catanzaro.

This variable is useful to get rid of shell color escapes in tests stderr files.

  • Scripts/webkitpy/port/gtk.py:

(GtkPort.setup_environ_for_server):

  • Scripts/webkitpy/port/wpe.py:

(WPEPort.setup_environ_for_server):

10:35 AM Changeset in webkit [237740] by aakash_jain@apple.com
  • 1 edit
    20 adds in trunk/Tools

Add Basic Django web app for OpenSource EWS
​https://bugs.webkit.org/show_bug.cgi?id=191091

Reviewed by Lucas Forschler.

  • BuildSlaveSupport/EWS_app/manage.py: Added, auto-generated by Django.
  • BuildSlaveSupport/EWS_app/EWS_app/init.py: Added, auto-generated by Django.
  • BuildSlaveSupport/EWS_app/EWS_app/settings.py: Added, auto-generated by Django.
  • BuildSlaveSupport/EWS_app/EWS_app/urls.py: Added, auto-generated by Django.
  • BuildSlaveSupport/EWS_app/EWS_app/wsgi.py: Added, auto-generated by Django.
  • BuildSlaveSupport/EWS_app/ews/init.py: Added.
  • BuildSlaveSupport/EWS_app/ews/admin.py: Added, auto-generated by Django.
  • BuildSlaveSupport/EWS_app/ews/apps.py: Added, auto-generated by Django.
  • BuildSlaveSupport/EWS_app/ews/migrations/init.py: Added.
  • BuildSlaveSupport/EWS_app/ews/tests.py: Added, auto-generated by Django.
  • BuildSlaveSupport/EWS_app/ews/models/init.py: Added.
  • BuildSlaveSupport/EWS_app/ews/models/patch.py: Added.
  • BuildSlaveSupport/EWS_app/ews/views/init.py: Added.
  • BuildSlaveSupport/EWS_app/ews/views/patch.py: Added.
10:20 AM Changeset in webkit [237739] by dbates@webkit.org
  • 4 edits in trunk/Source/WebCore

[iOS] Normalize character string based on key code
​https://bugs.webkit.org/show_bug.cgi?id=191120

Reviewed by Tim Horton.

Source/WebCore:

Following r236417 (​https://bugs.webkit.org/show_bug.cgi?id=189604) we always have a valid key code
for a special key (e.g. page up) and can simplify the character string normalization code.

No functionality changed. So, no new tests.

  • platform/ios/WebEvent.mm:

(normalizedStringWithAppKitCompatibilityMapping): Modified to take a key code and wrote algorithm
in terms of it.

(-[WebEvent initWithKeyEventType:timeStamp:characters:charactersIgnoringModifiers:modifiers:isRepeating:withFlags:keyCode:isTabKey:characterSet:]):
(-[WebEvent initWithKeyEventType:timeStamp:characters:charactersIgnoringModifiers:modifiers:isRepeating:withFlags:withInputManagerHint:keyCode:isTabKey:]):
Pass the key code for the event to normalizedStringWithAppKitCompatibilityMapping().

Source/WebCore/PAL:

Remove forward declarations that are no longer needed now that we check key code.

  • pal/spi/ios/UIKitSPI.h:
10:17 AM Changeset in webkit [237738] by dbates@webkit.org
  • 16 edits in trunk

[iOS] WebKit should dispatch DOM events when a modifier key is pressed
​https://bugs.webkit.org/show_bug.cgi?id=190487

Reviewed by Tim Horton.

Source/WebCore:

Add support for modifier flags change events.

  • platform/ios/PlatformEventFactoryIOS.mm:

(WebCore::modifiersForEvent): Modifier keys do not have an associated character and do not
participate in key repeat.
(WebCore::keyIdentifierForKeyEvent): Identify modifier keys, returning "Unidentified" if
the modifier key is unidentified. This matches the behavior on Mac.
(WebCore::keyForKeyEvent): Identify modifier keys, returning the empty string if the modifier
key is unidentified. This matches the behavior on Mac.
(WebCore::PlatformKeyboardEventBuilder::PlatformKeyboardEventBuilder):

  • platform/ios/WebEvent.h:
  • platform/ios/WebEvent.mm:

(-[WebEvent initWithKeyEventType:timeStamp:characters:charactersIgnoringModifiers:modifiers:isRepeating:withFlags:keyCode:isTabKey:characterSet:]):
(-[WebEvent initWithKeyEventType:timeStamp:characters:charactersIgnoringModifiers:modifiers:isRepeating:withFlags:withInputManagerHint:keyCode:isTabKey:]):
Modifier keys do not have an associated character and do not participate in key repeat.

(-[WebEvent _eventDescription]): Modified to print a description for a keydown or keyup event
that represents a modifier flags change.
(-[WebEvent characters]): Modifier keys do not have an associated character. Assert this
invariant as it is a programming error. On Mac, the same operation would result in a
NSInternalInconsistencyException exception being raised.
(-[WebEvent charactersIgnoringModifiers]): Ditto.

Source/WebKit:

Add support for modifier flags change events dispatched by UIKit.

  • Shared/ios/WebIOSEventFactory.mm:

(modifiersForEvent): Update for renamed enumerations.
(WebIOSEventFactory::createWebKeyboardEvent): Modifier keys do not have an associated
character and do not participate in key repeat.

  • UIProcess/Automation/ios/WebAutomationSessionIOS.mm:

(WebKit::WebAutomationSession::platformSimulateKeyboardInteraction): Update for renamed enumerations.

  • UIProcess/ios/WKContentViewInteraction.mm:

(-[WKContentView _interpretKeyEvent:isCharEvent:]): Return NO if the event represents a
modifier flags change to allow the default handling.

  • UIProcess/ios/WKKeyboardScrollingAnimator.mm:

(-[WKKeyboardScrollingAnimator keyboardScrollForEvent:]): Update for renamed enumerations.

Source/WebKitLegacy/mac:

Having Legacy WebKit return NO if the editing event represents a modifier flags change to
allow the default handling.

  • WebView/WebHTMLView.mm:

(-[WebHTMLView _handleEditingKeyEvent:]):

Tools:

Update the code for renamed enumerations. For WebKitTestRunner, removed comment that does not
match my observations when using the simulator to simulate iOS 12 and add support for testing
right-hand side modifiers.

  • DumpRenderTree/mac/EventSendingController.mm:

(modifierFlags):
(-[EventSendingController keyDown:withModifiers:withLocation:]):
(-[EventSendingController setTouchModifier:value:]):

  • WebKitTestRunner/ios/HIDEventGenerator.mm:

(hidUsageCodeForCharacter):

LayoutTests:

Update expected results of an existing test now that we identify modifier keys.

  • fast/events/ios/keydown-keyup-special-keys-in-non-editable-element-expected.txt:
9:55 AM Changeset in webkit [237737] by ajuma@chromium.org
  • 17 edits
    2 copies
    7 adds in trunk

Allow cross-document intersection observing
​https://bugs.webkit.org/show_bug.cgi?id=165746

Reviewed by Simon Fraser.

LayoutTests/imported/w3c:

Update expectations for tests that now pass.

  • web-platform-tests/intersection-observer/client-rect-expected.txt:
  • web-platform-tests/intersection-observer/cross-origin-iframe-expected.txt:
  • web-platform-tests/intersection-observer/iframe-no-root-expected.txt:
  • web-platform-tests/intersection-observer/timestamp-expected.txt:

Source/WebCore:

Add logic to compute the intersection between the viewport and an element in a
subframe.

Add a FloatRect version of ScrollView::rootViewToContents, and FloatRect versions
of the methods it calls.

Test: http/tests/intersection-observer/intermediate-frame-changes.html
Also covered by rebased tests in imported/w3c/web-platform-tests/intersection-observer.

  • dom/Document.cpp:

(WebCore::computeClippedRectInRootContentsSpace):
(WebCore::computeIntersectionState):
(WebCore::Document::updateIntersectionObservations):

  • page/FrameView.cpp:

(WebCore::FrameView::viewportContentsChanged):
(WebCore::FrameView::convertFromContainingViewToRenderer const):
(WebCore::FrameView::convertFromContainingView const):

  • page/FrameView.h:
  • platform/ScrollView.cpp:

(WebCore::ScrollView::viewToContents const):
(WebCore::ScrollView::contentsToView const):
(WebCore::ScrollView::rootViewToContents const):

  • platform/ScrollView.h:
  • platform/Widget.cpp:

(WebCore::Widget::convertFromRootView const):
(WebCore::Widget::convertFromContainingView const):

  • platform/Widget.h:

LayoutTests:

Add a test that cross-document intersection observations are triggered when an intermediate frame
scrolls. Skip the test on WK1 platforms where IntersectionObserver is not supported.

Add platform-specific baselines for tests involving iframe scrolling, which isn't supported on iOS.

  • http/tests/intersection-observer/intermediate-frame-changes-expected.txt: Added.
  • http/tests/intersection-observer/intermediate-frame-changes.html: Added.
  • http/tests/intersection-observer/resources/intermediate-frame.html: Added.
  • http/tests/intersection-observer/resources/nested-frame.html: Added.
  • platform/ios/imported/w3c/web-platform-tests/intersection-observer/cross-origin-iframe-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/intersection-observer/cross-origin-iframe-expected.txt.
  • platform/ios/imported/w3c/web-platform-tests/intersection-observer/iframe-no-root-expected.txt: Copied from LayoutTests/imported/w3c/web-platform-tests/intersection-observer/iframe-no-root-expected.txt.
  • platform/mac-wk1/TestExpectations:
  • platform/win/TestExpectations:
  • platform/wincairo/TestExpectations:
9:49 AM Changeset in webkit [237736] by Kocsen Chung
  • 2 edits in tags/Safari-607.1.12.1/Source/JavaScriptCore

Cherry-pick r237707. rdar://problem/45749094

Unreviewed, JavaScriptCore should only guarantee to produce a
modulemap if we are building for iOSMac.

  • Configurations/JavaScriptCore.xcconfig:

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@237707 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8:59 AM Changeset in webkit [237735] by Chris Dumez
  • 6 edits in trunk

[PSON] Reuse SuspendedPages' process when possible, for performance
​https://bugs.webkit.org/show_bug.cgi?id=191166

Reviewed by Geoffrey Garen.

Source/WebKit:

When process-swapping check if there is an existing SuspendedPage for the domain we're going to.
If there is, use this SuspendedPage's process for the navigation instead of a fresh new process.
This change should be beneficial for performance as it:

  • Avoids spinning up a new process (CPU & memory cost)
  • Likely better leverages caches since this process already loaded this domain in the past

Due to current limitations, using a SuspendedPage's proxy may consume the SuspendedPage, which
means that it can no longer be used for PageCache on history navigations. We need to do this when
the SuspendedPageProxy in question was created for the current WebPageProxy because:

  • This SuspendedPageProxy's process already has a suspended WebPage with this WebPageProxy's pageID

and

  • We do not currently support having more than one WebPage with a given pageID within a single WebProcess.
  • UIProcess/SuspendedPageProxy.cpp:

(WebKit::SuspendedPageProxy::SuspendedPageProxy):

  • UIProcess/SuspendedPageProxy.h:
  • UIProcess/WebProcessPool.cpp:

(WebKit::WebProcessPool::processForNavigationInternal):

Tools:

Add API test coverage.

  • TestWebKitAPI/Tests/WebKitCocoa/ProcessSwapOnNavigation.mm:
8:46 AM Changeset in webkit [237734] by Tadeu Zagallo
  • 4 edits in trunk/Source/JavaScriptCore

Metadata should not be copyable
​https://bugs.webkit.org/show_bug.cgi?id=191193

Reviewed by Keith Miller.

We should only ever hold references to the entry in the metadata table.

  • bytecode/CodeBlock.cpp:

(JSC::CodeBlock::finalizeLLIntInlineCaches):

  • dfg/DFGByteCodeParser.cpp:

(JSC::DFG::ByteCodeParser::parseBlock):

  • generator/Metadata.rb:
8:41 AM Changeset in webkit [237733] by pvollan@apple.com
  • 2 edits in trunk/LayoutTests

[Win] Layout Test fast/block/basic/inline-content-with-floating-image.html is failing
​https://bugs.webkit.org/show_bug.cgi?id=191194

Unreviewed test gardening.

  • platform/win/TestExpectations:
8:39 AM Changeset in webkit [237732] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebCore

Remove ENABLE_OPENCL fully
​https://bugs.webkit.org/show_bug.cgi?id=191172

Patch by Rob Buis <​rbuis@igalia.com> on 2018-11-02
Reviewed by Yusuke Suzuki.

No new tests since no change in functionality.

  • platform/graphics/filters/FilterEffect.h:

(WebCore::FilterEffect::hasResult const):
(WebCore::FilterEffect::applyAll):
(WebCore::FilterEffect::openCLImage): Deleted.
(WebCore::FilterEffect::setOpenCLImage): Deleted.

7:23 AM Changeset in webkit [237731] by Alan Bujtas
  • 7 edits
    2 adds in trunk

[LFC][IFC] Add support for intrinsic width calculation
​https://bugs.webkit.org/show_bug.cgi?id=191144

Reviewed by Antti Koivisto.

Source/WebCore:

This is the inline formatting version of the shrink-to-fit computation. It generates inline runs
and uses InlineLineBreaker to compute min/max width. This is very similar to ::layout.

Test: fast/inline/simple-shrink-to-fit-inline-block.html

  • layout/inlineformatting/InlineFormattingContext.cpp:

(WebCore::Layout::InlineFormattingContext::layout const):
(WebCore::Layout::InlineFormattingContext::computeWidthAndMargin const):
(WebCore::Layout::InlineFormattingContext::computeHeightAndMargin const):
(WebCore::Layout::InlineFormattingContext::layoutFormattingContextRoot const):
(WebCore::Layout::InlineFormattingContext::computeWidthAndHeightForReplacedInlineBox const):
(WebCore::Layout::InlineFormattingContext::collectInlineContent const):
(WebCore::Layout::InlineFormattingContext::instrinsicWidthConstraints const):
(WebCore::Layout::InlineFormattingContext::computeWidthAndHeightForInlineBox const): Deleted.

  • layout/inlineformatting/InlineFormattingContext.h:

Tools:

  • LayoutReloaded/misc/LFC-passing-tests.txt:

LayoutTests:

  • fast/inline/simple-shrink-to-fit-inline-block-expected.txt: Added.
  • fast/inline/simple-shrink-to-fit-inline-block.html: Added.
  • platform/ios/TestExpectations:
7:21 AM Changeset in webkit [237730] by Alan Bujtas
  • 2 edits in trunk/Source/WebCore

[LFC][BFC] BlockFormattingContext::instrinsicWidthConstraints logic should look similar to ::layout
​https://bugs.webkit.org/show_bug.cgi?id=191181

Reviewed by Antti Koivisto.

  • layout/blockformatting/BlockFormattingContext.cpp:

(WebCore::Layout::BlockFormattingContext::instrinsicWidthConstraints const):

6:41 AM Changeset in webkit [237729] by Alan Bujtas
  • 9 edits in trunk/Source/WebCore

[LFC] Align shrink-to-fit width computation with the layout implementation.
​https://bugs.webkit.org/show_bug.cgi?id=191179

Reviewed by Antti Koivisto.

There are many similarities between layout and shrink-to-fit.
They both operate on formatting roots only (shrink-to-fit -> float, out-of-flow, inline-block) and in both cases
the algoritm depends on what type of formatting context the element establishes.

This patch is in preparation for transforming the "shrink-to-fit" width computation to make it behave more like layout.
With this change the instrinsicWidthConstraints() computation happens in the formatting context that the element establishes (similar to layout).

  • layout/FormattingContext.cpp:

(WebCore::Layout::FormattingContext::computeOutOfFlowHorizontalGeometry const):

  • layout/FormattingContext.h:

(WebCore::Layout::FormattingContext::Geometry::outOfFlowHorizontalGeometry):
(WebCore::Layout::FormattingContext::Geometry::floatingWidthAndMargin):
(WebCore::Layout::FormattingContext::Geometry::outOfFlowNonReplacedHorizontalGeometry):
(WebCore::Layout::FormattingContext::Geometry::floatingNonReplacedWidthAndMargin):

  • layout/FormattingContextGeometry.cpp:

(WebCore::Layout::FormattingContext::Geometry::shrinkToFitWidth):
(WebCore::Layout::FormattingContext::Geometry::outOfFlowNonReplacedHorizontalGeometry):
(WebCore::Layout::FormattingContext::Geometry::floatingNonReplacedWidthAndMargin):
(WebCore::Layout::FormattingContext::Geometry::outOfFlowHorizontalGeometry):
(WebCore::Layout::FormattingContext::Geometry::floatingWidthAndMargin):

  • layout/blockformatting/BlockFormattingContext.cpp:

(WebCore::Layout::BlockFormattingContext::computeWidthAndMargin const):
(WebCore::Layout::BlockFormattingContext::instrinsicWidthConstraints const):

  • layout/blockformatting/BlockFormattingContext.h:
  • layout/inlineformatting/InlineFormattingContext.cpp:

(WebCore::Layout::InlineFormattingContext::layoutFormattingContextRoot const):
(WebCore::Layout::InlineFormattingContext::instrinsicWidthConstraints const):

  • layout/inlineformatting/InlineFormattingContext.h:
  • layout/inlineformatting/InlineFormattingContextGeometry.cpp:

(WebCore::Layout::InlineFormattingContext::Geometry::inlineBlockWidthAndMargin):

3:46 AM Changeset in webkit [237728] by Tadeu Zagallo
  • 3 edits in trunk/Source/JavaScriptCore

REGRESSION(r237547): Exception handlers should be aware of wide opcodes when JIT is disabled
​https://bugs.webkit.org/show_bug.cgi?id=191175

Reviewed by Keith Miller.

​https://bugs.webkit.org/show_bug.cgi?id=191108 did not handle the case where JIT is not enabled

  • jit/JITExceptions.cpp:

(JSC::genericUnwind):

  • llint/LLIntData.h:

(JSC::LLInt::getWideCodePtr):

3:04 AM Changeset in webkit [237727] by Philippe Normand
  • 5 edits in trunk

[Apache] Self-signed SSL certificate RSA key is considered too weak
​https://bugs.webkit.org/show_bug.cgi?id=191156

Reviewed by Michael Catanzaro.

Tools:

  • Scripts/webkitpy/common/system/pemfile.py:

LayoutTests:

Regenerate the self-signed SSL certificate with a 2048 bits RSA key.

  • http/conf/webkit-httpd.pem:
12:54 AM Changeset in webkit [237726] by graouts@webkit.org
  • 17 edits in trunk

[Web Animations] Make document.getAnimations() return declarative animations in the correct order
​https://bugs.webkit.org/show_bug.cgi?id=191153

Reviewed by Dean Jackson.

LayoutTests/imported/mozilla:

Mark progressions for the document.getAnimations() tests covering CSS Animations and CSS Transitions.
These tests are now in WPT and will be updated as part of a general WPT update for all of Web Animations
shortly. We will upstream the minor changes made at that point. Those changes were necessary because we
don't support the PseudoElement interface and instead we check which animation names and CSS properties
are applied to pseudo-elements rather than the pseudo-element's type.

  • css-animations/test_document-get-animations-expected.txt:
  • css-animations/test_document-get-animations.html:
  • css-animations/test_event-dispatch.html: Remove a comment that was added by mistake in a previous patch.
  • css-transitions/test_document-get-animations-expected.txt:
  • css-transitions/test_document-get-animations.html:

Source/WebCore:

The Web Animations spec has the notion of "composite order" which determines the order in which animations should
be returned when document.getAnimations() is called. The CSS Transitions and CSS Animations specifications also
determine the composite order of their respective animation classes, as well as the notion of "owning element",
the element that was the animation target when specified through style, prior to any manipulation via the Web
Animations API. We implement these two notions so that we can pass the document.getAnimations() tests for
declarative animations.

It is important that we have the notion of an "owning element" since when a CSS Transition or CSS Animation is
modified via the Web Animations API in a way that an animation created through CSS we must consider no longer
as a declarative animation but as a regular Web Animation. In this patch, we remove the DeclarativeAnimation's
target(), which returned a reference, to owningElement(), which returns a pointer and return nullptr once the
declarative animation has been modified.

In order to also keep a correct count of declarative animations applied to elements, we correctly add transitions
that have completed to a list of completed transitions, as specified by the CSS Transitions spec. We have had this
list declared in AnimationTimeline.h for a while but never actually did the work to add items to it. Thanks to that,
AnimationTimeline::updateCSSTransitionsForElement() now correctly accounts for completed transitions so that they
may be canceled if overridden with a new animation, correctly removing their "owning element".

Finally, now that we have saner lists of animations per classes, we can apply the right sorting routines to match
the composite order for CSS Transitions, CSS Animations and Web Animations, keeping a list of all animations created
in order as the final criterion for sorting.

  • animation/AnimationTimeline.cpp:

(WebCore::AnimationTimeline::animationTimingDidChange): Make sure this animation is tracked on the list of all animations
created for this timeline in the order in which they were created, which is needed to sort animations correctly when
document.getAnimations() is called.
(WebCore::AnimationTimeline::animationWasAddedToElement): We've removed the relevantMapForAnimation() method which we used
to determine which map we should add an animation to based on its class and instead have code that accounts for not just
the animation's class, but also whether it has an owning element since a valid owning element is required to qualify as
a CSS Transition or CSS Animation, regardless of the animation's original class.
(WebCore::removeAnimationFromMapForElement): Adding this helper to remove an animation from the provided animation map so
that animationWasRemovedFromElement() can call this with all of the various animation maps.
(WebCore::AnimationTimeline::animationWasRemovedFromElement): Check all of the various animation maps to see which may
contain the animation we're trying to remove as the owning element might have been cleared by the time this function is
called and looking at the animation's class would not be sufficient to determine which animation map the animation was in.
(WebCore::AnimationTimeline::removeDeclarativeAnimationFromListsForOwningElement): New function in which the logic from
animationWasRemovedFromElement() that dealt with removing animations from the list of running CSS Animations/Transitions as
well as completed CSS Transitions was factored out. This allowed us to also call this function from
DeclarativeAnimation::disassociateFromOwningElement().
(WebCore::AnimationTimeline::elementWasRemoved): We no longer need to remove an animation as canceling it will remove it
correctly when DocumentTimeline::updateAnimationsAndSendEvents() is called.
(WebCore::AnimationTimeline::updateCSSAnimationsForElement): Call cancelDeclarativeAnimation() instead of the removed
cancelOrRemoveDeclarativeAnimation() when a CSS Animation should be canceled.
(WebCore::AnimationTimeline::ensureRunningTransitionsByProperty): Now that we correctly remove transitions from the list
of running transitions once they've completed or have been canceled, we need a helper to always get a valid map of running
transitions for a given element as that map can be cleared while updateCSSTransitionsForElement() is running.
(WebCore::AnimationTimeline::updateCSSTransitionsForElement): Call cancelDeclarativeAnimation() instead of the removed
cancelOrRemoveDeclarativeAnimation() when a CSS Transition should be canceled. Additionally we always get the list of running
transitions for a given element as it can be cleared by a prior cancelation.
(WebCore::AnimationTimeline::cancelDeclarativeAnimation): New function replacing cancelOrRemoveDeclarativeAnimation() in which
we call the new DeclarativeAnimation::cancelFromStyle() function on the provided animation and then remove the animation from
all known lists, including the new list of all animations. We do this final part so that the animation gets re-added as if it
were a new animation since canceling a declarative animation via a style change removes its declarative-ness. This guarantees
that a declarative animation that is canceled through style but then resumed through the Web Animations API sorts after any
declarative animation with an owning element.
(WebCore::AnimationTimeline::relevantMapForAnimation): Deleted.
(WebCore::AnimationTimeline::cancelOrRemoveDeclarativeAnimation): Deleted.

  • animation/AnimationTimeline.h:

(WebCore::AnimationTimeline::timingModelDidChange): Deleted. This was left over in the patch where we implemented the "update
animations and send events" procedure.
(WebCore::AnimationTimeline::animations const): Deleted.

  • animation/CSSAnimation.cpp:

(WebCore::CSSAnimation::create): Some refactoring to make the handling of a declarative animation's owning element part of the
DeclarativeAnimation constructor.

  • animation/CSSTransition.cpp:

(WebCore::CSSTransition::create): Some refactoring to make the handling of a declarative animation's owning element part of the
DeclarativeAnimation constructor.

  • animation/DeclarativeAnimation.cpp:

(WebCore::DeclarativeAnimation::DeclarativeAnimation):
(WebCore::DeclarativeAnimation::tick): Make sure we disassociate from the animation's owning element once we transition from a
relevant state to an idle state. This will catch any changes made via the Web Animations API to a declarative animation when the
"update animations and send events" procedure is run.
(WebCore::DeclarativeAnimation::disassociateFromOwningElement): Remove this animation from the list of declarative animations on
the associated timeline and make owningElement() nullptr so that document.getAnimations() will know to sort this animation with other
Web Animations created via the Web Animations API.
(WebCore::DeclarativeAnimation::initialize):
(WebCore::DeclarativeAnimation::cancelFromStyle): New method called from AnimationTimeline::cancelDeclarativeAnimation() which
cancels the animation but also disassociates it from its owning element.
(WebCore::DeclarativeAnimation::invalidateDOMEvents):
(WebCore::DeclarativeAnimation::enqueueDOMEvent):

  • animation/DeclarativeAnimation.h:

(WebCore::DeclarativeAnimation::owningElement const):
(WebCore::DeclarativeAnimation::target const): Deleted.

  • animation/DocumentTimeline.cpp:

(WebCore::compareDeclarativeAnimationOwningElementPositionsInDocumentTreeOrder): Adding a new utility used when sorting both
CSS Transitions and CSS Animations by tree order when their owning element differ, with special logic for pseudo-elements.
(WebCore::DocumentTimeline::getAnimations const): Filter and sort animations according to their composite order.
(WebCore::DocumentTimeline::updateAnimationsAndSendEvents): Compile a list of transitions that move to a finished state to
the list of completed transitions so that they may be canceled correctly in AnimationTimeline::updateCSSTransitionsForElement().
(WebCore::DocumentTimeline::transitionDidComplete):

  • animation/DocumentTimeline.h:

LayoutTests:

Mark that the two document.getAnimations() tests for declarative animations are no longer flaky.

12:15 AM Changeset in webkit [237725] by zandobersek@gmail.com
  • 2 edits in trunk/Source/WebCore

PNGImageDecoder: report no repetition for non-animated images
​https://bugs.webkit.org/show_bug.cgi?id=191068

Reviewed by Michael Catanzaro.

When building with APNG support enabled, the
PNGImageDecoder::repetitionCount() method until now reported infinite
repetition count for all PNG images, even the ones that were not
animated. This is now changed so that no repetition is reported for
non-animated images.

  • platform/image-decoders/png/PNGImageDecoder.cpp:

(WebCore::PNGImageDecoder::repetitionCount const):

12:11 AM Changeset in webkit [237724] by zandobersek@gmail.com
  • 2 edits in trunk/Source/WebDriver

[WPE] WebDriver: support validating and parsing the 'certificates' capability
​https://bugs.webkit.org/show_bug.cgi?id=191014

Reviewed by Michael Catanzaro.

Adopt the same approach as in the GTK's WebDriverService implementation
for validating and parsing the 'certificates' WebDriver capability.

  • wpe/WebDriverServiceWPE.cpp:

(WebDriver::WebDriverService::platformValidateCapability const):
(WebDriver::WebDriverService::platformParseCapabilities const):

12:09 AM Changeset in webkit [237723] by Justin Fan
  • 21 edits
    9 copies
    2 moves
    5 adds in trunk

[WebGPU] Experimental prototype for MSL shaders
​https://bugs.webkit.org/show_bug.cgi?id=191084

Reviewed by Dean Jackson.

Source/WebCore:

Begin implementation for WebGPUDevice and WebGPUShaderModule and associated descriptor objects.

Test: webgpu/webgpu-basics.html
Test: webgpu/shader-modules.html

  • CMakeLists.txt:
  • DerivedSources.make:
  • Modules/webgpu/GPUDevice.cpp:

(WebCore::GPUDevice::createShaderModule const):

  • Modules/webgpu/GPUDevice.h:

(WebCore::GPUDevice::platformDevice const):

  • Modules/webgpu/GPUShaderModule.h:

(WebCore::GPUShaderModule::platformShaderModule const):

  • Modules/webgpu/GPUShaderModuleDescriptor.h:
  • Modules/webgpu/WebGPU.cpp:

(WebCore::WebGPU::requestAdapter const):

  • Modules/webgpu/WebGPUAdapter.cpp:

(WebCore::WebGPUAdapter::create):
(WebCore::WebGPUAdapter::createDevice):

  • Modules/webgpu/WebGPUAdapter.h:
  • Modules/webgpu/WebGPUAdapter.idl:
  • Modules/webgpu/WebGPUDevice.cpp:

(WebCore::WebGPUDevice::create):
(WebCore::WebGPUDevice::WebGPUDevice):
(WebCore::WebGPUDevice::createShaderModule const):

  • Modules/webgpu/WebGPUDevice.h:
  • Modules/webgpu/WebGPUDevice.idl:
  • Modules/webgpu/WebGPUShaderModule.cpp:

(WebCore::WebGPUShaderModule::create):
(WebCore::WebGPUShaderModule::WebGPUShaderModule):

  • Modules/webgpu/WebGPUShaderModule.h:
  • Modules/webgpu/WebGPUShaderModule.idl:
  • Modules/webgpu/WebGPUShaderModuleDescriptor.h:
  • Modules/webgpu/WebGPUShaderModuleDescriptor.idl:
  • Modules/webgpu/WebGPUSwapChain.cpp:

(WebCore::WebGPUSwapChain::configure):

  • Modules/webgpu/WebGPUSwapChain.h:
  • Modules/webgpu/WebGPUSwapChain.idl:
  • Modules/webgpu/cocoa/GPUDeviceMetal.mm:

(WebCore::GPUDevice::create):
(WebCore::GPUDevice::GPUDevice):

  • Modules/webgpu/cocoa/GPUShaderModuleMetal.mm:

(WebCore::GPUShaderModule::create):
(WebCore::GPUShaderModule::GPUShaderModule):

  • Sources.txt:
  • SourcesCocoa.txt:
  • WebCore.xcodeproj/project.pbxproj:
  • bindings/js/WebCoreBuiltinNames.h:
  • platform/Logging.h:

Source/WTF:

Disabling WebGPU on non-Cocoa platforms and iOS Simulator.

  • wtf/Platform.h:

LayoutTests:

  • webgpu/shader-modules.html: Added.
  • webgpu/shader-modules-expected.html: Added.
  • webgpu/webgpu-basics.html: Added.
  • webgpu/webgpu-basics-expected.html: Added.
12:04 AM Changeset in webkit [237722] by zandobersek@gmail.com
  • 2 edits in trunk/Tools

Unreviewed follow-up to r237678.

  • wpe/jhbuild.modules: The patch element apparently has to be a child

of the branch in order for Jhbuild to properly detect and apply the
patch file.

Note: See TracTimeline for information about the timeline view.